Arbitrage, Margins, and Efficient Markets

Markets clearly overreacted in the months following the Lehman bankruptcy. The same thing holds true for other financial crisis, such as the 1987 crash and Asian/Russian/LTCM crisis. Market overreaction is not limited to dramatic macroevents. If one looks at implied option volatility, it is clear that market participants predictably overreact/underreact to different types of news. This overreaction is one of the cornerstones of behavioral finance.

Behavioral finance is often seen as the antithesis of the efficient market hypothesis (EMH). One of the premises of the field is that markets cannot be efficient considering markets' tendency to overreact. As behavioral finance as a field of study has boomed in the last two decades, so has rejection of the EMH (a poll showed a strong majority of CFAs reject the theory). Much of the hysteria about the EMH comes from the fact that much financial theory is premised on an efficient market. But behavioral finance should not necessarily threaten this theory outright. The value of the behavioral finance perspective is it helps us figure out which markets are inefficient and why that might be.

It is generally accepted that efficient markets are good for investors. In an efficient market, a buyer can be sure his purchase is at fair value. Markets love certainty. Investors cannot trust markets if they are not efficient. (Check out this recent article in the Financial Times.) For that reason, there should be efforts to create structural changes in finance to make markets more efficient. In the last year, derivative regulation is a step in the right direction. The change in fair value accounting was a move in the wrong direction.

But one structural aspect that has not been addressed is the role of arbitrage. Arbitrage is a driver of efficiency in markets. It is an example of the beautiful way in which the invisible hand of mutual self-interest creates a surprisingly fair market. But in times of stress, arbitrage is no longer a source of efficiency, but a source of inefficiency. This inefficiency comes from the wrong premise that arbitrage requires no capital.

Consider the following example. A hedge fund decides to take advantage of a price differential between S&P500 futures on two different exchanges, buying the cheaper futures and selling the overvalued ones. From an economic perspective, he has locked in the spread between the two exchanges since they are fundamentally the same thing. However, if after the trade the differential increase, technically the transaction has a mark-to-market loss. Because the original price differential was miniscule in the first place, the trader probably levered up to get a better return. As markets are behaving irrationally, the differential could grow further, triggering more accounting losses. These losses require posting collateral, which is why arbitrage is not capital-free. As Keynes said, "the market can stay irrational longer than you can stay solvent," meaning eventually collateral calls can lead to default. (Schleifer and Vishny 1997)

LTCM manager John Meriweather described his arbitrage fund's strategy as "picking up nickels all over the world." Many of these nickels need to be picked up. They make the streets of finance more transparent. When the price differential between S&P500 futures on different exchanges increases, traders should be able to increase their positions and lock in more guaranteed profit rather than having to liquidate trades to post margin. When interest rates on treasuries go negative, as they were briefly after Lehman and during the Great Depression, market participants should be able to clear the nickels and bring yields to reality.

Considering most financial crises are exacerbated by an irrational market causing caustic margin calls, it is interesting to look at how policymakers have fought this effect in the past. The best example comes from the Panic of 1907. The Panic of 1907 was very similar in nature to the 2008 Financial Crisis--caused by a lack of trust in financial institutions, bringing stock prices down 50%. The issue in 1907 was liquidity--bank runs were draining money out of the system. J.P. Morgan famously locked the most important bankers in a room to come up with a solution. The solution was to replace clearinghouse deposits (basically mark-to-market margin) with bonds so banks would have enough cash.

What if this model could be adapted to modern day liquidity crises? (Leveraged) arbitrage is similar in nature to fractional reserve banking. In commercial banking, if every consumer demands their deposits at the same time, the bank will be insolvent. Similarly, if all arbitrage trades go wrong at the same time, a firm will not be able to post margin on all of its trades even though they are economically viable. This example illustrates that for arbitrage to be an effective driver of efficient markets, there must be a mechanism in place that guards arbitrageurs against bank runs. One good step would be counter-cyclical margin requirements like those seen in 1907.

Impact of Margin Requirements on OTC Derivatives

One of Geithner's six measures to contain systemic risk and prevent financial crisis is to process OTC derivatives through clearinghouses and to move a large portion of OTC derivatives onto exchanges. This change could have the following effects:

1. Exchanges' volume (and profit) will increase. (Considering the recent gains of the CME group and Deutsche Boerse, it seems the market agrees.) More importantly, their operations will evolve. Derivatives are an emerging technology with powerful risk managment applications. Like other new technologies, they still need time before users figure out how to use them safely. The dangerous application of derivatives for risk management was most evident in the role of AIG during the credit crisis. Many financial institutions hedged their exposure to bad credit with credit default swaps from AIG. Hence, it would appear that derivatives adequately distributed risk, when really they just concentrated all the risk in one counterparty. These CDS were also used to bypass capital requirements. Geithner's plan takes strong steps to prevent this misuse in the future by limiting counterparty risk.

2. But limiting counterparty risk also has negative side effects. Geithner's plan includes "robust margin requirements," to prevent firms like AIG from making hundreds of bets without any money down. While margin requirements discourage excessive speculation and go-for-broke strategies, it is also a drain of liquidity. Assets held on margin cannot be productive assets necessary for business operations. They must be passive assets that are not used by the company.

On an aggregate level, this represents a significant loss of liquidity in the financial system as it slows the velocity of money. Instead of sitting in idly, these funds could be used for capital expenditures that grow the economy. The margin requirements will have an especially strong effect on distressed or highly levered companies who do not have easy access to capital. Having to pay a margin for a simple, unrisky derivative like an interest rate swap makes these instruments significantly more expensive since liquidity holds an especially large premium for these companies. If it becomes impossible to make interest rate swaps without posting margin, many of these already vulnerable companies will simply take the risk. One company I was recently looking at has 8bn of floating debt and would have needed 400m margin in order to hedge the interest rate risk associated with this debt.

3. I predict interst rate swaps will still be available OTC, though I expect the fees will be higher to compensate for a lack of margin requirements. This will essentially be a liquidity premium. If the market develops in this way, it would actually benefit the banks by keeping OTC volume. Banks make very high profits off OTC derivatives from their ability to shave off a few basis points here and there. (JPM had 5bn in profits from OTC derivatives in 2008.) The profits from OTC derivatives could also be more evenly spread out among banks. Geithner's plan includes margin requirements for OTC dealers as they have greater systemic risk. This means the large dealers like JPMorgan might have to decrease their volume. Naturally, the extent to which this occurs depends on the exact margin rules and the degree to which OTC derivatives are moved to exchanges.