Seeking Alpha
Profile| Send Message|
( followers)  

Value-at-Risk, widely known as VaR, is one of the most widespread tools for measuring and managing financial risk. VaR is a predictive metric, expressed in just one number that gives an idea about the extent of risk in a given portfolio. Because of its simplicity, VaR has become the industry and regulatory standard, thus a large proportion of financial institutions and professional asset managers around the world employ VaR in their risk management process in one way or another. However, very few retail investors follow the suit.

In this article, I would like to discuss reasons why retail investors haven't adopted a tool that is so prevalent in the institutional investors' space. Among possible explanations of this phenomenon, three stand out as plausible ones:

1) Insufficient knowledge of VaR analysis;

2) Lack of resources to calculate VaR on a regular basis;

3) General disbelief in statistical measures as "they are all wrong and fail when you need them most".

Let's address these points one by one.

First, people tend to avoid what they don't understand. If an investor has never had exposure to the world of statistics, one is likely to be reluctant to use a tool such as VaR. However, there is good news for those without a degree in mathematics or econometrics - in its essence, VaR is very simple.

There is loads of material online that describes and analyses this statistical measure (a good starting point is this paper by Aswath Damodaran of NYU). In brief, VaR captures three elements: a specified level of loss in percentage terms, a fixed time period over which risk is assessed and a confidence level. If we consider a market standard 1-day VaR at a 99% confidence level, a reading of 2% means that on 99 out of 100 days an investor can expect the maximum loss in a portfolio not to exceed 2%.

Although it is useful to understand the computational process behind statistical metrics, it is not always essential. I would like to draw a parallel here with weather temperature forecasting: you don't really know how exactly it was calculated, you appreciate it's not precise but you still regularly check and cautiously trust it.

Second, even if investors are comfortable with the notion of VaR, it is usually not an easy task to calculate that single number on a daily basis. Historical data, computational power and software package requirements are often enough to deter retail investors from the use of statistical metrics like VaR. This is particularly challenging for active traders whose portfolio is different at the end of each trading day. Fortunately, technology has progressed and there are freely available online resources such as InvestSpy that do the work for you in a matter of seconds - you only need to specify security tickers and their weights in a portfolio. Therefore, scarcity of resources is no longer a real issue.

Third, some investors simply don't want to waste time dealing with statistical measures that have been proven wrong in a number of cases. For instance, correlations have a tendency to spike at the times of distress, diminishing the diversification benefit investors believe they have. To a certain degree, I agree with the skeptics. None of the risk measures should be followed blindfolded or assessed in isolation. But instead of avoiding them altogether, my suggestion is to use VaR for guidance rather than prescriptive action.

To summarize, there are good reasons why institutional investors pay attention to VaR. And if you don't do so, you may want check how much you have at stake as well. After all, even Warren Buffett's No.1 rule is to never lose money.

Source: Can Retail Investors Unlock Value In Value-At-Risk?