 Thank you all for being here and thank you Vice-Chancellor for those highly insightful comments. We appreciate it very much. This journey, this policy and economic theory and empirical work began for me more than 25 years ago. It had a fairly inauspicious beginning. For me personally going to work for John Dawkins where he announced that I only had one job which was to reintroduce university fees into Australian higher education. My first response to that was I wonder if it would be inappropriate if I quit this job after five minutes because I thought it would be difficult and unpopular and I actually didn't favour it anyway. But I did stay and I wrote an options paper and I don't know if those of you in the audience have ever written an options paper. There are ways of writing it where it's pretty clear what you think is the right way to go. You can do this quite honestly but there is a subtlety to it but John Dawkins got the subtlety fairly. Clearly he realised that I thought this was a pretty good idea to have a loan system which was unlike the kind of loan systems that exist in other countries which you usually run finance from banks guaranteed by governments and with the debt collected over time and the idea was to have a repayment in the future depending on income. That's very simply what it was. I can remember one meeting with him after I delivered this so called options paper where he said to me and he looked very displeased. I think that was his general way of looking but he was particularly displeased with this options paper and he asked me a series of questions. The first one was what will the students think about this and I said I think they'll hate it because they're not paying now and they will. And he said well what about the universities? Where will they be in this debate? And I said well they're kind of lefties in an old fashioned way and they'll probably dislike it quite a lot too and they're not getting the money directly. It's only going to the budget so I don't think you can count on them for support either. And then he said uh-huh. He said well how many countries actually do this? To which the answer was uh-huh. Poor's will none do this. Uh-huh. Well then how do you know it will work? Uh-huh. Poor's. I said well uh-huh. I don't know. It will work. And he was looking at me increasingly skeptically and then he said and this was the crunch question. Because why did they do this? Well they didn't do it for reasons of economic efficiency or fairness. They wanted the money and they wanted the money because the queues were very big and they couldn't as a government wanting to be seen as low tax and low spend they weren't prepared to increase university places uh-huh through taxation. So the killer question was after all of this he then said okay then so when do we get the money to which my response was well in about six or seven years. So this was not a highly comfortable experience for me and even less I think for him. Have you ever been in a conversation where after a little while you're quite sure from facial expressions and other body movements that the person you're talking to thinks that you are basically a complete moron. And that's what happened to me. He'd be more not to uh-huh not to employ me because he was told I was quote to academic and didn't quote get politics and all of those things were sort of right. But he knew it. At that time in his office that uncomfortable 40 minutes in his office he knew that. But in the end the HEX system suited him politically very well because it was a little teddy bear and they needed the big fight was within the party and the party had at least I would say at least 45% powerfully opposed to the reintroduction of university fees in part because of an affection for Gough Whitlam who had abolished university fees in 1974. So having it pay back depending only on your future income and importantly when your future income may have reached average earnings of all people working in the country was kind of made it quite a lot easier for John Dawkins. So I don't think it was the default insurance the consumption smoothing all of those things that the economists eventually worked out this this that's what this system was it wasn't that it meant that it could get through the Labor Party and it was a vote which was one for John Dawkins by the number of three votes out of about 290 on the floor of the Labor Platform. The basic idea of this book is extremely straightforward and to put it in conceptual terms governments have a monopoly about knowing citizens incomes and that monopoly has a legal jurisdiction which no other agency has and that's a critical point. Moreover the government has an income tax system which is very comprehensive very hard to avoid unless you leave the country or die young and that meant that the marginal costs of collection were extremely low and what that meant and I might tell you that we didn't get this at the time we kind of made it up it wasn't in the textbooks and it wasn't as if there was a group of very clever minds thinking it all the way through we didn't really start to understand the economic aspects of income contingent loans until much later and some of them until even through Joe Stiglitz through quite recently. Now it sounds like an odd thing but I think in that it's not left-wing or right-wing it's basically using government as an instrument. The government monopoly as an agency which effectively allows people to tax themselves in the future when their incomes are relatively high and deliver themselves resources when they need it now and it won't come from commercial banks because of the central capital market failure and it fits very easily and nicely into the whole concept of government as risk manager. There's a wonderful book by David Moss which is essentially about government role as a risk manager and we do it in all sorts of ways pension systems, road rules occupational health and safety regulations in Australia most clearly is the Medicare system this is an insurance system it's basically a government managing risk and income contingent loans essentially a small part of that and the work then in my head had gone beyond higher education but I had the privilege of working with a whole lot of Australian social scientists and these many applications some of which Ian mentioned. My first experience was with Linda Botterall an agricultural political scientist expert and when we discussed this it seemed to her like this was a pretty good fair thing to do farmers were getting extremely large amounts of money for drought even though they were very asset rich. This led to many other examples and I won't go through them but John Quiggan had a very good idea about the collection of low level criminal fines and we did that with some criminologists Richard Dennis and I I think with Clive Hamilton had a drink one night we were complaining about Layton Hewitt not having to repay any of his money I've got nothing against Layton Hewitt but it was $100,000 worth of taxpayers money in a grant form and he's done extremely well of course so Glenn Wither thought about it with respect to research and development investments and there were many more examples not long after that I got to know Tim Higgins who is kind of more or less dragging on the students over to my office and we are jointly supervising them in actuarial studies costing income contingent loans in childcare for example indigenous in investments and paid parental leave which Tim then did a very wonderful thesis on. This all came to fruition as the Vice Chancellor mentioned in a workshop which Joe Stiglitz and I convened through a wonderful institution called the International Economics Association and they all about three or four times a year get a group of people kind of privately to nut out a problem because we didn't really understand the important the really important microeconomic foundations of income contingent loans and that's what we wanted to do and of course once you ask Joe Stiglitz and he says yes then you get everybody else because they just want to hang out with him and he ran this workshop brilliantly and came up with a cold concept which Richard Dennis has taken the beyond what happened in Bangkok the whole concept of transactional efficiencies the governments can do things more effectively and administratively inexpensive ways that are also very fair and that to Joe and to Richard became the basic point. So I wanted to thank Joe so much and the International Economics Association all the people that I've been working with over the last 25 years what a lucky person I have been for that but I'll leave my most important thanks to Dr Tim Higgins. Sometimes in your life you make decisions which turn out to be really kind of canny and sometimes you don't. Now I've had one or two canny thoughts about academic work and the cannyist I had was to invite Tim to be a joint editor of the book because he was so superb. 22 chapters he went through every line in some chapters many many times he went through the maths of all of the chapters with our technical and found mistakes and confusion in very many of them so I accept Joe's chapter of course which was totally perfect. Although some of us thought while reading that chapter just to pity this guy doesn't really understand asymmetric information properly we better take him aside and give him a tutorial. So thank you Tim this is the greatest debt the book is so careful because of his input I feel very much in his debt and I'd like him to come and say a few words. Thank you. Thanks everyone. Thanks Bruce. Bruce is way too kind with his words. We all know that we wouldn't be here if it wasn't for his hoots by 25 years ago coming up with Hex in the first place. So we all owe a debt of gratitude to Bruce for that. I happen on to Hex as one of the first cohorts of student borrowers in 1990. Although my older brother went through with free education as Nick Barr likes to say free is another word for some other sucker pays and that sucker is of course the taxpayer many of whom don't go to university and don't graduate. Cost sharing between university students and all taxpayers can be argued for on the grounds of equity since large private benefits go to university graduates but it can also be argued for on the grounds of competing physical pressures. The difficult question is what is the right balance. I don't have an answer but I do have concerns about the effectiveness of income contingent loans if loan amounts run unchecked and the balances run unchecked. Income contingent loans work for a number of reasons and Joe is going to be talking more about this shortly. They provide a solution to imperfect credit markets by providing funds that would otherwise not be available. They provide a solution to imperfect insurance markets by not requiring repayment when income is low and critically since debt is recovered from future means as opposed to current means as Bruce has spoken about. Both rich and poor have an equal opportunity to participate and that's the kicker. But equal opportunity to participate might not turn into equal participation. Since I started university in 1990, hex tuition fees are more than doubled in real terms. So far as far as we know participation hasn't been affected. But I don't think anyone knows the impact that very large contingent debts will have on participation or future graduate welfare. So I hope at least for mine the government approaches their discussions with caution moving forward. But the book is a lot more than about higher education. An important message from the book emphasised by Joe but really taken up in the last chapter of the book by Richard Dennis is that lending and recovering contingent debt through our underutilised but efficient tax system in areas beyond higher education has the potential to lead to quite substantial welfare gains. And I think that's one of the key messages that comes out of the book. Before I pass to Joe, some personal comments. My father was a well-known economist. My mother, who is here today, is as well. And it was for those reasons that I avoided economics my entire life. But you can't fight your genes and I've found that out. And more than anyone, it has been Bruce who convinced me to come into the fold, so to speak. I'm an actuary. I did statistics. I did maths. I did actuarial studies. I skirted around the issue of economics. But Bruce has brought me in. More than anyone, I owe Bruce. If he is father of Hex, given my 10 years working with Bruce, co-supervising, co-authoring, co-advising government now, maybe one day I can claim to be son of Hex. Though son of Hex does sound a bit like a horror movie, son of Hex, particularly with the budget around the corner, so maybe second cousin of Hex would be better. Finally, although my father passed away 20 years ago, his passion for social equality shared by Bruce is certainly rubbed off on me. And this is why I'm so thankful and thrilled that Joe, the greatest proponent of equality of all, has agreed to be part of this book and this project and is here to share his thoughts with us today. Thank you. It's a real pleasure to be here and especially to talk about the idea of income contingent loans. I think it demonstrates the power of an idea coming from the middle of a big continent with few people in the southern hemisphere having global impact. Ideas do matter and this idea, as the book illustrates, has really spread around the world and there actually is a bill in the U.S. Congress to try to do something like it. Let me begin by describing the alternative, the kind of system of financing higher education that we have in the United States currently. What I want to try to convince you is it's a disaster and I know that your government wants to move in that direction. I find that a total mystery and I want to just very briefly describe some of what's been happening in the United States. Part to understand what's been happening in the United States when it has to understand the broader economic context, the fact that most people in the United States have not been participating in the economic growth that has occurred. That median income, median half above half below, today in the United States is lower than it was a quarter century ago. So there's been stagnation for decades in the United States. Stagnation in income but not stagnation in university fees. Those have continued to rise and they've risen particularly since the great recession of 2008. Part of the great recession was cutbacks in state and local spending, government spending more generally. Most of the stakes in the United States have balanced budget frameworks which mean that they are, they have no choice but when the revenues go down they have to cut back expenditures or raise taxes. And as you know raising taxes particularly in the recession is not very popular. The result of that was that the United States had a pretty big dose of austerity at the state and local level. Where this showed up ironically was in cutbacks and investments in the future, comebacks in education in particular higher education. So stakes support to higher education went way down. It was I think a massive policy mistake but with stakes support for higher education the only way that the stakes, I shouldn't say the only way, the main way that the stakes could square the circle was to raise tuition. And some of the stakes raised tuition by 50, 80%, 100%. These were massive increases in an era where wages incomes were going down. The result of that was that the only way that most students who made the mistake of choosing parents who were in the bottom half of the distribution could go to college was to borrow. But we didn't have an income contingent loan program so they had to borrow with un-contingent loans that had fixed obligations. And the result of that is that student debt increased enormously. Today student debt in the United States is well over 1.1 trillion dollars. And even for a rich country that's a lot of money. It's more than all of our credit card debt. It is beginning to have macroeconomic effects. It's having very big impact on young people's lives. They can't afford to buy a house because of the student debt. They can't afford to go into teaching or low paid occupations because they have to repay the debt. And of course to make things worse, our financial system persuaded Congress to pass one of the most iniquitous laws that you could imagine. It said that under our new bankruptcy law, you could, if you borrow too much to buy a yacht, you could discharge the debt in bankruptcy. If you borrow too much to go on a vacation, you could discharge the debt in bankruptcy. But if you borrow too much to get ahead in life, to get an education, so you could participate in the American dream, you could never discharge that debt. And if parents cosigned the loan of their child and their child died from an accident or from a disease, the parent could not discharge the debt no matter what happened. And as I traveled around the United States when I was discussing my book, The Price of Inequality, the most poignant stories were coming from young people and their parents exactly about this. It was even more difficult for those who wanted to get a graduate degree. Because by the time they graduated from college, they had an average $30,000 of debt, but many had $100,000 of debt. And then they said, you know, in the United States today, high youth unemployment, even those with a college degree can't get a job. They're working in a place like, you know, waitress, waitress, no future job prospects. And those who are in the bottom half of the population say, you know, I look around, they look around, they see the children of wealthy people going to graduate school because their parents are paying for it. They know they can't take on more debt. They see the children of the wealthy parents taking unpaid internship, getting skills, but they can't afford it. So the sense in the United States is that we have created a system that is totally unfair. And this inequity is showing up in the data. So we now realize that the notion of the American dream is really a myth. The ability of those at the bottom half to move up to the top is really minimal, much, much lower than in any of the other advanced countries. I joking say, we have much less opportunity than even then old Europe. Because we used to castigate old Europe as saying, you know, those are ossified countries. Turns out that America is more ossified. Statistically, what that means, I mean, obviously, there are examples of people who made it from the bottom to the top. But those are rare. And that's why they get so much attention. What economists and sociologists, social scientists mean by equality of opportunity is what are the chances? What are the prospects? And the life prospects of a young American are more dependent on the income and education of his parents than in any of the other advanced countries. And a primary reason for this is the way we have organized the finance of education. And particularly higher education. There are real problems with our elementary and secondary schools. But I'm focusing now on higher education. The result of this is that the fraction of those at the bottom half of our population who go to school, who go to university and graduate is about 8%. And so the reason for it is they can't afford it. At least it's one of the reasons. Can you imagine in the United States, the income of those at the bottom, a worker who is full-time, a full-time worker at the minimum wage, his income in the United States is under $15,000. Tuition at a school like Columbia is $50,000. And for a young person whose parents have been earning $15,000 to wrap his mind around the idea of getting a debt of $100,000 to get an education, the returns to which are risky is mind-boggling. And we've done studies that show that poor students are just not willing to undertake debt. And so if you force them to take debt, they won't go. There are a lot of discussion here in Australia looking at our best universities, Harvard, Columbia. Some people say you ought to, Australia ought to emulate that model. One has to understand a couple of things about these universities. Not a single good university in the United States is a for-profit university. The for-profit universities in which the price system works, who are trying to maximize their profits. The for-profit universities excel in two things. I've mentioned that in a couple of talks I've been giving around. If you've heard me, forgive me, but it's something I feel very strongly about. They excel in two things. One, in exploiting children from poor families. And they do a really good job with that. Oh, telling people that if you go to our school, you'll get a good job. In fact, if they go to the school, they typically don't graduate. They don't get an education. If they do graduate, they don't get a job. And that's why the repayment rate is so, so low. The other thing they excel in is lobbying Congress to be able to get government grants without any regulation. You know, a lot of us, for 20 years we've been agitating over this issue. This is not a new issue. We've been agitating over the fact that, you know, the view that we should be giving government loans, government programs, government grants, only to universities that have demonstrated some element of quality. That they can graduate students and the students that graduate can get jobs. We're not talking about high standards. We're talking about minimal standards. And they've been very successful in bribing Congress. Why did they bribe Congress? They're making millions, millions, billions of dollars off of it. And unfortunately, we have a Congress that is easily bribed. We had a student loan program that was a public-private partnership. A lot of public-private partnerships you may hear about. This is a wonderful public-private partnership where the government took the risk and the private sector took the profits. So what they did was really outrageous. They charged interest rates as if they were bearing the risk. But they weren't bearing any risk. And their confidence in collection was very low. And they had no incentive for collection because when the people didn't collect, the government bore the risk. It was a real system of corruption. It was highly profitable for the banks. And again, they used that not only, I mean, one of the sad things is that not only did they corrupt our political process, they corrupted some of our universities. Because it was so profitable to get students to borrow money, say at 6%, when you had zero risk, that they started bribing university officials. So it was a disaster. Finally, under Obama, we stopped this public-private partnership. The amount that the U.S. government saved, just the U.S. government saved from stopping to going to a publicly driven loan program. Not a well-designed loan program. I mean, not an income contingent, but just getting the private sector out of this, focusing on the transaction costs that have been over 10 years, the savings was $70 billion. That was a gift we were giving to the banks every year. No wonder with that amount of money they were willing, you know, with $70 billion, you can pay a lot of bribes and still wind up with a lot of money left over. And so that illustrates what I would call a failed financial model. It's a failed financial model, but it's also a failed broader economic model because it's contributed to the inequality in the United States and the inequality of opportunity. So when one thinks about the issue of income contingent loans, I think it's important to put it in a broader context of understanding the market failures by markets don't work well. The advantages that government provision has and the economic and social consequences of making sure that there is equality of opportunity and that income contingent loans are an important ingredient, not just in increasing opportunity, increasing equality, but also when increasing efficiency. Later on, I hope the discussion will extend to the number of other areas where the concept of income contingent loans can and should be applied. In fact, we should more broadly understand that there are very big advantages of government running certain basic financial transactions. This was the point that was made about the economies of scale and scope of the government, that it has to keep large administrative records in order to run a tax system, to run a social security system, that once you have in place this large system of collection and record keeping, there are big advantages of using that for other purposes. The marginal cost, the extra cost associated with using it for other purposes is very low. So to give you just one example of the advantages of the government in these kinds of financial transactions is social security, old age retirement, I guess you call it super old age retirement programs. Studies in the United States show that the cost of running social security, all the administrative costs, collecting the money, keeping the administrative records, all that is around 2% and that's about a tenth of what the private sector pays. In fact, that's a tenth of what well-run private sector firms do. There have been some experiments of privatizing social security, privatizing these retirement programs, turning them over to the private sector. The UK did a partial privatization and the result of that was it was shown that retirement, what retirees got as a result of their savings, the money they were putting aside was reduced by 40%. You have to understand the economic incentive. The private sector wants to maximize transaction costs. The private sector makes money off of transaction costs. That's their objective. Their objective isn't making people more secure, making people have greater security in their retirement. Their objective is making money and they do it whatever way they can. Now unfortunately America's private sector showed that it had no moral screw-ables. It would engage in discrimination, predatory lending, abusive credit card practices, market manipulation, whatever it took to make profits and I can go through all the very, very clever. Our best students were going into the financial sector because it takes a lot of ingenuity to do this in ways that keep yourself out of jail. A few of them went to jail but mostly they did maximize profits subject to the constraint of not going to jail. So we took our best students and that said, that's your life mission and they did very well. But actually our social security system is not only more efficient and actually is very responsive to the needs. We did a survey when I was in the Clinton administration of consumer responsiveness. And the Social Security Administration was in the top four of Americans, American Corporations. Much better than most of our airlines or other big companies. And actually we succeeded in making it very responsive to what we call the clients, our old age people. So you can succeed in having an efficient, very efficient public sector that is very consumer responsive. Well, the point of this is more than that. It is that once you've established this kind of infrastructure that collects the data and can enforce, delivers checks, can enforce tax collection, you ought to use this for other purposes. And that's one of the basic ideas behind income contingent loans. But there are some other ideas that I want to also talk about this morning. Before doing that, though, I want to echo one of the things that was just said and that is the importance of getting a balance between overall public funding and putting the burden on individuals to fund through income contingent loans. One can think in a way of the income tax system as a kind of income contingent loan. But it's across generations. This generation, those who are, have higher income largely because of better education, but sometimes because of luck. Sometimes because they've been able to persuade government to give them big gifts. Sometimes because they are able to figure out how to use monopoly power to make big profits. So there are a variety of reasons why people make high income. The idea is whatever the reason you're making high income, you ought to share a little of that high income to help finance other people's opportunity, the next generation's opportunity. And that's the way the traditional income tax system worked. And I think there's a lot of virtue in that system. But as Bruce pointed out, in recent decades, governments have faced, I think, a somewhat artificial stringency, fiscal stringency. The kind of debate you're having in Australia is really peculiar because your debt GDP ratio is 14%. A fifth of that of the United States. And there is no evidence that even debt GDP ratios much, much higher have any adverse effect on economic growth. The Rogoff-Reinhardt research that talked about there being a critical number, like 90%, you're nowhere near that. But that research has been totally discredited. No evidence of that. The United States' fastest economic growth occurred in the decades after World War II. And at the end of World War II, we had a debt GDP ratio of 130%. So high debt is not an impediment to strong economic growth. So this worry about your debt when it's at 10% of what the United States had at the end of World War II is really peculiar. And if there are cutbacks on investments in people or technology, the cost of the future are much greater than the non-existent cost from the debt. So the question of how much of education should be paid by the students and how much should be paid for a broader society is an important question. And one has to keep in mind that one should not put too big of a burden on the students today. But there's another aspect of that that I think one has to keep in mind. And that is that modern education at the university level, for the most part in our research universities like ANU and other research universities like Columbia, almost surely more than 50% of all the costs are really not associated with teaching, but with basic research. There are joint products, we sometimes call joint products, where combining teaching and research turns out to be very efficient. It's a good way of organizing both teaching and research. But the students shouldn't be obligated to subsidize our basic research. And yet the cost of education are largely, as I say, at least 50% related to that basic research. I won't describe how little I teach. I think it's a good system as I said. But I can't justify my salary in terms of teaching. And yet the students are obligated to pay for it. Now, our rich American universities manage this by very large endowments, very large charitable contributions that are made by a society that recognizes the value of higher education and basic research. So at Columbia, for instance, in the last few couple years, we've raised $6 billion of donations. And you can do a lot with $6 billion, you can certainly pay my salary. So this has been, but this is how we finance our basic research. Through these kinds of contributions, we created a, you know, funded a big new neuroscience laboratory. We do, you know, lots of things of that kind. We get a lot of support from other foundations that are focused on on a whole variety of issues. And we have a lot of wealthy people who are committed to the notion that there should be equal access for all. So we got one donation of $400 million dollars for scholarships for poor, for poor Americans to make sure that they could go to school. So that's the way the American system works. It's not nothing to do with the price system. It's not price signals that are working. And anybody that thinks about it as a, as a conventional market model, or you're just totally, totally wrong. But the point of all of this is that as you think about how you draw that balance between how much should be paid by the individual and how much should be paid by society, certainly in my judgment, the part that is attributable to basic research should be part of the burden that should be based on general taxpayers and not be put to the students. Well, I now want to switch a little bit to the basic theory of income contingent loans. And it's the basic theory of income contingent loans is arises out of an awareness that of a general theory of what we call market failures. economists have described what well functioning markets would look like. When I say well functioning, what I mean is they're mythical markets in which there was perfect information, perfect information never exists. But with imperfect information, markets often don't work very well. And over the last 40 years, we've come to understand the ways in which they don't work very well. So one of the consequences of imperfect information is that there are imperfect risk, or we can say insurance markets, and imperfect capital markets. So a general principle, for instance, is that we talk about to our students in the simplest economic courses is it makes sense to spread your income over your lifetime to smooth your consumption. And so the kind of paper you were referring to smoothing consumption is a basic principle. So that if you have a low income, you don't want to starve that period. Starving is not very good for future productivity and has all other it's uncomfortable and so forth. So you want to take income from periods when you have high income and keep it relatively level. And that requires borrowing. Now when we talk to our graduate students and teach them about this general theory about income smoothing, one of my students, a co co turned out to be a co author. On the basis of that went down to the local bank and said, I just learned about income smoothing consumption smoothing. Right now my income is very, very low. But when I get my PhD, my income is going to be much higher, not as high as if I went into finance. But but being a university professor, it's certainly lower than going to be higher than the $10,000 I get now, it'll increase by two, three, four, five fold. So can you help me smooth my consumption? He wanted to take out a lifetime, you know, alone. And the bank was distinctly uninterested in making him alone. His name was Michael Rothschild and he explained that he was a Roth child. And the banker, this was in Cambridge, the banker said, I don't know about the Rothschilds, I've only been in banking for 11 years. So anyway, the bottom line is he didn't get the he didn't get the loan. Well, we now understand that the theoretical reasons why there isn't that kind of consumption smoothing, we call that intertemporal smoothing that over time. That's one aspect of smoothing. But the other kind of smoothing is what we call interstate smoothing, that you can have a bad contingency or a good contingency, you could have your your consumption, you could be lucky and and get a good job and have a very high income, you could be unlucky and get a very bad job. And some of this is to do with your efforts, but a lot of it has to do with just luck. So you would like to get insurance against the bad outcomes. So you know, I'd rather I'd like to give up some income in the good stakes to make sure that I have some more income in the bag stakes. And kind of insurance that allows for, as I say, we call interstate smoothing, you can't buy insurance that kind. Just like you can have interstate intertemporal smoothing. There are very high costs for the inability to have engage in this interstate and intertemporal smoothing costs, both in terms of welfare, because there's they're going to be a lot more volatility in consumption. And people don't like that kind of volatility. And they, they, there are further consequences that I'll describe in a moment. So one way of thinking about income contingent loans is that they are a response to these imperfections and risking capital markets that with existing institutional arrangements, people can't smooth consumption over time. They can't smooth consumption across stakes of nature. And incantation loans improve matters. They're not perfect, but they improve matters relative to the current system. One of the reasons, of course, is that the absence of interstate smoothing of insurance contributes to the inability to have intertemporal smoothing over time. Because lenders worry that borrowers won't have the resources to pay. At the same time, this has very big implications for borrowers. Borrowers worry that with a conventional loan, there will be a variability in future incomes leading to a variability in consumption with hardship. What they have to pay is fixed. And if their income goes down, what they have left over is very small, could be nothing. And that's why the evidence is overwhelming that if you for if the only option they have is to borrow conventional loans, many poor people won't borrow. I was mentioning before, the kind of mindset of if you're born to poor parents, and you've seen them get a minimum way income of $15,000, you're going to be reluctant to take a debt of $100,000. I was a trustee at Amherst College, which is a small college with 1600 students, but a very wealthy college that has a lot of income. And we have what we call a needs buying admission, where we accept students and then we ask them what their parents' income are, and then we make up for the difference between our view of what they can afford and what the cost of the college is. So we try to say we're making it available to everybody, regardless of the income. And it's not a price system working. Let me make it clear. We don't believe in that. It's very strong in our view. We used to in that financial package that we gave young people, we used to have mostly scholarships, some work, but some loans, some conventional loans. And then about five years ago we decided to eliminate conventional loans. Why? Because we had done a study that showed that people from poor backgrounds just were so afraid of the conventional loans because of the hardship, the potential of hardship that they imposed that they wouldn't go to Amherst. And we said the cost of that to our objective of having economic diversity was too great. And so we said we'll just give them the money, not ask them to repay it. And it did help us achieve greater economic diversity. So there are some important economic consequences. This is not just a question of equity, as I said before. It leads to underinvestment in education, particularly of those who are not at the bottom. The only way they can afford to go to school is to take these huge debts and they don't want to do that. It means, as a result, that the bottom half of our population does not live up to their real potential. We're wasting our most valuable resources, our human resources. It means that when they do take out a loan, they can't choose the occupation for which they're best qualified or they're most interested and they can't take a low paying occupation like education or the ministry. They have to look for the highest paying job. It distorts the labor market. It means that there are lower levels of job search. They have to accept early job offers rather than continue to search for a better match because the risk of not getting a job is just too great. There are all kinds of real economic consequences in terms of economic efficiency from this system in the absence of income contingent loans because there isn't this adequate risk sharing across states. One way of thinking about income contingent loans is they represent, you might say, a compromise. There is some degree of interstate smoothing. When future income is very low, repayment is lowered. But most importantly, when future income is moderately low, the burden of payment is spread over more years. And that means that there is better intertemporal smoothing. So you get better interstate smoothing, better intertemporal smoothing. It's not perfect, but it is a marked improvement over existing arrangements. So the access to funds increase and the access to this kind of partial insurance increases both efficiency and equity. And finally, as I said before, it takes advantage of economies of scope and tax collection, reduces the cost of enforcement and increases the effectiveness of enforcement. The main application of this idea of income contingent loans, as it's been said, is to student loans. Australia has been a real model of success. The Conventional Loan Program in the United States, as I have already described, has been a disaster. And it's been one of the real reasons we have had such inequality. What I want to do in the remaining few minutes is to talk about another application. And I assume in the conference this afternoon there will be a discussion of many other applications. And the application I want to talk about is to unemployment. And it illustrates a lot of the principles that I've talked about. The basic idea is the following. That most people have short bouts of unemployment. And with short bouts of unemployment, they can absorb that in terms of their lifetime income. So if you're unemployed even for six months out of a, and you knew that that was all you were going to be unemployed, six months out of a working life, say 40 years, is one eightieth that you're losing, roughly. And one eightieth, you know, 1%, 2%, people could absorb that kind of bump in their lifetime income. So if you knew that you were only going to be unemployed once, it would make sense to, for instance, borrow against your social security, your retirement, against your future income. And a government-organized loan program would make a lot of sense. You could go to the social security, say, I want to borrow, I'll pay you back, here is the collateral of my retirement income. And the advantage of that would be that you would be able to smooth your income and it would have no adverse incentive effect. You would have no reason not to search intensely for a job. You would have no reason to accept the first job that came along simply because it's a job. So there are a lot of efficiency benefits from being able to smooth over time and use the government as part of your, as your bank, if you want to think about that way, as an efficient bank. It would have easy records to know how much you owed, it would have easy collection. So that would make a lot of sense. The problem is you don't know that you're going to be only unemployed one period for six months. There are some people who have bad luck and face repeated episodes of unemployment. And people would worry that if I borrowed against my social security, against my retirement, and I have repeated episodes of unemployment, that would mean if I had a system that was dependent on paying it back out of my old age, when I come to old age I'd be starving. And society worries about that. They want to make sure that you don't deplete your retirement account. Well, there's an easy way out of this. And that is income contingent unemployment programs. So that you have a mixture of insurance, some unemployment insurance, which is not repaid, but an income contingent loan program will lend you money and provided your history is not too bad, you pay it back. But if your history is very bad, what you pay back will be reduced. So that kind of mixture between a income contingent loan program and a formal unemployment insurance program is a way of getting better risk mitigation, risk over differing steaks of nature, different outcomes, possibility that you may have a long number of periods of unemployment, and better intertemporal smoothing, better smoothing over time. So the bottom line is that what we show, I'm not going to go through this, is that one can show that this kind of income contingent loan program for unemployment is much better than our current system of just unemployment insurance, and obviously much, much better than a non-contingent loan program. So that it is a way of improving efficiency and risk mitigation. And in paper in this volume, we describe the design of the optimal unemployment program, a mixture of income contingent loans and unemployment insurance. And we show how that kind of a program is much better, as I say, than existing programs. There's an element in the program that we describe that has a similarity to Singapore and Malaysia's Providence funds. Singapore and Malaysia have a system of a lifetime savings program, where in the case of Singapore, you had to put in just 42% of your income into this Providence fund. And then you could draw upon it for a whole variety of reasons. And if your necessities for health unemployment exceeded a certain amount, there was effectively a kind of insurance element in it. So this is really taking an idea that already exists, but actually extending it further. There's some complexities in the design of the program that I probably, you know, if there's some interest here, could talk about there are a whole variety of externalities that are critical in the design of the program related to effects of behavior at one time, or a program at one time, say, at early life on behavioral later life, or later life on early life. There are externalities between the loan instrument and the insurance instrument. There are externalities from private savings decisions to public decisions, and vice versa. So the complexity, the mathematical complexity of the paper arises from trying to identify these cross market, cross instrument, cross public and private sector externalities and trying to understand the implications of that for the design of the program. There are also some, we did, design features that we describe where peer monitoring can be used to improve fishing sea. There's some general principles we derive that the optimal unemployment program varies with age. There are the parameters that are critical in determining the design of the program, search elasticities, correlations between employment different periods. All of that is sort of all, you know, interesting analytic problems of how you design the optimal program. What I want to emphasize is, though there are real benefits to having this kind of income contingent loan program, it's not just better intertemporal smoothing, not just better interstate smoothing, but there are actually greater efficiency in job search. Individuals are willing to continue to search to find a better match with greater productivity. The problem with the existing arrangements is that when the unemployment insurance ends, people have a strong incentive to accept the first job. And that means they're less well matched and productivity is impaired. And also there are significantly greater transactional efficiencies. So unemployment is another example where appropriately designed income contingent loans can be an important part of the policy mix, improving societal well-being and economic efficiency. There are a whole other variety of context in which one ought to think of using the efficiency of government, the economies of scope of administration and enforcement for, instead of relying on the private financial sector. The private financial sector has shown that it excels in the ability to maximize transaction costs and engage in predatory and abusive behavior. And if you want to maximize that, that's one, you know, rely on the private sector. But at least for a wide class of loans, intertemporal smoothing, where the risk are relatively low, where there are assets that you can use to back the back at home, old age retirement collateral where you have sufficient assets, there's essentially almost no risk. And the government can really improve intertemporal smoothing enormously. But you can go beyond that to have some degree of interstate smoothing that is to say some risk mitigation. The boundaries of where you start having problems of our, I think our really interesting open question for future research. But I finally want to congratulate Bruce and his team for raising what is a really important new set of instruments and a new way of thinking about financial markets. And one which I think over the coming decades will have a profound effect on the way our economies function, increasing both efficiency and fairness in our system. Thank you. Peter Whiteford, Crawford School. I was very struck by the unemployment example, just to give you a brief background, in Australia, our benefit system is non-contributory, so we have unemployment benefits that are flat rate and there's no time limit on them. So you don't have to have contributed in order to get the benefit. In the most recent budget the government proposed that if you are under the age of six months a year waiting in advance, so you won't be eligible for the first six months of your unemployment. There's a lot of concerns about this, whether it actually happens will be interesting to see. But I was wondering in the light of what you have said about income contingent loans, when you compare that sort of proposal but also with our current system, which is flat rate and as I say not time limit on the new element of loans to get around this problem. Very much so. Let me just say that young people, the argument in the analysis, it makes sense to rely more on loans for young people. Because the notion is that they have a lifetime to help smooth. And so that if they're unemployed for say six months the the diminution in future consumption is going to be relatively low. I think it's really the proposal to say that you have to wait six months is a real hardship in several ways. It's not like most 23 year olds have a big bank account to draw upon. It's actually just to reverse. Somebody in their 40s or 50s ought to have a bank account. Somebody in their 20s won't have a bank account. So what you're telling them is, you know, sorry suffer. And, you know, I guess there are curtain traditions like saying suffering, there's redemption through suffering and but I've never bought into that particular view of life. The income contingent loan says that the government will be the risk of not being repaid is minuscule because the likelihood that they won't over the rest of their life get enough money to repaid the loan if it's a modest loan is close to zero. So that's exactly the kind of instincts where it would mitigate hardships. And finally let me say there are real real efficiency gains from doing it because in the absence of an income contingent loan of the kind that I just described individuals will feel forced to accept the first job or to accept a job that doesn't increase their skills and will therefore have adverse effects on their lifetime prospects for which society will pay a high cost, not only the individual. So it seems to me it's just the wrong way to go the proposal of the government and this. My name is Paul Hubbard. I'm a PhD student here in economics and it was thanks to something you wrote that stopped me from a career as a lawyer. So firstly, thank you. My question is this. You mentioned the lack of relationship between the price of tuition in universities and sort of the costs associated with that. Do you think that competition from online models or elsewhere can potentially bring that price closer to the marginal cost of certifying that a particular student has a particular skill? Yeah, so I think I think that what I was trying to argue is that in fact education is different from steel and from lots of other commodities where you think naturally there are competitive forces or there should be competitive forces that lead to efficiency. There are a lot of reasons why education is very different. One of the reasons is that the buyer is really uninformed about what he's getting. That if you knew what you were going to be taught in school you wouldn't have to go to school. So there's almost by definition an information asymmetry. There are some reputation issues that may mitigate that but particularly people from lower income strata and from less educated parents don't understand a lot of the characteristics and they're actually been resistant from the for-profit universities of getting good metrics that would be useful as providing the relevant information. At the same time I think at least all the universities that I know are very sensitive to cost. They're trying to get more efficient but trying to think about what are the things that are going to enhance the efficiency of the education process but not deteriorate the quality of the educational experience. Those are really hard. I think going forward that MOOCs and other online courses will become an important part of the education process but only part. I think it's still going to be the case that people want interaction with people with teachers. That some of the lectures will be given through MOOCs and there will be sections where people will interact. The division of labor may change and that is part of the ongoing changes in the education system. There are forces trying to make sure that education is getting more efficient and not really being driven through the price system. That's not the major mechanism through which competition exhibits is manifested in the education sector. Michael Harris I'm from ABIS which is a Government Research Agency but I actually want to talk about education and ask about education because that's obviously where this is stemmed from and a lot of your talk focused on it but I'm sort of mindful of the fact that there's a presumption that we go that there's an under-investment in education, particularly from poor people and that will benefit them in reducing equality and I just want to push back on that a little in the sense some things we know is that in particular higher education we know some of that is consumption as well as investment. Some of it obviously has private and public payoffs and HEX is designed to share that burden but also we've got human capital versus signalling arguments and stuff like that and it's not clear from PhD who develops human capital and goes into finance is in fact improving national productivity but if we gloss over all that then the inequality nail that we want to hit with our education hammer is just put more students through universities and I wonder whether there's a potential for the income contingent loan model to kind of create a tail wag the dog situation where the higher education sector gets driven by dragging more people through the door I think in Australia there's been some views that that's happened we've taken our a segregated or sorry a differentiated education system tried to make it homogenous so every university is meant to essentially have the same mission we push students through the door there's pressure then on the types of programs we offer the types of things we do and oddly enough in the United States which has a very highly differentiated system you get things like the adjunct problem of lots and lots of poorly paid people who do the teaching while the star researchers go off and do research and do very little teaching not pointing any fingers but I'm sure Columbia does things differently but I have friends overseas who struggle to do their PhDs and then struggle to make a living after it so is there a way what are the other inequality reducing things you might want to do at the same time as you look at education and not the one size fits all solution to an inequality problem well obviously one has to pay attention to the design of the university system and there has there have been I think some failures of the kind that you've described in the design of the university system still I think there are a couple facts one has to bear in mind one of which I already alluded to that no matter what you think about the quality of education social versus the private returns it is a little bit of a crime that those at the bottom half don't have access and that this is design the income contingent loan enhances the ability of those at the bottom to access the higher education and to me that's an important aspect of that and that can be separated from the design of the university system so I think one wants to keep those two things separate the second thing it has to do with the returns the private returns to education we know and this is part of what makes the lack of accessibility to higher education so poignant in the United States that those who do not get a college education have seen their income fall dramatically over the last 20 years median income of high school graduates I can't remember the exact number but it's like 20% below adjusted for inflation so they're caught now in this impossible situation where if they don't go to college they know that their life prospects are really dismal they're going to be worse off than their parents and the idea that every generation be at least as well off as your parents better off but the notion they know that they will be worse off than their parents will get a college education but on the other hand if they do get a college education they're saddled with these real burden of tax of debts and they're not sure they're going to be able to repay and there's still some uncertainty what I described as the average returns but there's a lot of variability in those returns and that's exactly that variability is the reason that incontinent loans are so important now you raised a lot of questions about the design of our university system and I think that we are both the United States all the advanced countries should be thinking a lot more about how we design the new university the new set of educational institutions and I think you're absolutely right the same kind of education is not appropriate for everybody we want to have very, very kinds of educational institutions there are some interesting going back to the question of MOOCs there's one university in the United States a state university Arizona State University that has become thoroughly committed to the issue of diversity and performance and efficiency so you know describing itself as a new American university so what it has succeeded in doing is getting a student body this is Arizona State University that is reflective of the population almost the only state that has succeeded in doing it it's done it through a process of inclusion rather than exclusion most of the universities try to exclude difficult populations they've gone out and recruited African-Americans, Hispanics but realize that retention rates are really a problem America has fairly good enrollment rates but retention rates are really miserable and they are not at the top of the OECD they're actually fairly mediocre and so they've said this is a real problem that our education system has to be able to retain these students who may not have the right background and they've shown that you can do it and good success they've obviously at the graduation ceremony last May it was really quite amazing it was held in a football stadium some 12,000 people graduating at one time and the school president asked how many students of this group were first-time college graduates that is to say their parents had not graduated 95% raised their hands he asked how many of them had worked their way through college again about 95% raised their hands they had really worked to get this kind of diversity and openness that was really successful it was a real commitment and then part of their success was using sophisticated online teaching tools with artificial intelligence so they could adjust the learning experience to every child so it was a much more adaptive learning experience they had a good computer department and they were really excited really interesting thing about the university you visited their budget the state allocation to the university had been cut in half and they doubled the number of students the faculty were all happy and the reason the faculty were happy is partly that they really felt committed to this mission it wasn't about selling more products it wasn't a price competition it was a social mission and it was really motivating the entire faculty to feel like they were actually doing something to help transform our society and it's a model that's going around to other universities thank you Mayor Duterte with the University of Canberra my question goes to the appropriate income unit we know that with HEX we've got the individual as the income unit and for good reasons I understand that it was based on the lack of evidence that there's widespread sharing of family income or indeed inequities if that doesn't happen so it's not an issue it's not a complexity I think when you're talking about unemployment in the American context with unemployment insurance the individual also is the unit in Australia as Peter explained we have family we have a different type of system based on benefits which isn't insurance based so we don't we have family as the unit for social security the family unit the marital unit the marital income so how would unemployment benefit work if we did have it in an income contingent context if I'm about to go to work and I've got a rocky relationship with my husband and I'm searching for work who's going to take the burden of that loan if we separate and may he not put pressure on me to take the first job well all the complexities of family relationships what I would say is that probably from a pragmatic point of view certainly reflecting American society that we no longer know what a family is so figuring that out would be too complex for us and that the individual there are all kinds of relationships all kinds of some states allow same sex marriage some do not but the people lived as if they were married so in the context of the United States I would take the view that it has to be the individual because we've messed up our legal frameworks to have a reasonable household unit as the basis but it seems to me that that there is marriage is a contract and you could say part of the understanding of the marriage would be you could have partial insurance across the family and that would determine both eligibility and repayment so as I say I haven't really thought about how you would do that except to say I know it wouldn't work in the United States but I just completed a PhD on foreign aid in the state building at ANU Professor, do you think there is a problem with the current universal higher education system this is a similar question which was raised in a broader context as you mentioned that it seems that higher education is not the key for better opportunities and especially when it comes I mean the problem to the relationship between education and market what you mentioned in the context of the US there is similar problems in other contexts especially in emerging markets do you think if there is market failure and there is need for intervention if so how well there clearly are important market failures in the education sector it's rife with market failures as I mentioned before people don't know what they're getting there's inability to ensure the risk of the return there are huge risks of the investment and there's lack of access to capital so it's a real good example of a sector in which there are multiple market failures there are other market failures that were alluded to before the fact that in signaling models social returns may markedly different from private returns so it is an enormously complex arena but I guess what I would say is that we do know this that overall that access to education access to education is one of the primary determinants to higher income you know there are other determinants if you're lucky enough to have a billionaire father that helps you don't need to go to school but if you don't have the good luck of having a billionaire father education is clearly one of the important ways of getting a higher income society's are getting increasingly complex and that complexity requires high levels of education now obviously the design of the education system today has to be different from what it was a command of greek and latin is not going to suffice in the modern world and we've changed our education system in 19th century so a command of the classics was what was required to get a good job now we expect people to have other skills and some of the people that are doing really well have all kinds of computer skills and technology and science so what are the skills that have high social returns or private returns are going to be constantly discussed and will differ in different countries and in different context and that's part of what education systems you know part of the problem of education systems while they're supposed to be about knowledge and learning they're very resistant to change and they actually are fairly conservative and that's they have to constantly be pushed to say okay well the things that were important to learn 50 years ago may not be the things that are important to learn now and that's going to be an ongoing discussion in terms of restructuring the education system restructuring not only in terms of content but in terms of the question of the modes of delivery that will also have to change so it ought to be and I think in many ways is a very dynamic sector and like chance to talk about the labor market changes in the labor market it used to be the case that when you graduate for instance GM ran its own school to train people to be auto engineers and it could do it because people who were in the auto industry in the GM it was a lifetime job they would make the investment and they get the return of investment over the next 40 years now the average duration of a job average person has about 7 jobs in their lifetime we've changed the structure of the labor market that has some profound implications that means that much of what we ought and something else that's changed very dramatically is the internet it used to be that our style was to try to stuff as much knowledge in the people's brain knowing that in 30 years it would be obsolete but then we could throw the people in the garbage dump at least that's what we do in America but now with the internet the issue is not stuffing your brains with knowledge because you can get instant access to a whole variety of knowledge through the internet the real question is the ability to access that to make judgments about what is good information, what is bad information to assess the relevance of it to put it all together the ability to engage in lifetime learning that you can constantly retool to rethink that's a very different model so that means that what the university does will have to change that more the education is going to happen after in some sense the university, universities will have to focus more on this ability to learn and the ability to have the context in which to assess various ideas that more of the learning will be done through MOOCs that will be focused on lifelong learning of skills after you're on the job because the employer won't be doing that for you because they know that you're only going to be there for a few years so there will be more burden placed on the individual in that lifelong learning process and less on the employer so there are really profound changes in the structure of our education learning system that are going to have to occur in the coming decades and we're just at the beginning of this process and I think it's going to be a very important one but whatever the form finance is going to be important and having a system of finance that is equitable is going to be critical Thank you Tom and thank you Joe and thanks everyone for coming along as my name is Richard Dennis I'm the director of the Australia Institute and we're co-hosting a workshop this afternoon which extends on the work of Bruce's book and all the things that Joe's been talking about today you know I run a think tank which is a strange thing to do but the purpose of a think tank is to build a bridge between ideas and policy and I think it's I'm particularly proud to be here today because I think the particular issue of Hex shows how ideas really can matter and to have someone Professor Stiglitz is standing to come here and say wow look what one idea in Australia did and look how that might shape things around the world well for me it's quite inspirational in terms of why I think ideas matter I'm an escapist from academia I still come and do a bit of teaching here but I guess I focus at the implementation and persuasion end of ideas but Bruce dragged me into this project and we've been talking about Hex for 15 years and there are different elements of it but I'm so glad that I was involved in the Bangkok event which in turn led to the book which in turn led to today so I just want to say something brief about Bruce and then something brief about Joe a lot of people have thanked Joe for getting them into economics today I'm not one of those I'd already gone into economics before I heard of Joe and I went to first year uni and I failed dismally it was 1988 it was the last year of free education and then Hex came along and I got all distinctions so thank you Bruce thank you if it hadn't been for you I don't know what would have happened but price as a signal clearly worked so so thanks to Bruce again for encouraging me to go to the event last year just to give you a quick example of the transactional costs that Joe was talking about in Australia we force people to spend 9% of their income buying superannuation it's any product we force people to buy 9% of their income and the private sector who administers that accumulates $22 billion a year, 1.5% of GDP in fees cost them 1.5% of GDP to manage the 9% of income that everyone here is forced to put in every Australian has a superannuation account every Australian has a superannuation account and it costs the private sector $22 billion to administer it Australian tax system every employee has a tax account and it costs the tax office $3.5 billion to administer the entire tax system costs 7 times more to administer compulsory superannuation than it costs to administer all the complexity of compulsory tax now from my point of view I'd like to pay 9% more tax and get a much more generous age pension but that would be inefficient so I agree with everything that you heard today about the potential if we actually sit back and think what are the economies of scope once you invent an incredible information and tax and revenue transaction mechanism what else can you do with it is a really interesting question for me now the hard part how do you say thank you to Professor Stiglitz in a way that he hasn't heard a million times before rightly and still does justice to his contribution not just here today but in doing what I said I set out to work in think tankery to take ideas into the world and change things I think Professor Stiglitz represents the best of academia in terms of having done great research talks to lots of people about it talks to politicians about it tries to change things so I thought how can I say thank you in a way you might remember and I realised well we've got a Nobel Prize winner here in Canberra and his name is Brian Schmidt and he won the Nobel Prize for physics for basically figuring out that the rate of expansion of the universe really was accelerating so I rang and you might not know this but Brian Schmidt is also a wine maker so not only not only does Brian know a thing or two about the universe and how it was created and where it's going but he knows a thing or two about making wine so I asked Brian can I get a bottle of your wine to give to Professor Stiglitz he said Richard on one condition he said I'll give him a bottle of wine and I'll if I don't paraphrase him he said if he can explain to me why bank CEOs earn so much I'll explain to him how the universe really works