1 post tagged “stiglitz”

Securitization has become a panacea on financial markets.

Indeed, it is predicated on the simple idea that balance sheets are more often than not the best parking for risks. In other (shareholders) words, they are the most costly parking you can find. Hence, commercial banks with the structuring help of Wall Street firms have "shipped" assets and liabilities that were so far stored in their balance sheets to investors.

It sounds like a marvelous and lucrative never ending innovation spiral à la Merton where banks can do more business, investors have access to a wider spectrum of securities and risks are more efficiently shared.

Too good to be true: The subprime mess is a wake up call, one that calls for deep thinking.

There are some things that do not change. In October 1987, the equity market crashed and portfolio insurance was designated as the culprit. Portfolio insurance used to be considered as the solution to downside risk protection. Investors hate downside risk even more than they love upside. University of Berkeley mavericks, Hayne Leland and Mark Rubinstein (co-founders if LOR), came up with a technique that promised to replicate what a long put option does, namely pay off, when the market goes south. The "trick" was simply to use Black-Scholes-Merton option hedging argument to replicate the put option payoff. In other words, they were selling stock index futures (mechanically/dynamically) and going long T-bills (cash) when the equity market was tanking. When the market was rallying up, they did the opposite. That's why when the market crashed in October 87, LOR was accused of provoking/amplifying the downward trend. The argument is a bit odd: Mark Rubinstein rightly pointed out that nobody praised LOR when the market was going up for making the growth even stronger.

The truth is that while portfolio insurance is in effect more of car insurance than earthquake insurance type there is an important market item that it paradoxically "destroys": Information. Indeed, when lots of investors are willing to buy put options bidding the price up they signal their bearishness to the market. What portfolio insurance does is to break down a single (put) transaction into two separate transactions: stock index futures and T-bills. Information about market expectations may get distorted in the process as it is hard for the rest of the market to decipher the initial intent. Worse, as shown by Sanford Grossman and others when portfolio insurers sell mechanically to dynamically replicate the target put, people may mistakenly think that this signals bad news. Asymmetric information problems have increased and we know both from the seminal works of George Akerlof, Michael Spence and Joseph Stiglitz and casual business experience that this is a sure recipe for trouble.

Sadly enough, the same holds true with securitization as witnessed by the subprime mess. Bad loans have been securitized. As a result the shareholders of the originating banks do not bear the consequences, good or bad, of their loan activity anymore. They have no incentive anymore to monitor these loans that get "diluted" in securitization vehicles. Hence while risks seem to have been more efficiently cut into pieces and shared the overall situation has worsened drastically because asymmetric information problems are now more toxic (1). Asymmetric information is a market killer and we end up collectively harvesting what others have planted.

Not convinced. Well, think of what microfinance does. It does exactly the opposite: It reduces information asymmetries by putting monitoring and safety devices in the micro-lending process: Lend only to women, in a village where everybody knows each other, make the borrowers collectively liable when one of them defaults etc... This is why it is successful.

So, next time innovation in the form of sophisticated securitization knocks at your door, ask yourself whether information and incentives have been reduced or not before joining the bandwagon. If portfolio insurance and the subprime can teach us one thing or two, it is precisely this.

(1) Not to mention the fact that it becomes very hard, almost impossible, to restructure the (securitized) loans when disaster strikes.