DENNISS, Dr Richard, Executive Director, Australia Institute

RICHARDSON, Mr David, Senior Research Fellow, Australia Institute

Committee met at 08:59.

CHAIR ( Senator Back ): Good morning and welcome to the Senate Education, Employment and Workplace Relations References Committee and its inquiry into the adequacy of the allowance payment system for jobseekers and others, the appropriateness of the allowance payment system as a support into work and the impact of the changing nature of the labour market, which was referred to the committee on 26 June 2012 for inquiry and report. The committee is scheduled to report on 29 November 2012. The terms of reference for the inquiry can be found on the committee's website and are available today from the committee secretariat.

Before the committee starts taking evidence, I advise that all witnesses appearing before the committee are protected by parliamentary privilege with respect to their evidence. This gives them special rights and immunities because people must be able to give evidence to committees without prejudice to themselves. Any act which disadvantages a witness as a result of evidence given before the Senate or any of its committees is treated as a breach of privileges. Witnesses may request that part or all of their evidence is heard in private. However I remind witnesses that giving false or misleading evidence to the committee may constitute a contempt of the Senate.

I welcome Mr Richardson and Dr Denniss from the Australia Institute. I thank you for your submission, which we have got as submission No. 28, and for appearing today. Do you have any alterations, amendments or deletions that you want to make to the submission? No? I invite you to make a brief opening statement after which I will go to Senator Gallacher first for questions.

Dr Denniss : We will both make quite brief opening statements, if that is all right. The first point is, I think, the most important. I think that unemployment benefits in Australia need to rise. I do not think that they are adequate for the purpose for which they are designed and that is to give people the income they require to sustain themselves during periods of unemployment.

The benefits were designed with short periods of unemployment in mind, but the nature and structure of our economy is now such that some people are on those benefits for a very long period of time and after a long period of real wage growth the disparity between unemployment benefits and the minimum wage, for example, and certainly average weekly earnings, has risen quite substantially and, in turn, the relative impact on a person of losing their job today is far greater than it was 20 years ago.

For example, a person who goes to work and earns $60,000 a year finds out that through no fault of their own that their own factory has shut down. They now find themselves living on a very modest income and, if they are expected to live on that for a long period of time, our research shows that they tend to not spend money on health care, that they are likely to drop out of education or not pursue training options because the income is so low. We are actually creating a situation through poverty where the circumstances of that poverty are likely to become entrenched. Even simple things like the ability to clothe themselves well and get regular haircuts et cetera are discretionary expenses for people on very low incomes, and this can potentially impact on their capacity to get a job.

The final point I would like to make before handing over to Dave is that the mining boom has not delivered, and cannot deliver, benefits for the unemployed. While it has created some opportunities for some people to leave the state of unemployment and get jobs in mining or elsewhere, if there is one group in society that clearly is not benefiting from the mining boom, it is those on unemployment benefits. Indeed, if they live in regional areas near mining communities, the cost of living is likely to be rising far faster than the national average. We do not collect data on the cost of living outside capital cities but we know that the cost of living in some of these communities has risen far faster. So I fear that we are often blaming the victim and, while we are presenting to you evidence about why we think it is affordable to increase unemployment benefits, I would note that I have seen no strong evidence to support the claim that if we increase unemployment benefits, people will stop looking for work. The fact that it has been repeated again and again is not the same thing as strong supporting evidence, and there is strong empirical evidence to support that.

Mr Richardson : I would like to draw the committee's attention to the work of the OECD in this area concerning what happens when somebody loses his job and goes onto unemployment benefits, and how much of their income is replaced.

Every year the OECD compares the average wage earned in OECD countries with the replacement rate when you go on unemployment benefits or the equivalents around the world. Every year, out of the 29 OECD countries surveyed, we come in at bottom on that measure. The next lowest is Greece with 36 per cent, going up to Luxembourg with 87 per cent. In the submission we also look at some of those countries that have much lower unemployment than Australia. Each of those OECD countries with lower unemployment has a much higher replacement rate than the Australian figures.

This also leads us to question the role of incentives. People assert that a higher replacement rate when you are unemployed acts as some sort of incentive to become poor. That seems very questionable, when you look at the international evidence.

I want to also draw your attention to the wider macro-economic context in which we are looking at this matter. Our policy is designed to hit about five per cent unemployment. We know that five per cent unemployment is not a random thing. The likes of us are not going to be affected by unemployment. Instead, we know that year in, year out it is roughly the same people, people whom we describe as being part of an underclass who, when they do get out of unemployment, get into low-paid, insecure casual work, often for cash. What we in Australia are effectively doing is idling about five per cent of the workforce, five per cent of the most disadvantaged people and doing that for macro-economic purposes and then often turning around and blaming them for not having the incentive to find work. I will conclude there.

Senator GALLACHER: Just on the overseas comment, the committee has been apprised by the Parliamentary Library that perhaps that is not really an apple with apple comparison because most of the overseas figures include a proportion of insurance paid by the worker prior to losing their job. Do you have a comment on that?

Mr Richardson : That is true. A lot of those overseas schemes are insurance sorts of arrangements. But in practice it is a tax on the employer, the social security contributions which go towards financing the unemployment benefit scheme, retirement incomes and sickness as well in some places.

Different arrangements have evolved in every country. We are one of the few that has the sort of generally funded model that provides for minimum entitlement for all but at a very low level.

Senator GALLACHER: What would you say to the proposition that, as we are told, 60 per cent of people move out of Newstart within 12 months—and that is one of the aims of Newstart and the department says that Newstart is working because 60 per cent of people move out—and for the remaining 40 per cent who continue, basically that is because of a failing of society in other areas, education, family background, parental upbringing and that those people will struggle in any situation, regardless of how much help they get.

Dr Denniss : There is no doubt that health, education and a range of other services may contribute to that. Typically these analyses completely exclude geography and the family circumstances of people. Imagine that I have been working for a small manufacturer in a regional town. That manufacturer, thanks to a high exchange rate, has gone out of business and I have found myself unemployed for the first time in my life. Let us assume that unfortunately my partner worked at the same place, so our family income has collapsed. While it is true that I might be able to leave that town, my children might be finishing school or I might be caring for an elderly parent. It is entirely inconceivable that we can simply look at someone's education and ask, 'Should they still be on Newstart or not? While young people with no family commitments might feel like moving to Karratha to find themselves a high-paid job, what would happen if middle-aged people with caring responsibilities for their children, a parent or potentially both were to pack up and move out of town? Who would they ring to say, 'Someone else can now care for my mum'? Who would they ring to say, 'My kids are two years away from finishing school; I do not want to disrupt them. Can someone else please look after them because I am going to move to remote Western Australia where there is no high school'?

There are very, very good reasons why people might be reluctant to move to find work. When we look at things like tenure of people on Newstart allowance and we do not account for geography and family circumstance then we are probably missing, in many cases, the main determinant of their situation. I think we should reflect very hard on what signal we want to send to these people. Would the government or the department like to tell people, 'In order to find a job you should deprioritise your care commitments to your children and your parents'? I have never heard anyone say that, but that is what the pursuit of a job in a faraway region often requires of people.

Senator GALLACHER: Probably many members of the defence forces would view that statement about moving children vastly differently, because they move every two years and their kids go about their business normally. But geographical zones of high unemployment and generational welfare are obviously comprising a significant proportion of the long-term unemployed. There are a number of zones around Australia. One of mine is Playford in Adelaide. It has been targeted by the Department of Human Services for some activities to try to improve three generations of welfare. The briefing I had from Centrelink indicated that the people they were most worried about were the grandparents who were in their late 30s. They are facing a considerably long period of looking after a child who has a child. Do you have a view on any of that or how that can be ameliorated or fixed?

Mr Richardson : There are two important things. The OECD does have lots of evidence that well-designed training, education and work skill programs are effective. The other thing, too, is that if you want to, for example, put a number on the underclass in Australia, which would comprise a lot of these people that we are talking about, that would move with the business cycle. The people which do not look job ready, for whatever reason, go up and down with the business cycle. In the early 1990s there were a lot more than there are today. And today there seem to be more than we had five years ago. The same is true of other countries around the world. What looks hardcore unemployable moves with the state of the economy.

Senator GALLACHER: So the central premise of all the people who have submitted to this inquiry or committee is that $50 is going to improve things. What about the contention that the $31 that people are allowed to earn prior to the tapering off of their Newstart should be reviewed, that there should be movement in that area because $31 barely equates to two hours at the minimum wage? What if people were allowed to earn more without reducing their Newstart? Would that be a potential solution?

Dr Denniss : I do not see them as mutually exclusive. Some people do not have access to two hours or three hours work. It might not be a lot but in some cases it does not exist. Again, if you had to travel a long way to get two hours work and you did not have a car, and, even if you did, according to our research previously it is very hard to maintain a car on unemployment benefits, you would find it very difficult. If the car breaks down or it needs to be regularly serviced, they are the kinds of discretionary expenditures that people often have to let go. So my view is that the unemployment benefit needs to increase, but I do think that you could simultaneously shift the taper rates or the thresholds at which they kick in. There is an irony, I suppose, in that. While there is not a lot of evidence to suggest that these incentive effects are the main reason people are unemployed, we have heard from the minister and lots of other people that we have to keep the dole very low because people respond to incentives. Similarly, we give them quite a disincentive to go and try and get five or six hours work because we take money from them so quickly. So I think these incentive effects are exaggerated. I think that if we are going to tell ourselves that the incentive is so important that we have to keep the unemployment benefit very low, if we are going to believe that then I think it makes sense to loosen the taper rate.

Mr Richardson : I might throw a bit of a fly in the ointment here. If you are going for a job for a couple of hours a week, you are probably in the cash economy and you have got big problems then. If you declare to Centrelink that you are earning $30-odd a week, they want to see your pay slip. Then you are in big trouble. You have told them that you are working, but you cannot prove how much you are earning. I know this from acquaintances. So you are effectively in a very hard spot. You either have to withhold the information from Centrelink or declare it and get nothing as you move into casual, low-paid cash-economy work. I do not know the answer to that, but it seems a terrible dilemma to put our kids through—kids especially.

Senator McKENZIE: When we talk of geography as a determinant, I am just wondering whether you can give us the evidence particularly with regard to this particular issue.

Dr Denniss : I can furnish the committee with some evidence subsequently.

Senator McKENZIE: That would be great.

Dr Denniss : But to give you an indication, there are unemployment rates published by what are called 'small area labour markets'—so, almost postcard size—and across New South Wales you will find disparities in the unemployment rate ranging from two per cent in North Sydney to 12 per cent on parts of the North Coast and north-west New South Wales, so six times higher unemployment rates. Let us understand that across New South Wales you have identical wage rates as in minimum wage rates, you have an identical industrial relations system, and you have an identical Centrelink payment. Colleagues who are economists—and we are both economists—are obsessed with certain things like the size of the unemployment benefit or the design of the industrial relations system, and they are very bad at explaining why exactly the same instruments drive such different results in different communities. If unemployment can reach two per cent in North Sydney or three per cent in Canberra, then it is quite odd to think that there is some structural legal impediment to unemployment being much lower across the country.

Clearly some regions can get very low unemployment with the policies we have in place. Clearly other regions cannot. What are the variables that explain the difference? I hate to say it but are there lots of investments there? Are there lots of jobs there? Are other people who do not live in that community investing new money in those communities to build things and create things? Often not.

Senator McKENZIE: If you can provide the research around that then that would be fantastic. Could you briefly supply us with some comments around the relationship between Newstart and other allowances such as family tax benefit et cetera and what implications they might have on it being enough or not?

Dr Denniss : It is not my area of expertise but, clearly, unemployment benefits are the lowest of the benefits that we pay to adults in Australia. Clearly that is by design. Clearly that is in part based on the assumption that you should not be on unemployed benefits for a long period of time. People accept that if you are on a disability pension for example then you might be on that for the rest of your life. We do need to accept that many people wind up on unemployment benefits, I would argue, through no fault of their own, for very long periods of time. I think the premise that underpins the idea that the unemployment benefit should be the lowest is a flawed premise. It is a premise that works for educated people in capital cities for whom lots of job opportunities arise within the transport network that they can avail themselves of. It is not the premise that applies if you used to work in a small manufacturing part of western Victoria.

Mr Richardson : The OECD figures that I mentioned before show us if there is, for example, a single parent with a couple of children, we then do better than Turkey, Greece and Korea among OECD countries. So, yes, the replacement rate goes up a bit where there are children involved on average.

Senator McKENZIE: When we are comparing that figure, are we only comparing the Newstart allowance with Turkey and Greece or the entire package?

Mr Richardson : We are comparing the entire package. I could have brought along the OECD stuff but it is a messy spread sheet. I can still supply it though.

Dr Denniss : There is no doubt that some countries have insurance type schemes which allow them to be more generous but we have to separate two entirely different things. One, how do we fund unemployment benefits and, two, what is the economic effect of these benefits? If the economic effect that we are concerned about is the disincentive effect, if the economic effect we are worried about is that if it unemployment benefits are too high people will not look for work then it is irrelevant whether it is a contribution funded scheme or a tax funded scheme. The only thing that is relevant is how much money do I get if I work and how much money do I get if I do not work?

While the committee is right to inquire into how other schemes might be funded, this argument that if the unemployment benefit was a bit higher people would not look for work is still negated by the international evidence that shows that in countries where the replacement rate might be 60 or 70 per cent, people still go and look for work. The only way to get your hands on unemployment insurance in these countries is to be unemployed. You do not to draw your money out. If you work all through your life, you do not get to say at the end: can I have that contribution I made?

The economic decision facing an individual is what are the costs of going to work and what are the benefits of going to work and whether or not that 60 per cent replacement rate is funded through contributions or not. If you think it is all about incentives—which I do not but the minister does and others do—then it is irrelevant how it is funded.

CHAIR: Except, with respect, that there is a timing issue on it, isn't there? The information we have been given with regard to the OECD is that, for the first six months, Australia's circumstance might appear to be lower than the others because of this insurance factor but, after six months, when the insurance ceases in other countries, the Australian level remains the same. It tapers off very dramatically in countries which are insurance based, because the insurance policy effectively ceases at the end of six months and therefore the government support ceases. I think we need to compare that longer term for those who do not get back into work in that first six-month period. Senator McKenzie?

Senator McKENZIE: I would like some commentary around that so we can get some understanding of apples and apples.

Dr Denniss : I agree. I do not know that it is six months in every country but, in most countries that have that scheme, that is the kind of trajectory that you expect. But the point is that there are lots of people in those countries who, during that period where they are getting 60 or 70 per cent of their previous earnings, are still finding new jobs for themselves. The people do not wait till that six months.

CHAIR: Obviously not, and there is a tremendous incentive for them not to do so, as indeed there is everywhere.

Dr Denniss : But there is a tremendous incentive at months 2, 3 and 4. I agree the incentive rises after six months but, if you look at the international evidence, you will still find that, in countries in which it does drop off after six months, most people are finding themselves jobs at months 2, 3 and 4. If it were the life of Riley living on 60 per cent of your previous earnings, everyone would wait until month 6. This is my point. I think we are overemphasising this incentive effect. Most people's incentive is to get themselves another job. In countries where there are much higher replacement rates than ours, there is still the incentive and most people in those countries are still finding themselves new jobs way before this very generous period ends. That is for the very obvious reason that 60 per cent of your earnings is 40 per cent less than it used to be, and that is not helping your mortgage and that is not helping your family, and, secondly, most people want to find jobs.

Senator McKENZIE: We have heard a bit of evidence around the importance of skill sets in assisting people who have been unemployed, particularly the long-term unemployed, to get and retain work. Do you have a view on work experience in whatever form it may take?

Mr Richardson : There is evidence that different types of labour market programs have different levels of effectiveness. I have not looked at this sort of stuff much lately but, particularly for women, labour market programs seem to be effective—work experience sorts of things. But, with unemployment benefits at the levels we have in Australia, there are all sorts of problems. If you have a year on low income, maintaining the appearance that you look work-ready is very hard and so is getting to interviews and things like that. The transport problem can be terrible. While labour market programs and education and training—vocational training sorts of things—are very important, so too is helping people to keep up their appearance and their, if you like, pro-social attitudes. It is very easy for somebody to try and maintain their self-esteem by adopting the unemployed 'out of it' sort of lifestyle.

Dr Denniss : The system works well for the majority of people. It is designed to work well for the majority of people. What we need to do is think about the substantial minority of people that it does not work well for.

Almost by definition, the people who are most likely to stay unemployed for long periods of time are the people who for whatever reason, whether it is their skills or education or whether it is where they live, are the ones least helped by the system we have. We have a pretty good system, but if we want to ramp up the pressure on the people who fall through the cracks I think we should reflect on why it is that a predictable set of people seem to fall through those cracks.

There are a range of explanations, but my view is that for people who have been unemployed for six months or certainly 12 months we need to have a radically different approach to identifying their needs and identifying where it is that we can help them. I think the formulaic, across-the-board stuff works well for the majority. Giving the people who that has not worked for a larger dose of the treatment that has not worked is the triumph of hope over experience. I think we need to reflect on what it is about them. As I said, if the problem is geography then all the training programs in the world will not help.

Senator McKENZIE: I did not mean training programs; I meant work experience with employers.

Dr Denniss : But similarly I think we need to ask: what is it that is holding you back? As a rule I think work experience is very useful, in part because it helps to overcome the barriers in the employers' minds. A lot of employers are quite concerned about employing someone who has been unemployed for 12 months. If someone comes to a job interview and is competing against three other people, one of whom has just moved into town, one of whom has just finished school and one of whom has been unemployed for 12 months, a rational employer would think, 'In every other interview you have sat in for the last 12 months, someone sitting in my seat has seen something that I haven't seen. So, all other things being equal, I'm not going to bet on you.'

Work experience is very useful for confidence and experience for employees; it is also a low-cost way to say to employers: 'You can get a good look at this person. Even though on paper or in a job interview perhaps you would not have put them at the top of the list, they are pretty good. They fit in.' We have to understand there are structural impediments for employers who literally see long-term unemployment as an adverse signal.

Senator GALLACHER: There are 60,000 Indigenous recipients of Newstart, and amongst the 60,000 there would be some chilling examples of that geography you are talking about, where there is absolutely no industry, no opportunity to work and they are on Newstart. Do you have a comment on that?

Dr Denniss : I think it is a very vexed issue. I notice that when it comes to things like the Murray-Darling River and water flowing down it we are very conscious that some people say: 'I've lived in this particular town for generations. I like living here. This is where I should live. How is the government going to help me stay here?' You have Indigenous communities where people say: 'We've lived here for tens of thousands of years. I want to stay here.' Sometimes I do not know that we are entirely consistent with our attitudes.

I do not think there is a simple solution to employment creation in remote Indigenous communities. If there were, people would have done it by now. There is evidence that CDEP seemed to have had some good results, but it does not work everywhere for everyone. Again, I think we need to be idiosyncratic about these things. There are unique problems for employment creation in those areas and in turn we need flexible solutions.

Senator SIEWERT: I want to go to the reverse of the low payments for Newstart being an incentive to work. The flip side, which a lot of witnesses have articulated, as you do in your submission, looks at it as a disincentive. A lot of people say the rates being low is another barrier to finding work. Could you explain that a little bit more for us, please?

Dr Denniss : We have done survey research that shows, as I said before, that when people are living on such low incomes they are forced to make big cuts to their expenditure, including cuts to their expenditure on health and cuts to their expenditure on their own education. According to our survey, there are people who have ceased education because they could not afford it while they were on such low benefits.

As Dave alluded to before, even just maintaining the level of appearance that is required for some jobs requires a budget for clothing and a budget for haircuts—not that that affects me too drastically, but I hear that others look a bit unkempt if they do not stay on top of these things. So there are those direct barriers, but I think there are also motivation barriers that we should not be blind to. Being a very low-income earner who has to think about being unemployed as the main defining characteristic is not, perhaps, the best mindset to have when you are walking into a job interview to show someone that you are can-do and full of energy. So I think there are all sorts of barriers.

Senator SIEWERT: Thank you. I want to go to a couple of issues that came up yesterday in the inquiry. We are looking at the big picture. When we talk about how much it will cost to increase Newstart, there is a price tag of about $1.5 billion, but we had a lot of people from the not-for-profit sector yesterday and we have more today, and it is quite obvious that they are picking up the gap: they are increasing emergency relief, they are doing a lot of financial counselling and those sorts of things. So that is a massive amount of money that the not-for-profit sector is putting in to support people on Newstart because Newstart is so low. Have you done any work on how much that would be? I know there is a study from a couple of years ago that looks at what the not-for-profit sector contributes to the economy. Is there any specific work done around what the not-for-profit sector contribute to the unemployment system by supporting people?

Dr Denniss : I am not aware of any, but there might be. It is a very good question. There is no doubt that a lot of people literally cannot afford to live on the unemployment benefit, and what that means is that either civil society steps in to pick up the difference or, often, family members do, which again brings back this issue of, 'Let's not treat everyone the same.' Some people do not have family that can help, and some people do not live near parts of civil society that can help, so we have quite an uneven—I would say inequitable—system. When we know that the dole is so low that it requires a top-up, some people are in a better position to be topped up than others. As for the cost to civil society, I think it is substantial. I think there is the financial support that people can and do access, but there is also the time—the volunteer time and organisational time—which, if it were not spent picking up for an inadequate benefit, could be invested in other parts of the community, whether it is aged care or young kids. There is only so much capacity in the not-for-profit sector, and if an unemployment benefit is so low that it requires a lot of investment from that then that is problematic. One other thing I would add is that, if anyone else were asking the Commonwealth government for $1½ billion worth of support, they would probably have gone off and commissioned some economic modelling showing all the spillover benefits associated with spending that $1½ billion.

Senator SIEWERT: That was going to be my next question: how much would it benefit the economy if we spent that money?

Dr Denniss : Exactly. Putting $1½ billion into the pockets of the poorest people and the poorest communities in Australia will unambiguously improve the economic circumstances of those communities. More than anyone else in society, they will spend every cent they get. More than anyone else, they will spend it in communities that could do with a bit of an injection. Some of that money—not much, but some—will come back to the government in the form of increased tax revenue, and some of that money will actually create jobs in those communities, which will reduce other unemployment benefits. We have not done the modelling on that. I am just making the point that more powerful and perhaps better funded advocacy groups would probably come and ask government for money with a very clear indication of what those spillover benefits are. All I am telling you is that those benefits are not zero just because no-one has gone and estimated them.

Senator SIEWERT: We are also hearing—it is in the terms of reference, so obviously people are addressing it—that while, the employment services seem to be adequate for a cohort of people, for streams 3 and 4 there has been an argument that we need further streams and that 3 and 4, for example, are not meeting the needs of some of the people with complex barriers.

Have you done any modelling around what sort of investment we would need to put into employment services to meet the needs of people with the most complex barriers?

Dr Denniss : Not modelling, no; but that is effectively what I was alluding to before when I said we need to be far more flexible. I think we need to be almost entirely individual-focused—

Senator SIEWERT: So a case management approach.

Dr Denniss : I think some would argue that streams 3 and 4 have got a lot of case management, but it starts with the bureaucracy. I think you need to turn that on its head and say that people who have found themselves in this situation need to go into a very customised case management approach with a budget for assisting their gaps and advice that is tailored to their personal circumstances rather than what I suspect is a bit more formulaic than that.

Senator SIEWERT: I have some questions that I will put on notice, but I want to go to a question that was raised yesterday, and I note it has a bit of pick-up in the media today; that is the concept of using an insurance based model. You alluded to it earlier in the OECD models; could you expand on it a little bit?

Dr Denniss : Yes. I think if individuals want to take out insurance against the risks they should, but I do not think the government should see insurance as solving a societal problem. Perhaps excessively confidently, I am sure I could get cheap, good unemployment insurance. I think a private insurer would look at me and think, 'You're at low risk of not being able to find a job.' The people who need unemployment insurance the most will either find it the hardest to get or they will have to pay the most for it. That is what insurance does. If you think I am going to crash my car you are going to charge me a lot more for my insurance. You see the ads on television for pensioners cut-price car insurance. 'If you promise that you do not drive very much, we'll give you cheap insurance. So there is that element. There is also the element that if you have some sort of contributions-type approach—people who do not work, people who work casually or intermittently by definition are the people who are going to make the smallest contributions to any compulsory insurance model. Insurance has a role to play but insurance cannot make complex social problems go away, and there is an economic literature about adverse selection and there is an economic literature about moral hazard. I think both of those would suggest that unemployment insurance is not going to be a very efficient private model. Dave, would you add to that?

Mr Richardson : I would like to draw an analogy with superannuation, which is a mature insurance scheme now—effectively insurance. According to ABS figures, well over 90 per cent of employees have superannuation paid on their behalf. But if you look at women on low incomes in casual work, it is around 40 per cent. Exactly the same thing would happen if you had a contributory unemployment insurance arrangement. Sure, the people with good, full-time permanent work would be covered, but the people who need it—and the people who turn up in the unemployment figures from year to year, with intermittent spells out of the workforce and in poor jobs—are going to have nothing by way of a contribution history to support themselves. Even if you had an insurance arrangement, you would still need the residual scheme.

CHAIR: Could I put the view to you—with your economists hats on—as an extension of what you have just said, Mr Richardson, in agreeing with you as we look into the future of an ageing population: is it not the case then, as more people do retire with their superannuation being adequate, that will leave enough in the pool for those who you quite rightly identify will not have sufficient? I guess equally I ask the question with regard to an insurance-type scheme: yes, Dr Denniss, I agree with you. Those of us who are fully employed are far less likely to be costing the system. Is it not the case then that that leaves more in the pool to allocate to those who we know are falling through?

I ask you the question in this context—and I hope I do not ask it of everyone during the day, but you are economists. This, in summary, is the breakdown of the expenditure budget for Australia for this current year: 33 per cent, social security and welfare; 16 per cent, health; eight per cent, education; four per cent, defence. We know that we cannot leave defence where it is because it has been decimated this year, so it has to go up. We know there is always going to be an increasing demand on education. We know we have got an ageing population and that the 16 per cent for health is going to go up. The 33 per cent for social security and welfare is about $130 billion; I do not know that we are going to be able to dig more into that pool. That causes me to ask, within that $130 billion, where can we look to make adjustments? I put this question yesterday. If we cut five per cent out of administration and left the overall amount in, that is $6.5 billion that can go to the very people that you are eloquently speaking about. Can you give your response to those points?

Dr Denniss : Yes, I would broaden the net slightly. We published a paper on this a week or 10 days ago. According to Treasury's forecasts, the costs of tax concessions for superannuation will rise from $30 billion a year—it is called self-funded retirement but it costs the taxpayer $30 billion a year—to $45 billion a year in three years time. That is a $15 billion a year growth in so-called self-funded retirement. Treasury also estimate that 37 per cent of that money goes to the wealthiest five per cent of people, none of whom will be eligible for the age pension. So we know that that incredible spending and that incredibly rapid growth in expenditure is not 'taking pressure' off the age pension, because most of it is going to people who will never get it. So I would suggest some simple changes to the incredibly generous tax concessions for superannuation would give you all the money you need to pay for Gonski and the National Disability Insurance Scheme and to increase unemployment benefits, with change left over for Denticare.

CHAIR: So are you suggesting a change in that 33 per cent?

Dr Denniss : I am saying that if you treat tax concessions for superannuation as part of the welfare budget, which they are, I would look at the fastest growing area of that, which is tax concessions for superannuation.

CHAIR: At what point does that become an incentive for people who, as you say, will never draw the age pension? At what point does your proposed change get to a trigger where those who are making this provision for their own retirement make adjustments with their accountants to the extent that they do end up recipients of the age pension?

Dr Denniss : I do not know that anyone is going to voluntarily impoverish themselves—

CHAIR: No, not impoverish, just make adjustments.

Dr Denniss : What we know is that if tax concessions for superannuation were not so generous then they would probably invest in other asset classes. We are talking about high-income people here preparing for their retirement. The question is not, 'Do they do it?' but 'How do they do it?' They do it in the most tax-effective vehicle available, and that is superannuation. If they were instead to invest in investment properties or buy shares directly, the money is still there; it is just the taxpayer contribution—which is how you have to see the cost of the tax expenditure—would be a lot lower.

CHAIR: I would be very keen for the committee to receive that paper. I thought the adjustments had already been made as to how much a person can put into their superannuation fund. It is down to $25,000 a year?

Dr Denniss : Yes. The voluntary caps have been lowered, which is good policy. We have made the tax concessions more generous for those at the bottom, which is good policy. But if I earn half a million dollars a year the taxpayer is putting about $15,000 a year into my retirement income.

CHAIR: Does it count that person who only puts $25,000 into their superannuation?

Dr Denniss : That is the voluntary cap over your nine per cent.

CHAIR: Yes.

Dr Denniss : I am just talking about the nine per cent compulsory superannuation. If I am on the minimum wage the taxpayer is putting in about $500 a year. So if Dave earns $500,000 and I earn $37,000, the taxpayer is putting $15,000 a year into his retirement income and $500 into mine.

CHAIR: We would be very pleased to get a copy of that paper if we could. Time has beaten us. I thank you very much for your submission and your evidence.