Understand the difference between luck and skill to assess your returns

By Shyam Sunder   – Managing Director– PeakAlpha Investments, Livemint article posted 18 Dec 2020

Determine the result, ascertain its cause, assess the momentum to better predict performance

For all the talk, the pretence of knowledge, even the condescension she had got from her husband in terms of investment skill, she had beaten his pants off. Her investment had grown by 28%, whereas her husband had lost money. Here is the interesting part. Both had invested equal amounts of money and in the same scheme. The only difference was that the husband invested in February this year, whereas the wife invested in April. The husband had caught the crest of the wave, whereas the wife had caught the trough that followed a few months later. Was that skill or luck?

At the best of times, and armed with sophisticated attribution models, separating luck from skill in investment performance is difficult. Anyone who has invested across market cycles will know that some decisions that were backed by robust and thorough analysis suffered in comparison to other decisions that were made with the most cursory analysis. Having tracked the performance of mutual fund schemes and fund managers closely for 15 years, we have seen superstar fund managers from one era scrape the bottom of the barrel in another and vice-versa.

Yet, any endeavour of investing over the long term would be significantly hampered without an ability to analyze performance and draw the right conclusions. Here are some of the obvious questions that need to be addressed. First, the result—how well did your investment decisions fare? Second, the cause—how much of the result can be attributed to luck and how much to skill? Third, the “so what”—given the past, what should you do going forward? Let us look at each of these.

The result can be described in a few different ways. First, in rupees—how much money did you gain or lose? Second, as a percentage of the amount invested, or absolute return. Third, as a percentage after adjusting for time invested. Surely, a 20% return in one year is better than a 20% return in three years, right? This measure is usually called annualized return. Fourth, as a comparison to a benchmark. If one made 18% return where the benchmark made 20%, is that better than another making 16% where the benchmark made only 14%? Another way of looking at the same thing is to adjust the return for the amount of risk taken to achieve that return, something we refer to as the Greek letter alpha. We all need some measure of compensation for taking additional risk, often known as risk premium.

These measures do sound a bit like the blind men and the elephant, where they all came up with different assessments. Yet, people have many questions regarding their investments, and each measure above answers a different question. If you are looking to compare investment performance, using alpha and comparing against the benchmark are useful places to begin.

The cause is trickier, because here one needs to separate luck from skill. Timing is one of the reasons that can drive improper conclusions, as we have already seen. Further, when one is comparing two investment experiences, it is important to ensure it is an “apples to apples” comparison. Comparisons between fundamentally different instruments that are likely to behave differently are flawed, such as a mutual fund with a unit-linked insurance plan (Ulip), a hybrid scheme with an equity scheme, or a large-cap with a mid-cap. The third is to distinguish between consistency and “flash-in-the-pan” outcomes. Superior performance must be repeatable to be claimed as a skillset.

Finally, the “so-what”. To what extent can our conclusions from these comparisons be extended into the future? When is change necessary? If you are happy with the result, should you stay the course? If your results aren’t great, do you need to make changes? If our goal is to predict what will deliver superior returns in the year ahead, let us reflect on the inputs. While there are many factors, here are some to consider. The investor’s skill, whether it is regarding instrument selection, asset allocation or execution is likely to persist. The processes followed, whether they be research, investment or risk management, are also contributors. However, it is important to assess whether the investment environment has changed. If so, what worked in the past may not work in the future. Winners in a bull market are often different from winners in a flat market.

I recommend the above three-step process—determine the result, ascertain its cause, assess its momentum—as a way to achieve greater predictability in investment performance. At the very least, it will establish a lingua franca for its discussion.