Joseph Stiglitz, professor of economics at Columbia University and former chief economist of the World Bank, is now best known as the most famous academic critic of globalisation. However, he won the 2001 Nobel Prize for pioneering work on information asymmetries – how markets misbehave when some participants have more information than others.
He has applied insights from this research to his campaign to introduce default options for pension plans. Savers must have freedom of choice, he believes, but he also believes that the default option should be truly suited to them. His research has now moved on to radical ways of designing default options that are tailored to individual needs. His Nobel-winning research also helps to understand the credit crisis, as the recent market turmoil was often bedevilled by a lack of information.
JOHN AUTHERS: Why do you support a system of default options? Doesn’t it limit people’s choices?
JOSEPH STIGLITZ: It is a recognition that individuals typically don’t have the basis for making an informed judgment. You could make the case stronger than that. The economic theory of rationality is really informed by the notion that people learn from experience what their preferences are. How do you decide if you like red lettuce or green? You taste them and you find out. If red lettuce is more expensive, you decide whether it is worth that to you. After repeated buying, maybe you see you made a mistake.
But the nature of the decision on saving for retirement means that that is no longer an option. You can’t go through your life and say: “I’ve saved too little; I’ll save more in the next life.” There’s no way that people can learn. It’s even impossible for people to learn from their parents or from society. Social learning isn’t going to work because the world today is so different from the world of our parents. The social security system in the US is stronger. In other countries it’s weaker. Longevity is different. Markets are different.
That makes this decision very difficult. The only way that people can learn is by going through simulation models and more complex optimisation models and seeing what would happen if I saved less or if I saved more. There’s a sense in which there’s a real need for advice. That’s the basic framework.
Moreover, more recent work on behavioural economics (for which Daniel Kahneman received the Nobel Prize in 2002) has shown that defaults matter. An employer who gives his workers three different choices, but randomly assigns to one group one default and another default to another group, will find that workers will systematically pick the default.
JA: OK, this choice is difficult, but aren’t you limiting it?
JS: The second important concept is that it’s a default, not a compulsion. You still leave people with choices. If you are more worried about having nothing for your retirement, you can save more. If you think you are likely to have cancer and you are going to die, you don’t need to save so much. You know more about your characteristics and family situation than we do. But you have a norm. You make judgments on the level of savings and risk taken in relation to that norm.
JA: Lifecycle and target-date funds, which automatically switch from equities and bonds as you get older, are growing in popularity. Is that how you would design the default option? What else do you need to take into account?
JS: It’s a science or practice which is just beginning to develop and much of the financial literature has been limited in its approach in that it has only looked at financial wealth, not other forms of wealth like human capital and housing capital – and for most people those are major parts of their capital.
One of the standard wisdoms of asset allocation is that you should take less risk as you get older, and put more in bonds and less in stocks. There’s wisdom about the ratio of bonds in your portfolio as you get older. In fact, if you look at your overall portfolio, there are a number of criticisms of that particular result. If you looked at it from a total portfolio perspective, then the risk associated with your human capital diminishes to zero as you approach retirement. At 25, you don’t know where you are going to end up. At 65 or 70, there’s very little risk. So if you look at it from this broader perspective, risk is diminishing from your overall portfolio simply because of the human capital. Retirement experts usually only look at one part of wealth management. There’s no simple formula, and I don’t think there ever will be.
JA: If that’s the case, how are you going to come up with an acceptable default?
JS: One objective of my research programme is to try to see if there are broad patterns among individuals. Out of these broad patterns it might be possible to identify individuals who would represent key norms. You would still have a range of options.
Right now the best that most companies offer is that everyone is identical, and you get one norm and a choice or two around that. My programme would say that: “We recognise there are differing situations by family and wealth accumulation and here are some options. Maybe five or ten.”
This is where the research programme is going. We aren’t there yet. We are trying to see empirically, looking across households, if there are in fact clusters of types of people that one can identify. Obviously, we know that people who are single are different from people who are married, and so on. Most of the default approaches don’t have that type of refinement. They are mostly on age.
JA: This is quite a radical rethink. Is the fund management industry going to go along with this?
JS: If you are managing these magnitudes of funds then you want to help your clients. In the long run it will be good for business but in the short run it’s almost a sense of corporate social responsibility.
JA: A critical issue for fund managers at the moment is the ongoing credit crisis. How do you explain what has happened?
JS: I totally predicted this. Securitisation was pushed because of its advantages in risk diversification. But I emphasised in some of my work on securitisation that you have to offset that advantage with the awareness that you are creating an agency problem. And you are creating a potential for asymmetries of information.
In the old days, it was the banks that originated loans and kept the loans. But once you went to securitisation you created the possibility of the originator having different information from the buyer. Not only is there information asymmetry but in this context there are perverse incentives. The originator has an incentive to provide distorted information. The buyers should have been aware of this, but it’s quite apparent that they weren’t as aware of this as they should have been.
This was partly because they bought into the notion of risk diversification – they thought they didn’t need to worry about it because of the law of large numbers. But the law of large numbers says only that you don’t need to worry about a single one; you do need to worry about systemic risks. And securitisation helped create systemic risks.
Moreover, securitisation made restructuring loans (when the borrower couldn’t repay) difficult, if not impossible. In the old days, when the bank originated the mortgage and kept it, when its customer ran into difficulties, it was a much easier matter to restructure. With mortgages sliced and diced, spread among dozens, perhaps hundreds, of “investors”, restructuring is virtually impossible (and made more difficult by some of the provisions in the loan agreements). That’s why the expectation is that there will be 1.7m foreclosures in the coming year – or more. And this will reinforce the downward spiral in the housing market; forced sales will drive down prices.
JA: So can securitisation continue in its current form? What does it take to sort out this mess?
JS: At a minimum, it requires greater transparency. For American financial markets, this is the third scandal in 20 years. And that should itself be telling a story. They’ve gone from one scandal to the next with booms in between each one, and moving from one market to another. It highlights the need for stronger regulation, the importance of asymmetric information, and the role of greed in financial markets.
In the scandals of the late 1990s and in this scandal, there were elements of informed parties taking advantage of the less informed. In this crisis, there’s a clear issue of predatory lending. That’s an underlying problem and the most important social aspect of this.
JA: Are there any lessons for fund managers trying to draw up default savings options?
JS: Yes. One of the problems with subprime mortgages is that they have made these negotiations more difficult; another is the lack of transparency, as they’ve hidden these products into complex products where they slice and dice the risk. They sliced and diced it and then they reassembled it. It wasn’t clear in the process of slicing and dicing and reassembling that they were actually adding much tailoring to the different risk needs of different groups. Everyone was focusing on diversification.
Now we need to begin doing slicing and dicing not to generate revenues for the investment banks but to generate products that really are tailored for the different needs of the different groups. Those different needs aren’t the kinds of second order refinements that the investment bankers were doing. They were very high refinements, but we are talking about very basic needs.
JA: So whose responsibility should it be to offer the defaults? The fund manager’s?
JS: That’s right. You have to realise that there’s a supply chain here and you are trying to intermediate between the needs of the individual investors, and the suppliers who have products out there. In a competitive market, what the sellers of mutual funds need to do eventually is say: “These are the kinds of risk profiles that we need.”
The sellers are at the interface. They are the people who are making a market in a way. Their job is to try to assess the needs of the market and then look out for what are the potential offerings on the supply side.
