RISK TO TRADING STRATEGIES

In investment management, risk is a primary concern. The majority of trading strategies are not risk free but rather subject to various risks. It is important to be familiar with the most common risks in trading strategies. By understanding the risks in advance, we can structure our empirical research to identify how risks will affect our strategies. Also, we can develop techniques to avoid these risks in the model construction stage when building the strategy.

We describe the various risks that are common to factor trading strategies as well as other trading strategies such as risk arbitrage. Many of these risks have been categorized in the behavioral finance literature.⁹ The risks discussed include fundamental risk, noise trader risk, horizon risk, model risk, implementation risk, and liquidity risk.

Fundamental risk is the risk of suffering adverse fundamental news. For example, say our trading strategy focuses on purchasing stocks with high earnings-to-price ratios. Suppose that the model shows a pharmaceutical stock maintains a high score. After purchasing the stock, the company releases a news report that states it faces class-action litigation because one of its drugs has undocumented adverse side effects. While during this period other stocks with high earnings-to-price ratio may perform well, this particular pharmaceutical stock will perform poorly despite its attractive characteristic. We can minimize the exposure to fundamental risk within a trading strategy by diversifying across many companies. Fundamental risk may not always be company specific, sometimes this risk can be systemic. Some examples include the exogenous market shocks of the stock market crash in 1987, the Asian financial crisis in 1997, and the tech bubble in 2000. In these cases, diversification was not that helpful. Instead, portfolio managers that were sector or market neutral in general fared better.

Noise trader risk is the risk that a mispricing may worsen in the short run. The typical example includes companies that clearly are undervalued (and should therefore trade at a higher price). However, because noise traders may trade in the opposite direction, this mispricing can persist for a long time. Closely related to noise trader risk is horizon risk. The idea here is that the premium or value takes too long to be realized, resulting in a realized return lower than a target rate of return.

Model risk, also referred to as misspecification risk, refers to the risk associated with making wrong modeling assumptions and decisions. This includes the choice of variables, methodology, and context the model operates in. There are different sources that may result in model misspecification and there are several remedies based on information theory, Bayesian methods, shrinkage, and random coefficient models.¹⁰

Implementation risk is another risk faced by investors implementing trading strategies. This risk category includes transaction costs and funding risk. Transaction costs such as commissions, bid-ask spreads and market impact can adversely affect the results from a trading strategy. If the strategy involves shorting, other implementation costs arise such as the ability to locate securities to short and the costs to borrow the securities. Funding risk occurs when the portfolio manager is no longer able to get the funding necessary to implement a trading strategy. For example, many statistical arbitrage funds use leverage to increase the returns of their funds. If the amount of leverage is constrained then the strategy will not earn attractive returns. Khandani and Lo confirm this example by showing that greater competition and reduced profitability of quantitative strategies today require more leverage to maintain the same level of expected return.¹¹

Liquidity risk is a concern for investors. Liquidity is defined as the ability to (1) trade quickly without significant price changes, and (2) the ability to trade large volumes without significant price changes. Cerniglia and Kolm discuss the effects of liquidity risk during the “quant crisis” in August 2007.¹² They show how the rapid liquidation of quantitative funds affected the trading characteristics and price impact of trading individual securities as well as various factor-based trading strategies.

These risks can detract or contribute to the success of a trading strategy. It is obvious how these risks can detract from a strategy. What is not always clear is when any one of these unintentional risks contributes to a strategy. That is, sometimes when we build a trading strategy we take on a bias that is not obvious. If there is a premium associated with this unintended risk then a strategy will earn additional return. Later the premium to this unintended risk may disappear. For example, a trading strategy that focuses on price momentum performed strongly in the calendar years of 1998 and 1999. What an investor might not notice is that during this period the portfolio became increasingly weighted toward technology stocks, particularly Internet-related stocks. During 2000, these stocks severely underperformed.