Potential Smartism #1: The champagne glass theory of economic bubbles

Alright here we go. This is a site essentially to house my working theory’s on economics, finance, sports, incentives, morality, Dr. Pepper, puppies, why brussel sprouts are gross (I know right?) etc. It will probably be a mile wide and a couple inches deep, but with time, OH YES WITH TIME, hopefully it can grow into a grand unified theory on how awesome everything could be.

The quest to universalize and diagnose bubbles has been a “holy grail” of policy making since the Dutch Tulip Mania. Ben Bernanke and Alan Greenspan thought they had it solved in the early 2000s with “the great moderation”, where volatility had dropped and recessions were much less severe for a roughly 20 year period. They were under the opinion that financial theory and regulation had gotten so effective and the markets so efficient that there could no longer be bubbles! Of course this was proven false.

Here is my take on bubbles. Something happens to distort normal market incentives and the risk / return framework (could be regulatory related see Potential Smartism #2: the Spinning Top Theory of Regulation). Smart people (rational actors) recognize and take advantage of the distortion. They are followed by sillier people (irrational actors) who see profits but not risks. Rational actors exit the situation when the distortion is corrected or systemic issues add enough risk to balance the equation. Irrational actors remain in and often lever up to recapture profits from boom time. Then there is a catalyst, the bubble bursts and irrational actor’s exuberance turns to panic to get out. Rinse repeat. Similar to how champagne glasses produce bubbles because of imperfections in their surface, it is when there is an imperfection in the symmetrical risk / return framework that bubbles occur in economies. If we try and minimize distortions and leverage then we should be able to minimize the impact of these bubbles.

Champagne Glass Model

There is a symmetrical risk / return framework that in an open market setting generally holds. If you are equally exposed to the upside and the downside of any investment then it stands to reason that the riskier the investment the more you will need to be compensated for it. The supply of investable capital should logically decrease as the risk increases and the capital able to tolerate that risk wanes. This, in theory, should keep a leash on the riskiest parts of the market. It’s when there is a distortion in this risk / return framework that things go crazaaayyyy.

Initially, whatever the distortion is, you are generally able to earn larger rewards than the risk should entitle you too (or the same rewards for lower risk). In the lead up to 2008 there were two distortions:

1. Banks could earn high interest from lending to risky subprime lenders, but the downside risk was trimmed by federal backing (both explicit and implicit). If you can get all of the reward, yet lay off the risk onto the government (read: taxpayer) then you have a distortion. It has long been a political aim to increase the home ownership rate in the US, so you can see the motivation for introducing this distortion into the market.

2. There were unnaturally high ratings on the Mortgage Backed Securities eventually created from these mortgages. These unnaturally high ratings allowed banks to borrow against them at a low rate, while selling them at a high rate. The high ratings arose from misaligned incentives stemming from the issuer pays model of rating. If the person paying you is asking for a AAA rating, the incentive system is there to not do a full independent analysis to ensure future business. Add to this the fact that the US government mandated that pension plans invest in only AAA assets, which effectively gave a seal of approval to the ratings agency ratings. After all if the government mandates that we follow these ratings, they must be right, right?

When the distortion is identified, smart players (fancily termed: rational actors) understood that there were excess profits to be made in this distortion and they endeavoured to do it over and over and over again in a linear programming, maximization type equation. Less smart players (irrational actors) see profits being made and follow suit, not necessarily understanding the distortion or the underlying incentives and motivations of the rational actors, but making money regardless.

A rational actor would cease this action once the excess profits are no longer there and the distortion is corrected. This could happen either from returns being reduced, or from risk being increased. One way that the distortion could have been resolved in 2008 was for the government to reduce its role in the housing market and return to a market focused approach, that didn’t happen. What balanced the equation was systemic risk that arose from the fact that the size of this market was overwhelming the governments ability to backstop it. If everyone was rational then when this systemic risk balanced the return, this would be where the bubble ends. Housing prices would remain elevated and it wouldn’t be great, but this would seem to be a “non-cataclysmic” type bubble.

In real life though, at this tipping point not everyone exits the market (see Michael Lewis: The Big Short for a cool treatment on really seeing the difference between rational and irrational in a case like this). The irrational actors see what used to be outsized profits receding and, lacking the ability to find their own distortions, proceed to use the only mechanism they know how to regain their profits: leverage.

This last stage is the most damaging portion of the bubble. The rational actors would acknowledge that adding leverage to an investment that already wasn’t earning excess profits anymore would earn them an economic loss, but the irrational actors keep fuelling. They can fuel for quite some time depending on the distribution between rational and irrational and the amount of leverage available. Finally, some catalyst punctures the bubble, leveraged losses start to pour in, and the irrational actors flood to the exits en masse creating panic and severe downturns.

Given that governments have a) the ability and b) the political desire to distort markets, the largest bubbles seem to form from government intervention. In 2008, the distortion was from the government combined with the rating agencies[1].  In the South Sea bubble of 1711, the British government debt exchange program provided a massive source of demand for the South Sea Company’s shares. This demand, done to try and reduce crown debt, caused enormous price appreciation independent of the underlying economics. Initially everyone made money as the returns were artificially high, however at a certain point it was unsustainable and the rational actors cashed out while the irrational actors sold their houses and took on debt to buy more shares. It all ended in tears.

It does not need to be government made though. In the tech bubble of 2000 the distortion was the completely incomprehensible impact of the internet, which distorted the perceived return potential for a given level of risk. Perceptions change quicker than government actions though, so when people realized that “eyeball share” did not equal cash, we saw a much quicker, and less severe boom/bust cycle.

In terms of policies that could limit the damage of bubbles, the largest one is to NOT enact policies that create these excess profit potentials and keep the upside and downside symmetrical. This is understandably difficult as regulation is very complicated and there will always be tradeoffs. We cannot legislate risk tolerance or prudence, but we can legislate leverage restrictions on the intermediaries that transmit these distortions around, namely financial institutions. This would have the effect of at least containing the ability of the largest irrational actors to continue to inflate after the systemic issues have eliminated excess profit.

It’s not perfect (still not sure what the hell people were doing in the tulip bulb mania in Holland), but its an interesting way of thinking about bubbles and what pushes them through the cycle. I’m interested in looking back at other bubbles and seeing how they fit into this model.


One thought on “Potential Smartism #1: The champagne glass theory of economic bubbles

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s