This one tool can separate skilled investors from lucky ones

By Dan the Quant

Investing in the stock market is about making predictions about the performance of companies in a certain horizon. By nature, predictions are uncertain, but in the world of investing, uncertainty is even more pronounced because prices of company stocks depend on the collective perceptions and predictions of all humans participating in their respective markets. And as we frequently say in Spanish: each head is its own world.

Many people try to make a living out of building portfolios of stocks and selling them to investors. They claim to have exceptional skill at picking the most successful stocks and deciding how much of each stock to own. The problem with this declaration is that it’s very difficult to prove or disprove because getting good returns for a few years can result from good luck, good skill or a combination of both.

So, how do we identify superior portfolios and distinguish them from inferior ones; especially when we are making a decision today and we won’t have information about performance until several years in the future? How do we know if fund managers are doing the best they can to get us good returns on our investment?

Enter risk models. Since 1952, Markowitz published a paper explaining how selecting a portfolio is a trade off between return and risk. In the finance world it is known that there is no free lunch and therefore any return on investment comes from exposing the capital to certain risks. This is generally true except when there is a mispricing induced for instance by uninformed market participants. In this case a person who owns an asset believes the asset is worth less than other people are willing to pay for it. If a very smart trader gets this information, this person will immediately buy it from the uninformed participant and also immediately sell it to the person who are willing to pay more for it making and instantaneous profit.

Mispricings are important because they can occur and they can subsist for long periods of time if most people participating in the market misinterpret the information about certain securities. Now, if a misprice exists, this mean that the relation between risk and return is no longer the same. It’s possible that some people need to risk more for an expected return if the asset is overpriced or risk less for the same expected return if the asset is underpriced.

Here is where risk models help us make decisions. A risk model allows us to determine the risk of each security in our analysis. If we expect two stocks to provide the same return in a 20 year horizon, but one of them has more risk than the other, then we know the most rational decision would be to prefer more of the less risky one. Notice that I didn’t say all from the less risky one because there is value in not putting all your eggs in the same basket.

Risk models also help us understand the correlation between two stocks and consider this measurement while making decisions. For instance, if one of the stocks performs much better in economic recessions (i.e. basic products are demanded more in hard times than luxury ones, therefore Walmart and the likes excel during these times) and the other one performs better during market booms, it is very likely that these two are negatively correlated. If this is the case it’s a good idea to own both so one can provide us with good returns when the other one can’t.

One more benefit of risk models is that they can help analyzing what risks are portfolios more exposed at different points in time. If we run a certain portfolio on a risk model, the model can attribute the total risk to different factors and this can make us more conscientious of the risks we are facing on our investment. If we are familiar with the kind of risk we are holding, it’s less likely that we panic when thing go against us. And if you are investing for the long term, things WILL go against us eventually. Risk models help us asses if it’s worth waiting the downturn or if it’s better to make some corrections in our portfolio for the incoming months or years.

Risk models are the tool of choice of professional investors because they help us understand the relation between risk and return, they estimate the benefits of diversification and identify to what of the known risk factors is our investment more exposed. For the previous reasons, I invite investors to learn more about risk models and upgrade their game while making investment decision making a more scientific endeavor.

1 Comment