Note: Where available, the PDF/Word icon below is provided to view the complete and fully formatted document
Cashless debit card extended following positive independent evaluation



Download PDFDownload PDF

The Hon Christian Porter MP Minister for Social Services

The Hon Alan Tudge MP Minister for Human Services

MEDIA RELEASE 14 March 2017

CASHLESS DEBIT CARD EXTENDED FOLLOWING POSITIVE INDEPENDENT EVALUATION The Government has agreed to extend the Cashless Debit Card trial sites in Ceduna, South Australia and East Kimberley, Western Australia due to the strong independent evaluation results, released today and in consultation with community leaders.

The Wave 1 Report of the independent evaluation being undertaken by ORIMA Research concluded that “overall, the [trial] has been effective to date… in particular, the trial has been effective in reducing alcohol consumption, illegal drug use and gambling - establishing a clear ‘proof-of-concept’.”

The Cashless Debit Card aims to reduce the devastating effects of welfare fuelled alcohol, drug and gambling abuse. Over time it is hoped the card will assist people to break the cycle of welfare dependency by stabilising their lives and helping them into employment.

The Report found “most stakeholders felt that excessive alcohol consumption was at a “crisis point” and was having wide-ranging negative impacts on individuals, their families and the community.”

Under the current trial, 80 per cent of welfare payments are placed onto a recipient's card, with the remaining 20 per cent placed into their regular bank account.

The trial has consisted of 3 parts - a Cashless Debit Card, comprehensive support services to help people break their addictions, and a community leadership group to guide the design and implementation.

The Report outlines key results across the two trial sites including:

 Alcohol - on average, of trial participants surveyed who reported that they do drink alcohol, 25% of participants and 13% of family members reported drinking alcohol less frequently, whilst 25% of participants reported engaging in binge drinking less frequently.

 Gambling - on average, of trial participants surveyed who reported they do gamble, 32% of participants and 15% of family members reported gambling less.

 Drug use - on average, of trial participants surveyed who reported using illegal drugs before the trial commenced, 24% reported using illegal drugs less often.

In addition, the evaluation data states a significant proportion (31%) of the participants surveyed indicated they had been better able to care for children and save more money.

2

Reductions in alcohol consumption, illegal drug use and gambling have been “largely driven by the impact of the debit card quarantining mechanism and not by the additional services provided,” according to the Report.

The Report supports other data from local partners and anecdotal feedback:

 The number of pick-ups made by the Kununurra Miriwoong Community Patrol Service for Alcohol in January 2017 was 19 per cent lower than in January 2016.

 Monthly poker machine revenue in Ceduna and surrounding local government areas in January 2017 is 12 per cent lower compared to January 2016.

 Admissions to the Wyndham Sobering-Up Unit in September 2016 were 49 per cent lower than before the trial began in September 2015.

 The senior medical officer in the East Kimberley has reported a “dramatic reduction in alcohol related presentations to the emergency department”

 The Ceduna mayor says that “it is the quietest the town has been.”

 Retailers in both sites report an increase in white goods, clothes, food and household items purchased since the introduction of the card.

Minister for Human Services, Alan Tudge, worked with the community leaders on the design and implementation of the trial and believes the results support an extension of the card.

“The card is a not a panacea, but it has led to stark improvements in these communities. There are very few other initiatives that have had such impact.

“A large part of the success has been the close working relationship with local leaders, who have co-designed and implemented the trial with us. The South Australian and Western Australian State Governments have also been very supportive.

“There is still a lot of work to do, but if we can continue on this path, then over time we can make these communities safe, healthy and prosperous once again,” Minister Tudge said.

The extension of the card will allow the Government to make fully informed decisions about the future of welfare conditionality. The final evaluation report by ORIMA Research is due mid-2017.

Attachments: Initial Conditions Report; Wave 1 Interim Evaluation Report - ORIMA Research

Media contacts: Kevin Donnellan (Minister Porter) 0408 491 956

Rachelle Miller (Minister Tudge) 0475 804 886

Department of Social Services

Evaluation of the Cashless Debit Card Trial - Initial Conditions Report

Table of Contents

Executive Summary ..................................................................................... i

A. Introduction .............................................................................................................. i

B. Demographic profile of the trial communities ......................................................... i

C. Initial data - debit card roll out ............................................................................... ii

D. Stakeholder views of pre-CDCT conditions ............................................................. ii

E. Awareness, understanding and expectations of the CDCT .................................... iv

F. Conclusions .............................................................................................................. v

I. Introduction .......................................................................................... 1

A. Background .............................................................................................................. 1

B. Evaluation framework ............................................................................................. 2

C. Qualitative research methodology .......................................................................... 3

D. Presentation of findings ........................................................................................... 5

E. Quality assurance..................................................................................................... 5

II. Demographic profile of the trial communities ....................................... 6

A. About this chapter ................................................................................................... 6

B. Total population....................................................................................................... 6

C. Indigenous population ............................................................................................. 6

D. Labour force status .................................................................................................. 7

E. Age distribution ....................................................................................................... 8

F. Early Childhood Development ................................................................................. 8

G. Income distribution ............................................................................................... 10

III. Initial data - cashless debit card roll out ............................................. 11

A. About this chapter ................................................................................................. 11

B. Progressive roll out ................................................................................................ 11

C. Proportion of CDCT trial site populations with a CDC ........................................... 12

D. Income Support Payments (ISPs) paid via the CDC ............................................... 15

IV. Stakeholder views of pre-CDCT conditions .......................................... 20

A. About this chapter ................................................................................................. 20

B. Alcohol consumption and impacts ........................................................................ 20

C. Illicit drug consumption and impacts .................................................................... 24

D. Gambling activity and impact ................................................................................ 25

E. Awareness and usage of support services ............................................................. 27

F. Crime, safety and security ..................................................................................... 29

G. Other significant community experiences and concerns ...................................... 32

H. Summary ratings of initial conditions .................................................................... 34

V. Awareness, understanding and expectations of the CDCT ................... 35

A. About this chapter ................................................................................................. 35

B. Stakeholders’ awareness and understanding of CDCT .......................................... 35

C. ISP recipients’ awareness and understanding of CDCT ......................................... 37

D. Stakeholders’ expectations of the CDCT ............................................................... 37

VI. Baseline Administrative Data .............................................................. 40

VII. Conclusion .......................................................................................... 42

Appendices

Appendix A: Evaluation Framework

i

Executive Summary

A. Introduction

The Australian Government is undertaking a Cashless Debit Card Trial (CDCT) to deliver and manage income support payments (ISPs), with the aim of reducing levels of community harm related to alcohol consumption, drug use and gambling.

In the CDCT, a proportion of an individual’s ISP is directed to a restricted bank account, accessed by a debit card (not allowing cash withdrawals). Participation in the CDCT is mandatory for all working age ISP recipients who live in the selected trial sites. In addition, wage earners, Age Pensioners and Veterans Affairs Pensioners who live in the trial sites can opt-in to the CDCT.

To date, the CDCT is being implemented in Ceduna and Surrounds in South Australia and Kununurra / Wyndham (East Kimberley) in Western Australia.

The Department of Social Services (DSS) commissioned ORIMA Research to conduct an independent evaluation of the CDCT. This report is focused on identifying initial conditions prevailing in the trial sites before the implementation of the CDCT.

The report is primarily based on the findings of qualitative research (interviews and focus groups) with key stakeholders in each of the trial sites. It also includes some coverage of administrative data that was available at the time of report writing.

A total of 37 stakeholders (members of regional leadership groups as well as government and non-government service providers) participated in the qualitative research, which was conducted between 21 April and 26 May 2016.

B. Demographic profile of the trial communities

The 2011 Census found that the total population of Ceduna and Surrounds was 4,221, of which 2,289 people lived in the town of Ceduna. The total population of the East Kimberley was 6,950, including:

 5,525 people living in Kununurra, and

 1,003 people living in Wyndham.

Around one-third of the population in each trial area identified as being of Aboriginal and / or Torres Strait Islander origin in the 2011 Census, compared with 2.7% of the overall Australian population. Most of the Indigenous people (62%) in Ceduna and Surrounds lived in communities outside of the Ceduna urban area.

ii

C. Initial data - debit card roll out

Cashless debit cards (CDCs) were progressively distributed to eligible ISP recipients in Ceduna and the East Kimberley. CDCs were distributed to eligible ISP recipients mainly between mid-April and end-May 2016 in Ceduna and over the month of June 2016 in East Kimberley.

As at 4 October 2016, 785 residents of Ceduna and Surrounds (around 26% of the total working age population) and 1,225 residents of East Kimberley (26% of the working age population) had received an ISP via a CDC1.

In both locations, 42% of Aboriginal and / or Torres Strait Islander residents had received an ISP via a CDC compared with around 5% of non-Indigenous residents. This reflects the fact that a large majority of ISP recipients (73% in Ceduna and Surrounds and 86% in East Kimberley)2 were Indigenous people. The disproportionately high share of Indigenous people in the ISP recipient population reflected their relatively high levels of socio-economic disadvantage.

D. Stakeholder views of pre-CDCT conditions

Alcohol consumption and impacts

Overall, the research found that alcohol consumption was the most concerning issue for stakeholders across both trial sites, in comparison to gambling and drug use. Most stakeholders felt that excessive alcohol consumption was at a “crisis point”, and was having wide-ranging negative impacts on individuals, their families and the community.

A few stakeholders believed that the levels of alcohol consumption had reduced since the introduction of alcohol restrictions in these communities in late 2015. However, a few other stakeholders felt that such reductions were likely to be only temporary based on the perceived impacts of previous alcohol restriction arrangements3.

Illicit drug consumption and impacts

Overall, stakeholders across both trial locations reported that, in comparison to alcohol consumption, usage of illicit drugs was less widespread. Although most stakeholders considered the excessive consumption of alcohol to be a greater issue, they still reported that drug use was of concern as they saw it as a potential issue that was likely to increase into the future.

1 Statistics based on Indue Cashless Debit Card transactions data 2 Source: Department of Social Services, March 2015. 3

These perceptions were consistent with the findings of two recent reviews of alcohol sale restrictions in Ceduna and Kununurra/ Wyndham, which found evidence of significant levels of circumvention (particularly creation/ growth of a secondary supply market) and displacement (e.g. shift of drinking from public to private spaces) activities: Office of the Liquor and Gambling Commissioner of South Australia (2011), Report on Liquor Sales within the Ceduna Region; Western Australian Alcohol and Drug Authority (2012), The Impact of Liquor Restrictions in Kununurra and Wyndham (Twelve Month Review).

iii

Marijuana was reported as being the most commonly used drug (other than alcohol). In comparison to alcohol and other illicit forms of drugs, stakeholders felt that marijuana had less of an impact on the wider community as it tended not to lead to “aggressive” and violent behaviours.

Overall, amphetamine usage was reported by stakeholders as being less common than marijuana. However, many stakeholders (especially in Ceduna) indicated that usage of amphetamines, in particular methamphetamine (i.e. “ice”), had increased over the last 12 months as it had become more readily available. Stakeholders reported that due to their higher cost, amphetamines were mainly used by adults who were working full-time. Whilst not widely used drugs, amphetamines were considered to be particularly harmful (especially when “mixed” with alcohol) as use often resulted in aggressive and violent behaviours, and thus the impacts on others in the community were perceived to be quite severe.

Gambling activity and impact

Overall, most stakeholders in Ceduna and a few stakeholders in Kununurra and Wyndham reported that excessive gambling was prevalent in their community. Gambling behaviours differed between the two sites, with gambling via electronic gaming machines (“pokies”) prevalent in Ceduna, but not available in Kununurra and Wyndham. Excessive gambling in the East Kimberley was perceived by stakeholders there to be primarily based on informal gambling activities (e.g. card games).

The research found that most stakeholders in Kununurra and Wyndham did not hold serious concerns about the impacts of gambling in their communities, particularly compared to that of alcohol. In contrast, many stakeholders in Ceduna felt that gambling (particularly the “pokies”) was a serious issue in their community, similar to alcohol consumption.

Awareness and usage of support services

Overall, stakeholders in both trial areas reported that there was a large number of family and support services available in their community. These included:

 Short term / relief services - e.g. accommodation services (e.g. the Sobering Up Unit and the town camps), meal services, food vouchers and food hampers, and shower and laundry services;

 Longer term rehabilitation and counselling services - drug and alcohol counselling and rehabilitation, financial counselling and planning, and family counselling; and

 Indigenous specific and mainstream services.

Most stakeholders also felt that there was good awareness of these services in the community, including amongst trial participants.

The inclusion of additional services (particularly drug and alcohol, mental health and financial counselling services) as part of the CDCT was considered “very important” in ensuring that adequate care and support was provided for CDCT participants - especially, for those who may experience “withdrawals” as a result of reduced alcohol / drug consumption. However, the research found that, at the time of the research, there was limited awareness amongst most stakeholders about what extra services would be provided / funded in the CDCT - particularly in Kununurra and Wyndham, where stakeholders from support services reported they had not been informed regarding any additional funding, and were unsure whether or not they would receive extra funding.

iv

Crime, safety and security

Overall, across both trial sites stakeholders indicated that the excessive use of alcohol, drugs and / or gambling contributed to high levels of crime and / or violence in their communities. Most stakeholders felt that alcohol was the predominant cause of many of these behaviours, particularly those where violence was involved (e.g. assaults).

Most stakeholders also perceived that the general sense of safety and security in their communities had gradually eroded, predominantly due to the excessive consumption of alcohol and its resulting impacts. Drug use and excessive gambling were also identified as contributing factors.

Many stakeholders felt that criminal and violent behaviours were under-reported and unprosecuted in the trial sites. As such, they believed that the crime statistics for the trial sites would be considerably lower than the actual number of incidents occurring on a daily basis. Additionally, some felt that crime statistics were likely to reflect policing strategies (e.g. periodic focus on specific criminal issue / “blitzes”) and as such may not accurately reflect the true nature of criminal incidents in the communities.

Other significant community experiences and concerns

The research found significant concern among many stakeholders about the social, financial, housing and schooling impacts on their communities as a result of excessive alcohol consumption (and to a lesser extent illicit drug use and gambling).

E. Awareness, understanding and expectations of the CDCT

The research found that there was generally good awareness and understanding of the CDCT amongst stakeholders in the trial sites. Community leaders tended to have a better and more detailed understanding of the CDCT processes than other stakeholders.

Most stakeholders felt that the CDCT had been well communicated, overall, to their organisation by DSS and felt adequately informed. However, a few stakeholders in Ceduna felt that services in adjacent / nearby areas needed to be better informed about the trial. These stakeholders reported knowing of some services in nearby areas that had dealings with trial participants who had left Ceduna, but had not been aware of the trial.

Stakeholders reported that while most ISP recipients had known that the CDCT was occurring, many had shown limited interest in the trial and had not attended information sessions that were held prior to the rollout. As a result, stakeholders indicated that some trial participants had a limited understanding about the details of card usage and logistics.

The research also identified a number of stakeholder concerns around implementation issues / difficulties with the debit card, which was being rolled out during the time of the research fieldwork. These concerns related to card activation, impact on attendance at cash-only events, communicating the CDCT to clients in remote communities and with limited literacy, ability to facilitate private rental arrangements and funds transfer / direct debit limitations.

v

Across both trial locations, most stakeholders felt strongly that there was a need for something to be done to address the high levels of alcohol consumption and, to a lesser extent, illicit drug usage and gambling in the community and their associated harms. Many also felt that a new approach was required to address these issues as current and previous programs and services had not reduced these behaviours. As such, most stakeholders were broadly supportive of the CDCT. However, perceptions in relation to the likely effectiveness of the trial were mixed.

F. Conclusions

The initial conditions qualitative research with stakeholders in Ceduna, Wyndham and Kununurra found widespread local concern about high levels of alcohol consumption and, to a lesser extent, illicit drug use and gambling activity.

Stakeholders indicated that these issues had become progressively worse over the past 5-10 years and that the local communities were experiencing significant adverse impacts.

In particular, most stakeholders felt that excessive alcohol consumption was at a “crisis point”, and was having wide-ranging negative impacts on individuals, their families and the community.

Most stakeholders who participated in the research felt strongly that there was a need for something to be done to address these issues and were broadly supportive of the CDCT.

1

I. Introduction

A. Background

The Australian Government is undertaking a Cashless Debit Card Trial (CDCT) to deliver and manage income support payments (ISPs), with the aim of reducing levels of community harm related to alcohol consumption, drug use and gambling. This initiative has been informed by a recommendation in Andrew Forrest’s Creating Parity report4. It has also been informed by lessons learned from previous income management trials.

In the CDCT, a proportion of an individual’s ISP is directed to a restricted bank account, accessed by a debit card (not allowing cash withdrawals). Participation in the CDCT is mandatory for all working age ISP recipients who live in the selected trial sites. In addition, wage earners, Age Pensioners and Veterans Affairs Pensioners who live in the trial sites can opt-in to the CDCT.

To date, the CDCT is being implemented in Ceduna and Surrounds in South Australia and Kununurra / Wyndham (East Kimberley) in Western Australia. The sites were proposed by community leaders in these locations and the CDCT has been developed via a collaborative process involving local community leaders, local and state government agencies and Australian Government agencies (led by the Department of Social Services - DSS).

The two CDCT sites have experienced high levels of community harm related to alcohol consumption, drug use and gambling. In its submission to a Senate Committee Inquiry into the Social Security Legislation Amendment (Debit Card Trial) Bill 2015, the Ceduna District Council noted that its “community has a long-standing problem associated with substance abuse, particularly of alcohol. In common with some other communities we also have issues with drug and gambling addiction.”5 Similarly, recent WA State Government agency reports have identified relatively high levels of harm related to alcohol consumption and drug use in the East Kimberley (and wider Kimberley) region, including:

 Between 2005 and 2009, per capita alcohol-related hospitalisations for the Shire of Wyndham-East Kimberley were 4.7 times higher than the WA State average6;

 Between 1999 and 2007, per capita alcohol-caused deaths in the Kimberley region were 2.9 times higher than the State average7; and

 In 2013, the per capita incidence of drug offences in the Kimberley region was 1.7 times higher than the State average8.

4 Forrest, A. (2014). The Forrest Review: Creating Parity. Commonwealth of Australia, Canberra. 5 District Council of Ceduna (2015), Submission to Senate Standing Committee on Community Affairs,

Inquiry into the Social Services Legislation Amendment (Debit Card Trial) Bill 2015, 18 September 2015 6 WA Drug and Alcohol Office and Epidemiology Branch of Department of Health WA (2011), Alcohol-

related hospitalisations and deaths: Kimberley 7 Epidemiology Branch of Department of Health WA and the Cooperative Research Centre for Spatial

Information (2011), Health status report on alcohol deaths - drug-caused for the Kimberley Health Region.

2

While only around one-third of the population in each trial area identified as being of Aboriginal and / or Torres Strait Islander origin in the 2011 Census, a large majority of ISP recipients (73% in Ceduna and Surrounds and 86% in East Kimberley)9 are Indigenous people. The disproportionately high share of Indigenous people in the ISP recipient population reflects their relatively high levels of socio-economic disadvantage. In turn, these reflect a range of general, long-term historical factors in Australia that have driven significant gaps between the education, health, social and economic outcomes for non-Indigenous Australians and those for Indigenous Australians10.

The main elements of the CDCT include:

 A cashless card, delivered by a commercial provider (Indue Ltd);

 80% of income support payments to be placed into a restricted account linked to the cashless card (100% of lump sum payments and arrears payments);

 The percentage of funds accessible in an unrestricted manner (e.g. as cash) may be varied by local community panels;

 Alcohol and gambling (excluding lotteries) will not be able to be purchased with the card, and no cash will be able to be withdrawn from the card;

 CDCT participants who move away from the trial sites will remain participants in the CDCT; and

 Up to three sites will operate for 12 months, with a staggered rollout from March 2016.

DSS commissioned ORIMA Research to conduct an independent evaluation of the CDCT. This report presents the initial findings of that evaluation process. It is focused on identifying initial conditions prevailing in the trial sites before the implementation of the CDCT.

B. Evaluation framework

This report is the first in a series of three evaluation reports. It is primarily based on the findings of qualitative research (interviews and focus groups) with key stakeholders in each of the trial sites. It also includes some coverage of administrative data that was available at the time of report writing.

The evaluation will be based on evidence collected via a range of data sources, including:

 Three waves of qualitative research with on the ground stakeholders (i.e. initial conditions, wave 1 and wave 2);

 Two waves of post-implementation quantitative research amongst CDCT participants and their families, as well as non-participant community members;

 Department of Human Services (DHS) administrative data;

 State government secondary data; and

 Unidentifiable data from the DSS welfare card ‘inbox’ and hotline - provided to ORIMA Research via summary tables and de-identified comments.

8 WA Drug and Alcohol Office (2015), Alcohol and Other Drug Indicators Report - Kimberley Health Region 9 Source: Department of Social Services, March 2015. 10

See Department of the Prime Minister and Cabinet (2016), Closing the Gap Prime Minister’s Report.

3

It should be noted that pre-implementation baseline primary research with potential CDCT participants and the broader community was not possible due to the timing of evaluation commissioning and the time required to obtain ethical clearance from a Human Research Ethics Committee prior to the conduct of such research.

The Evaluation Framework (presented in Appendix A) for the project outlines in detail the evaluation’s scope, key questions concerning impacts and higher-level process issues, the evaluation design and methodologies, data sources and specific data to be used or generated by this project.

C. Qualitative research methodology

A total of 37 stakeholders participated in the qualitative research, which was conducted between 21 April and 26 May 2016 (across Ceduna, Wyndham and Kununurra) via:

 Two focus groups with members of the regional leadership groups;

 Nineteen face-to-face interviews with members of the regional leadership groups and stakeholders from government and non-government service providers; and

 Ten telephone interviews with members of the regional leadership groups and stakeholders from government and non-government service providers.

Table 1 shows the research design and locations adopted for the research. Table 2 and Table 3 overleaf present the full list of organisations that research participants represented.

Table 1: Qualitative research design - initial conditions research

Research location: Ceduna Kununurra / Wyndham TOTAL

Regional leadership group representatives

3 x IDIs

1 x Telephone IDI

n=4

2 x FG

n=8

4 x IDIs

n=4

2 x FG

n=8

7 x IDIs

1 x Telephone IDI

n=7

Service provider representatives

5 x IDIs

6 x Telephone IDIs

n=11

7 x IDIs

3 x Telephone IDIs

n=9

12 x IDIs

9 x Telephone IDIs

n=20

Total number of groups / interviews 8 x IDIs 7 x Telephone IDIs n=15

2 x FG 11 x IDIs

3 x Telephone IDIs n=22

2 x FG 19 x IDIs

10 x Telephone IDIs n=37

4

Table 2: Qualitative research - list of organisations that research participants represented - Ceduna and Surrounds

Table 3: Qualitative research - list of organisations that research participants represented - Kununurra / Wyndham

11 Minimum of 3 contact attempts made via phone and / or email.

Interviewed Contacted11 - not interviewed

 Regional leadership group representatives  Ceduna Aboriginal Corporation  Scotdesco  District Council of Ceduna  Families SA  Ceduna Area School  Aboriginal Drug and Alcohol Council  Save the Children  Centacare Catholic Family Care  Housing SA  Ceduna Youth Club  SA Police: Ceduna  Foodland  Not willing to be identified x 2

 Koonibba Community  Oak Valley (Maralinga) Inc.  Yalata Community  Red Cross

 Ceduna Hospital  Family Violence Legal Service  Ceduna Koonibba Aboriginal Health Service  Complete Personnel  Ngura Yadurirn Children and Family Centre  Eyre Futures

Interviewed Contacted - not interviewed

 Regional leadership group representatives  Kununurra Empowered Communities  Wunan Foundation  MG Corporation  Waringari Aboriginal Corporation  Kununurra Chambers of Commerce and

Industry

 WA Police: Kununurra  Kimberley Mental Health and Drug Service  Dept of Corrective Service - Youth

Justice Services  Save the Children  Kununurra Local Drug Action Group  Department of Social Services  Ngnowar Aerwah Aboriginal

Corporation  Wyndham Early Learning Activity Centre  Wyndham District High School  WA Police: Wyndham  Wyndham District Hospital  Shire of Wyndham East Kimberley

 Gelganyem Trust  Kununurra District Hospital  Community Housing Limited  St John’s Ambulance  WA Housing  Kimberley Community Legal Services Inc.  Wyndham Community Club  East Kimberley Job Pathways

5

D. Presentation of findings

The research was qualitative in nature and hence, the results and findings are presented in a qualitative manner. This research approach does not allow for the exact number of participants holding a particular view on individual issues to be measured. This report, therefore, provides an indication of themes and reactions among research participants rather than exact proportions of participants who felt a certain way. The following terms used in the report provide a qualitative indication and approximation of size of the target audience who held particular views:

 Most—refers to findings that relate to more than three quarters of the research participants;

 Many—refers to findings that relate to more than half of the research participants;

 Some—refers to findings that relate to around a third of the research participants; and

 A few—refers to findings that relate to less than a quarter of research participants.

The most common findings are reported except in certain situations where only a minority has raised particular issues, but these are nevertheless considered to be important and to have potentially wide-ranging implications / applications.

Quotes have been provided throughout the report to support the main results or findings under discussion.

We acknowledge and understand that Aboriginal and / or Torres Strait Islander people is the preferred term when referring to Indigenous Australians. However, in this report we have opted to use the term Indigenous participants when referring to Aboriginal and / or Torres Strait Islander participants for brevity of readership.

E. Quality assurance

The project was conducted in accordance with international quality standard ISO 20252 and the Australian Privacy Principles contained in the Privacy Act 1988 (Cth.).

6

II. Demographic profile of the trial communities

A. About this chapter

This chapter presents contextual demographic data for the Ceduna and East Kimberley CDCT sites. All data presented in the chapter has been sourced from the last ABS Census (2011).

B. Total population

The 2011 Census found that the total population of Ceduna and Surrounds12 was 4,221, of which 2,289 people lived in the town of Ceduna.

The total population of the East Kimberley13 was 6,950, including:

 5,525 people living in Kununurra; and

 1,003 people living in Wyndham.

In both Ceduna and Surrounds (50.3%) and East Kimberley (52.6%), the proportion of residents who were male was a little higher than the national average (49.4%).

C. Indigenous population

Of the people living in Ceduna and Surrounds during the 2011 Census, 4,015 (95%) stated whether or not they were of Aboriginal and / or Torres Strait Islander origin. Of this group, 1,245 (31%) identified as being of Aboriginal and / or Torres Strait Islander origin. Most of these Indigenous people (773) lived in communities outside of the Ceduna urban area. Twenty one per cent of the Indigenous population in the area spoke a language other than English and 1% did not speak English well.

Of the 6,950 people living in the East Kimberley during the 2011 Census, 6,304 (91%) stated whether or not they were of Aboriginal and / or Torres Strait Islander origin. Of this group, 2,068 (33%) identified as being of Aboriginal and / or Torres Strait Islander origin. Sixteen per cent of the Indigenous population in the area spoke a language other than English and 1% did not speak English well.

12 Ceduna and Surrounds is comprised of the Local Government Area of Ceduna and the following Geographical Areas from Statistical Area Level 1 (SA1s): 40601113409, 40601113410, 40601113501 and 40601113502.

13 East Kimberley is comprised of the following SA1s: 5120801, 5120802, 5120804, 5120805, 5120807, 5120808, 5120810, 5120811, 5120812, 5120814, 5120815, 5120816, 5120817, 5120818.

7

Nationally, 2.7% of the Australian population identified as being of Aboriginal and/ or Torres Strait Islander origin in the 2011 Census. Thirteen per cent of the Australian Indigenous population spoke a language other than English and 2% did not speak English well.

D. Labour force status

The 2011 ABS Census found that of the working age population14 living in Ceduna and Surrounds:

 65% were employed;

 3% were unemployed; and

 32% were not in the labour force.

Of the working age population living in the East Kimberley:

 74% were employed;

 3% were unemployed; and

 22% were not in the labour force.

Nationally, in the 2011 Census, 61% of the Australian working age population were employed, 4% were unemployed and 35% were not in the labour force.

In Ceduna and Surrounds, the agriculture sector was the largest employer of non-Indigenous persons (followed by health care and social assistance), while the health care and social assistance sector was the largest employer of Indigenous persons.

In the East Kimberley, the construction sector was the largest employer of non-Indigenous persons (followed by health care and social assistance), while the health care and social assistance sector was the largest employer of Indigenous persons.

14 Working age population is defined for the purposes of this report as people aged 15-64 years.

8

E. Age distribution

Figure 1: Age Distribution — Population residing in CDCT trial sites below shows that the population of Ceduna and Surrounds in 2011 had a similar age distribution to that of Australia as a whole, while that of East Kimberley had a relatively high proportion of people of working age.

Figure 1: Age Distribution — Population residing in CDCT trial sites

Source: ABS Census 2011.

F. Early Childhood Development

The Australian Early Development Census (AEDC) is conducted every three years and has occurred in 2009, 2012 and 2015. The AEDC measures the development of children in Australia in their first year of full-time school. The AEDC is considered to be a measure of how well children and families are supported from conception through to school age.

AEDC data is collected using an Early Development Instrument (completed by each child’s teacher) that consists of approximately 100 questions across five key domains, which are closely linked to:

1. physical health and wellbeing;

2. social competence;

3. emotional maturity;

4. language and cognitive skills (school-based); and

5. communication skills and general knowledge.

9

AEDC domain scores are calculated for each domain for each child where enough valid responses have been recorded.15 In 2009, domain cut-off scores were established and children falling below the 10th percentile in a domain are categorised as 'developmentally vulnerable'. The percentage of children assessed as developmentally vulnerable on two or more domains provides a summary indicator of developmental vulnerability of young children in the community or population group being considered.

Table 4 summarises the AECD findings for the CDCT sites and provides corresponding national, Indigenous and non-Indigenous findings. It shows that in 2015:

 Indigenous children accounted for 5.5% of Australian children in their first year of school who were assessed in the AEDC process

 26.2% of Indigenous children were assessed as developmentally vulnerable in two or more AEDC domains, compared to 10.2% of non-Indigenous children

 the proportion of children assessed as developmentally vulnerable in two or more domains in East Kimberley (27.0%) and Ceduna & surround (19.6%) were higher than hypothetical rates (of 19.6% and 17.3%, respectively) controlling for the higher-than-average proportion of Indigenous children in these communities.16

Table 4: 2015 AEDC findings

Population group / Community Children with valid domain scores (#)

Indigenous children (%)

Developmentally vulnerable in two or more domains (DV2%)

Hypothetical DV2% allowing for indigenous % (%)

Australia 286,616 5.5 11.1 -

Indigenous Children 15,875 100 26.2 -

Non-Indigenous Children 270,741 0 10.2 -

East Kimberly (CDCT Site) 126 58.7 27 19.6

Ceduna & Surrounds (CDCT Site) 56 44.6 19.6 17.3

Source: AEDC Community Profile and National Report, 2015.

15 Domain information about children with special needs is not included in the AEDC results. 16 These hypothetical DV2% rates were derived by assuming that Indigenous and non-Indigenous DV2%

rates in these communities were equal to the national averages (of 26.2% and 10.2%, respectively) and applying the Indigenous children proportions for each of these communities.

10

G. Income distribution

Figure 2 below shows that the population of Ceduna and Surrounds in 2011 had a total annual personal income distribution that was skewed towards lower and middle income brackets compared to that of Australia as a whole. In contrast, the income distribution of East Kimberley was skewed towards higher income brackets.

Figure 2: Total Annual Personal Income Distribution — Population residing in CDCT trial sites

Source: ABS Census 2011.

0%

2%

4%

6%

8%

10%

12%

14%

16%

$104,000 or more $78,000-$103,999 $65,000-$77,999 $52,000-$64,999 $41,600-$51,999 $31,200-$41,599 $20,800-$31,199 $15,600-$20,799 $10,400-$15,599

$1-$10,399 Nil income

Negative income

Australia

Ceduna & Surrounds

East Kimberly

11

III. Initial data - cashless debit card roll out

A. About this chapter

This chapter presents initial data concerning the distribution of cashless debit cards to eligible persons in the Ceduna and East Kimberley CDCT sites. All data presented in the chapter has been sourced from the Department of Human Services.

B. Progressive roll out

Cashless debit cards (CDCs) were progressively distributed to eligible Income Support Payment (ISP) recipients in Ceduna and the East Kimberley. Figure 3 below shows that CDCs were distributed to eligible ISP recipients mainly between mid-April and end-May 2016 in Ceduna and over the month of June 2016 in East Kimberley.

Figure 3: Number of persons paid an ISP via a CDC

Source: Department of Human Services.

As at 2 October 2016, a total of 2,115 persons had been paid an ISP via a CDC of which:

 757 were residents of Ceduna and Surrounds;

 43 were residents of Ceduna and Surrounds at the time of CDC eligibility assessment (15 March 2016) and had subsequently moved out of area;

 1,247 were residents of the East Kimberley;

 63 were residents of the East Kimberley at the time of CDC eligibility assessment (26 April 2016) and had subsequently moved out of area; and

 1,181 (56%) were female and 934 (44%) were male.

12

C. Proportion of CDCT trial site populations with a CDC

As at 4 October 2016, 757 residents of Ceduna and Surrounds had received an ISP via a CDC - this represents:

 around 18% of the total resident population of Ceduna and Surrounds17, and

 around 27% of the total working age18 resident population of Ceduna and Surrounds19

As at 4 October 2016, 1,247 residents of the East Kimberley had received ISPs via a CDC - this represents:

 around 18% of the total resident population of East Kimberley20, and

 around 25% of the total working age resident population of East Kimberley21.

Figure 4 shows the proportion of CDCT area residents who had received an ISP via a CDC by age group.

17 In 2011, the population of Ceduna and Surrounds was 4,221 persons. 18 In this report, working age population has been defined as those aged 15 to 64 years. 19

In 2011, the working age population of Ceduna and Surrounds was 2,823 persons 20 In 2011, the population of East Kimberley was 6,950 persons. 21

In 2011, the working age population of the East Kimberley was 4,982 persons.

13

Figure 4: Proportion of CDCT Area Residents paid an ISP via a CDC, by Age Group

Figure 5 shows the proportion of CDCT area residents who had received an ISP via a CDC by Indigenous status.

 In Ceduna and Surrounds, 45% of Aboriginal and / or Torres Strait Islander residents had received an ISP via a CDC compared with 6% of non-Indigenous residents.

 In the East Kimberley, 24% of Aboriginal and / or Torres Strait Islander residents had received an ISP via a CDC compared with 10% of non-Indigenous residents.

10%

37%

28%

21%

8%

10%

27%

22% 23%

11%

0%

5%

10%

15%

20%

25%

30%

35%

40%

Less than 25 25 - 35 35 - 45 45 - 60 Over 60

Ceduna & Surrounds East Kimberley

14

Figure 5: Proportion of CDCT Area Residents paid an ISP via a CDC, by Indigenous Status

45%

6%

11%

49%

5% 4%

0%

10%

20%

30%

40%

50%

60%

Indigenous Non-Indigenous Unknown

Ceduna & Surrounds East Kimberley

15

D. Income Support Payments (ISPs) paid via the CDC

As at 4 October, a total of $10.4 million of Income Support Payments (ISPs) had been paid via the CDC to CDCT participants, while around $2.6 million22 of ISPs were accessible as cash23.

Figure 6 shows that as at 4 October 2016:

 CDCT participants in Ceduna and surrounds were paid at total of $4.1 million of ISPs via their CDCs and around $1.0 million accessible as cash; and

 CDCT participants in East Kimberley were paid a total of $6.2 million of ISPs via their CDCs and $1.6 million accessible as cash.

Figure 6: Income Support Payments (ISPs) via CDC (as at 4 October 2016)

ISP paid via CDC ISP accessible as cash* Total ISP

Ceduna & surrounds $4,130,112 $1,032,528 $5,162,639

East Kimberley $6,229,961 $1,557,490 $7,787,451

Total $10,360,072 $2,590,018 $12,950,090

22 Assumes all CDC holders have default 80% of ISP paid via CDC. As at 4 October 2016: 100% of CDCT participants in the East Kimberly had the default 80% of their ISP paid via their CDC; and over 95% of CDCT participants in Ceduna and surrounds had the default 80% of their ISP paid via the CDC (with the

remaining 5% being paid between 50% and 70% of their ISP via their CDC - yielding an average proportion of ISP paid via the CDC of 79%). 23 Excludes royalty payments and other cash payments (e.g. emergency assistance payments) made to CDCT

participants. The Department identified five (5) CDCT participants who reported royalty payments totalling around $10,000 per annum, which represents less than 0.1% of total ISPs to the CDCT communities.

$4,130,112

$6,229,961

$0

$3,000,000

$6,000,000

$9,000,000

Ceduna & surrounds East Kimberley

ISP paid via CDC ISP accessible as cash*

16

*based on assumption that all CDC participants have default 80% of ISP paid via CDC.

Figure 7 and Figure 8 show the total value of ISPs delivered via the CDC (as at 4 October 2016) to CDCT participants in Ceduna and surrounds and East Kimberley, respectively.

Figure 7: ISPs via CDC, Ceduna & surrounds, by ISP type

*Other ISP types paid via CDC include: Family Tax Benefit; Sickness Allowance; ABSTUDY; Partner Allowance; Widow Allowance; and Maternity Immunisation Allowance.

Figure 8: ISPs via CDC, East Kimberley, by ISP type

*Other ISP types paid via CDC include: ABSTUDY; Widow Allowance; Sickness Allowance; Age Pension; and Maternity Immunisation Allowance.

$1,834,644

$1,801,659

$1,333,362

$520,454

$333,270

$188,869

$168,777

$48,925

$0 $500,000 $1,000,000 $1,500,000 $2,000,000

Parenting Payment Single (n=198)

Newstart Allowance (n=525)

Disability Support Pension (n=249)

Parenting Payment Partnered (n=68)

Carer Payment (n=44)

Family Tax Benefit (n=24)

Youth Allowance (n=98)

Other* (n=19)

17

Figure 9 shows the total value of ISPs delivered via the CDC (as at 4 October 2016) to CDCT participants with ATSI and non-ATSI status.

Figure 9: ISPs via CDC, by ATSI status

Ceduna & surrounds East Kimberley Total

ATSI CDC holders $3,101,160 $5,185,600 $8,286,760

Non-ATSI CDC holder $929,453 $954,275 $1,883,729

ATSI status unknown $99,499 $90,085 $189,584

Total $4,130,112 $6,229,961 $10,360,072

$3,101,160

$5,185,600

$929,453

$954,275

$0

$1,000,000

$2,000,000

$3,000,000

$4,000,000

$5,000,000

$6,000,000

$7,000,000

Ceduna & surrounds East Kimberley

ATSI CDC holders Non-ATSI CDC holder ATSI status unknown

18

Figure 10 shows the total value of ISPs (as at 4 October 2016) to male and female CDCT participants. Around two-thirds of ISPs were paid to female CDCT participants:

 66% of ISPs paid in Ceduna were to female CDCT participants; and

 66% of ISPs paid in East Kimberley were to female CDCT participants.

Figure 10: ISPs via CDC, by CDC holder gender

Ceduna & surrounds East Kimberley Total

Female $2,714,504 $4,324,262 $7,038,766

Male $1,415,608 $1,905,699 $3,321,307

Total $4,130,112 $6,229,961 $10,360,072

Figure 11 and Figure 12 shows the total value of ISPs (as at 4 October 2016)

 Ceduna and surrounds; and

 East Kimberley

19

Figure 11: ISPs via CDC, Ceduna and surrounds, by CDC holder age

Figure 12: ISPs via CDC, East Kimberley, by CDC holder age

20

IV. Stakeholder views of pre-CDCT conditions

A. About this chapter

This chapter presents research findings relating to stakeholders’ views about the on-the-ground conditions before the Cashless Debit Card Trial (CDCT) across the trial locations.

It presents stakeholders’ observations and perceptions about alcohol consumption and its impacts, illicit drug usage and its impact, and gambling and its impacts. The chapter presents their views about the communities’ awareness and usage of on-the-ground support services, as well as observations about crime, safety and security. Finally, the chapter covers stakeholders’ perceptions about other significant community experiences and concerns.

B. Alcohol consumption and impacts

Overall, the research found that alcohol consumption was the most concerning issue for stakeholders across both trial sites, in comparison to gambling and drug use. Most stakeholders felt that excessive alcohol consumption was at a “crisis point”, and was having wide-ranging negative impacts on individuals, their families and the community

“It’s a social catastrophe, nearly everything is linked to alcohol. It’s at the core of nearly every problem”—Kununurra

Overall, most stakeholders reported that excessive consumption of alcohol was prevalent in their town, and that this had increased over time (particularly in the past 5-10 years). Stakeholders stated that this was evidenced by:

 Visible public drunkenness;

 Increasing numbers of people who needed help and assistance as a result of alcohol use;

 Family and community concern and discussions about the adverse effects of high alcohol consumption; and

 Extremely high blood alcohol content (BAC) readings of people presenting to hospitals.

“I’ve been here 40 years and I’ve seen the rise of chronic alcohol abuse”—Ceduna

“0.3, 0.4 alcohol [blood alcohol concentration] doctors say they should technically be dead”— Ceduna

A few stakeholders believed that the levels of alcohol consumption had reduced since the introduction of alcohol restrictions in these communities. However, a few other stakeholders felt that such reductions were likely to be only temporary, based on the perceived impacts of previous alcohol restriction arrangements24. The restriction arrangements reported by stakeholders included:

24 These perceptions were consistent with the findings of two recent reviews of alcohol sale restrictions in Ceduna and Kununurra/ Wyndham, which found evidence of significant levels of circumvention

21

 Dry areas and alcohol restrictions in Ceduna25 - the Ceduna District Council, South Australian Police and local alcohol licensees and the Office of the Liquor and Gambling Commissioner have introduced a range of measures in relation to responsible service, sale and consumption of alcohol. These include the introduction of Dry Areas, restrictions on sales of certain types of alcohol and the introduction of ID Tect machines; and

 The Takeaway Alcohol Management System (TAMS) in Wyndham and Kununurra26 - the Kununurra / Wyndham Alcohol Accord has implemented a 12 month trial of TAMS which began on 14 December 2015. This system limits individuals’ daily alcohol purchases by using scanning technology of their personal identification.

“TAMS has made a difference - I base that on how people behave around the ‘Big Croc’, that’s mainly where antisocial behaviour occurs”—Wyndham

Stakeholders reported that excessive alcohol consumption was common amongst people from:

 Both genders;

 Both Indigenous and non-Indigenous backgrounds;

 A range of ages - stakeholders reported that while frequent excessive alcohol consumption was more common amongst those over 18, it was common for children to begin drinking at 14-15 years of age with the behaviour increasing as they got older; and

 Local ‘dry’ communities - who typically travelled into town in order to consume alcohol.

“It’s not a racial thing... non-Indigenous have their issues too”—Wyndham

“They’ll be out camping and drinking and then they’ll come into Ceduna pissed”—Ceduna

“Port was the drink of choice - as it’s cheap and very high alcohol content”—Ceduna

“Through circumstance there’s a lot of drinking in public. People who are itinerate don’t have a place to stay and drink”—Ceduna

For people consuming excessive amounts of alcohol, stakeholders reported that the most commonly consumed beverages were low cost and had high alcohol content. However, when lump sum payments were available in the community (e.g. royalties) more expensive beverages were consumed, particularly spirits.

Overall, stakeholders reported that alcohol consumption amongst Indigenous community members was more noticeable than that of non-Indigenous community members. This was attributed to a more communal nature of drinking among Indigenous community members, as well as housing

(particularly creation/ growth of a secondary supply market) and displacement (e.g. shift of drinking from public to private spaces) activities: Office of the Liquor and Gambling Commissioner of South Australia (2011), Report on Liquor Sales within the Ceduna Region; Western Australian Alcohol and Drug Authority (2012), The Impact of Liquor Restrictions in Kununurra and Wyndham (Twelve Month Review). 25

More detailed information about dry areas and alcohol restrictions in Ceduna and surrounding areas can be found at http://www.ceduna.sa.gov.au/dryzoneandalcoholrestrictions. 26 More detailed information about TAMS in Kununurra / Wyndham and surrounding areas can be found at

http://www.rgl.wa.gov.au/maps/Restrictions/KununurraWyndham.pdf and http://www.swek.wa.gov.au/tams.aspx.

22

constraints (e.g. overcrowding) and the prevalence of people from outside the trial sites visiting these communities. This meant that there was a greater likelihood for Indigenous community members to consume alcohol in public and / or highly visible places such as parks, on the street and at “party houses” (i.e. houses where large groups of people regularly congregated to consume alcohol). In contrast, stakeholders reported that non-Indigenous community members generally consumed alcohol privately.

“There’d be white fellas that are as big an alcoholic but you don’t see them because they’ve got a home and they’re more private”—Ceduna

As such, while stakeholders reported that the consequences of drinking were similar and evident across both groups, they generally found it more difficult to report on the specific timing and frequency of alcohol consumption of non-Indigenous people.

Stakeholders reported that there were several patterns of excessive alcohol consumption that were observed amongst Indigenous community members, including:

 Regular binge drinking - i.e. consuming very large amounts of alcohol several times a week;

 Irregular binge drinking - stakeholders indicated that there were some people living in Indigenous communities who stayed in town for periods of time to drink before returning to their dry communities and / or ‘going bush’ and / or visiting others; and

 Dependant / continuous drinking - stakeholders reported that there were some community members who were highly dependent on alcohol and “continuously” intoxicated.

“… some people would be intoxicated all the time… they’re seasoned, hardened alcoholics”— Ceduna

In terms of the timing of alcohol consumption, stakeholders reported that:

 Excessive drinking occurred throughout the week;

 However, in Kununurra and Wyndham stakeholders reported a decrease in consumption during periods when bottle shops were closed (e.g. on Sundays and during police-enforced Liquor Act closures); and

 It was common for people to begin drinking as soon as bottle shops opened in the morning (i.e. 10am) - this was also observed by researchers conducting fieldwork at the trial sites.

“Bottle shops are closed on Sundays and this place is a ghost town then”— Wyndham

Stakeholders reported that while the level of alcohol consumption was high throughout the year, it tended to increase further during events (e.g. football games) and communal gatherings (e.g. funerals).

“For funerals people drink to excess”—Wyndham

Stakeholders across the trial sites reported that in their communities, the excessive consumption of alcohol caused:

 a range of injuries, both directly to the individual consuming the alcohol and to others;

 longer-term adverse health impacts to the individual and their unborn / new born and older children; and

23

 negative social impacts for the community.

“Tourists were scared, on the grey nomad networks they were saying ‘give Ceduna a miss’”— Ceduna

The following alcohol-related injuries to the individual were commonly reported by stakeholders:

 Fatalities and injuries sustained while intoxicated (e.g. from falls and fights);

 Fatalities and injuries as a result of drunk driving or being near moving vehicles while intoxicated;

 Exacerbation of mental illness - leading to self-harm and suicide; and

 Alcohol poisoning.

“There’s been a lot of road accidents with people travelling to get more alcohol”—Wyndham

“There’s a high level of suicide attempts, a couple of people every week”—Kununurra

Alcohol-related injuries to others were frequently reported by stakeholders as being sustained from intoxicated people in the community via widespread:

 Domestic violence;

 Rape and sexual violence, assaults and abuse; and

 Physical “outbursts” / assaults / violent behaviour.

“Family violence is huge… people get pissed and fight in front of the kids”—Ceduna

The research also identified a range of longer-term health impacts of excessive alcohol consumption based on feedback from stakeholders, including:

 Memory loss and “confusion”;

 Physical illness / conditions - e.g. cancers, high blood pressure, kidney damage, liver failure, stomach / digestion problems, diabetes, etc;

 Mental illness / conditions - e.g. anxiety and depression; and

 Foetal Alcohol Syndrome and learning difficulties in children.

“Because of the amount of liver failure there’s quite a number of people in dialysis here in Wyndham and Kununurra”—Wyndham

“We see kids with foetal alcohol syndrome too… it affects their behaviour and their concentration at school”—Kununurra

24

C. Illicit drug consumption and impacts

Overall, stakeholders across both trial locations reported that, in comparison to alcohol consumption, usage of illicit drugs was less widespread. Although most stakeholders considered the excessive consumption of alcohol to be a greater issue, they still reported that drug use was of concern as they saw it as a potential issue that was likely to increase into the future.

Marijuana was reported as being the most commonly used drug (other than alcohol). Stakeholders indicated that usage was relatively widespread and that the drug had “always been present” in communities. Marijuana usage was reported as being more prevalent amongst younger community members, including those below 18 years of age. Some stakeholders reported incidents of children using marijuana from as young as 10 years of age. In comparison to alcohol and other illicit forms of drugs, stakeholders felt that marijuana had less of an impact on the wider community as it tended not to lead to “aggressive” and violent behaviours. Rather, a few stakeholders reported that marijuana use was more commonly associated with low levels of motivation to find paid employment, as well as low levels of school engagement and performance amongst children.

“Number one is alcohol and then cannabis”—Kununurra

“Ganja has been around and a lot of people handle it”—Wyndham

“Marijuana is a big thing, the kids are using it”—Wyndham

“Marijuana mellows you out”—Kununurra

Overall, amphetamine usage was reported by stakeholders as being less common than marijuana. However, many stakeholders (especially in Ceduna) indicated that usage of amphetamines, in particular methamphetamine (i.e. “ice”), had increased over the last 12 months as it had become more readily available. Stakeholders reported that due to its higher cost, amphetamines were mainly used by adults who were working full-time. Whilst not a widely used drug, it was considered to be particularly harmful (especially when “mixed” with alcohol) as it often resulted in aggressive and violent behaviours, and thus the impacts on others in the community were perceived to be quite severe.

“Ice has started to creep in”—Kununurra

“It’s tradies [using amphetamines] you know, young guys who earn a lot of money”—Ceduna

“Ice takes everything to a different level with the aggression”—Ceduna

Based on feedback received from stakeholders, amphetamines appeared to be more widely available and used in Ceduna than in the other two trial sites. The research suggested that this was due to Ceduna’s proximity to the highway, which stakeholders reported as providing a “convenient supply route” into the area.

In contrast, in the smaller, more isolated communities (i.e. outside of Kununurra and Ceduna) amphetamine use was reported as being uncommon and generally not perceived to be an issue of concern. Stakeholders noted that availability of amphetamines was limited in smaller, more remote communities. Furthermore, they felt that the small size and isolated nature of these communities

25

made it relatively easy for authorities and community leaders to control and monitor the drug situation. In addition, these communities generally had less money available to purchase higher cost substances.

“Being a small community, everyone knows what’s going on”—Wyndham

A few instances of heroin usage were cited by some stakeholders, however usage of this form of illicit substance was perceived to be less prevalent in comparison to usage of other substances - primarily due to cost reasons.

Stakeholders reported that drug use was evident in both the Indigenous and non-Indigenous populations in the communities.

While stakeholders reported that some children were using marijuana in public areas, overall drug usage generally did not occur in public places as consumption was illegal. As the usage of the other forms of illicit drugs was less visible, stakeholders were less able to comment on consumption patterns.

Stakeholders that worked in drug related support services noted that amongst users of drugs, consumption tended to be regular and ongoing due to dependency. The binge patterns that were evident with alcohol were generally not seen.

D. Gambling activity and impact

Overall, most stakeholders in Ceduna and a few stakeholders in Kununurra and Wyndham reported that excessive gambling was prevalent in their community. The South Australian Attorney General reports that poker machine revenue in the Ceduna region for August 2014 was $437,64627, which was estimated to be 5.5% of Ceduna’s total monthly income28 .

Gambling behaviours differed between the two sites, with gambling via electronic gaming machines (‘pokies’) prevalent in Ceduna, but not available in Kununurra and Wyndham. Excessive gambling in the East Kimberley was perceived by stakeholders there to be primarily based on informal gambling activities (e.g. card games). The research found that most stakeholders in Kununurra and Wyndham did not hold serious concerns about the impacts of gambling in their communities, particularly compared to that of alcohol. In contrast, many stakeholders in Ceduna felt that gambling (particularly the ‘pokies’) was a serious issue in their community, similar to alcohol consumption.

“Gambling pales in comparison to alcohol”—Kununurra

“A lot of people here wish the town never got pokies”—Ceduna

Stakeholders reported that excessive behaviours were evidenced by:

 Individuals’ reported expenditure on gambling when presenting to financial counselling services;

 Individuals accessing support services to meet basic needs (e.g. meals and food vouchers);

27 Source: SA Attorney-General 28 Source: ABS National Regional Profile: Ceduna (last updated 23/09/2016) - Estimates of Personal Income

- Total annual income (excl. Government pensions and allowances): $95m

26

 Individuals in government assisted housing not being able to meet rental repayments;

 Presentation of unsupervised children at support services;

 Observed neglect of children (e.g. children not being adequately fed and cases of children being locked in cars during gambling sessions); and

 Direct observation of extended amounts of time spent gambling (e.g. individuals arriving at pokie venues and remaining all day).

“We’ve had some clients who admit they go to the TAB and we have to work out a budget”— Wyndham

“They will call us for an order from the supermarket to feed their families”—Ceduna

“There have been cases where kids have been locked in cars and the police have been called”— Ceduna

Generally, stakeholders who worked in financial and family support services and / or who dealt with clients’ financial issues (e.g. housing services) were better able to assess the impact that excessive gambling had on individuals. Many other stakeholders found it difficult to comment on the impacts of excessive gambling, as they were not privy to individuals’ financial circumstances.

Overall, stakeholders reported that both unregulated and regulated gambling were common amongst adults in their towns, including:

 Unregulated card games - reported by stakeholders as being more common amongst older Indigenous females;

 Electronic gaming machines - which only occurred in Ceduna due to government restrictions in WA which meant that these were not available in Kununurra and Wyndham. Usage of pokies was reported as being high across all demographic types of people (i.e. males vs females, young vs older, Indigenous vs non-Indigenous);

 ‘Scratchies’ - appeared to be more common amongst females;

 TAB - reported as being more common amongst males;

 Online gambling (e.g. sports betting) - which was commonly reported in Ceduna as an activity that was more prevalent amongst males.

“All the elderly ones do gambling, but it’s just cards”—Wyndham

“Also scratchies, people buying hundreds of dollars of them and scratching them on the footpath”—Kununurra

“There’s gamblers who get up at midnight and gamble online… by the next morning they’ve lost all their money”—Ceduna

The research found that there was not always a linkage between excessive gambling behaviours and excessive alcohol consumption and / or drug use. Stakeholders reported that while some people in the trial sites engaged in excessive gambling in addition to excessive alcohol consumption and / or drug use, others only engaged in excessive gambling behaviours.

27

Stakeholders reported that most regulated forms of gambling were conducted in venues (e.g. pubs and clubs).

In contrast, unregulated gambling (e.g. card games) was reported to occur in private houses - stakeholders noted that there was often an unofficially designated house for such activities. In Kununurra and Wyndham, card games were also reported to take place in parks, which authorities received complaints about from the public.

Most stakeholders reported that gambling behaviours occurred at all times throughout the day and throughout the year. However, the research found that regulated gambling took place less often in Wyndham, as the TAB (the only venue in town) was only open on limited days / time.

“There’s a card house, mainly elderly ladies and it’s very controlled”—Wyndham

“I think gambling at the park has increased”—Kununurra

E. Awareness and usage of support services

Overall, stakeholders in both trial areas reported that there was a large number of family and support services available in their community. These included:

 Short term / relief services - e.g. accommodation services (e.g. the Sobering Up Unit and the town camps), meal services, food vouchers and food hampers, and shower and laundry services;

 Longer term rehabilitation and counselling services - drug and alcohol counselling and rehabilitation, financial counselling and planning, and family counselling; and

 Indigenous specific and mainstream services.

“The obvious ones are the District Health Service, the Aboriginal Health Service - they run the Sobering-up Centre”—Ceduna

Most stakeholders also felt that there was good awareness of these services in the community, including amongst trial participants.

Stakeholders reported that there was high usage of services providing immediate relief. This was believed to be particularly the case among those who consumed excessive amounts of alcohol, drugs or gambled frequently and their families - as many of these community members had limited funds available for basic needs due to these behaviours.

Some stakeholders in Ceduna also noted that the meal and accommodation services were often accessed by people living in the neighbouring Indigenous communities who spent periods in Ceduna on “drinking binges” and facilitated / encouraged this behaviour.

“There’s an argument that says we’ve allowed all the drinkers to come to Ceduna because of all the services… like a holiday camp for drinkers… they’ll take the bits they need, like go to the day centre and get a feed”—Ceduna

In contrast, stakeholders reported that there was generally limited engagement with longer-term assistance / services such as rehabilitation / counselling services. Stakeholders from these services indicated that they received limited self-referrals.

28

Across both trial sites, many stakeholders felt that the available range of services in their towns lacked coordination with each other, and “operated in silos”. This was felt to negatively impact the strategic approach to case management and the pathways into longer-term support programs. However, a few stakeholders in Ceduna felt the implementation of the Ceduna Services Reform had improved the coordination of service delivery and were expecting that this initiative would improve integration of services.

“A lot of services aren’t communicating with partners and other people in the town, they are working in silos”—Wyndham

“There’s much better value coming up from those services now”—Ceduna

The inclusion of additional services (particularly drug and alcohol, mental health and financial counselling services) as part of the CDCT was considered “very important” in ensuring that adequate care and support was provided for CDCT participants - especially, for those who may experience “withdrawals” as a result of reduced alcohol / drug consumption.

However, the research found that, at the time of the research, there was limited awareness amongst most stakeholders about what extra services would be provided / funded in the CDCT - particularly in Kununurra and Wyndham, where stakeholders from support services reported they had not been informed regarding any additional funding, and were unsure whether or not they would receive extra funding.

Drug and alcohol counselling, treatment and rehabilitation

The research found that overall there was limited use of drug and alcohol support services across the trial sites.

For example, many stakeholders in Ceduna reported high usage of daily support services offered by the Sobering-up Unit and Day Centre (e.g. accommodation, shower, laundry and meal services), but very limited usage of the treatment, rehabilitation and counselling services offered by these facilities and other drug and alcohol services.

“No one talks to the counsellors at the day centre, they just go there for a feed and a wash”— Ceduna

Family support

Stakeholders reported some usage of family support services by trial participants. In particular, stakeholders reported high demand and use of:

 Homelessness programs; and

 Domestic violence services.

However, stakeholders reported much of the use of these services occurred as a result of outreach, referrals (e.g. families identified as at-risk of child protection removal) and / or court orders (e.g. to attend counselling services) rather than self-referral.

29

“If we have a domestic at night, we go in the next morning and separate them for 24 hours and to reinforce that we have counselling”—Wyndham

Financial counselling and support

Most stakeholders reported high usage of financial support services that provided immediate relief (e.g. food vouchers and hampers).

“We’ve noticed a steady stream of clients asking for food hampers”—Ceduna

F. Crime, safety and security

Overall, across both trial sites stakeholders indicated that the excessive use of alcohol, drugs and / or gambling contributed to high levels of crime and / or violence in their communities. Most stakeholders felt that alcohol was the predominant cause of many of these behaviours, particularly those where violence was involved (e.g. assaults).

“The police and St John’s workload decreases dramatically on Sundays when the [alcohol] stores are closed which shows alcohol is the key driver”—Kununurra

Most stakeholders also perceived that the general sense of safety and security in their communities had gradually eroded, predominantly due to the excessive consumption of alcohol and its resulting impacts. Drug use and excessive gambling were also identified as contributing factors.

Violent and criminal behaviours

Specific violent and / or criminal behaviours reported in trial sites by stakeholders included:

Assault - which reportedly occurred due to the increased aggression and lowered inhibitions associated with high alcohol consumption and illicit drug use (particularly amphetamines). Types of assaults included:

 Domestic violence / spousal abuse - stakeholders reported that this was very common, and also occurred due to arguments about family finances due to excessive gambling and alcohol consumption. Law enforcement stakeholders indicated that a large volume of their call-outs were related to domestic violence, however they (and other stakeholders) noted that “a lot of domestic violence went unreported” and thus expected that rates would be “significantly higher” than shown by data;

“Domestic violence is the number one issue we deal with… it’s 22% higher than last year”— Kununurra

 Fights between people in the trial sites - stakeholders reported that it was common for fights to breakout between people when intoxicated and that these would often result in physical injuries. Stakeholders reported that some of these fights were part of long-term inter-family disagreements, particularly amongst the Indigenous community members;

“They’re beating the shit out of each other in the main street”—Ceduna

30

 Elder abuse - stakeholders reported that there were some instances where older people in the community were assaulted in an attempt to obtain money, goods or liquor from them;

 Physical abuse of children; and

 Sexual assault and rape.

“Throughout a year a whole heap of young girls being sexually abused and raped”—Wyndham

Burglaries, robberies and thefts - of money, food, liquor, vehicles and personal property. Stakeholders cited that such items were stolen from private properties and shops;

Vandalism - this included property damage, damage to motor vehicles and graffiti;

“We get bursts of graffiti from kids too”—Ceduna

Driving under the influence of drugs and / or alcohol - as discussed earlier in the Chapter, injuries and fatalities from drink driving or pedestrians being intoxicated were reported by many stakeholders as being commonplace;

“A lot don’t drive but they’ll get on the highway when they’ve been drinking all day”—Wyndham

Prostitution - some stakeholders reported that some people in the community resorted to prostitution for additional income in order to gamble and / or purchase alcohol or drugs; and

Public intoxication - this was reported to be widespread across the trial sites. However, some stakeholders in Ceduna felt that the community had “quietened down” since the introduction of alcohol restrictions.

“We noticed a big reduction in alcohol fuelled violence [after] there were some liquor licencing restrictions”—Ceduna

In addition, stakeholders reported that the excessive use of alcohol (and to a lesser extent illicit drug use and gambling) indirectly caused violent and criminal behaviours amongst children / minors in the community. These included:

 Burglaries, robberies and thefts - many stakeholders reported that the children of those who abused alcohol, drugs and / or gambling were involved in stealing money, food or other goods. These stakeholders reported that this occurred as a result of parents spending excessive amounts of money on alcohol, drugs and / or gambling, which left insufficient money for groceries, toys and other necessary household items; and

 Assaults / violent behaviour by children - a few stakeholders reported that children had been violent toward their parents in order to obtain money from them. Stakeholders reported that these children had reduced access to money and goods as a result of their parents’ excessive spending on alcohol, drugs and / or gambling.

“Kids are breaking into homes looking for money”—Wyndham

31

Safety and security

Most stakeholders felt that the excessive consumption of alcohol (and to a lesser extent illicit drug use and gambling) contributed to a low sense of community safety in the trial sites.

“For residents, you don’t feel safe, you’ve got to be constantly aware”—Kununurra

It was reported that members of the community, particularly women, children, elderly people, as well as visitors / tourists to the trial sites often “did not feel safe" as a result of:

 High incidence of violence and crime - as discussed above;

 Large numbers of intoxicated people in the trial sites, who were often “rowdy” (i.e. yelling and / or swearing);

“In the streets you see drunken people yelling and carrying on”—Ceduna

 Humbugging of people at ATMs and / or outside stores - it was reported that tourists were often targeted when they were doing their banking transactions;

“It’s intimidating and frightening, black fellas who are drunk asking you for money and cigarettes”—Ceduna

 Verbal abuse - a few stakeholders reported incidents where they had witnessed or experienced verbal abuse as a result of refusing requests for cash from people in the streets and when local businesses conducted bag checks; and

 Groups of children roaming the streets - some stakeholders reported that there were groups of children / minors who roamed the streets at night as they did not feel safe in their homes due to groups of adults, including strangers, drinking (i.e. at the ‘party houses’ previously mentioned). Stakeholders reported that while the children themselves were not safe, they also made others in the community feel unsafe. These children were also reported as sometimes being involved in crimes.

The kids wander the streets and get up to no good”—Wyndham

“Some of the kids, I don’t think they feel safe at home… that’s why they’re roaming the streets”—Wyndham

“Some people are afraid because of the kids that roam… they’re sometimes involved in incidents”—Kununurra

Crime Statistics

Many stakeholders felt that criminal and violent behaviours were under-reported and unprosecuted in the trial sites. As such, they believed that the crime statistics for the trial sites would be considerably lower than the actual number of incidents occurring on a daily basis. Additionally, some felt that crime statistics were likely to reflect policing strategies (e.g. periodic focus on specific criminal issue / “blitzes”) and as such may not accurately reflect the true nature of criminal incidents in the communities.

32

“It could be the way we’re receiving information about it [domestic violence], we see more reports from third parties”—Ceduna

G. Other significant community experiences and concerns

The research found significant concern among many stakeholders about the social, financial, housing and schooling impacts on their communities as a result of excessive alcohol consumption (and to a lesser extent illicit drug use and gambling).

Many stakeholders commonly noted a range of social impacts associated with excessive alcohol consumption, illicit drug use and / or gambling, including:

 Family arguments, disputes and “fights”;

 Unemployment or under-employment;

 Humbugging; and

 Abuse and / or intimidation of more vulnerable members of the community.

“There’s fighting with the alcohol… assaults”—Wyndham

“My wife came home and said she got harassed out the front of the Foodland”—Ceduna

They also identified financial impacts on the individuals, their families and communities as a result of significant expenditure on alcohol, drugs and / or gambling on an ongoing basis, including:

 Accumulation of and inability to pay fines, which in some cases has led to incarceration of individuals; and

 Inability to fund basic living requirements including food, clothing, hygiene requirements, rent, bills / utilities and transportation.

A lot of money is going on grog which means less money for groceries”—Kununurra

In addition, many stakeholders were concerned about housing challenges facing their communities as a result of overcrowding and inability to meet financial responsibilities associated with securing permanent housing. While some stakeholders felt that poor housing access was due to insufficient affordable housing stock in those communities, others disagreed. They felt that lack of “sobriety”, “clear headed thinking” and “motivation” - as a result of excessive alcohol consumption and / or gambling - had restricted opportunities for employment and financial stability which were perceived to be necessary prerequisites to securing stable housing.

“We need more housing”—Kununurra

Finally, there was widespread concern among most stakeholders about the impact that excessive alcohol consumption (and to a lesser extent illicit drug use and gambling) were having on children.

“It’s dysfunctionality… kids come to school without breakfast and with no lunch packed”— Ceduna

Such concerns were primarily in relation to poor parenting / neglect of family responsibilities and lack of engagement, especially in relation to:

33

 School attendance, engagement and performance;

 Positive parental / familial role-modelling; and

 Being able to properly nurture, care and protect children from harm and abuse associated with lack of safety and security in their environment (as discussed above).

“Kids don’t go to school”—Kununurra

“A lot of kids don’t want to be at home… if there’s strangers partying there”—Wyndham

34

H. Summary ratings of initial conditions

Stakeholders participating in the research completed a short questionnaire which asked them to rate the prevalence and severity of issues in their local community as well as aspects of community functioning. Average ratings provided by participants are presented in Table 5 below.

Table 5: Stakeholders’ average ratings of severity of issues and community functioning

(n=31)

Kununurra / Wyndham Ceduna

How much of an issue are each of the following in the local community? (Average ratings on a scale of 0 - Not at all to 10 - Extremely severe) Alcohol abuse 8.3 7.7

Drug use 7.2 7.0

Gambling 6.6 8.3

Violence and other crimes 8.1 7.4

Street begging 4.8 6.0

Humbugging 5.7 6.1

Harassment, abuse, intimidation 5.9 6.5

How well is the local community performing on each of the following aspects? (Average ratings on a scale of 0 - Very poorly to 10 - Very well)

Ability to afford basic household goods 3.7 4.6

Paying bills 3.4 5.5

Employment 3.4 3.8

Education / training 3.2 3.9

Nutrition 3.1 4.4

Health and wellbeing 3.3 4.8

Community pride 4.4 4.9

Community safety 4.1 4.5

35

V. Awareness, understanding and expectations of the CDCT

A. About this chapter

This chapter presents research findings relating to awareness, understanding and expectations of the CDCT amongst stakeholders across the trial locations. It also reports on feedback stakeholders provided about ISP recipients’ awareness and understanding of the trial.

B. Stakeholders’ awareness and understanding of CDCT

Overall, the research found that there was generally good awareness and general understanding of the CDCT amongst stakeholders in both trial sites. Community leaders tended to have a better and more detailed understanding of the CDCT processes than other stakeholders.

Most stakeholders were aware of the following:

 The trial was mandatory for all ISP recipients (other than Age Pensioners and Veterans’ Affairs Pensioners);

 However, one Indigenous leader in Wyndham thought that the trial was an Indigenous specific measure;

 That Age Pensioners and Veterans’ Affairs Pensioners could elect to participate in the trial on a voluntary basis;

 Payment conditions - i.e. that 80% of trial participants’ income support payments would be paid to the cashless debit card and 20% into their regular bank account;

“80% will be reasonable to buy food and clothes and everything they need to become a strong family”—Kununurra

 Arrangements for altering payment conditions - i.e. that trial participants would be able to make an application to a community panel to increase the percentage of their payments received as cash;

“We have this panel that can decide whether we go from 80-20 to 50-50”—Ceduna

 Card restrictions - i.e. the card could not be used to purchase alcohol or gambling products or to withdraw cash; and

 That additional funding was being provided for support services in trial locations.

“Programs are going to be funded as part of the trial is a good thing”—Wyndham

However, some stakeholders had a more limited understanding of the details of this (e.g. how many additional drug and alcohol workers would be funded, etc.)

36

In contrast, the research found that most stakeholders’ had limited knowledge of specific operational / functional elements of the card and trial (e.g. how the card would operate, how participants could view card balances and assessment criteria used by the community panel). However, most stakeholders were aware of who had responsibility for supporting the rollout / operation of the card in their trial site and indicated that they would refer their client queries about such matters to these organisations.

Overall, most stakeholders felt that the CDCT had been well-communicated to their organisation by DSS and felt adequately informed. However, a few stakeholders in Ceduna felt that services in adjacent / nearby areas needed to be better informed about the trial. These stakeholders reported knowing of some services in nearby areas that had dealings with trial participants who had left Ceduna, but had not been aware of the trial.

“They needed to send out information to Port Lincoln. Port Lincoln Aboriginal Health Services had people come through at the time of the roll-out”—Ceduna

The research also identified a number of stakeholder concerns around implementation issues / difficulties with the debit card, which was being rolled out during the time of fieldwork. These included:

 Card activation - a few stakeholders reported CDCT participants were attempting to use cards that had not yet been activated due to some confusion around the staggered starting date of the trial;

 Concerns that CDCT participants would be unable to access funds to attend and / or spend at specific cash only events;

“[People ask] ‘when they’re selling a ticket and they don’t have EFTPOS, how am I going to get it?’ ”—Wyndham

 Communicating the CDCT to clients in remote communities and with limited literacy - many stakeholders noted that a substantial effort was required to inform and ensure understanding of the CDCT amongst some of their clients living in remote communities who were limited in terms of their English literacy and access to the internet;

 Private rent arrangements - a few stakeholders dealt with clients who rented privately and had difficulties arranging for rental payments to be taken from their cashless debit card account and / or were soon to begin the trial but were unsure about how these payments would be made;

 Concerns regarding funds transfer limitations - one stakeholder reported a client experienced “high anxiety” due to uncertainty about how to repay personal loans to family / friends given the transfer limitations;

“Our last session all she talked about was the card… she had high anxiety about how she was going to pay her mother back with it and how to put money away for her kids”—Ceduna

 Direct debit limitations - the card would only facilitate direct debits via a card number and did not allow direct debits to be set-up using a BSB (i.e. electronic transfer). Some stakeholders noted that this limitation impacted trial participants’ ability to set-up regular payment arrangements (e.g. car repayments) as not all businesses offered this form of direct debit. Furthermore, some organisations charged extra fees when direct debits were made via a card. A few stakeholders also expressed concerns that they had not been made aware of this limitation and had been informing the community that the card would support all direct debit arrangements; and

37

“Direct debits we thought would be no problem. Someone who’s turned his life around can’t make his car repayments”—Ceduna

“We’ve gone and told people in the community that direct debits won’t be a problem… now we look like idiots”—Ceduna

 Confusion regarding account selection for EFTPOS transactions - a few stakeholders noted that some CDCT participants were unsure about whether to select ‘savings’ or ‘cheque’ when paying for goods using EFTPOS. However, these stakeholders noted that this had not caused any concerns as it was easily remedied by staff at point-of-sale.

C. ISP recipients’ awareness and understanding of CDCT

Stakeholders reported that while most ISP recipients had known that the CDCT was occurring, many had shown limited interest in the trial and had not attended information sessions that were held prior to the rollout. Stakeholders indicated that these people had only begun to engage with trial information once the rollout had begun / was about to begin and had become more relevant. As a result, stakeholders indicated that some trial participants had a limited understanding about the details of card usage and logistics. As such, one stakeholder felt additional face-to-face information sessions would be beneficial to allow ISP recipients to ask questions and voice any concerns they had regarding card logistics.

“… they’re resistant to engage with it until it happens”—Ceduna

In addition, a few stakeholders in Ceduna indicated that some ISP recipients had their payment suspended for failing to meet participation requirements of the Community Development Programme. These clients had thought that this was due to problems with the cashless debit card as the timing had coincided with the rollout of the cards.

“There’s a new provider that has come in and they are enforcing the cut-off and they think that the cashless debit card isn’t working”—Ceduna

D. Stakeholders’ expectations of the CDCT

Across both trial locations, most stakeholders felt strongly that there was a need for something to be done to address the high levels of alcohol consumption and, to a lesser extent, illicit drug usage and gambling in the community and their associated harms. Many also felt that a new approach was required to address these issues as current and previous programs and services had not reduced these behaviours.

“We’re treading water, just surviving”—Kununurra

“Rehab’s not working but we keep spending”—Wyndham

As such, most stakeholders were broadly supportive of the CDCT. However, perceptions in relation to the likely effectiveness of the trial were mixed. The research found that:

 Some stakeholders felt strongly that the CDCT would have a positive impact on reducing alcohol consumption, illicit drug usage and gambling - these tended to be stakeholders who had been

38

involved in initiating the trial in their community (i.e. members of the regional leadership groups); and

 Some others were less confident about the extent to which the CDCT would address these issues. These participants reported that, while they were “hopeful” that the trial would have a positive impact, it was “too early to say” whether or not the CDCT would reduce these issues.

Stakeholders reported that they were expecting and / or hoping that there would be a range of positive outcomes, for individual CDCT participants and their families, as well as the broader community as a result of the trial reducing the consumption of alcohol, illicit drug use and gambling. These included (as discussed in Chapter IV):

 A reduction in the amount of domestic violence, crime, assaults and self-harm;

 A reduction in street drinking and conflict;

 A decrease in the humbugging of women and the elderly in Indigenous communities;

 “Stronger families” and improved outcomes for children, in relation to safety, health / nutrition and school attendance and engagement;

“There’s a culture in the communities of men taking money from women when they don’t have any… there’s a strong hope that this will be addressed”—Ceduna

 One stakeholder felt it was important to assess the type of attendance data that would be used for decisions in altering the 80 / 20 arrangements with the community panel. This stakeholder felt many parents had become “savvy” at explaining their children’s absence from school and therefore felt that to be effective at increasing school attendance the community panel should use “bums on seats” data rather than explained / unexplained absences data;

 In Indigenous communities around Ceduna, more people returning to their community (i.e. not staying in town to drink) and taking an interest in improving their towns / communities; and

 An increase in the uptake of longer-term counselling / rehabilitation services and a decrease in the use of crisis / short-term services (e.g. meal and short-term accommodation services).

Despite most stakeholders being generally supportive of the trial, there were some concerns about specific aspects of the trial, such as:

 The potential for adverse consequences - some stakeholders were also expecting and / or concerned about a range of negative impacts that the trial may have. Most commonly, that some ISP recipients would try to access cash and / or alcohol and drugs in other ways, which would negatively impact the community (e.g. increased humbugging / harassment, prostitution and petty theft);

“I’m worried about the impact on prostitution and petty crime”—Kununurra

 The community panel arrangements - a few stakeholders were concerned about the extent of personal information / data the panel would have access to when assessing applications to alter the cash component paid. These stakeholders were particularly concerned that local community members would have access to this information given the small size of the trial communities; and

 Additional support services were not in place at the beginning of the trial - a few stakeholders were concerned that extra support services were not in place at the commencement of the trial. These stakeholders felt that it was essential that these services were in operation to support

39

participants through initial withdrawal periods and to enable the trial’s impact to be properly evaluated. A few stakeholders also noted that due to the lack of funding certainty for service providers, there would be significant delays in establishing extra services / supports once the contracts were in place.

“In reality if [support services] are not on the ground when the card is running how do you assess it?”—Kununurra

“Nobody signed off on anything… you have to recruit and train staff. It could take you 6-10 weeks”—Kununurra

In addition, a few stakeholders expressed personal views that that the trial did not address the “root causes” of the high rates of alcohol and drug use in the communities and were critical that the CDCT would only be “a short-term fix”.

“These are deep seated problems... you’re dreaming if you think it’ll fix everything”— Ceduna

40

VI. Baseline Administrative Data The South Australian and Western Australian State governments have provided the CDCT evaluation with a range of administrative data relating to social harm for the CDCT sites. Table 6 (below) and Table 7 (over page) presents baseline values29, where available, for these social harm indicators for Ceduna and surrounds and the East Kimberley CDCT sites, respectively.

Table 6: SA state government baseline social harm data for Ceduna and surrounds

Explanation of Data Frequency Baseline

South Australian Police - Number of Police Reports (Eyre local Service Area - wider than trial area)

Murder, homicide and related offences Quarterly 0.44 per month (Jun-15 to Feb-16)

Acts Intended To Cause Injury (i.e. assault) Quarterly 84.11 per month (Jun-15 to Feb-16)

Sexual Assault And Related Offences Quarterly 5.67 per month (Jun-15 to Feb-16)

Robbery And Related Offences Quarterly 0.89 per month (Jun-15 to Feb-16)

Other Offences Against The Person Quarterly 7.56 per month (Jun-15 to Feb-16)

Emergency Department admissions

Number of emergency department admissions Quarterly 21.17 per month (Sep-15 to Feb-16)

Department for Communities and Social Inclusion - Ceduna Service Reform

Sobering Up Unit (SUU) admissions Quarterly 214.75 per month (Jul-15 to Feb-16)

Sobering Up Unit (SUU) - discharges at risk Quarterly 32.88 per month (Jul-15 to Feb-16)

Sobering Up Unit (SUU) - Blood Alcohol Content on admission Quarterly average 0.259 (Jan-16 and Feb-16)

Sobering Up Unit (SUU) - Blood Alcohol Content on discharge Quarterly average 0.106 (Jan-16 to Feb-16)

Drug and Alcohol Services SA (DASSA) outpatient counselling

Total attendances Quarterly 30.25 per month (Jul-15 to Feb-16)

Proportion of attendances where alcohol was the principal drug of concern Quarterly 56.48% average (Jul-15 to Feb-16)

Total number of new treatment episodes Quarterly 7 per month (Jul-15 to Feb-16)

Proportion of new treatment episodes where alcohol was the principal drug of concern Quarterly 45.62% monthly average (Jul-15 to Feb-16)

Department for Communities and Social Inclusion

Not eligible for Transitional Centre Quarterly 15.5 monthly average (Jul-15 to Feb-16)

Number of apprehensions under the Public Intoxication Act Quarterly 37.92 per month (Mar-15 to Feb-16)

Mobile Assistance Patrol (MAP) clients Quarterly 468.5 per month (Jul-15 to Feb-16)

SA Attorney-General

Poker Machine Revenue Quarterly $381,257 average monthly expenditure (Jul-13 to Feb-16)

Department for Communities and Social Inclusion - Housing SA

Proportion of Tenants with debt Quarterly 48% quarterly average for Q2 & Q3 2015/16

Total Customer (Tenants) Debt ($) Quarterly $253,356 quarterly average for Q2 & Q3 2015/16

Acts Intended To Cause Injury (i.e. assault) Quarterly 2 quarterly average for Q2 & Q3 2015/16

Number of Support Periods (counts represent a client's Intake) Quarterly 424 quarterly average for Q2 & Q3 2015/16

Number of clients (all client counts are unique) Quarterly 400.5 quarterly average for Q2 & Q3 2015/16

Proportion of clients where Domestic Violence issue was identified Quarterly 15.7% quarterly average for Q2 & Q3 2015/16

Proportion of clients where Drug/Alcohol issue was identified Quarterly 3.0% quarterly average for Q2 & Q3 2015/16

Department for Education and Child Development

Collated attendance figures for the Koonibba, Oak Valley, and Yalata Anangu schools (data for Ceduna will be provided through the MySchool system every six months)

End of each school term. MySchool data by March 2017 66.0% term average 2015

29 The baseline values reported in Table 6 have been calculated by the Department of Social Services based on data provided by the SA and WA governments.

41

Table 7: SA state government baseline social harm data for Ceduna and surrounds

Explanation of Data Frequency Baseline

Western Australia Police (WAPOL)

Kununurra Verified Assaults Quarterly 9 (July 15)

Kununurra Verifed Burglary Quarterly 4 (July 15)

Kununurra Verified Domestic Violence Assault Quarterly 25 (July 15)

Kununurra Verified Theft Quarterly 10 (July 15)

Kununurra Police attended Domestic Violence Reports Quarterly 46.3 per month (between May 15 and Jul 15)

Wyndham Verified Assault Quarterly 1 (July 15)

Wyndham Verified Domestic Violence Assault Quarterly 3 (July 15)

Wyndham Verified Burglary Quarterly 0 (July 15)

Wyndham Verified Theft Quarterly 2 (July 15)

Wyndham Domestic Violence - Police attended incidents Quarterly 7.33 per month (between May-15 to Jul-15)

Western Australian Department for Child Protection and Family Support

Substantiated safety and wellbeing assessments Quarterly Not yet available

Mandatory Child Protection Reports Received Quarterly Not yet available

Number of children in care as at last day of the month Quarterly Not yet available

Western Australian Housing Authority

Disruptive Tennancy Complaints Quarterly Not yet available

Western Australian Health (KNX hospital)

Emergency Presentations Monthly Not yet available

St John Ambulance

Total Call Outs Monthly 351 (May-Jul 14); 464 (May-Jul 15); 84 (1-22 Aug 15)

Alcohol Only Related Call Outs Monthly 7 (May -Jul1 14); 21 (May-Jul 15); 3 (1-22 Aug 15)

Assault Related Call Outs Monthly 23 (May-Jul 14); 22 (May-Jul 15); 11 (1-22 Aug 15)

Stabbing Related Call Outs Monthly 4 (May-Jul 14); 3 (May-Jul 15); 0 (1-22 Aug 15)

The Department of Education and Catholic Schools

Aboriginal Male Attendence Quarterly 57% (May to June 15)

Aboriginal Female Attendance Quarterly 56% (May to June 15)

Non-Aborignal Male Attendance Quarterly 90% (May to June 15)

Non-Aboriginal Female Attendance Quarterly 91% (May to June 15)

Western Australian Deptarment of Aboriginal Affairs - Kununurra Waringarri Aboriginal Corporation Patrol Service (Kununurra)

Total Number of People picked up by Kununurra Miriwoong Community Patrol Service for Alcohol Monthly 494 per month (Jan 16 - Feb 16); 541 (Jan 15- Jun 15)

Total People Refered to Sobering Up Shelter Moongoong Sober Up Shelter (Kununurra) Monthly 190 (Jan 15 - Jun 15; 153 (Jan 16 - March 16)

Western Australia Dept. Child Protection and Family Support (Wyndham)

Total Assisted by Women’s Crisis Centre Monthly Not yet available

Western Australian Mental Health Commission - Ngnowar-Aerwah Aboriginal Corporation (Wyndham)

Total assisted by the Sobering Up Shelter Monthly 97 (April 15); 54 (May 15); 71 (Jun 15)

Total Assisted by the Night Patrol Monthly Average 250 (Apr 15 - Jun 15)

The Drug and Alcohol Office of Western Australia (also Commonwealth Funded) - Ngnowar-Aerwah Aboriginal Corporation (Wyndham)

Total assisted by the Kimberley Mental Health and Drug Service TBD Not yet available

Women’s Safe House (Kununurra)

Estimated number of clients TBD Not yet available

Western Australian Police

Drunk related behaviours (driving, drunk and disorderly etc.) Quarterly Not yet available

The baseline value percentages in Table 6 and Table 7 are derived from the average of totals for the stated time periods. Where historical data is not yet available, current figures are included, noting that as further data is provided the baseline data will be updated. At this stage, the evaluation considers that it would be premature to draw conclusions from initial monthly movements in the above indicators, which may too be volatile on a monthly basis and subject to seasonality. Trend movements in the above indicators will be analysed and reported during the course of the evaluation. In addition, the department is sourcing administrative data reports as input to the evaluation, which present the number of clients and level of services provided by Commonwealth organisations delivering services in the CDCT sites.

42

VII. Conclusion The initial conditions qualitative research with stakeholders in Ceduna, Wyndham and Kununurra found widespread local concern about high levels of alcohol consumption and, to a lesser extent, illicit drug use and gambling activity.

Stakeholders indicated that these issues had been becoming progressively worse over the past 5-10 years and that the local communities were experiencing significant adverse impacts. These were commonly identified in relation to:

 the health of adults and children in the communities (e.g. a range of injuries and longer-term health issues such as anxiety, depression, cancer, high blood pressure, Foetal Alcohol Syndrome);

 safety and security (e.g. domestic violence, sexual violence, assaults and harassment / intimidation);

 financial problems (e.g. inability to pay fines, inability to fund basic living expenses for items such as food, clothing, rent and utilities);

 social problems such as family arguments / disputes, unemployment / underemployment and humbugging;

 inability to secure stable housing;

 living in overcrowded housing conditions; and

 adverse impacts on the wellbeing of children as a result of poor parenting / neglect of family responsibilities and lack of engagement (e.g. lower school attendance and engagement, poor educational outcomes and poor nutrition).

In particular, most stakeholders felt that excessive alcohol consumption was at a “crisis point”, and was having wide-ranging negative impacts on individuals, their families and the community.

Overall, the research found that there was generally good awareness and general understanding of the CDCT amongst stakeholders in both trial sites. Community leaders tended to have a better and more detailed understanding of the CDCT processes than other stakeholders.

Across both trial locations, most stakeholders felt strongly that there was a need for something to be done to address the high levels of alcohol consumption and, to a lesser extent, illicit drug usage and gambling in the community and their associated harms. Many also felt that a new approach was required to address these issues as current and previous programs and services had not reduced these behaviours.

As such, most stakeholders were broadly supportive of the CDCT. However, perceptions in relation to the likely effectiveness of the trial were mixed.

Appendix A: Evaluation Framework

Australian Government Department of Social Services

Cashless Debit Card Trial: Evaluation Framework

Contents

1 Executive Summary ............................................................................... 1

2 Introduction .......................................................................................... 2

2.1 Objective of the framework ..................................................................................... 2

2.2 The Cashless Debit Card Trial................................................................................... 2

2.3 Contextual factors.................................................................................................... 4

2.4 Ethics clearance and approval ................................................................................. 5

3 Evaluation scope and key measures ...................................................... 6

3.1 Introduction ............................................................................................................. 6

3.2 CDCT Evaluation Program Logic ............................................................................... 7

3.3 Key Performance Indicators .................................................................................. 11

4 Data Collection Approach .................................................................... 20

4.1 Introduction ........................................................................................................... 20

4.2 Qualitative research with on the ground observers/stakeholders ....................... 20

4.3 Quantitative research ............................................................................................ 21

Recruitment and training of interviewers ......................................................................... 23

Fieldwork management ..................................................................................................... 24

4.4 Collation and analysis of administrative data........................................................ 25

CDCT Comparison Sites ..................................................................................................... 25

5 Timing of evaluation reporting ............................................................ 28

6 Challenges in evaluating the Cashless Debit Card Trial ........................ 29

A1

1 Executive Summary ORIMA Research has been commissioned by the Department of Social Services (DSS) to evaluate the Cashless Debit Card Trial (CDCT) in South Australia (SA) and Western Australia (WA).

The aim of the CDCT is to reduce the levels of harm associated with alcohol consumption, illicit drug use and gambling within the communities of Ceduna and Surrounds in SA and East Kimberley in WA (Kununurra and Wyndham). These sites were proposed by local community leaders and the CDCT has been developed via a collaborative process involving local community leaders, local and state government agencies and Australian Government agencies (led by DSS). The two CDCT sites have experienced high levels of community harm related to alcohol consumption, drug use and gambling.

The overall objective of this evaluation is to assess the effectiveness of the CDCT. This document specifies the design framework for the evaluation.

The evaluation design is based on a multi-staged and multi-method approach including desk research, qualitative research, quantitative research and analysis of administrative and program data. The evaluation will consist of six key (and sometimes overlapping) phases:

1. Project Inception meetings and set up (including initial desktop program scoping, consultation with community representatives and leadership, development of the Program Logic (PL), Key Performance Indicators (KPIs) and Theory of Change (TOC), ethics approval);

2. Three waves of qualitative research with observers / on-the-ground stakeholders (named initial conditions, wave 1 and wave 2);

3. Two waves of quantitative research (termed waves 1 and 2) amongst CDCT participants and their families, as well as non-participant community members;

4. Collation and analysis of administrative data from the Department of Human Services (DHS), Indue Ltd, State Government agencies and local service providers (with comparison between CDCT Trial sites and non-CDCT comparison sites where applicable);

5. Ongoing monitoring of the DSS CDCT ‘inbox’ and hotline; and

6. Interim and final reporting.

A2

2 Introduction

2.1 Objective of the framework

The evaluation of the Department of Social Services’ (DSS) Cashless Debit Card Trial (CDCT) is being conducted by ORIMA Research, an independent specialist social and government research and evaluation service provider. The overall objective of the evaluation is to assess the effectiveness of the CDCT.

This document presents the design framework for the evaluation.

This evaluation framework will:

 Describe the Cashless Debit Card Trial program and what will be evaluated;

 Help to develop sound evaluation plans and implementation of evaluation activities;

 Articulate the program goals and measurable short, medium and long-term objectives;

 Define relationships among inputs, activities, outputs, outcomes and impacts; and

 Clarify the relationship between program activities and external factors.

2.2 The Cashless Debit Card Trial

The Australian Government is undertaking the CDCT to deliver and manage income support payments (ISPs) in order to reduce levels of community harm related to alcohol consumption, drug use and gambling. This initiative has been informed by a recommendation in Andrew Forrest’s Creating Parity report.30 It has also been informed by lessons learned from previous income management (IM) trials.

In the CDCT, a proportion (from 50 to 80 per cent) of an individual’s ISP is directed to a restricted bank account, accessed by a debit card (not allowing cash withdrawals). This debit card cannot be used at merchants who sell alcohol and gambling related products.31

Participation in the CDCT is mandatory for all working age ISP recipients who live in the selected Trial sites. In addition, wage earners, Age Pensioners and Veterans’ Affairs Pensioners who live in the Trial sites can opt-in to the CDCT.

30 Forrest, A. (2014). The Forrest Review: Creating Parity. Commonwealth of Australia, Canberra.

31 Merchants within Trial locations who sell both excluded and allowable goods are involved in individual mixed merchant agreements. Lottery purchases are permissible.

A3

To date, the CDCT is being implemented in Ceduna and Surrounds32 in South Australia (SA) and Kununurra / Wyndham (East Kimberley)33 in Western Australia (WA). These sites were proposed by local community leaders and the CDCT has been developed via a collaborative process involving local community leaders, local and state government agencies and Australian Government agencies (led by DSS). The two CDCT sites have experienced high levels of community harm related to alcohol consumption, drug use and gambling.

To support the CDCT implementation, DSS has worked with the SA and WA State Governments, community agencies and Indigenous leadership to supplement the social services being provided to the Trial areas. Additional services that have been provided at the Trial sites are listed below:

 Kununurra/Wyndham

 AOD Brokerage Fund

 Substance abuse rehabilitation support for adolescents

 ‘One family at a time’ program

 ‘A Better Life’ program

 Children and Parenting Services (CaPS)

 Improved financial counselling

 Ceduna and Surrounds

 Alcohol and Other Drug Outreach Workers

 Ceduna 24/7 Mobile Outreach ‘Street Beat’

 Brokerage Fund

 Domestic Violence: Family Violence Prevention Legal Services

 Mental Health support services

 A Better Life (ABLe)

 Financial counselling and support services

 Additional aftercare support service

 Outreach and transport support services (Mobile Assistance Patrol)

32 The Ceduna and Surrounds Trial site is defined by the town of Ceduna (meaning the area of the District Council of Ceduna as defined in accordance with the Local Government Act 1999 (SA); and the surrounding region of Ceduna, which is composed of and limited to the ABS 2011 Australian Statistical

Geography Standard (ASGS) Statistical Area Level 1s (SA1) of 40601113409, 40601113410, 40601113501 and 40601113502.

33 The Wyndham/Kununurra Trial site is situated in the East Kimberley region of Western Australia. The Trial site, incorporating communities within the postcode regions 6740 and 6643, comprises a number of SA1s.

A4

The main elements of the Trial include:

 Co-design with local community reference groups in the Trial sites;

 A cashless debit card, delivered by a commercial provider (Indue Ltd);

 80 per cent of welfare payments to be placed into a restricted account linked to the cashless card (100% of lump sum payments and arrears payments);

 The quarantined percentage may be varied by local leadership boards to a base level of 50 per cent;

 Alcohol and gambling (excluding lotteries) will not be able to be purchased with the card, and no cash will be able to be withdrawn from the card;

 The debit card and associated services will be provided by the commercial partner who will provide support to participants via a customer contact centre, a mobile phone app and text alerts to keep people informed;

 The optional operation of a community panel in each Trial site;

 All working age income support recipients in selected Trial locations will be included in the Trial. Those who move from the Trial location elsewhere will remain participants in the Trial;

 Aged and Veterans pensioners and wage earners may opt-in to participate;

 Up to three sites will operate for 12 months, with a staggered rollout from March 2016; and

 The individuals impacted have been informed about the Trial by DSS through direct consultation, a community reference group and community members who were involved in the consultation phase. In addition, public information sessions have been held in Ceduna and the East Kimberley, and local Indigenous organisations have been highly involved in informing participants about the Trial.

2.3 Contextual factors

This document has been informed by feedback from:

 respected academics and commentators with expertise in conducting research and evaluations involving Aboriginal and Torres Strait Islander Peoples (via an expert panel convened by the Department of Social Services);

 leaders and representatives of Aboriginal corporations and community organisations in the Ceduna and Surrounds and East Kimberley regions; and

 officers of Australian and State Government agencies with on-the-ground experience in the CDCT sites.

The evaluation design is largely based on measuring the views and reported experiences of several stakeholder segments:

 Local observers and on-the-ground stakeholders in the CDCT sites - community leaders, as well as government and non-government service providers;

 CDCT participants;

 CDCT participants’ families; and

A5

 Other members of the general community living in the CDCT sites.

The evaluation design takes into account two important contextual issues:

1. A need for the evaluation to assess the impact of CDCT on individual and community functioning taking into account the impact of factors other than the CDCT which may also affect its planned outcomes; and

2. DSS needs ‘real-time’ early warning of any issues and problems uncovered by ORIMA Research. These need to be communicated in a timely manner to the Department as the evaluation progresses. In practice, this will take place over the three two-week periods during which the ORIMA Research qualitative team is on the ground at each location, as well as the two two-week periods during which ORIMA specialist Indigenous interviewers are on the ground at each location, and as any issues are identified through data provided to ORIMA Research via the DSS CDCT email ‘inbox’.

2.4 Ethics clearance and approval

ORIMA Research will develop ethical protocols in accordance with Human Research Ethics Committee (HREC) requirements and obtain ethics clearance for the research involving CDCT participants, their family members and non-participants in the relevant communities. It will not be necessary to obtain ethics approval for collecting data amongst observer groups, including community leaders. ORIMA Research will use the services of the Bellberry Human Research Ethics Committee to ethically review and provide approval for the methodology, interview questions, reimbursement of research participants, consent forms, and information sheets.

A6

3 Evaluation scope and key measures

3.1 Introduction

In this evaluation, the Program Logic methodology has been used to establish the scope of the evaluation and the key performance indicators that will inform an assessment of the effectiveness of the CDCT. If the outputs, short-term outcomes and medium-term outcomes specified in the CDCT Evaluation Program Logic are achieved, this will indicate that the CDCT has been effective. In order to measure the extent of effectiveness, each individual output and outcome has been translated into one or more Key Performance Indicators (KPIs), which have been operationalised very specifically and are measurable via existing or new data sources.

The CDCT Program Logic also identifies a range of potential longer-term outcomes and impacts of the CDCT that are outside of the scope of the evaluation because the expected timeline for their realisation extends beyond that of the evaluation.

The key evaluation questions are:

3. What have been the effects of the CDCT on program participants, their families and the broader community?

 Have there been reductions in the consumption of alcohol, illegal drug use, or gambling?

 Has there been a reduction in crime, violence and harm related to these behaviours?

 Has there been an increase in perceptions of safety in the Trial locations?

 Have there been any other positive impacts (e.g. increase in school attendance, increase in self-reported well-being, reduction in financial stress)?

4. Have there been any circumvention behaviours (e.g. participants selling goods purchased with cashless debit cards to obtain more cash, increase in humbugging or theft) that have undermined the effectiveness of the CDCT?

5. Have there been any other unintended adverse consequences (e.g. feelings of shame, social exclusion)?

6. What lessons can be learnt throughout the Trial to improve delivery and to inform future policy?

 How do effects differ among different groups of participants (e.g. men compared to women, people from different age groups)?

 Where has the Trial worked most and least successfully?

 To what extent can any changes be attributed to the Trial as opposed to external factors such as alcohol restrictions?

 Can the contribution of the debit card be distinguished from that of the additional services in the Trial locations provided via the CDCT support package?

A7

3.2 CDCT Evaluation Program Logic

In consultation with DSS, a CDCT Program Logic was developed for the purposes of the evaluation. The CDCT Evaluation Program Logic uses a Theory of Change approach to articulate the objectives of the Trial, and to trace the links between program activities and these objectives. The Program Logic clearly specifies hypothesised or desired (as opposed to actual) outcomes.

There are five major components to the Program Logic (see Figure 13 on page 8). Starting from the left and moving right, we begin with the program inputs. These are the resources and infrastructure that are essential for program activities to occur. The inputs support the program activities - the specific actions that make up the program. These activities will produce or create a series of immediate outputs. The outcomes are the intended changes in the communities as a result of the program. For the purpose of the CDCT, these are divided into short-term outcomes (changes in behaviour, attitudes and perceptions achieved by 3 months of Trial launch), medium-term outcomes (changes in behaviour, attitudes and perceptions achieved by 12 months) and long-term outcomes (changes in state achieved in two or more years). Finally, the Program Logic articulates the intended impact of the CDCT, ‘safer families and communities’ - as the intended societal change but, like the long-term outcomes, is not included in the scope of the evaluation as it lies beyond the timeframe of the evaluation.

The core causal relationship is presented in the centre of the Theory of Change diagram (see page 9). As access to cash is restricted to 20% of Trial participants’ income support payments, participants are expected to have less money to purchase alcohol and drugs, as well as to gamble. This restriction is therefore expected to lead to less alcohol consumption, less drug use and less gambling, in both the short- and medium-term. The reduction in alcohol consumption and drug use is expected to lead to less alcohol- and drug-fuelled violence, fewer accidents and fewer injuries. Over time, this process is expected to lead people at the Trial locations feeling safer in their homes and communities and feeling prouder of their communities.

Figure 13: Program Logic - Cashless Debit Card Trial

INPUTS

Budget funding

Relevant legislation

Program precedents and evaluation

evidence

DSS

administrati on

Input from DHS, PM&C

Partnership with local leadership

Indue Ltd service provision

Existing community social services

Potential program circumventions Program circumventions may

directly undermine the program logic

 People use mixed merchants in other communities  Housing / other expenses money transfer facility is abused  Humbugging increases or changes in nature  Secondary, barter or black market for goods

develops  Prostitution and other undeclared paid work increases  Theft and burglary increases  Consumption switches to cheaper illicit drugs

and/or cheaper alcohol products (e.g. home brews)  Unregulated forms of gambling increase  Mixed merchant staff compelled not to, unwilling or unable to comply  Merchants process cash refunds for legitimate card purchases

Potential adverse consequences  Household spending complicated due to one card per client  Previously responsible households feel

disenfranchised  Feelings of shame are associated with card usage / social exclusion occurs or increases  Business owners and/or workers are abused or

intimidated  Reduced retail competitiveness leads to higher prices  Recipients financially worse off due to loss of interest

on savings or new 3rd party fees  Card skimming scams occur  Privacy breaches occur  Vulnerable people, including age pensioners, are

victimised / experience more humbugging  People leave communities altogether to avoid the Trial

Potential spill-over benefits

 Short-term: Spending higher % of income on non-harmful / productive goods & services (e.g. food, rent, bills)  Short-term: Spending higher % of time on healthy activities (e.g. physical activity, education & training)  Short-term: Less humbugging  Short-term: Greater motivation to be in paid employment  Medium-term: Improved IT/PC literacy  Medium-term: Better nutrition (children & adults)  Medium-term: Greater ability to budget and save  Medium-term: More time spent on education (child school

attendance, parental involvement)  Medium term: Increased sense of community pride  Medium-term: Greater housing stability  Long-term: Improved educational outcomes and later school

leaving age

Note: While both potential spill-over benefits and adverse consequences will be measured as part of the evaluation, they are not performance indicators

Outside scope of current evaluation

Safer

families and communities

Reducing welfare-fuelled violence arising from

alcohol and other drug use and gambling

IMPACT

Intended societal change

ACTIVITIES

2. Public awareness & information activities launched

1. Community leaders consulted and co-design initiative

5. Financial processes including merchant lists and system integrity managed

3. Card developed and distributed by commercial provider

4. Existing public funding for support services supplemented

7. DSS ongoing monitoring

6. Assessment activity of community panels

OUTPUTS

2. Clients aware of the trial and understand how it works

1. Community leaders endorse an initiative that is tailored to the community

5. Clients have access to an adequate range of merchants

3. Clients cannot spend more than 20% of their income support payment on alcohol, drugs and/or gambling

4. Community support services are available

7. Ongoing adjustments are made to program where relevant

6. Appropriate adjustments are made to income restrictions on case-by-case basis

OUTCOMES

Change in communities as a result of the trial program

Medium-term ∆s in behaviour, attitudes & perceptions

(By 12 months)

Less violent behaviour

Sustained lower alcohol consumption

Less criminal behaviour

Sustained decreased gambling

Fewer alcohol-related injuries

Sustained lower illicit drug use

Increased sense of safety

Short-term

∆s in behaviour, attitudes & perceptions

(By 3 months)

Lower alcohol consumption

Use of drug & alcohol treatment, counselling & rehab programs

Awareness and use of family support and financial support services

Lower illicit drug use

Decreased gambling

Use of family support and financial support services

Awareness of drug & alcohol treatment, counselling & rehab programmes

Long-term ∆s in state (By 2 years or more)

Safety of women and children ↑

Increased community, personal and children’s well-being (inc. school attendance)

More powerful community expectations and norms for alcohol use, drug

use, gambling and violence

More powerful community expectations and norms around housing and

schooling

Community safety &security ↑

Figure 14: Theory of Change

Greater awareness of drug / alcohol treatment programmes

Greater awareness of other family support services

Greater awareness of financial support services

Use of alcohol / drug / gambling treatment programmes

increases

Use of other family support services increases

Use of financial support services increases

Less alcohol & drug use

Less gambling

Safer families and communities

Fewer

accidents and injuries

Fewer physical assaults in the community

Fewer physical assaults in the home

Sustained decrease in alcohol & drug use

Less theft & burglary

Decreased need for money to purchase drugs and alcohol and

gambling

Less violent crime

Sustained lower gambling

Fewer victims of domestic violence

Fewer victims of crime & violence

Community norms support zero tolerance for alcohol/drug/gambling-fuelled violence

People feel safer at home

People feel safer in community

People feel prouder of their community

Community leadership is supportive of the initiative

Community leadership are partners in designing the initiative

Reduced restrictions granted by Community Panels where

needed

Existing community support services

topped up

Access to cash is restricted to 20% of

income support payment

Community support services promoted

Less money to purchase alcohol, drugs & gambling

Access to support services is increased

Need for support services is increased

A10

As highlighted in the Program Logic diagram (Figure 13), ultimately this process is expected to lead to positive long-term outcomes in the areas of improved community safety and general well-being, as well as more powerful community expectations and norms in relation to alcohol use, drug use, gambling, violence, housing and schooling. A key long-term outcome is expected to be greater safety for women and children. Women and children could also benefit in the short-medium term (see potential spill-over benefits in theProgram Logic - Cashless Debit Card Trial diagram) from having more money for food, greater housing stability and more parental involvement in children’s education.

The Theory of Change diagram also highlights important elements that are expected to support the core process outlined above. These include greater access to community support services (drug and alcohol treatment, family support, financial support), and the partnership / co-design role of community leadership. An important component of the latter role is the ability of local leadership boards to vary an applicant’s restricted amount of payment so that it is lower than 80 per cent of their total ISP (but no lower than 50 per cent). This flexibility is expected to build community acceptance of the Trial and to help reduce any unintended adverse effects of the Trial.

In relation to support services, it should be noted that not all Trial participants are expected to access these services and that the Trial is expected to have positive impacts irrespective of the take-up of these services. Further, fewer people using some services in the longer term could indicate Trial success. For example, fewer people may use sobering up services, because they no longer need to.

The CDCT Evaluation Program Logic also makes explicit reference to a series of potential program circumventions. These potential circumventions are based on experience with previous IM programs. 34 They will be important to monitor because if they occur, they could directly undermine the Theory of Change and help explain why outcomes have not been achieved.

Finally, the Program Logic also highlights a number of potential spill-over benefits and adverse consequences. The hypothesised spill-over benefits are potential ways in which the program could benefit the community above and beyond the program outcomes. These potential benefits, while premised on previous experience with IM programs, are not seen as being central to the Trial’s objectives. Their achievement will be important to monitor and record, but whether or not they are achieved is not an indication of the success or failure of the Trial. Conversely there are a number of potential adverse consequences that could occur as secondary effects. These too will be important to monitor because it is possible for the Trial to create unintended negative consequences while at the same time achieving its stated objectives.

34 See: Deloitte Access Economics (2015) Consolidated Place Based Income Management Evaluation Report 2012-2015, for DSS; DSS Evaluation Hub (2014) A Review of Child Protection Income Management in Western Australia: Final report; ORIMA Research (2010) Evaluation of the Child Protection Scheme of

Income Management and Voluntary Income Management Measures in Western Australia, for Commonwealth Department of Families, Housing, Community Services and Indigenous Affairs (FaHCSIA); Social Policy Research Centre (2010) Evaluation Framework for New Income Management, for FaHCSIA; Social Policy Research Centre (2014) Voluntary Income Management in the Anangu Pitjantjatjara Yankunytjatjara (APY) Lands, for DSS; Social Policy Research Centre (2014) Evaluating New Income Management in the Northern Territory: Final Evaluation Report, for DSS

A11

3.3 Key Performance Indicators The Program Logic and the underlying Theory of Change led to the development of a series of Key Performance Indicators (KPIs) that will drive evaluation of the effectiveness of the Cashless Debit Card Trial. The specific KPIs developed for this evaluation are detailed in the following pages.

Figure 15: Performance Indicators

# community leaders who endorse program

% participants who understand card conditions

% of participants in Trial locations sent card

% of distributed cards that are activated

80% of income support payments are quarantined

# support services available in community

% participants with reasonable access to merchants and products

# community leaders who believe appropriate adjustments are made

to restrictions

Support of Community Leaders

Frequency of use/volume consumed of drugs and alcohol

Frequency/volume of gambling and associated problems

% aware of drug & alcohol support services

Usage of drug & alcohol support services

% aware of family & financial support services

Usage of family & financial support services

Support of Community Leaders

Frequency of use/volume consumed of drugs and alcohol

Frequency/volume of gambling and associated problems

Incidence of violent & other types of crime and violent behaviour

Drug/alcohol-related injuries and hospital admissions

% reporting feeling safe in the community

% reporting feeling safe at home

Outputs Short-term Outcomes Medium-term Outcomes

• School attendance rates (State administrative data) • Child protection substantiations (State administrative data) • Disruptive behaviour in public housing (State administrative data) • Rent arrears in public housing (State administrative data) • Debit Card account balances (DHS data) • Crisis payment applications (DHS data) • Reported ability to afford basic needs (survey and stakeholder interviews) • Reported incidence of humbugging (survey and stakeholder interviews) • Reported ability to save money (survey and stakeholder interviews) • Reported job search activity (survey and stakeholder interviews) • Reported capacity to care for children (survey and stakeholder interviews) • Reported engagement in children’s education (survey and stakeholder interviews) • Self-reported well-being (survey) • Reported sense of community pride (survey and stakeholder interviews)

Measuring spill-over benefits / Adverse consequences (these will be measured but are not Performance Indicators)

• Reported practical difficulties using the card (survey and stakeholder interviews) • Reported extent of negative financial control consequences, including less disposable income, inability to purchase basic household goods, or feelings of disempowerment (survey and stakeholder interviews) • Reported feelings of shame or experiences of exclusion (survey and stakeholder interviews) • Reported street begging, humbugging, harassment, abuse or intimidation by others (survey and stakeholder interviews) • Reported increases in product pricing, merchants imposing minimum purchase requirements or surcharges (survey and

stakeholder interviews) • Incidence of privacy breaches, stolen cards, or skimming (DHS / Indue data) • Reported circumvention behaviours (survey and stakeholder interviews), including: o replacing alcohol or drugs with cheaper products

o making purchases of alcohol / drugs outside community o pooling funds with others to make purchases o using money transfer facility to obtain cash o obtaining cash or goods-in-kind from other household members o engaging in bartering or other secondary market activities o undeclared cash-in-hand work o merchant non-compliance • Suspected merchant non-compliance events (Indue)

A12

Table 8: Output Performance Indicators

Performance Indicator Specification Target Timeframe Data Sources Definitions/comments

Number of

community leaders who endorse

program

Number of community leaders who:

 feel program design is appropriate for their community characteristics  believe program will be / is a good thing for their community  speak positively about program  believe Trial parameters were developed using a co-

design approach

Not applicable Within one month of

program launch (initial conditions), repeated at Wave 1 and Wave 2

Qualitative research with community leaders

Community leaders defined as members of regional leadership groups

Qualitative indication of number: all, most, many, some, few

% participants who understand card conditions

% of participants who are aware:

 How much of their welfare income is quarantined in terms of cash withdrawals  What they can and cannot purchase on the card  Which merchant types they can and cannot use the

card at

 They can use the card wherever Visa is accepted, including online (except where a Merchant is blocked)  They can use the card to make online payment transfers for housing and other expenses, and to pay

bills

 What to do if the card is lost or stolen

Not applicable Self-reported at Wave 1

and Wave 2

Survey of Trial participants Not applicable

% of participants in Trial locations sent card

% of compulsory Trial participants sent a debit card 100% Within two months of program launch

Indue / DHS Client database

Not applicable

A13

Performance Indicator Specification Target Timeframe Data Sources Definitions/comments

% of distributed cards that are

activated

Of all cards distributed to participants, % of these that are activated 95% Within one

month of receiving card

Indue 5% margin allowed for

people moving in and out of income support payments

80% of income

support payments are quarantined

Income support payments are quarantined and 20% are received in cash (excluding approved adjustments) 100% of recipients

Within two months of program launch

DHS Client database

Not applicable

# support services available in

community

# and type of additional support services in operation as planned 100% Within three

months of program launch

DSS provided Need for services is expected to develop over the first 3 months of the program

% participants with reasonable access to merchants and products

Excluding the purchase of alcohol and gambling % of participants who agree that they can still shop where and how they usually shop

% reporting concerns over access to allowable products

90%

10% maximum

Self-reported at Wave 1 Survey of Trial participants

Not applicable

# community

leaders who believe appropriate adjustments are made to income restrictions on a case-by-case basis

Number of community leaders who believe community panels are assessing applications in a timely, consistent and fair manner

Number of community leaders who believe community panels are making just and reasonable decisions about changing percentage of welfare payments quarantined

Most Within one

month of program launch (initial conditions), repeated at Wave 1 and Wave 2

Qualitative research with community leaders

Community leaders defined as members of regional leadership groups

Qualitative indication of number: all, most, many, some, few

A14

Table 9: Short-term Outcome Performance Indicators35

Performance Indicator

Specification Target Timeframe Data Sources Definitions/comments

Frequency of use /volume consumed of drugs and alcohol

 Number of times alcohol consumed by participants per week  % of participants who say they have used non-prescription drugs in the last week  Number of times per week spend more than $50 a

day on drugs not prescribed by a doctor  Number of times per week have six or more drinks of alcohol at one time (binge drinking)  % of participants, family members and general

community members reporting a decrease in drinking of alcohol in the community since commencement of Trial  Number of on-the-ground stakeholders reporting a decrease in drinking of alcohol in the community since commencement of Trial

Many As self-

reported at Wave 1

Survey of Trial participants

Survey of families

Survey of community members

Qualitative research with stakeholders

No targets specified for survey data due to absence of baseline (pre Trial) survey

On-the-ground stakeholders defined as members of the regional leadership groups and observers from government and non-government service providers based in the Trial areas

For stakeholders, qualitative indication of number: all, most, many, some, few

Frequency/volume of gambling and  Number of times Trial participants engage in gambling activities per week

Many As self-

reported at Survey of Trial No targets specified for survey data due to

35 Following the finalisation of the Evaluation Framework it was agreed that the Support of Community Leaders should also be considered as a short-and-medium-term outcome as well as an output measure. In practice these will be addressed in the Output Performance Indicators section, but their importance as an outcome is noted here.

A15

Performance Indicator

Specification Target Timeframe Data Sources Definitions/comments

associated problems

 Number of days a week spend three or more hours gambling  Number of days a week spend more than $50 gambling  % of participants indicating that they gamble more

than they can afford to lose or borrow money or sell things to gamble  % of participants, family members and general community members reporting a decrease in

gambling in the community since commencement of Trial  Number of on-the-ground stakeholders reporting a decrease in gambling and associated problems in the

community since commencement of Trial  EGM (‘poker machine’) revenue in Ceduna and Surrounds

Lower than before Trial Wave 1 participants

Survey of families

Survey of community members

Qualitative research with stakeholders

absence of baseline (pre Trial) survey

For stakeholders, qualitative indication of number: all, most, many, some, few

Gambling revenue data only available in SA (not WA)

% aware of drug and alcohol support services

% participants who are aware of drug and alcohol support services available in their community Not applicable

As self-reported at Wave 1

Survey of Trial participants No sound evidentiary basis for setting a target

% aware of financial and family support services

% participants who are aware of financial and family support services (including domestic violence support services) available in their community

Not applicable As self-reported at

Wave 1

Survey of Trial participants No sound evidentiary basis for setting a target

Usage of drug and alcohol support services

 % of participants who have ever used drug and alcohol support services  Number of times services used per participant  Intention to / likelihood of using service in future  Number of people in community using services

Higher at Wave 2 than at Wave 1 (statistically

As self-reported at Wave 1

Trial period compared

Survey of Trial participants

Department of Social Services (based on data from service

Not applicable

A16

Performance Indicator

Specification Target Timeframe Data Sources Definitions/comments

significant)

Higher than before Trial

with 12 months prior to Trial launch

providers and State Government agencies)

Usage of financial and family support services

 % of participants who have ever used financial or family support services (including domestic violence support services).

 Number of times services used per participant  Intention to / likelihood of using service in future  Number of people in community using services

Higher at Wave 2 than at Wave 1 (statistically significant)

Higher than before Trial

As self-reported at Wave 1

Trial period compared with 12 months prior to Trial launch

Survey of Trial participants

Department of Social Services (based on data from service providers and State Government agencies)

Not applicable

A17

Table 10: Medium-term Outcome Performance Indicators36

Performance Indicator

Specification Target Timeframe Data Sources Definitions/comments

Frequency of

use/volume consumed of drugs and alcohol

See short-term indicators of frequency of use / volume consumed of drugs and alcohol Frequency/volume not higher at

Wave 2 than at Wave 1

Wave 2 Not applicable Not applicable

Frequency/volume of gambling and associated problems

See short-term indicators of frequency/volume of gambling and associated problems Frequency/volume not higher at

Wave 2 than at Wave 1

Wave 2 Not applicable Not applicable

Incidence of violent and other types of crime and violent behaviour

 Police reports of assault and burglary offences; drink driving / drug driving; domestic violence incidence reports; drunk and disorderly conduct; outstanding driving and vehicle fines.

 % of participants, family members and the general community who report being the victim of crime in the past month

 % of participants, family members and the general community who report a decrease in violence in the community since commencement of Trial

 Number of on-the-ground stakeholders reporting a decrease in violence in the

Lower than before Trial Trial period compared

with 12 months prior to Trial launch

As self-reported at Wave 1 and Wave 2

SA and WA Police

Surveys of Trial participants, families and community members

Qualitative research with stakeholders

On-the-ground stakeholders defined as members of the regional leadership groups and observers from government and non-government service providers based in the Trial areas

For stakeholders, qualitative indication of number: all, most,

36 Following the finalisation of the Evaluation Framework it was agreed that the Support of Community Leaders should also be considered as a short-and-medium-term outcome as well as an output measure. In practice these will be addressed in the Output Performance Indicators section, but their importance as an outcome is noted here.

A18

Performance Indicator

Specification Target Timeframe Data Sources Definitions/comments

community since commencement of Trial many, some, few

Drug/alcohol-related injuries and hospital admissions

 Drug / alcohol-related hospital admissions / emergency presentations / sobering up service admissions

 % of participants / family members who say they have been injured after drinking alcohol / taking drugs in the last month

Lower than before Trial

Not higher at Wave 2 than at Wave 1

Trial period compared with 12 months prior to Trial launch

As self-reported at Wave 1 and Wave 2

Department of Premier and Cabinet SA, WA Health, Department of Social Services (based on data provided by local sobering up services)

Surveys of Trial participants and families

Not applicable

% reporting feeling safe in the

community

% of participants, family members and other community members who report feeling safe in their community

Higher at Wave 2 than at Wave 1 (statistically significant)

As self-reported at Wave 1 and Wave 2

Surveys of Trial participants, families and community members

Not applicable

% reporting feeling safe at home % of participants, family members and other community members who report feeling safe at

home

Higher at Wave 2 than at Wave 1 (statistically significant)

As self-reported at Wave 1 and Wave 2

Surveys of Trial participants, families and community

Not applicable

A19

Performance Indicator

Specification Target Timeframe Data Sources Definitions/comments

members

A20

4 Data Collection Approach

4.1 Introduction

Data collection for the evaluation is based on a multi-staged and multi-method approach including:

1. Three waves of qualitative research with observers / on-the-ground stakeholders (named initial conditions, wave 1 and wave 2);

2. Two waves of quantitative research (termed waves 1 and 2) amongst CDCT participants and their families, as well as non-participant community members; and

3. Collation of administrative data from the Department of Human Services (DHS), Indue Ltd, State Government agencies and local service providers.

4. Ongoing monitoring of the DSS CDCT ‘inbox’ and hotline.

Prior to commencing data collection, ORIMA Research will visit Ceduna, Kununurra and Wyndham. During the visits we will consult with local community representatives and other relevant stakeholders:

 Regarding the proposed evaluation / research plan and its implementation;

 To gain any feedback and answer questions representatives and other stakeholders have about the evaluation;

 To seek advice about issues such as the nature of the reimbursements to be provided to survey respondents, focus group attendees and individual interview participants; and

 To gain views on the profile of appropriate interviewers to be used by ORIMA Research.

4.2 Qualitative research with on the ground observers/stakeholders

Interviews and focus groups will be conducted in Kununurra/Wyndham and Ceduna and Surrounds around the time of the Trial launch (as well as at two-post launch points) with relevant observer groups and on-the-ground stakeholders (members of regional leadership groups as well as government and non-government service providers). The initial round of research will be used to gain a detailed understanding of on-the-ground conditions prior to the Trial, as well as gather insights the community and stakeholders might have about the Trial itself. The second and third rounds of research will focus on how the Trial has impacted individuals and the broader community, relating to the area of expertise on which the observers are able and qualified to answer. Stakeholders will be selected for participation in the research based on their capacity to provide informed feedback relevant to the CDCT. Selection will be informed by desk research, the outcomes of the pre-fieldwork consultations and consultations with the Evaluation Steering Committee.

A21

Table 11: Interviews and focus groups with observers / on-the-ground stakeholders

Who will we talk to? Researched how?

How many?

When? (Ceduna / Kununurra / Wyndham)

Observers / on the ground stakeholders:

 Regional Leadership Groups; and  Government and non-government service providers

 4 group

discussions  10 individual interviews

At three points:

 Initial conditions (April/May 2016),  Wave 1 (August/September 2016), and  Wave 2 (February/March

2017).

(Total 75 people per site, 25 per visit)

4.3 Quantitative research

Two waves of quantitative, face-to-face survey interviews will be undertaken with CDCT participants, family members of CDCT participants and other community members in both CDCT locations. The first wave will occur between August and September 2016, while the second wave will occur between February and March 2017. These interviews will provide information (stated behaviours, perceptions and observations) on the impact of the CDCT on participants, their families and the communities. The survey findings will be analysed in the context of the findings of other evaluation data collection mechanisms and with appropriate regard for the limitations inherent in self-reported, survey-based feedback.

Over the two survey waves, ORIMA Research will conduct a total of 1,350 face-to-face interviews across the two CDCT locations covering a longitudinal sample of CDCT participants and family members (same people interviewed across the two waves) and a non-longitudinal sample of other community members, as shown in the table below.

Table 12: Face-to-face interviews with CDCT participants, families and community members

Who/what Wave 1

N

(August/September)

Wave 2 N

(February/March)

CDCT participants 325 200^

CDCT participants’ families:

 Partners, siblings, significant others

30 20^

Non-participant community members 50 50#

Total/site N = 405 N = 270

Total across 2 CDCT sites (Ceduna and Kununurra/Wyndham)

N = 810 N = 540

^ Lower N at Wave 2, due to expected attrition #

Independent sample, i.e. not longitudinal

A22

Wave 1 data collection will be conducted as an intercept survey in the vicinity of a range of locations (e.g. outside venues and central meeting points such as the Kununurra Community Resource Centre, local shopping centres, Centrelink, Ceduna Aboriginal Arts and Cultural/Language Centre, etc.), using a systematic and unbiased selection process: approaching every third or fourth person encountered in each location.

The second wave of research (Wave 2) will be conducted face-to-face, but primarily by appointment as Wave 1 interviewers will collect the contact details of most Wave 1 respondents (CDCT participants and family members) and these will then be followed up at Wave 2. Non-participant community members will be interviewed via an intercept survey in Wave 2 (same approach as in Wave 1).

Initial selection of survey respondents via systematic intercept sampling at neutral public places is the most statistically robust sampling approach that is available for the study. Cultural sensitivities preclude the adoption of a door-to-door household survey. Legal privacy constraints preclude the selection of a probability sample from Department of Human Services (DHS) administrative data on CDCT participants. Lack of access to landline and mobile telephones as well as cultural barriers to participating in a telephone interview mean that probability based sampling from local telephone number listings would lead to considerable statistical coverage bias.

A number of research design features will minimise the extent of coverage bias (i.e. the extent to which members of the target underlying population have a zero probability of selection):

 Overcoming cultural engagement barriers by conducting fieldwork using an interviewing team of local Indigenous interviewers, experienced Indigenous interviewers from outside of the local area (this will address barriers that are likely to arise for some respondents in relation to sharing personal information with local people who may be connected socially with them), and an experienced ORIMA non-Indigenous field manager;

 Selection of appropriate intercept locations based on advice from local stakeholders and pre-fieldwork observation / site inspection by senior ORIMA personnel;

 In each fieldwork location a marquee will be set-up for interviews to be conducted in an environment that maximises interviewer and interviewee privacy, safety and confidentiality (this will minimise barriers that may arise due to fear of lack of privacy or harassment as a result of participating in the survey);

 Promotion of the value and bona fides of the survey via pre-fieldwork communications (via local community organisations and service providers); and

 Conducting the survey fieldwork over a two-week period in each location, which will minimise the risk of failing to provide an opportunity for members of the target population to come across the interviewing team.

Identity and contact information will be obtained from survey respondents in the first wave of the survey (primarily to enable follow-up interviews in the second wave for CDCT participants and family members of CDCT participants). This information will be verified via inspection of a form of proof of identification (e.g. debit card or driver’s licence). This measure will minimise the risk of people attempting to participate in the survey on more than one occasion in each wave of the survey. In addition, at the data processing stage, survey responses will be checked for duplicate identification details and any duplicates identified will be removed from analysis.

A23

Notwithstanding the abovementioned measures it is likely that the sample selection process will produce a degree of sample selection bias (in the sense that the probability of selection will differ across the target population). In addition, it is expected that there may be differential non-response rates among different groups within the target population. We will control for these issues at the data analysis stage via weighting the raw survey results using population parameters obtained from DHS administrative data and ABS population data. This form of weighting (known as calibration) will effectively deal with these issues and associated measurement biases (at the cost of a reduction in effective sample size - i.e. higher degree of sampling error / lower level of statistical precision).

The sample sizes for the study have been selected based on the following considerations:

 Available resources and constraints;

 Requirement to obtain statistically precise findings in relation to CDCT participants:

 at the aggregate level (i.e. estimates relating to the total CDCT participant population);

 at the level of each of the CDCT sites (Ceduna and Kununurra/Wyndham) - with each site of separate and equal analytical importance;

 separately for men and women; and

 separately for Indigenous and non-Indigenous participants;

 Requirement to obtain indicative (unbiased but not statistically precise) findings in relation to CDCT participants’ families and other community members; and

 Desirability of minimising the overall study burden placed on CDCT participants, their families and their local communities.

Recruitment and training of interviewers

ORIMA Research will deploy an interview team at each location that will comprise:

 ORIMA’s fieldwork manager (a highly experienced non-Indigenous person);

 Two experienced interviewers from ORIMA’s specialised Indigenous interviewers who are not based in the CDCT communities (both are Indigenous people); and

 Two Indigenous people recruited from the local community and trained for the purposes of this project.

By having a mixed team of existing and new interviewers, we will provide a supportive environment for our interviewers to share learnings, experiences and strategies to facilitate skill development and minimise any challenges and potential harm from the interview process. Our existing interviewers are older, well respected community members and have considerable interview experience.

To recruit local Indigenous interviewers, ORIMA Research will actively network with community-based groups within the region(s) where the interviewing is required.

ORIMA Research will conduct initial training with all new fieldworkers following their selection from the recruitment process. As a minimum, training will include:

 the general principles of market, opinion and social research;

 ethical requirements, including respondent safeguards and data protection issues;

A24

 the treatment of children or any vulnerable respondents they may encounter;

 interviewing skills and/or other relevant techniques; and

 interview role playing.

The ORIMA Research fieldwork manager will accompany interviewers on each day of fieldwork with feedback provided to the interviewers as required.

Initial training will last for at least six hours and will cover:

 a structured training session that covers the points described above;

 tablet operations and software training;

 practice interviews with other interviewers or ORIMA Research staff; and

 coaching (including conducting interviews that are observed by the ORIMA data collection manager).

Fieldwork management

In each fieldwork location a marquee will be set-up for interviews to be conducted in an environment that maximises interviewer and interviewee privacy, safety and confidentiality. Such a process ensures that both interviewers and interviewees are not easily visible or identifiable to the wider community. Interviews will be conducted via Computer Assisted Personal Interviewing (CAPI), whereby answers to interview questions will be entered into a tablet computer by the interviewers.

Our procedures will include:

 Conducting a full-day training workshop at each survey site for the interviewing team;

 Having our highly experienced national fieldwork manager for initial and on-going interviewer training as well as support throughout the fieldwork;

 Interviewers will be observed in field and receive feedback from validation of their work (a minimum of 10% of interviews will be observed by our fieldwork manager);

 Conducting daily briefings to ensure that any potential issues or concerns are proactively addressed and allowing opportunities for feedback on skill enhancement/development;

 Conducting an end of fieldwork debriefing process which incorporates strategies for addressing any current and anticipated sensitivities and concerns (e.g. how to deal with interviewees who may raise the subject matter with interviewers after the fieldwork period); and

 Having an established network of supportive relationships with key community leaders and stakeholders on-the-ground for our interviewers to access on a needs basis.

Interviewers will be supplied with:

 an ORIMA ID, which includes a validity period and the contact details for ORIMA Research;

 a tablet computer on which to conduct interviews; and

 brief notes, a hard copy questionnaire, information sheets on support services available at each site and reimbursements.

A25

For each wave of research, respondents will receive a voucher to compensate them for their time ($30 value in Wave 1 and $50 value in Wave 2). The vouchers will be sourced from local services. For example, in Oak Valley we have arranged for the vouchers to be provided through the Oak Valley Outback store to enable purchase of items from this local store. Similarly, in other locations we plan to use local food stores and services for the provision of these vouchers.

4.4 Collation and analysis of administrative data

ORIMA Research will collate and conduct analysis of relevant administrative / secondary data. Wherever possible, the data will be compared at two time points - at Baseline (12 months prior to Trial launch) and at Wave 2 (10-12 months into the Trial), i.e. a pre-post Trial comparison. A listing of data sources and key areas of interest is shown in the table below and reflects the earlier outlined KPIs and indicators of potential spill-over benefits and adverse consequences.

Table 13: Analysis of Administrative / secondary data

How/What When / Evaluation phase

Analysis of administrative / secondary data:

 DHS data on proportion of income support payments to Trial participants that are quarantined and number of crisis payment applications

 Indue (card provider) data on activation and usage of the card, including account balances  Data collated by DSS from State and NGO service providers on number of people using drug and alcohol support

services and family/financial support services  Available State Government data. For example:  Police reports of assault and burglary offences; drink driving / drug driving; domestic violence incidence

reports; drunk and disorderly conduct; outstanding driving and vehicle fines.  School attendance rates  Child protection substantiations  Disruptive behaviour in public housing  Rent arrears in public housing  Drug / alcohol-related hospital admissions / emergency presentations / sobering up service admissions

Collated throughout Trial period

Collated throughout Trial period

Collated and compared at two points:

 Baseline (12 months preceding the Trial)

 Wave 2 (10-12 months post-launch)

CDCT Comparison Sites

Movements in statistics (e.g. changes in drug / alcohol-related hospital admissions) that will be used in assessing the impact of the CDCT could occur due to either the impact of the CDCT or other (external) factors (e.g. decrease in the general availability of certain kinds of illicit drugs in Australia). In order to assess the possible impact of these external factors (so as to better estimate the impact of the CDCT), wherever possible, movements in Trial site statistics will be compared with those in

A26

comparable locations where the CDCT has not been implemented. The latter will provide an indication of what would have happened in the Trial sites in the absence of the CDCT.

A27

These comparison sites do not represent perfect “control sites” and differences in movement of community statistics over the CDCT period cannot be solely attributed to the impact of the CDCT. Nevertheless, it is the intention that these comparison sites be similar in character to the CDCT sites (in terms of underlying demographic and socio-economic characteristics) and that comparing the movement in community statistics of the CDCT and comparison sites would usefully supplement the other information gathered over the course of the evaluation.

The South Australian and Western Australian State Governments have suggested comparison areas for Ceduna and Surrounds and the East Kimberley (or Kununurra/Wyndham), respectively, and have agreed to provide relevant data for these comparison areas. In particular:

 the South Australian State Government has suggested that Coober Pedy and Port Augusta be used as comparison sites for the Ceduna and Surrounds CDCT site; and

 the Western Australian State Government has suggested that Derby be used as the comparison site for the East Kimberley CDCT site.

We consider that the proposed comparison sites are appropriate given that they are similar in character to the CDCT sites in terms of underlying demographic and socio-economic characteristics.

In terms of the South Australian CDCT and comparison sites, in 2011:

 Ceduna had a usual resident population of approximately 4,200, of which approximately 30% were Indigenous;

 Coober Pedy had a usual resident population of approximately 1,500, of which approximately 20% were Indigenous; and

 Port Augusta had a usual resident population of approximately 13,000, of which approximately 20% were Indigenous.

The Socio-Economic Indexes for Areas (SEIFA, based on 2011 Census data) for Ceduna, Coober Pedy and Port Augusta indicate that all are relatively disadvantaged. All three have similar proportions of the population who are Indigenous. However, compared to Ceduna, Coober Pedy has less than half the population, while Port Augusta has almost four times the population. Although local issues facing these three communities differ, Coober Pedy has similar liquor restrictions in place as Ceduna. We consider that Coober Pedy would serve as an appropriate primary comparison site for Ceduna and Port Augusta could serve as a useful secondary comparison site. Having a secondary site may assist where data for the primary site (Coober Pedy) is unavailable, unreliable and/or not suitable for comparison purposes . Moreover, Port Augusta has a range of similar services (e.g. Sobering Up unit) as Ceduna, potentially making extra comparison data available.

In terms of the Western Australian CDCT and comparison sites, in 2011:

 Kununurra had a usual resident population of approximately 7,800, of which approximately 40% were Indigenous; and

 Derby had a usual resident population of approximately 3,300, of which approximately 45% were Indigenous.

Geographically, Derby and Kununurra are both located in the Kimberley region of WA. Kununurra and Derby are both relatively disadvantaged with similar SEIFA values. Taken in conjunction with their geographic proximity and indigenous population ratios, this indicates that Derby represents a reasonable comparison site for the Kununurra CDCT site.

A28

One of the important considerations for the evaluation will be the question of ‘attribution’ of any changes observed to the CDCT. The research design is intended to yield a range of data which, collectively, will reveal if there has been a change in the trial communities. The comparison sites will assist in interpreting any such changes and understanding whether they are broader effects that just happen to affect the trial communities, or localised to the area where the trial is occurring.

The trial sites involve both the introduction of the cashless debit card itself, but also the increased provision of support services. This makes it more difficult to identify what is the impact (if any) of the debit card, what is the impact of the additional services, and what is the impact of the combination. As there are no comparison sites where only one or the other of the interventions has been trialled, we need to use more indirect ways to tease out the distinction. Qualitative information will assist this, and this will be supported by administrative data about service use which is made available to the evaluation. However, the main way of examining the effect of the debit card itself may ultimately come from examining any differences between CDCT participants in the survey who have used or not used the services available.

5 Timing of evaluation reporting Key reporting milestones are as follows:

 An Initial Conditions report by July 2016;

 A Wave 1 Interim Report by December 2016;

 A Wave 2 Interim Report by May 2017; and

 A Final Report by June 2017.

A29

6 Challenges in evaluating the Cashless Debit Card Trial All evaluations face a number of conceptual and practical challenges that need to be addressed in order to observe processes and measure impacts accurately. This evaluation presents a number of significant challenges, some of which are generic to Indigenous research, while others are particular to the income payment quarantining context. Below we have outlined some of the main challenges we foresee, taking into account the contextual environment and objectives of the evaluation.37

Table 14: Key challenges and considerations specific to the project

Challenge/ consideration

How we will address this challenge / consideration

Maintaining sensitivity with at‐risk families This project will need to be highly sensitive to issues of perceived coercion and government and research intrusion into families’ time and

personal environment. For both Indigenous and non‐Indigenous families, the evaluation will need to be responsive to factors such as socio-demographic characteristics, previous experience with government agencies, and potentially low engagement with social research.

Ensuring independence between the evaluator and the Trial design and implementation teams

At all times, the ORIMA Research analysis and reporting team will remain at arm’s length from the design and program implementation teams. All liaison and necessary communication will be conducted via the Department's Evaluation Unit which is responsible for managing the evaluation within DSS and / or the Department’s on the ground contact officers. Issues identified by ORIMA Research around Trial implementation and the Debit Card program will be raised directly with the Department and any response / further communication with the program implementation and design teams will be left strictly to the Department.

Logistical challenges of the research fieldwork The need for the evaluation to stand up to robust scrutiny and to ascertain differences between audience segments will demand a

substantial evaluation program in terms of sample size across both locations. The fact that much of the research fieldwork will need to be

37 This chapter has been informed by the following income management program evaluation reports: Deloitte Access Economics (2015) Consolidated Place Based Income Management Evaluation Report 2012-2015, for DSS; DSS Evaluation Hub (2014) A Review of Child Protection Income Management in

Western Australia: Final report; ORIMA Research (2010) Evaluation of the Child Protection Scheme of Income Management and Voluntary Income Management Measures in Western Australia, for Commonwealth Department of Families, Housing, Community Services and Indigenous Affairs (FaHCSIA); Social Policy Research Centre (2010) Evaluation Framework for New Income Management, for FaHCSIA; Social Policy Research Centre (2014) Voluntary Income Management in the Anangu Pitjantjatjara Yankunytjatjara (APY) Lands, for DSS; Social Policy Research Centre (2014) Evaluating New Income Management in the Northern Territory: Final Evaluation Report, for DSS.

A30

Challenge/ consideration

How we will address this challenge / consideration

undertaken in the East Kimberley (which is largely inaccessible during the wet season) adds a further element of logistical difficulty to the evaluation. The resource demands of the project will be compounded by the geographic remoteness of the research locations, and consequent time‐consuming nature of travel to, from and within these areas. Furthermore, based on prior experience, we expect that in these areas significant time will be spent building rapport in communities prior to conducting fieldwork, as well as in unplanned for ‘downtime’. Considerable time, effort and logistical resources will therefore need to be brought to bear to successfully arrange and conduct the evaluation program in the time available. These factors have, in part, informed our decision to recruit local field workers and interviewers.

The sensitivity of the subject matter From our experience with similar evaluations, as well as with other studies targeting income support recipients, it is clear that collecting

representative information from all of the target audiences in this evaluation will present a challenge. Financial matters can be sensitive for some people to discuss - overlaying these issues with cultural factors in relation to gender roles, child neglect issues and the historically often difficult relationship between Indigenous communities and government, creates a potentially difficult mix. These issues should not be avoided, but rather recognised and dealt with appropriately to ensure the research design and data collection approaches are developed so as to ensure these issues do not obstruct the collection of high quality, reliable data or create any additional discomfort for the community and individuals involved. In addition to evaluation design issues, a sound understanding of the multiple factors ‘external’ to the CDC Trial itself, but nevertheless capable of impacting on the evaluation outcomes, will be vital. For instance, it will be critical for the researchers involved in conducting the qualitative research to establish credibility in the areas of questioning in order to have a robust dialogue that will elicit rich and detailed information from participants. This in turn will depend on the evaluation team having an understanding of the broader issues in relation to Indigenous welfare and disadvantage in general and welfare quarantining in particular, so that the collection, synthesis and interpretation of data and the subsequent development of recommendations is appropriate and comprehensive.

Difficulty of

‘attribution’ and

isolating Trial impact on participants from

impact of other

concurrent factors

One of ORIMA Research’s responses to this challenge is to deploy a number of independent data sources on trial impact and participant experiences. If all or most data sources are pointing to a specific set of conclusions, it provides stronger evidence of impact than one data source. Thus, survey feedback from Trial participants, feedback from local leaders and stakeholders, and administrative data will all be deployed to assess both total and disaggregated impact of all the Trial and non-Trial changes taking place in local communities. Administrative data will also be compared against corresponding data

A31

Challenge/ consideration

How we will address this challenge / consideration

in appropriate non-trial or comparison areas in SA and WA to help assess the impact of non-CDCT factors on movements in Trial site statistics. The evaluation will therefore use several sources of complementary qualitative and quantitative information and will ‘triangulate’ the data sources to both verify the consistency of data collected, and to understand the potential impact and contribution of other factors on the Trial sites and the participants. Using a longitudinal data collection approach means we can isolate the impact of the CDCT on Trial participants on a ‘case-by-case’ basis. Self-reports from individuals on the Trial will tell us what they are doing and experiencing in response to the Trial itself and what, if any, changes in their lives are taking place in response to provision of new support services for example. These self-reports will of course be checked on an aggregate level when we look at service usage data. All these ‘case studies’ will then be ‘aggregated up’ to give us a clear picture of precisely what (in the mix of changes taking place in each Trial community) is and is not impacting on Trial participants (as well those not on the Trial). This approach is important for the evaluation in order to assess and isolate the individual contribution of the Debit Card to individual and community functioning, while simultaneously acknowledging and isolating other factors.

Developing practical strategies and

recommendations to inform any future

rollout of income

quarantining programs

Notwithstanding the complexity of the contextual environment within which the evaluation is being conducted, the success of the evaluation program will hinge on the evaluation team’s effectiveness in being able to clearly and succinctly synthesise, interpret and analyse the feedback elicited from respondents. The ability to subsequently develop practical, clear guidance to inform the evaluation and potential subsequent rollout of CDCT on a broader basis will be a critical success factor. The lessons learned from previous complex evaluations have informed the design of and our overall approach to this evaluation.

Table 15: Generic challenges and considerations

Challenge/ consideration

How we will address this challenge / consideration

Maintaining engagement and

involvement of all

stakeholder agencies

Due to the range of stakeholders involved in this project, maintaining communication, awareness and engagement will be critical to the project’s success. Clear lines of reporting between the Departmental project team, consultancy team and other stakeholders will be essential and all stakeholders will need to have a shared understanding of the roles of the different agencies and their staff.

Questionnaire and discussion guide

techniques do not

The very high level of questionnaire and discussion guide design experience within ORIMA Research makes it unlikely that there will be any serious problems with wording or design of the evaluation materials. The survey and discussion guides will be drafted by senior

A32

Challenge/ consideration

How we will address this challenge / consideration

answer objectives members of the project team and overseen by the project manager, to ensure they meet need and facilitate participation across a spectrum of the interview and group participants.

Outputs do not meet the Department and Steering Committee’s expectations

Ongoing communication with the Department and an effective inception / start-up workshop will be critical to ensuring that the deliverables meet expectations. We feel that the amount of contact we will have with the Department throughout this project will ensure that our outputs meet expectations. All outputs will be submitted in draft form to be agreed with the Department and the frequent contact up to this point means the Department will already have a good understanding of the emerging findings. In addition, each deliverable is subject to Quality Assurance and oversight from at least one Director of ORIMA Research. In this case, Szymon Duniec will provide both strategic project oversight and approve all deliverables prior to these being forwarded to the Department. This is another significant step in our approach to minimising risks of any project.

Timetable slippage A strong evaluation team has been assembled with individual roles defined, led by a highly experienced and senior Associate Partner. The scale of ORIMA Research resources also means that this is not a serious risk. Adequate moderating and interviewing resources will be allocated to ensure that fieldwork is finished to schedule. In addition, ensuring high quality recruitment at the outset will assist in delivering the quantitative fieldwork within the required timeframe. The timetable we have proposed is achievable but is contingent on all parties adhering to milestone dates. In meeting our commitment to the timetable we will provide regular updates to the Department on progress vs milestones achieved and monitor fieldwork closely. We aim for transparency with our stakeholders so that if problems with the timetable emerged, these will be shared. There would be three main recovery options depending on the reason for the slippage:

 Increasing the size of the project team;  Drawing additional resources on tasks such as discussion guide and data analysis or report writing; and  Assigning more senior resources to the team if the timetable

slippage is due to unforeseen circumstances.

Department of Social Services

Cashless Debit Card Trial Evaluation

Wave 1 Interim Evaluation Report

February 2017

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

About this Report

This is the Wave 1 Interim Evaluation Report of the Cashless Debit Card Trial (CDCT) being conducted in Ceduna and Surrounds (South Australia; SA) and in the East Kimberley (EK) region (Western Australia, WA).

Initial Conditions Report

Wave 1

Interim Evaluation Report

Wave 2

Final Evaluation Report

July 2016 January 2017 Anticipated June 2017

Qualitative research with 37 stakeholders and community leaders in the Trial communities

Qualitative research with 73 stakeholders and community leaders in the Trial communities + quantitative surveys with 552 participants, 78 family members of participants and 110 general community

members (non-Trial participants)

+ administrative data

Qualitative research + quantitative research + administrative data

Details To Be Confirmed

This report consists of several layers of information and data, suited to different readers and purposes. As these layers build on top of each other, some content is repeated across multiple layers as relevant. Readers are suggested to utilise the layer(s) most suited to their needs, and to seek more detailed data from deeper layers as and when required.

The layers are:

1. Executive Summary. A brief narrative summary of the CDCT and its objectives, and the key findings from the Wave 1 Interim Evaluation Report (Part I).

2. Overview of Performance against the KPIs. A summary of key survey results, qualitative observations and administrative data which specifically relate to the Evaluation Key Performance Indicators (KPIs) of the CDCT, including an overview table of KPIs (Part III).

3. Response to Evaluation Questions. A discussion of the broader Evaluation Questions, drawing together and considering evidence from all data sources as they relate to these questions (Part IV).

4. Conclusions. A succinct statement of the conclusions as at the Wave 1 Interim Evaluation (Part V).

5. Quantitative Survey Results. The detailed survey results in chart and table form, with minimal commentary. These results are presented in two sections - those relating to KPIs (Appendix A) and those relating to other facets of the CDCT (Appendix B).

6. Qualitative Research Summary Reports. Detailed descriptive results from the qualitative research with stakeholders and community leaders in each of the Trial sites (Appendix C).

Information on the evaluation methodology can be seen in Part II, and in Appendix D: Organisations Interviewed and Contacted in Qualitative Research.

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Contents

I. Executive Summary ............................................................................................... 1

II. Background ........................................................................................................... 6

1. Overview of the Cashless Debit Card Trial ............................................................... 6

2. Role of the Evaluation .............................................................................................. 7

3. Sources of Data ........................................................................................................ 8

III. Overview of Performance against KPIs .................................................................. 15

IV. Responses to Evaluation Questions ....................................................................... 19

1. What have been the effects of the CDCT on program participants, their families and the broader community? .................................................................. 20

2. Have there been any circumvention behaviours? ................................................. 34

3. Have there been any other unintended adverse consequences? ......................... 36

4. What lessons can be learnt throughout the Trial to improve delivery and to inform future policy? ............................................................................................. 38

V. Conclusions .......................................................................................................... 44

Appendix A: Performance Indicators and Wave 1 Results ............................................. 45

1. Output Performance Indicators ............................................................................. 45

2. Short-Term Outcome Performance Indicators ...................................................... 73

3. Medium-Term Outcome Performance Indicators ............................................... 100

Appendix B: Non-Performance Indicator Results ........................................................ 119

Appendix C: Qualitative Summary Reports................................................................. 147

Ceduna and Surrounds ............................................................................................... 147

East Kimberley ............................................................................................................ 159

Appendix D: Organisations Interviewed and Contacted in Qualitative Research ......... 173

Appendix E: Interview Questionnaire Results ............................................................. 175

1

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

I. Executive Summary

Background

With support from the Department of the Prime Minister and Cabinet (PM&C), and developed in close consultation with local community leaders, local and state government agencies and other Australian Government agencies, the Department of Social Services (DSS) is conducting a 12-month trial of a Cashless Debit Card for income support payments (ISPs) in two regional communities.

The Cashless Debit Card Trial (CDCT) aims to reduce the levels of harm associated with alcohol consumption, illicit drug use and gambling by limiting Trial participants’ access to cash and by preventing the purchase of alcohol or gambling products (other than lottery tickets). Between 50% and 80% of CDCT participants’ ISPs are directed to a restricted bank account, accessed by the debit card, with the remainder of these payments accessible through a normal (unrestricted) bank account. Participation in the Trial is mandatory for all working age ISP recipients in the selected Trial sites. Wage earners, Age Pensioners and Veterans’ Affairs Pensioners who live in the Trial sites can opt in to the CDCT1. To support the implementation of the Trial, DSS worked with the South Australian and Western Australian state governments, community agencies and local Indigenous leadership to supplement the support services being provided in the Trial areas with significant further investment.

The Trial commenced in Ceduna and Surrounds (South Australia, SA) on 15 March 2016; and in the East Kimberley (EK) region (Western Australia, WA) on 26 April 2016.

Three evaluation reports are planned across the period of the Trial, with this being the second of these. It is based on data collected during the first six months of the Trial (up to 4 October 2016).

Initial Conditions Report

Wave 1

Interim Evaluation Report

Wave 2

Final Evaluation Report

July 2016 January 2017 Anticipated June 2017

Qualitative research with 37 stakeholders and community leaders in the trial communities

Qualitative research with 73 stakeholders and community leaders in the Trial communities + quantitative surveys with 552 participants, 78 family members of participants and 110 general community

members (non-Trial participants) + administrative data

Qualitative research + quantitative research + administrative data Details To Be Confirmed

Prior to the evaluation commencing a Program Logic for the CDCT was developed. From this a series of Key Performance Indicators (KPIs),2 and where relevant targets, were agreed against which its performance would be primarily evaluated. These were the basis of the Final Evaluation Framework, which also received an ethics approval.

1 N=5 Trial participants opted into the CDCT overall. As at 7 October 2016, only N=2 (both in East Kimberley) were recorded as being ‘ON’ the CDCT.

2 The KPIs can be seen in the CDCT Evaluation Framework which was included as an appendix to the Initial Conditions Report, Appendix A of this report and in CDCT Evaluation Framework Summary available on the DSS website.

2

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

A number of broader Evaluation Questions were also identified as themes to provide greater context and understanding of how and why the CDCT may have an impact, and to assist in fine-tuning or improving its implementation depending on what was learned.

Overview of Findings against KPIs The findings indicate that, overall, the CDCT has been effective to date in terms of its performance against the KPIs established in the evaluation framework. At this early stage of the CDCT, the KPIs underpinning this overall effectiveness assessment were those relating to CDCT program outputs and short term outcomes (ie: those expected to have occurred by 3 months of full implementation). Performance against these KPIs ranged from partially to fully effective. Key findings in relation to these KPIs are set out below.

Output KPIs - performance rated fully effective/ KPI target achieved  All community leaders (members of regional leadership groups) who participated in the Wave 1 qualitative research were supportive of the CDCT. See Appendix D of full report for list of participating leaders  Cashless debit cards (CDCs) were provided to participants and activated in a timely manner,

with nearly all CDCT participants making their first CDC purchase within one month of the first income support payment (ISP) into their CDC account.  DHS data indicates that quarantining of ISPs via CDC accounts has been effective, with a large majority of ISP payments to CDC accounts in the early stages of the Trial (98% of total value

up to end June 2016) being a result of quarantining at the rate of 80%. The data indicates that 2% of the total value of ISP payments to CDC accounts during this period was made to CDCT participants with an approved lower rate of quarantining.

Output KPIs - performance rated partially effective/ KPI target not achieved  Participant understanding of CDC conditions and functionality has improved over time. In the Wave 1 survey, nearly all participants understood that people cannot buy alcohol with the CDC. However, the Wave 1 evaluation research found that significant awareness/

understanding gaps remained.  A large majority of Trial participants surveyed (78% average across the two Trial sites) reported that they had not changed where and how they shopped since the Trial commenced, indicating that they had reasonable access to merchants and products via the CDC. A minority

(18% average across the two Trial sites) expressed concerns about constraints on their ability to access allowable goods and services via the CDC. These proportions were close to KPI target levels (90% and no more than 10%, respectively). Problems reported by participants and stakeholders primarily related to participants’ constrained ability to make allowable purchases in settings where cash was the normal payment medium.  Most community leaders in both Trial sites considered that community panels had not been

established in a timely manner (in Ceduna the panel was established shortly after trial commencement but not at commencement; and in EK the panel had not been established by the time of the Wave 1 fieldwork). These panels were developed at the local community level.

Outcome KPIs - performance rated fully effective/ KPI target achieved  Indicators relating to alcohol consumption showed positive interim results. The Wave 1 survey found that (on average across the two Trial sites) 25% of CDCT participants and 13% of their family members reported drinking alcohol less frequently since the Trial commenced

(with around 1-2% in each case reporting that they were drinking more frequently). Moreover, 25% of CDCT participants reported engaging in binge drinking less frequently since having a CDC, while only 3% reported binge drinking more frequently. Many EK stakeholders

3

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

and some Ceduna stakeholders who participated in the Wave 1 qualitative research had noticed positive changes since the commencement of the Trial that were indicative of lower levels of alcohol consumption in their communities (particularly levels of problematic consumption). In addition, in the Wave 1 survey, substantial proportions of non-participant community members (41% average across the two Trial sites), CDCT participants (24%) and CDCT participants’ family members (28%) reported that they had noticed a reduction in the drinking of alcohol in their community since the Trial started. However, it should be noted that seasonal factors (particularly the cold and wet winter conditions in Ceduna) may have resulted in less drinking of alcohol in public spaces and hence influenced these perceptions.  Indicators relating to illegal drug use showed some improvement at the Interim Evaluation

stage. The Wave 1 survey found that (on average across the two Trial sites), around a quarter of CDCT participants who reported using illegal drugs before the Trial commenced indicated that they had been using illegal drugs less often since becoming CDCT participants.  Indicators relating to gambling also recorded positive short-term outcome results. The Wave 1 survey found that (on average across the two Trial sites) 28% of non-participant community members, 27% of CDCT participants and 28% of CDCT participants’ family members had noticed a reduction in gambling in their community since the Trial started.

Outcome KPIs - performance rated partially effective/ KPI target not achieved  Awareness was limited among CDCT participants in relation to local drug and alcohol support services, as well as financial and family support services.

Interim Responses to Evaluation Questions

What have been the effects of the CDCT on program participants, their families and the broader community?

Both quantitative and qualitative evidence indicates that the first few months of the CDCT has seen a reduction in all three targeted behaviours - alcohol consumption, gambling and use of drugs. Among the 66% of participants who reported drinking alcohol, gambling or taking illegal drugs before or during the Trial, one third (33%) reported a reduction in at least one of these behaviours.

At the time of the Wave 1 data collection, there was some preliminary evidence to suggest that there has been a reduction in crime, violence and harm related to alcohol consumption, illegal drug use and gambling since the Trial commenced. These are expected to be impacted more in the medium and longer term, but police crime statistics together with the reported perceptions of non-participant community members, community leaders and other stakeholders did provide some preliminary evidence of a reduction in crime and violence in Ceduna and Surrounds. It was anticipated before the Trial that there was some risk of an increase in crime when the CDCT started as a way of people seeking to obtain cash, but in the main, at Wave 1 this was not considered to have eventuated. Administrative data relating to sobering up services in the Trial communities showed some positive preliminary signs in relation to problematic alcohol consumption. In particular, sobering up services in East Kimberley recorded lower numbers of cases post Trial than at pre-Trial baseline. In addition, alcohol and drug related referrals to the Kimberley Mental Health and Drug Service declined after the commencement of the Trial.

There was little evidence of change in perceptions of safety. In Kununurra (in EK) it was felt that there may be fewer intoxicated people in the parks, making them safer, but that night time was still problematic.

4

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

There have been some other positive impacts observed in the community. Overall, stakeholders and community leaders felt the Trial has had some positive impacts on participants’ financial capacity (eg: better able to save, money available to spend on children, fewer requests for emergency funds), as well as nutrition and health within their communities (eg: purchasing of more food, school lunches, sobriety, and engagement with programs). In particular, there was a notable increase in East Kimberley community leaders’/stakeholders’ average ratings (on a scale of 0 to 10) in relation to the ability of people in their community to afford basic household goods (3.7 to 5.6) and pay bills (3.5 to 5.5) as well as nutrition in the community (3.2 to 4.6). A significant proportion of participants in the survey indicated they had been able to save more money (31%), care for children better (31%) and improve at using technology (21%).

Overall, perceptions of the impact of the Trial varied between those involved in it and those in the general community. More participants said the CDCT had made their lives worse than made it better (49% compared to 22%). Family members of trial participants gave a similar pattern of answers (37% and 27%). However, non-participants had the reverse perception, with 46% saying the Trial had made life in their community better, and only 18% that it had made life worse. Non-participants in East Kimberley were somewhat more positive than those in Ceduna.

Segmenting participants by self-reported behaviour change across the three target behaviours - alcohol consumption, gambling or illegal drug use - showed that participants who reported positive behaviour change on at least one of the three target behaviours were more likely to say that the Trial has made their lives better (30%), compared to those who reported no change (22%). No participants who reported negative behaviour change (more) said that the Trial had made their lives better, though this was a very small group (n=8).

Have there been any circumvention behaviours that have undermined the effectiveness of the CDCT?

Community leaders and other stakeholders interviewed at Wave 1 indicated that they had heard of various CDCT circumventions having occurred. However, they were unable to comment on how widespread such practices were, and it was not possible to quantify the extent of these reported circumventions. It is likely that neither successful circumventions nor the existence of some sources of income outside of the Trial (such as royalties or emergency assistance payments) could have replaced more than a small proportion of the total value of ISPs quarantined by the CDCT.

Have there been any other unintended adverse consequences?

Perceptions of the impact of the CDCT on humbugging3 has been varied. Stakeholders and community leaders generally felt humbugging had reduced. For example, people who were known to be on the Trial now were either not asked or had an easy answer to decline providing money by stating that they had limited cash because of the Trial. However, stakeholders noted that people known to have access to cash (eg: age pensioners) may be more likely to now be targeted. In contrast, the survey showed that participants and family members both felt that the overall level of humbugging had gone up since the Trial started. This was higher in EK than in Ceduna, but in both sites more people in these groups thought humbugging had increased than thought it had decreased.

3 Making unreasonable financial demands on family members or other local community members.

5

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

A few stakeholders in the qualitative research felt that some CDCT participants felt a sense of shame or stigma associated with having a CDC (especially those who felt they were already managing their money appropriately). However, in the quantitative survey only 6% of all participants explicitly raised stigma or shame associated with the card as an issue.

Beyond that, there have been some issues related to the experience of participants using the cards. Typically these related to specific transactions that were difficult without cash, and to effective use of the card itself and its available features.

What lessons can be learnt to improve delivery and to inform future policy?

Given the early stage of the Trial, it is promising that there are signs of the CDCT working in both sites. It was not possible at this stage of the evaluation to reliably assess where the Trial has worked most and least successfully. In the survey data, demographically there were only fairly minor variations seen in the responses of participants. Overall, the pattern of responses varied little by gender, although men were significantly more likely than women to believe that the CDCT has made their lives worse. There was a somewhat more variation seen across age groups. In particular, the 18-24 age group showed generally the most positive profile of changes, and it was the 55+ age group who were the least positive about the effect of the Trial on their lives.

Given the absence of material changes in other influential factors and conditions, the positive short term impacts reported since the commencement of the CDCT appear likely to be largely attributable to the Trial. Moreover, as the majority of participants had not used any of the existing and additional support services provided in Trial sites, the Wave 1 survey results indicate that the debit card itself had a separate and significant impact on participants’ behaviours. The survey results were suggestive of an additive effect of services on the small proportion of the population using them, but that this was only a relatively small effect for a relatively small proportion of the total participant population.

The Interim Evaluation findings show there are opportunities for improvement in the implementation of the CDCT during the remainder of the Trial period. Initiatives that could be considered include ongoing communications, and continuing to identify and communicate solutions to functionality issues.

Conclusions 1. Overall, the CDCT has been effective to date in terms of its performance against the key performance indicators (KPIs) established in the evaluation framework. 2. In particular, the Trial has been effective in reducing alcohol consumption, illegal drug use

and gambling - establishing a clear ‘proof-of-concept’ and meeting the necessary preconditions for the planned medium-term outcomes in relation to reduced levels of harm related to these behaviours. 3. The Interim Evaluation findings indicate that the reductions in these behaviours have been largely driven by the impact of the debit card quarantining mechanism and not by the additional services provided via the CDCT package or factors external to the CDCT. 4. At this interim stage there is only limited evidence of early impacts on crime, violence, injuries and perceptions of safety - though as medium-term outcomes these were not expected to be seen in this timeframe and will be a focus of Wave 2 of the evaluation.

6

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

II. Background

1. Overview of the Cashless Debit Card Trial

The Cashless Debit Card Trial (CDCT) is a co-designed program developed through collaboration between government and two local communities. The aim of the CDCT is to reduce the levels of harm associated with alcohol consumption, illicit drug use and gambling within the communities of Ceduna and Surrounds in South Australia and East Kimberley in Western Australia (Kununurra and Wyndham). Both communities are relatively small (with populations of around 3,000 and 6,000 respectively) and geographically remote. Such remote sites in Australia typically have significant economic and social challenges, but their relative isolation does allow them to be more effective test sites.

The Trial has been led by the Department of Social Services (DSS), with support from the Department of the Prime Minister and Cabinet (PM&C), and developed in close consultation with local community leaders, local and state government agencies and other Australian Government agencies. Trial participants have been issued with a debit card which cannot be used to buy alcohol, gambling products (with the exception of lottery tickets) or to withdraw cash. Eighty percent of a Trial participant’s income support payments (ISPs) are placed into a restricted account linked to the cashless card (100% of lump sum payments and arrears payments), with the remainder of these payments accessible through a normal (unrestricted) bank account. The percentage of funds accessible in an unrestricted manner (e.g. as cash) may be varied by local community panels, up to 50%.

Participation in the Trial is mandatory for all working age ISP recipients in the selected Trial sites. In addition, wage earners, Age Pensioners and Veterans’ Affairs Pensioners who live in the Trial sites can opt in to the CDCT.

To support the implementation of the Trial, DSS worked with the South Australian and Western Australian State Governments, community agencies and local Indigenous leadership to supplement the support services being provided in the Trial areas with significant further investment.

The Trial commenced in Ceduna and Surrounds on 15 March 2016; and in East Kimberley on 26 April 2016.

Left: Kununurra Above: Wyndham Right: Ceduna

Ceduna South Australia

Kununurra & Wyndham Western Australia

7

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

2. Role of the Evaluation

Framework

ORIMA Research has been commissioned by (DSS) to independently evaluate the Trial in both locations using qualitative and quantitative research methods.

ORIMA Research has developed a formal evaluation framework which specifies the scope of the evaluation and the key performance indicators (KPIs) that will lead its assessment of the effectiveness of the CDCT.

The overall evaluation design and process has been informed by feedback from:

 respected academics and commentators with expertise in conducting research and evaluations involving Aboriginal and Torres Strait Islander people, as expert advisors to the Steering Committee;

 leaders and representatives of Aboriginal corporations and community organisations in the Ceduna and Surrounds and East Kimberley regions; and

 officers of Australian and state government agencies with on-the-ground experience in the Trial sites.

Objective

The overall objective of the evaluation is to assess the effectiveness of the CDCT against agreed KPIs.

Broader evaluation questions also include:

1. What have been the effects of the CDCT on program participants, their families and the

broader community?

 Have there been reductions in the consumption of alcohol, illegal drug use, or gambling?

 Has there been a reduction in crime, violence and harm related to these behaviours?

 Has there been an increase in perceptions of safety in the Trial locations?

 Have there been any other positive impacts (e.g. increase in self-reported well-being,

reduction in financial stress)?

2. Have there been any circumvention behaviours (e.g. participants selling goods purchased

with cashless debit cards to obtain more cash, increase in humbugging or theft) that have

undermined the effectiveness of the CDCT?

3. Have there been any other unintended adverse consequences (e.g. feelings of shame, social

exclusion)?

4. What lessons can be learnt throughout the Trial to improve delivery and to inform future

policy?

8

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

 How do effects differ among different groups of participants (e.g. men compared to

women, people from different age groups)?

 Where has the Trial worked most and least successfully?

 To what extent can any changes be attributed to the Trial as opposed to external

factors such as alcohol restrictions?

 Can the contribution of the debit card be distinguished from that of the additional

services in the Trial locations provided via the CDCT support package?

3. Sources of Data

The evaluation is based on data from three principal sources:

 Administrative data;

 Interviews and focus groups with community leaders and stakeholders (qualitative research); and

 Face-to-face interviews with Trial participants, family members of Trial participants and other non-participant community members residing in the Trial sites (quantitative survey).

This Interim Evaluation Report presents findings from all of these sources. It is based on data collected during the first six months of the Trial (up to 4 October 2016).

Administrative data

The administrative data presented in this Interim Evaluation Report includes income support payment data from the Department of Human Services and Indue, as well as other available data from service providers and state government agencies.

Interviews and focus groups with community leaders and other stakeholders

To date, interviews4 and focus groups with community leaders and other on-the ground stakeholders in the Trial sites have been conducted in the Trial communities at two points in time:

 Pre-Trial launch - conducted between 21 April and 26 May 2016 across Ceduna and Surrounds and East Kimberley; and

 At Wave 1 - conducted between 15 August and 15 September 2016 in Ceduna, and between 12 September and 4 October in East Kimberley.

4 Interviews were conducted either face-to-face or via telephone.

9

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Stakeholders were selected for participation in the research based on their capacity to provide relevant and informed feedback. Selection was informed by desk research, the outcomes of the pre-fieldwork consultations and discussions with the Evaluation Steering Committee.

Table 1: Number of community leaders and stakeholders5 participating in the research

Ceduna and Surrounds6

East Kimberley7 Total

Pre-Trial launch 15 22 37

Wave 1 33 40 73

In Wave 1 there were 28 community leaders and stakeholders who were contacted but not interviewed. Of these, only 16 declined to participate, with the others being cases where an interview at a mutually suitable time was not able to be organised.

All qualitative research was conducted by ORIMA’s specialist qualitative research team. This team has extensive experience conducting research with Indigenous people and in regional Australia, and has participated in cultural awareness training sessions.

The research was qualitative in nature, and hence the results and findings are presented in a qualitative manner. This research approach does not allow for the exact number of participants holding a particular view on individual issues to be measured. This Interim Evaluation Report, therefore, provides an indication of themes and reactions among research participants rather than exact proportions of participants who felt a certain way. The following terms used in this Interim Evaluation Report provide a qualitative indication and approximation of size in relation to the proportion of research participants who held particular views:

 Most—refers to findings that relate to more than three quarters of the research participants;

 Many—refers to findings that relate to more than half of the research participants;

 Some—refers to findings that relate to around a third of the research participants; and

 A few—refers to findings that relate to less than a quarter of research participants.

Face-to-face interviews with the Trial communities

Two waves of face-to-face survey interviews were planned to be undertaken with Trial participants, their families and other community members.

The first wave of survey fieldwork was conducted in Ceduna and Surrounds from 17-28 August 2016. This included day visits to Thevenard, Yalata and Oak Valley. A second visit to Yalata was unable to be completed due to a death in the community.

Wave 1 survey fieldwork was conducted in East Kimberley from 12-23 September 2016. Interviews were conducted in Kununurra, Wyndham, Glen Hill, Cockatoo Springs, Mirima and Nulleywah.

5 Please refer to Appendix D: Organisations Interviewed and Contacted in Qualitative Research for further detail 6 Includes participants in Ceduna, Koonibba, Scotdesco and Yalata 7

Includes participants in Kununurra and Wyndham

10

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

This Interim Evaluation Report presents the findings of the first survey wave.

The surveys were conducted by ORIMA’s Indigenous Fieldforce, consisting of trained Indigenous interviewers supported by other experienced researcher interviewers and some local Indigenous people in support roles. A local cultural awareness session was conducted with the initial interviewing team and the field manager before interviewing commenced.

The surveys used a systematic intercept sampling methodology. High traffic sites around the communities were identified. The interviewing teams were then rostered to fixed locations or roving teams for specified times. During scheduled sessions interviewers, and in some cases dedicated ‘interceptors’, approached every Xth person who passed by a designated point to conduct an interview. The frequency was adapted to suit traffic volumes, but never dropped below every 2nd person. This approach is commonly used in intercept interviewing methodologies to assist in randomising the sample of participants, allowing more confident extrapolation to the wider population of interest. People who agreed to participate in the survey were then screened into the Participant, Family Member or Non-Participant surveys. Quotas for family members and non-participants were expected to be filled quickly, and once full only participants were screened in to an interview.

Table 2: Wave 1 Starting Maximum Sample Size Quotas Ceduna East Kimberley Total

Trial participants 325 325 650

Family members of Trial participants 30 30 60

Non-participants of the Trial 50 50 100

Total 405 405 810

Despite the much smaller overall population, the same nominal maximum quotas were set in Ceduna as East Kimberley. Once actual achieved numbers were known in Ceduna, then the target numbers were adapted in EK to suit. (Wave 2 targets will be determined prior to that phase commencing.)

In total 286 interviews were achieved in Ceduna during the specified fieldwork period. A larger interviewing team was deployed to EK in response to the observed response rates and other challenges relating to the nature of the CDCT which emerged while interviewing in Ceduna. This enabled interviewers to operate in larger teams, and to increase the number of interviews which could be achieved. A total of 454 interviews were completed in EK.

Table 3: Wave 1 Sample Sizes of survey respondents Ceduna East Kimberley Total

Trial participants 196 356 552

Family members of Trial participants 32 46 78

Non-participants of the Trial 58 52 110

Total 286 454 740

The family and non-participant quotas were all achieved and in some cases exceeded, and a total of 552 CDCT participants were interviewed across the two sites.

11

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Participation rates in the quantitative surveys were reasonable for an intercept methodology, with refusals somewhat lower in Ceduna than EK. In EK the proportion of refusals and the proportion of people who agreed to be surveyed were approximately equal, whereas in Ceduna intercepted people were around three times more likely to agree to be interviewed than to refuse.

Table 4: Wave 1 Number of Refusals and screen-outs by methodology Ceduna East Kimberley Total

Qualitative Refusals 4 12 16

Quantitative survey n % n % n %

Completes 286 31% 454 15% 740 19%

Refusals 89 10% 444 15% 533 13%

Screen-outs (total) 560 60% 2157 71% 2717 68%

Under 18 17 2% 93 3% 110 3%

Already completed 129 14% 630 21% 759 19%

Tourist / out of area 221 24% 621 20% 842 21%

Language 12 1% 11 0% 23 1%

Can’t be interviewed 14 1% 63 2% 77 2%

Other 167 18% 739 24% 906 23%

Total intercepts 935 3055 3990

Weighting

Survey data is typically weighted to balance obtained samples against known population characteristics. This maximises the confidence with which results can be extrapolated to the wider population.

In this case, two weighting approaches were employed. First, separate weights were created for the participant results in each Trial location, and then an additional weight was created for the calculation of aggregate results across both Trial sites.

For the two individual trial sites:

 For participants, the survey results were weighted independently for Ceduna and East Kimberley to enable analysis at each site. This weighting aligned the distribution of respondents with that of their respective population distributions of CDCT participants on three known population characteristics - age, gender and Indigenous / non-Indigenous origin. The benchmark population distribution data was provided by DHS.  Results labelled Ceduna Participant or East Kimberley Participant have been weighted in this

way.

 The Family and Non-Participant sub-groups across sites were not weighted due to low sample sizes.

12

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

In order to provide an overall aggregate measure across both sites, an additional step in the weighting was needed to balance the different sample sizes at the two sites. Despite the different population sizes, equal weight was given to both locations - so that they each contributed 50% of the overall result reported. This location weight was applied on top of the individual participant weighting created for the calculation of results at each site.

 Results labelled Participant Average have been weighted in this way.

 The Family and Non-Participant sub-groups were also weighted equally across sites to give the Family Average and Non-Participant Average results.

Statistical precision

Table 5 provides indicative confidence intervals (at the 95% level of statistical confidence) for different response sizes within the survey, allowing for the impact of weighting as outlined above.

For this survey, overall percentage results for questions answered by at least 500 respondents have a degree of sampling error (i.e. confidence interval) at the 95% level of statistical confidence of +/- 5 percentage points (pp). That is, there is a 95% probability (abstracting from non-sampling error) that the percentage results will be within +/- 5pp of the results that would have been obtained if the entire target population had responded.

Table 5: Indicative confidence intervals - 95% confidence level Response size (n) Statistical precision (percentage points)

500 +/- 5pp

350 +/- 6pp

200 +/- 8pp

150 +/-9pp

100 +/- 11pp

80 +/- 13pp

40 +/- 18pp

Higher degrees of sampling error apply to questions answered by fewer respondents and to results for sub-groups of respondents. This is important, because it impacts on the statistical significance of observed differences. In general terms, the smaller the sample size, the larger the difference needs to be in order to be statistically significant (ie: to enable us to conclude that the observation is likely to be a real difference and not just due to natural variation in the sample).

13

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

In reality, testing statistical significance is a complex calculation, and the table above is just a guide to understanding how it varies based on sample size. A crude way of conceptualising significance testing is that for a result to be statistically significant, the difference between two numbers needs to be several percentage points in excess of the statistical precision figure shown.

There are several further technical considerations:

i. We use the 95% confidence level for determining significance. This is a commonly used threshold in social research, and means that 95% of the time a difference which exceeds this threshold should indicate a real difference and not just natural variation.

ii. The statistical precision shown above is for results of 50%. As the results being examined become higher or lower, the confidence intervals narrow somewhat. In practical terms this means that the absolute difference between two results needed to be statistically significant is smaller the closer the numbers involved get to 0% or to 100% (eg: at 10% or 90%, the difference needed to be statistically significant is just over half what is needed for a significant difference to 50%).

iii. Weighting data also affects the ‘effective sample size’. The more weighting applied, the lower the effective sample size for the calculation of statistical significance. Here, a design effect of 1.30 has been applied to allow for the effect of the weighting required. This scaling means that somewhat larger differences are required before the threshold for statistical significance is reached.

4. Ethics Approval and Quality Assurance

The Bellberry Human Research Ethics Committee (HREC) reviewed this project in accordance with the National Health and Medical Research Council’s National Statement on Ethical Conduct in Human Research. The Bellberry HREC is constituted and operates in accordance with the National Statement. The Bellberry HREC approved the project on 8 August 2016.

The project was conducted in accordance with international quality standard ISO 20252.

14

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Information on this page is unweighted. Data / charts in the remainder of the Interim Evaluation Report have been weighted to population proportions for analysis purposes, unless otherwise stated.

15

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

III. Overview of Performance against KPIs The evaluation framework specifies a range of Key Performance Indicators (KPIs) to be used for the assessment of the effectiveness of the CDCT, and where relevant, targets against which they should be assessed. This chapter presents an overview of the Interim Evaluation findings in relation to the performance of the CDCT against these KPIs. Detailed results for each KPI are presented in Appendix A.

Overall, the Interim Evaluation (Wave 1) findings indicate that the CDCT has been effective to date in terms of its performance against the evaluation KPIs. At Wave 1, with data having been collected less than 6 months after the commencement of the Trial, the KPIs underpinning the overall effectiveness assessment were those relating to CDCT program outputs and short term outcomes (those expected to have occurred by 3 months of full implementation). Performance against these KPIs is summarised below.

Output KPIs

Performance against output KPIs ranged from partially to fully effective. Specifically:

Fully effective/ KPI target achieved

 All community leaders (i.e. members of regional leadership groups)8 who participated in the Wave 1 qualitative research were supportive of the CDCT, consistent with the findings of the Initial Conditions qualitative research.

 The interim findings did not find any evidence of compulsory Trial participants not being provided with cashless debit cards (CDCs) in a timely manner.  Nearly all CDCT participants (97%) made their first CDC purchase within one month of the first income support payment (ISP) into their CDC account. Of the $10.5 million in ISPs deposited

into CDC accounts on or before 30 September 2016, $10.0 million (95%) was spent on purchases using a CDC.  DHS data indicates that quarantining of ISPs via CDC accounts has been effective, with a large majority of ISP payments to CDC accounts in the early stages of the Trial (98% of total value

up to end June 2016) being a result of quarantining at the rate of 80%.

Partially effective/ KPI target not achieved

 Participant understanding of CDC conditions has improved over time. The Wave 1 survey found that a large majority (80% or more) of CDCT participants understood (in general terms) what goods/ services can be purchased and the merchant types where the card can be used. In particular, nearly all participants (98% average across the two Trial sites) understood that people cannot buy alcohol with the CDC. Additionally, most participants (91% on average) were aware that the card cannot be used to make bets or for other types of gambling. However, awareness gaps remain, particularly in relation to:

o what to do if the card is lost or stolen; o that the card can be used in most places where VISA is accepted;

8 Please refer to Appendix D: Organisations Interviewed and Contacted in Qualitative Research for further detail

16

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

o that the CDC can be used to make online payment transfers to pay bills; o how to set up rent payments; o how to use the card online; and o how to check balances.  A large majority of Trial participants (78% average across the two Trial sites) surveyed

indicated that they had not changed where and how they shopped since the Trial commenced. A minority (18% average across the two Trial sites) expressed concerns about constraints on their ability to access allowable goods and services via the CDC. These proportions were close to KPI target levels (90% and no more than 10%, respectively). Problems reported by participants and stakeholders primarily related to participants’ constrained ability to conduct legitimate purchases in settings where cash was the normal payment medium.  Most community leaders who participated in the Wave 1 qualitative research considered that

the community panel arrangements were an appropriate and reasonable mechanism for adjusting CDCT quarantining restrictions. However, they felt that community panels had not been established in a timely manner. In Ceduna the panel commenced shortly after the commencement of the Trial but was not active at commencement, while in East Kimberley, the community panel had not been established at the time of the Wave 1 qualitative research fieldwork.

Outcome KPIs

Performance against short-term outcome KPIs also ranged from partially to fully effective. Specifically:

Fully effective/ KPI target achieved

 Indicators relating to alcohol consumption showed positive interim results. o The Wave 1 survey found that (on average across the two Trial sites) 41% of non-participant community members, 24% of CDCT participants and 28% of CDCT participants’ family members had noticed a reduction in the drinking of alcohol in

their community since the Trial started. In each case, this was significantly higher than the proportion who felt that the drinking of alcohol had increased (7%, 16% and 10% respectively). However, it should be noted that seasonal factors (particularly the cold and wet winter conditions in Ceduna) may have resulted in less drinking of alcohol in public spaces and hence influenced these perceptions. o Similarly, the survey found that (on average across the two Trial sites, excluding those

who Refused or said Not Applicable) 25% of CDCT participants and 13% of their family members reported drinking alcohol less frequently since the Trial commenced (with around 1-2% in each case reporting that they were drinking more frequently). Moreover, 25% of CDCT participants reported engaging in binge drinking less frequently since having a CDC, while only 3% reported binge drinking more frequently. o Many East Kimberley stakeholders and some Ceduna stakeholders who participated

in the Wave 1 qualitative research had noticed positive changes since the commencement of the Trial that were indicative of lower levels of alcohol consumption in their communities (particularly levels of problematic consumption).  Indicators relating to illegal drug use showed some improvement at the Interim Evaluation stage. o The Wave 1 survey found that (on average across the two Trial sites, excluding those who Refused or said Not Applicable), 24% of CDCT participants who reported using illegal drugs before the Trial commenced indicated that they had been using illegal

17

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

drugs less often since becoming CDCT participants. This was significantly higher than the 3% of these participants who reported using illegal drugs more frequently.

 Indicators relating to gambling also showed positive short-term outcome results. o Electronic Gaming Machine (poker machine) revenue in Ceduna and Surrounds in the period after Trial commencement (April-August 2016) was 15% lower than in the same months in 2015. o The Wave 1 survey found that (on average across the two Trial sites) 28% of non-

participant community members, 27% of CDCT participants and 28% of CDCT participants’ family members had noticed a reduction in gambling in their community since the Trial started. In each case, this was significantly higher than the proportion who felt that gambling had increased (4%, 8% and 7% respectively). o On average across the two Trial sites, excluding those who Refused or said Not

Applicable, 32% of all CDCT participants and 15% of their family members reported gambling less frequently since the Trial commenced (with only 4% of participants and 3% of family reporting that they were gambling more frequently). o Survey-based indicators of problem gambling also showed improvement, with (on average across the two Trial sites, excluding those who Refused or said Not Applicable) 27% of all CDCT participants reporting that they less frequently spent more than $50 a day on gambling and 18% reporting that they less frequently bet more than they could afford to lose.

Partially effective/ KPI target not achieved

 Awareness of local drug and alcohol support services among CDCT participants ranged from 40% in Ceduna and Surrounds to 56% in East Kimberley. Awareness was higher among those who reported using illegal drugs or using prescription drugs for non-medical reasons (68%9 and 65%, respectively) and those who reported that they drank alcohol (48% and 57%).

 Awareness of local financial and family support services among CDCT participants was lower, with 33% of those in Ceduna and Surrounds and 37% of those in East Kimberley reporting awareness. Awareness of financial support services among those who reported that they had experienced financial difficulties in the past 3 months was in line with that of the overall CDCT participant population (32% and 33% respectively).

 On average across the two Trial sites, 15% of CDCT participants reported ever having used a drug or alcohol support service, with around half of these having done so within the last 3 months. This is a baseline measure (not an indication of effectiveness) that will be assessed again during Wave 2 survey fieldwork (February-March 2017).

 In each Trial site, 17% of CDCT participants reported ever having used a financial or family support service, with around half of these having done so within the last 3 months. This is a baseline measure (not an indication of effectiveness) that will be assessed again during Wave 2 survey fieldwork (February-March 2017).

9 This figure should be interpreted with caution due to the low base of respondents n=8

18

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

19

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

IV. Responses to Evaluation Questions The key evaluation questions are:

1. What have been the effects of the CDCT on program participants, their families and the broader community?

 Have there been reductions in the consumption of alcohol, illegal drug use, or gambling?

 Has there been a reduction in crime, violence and harm related to these behaviours?

 Has there been an increase in perceptions of safety in the Trial locations?

 Have there been any other positive impacts (e.g. increase in self-reported well-being, reduction in financial stress)?

2. Have there been any circumvention behaviours (e.g. participants selling goods purchased with cashless debit cards to obtain more cash, increase in humbugging or theft) that have undermined the effectiveness of the CDCT?

3. Have there been any other unintended adverse consequences (e.g. feelings of shame, social exclusion)?

4. What lessons can be learnt throughout the Trial to improve delivery and to inform future policy?

 How do effects differ among different groups of participants (e.g. men compared to women, people from different age groups)?

 Where has the Trial worked most and least successfully?

 To what extent can any changes be attributed to the Trial as opposed to external factors such as alcohol restrictions?

 Can the contribution of the debit card be distinguished from that of the additional services in the Trial locations provided via the CDCT support package?

This chapter directly addresses each of these evaluation questions by drawing on all available data sources. The primary data sources are:

a. The qualitative research with stakeholders and community leaders b. The quantitative data from surveys conducted with participants, family members of participants and non-participant members of the Trial communities

The Interim Evaluation also makes some use of administrative data made available from Australian Government agencies, state government agencies, service providers and other local sources. Only limited reference is made to these sources at this interim stage, as this data has short and varied timeframes, often small base sizes, and is subject to unknown seasonal variations. It is included here to provide additional insight beyond the perceptions of people in the communities, and will play a greater role in the Final Evaluation Report once its reliability and usefulness can be better established.

20

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

1. What have been the effects of the CDCT on program participants, their families and the broader community?

Have there been reductions in the consumption of alcohol, illegal drug use, or gambling?

Both quantitative and qualitative evidence indicates that the first few months of the CDCT has seen a reduction in all three target behaviours.

When asked about changes in their alcohol consumption, gambling or illegal drug use since the commencement of the Trial, almost one quarter of CDCT participants (on average across the two Trial sites) reported a reduction in at least one of these behaviours (see Figure 1.) In contrast, just 2% of participants claimed to have experienced solely an increase (ie: doing at least one of the three target behaviours more and none of them any less). 43% reported no change and 34% reported that they did not drink alcohol, gamble or take illegal drugs before or after the Trial. Among the 66% of participants who reported drinking alcohol, gambling or taking illegal drugs before or after the Trial, one third (33%) reported a reduction in at least one of these behaviours.

Figure 1: Self-reported changes in alcohol consumption, gambling or illegal drug use Base: Participants currently on the Trial - average across the two Trial sites. Excludes those who say ‘Refused’ or ‘Can’t Say’ across all three measures (n=2).

Q44a (P) / Q44c (P) / q44g (P). Lately, have you done any of these things? Drunk grog or alcohol; Gambled; Used an illegal drug like benzos, ice, marijuana or speed. N=546

Alcohol

Alcohol is a (mostly) legal purchase which can be directly impacted by the CDCT’s mechanism of limiting access to cash and preventing use of the debit card to purchase it. Qualitative feedback from

21

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

community leaders and other stakeholders is alcohol consumption appears to be lower and less visible. There is a sense that people are drinking less per person per day, and stakeholders in alcohol-related organisations and service providers report patterns of observations consistent with this (e.g. sobering up facilities, ambulance, police).

Community leaders and stakeholders ratings to a short questionnaire in the qualitative research indicated that alcohol abuse had reduced in their local community between pre-Trial and Wave 1 - in Ceduna from 7.4 to 7.0 out of 10 and in East Kimberley (EK) from 8.3 to 6.8 out of 10 (based on average ratings on a scale of 0 (not at all) to 10 (extremely severe)).

The survey data supports these observations and perceptions. Consumption of alcohol still occurs, with 25% of Ceduna participants and 46% of EK participants reporting they consume alcohol at least weekly. However, virtually no participants or family members reported drinking more than before the Trial started, and, on average across the two Trial sites, 25% of participants and 13% of family members of participants interviewed said they now drank less (Figure 2 - note excludes ‘Refused’ and ‘Not Applicable’). Very few said they drank more, with a net positive change of 24 percentage points (pp) seen amongst participants and 11pp amongst family members.

Figure 2: Change in behaviour since Trial started: Drunk grog or alcohol (% of respondents) Base: Participants currently in Trial and Family. Excludes ‘Refused’ and ‘Not applicable - did not do activity before’.

Q44a (P) / Q26a (F). Since being on the Cashless Debit / Indue Card have you done each of the following more often, less often or the same as before: Drunk grog or alcohol? **Significantly different to zero at the 95% level *Significantly different to zero at the 90% level CAUTION: Note small base sizes for the Family group

Amongst participants, where the question was also asked about having six or more drinks on one occasion, a similar proportion of those who drank alcohol before the Trial said they did this less since

22

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

the Trial commenced (25% - Figure 3). Again, with very few who reported doing this more, there was a net positive change of 22pp.

Figure 3: Change in behaviour since Trial started: Had six or more drinks of grog or alcohol at once (% of respondents) Base: Participants currently in Trial. Excludes ‘Refused’ and ‘Not applicable - did not do activity before’.

Q44b (P). Since being on the Cashless Debit / Indue Card have you done each of the following more often, less often or the same as before: Had six or more drinks of grog or alcohol at one time? **Significantly different to zero at the 95% level *Significantly different to zero at the 90% level

23

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Overall, the survey showed that for all three groups surveyed, there was also a net perception of less drinking in the community since the Trial started. Figure 4 illustrates that there were small proportions who felt they saw more drinking, but in all cases this was outweighed by the proportion who noticed less drinking. Interestingly, the perception that reductions outweighed increases was least common amongst participants (net 8 percentage point (pp) improvement on average across the two Trial sites), somewhat higher amongst family members (18pp) and higher still amongst non-participants (33pp). However, it should be noted that seasonal factors (particularly the cold and wet winter conditions in Ceduna) may have resulted in less drinking of alcohol in public spaces and hence influenced these perceptions.

Figure 4: Noticed a change in drinking of alcohol or grog in the community since the Trial started (% of respondents) Base: Participants, Family and Non-Participants. Excludes ‘Refused’.

Q42a (P) / Q24a (F) / Q16a (NP). Since the Cashless Debit Card / Indue Card Trial started in your community have you noticed more, less or the same amount of: Drinking of alcohol or grog in the community? **Significantly different to zero at the 95% level *Significantly different to zero at the 90% level

Illegal drugs

Use of illegal drugs is difficult to reliably assess due to the illegal and therefore clandestine nature of the behaviour. However, again there is a set of positive indications. Theoretically, being an illegal and therefore predominantly cash-based trade, the reduction in available cash should have the effect of making it generally harder to buy or sell drugs (this is fundamental to the CDCT Program Logic). While most stakeholders and community leaders generally didn’t feel they could comment with authority on drug use, there was a general sense, especially in EK, that this behaviour had decreased.

24

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Community leaders and stakeholders’ ratings to a short questionnaire in the qualitative research indicated that drug use problems had remained stable in Ceduna (6.8 out of 10 pre-Trial, 6.7 Wave 1) and reduced in East Kimberley - from 6.9 pre-Trial to 5.6 out of 10 Wave 1 (based on average ratings on a scale of 0 (not at all) to 10 (extremely severe)).

A few stakeholders identified the following anecdotes as evidence:

 Drug dealers finding it harder to sell their products since the Trial started; and

 Known participants who had reduced the amount of drugs they consumed since the Trial.

The quantitative survey data gives us an evidence base to suggest that this is indeed the case. Excluding those respondents who ‘Refused’ or said ‘Not Applicable - did not do activity before’, an average of 24% of participants across the two Trial sites said they used less illegal drugs than before the Trial (a net improvement of 21pp - see Figure 5) as did 27% of family members (also a net improvement of 21pp - although this was not statistically significant due to very small base size for this sample). There was also a net 12pp improvement in the number who reported spending more than $50 a day on illegal drugs.

Figure 5: Change in behaviour since Trial started: Used an illegal drug (% of respondents)

Base: Participants currently in Trial and Family. Excludes ‘Refused’ and ‘Not applicable - did not do activity before’.

Q44g (P) / Q26c. Since being on the Cashless Debit / Indue Card have you done each of the following more often, less often or the same as before: Used an illegal drug like benzos, ice, marijuana or speed? **Significantly different to zero at the 95% level *Significantly different to zero at the 90% level CAUTION: Note very small base size for Family group

25

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 6: Change in behaviour since Trial started: Spent more than $50 a day on illegal drugs (% of respondents) Base: Participants currently in Trial. Excludes ‘Refused’ and ‘Not applicable - did not do activity before’.

Q44h (P). Since being on the Cashless Debit / Indue Card have you done each of the following more often, less often or the same as before: Spent more than $50 a day on illegal drugs like benzos, ice, marijuana or speed? **Significantly different to zero at the 95% level *Significantly different to zero at the 90% level

Gambling

Similar patterns are also seen with respect to gambling. Qualitatively, stakeholders and community leaders found informal and online gambling difficult to confidently comment on, as it tends to occur in private residences and is not a highly visible activity. Again though, they did have anecdotes to tell about perceived positive impacts.

In Ceduna there is administrative data from the SA Attorney General showing a reduction in Electronic Gaming Machines (EGM) revenue compared to the same time in the previous year (see STO PI#2 in Appendix A). It has also been reported by stakeholders and community leaders that there has been a substantial reduction in money and time spent on legalised gambling in the form of EGMs at the Ceduna Hotel based on their direct observations, feedback from clients / the community and citation of the Attorney General’s report. This was also supported by the observations of the qualitative researchers by comparison to the Initial Conditions visit. However, there is a strong seasonal pattern in the EGM data, and the magnitude of the effect of the CDCT will need to be monitored closely in the Final Evaluation Report.

In EK it was reported that card houses were smaller and running less often, that there may have been fewer card games in public places, and individual stories of people who were now gambling less.

26

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Community leaders and stakeholders’ ratings to a short questionnaire in the qualitative research indicated that problematic gambling had reduced in their local community between pre-Trial and Wave 1 - in Ceduna from 7.7 to 6.5 out of 10 and in East Kimberley from 6.7 to 5.0 out of 10 (based on average ratings on a scale of 0 (not at all) to 10 (extremely severe)).

As with drugs, this perception was backed up by the behaviours and perceptions reported in the quantitative surveys. Asked whether they had seen more or less gambling in the community since the Trial started, an average net improvement of 19-24pp was seen across all three groups surveyed (see Figure 7).

Figure 7: Noticed a change in gambling in the community since the Trial started (% of respondents) Base: Participants, Family and Non-Participants. Excludes ‘Refused’.

Q42c (P) / Q24c (F) / Q16c. Since the Cashless Debit Card / Indue Card Trial started in your community have you noticed more, less or the same amount of: Gambling in the community? **Significantly different to zero at the 95% level *Significantly different to zero at the 90% level

On average across the two Trial sites (excluding those who Refused or said Not Applicable)- 32% of participants and 15% of family members interviewed indicated they were gambling less. Of those who didn’t refuse or say it was not applicable to them, 27% of participants said they spend more than $50 a day on gambling less often since the Trial (Figure 8-Figure 9).

27

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 8: Change in behaviour since Trial started: Gambled (% of respondents) Base: Participants currently in Trial and Family. Excludes ‘Refused’ and ‘Not applicable - did not do activity before’.

Q44c (P) / Q26b (F). Since being on the Cashless Debit / Indue Card have you done each of the following more often, less often or the same as before: Gambled? **Significantly different to zero at the 95% level *Significantly different to zero at the 90% level

Figure 9: Change in behaviour since Trial started: Spent more than $50 a day on gambling (% of respondents) Base: Participants currently in Trial. Excludes ‘Refused’ and ‘Not applicable - did not do activity before’.

Q44d (P). Since being on the Cashless Debit / Indue Card have you done each of the following more often, less often or the same as before: Spent more than $50 a day on gambling? **Significantly different to zero at the 95% level *Significantly different to zero at the 90% level

28

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Has there been a reduction in crime, violence and harm related to these behaviours?

At the time of the Wave 1 data collection, there was only limited evidence to suggest that there has been a reduction in crime, violence and harm related to alcohol consumption, illegal drug use and gambling since the Trial commenced.

Administrative data relating to sobering up services in the Trial communities showed some positive preliminary signs in relation to problematic alcohol consumption. In particular, sobering up services in East Kimberley recorded lower numbers of cases post Trial than at pre-Trial baseline. In addition, alcohol and drug related referrals to the Kimberley Mental Health and Drug Service declined after the commencement of the Trial.

Police administrative data for Ceduna and Surrounds showed some preliminary evidence of a downward trend in the incidence of crime particularly in terms of violent behaviour such as the acts intended to cause injury statistics (see Figure 83, Appendix A: Performance Indicators and Wave 1 Results). However, police statistics for EK did not show any evidence of a decline in crime there (see Figure 84 and Figure 85, Appendix A: Performance Indicators and Wave 1 Results) , with the data available for EK covering only a short period of time at the very early stages of the Trial and able to use only a point-in-time baseline. The individual indicators have small base sizes and move in different ways, making it hard to discern any clear overall pattern.

At the Initial Conditions stage it had been anticipated by some stakeholders and community leaders in both Trial sites that there may actually be an increase in crime when the CDCT started as a way of people seeking to get cash (this was also raised as a possibility in the Forrest Review). In the main, this has not eventuated - though in Kununurra there was a perception amongst some stakeholders that there had been an increase in crimes committed by young children seeking to get cash that they can no longer get from their parents while on the CDCT.

Community leaders and stakeholders’ ratings to a short questionnaire in the qualitative research indicated that violence and other crimes had slightly reduced in their local community between pre-Trial and Wave 1 - in Ceduna from 7.0 to 6.2 out of 10 and in East Kimberley from 8.0 to 6.3 out of 10 (based on average ratings on a scale of 0 (not at all) to 10 (extremely severe)).

In East Kimberley, a couple of positive changes were cited by some stakeholders including:

 A decrease in vandalism of ATMs; and  A reduction in the number of injuries indicative of domestic violence presenting at the hospital.

The Wave 1 survey has established a range of benchmarks about individual experiences of violence and crime that will be revisited at Wave 2 (see Figure 86-Figure 88, Appendix A: Performance Indicators and Wave 1 Results). As with alcohol consumption and gambling, the survey also asked respondents about their perceptions of whether violence in the communities has increased, decreased or stayed the same since the commencement of the Trial. The results (see Figure 10) show that perceptions were mixed, with non-participant community members being more likely to report a decrease in violence than CDCT participants and their family members.

29

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 10: Violence noticed in the community since the Trial started (% of respondents) Base: Participants, Family and Non-Participants. Excludes ‘Refused’.

Q42b (P) / Q24b (F) / Q16b (NP). Since the Cashless Debit Card / Indue Card Trial started in your community have you noticed more, less or the same amount of: Violence in the community? **Significantly different to zero at the 95% level *Significantly different to zero at the 90% level

Has there been an increase in perceptions of safety in the Trial locations?

As there has been no clear impact (positive or negative) yet on crime and violence, it is not surprising that there is also no strong evidence yet of a change in perceptions of safety.

Qualitatively, there is little sense of change in this area, with both Ceduna and Wyndham stakeholders and community leaders reporting ‘no change’. Those in Kununurra suggested that there may have been some gains during the daytime (for example, fewer intoxicated people present in the parks, which made them more available to others to use), but that night time was still problematic.

Community leaders and stakeholders’ ratings to a short questionnaire in the qualitative research indicated that community safety had sightly increased in their local community between pre-Trial and Wave 1 - in Ceduna from 4.6 to 5.0 out of 10 and in East Kimberley from 4.2 to 5.2 out of 10 (based on average ratings on a scale of 0 (very poor) to 10 (very well)).

The Wave 1 survey has established benchmarks for feelings of safety in the community, and these will be compared to equivalent figures at Wave 2. The Wave 1 survey figures are consistent with what was drawn from the qualitative interviews - with feelings of safety in the home and on the streets during the day generally very high, but much lower in the streets at night. The survey data also showed that the EK results were considerably lower than Ceduna at night10, which also supports the views expressed in the qualitative interviews and groups.

10 Later Trial commencement in East Kimberley may partly explain this difference.

30

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Have there been any other positive impacts?

The Program Logic highlights a number of potential spill-over benefits and adverse consequences. The hypothesised spill-over benefits are potential ways in which the program could benefit the community above and beyond the program outcomes. These types of potential benefits are not seen as being central to the Trial’s objectives. Their achievement will be important to monitor and record, but whether or not they are achieved is not an indication of the success or failure of the Trial.

At Wave 1 there is considerable data to show that there are other positive impacts being seen at an individual level across the Trial sites.

Overall, stakeholders and community leaders felt that the Trial has had some positive impacts on participants’ financial capacity, as well as nutrition and health within the community.

Community leaders and stakeholders’ ratings to a short questionnaire in the qualitative research indicated that (based on average ratings on a scale of 0 (very poor) to 10 (very well)):

1. Ability to afford basic household goods had increased in their local community between pre-Trial and Wave 1 - in Ceduna from 4.4 to 5.6 out of 10 and in East Kimberley from 3.7 to 5.6 out of 10.

2. Ability to pay bills had increased in their local community between pre-Trial and Wave 1 - in Ceduna from 4.3 to 5.0 out of 10 and in East Kimberley from 3.5 to 5.5 out of 10.

3. Nutrition had increased in their local community between pre-Trial and Wave 1 - in Ceduna from 4.2 to 4.4 out of 10 and in East Kimberley from 3.2 to 4.6 out of 10.

4. Health and wellbeing had increased in their local community between pre-Trial and Wave 1 - in Ceduna from 4.4 to 4.7 out of 10 and in East Kimberley from 3.5 to 4.5 out of 10.

Specific, qualitative observations from stakeholders and community leaders include:

 Greater purchasing of food - people are observed with ‘trolleys of food rather than bags’.

 More money seems to be being spent on purchases for children - such as food (e.g. for school lunches), clothes, shoes, treats and toys11.

 Some people appear to now be able to save and make larger purchases of appliances and cars.

 Fewer requests for emergency food or money.

11 Indue data on purchases at the level of merchant categories exists (eg: approximately half of expenditure through Indue cards in EK is classified as “Grocery Stores and Supermarkets”). However, this does not have any pre-Trial baseline and does not go to the level of items purchased. Analysis of this data over the longer

timeframe of the full Trial period may allow the identification of macro changes in purchasing patterns across merchant types during the duration of the Trial, but direct pre-Trial comparison data is not anticipated and only where merchant categories are very narrow will this provide any information about specific products.

31

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

 Possible improvements in IT skills12.

 Requests for work (especially for cash jobs).

 More engagement with programs (self-referrals, more persistence).

 Stories of individuals who have been sober or off drugs ‘for the first time’ and starting to do more positive and constructive things with the time and functionality they now have.

 Fewer alcohol-related injuries or ambulance call-outs, and fewer people discharging themselves early from hospital against medical advice (EK).

Quantitative survey results include:

 31% of participants and 23% of family members interviewed said they have been able to save more money than before the Trial.  Of the participants and family members with children - 31% of participants and 30% of family members said they have been better able to care for their children since the

Trial started, and 16% of participants and 7% of family members said they had been more involved in their children’s homework and schooling; and  21% of participants said they had got better at things like using a computer, the internet or a smartphone.

All of these results support the observations of the stakeholders and leaders in the qualitative stages.

Perceptions of the impact of the CDCT on humbugging has been varied. Overall, stakeholders and community leaders felt that the Trial has had a positive impact on reducing humbugging and street begging. For example, people who were known to be on the Trial now were either not asked or had an easy answer to decline providing money by stating that they had limited cash because of the Trial. Another positive effect described was that it had become easier to say no to being humbugged because it was clearer that the money was being sought for alcohol, drugs or gambling and not for food or looking after children.

Community leaders and stakeholders’ ratings to a short questionnaire in the qualitative research indicated that (based on average ratings on a scale of 0 (not at all) to 10 (extremely severe)):

1. Humbugging had reduced in their local community between pre-Trial and Wave 1 - in Ceduna from 6.3 to 4.9 out of 10 and in East Kimberley from 5.9 to 4.7 out of 10.

2. Street begging had slightly reduced in their local community between pre-Trial and Wave 1 - in Ceduna from 5.4 to 4.0 out of 10 and in East Kimberley from 5.0 to 3.9 out of 10.

However, people who were known to have access to cash - such as Age Pensioners - were perceived by some stakeholders to now more likely to be targeted.

12 Improvement in IT skills was envisaged in the evaluation framework as a potential spill-over effect. These spill-over benefits are potential ways in which the program could benefit the community above and beyond the program outcomes. These potential benefits, while premised on previous experience with Income

Management programs, are not seen as being central to the Trial’s objectives.

32

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

In contrast to the qualitative findings, the survey showed that participants and family members both felt that the overall level of humbugging had gone up since the Trial started - though non-participants were more neutral in their views on this. This was higher in EK than in Ceduna, but in both cases more people in these groups thought humbugging had increased than thought it had decreased - with net scores of -17pp and -21pp respectively (Figure 11).

Furthermore, male participants were more likely than females to report a negative change in humbugging (net change -21pp versus -14pp female) and older participants (aged 45-54 years) and those aged 25-34 years were also more likely than others to report this negative change (-19pp to -20pp).

Figure 11: Noticed more humbugging or harassment for money since the Trial started (% of respondents) Base: Participants, Family and Non-Participants. Excludes ‘Refused’.

Q42d (P) / Q24d (F) / Q16d. Since the Cashless Debit / Indue Card started in your community, have you noticed more, less or the same amount of: Humbugging or harassment for money? **Significantly different to zero at the 95% level *Significantly different to zero at the 90% level

Amongst family members, 27% said the Trial had made their family’s life better and 37% that it had made it worse (net -10pp, see

Figure 12). Across participants interviewed, 22% said it had made their lives better and 49% that it had made their lives worse (net change -26pp). These figures were fairly consistent across the two Trial sites.

Segmenting participants by self-reported behaviour change across the three target behaviours - alcohol consumption, gambling or illegal drug use - allows for further exploration of the results (please see Figure 1 for further detail on these groups.) As may be expected, participants who reported positive behaviour change on at least one of the three target behaviours were more likely to say that

33

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

the Trial has made their lives better (30%, net change -13pp), compared to those who reported no change (22%, net change -28pp). No participants who reported negative behaviour change (more) said that the Trial had made their lives better, though this was a very small group (n=8).

In contrast, those participants who said that they had done at least one of the target behaviours more (but none less) or that they had experienced no change were more likely to report that the Trial had made their lives worse ((67%13 and 50% respectively, versus 43% of those who reported positive change).

Figure 12: Impact of the Trial on your life / your family’s life Base: Participants currently in Trial and Family. Excludes ‘Refused’.

Q45 (P) / Q27 (F). Would you say the Cashless Debit / Indue Card has made your life / your family’s life..?

However, non-participants provided quite different responses. Across both sites 46% of the non-participants said the Trial had made life in their community better, and only 18% that it had made life worse - a net change of +28pp. Non-participants in EK (+35pp) were somewhat more positive than those in Ceduna (+21pp).

Respondents who said that the Trial had made their lives better or worse were asked to provide some information about how. The most common reasons that Trial participants and family members gave for their lives being ‘a bit better’ or ‘a lot better’ included:

 More money to spend on other things (e.g. food/rent/clothes) (n=71 participants, n=18 family); and

13 This figure should be interpreted with caution due to the low base of respondents n=8

34

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

 Saving money (easier)/ Keeping more money (n=43 participants, n=5 family).

The most common reasons that participants and family members provided to explain why their lives were ‘a bit worse’ or ‘a lot worse’ related to:

 Not being able to spend money on things you need to (e.g. bills, appointments) or want (e.g. personal items) (n=74 participants, n=8 family).

Trial participants who said that their lives were ‘a bit worse or ‘a lot worse’ also frequently mentioned reasons related to not being able to get cash out / having no cash (n=60), whilst family members also mentioned:

 Not being able to send money to kids/family/friends or buy them presents/ go on excursions (e.g the show) (n=7); and

 Not being able to see how much money you have/ hard to use/ hassle/ don’t know where money goes (n=7).

Non-participants who reported that the Trial had made life in their community better (a lot better or a bit better), mentioned the following reasons:

 More money to spend on other things (e.g. food/rent/clothes) (n=22); and

 Less drinking / violence / people on streets / public intoxication (n=15).

Of those who said that the Trial had made life in their community worse, the most common reason was due to the perception of more stealing and more humbugging (n=10).

2. Have there been any circumvention behaviours?

The CDCT Evaluation Program Logic makes explicit reference to a series of potential program circumventions. They will be important to monitor because if they occur, they could directly undermine the Theory of Change and help explain why outcomes have not been achieved.

The evidence at Wave 1 relating to these types of circumventions comes mostly from the qualitative interviews with community leaders and other stakeholders. Circumventions are difficult to quantify, but amongst the types which have been reported as being known or seen first-hand are:

 Purchasing goods and returning them for cash14;

 Merchants overcharging for a product or services and then refunding the difference in cash or using the difference to buy alcohol for Trial participants;

 Buying legitimate products on the Indue card for other people, and being reimbursed with cash (sometimes for less than the full value);

14 Note that this cannot be done under the epayments code. This code states that refunds must be made to the method that was used to purchase the goods.

35

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

 Buying products which are then sold for cash (again, often at reduced value);

 The black market or ‘sly grogging’ by those who can access alcohol;

 Gambling using non-cash wagers, including giving others the use of CDCs as payment for a lost bet;

 Figuring out secondary purchases that enabled access to cash or prohibited purchases (e.g. the purchase of Paysafe cards was possible with the Indue card at some merchants in Ceduna, which could then be used for online gambling); and

 Transfer systems that enabled secondary access to funds as cash (e.g. by making spurious transfers for ‘rent’ or to BPAY biller accounts that may have even been set up by themselves, and then withdrawing cash from the end receiving account).

Beyond these attempted circumventions, other strategies to get around the reduced cash available to participants reported by stakeholders included:

 Changing the targets of humbugging to those who can access cash; and

 Illegal and undesirable behaviours in order to obtain cash, including crime and a couple of examples of suspected prostitution were reported.

Taxi drivers in Kununurra were identified by some stakeholders as having long been a player in circumventions for previous systems, and that they continued to be in the CDCT.

Another possible factor which, while not a circumvention per se, could impact the Trial is the influx of cash from other sources. It is hypothesised that this may happen in the form of royalties paid, inheritances and tax or superannuation payments, and even non-quarantined payments such as emergency assistance. It is difficult to assess the impact of these, because their quantum is not knowable. Disclosed royalties were believed to be only in the order of $12,000 (at EK), but there were unverified stories of six-figure amounts flowing into the communities from time to time depending on individual circumstances (mainly due to inheritances and / or insurance payments).

However, to put that in context, while those lump-sum events do offer access to cash, they are likely to represent only a very small proportion (and as one-off payments only) of the approximately $10.5 million which has been quarantined via ISPs deposited into CDC accounts on or before 30 September 2016. The apparent magnitude of likely circumventions also appears to be small compared to this total quarantined payments figure.

36

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

3. Have there been any other unintended adverse consequences? A number of potential adverse consequences that could occur as secondary effects of the Trial were identified in the Program Logic. These are important to monitor because it is possible for the Trial to create unintended negative consequences while at the same time achieving its stated objectives.

A few stakeholders in the qualitative research reported that some CDCT participants who thought they spent their money appropriately felt as though they were being penalised and/or discriminated against by being forced to participate. These CDCT participants reportedly felt that there was a stigma and sense of shame associated with having a CDC. Through open-ended comments, only 6% of all participants explicitly raised ‘stigma’ or ‘shame’ associated with the card as an issue.

Beyond that, there have been some issues related to the experience of participants using the cards. Some of the types of situations which have caused challenges for participants include:

 Being able to transfer money to children that are away at boarding schools15;

 Being able to participate in the ‘second hand’ market for used goods;

 Being able to make small transactions at fundamentally cash-based settings16 (e.g. fairs, swimming pools, canteens);

 Being able to make purchases from merchants or services where EFT facilities were unavailable;

 Being told by a merchant out of the area that they cannot accept this card17; and

 Effectively setting up automatic payments and other transactions on their cards18.

Many of these issues are known to DSS, and are either actually achievable with the Indue card or solutions have been developed. However, these were still perceived as issues for some participants.

People who are on the CDCT are heavily reliant on the technology being available to use their cards. Where EFTPOS terminals or other alternatives are not available, they are limited in their ability to make purchases. There was some concern among stakeholders that this leaves them potentially more vulnerable to things like power outages than non-participants might be. This was raised as a concern in the lead-in to the wet season in EK.

15 The Department can increase other external transfer limits (default $200) upon reasonable proof and can place children on an Indue Card at boarding school.

16 The Department advised that the need to have access to cash for such purposes was acknowledged in the co-design process with community leaders and is why 20% of payments are not quarantined.

17 The Department indicated that the CDC has complete coverage of merchants that do not have alcohol or gambling as the main source of business.

18 The Department acknowledged that there were significant issues with the set-up of automatic payments and other transactions at the start of the trial. It advised that this has since been fixed (as of August 2016).

37

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

It was considered a possible risk that people could leave the area before or after the commencement of the Trial, but there is little evidence of this happening to any significant extent.

Figure 13 shows that the number of ISP recipients who moved out of the CDCT sites in the June Quarter of 2016 (i.e. since the introduction of the CDC) was broadly in line with historical experience. It shows that there is historically a steady level of movement out of both sites. There was a larger number who moved out of Ceduna and Surrounds in the March Quarter 2016 compared to the same quarter in 2015, but only by a small proportion compared to the typical quarterly number or compared to the March Quarter 2014. Overall the chart suggests that the CDC has not had a major impact on the number of ISP recipients moving out the CDCT sites. There may have been a small increase immediately prior to the Trial commencing in SA, though given the level of fluctuations observed across quarters this cannot be attributed confidently to the Trial from this data alone.

Figure 13: ISP recipients moving out of the CDCT sites19

19 Source: data provided by DHS on all Centrelink customers in receipt of an Income Support Payment whose address has changed from in the defined community, to out of the defined community during a quarter. The data was extracted in September 2016 and so this Figure presents ISP recipient movement statistics up to the June Quarter 2016 (JQ-2016) - the most recent quarter for which complete data was available at the time of data extraction.

38

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

4. What lessons can be learnt throughout the Trial to improve delivery and to inform future policy?

How do effects differ among different groups of participants (e.g. men compared to women, people from different age groups)?

In the survey data, demographically there were only fairly minor variations seen in the responses of participants (see Tables in Appendix B). Overall, the pattern of responses varied little by gender, although men were significantly more likely than women to believe that the CDCT has made their lives worse. There was a somewhat more variation seen across age groups. In particular, the 18-24 age group showed generally the most positive profile of changes, and it was the 55+ age group who were the least positive about the effect of the Trial on their lives.

Where has the Trial worked most and least successfully?

Given the early stage of the Trial, it is promising that there are signs of the CDCT working in both sites. It is not possible at this stage of the evaluation to reliably assess where the Trial has worked most and least successfully.

To what extent can any changes be attributed to the Trial as opposed to external factors such as alcohol restrictions?

The Trial sites are complex communities. The CDCT is one facet of a larger suite of interventions and activities operating in each of them. This Trial is not an experimental design which allows the isolation of the different strategies, or where the specific effects of the Cashless Debit Card concept can be unequivocally identified.

However, on the face of it, the positive short-term impacts since the CDCT Trial commenced in relation to alcohol consumption, illegal drug use and gambling appear likely to be largely attributable to the CDCT. This is because available evidence (based on stakeholder interviews - see Appendix C) indicates that there has not been a significant contemporaneous change in potentially influential external factors/ conditions.

This preliminary assessment will be rigorously tested after Wave 2 of evaluation data collection when relatively long data series will be available for both the CDCT sites and comparison sites in both SA and WA where the CDCT is not operating.

Can the contribution of the debit card be distinguished from that of the additional services in the Trial locations provided via the CDCT support package?

The measures at the Trial sites include a combination of the CDC itself and a range of additional services. There is not an experimental design where these two facets are also implemented separately

39

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

or in a staggered timing, and so there is no definitive way of isolating their individual effects. However, there is a way within the evaluation design to investigate the relative contributions of the two facets.

The mechanism by which the CDC part of the Trial should have an effect on alcohol consumption, drug use and gambling is very direct, and is not reliant on the provision of additional services. Rather, the role of those services is more additive to assist individuals adapt to the changes the CDC causes in a positive way. In this sense, the CDC could and should be expected to have a distinct effect in its own right.

The best way to examine this hypothesis in this Trial configuration is to look at those individuals who had used the available services, and compare them to those who had not. This classification of participants cannot be perfect, as concepts such as use of particular types of services are hard to definitively measure in a large-scale survey. However, the survey did ask participants whether they were aware of any services and whether they had used any across two broad categories - drug and alcohol services, and financial services. This gives us some ability to isolate and differentiate between these groups of interest.

The tables below show the proportion of all participants who reported they had used services from either of these categories in two timeframes. The first table is those who have ‘ever’ used a service, and the second is just those who have done so ‘within the past 3 months’ or approximately since the commencement of the Trial - and therefore who are the primary group for examination.

There was little overlap between users of drug and alcohol services and users of financial services, but usage of either category is in a small minority of the interviewed participants. Less than a quarter had ever used a service across either category, and just 12% said they had in the past 3 months (P3M)20. This immediately tells us that the provision of services can be making only a relatively small contribution to the total effect of the CDCT, as the great majority of participants have simply not been exposed to the services.

Table 6: Proportion of participants ever using support services Drug and alcohol

services

Financial services

Either Drug and alcohol or Financial

Status*

% all

participants (weighted)

Sample size (Unweighted)

% all

participants (weighted)

Sample size (Unweighted)

% all

participants (weighted)

Sample size (Unweighted)

Ever used 13% 69 15% 78 24% 125

Not used 85% 474 83% 466 76% 424

Refused 1% 8 1% 7 <1%^ 2

* Self-reported. Also note that n=1 participant did not respond to these questions. ^Refused in both categories.

20 It was anticipated that DEX administrative data on service usage may provide a way of establishing a baseline of pre-Trial service usage and then examining changes in usage as the CDCT operated. However, examination of the available DEX data reveals that not all service providers fully utilised the DEX facility to report their activity data in the July through

December 2015 period. Without this baseline data, it is not possible to reliably ascertain the extent to which service usage has increased or decreased since the introduction of the CDCT.

40

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Table 7: Proportion of participants using support services in the past 3 months Drug and alcohol

services

Financial services

Either Drug and alcohol or Financial

Status*

% all

participants (weighted)

Sample size (Unweighted)

% all

participants (weighted)

Sample size (Unweighted)

% all

participants (weighted)

Sample size (Unweighted)

Used past 3 months

7% 32 8% 38

12% 61

Not used past 3 mths 92% 511 91% 506

87% 488

Refused 1% 8 1% 7 <1%^ 2

* Self-reported. Also note that n=1 participant did not respond to these questions. ^Refused in both categories.

The distributions provide indicative sample sizes for an exploration, though the samples for the last three months are very small. Because the two categories are quite different, it makes sense to look at them separately as well as to integrate them into a single compound variable.

Using these categories, we can then look at the key survey questions which ask about changes to behaviours since the commencement of the Trial. There are a range of patterns we could see in this analysis:  If we see that it is only participants who have used services showing changes, then we

would infer that the CDC may be having little independent effect.

 If there are no differences between those using services and those who are not, then we would infer that the services may be having little independent effect.

 If there are effects seen for those who have used services and different effects seen for those who have not used services, then we would infer that both approaches are likely to be having some separate effect.

It is the third of these possibilities that is evident in the results, though they suggest that the contribution of services seems to be much less than the contribution of the CDC itself.

Table 8: Reported behaviour change across service usage segments

Used in past 3 months Drug and alcohol services Financial services Either Drug/Alcohol OR Financial

Since being on the CDCT

Used P3M

Not used P3M Used P3M

Not used P3M Used P3M

Not used P3M

Percent “yes”: n=11-31 n=238-504 n=19-38 n=231-498 n=27-60 n=224-481

You’ve been able to save more money than before [FIN] 31% 31% 42% 31% 34% 31%

You’ve been better able to care for your child/ren 54% 29% 35% 31% 29% 31%

You’ve got more involved in your children’s homework and school 43%* 13% 30% 14% 22% 15%

I’ve got better at things like using a computer, the internet or a smartphone 11% 22%* 40%* 20% 26% 21%

NOTE: * = Statistically significant difference at the 95% confidence level

[FIN]: Financial statement, [D&A]: Drug and Alcohol statement.

41

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Continued from previous page

Used in past 3 months Drug and alcohol services Financial services Either Drug/Alcohol OR Financial

Since being on the CDCT

Used P3M

Not used P3M Used P3M

Not used P3M Used P3M

Not used P3M

Net positive change: n=31 n=508-509 n=38 n=502-503 n=60 n=485-486

Drunk grog or alcohol [D&A] 23% 14% 24% 14% 20% 14%

Had six or more drinks of grog or alcohol at one time [D&A] 11% 11% 7% 12% 9% 12%

Gambled [FIN] 8% 8% 2% 8%* 5% 8%

Spent more than $50 a day on gambling [FIN]

3% 5% 10% 4% 6% 4%

Bet more than you can really afford to lose [FIN] 3% 2% 5% 2% 3% 2%

Had to borrow money or sell things to get money to gamble [FIN] 3% 1% 5% 1% 3% 1%

Used an illegal drug like benzos, ice, marijuana, or speed [D&A] 6% 3% 7% 3% 6% 3%

Spent more than $50 a day on illegal drugs like benzos, ice, marijuana, or speed [D&A] 6% 2% 0% 2%* 2% 1%

NOTE: * = Statistically significant difference at the 95% confidence level

[FIN]: Financial statement, [D&A]: Drug and Alcohol statement.

The small group of participants who had used services in the past 3 months showed a positive trend but few statistically significant differences from those who had not. This is suggestive of an additive effect of services on the small proportion of the population using them, but that this is only a relatively small effect for a relatively small proportion of the total participant population.

In these tables, the ‘not used’ columns are the closest available proxy for the CDC without services. It shows that without use of services, there have been positive effects of the CDCT on most of these behaviours. From this we can infer that a CDC that does not have additional supporting services would still be expected to impact on the targeted behaviours.

Sample sizes are small, but the general trend in the results was for those participants who reported using a service to be slightly more likely to report positive impacts on behaviours.

This data from the survey is congruent with expectations of the CDC Program Logic, and consistent with the general qualitative feedback from the Trial sites. At this interim stage, the CDC component of the Trial does appear to have an effect independent of the services provided around it.

Those services may have a small complementary role of enhancing the effects of the CDC, but this is a relatively smaller effect and limited to the small proportion of the population who access the services.

42

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Program implementation issues and potential improvements

Many participants in the Trial have reported issues and problems at times - with nearly half of all participants interviewed answering yes to the question “Have you had any problems using the card?” (46%, 49% at Ceduna and 43% at EK).

Many of these are likely to be user errors (for example: using a wrong PIN in the case of ‘failed’ transaction attempts) or failures to understand the features and capabilities of the card, while others likely reflect imperfect knowledge and systems amongst some merchants. Some also do reflect limitations of the card and characteristics of its operations.

Regardless of their source, this section seeks to identify where participants have been having issues, drawing on both qualitative observations of participants and feedback from the participants’ themselves. This will enable DSS to target either improvements or communications which can enhance user experiences over the remainder of the Trial. Some of these have been previously brought to the attention of DSS and may already have been addressed.

 Balances21 - card balances are important for people who are operating sometimes on a cashflow availability basis. The concept of balances is not intuitive or familiar for all Trial participants, in particular the distinction between the ‘account / current balance’ and the ‘available balance’.

 More education about how to interpret balances would be beneficial.

 Automatic payments - automatic payments are causing issues for some participants. In many cases it seems likely to be that they are set up incorrectly (e.g. not correctly synced with incoming payments or due dates). On occasions, this seems to be resulting in participants incurring dishonour or late payment fees, which a) costs them available money, and b) can appear to be unauthorised withdrawals. Some participants lack confidence in the security of their money as a result of hearing about or experiencing these occurrences (and some even reported spending much or all of their money as soon as it comes in just to make sure it is not taken away).

Proactive steps to assist participants set up and check these types of payments may be advantageous, and could be particularly targeted at the point of activation.

 Acceptance of the card - the experience of participants is that acceptance is not universal, though technically it should be. There are stories of the cards being declined at stores both within and outside the Trial sites, and though these might be relatively isolated and often due to the user, some cases do seem to involve merchants telling cardholders that they cannot use the particular card.

Additional education to assist participants be more confident about using their card may assist them in trying a second time if they make an error, or being more assertive with merchants. Ensuring merchants in the Trial sites know how to use the cards and to assist participants should they make an error would also be potentially useful.

21 From February 2017 balance checking will be available at selected ATMs

43

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

 Accessing the legitimate “cash economy” - while one of the main intentions of the CDCT is to limit access to cash, this does impose some limitations on legitimate places that participants can make purchases when the available cash component of their payments is not available. Examples cited include canteens, schools, swimming pools, carnivals, petrol stations and second hand private sales; as well as some specific merchants who did not accept card payments (e.g. a funeral director). These limitations are a natural consequence of the Trial, but are also a source of frustration for participants, who can feel ‘discriminated’ against by their inability to access these.

 Rents - accommodation payments often represent substantial proportions of a welfare recipient’s regular income. Some stories were provided of participants whose private landlords would only accept payment in cash or forms that were not easily able to be met by CDCT participants, resulting in difficulties meeting tenancy requirements22.

22 Please note that informal rent arrangements are possible through the Cashless Debit Card hotline if proof is presented.

44

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

V. Conclusions 1. Overall, the CDCT has been effective to date in terms of its performance against the key performance indicators (KPIs) established in the evaluation framework.

2. In particular, the Trial has been effective in reducing alcohol consumption, illegal drug use and gambling - establishing a clear ‘proof-of-concept’ and meeting the necessary preconditions for the planned medium-term outcomes in relation to reduced levels of harm related to these behaviours.

3. The Interim Evaluation findings indicate that the reductions in these behaviours have been largely driven by the impact of the debit card quarantining mechanism and not by the additional services provided via the CDCT package or factors external to the CDCT.

4. At this interim stage there is only limited evidence of early impacts on crime, violence, injuries and perceptions of safety - however, these medium-term outcomes were not expected to be seen in this timeframe and will be the focus of Wave 2 of the evaluation.

45

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Appendix A: Performance Indicators and Wave 1 Results

1. Output Performance Indicators

Output PI#1: Number of community leaders who endorse programme

Figure 14: Output PI #1

Performance Indicator Interim Rating

Number of community leaders who endorse programme Achieved

Theme Details

Specification

Number of community leaders who:

 feel programme design is appropriate for their community characteristics  believe programme will be / is a good thing for their community  speak positively about programme  believe Trial parameters were developed using a co-design

approach

Target NA

Timeframe

Within one month of programme launch (Initial Conditions), repeated at Wave 1 and Wave 2

Data Sources Qualitative research with community leaders

Definitions / Notes Community leaders defined as members of regional leadership groups.

The community leaders initially strongly supported the introduction of the CDCT, and at the Interim Evaluation point this was still the case.

46

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Evidence

Qualitative research with stakeholders

The qualitative research identified that all participating community leaders endorsed the CDCT. Specifically, all community leaders interviewed (see Table 1 for number interviewed):

 felt the programme design was appropriate for their community characteristics;

 believed the CDCT will be a good thing for their community - most were expecting to see the evidence of this in the medium to longer term;

 generally spoke positively about the CDCT; and

 believed Trial parameters were developed using a co-design approach - most leaders had been involved in the design of the Trial, and felt there were adequate opportunities for input from the community.

47

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Output PI#2: Percent of participants who understand card conditions

Figure 15: Output PI #2

Performance Indicator Interim Rating

Percent of participants who understand card conditions Partial

Theme Details

Specification

Percent of participants who are aware:

 how much their welfare income is quarantined in terms of cash withdrawals  what they can and cannot purchase on the card  which merchant types they can and cannot use the card at  they can use the card wherever VISA is accepted including

online (except where a merchant is blocked)  they can use the card to make online payment transfers for housing and other expenses, and to pay bills  what to do if the card is lost or stolen

Target NA

Timeframe Self-reported at Wave 1 and Wave 2

Data Sources

Survey of Trial participants Additional Sources:

Survey of families

Survey of community members

Indue data (Declined transactions, by decline reason, Use of online transfers, Number of participants provided with replacement and temporary cards)

Survey Questions

Participant questionnaire: q16/17, q20a-c, 21a-d

Family questionnaire: q14a-b, q15c-f

Non-Participant questionnaire: q12ia-b, q12iic-f

Definitions / Notes Not applicable

Participant understanding of CDC conditions has improved over time. The Wave 1 survey found that a large majority (80% or more) of CDCT participants understood (in general terms) what goods/ services can be purchased and the merchant types where the card can be used. In particular, nearly all participants (98% average across the two Trial sites) understood that people cannot buy alcohol with the CDC. However, awareness gaps remain, particularly in relation to:

 what to do if the card is lost or stolen;  that the card can be used in most places where VISA is accepted;  that the CDC can be used to make online payment transfers to pay bills;  how to set up rent payments;  how to use the card online; and  how to check balances.

48

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Evidence

Qualitative research with stakeholders

Stakeholders reported there had been some initial “teething problems” while participants became familiar with the CDC (including using a PIN, the requirement to select the ‘credit’23 account and setting up rental transfers). However, they noted that participant understanding of card conditions and functionality had improved over time.

In both East Kimberley and Ceduna, some stakeholders considered that there was a need for additional communication around certain aspects of the CDC, as they were causing confusion or were less well understood. These included:

 how to set up rent payments;

 how the card worked online; and

 how to check balances.

Evidence from external data sources

Analysis of CDC purchase24 transactions that have been declined indicates that participant understanding of CDC conditions improved over time. The following chart shows that the proportion of CDC purchases that have been declined has fallen gradually over time. During the CDC roll-out period, the average CDC transaction decline rate was 16.5%. Subsequently, the average CDC transaction decline rate has been 14.3%. It should be noted that the first two weeks (indicated by dashed lines in) had atypically low transaction volumes - with less than 500 transactions, compared to an average of over 12,000 CDC purchase transactions per week subsequently.

23 Note that card functionality has since been upgraded so that from December 2016 participants can select ‘savings’, ‘cheque’ or ‘credit’.

24 Excludes ISP deposits into CDC accounts.

49

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 16: Declined CDC transactions

Table 9 shows the reasons for declined CDC transactions over the period April to September 2016.

 Card user error accounts for 86% of declined CDC transactions.

 The main reason for transaction declines (accounting for over half of CDC transaction declines) is the CDC cardholder seeking to make a purchase when there are insufficient funds available. Interviews with CDCT participants found that some who did not know how to obtain an account balance simply attempted to make a purchase in order to ascertain whether they had been paid. Interviews also found that there was some confusion about the difference between the “current” and “available” CDC balance and this may have contributed to the number of declined CDC transactions.

 The second most common reason for transaction declines (accounting for around one quarter of transaction declines) is the CDC cardholder providing an incorrect PIN.

 The third most common reason for transaction declines (accounting for 6% of transaction declines) is the CDC cardholder seeking to make a transaction that exceeds the CDC transaction value limit.

 Attempts to use the CDC at prohibited merchants or terminals25 accounted for 8% of declined CDC transactions.

 Other CDC transaction declines26 accounted for the remaining 6% of declined CDC transactions.

25 The reason for these CDC transaction declines was that they were attempted on blocked terminals associated with a merchant trading under a merchant category code which indicates they sell prohibited goods such as alcohol or gambling products.

26 Reasons for these CDC transaction declines include: advised to reject; card listed as restricted; and card listed as stolen.

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

3-Apr-16 10-Apr-16 17-Apr-16 24-Apr-16 1-May-16 8-May-16 15-May-16 22-May-16 29-May-16 5-Jun-16 12-Jun-16 19-Jun-16 26-Jun-16

3-Jul-16 10-Jul-16 17-Jul-16 24-Jul-16 31-Jul-16 7-Aug-16 14-Aug-16 21-Aug-16 28-Aug-16 4-Sep-16

11-Sep-16 18-Sep-16 25-Sep-16

Proportion of CDC purchase transactions declined

Week commencing

Proportion of CDC purchase transactions

CDC Roll-out period

50

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Table 9: Reasons for declined CDC transactions (April to September 2016)

Reason for transaction decline

Declined transactions % of declined transactions

% of CDC purchase transactions*

Card user errors 32,237 86.2% 10.9%

Insufficient funds 20,735 55.5% 7.0%

Incorrect PIN entry 8,416 22.5% 2.8%

Cardholder exceeding withdrawal limit 2,189 5.9% 0.7%

Card listed as lost 897 2.4% 0.3%

Attempts to use card for prohibited purchases 3,022 8.1% 1.0%

Other reasons 2,121 5.7% 0.7%

Total 37,380 100% 12.6%

Source: Indue data. * CDC transactions excluding deposits of ISPs.

The fact that attempts to use the card for prohibited purchases accounted for only 1% of all CDC purchase transactions (8% of all declined transactions) up to 30 September 2016, indicates a high level of understanding among CDCT participants of what cannot be purchased with the card.

Quantitative research with Participants, Families and Non-Participant stakeholders

The survey found that a large majority of participants were able to specify the proportion of their ISP payment that has been quarantined via a CDC account (see Figure 17). Of those who were unable to specify a proportion, a large majority understood that ‘most’ or ‘almost all’ of their ISP has been quarantined (see Figure 18).

Figures 19-22 indicate that a very large majority of CDCT participants understood:

 what people can and can’t buy with the CDC (in particular, nearly all participants understood that people can’t buy alcohol with the CDC); and  in general terms, the merchant types they can and cannot use the card at.

However, Figures 23-25 show that significant awareness gaps were found in relation to:

 what to do if the card is lost or stolen (particularly among Ceduna participants);  that the card can be used in most places where VISA is accepted; and  that the card can be used to make online payment transfers to pay bills.

51

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 17: Proportion of Centrelink payment that goes onto the card Base: Ceduna, East Kimberley and Total Participants. Excludes ‘Refused’.

Q16 (P). How much of your Centrelink payment goes on the Cashless Debit Card / Indue Card?

52

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 18: Proportion of Centrelink payment that goes onto the card if provided in $’s or don’t know Base: Ceduna, East Kimberley and Total Participants who answered ‘provided in $’ or ‘don’t know’ at Q16.

Q17 (P). (IF RESPONSE TO Q16: PROVIDED IN $ OR DON’T KNOW) Is it..?

Figure 19: Knowledge of what you / people can and can’t buy with the card Base: Participants, Family and Non-Participants.

Q20a (P) / Q14a (F) / Q12ia (NP). Do you know what you can and can’t buy with the card?

53

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 20: Knowledge that you can’t buy alcohol or grog with the card Base: Participants, Family and Non-Participants.

Q21a (P) / Q15c (F) / Q12iic (NP). Before this survey did you know that: You can’t buy alcohol or grog with the card?

Figure 21: Knowledge that you can’t use the card to make bets or for other types of gambling Base: Participants, Family and Non-Participants.

Q21b (P) / Q15d (F) / Q12iid (NP). Before this survey did you know that: You can’t use the card to make bets or for other types of gambling?

54

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 22: Knowledge of the types of places where you / people can and can’t use the card Base: Participants, Family and Non-Participants.

Q20b (P) / Q14b (F) / Q12ib (NP). Do you know the types of places or where you can and can’t use the card?

Figure 23: Knowledge of what to do if the card is lost or stolen Base: Participants.

Q20c (P). Do you know what to do if the card is lost or stolen?

55

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 24: Knowledge that you can use the card in most places where VISA is accepted Base: Participants, Family and Non-Participants.

Q21c (P) / Q15e (F) / Q12iie (NP). Before this survey did you know that: You can use the card in most places where VISA cards are accepted, including online or on the internet?

Figure 25: Knowledge that you can use the card to make online payment transfers to pay bills Base: Participants, Family and Non-Participants.

Q21d (P) / Q15f (F) / Q12iif (NP). Before this survey did you know that: You can use the card to make online payment transfers to pay bills, for housing and other expenses?

56

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Output PI#3: Percent of participants in Trial locations sent card

Figure 26: Output PI #3

Performance Indicator Interim Rating

Percent of participants in Trial locations sent card Achieved

Theme Details

Specification Percent of compulsory Trial participants sent a debit card

Target 100%

Timeframe Within two months of program launch

Data Sources Indue / DHS Client database

Definitions / Notes Not applicable

There is no evidence that relevant compulsory Trial participants have not been sent cards, but the evaluation cannot confirm 100% was achieved.

Evidence

Evidence from external data sources

Income Support Payment (ISP) data may be used to assist in the assessment of this performance indicator, in the absence of specific data demonstrating the percent of participants in Trial locations who were sent the card.

The official start dates for the CDCT were:

 15 March 2016 in Ceduna and surrounds; and

 26 April 2016 in the East Kimberley.

Cashless debit cards (CDCs) were progressively distributed to eligible ISP recipients:

 between mid-April and end-May 2016 in Ceduna and surrounds; and

 over the month of June 2016 in East Kimberley.

The number of ISP recipients participating in the CDCT changes each week as people move on and off various ISPs. DHS data indicates that, as at end June 201627 (i.e. around three and a half months after the official start date in Ceduna and surrounds and two months after the official start date in the East Kimberley):

27 The first CDCT participant listing provided to the Evaluation Team.

57

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

 of the 1,972 ISP recipients in the CDCT sites who had been assessed as eligible to be placed on the CDCT, 1,837 (93%) had ISPs paid into their CDC:

 of the 765 ISP recipients in Ceduna & surrounds who had been assessed as eligible to be placed on the CDCT, 728 (95%) had ISPs paid into their CDC; and

 of the 1,207 ISP recipients in the East Kimberley who were eligible to be placed on the CDCT, 1,109 (92%) had ISPs paid into their CDC.

The fact that less than 100% of those initially deemed eligible for the CDCT have received ISP payments via a CDC reflects the fact that people may move on and off ISPs and a number of ISP recipients initially assessed to be eligible for the CDCT would have moved off ISP before the official start of the CDCT. In this regard, of the 1,972 persons deemed eligible to be placed on the CDCT as at end-June 2016:

 92 (4.7% of those originally deemed eligible for the CDCT) had had their payments cancelled; and

 148 (7.5% of those originally deemed eligible for the CDCT) had their ISPs suspended.

58

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Output PI#4: Percent of distributed cards that are activated

Figure 27: Output PI #4

Performance Indicator Interim Rating

Percent of distributed cards that are activated Achieved

Theme Details

Specification

The percentage of CDCT participants who make their first CDC purchase within one month of the first ISP into their CDC account.

Target 95%

Timeframe Within one month of receiving card

Data Sources

Indue Survey of Trial Participants

Questions Participant questionnaire: q13

Definitions / Notes 5% margin allowed for people moving in and out of income support payments

97% of CDCT participants made their first CDC purchase within one month of the first ISP payment into their CDC account.

Evidence

Quantitative research with Participants, Families and Non-Participant stakeholders

Figure 28: Activated and used card Base: Participants.

Q13 (P). And have you activated your Cashless Debit / Indue Card and started using it to buy things?

59

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Evidence from external data sources

As at 30 September 2016, 2,125 CDCT participants had ISPs deposited into a CDC account since the introduction of the CDC in April 2016. Over this period, these CDCT participants had been issued a total of 4,716 CDCs, with some being issued more than one card due to, for example, the loss of the originally issued card or the provision of a temporary card. Figure 29 shows the number of cards issued to individual CDCT participants. Of the 2,125 CDCT participants who had ISPs paid into their CDC account at or before 30 September 2016:

 1,256 (59%) had been issued one CDC;  730 (34%) had been issued between two and five CDCs; and  139 (7%) had been issued more than five CDCs. The maximum number of cards issued to a single CDCT participant over this period was 16.

Figure 29: Number of CDCs issued to individual CDCT participants

Of the 4,716 CDCs that had been issued to CDCT participants as at 30 September 2016:

 3,917 (83%) had been activated by CDCT participants, and  3,495 (74%) had been activated by CDCT participants within one month.

However, as noted above, a number of CDCT participants have been issued multiple CDCs. While CDCT participants may not have activated all cards issued to them, they may nevertheless been able to access the ISPs deposited into their CDC account via subsequently activated CDCs. Of the 3,917 CDCs activated by CDCT participants for whom ISPs were deposited into their CDC account on or before 30 September 2016:

 3,594 (92%) were activated within a month;  3,752 (96%) were activated within two months; and  3,792 (97%) were activated within three months.

Of the 2,125 CDCT participants who had ISPs deposited into a CDC account on or before 30 September 2016, 2,089 (98%) had used a CDC to make purchases over this period. This is consistent with the Wave 1 survey results outlined above, which found that 98% of survey participants reported they had activated and used the CDC to buy things.

0

200

400

600

800

1,000

1,200

1,400

1 2 3 4 5 6 7 8 9 10+

Number of CDCs issued (at or before 30 Sep 2016)

# of CDCT participants

60

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 30 shows the cumulative proportion of CDCT participants who had made their first CDC purchase within one month of their first ISP being deposited into their CDC account. It shows that:

 88% of CDCT participants made their first CDCT purchase within one week of their first ISP being deposited into their CDC account;  93% had made their first CDC purchase within two weeks;  95% had made their first CDC purchase within three weeks; and  97% had made their first CDC purchase with one month.

Figure 30: Proportion of CDCT participants who made their first CDC purchase within one month of their first ISP payment into their CDC account

Most CDCT participants spend almost all of the ISPs deposited into their CDC account within a week. As a consequence, there is a close correlation between total ISPs deposited into CDC accounts and total purchases made with CDCs. Figure 31 shows the total ISPs deposited into CDC accounts and total purchases using CDCs on a daily basis.

88%

93% 95%

97%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Same day + 1 week + 2 weeks + 3 weeks + 1 month

Time between first ISP deposit and first CDC purchase

Cumulate % of CDCT participants

61

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 31: Total ISPs paid into CDC Accounts vs total purchases using CDCs (daily)

 Most individual CDC account balances peak on “pay day” and then quickly decline to a low level prior to the next pay day. Given that around two-thirds of ISPs are deposited in CDC accounts on Wednesday through Friday, this intra-week pattern is also evident for the total CDC balances in both CDCT sites. Figure 32 shows the total CDC Account balances for CDCT participants in the two CDCT sites.28

Figure 32: Total balance of CDC Accounts (daily)

28 The spike in CDC account balances in July 2016 reflects the end of financial year reconciliation payments for Family Tax Benefit made to CDCT participants.

$0

$50,000

$100,000

$150,000

$200,000

$250,000

1-Apr-16 8-Apr-16 15-Apr-16 22-Apr-16 29-Apr-16 6-May-16 13-May-16 20-May-16 27-May-16 3-Jun-16 10-Jun-16 17-Jun-16 24-Jun-16

1-Jul-16 8-Jul-16 15-Jul-16 22-Jul-16 29-Jul-16 5-Aug-16 12-Aug-16 19-Aug-16 26-Aug-16 2-Sep-16 9-Sep-16

16-Sep-16 23-Sep-16 30-Sep-16

Total ISPs deposited into CDC Accounts

Total CDC purchases

$0

$100,000

$200,000

$300,000

$400,000

$500,000

$600,000

$0

$50,000

$100,000

$150,000

$200,000

$250,000

$300,000

1-Apr-16 8-Apr-16 15-Apr-16 22-Apr-16 29-Apr-16 6-May-16 13-May-16 20-May-16 27-May-16 3-Jun-16 10-Jun-16 17-Jun-16 24-Jun-16

1-Jul-16 8-Jul-16 15-Jul-16 22-Jul-16 29-Jul-16 5-Aug-16 12-Aug-16 19-Aug-16 26-Aug-16 2-Sep-16 9-Sep-16 16-Sep-16 23-Sep-16 30-Sep-16

Ceduna & surrounds (LHS axis) East Kimberley (RHS axis)

62

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Of the $10.5 million in ISPs deposited into the CDC accounts of CDCT participants on or before 30 September 2016, $10.0 million (95%) has been spent on purchases using a CDC.

The above evidence indicates that:

 CDCT participants have had timely access to the ISPs deposited into their CDC accounts; and  the vast majority of CDCT participants have activated their CDCs and have started to make purchases with their CDCs.

63

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Output PI#5: 80% of income support payments are quarantined

Figure 33: Output PI #5

Performance Indicator Interim Rating

80% of income support payments are quarantined

Achieved

Theme Details

Specification

Income support payments are quarantined and 20% are received in cash (excluding approved adjustments)

Target 100% of recipients

Timeframe Within two months of program launch

Data Sources

DHS Client database (Regular payment quarantine percentage variable) Survey of Trial participants

Questions Participant questionnaire: q16/17

Definitions / Notes Not Applicable

DHS data indicates that 98% of the total value of ISP payments made to CDC accounts up to end June 2016 was paid to CDCT participants for whom (the default) 80% of their ISP had been paid to these accounts. The data indicates that 2% of the total value of ISP payments to CDC accounts during this period was made to CDCT participants with an approved lower rate of quarantining. The majority of participants surveyed could nominate 80% as the quarantined proportion.

Evidence

Evidence from external data sources

Figure 34 shows the number of people receiving ISPs and making purchases via the CDC over the period April through September 2016. It is evident that after the initial roll-out of the CDC over April through June 2016, the number people receiving ISPs via the CDC has plateaued at around 1,850 each month.

The number of people receiving ISPs via the CDC fluctuates as people move on and off various benefits. In any given month, ISP payments may be made to people currently entitled to ISPs or those who have been entitled to an ISP at some point in that month but who have had their payment cancelled or suspended. Of the approximately 1,850 persons receiving their ISPs each month via the CDC since July 2016, typically around 85% are currently entitled to an ISP and around 15% have had their ISP cancelled or suspended during the month.

64

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 34: Number of people using the CDC

Apr-16 May-16 Jun-16 Jul-16 Aug-16 Sep-16

Paid ISP via CDC 522 1,595 1,798 1,873 1,855 1,840

Used CDC for purchase 424 1,420 1,741 1,867 1,867 1,849

As at end-June 201629 (i.e. at the end of the CDC roll-out period and almost two months after the first ISPs were made via the CDC) a total of $3.12 million in ISPs had been delivered via the CDC. Of this, $3.05 million (98%) was paid to CDC participants for whom (the default) 80% of their ISP was delivered via the CDC. That is, only 2% of the total value of ISPs made via the CDC had been paid to CDCT participants who had successfully applied to a Community Panel for a reduction in their CDC quarantine (to between 50% and 70%). However, it should be noted that, as at end-June 2016, the East Kimberley Community Panel (which will assess individual requests for variations in the CDC quarantine percentage on a case-by-case basis) had not yet been established and so no requests for CDC quarantine reductions had been assessed for the East Kimberley.

Figure 35 shows the total monthly value of CDC transactions over the six-month period April to September 2016:

 The value of CDC transactions rose steadily over the CDC roll-out period of April to June 2016.

 The spike in CDC transactions in July 2016 reflected the payment of End of financial year reconciliation payments for Family Tax Benefit in a number of ISPs30.

 Most ISPs delivered via the CDC are spent within the same month.

29 Based on DHS data extracted 29 June 2016. 30 End-of-year top-ups (paid in July) have a significant impact on Carer Payments, Family Tax Benefit payments,

Parenting Payments

0

500

1,000

1,500

2,000

Apr-16 May-16 Jun-16 Jul-16 Aug-16 Sep-16

People Using CDC, April to September 2016

Paid ISP via CDC

Used CDC for purchase

CDC Roll-out period

65

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 35: Total value of CDC transactions

Apr-16 May-16 Jun-16 Jul-16 Aug-16 Sep-16

Total value of ISPs paid via CDC $286,610 $1,057,719 $1,802,274 $3,125,108 $2,079,518 $2,058,492 Total value of CDC purchases

$187,255 $927,851 $1,688,748 $3,000,069 $2,141,556 $1,998,108

Quantitative research with Participants

Figure 36: Proportion of Centrelink payment that goes onto the card31 Base: Ceduna, East Kimberley and Total Participants. Excludes ‘Refused’.

Q16 (P). How much of your Centrelink payment goes on the Cashless Debit Card / Indue Card?

$0

$500,000

$1,000,000

$1,500,000

$2,000,000

$2,500,000

$3,000,000

$3,500,000

Apr-16 May-16 Jun-16 Jul-16 Aug-16 Sep-16

Total Value of CDC Income and Expenditure

Total value of ISPs paid via CDC

Total value of CDC purchases

CDC Roll-out period

66

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 37: Proportion of Centrelink payment that goes onto the card- provided in $’s or don’t know31 Base: Ceduna, East Kimberley and Total Participants who answered ‘provided in $’ or ‘don’t know’ at Q16.

Q17 (P). (IF RESPONSE TO Q16: PROVIDED IN $ OR DON’T KNOW) Is it..?

31 Customers with active Centrepay deductions have a lower net percentage of funds paid to their CDC, as Centrepay is deducted from the default 80% prior to quarantining of funds. This may account for some of those customers who believe that an amount lower than 80% is being paid to their cards.

67

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Output PI#6: Number of support services available in community

Figure 38: Output PI #6

Performance Indicator Interim Rating

Number of support services available in community Not assessed

Theme Details

Specification

Number and type of additional support services in operation as planned

Target 100%

Timeframe Within three months of program launch

Data Sources DSS provided

Definitions / Notes Not Applicable

Data to assess this output performance indicator was not available at the Interim Evaluation point.

Evidence

Evidence from external data sources

Usage of the support services was to be assessed using Data Exchange (DEX) data to be provided by DSS. At face value, the DEX data suggests that usage of specified support services more than doubled over the six months to June 2016 (which encompasses the CDC roll-out period) compared with the six months to December 2015 (which precedes the CDC roll-out):

 The total of client residents reported as using support services in Ceduna and Surrounds rose from 163 over the six months to December 2015 to 348 over the six months to June 2016.

 The total of client residents reported as using support services in East Kimberley rose from 49 over the six months to December 2015 to 118 over the six months to June 2016.

However, this sharp increase in reported usage reflects the fact many providers only started using the DEX reporting facility in the six months to June 2016. Consequently, the six months to December 2015 does not provide a reliable benchmark against which to assess service usage rates.

The Department is currently exploring the possibility of identifying those services in DEX that have: 1. Fully reported service provision via DEX in the six months to December 2015; and 2. Received additional funding to support the implementation of the CDCT.

For the above reasons, the Department agreed that it was not advisable to include service usage statistics in the Wave 1 Interim Evaluation Report.

68

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Qualitative research with stakeholders

Many service providers reported that the additional service funding was not provided early enough to allow for services to be ready at the beginning of the Trial, particularly given the significant challenges associated with delivering new programs and support services in a remote location.

69

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Output PI#7: Percent of participants with reasonable access to merchants and products

Figure 39: Output PI #7

Performance Indicator Interim Rating

Percent of participants with reasonable access to merchants and products Partial

Theme Details

Specification

Excluding the purchase of alcohol and gambling:

 percent of participants who agree that they can still shop where and how they usually shop  percent reporting concerns over access to allowable products

Target 90% agree point 1, and maximum 10% reporting at point 2

Timeframe Self-reported at Wave 1

Data Sources Survey of Trial participants

Questions q22

Definitions / Notes Not Applicable

A large majority of Trial participants (78% average across the two Trial sites) surveyed indicated that they had not changed where and how they shopped since the Trial commenced. A minority (18% average across the two Trial sites) expressed concerns about constraints on their ability to access allowable goods and services via the CDC. These proportions were close to KPI target levels (90% and no more than 10%, respectively). Problems reported by participants and stakeholders primarily related to participants’ constrained ability to conduct legitimate purchases in settings where cash was the normal payment medium32.

Evidence

Qualitative research with stakeholders

Overall, most stakeholders reported that CDCT participants had adapted to the conditions of the card, and were generally not concerned about accessing allowable products. However, the following were identified as limiting CDCT participants’ ability to access allowable products:

32 The Department advised that 20% of payments are not quarantined due to the need for participants to have access to cash for legitimate purposes. The CDC also allows for $200 to be transferred externally per 28 days. If a legitimate reason can be provided for requiring this limit to be increased then DSS can assist in

this.

70

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

 Some stakeholders in Ceduna felt that CDCT participants were adversely impacted by their inability to make cash-based purchases in some situations, including from second hand websites, sports club canteens, tickets for community events (e.g. Oyster Fest and NAIDOC);

 A couple of stakeholders in East Kimberley reported that they had heard there were some stores where the card was not accepted / did not work (e.g. petrol stations outside of the Trial site and in Derby); and

 One stakeholder in East Kimberley was concerned about the potential impact of blackouts during the upcoming wet season, which usually put EFTPOS terminals out of operation. This stakeholder noted that people were usually able to withdraw cash from banks in this situation, but was concerned about what would happen this wet season with people only able to access 20% cash. This stakeholder reported that this issue could be compounded when the DSS office is closed for Christmas for 2 weeks33.

Quantitative research with Participants

Figure 40: Changed where or how you shop since using the card Base: Participants.

Q22 (P). Since you started using the card, have you had to change where or how you shop?

33 The Department has contingency arrangements for emergency situations and the Christmas shutdown period. These include assistance through the DSS and Indue hotlines as well as officers on the ground and in Canberra who have the ability to transfer money to external accounts in such situations.

71

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 41: Problems had using the card - open ended Base: Participants who have had problems using the card.

Q15 (P). Have you had any problems using the Cashless Debit / Indue Card - Yes. Please tell me about these problems. Unweighted n=245

72

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Output PI#8: Number of community leaders who believe appropriate adjustments are made to income restrictions on a case-by-case basis

Figure 42: Output PI #8

Performance Indicator Interim Rating

Number of community leaders who believe appropriate adjustments are made to income restrictions on a case-by-case basis Partial

Theme Details

Specification

Number of community leaders who:

 believe community panels are assessing applications in a timely, consistent and fair manner  believe community panels are making just and reasonable decisions about changing percentage of welfare payments

quarantined

Target Most

Timeframe

Within one month of programme launch (Initial Conditions), repeated at Wave 1 and Wave 2

Data Sources

Qualitative research with community leaders

Definitions / Notes Nil

The Community Panels have been late to commence in both Trial sites, limiting their effectiveness. Once established the Ceduna Panel has been useful, but the EK Panel only became functional after the Interim Evaluation was completed.

Evidence

Qualitative research with stakeholders

Overall, the qualitative research found that most community leaders felt the community panel arrangements were an appropriate and reasonable method for adjusting income restrictions. However, community leaders identified that community panels had not been established in a timely manner. Specifically:

 In Ceduna, the panel was not established for the start of the Trial; and

 In East Kimberley, the panel had still not been established at Wave 1.

Given that the community panels have not been up and running, there is limited evidence to provide feedback / response against this output measure.

73

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

2. Short-Term Outcome Performance Indicators

STO PI#1: Frequency of use / volume consumed of drugs & alcohol

Figure 43: STO PI #1

Performance Indicator Interim Rating

Frequency of use / volume consumed of drugs & alcohol Achieved

Theme Details

Specification

 Number of times alcohol consumed by participants per week  Percent of participants who say they have used non-prescription drugs in the last week  Number of times per week spend more than $50 a day on

drugs not prescribed by a doctor  Number of times per week have six or more drinks of alcohol at one time (binge drinking)  Percent of participants, family members and general

community members reporting a decrease in drinking of alcohol in the community since commencement of Trial  Number of on-the-ground stakeholders reporting a decrease in drinking of alcohol in the community since

commencement of Trial #

Target NA, other than # target = ‘many’

Timeframe As self-reported at Wave 1

Data Sources

Survey of Trial participants Survey of families Survey of community members Qualitative research with stakeholders

Survey Questions

Participant questionnaire: q25a-b/h-j, q42a, q44a-b/g-h

Family questionnaire: q17a-b/h-j, q24a, q26a/c

Non-Participant questionnaire: q16a

Definitions / Notes

On-the-ground stakeholders defined as members of the regional leadership groups and observers from government and non-government service providers based in the Trial areas

For stakeholders, qualitative indication of number: all, most, many, some, few

Overall, the KPI measures indicate that the Trial has been effective in reducing alcohol and illegal drug consumption.

74

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Evidence

Ceduna - Qualitative research with stakeholders

Frequency of use / volume consumed of alcohol

Many stakeholders were reluctant to indicate that the CDCT had reduced alcohol consumption in Ceduna and surrounding areas as they felt that there were no direct, visible changes to alcohol consumption since the start of the Trial. However, some had noted broader changes that could suggest an indirect impact on reducing alcohol consumption (e.g. more money directed to other purposes and greater engagement with programs).

The few stakeholders who had noted direct positive impacts of CDCT in relation to alcohol consumption reported that these included:

 Less visibility / fewer people consuming alcohol in public. This was evidenced by:

 A few stakeholders seeing and hearing fewer intoxicated people on the streets;

 Fewer ambulance callouts to public places. While call-out figures were unchanged, there were more to private addresses since the Trial begun;

 Fewer people drinking in front and back gardens in one remote community (Koonibba);

 More sobriety - evidenced by:

 A substantial increase (around 50%) in the number of people attending the lunch service at the ADAC day centre (previously most clients only attended breakfast);

 More people attending services at CAC while sober;

 Seeing specific individuals sober - a few stakeholders reported seeing certain individuals in a sober state “for the first time”; and

 Reduced emergency presentations related to alcohol consumption during the end of the financial year period - there tended to be an influx in hospital presentations associated with excessive alcohol consumption as lump sum payments received at this time of year (e.g. school kids bonus) were spent on alcohol, but this year no spike had been noticed by hospital staff.

Frequency of use / volume consumed of drugs

Overall, consistent with the Initial Conditions Report, stakeholders found it more difficult to comment on levels of drug use in the community as substance usage was less visible. There were minimal positive impacts of the CDCT on drug use identified by stakeholders other than a report by one stakeholder who had spoken with a drug dealer (who was on the CDCT) who had not been able to buy wholesale drugs because he no longer had the supply of cash. Another stakeholder reported hearing reports that it was harder to access marijuana in some of the remote Trial communities outside of Ceduna, however, as there had been some recent arrests of drug suppliers it was unclear if this was as a result of these arrests or the CDCT.

75

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

East Kimberley - Qualitative research with stakeholders

Frequency of use / volume consumed of alcohol

While some stakeholders felt that there had been minimal / no change in alcohol consumption, many others had noticed changes at an individual and / or community level since the start of the CDCT.

The stakeholders who had noted positive impacts of CDCT in relation to alcohol consumption reported that these included34:

 Feedback from Sobering Up Units (SUUs) in Kununurra and Wyndham indicated that there had been:

 A decrease in the number of people being picked up as well as using the SUU:

 A change in the types of people being picked up by the SUU - now more likely to be people from outside the Trial area;

 Less visibility / fewer people consuming alcohol in public - some stakeholders reported seeing and hearing fewer intoxicated people in public places (i.e. on the streets in town, in parks and on the side of the road in the early mornings);

 One stakeholder had observed that binge drinking patterns had changed amongst some of the CDCT participants that they had dealings with and that these sessions were now of a shorter duration;

 Hospital related changes:

 Admissions due to alcohol-related presentations had decreased;

 A noticeable decrease in rowdy and abusive behaviour towards Accident and Emergency staff since the Trial. A stakeholder from the hospital estimated it to have reduced by around half;

 Ambulance-related changes:

 Decrease in primary call-outs (from 107 in August-September 2015 to 73 in August-September 2016);

 Fewer call-outs for alcohol-related injuries;

 Substantially decreased sales at a Kununurra bottle shop - this decrease was reported to have occurred since the Trial started, as well as at the comparative point from last year;

 This had resulted in a need to reduce casual staff (who were mostly backpackers);

 More sobriety - evidenced by:

 A local football coach reported that several players who had previously been unable to play / not trained effectively due to frequent intoxication had reduced their alcohol consumption and significantly improved their behaviour and their commitment and performance to the game, resulting in noticeable “transformation” of their lives; and

34 For supporting data see Figure 93 under MTO PI#4: Drug/alcohol-related injuries and hospital admissions.

76

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

 A few stakeholders reported seeing individuals they knew to previously be high alcohol users who were now more regularly sober and seen to be spending their money on food, groceries and household items.

Frequency of use / volume consumed of drugs

Overall, consistent with the Initial Conditions Report, stakeholders found it more difficult to comment on levels of drug use in the community - primarily because such behaviours were not as visible and the impacts of using marijuana (the main drug used) were not as overt / aggressive as alcohol.

The following positive impacts of the CDCT on drug use were identified by the qualitative research:

 A couple of stakeholders had directly spoken with CDCT participants who had reduced and / or stopped their drug use as a result of the CDCT, including:

 A CDCT participant who had previously been addicted to methamphetamines (“ice”) but had stopped using ice due to limited access to cash;

 A family who was now consuming less marijuana due which had allowed them to spend more money on clothes and food for their child and were supportive / happy about the Trial as a result; and

 A few stakeholders felt that the frequency of marijuana usage had reduced due to limited access to cash.

77

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Quantitative research with Participants, Families and Non-Participants

Figure 44: Behaviours done lately: Have grog (a drink containing alcohol) Base: Participants and Family. Excludes ‘Refused’.

Q25a (P) / Q17a (F). Lately, have you done any of these things? Have grog (a drink containing alcohol).

Figure 45: Behaviours done lately: Have six or more drinks of grog / alcohol at one time Base: Participants and Family. Excludes ‘Refused’.

Q25b (P) / Q17b (F). Lately, have you done any of these things? Have six or more drinks of grog / alcohol at one time.

78

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 46: Behaviours done lately: Use an illegal or prescription drug for non-medical reasons. Base: Participants and Family. Excludes ‘Refused’.

Q25h (P) / Q17h (F). Lately, have you done any of these things? Use an illegal drug or a prescription medication for nonmedical reasons.

Figure 47: Behaviours done lately: Spend more than $50 a day on drugs not prescribed by a doctor. Base: Participants and Family. Excludes ‘Refused’.

Q25i (P) / Q17i (F). Lately, have you done any of these things? Spend more than $50 a day on drugs not prescribed by a doctor.

79

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 48: Behaviours done lately: Borrow money or sell things to buy alcohol/drugs. Base: Participants and Family. Excludes ‘Refused’.

Q25j (P) / Q17j (F). Lately, have you done any of these things? Borrow money or sell things to get money to buy alcohol / drugs. *Question not asked of Ceduna Family respondents.

Figure 49: Noticed a change in drinking of alcohol or grog in the community since the Trial started Base: Participants, Family and Non-Participants. Excludes ‘Refused’.

Q42a (P) / Q24a (F) / Q16a (NP). Since the Cashless Debit Card / Indue Card Trial started in your community have you noticed more, less or the same amount of: Drinking of alcohol or grog in the community? **Significantly different to zero at the 95% level *Significantly different to zero at the 90% level

80

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 50: Change in behaviour since Trial started: Drunk grog or alcohol (% of respondents) Base: Participants currently in Trial and Family. Excludes ‘Refused’ and ‘Not applicable - did not do activity before’.

Q44a (P) / Q26a (F). Since being on the Cashless Debit / Indue Card have you done each of the following more often, less often or the same as before: Drunk grog or alcohol? **Significantly different to zero at the 95% level *Significantly different to zero at the 90% level

Figure 51: Change in behaviour since Trial started: Had six or more drinks of grog or alcohol at one time (% of respondents) Base: Participants currently in Trial. Excludes ‘Refused’ and ‘Not applicable - did not do activity before’.

Q44b (P). Since being on the Cashless Debit / Indue Card have you done each of the following more often, less often or the same as before: Had six or more drinks of grog or alcohol at one time? **Significantly different to zero at the 95% level *Significantly different to zero at the 90% level

81

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 52: Change in behaviour since Trial started: Used an illegal drug (% of respondents) Base: Participants currently in Trial and Family. Excludes ‘Refused’ and ‘Not applicable - did not do activity before’.

Q44g (P) / Q26c. Since being on the Cashless Debit / Indue Card have you done each of the following more often, less often or the same as before: Used an illegal drug like benzos, ice, marijuana or speed? **Significantly different to zero at the 95% level *Significantly different to zero at the 90% level

Figure 53: Change in behaviour since Trial started: Spent more than $50 a day on illegal drugs (% of respondents) Base: Participants currently in Trial. Excludes ‘Refused’ and ‘Not applicable - did not do activity before’.

Q44h (P). Since being on the Cashless Debit / Indue Card have you done each of the following more often, less often or the same as before: Spent more than $50 a day on illegal drugs like benzos, ice, marijuana or speed? **Significantly different to zero at the 95% level *Significantly different to zero at the 90% level

82

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

STO PI#2: Frequency of use / volume of gambling and associated problems

Figure 54: STO PI #2

Performance Indicator Interim Rating

Frequency of use / volume of gambling and associated problems Achieved

Theme Details

Specification

 Number of times Trial participants engage in gambling activities per week  Number of days a week spend three or more hours gambling  Number of days a week spend more than $50 gambling  Percent of participants indicating that they gamble more

than they can afford to lose or borrow money or sell things to gamble  Percent of participants, family members and general community members reporting a decrease in gambling in the

community since commencement of Trial  Number of on-the-ground stakeholders reporting a decrease in gambling and associated problems in the community since

commencement of Trial #  EGM (‘poker machine’) revenue in Ceduna and Surrounds*

Target NA, except # target = ‘many’ and * target = lower than before Trial

Timeframe As self-reported at Wave 1

Data Sources

Survey of Trial participants

Survey of families

Survey of community members

Qualitative research with stakeholders

SA Poker Machine Revenue data

Survey Questions

Participant questionnaire: q25c-g, q42c, q44c-f

Family questionnaire: q17c-g, q24c, q26b

Non-Participant questionnaire: q16c

Definitions / Notes

On-the-ground stakeholders defined as members of the regional leadership groups and observers from government and non-government service providers based in the Trial areas

For stakeholders, qualitative indication of number: all, most, many, some, few

Administrative data on EGMs in SA showed a decline in EGM revenue in Ceduna and Surrounds from the previous year and all survey measures showed positive impacts on behaviours and perceptions. Consistent with these findings, stakeholders in Ceduna and Surrounds considered that

83

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

the CDCT has had a significant impact in relation to reducing the amount of time and money spent on EGMs. Stakeholders in East Kimberley generally felt that the CDCT had not had a significant impact on gambling, but were unsure about the impact on informal card playing (the main form of gambling).

Evidence

Ceduna - Qualitative research with stakeholders

Overall, stakeholders reported that CDCT had a significant impact on reducing the amount of money and time spent on formal gambling in Ceduna, particularly on Electronic Gaming Machines (EGMs).

The following positive impacts in relation to gambling were identified by the qualitative research:

 A significant positive impact on the CDCT in relation to gambling was noted in the reduction in use of EGMs, as evidenced by:

 Fewer people seen in EGM venues - which was reported by some stakeholders and was observed by ORIMA researchers during the fieldwork period; and

 A reported 15% reduction in gambling spend in the Eyre Peninsula region - as reported by the Gaming Authority. In addition, a couple of stakeholders reported that in Ceduna specifically, the reduction was 30%.

Many stakeholders reported there had been no change in the amount of informal card-playing and online gambling occurring in Ceduna. However, these stakeholders noted these forms of gambling were more difficult to report on as they usually occurred in private residences and were therefore less visible.

East Kimberley - Qualitative research with stakeholders

Overall, the CDCT was not felt to have had a significant impact on reducing the amount spent on gambling, although stakeholders found it difficult to comment on changes to card playing (the main form of gambling), as it was less visible.

The following positive impacts of the CDCT in relation to gambling were identified by the qualitative research:

 A reduction in the number of people playing bingo in Wyndham - one stakeholder reported that the venue had such a reduction that they had almost been unable to obtain their licence for their regular bingo night;

 A couple of stakeholders reported that people they personally knew found it “too difficult” / “frustrating” to play cards due to limited access to cash;

 A reduction in card games in public places - identified by a couple of stakeholders; and

 A reduction in the amount of money being spent in card games - one stakeholder reported that one of his organisation’s staff members was previously spending “all of her money”

84

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

gambling and since the CDCT had been spending more money on food as she was unable to afford to continue gambling.

Evidence from External data sources

Figure 55: Gambling - SA Attorney General - EGM (Poker Machine) Revenue (LGAs of Ceduna, Streaky Bay, Le Hunte, Elliston and Lower Eyre Peninsula)

85

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Quantitative research with Participants, Family and Non-Participants

Figure 56: Behaviours done lately: Gamble Base: Participants and Family. Excludes ‘Refused’.

Q25c (P) / Q17c (F). Lately, have you done any of these things? Gamble.

Figure 57: Behaviours done lately: Spend three or more hours a day gambling Base: Participants and Family. Excludes ‘Refused’.

Q25d (P) / Q17d (F). Lately, have you done any of these things? Spend three or more hours a day gambling.

86

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 58: Behaviours done lately: Spend more than $50 a day on gambling Base: Participants and Family. Excludes ‘Refused’.

Q25e (P) / Q17e (F). Lately, have you done any of these things? Spend more than $50 a day on gambling.

Figure 59: Behaviours done lately: Gamble more than you can afford to lose Base: Participants and Family. Excludes ‘Refused’.

Q25f (P) / Q17f (F). Lately, have you done any of these things? Gamble more than you can afford to lose.

87

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 60: Behaviours done lately: Borrow money or sell things to get money to gamble Base: Participants and Family. Excludes ‘Refused’.

Q25g (P) / Q17g (F). Lately, have you done any of these things? Borrow money or sell things to get money to gamble.

Figure 61: Noticed a change in gambling in the community since the Trial started Base: Participants, Family and Non-Participants. Excludes ‘Refused’.

Q42c (P) / Q24c (F) / Q16c. Since the Cashless Debit Card / Indue Card Trial started in your community have you noticed more, less or the same amount of: Gambling in the community? **Significantly different to zero at the 95% level *Significantly different to zero at the 90% level

88

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 62: Change in behaviour since Trial started: Gambled (% of respondents) Base: Participants currently in Trial and Family. Excludes ‘Refused’ and ‘Not applicable - did not do activity before’.

Q44c (P) / Q26b (F). Since being on the Cashless Debit / Indue Card have you done each of the following more often, less often or the same as before: Gambled? **Significantly different to zero at the 95% level *Significantly different to zero at the 90% level

Figure 63: Change in behaviour since Trial started: Spent more than $50 a day on gambling (% of respondents) Base: Participants currently in Trial. Excludes ‘Refused’ and ‘Not applicable - did not do activity before’.

Q44d (P). Since being on the Cashless Debit / Indue Card have you done each of the following more often, less often or the same as before: Spent more than $50 a day on gambling?**Significantly different to zero at the 95% level

89

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 64: Change in behaviour since Trial started: Bet more than you can really afford to lose (% of respondents) Base: Participants currently in Trial. Excludes ‘Refused’ and ‘Not applicable - did not do activity before’.

Q44e (P). Since being on the Cashless Debit / Indue Card have you done each of the following more often, less often or the same as before: Bet more than you can really afford to lose? **Significantly different to zero at the 95% level

Figure 65: Change in behaviour since Trial started: Had to borrow money or sell things to gamble (% of respondents) Base: Participants currently in Trial. Excludes ‘Refused’ and ‘Not applicable - did not do activity before’.

Q44f (P). Since being on the Cashless Debit / Indue Card have you done each of the following more often, less often or the same as before: Had to borrow money or sell things to get money to gamble? **Significantly different to zero at the 95% level

90

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

STO PI#3: Percent aware of drug & alcohol support services

Figure 66: STO PI #3

Performance Indicator Interim Rating

Percent aware of drug & alcohol support services Partial

Theme Details

Specification

Percent participants who are aware of drug and alcohol support services available in their community

Target NA

Timeframe As self-reported at Wave 1

Data Sources Survey of Trial participants

Survey Questions q32a

Definitions / Notes Services list will be provided by DSS / DHS.

Awareness of local drug and alcohol support services among CDCT participants ranged from 40% in Ceduna and Surrounds to 56% in East Kimberley. Awareness was higher among those who reported using illegal drugs or using prescription drugs for non-medical reasons (68%35 and 65%, respectively) and those who reported that they drank alcohol (48% and 57%).

35 This figure should be interpreted with caution due to the low base of respondents n=8

91

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Evidence

Quantitative research with Participants

Figure 67: Self-reported and back-coded awareness of drug and alcohol support services in local area before survey Base: Participants. Excludes ‘Refused’.

Q32a (P). Before this survey were you aware of any drug and alcohol support services in your local area? *Post survey, results were back-coded to represent those who said they were aware, and could accurately name a service. Those who could not were recoded as unaware.

40

59

2

CEDUNA PARTICIPANT ( N = 1 9 3 )

BACKCODED:

34% AWARE* 48 50

2

PARTICIPANT AVERAGE ( N = 5 4 8 )

BACKCODED:

40% AWARE* 56

41

3

EAST KIMBERLEY PARTICIPANT ( N = 3 5 5 )

YES, I WAS AWARE NO, I WASN’ T AWARE UNSURE

BACKCODED:

46% AWARE*

92

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

STO PI#4: Percent aware of financial & family support services

Figure 68: STO PI #4

Performance Indicator Interim Rating

Percent aware of financial & family support services Partial

Theme Details

Specification

Percent participants who are aware of financial and family support services available in their community

Target NA

Timeframe As self-reported at Wave 1

Data Sources Survey of Trial participants

Survey Questions q37a

Definitions / Notes Services list will be provided by DSS / DHS.

Awareness of local financial and family support services among CDCT participants was lower, with 33% of those in Ceduna and Surrounds and 37% of those in East Kimberley reporting awareness. Awareness of financial support services was similar among those who reported that they had experienced financial difficulties in the past 3 months (32% and 33% respectively).

93

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Evidence

Quantitative research with Participants

Figure 69: Self-reported and back-coded awareness of financial and family support services in local area before survey Base: Participants. Excludes ‘Refused’.

Q37a (P). Before this survey were you aware of any financial and family support services in your local area? *Post survey, results were back-coded to represent those who said they were aware, and could accurately name a service. Those who could not were recoded as unaware.

33

65

2

CEDUNA PARTICIPANT ( N = 1 9 3 )

35

63

2

PARTICIPANT AVERAGE ( N = 5 4 9 )

37

61

2

EAST KIMBERLEY PARTICIPANT ( N = 3 5 6 )

YES, I WAS AWARE NO, I WASN’ T AWARE UNSURE

BACKCODED:

33% AWARE*

BACKCODED:

29% AWARE*

BACKCODED:

31% AWARE*

94

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

STO PI#5: Usage of drug & alcohol support services

Figure 70: STO PI #5

Performance Indicator Interim Rating

Usage of drug & alcohol support services Partial

Theme Details

Specification

 Percent of participants who have ever used drug and alcohol support services  Number of times services used per participant  Intention to / likelihood of using service in future

 Number of people in community using services

Target

Above points 1-3: Higher at Wave 2 than at Wave 1 (statistically significant) Above point 4: Higher than before Trial

Timeframe

Above points 1-3: As self-reported at Wave 1 Above point 4: Trial period compared with 12 months prior to Trial launch

Data Sources

Above points 1-3: Survey of Trial participants

Above point 4: Department of Social Services (based on data from service providers and State Government agencies)

Survey Questions q33-34, q36

Definitions / Notes NA

On average across the two Trial sites, 15% of CDCT participants reported having ever used a drug or alcohol support service. Nearly half of these had done so within the last three months. Seven percent of Ceduna and Surrounds participants and 17% of East Kimberley participants indicated an intention to use these services in future.

Evidence

Evidence from external data sources

The Department agreed that due to unverified data, it was not advisable to include service usage statistics in this Wave 1 Interim Evaluation Report. Supporting evidence is outlined in STO PI#1.

95

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Quantitative research with Participants

Figure 71: Self-reported usage of local or other alcohol or drug support services Base: Participants. Excludes ‘Refused’.

Q33 (P). Have you ever used these local services or other services that help people to deal with problems related to alcohol or drug use?

Figure 72: Last time got help from an alcohol or drug support service Base: Participants who have ever used alcohol or drug services at q33 (self-reported).

Q34 (P). When was the last time that you got help from an alcohol or drug support service?

14

24

61

CEDUNA PARTICIPANT ( N = 1 9 3 )

15

32

53

PARTICIPANT AVERAGE ( N = 5 4 7 )

16

39

44

EAST KIMBERLEY PARTICIPANT ( N = 3 5 4 )

YES NO UNAWARE DON’ T KNOW / NOT SURE

96

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 73: Average number of times got help from an alcohol or drug support service Base: Participants who have used an alcohol or drug support service in the past 12 months. Excludes ‘Refused’.

Q35 (P). How many times did you get help from an alcohol or drug support service in the past year? *Note low n

Figure 74: Likelihood of trying to get help from an alcohol or drug support service in future Base: Participants. Excludes ‘Refused’.

Q36 (P). How likely is it that you will try and get help from an alcohol or drug support service in the future?

Ceduna Participant (n=13)*

16x

East Kimberley Participant (n=28)*

Participant Average (n=41)*

13x 15x

97

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

STO PI#6: Usage of financial & family support services

Figure 75: STO PI #6

Performance Indicator Interim Rating

Usage of financial & family support services Partial

Theme Details

Specification

 Percent of participants who have ever used financial or family support services  Number of times services used per participant  Intention to / likelihood of using service in future

 Number of people in community using services

Target

Above points 1-3: Higher at Wave 2 than at Wave 1 (statistically significant) Above point 4: Higher than before Trial

Timeframe

Above points 1-3: As self-reported at Wave 1 Above point 4: Trial period compared with 12 months prior to Trial launch

Data Sources

Above points 1-3: Survey of Trial participants

Above point 4: Department of Social Services (based on data from service providers and State Government agencies)

Survey Questions q38-39, q41

Definitions / Notes NA

In each Trial site, 17% of interviewed CDCT participants reported having ever used a financial or family support service. A greater proportion of these had done so in the last 12 months compared to for drug and alcohol services, but a similar proportion (about half) had done so in the last three months. Stated intention to use these services in future was slightly higher.

Evidence

Evidence from external data sources

The Department agreed that due to incomplete and unverified data, it was not advisable to include service usage statistics in this Wave 1 Interim Evaluation Report. Supporting evidence is outlined in STO PI#1.

98

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Quantitative research with Participants

Figure 76: Self-reported and back-coded usage of financial and family support services in local area Base: Participants. Excludes ‘Refused’.

Q38 (P). Have you ever used these local services or other services that help people deal with financial or family problems?

Figure 77: Last time got help from a financial or family support service Base: Participants who have ever used financial or family services at q38 (self-reported).

Q39 (P). When was the last time that you got help from a financial or family support service?

17

18

65

PARTICIPANT AVERAGE ( N = 5 4 8 )

17

17

66

CEDUNA PARTICIPANT ( N = 1 9 3 )

17

18

64

EAST KIMBERLEY PARTICIPANT ( N = 3 5 5 )

YES NO UNAWARE DON’ T KNOW / NOT SURE

99

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 78: Average number of times got help from financial or family support service Base: Participants who have used an alcohol or drug support service in the past 12 months. Excludes ‘Refused’.

Q40 (P). How many times did you get help from financial or family support service in the past year? *Note low n

Figure 79: Likelihood of trying to get help from a financial or family support service in future Base: Participants aware of when they last used financial or family services at q39. Excludes ‘Refused’.

Q41 (P). How likely is it that you will try and get help from a financial or family support service in the future?

Ceduna Participant (n=20)*

6x

East Kimberley Participant (n=39)*

Participant Average (n=59)

7x 7x

100

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

3. Medium-Term Outcome Performance Indicators

MTO PI#1: Frequency of use / volume consumed of drugs and alcohol

Figure 80: MTO PI #1

Performance Indicator Interim Rating

Frequency of use / volume consumed of drugs and alcohol Not relevant at interim point

Theme Details

Specification

 Number of times alcohol consumed by participants per week  Percent of participants who say they have used non-prescription drugs in the last week  Number of times per week spend more than $50 a day on

drugs not prescribed by a doctor  Number of times per week have six or more drinks of alcohol at one time (binge drinking)  Percent of participants, family members and general

community members reporting a decrease in drinking of alcohol in the community since commencement of Trial  Number of on-the-ground stakeholders reporting a decrease in drinking of alcohol in the community since

commencement of Trial

Target Frequency / volume not higher at Wave 2 than at Wave 1

Timeframe As self-reported at Wave 2

Data Sources

Survey of Trial participants

Survey of families

Survey of community members

Qualitative research with stakeholders

Survey Questions

Participant questionnaire: q25a-b/h-j, q42a

Family questionnaire: q17a-b/h-j, q24a

Non-Participant questionnaire: q16a

Definitions / Notes NA

Evidence

The quantitative measures from STO#1: Frequency of use / volume consumed of drugs and alcohol will be reported here at Wave 2.

101

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

MTO PI#2: Frequency of use / volume of gambling and associated problems

Figure 81: MTO PI #2

Performance Indicator Interim Rating

Frequency of use / volume of gambling and associated problems Not relevant at interim point

Theme Details

Specification

 Number of times Trial participants engage in gambling activities per week  Number of days a week spend three or more hours gambling  Number of days a week spend more than $50 gambling  Percent of participants indicating that they gamble more

than they can afford to lose or borrow money or sell things to gamble  Percent of participants, family members and general community members reporting a decrease in gambling in the

community since commencement of Trial  Number of on-the-ground stakeholders reporting a decrease in gambling and associated problems in the community since

commencement of Trial  EGM (‘poker machine’) revenue in Ceduna and Surrounds

Target Frequency / volume not higher at Wave 2 than at Wave 1

Timeframe As self-reported at Wave 2

Data Sources

Survey of Trial participants

Survey of families

Survey of community members

Qualitative research with stakeholders

SA Poker Machine Revenue data

Survey Questions

Participant questionnaire: q25c-g, q42c, q44c-f

Family questionnaire: q17c-g, q24c, q26b

Non-Participant questionnaire: q16c

Definitions / Notes NA

Evidence

The quantitative measures from STO#2: Frequency of use / volume of gambling and associated problems will be reported here at Wave 2.

SA Poker Machine Revenue data supporting this performance indicator is outlined earlier, in STO PI#2.

102

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

MTO PI#3: Incidence of violent & other types of crime and violent behaviour

Figure 82: MTO PI #3

Performance Indicator Interim Rating

Incidence of violent & other types of crime and violent behaviour Not relevant at interim point

Theme Details

Specification

 Police reports of assault and burglary offences; drink driving/drug driving; domestic violence incidence reports; drunk and disorderly conduct; outstanding driving and vehicle fines.

 Percent of participants, family members and the general community who report being the victim of crime in the past month

 Percent of participants, family members and the general community who report a decrease in violence in the community since commencement of Trial

 Number of on-the-ground stakeholders reporting a decrease in violence in the community since commencement of Trial

Target Lower than before Trial

Timeframe

Above point 1: Trial period compared with 12 months prior to Trial launch

Above points 2-4: As self-reported at Wave 1 and Wave 2

Data Sources

Above point 1: SA and WA Police

Above points 2-4: Surveys of Trial participants, families and community members; and

Qualitative research with stakeholders

Survey Questions

Participant questionnaire: q29b,d-e, q42b

Family questionnaire: q21b,d-e, q24b

Non-Participant questionnaire: q13a, q13c-d, q16b

Definitions / Notes

On-the-ground stakeholders defined as members of the regional leadership groups and observers from government and non-government service providers based in the Trial areas

For stakeholders, qualitative indication of number: all, most, many, some, few

As a medium-term outcome this performance indicator is not evaluated at this interim stage. Evidence related to this topic suggests there are few early signs of movement.

103

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Evidence

Ceduna - Qualitative research with stakeholders

At the Initial Conditions reporting stage, many stakeholder participants had expected that there would be a significant increase in petty crime and domestic violence at the start of the Trial. In Wave 1, these participants reported that this had not eventuated, and as such were “relieved”.

Overall, there was no significant positive change noted in relation to crime, violence and harm. However, a few stakeholders identified some increase in domestic violence / intervention orders - although it was not clear whether this was due to changes in reporting requirements or increased community awareness, understanding and willingness to take action, or to the CDCT.

A couple of store owners reported that there had been an increase in thefts from their stores, including:

 Shoplifting, including by intoxicated adults; and / or

 Children stealing items (e.g. lollies, hair dye, etc.).

In addition, a few stakeholders reported an increase in the following illegal / harmful behaviours, however it was not clear whether such behaviours were related / attributable to the CDCT:

 A few stakeholders reported they had heard about an increase in break-ins in Ceduna where laptops, jewellery and cash had been stolen;

 However, Police reported that they had not seen an increase in crime committed by CDCT participants on the ground and that the recent incidents that they had dealt with had been committed by juveniles and people from outside the Trial area.

 A few stakeholders from remote communities reported hearing about break-ins in their community (i.e. outside Ceduna) but thought that these crimes were committed by people from out of town (i.e. not local) visiting for funerals and other events;

 A couple of stakeholders reported that arguments were increasing between community members. Specifically, community members who received less money through payments such as Newstart were reported to be asking others who received larger payments (e.g. Disability Support Pension or Parenting Payments) for cash; and

 One service provider stakeholder reported that there had been five cases of prostitution among their clients from remote communities outside of Ceduna. However, this occurrence was not mentioned by other stakeholders.

104

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

East Kimberley - Qualitative research with stakeholders

At the Initial Conditions reporting stage, many stakeholder participants had expected that there would be a significant increase in petty crime amongst CDCT participants at the start of the Trial. While an increase of such crime was not evident among CDCT participants, it was commonly reported by stakeholders to be occurring among children as a result of reduced access to cash.

Overall, there was no significant positive change noted in relation to crime, violence and harm among CDCT participants. However, a couple of positive changes were noticed / reported by stakeholders, including:

 A decrease in vandalism of ATMs - noticed by a few stakeholders;

 A decrease in crime (including alcohol related incidents) in Wyndham - although it was reported that there were many other contributing factors (e.g. TAMS) and it was felt that this could not be directly attributed to CDCT; and

 A reduction in the number of injuries indicative of domestic violence presenting at the hospital reported by the Accident and Emergency department.

 However, a few stakeholders identified some increase in domestic violence / intervention orders - although it was not clear whether the increase was due to changes in reporting requirements, the policing approach or increased community awareness, understanding and willingness to take action.

In addition, many stakeholders reported an increase in the following illegal / harmful behaviours among young people / children:

 Robberies / thefts from cars / vehicles and dwellings; and

 Petty crime (e.g. pickpocketing and “snatch and grab”).

105

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Evidence from external data sources

Figure 83: Crime statistics - South Australian Police - Ceduna LGA

106

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 84: Criminal behaviour data - Western Australian Police - Kununurra

107

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 85: Criminal behaviour data - Western Australian Police - Wyndham

108

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Quantitative research with Participants, Families and Non-Participant stakeholders

Figure 86: Beaten up, injured or assaulted in the past month Base: Participants, Family and Non-Participants. Excludes ‘Refused’.

Q29b (P) / Q21b (F) / Q13a (NP). In the past month have you been: Beaten up, injured, or assaulted?

Figure 87: Robbed in the past month Base: Participants, Family and Non-Participants. Excludes ‘Refused’.

Q29d (P) / Q21d (F) / Q13c (NP). In the past month have you been: Robbed?

109

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 88: Threatened or attacked with a gun, knife or other weapon in the past month Base: Participants, Family and Non-Participants. Excludes ‘Refused’.

Q29e (P) / Q21e (F) / Q13d (NP). In the past month have you been: Threatened or attacked with a gun, knife or other weapon?

Figure 89: Violence noticed in the community since the Trial started Base: Participants, Family and Non-Participants. Excludes ‘Refused’.

Q42b (P) / Q24b (F) / Q16b (NP). Since the Cashless Debit Card / Indue Card Trial started in your community have you noticed more, less or the same amount of: Violence in the community? **Significantly different to zero at the 95% level *Significantly different to zero at the 90% level

110

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

MTO PI#4: Drug/alcohol-related injuries and hospital admissions

Figure 90: MTO PI #4

Performance Indicator Interim Rating

Drug/alcohol-related injuries and hospital admissions Not relevant at interim point

Theme Details

Specification

 Drug / alcohol-related hospital admissions / emergency presentations / sobering up service admissions

 Percent of participants / family members who say they have been injured after drinking alcohol / taking drugs in the last month

Target Lower than before Trial

Timeframe

Above point 1: Trial period compared with 12 months prior to Trial launch

Above point 2: As self-reported at Wave 1 and Wave 2

Data Sources

Above point 1: Department of Premier and Cabinet SA, WA Health, Department of Social Services (based on data provided by local sobering up services)

Above point 2: Surveys of Trial participants and families

Survey Questions

Participant questionnaire: q29h

Family questionnaire: q21h

Definitions / Notes NA

As a medium-term outcome this performance indicator is not evaluated at this interim stage. Early administrative data relating to sobering up services in the Trial communities are suggestive of positive early signs, but this will be more meaningfully examined at Wave 2.

Evidence

Evidence from external data sources

111

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 91: Problematic Alcohol consumption / intoxication data - Ceduna - Department for Communities and Social Inclusion and Drug and Alcohol Services

112

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 92: Sobering Up Unit (SUU) - Ceduna - Department for Communities and Social Inclusion

113

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 93: Problematic alcohol consumption / intoxication data Kununurra Waringarri Aboriginal Corporation Patrol Service - Ngnowar-Aerwah Aboriginal Corporation (Wyndham)

114

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Quantitative research with Participants, Families and Non-Participant stakeholders

Figure 94: Injured or had an accident after drinking alcohol/ grog or taking drugs in the past month Base: Participants and Family. Excludes ‘Refused’.

Q29h (P) / Q21h (F). In the last month have you been: Injured or had an accident after drinking alcohol or taking drugs?

115

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

MTO PI#5: Percent reporting feeling safe in the community

Figure 95: MTO PI #5

Performance Indicator Interim Rating

Percent reporting feeling safe in the community Not relevant at interim point

Theme Details

Specification

 Percent of participants, family members and other community members who report feeling safe in their community

Target NA

Timeframe As self-reported at Wave 1 and Wave 2

Data Sources

Surveys of Trial participants, families and community members

SA and WA Police crime data

Survey Questions

Participant questionnaire: q31a-b

Family questionnaire: q23a-b

Non-Participant questionnaire: q15a-b

Definitions / Notes NA

As a medium-term outcome this performance indicator is not evaluated at this interim stage. At Wave 1 perceptions of safety on the streets of the Trial communities were high during the day, but considerably lower at night time - especially in East Kimberley.

Evidence

Evidence from external data sources

SA and WA Police crime data supporting this performance indicator is outlined earlier, in MTO PI#3.

116

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Quantitative research with Participants, Families and Non-Participant stakeholders

Figure 96: Reports of feeling either very safe or safe on the streets during the day or at night Base: Participants, Family and Non-Participants. Excludes ‘Refused’.

Q31a-b (P) / Q23a-b (F) / q15a-b (NP). Do you feel safe or unsafe on the streets of your community during the day / at night?

117

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

MTO PI#6: Percent reporting feeling safe at home

Figure 97: MTO PI #6

Performance Indicator Interim Rating

Percent reporting feeling safe at home Not relevant at interim point

Theme Details

Specification  Percent of participants, family members and other

community members who report feeling safe at home

Target NA

Timeframe As self-reported at Wave 1 and Wave 2

Data Sources

Surveys of Trial participants, families and community members

SA and WA Police crime data

Survey Questions

Participant questionnaire: q31c

Family questionnaire: q23c

Non-Participant questionnaire: q15c

Definitions / Notes NA

As a medium-term outcome this performance indicator is not evaluated at this interim stage. At Wave 1 the sense of safety in the home was high.

Evidence

Evidence from external data sources

SA and WA Police crime data supporting this performance indicator is outlined earlier, in MTO PI#3.

118

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Quantitative research with Participants, Families and Non-Participant stakeholders

Figure 98: Reports of feeling either very safe or safe at home Base: Participants, Family and Non-Participants. Excludes ‘Refused’.

Q31c (P) / Q23c (F) / q15c (NP). Do you feel safe or unsafe at home?

119

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Appendix B: Non-Performance Indicator Results

Questionnaire Section A: Demographics (Unweighted)

Ninety-nine percent of all participants said that they currently have an Indue / Cashless Debit card in their name, whilst the remaining 1% had used one in the past. No family or non-participant respondents reported that they had ever had this card in their name.

All of the respondents in the family group reported that someone in their immediate family (that they live with) has the card. No respondents from the non-participant group claimed to live with anyone who has the card.

Figure 99: Age

Base: Participants, Family and Non-Participants.

Q1/1a (P) / Q1/1a (F) / Q1/1a (NP). How old are you? Unweighted

NOTE: The systematic intercept methodology used for the surveys produced a sample that was broadly in line with CDCT participant ages in both locations. In both locations the 18-34 age group was slightly under-represented in the raw sample (by -5% in Ceduna and -7% in EK), which is a typical characteristic of most survey samples in any context due to younger people generally being less

willing to participate. The 55+ age groups were within +3% (Ceduna) and -2% (EK) of the participant population. The raw participant samples were weighted to CDCT participant population benchmarks on age, gender and Indigenous / non-Indigenous proportions for analysis, correcting these slight age variations for the statistical analysis.

120

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 100: Gender

Base: Participants, Family and Non-Participants.

Q4 (P) / Q4 (F) / Q4 (NP). Gender. Unweighted

Figure 101: Relationship to family member on card Base: Family.

Q6 (F). What is your relationship to them? Unweighted

121

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 102: Born in Australia (% yes) Base: Participants, Family and Non-Participant.

Q5 (P) / Q7 (F) / Q6 (NP). Were you..? Unweighted

Figure 103: Aboriginal or Torres Strait Islander origin (% yes) Base: Participants, Family and Non-Participants.

Q6 (P) / Q8 (F) / Q7 (NP). Are you of Aboriginal or Torres Strait Islander origin? Unweighted

122

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 104: Which of the following best describes your origin? Base: Participants, Family and Non-Participants of Aboriginal and/or Torres Strait Islander origin.

Q6a (P) / Q9 (F) / Q8 (NP). Which of the following best describes your origin? Unweighted

Figure 105: Location of Ceduna respondents Base: Ceduna Participants, Family and Non-Participants.

Q7 (P) / Q10 (F) / Q9 (NP). What town, suburb or community do you usually live in? Unweighted

123

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 106: Location of East Kimberley respondents Base: East Kimberley Participants, Family and Non-Participants.

Q7 (P) / Q10 (F) / Q9 (NP). What town, suburb or community do you usually live in? Unweighted

Figure 107: Do you care for a child who is less than 18 years old (% yes) Base: Participants and Family.

Q9 (P) / Q12 (F). Do you care for, or look after, a child who is less than 18 years old? Unweighted

124

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Questionnaire Section B: Profile of Debit Card Trial Participation

Figure 108: Type of card Base: Participants.

Q11 (P). What type of Cashless Debit Card Trial are you currently on?

Figure 109: Experienced problems with the card Base: Participants.

Q14 (P). Have you had any problems using the Cashless Debit / Indue Card?

125

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 110: Asked Community Panel for a review Base: Ceduna Participants.

Q17a (P). Have you asked the Community Panel to review how much of your Centrelink money goes onto the Cashless Debit / Indue Card?

Figure 111: Community Review resulted in a change Base: Ceduna Participants.

Q17b (P). Did the amount or per cent of your Centrelink money that goes onto the Cashless Debit / Indue Card change after the Community Panel reviewed you?

126

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 112: Problems with the Community Review panel or process Base: Ceduna Participants.

Q17c (P). Did you have any problems with the Community Panel or the process?

Figure 113: Live with anyone else who is in the Trial Base: Participants.

Q18 (P). Do you live with anyone else who is in the Cashless Debit Card / Indue Card Trial or who has a Cashless Debit Card / Indue Card?

127

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 114: Other people you live with who are on the card Base: Participants.

*Includes Husband, Wife, Boyfriend, Girlfriend, Defacto Male Partner, Defacto Female Partner. #Child (and Defacto Male / Female partner) not asked of Ceduna Participants, but was back-coded where possible. Q19 (P). What is your relationship to them? Would they be your..

Figure 115: Knowledge that all people receiving Centrelink payments in the area have a big part of their payments put on the card Base: Family and Non-Participants.

Q15a (F) / q12iia (NP). Before this survey did you know that: All people receiving Centrelink payments who live in this area apart from aged pensioners have a big part of their payments put onto this card?

128

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 116: Knowledge that wage earners and other pensioners can choose to get a card Base: Family and Non-Participants.

Q15b (F) / q12iib (NP). Before this survey did you know that: Wage earners, aged pensioners and veterans pensioners who live in this area can choose to get one of these cards?

129

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Questionnaire Section C: Profile of Current Behaviour and Attitudes

Financial

Figure 117: Run out of money to buy food in the last 3 months Base: Participants and Family. Excludes ‘Refused’.

Q24i (P) / Q16i (F). In the last 3 months, how often, if at all, did you: Run out of money to buy food?

130

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 118: Not have money to pay rent / mortgage in the last 3 months Base: Participants and Family. Excludes ‘Refused’.

Q24ii (P) / Q16ii (F). In the last 3 months, how often, if at all, did you: Not have money to pay rent or your mortgage on time?

Figure 119: Not have money to pay another type of bill in the last 3 months Base: Participants and Family. Excludes ‘Refused’.

Q24iii (P) / Q16iii (F). In the last 3 months, how often, if at all, did you: Not have money to pay some other type of bill when it was due?

131

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 120: Run out of money to pay for things for your children’s schooling in the last 3 months Base: Participants and Family. Excludes ‘Refused’.

Q24iv (P) / Q16iv (F). In the last 3 months, how often, if at all, did you: Run out of money to pay for things that your child / children needed for school, like books?

Figure 121: Run out of money to pay for essential non-food items for your children Base: Participants and Family. Excludes ‘Refused’.

Q24v (P) / Q16v (F). In the last 3 months, how often, if at all, did you: Run out of money to pay for essential (non-food) items for your children, such as nappies, clothes and medicine?

132

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 122: Borrow money from family or friends in the last 3 months Base: Participants and Family. Excludes ‘Refused’.

Q24vi (P) / Q16vi (F). In the last 3 months, how often, if at all, did you: Borrow money from family or friends?

Figure 123: Run out of money because you had given money to friends or family in the last 3 months Base: Participants and Family. Excludes ‘Refused’.

Q24vii (P) / Q16vii (F). In the last 3 months, how often, if at all, did you: Run out of money because you had given money to friends or family?

133

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 124: Currently looking for a job Base: Participants and Family.

Q26 (P) / Q18 (F). Are you currently looking for a job or paid work?

Figure 125: Number of hours usually spend on trying to get a job Base: Participants and Family. Excludes ‘Refused’.

Q27 (P) / Q19 (F). Usually, how many hours a week would you spend on trying to get a job or paid work?

134

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Family

Figure 126: Children go to school

Base: Participants and Family who care for a child / children. Excludes ‘Refused’.

Q28 (P) / Q20 (F). Do any of the children you care for go to school?

Figure 127: Check on homework or help out with other things to do with school Base: Participants and Family. Excludes ‘Refused’.

Q28a (P) / Q20a (F). Usually, do you check to make sure that the children are doing their homework or help out with other things to do with school?

135

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Recent experiences

Figure 128: Arrested by the Police in the past month Base: Participants and Family. Excludes ‘Refused’.

Q29a (P) / Q21a (F). In the past month have you been: Arrested by Police?

Figure 129: Harassed in the past month Base: Participants, Family and Non-Participants. Excludes ‘Refused’.

Q29c (P) / Q21c (F) / Q13b (NP). In the past month have you been: Harassed?

136

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 130: Been homeless or had to sleep rough in the past month Base: Participants and Family. Excludes ‘Refused’.

Q29f (P) / Q21f (F). In the past month have you been: Homeless or had to sleep rough?

Figure 131: Humbugged or pressured by family or friends to give them money in the past month Base: Participants, Family and Non-Participants. Excludes ‘Refused’.

Q29g (P) / Q21g (F) / Q13e (NP). In the past month have you been: Humbugged or pressured by family or friends to give them money?

137

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Community Pride

Figure 132: Participants’ feelings toward the community in which you live Base: Participants.

Q30 (P). Do you feel proud or ashamed of the community in which you live?

Figure 133: Families’ feelings toward the community in which you live Base: Family.

Q22 (F). Do you feel proud or ashamed of the community in which you live?

138

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 134: Non-Participants feelings toward the community in which you live Base: Non-Participants.

Q14 (NP). Do you feel proud or ashamed of the community in which you live?

139

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Questionnaire Section D: Opinions of the impact of the Debit Card Trial

Figure 135: Noticed more humbugging or harassment for money since the Trial started Base: Participants, Family and Non-Participants. Excludes ‘Refused’.

Q42d (P) / Q24d (F) / Q16d. Since the Cashless Debit / Indue Card started in your community, have you noticed more, less or the same amount of: Humbugging or harassment for money? **Significantly different to zero at the 95% level *Significantly different to zero at the 90% level

140

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 136: Been able to save more money than before since Trial started Base: Participants currently in Trial and Family. Excludes ‘Refused’.

Q43a (P) / Q25a (F). Since being on the Cashless Debit / Indue Card have these happened to you: You’ve / the family has been able to save more money than before?

Figure 137: Been better able to care for your child/ren since Trial started Base: Participants currently in Trial and Family who care for a child / children. Excludes ‘Refused’.

Q43b (P) / Q25b (F). Since being on the Cashless Debit / Indue Card have these happened to you: You’ve / the family has been better able to care for your child/ren?

141

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 138: Got more involved in your children’s homework and school since Trial started Base: Participants currently in Trial and Family who care for a child / children. Excludes ‘Refused’.

Q43c (P) / Q25c (F). Since being on the Cashless Debit / Indue Card have these happened to you: You’ve / the family has got more involved in your children’s homework and school?

Figure 139: Got better at things like using a computer, the internet or a smartphone since Trial started Base: Participants currently in Trial. Excludes ‘Refused’.

Q43d (P). Since being on the Cashless Debit / Indue Card have these happened to you: I’ve got better at things like using a computer, the internet or a smartphone?

142

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 140: Impact of the Trial on your life / your family’s life Base: Participants currently in Trial and Family. Excludes ‘Refused’.

Q45 (P) / Q27 (F). Would you say the Cashless Debit / Indue Card has made your life / your family’s life..?.

Figure 141: Impact of the Trial on your child’s life / children’s lives Base: Participants currently in Trial. Excludes ‘Refused’.

Q47 (P). Would you say, the Cashless Debit / Indue Card has made your child’s life / children’s lives..?

143

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Figure 142: Impact of the Trial on your community Base: Family and Non-Participants. Excludes ‘Refused’.

Q29 (F) / Q17 (NP). Would you say, the Cashless Debit / Indue Card has made life in your community..?

Figure 143: Recommend card to others Base: Participants currently in Trial. Excludes ‘Refused’.

Q48 (P). Have you told anyone who doesn’t have a Cashless Debit / Indue Card to get one, or do you plan to?

144

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Further Analysis: Key questions by key demographics

Table 10: Key questions split by age

Age group

Since being on the CDCT 18-24 25-34 35-44 45-54 55-64

Percent “yes”: n=37-80 n=79-127 n=72-121 n=44-135 n=19-79

You’ve been able to save more money than before [FIN] 41% 32% 28% 33% 23%

You’ve been better able to care for your child/ren 29% 29% 37% 30% 28%

You’ve got more involved in your children’s homework and school 3% 11% 26% 23% 16%

I’ve got better at things like using a computer, the internet or a smartphone 24% 30% 23% 15% 12%

Net positive change: n=80 n=130 n=121-122 n=136 n=79

Drunk grog or alcohol [D&A] 18% 13% 19% 14% 10%

Had six or more drinks of grog or alcohol at one time [D&A] 15% 7% 12% 15% 11%

Gambled [FIN] 10% 8% 8% 8% 5%

Spent more than $50 a day on gambling [FIN] 8% 3% 6% 3% 4%

Bet more than you can really afford to lose [FIN] 5% -1% 2% 3% 4%

Had to borrow money or sell things to get money to gamble [FIN] 2% 1% 2% 0% 1%

Used an illegal drug like benzos, ice, marijuana, or speed [D&A] 7% 1% 6% 2% 4%

Spent more than $50 a day on illegal drugs like benzos, ice, marijuana, or speed [D&A] 6% 1% 0% 1% 0%

Net positive change: n=80 n=130 n=122 n=136 n=79

Would you say the Cashless Debit Card / Indue Card has made your life better -27% -33% -12% -16% -45%

[FIN]: Financial statement, [D&A]: Drug and Alcohol statement.

145

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Table 11: Key questions split by gender

Gender

Since being on the CDCT Female Male

Percent “yes”: n=200-341 n=51-201

You’ve been able to save more money than before [FIN] 33% 29%

You’ve been better able to care for your child/ren 30% 36%

You’ve got more involved in your children’s homework and school 13% 23%

I’ve got better at things like using a computer, the internet or a smartphone 23% 20%

Net positive change: n=345-346 n=201

Drunk grog or alcohol [D&A] 13% 17%

Had six or more drinks of grog or alcohol at one time [D&A] 10% 14%

Gambled [FIN] 10% 5%

Spent more than $50 a day on gambling [FIN] 4% 5%

Bet more than you can really afford to lose [FIN] 2% 2%

Had to borrow money or sell things to get money to gamble [FIN] 1% 1%

Used an illegal drug like benzos, ice, marijuana, or speed [D&A]

3% 5%

Spent more than $50 a day on illegal drugs like benzos, ice, marijuana, or speed [D&A] 1% 1%

Net positive change: n=346 n=201

Would you say the Cashless Debit Card / Indue Card has made your life better -25% -28%

[FIN]: Financial statement, [D&A]: Drug and Alcohol statement.

146

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Table 12: Key questions split by those who care for a child/children and those that don’t

Care for child

Since being on the CDCT Yes No

Percent “yes”: n=233-251 n=295

You’ve been able to save more money than before [FIN] 27% 35%

You’ve been better able to care for your child/ren 31% -

You’ve got more involved in your children’s homework and school 16% -

I’ve got better at things like using a computer, the internet or a smartphone 22% 21%

Net positive change: n=250-251 n=296

Drunk grog or alcohol [D&A] 14% 16%

Had six or more drinks of grog or alcohol at one time [D&A] 11% 13%

Gambled [FIN] 8% 7%

Spent more than $50 a day on gambling [FIN] 2% 6%

Bet more than you can really afford to lose [FIN] 2% 3%

Had to borrow money or sell things to get money to gamble [FIN] 0% 3%

Used an illegal drug like benzos, ice, marijuana, or speed [D&A]

5% 2%

Spent more than $50 a day on illegal drugs like benzos, ice, marijuana, or speed [D&A] 1% 2%

Net positive change: n=251 n=296

Would you say the Cashless Debit Card / Indue Card has made your life better -27% -26%

[FIN]: Financial statement, [D&A]: Drug and Alcohol statement.

147

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Appendix C: Qualitative Summary Reports

Ceduna and Surrounds

Methodology

 In-depth interviews and focus groups - n=33

 Dates: Monday 15 August - Thursday 15 September

 Locations: Ceduna, Koonibba, Scotdesco and Yalata

Overall impact of Trial

Overall, most participants did not believe that there had been significant change / impact of the CDCT, however upon further probing there was evidence of some change starting to emerge. Most of these participants felt it was “too early” in the Trial to determine any considerable change, and were expecting to only see changes in the longer term (i.e. another 6 months or so).

Some participants indicated that there were other external factors which may have impacted on data / behaviour over the Trial period, beyond the impact of CDCT itself, such as:

 Weather - it was noted that the seasonal effect of a cold and wet winter could be contributing to less visible public activities (e.g. public drinking and gambling);  Lump sum payments - it was reported that there had been an “influx of cash” into the community via tax payments, superannuation payments (a $200,000+ payment was

reported) and inheritances, and this has potentially contributed to abnormal expenditure patterns (e.g. alcohol and large purchases); and  Alcohol restrictions - some stakeholders felt that the introduction of alcohol restrictions in 2015 had already caused a noticeable change in the community and that this was the cause

for the visible reduction in alcohol consumption rather than the CDCT itself (e.g. a health worker participant reported a substantial drop of intoxicated presentations since the changes to liquor laws, but felt that it was not as considerable since the CDCT).

148

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Alcohol consumption

Many stakeholders were reluctant to indicate that the CDCT had reduced alcohol consumption in Ceduna and surrounding areas as they felt that there were no direct, visible changes to alcohol consumption since the start of the Trial. However, some had noted broader changes that could suggest an indirect impact on reducing alcohol consumption (e.g. more money directed to other purposes and greater engagement with programs).

The few stakeholders who had noted direct positive impacts of CDCT in relation to alcohol consumption reported that these included:

 Less visibility / fewer people consuming alcohol in public. This was evidenced by:  A few stakeholders seeing and hearing fewer intoxicated people on the streets;  Fewer ambulance callouts to public places. While call-out figures were unchanged, there were more call-outs to private addresses since the Trial had begun;  Fewer people drinking in front and back gardens in one remote community (Koonibba);

 More sobriety - evidenced by:  A substantial increase (around 50%) in the number of people attending the lunch service at the ADAC day centre (previously most clients only attended breakfast);  More people attending services at CAC while sober;  Seeing specific individuals sober - a few stakeholders reported seeing certain

individuals in a sober state “for the first time”; and  Reduced emergency presentations related to alcohol consumption during the end of the financial year period - there tended to be an influx in hospital presentations associated with excessive alcohol consumption as lump sum payments received at this time of year (e.g.

school kids bonus) were spent on alcohol, but this year no spike had been noticed by hospital staff. The negative impacts of alcohol consumption on health and wellbeing was still a concern and the recent spate of deaths in the community were cited as examples of this. Two out of three deaths in Koonibba in the weeks leading up to the evaluation fieldwork were reported as being related to alcohol. However, these were believed to be long-term health impacts (e.g. heart disease) that the CDCT was unlikely to influence and impact in the short period.

Drug use

Overall, consistent with the Initial Conditions Report, stakeholders found it more difficult to comment on levels of drug use in the community as substance usage was less visible. However, concern about methamphetamines (“ice”) had significantly heightened among the stakeholders compared to when they were interviewed at the Initial Conditions stage of the research (i.e. in April 2016). Most stakeholders consistently reported ice as being more prominent and easily available at Wave 1 compared to the Initial Conditions stage.

149

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

There were minimal positive impacts of the CDCT on drug use identified by stakeholders other than a report by one stakeholder who had spoken with a drug dealer (who was on the CDCT) who had not been able to buy wholesale drugs because he no longer had the supply of cash. Another stakeholder reported hearing reports that it was harder to access marijuana in some of the remote Trial communities outside of Ceduna, however, as there had been some recent arrests of drug suppliers it was unclear if this was as a result of these arrests or the CDCT.

Many stakeholders were concerned about ice becoming increasingly available and used (including a few incidents in remote communities, e.g. Yalata). Stakeholders were particularly concerned that it may proliferate very rapidly in the community as it was reported to be very low cost (e.g. $10) so access to cash was not a barrier to use. One stakeholder noted that they had received increased requests from health services in the Ceduna Trial area to deliver staff training sessions in relation to methamphetamines. However, most stakeholders felt that the growth in this problem was consistent with other areas (i.e. it wasn’t confined to Ceduna or related to the CDCT). From a policing perspective it was not reported to be an issue of major concern, and most incidents dealt with related to ice usage did not occur amongst CDCT participants.

Some stakeholders reported that marijuana was being laced with ice to get people “hooked”. This was evidenced by a stakeholder from a health service who had new presentations of people behaving in a way that indicated ice use, but who reported only taking marijuana.

In terms of the CDCT, ambulance presentations demonstrated that there hasn’t been a significant change in marijuana and other drug use. A few stakeholders reported that abuse of prescription drugs was still an issue.

Gambling

Overall, stakeholders reported that CDCT had a significant impact on reducing the amount of money and time spent on legalised gambling in Ceduna, particularly on Electronic Gaming Machines (EGMs).

The following positive impacts in relation to gambling were identified by the qualitative research:

 A significant positive impact on the CDCT in relation to gambling was noted in the reduction in use of EGMs, as evidenced by:  Fewer people seen in EGM venues - which was reported by some stakeholders and was observed by ORIMA researchers during the fieldwork period; and

 A reported 15% reduction in gambling spend in the Eyre Peninsula region - as reported by the Gaming Authority. In addition, a couple of stakeholders reported that in Ceduna specifically, the reduction was 30%. Many stakeholders reported there had been no change in the amount of illegal card-playing and online gambling occurring in Ceduna. However, these stakeholders noted these forms of gambling were more difficult to report on as they usually occurred in private residences and were therefore less visible.

150

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Crime, violence and harm

At the Initial Conditions reporting stage, many stakeholder participants had expected that there would be a significant increase in petty crime and domestic violence at the start of the Trial. In Wave 1, these participants reported that this had not eventuated, and as such were “relieved”.

Overall, there was no significant positive change noted in relation to crime, violence and harm. However, a few stakeholders identified some increase in domestic violence / intervention orders - although it was not clear whether this was due to changes in reporting requirements or increased community awareness, understanding and willingness to take action, or to the CDCT. One stakeholder reported that recent deaths related to domestic violence had increased the community’s willingness to report instances of domestic violence and had led to an increase in the number of calls the police received relating to domestic violence. Likewise, the Police and ambulance also had more rigorous domestic violence reporting requirements in recent times.

A couple of store owners reported that there had been an increase in thefts from their stores, including:

 Shoplifting, including by intoxicated adults; and / or  Children stealing items (e.g. lollies, hair dye, etc.) from the local store in Yalata since the Trial commenced as they no longer had access to cash from their parents. However, the store had taken successful steps to reduce this (e.g. reporting incidents to the school) and such

measures had addressed the issue. In addition, a few stakeholders reported an increase in the following illegal / harmful behaviours, however it was not clear whether such behaviours were related / attributable to the CDCT:

 A few stakeholders reported they had heard about an increase in break-ins in Ceduna where laptops, jewellery and cash had been stolen;  However, Police reported that they had not seen an increase in crime committed by CDCT participants on the ground and that the recent incidents that they had dealt

with had been committed by juveniles and people from outside the Trial area.  A few stakeholders from remote communities reported hearing about break-ins in their community (i.e. outside Ceduna) but thought that these crimes were committed by people from out of town (i.e. not local) visiting for funerals and other events;  A couple of stakeholders reported that arguments were increasing between community

members. Specifically, community members who received less money through payments such as ISP Newstart Allowance were reported to be asking others who received larger payments (e.g. Disability Support Pension or Parenting Payments) for cash; and  One service provider stakeholder reported that there had been five cases of prostitution among their clients from remote communities outside of Ceduna. However, this occurrence was not mentioned by other stakeholders.

151

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Safety

Overall, no change was reported by stakeholders in relation to personal and community safety since the Trial began.

Other impacts

Overall, the qualitative research identified that to date, CDCT has had a more visible / noticeable impact on financial, wellbeing and parenting outcomes than on alcohol and drug consumption. While stakeholders were able to identify a range of impacts relating to finances, wellbeing and parenting, they were less able to identify specific direct changes to alcohol and drug consumption.

Financial

Overall, many stakeholders had noticed changes that indicated that the Trial was having a positive impact on CDCT participants’ financial circumstances. These included:

 Many stakeholder participants reported noticing changes in household consumption, believed to be due to people having more disposable income. Changes were particularly noted in relation to increased expenditure on the following items:  Clothing - evidenced by:

 A reported increase in sales at the Ceduna “op shop”;  Children attending school and day care with new clothes and shoes - particularly amongst families who previously rarely had new items;  Cars - evidenced by several community members purchasing cars. In one instance, this included having funds available to travel to Adelaide to make the purchase;  Furniture and large household appliances (e.g. baby equipment, lounge suites, tables and whitegoods) - evidenced by:  Increased usage of community trailers / trucks for transporting these items

from Ceduna to communities;  Increased deliveries of such items to communities through online shopping;  Groceries - evidenced by:  Stakeholders reported sighting CDCT participants purchasing and consuming

more food items, general household items (e.g. cleaning products) and personal hygiene items;  An increase in the number of deliveries per week to a remote community from the food truck (increased from one to two deliveries a week);  Increased revenue at a sports club’s canteen—as reported by one

stakeholder;

 “Big ticket” items (e.g. iPhones, Yamaha keyboard and car spotlights) - reported by an Outback store owner who reported an increase in demand and inquiries about such items. In addition, the store’s data showed a substantial increase in sales in the relevant category;

152

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

 A few community leaders and financial service providers indicated that they knew of community members who had been able to save money for the first time, or save more money than usual;

 A few stakeholders (including financial counsellors, and services providing homelessness programs) reported a decrease in requests for emergency food hampers, and a decrease in clients financially defaulting (e.g. loan repayments); and

 A few stakeholders indicated that requests for emergency / crisis financial counselling intervention had reduced since the introduction of CDCT.

Some stakeholders felt that there had been an increase in the humbugging of Age Pensioners. This was perceived to be evidenced by an increase in elderly people attending an aged care health service in Yalata to avoid being humbugged. However, some other stakeholders noted that humbugging was always a significant issue and did not think that this behaviour / occurrence had changed but rather was more noticeable as CDCT participants were less likely to be humbugged now.

Social impacts

The qualitative research identified the following positive social impacts as a result of the CDCT:

 Improved IT literacy - one stakeholder reported that the requirement of CDCT for participants to use the internet and computers (e.g. to access card balances) had resulted in improved IT skills, including amongst older community members who previously had very low IT literacy36;

 Increased requests for employment opportunities - one stakeholder from a remote community outside of Ceduna reported that more people had been requesting employment in order to earn cash; and

 Increased sense of community pride - reported by some stakeholders. The qualitative research identified the only negative social impact of CDCT as being stigma. A few stakeholders reported that community members who thought they spent their money appropriately felt as though they were being “penalised” and / or “discriminated against”.

Housing impacts

Most participants reported there had been no considerable impact on housing since the commencement of CDCT, with overcrowding and rough sleeping still apparent. Most participants noted that this was still a key issue for the region, particularly given the wide-ranging flow-on effects (including health, hygiene, financial, employment).

Parenting and family impacts

36 The Department advised that the Australian Government had invested over $100,000 in internet infrastructure in Ceduna and surrounding communities to support the CDCT.

153

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Overall, stakeholders had observed that the Trial had some positive impacts in relation to parenting / family wellbeing, including:

 Increased purchasing of “treats” (e.g. lollies) by and for children - reported by a few stakeholders;  Some stakeholders working at schools and childcare services noted that some children:  Were attending school with packed lunches and appeared to be less hungry; and

 Had new clothes / shoes that they were “proud” of;  The above was noted to be occurring among families who wouldn’t normally have new items; and  A few stakeholders who ran parenting programs / classes had noticed greater participation

and engagement among mothers. For example, one stakeholder reported mothers attending multiple, follow-on classes in a week. While some stakeholders working in schools / child care centres noticed some improvement in attendance, they were reluctant to attribute this solely to the CDCT. These stakeholders felt that there were other contributing factors, including specific school initiatives and the weather / seasonal effects.

Wellbeing impacts

The qualitative research identified the following positive wellbeing impacts:

 Improved nutrition - due to more money being spent on food, as reported by many stakeholders. This was evidenced by stakeholders seeing more people with “full” shopping trolleys; and

 A few stakeholders reported that there were several individuals in the community who were known for severe alcohol abuse who had reduced their alcohol consumption and had actively sought health services since the commencement of the CDCT. The qualitative research identified the only negative wellbeing impact of the CDCT as being increased purchases of cigarettes - one stakeholder reported that CDCT participants who had previously been on the Basics card (which restricted tobacco purchasing) were now purchasing more cigarettes.

Support services

Overall, the research identified good awareness amongst stakeholders of welfare services, Centacare, ABLE, financial counselling services and MAP / street beat. However, beyond these services there was limited understanding of the range of new / revamped services being implemented as part of CDCT.

Many stakeholders reported that the Ceduna service reform was beginning to have a positive impact, with improvements to:

 The communication and linkages between services providers; and

154

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

 The assistance provided to clients, including referral to other services as required.

Stakeholders identified the following as needs / gaps in service provision or where additional services were needed to meet demand or need:

 Diversionary programs to provide more opportunity and activities  Financial counselling  Domestic violence services including safe houses and community education  Mental health  Homeless / transitional accommodation - including for those entering and leaving

rehabilitation clinics  Programs for men  Programs for 8-12 year olds  Support services that are located / based in remote communities and not just in Ceduna

Circumvention behaviours

Some stakeholders reported that they had heard reports of the following circumvention behaviours, from their clients and / or other community members (however, stakeholders were unable to assess how widespread such practices were):

 Some stakeholders reported Trial participants were selling goods for cash below their value. Examples given included the sale of fishing rods, phone credit and groceries;  One Ceduna store owner identified the purchase of ‘Paysafe’ cards for online gambling as a potential circumvention practice, as Paysafe cards were not restricted by the Indue card and

these PaySafe cards could be used for online gambling. While this stakeholder had decided not to sell Paysafe cards to Trial participants, cards were reportedly available for purchase at other stores;  One stakeholder reported that CDCT participants had been initially accessing prohibited items online (e.g. online gambling, obtaining alcohol through Chrisco hampers), however the stakeholder reported that this had been addressed by the Department;  A few stakeholders reported that some CDCT participants had created BPAY biller accounts and transferred money from their Indue cards to these accounts to later withdraw as cash;  A couple of stakeholders had heard that clients were transferring money to other family members for rent, and these family members were subsequently providing them with the cash;  One stakeholder reported that CDCT participants had set up an account at a local store and attempted to seek a refund in cash from the store at another outlet / location - however, it was noted that the particular store had been recording the payment method on the store accounts to prevent this; and

155

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

 A few stakeholders had heard of local businesses overcharging / processing fake service transactions on Indue cards in return for cash (e.g. hotel room charged at $150 and CDCT participant given $100 cash back). In addition, stakeholders identified a range of practices that allowed CDCT participants to access additional cash / goods, however these were not specific to the Indue card, and had been occurring prior to CDCT. These included:

 Humbugging;  Taxi drivers accepting cards with no balance as payment, with the intention of withdrawing cash when the next Centrelink payment was received. The cardholder would then cancel the card and replace it with a new card; and

 Card sharing - using other people’s cards to purchase items.

Perceptions of card implementation

Overall, many stakeholders felt that the level of community concern about the Trial had decreased and some people were now accepting of, and adjusting to, the card. However, some CDCT participants were reportedly “just putting up with it for a year”, and a few stakeholders felt that there may be a backlash if it became permanent or the Trial period was extended. These stakeholders felt that the Department should start educating the community if there was a possibility of the card continuing to minimise potential future hostility.

Overall, most service providers felt that they had been well informed about the Trial and received enough information to be able to assist with client queries. These stakeholders reported there had been some initial “teething problems” while participants became familiar with the card (including using a PIN, the requirement to select the ‘credit’ account and setting up rental transfers), however understanding of the card had improved over time. Specific positive feedback about the Trial implementation included:

 Effective rollout and communication in remote communities prior to Trial - most stakeholders felt that remote communities had been very well informed due to the Department’s frequent visits to communities prior to and during implementation. A couple of stakeholders reported that the “door-knocking” approach of houses both with Trial participants and non-Trial participants was effective;

 Store owners reported that they were well-informed of the Trial and card functionality and as a result had been able to assist in educating CDCT participants on how to use the card when making purchases;

 Distributing the card in conjunction with money management / budgeting advice in communities outside of Ceduna was perceived to be very effective;  However, stakeholders felt this was not as effective in Ceduna where not everyone was given access to such advice / service;

 Having a direct contact phone line to Indue for card support was considered effective as it minimised waiting times;

156

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

 Gradual rollout of the card - a couple of stakeholders felt it was effective to have different Trial start dates for different CDCT participants, to minimise the impact on any one day; and  Having “one-stop-shops” (e.g. CAC) to assist with checking balances, replacing cards and accessing PINs was felt to be effective.

Stakeholders identified the following negative feedback / difficulties with the Trial implementation:

 While the community had been well informed about the Trial and consultation had occurred, there was a perception amongst some stakeholders and amongst some in the community that “the decision had already been made” about adopting the Trial when the consultation occurred;

 Some stakeholders felt that the Trial could have been better communicated to participants and the broader community in Ceduna area (e.g. via TV or radio advertising37);  A few stakeholders felt that the usage of written communication materials (e.g. flyers and posters) was not an effective strategy due to low literacy levels amongst

the population of cardholders;  While the Department of Social Services and the Department of the Prime Minister and Cabinet reportedly worked well together in implementing the Trial and were perceived to have a strong on-the-ground presence in the remote communities, a few stakeholders felt

that Centrelink / the Department of Human Services appeared to be “missing” and less willing to directly engage with communities about the Trial38;  Stakeholders felt it was particularly important for Centrelink to be seen as supportive and willing to assist with the Trial and have an on-the-ground presence in

communities, as CDCT participants were more familiar with Centrelink and many had specific questions about their payment arrangements;  Community panels not up and running - some stakeholders noted that the panel was not set up and fully operational at the start of the Trial;  High turnover of staff in service provision and government agencies - this was felt to impact on efficiency of processes due to lack of consistency, learning and relationship development;  One stakeholder reported that having a large number of email addresses created on shared computers in the Oak Valley Corporation office had created a Google security alert and the computers had been locked, although this has since been resolved;  Many service providers reported that the additional service funding was not provided early enough to allow for services to be ready at the beginning of the Trial, particularly given the significant challenges associated with delivering new programs and support services in a remote location. Specific challenges included: o Staff recruitment - there was a lack of staff with appropriate level of skills / qualification in the local area and it was difficult to attract staff to the location;

37 The Department advised the evaluators that it had engaged in radio advertising in each location. 38 Note that the trial was designed to minimise the involvement of the Department of Human Services in

quarantining arrangements.

157

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

 It was also reported that two staff from the ABLE program did not want to continue working on the program due to their personal views about the CDCT; and o Staff accommodation (for staff relocating); and  Some service providers also reported that the duration of the funding for additional services

/ programs in general is too short (only 1 -2 years) and doesn’t allow for effective set-up and relationship building and therefore potentially constrained sustainable and long-term outcomes from being achieved.

Issues with the Indue card / support processes:

The research identified that stakeholders perceived a number of issues with the Indue card and / or its support processes. While stakeholders acknowledged that some of these were initial “teething problems”, it was unclear the extent to which these had been resolved.

 A few stakeholders reported that there was a lag time between when payments appeared on online balances and when they were able to be accessed. Stakeholders reported that participants had tried to purchase goods with the card when their online account showed they had available funds but had been unable to do so. One store manager reported sighting a participant’s online balance when this has occurred, and confirmed it appeared to have funds available39;

 In remote communities, the reliance on online accounts to check card balances caused some problems due to limited internet access, digital literacy and familiarity (including the need to remember passwords and login details). As a result some CDCT participants had difficulty accessing and keeping track of their card balances40;

 DSS is currently implementing more technological support to allow people to check their balances (e.g. iPads in shops), however a few stakeholders felt that this may not fully resolve the issue as participants would still need to remember login details;

 ATM balance checking is also being implemented, available from mid-December 2016.  The reliance of mobile phones to receive PINs was an issue, particularly in remote communities as:

 Oak Valley has no phone reception - as a result, departmental and service provider staff had been driving out of the community to areas with phone reception to receive text messages for CDCT participants. This had caused significant delays for CDCT participants in being able to access card funds;

 Some community members didn’t own mobile phones / had lost phones and / or often had no phone credit;

39 The Department advised the evaluators that this issue was the result of banks ignoring requests from DHS to delay payments for Centrelink (of either 2 or 4 working days) and that Indue is not at fault. Ignoring this request results in participants receiving money earlier than they should have.

40 The Department advised that other balance checking is available, including phone and SMS alerts.

158

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

 CDCT participants in communities with more limited English levels had difficulty accessing support via the Indue phone line due to the language barrier. Some of these CDCT participants had tried to use other community members as translators, however these third parties were unable to deal with Indue on their behalf due to privacy rules / ID checks required. As a result, one Age Pensioner who was voluntarily on the card had removed herself from the Trial; and

 One stakeholder reported that they had recently heard that there was a 30 minute period of time immediately after welfare payments were received when the full amount of welfare payments were accessible in cash (i.e. before 80% was transferred to Indue card accounts) and some Trial participants had become aware of this and had been queuing up to withdraw cash from ATMs at 3am41. This stakeholder had not had this confirmed and was currently investigating the issue.

Other areas for improvement

The research identified the following additional areas for improvement:

 Ongoing communication - some stakeholders reported that communication and education about the card reduced / stopped once the Trial had begun and felt that there was a need for ongoing communication to:  Educate some Trial participants who had not engaged with pre-Trial

communications;  Consolidate understanding of the card, particularly as there were some functionality aspects that were causing confusion / were less well-understood (e.g. how to set up

rent payments, online functionality and how to check balances);  Clarify misconceptions - some Trial participants who had payments suspended due to not meeting CDP participation requirements mistakenly thought it was as a result

of the CDCT. There is potential for this issue to also arise with forthcoming welfare payment changes;  Forms for applying to the community panels to have payment arrangements altered - a few stakeholders reported that some of their clients had found these lengthy and difficult to understand, particularly given limited levels of literacy in remote communities;  One stakeholder suggested placing computer terminals and staff in frontline services in Ceduna (e.g. Day Centre) to provide Trial participants with better access to card support and online card services, particularly as some clients are not comfortable / willing to go into a Centrelink office for support; and  Some stakeholders felt that CDCT participants were adversely impacted by their inability to make cash-based purchases in some situations, including from second hand websites, sports club canteens, tickets for community events (e.g. Oyster Fest and NAIDOC).

41 The Department advised the evaluators that this was not technically feasible.

159

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

East Kimberley

Methodology

 In-depth interviews and focus groups - n= 40

 Dates: Monday 12 September - Tuesday 4 October

 Locations: Kununurra and Wyndham

Overall impact of Trial

Overall, stakeholders’ perceptions of the impact of the CDCT were mixed based on the client base that stakeholders were dealing with. Overall, those stakeholders dealing with the most high needs / disadvantaged cases felt the CDCT had made less of an impact than those stakeholders who dealt with a broader range of clients.

Many stakeholders had noticed some early positive changes / impact of the CDCT, however most felt that it was “too early” in the Trial to determine how widespread such impacts would be. In addition, most felt that it was too early to determine whether there would be any considerable change in relation to longer term issues (e.g. employment, health and housing) and were expecting to only see changes in relation to these issues in the longer term (i.e. in another 6-12 months).

Some stakeholders indicated that there were other external factors which may have impacted on data / behaviour over the Trial period, beyond the impact of CDCT itself, such as:

 Takeaway Alcohol Management System (TAMS) and other alcohol restrictions (including restricted liquor store operating hours, and homeowner nominated alcohol restricted premises);

 Fewer people in town as a result of people leaving / being away during the dry season;  Lump sum payments - some visitors to the community had received royalty payments and this had encouraged some “party houses” to continue. Stakeholders reported that three royalty payments from neighbouring communities outside the Trial site had been received

since the Trial began;  Weather - while not as big an issue as Ceduna, a few stakeholders indicated that the change in seasons from dry to wet had the potential to have some impact on Trial outcomes;  As identified in the Initial Conditions Report, Western Australia has restricted Electronic

Gaming Machine (EGM) licensing, and there were no EGMs in the Trial sites in WA;  Changes to health servicing via hospitals - since the Trial began, the Accident and Emergency department in Kununurra Hospital had ceased its GP service;  Changes in police policing approach / requirements, including:

o Stricter domestic violence reporting requirements, and an increased focus on domestic violence by the new Police Superintendent;

160

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

o A new Police Officer in Charge in Wyndham with a more active focus on reducing public intoxication; and o Police efforts to increase communication and education about crime rates and statistics;  Distribution of a large sum of money ($300,000-$400,000) by an Indigenous Corporation -

one stakeholder reported that community members had placed purchase orders to buy items (e.g. furniture, clothing and shoes) during the Trial; and  Closure of a mine site near Wyndham - which stakeholders reported had resulted in higher unemployment and less money in town.

Alcohol consumption

While some stakeholders felt that there had been minimal / no change in alcohol consumption, many others had noticed changes at an individual and / or community level since the start of the CDCT.

The stakeholders who had noted positive impacts of CDCT in relation to alcohol consumption reported that these included:

 Feedback from Sobering Up Units (SUUs) in Kununurra and Wyndham indicated that there had been42:  A decrease in the number of people being picked up as well as using the SUU;  A change in the types of people being picked up by the SUU - now more likely to be

people from outside the Trial area;  Less visibility / fewer people consuming alcohol in public - some stakeholders reported seeing and hearing fewer intoxicated people in public places (i.e. on the streets in town, in parks and on the side of the road in the early mornings);  One stakeholder had observed that binge drinking patterns had changed amongst some of

the CDCT participants that they had dealings with and that these sessions were now of a shorter duration;  Hospital related changes:  Admissions due to alcohol-related presentations had decreased;

 A noticeable decrease in rowdy and abusive behaviour towards Accident and Emergency staff since the Trial. A stakeholder from the hospital estimated it to have reduced by around half;  Ambulance-related changes:

 Decrease in primary call-outs (from 107 in August-September 2015 to 73 in August-September 2016);  Fewer call-outs for alcohol-related injuries;

42 For supporting data see Figure 93 under MTO PI#4: Drug/alcohol-related injuries and hospital admissions

161

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

 Substantially decreased sales at a Kununurra bottle shop - this decrease was reported to have occurred since the Trial started, as well as at the comparative point from last year;  This had resulted in a need to reduce casual staff (who were mostly backpackers);

 More sobriety - evidenced by:  A local football coach reported that several players who had previously been unable to play / not trained effectively due to frequent intoxication had reduced their alcohol consumption and significantly improved their behaviour and their commitment and

performance to the game, resulting in noticeable “transformation” of their lives; and  A few stakeholders reported seeing individuals they knew to previously be high alcohol users who were now more regularly sober and seen to be spending their money on

food, groceries and household items.

Drug use

Overall, consistent with the Initial Conditions Report, stakeholders found it more difficult to comment on levels of drug use in the community - primarily because such behaviours were not as visible and the impacts of using marijuana (the main drug used) were not as overt / aggressive as alcohol.

The following positive impacts of the CDCT on drug use were identified by the qualitative research:

 A couple of stakeholders had directly spoken with CDCT participants who had reduced and / or stopped their drug use as a result of the CDCT, including:  A CDCT participant who had previously been addicted to methamphetamines (“ice”) but had stopped using ice due to limited access to cash;

 A family who was now consuming less marijuana which had allowed them to spend more money on clothes and food for their child and were supportive / happy about the Trial as a result they; and  A few stakeholders felt that the frequency of marijuana usage had reduced due to limited

access to cash.

The following negative feedback in relation to drug use were identified by the qualitative research, although these reports were not observed amongst CDCT participants:

 One stakeholder in Wyndham reported more young people had told her they were using marijuana, however he / she did not think this increase was a result of the CDCT but rather due to an increase in self-reporting from improved efforts to build more open relationships with young people by her organisation; and

 A couple of stakeholders reported that there had been an incident involving young people using sprays (i.e. aerosols). These stakeholders noted that they had counselled these children and had not heard of any other cases since. Unlike Ceduna, and consistent with the Initial Conditions Report, ice usage was felt to be contained to the non-CDCT / working population, and not a key area of concern. However, one hospital

162

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

stakeholder in Kununurra noted that they had begun to see a small number of cases over the last 6-12 months when previously they had no incidents.

Gambling

Overall, the CDCT was not felt to have had a significant impact on reducing the amount spent on gambling, although many participants noted that gambling was not a key concern for the region. The main form of gambling was in relation to card playing among elderly women and was not believed to have as adverse impacts as alcohol consumption. In addition, stakeholders found it difficult to comment on changes to card playing (the main form of gambling), as it was less visible.

The following positive impacts of the CDCT in relation to gambling were identified by the qualitative research:

 A reduction in the number of people playing bingo in Wyndham - one stakeholder reported that the venue had such a reduction that they had almost been unable to obtain their license for their regular bingo night;

 A couple of stakeholders reported that people they personally knew found it “too difficult” / “frustrating” to play cards due to limited access to cash;  A reduction in card games in public places - identified by a couple of stakeholders; and  A reduction in the amount of money being spent in card games - one stakeholder reported

that one of his organisation’s staff members was previously spending “all of her money” gambling and since the CDCT had been spending more money on food as she was unable to afford to continue gambling. The only negative impact identified by the qualitative research in relation to gambling was a few stakeholders who reported that they had seen or heard CDCT participants now “buying-in” to card games with their Indue card, and giving others their PIN to access funds.

Crime, violence and harm

At the Initial Conditions reporting stage, many stakeholder participants had expected that there would be a significant increase in petty crime amongst CDCT participants at the start of the Trial. While an increase of such crime was not evident among CDCT participants, it was commonly reported by stakeholders to be occurring among children as a result of reduced access to cash.

Overall, there was no significant positive change noted in relation to crime, violence and harm among CDCT participants. However, a couple of positive changes were noticed / reported by stakeholders, including:

 A decrease in vandalism of ATMs - noticed by a few stakeholders;

163

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

 A decrease in crime (including alcohol related incidents) in Wyndham - although it was reported that there were many other contributing factors (e.g. TAMS) and it was felt that this could not be directly attributed to CDCT; and

 A reduction in the number of injuries indicative of domestic violence presenting at the hospital reported by The Accident and Emergency department.  However, a few stakeholders identified some increase in domestic violence / intervention orders - although it was not clear whether the increase was due to changes

in reporting requirements, the policing approach or increased community awareness, understanding and willingness to take action. In addition, many stakeholders reported an increase in the following illegal / harmful behaviours among young people / children:

 Robberies / thefts from cars / vehicles and dwellings - stakeholders reported that in these cases young people were in search of cash; and  Petty crime (e.g. pickpocketing and “snatch and grab”) - stakeholders reported that children on bikes were often the perpetrators in these cases.

Safety

Overall, some stakeholders reported improvements in the safety of public places, particularly during the day in Kununurra. A few stakeholders thought White Gum Park was safer as a result of decreased alcohol-fuelled behaviours - a couple noted that usage of the park had changed since CDCT began as a result, with more families using it for picnics / BBQs, and riding bikes through the skate park. Other stakeholders indicated that safety was still an issue, especially at night, due to “roaming groups of children” and crime among young people (see above).

Most stakeholders did not feel safety was an issue in Wyndham during the day, although a few felt it was unsafe at night (particularly for non-locals). However, these participants did not report any change in safety in Wyndham.

Other impacts

In addition to direct impacts of CDCT on alcohol and drug consumption, stakeholders identified a range of impacts on financial, wellbeing and parenting outcomes.

Financial

Overall, many stakeholders had noticed changes that indicated that the Trial was having a positive impact on CDCT participants’ financial circumstances. These included:

 Many stakeholder participants reported noticing changes in household consumption, believed to be due to more disposable income. Changes were particularly noted in relation to increased expenditure on the following:

164

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

 Clothing - evidenced by:  Children attending school and day care with new clothes, uniforms and shoes - particularly amongst families who previously rarely had new items;  Increased sales of school sports team jerseys;  Community members observed in new clothes;

 Cars - a couple of stakeholders reported knowing of people who had been able to buy a car, and noted that this was due in part to the Indue card;  Furniture and large household appliances - evidenced by a stakeholder seeing “many purchases” from Retravision on clients’ Indue card statements;  Groceries - some stakeholders, including store owners, reported sighting Trial

participants purchasing and consuming more food items, “luxury items” (e.g. prawns) and more expensive brands of groceries;  Food - one local takeaway food store in Kununurra that was frequented by Trial participants reported that:

 Store traffic had increased since the Trial;  Families who previously only purchased lunch a couple of times per week were buying lunch more frequently;  Some families who previously had not been able to buy children “treats”

were now buying these items from the store;  “Big ticket” items (e.g. iPhones and bicycles) - reported by one stakeholder who had seen children in town with these items;  School expenses - one school-based stakeholder reported that the school usually had to

financially support families to allow children to attend Year 6 camp. This year, all families were able to pay for it themselves without financial support, and children had been given pocket money for spending;  A change in spending / consumption patterns during the end of financial year period - a few stakeholders reported that when family bonus payments had been received they had been spent on larger purchases and items for children (e.g. scooters - which had sold out of the local shop) when in previous years such lump sum payments were spent on alcohol;  A few stakeholders indicated that they knew of members of the community who had been able to save money for the first time, or save more money than usual; and  A few stakeholders reported increased access to, and demand for financial counselling services.

However, the qualitative research identified the following negative financial impacts:

 Stakeholders reported that some people were circumventing the card by bartering items for cash at a reduced rate. As a result of this reduced bartering rate, these people had decreased disposable income;

 Some stakeholders reported that there had been an increase in humbugging, particularly of Age Pensioners;

165

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

 However, some stakeholders reported that this had decreased since the beginning of the Trial as some Age Pensioners had voluntarily gone onto the CDCT and / or now told family members that they were on the card to avoid being humbugged; and  Stakeholders reported that taxi drivers had been significantly inflating flat fees for some

longer trips for Trial participants.

Social impacts

The qualitative research identified the following positive social impacts as a result of the CDCT:

 A few stakeholders reported that more people were seeking employment and cash-in-hand work (both ad hoc and regular);  One stakeholder reported an improvement in some CDCT participants’ willingness to engage in training programs and had been able to engage participants in security training

certificates as a result of Trial funding;  A few stakeholders reported cases of people who had begun to contribute / participate in community activities, including:  A member of local sports team who had previously been unable to play due to frequent

intoxication, who was now one of the team’s best players; and  A couple who previously consumed significant volumes of alcohol, who were now sober and had cleaned a public reserve and built children’s play equipment.

The qualitative research identified the following negative social impacts related to the CDCT:

 Stigma - a few stakeholders reported that community members who thought they spent their money appropriately felt as though they were being “penalised” and / or “discriminated against”. Stakeholders cited the colour of the card as a key issue enabling cardholders to be easily identified and “discriminated against”;

 One stakeholder reported that some families on the CDCT had taken elderly family members out of aged care facilities in order to gain access to their cash; and  One stakeholder had heard reports that a few Age Pensioners had volunteered for the Trial and their family members had pressured them to withdraw and / or had taken them to

Centrelink to force them to withdraw.

In relation to the impact of the CDCT on the level of humbugging in the community, stakeholder views were mixed. While some stakeholders reported that humbugging had increased since the Trial, others thought it had decreased and / or felt that there had been no change. A couple of stakeholders also noted that humbugging behaviours had changed as a result of the Trial, with people now being humbugged for cash and to place bets at the TAB instead of for food.

Housing impacts

Most participants reported there had been no considerable impact on housing as a result of the CDCT.

166

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Parenting and family impacts

Overall, some stakeholders had observed that the Trial had some positive impacts in relation to parenting / family wellbeing, generally in relation to the increased expenditure on children. Positive parenting / family impacts included:

 Some stakeholders working at schools and childcare services and / or involved in the delivery of family programs noted that:  Some children were now attending school with packed lunches;  Some children had new clothes / shoes that they were “proud” of;

 The above were noted to be occurring among families who wouldn’t normally bring food to school or have new items;  When they were conducting client visits, some families appeared to have more food in their homes;  One school-based stakeholder reported that children had been given pocket money

for spending on school camp which had not occurred in previous years;  Increased purchasing of “treats” (e.g. lollies) by and for children - reported by a few stakeholders. A couple of stakeholders reported that some toys (including scooters), had sold out at the local store; and  One stakeholder reported that the recent school athletics carnival had its highest ever

attendance (approximately 1,000), and behaviour was considerably better than previous years with no fights or issues relating to alcohol.

Wellbeing impacts

The qualitative research identified the following positive wellbeing impacts as a result of the CDCT:

 Hospital related changes:  Fewer presentations of injuries and illnesses;  A 50% reduction in people discharging themselves against medical advice when admitted to the hospital ward from Accident and Emergency;

 Ambulance-related changes:  A decrease in primary call-outs (from 107 in August-September 2015 to 73 in August-September 2016);  Fewer call-outs for alcohol-related injuries;

 Improved nutrition (for both adults and children) due to more money being spent on food, and higher quality food being purchased. This was evidenced by:  Many stakeholders seeing community members with “full” shopping trolleys, or shopping bags - particularly amongst those who would normally be seen intoxicated

and without food;  Increased sales at a takeaway food shop;  A store owner reporting more money being spent on premium brands and “luxury”

items;

 Increased usage of health services, including:

167

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

 Self-referrals to a mental health service in Kununurra had increased considerably. In particular, the demand for psychology services at OVAHS had increased so significantly that in order to meet the increase in demand they had reduced consultation times (from 2 hours - 30mins) and frequency (e.g. from three times per week to twice per week per patient); and

 New types of people / clients accessing general health services - a couple of stakeholders felt that this was as a result of people having more sober time, and realising that they required these services; and  More time being spent on healthy activities, including sports (football), fishing trips, hunting

trips and training programs.

The only negative wellbeing impact of the CDCT in relation to wellbeing identified by the qualitative research was fewer people from outside the Trial site were reported to be attending a rehabilitation clinic located in the Trial site, as they were concerned they would be put onto the CDCT.

Support services

Overall, most stakeholders (other than those directly delivering additional services) had limited awareness and understanding of support services that were being provided through the Trial in Kununurra and Wyndham. Specifically:

 In Kununurra, there was some awareness that OVAHS and BOAB Health had received additional funding for services as a part of the Trial. However, there was limited awareness of what specific services / programs were being implemented (other than amongst service providers delivering the specific service);

 Ngnowar Aerwah Aboriginal Corporation was reported to have a “flexible” amount of funding to deliver services in Wyndham. This flexibility was felt to be effective as it allowed the Corporation to respond and tailor services to meet specific needs and requests;  However, it was reported that only some of the available funding had been accessed

and that it had not been fully utilised to date. In addition, community leaders in Wyndham were unaware of how this funding was being used; and  Some stakeholders reported that an intensive family program ‘One family at a time’ delivered through Waringari had been expanded to Wyndham as a part of the Trial funding.

Stakeholders identified the following as needs / gaps in service provision or where additional services were needed to meet demand or need:

 Mental health  Youth programs (including diversion and support), especially for ages 9 - 16  Diagnosis of Foetal Alcohol Spectrum Disorders (FASD)  Staffing of Wyndham services

168

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

 Transitional accommodation and holistic support to assist people leaving rehabilitation to re-enter the community and maintain changes in their behaviour;  Rehabilitation - stakeholders reported that this service was only available in Wyndham, not Kununurra.  Diversionary programs  Employment programs

Circumvention behaviours

Some stakeholders reported that they had heard reports of the following circumvention behaviours, from their clients and / or other community members (however, stakeholders were unable to assess how widespread such practices were):

 Some stakeholders reported Trial participants were selling goods for cash below their value. Examples given included groceries, toys, petrol and cigarettes;  A couple of stakeholders had heard that clients were transferring money to other family members for rent, and these family members were subsequently providing them with the

cash;

 Many stakeholders reported that merchants (i.e. taxis) were offering cash back at a reduced rate (e.g. charging the cardholder $100 and giving them $70 cash) and / or were buying alcohol on behalf of cardholders;

 “Sly grogging” - i.e. people buying large amounts of alcohol and on-selling this (at a marked-up rate) to CDCT participants via their Indue card by putting through their alcohol purchases as fake transactions (e.g. taxi fares). In some instances taxis were reported to be paying others to buy alcohol on their behalf to overcome TAMS restrictions and because some local bottle shops refused to sell alcohol to taxis due to such practices; and

 A couple of stakeholders reported that they had heard of some incidents of taxi drivers taking advantage of young female CDCT participants (i.e. providing rides in exchange for sexual favours). However, it was reported that such incidents had also occurred before the Trial began.

Perceptions of card implementation

Overall, many stakeholders felt that the level of community concern about the Trial had decreased and some people were now accepting and adjusting to the card. These stakeholders noted that some of their clients / community members reported that they had experienced some positive outcomes as a result of CDCT.

Overall, most service providers felt that they had been well informed that the Trial was commencing. However, many stakeholders felt that the level of community education about the Trial could have been improved, especially in relation to:

 The mechanics of the card;  What action needed to be taken to transition existing payment arrangements; and

169

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

 Where to get assistance / help (i.e. local Indue office). Stakeholders reported there had been some initial “teething problems” while participants became familiar with the card (including using a PIN, the requirement to select the ‘credit’ account and setting up rental transfers). While understanding of the card had improved, there were still issues being experienced.

Specific positive feedback about the Trial implementation included:

 Stakeholders involved with the implementation group reported that the State and Commonwealth Governments worked effectively together and actively engaged with local Indigenous leaders;

 The involvement of local Indigenous corporations in educating and engaging the local community as well as using these corporations to support the card implementation (e.g. for card distribution) was felt to be effective and important;  Stakeholders reported that many community members had their mail sent to these

corporations, so it was a logical and familiar place to distribute many cards;  Many stakeholders reported that community leaders had been willing to show their support / back the Trial and engage in direct forums with community members who held negative views / concerns about the CDCT. These stakeholders felt that this was important to show

that the local community was supportive of the Trial;  However, a few stakeholders felt that these forums / sessions had been “defensive” and community leaders had not actively listened and / or sought answers to people’s concerns and questions;  A couple of stakeholders reported that information sessions had been held with older

children in schools (i.e. aged 16+), and felt this was a good opportunity for students to ask questions; and  The on-the-ground presence of DSS - stakeholders reported that this enabled easy communication and resolution of issues. Stakeholders identified the following negative feedback / difficulties with the Trial implementation:

 Overall, stakeholders felt the communication about the card had been “rushed”, with limited lead-up time given before the start of the Trial. Understanding of the card was felt to be better in Wyndham, as it was a smaller town and was easier to communicate to a smaller population. However, communication in Kununurra appeared to have been less effective;

 Most communication was reported to have been passive (i.e. requiring Trial participants to actively approach / engage) - some stakeholders felt there should have been more active communication including door-knocking and local advertising - particularly as many people impacted by the Trial did not attend such sessions and / or despite attending did not understand the mechanics of the card;

 Specific aspects of the Trial that stakeholders felt CDCT participants had not been aware of and / or were not well-understood included:  That arrangements needed to be made to move automatic Centrelink deductions (e.g. rent and payments for schools) - it was reported that some

170

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

CDCT participants had assumed that these direct payments would automatically transfer across to their Indue accounts. As a result some experienced fines, late fees / interest charges and / or “children went without food at school”;  Understanding what could be bought on the card - a few Trial participants

had initially spent all their cash component straight away as they did not understand what they could buy on the card;  Understanding of card balances - some CDCT participants thought they were being charged fees and / or “missing money”, which appeared to be

due to confusion between their ‘current balance’ and ‘available balance’. Stakeholders indicated that this was still an ongoing issue requiring education;  Where to access help / support for their Indue card / account - some stakeholders reported that there was limited awareness of the availability of the local Indue office providing card support;  The availability of financial counselling that was implemented as part of the Trial.  Engagement with Trial participants who spoke English as a second language could have been improved - some of the information / printed materials initially produced were “too complicated” and had to be revised for those who spoke English as a second language.  The requirement for cardholders to link the card to an email address caused difficulties / was not simple as many did not previously have an email address and were unfamiliar with technology.  The panel was not established at the start of Trial and was still not established at Wave 1 of this Evaluation43;

 One stakeholder also felt that the change in the payment split variation from the arrangements that were initially communicated to participants (i.e. change from 50-50% to 70-30%) was likely to cause annoyance and anger amongst some Trial participant.  Additional ‘wrap around’ support services were not in place for the commencement of the

Trial as services providers had only received certainty in their funding arrangements very close to the commencement of the Trial;  Service providers indicated that they required a lead up period of approximately three months to ensure that they had adequate time to have additional programs /

services established, especially given the staffing challenges in remote locations; and  A few service providers also noted that it was important that they be given adequate notice if their funding was to continue and / or cease after the Trial period for

planning

43 Note that the community panel commenced in October 2016.

171

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Issues with the Indue card / support processes:

The research identified that stakeholders perceived the following issues with the Indue card and / or its support processes (while it was acknowledged that some of these issues had been initial “teething” / educational problems, it was unclear the extent that they had been addressed):

 A couple of stakeholders reported that there was a group of CDCT participants who were unfamiliar with the concept of using a debit card and were losing their cards very frequently (e.g. once every one-two weeks). These participants were previously able to have a 3rd party hold their card and PIN and withdraw cash once a week (e.g. banks or Aboriginal corporations);

 A couple of stakeholders reported that they had heard there were some stores where the card was not accepted / did not work (e.g. petrol stations outside of the Trial site and in Derby);

 The reliance on online accounts to check card balances caused some problems due to limited internet access, digital literacy and familiarity (including the need to remember passwords and login details). As a result some CDCT participants had difficulty accessing and keeping track of their card balances; and

 The 20% cash component was available in Trial participants’ accounts before the 80% card component - one stakeholder reported that as a result of this some Trial participants ran out of their cash component as they needed to go shopping immediately after receiving their welfare payment, and had to spend their cash component as the card funds were not yet available.

In addition, one stakeholder was concerned about the potential impact of blackouts during the upcoming wet season, which usually put EFTPOS terminals out of operation. This stakeholder noted that people were usually able to withdraw cash from banks in this situation, but was concerned about what would happen this wet season with people only able to access 20% cash. This stakeholder reported that this issue could be compounded when the DSS office is closed for Christmas for 2 weeks.

Other areas for improvement

Stakeholders identified the following additional changes / suggestions to improve the CDCT:

 Ongoing communication - some stakeholders reported that communication and education about the card reduced / stopped once the Trial had begun and felt that there was a need for ongoing communication to:  Educate some Trial participants who had not engaged with pre-Trial

communications;  Consolidate understanding of the card, particularly as there were some functionality aspects that were causing confusion / were less well-understood (e.g. how to set up

rent payments, online functionality, understanding of card balances);  To inform Trial participants of new arrangements (e.g. when the Community Panel begins operating);

172

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

 Clarify misconceptions - a few elders sitting on boards had confused changes in the payment arrangements for sitting fees (which had recently begun to be counted as income by Centrelink) and “blamed” the CDCT for reducing their income support payments;  Adopting a range of different coloured cards - some stakeholders felt that the stigma of the

card would be reduced considerably if Trial participants had different coloured cards, and were able to choose their own colour as they would be less identifiable;  Increased flexibility in discretionary cash - a few stakeholders felt that their needed to be greater flexibility to access larger amounts of discretionary cash as $200 per month was not

enough in some circumstances (e.g. to pay back personal cash loans and to send money to children at boarding schools); and  More regularity in the discretionary cash component - a few stakeholders reported that having more regular access but smaller amounts of discretionary cash (i.e. $50 per week

rather that $200 per month) would be easier for Trial participants to budget / manage their cash component.

173

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Appendix D: Organisations Interviewed and Contacted in Qualitative Research Ceduna and Surrounds

44 The number of organisations that participated in the evaluation does not equal the number of participants

interviewed because in some cases multiple people from the same organisation were interviewed and n=6 organisations from Ceduna did not consent to being identified.

45 Organisations were contacted to participate at least three times.

Participating organisations44

Ceduna Aboriginal Corporation Aboriginal Family Support Services Ceduna District Health Service

Scotdesco Save the Children SA Police

Koonibba Aboriginal Community Council Inc. Ngura Yadurirn Children and Family Centre

Department of Prime Minister and Cabinet

Yalata Community Inc. Centacare Catholic Family Services Yalata Outback Store

District Council of Ceduna Able Program (through Centacare) SA Ambulance Service

Families SA Housing SA IGA Thevenard

Ceduna Area School Ceduna Youth Hub Tullawon Health Service

Aboriginal Drug and Alcohol Council Ceduna Foodland

Declined invitation to participate

Life Without Barriers Uniting Care Wesley Joanne’s Anglican Op Shop

Eyre Futures Red Cross

Contacted45 but not reached

Oak Valley Inc.

174

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

East Kimberley

46 The number of organisations that participated in the evaluation does not equal the number of participants

interviewed because in some cases multiple people from the same organisation were interviewed and n=4 organisations from Kununurra / Wyndham did not consent to being identified.

47 Organisations were contacted to participate at least three times.

Kununurra

Participating organisations46

Gelganyem Trust Save the Children Wunan Foundation

Ord Valley Aboriginal Health Service WA Reform Unit Waringari Aboriginal Corporation

MG Corporation Kununurra District High School St John’s Ambulance

Kununurra Crisis Accommodation Centre East Kimberley Chamber of Commerce and Industry

Kimberley Mental Health and Drug Service

WA Attorney General’s Department

WA Country Health Service Gulliver’s Tavern

Legal Aid WA Department of Child Protection

and Family Services

WA Police - Kununurra WA Housing Authority

Contacted47 but not reached

St Joseph’s Primary School Kununurra

Wyndham

Participating organisations4

Ngnowar Aerwah Aboriginal Corporation Kimberley Land Council Board of Directors

Ngnowar Aerwah Sobering Up Shelter

Wyndham District High School Joongari House Wyndham Early Learning Activity

Centre (WELA)

East Kimberley Job Pathways WA Police - Wyndham Wyndham Supermarket

Declined invitation to participate

Wyndham District Hospital Wyndham Community Club

Contacted5 but not reached

Shire of Wyndham East Kimberley

175

DSS Cashless Debit Card Trial Wave 1 Interim Evaluation Report

Appendix E: Interview Questionnaire Results Average ratings across stakeholders who completed the interview questionnaire

Location East Kimberley Ceduna

Question Indicator

Initial conditions48

n=23

Wave 1

n=36

Initial conditions13

n=19

Wave 1

n=31

How much of an issue are each of the following in the local community?

(Average ratings on a scale of 0 - Not at all to 10 - Extremely severe)

Alcohol abuse 8.3 6.8 7.4 7.0

Drug use 6.9 5.6 6.8 6.7

Gambling 6.7 5.0 7.7 6.5

Violence and other crimes 8.0 6.3 7.0 6.2

Street begging 5.0 3.9 5.4 4.0

Humbugging 5.9 4.7 6.3 4.9

Harassment, abuse, intimidation 5.8 4.4 5.9 4.3

How well is the local community performing on each of the following aspects?

(Average ratings on a scale of 0 - very poorly to 10 - very well)

Ability to afford basic household goods 3.7 5.6 4.4 5.6

Paying bills 3.5 5.5 4.3 5.0

Employment 3.4 3.6 3.6 3.5

Education / training 3.6 4.5 3.9 4.3

Nutrition 3.2 4.6 4.2 4.4

Health and wellbeing 3.5 4.5 4.4 4.7

Community pride 4.3 5.0 4.9 5.1

Community safety 4.2 5.2 4.6 5.0

48 Some participants in the evaluation who were not interviewed for the Initial Conditions Report completed a questionnaire retrospectively. These average ratings include retrospective responses.