Saturday, July 10, 2010

The destruction of prudence: understanding the global financial crisis

Prof. Russ Roberts of George Mason University has produced the best single analysis (pdf) (via) of the global financial crisis (GFC) I have read: empirically grounded, with careful, perceptive analysis leavened with moral outrage. (And if you do not have some level of moral outrage, you do not know what went on.)

Over the last 15 years or so, I have had various conversations with people about how the IMF has been injecting moral hazard into the international financial system by acting as “welfare for Wall St” and bailing out banks who loan to developing countries: bailouts which have usually involved major costs to the taxpayers of said countries.

This turns out to be a more general problem. President Bush put it nicely in conversation with the Chairman of the Federal Reserve and his Treasury Secretary:
Someday you guys are going to have to tell me how we ended up with a system like this. I know this is not the time to test them and put them through failure, but we’re not doing something right if we’re stuck with these miserable choices.
(I wonder if he did ever get a good answer.)

Roberts is about finding underlying causes:
Yes, deregulation and misregulation contributed to the crisis, but mainly because public policy over the last three decades has distorted the natural feedback loops of profit and loss. As Milton Friedman liked to point out, capitalism is a profit and loss system. The profits encourage risk taking. The losses encourage prudence. When taxpayers absorb the losses, the distorted result is reckless and imprudent risk taking.
In other words, a system not merely suffering from moral hazard, but increasingly built on it:
The most culpable policy has been the systematic encouragement of imprudent borrowing and lending. That encouragement came not from capitalism or markets, but from crony capitalism, the mutual aid society where Washington takes care of Wall Street and Wall Street returns the favor. Over the last three decades, public policy has systematically reduced the risk of making bad loans to risky investors. Over the last three decades, when large financial institutions have gotten into trouble, the government has almost always rescued their bondholders and creditors. These policies have created incentives both to borrow and to lend recklessly.
At the level of social systems, there is a very large difference between reducing risk and suppressing risk. Reducing risk means one is lowering the level of risk in the general system. One can see that happening over time with the downward historical trends in interest rates, for example.

Suppressing risk
Suppressing risk shifts the risk from certain actions (and thus particular agents) to somewhere else: generally into more concentrated forms in the future. From the beginning of Federation until the early 1980s, Australia was under the Deakinite policy regime of industry protection, wage arbitration, state paternalism, white Australia and imperial benevolence.

Each part of the Deakinite regime was intended to be about reducing risk. Industry protection blocked foreign competition, wage arbitration blocked (low) wage competition, state paternalism reduced risks of aging and ill-health, white Australia blocked competition from “tropical zone” migrants and imperial benevolence sought the protection of friendly Great Power (first Britain, then the US). White Australia was the first part of the policy regime to disappear, being abandoned in the 1966-72 period. It arose because, during the nineteenth and first part of the twentieth century, there were two global labour flows. “Temperate zone labour” from Europe to North America and the Antipodes and “tropical zone labour”, particularly from China and India to various European colonies. It was an urgent demand of the North American and Antipodean working class that it be shielded from competition from “tropical zone labour”, often using the available prop of racial ideology.

As a policy regime, the Deakinite policy regime suppressed risk rather than reducing it in any systemic sense. And it suppressed risk by building a lot of rigidities into the economy. Consequently, the Australian economy handled external economic shocks badly because it lacked flexibility. The greatest achievement in the dismantling of most of the Deakinite policy regime has to been create a much more flexible Australian economy, which has handled both the 1997 Asian Crisis and the more recent GFC and Great Recession not merely well, but better than comparable countries.

[Under the Deakinite regime, regulation and other government intervention blocked the flow of information by blocking action: which inhibited the development of better ways of doing things and effective action in response to sudden changes. The latter in particular led to the Australian economy tending to deal poorly with economic shocks in the period.]

What the history of bailouts and implicit government guarantees that Roberts sets out did is suppressed risk by reducing risk for individual actors thereby massively increasing the level of risk in the financial system as a whole. Suppressing risks for individual acts, which therefore discouraged attention to risk, massively increased the level of systemic risk. [The pattern of action did not reflect the risks inherent in particular investments, as it suppressed seeking such information and acting upon it. What it actually reflected was the implicit government guarantees. Taxpayers were the "backstops" for the actions of others.]

Destroying prudence
Roberts sets out the perverse incentives created:
The punishment of equity holders is usually thought to reduce the moral hazard created by the rescue of creditors. But it does not. It merely masks the role of creditor rescues in creating perverse incentives for risk taking.
These perverse incentives became pervasive, which is precisely the problem:
Because of the large amounts of leverage—the use of debt rather than equity—executives can more easily generate short-term profits that justify large compensation. While executives endure some of the pain if short-term gains become losses in the long run, the downside risk to the decision-makers turns out to be surprisingly small, while the upside gains can be enormous. Taxpayers ultimately bear much of the downside risk. Until we recognize the pernicious incentives created by the persistent rescue of creditors, no regulatory reform is likely to succeed.
Almost all of the lenders who financed bad bets in the housing market paid little or no cost for their recklessness. Their expectations of rescue were confirmed.
And what we do is make it easy to gamble with other people’s money—particularly borrowed money—by making sure that almost everybody who makes bad loans gets his money back anyway. The financial crisis of 2008 was a natural result of these perverse incentives.
Roberts uses the analogy of a poker game with Uncle Sam watching (sometimes changing the rules, sometimes guaranteeing players) to illustrate the problems very effectively.

What was going on was not “managing” the market system, it was a profound subverting of it:
Capitalism is a profit and loss system. The profits encourage risk taking. The losses encourage prudence. Eliminate losses or even raise the chance that there will be no losses and you get less prudence. So when public decisions reduce losses, it isn’t surprising that people are more reckless.
Read More...There is also another factor which is implicit in Roberts’ analysis, but he does not bring out. If access to decision-makers increases one’s returns, that will advantage firms with access to decision-makers. The flooding of funds into the financial system and financial markets is hardly surprising in the circumstances. An expansion particularly driven via the underpinning of massive levels of leveraging. Thus:
Without extreme leverage, the housing meltdown would have been like the meltdown in high-tech stocks in 2001—a bad set of events in one corner of a very large and diversified economy.
This was a crisis building for a long time:
The [1984] rescue of Continental Illinois and the subsequent congressional testimony sent a signal to the poker players and those that lend to them that lenders might be rescued.
A signal that was reinforced again and again by subsequent government action:
Continental Illinois was just the largest and most dramatic example of a bank failure in which creditors were spared any pain. Irvine Sprague, in his 1986 book:
“Of the fifty largest bank failures in history, forty-six—including the top twenty—were handled either through a pure bailout or an FDIC assisted transaction where no depositor or creditor, insured or uninsured, lost a penny.”
The 50 largest failures up to that time all took place in the 1970s and 1980s. As the savings and loan (S&L) crisis unfolded during the 1980s, government repeatedly sent the same message: lenders and creditors would get all of their money back. Between 1979 and 1989, 1,100 commercial banks failed. Out of all of their deposits, 99.7 percent, insured or uninsured, were reimbursed by policy decisions.
This had utterly predictable effects on the attention of actors in the financial system to levels of risk:
… all profit and no loss make Jack a dull boy.
[The information role of markets and prices were systematically distorted.] It was the systematic destruction, by government policy, of prudence in the financial system:
Each case seems different. But there is a pattern. Each time, the stockholders in these firms are either wiped out or see their investments reduced to a trivial fraction of what they were before. The bondholders and lenders are left untouched.
Tracing the effects of this does have some evidentiary problems. No one is likely to say “yes, I was reckless!” But:
While direct evidence is unlikely, the indirect evidence relies on how people generally behave in situations of uncertainty. When expected costs are lowered, people behave more recklessly.
This was a problem not only in the US. The UK authorities were playing the same game:
The only difference between this scenario in the United Kingdom and the one in the United States is that in the U.S. the Fed came to the rescue and the executives, for the most part, kept their bonuses.
That equity was not guaranteed made much less of a difference than one would imagine. As Roberts points out, equity investors tend to diversify and could use bonds to reduce downside risk.
Roberts cites a study on the perverse incentives facing executives and concludes that:
This kind of looting and corruption of incentives is only possible when you can borrow to finance highly leveraged positions. This in turn is only possible if lenders and bondholders are fools—or if they are very smart and are willing to finance highly leveraged bets because they anticipate government rescue.
He then considers the case of a couple of the CEOs whose firms were rescued by the government (i.e. the taxpayer):
When we look at Cayne and Fuld, it is easy to focus on the lost billions and overlook the hundreds of millions they kept. It is also easy to forget that the outcome was not preordained. They didn’t plan on destroying their firms. They didn’t intend to. They took a chance. Maybe housing prices plateau instead of plummet. Then you get your $1.5 billion. It was a roll of the dice. They lost.
When Cayne and Fuld were playing with other people’s money, they doubled down, the ultimate gamblers. When they were playing with their own money, they were prudent. They acted like bankers. (Or the way bankers once acted when their own money or the money of their partnership was at stake.)
In other words, behaviour was different in the realms were public policy was not systematically destroying prudence.

Roberts examines various explanations offered for the GFC, concluding that:
These explanations all have some truth in them. But the undeniable fact is that these allegedly myopic and overconfident people didn’t endure any economic hardship because of their decisions. The executives never paid the price. Market forces didn’t punish them, because the expectation of future rescue inhibited market forces. The “loser” lenders became fabulously rich by having enormous amounts of leverage, leverage often provided by another lender, implicitly backed with taxpayer money that did in fact ultimately take care of the lenders.
The systematic destruction of prudence as moral outrage.

Perverse incentives
Roberts then examines the (largely regulation driven) perverse incentives in housing finance and purchase in various markets before examining Fannie Mae and Freddie Mac:
But between 1998 and 2003, Fannie and Freddie played an important role in pushing up the demand for housing at the low end of the market. That in turn made subprime loans increasingly attractive to other financial institutions as the prices of houses rose steadily.
Fannie and Freddie had particular value for policy makers:
… one other group sitting at the table playing with other people’s money: politicians. Politicians are always eager to spend other people’s money. It’s what they do for a living. But it’s an even better deal for politicians if they can hide the fact that they’re spending other people’s money or delay when the bill comes due. That’s what they did with Fannie and Freddie.
We can see clearly from Roberts’s analysis that the suppression of risk actually massively increased the total level of risk in the system. In the housing market, there were massive increases in riskier housing loans.

The story Roberts tells is one of consistent rationality creating perverse incentives, rather than having to posit various levels of irrationality. In particular, he asks the excellent question of why specific types of assets were invested in and not others.

Roberts shows quite clearly that the issue is not merely problems with computer models used in assessing risks, it is the incentives driving the creation and use of models.

How was prudence destroyed? By making the taxpayers the backstop, the guarantors of what wealthy, powerful and connected people were doing:
As in the Fannie and Freddie story, the firms aren’t the real financers of the salaries associated with picking up nickels. The taxpayers ultimately fund picking up of nickels, and the taxpayers get flattened.
In Yoram Barzel’s property rights analysis of firms, the boundary of the firm is the boundary the guarantee offered by the equity capital. The owner of the risk exposure in lending is whoever provides the debt guarantee. That turned out to be the taxpayers. The system systematically nationalised risk, and operated as chaotically as any command economy does. [For it had the same underlying structure: those making the decisions about capital did not have to bear the costs of what they did while information was being systematically blocked or distorted.]

In particular, it destroyed prudence.

Bloating perversity
Part of what went on was the creation of complex financial instruments whose attributes were not clear to the “owners” of said instruments. But, if someone else is providing the ultimate guarantee, how hard are you going to bother to look (or even worry)?

Roberts’ analysis leads him to a very scary place:
An unpleasant but unavoidable conclusion of this paper is that Wall Street was (and remains) a giant government-sanctioned Ponzi scheme.
But, if one wants to know how the financial system became so huge, then this analysis explains why. The advantages of access to policy makers both causing, and combining with, the implicit (and increasingly explicit) taxpayer guarantee. As Roberts notes:
There is an old saying in poker: If you don’t know who the sucker is at the table, it’s probably you. We are the suckers. And most of us didn’t even know we were sitting at the table.
This is not a good place for democracy to be:
Rescuing rich people from the consequences of their decisions with money coming from average Americans is bad for democracy.
Just as the IMF has been bad for global governance because it has been doing the same thing, only more so, and gouging taxpayers a lot poorer than the average American taxpayer.

The consequences for democracy are indirect, the consequences for the economy much more direct:
Rescuing people from the consequences of their decisions is bad for capitalism.
What we do not need is replacing “bad” people with “good” ones. We need better ideas and much better public policy incentives.

Maybe it is as simple, as the cases of Canada and Australia suggest, of just having good prudential regulation in the first place. (A model, which does not, alas, work for the IMF.)

The massive destruction of prudence, and the subsequent creation of high levels of systemic risk (including sovereign debt issues—themselves an issue of a lack of fiscal prudence—which Canada and Australia have also managed to avoid), perhaps explain the fearfulness of monetary and banking authorities, such as the Bank for International Sentiments recently endorsing fiscal and monetary tightening in a situation of serious deflationary pressures. (Since one would have to be something of an obsessive idiot to be worrying about inflation in the current situation: though (re)fighting the last war you won has notoriously held a certain attraction.)

Roberts quotes Milton Friedman putting the point well about not just hankering for the “right” people:
The way you solve things is to make it politically profitable for the wrong people to do the right things.
That is, after all, the way Madison designed the constitutional structure of the American Republic. What is needed is ways of discouraging politicians from continuing to systematically destroy prudence in the financial system.

ADDENDA: Roberts' analysis can be taken as something of a case study for Jeffrey Friedman's argument (pdf) (via) for the cognitive superiority of markets over politics.

FURTHER ADDENDA: Having now read Hayek's The Meaning of 'Competition' (which, Jeffrey Friedman is completely correct, should be read in conjunction with Hayek's The Uses of Knowledge in Society), I have extended the post somewhat by the sections in [square brackets].

No comments:

Post a Comment