“Where’s the Risk?” : Geithner's Talk

Via Justin Fox, another post from Timothy Geithner, this one called "Liquidity Risk and the Global Economy:

"Although there is much that is positive in the world today, there is little reason to believe we have entered a new era of permanent stability. Financial innovation and global financial integration do not offer the prospect of eliminating the risk of asset price and credit cycles, of manias and panics, or of shocks that could have systemic consequences."

Nor will they. However, we can do better.

"Over the past 25 years or so, we have seen a significant number of episodes of financial shocks, in U.S. markets and globally. Although more different than similar in their nature and impact, they had some common features. They were unanticipated. The causes varied, with some associated with a substantial deterioration in the real economy, and others not. And they typically involved the dynamic in which a sharp change in risk perception results in a fall in asset prices and a sharp reduction in market liquidity, and an increase in correlations across asset classes. As market participants move to protect themselves against further losses, by selling positions, requiring more margin, hedging against further declines, the shock is amplified and the brake becomes the accelerator."

I agree that the sharp change in risk perception is key.

"All markets are vulnerable to this dynamic. The degree of vulnerability is not something we can measure with any confidence, for it depends not just on the how market participants behave in the event, but on the complex interaction of a number of different factors. The vulnerability depends, for example, on the size of the asset price misalignment, the conditions that produced it, and the magnitude of the shock to confidence in economic fundamentals. It depends on the scale of leverage in the system, and the incidence of leverage or concentration of exposure to different risk factors. It depends on the diversity of exposures or positions of different financial institutions. It depends on the presence or absence of distortions or market failures, moral hazard being the typical problem. It depends on perceptions of the capacity for monetary policy and other policies to adjust to mitigate the damage."

I like that he focuses on how market participants behave. This is the most important factor.

"The dramatic changes we’ve seen in the structure of financial markets over the past decade and more seem likely to have reduced this vulnerability. The larger global financial institutions are generally stronger in terms of capital relative to risk. Technology and innovation in financial instruments have made it easier for institutions to manage risk. Risk is less concentrated in the banking system, where moral hazard concerns and other classic market failures are more likely to be an issue, and spread more broadly across a greater diversity of institutions."

This is where there's a big disagreement. I feel that government actions, particularly the S & L crisis, had reduced moral hazard to almost nil. The financial institutions believed, correctly, I'd point out, that the government had given both implicit and explicit guarantees to intervene in a financial crisis. There was consequently more risk, because there was no Plan B. It was the government or nothing if things spiraled out of control. Consequently, businesses focused on the more limited crises that might effect them, and ignored the major ones. While this might sound rational, it isn't, because it's impossible to tell when a small problem becomes a major one until you've let it happen. The tendency is, as he has described earlier, to ignore more and more of the serious risk as time goes by, and so fall asleep at the switch, having inadvertently narrowed your range of focus.

"And yet this overall judgment, that both financial efficiency and stability have improved, requires some qualification. Writing a decade ago about the history of the financial shocks of the 1980s and early 1990s, Jerry Corrigan argued that these same changes in financial markets we see today, though less pronounced then than now, created the possibility that financial shocks would be less frequent, but in some contexts they could be more damaging. This judgment, that systemic financial crises are less probable, but in the event they occur could be harder to manage, should be the principal preoccupation of market participants and policymakers today."

Actually, that's a version of what I just said.

"What factors might contribute to this risk, a risk that could be described as the possibility of longer, fatter tails? One reason is a consequence of consolidation. The major banks are larger and stronger today, and it would take a much larger shock to make them vulnerable than was true in the past. But in that event, the consequence of their failure could be greater for markets."

See, I don't agree. They'd understood the meaning of "Too big to fail" in the S & L crisis. They simply worried less about the consequences of massive failure, than about more mundane and contained crises. They were buying a kind of insurance policy from the government by being big. So, I agree with the point, but take a different view of it. They were stronger in smaller crises, much weaker in anything of major import.

"Another reason is the consequence of leverage. Leveraged arbitrage activity, so some of the literature suggests, is likely to reduce volatility in normal times and increase it in times of stress, because of the greater financial constraints faced by leveraged funds relative to larger, more diversified banks and investment banks. Whether this matters in a systemic sense or not depends on the heterogeneity of funds and how correlated their exposures are with those of the major banks and investment banks."

See, to me, leverage means more risk. Period. I don't buy the lessening of risk scenario. The risk always has to come back to the originator, and to them leverage is added risk.

"A third reason is the consequence of long periods of low losses and low volatility. When markets have been through a sustained period of relative stability and low credit losses, this reduces the estimates of potential losses produced by conventional risk management tools. Like gravity, this force is hard to counteract. And when we’ve had a long period of low realized losses, rapid change in markets, dramatic growth in new instruments, and larger potential leverage, this creates a greater possibility of unanticipated losses, and therefore a greater potential for the trend–amplifying, positive feedback effects that have characterized the major asset price shocks of 1987 and 1998."

Because this is the tendency, this is when you should be scared. It's like value investing. Be most afraid in these types of periods. If value investors can do it, we can as well.

"Policymakers do not have the capacity to eliminate the risk of excess leverage or asset price misalignments, nor do we have the ability to act preemptively to diffuse them. And yet policy can play a very important role in limiting the vulnerability of financial markets to the risk that a shock will pose greater potential risks to the stability of the financial system and to economic growth. A few points about both monetary policy and supervision in this context."

It might not be fair, but given what we've just been through, this sounds like a nostrum.

"This approach to supervision is designed to look ahead to try to ensure the adequacy of capital and liquidity, and the sophistication of risk management discipline relative to risk, over time and at all points in the cycle. I outlined earlier the reasons why monetary policy has only limited capacity to deal preemptively with financial bubbles. Similarly, I see little prospect that supervision will have the capacity to identify and address potential concentrations in exposure to individual risk factors, whether through changes to capital charges or other means. Focusing on the quality and strength of cushions against tail risk is the best way for supervision to be countercyclical rather than procyclical."

This I disagree with. It should be possible to significantly decrease risk, with a wide mouthed funnel approach to investments.

"The conditions we see prevailing in global financial markets today reflect a range of different factors, some fundamental and others that are less likely to be enduring. The most effective thing that policymakers and market participants can do in what is a necessarily uncertain world is to work to ensure that the shock absorbers are strong relative to the range of potential economic and financial outcomes."

I'm not sure that we want to rely on shock absorbers. They're going to have to be pretty large unless you can lessen the load.

It's very true that, even here, he doesn't agree with me about implicit and explicit government guarantees increasing risk, or mention fraud very much, which to me is also important. I also think that my Bagehot's Principles would be a serious deterrent to future crises.