Note: Where available, the PDF/Word icon below is provided to view the complete and fully formatted document
Data-matching Program (Assistance and Tax) Act - Report - Department of Social Security and the Data-matching Agency - 1993-94


Download PDF Download PDF

->n AUSTRALIA,.,'·'

D a t a - m a t c h i n g P r o g r a m , R e p o r t o n p r o g r e s s , O c t o b e r 1 9 9 4

Department of Social Security and the Data-Matching Agency

Data-matching Program

Report on progress October 1994

Australian Government Publishing Service Canberra

© Commonwealth of Australia 1994

ISBN 0 644 35317 1

This work is copyright. Apart from any use as permitted under the Copyright Act 1968, no part may be reproduced by any process without prior written permission from the Australian Government Publishing Service. Requests and inquiries concerning reproduction rights should be directed to the Manager, Commonwealth Information Services, Australian Government Publishing Service, GPO Box 84, Canberra ACT 2601.

Produced by the Australian Government Publishing Service

The Hon Peter Baldwin MP Minister for Social Security Parliament House CANBERRA ACT 2600

D epartm ent of Social Security Box 7788, C a n b erra Mall Centre, ACT 2610 Telephone: (06) 244 6400 · Fax: (06) 244 7983

Dear Minister

In accordance with section 12 (2 A) of the Data-matching Program (Assistance and Tax) Act 1990, 1 present the fourth report on progress with the operation of the Program which covers both the Data-Matching Agency and the Department as a participating source agency.

This report demonstrates the Department's continuing commitment to strengthening the controls over incorrect social security and other payments through the Data- matching Program. In the year under review the Department has consolidated its efforts by making enhancements to individual matching projects. These enhancements

are helping to target incorrect payments more effectively using the Program. The Department also has finished collecting tax file numbers from existing clients and their partners which means that income matching is now more comprehensive.

The Department has continued to promote awareness of the Data-matching Program and its use of tax file numbers to deter people from continuing to claim or attempting to claim social security payments to which they are not entitled. As people become more aware of the increased likelihood of detection and decide to comply voluntarily

with the system, the level of incorrect payment available for detection will reduce affecting the level of direct savings. Direct savings in public expenditure continue to be realised at relatively low cost, but in future years the Program, if continued, is likely to be of most value for its deterrent effect.

The Department has maintained its strong record on protecting the privacy of individuals. Much of its success in this area can be attributed to the close working relationship the Department has with the Privacy Commissioner and his staff. In the interests of privacy and security of personal information the Regional Office network

has made a major effort to remove tax file number records from clients' paper files. The aim is to ensure that the Department holds tax file number records only in encoded form on its computer system.

Yours sincerely

A S BLUNN

CONTENTS

Executive summary vii

Introduction 1

Chapter 1 — Overview of the Data-matching Program 5 Fraud control strategy 5

Introduction of the Data-matching Program 7

How the Program works 8

Privacy safeguards 16

Chapter 2 — Program enhancements 19

Evaluation program 19

Improvements to families end of year processing 20 Revised strategy for reviewing personal identity 20 discrepancies

Chapter 3 — Legislative issues 23

Legislative amendments 23

Future changes to the Act 25

Chapter 4 — Assessments of the Program 27

Ongoing assessments 27

Australian National Audit Office efficiency audit 28

Chapter 5 — Action taken on discrepancies 35

Data-Matching Agency input and output 35

Action taken on discrepancies by DSS 37

Chapter 6 — Savings in outlays and other benefits 45 Variations to original estimates 45

Indirect savings — voluntary compliance 45

Direct savings — methodology 48

Actual savings achieved in 1993-94 50

Estimated direct savings for 1994— 95 and beyond 52 Total actual and estimated savings for other agencies 53 Summary of direct and indirect savings 53

Other benefits 53

vi Data-matching Program

Chapter 7 — Operating costs 55

Methodology 55

Administrative costs 56

Salary costs 56

Total actual and estimated operating costs 58

Chapter 8 — Costs and benefits 59

Opportunity costs 59

Upward variations 61

DSS total costs 61

DSS direct cost-benefit summary 61

Total costs for all agencies 63

Cost-benefit summary for all agencies 63

Appendices Appendix A Data-matching Program cycle 65

Appendix B Data-Matching Agency input and output 67 Appendix C Action taken on discrepancies by DSS 71 Appendix D Costs and benefits 79

Glossary 85

Chronology 89

EXECUTIVE SUMMARY

The Data-matching Program is an important control measure which helps the Department to maintain the integrity of the

social security system and be accountable for the funds it distributes. A key objective is to

deter incorrect payment by encouraging voluntary compliance. The Department has operated the Program with

due regard for the privacy and security of personal information. (Chapter 1)

A comprehensive and continuous evaluation process is undertaken by the Department to ensure that the information

produced about possible incorrect payments is as accurate as possible. This close scrutiny has shown a steady

improvement in results. In 1993-94 payments were reduced or cancelled in 32.7 per cent of completed reviews where the client was notified of a discrepancy, an

improvement on the 22.7 per cent in 1992-93. The rate of overpayments detected has improved from 29 per cent in 1992-93 to 41.6 per cent in

1993-94. (Chapter 2 and 5)

In July 1994 the Social Security Act 1991 was amended to make the intention of the legislation clear that people who do not

have a tax file number and who have no intention of applying for one will not receive payments. This change was in response to a

decision by the Administrative Appeals Tribunal in July 1993 which held that the Act, as it stood, did not apply to people who did not have a tax file number and had no intention of applying for one. (Chapter 3)

The Australian National Audit Office (ANAO) conducted an efficiency audit of the Data- matching Program during 1993.

Its report was tabled in Parliament in October 1993. The report indicated that the Program is a 'valuable

compliance and control technique' which addresses areas of risk that other review activities do not cover as

effectively. It also confirmed that the Department and the Data-Matching Agency effectively fulfil their obligations

in relation to privacy. The ANAO made a number of recommendations for operational improvements and reporting

which have been substantially

viii Data-matching Program

met by the Department. (Chapter 4)

Significant direct savings in social security outlays have been made through the Data-matching Program. In 1993-94 the Department saved $79.8 million and currently estimates savings of $58.3 million in 1994-95, $41.9 million in 1995-6 and $39.3 million in 1996-97 and

1997-98. The savings are substantially higher when the indirect effect of increased voluntary compliance is added to the equation. (Chapter 6)

The direct savings are achieved in a cost-effective manner. Costs are relatively low — $16 million in 1993-94 and $19.3 million in 1994— 95 and

$18.1 million in 1995-96 and beyond. (Chapters 7 and 8)

When the full costs of the Program are set against the full benefits the value of the Program remains. In later years the ratio of costs to benefits is 1:7 and when the other participating agencies are taken into account it becomes 1:6. (Chapter 8)

INTRODUCTION

In the 1990-91 Budget, the Government announced that most people claiming or receiving Government financial

assistance would need to provide a tax file number as a condition for receipt of that

assistance. It also announced the introduction of a program of data-matching between agencies making financial assistance payments and the Australian Taxation Office.

From Budget night a six week Amnesty came into effect. It gave people a last chance to advise the Department of a change in their circumstances without penalty before the new requirements to provide tax file numbers and the new Data- matching Program came into effect.

These initiatives required supporting legislation, including the Data-matching Program (Assistance and Tax) Act 1990 which received Royal Assent on 23 January 1991. This Act provided authority for the establishment of a Data-

Matching Agency to match data provided by agencies participating in the Data- matching Program. These

agencies are known as source agencies.

Under guideline 12 of the original Schedule to the Data- matching Program (Assistance and Tax) Act 1990, the Data- Matching Agency and each source agency were required to table a comprehensive report in both Houses of Parliament six months after the first data- matching cycle began. The Schedule specified that the following details were to be included in the report:

. actual costs and benefits flowing from the Program;

. any non-financial factors that were considered relevant;

• difficulties in the operation of the Program and how these were overcome;

. the extent to which

internal audits or other forms of assessment were undertaken by agencies and their outcome; and

2 Data-matching Program

• such other matters as:

- the total number of

matches undertaken;

- the number of

discrepancies;

- the proportion of

discrepancies which resulted in action being taken;

- the number of

instances where subsequent action was taken;

- the number of cases

in which action proceeded despite a challenge as to the accuracy of the data;

- the proportion of

discrepancies which did not proceed to action after the individual was contacted; and

- the number of cases

where successful recovery action occurred.

In accordance with this requirement, on 5 November 1991 the Minister for Social Security tabled a

report on his Department's progress with the operation of the Program in its first six months.

Section 12 of the Data-matching Program (Assistance and Tax) Act 1990 also required the Privacy Commissioner to

replace the interim guidelines in the Schedule to the Act with a final version. He did this on 27 September 1991. The revised guidelines specified that another report must be tabled by the Data-Matching Agency and each source agency covering the Program's first eighteen months. The Privacy Commissioner requested second reports because the Program's incremental implementation and the short period it had been operating at the time of the first reports meant that he and the agencies had been unable to assess adequately some aspects of the Program.

The matters to be reported on by the agencies remained the same as those required for the first reports. The second report covering the Data-Matching Agency and the Department of Social Security was tabled on 15 October 1992. This report, and those from other source agencies, provided the basis for debate in Parliament on whether it should allow the Program to continue beyond

Introduction 3

23 January 1993. The result was an amendment to the Data- matching Program (Assistance and Tax) Act 1990 to allow the Program to continue for a further

12 months to 22 January 1994. The amendment also included the requirement for further reports to be tabled in Parliament by the end of

October 1993. This was done and Parliament subsequently passed amending legislation extending the Program for a

further two years to 22 January 1996. The reporting requirement was retained with reports to be tabled in

Parliament by the end of October 1994 and 1995. This report has been produced in accordance with that

requirement.

This document covers the activities of the Department of Social Security as a source agency and those of the Data-

Matching Agency. It also provides details of the combined costs and benefits for all participating agencies. Each

agency has still included full details of the costs and benefits associated with its participation in the Program in its separate

report.

The reports tabled in 1992 and 1993 contained comprehensive assessments of the implementation and operation of

the Program since its commencement. This document focuses mainly on the 12 months since the last report, although

some explanatory material has been retained to enable the reader to understand the background to the Program. For

a full analysis this document should be read in conjunction with the earlier reports.

Enquiries about the report should be directed to Lisa Cahill, Security, Fraud and Control Division, telephone

(06) 244 7429.

CHAPTER 1

OVERVIEW OF THE

DATA-MATCHING PROGRAM

Fraud control strategy

One of the Department's aims is to maintain the integrity of the social security system and be

accountable for the funds it distributes. The Data-matching Program helps the Department

to achieve this aim by:

• encouraging voluntary compliance — this includes deterring people from attempting to claim payments to which they

are not entitled, the voluntary surrender of payments to which people

may not be entitled and the voluntary disclosure to the Department of changes in circumstances

which affect rates of payment;

• detecting people who are receiving incorrect payments from an income support agency;

. verifying the accuracy of incomes declared to agencies which make

income support payments; and

• detecting fictitious or assumed identities.

The Program also provides a mechanism which can be used to:

• identify social security clients who appear to be eligible for, but are not receiving, additional

Family Payment; and

. identify clients who are receiving less than their proper entitlements.

The strategy the Department employs to maintain the integrity of the social security system is set out in its Fraud Control

Plan. The plan is revised each financial year and specifies the key compliance activities to be undertaken and areas where the

Department intends to improve its performance.

The Data-matching Program plays an important role in

6 Data-matching Program

detecting dual payments and undeclared or understated income which cannot be detected by other control measures. Its introduction has meant that the Department now has a comprehensive system of controls that enable it to detect a greater number of incorrect payments and frauds than in the past.

As discussed in the Department’s 1993 report, the 1980s was a period in which the Department made significant advances in its methods of controlling incorrect payment and fraud. For example, the Department developed risk-based strategies to select for review clients whose characteristics were similar to those of clients found to be paid incorrectly in the past. It supplemented this with other compliance activity such as the introduction of mobile review teams.

While risk-based review techniques are very effective, their selective nature means that they leave some residual incorrect payments undetected. Risk-based strategies generate a high level of review activity for the unemployed and sole parents because their circumstances can be volatile and there is a strong possibility of changes in circumstances occurring which may affect eligibility. In

contrast, the circumstances of the aged and disabled are less subject to changes that may affect eligibility although small changes in rate of payment do occur. Similarly, basic Family Payment has traditionally had a low rate of incorrect payment, mainly because payments are affected by comparatively fewer changes to clients' circumstances. This makes it difficult to devise cost-effective risk-based strategies to detect incorrect payments in this particular client population.

Matching through the Data- matching Program is the most effective way of reviewing Family Payment clients and is also an important mechanism for reviewing age and disability support pensioners. This is significant given the size of these client populations. For example, there are about

1.6 million age pensioners (including their wives and carers), 545 000 disability support pensioners (including their wives and carers) and

1.8 million Family Payment clients. Together they make up 84 per cent of social security clients and account for 64 per cent of social security outlays.

The Department first began using computer data-matching techniques in the late 1970s and 1980s. Advances in technology

The Data-matching Program 7

have enabled the Department to expand the role of data-matching in the 1990s. (Between 1988-89 and 1993-94 the

number of compliance reviews completed by the Department increased by 208 per cent. At the same time, the proportion of

reviews undertaken as a result of data-matching increased from a minimal level to 36 per cent.) The advantage of data-matching by computers is that it enables the Department to check its client population

comprehensively against a number of risks and to confirm the correctness of payments where no discrepancies are

detected. This means that the Department is able to detect a wider range of incorrect payments than if it relied solely

on other techniques such as risk- based methods of review.

The Department undertakes a number of different types of matching. For example, it matches Employment

Declaration Form data provided by the Australian Taxation Office (ATO) to detect instances in which clients have

commenced work and have not declared their income to the Department or have understated the amount. It is important that

these other types of matching, which do not use tax file numbers, are not confused with the subject of this report.

Introduction of the Data- matching Program

Legislative changes

The introduction of the Data- matching Program required supporting legislation as well as amendments to the Social Security Act 1991.

The legislation covering the Program is the Data-matching Program (Assistance and Tax) Act 1990 (the Act). This

Act, which received Royal Assent on 23 January 1991, provides the authority for the matching of specified data held

by source agencies.

The Social Security Act 1991 was also amended so that most recipients of social security payments must provide tax file

numbers as a condition for receipt of those payments. Tax file numbers are an essential part of the Program since they

provide an efficient means for paying agencies to check that the identity and income information given to them by

clients is consistent with the information they give the ATO.

Tax file number collection

The Department started collecting tax file numbers from its five million clients and their

8 Data-matching Program

partners in June 1991 and completed the task in April 1994. About 400 000 clients have been exempted from the requirement to provide tax file numbers because of their age (80 years and over) or other reasons (for example, the client is profoundly disabled).

A detailed description of the stages of the tax file number collection is contained in the 1993 report. The stages are also listed in the chronology at the end of this report.

If clients or their partners decline to provide their tax file numbers, their social security payments are stopped. This has resulted in savings to the Department and is discussed in detail in chapter 6.

How the Program works

The Data-Matching Agency and source agencies

Section 4 of the Act provides for the establishment of a Data- Matching Agency to match data supplied to it by source agencies.

Source agencies are those Commonwealth Government agencies which supply data to the Data-Matching Agency for the purposes of data-matching.

They are the ATO and assistance agencies. The assistance agencies are the:

• Department of Social Security (DSS);

• Department of

Employment, Education and Training (DEET);

• Department of Veterans' Affairs (DVA); and

. Department of Housing and Regional Development (DHRD) (First Home Owners Scheme only).

Because DSS is the principal agency involved in the Data- matching Program, the Secretary of the Department was given responsibility under the Act for setting up the Data-Matching Agency. It comprises officers of DSS who are responsible for the matching of data under the Act. While operating as part of the Data-Matching Agency, these officers do not have access to social security client information other than that provided to the Data-Matching Agency by DSS in its capacity as an assistance agency.

Data used in the Program

The data provided by the assistance agencies for the Data-

The Data-matching Program 9

matching Program includes details of identity and declared income for clients and, where applicable, their partners, children and parents. In the case of the First Home Owners Scheme, identity details of all applicants (including joint applicants) are provided.

The assistance agencies also provide details of former clients who owe a debt to the

Commonwealth. In addition, DVA provides records of its former clients who are now deceased.

The ATO provides details of identity and taxable income for people in receipt of income support payments and, where

relevant, similar details for their partners and parents. .

Data-matching cycles

The Act specifies that a data- matching cycle must be completed within two months of its commencement and that no

more than nine cycles may be conducted each year. A new cycle cannot commence until the previous one has finished. The Data-Matching Agency has complied with these requirements. In practice, it usually conducts five cycles a year as some time must be allowed for program development and maintenance.

Numerous checks are undertaken during the six steps of a data-matching cycle.

Step 1

In step 1 the Data-Matching Agency checks the validity of the tax file number information given to it by the assistance agencies. It does this by testing the number against a tax file number algorithm provided by the ATO. If an invalid number is found, the Data-Matching Agency compares the identity

data for that client against reference identity data provided by the Australian Electoral Commission and the Health

Insurance Commission. Client records with invalid tax file numbers are then referred back to the relevant assistance

agencies with information on whether the records appeared on Australian Electoral Commission and/or Health Insurance

Commission files. Only records with valid tax file numbers continue through the cycle.

Step 2

In step 2 the Data-Matching Agency extracts from the data provided to it valid tax file numbers and the associated assistance agency identification numbers and passes this

information to the ATO. The assistance agency identification

10 Data-matching Program

numbers are the numbers which the assistance agencies assign to each record.

Step 3

In step 3 the ATO uses the tax file number information given to it by the Data-Matching Agency to extract taxable income and personal identity data for each record. It then passes this information back to the Data- Matching Agency.

Step 4

In step 4 the Data-Matching Agency undertakes identity matching. In this process, the Data-Matching Agency compares the identity information given to it by the ATO with identity information for clients provided to it by the assistance agencies. The purpose of identity matching is to find people who are claiming a payment to which they are not entitled using an assumed or false identity. Although people engaging in tax evasion through the use of a false identity may also come to notice during this process, the ATO has not

identified any such cases to date.

Any cases which fail the identity matching process are compared against reference identity data provided by the Australian Electoral Commission and the

Health Insurance Commission. These records are then referred back to the relevant source agencies with information on whether the records appeared on Australian Electoral Commission and/or Health Insurance Commission files. Only records with confirmed identities continue through the cycle.

Step 5

In step 5 the Data-Matching Agency undertakes payment matching and income matching.

In the payment matching process, data provided to the Data-Matching Agency by assistance agencies are compared to detect people in receipt of two payments from different agencies where the receipt of one of the payments precludes or limits the amount of the other payment (for example, AUSTUDY from DEBT and Job Search Allowance from DSS).

Payment matching can also bring to light people who are receiving a payment as a single person but eligibility for a grant from the First Home Owners Scheme indicates that they are married or living in a marriage-like relationship.

Details of debts owed to the Commonwealth are also included in payment matching to

The Data-matching Program 11

identify whether people may now be in receipt of a payment from another assistance agency

that can be withheld to repay the debt.

In the income matching process, the Data-Matching Agency compares income details of

assistance agency clients (and partners and parents where applicable) with taxable income details provided by the ATO.

The purpose of this matching is to identify clients who have misinformed one or more agencies about their or their

partners' or parents' income and so may be receiving incorrect assistance agency payments or paying less tax than they should.

The Data-Matching Agency also compares income details of clients and partners with their taxable income details recorded by the ATO to identify clients

whose income is such that they may have an entitlement to additional Family Payment but are not receiving it.

Until December 1992, the Act stated that step 5 must be completed within 24 hours. In December 1992 the Act was

amended to allow step 5 to be completed within seven days. The Data-Matching Agency is given exclusive access to the

computer for a weekend to undertake step 5 processing. In

1993-94, step 5 processing for all cycles was completed during the allotted weekend and took an average of 30 hours to complete.

Personal information used in a data-matching cycle that does not indicate a discrepancy is destroyed by the Data-Matching Agency within 24 hours of the end of step 5.

Step 6

In step 6 the Data-Matching Agency returns the raw output to the source agencies. Under guideline 6 of the Schedule to the Act, the Data-Matching Agency must do this within

seven days of the completion of step 5.

A comprehensive report on the outcome of each cycle is provided to the agencies and the Privacy Commissioner.

The data-matching cycle is described diagrammatically in appendix A.

Agency follow-up procedures

Upon receipt of the raw output from the Data-Matching Agency, the Department undertakes an extensive process

of testing the data before releasing the cases to the

12 Data-matching Program

Department's Area and Regional Office network for review.

Staff conducting the reviews adhere to established clerical procedures which were developed in consultation with the Privacy Commissioner.

Staff initially examine the client's computer record and, if necessary, the paper file. This is to ensure that the information provided by the Data-Matching Agency does relate to the client in question. The file is also examined to see if there is any explanation for the discrepancy — for example, the client may

have advised of a change in circumstances after the start of a data-matching cycle. If additional information is

required to conduct the review the Department liaises with the other agency which is a party to the match.

Where, following these checks, there still appears to be a discrepancy, a notice of proposed action is sent to the

client. The client then has 28 days to show cause why the proposed action should not be taken.

The following case studies illustrate the steps involved in conducting reviews generated by the Data-matching Program itself or the associated tax file

number collection. The names of the people concerned have been changed.

Case study 1

Hilary was a client of the Department for 11 years. She was paid Widow Pension.

The Data-matching Program indicated that Hilary may have been in receipt of a Service Pension from DVA under her maiden name. Under the Social Security Act 1991, Widow Pension is not payable to a person who is receiving a DVA Service Pension.

The Department checked Hilary's computer record and paper file. These checks indicated that she had not advised the Department that she was in receipt of a DVA Service Pension.

The Department then requested further information from DVA. It also wrote to Hilary, in accordance with the Data- matching Program (Assistance and Tax) Act 1990, telling her what information it had and that if the information was correct this might result in her pension being stopped and an overpayment being raised.

Hilary was given 28 days to respond to the letter. She

The Data-matching Program 13

replied to the letter, acknowledging that she had been in receipt of a DVA Service Pension since 1977. The information from DVA

confirmed this. The Department stopped Hilary's Widow Pension and raised an overpayment of $70 000. Recovery of the overpayment has been suspended until Hilary has finished repaying an existing

$13 000 debt owed to DVA. The case has been referred to the Director of Public

Prosecution (DPP) for a decision on whether prosecution should proceed.

Case study 2

Kimberley was being paid basic Family Payment for four children. In September 1993, the

Department sent Kimberley an invitation to apply for additional Family Payment. This followed matching with the ATO which indicated that her and her husband's taxable income was such that she may have had an entitlement to additional Family Payment.

Kimberley lodged an application and was granted additional Family Payment for her four children, amounting to

$298.60 a fortnight. She had not applied for additional Family Payment in the past because she

had believed (mistakenly) that she was not eligible because of compensation payments being paid to a member of the family.

Case study 3

Amanda was a client of the Department for 11 years. She was being paid basic Family Payment for one child.

The Data-matching Program indicated that the combined income of Amanda and her husband was such that she may not have been entitled to basic

Family Payment.

The Department checked Amanda's computer record and paper file for an indication of the source of the discrepancy

between the income declared to the Department and the taxable income information held by the ATO. These checks showed

that Amanda and her husband were self-employed and had been receiving basic Family Payment in 1993 on the basis of

an estimate of her and her husband’s income for the 1991-92 financial year. It appeared that their actual

income had exceeded their estimated income by more than 25 per cent. (Under the Social Security Act 1991, a person's Family Payment entitlement must be recalculated if they are being paid on the basis of an

14 Data-matching Program

estimate of their income and their actual income subsequently exceeds their estimated income by 25 per cent or more.)

The Department then wrote to Amanda, in accordance with the Data-matching Program (Assistance and Tax) Act 1990, telling her what information it had and that if the information was correct this might result in her Family Payment being

stopped and an overpayment being raised.

Amanda was given 28 days to respond to the letter. She contacted the Department and confirmed that the Department's information was correct. She advised that it had been an honest mistake and was keen to repay any money she owed to the Department as a result of the mistake. The Department

stopped Amanda's basic Family Payment and raised an overpayment of $3 800. The overpayment is being recovered by cash instalments of about

$1 000 a month.

Case study 4

Michael was a client of the Department from 1990 to 1993. He was paid Job Search Allowance and then Newstart Allowance.

The Data-matching Program detected a discrepancy between the amount of income declared to the Department and to the ATO.

The Department first checked his computer record and paper file for an indication of the

source of the discrepancy between the income declared to the Department and the taxable income information held by the ATO. These checks showed that the Department had no record of Michael's additional income as recorded with the ATO.

The Department then sought more information from the ATO about the amount, source and period of Michael’s income. It then sent Michael a letter, in accordance with the Data- matching Program (Assistance and Tax) Act 1990, which told him the information the Department had and that, if he could not provide a satisfactory explanation, his allowance might be stopped and an overpayment might be raised.

Michael contacted the Department and stated that, while he had received distributions from a family tmst, he had not declared them because he did not know at the time that there was a family tmst and that he had received

The Data-matching Program 15

distributions. The Department stopped his Newstart Allowance and raised an overpayment of $17 700.

Michael appealed against the Department's decision to the Social Security Appeals Tribunal (SSAT). The SSAT affirmed the

Department's decision. Michael then appealed to the Administrative Appeals Tribunal (AAT) which also affirmed the

Department's decision (AAT Reference No. V93/923). The AAT accepted that while Michael had not been aware of the trust or the distributions made to him by the trust, he had in fact made false statements (in the sense that they were incorrect) which misrepresented his financial position. Accordingly, the amount

overpaid to Michael was a debt due to the Commonwealth.

In handing down its decision, the AAT made the following comment.

We feel we should point out that it appears to us that were it not for the Data- matching Program

(Assistance and Tax) Act 1990 there is no way in which the Department would have become aware

of [Michael's] income from the Trust. That legislation is probably the reason why we have recently noticed an

increase in the number of Social Security matters involving money held on tmst by or for a person who

claims to be entitled to payments under the [Social Security] Act.

The Department recovered the overpayment by serving a garnishee notice on the tmst.

Case study 5

Alison was a client of the Department for 10 years. She was paid Disability Support Pension. Her husband, Billy, had advised the Department that he had a small income from investments but this did not

affect the amount of Alison’s pension.

When the Department asked Alison to provide her and her husband's tax file numbers, Alison advised the Department

that she would not be providing their tax file numbers and that she no longer wished to receive Disability Support Pension.

Because Alison surrendered her payment after she received a tax file number request, the

Department decided to check that Alison's past entitlement to Disability Support Pension had been correct.

The Department contacted the ATO for information about

16 Data-matching Program

Alison and Billy's income. The ATO indicated that Billy had declared earnings from employment in his last tax return. The Department then contacted Billy's employer to establish the amount, source and period of Billy's income. These

checks confirmed that Billy had been working for most of the time that Alison had been receiving the pension and that the amount of his income was such that Alison had not been entitled to receive Disability

Support Pension.

The Department then interviewed Alison and asked her to explain the discrepancy. She acknowledged that Billy had been working and that they had not advised the Department. The Department raised an overpayment for $47 400. The overpayment is being recovered by cash instalments at the rate of $130 per fortnight. The case was not referred to the DPP because of Alison's poor health.

Full details of the action taken on discrepancies by the Department are contained in chapter 5.

Privacy safeguards

The Data-matching Program operates within a framework of comprehensive and strict privacy safeguards which cover the

collection, storage, use and disclosure of personal information. The Department has been careful to adhere to these requirements. In 1993 the Australian National Audit Office confirmed that the Department has effectively fulfilled its privacy obligations and has responded appropriately to operational matters raised by the Privacy Commissioner. The Privacy Commissioner has also been satisfied with the Department's treatment of privacy concerns. There have been no formal complaints to the Privacy Commissioner about the Department's handling of personal information under the Data-matching Program.

The main safeguards associated with the Program ensure that:

. source agency data is not

held centrally for any longer than is necessary;

. source agencies cannot link or merge the information used in the Program to create a new permanent database of information;

. the source agency data that is used is as up to date as possible;

. the data received and generated by the Data-

The Data-matching Program 17

Matching Agency is protected by strict physical and system security arrangements;

. clients are aware of the existence of the Program;

. clients are contacted only when the discrepancy cannot be explained following an examination

of the Department's records;

. individuals affected by the Program are protected by a number of safeguards, including the destruction of personal information which is no longer needed; and

. tax file number

information is protected.

On the last point, the Department's objective is to have tax file number information in one place only — on its

computer system in an encoded form. Prior to the introduction of the Program, many tax file numbers provided by clients

incidentally when establishing proof of income were placed on clients'paper files. In accordance with the

Department's commitment to the Privacy Commissioner, these tax file numbers are being progressively removed by staff

in the Department's 220 Regional Offices. Because of their efforts during the last twelve months, the number of tax file numbers held on paper files has substantially decreased.

Detailed information about how DSS ensures that it fulfils the above requirements is contained in the 1993 report.

CHAPTER 2

PROGRAM ENHANCEMENTS

Evaluation program

The Department, in consultation with the participating agencies, evaluates the results achieved by the Program on a cycle by cycle basis. The evaluation process is

comprehensive and continuous and aims to ensure that the output from cycles is as refined and accurate as possible.

In the last twelve months the Department has also implemented a formal evaluation program to ensure that each

individual project is analysed in an efficient and structured way. (Projects are the matching routines undertaken to detect various discrepancies — for

example, one project matches clients in receipt of Job Search Allowance against clients in receipt of AUSTUDY.) The aim

of the evaluation program is to analyse past review work to assess the effectiveness of projects. As a result of the program, the Department has made a number of enhancements which include the following.

• The Department has

deleted projects which

address risks which are no longer present. For example, a project which addressed the risk that two people might be receiving basic Family Payment for the same child was discontinued after integration of the Department's databases eliminated that risk.

. Some projects have

proven to be more effective when they are conducted at particular times of the year. As a

result, the Department has decided to run these projects at the relevant time(s) of the year rather than in every data- matching cycle.

. It has made a number of

changes to some projects to ensure that they more accurately reflect income limits for payments.

The changes made as a result of the Department's evaluation program have resulted in a steady improvement in the

capacity of the Program to

20 Data-matching Program

detect incorrect payments. For example, the rate of downward variations detected by the Data- matching Program has increased from 22.7 per cent of completed reviews where the client was notified of the discrepancy in

1992- 93 to 32.7 per cent in 1993- 94. Similarly, the rate of overpayments detected has improved from 29 per cent of completed reviews where the client was notified of the discrepancy in 1992-93 to 41.6 per cent in 1993-94.

Improvements to families end of year processing

The Department intends to use the Data-matching Program to facilitate its 1994 end-of-year families processing. In the past about 300 000 additional Family Payment clients have been required to provide a copy of their Tax Notices of Assessment before payment could continue in the next calendar year. This made it difficult for clients to comply if they had not yet obtained their Tax Notices of Assessment for the last financial year or could not find them. It also resulted in dismption to payments for clients who forgot to provide their Tax Notices of Assessment. The processing of these cases also created a significant workload for staff in

the Department's Regional Office.

In 1994 additional Family Payment clients will be asked to provide a statement of their income, rather than a copy of their Tax Notices of Assessment. About

150 000 basic Family Payment clients with an income of $50 000 or more will also be asked to provide a statement of their income. The client's stated income will be verified electronically by the Data- matching Program and clients

will be selected for review if their combined taxable income as declared to the Australian Taxation Office (ATO) is in excess of the allowable limits. This will ensure that the integrity of the Department's data is not compromised as a result of the streamlined procedure.

Revised strategy for reviewing personal identity discrepancies

Personal identity discrepancies are identified in step 4 of a data- matching cycle. As explained in chapter 1, the purpose of identity matching is to find cases in which people claim a payment to which they are not entitled or evade tax by using an assumed or false identity.

Program enhancements 21

Selections were first issued to the Department's Area Offices for examination in December 1991. Experience with these selections indicated that the vast majority of these cases do not involve fraudulent behaviour and are explained easily. Analysis indicates that

about 25 per cent of the discrepancies are due to these clients using tax file numbers that belong to another person (for example, a family member). Most of the remaining discrepancies occur because agencies have recorded personal identity information differently — an agency may have spelt a

name incorrectly (for example, 'Mary' has been recorded as 'Mray') or an agency may have recorded a person’s preferred name rather than their actual name (for example, the person's

real name is Fred but he prefers to use the name Bluey).

After considering a number of options, the Department decided on a new approach for correcting the personal identity

data for those clients who had apparently provided the wrong tax file number. The approach adopted involves writing to these clients and advising them that they appear not to have provided the correct tax file number. They are then asked either to provide the correct tax

file number or, if they do not

have a tax file number or do not know what it is, to complete a Tax File Number

Application/Enquiry Form. (The completed form is sent to the ATO which then advises both the Department and the client of the correct tax file number.) Clients who do not respond to the request for the correct tax file number will have their identity and income details investigated thoroughly.

About 23 000 cases were sent to the Department's Area Offices for review in late June 1994. As clients are reviewed and provide their correct tax file numbers, their records are then able to pass through to step 5 of a data-

matching cycle.

The Department is also working with the ATO to determine the most effective strategy for resolving discrepancies which

occur because of the different recording of personal identity information by agencies.

CHAPTER 3

LEGISLATIVE ISSUES

Legislative amendments

Most of the following amendments were contained in the Social Security (Budget and Other Measures) Legislation Amendment Act 1993 which

received Royal Assent on 24 December 1993. There was also an amendment to the Social Security Act 1991 in the Social Security Legislation Amendment Act (No. 2) 1994 which received

Royal Assent on 12 July 1994.

The definition of personal assistance has also been amended a number of times to incorporate new social security

payments such as the Home Child Care Allowance.

Details of partners

In December 1993 the Act was amended to allow the Australian Taxation Office (ATO) to give the Data-Matching Agency the names of any partners of clients

or partners of parents of clients and not just those in respect of whom a spouse rebate is claimed.

The amendment was of particular value to the Department of Employment, Education and Training (DEBT) which uses this information to check that the marital status of clients’ parents is as declared where clients have provided the income information of a single parent only.

Quoting tax file numbers back to the ATO

The Act places limits on the use of tax file numbers within a data-matching cycle and in follow up work. In the past, this

included a prohibition on assistance agencies quoting clients' tax file numbers back to the ATO when following up

discrepancies. Because the tax file number is the ATO's client reference number, this made it very difficult in some cases for

the ATO to identify the correct client when responding to requests from assistance agencies. In December 1993 the

Act was amended to allow assistance agencies to quote clients' tax file numbers back to the ATO. This has enhanced the administration of the Program by

24 Data-matching Program

allowing the ATO to identify the correct clients more easily.

Compromised tax file numbers

A compromised tax file number is one that may have become known to an unauthorised third party. This can occur, for example, when documentation containing a person's tax file number is lost or stolen. In these situations, the ATO gives the person two options. One option is to have a new tax file number allocated to them; the other is to continue to use their existing number. If they choose to keep their existing tax file number, the ATO flags their number as compromised on its computer system. This indicates that there is a possibility that someone other than that person may be using that number.

The Act was amended in December 1993 to allow the ATO to inform the Data- Matching Agency which tax file numbers in a cycle are compromised. This will be implemented from cycle 3/1994. Although only a very small number of clients are affected, this information will allow the Department to conduct its investigations of these clients in a way which reflects the possibility that there may be fraud involved (that is, a client

may be using another person's tax file number or another person may be using a client's tax file number).

Payments made to clients after a notice of proposed action has been sent

Section 11 of the Act provides that clients have up to 28 days in which to explain why action should not be taken against them. It also provides that l payments are to continue during the 28 days.

In accordance with a suggestion by the Department of Finance, the Act was amended in December 1993 to provide explicit authority for continued payment during the period of up to 28 days available to clients to

show cause why their payments should continue.

Change to tax file number requirements

A decision by the Administrative Appeals Tribunal in July 1993 indicated that the tax file number requirement in the Social Security Act 1991 did not apply to people who did not have a tax file number and had no intention of obtaining one. The decision meant that a very small number of people may have been able to avoid the intention of the legislation which was to make

Legislative issues 25

the provision of a tax file number a condition of payment.

The Social Security Act 1991 was amended in July 1994 to ensure that people who do not have a tax file number and have no intention of obtaining one

will not receive payments. The Act now makes it clear where a person has been requested to provide their tax file number, the

decision whether to comply with the request rests with the person, in full knowledge that failure to

satisfy the request will deny them payment. This means that the Secretary may now stop payments to any person who chooses not to apply to the ATO for a tax file number and provide

a written statement of the number to the Department.

Future changes to the Act

The following changes to the Act may be put forward for consideration in the future.

Withdrawal of First Home Owners Scheme data

In the 1990 Budget, the Government announced that the First Home Owners Scheme would end in August 1995.

However, under the legislative provisions governing First Home Owner Scheme payments, no

data will be obtained from existing clients after

August 1994. This means that all personal data held for these clients will become progressively outdated. As it does so, it will become of only limited value to the assistance agencies.

For this reason, the Minister for Housing, Local Government and Human Services has sought the withdrawal of the Department of Housing and Regional Development (DHRD) from the Program. The request is supported by the other agencies involved in the Program. The withdrawal of DHRD from the Program is expected to have a minimal effect on savings for the Program — for example, the Department of Social Security (the greatest user of First Home Owner Scheme data within the Program) estimates that it will lose less than $50 000 a year.

The First Home Owner Scheme data ceased to be used in data- matching cycles from June 1994. A proposal to formalise DHRD's

withdrawal from the Program by removing references to it from the Act is expected to be introduced in the 1994 Spring

Sittings of Parliament.

Improved linking of records

Agency identification numbers are assigned by the assistance agencies to all client records

26 Data-matching Program

and, where applicable, to the records of their partners, parents and children. For the purposes of a data-matching cycle, it is essential that all such numbers are unique. However, often two or more people will share the same agency identification number. For example, a child's record and its parent's record may share the same number.

If the same agency identification number applies to more than one person, it can create problems for the Data-Matching Agency when it tries to link data from the ATO to the correct person. This occurs because the Act provides that the only means of identifying the destination of the ATO data is to use the agency identification number. For example, where both parent and child share the same agency identification number, the Data- Matching Agency must determine whether the ATO income data belongs to the parent or child.

To get around this difficulty, the Privacy Commissioner's Office agreed that the Data-Matching Agency could add two numerals to the end of every agency identification number. The two numerals are in sequence. This means that if, for example, the same agency identification number is shared by two records, the Data-Matching

Agency records the agency identification number with '01' added at the end for the first record and with '02' added at the end for the second record. This allows the Data-Matching Agency to attach the ATO data to the correct record with greater accuracy and efficiency.

A proposal to spell out this practice explicitly in the Act is being considered.

Definition of spouse

An amendment to the Data- matching Program (Assistance and Tax) Act 1990 is proposed so that the definition of 'spouse' will be taken to have the same meaning as 'partner' in the Social Security Act 1991. At present the definition of spouse assumes that the same word is used in all other relevant legislation. Although that was the case when the Data-matching Program (Assistance and Tax) Act 1990 was first enacted, it ceased to be the case when the Social Security Act 1991 commenced in July 1991.

CHAPTER 4

ASSESSMENTS OF THE PROGRAM

During 1993-94 the Department has continued to monitor operation of the Program closely in consultation with the Privacy

Commissioner. Assessments have taken a variety of forms including reports by the Privacy Commissioner and an efficiency

audit by the Australian National Audit Office (ANAO). The assessments, in conjunction with other activities such as staff

training, have been important in ensuring that the Program is operating in accordance with the law and with strict adherence to

privacy safeguards and principles.

Details of early assessments of the Program are contained in the Department's 1993 report.

Ongoing assessments

Parliamentary scrutiny

In accordance with the legislation, the Department and the Data-Matching Agency tabled a third report on progress

with the Program in Parliament on 19 October 1993. This report, along with those of the other participating agencies, was

a factor considered by Parliament when it decided to continue the Program for another two years until January 1996.

Privacy Commissioner

The Department consults regularly with the Privacy Commissioner and his staff on a variety of issues.

The Privacy Commissioner has taken a very active role in monitoring the implementation and operation of the Program.

In 1993-94 his officers visited a total of 51 offices from all participating agencies, 38 of which were Department of

Social Security (DSS) offices. (The visits are listed in the chronology at the end of this report.) In the course of their visits, they report to local management on any matters of

concern. At the end of their visits, they give a full report to managers in the Department's

National Administration. Points raised have included staff not complying fully with the

procedures in some instances and the identification of areas in which a change to the

28 Data-matching Program

procedures would be appropriate. From time to time the Privacy Commissioner provides the Department with other reports on the appropriateness and effectiveness of its data- matching procedures. The Department makes every effort to address quickly any concerns raised.

Monitoring of results

The Department has on-line result recording and management information systems which enable ongoing monitoring of results. The Department produces monthly result reports for management from these systems. These reports enable managers to monitor progress with the Data- matching Program review work. The Privacy Commissioner's Office is provided with a copy of the report each month.

Quality assurance

The Department added the Data- matching Program to its National Quality Assurance Package in 1993. The purpose of the Package is to ensure that staff are conducting reviews correctly and to identify any areas of concern which require attention.

Every three months the Department selects an average

of 20 cases from each Regional Office (about 3 300 cases nationally) for quality assurance checks. The cases selected include both payment and income matching cases.

The first sample of cases for review were released in August 1993. At 30 June 1994 three more samples had been completed. Quality assurance reviews are conducted by senior staff in the Regional or Area Offices using a copy of the original selection printout. Staff record their responses to the survey on a computer-based system. The questions are designed to check that the original review was conducted in accordance with the Data- matching Program (Assistance and Tax) Act 1990 and with other relevant Departmental procedures. The results of the samples are then analysed and action taken to address any areas of concern identified by the sample. A copy of the results is provided to the Privacy Commissioner's Office.

Australian National Audit Office efficiency audit

The ANAO conducted a full efficiency audit of the Program from June to September 1993. The Auditor-General's final report, entitled Department o f

Assessments of the Program 29

Social Security — Data- matching, was tabled in Parliament in October 1993.

The AN AO’s report indicated that the Program is a 'valuable compliance and control technique' which addresses areas

of risk that other review activities do not cover as effectively. It noted that the savings originally estimated had

not been achieved. The ANAO also confirmed that the Department and the Data- Matching Agency effectively

fulfil their obligations under the Act and other privacy guidelines.

The ANAO stated that there were a number of 'significant strengths in the Department's management of its

responsibilities under the Act'. It also suggested some areas in which the Department could improve its performance. The

Department provides a report on its progress with the ANAO's recommendations to the Minister for Finance on a quarterly basis.

The following comments explain the current status of each of the recommendations. For ease of reference, the ANAO's

recommendation numbers are noted below each recommendation and these numbers are also used within the

text.

A cost-benefit analysis should be included in the reports to Parliament (ANAO Recommendation 1)

Both the 1993 report and this report contain a cost-benefit analysis for the Program.

Costs incurred by the Data- Matching Agency and DSS to match data under the Program should be identified and charged to

participating agencies (ANAO Recommendation 2)

The Department has had preliminary discussions with the Department of Finance about cost recovery and raised this

issue with the participating agencies.

Work to date has focused on identifying sources of costs and methods of apportioning those costs between agencies. The

Department has also examined the appropriateness of granting the Program an exemption from cost recovery arrangements.

The question of an exemption is a complex one and discussions with the Department of Finance are in progress.

Individual project costs should be recorded and reported (ANAO Recommendation 3)

The Department disagreed that recording individual project costs would help identify cost- effective projects. There are

30 Data-matching Program

about 80 different projects and the relative success of each can already be assessed in terms of cancellations, downward variations and overpayments. Each individual project is also analysed as part of the Department's formal evaluation program (see chapter 2) and projects are deleted, run less often or amended as a result of this analysis. The Department also considers that the work required for an accurate attribution of costs and benefits to individual projects would not be cost-effective.

Explanatory narrative should be included when reporting savings estimates (ANAO Recommendation 4)

Both the 1993 report and this report contain explanatory narrative about the rationale and methodology used for calculating savings estimates.

A discount factor should be incorporated into the savings estimates (ANAO Recommendation 5)

The savings estimates reported in both the 1993 report and this report incorporate a discounting factor to reflect the expected reduction in the value of future identified overpayments.

Confirm that the carry-over factors used to project forward savings are still valid (ANAO Recommendation 6)

A departmental study in 1992 confirmed the validity of claiming savings over 26 fortnights in Job Search, Newstart and Sickness Allowance and Special Benefit payments.

The decision to claim savings for 52 fortnights in pension and family payments was based on a consideration of factors likely to preclude clients from these payments. In fact 52 fortnights may be conservative because, on average, pension and family clients remain in pay for longer durations than other client groups. This approach was agreed with the Department of Finance.

The carry-over factors are also discussed in chapter 6.

A more considered estimate of the deterrent effect of the Program should be included in the cost- benefit analyses (ANAO Recommendation 7)

Both the 1993 report and this report include an estimate of the deterrent effect of the Program and outline the basis for calculating the estimate.

Assessments of the Program 31

Extent to which original estimates have been reconsidered or revised should be reported and performance benchmarks should be

established (ANAO Recommendation 8)

Full details of the reasons why the original estimates were not achieved were contained in the 1993 report. This report

concentrates on the capacity of the Program in the current climate to achieve the revised estimates made for it.

The Department rejected the second part of this recommendation which relates to the setting of performance benchmarks for the Program. The Department makes

decisions about the effectiveness of the projects it mns in the Program drawing on ongoing analysis and feedback received from its Area and Regional Offices. This has proven to be an effective means of monitoring the efficiency of the Program. In

addition, the assessment of the value of particular projects and the Program as a whole is too complex a matter to be achieved

through the application of benchmarks. This is because benefits from the Program range from the clear-cut savings from

recovered overpayments and reduced or cancelled payments to less tangible savings from voluntary compliance and knowledge about the relative

vulnerability of the Department to certain categories of risk. Also the ANAO acknowledged the limitations of benchmarks

derived from comparative analysis of various forms of compliance reviews.

Priority should be given to reducing the variability of performance between offices (ANAO Recommendation 9)

The Department has analysed review performance to determine reasons for variations in results between its Area and

Regional Offices. The Department has found insufficient internal differences to explain the variations

adequately. Rather, the larger part of the differences in results appears to be due to external

factors such as the state of the local labour market and income levels. However, the Department has continued to monitor results very closely.

Because there are still significant variations between some offices, the Department intends to undertake further analysis to

ensure that its initial conclusion (that is, that the differences are primarily due to external factors) is correct.

32 Data-matching Program

The project management process should be more formal and project- specific and should include the development of implementation plans and performance benchmarks for individual projects (ANAO Recommendation 10)

The Department agreed that the management process should be more project specific. The formal evaluation program, which was implemented in 1994, is an integral part of that process (see ANAO Recommendation 12 for further details).

However, the Department disagreed with the remaining parts of the recommendation. The Department does not consider it practical or necessary to establish comprehensive project implementation plans for each of the 70 to 80 projects typically included in a data- matching cycle. To do so would require additional staffing resources to develop and maintain such plans and it is not clear that the work would assist decisions on the refinement of projects. The reasons for rejecting the setting of performance benchmarks are set out under ANAO Recommendation 8.

Identification of emerging risk areas should be done in wider consultation with external agencies, possibly by forming an advisory committee comprising members from the Australian Federal Police, the Australian Institute of Criminology, the Australian Council of Social Services, relevant academics, etc

(ANAO Recommendation 11)

The Department disagreed with this recommendation. This was because the identification of risks requires intimate knowledge of the social security system and eligibility criteria as well as the systems used to deliver payments. The Department already maintains links with its international counterparts. It also has a memorandum of understanding with the Australian Federal Police and well-developed liaison arrangements with that organisation. The Department also maintains an active dialogue with welfare and social research bodies.

A detailed review of individual projects should be undertaken (ANAO Recommendation 12)

As discussed in chapter 2, the Department has implemented a formal evaluation program to ensure that each individual project is analysed in an efficient and stmctured way. The aim of the evaluation program is to refine selections further so that

Assessments of the Program 33

incorrect payments are targeted more effectively and clients are contacted only when all simple explanations for discrepancies

have been eliminated.

Refinement of selections should be done as part of the processing during data-matching cycles (ANAO Recommendation 13)

The Department agrees that the refinement of selections should, as far as possible, be done by the Data-Matching Agency as part of the data-matching cycle,

rather than by the Department upon receipt of the raw output from the Data-Matching Agency. One exception is the

removal of reselections from the raw output which can only be done by the receiving agency.

Since the audit, the Department has, to a large extent, incorporated the refinements into

data-matching cycles. The Department will continue to look for opportunities for building further refinements into the

processing.

A more focused national approach should be applied to the analysis of deletion practices (ANAO Recommendation 14)

Area and Regional Office staff may decide to delete a review because:

• the match relates to

two separate people;

• a review was recently

conducted for similar reasons; or

• the match should have

been sent to another agency.

The number of deletions has decreased markedly over time. In 1992-93, nine per cent of selections released were deleted.

In 1993-94, only three per cent of selections released were deleted. This is a direct result of two processes.

. The reselection mles were changed so that a person who is selected for review by the Data-matching

Program cannot be selected for review for the same reason during the next twelve months.

Previously a person could be selected for review for the same reason up to three times a year. This

occurred if the earlier review had not been completed and the client's record had not yet been

corrected.

. The continual evaluation of projects to identify ways to improve the quality of selections

34 Data-matching Program

released for review has also contributed to the decline in the number of reviews deleted.

The Department should examine the implications of the high proportion of notices of proposed action which are successfully disputed for certain projects (ANAO Recommendation 15)

As discussed in chapter 1, staff conducting Data-matching Program reviews make preliminary checks then send a notice of proposed action to the client. Clients have 28 days to show cause why the proposed action should not be taken. Some clients may dispute the accuracy of the information contained in the notice.

The Department examined the reasons for variation between projects in the proportion of notices of proposed action which are successfully disputed. Its analysis showed that there was considerable variation between offices in what was recorded as a dispute. As a result, the Department formulated a more precise definition. This definition was included in a revised version of the Data- matching Program Clerical Procedures Handbook released

in March 1994 and was also brought to the attention of review staff. Despite this, the Department believes that some

degree of variation between projects is to be expected, given the different risks which each project addresses. The Department will continue to monitor the number of disputes as part of its normal management practices.

Documentation should be sighted to confirm that the tax file number information supplied belongs to the client

(ANAO Recommendation 16)

The Department rejected this specific recommendation but agreed that an alternative strategy for handling personal identity discrepancies was needed. The Department's new approach to these cases is discussed in detail in chapter 2.

CHAPTER 5

ACTION TAKEN ON DISCREPANCIES

This chapter contains the statistical details required by guideline 12(v) of the Schedule to the Data-matching Program

(Assistance and Tax) Act 1990. The information is divided into two main parts. The first part, which has been compiled for the Data-Matching Agency, details the output from data-matching

cycles. The second part details the action taken on discrepancies by the Department

of Social Security (DSS).

Data-Matching Agency input and output

The legislation requires that the following details of the information processed by the Data-Matching Agency be

provided:

. the total number of

matches undertaken;

. the proportion of matches that result in discrepancies; and

. the number of

discrepancies.

'Discrepancy' is defined in guideline 2.2 of the Schedule to the Act. It refers to a result of the Program which warrants further action by a source

agency for the purposes of giving effect to the Program. Discrepancies can result from step 1 (invalid tax file numbers),

step 4 (identity matching) or step 5 (payment and income matching).

'Matches undertaken’ is not defined in the Act. For the purpose of this report, 'matches undertaken' refers to the total number of records received by the Data-Matching Agency from

assistance agencies after they have been separated into individual records for clients,

partners, children, parents, maiden names and aliases.

Before commenting on the output from each data-matching cycle, some points of clarification are necessary. A

comparison cannot be drawn between cycles on the numbers in isolation. This is because cycles have varied in terms of:

• the number of tax file

numbers collected;

• the type and volume of

information provided by assistance agencies to the Data-Matching Agency;

• the financial year data

provided by the Australian Taxation Office;

• the business mles used to

determine the types of matching;

• the matching projects run; and

• the times of the year the

cycles are run.

On rare occasions, technical problems during the course of a cycle have also had a bearing on results.

To ensure that each cycle is conducted as effectively as possible, the Data-Matching Agency seeks feedback from source agencies on the results of their examination of discrepancies produced through the Data-matching Program. The Data-Matching Agency undertakes refinement of its matching programs between cycles.

36 Data-matching Program

Table 1 of appendix B indicates that the total number of matches undertaken in the five data- matching cycles in 1993-94 was 72.4 million, comprising 59.3 million matches for DSS and 13.1 million matches for other assistance agencies (the Department of Veterans' Affairs, the Department of Employment Education and Training and the Department of Housing and Regional Development). These matches produced a total of 2.8 million cases in raw output, representing 3.8 per cent of all matches. Output for DSS totalled 1.9 million cases (3.3 per cent of DSS matches), while that for other assistance agencies totalled 0.8 million cases (6.2 per cent of other agency matches). It should be noted that the number of matches does not correlate directly with the number of clients affected because some of the matches are duplicate records (for example, a discrepancy involving a client with both a married and maiden name recorded will be ouput as two matches, once for each name). Subsequent refinement by the relevant agency ensures that only one selection for each client is released for review.

The Data-Matching Agency's input and output for each assistance agency for 1993-94 can be found in appendix B.

Action taken on discrepancies 37

Action taken on discrepancies by DSS

The legislation requires the following details of the action taken on discrepancies to be reported by each source agency.

According to guideline 2.2 of the Schedule to the Act, 'action' refers to those activities set out in section 10 of the Act. They

are as follows:

. the cancellation or suspension of personal assistance being given to a person;

• the rejection of a claim for

personal assistance;

. a change to the rate or

amount of personal assistance being given to, or claimed by, a person;

. the recovery of an

overpayment of personal assistance;

• the investigation of the possible commission of an offence;

. informing a person that they may be entitled to personal assistance or granting a claim for

personal assistance; or

• correcting the personal identity data on a client.

The last two types of action are somewhat different in character to the first five types of action listed above. For this reason, they are dealt with separately at page 41. The following section refers only to the first five types of action.

The proportion of discrepancies which resulted in action being taken

Preliminary action taken on discrepancies

In 1993-94, the Department received a total of 1 628 111 selections as raw output from the Data-Matching

Agency. (This figure excludes output to be examined to determine cases in which additional Family Payment may

be payable).

Of these, 69 393 selections were released to the Department's Area and Regional Offices for review. There are a number of

reasons why the remaining 1 558 718 selections were not released for review:

• Most of the personal

identity discrepancies identified in 1993-94 were withheld from release pending the

38 Data-matching Program

development of a new strategy for resolving these discrepancies (see chapter 2). These selections accounted for about 88 per cent of the output not released for review.

While most of the refinement of selections is done by the Data- Matching Agency as part of the data-matching cycle, some refinements can only be done by the Department upon receipt of the raw output from the Data-Matching Agency. This refinement accounted for about seven per cent of the output not released for review.

Some of the output duplicated that produced in previous cycles and it was not appropriate for it to be released again for review (this occurs when reviews have not been completed before the start of a new cycle). Duplicate selections accounted for about four per cent of the output not released for review.

Some of the output were duplicate records. Duplicate records accounted for about

one per cent of the output not released for review.

Upon receipt of the selections the Area or Regional Office staff check client records and, if necessary, confirm the information with source agencies. In 1993-94, staff decided, as a result of these checks, not to contact the client in 21 846 cases (31.5 per cent). In 20 005 of these cases, this was because the checks indicated that:

• the client had provided

information about the change in their circumstances (where the information was received after the start of the relevant data-matching cycle);

. the client had been

recently reviewed (where the client's payment was corrected after the start of the relevant data-matching cycle); or

• the match was not a valid

match (that is, the information related to two different people).

In the remaining 1 841 cases, the client was not sent a notice of proposed action because it might have prejudiced the effectiveness of the

Action taken on discrepancies 39

investigation. This exception to the notification requirement is permitted under section 11 (4) of the Act and is intended to be

used only when staff suspect that the client is engaged in fraudulent activity. However, the Department's analysis

indicates that this figure is overstated due to coding errors and that the actual number of cases where the issue of a letter

might have prejudiced the effectiveness of the investigation is likely to be only a small subset of these cases. The

Department has taken steps to ensure that the situation is remedied. The results of the reviews for this group are

discussed below.

A total of 20 144 cases (29 per cent) had been completed following the issue of a notice of proposed action. There were a further 637 cases (0.9 per cent) where action had been taken as a result of a review but the result system did not show that a letter had been sent or that a letter had not been sent because it might have prejudiced the effectiveness of the

investigation. A system change in August 1992 was intended to remedy this problem by requiring staff to indicate that a

letter has been sent (or that an exemption from sending a letter applied) in order to record a result. It appears that this

control has not worked for these cases and the Department is investigating the cause with a view to fixing the problem.

There were also 26 766 cases (38.6 per cent) which were awaiting review or had not been finalised.

Results of completed reviews

Table 1 of appendix C provides details of the results of completed reviews by type of discrepancy for 1993-94. It

indicates that the most successful components of the Program are payment and income matching, with the largest proportion of the savings due to income matching.

Table 1.1 shows the results for the 20 144 reviews where the client was notified of the discrepancy and their eligibility and entitlement was checked. It

shows that these reviews resulted in the stopping or downward variation of payments to 6 596 clients (32.7 per cent).

Overpayments were identified in 8 389 cases (41.6 per cent).

Table 1.2 shows the results for the 1 841 reviews where a notice of proposed action was not sent because it might prejudice the effectiveness of the investigation. As noted above, the Department believes that this

40 Data-matching Program

figure is inflated due to coding errors. The table shows that these reviews resulted in the stopping or downward variation in payments to 241 clients (13.1 percent). Overpayments were identified in 148 cases (8 per cent).

Table 1.3 shows the results for the 637 cases where action had been taken as a result of a review but the result system did not show that a letter had been sent or that a letter had not been sent because it might have prejudiced the effectiveness of the investigation. It shows that these reviews resulted in the stopping or downward variation of payments to seven clients (1.1 per cent). Overpayments were identified in 636 cases (99.8 per cent).

Overall, 10 388 of the cases completed (45.9 per cent) resulted in the stopping or downward variation of payments and/or the identification of an overpayment. In the remaining

12 234 of the cases completed (54.1 per cent), no action was taken because:

• the client provided

additional information which confirmed that their existing rate of payment was correct; or

. the client had information which needed to be corrected such as name or date of birth or tax file number which did not

lead to any variation to their rate of payment.

Table 2 of appendix C shows the results of all completed reviews by payment type. It indicates that the Program is most effective at reviewing Family Payment clients. Table 2.1 shows the results for those reviews where a notice of proposed action was sent. Table 2.2 shows the results for those reviews where a notice of proposed action was not sent because it might prejudice the effectiveness of the investigation. Table 2.3 shows the results for those reviews where action had been taken as a result of a review but the result system did not show that a letter had been sent or that a letter had not been sent because it might have prejudiced the effectiveness of the investigation.

Upward variations

Sometimes Data-matching Program reviews bring to notice clients who have been receiving less than their full entitlement. In 1993-94, payments to 247 clients were varied upward, amounting to $81 510 in outlays.

Action taken on discrepancies 41

The Department also identified 49 clients who had, at some time in the past, been underpaid to the value of $25 558 in outlays.

Challenges to the accuracy of the data

Of the 20 144 clients notified of a discrepancy in 1993-94, 1 429 clients (7.1 per cent) challenged the accuracy of the

information on which the Department based its proposed action. In 590 of these cases (41.3 per cent), the Department proceeded with action despite the challenge. In most cases

clients challenged the information based on a lack of understanding of the conditions of eligibility for payment.

The proportion of discrepancies which resulted in people being informed that they may be entitled to personal assistance or being granted a claim for personal assistance

Clients identified by the Data- matching Program as possibly eligible to receive additional Family Payment are sent an

invitation to claim with a simplified claim form on the back. The wording of the invitation to apply was

developed in consultation with the Privacy Commissioner.

The clients' records are flagged when an invitation to claim additional Family Payment has

been issued. This prevents clients from being selected again on the basis of later matches conducted during the same calendar year.

In 1993-94, about 46 300 clients were sent invitations. Due to technical problems, information on the number of people who

lodged claims for, or were granted, additional Family Payment is not currently available.

The proportion of discrepancies which resulted in the correction of personal identity data

The Department is able to use the Program to correct the personal information it holds on its clients where it is clearly

incorrect. For example, a discrepancy may indicate, among other things, that the client has notified a change of

address to another agency but not to DSS. The Department is then able to correct its record for that client. This gives the

Department's client information data-base greater integrity. It is also consistent with the Privacy Act 1988 which requires

data held by an agency to be accurate, relevant, up-to-date, complete and not misleading.

42 Data-matching Program

Using the Program to correct client information is treated as part of general Departmental transactions. As a result, separate statistical data on such transactions generated by the Program is not available.

The number of instances of subsequent action being taken

'Subsequent action' is not defined in the legislation. For the purposes of this report it is taken to refer to:

• the number of appeals lodged against action taken;

. the number of

prosecutions undertaken; and

. the investigation of multiple frauds.

Number of appeals lodged against action taken

Social security clients dissatisfied with departmental decisions can use a comprehensive system of review and appeal. Clients who are dissatisfied with a decision are encouraged in the first instance to discuss the matter with the original decision-maker in the Regional Office. Where the decision is not changed, the client may ask for the matter to

be reviewed by an Authorised Review Officer. Authorised Review Officers provide a quick, fair and informal process for reviewing disputed decisions.

Where a client remains dissatisfied, a further right of appeal exists to the Social Security Appeals Tribunal, the Administrative Appeals Tribunal and, on a point of law, the Federal and High Courts.

The Department's appeals system was modified in January 1993 to facilitate monitoring of appeal activity. However, due to coding errors, this modification has, in practice, proved to be of little value. It is therefore not possible to be precise about the number of appeals finalised by the Authorised Review Officers, Social Security Appeals Tribunal and the Administrative Appeals Tribunal. The Department has examined a sample of appeals and, based on the findings from this analysis, estimates that fewer than 100 appeals associated with the Data- matching Program were finalised in 1993-94. This suggests that the Data-matching Program generates few appeals.

Action taken on discrepancies 43

Number of prosecutions undertaken

Reviews of entitlement by the Department may bring to notice cases where an offence has been committed against either the Social Security Act 1991 or the

Crimes Act 1914. The Department's role in the prosecution process is to investigate cases in which it

appears an offence has been committed and to forward these cases to the Director of Public Prosecutions (DPP) for a decision on whether prosecution action should proceed. (If a case is sufficiently serious it is referred to the Australian Federal Police for investigation.)

In 1993-94, the DPP prosecuted 242 cases identified through the Data-matching Program. The Department believes that this figure is inflated by the inclusion

of prosecutions as a result of other data-matching activities undertaken by the Department.

Investigation of multiple frauds

No identity frauds were detected by the Program in 1993-94. This is because the vast majority of personal identity

discrepancies do not involve fraudulent behaviour and result in the updating of personal records only. (In most cases,

clients are either mistakenly

using tax file numbers that belong to another family member or the agencies have recorded personal identity

information differently.) The Department's new strategy for reviewing personal identity discrepancies, which is discussed in chapter 2, should help identify more easily those

cases which require identity and income details to be investigated thoroughly.

However, the Department does have strategies in place outside the Data-matching Program for

detecting staff or clients involved in identity fraud. This work involves analysis of the client data held on departmental

systems and the staff transactions made. Enhanced efforts in 1993-94 and the

introduction of new techniques as announced in the 1993 Budget demonstrate the Department's commitment in this

area. These efforts are proving successful. In 1993-94, the Department identified 45 payment frauds amounting to more than $3 million. These

include seven internal cases which were prosecuted successfully. A further three cases of internal fraud were awaiting court action.

44 Data-matching Program

The number of cases where successful recovery action occurred

The Department recovers the majority of debts in two ways.

. If the debtor has a

continuing entitlement to a social security payment, it is normal practice for an amount to be withheld from the payment to recover the debt.

. If a debtor is no longer in

receipt of a payment from the Department, staff negotiate with the debtor to determine a form of repayment (such as regular cash payments) that will ensure that the money overpaid is

recovered as promptly as possible.

It is important to point out that recovery of a debt can take place over a period of years and that the number and value of debts raised in a year does not necessarily correspond with the number and value of recoveries.

In 1993-94, recovery action was finalised in 1 023 cases and is continuing in 1 694 cases.

CHAPTER 6

SAVINGS IN OUTLAYS AND OTHER BENEFITS

Variations to original estimates

In its 1993 report the Australian National Audit Office (ANAO) noted that savings achieved have been substantially less than what

was originally estimated in 1989-90. It is important to realise that the Data-matching Program operates within a dynamic environment. Many factors on which the original estimates were based have

changed — economic conditions, the better than expected performance of other complementary control measures

and the numbers of people who left the social security system at the time of the 1990 Social Security Amnesty and the tax file number collection. Readers

who have an historical interest in the variations to original estimates should refer to the detailed discussion in the

Department's 1993 report.

The Data-matching Program is an important compliance measure which achieves direct savings bolstered considerably

by the indirect savings associated with an increased level of voluntary compliance with the social security system.

Indirect savings — voluntary compliance

A detailed analysis of the level of voluntary compliance generated by the tax file number requirements and the Data- matching Program is contained

in the Department's 1993 report. The main points are repeated here because the effect of these measures appears to be

substantial.

Encouraging voluntary compliance with the law is an essential part of minimising misuse of the social security

system in a cost-effective way. It results in a reduction of incorrect payments without the use of compliance and control resources and at no cost to the taxpayer.

In the social security system, voluntary compliance can be considered in three categories.

46 Data-matching Program

• In category 1 voluntary

compliance is evident when people are deterred from attempting to claim payments to which they

are not entitled.

• Category 2 includes the

voluntary surrender of payments to which people may not be entitled.

. Category 3 includes the voluntary disclosure to the Department of changes in circumstances which affect rates of payment (that is, without the prompt of a review or other contact by the Department).

As indicated above the tax file number requirement and community awareness of the Data-matching Program is likely to have contributed to the level of voluntary compliance with the social security system.

The Department raised community awareness of these measures through extensive publicity and letters to all social security clients who were required to provide tax file numbers. In 1993-94 the Department continued its efforts to raise the level of awareness about data-matching and tax file numbers. An insert explaining the activities the Department

undertakes to detect incorrect payments and fraud was included in mailouts to all Job Search and Newstart Allowance recipients over a fortnight in June 1994. The Department plans to issue similar information in planned mailouts to Family Payment clients from September 1994. Details of the publicity undertaken are contained in the chronology which can be found at the end of this report and the earlier publicity was discussed in the

1993 report.

Savings resulting from people deciding not to claim, or to voluntarily surrender, payments to which they are not entitled (categories 1 and 2) are difficult to quantify because they require assumptions to be made about behavioural responses. Nevertheless, it seems reasonable to use past experience as a basis for calculating the extent of voluntary compliance.

To date the numbers in these two categories have been evidenced by the cancellation or voluntary surrender of payments as a result of the 1990 Amnesty and the tax file number collection. The people concerned are for the most part those who comply with mles and processes if they believe there is a likelihood of detection. While

Savings in outlays and other benefits 47

the particular individuals may have left the social security system at some point anyway, it is reasonable to assume that they

would have been 'replaced' by others with the same characteristics were it not for the Data-matching Program.

Therefore, the total number of people who 'left' the social security system because of the 1990 Amnesty and the

subsequent tax file number collection can be taken as the number of people who at any one time would have been in the

social security system were it not for the deterrent effect of the Program.

The Government announced the Amnesty in the 1990-91 Budget. It gave people a last chance to advise the Department of a change in their circumstances without penalty before the new requirements to provide tax file numbers and the new Data- matching Program came into effect. The Amnesty resulted in 4 900 people voluntarily

surrendering their payments.

A total of 4 100 people voluntarily surrendered their payments and payments to a further 31 400 people had their payments stopped as a result of the tax file number collection.

Annual expenditure in current terms for the above groups would be about $90 million. This figure indicates an annual

amount which, in addition to direct savings from matching, could be expected to be paid incorrectly to clients if the Data- matching Program did not exist.

In other words, in addition to the direct savings from review activity the indications seem to be that the Data-matching Program generates savings of about $90 million a year from increased voluntary compliance with the social security system.

An Australian National University study, undertaken by Professor R A Weatherley, indicates that the Program has also had a significant impact on the increased disclosure of

changes in circumstances which affect the rate of payment (defined as category 3 earlier). The study found that when

asked if they would disclose changes in their income, people generally said that they would. These people cited the deterrent effect of the Department's control measures, in particular the tax file number requirement.

From analysis undertaken the Department believes that this effect (category 3) resulted in savings of about $9 million in

1992-93. A further $23 million in annual savings resulted from

48 Data-matching Program

the voluntary disclosure of changes in circumstances during the 1990 Amnesty. This suggests that the amount of savings generated by the disclosure of changes in circumstances as a result of the Program is about $32 million a year.

This figure of $32 million is likely to be understated when regard is had to changes which have occurred in the Age

Pension area. An analysis of the deeming initiative — A n Evaluation of the Impact of "Deeming" on Pensioner Investment Behaviour', Social Security Journal, March 1993, pp 43-53 — reported that

savings from increases in declared pensioner interest income were about $117 million a year. The report noted that some part of the increase in declared assets and income by pensioners may have been due to the Data-matching Program and the tax file number collection.

While voluntary compliance is encouraged as the cheapest and best form of protection for the integrity of the social security system, it does reduce the number of incorrect payments which remain to be detected through the Data-matching Program. This means that the Data-matching Program is less

effective in terms of its capacity to capture savings in outlays directly from matching reviews. Nevertheless, the cost in increased program outlays from abolishing the Program would obviously be greater than the savings directly achieved by the Program. This cost may be an amount greater than $120 million a year.

Direct savings — methodology

There are three direct savings components from the Data- matching Program:

• savings related to tax file

number requirements, including:

- payments stopped

following tax file number request mailouts;

- payments

voluntarily surrendered by clients when asked for their tax file numbers;

- overpayments

identified as a result of reviewing the past entitlement of clients whose payments were

Savings in outlays and other benefits 49

stopped for failure to comply with tax file number

requirements;

- overpayments

identified as a result of reviewing the past entitlement of former clients who voluntarily surrendered their payments at the time of the tax file number collection; and

- payments to new

claimants stopped for failure to comply with tax file number

requirements;

. downward variations in rate or the stopping of payments due to Data- matching Program

reviews; and

. overpayments detected due to Data-matching Program reviews.

The Department calculates savings for clients whose payments are stopped or reduced and not restored within

six weeks.

For pensioners and Family Payment clients, the Department

calculates savings on the basis that these clients will not come back into pay or assume the

same rate of payment for 52 fortnights. For clients in receipt of a benefit or allowance, it is assumed that they will not come back into pay or assume the same rate of payment for 26 fortnights. This means, for example, stopping a pension payment in 1993-94 will produce direct savings in that year, in 1994— 95 and also in

1995-96.

The AN AO made a recommendation in its 1993 report on the Program that the Department validate the carryover factors used to project forward savings. The

Department has decided not to proceed with a review at this time because a departmental study in 1992 confirmed the

validity of claiming savings over 26 fortnights in Job Search, Newstart and Sickness Allowance and Special Benefit

payments. The Department of Finance agreed with this approach. It is important to point out that the 52 fortnights

claimed for pensioners and Family Payment clients may be conservative because, on average, these clients remain in

pay for much longer durations than other client groups.

50 Data-matching Program

The savings also assume that on average the Department will recover 70 per cent of the total amount identifed through the Program as overpaid. This

reflects the fact that it is not possible or not cost-effective to recover all moneys overpaid.

Savings that are generated through the tax file number collection and the stopping of payments at the new claim stage for failure to comply with tax file number requirements, are distributed across financial years from the date that payments are stopped.

Actual savings achieved in 1993-94

In 1993-94 the Department saved $47.2 million through the collection of tax file numbers and $32.6 million through the stopping and variation of payments and the identification of overpayments — a total of $79.8 million.

The Department estimated that it would save $84.8 million in 1993-94. Actual savings are six per cent below the estimate. The reason for this reduction lies mainly in the review of the past entitlement of clients whose payments were stopped due to the tax file number collection. The Department estimated it

could save up to $5 million but, because of the developmental work required before the review process could begin,

$1.1 million was saved in 1993- 94. The Department expects to recoup the remaining portion in 1994— 95.

The estimate also included the expected savings from stopping of payments to Family Payment clients as a result of non­ compliance with the final tax file number collection mailout. The

Department estimated that it would stop payments to 8 400 clients but the actual number of payments stopped was about 1 800. This resulted in a reduction in the actual savings for 1993-94 of $1.8 million. The estimates for

1994— 95 and 1995-96 have also been revised to take account of this effect.

Actual savings by social security program for 1993-94 are set out in appendix D, table 1.

Actual savings from tax file number requirements

New claimants

The Department made a change to its computer systems from February 1993 to stop payments to new clients who do not comply with tax file number requirements. The number of

Savings in outlays and other benefits 51

clients whose payments were stopped in 1993-94 by social security program were:

• Program 1 — 61 clients;

. Program 2 — 49 clients;

. Program 3 — 163 clients;

• Program 4 — 570 clients; and

. Program 5 — 135 clients.

Stopping payments to these new clients achieved direct savings in 1993-94 of $3.5 million, $4.1 million in 1994-95 and

$824 000 in 1995-96.

Current clients

The strategy adopted for the collection of tax file numbers resulted in different client groups having their payments stopped at

different times between September 1991 and May 1994.

Details of the numbers of clients in the various social security programs whose payments were stopped and the associated

savings in outlays to June 1993 was provided in the Department's 1993 report. The information below outlines the

action taken in the past 12 months to finalise the tax file number collection and the

payments stopped due to this action.

In February 1994 the Department issued advices to 42 000 Family Payment clients requesting their partners' tax file numbers. In May 1994 the

8 617 clients who had not responded were advised that their Family Payment had been

stopped. Most clients have since complied with the requirement and payments were reinstated without loss. About

1 800 clients remain out of pay and the Department intends to review their past entitlements.

Stopping payments to these clients has resulted in direct savings spread over three years — $314 708 in 1993-94,

$2.1 million in 1994-95 and $1.7 million in 1995-96.

Investigation of previous entitlement of clients whose payments were stopped through the tax file number collection

In 1993-94 the Department continued checking the past entitlement of clients whose payments were stopped for

failure to comply with tax file number requirements. In total 1 267 reviews were conducted resulting in 124 overpayments to the value of $1.5 million

(70 per cent of which is claimed as savings).

52 Data-matching Program

Actual savings from review activity

In 1993-94, 6 844 clients had their rate reduced or their payments stopped. The total reduction was $512 298 a fortnight.

The Department also identified 9 173 clients who had been overpaid. The total amount of overpayments identified was $16 million (70 per cent of this amount is claimed as savings).

Estimated direct savings for 1994-95 and beyond

In its 1993 review of the Program the Department based the savings estimates for review activity on experience with the Program. The 1992-93 results were used as the basis for extrapolating savings for future years. The assumption was that five matching cycles would be conducted each year and that the mix of reviews undertaken would be similar in future years.

A review of the estimates has confirmed that the experience in 1992-93 is still a reasonable basis for projecting future direct savings for the Program although adjustment needs to be made for factors such as the lower than expected number of Family Payment clients whose payments

were stopped for failure to comply with requests for their partners' tax file numbers. This reduced actual savings in

1993- 94, an effect which will flow on through to savings in 1994— 95 and 1995-96 (see page 50). (This effect is

reflected in the estimates for these years.)

The amount of debt being recovered is another area the Department will be examining in 1994—95. A discount of 30 per cent of debts was made for the overpayment component of

savings reported in 1993 and this year. Through its Debt Management Information System the Department will be in a position to make further assessments of the value of overpayments being claimed following the completion of reviews.

It is important to point out, however, that although the estimates reflect the Department's best judgment of

likely outcomes, actual results may be affected by administrative, technical or social factors, the advent and effect of which cannot be forseen when the estimates are produced. For these reasons the

Department will monitor closely achievement towards the estimated savings for 1994—95 and if necessary may seek a

Savings in outlays and other benefits 53

revision of them to take account of any significant changes. At this stage, if the Program continues beyond

23 January 1996, the Department estimates that it will save $58.3 million in 1994— 95, $41.9 million in 1995-96,

$39.3 million in 1996-97 and 1997-98 (see appendix D, table 1).

These estimates comprise:

. savings associated with the tax file number requirements, which are expected to be

$18.5 million in 1994— 95 and $1.7 million in 1995-96;and

. ongoing savings from the operation of the Program, which are expected to be $39.8 million in 1994-95,

$40.2 million in 1995-96, $39.3 million in 1996-97 and 1997-98.

It is important to point out that even if these estimates are not realised the Program appears to be valuable for the level of

voluntary compliance it encourages as discussed on pages 45-8.

Total actual and estimated savings for other agencies

The other participating agencies saved a total of $9 million in 1993-94 and expect to save $6.6 million in 1994— 95,

$4.9 million in 1995-96, $3.9 million in 1996-7 and 1997-98.

The total actual and estimated savings for each agency is set out in appendix D, table 2.

Summary of direct and indirect savings

As outlined above there are significant direct savings in outlays being achieved through the Data-matching Program.

Ongoing savings from matching are estimated to be about $40 million a year.

In addition to this, voluntary compliance with the social security system may be greater than $120 million a year.

Other benefits

There are three other factors which should be considered in any analysis of the value of the Program. They are:

54 Data-matching Program

• enhanced working relationships between agencies;

. the impact of using the

Data-matching Program to identify residual eligibility in the community for additional Family Payment; and

. identification of cases where the rate of payment is lower than it should be.

These other benefits were discussed in more detail in the Department's 1993 report.

CHAPTER 7

OPERATING COSTS

Methodology

The material in this chapter shows the actual direct costs for 1993-94 and the estimates for the current year and beyond

(were the Program to be continued). Readers with an interest in the full cost-benefit information for the Program

since it began should read this material in conjunction with the 1993 report.

The methodology used by the Department for this analysis follows the approach suggested by the Department of Finance in its publications — Guidelines for Costing o f Government

Activities (1991) and Value for your IT Dollar — Guidelines for Cost-Benefit Analysis of Information Technology

Proposals (1993).

The main cost components which have been included are:

. administrative costs;

. salary costs; and

. opportunity costs.

The salary costs include two components:

. direct operating salary; and

. overheads.

Overheads include all other expenditure for undertaking the work associated with the

Program. The Department of Finance figure of 154.4 per cent of direct operating salary has been used to calculate the

overheads. The overhead components are described in detail in the 1993 report.

The Department's Regional Staffing Model was also used to identify the time it takes to undertake a review (including preparation and finalisation time

and the time it takes to delete a review) and this was validated through discussions with network staff involved in this

activity.

Actual reviews undertaken in 1992-93 and 1993-94 were used for the workload analysis and work measurement

assessment of review activity.

56 Data-matching Program

Estimates of the number of prosecutions referred to the Director of Public Prosecutions and the number of debts raised were used to cost the network effort for these activities.

Administrative costs

The direct administrative costs fall into the following categories:

• computer and associated costs (the operation of the computer equipment used to run the Data-matching

Program cycles has some ongoing administrative costs associated with computer hardware and software maintenance);

• printing of reports to

Parliament; and

• tax file number collection and associated processing.

In total the administrative costs of the Program were $ 106 400 in 1993-94 and are expected to be $69 600 in 1994-95 and

1995- 96 and $65 100 in 1996- 97 and beyond. The reduction in administrative costs is due to the completion of the tax file number collection.

These costs are set out in appendix D, table 3.

Salary costs

The main salary costs associated with the Program are:

• those associated with the management and support of the Program within the Department's National Administration; and

• work associated with the Regional and Area Office review activity and management and coordination of this activity.

In total the salary costs of the Program were $10.1 million in 1993-94 and are expected to be $11.2 million in 1994— 95 and $9.9 million in 1995-96 and beyond.

Appendix D, table 3, sets out the salary cost components.

National Adm inistration operating costs

The main tasks for National Administration staff are:

• the operation and management of the Data- matching Program (including computer support);

. systems work associated with the Data-matching

Operating costs 57

Program cycles and the distribution of output to the Department's network for review;

. development and support of policy and procedures for undertaking Data- matching Program review work;

. compliance with the privacy and security safeguards associated with the operation of the

Program;

• support of management information systems to monitor and evaluate performance; and

• reporting to the Privacy Commissioner and to Parliament on the operation of the Program.

The salary and overhead cost of these activities was $ 1.7 million in 1993-94 and is expected to $1.5 million in 1994— 95 and in

each forward year.

Network operating costs

The direct network operating costs are the salaries of Area and Regional Office staff involved in work directly associated with the Data-

matching Program including:

• review activity;

• checking and determining review recommendations;

. preparing cases for

referral to the Director of Public Prosecutions;

• raising overpayments;

. undertaking debt recovery action;

. processing tax file number applications;

. restoring payments to new clients whose payments were stopped for non­ compliance with tax file number requirements and who have subsequently

complied; and

. coordination and management of these activities including quality assurance, training and

performance monitoring.

In 1993-94 the cost of undertaking the review work was $8.4 million. In 1994-95 it is expected to be $9.7 million

and $8.4 million in 1995-96 and beyond.

58 Data-matching Program

Assistance to the Privacy Commissioner

The Department provides supplementation to the Privacy Commissioner to facilitate his monitoring of the operation of the Data-matching Program. The ongoing operating cost of this assistance was $180 000 in

1993-94. The cost in 1994-95 and beyond is estimated at $268 000 a year.

Total actual and estimated operating costs

In total the costs associated with the operation of the Data- matching Program were

$10.4 million in 1993-94 and are expected to be $11.5 million in 1994— 95 and $10.2 million in 1995-96 and beyond.

CHAPTER 8

COSTS AND BENEFITS

Opportunity costs

The Department sought no additional funds to operate the Data-matching Program (apart from the allocation in the

1990-91 Budget for the acquisition of computing equipment to undertake the matching). The development and review activity was resourced by forgoing other activities and diverting the available resources to the Program. The opportunity cost

of the Program is therefore the value of the benefits forgone if the resources had been put to the

most likely alternative use.

Review activity

The Data-matching Program plays an important role in screening all client groups and, for the risks it targets, acts as a filter to detect incorrect payments which are not detected by earlier filters.

Given the need to have an adequate screening of all client populations, it is highly likely that resources would have

continued to be deployed on

reviews of Age Pension. As a result the most likely alternative use for the resources devoted to the Data-matching Program

would have been additional Pensioner Entitlement Reviews (PERs). In 1989-90 the Department undertook 440 000 PERs. This fell to about 115 000 in 1990-91 because pensioners received an investment review form as part of the deeming initiative. In

1991-92 the impact of the Data- matching Program was foreshadowed and

138 000 PERs were undertaken. In 1992-93 the number of PERs fell to 72 000 and in 1993-94 82 000 PERs were conducted.

In compliance terms the Data- matching Program is a better targeted mechanism because it is

designed to check pensioners' income and identity on the basis of objective data.

The opportunity cost of undertaking the Data-matching Program reviews has been calculated on the assumption that an additional 200 000 PERs would have been undertaken in a

full year had the Program not

60 Data-matching Program

been in place. The estimated net results of the PERs that would have been undertaken is therefore the opportunity cost.

Last year the Department estimated the additional 200 000 PERs would save

$2.9 million in 1993-94. In the forward years the estimates were $3.7 million in 1994— 95, $3.2 million in 1995-96 and beyond. Based on PER results achieved in 1993-94, an additional 200 000 PERs may have produced savings of

$5.4 million. Extrapolating this to forward years, savings could be even higher. However, as PERs are targeted reviews, an increase in the number of PERs by an additional 200 000 would give a lower saving per review. Because the level of savings from PERs depends very much on the economic climate, particularly interest rates, it is difficult to predict future year savings. This is demonstrated by the increase in savings between 1992-93 and 1993-94. For the purposes of this analysis an opportunity cost of

$7.5 million in each forward year has been included in total costs.

Management and development

The management and development of the Data- matching Program was

undertaken primarily in the Department's Security, Fraud and Control Division in National Administration. The resources allocated to the Program were diverted from various other tasks within National Administration. The resources devoted to managing and coordinating the work in Area and Regional Offices were also diverted from other tasks.

It is difficult to be precise about the alternative uses for these resources. The Department may have used the additional programming capacity to enhance its review result recording systems or to develop other mechanisms for detecting incorrect payment. Alternatively the resources might have been deployed on work not associated with compliance.

Because it is not possible to determine the most likely alternative uses for the National Administration resources, it is also not possible to quantify the benefits which may have flowed from any alternative use. As a result the opportunity cost of deploying the $1.5 million a year the Department spends on managing and coordinating the Data-matching Program has not been calculated for this analysis. However, these costs have been included as part of the costs

Costs and benefits 61

discussed earlier and form part of the final cost-benefit analysis.

Upward variations

Sometimes Data-matching Program reviews bring to notice clients who have been receiving less than their full entitlement.

Although this is considered a benefit to the Program because people are being restored their full entitlement, it is an increase

in outlays. This is not the same as the use of the Program to specifically identify families in need. It is simply an incorrect payment identified through the

normal processing. As a result the amount of upward variations in payment has been added to the total costs of the Program.

DSS total costs

A summary of the total costs of the Program is shown in appendix D, table 4. The total cost of the Data-matching

Program was $16 million in 1993-94 and is expected to be $19.3 million in 1994-95 and $18.1 million in 1995-96 and beyond.

DSS direct cost-benefit summary

When the costs and the benefits (direct savings) for DSS are

brought together the net benefits are significant. In 1993-94 the net benefit of the Program (when costs are subtracted from benefits) was $63.7 million and is expected to be $39 million in

1994- 95, $23.9 million in 1995- 96 and $21.2 million in 1996- 97 and 1997-98.

The ratio of costs to direct benefits for DSS is favourable. In 1993-94 the Program

returned a ratio of 1:5. In 1994-95 the Department expects the ratio to be 1:3 reducing to 1:2 in 1995-96,

1996-97 and 1997-98.

The direct cost-benefit summary for DSS is contained in appendix D, table 5.

Voluntary compliance

While the direct savings are declining, the Program is still worthwhile because of the effect it has on encouraging voluntary

compliance with the social security system.

In chapter 6 the benefits associated with an increase in the level of voluntary compliance brought about by the

introduction of the Program were discussed. The Department suggests that in total voluntary compliance of more than $120 million a year. However, in the interests of

62 Data-matching Program

making the estimate conservative appendix D, table 6, shows the effect of including $90 million from

voluntary compliance into the cost-benefit equation. Care has been taken to avoid double­ counting with the implementation savings. (Under

the methodology agreed with the Department of Finance and outlined in chapter 6, some savings from cancellations

resulting from the tax file number collection are included in the implementation savings up to 1995-96.)

With this conservative recognition of voluntary compliance, in 1995-96, 1996-97 and 1997-98 the ratio of costs to benefits becomes 1:7.

Current prices

For the 1993 report the Australian National Audit Office suggested to the Department that the costs and benefits should be shown in current prices (1993-94). As this report presents costs and benefits from

1993-94 onwards and does not include previous years no adjustment was necessary.

Computer costs amortised

The Australian National Audit Office also suggested that computer costs associated with

implementation be amortised over a period of seven years to account for standard depreciation of the equipment. The Department considers that amortisation is relevant when costing a new proposal but in this case such an adjustment only serves to artificially increase the savings in the early years and decrease savings in later years. Nevertheless for illustrative purposes the effect of amortising these costs was shown in the 1993 report and is also shown in appendix D, table 7.

Cost-benefit per record checked

The Australian National Audit Office in its 1993 review of the Program suggested that the cost per review be calculated. The Department included this calculation in its 1993 report and has revised the calculation discussed below.

In a full year about 60 million records are checked through the Data-matching Program. To do this checking the average cost is about 17 cents a record and the direct benefit is 65 cents a record in later years. This shows that the Data-matching Program is an inexpensive means of checking large client populations to verify entitlements and detect incorrect

Costs and benefits 63

payments. Appendix D, table 8, illustrates this outcome.

Total costs for all agencies

Appendix D, table 9, shows the total costs of the Data-matching Program for all agencies.

Total costs for all agencies, including DSS, were $21.2 million in 1993-94 and are expected to be $23.3 million in 1994-95 and $22 million in

1995-96 and beyond.

Cost-benefit summary for all agencies

When the other agencies costs and benefits are included the value of the Program remains. The cost-benefit of the Program for all agencies is illustrated in

appendix D, table 10.

DATA-MATCHING PROGRAM CYCLE

APPENDIX A

Step 1 Step 2 Step 3 Step 4 Steps

DMA undertakes ATO extracts identity matching identity and incom e details

If discrepant

TFN & agency identification no. to ATO

DMA undertakes paym ent matching

A ssistance agencies

A ssistance ag en cies

DMA undertakes incom e matching

DEET 1

AEC/HIC

1 identity checks

AEC/HIC identity checks

DMA validates TFNs

If invalid

D V A

DSS

DHRD

Step 6

DSS

DVA

DEET

DHRD

I— ATO

DATA-MATCHING AGENCY

INPUT AND OUTPUT

APPENDIX B

Table 1: Data-Matching Agency input and output for all agencies in 1993-94 (a)

Agency Input from

assistance agency

Matches undertaken DMA output (b)

DMA output as percentage of total matches

DSS 59 524 310 59 286 120 1 944 091 3.3

DVA 3 421 200 3 721 514 120 763 3.2

DEBT 9 429 858 9 045 528 590 442 6.5

DHRD 255 371 379 822 20 980 5.5

ATO Not applicable Not applicable 86 670 Not applicable

Total 72 630 739 72 432 984 2762 946 3.8

(a) The statistics refer to the total number of records, not the total number of clients. Some clients have more than one record. For example, clients with both a married and maiden name recorded will have two records.

(b) Due to some programming problems, the number of personal identity discrepancies identified in step 4 of cycle 5/1993 was significantly higher than usual. The output from this step was later destroyed by the agencies. It should also be noted that the output for the ATO in 1993-94 is from two cycles only.

68 Data-matching Program

Table 2: Data-Matching Agency input and output for DSS in 1993-94 (a)

Cycle Input from

number/ assistance

year agency

Matches undertaken

DMA output DMA output as percentage of total matches

3/1993 12 167 238 12 139 911 329 260 (b) 2.7

4/1993 12 091 880 12 066 088 237 972 2.0

5/1993 12 160 564 12 132 645 731 677 6.0

1/1994 11 380 641 11 329 140 276 547 (b) 2.1

2/1994 11 723 987 11 618 336 368 635 (b) 3.2

Total 59 524 310 59 286 120 1 944 091 3.3

(a) The statistics refer to the total number of records, not the total number of clients. Some clients have more than one record. For example, clients with both a married and maiden name recorded will have two records.

(b) Included in these cycles are clients identified as possibly being eligible for additional Family Payment. The numbers for each cycle are as follows: 80 375 cases in cycle 3/1993; 106 290 cases in cycle 1/1994; and 129 315 cases in cycle 2/1994.

Table 3: Data-Matching Agency input and output for DVA in 1993-94(a)

Cycle number/ year

Input from assistance agency

Matches undertaken

DMA output DMA output as percentage of total matches

3/1993 688 716 747 722 9 135 1.2

4/1993 686 325 745 850 7 913 1.1

5/1993 684 448 744 748 81 296 10.9

1/1994 681 933 742 397 12 831 1.7

2/1994 679 778 740 797 9 588 1.3

Total 3 421 200 3 721 514 120 763 3.2

(a) The statislics refer to the total number of records, not the total number of clients. Some clients have more than one record. For example, clients with both a married and maiden name recorded will have two records.

Appendices 69

Table 4: Data-Matching Agency input and output for DEBT in 1993-94(a)

Cycle number/ year

Input from assistance aeency

Matches undertaken DMA output DMA output as percentage of

total matches

3/1993 1 248 762 1 237 328 126 771 10.3

4/1993 1 351 646 1 341 440 121 080 9.0

5/1993 2 599 224 2 244 334 95 122 4.2

1/1994 1 906311 1 901 065 101 806 5.4

2/1994 2 323 915 2 321 361 145 663 6.3

Total 9 429 858 9 045 528 590 442 6.5

(a) The statistics refer to the total number of records, not the total number of clients. Some clients have more than one record. For example, clients with both a married and maiden name recorded will have two records.

Table 5: Data-Matching Agency input and output for DHRD in 1993-94 (a)

Cycle number/ year

Input from assistance aeencv

Matches undertaken

DMA output DMA output as percentage of total matches

3/1993 66 064 98 125 3 304 3.4

4/1993 61006 90 683 2 502 2.8

5/1993 52 862 78 660 11 468 14.6

1/1994 42 737 63 666 2 052 3.2

2/1994 32 702 48 688 1654 3.4

Total 255 371 379 822 20 980 5.5

(a) The statistics refer to the total number of records, not the total number of clients. Some clients have more than one record. For example, clients with both a married and maiden name recorded will have two records.

70 Data-matching Program

Table 6: Data-Matching Agency output for the ATO in 1993-94(a)

Cycle number/ year

DMA output

3/1993 0

4/1993 38 497

5/1993 48 173

1/1994 0

2/1994 0

Total 86 670

(a) The statistics refer to the total number of records, not the total number of clients. Some clients have more than one record. For example, clients with both a married and maiden name recorded will have two records.

ACTION TAKEN ON DISCREPANCIES BY DSS

APPENDIX C

Table 1: Results of completed reviews by type of discrepancy for 1993-94

Type of Reviews Downward variations (including Identified overpayments

discrepancy completed cancellations)

$

Invalid TFN 175 7 4.0

Identity 430 18 4.2

Payment 5 957 1 558 26.2

Income 16 060 5 261 32.8

Total 22 622 6844 30.3

2 247 0 0.0 0

5 505 11 2.6 114 410

204 471 3 147 52.8 4 255 994

300 075 6 015 37.5 11 652 939

512 298_________ 9 173 40.5 16 023 343

tv>

o

Table 1.1: Results of completed reviews where a notice was sent by type of discrepancy for 1993-94

Type of Reviews Downward variations (including Identified overpayments

discrepancy completed cancellations)

where a notice was sent No. No. % $p.f. No. %

Invalid TFN 175 7 4.0 2 247 0 0.0

Identity 424 13 3.1 3 847 5 1.2

Payment 4 945 1428 28.9 175 195 2 636 53.3

Income 14 600 5 148 35.3 284 636 5 748 39.4

Total 20 144 6 596 32.7 465 925 8 389 41.6

1

o

50 488 3 735 129 11 186 094

14 971 711

:hing Program

Table 1.2: Results of completed reviews where a notice was not sent by type of discrepancy for 1993-94

Invalid TFN 0 0 0.0 0 0 0.0

Identity 0 0 0.0 0 0 0.0

Payment 567 128 22.6 28 801 67 11.8

Income 1 274 113 8.9 15 439 81 6.4

o o

166 279 186 737

353 016

Table 1.3: Results of other completed reviews by type of discrepancy for 1993-94 (a)

Type of discrepancy

Other

completed reviews

Downward variations (including cancellations)

Identified overpayments

No. No. % $p.f. No. % $

Invalid TFN 0 0 0.0 0 0 0.0 0

Identity 6 5 83.3 1 658 6 100.0 63 922

Payment 445 2 0.4 475 444 99.8 354 586

Income 186 0 0.0 0 186 100.0 280 108

Total 637 7 1.1 2 133 636 99.8 698 616

(a) These are cases where action has been taken as a result of a review but the result system did not show that a letter had been sent or that a letter had not been sent because it might have prejudiced the effectiveness of the investigation.

Table 2: Results of completed reviews by payment type for 1993-94

Payment Reviews Downward variations (including Identified overpayments

type completed cancellations)

No. No. % $P.f. No. %

Program 1 Age 4 427 733 16.6 53 170 712 16.1

Sub-total 4 427 733 16.6 53 170 712 16.1

Program 2 Disability 1 820 229 12.6 24 639 346 19.0

Sickness 210 46 21.9 11 516 93 44.3

Sub-total 2 030 275 13.5 36 155 439 21.6

Program 3 Job Search 3 484 465 13.3 102 793 1 373 39.4

Newstart 2 970 390 13.1 85 598 1 434 48.3

Sub-total 6 454 855 13.2 188 391 2 807 43.5

Program 4 Families 8 513 4 844 56.9 207 046 4 867 57.2

Sole Parent 949 104 11.0 23 073 262 27.6

Sub-total 9 462 4 948 52.3 230 119 5 129 54.2

Program 5 Special 102 9 8.8 2 081 47 46.1

Widow 147 24 16.3 2 383 39 26.5

Sub-total 249 33 13.3 4 464 86 34.5

Total 22 622 6 844 30.3 512 298 9 173 40.5

1

2 918 628 2 918 628

2 004 766 184 440 2 189 206

1 647 205 4 425 888 6 073 093

3 058 699 1 345 664 4 404 363

122 510 315 543 438 053

16 023 343

Table 2.1: Results of completed reviews where a notice was sent by payment type for 1993-94

Payment type

Reviews

completed where a notice was sent

Downward variations (including cancellations)

Identified overpayments

No. No. % $p.f. No. % $

Program 1 Age 3 737 692 18.5 48 641 629 16.8 2 729 217

Sub-total 3 737 692 18.5 48 641 629 16.8 2 729 217

Program 2 Disability 1 599 222 13.9 22 742 318 19.9 1 859 551

Sickness 174 39 22.4 9 459 76 43.7 157 982

Sub-total 1 773 261 14.7 32 201 394 22.2 2 017 533

Program 3 Job Search 2 852 389 13.6 81 236 1 100 38.9 1 451 220

Newstart 2713 366 13.5 77 512 1 331 49.1 4 194 072

Sub-total 5 565 755 13.6 158 748 2431 43.7 5 645 292

Program 4 Families 8 036 4 761 59.2 202 100 4611 57.4 2 915 395

Sole Parent 818 95 11.6 20 055 246 30.1 1 235 703

Sub-total 8 854 4 856 54.8 222 155 4 857 54.9 4 151 098

Program 5 Special 86 8 9.3 1798 43 50.0 120 043

Widow 129 24 18.6 2 383 35 27.1 308 528

Subtotal 215 32 14.9 4 181 78 36.3 428 571

Total 20 144 6 596 32.7 465 925 8 389 41.6 14 971 711

Table 2.2: Results of completed reviews where a notice was not sent by payment type for 1993-94

Payment type

Reviews completed where a notice was not sent

Downward variations (including cancellations)

Identified overpayments

No. No. % $p.f. No. % $

Program 1 Age 615 37 6.0 3 454 8 1.3 26 425

Sub-total 615 37 6.0 3 454 8 1.3 26 425

Program 2 Disability 199 7 3.5 1 897 6 3.0 63 993

Sickness 23 7 30.4 2 057 4 17.4 12 549

Sub-total 222 14 6.3 3 954 10 4.5 76 542

Program 3 Job Search 382 74 19.4 20 957 23 6.0 28 300

Newstart 168 24 14.3 8 086 14 8.3 58 783

Sub-total 550 98 17.8 29 043 37 6.7 87 083

Program 4 Families 312 83 26.6 4 946 91 29.2 77 496

Sole Parent 116 8 6.9 2 560 2 1.7 85 470

Sub-total 428 91 21.3 7 506 93 21.7 162 966

Program 5 Special 12 1 8.3 283 0 0.0 0

Widow 14 0 0.0 0 0 0.0 0

Sub-total 26 1 3.8 283 0 0.0 0

Total 1 841 241 13.1 44 240 148 8.0 353 016

Table 2.3: Results of other completed reviews by payment type for 1993-94 (a)

Payment Other reviews Downward variations (including Identified overpayments

type completed cancellations)

No. No. % $p.f. No. % $

Program 1 Age 75 4 5.3 1 075 75 100.0 162 986

Sub-total 75 4 5.3 1 075 75 100.0 162 986

Program 2 Disability 22 0 0.0 0 22 100.0 81 222

Sickness 13 0 0.0 0 13 100.0 13 909

Sub-total 35 0 0.0 0 35 100.0 95 131

Program 3 Job Search 250 2 0.8 600 250 100.0 167 685

Newstart 89 0 0.0 0 89 100.0 173 033

Sub-total 339 2 0.6 600 339 100.0 340 719

Program 4 Families 165 0 0.0 0 165 100.0 65 808

Sole Parent 15 1 6.7 458 14 93.3 24 491

Sub-total 180 1 0.6 458 179 99.4 90 299

Program 5 Special 4 0 0.0 0 4 100.0 2 467

Widow 4 0 0.0 0 4 100.0 7 015

Sub-total 8 0 0.0 0 8 100.0 9 482

Total 637 7 1.1 2 133 636 99.8 698 616

(a) These are cases where action has been taken as a result of a review but the result system did not show that a letter had been sent or that a letter had not been sent because it might have prejudiced the effectiveness of the investigation.

APPENDIX D

COSTS AND BENEFITS

Table 1: DSS direct savings by social security program

Program 1993-94

$'000 1994-95 $000 1995-96

$'000 1996-97 $'000 1997-98

$000

Income security for the retired 8 951.5 8 985.2 8 356.7 8 095.7 8 095.7

Income security for people with disabilities and the sick 6 269.6 5 634.2 5 053.7 4911.5 4911.5

Income security for the unemployed 13 440.1 14 427.9 14 427.9 14 427.9 14 427.9

Income security for families with children 47 745.0 27 625.6 12 794.0 10 582.9 10 582.9

Provision for special circumstances 3 345.9 1 615.0 1 301.5 1 232.9 1 232.9

Total 79 752.1 58 287.9 41 933.8 39 250.9 39 250.9

Table 2: All direct agencies' savings

1993-94 $'000 1994-95 $'000

1995-96 $'000 1996-97 $'000

1997-98 $'000

DSS 79 752.1 58 287.9 41 933.8 39 250.9 39 250.9

DVA 3 857.5 3 253.0 1 964.0 1 351.0 1 351.0

DEBT (a) 4 784.0 3 293.3 2 927.4 2 561.5 2 561.5

DHRD 348.5 45.0 - - -

ATO - - - - -

Total 88 742.1 64 879.2 46 825.2 43 163.4 43 163.4

(a) Forward estimates for DEBT show reducing savings for all outyears. The reduction is expected to follow as a consequence of stronger DEET/DSS front end controls to be introduced in January 1995.

80 Data-matching Program

Table 3: DSS direct costs

1993-94 $'000 1994-95 $'000

1995-96 $'000 1996-97 $'000

1997-98 $'000

Administrative costs

Computer and associated costs 65.1 65.1 65.1 65.1 65.1

Reports to Parliament 4.4 4.5 4.5 - -

Tax file number collection and processing 36.9 - - - -

Sub-total 106.4 69.6 69.6 65.1 65.1

Salary and overheads

National Administration 1 724.2 1 491.3 1 462.0 1 462.0 1 462.0

Area and Regional Offices 8 361.0 9 680.2 8 429.4 8 429.4 8 429.4

Sub-total 10085.2 11 171.5 9 891.4 9 891.4 9 891.4

Assistance to the Privacy Commissioner 180.0 268.0 268.0 268.0 268.0

Total 10 371.6 11 509.1 10 229.0 10 224.5 10224.5

Table 4: DSS total costs

1993-94 1994-95 1995-96 1996-97 1997-98 $'000 $'000 $'000 $'000 $'000

Operating costs 10 371.6 11 509.1 10 229.0 10 224.5 10 224.5

Opportunity costs 5 440.1 7 500.0 7 500.0 7 500.0 7 500.0

Upward variations and underpayments 204.4 288.7 343.4 343.4 343.4

Total 16 016.1 19 297.8 18 072.4 18 067.9 18 067.9

Appendices 81

Table 5: DSS direct cost-benefit summary

1993-94 $'000 1994-95 $'000

1995-96 $'000 1996-97 $'000

1997-98 $'000

Benefits 79 752.1 58 287.9 41 933.8 39 250.9 39 250.9

Costs 16 016.1 19 297.8 18 072.4 18 067.9 18 067.9

Net benefits (a) 63 736.0 38 990.1 23 861.4 21 183.0 21 183.0

Cost-benefit ratio (b) 1 :4.98 1 : 3.02 1 :2.32 1 :2.17 1 : 2.17

(a) Calculated by subtracting costs from benefits.

(b) Calculated by dividing benefits by costs.

Table 6: DSS cost-benefit summary (including savings associated with voluntary compliance (a))

1993-94 $'000 1994-95 $'000

1995-96 $'000 1996-97 $'000

1997-98 $'000

Benefits Direct savings Voluntary compliance Total

79 752.1 42 875.7 122 627.8

58 287.9 71 500.0 129 787.9

41 933.8 88 300.0 130 233.8

39 250.9 90 000.0 129 250.9

39 250.9 90 000.0 129 250.9

Costs 16 016.1 19 297.8 18 072.4 18 067.9 18 067.9

Net benefits 106 611.7 110 490.1 112 161.4 111 183.0 111 183.0

Cost-benefit ratio 1 :7.66 1 :6.73 1 :7.21 1 :7.15 1 : 7.15

(a) This summary includes an estimate of the voluntary compliance effect discussed in chapter 6.

82 Data-matching Program

Table 7: DSS cost-benefit summary (including savings associated with voluntary compliance with computer costs amortised (a))

1993-94 $'000 1994-95 $'000

1995-96 $'000 1996-97 $'000

1997-98 $'000

Benefits 122 627.8 129 787.9 130 233.8 129 250.9 129 250.9

Costs 17 275.1 20 556.8 19 331.4 18 072.1 18 068.6

Net benefits 105 352.7 109 231.1 110 902.4 111 178.8 111 182.3

Cost-benefit ratio 1:7.1 1 : 6.31 1 :6.74 1 :7.15 1 :7.15

(a) The computer cost component of total costs has been amortised across financial years.

Table 8: DSS cost-benefit per record checked

1993-94 1994-95 1995-96 1996-97 1997-98

Costs ($'000) 10 371.6 11 509.1 10 229.0 10 224.5 10 224.5

Benefits ($’000) 36 250.2 35 819.4 39 803.7 39 250.9 39 250.9

Records checked (Ό00) (a) 60 000.0 60 000.0 60 000.0 60 000.0 60 000.0

Cost per record checked ($) (b) 0.17 0.19 0.17 0.17 0.17

Benefit per record checked ($) (c) 0.60 0.60 0.66 0.65 0.65

(a) Refers to the estimated number of records that will be checked electronically in the data-matching process.

(b) Calculated by dividing total costs by the number of records checked.

(c) Calculated by dividing total benefits by the number of records checked.

Appendices 83

Table 9: All agencies' costs

1993-94 $'000 1994-95 $'000

1995-96 $'000 1996-97 $'000

1997-98 $'000

DSS 16 016.1 19 297.8 18 072.4 18 067.9 18 067.9

DVA 1 243.4 240.0 220.0 220.0 220.0

DEBT (a) 630.7 567.6 536.1 473.0 473.0

DHRD 64.0 52.0 - - -

ATO 3 234.3 3 182.3 3 182.3 3 182.3 3 182.3

Total 21 188.5 23 339.7 22 010.8 21 943.2 21 943.2

(a) Forward estimates for DEBT show reducing costs for all outyears. The reduction is expected to follow as a consequence of stronger DEET/DSS front end controls to be introduced in January 1995.

Table 10: All agencies' cost-benefit summary (including savings associated with voluntary compliance with DSS computer costs amortised)

1993-94 $'000 1994-95 $'000

1995-96 $'000 1996-97 $'000

1997-98 $'000

Costs DSS Other agencies Total

17 275.1 5 172.4 22 447.5

20 556.8 4 041.9 24 598.7

19 331.4 3 938.4 23 269.8

18 072.1 3 875.3 21 947.4

18 068.6 3 875.3 21 943.9

Benefits DSS Other agencies Total

122 627.8 8 990.0 131 617.8

129 787.9 6 591.3 136 379.2

130 233.8 4 891.4 135 125.2

129 250.9 3 912.5 133 163.4

129 250.9 3 912.5 133 163.4

Net benefits 109 170.3 111 780.5 111 855.4 111 216.0 111 219.5

Cost-benefit ratio 1 : 5.86 1 : 5.54 1 : 5.81 1 : 6.07 1 : 6.07

GLOSSARY

Action

Action is defined in guideline 2.2 of the Schedule to the Data-matching Program (Assistance and Tax) Act 1990. It refers to the actions set out in section 10 of the Act. These are:

• the cancellation or suspension of personal assistance being given to a person;

• the rejection of a claim for personal assistance;

• a change to the rate or amount of personal assistance being given to, or claimed by, a person;

• the recovery of an overpayment of personal assistance;

• the investigation of the possible commission of an offence;

• informing a person that they may be entitled to personal assistance;

• granting a claim for personal assistance;

• the correction of personal identity data on clients; or

• in the case of the tax agency, the issue of an assessment or an amended assessment of tax by the Australian Taxation Office.

Assistance agency

The term 'assistance agency' is defined in the Data-matching Program (Assistance and Tax) Act 1990. An assistance agency is a Commonwealth Government agency which provides financial assistance to individuals and which supplies data to the Data-Matching Agency for the purposes of data-matching. Assistance agencies are the Departments of Social Security, Veterans' Affairs, Employment, Education and Training and Housing and Regional Development (First Home Owners Scheme only).

86 Data-matching Program

Business rules

The business rules form a technical document used by the Data-Matching Agency for the comparison and matching of data carried out in step 5 of the data-matching cycle. The rules specify the types of records to be matched, how the records will be matched and the refinement criteria that should be applied. The rules are continually refined to improve the matching process.

Data-Matching Agency

The Data-Matching Agency comprises officers of the Department of Social Security who are responsible for the matching of data under the Data-matching Program (Assistance and Tax) Act 1990.

Data-matching cycle

A matching cycle has six steps. Each step is described in detail in chapter 1. Each cycle takes approximately six weeks to complete.

Discrepancy

Discrepancy refers to a result of a data-matching cycle which warrants further action by any source agency for the purposes of giving effect to the Program. Discrepancies are detected in steps 1 (invalid tax file number), 4 (identity matching) and 5 (payment and income matching) of a data-matching cycle.

Identity matching

Identity matching is done by the Data-Matching Agency as part of step 4 of a data-matching cycle. Identity matching involves the matching of personal identity data provided by the Australian Taxation Office with family identity data provided by the assistance agencies.

Income matching

Income matching is done by the Data-Matching Agency as part of step 5 of a data-matching cycle. It matches the taxable income details obtained from the Australian Taxation Office with income details given by clients to the assistance agencies. The Data-Matching Agency also tests whether income thresholds for payments have been exceeded and not reported to the assistance agencies or a client's taxable income is below the limits set for additional Family Payment.

Glossary 87

Partner

For social security purposes, a partner is a person who is either legally married to, or living in a marriage-like relationship with, a social security client and is not living separately and apart on a permanent basis.

Payment matching

Payment matching is done by the Data-Matching Agency as part of step 5 of a matching cycle. Payment matching involves the matching of family identity data given by assistance agencies to find out if personal assistance is being, or has been, given to, or is being claimed by, people who may not be entitled to it.

Program Protocol

Under the scheduled guidelines to the Data-matching Program (Assistance and Tax) Act 1990, the Data-Matching Agency was required to prepare a Program Protocol. This document outlines the nature and purpose of data-matching. It is available to the public from the Privacy Commissioner on request.

Source agency

The term 'source agency' is defined in the Data-matching Program (Assistance and Tax) Act 1990. A source agency is either the Australian Taxation Office or an assistance agency.

Subsequent action

Subsequent action refers to:

• the number of appeals lodged against action taken;

• the number of prosecutions undertaken; and

• the investigation of multiple frauds.

Tax file number algorithm

Tax file number algorithm refers to a systems routine supplied by the ATO to the assistance agencies which is used to check the validity of the tax file numbers provided by clients.

88 Data-matching Program

Voluntary compliance

Voluntary compliance refers to the compliance with the law by income support recipients and potential claimants which results from knowing that:

• there is a high probability that incorrect payments will be detected;

• they will be required to repay any debt; and

• they may be prosecuted if they attempt by fraud or misrepresentation to obtain payments to which they are not entitled.

This has the effect of people advising changes to their circumstances as they occur, voluntarily surrendering their payment when they are no longer entitled to it or not seeking to claim a payment to which they are not entitled.

CHRONOLOGY

22 August 1990 Budget announcement of the extension of the requirement to provide tax fde numbers and the introduction of a program of data-matching between agencies making assistance payments and the ATO

Social Security Amnesty began and ran to 2 October 1990. It gave people a last chance to notify changes in circumstances without penalty before the tax fde number requirements and the Data-matching Program came into effect

19 October 1990 Senate Standing Committee on Legal and Constitutional Affairs hearing

9 November 1990 Senate Standing Committee on Legal and Constitutional Affairs hearing

16 November 1990 Senate Standing Committee on Legal and Constitutional Affairs hearing

26 November 1990 Senate Standing Committee on Legal and Constitutional Affairs hearing

3 December 1990 Senate Standing Committee on Legal and Constitutional Affairs hearing

28 December 1990 Social Security and Veterans' Affairs Amendment Act (No. 2) 1990 received Royal Assent

9 January 1991 Social Security Legislation Amendment Act 1990 received Royal Assent

23 January 1991 Data-matching Program (Assistance and Tax) Act 1990 received Royal Assent

6 February 1991 Request for tender for new computer equipment issued

20 March 1991 Closing date for tenders

1 April 1991 Program Protocol and draft Technical Standards Report lodged with the Privacy Commissioner

18 April 1991 Cycle 1/1991 commenced on existing computer equipment

May 1991 Advice to age pensioners appeared in Age Pension

News

May 1991 Staff information video on tax file number/data-

matching produced

90 Data-matching Program

18 May 1991 Step 5 of cycle 1/1991 run

24 May 1991 Cycle 1/1991 completed

24 & 28 May 1991 General advertisement announcing tax file number/data-matching appeared in the press

30 May 1991 Advertisement advising sole parent pensioners of

tax file number data collection appeared in the press

June 1991 Successful tenderer (Teradata) for new computer

equipment advised

2 June 1991 Repeat advertisement advising sole parent

pensioners of tax file number data collection appeared in the press

'Announcement' advertisements aired on ethnic and Aboriginal radio

9 June 1991 Repeat 'announcement' advertisements aired on

ethnic and Aboriginal radio

18 June 1991 Teradata equipment received in the Department

22-25 June 1991 Mailout of 'authority' letters (initial request letters) to 74 000 sole parent pensioners

24 June 1991 Repeat advertisement advising sole parent

pensioners of tax file number data collection appeared in the press to coincide with the mailout

27 June 1991 Social Security Legislation Amendment Act (No. 2) 1991 received Royal Assent

1 July 1991 Final Technical Standards Report lodged with the

Privacy Commissioner

1 August 1991 Cycle 2/1991 commenced on existing computer equipment

4-11 August 1991 Advertisement advising invalid pensioners of the tax file number data collection appeared in the press

Advertisement advising age pensioners of the tax file number data collection appeared in the press

14 August 1991 Acceptance testing completed on the new computer equipment

23-29 August 1991 Mailout of 'authority' letters (informal request letters) to 254 000 age, invalid and widow pensioners

Chronology 91

23-29 August 1991 Mailout of formal tax file number request letters to 16 100 sole parent pensioners who did not respond to the ’authority' letter issued in June 1991

31 August 1991 Step 5 of cycle 2/1991 run

6 September 1991 Cycle 2/1991 completed

16 September 1991 Mailout of informal request letters to 1.4 million other pensioners with Consumer Price Index mailout

September 1991 Mailout of stop payment letters to 5 000 sole parent pensioners who did not respond to the formal request letters in August 1991 advising them that payment of their Sole Parent Pensions

would stop

October 1991 Mailout of informal requests to 200 000 additional Family Payment (AFP) clients as part of AFP review

5 November 1991 Data-matching Program — Report on Progress October 1991 tabled in Parliament

November 1991 Mailout of stop payment letters to 200 sole parent pensioners who did not respond to the formal request letters in August 1991 or to the stop payment letters in September 1991 advising them that payment of their basic Family Payment would

also stop

25 November 1991 Cycle 3/1991 commenced on new computer equipment

Social Security Legislation Amendment Act (No. 3) 1991 received Royal Assent

13 December 1991 Social Security Legislation Amendment Act (No. 4) 1991 received Royal Assent

16 December 1991 Step 5 of cycle 3/1991 run

20 December 1991 Cycle 3/1991 completed

December 1991 Mailout of informal requests to 1.13 million basic Family Payment recipients with advice about Consumer Price Index changes and new thresholds for payment

30 January 1992 Cycle 1/1992 commenced

6 February 1992 Tax file numbers encoded on the Department's on-line system of client records

21 February 1992 Step 5 of cycle 1/1992 run

28 February 1992 Cycle 1/1992 completed

92 Data-matching Program

5 March 1992 Cycle 2/1992 commenced

19 March 1992 On-line result reports available to Area and

Regional Offices on fortnightly basis

late March 1992 Advice to age pensioners appeared in Age Pension News

31 March 1992 Tax file number/data-matching segment on the Department's talk-back satellite link program

3 April 1992 Step 5 of cycle 2/1992 run

6 April 1992 Mailout of 131 400 informal request letters to

Special Benefit recipients, pensioners with non­ client partners and Job Search, Newstart and Sickness Allowance recipients who had not provided their partners' tax file numbers

9 April 1992 Cycle 2/1992 completed

13 April 1992 Cycle 3/1992 commenced

27 April 1992 Mailout of formal request letters to

243 000 pensioners who did not respond to August or September 1991 mailout

16 May 1992 Step 5 of cycle 3/1992 run

18 May 1992 Mailout of formal requests to 286 000 basic Family

Payment recipients who did not respond to the informal request letters in December 1991

Cycle 3/1992 completed

9 June 1992 Cycle 4/1992 commenced

15 June 1992 Mailout of 19 700 formal request letters to Special

Benefit recipients, pensioners with non-client partners and Job Search, Newstart and Sickness Allowance recipients who had not provided their partners' tax file numbers

Mailout of informal request letters to 8 800 pensioners requesting their partners' tax file numbers

30 June 1992 Social Security Legislation Amendment Act 1992 received Royal Assent

11 July 1992 Step 5 of cycle 4/1992 run

16 July 1992 Cycle 4/1992 completed

27 July 1992 Cycle 5/1992 commenced

Chronology 93

27 July 1992

10 August 1992

29 August 1992

September 1992

4 September 1992

7 September 1992

14 September 1992

24 September 1992

25 September 1992

late September 1992

6 October 1992

15 October 1992

14 November 1992

19 November 1992

Mailout of reminder request letters to 43 000 pensioners who did not respond to the formal request letters in April 1992

Mailout of reminder letters to 83 700 basic Family Payment recipients who did not respond to the formal request letters in May 1992

Mailout of invitations to apply for additional Family Payment to 450 clients

Step 5 of cycle 5/1992 run

Article appeared in Age Pension News thanking pensioners for responding promptly to the request for their tax file numbers

Cycle 5/1992 completed

Mailout of stop payment letters to 5 100 pensioners who did not respond to the reminder letters in July 1992

Cycle 6/1992 commenced

Mailout of invitations to apply for additional Family Payment to 26 000 clients

Mailout of further letter to 1 700 pensioners who did not respond to the letter sent on 4 September 1992

Step 5 of cycle 6/1992 run

Cycle 6/1992 completed

Mailout of stop payment letters to 36 900 basic Family Payment clients who did not respond to the reminder letters in August 1992

Mailout of invitations to apply for additional Family Payment to 23 500 clients

Cycle 7/1992 commenced

Data-matching Program — Report on Progress October 1992 tabled in Parliament

Step 5 of cycle 7/1992 run

Cycle 7/1992 completed

Student Assistance Amendment Act 1992 (containing an amendment to the Data-matching Program (Assistance and Tax) Act 1990) received Royal Assent

94 Data-matching Program

December 1992 Article appeared in Age Pension News giving pensioners information about the Department's fraud control strategy, including the Data- matching Program

21 December 1992 Data-matching Program (Assistance and Tax) Amendment Act 1992 received Royal Assent

18 January 1993 Mailout of stop payment letters to 1 800 clients (comprising Special Benefit recipients, pensioners with non-client partners and Job Search, Newstart and Sickness Allowance recipients who had not provided their partners' tax file numbers) who did

not respond to the formal request letter in June 1992

3 February 1993 Cycle 1/1993 commenced

13 March 1993 Step 5 of cycle 1/1993 run

22 March 1993 Mailout of formal request letters to

39 200 pensioners, 6 000 Job Search, Newstart and Sickness Allowance and Special Benefit recipients and 234 300 Family Payment clients

who were recorded as not having complied with the tax file number requirement

24 March 1993 Cycle 1/1993 completed

7 April 1993 Cycle 2/1993 commenced

3 May 1993 Mailout of stop payment letters to 1 600 Job

Search, Newstart and Sickness Allowance and Special Benefit recipients and 33 500 Family Payment clients who did not respond to the final request letter in March 1993

15 May 1993 Step 5 of cycle 2/1993 run

Mailout of invitations to apply for additional Family Payment to 87 500 clients

28 May 1993 Cycle 2/1993 completed

14 June 1993 Mailout of stop payment letters to about

4 000 pensioners who did not respond to the final request letter in March 1993

Chronology 95

June 1993 Introduction of automated process for stopping

payments to new claimants for Job Search, Newstart and Sickness Allowance and Family Payment who do not comply with the tax file number requirement within 28 days

July 1993 Privacy Commissioner's staff visited two offices in

Western Australia

8 July 1993 Cycle 3/1993 commenced

31 July 1993 Mailout of invitations to apply for additional

Family Payment to 10 600 clients

14 August 1993 Step 5 of cycle 3/1993 run

21 August 1993 Cycle 3/1993 completed

24 August 1993 Data-matching Program Quality Assurance module — first sample of selections issued to the network for checking

7 September 1993 Cycle 4/1993 commenced

18 September 1993 Introduction of automated process for stopping payments to new claimants for pension who do not comply with the tax file number requirement within 28 days

September 1993 Article appeared in Age Pension News giving pensioners information about the largest fraud uncovered by the Data-matching Program

Privacy Commissioner's staff visited five offices in Queensland, four offices in New South Wales and four offices in South Australia

16 October 1993 Step 5 of cycle 4/1993 run

19 October 1993 Data-matching Program — Report on Progress October 1993 tabled in Parliament

26 October 1993 Australian National Audit Office's Report No. 7 of 1993-94, Department o f Social Security — Data- matching, tabled in Parliament

27 October 1993 Cycle 4/1993 completed

Cycle 5/1993 commenced

96 Data-matching Program

27 November 1993 Step 5 of cycle 5/1993 run

3 December 1993 Cycle 5/1993 completed

8 December 1993 Data-matching Program Quality Assurance module — second sample of selections issued to the network for checking

24 December 1993 Social Security (Budget and Other Measures) Legislation Act 1993 (containing amendments to the Data-matching Program (Assistance and Tax) Act 1990) received Royal Assent

30 January 1994 Cycle 1/1994 commenced

February 1994 Privacy Commissioner's staff visited four offices in Victoria and three offices in Tasmania

22 February 1994 Data-matching Program Quality Assurance module — third sample of selections issued to the network for checking

March 1994 Privacy Commissioner's staff visited six offices in

New South Wales and one office in Queensland

early March 1994 Mailout of formal requests to 42 000 Family Payment clients asking them to provide tax file numbers for their partners

5 March 1994 Step 5 of cycle 1/1994 run

11 March 1994 Cycle 1/1994 completed

20 April 1994 Cycle 2/1994 commenced

May 1994 Privacy Commissioner's staff visited four offices in

New South Wales and five offices in Queensland

early May 1994 Mailout of invitations to apply for additional Family Payment to 35 700 clients

6 May 1994 Mailout of stop payment letters to 8 600 Family

Payment clients who did not respond to the final request letter in March 1994

14 May 1994 Step 5 of cycle 2/1994 run

20 May 1994 Cycle 2/1994 completed

Chronology 97

25 May 1994 Data-matching Program Quality Assurance

module — fourth sample of selections issued to the network for checking

June 1994 Mailout of 918 900 leaflets about all data-

matching activities undertaken by the Department (including the Data-matching Program) to Job Search and New Start Allowance recipients

27 June 1994 Cycle 3/1994 commenced

early July 1994 Mailout of invitations to apply for additional

Family Payment to about 17 000 clients

12 July 1994 Social Security Legislation Amendment Act (No. 2) 1994 received Royal Assent

23 July 1994 Step 5 of cycle 3/1994 run

26 July 1994 Cycle 3/1994 completed

August 1994 Privacy Commissioner's staff visited offices in

Western Australia

22 August 1994 Data-matching Program Quality Assurance module — fifth sample of selections issued to the network for checking

12 September 1994 Cycle 4/1994 commenced

11 October 1994 Data-matching Program— Report on Progress October 1994 scheduled to be tabled in Parliament

15 October 1994 Step 5 of cycle 4/1994 to be run

21 October 1994 Cycle 4/1994 to be completed

31 October 1994 Cycle 5/1994 to commence

23 November 1994 Data-matching Program Quality Assurance module — sixth sample of selections to be issued to the network for checking

3 December 1994 Step 5 of cycle 5/1994 to be run

9 December 1994 Cycle 5/1994 to be completed

98 Data-matching Program

October 1995

22 January 1996

Data-matching Program — Report on Progress October 1995 scheduled to be tabled in Parliament

Sunset clause in Data-matching Program (Assistance and Tax) Act 1990 is scheduled to take effect

~ * v ^

_ T H E PARLIAMENT OF THE COMMONWEALTH OF AUSTRALIA

PARLIAMENTARY PAPER No. 2 0 5 of 1894

ORDERS) TO BE PRWTED ISSN 0727418