Saturday, December 8, 2012

The Leverage Cycle

My latest Bloomberg column will appear sometime Sunday night, I expect. I wanted to give readers a few links here to various key papers of John Geanakoplos on the leverage cycle, as well as a little further discussion of a few points.

First, this is the most detailed paper Geanakoplos has published (as far as I know) describing the leverage cycle -- the natural feedback process that repeatedly drives economies through cycles in which leverage rises, driving increasing asset prices, and then falls as investors become uncertain, more cautious, and demand more collateral, Prices then crash down accordingly. His argument is that leverage (determined by collateral rates) is a key macroeconomic variable completely independent from interest rates, and often just as important to the economy at large. In particular, increasing (decreasing) leverage is one key direct cause of increasing (or decreasing) prices. As evidence, look at the figure below for housing prices from 2000 through 2009. It shows how the average down payment required for sub-prime mortgages went up and then down, in both cases just in advance of housing prices. (Okay, this isn't proof of a causal link, I suppose, but it's enough to convince me.)

But before reading the "serious" paper I recommend first reading the text of this talk that Geanakoplos gave two years ago in Italy. It's much less formal and makes all the main points in a clear way.

From a practical point of view, I think two things stand out to me in his arguments:

First, given the clear importance of leverage in driving economic outcomes, it is quite remarkable that the Federal Reserve Bank has not in the past made any systematic effort to collect the kind of data it would need to monitor average collateral rates in the economy. Between 2000 and 2005, for example, no one at the Fed was going to banks and collecting information on how much collateral they were demanding when lending. This was just not considered a crucial macroeconomic variable. Judging from the tone in his talk, Geanakoplos finds this pretty amazing too. He mentions that the Fed contacted him in 2008 or so to get hold of data of this kind that he had collected. It seems that the Fed has now accepted the systemic importance of leverage and is seriously considering including leverage as a key variable in future macroeconomic monitoring. Whether banks and hedge funds will be required to report leverage levels, I don't know, but this idea is at least on the table. Good thing, I think.

A second interesting point is one that seems kind of obvious in retrospect. The leverage that drives the rise of prices in Geanakoplos's picture is leverage in long positions, which enables optimistic investors to buy more than they would otherwise be able to buy. Leverage in the opposing sense, short leverage allowing pessimists to speculate on a collapse in the market, would act to depress prices. Hence, it is probably more than a little significant that credit default swaps (CDS) for mortgage backed securities were created in 2005. These most likely acted as a key trigger of the beginning of the crash. As Geanakoplos writes,
In my view, an important trigger for the collapse of 2007–9 was the introduction of CDS contracts into the mortgage market in late 2005, at the height of the market. Credit default swaps on corporate bonds had been traded for years, but until 2005 there had been no standardized mortgage CDS contract. I do not know the impetus for this standardization; perhaps more people wanted to short the market once it got so high. But the implication was that afterward the pessimists, as well as the optimists, had an opportunity to leverage. This was bound to depress mortgage security prices. ... this, in turn, forced underwriters of mortgage securities to require mortgage loans with higher collateral so they would be more attractive, which, in turn, made it impossible for homeowners to refinance their mortgages, forcing many to default, which then began to depress home prices, which then made it even harder to sell new mortgages, and so on. I believe the introduction of CDS trading on a grand scale in mortgages is a critical, overlooked factor in the crisis. Until now people have assumed it all began when home prices started to fall in 2006. But why home prices should begin to fall then has remained a mystery.

...Of course, if CDS were introduced from the beginning, prices would never have gotten so high. But they were only introduced after the market was at its peak.
Not that the crisis wouldn't have happened in the absence of CDS contracts, but they probably hastened the collapse.

Finally, on a related matter, I think many people may find good value in this article by Ray Dalio, head of Bridgewater Investments. This is Dalio's attempt to give a simple explanation of "how the economy works" and he puts the expansion and contraction of credit at the very center. He is essentially making much the same argument as Geanakoplos, but in a less formal way. Fun to read and very instructive, in my opinion.