This is just an attempt to systematize my thinking about the causes of the financial crash and subsequent recession of 2007-8. As I understand it, the core of the problem was the way moral hazard interacted with the principal agent problem. To clarify those technical terms, moral hazard is the concept that if an individual or group is assured that if they collapse or otherwise fail they will be protected from the fallout this assurance will disincentive trying especially hard to prevent failure. The principal-agent problem is the issue arising from a situation in which an agent – say a CEO – has different incentives than their principal – say the owners of that company. In this case, the CEO wants to maximize his returns in the short run, while the owners want to maintain the long term value of the company. This may cause the agent to pursue goals different from or even contrary to the principal’s.
In the decades preceding the crisis, the federal government began sending strong signals to the market that the failure of huge institutions wouldn’t be tolerated – that is to say, those that had lent money to those institutions would be secure. Russ Roberts does a good job covering the history of bailouts, beginning with the protection of the creditors of Continental Illinois through the Savings and Loans Crisis and finally to the bailout in 1998 of Long Term Capital Management. The cumulative effect of these bailouts was to tell creditors that if they wanted to lend to large financial institutions, those bets would be safe because they were in essence going to be backed by the US government. This allowed banks to borrow large amounts of money for very low interest rates, which in turn allowed them to take highly leveraged bets (that is, making bets with borrowed money).
At the same time, the payment structure of these banks encouraged leverage in several ways. One was that compensation was geared towards huge rewards when leveraged bets paid off and small downsides when they failed. Bonuses and other incentivized pay skyrocketed when a bet paid off; if it failed limited liability shielded the trader from having to pay the downside. Likewise, because executives were often paid in stock, they were encouraged to run up the stock price in the short run rather than nurse the long term value of the company. Given the enormous access to capital that these banks had – because the capital was essentially subsidized through implicit government backing – the executives were able to make catastrophically leveraged investments which, when they went well, paid off hugely and when they went bust they were able to walk away from (this is chillingly similar to the American mortgage market, where regulation allowed borrowers to simply walk away from mortgages they were unable or unwilling to pay).
And there’s an even greater structural problem. As Tyler Cowen points out, the returns from finance are so vast on an individual level that the downside doesn’t compensate. As he puts it, “we don’t know how to punish people in a manner consistent with the rising size of absolute rewards.” Russ Roberts in the piece above says that capitalism is a system of profit and loss, and that the government protection of downside is removing the element of loss, thereby fundamentally destabilizing capitalism. As the above analysis indicates, that’s right, but it may not be the whole problem. When a manager at an investment bank can expect millions of dollars of compensation in an average year, if they play their cards right, then the downside of losing their job just isn’t enough to make them very cautious. This causes them to, in Cowen’s phrase, short volatility – that is, make giant bets that what we would normally expect to happen happens. In an average year they’ll win the bet and reap huge returns. In a bad year, the bet explodes and possibly the company goes bankrupt, but the investor made so much money in the good years that the downside just doesn’t hurt that much.
Of course, all of these factors work together. The creditor bailout provides easy money to people who for all the factors mentioned above have strong incentives to place absurdly leveraged bets on the status quo continuing – the stock market will go up, the Pirates won’t win the World Series, etc. In our case, in the 2000s the market bet hugely and dangerously that real estate prices would go up. One year, they fell, and the financial system fell with them.
This leaves the question – where do these root causes stem from, and what could we possibly think about doing about them? I’ll try to write about those in a future post.
Picture: My photo, available under CC licence.
(Thanks to Sam and Gabe for useful conversations on the topic. Also for conversations with my dad, but I don’t want to make this sound like a book dedication or anything.)