Hedge Fund Strategies

It is often asserted that large size limits returns for various hedge fund strategies. While long/short managers sometimes claim that this may be the case for other strategies, they often argue that it is not for themselves because of the huge size and liquidity of the equity markets. The available evidence suggests otherwise. Clearly the returns of smaller funds are more dispersed and somewhat higher than those of larger funds, but the dispersion may come as a consequence of greater risk. Nonetheless, on a pure return basis, smaller long/short managers appear to offer somewhat better potential for outperformance.

In addition, the long/short strategy of numeric uses 90% of the invested capital to buy stocks long but since it sells simultaneously bad stocks short with a similar value, numeric's cash account is debited and credited by approximately the same amount. This leaves close to 100% of the invested capital as cash in the account of the portfolio. Therefore the portfolio can generate additional returns from proper cash management. On a first glance this might look as free return in the sense of return that comes without any risk because numeric could just invest the cash in treasury bonds and provide risk-free return. Looking deeper at this cash however reveals yet a different source of risk: The portfolio has inherent bankruptcy risk, if there is a shortage of liquidity if all the owners of the securities from whom numeric has borrowed the shares to short sell them want to sell their shares on the market which means that numeric had to give those stocks back. In order to do so, numeric had to buy the shares on the market. Especially if there have been losses on the short part of the portfolio, i.e. if the share price of the stocks that were short sold by numeric increased, numeric will not have enough money to cover those share purchases. While the risk on long positions is limited since in the worst case scenario the stock has just lost all its value, losses on the short positions are not capped at all, because the losses grow as the stock price increases and there is no cap on the stock price. For those reasons it is very important, that numeric has a good exit strategy if news on a stock change. At any given point in time numeric has to be able to cover a significant amount of all its positions. This means, that if the predicted development of a stock from the model change the classification of the stock over night from a bad stock into a good stock it is likely that the stock price is going to increase in the next few days. In this situation numeric has to get rid of the exposure to the short position in this stock. One way to do this is to buy the same amount of stocks on the market in order to fully cover the short position. Unfortunately, this works only with limited volume. As soon as the volume of stocks numeric wants to buy in order to cover a short position, exceeds the threshold from being immaterial to being material

...