Hearing on Systemic Risk and Regulation
Testimony of
Alex J. Pollock
Resident Fellow
American Enterprise Institute
To the Committee on Financial Services
United States House of Representatives
Hearing on Systemic Risk and Regulation
October 2, 2007
Mr. Chairman, Ranking Member Bachus and members of the Committee, thank you for the opportunity to be here today. I am Alex Pollock, a Resident Fellow at the American Enterprise Institute, and these are my personal views. Before joining AEI, I spent 35 years in banking, including twelve years as President and CEO of the Federal Home Loan Bank of Chicago. I am a Past President of the International Union for Housing Finance and a director of three companies in financial businesses.
My career has included many credit cycles which involved issues of systemic risk, from the credit crunch of 1969, the commercial paper panic of 1970, and the real estate investment trust collapse of 1975 (in which the entire commercial banking system was thought by some to be insolvent) to the current example of the credit panic triggered by the ongoing subprime mortgage and housing bust, and a number of others in between. Moreover, I have studied the long history of such financial events and their recurring patterns.
Booms and Busts in Context
To begin with, let me try to put the issues of financial booms and busts and the related question of systemic risk in context.
The fundamental principle is that long term growth and the greatest economic well being for ordinary people can only be created by market innovation and experimentation. Markets for goods and services must be accompanied by markets in financial instruments, which by definition place a current price on future, thus inherently uncertain, events. This much is obvious but easy to forget when addressing the results of a bust with the benefit of hindsight, when it seems like you would have to stupid to make the mistakes that smart people actually made.
Dealing with putting prices on the inherently uncertain future, all financial markets are constantly experimenting with how much risk there should be, how risks are distributed, what the price of risk-bearing should be, and how risk trades off with financial success or failure. Should individuals and institutions be free to take financial risks if they want to? Yes, they should.
In the boom, many people succeed, just as many people succeeded for a long time in the subprime mortgage and housing boom. This success gets extrapolated, supports optimism and makes lenders and investors, including private pools of capital, confident. Lender and investor confidence tends to the underestimation of risks, in particular, the risk that the price of the asset in favor, most recently houses, could fall or fall very much; and underestimation of the risk that if prices fall, especially in a leveraged sector, asset and credit markets could become illiquid.
In my opinion, the principles stated by the President’s Working Group for private pools of capital are professional and sensible. But even if everybody followed them, we would not avoid the inevitable times of financial turbulence.
We know for certain that markets will create long term growth and also cyclical booms and busts, but just what or when the outcome of a particular innovation will be cannot be known in advance. It can only be discovered by running the market experiment, as so brilliantly discussed in Friedrich Hayek’s “Competition as a Discovery Procedure.”
How hard it is to outguess this discovery procedure is shown by the fact that a mere three months ago, the financial and economic world was constantly treated to statements by very intelligent and well-informed people that there was “abundant liquidity” or even a “flood of liquidity,” which would guarantee a firm market bid for risky assets and narrow spreads. Then we were suddenly confronted with a lack of bids, nonfunctioning markets and the “evaporation of liquidity.”
Likewise, some very intelligent and well-informed people said, up until August, that the subprime mortgage bust would be “contained” and not cause wider financial or economic problems. Now we have had a subprime-induced credit panic and an ongoing credit crunch, with falling house prices, but the stock market has gone back up to near its high. How do we interpret that?
A fundamental point is that markets are recursive. Whatever opinions influence buying and selling and hedging, whatever models of financial behavior are relied on, whatever is done to regulate them, are all fed back into the system of interactions and change behavior in unpredictable ways. Thus models of financial behavior, themselves changing the market, tend to become less effective or obsolete, as did subprime credit models.
Regulations likewise change financial behavior, are arbitraged, and may end up producing the opposite of their intent. This is why regardless of what any regulator or legislator may do, markets will always create however much risk they want. Then when the bust has begun, regulatory actions to reduce or control risk may turn out to be procyclical, reinforcing the downward momentum.
Models
To successfully avoid booms and busts, regulatory operations or market actors would have to know the future. They often attempt to do so through creating models.
Of course, there is always a difference between financial models, however mathematically refined, and financial reality. This is so whether the models are those of Wall Street “rocket scientists” structuring securities, credit rating agencies, hedge funds or other private pools of capital, sophisticated institutions, the Federal Reserve or other regulators, or investment analysts. Finance cannot in principle be turned into physics.
John Maynard Keynes memorably observed that a prudent banker is one who goes broke when everybody else goes broke. One way to do this is to use models with the same assumptions that everybody else has. Then you can be confident when everybody else is confident and afraid when everybody else is afraid. (We can be skeptical of the models approach of Basel II in this respect.)
Once a Decade, On Average
The classic patterns of booms based on credit overexpansions and their following busts are colorfully discussed by such students of financial cycles as Charles Kindleberger, Walter Bagehot and Hyman Minsky.
Kindleberger, surveying several centuries of financial history, observed that financial crises and scandals occur, on average, about once every ten years. This matches my own experience. Every bust is followed by reforms, but the next bust arrives nonetheless. Still the trend of market innovation and long term growth continues.
The increased risk accumulated in credit overexpansions ultimately comes home to roost and prices of the favored asset fall. There is a hangover of defaults, failures, dispossession of unwise or unlucky borrowers, revelations of frauds and swindles (always), and then the search for the guilty. There is a sharp restriction of credit. For example, the chief executive of Countrywide recently announced, “We are out of the subprime business.”
There is a generalized retreat from risky assets, and a new danger arises: fire sales of assets turning into a debt deflation and the ruin of the financial system—systemic risk has arrived. Our students of financial cycles all support government intervention to stabilize the downward momentum. This is the correct answer as long as it is temporary.
To come to the current situation, it is evident that the present combination of the excess inventory of houses and condominiums, with the rapid restriction of mortgage credit—in other words, increased supply plus falling demand, equals a trend of falling house prices. The models used to analyze, rate and price subprime securitizations include as a key factor house price appreciation (“HPA” in the trade jargon). Now that we have house price depreciation, what will happen if prices fall much more and much more broadly than the models, the investors, the lenders and the regulators thought they could?
Unfortunately, a vicious cycle of falling house prices, more defaults, further credit tightening, less demand, further falls in prices, more defaults, and so on, is possible for a while, though of course not forever. Financial market result: Fear.
The fear is increased by great uncertainty about the value of subprime securities if no one wants to buy them anymore. What are they worth as assets to an investor, notably a leveraged investor? What are they worth as collateral to a lender—especially a very risk-averse repo dealer or commercial paper buyer?
Greater disclosure and transparency are reasonably suggested, although financial accounting, at least, is never truly “transparent.”
For example, what does “value” even mean when there are few or no buyers? How can assets be marked to market if there is no active market? Should everybody’s portfolio be marked to fire sale prices, or instead to some estimate of intrinsic value? Who is actually broke and who isn’t? The answers to these classic questions of the bust are never clear, except in retrospect.
Liquidity
As Bagehot wrote, “Every great crisis reveals the excessive speculations of many houses which no one before suspected.” So has our current bust, and these unpleasant surprises reinforce the uncertainty make about who is broke and who isn’t (perhaps including yourself). With this uncertainty and personal as well as institutional risk, everyone becomes conservative at once. When all investors and lenders, institutionally and personally, try quite logically to protect themselves by avoiding risk, the result is to make liquidity disappear and to put the whole at risk. Note that possibility of regulatory or political punishment arising from the search for the guilty will increase the risk aversion.
In other words, it is belated risk aversion which creates systemic risk. To understand why this can happen, we have to see that “liquidity” is not a substance which can “flow,” be a “flood,” “slosh around,” or be “pumped” somewhere, to use a number of misleading expressions.
In fact, liquidity is a figure of speech. It is verbal shorthand for the following situation:
-A is ready and able to buy an asset from B on short notice
-At a price B considers reasonable
-Which usually means C has to be willing to lend money to A
-And if C is a dealer, both A and C have to believe the asset could readily be sold to D
-Which means A and C believe there is an E willing to lend money to D.
Good times, a long period of profits, and an expansionary economy induce financial actors and observers to take this situation, “liquidity,” too much for granted, so liquidity comes to be thought of as how much you can borrow. When the crisis comes, it is found to be about what happens when you can’t borrow, except from some government instrumentality.
At this point we have arrived at why central banks exist. The power of the government, with its ability to compel, borrow, tax, print money, and credibly guarantee the payment of claims, can intervene to break the everybody-stops-taking-risk-at-once psychology of systemic risk.
The key is to assure that this intervention is temporary, as are credit panics by nature. As historically recent examples of government interventions in housing busts, since 1970 we have had the Emergency Home Finance Act of 1970, the Emergency Housing Act of 1975, the Emergency Housing Assistance Act of 1983, and the Emergency Housing Assistance Act of 1988. (I do not count the Hurricane Katrina Emergency Housing Act of 2005, a special case.) This is in line with Kindleberger’s estimate of about once a decade on average, and an emergency housing act of 2007 would fit the pattern.
The liquidity crunch won’t last forever. Large losses will be taken, the market get used to the idea, who is broke and who isn’t sorted out, failures reorganized, risks reassessed, models rewritten, and revised clearing prices discovered. A, B, C, D and E will get back into business trading and lending to each other again.
Liquidity will return reasonably quickly for markets in prime instruments. One long time observer of finance, whose insights I value, has predicted that “the panic about credit markets will be a memory by Thanksgiving.”
I believe this is probably right; however, the severe problems with subprime mortgages and securities made out of them, related defaults and foreclosures, and falling house prices will continue long past then. They will continue to cause macroeconomic drag and financial difficulties, but the moment of systemic panic will have passed.
The “Cincinnatian Doctrine”
In conclusion, my view is that it is not possible to design society, no matter what regulatory systems may be implemented, to avoid financial booms and busts and their resulting risk of systemic panics. We do need temporary interventions of the government periodically, when the financial system is threatened by a downward spiraling debt deflation. In other words, booms and busts are endemic to market economies with financial markets in which people are free to take risks and engage in borrowing against the uncertain future. They are a price well worth paying in return for the innovation and growth only such markets can create.
In normal times, that is, about 90% of the time, we predominately want the economic efficiency, innovation, productivity and the resulting well-being for ordinary people produced by competitive markets. But when the financial system hits its inevitable periodic crises, about 10% of the time, the intervention of the government is often necessary. This intervention should be temporary. If prolonged, it will tend to cartels, bureaucracy, less innovation, and less growth. In the extreme, of course, it becomes socialist stagnation.
Thus I suggest a 90%-10% policy mix. I have elsewhere explored this idea as the “Cincinnatian Doctrine.”
In the wake of every bust, various plans are put in place to prevent all future ones, but the next bust arrives in about ten years anyway. Such plans suffer from the assumption that financial group behavior is mechanistic and can be addressed by designing mechanisms. In fact, it is organic, creative, recursive and emergent. That is the source of its strength in creating wealth, also of its weakness in getting periodically carried away. I do not believe any regulatory structures can alter these fundamental characteristics.
Thank you again for the opportunity to share these views.