June 2, 2009
I am writing to comment on the SEC proposal to reinstate price tests to limit abusive short selling.
My background is both in industry and academic. In industry, I worked as an equity derivatives strategist at hedge fund Long-Term Capital Management from 1995 through 2000. From 2000 to 2003, I was a proprietary trader and researcher at Morgan Stanley's Equity Trading Lab. That group was one of the industry pioneers to demonstrate the power of algorithmic trading and the utility of market microstructure research. The infrastructure we developed is now used by the entire firm and, my proprietary work was very profitable for the firm.
I left industry to get my PhD at the University of Chicago and am now an assistant professor of finance at the University of Illinois at Chicago. At UIC I teach courses on my research areas: investment, market microstructure (i.e. the details of how markets work), and commodities.
Because of my background, I am well aware of the effects of price tests for short sales. Price tests, such as uptick, modified uptick, and bid tests, introduce inefficiency to the market. I believe the cost of this inefficiency will be borne mostly by small investors. I also believe we are ignoring what drives abusive short selling. The looming tragedy is that abusive short selling may be curbed by other regulation -- regulation we should have anyway.
Short sellers profit when a stock loses value. Accounting fraud at Enron was first publicized by a watchdog investor. This person watched for apparent misdeeds and publicized them. His work led to uncovering the fraud and saving future investors from losses. Some watchdogs get paid for their hard work by selling short the stock of companies like Enron. They get paid for their investigative work if they are correct and the stock declines in value.
Other short sellers constantly guide prices back towards sensible values and free investors from worrying about whether prices at any time are grossly incorrect. This small benefit accrues to millions of investors every day. Many firms that do this are based in Chicago and evolved from erstwhile futures pit traders. That background makes these firms acutely aware of risk and liquidity. The result: In 2008 when many banks and hedge funds incurred massive losses, Chicago market makers and "prop shops" had their most profitable years ever. Caring about liquidity and risk management paid handsomely -- and would seem to be a model for the finance industry at large. However, stock and options market makers require exemptions from price tests to be able to continually provide liquidity. Without such exemptions, we will be punishing the one area of our finance industry which performed admirably in the recent crisis.
Requiring short sales to occur subject to some price test, by design, slows price declines. This means prices will take longer to adjust to new, important negative information. However, positive information will be incorporated quickly into prices. This asymmetry means few investors will buy undervalued stocks -- while some will buy overvalued stocks. I also suspect this asymmetry will make the information contained in order flows more valuable. That also disadvantages small investors since only broker-dealers are privy to such order flows. If trustworthy prices attract investors, price tests will result in less trustworthy prices and discourage liquidity.
I also know from personal experience that price tests will not stop the effects of short selling. One simple way around price tests exists for many stocks: synthetic shorting. To synthetically short a stock, I find future on stock indices containing that stock. If I buy the stocks in the index proportionally to their index weight, my holdings are equivalent to holding the index. I can then sell an index futures contract, eliminating my exposure to the stock index. If I then sell one of the stocks, I will profit when that stock declines. I am effectively short the stock. This sale, since I hold the stock, is a "long" sale and would not be subject to price tests.
Synthetically shorting stocks is less efficient than normal shorting: It ties up more capital than would otherwise be necessary to short a stock. However, many broker-dealers have businesses which already require such inventories of stocks. For these market participants, the costs of synthetic shorting are trivial. Thus price tests will reduce the ability of small investors to sell stocks short however, price tests will have, at most, a small effect on institutional traders.
Most SEC communication about short selling seems to indicate is a desire to rein in abusive short sellers. However, price tests do nothing to address non-price-related problems. One such problem is flawed capital adequacy calculations. Currently, if a short sale is opened and closed within a day, many brokerages require no commitment of money from the short seller. This stands in contrast to buyers who are committing capital and long sellers who have previously committed capital.
Conversations with brokers indicate that many abusive short sellers could not operate if they had to commit capital for some period of time. Further, brokers expressed frustration that many abusive short sellers violate contractually-agreed risk limits during the day. While these short sellers are within their limits by the end of the day, they exceed their limits during the day, perhaps as they attempt to manipulate prices.
These clients hold overly large positions relative to their capital. That disparity puts their brokers at greater risk. Requiring capital commitment for trading will curb abusive short selling and reduce the risk to brokers. While the details will require some care, capital commitment is sensible regulation we should already have: Nobody should be able to trade in a stock without committing money for some amount of time.
Subjecting short sales to price tests will slow the spread of information and create less trustworthy prices. Perhaps these tests are needed however, we should first insist that all traders commit capital to their trades. We should also see if that minimal and sensible regulation -- regulation we should already have -- fixes the problem of abusive short selling. The alternative is to curb short sales by disadvantaging small investors in favor of less honest prices, lower liquidity, and preserving flawed capital requirements for traders.