UNH Media Relations
EDITORS AND REPORTERS: Goldberg can be reached at 603-862-3385 and firstname.lastname@example.org.
Roland H. O’Neal Professor and Associate Professor of Economics
University of New Hampshire
Reform measures aimed at reducing systemic risks in the financial system must be a core element of the overall regulatory reform process now underway. Chairman Bernanke’s call for such measures is an important step forward. However, like the major reform proposals that have been advanced recently—for example, by the Financial Stability Forum (FSF) and the Group of Thirty (G30)—Bernanke does not mention that systemic risks in the financial system stem in large measure from boom-and-bust fluctuations in asset markets. Instead, he emphasizes the need to deal with a lack of transparency, over-leverage in the system, too-big-to-fail financial institutions, and the procyclicality of capital and regulatory requirements. To be sure, addressing these problems will help to reduce systemic risks.
But, no one would deny that the sharp downswings in housing and equity prices, which followed long upswings that moved excessively far above historical benchmark levels, helped to trigger and fuel the crisis. As the downswings continue, there is a real danger that they may also become excessive and drag the economy and the financial system into an even deeper crisis.
The connection between excessive price swings in asset markets and financial risk and crisis suggests that understanding these price swings should play a central role in shaping current efforts to re-regulate the financial system. However, for the most part, this has not been the case.
There is a feeling that something should be done about excessive asset price swings. But, should policy officials really intervene to influence prices, given the widespread view among most that markets are vastly superior in allocating scarce capital, and if so, how should they do it? History and experience can help in thinking about these questions, but to understand the rationale and future consequences of policy actions, we need a conceptual framework.
To be useful, this framework needs to recognize the inherent dynamism of modern economies; that individuals search for innovative ways to forecast the future and make decisions, that the social context, including institutions and policy, also changes in novel ways.
Yet, contemporary economic models presume the exact opposite. They presume that the future unfolds according to a mechanical rule. In these models, as long as all market participants are “rational,” and transparency and a few other ideal conditions are met, the market will set prices, save for random deviations, to be consistent with the “true” longer-term fundamental values of all assets.
Economists do recognize that markets undergo long swings. But the belief that unimpeded markets composed of “rational” individuals get asset prices exactly right has led to the view that protracted departures from this normal state are “bubbles” that arise only because market participants fall prey to irrationalities, herding instincts, or reliance on technical rules.
Economists’ bubble models thus lead to an extreme view of the role of swings in capitalist economies: they imply that swings are unrelated to the movements of fundamentals and, as a result, serve no useful social function.
This standard theory gives rise to two extreme views of policy in asset markets.
Either the state should leave markets unimpeded, other than ensuring transparency,
and eliminating other market failures. Or, it should extinguish asset-price
swings as soon as they arise, even if this requires massive intervention.
Imperfect Knowledge Economics provides an alternative view that implies that price swings play an indispensible role in guiding society to allocate capital to alternative projects and companies. Psychological elements and technical trading may play a role. But, markets undergo swings even if everyone bases their trading solely on fundamental factors. Price swings lie at the heart of what markets do, so eliminating them as soon as they appear, as the bubble models would have us do, makes little sense.
Markets are not perfect, however, and as everyone else, participants must cope with imperfect knowledge about how to interpret fundamental factors in forecasting future returns. This can be shown to imply that price swings can sometimes become excessive, in the sense that participants may bid prices far from levels that are consistent with the longer-term values of assets. It is this possibility of excess that provides the rationale for policy intervention in asset markets. It also has important implications for how regulators should measure and manage systemic risk in the financial system.
Imperfect Knowledge Economics has a number of implications for prudential policy in financial markets. On the one hand it acknowledges that, within a reasonable range, the market would do a far superior job, though not perfect, in setting prices. On the other hand, it recognizes that price swings can become excessive and that this excess is socially costly. Consequently, policy officials need to be on guard for excessive swings in asset markets and be ready to dampen them with a set of prudential measures. The framework suggests an array of such measures.
Of course, to implement prudential policies, officials must be able to ascertain, with reasonable confidence, when prices have moved too high or too low from levels that are consistent with longer-term prospects. That is, they need what we call a “guidance range” of non-excessive values. If prices fluctuate inside this range, then policymakers should allow them to do so freely. However, if they move beyond this range, then prudential measures aimed at dampening these excessive movements should kick in gradually.
The idea is not confine asset prices to any target ranges; such polices in currency markets have been shown to always fail in crisis. But, although we cannot eliminate excessive price movements, we can dampen them. Unfortunately, because contemporary theory presumes that “true” fundamental values are known, we are seriously behind in our understanding of how to formulate guidance ranges in various markets, and how to formulate the measures that kick in when excess arises.
Just to illustrate this point, it was clear to many observers, most notably Alan Greenspan, Robert Shiller, and John Campbell, that by 1998 the long swing in U.S. equities had become excessive. If at that time, the Federal Reserve had had the research and operational procedures needed to announce a guidance range, and had implemented the tools that we discuss in a recent paper, they would have likely dampened the upswing and reduced systemic risks in the system.
These considerations pertain not only to what we could have done in the past or perhaps what we should do in the future. We now face an analogous danger; the current downswings in housing and equity prices could go too far from below. It is important that officials have measures in place now to help dampen such excessive movements if they come to pass.
In addition to regularly announcing policy ranges, officials could vary margin and capital requirements or rely on other tools if the downswings do become excessive. To dampen an excessive downswing, policy should encourage the trading behavior of bulls, who are helping to bid prices up, and discourage the behavior of bears, who are betting on a continuation of the downswing. For example, when prices move below the guidance range, margin and capital requirements should be lowered for bulls and raised for bears.
Of course, the opposite should be done in situations in which prices have moved beyond the guidance range from above. For example, margin requirements in these situations should be increased for the bulls and decreased for the bears.
This policy would be easy to implement and make transparent. But, it has been overlooked in policy discussions. Perhaps this is because the vast majority of economic models account for asset price fluctuations with just one set of views, that of the so-called “representative agent.” Such models are unable to target policies differently to bulls and bears.
A general lesson here is that there is a problem with restrictions on short-selling, such as the uptick rule, and other such measures that do not differentiate between bulls and bears and whether the long swing is excessive from above or below. Such policies could actually lead to greater instability not less.
Bernanke’s call to address systemic risks in the financial system is an important one. However, until reforms recognize that these risks stem in large part from excessive asset price swings, they will be woefully inadequate.