How to bet against a bubble: the Ultimate Bottleneck strategy
May 25, 2010
The other night I had a conversation with a financier that offers some pointers for spotting trouble in the sovereign crisis. How, we wondered, does one bet against a bubble?
It is not at all obvious. There is the problem of betting too early. We recalled the case of an ex-colleague of mine, economics professor and expert on real estate. He sold his flat in Barcelona in the year 2002, on the conviction that real estate was overheated. He lost five years of upside, and would still be up today if he had not sold. Now, this is a real problem. You not only have to be right, but also timely. How do you pull that off?
After a bit of joking (and the first pint) we came up with what that could be called the “Ultimate Bottleneck” strategy. Say you think people are irrationally buying houses in the hope of reselling them for a profit… even as prices are far too high. When will that stop? Well, it’s difficult to know, because there is a positive feedback loop. Banks lend home buyers money to do the gamble, which drives prices up. As they go up, less credit worthy borrowers want houses. In 30 years, when the houses are sold, their investment wont be recouped. But for now, betting that the trend will reverse is very dangerous… because you’re betting against feedback loop. Feedback removes the “rubber band” that would otherwise pull prices back to normal.
So the right question should not be when will people regain sanity but rather, “when is this loop sure to stop”? When is it that a society of deranged individuals (even if such a thing existed) would no longer be able to sustain price growth? And the answer we came up with is — when people can no longer pay the monthly payments. The strategy is then clear: if you are in a bubble, assume it will continue for as long as the feedback loop holds. Locate the crucial juncture points in the loop. Determine which the weakest linkage. And use it to time the bet.
Consider the beauty of this reasoning. It is clear, measurable, tangible. And it connects beliefs about the long term with the reality of short term. I saw a similar dynamic in my study of securities analysts during the dot-com bubble, done with Raghu Garud. The debate over the long term future of Amazon.com (say, long-term revenue, or margin) was shaped by the analysts interpretation of short term events such as third-quarter sales. This temporal dynamic in the controversy over frames has not been explored elsewhere in the literature, and I think it is crucial.
The application to Greece and other sovereign debt countries is clear: watch out for the auctions of bonds. The application to Spain and house prices is equally interesting: prices have not come down as much as they should. With this in mind, it is interesting that the auction of treasury bonds in Spain did not go well at all last week. According to the Financial Times,
Spain came close to its first debt auction failure on Tuesday, highlighting the funding problems for weaker eurozone economies (…) The government’s difficulties in selling €6.44bn ($7.96bn) in one-year and 18-month bills sparked worries over its 10-year debt auction on Thursday (…) Steven Major, head of fixed income at HSBC, said: “The Spanish auction was very disappointing and does not bode well for further issuance. It’s becoming more apparent just how difficult it is for Spain, which is a big worry so soon after the launch of the international rescue package.
socializing finance 
May 25, 2010 at 2:53 pm
Daniel, “finding the weakest link” might be a good strategy for making a bet, but it doesn’t go very far in uncovering *why* a particular bubble pops when it does.
More important than people failing to pay back their loans is the fact that this event becomes *publicly visible* to all (usually through the news). Unless that event itself becomes rendered visible, the bubble can keep on growing. For the purposes of making bets, it’s not that the event happens — it’s that the event happens and is rendered visible. However, precisely when that event will become rendered visible is quite a bit more uncertain than when the event itself will happen.
Echoing my comment from several weeks back, I think this is another case where it’s very important to pay attention to how the material infrastructure of the market shapes actors’ higher order beliefs. If you look at the economic theory of “rational” bubbles (where people try to time the market as you say), those bubbles persist because there is a lack of common knowledge about the bubble’s existence (Abreu and Brunnermeier 2002; 2003, Allen et. al. 1993).
So while each individual may believe that the bubble exists, each person might also believe that somebody else doesn’t realize there’s a mispricing. According to this theory, the bubble “pops” when actors’ beliefs are shocked back into alignment, usually by some kind of publicly observable event — say, news coverage of people defaulting on their loans.
The problem with economic theory is that its treatment of bubbles lacks methodological symmetry. Bubbles are aberrations where actors’ higher order beliefs become misaligned, a deviation from the “natural” behavior of markets. But why is it that accurate pricing can happen at all in the first place? I think we have to look very carefully at how market infrastructure keeps these beliefs aligned sufficiently well for prices to have facticity.
May 25, 2010 at 6:07 pm
Taylor — You have a very good grip on the rational bubbles literature, but I disagree. I agree that a galvanizing event helps align frames, but I would add that such event is not key.
Here’s why. Once the loop is broken, prices deflate by themselves. E.g., the US real estate market. Prices began leveling off when houses were being sold to unemployed buyers who could not even meet *their first monthly payment.* Once house prices stop rising, houses that looked cheap under the expectation of a 10 percent annual price rise suddenly look overvalued.
Second, why blog about hedge fund strategies? I began thinking of shorting strategies back in 2003, when I moved from New York to Spain and found that as a professor I could barely afford renting an apartment in Barcelona. Thinking of shorting gave me mental relief. More recently, it’s become a way of articulating an understanding of bubbles. And I share my thoughts because among other reasons, it is the difficulties of shorting bubbles that contribute to their persistence.
May 25, 2010 at 6:40 pm
Hmm, that’s a good point. Frames can be re-aligned in a variety of ways, one of which is (dramatically) through some event. I think you’re right that the ways in which this can happen are more diverse than the rational bubbles literature suggests.
But I think you’re switching the effect with the cause: you’re saying the price began leveling off, which made the houses look over-valued. But then why did the price start dropping in the first place? Didn’t someone have to believe that the houses were over-valued for the prices to begin falling in the first place?
In other words, how do you connect the unemployed buyers not making their payments with the resulting price change? Doesn’t that require the event to be rendered visible to the real estate market? So while it might not be a dramatic “event,” that information has to cross from the banks/investors (or whoever receives the payments) over to the real-estate speculators, and a sufficient number of those speculators needed to believe that others will receive and act on the same information. So what was the information conduit? Could it have been the ABX index? My understanding (it’s limited) of the history is that the housing bubble popped only after the CDS credit indices were imported over to ABS CDOs.
Also, I think it’s interesting to blog about financial strategy — I’m just saying that to get bubble strategy right I think we need to think about these issues of visibility, frame-alignment, and how the material infrastructure of the market shapes those. Addressing that could allow SSF to locate “alpha” in a way that financial economics can’t.
May 27, 2010 at 10:35 pm
Taylor — our (excellent) exchange so far came back to me tonight at a dinner with three economists. A banker, a professor at Oxford and a regulator at a prominent US agency in New York. The chat went back to Bear, and how the bank failed in its last 72 hours.
One of the key issues that led to the downfall was the insistence within the top management team in evaluating the situation in terms of solvency. Bear was fine long term. What the bank missed was the short term problem of the counterparts and repo market. If counterparts stop lending, the bank would go down. Obviously, whether they would do that or not partly depends on the long term perspectives of the bank. But, crucially, not just on that.
August 7, 2010 at 4:59 am
[…] If there is one message coming out of the crisis, it is that we live in a bubble economy. Bubbles are, as I see it, the central intellectual challenge of our time. And instead of telling the government to go and fix it for us, I’d rather devise strategies that allow actors to deal with bubbles on an individual basis. I am offering some ideas here, based on my research with David Stark. (This, by the way, is the second of a series started here). […]