Recent Items

Thinking the Unthinkable: Regulating the Brave New World of Finance

Posted on by

Earlier this week, I sought to frame the prevailing views of what the supervising adults, namely central bankers, should do about the turmoil in the financial markets. They break down into four groups (names of representative spokesmen included):

The keep the party going types (Jim Cramer and his less histrionic brethren) who argue that markets should be stabilized at any cost.

The cold water Yankees (Nouriel Roubini, Marc Faber Andy Xie, Michael Panzner) who think the problems have only started and providing any more cash to the miscreants (or anyone within hailing distance of them, which is basically the entire financial system) is a Bad Idea and Will Only Make Things Worse.

The realists (heavily representation at the Financial Times, including the estimable Martin Wolf and Paul de Grauwe) who agree things are a mess, but also argue that central bankers can’t sit on their hands in a crisis. They hew to the idea presented by Walter Bagehot in the 19th century, that central bankers should lend at penalty rates against good collateral (meaning in normal markets). This gets a bit tricky if you are talking about collateral with a lot of embedded leverage. A modern variant comes from Willem Buiter and Anne Sibert, who want central bankers to act as market makers of the last resort.

The reformers (former central banker Ian MacFarlane, Henry Kaufman, Steven Roach) who see the turmoil as evidence that our current regulatory regime is outdated. However, regulation is such a dirty word that they believe the current problems have to become much worse for there to be sufficient political will to take on such a challenge. Kaufman, as befits his 1980s sobriquet, “Dr. Doom,” is particularly pessimistic, since he feels that the economic discipline has become too specialized to have an adequate grasp of the issues.

Even though I am largely on the same page as the cold water Yankees in terms of how bad the downside could be, I believe that the reformist camp is correct. While booms and busts are an inevitable part of capitalism, the very reason we have financial regulation is to mitigate those swings.

It’s an open question as to whether the history of the 1992-2007 will be seen as one of stability, which is the current view, or rewritten to emphasize serial asset bubbles that eventually became so large as to exact a large cost on the real economy.

In recent weeks, as the carnage in the markets has worsened, the use of the dreaded “R” word by the media has increased. However, the worry is that Congress will return this fall, eager to have some public hangings to satisfy the blood lust of newly impoverished homeowners, schedule some hearings, summon some famous people and call them bad names for ten minutes, enact some hasty legislation that will at best address surface problems, and declare victory.

The only axis on which that approach might work is consumer protection. There has already been discussion of possible remedies and some states have enacted legislation that might serve as models. Any reform of the credit markets, in which a tremendous amount of activity has moved well beyond the Fed’s reach and therefore, to some extent, even its understanding, is a very tricky affair. It’s important to get it right. This process is not only politically daunting, (expect piercing cries and doom saying from the financial community), but intellectually demanding as well.

Now many may argue that that task is impossible, but recall that the Securities Act of 1933 and the Securities Exchange Act of 1934 were formed almost out of whole cloth, yet have proven remarkably robust and enduring. It is possible, though rare, to take great leaps forward with regulatory frameworks.

To come up with appropriate new rules, it is vital to define the problem or problems need to be addressed. Per Kaufman’s concerns, that isn’t as easy as it might seem. Framing the problem well, conducting analysis and investigation, and seeing if they point to other hypotheses and avenues for investigation is vital. And for issues of this importance and complexity, they will need to be attacked from various angles. In addition, the problem statement is likely to be reformulated as understanding advances.

In other words, expecting a one-shot investigation, no matter how well staffed, to do the job is unrealistic. Coming to grips with such a hairy, multifaceted problem is likely to be an iterative process. (And note we haven’t considered the possible dead body in the room, namely, that any approach will likely require either much greater international cooperation or an international body, raising issues of national sovereignty. We may need a real train wreck to get past that hurdle).

It’s encouraging that some good minds are already starting to grapple with problem definition. Unfortunately, the discussion is taking place on the other side of the Atlantic, where regulation is held in higher esteem than here. But ideas are fungible and will hopefully capture the attention of thoughtful people here.

A couple of examples, one from VoxEu, “Subprime crisis: Greenspan’s Legacy,” by Tito Boeri and Luigi Guiso .

The article makes an elegant and compelling argument. The subprime crisis had three causes, namely:

The low financial literacy of US households;
The financial innovation that has resulted in the massive securitisation of illiquid assets, and;
The low interest rate policy followed by Alan Greenspan’s Fed from 2001 to 2004.

The authors clearly state that the first two issues were necessary but not sufficient to cause the housing mess, and they make a cogent and critical review of Greenspan’s actions.

The comments on Greenspan, while well done, are in line with what is rapidly becoming conventional wisdom. But this section is useful to keep in mind when thinking about reform:

The first ingredient of the crisis is a blend of bad information, financial inexperience and myopia of consumers/investors. They fell for the prospect of getting a mortgage at rates never seen before and then extrapolating these rates out for thirty years. This myopia was encouraged and indeed exploited by banks and other lenders eager to attract and retain clients. This is surprisingly similar to what has been seen in the past when banks and intermediaries have advised their clients to invest in financial assets ill-suited to their ability to bear risk. In both cases, a biased advisor is the reflection of a clear conflict of interest in the financial industry. Financial literacy is low not only in financially backward countries (as one would expect), but also in the US. Only two out of three Americans are familiar with the law of compound interest; less than half know how to measure the effects of inflation on the costs of indebtedness. Financial literacy is particularly low among those who have taken out subprime mortgages. The intermediaries exploited this financial illiteracy.

Another good article, which explicitly looked at the question of regulation, came from Clive Crook of the Financial Times. He goes where US financial writers fear to tread, namely, questioning the benefits of financial innovation:

The harder question is whether new rules are needed for the wider financial system. On the face of it, the answer is Yes. One rationale for excluding non-bank lenders from Fed scrutiny is that they pose no systemic risk. So much for that. Wall Street financed the subprime boom by buying the loans – repackaged as securities, stamped AAA by the credit-rating agencies – and selling them on. This model, of course, made the original lenders even less attentive to loan quality. On the other hand, it spread the risk throughout the system, which was also thought to be a good thing – until the loans started to go bad. Then, it turned out, investors wanted to know where the risk was and nobody could say. Arriving as if from nowhere, that fear led to the freezing up of the credit system.

How are regulators to grapple with this? If the opacity of the system is the problem, then new scrutiny and disclosure requirements for secretive investors such as hedge funds and private-equity firms must be part of the remedy. But it could be that complexity, more than lack of transparency in its own right, is the issue. The accelerated pace of financial innovation and the ever-proliferating complication of modern financial instruments seem to defy the ability even of the products’ designers to fathom what is going on. And the new instruments are often thinly traded, if at all, so values are guessed by simulation or calculation, not in the market. Sophisticated investors are left poorly informed about the risks they are bearing; unsophisticated investors have not got a clue. Desirable as fuller disclosure by hedge funds and private equity firms may be, it is hard to believe that it will be enough.

In other words, financial innovation itself is the problem. This poses a dilemma. The benefits of modern finance are real: as its champions rightly say, it deserves much of the credit for the relative stability of the world economy in the past two decades. Stifling this innovation, or attempting to manage it, looks unpromising.

Part of the answer – and, along with fuller scrutiny, perhaps the best that can be done – is to create a climate where excessive risk-taking is more effectively discouraged, and punished when things go wrong. Here the role of the Fed is crucial, both in the boom phase of speculative cycles and in the bust. Fast-rising house and other asset prices have been buoyed by very low interest rates. It was enough for the Fed that consumer-price inflation was low; asset prices, in their own right, were not its concern. This set the scene for America’s remarkable debt-fuelled house-price surge – whose inevitable end was the proximate cause of the subprime collapse. The Fed’s long-maintained reluctance to weigh asset prices in its monetary policy calculations needs to go.

Then, when financial markets seize up, the Fed must take care, as far as possible, to avoid bailing out the culprits. As the economists, Willem Buiter and Anne Sibert, have argued, the Fed was wrong to cut the discount rate last week, and will compound the error if it soon cuts the more important Fed funds rate as many now expect – unless there is evidence of harm spreading to the real economy. Instead, honouring Walter Bagehot’s maxim, it should provide liquidity at a penalty rate (against conservatively valued collateral) to those so lacking in liquidity that they are willing to pay it. That memorably costly help should be available not just to banks, as now, but to hedge funds and other financial firms willing to accept the Fed’s terms.

It is a cliché, but nonetheless true, that the end of each financial crisis sows the seeds of the next. Better regulation has a place, but the Fed is the key to attacking that cycle.

While this is a good start, Crook’s ideas are an extension of well established principles, namely, disclosure and central banks as stewards of stability and lenders of the last resort. We may need thinking along new lines to devise enduring remedies.

Print Friendly
Twitter0DiggReddit0StumbleUpon0Facebook0LinkedIn0Google+0bufferEmail

6 comments

  1. Anonymous

    What happened to the names in the “Realist” paragraph? All wrong apart from Walter Bagehot, against all the odds!

    Carping aside, great post and great blog generally. Thanks for the thought you put into your posts. Charlotte, London

  2. Yves Smith

    Charlotte,

    Thanks for your comment. Af first I wrote a long explanation, but now I see you mean I put in the wrong first names for Wolf and de Grauwe. How embarrassing.

    Regulars readers have probably noticed that proofreading is not one of my strengths, and that skill (to the extent I have it) deteriorates even more when I post after midnight, which I often do.

    Thanks for the catch. I’ve corrected the post.

  3. a

    The key is to have financial crises more frequently, so that errors can be detected more quickly and eliminated sooner. The U.S. Forest Service used to think that wildfires were bad and tried to stamp them out, but then it realized that those wildfires consumed the tinder which, if left standing, would allow less frequent but Really Big wildfires.

  4. Anonymous

    “The rich are not like you or me. They need the government’s assistance to get by.”

    I would argue that there has been a continuous recession since Bush’s first enormous $2.xBillion tax cut. And since then, consumers have been propping the economy up with buying things on credit.

    Yesterday, I was reading The Big Picture, and he was discussing unemployment rates. Officially the govt says it’s 4.5% (and he dismisses that saying otherwise there would a labor shortage), but unofficially he estimates it to be 8% if figured the same way as Europe. IMO, I would hazard a guess that it might be closer to 14% since under-employment (less than 40 hours / week) isn’t counted (even the BLS says the average workweek is 33-34 hrs /wk.

    Which brings me to my point. I agree with the Cold Water Yankees (Roubini) and the Reformers (Kaufman). I like what Schiff has to say here, US a paper tiger.

    I also think we’re in unchartered territory and the US doesn’t have the smarts, the wherewithall, or the will to do anything about it, as FDR did back in the 1930s. And with that, I think the US, as we know it, is over. Beyond that, I can’t imagine what it could do to save itself, especially in this day of corporate media and poll-driven pols. Needless to say, I think there will be a lot of people who die broke out of this mess.

  5. Anonymous

    An idea for regulation:

    The problem here seems to be not that ratings were wrong — which would require a simple repricing of risk, but that the ratings were the only thing investors were relying on to determine the value of the assets.

    Ratings agencies developed to assess the default risk of assets that were already trading on markets. It seems as if in current (or should that be former) financial markets, they are more or less setting the parameters of what can trade on the market.

    Maybe rating agencies should not be allowed to rate new types of assets (yes, this needs definition) until they have been trading on markets for 5 or 10 years.

Comments are closed.