Einstein defines insanity as “doing the same thing over and over and expecting different results”. We would also add “doing the same thing in a completely different environment and expecting similar results”.
Private Equity has emerged as a mature industry throughout the last decades. Best practices have been created and the model has iterated and perfected. Nowadays, there’s a textbook on global practices that managers replicate.
However, those best practices were developed having a very specific background developed western countries.
It is easy to understand that the US has a very different macroeconomic dynamic than Latin America. Why then should Latam managers just copy the same investment strategy used elsewhere? Our economies are much more volatile and sources of inefficiency much wider than what is seen in other parts of the planet. We argue that this should lead to a different investment strategy as well.
In this letter, we will try to explain what we consider to be the main flaws of PE players that lead to suboptimal portfolio and why these flaws are even more harmful in Latam. The problems we see are (i) flawed analysis of the likelihood of extreme scenarios and (ii) risk aversion due to allocation constrains.
The vast majority of Private Equity managers model their returns through answering the question: “what is the outcome we expect?”. The answer tends to be only what is most likely to happen and disregard scenarios with low probability of materialization. We believe that not considering all possible options when making an investment decision is a huge error. One should be more concerned with the range than the mean.
As an example, during the COVID crisis we were discussing what would happen to Latin America and the market in the future. Our scenarios went from a small contamination wave in which everything would return to normal in one month to our society being completely transformed into what we described as a new feudal system. Obviously these two extremes presented an extremely low probability, but they could have impacted tremendously the economy. If we stick only to scenarios with relevant probabilities, several scenarios would be discarded, leading to a flawed analysis process.
We usually see extreme scenarios being disregarded when we receive investment opportunities that present base, optimistic and pessimistic scenarios. Our perceptions is that the optimistic and pessimistic scenarios are usually poorly priced and analyzed. Generally, the variation between the scenarios is lower than the real optimistic and pessimistic scenarios That is, the optimistic scenario is slightly better than the base and the pessimistic scenario shows mild but still positive returns.
Our studies, on the other hand, show that extreme scenarios are actually much more likely and relevant According to our Spectra Insper database, 12% of the transactions in the region have returned > 5 x MOIC and 22% returned 0 x. Moreover, the > 5 x outliers represent 60% of the total capital returned to LPs.
Source: Spectra Insper Database. Histogram of PE/VC gross MOIC in Brazil in USD 1994 until March 2018.
That said, we have never received an investment memorandum that presents a pessimistic scenario where the company goes bankrupt. A better understanding of the probabilities of extreme scenarios is key to a better allocation.
An expected investment return stems from the sum of the probabilities of each scenario. Since almost no GP presents those extreme scenarios in their investment thesis, this makes the expected return to be biased. Either GPs are taking higher risks than mapped because they are not taking into account the probability of the thesis to be much worse than projected, or they are passing good opportunities because they didn’t understand the real optimistic case.
The main reason why we believe that these extreme scenarios we see locally occur more often than projected by GPs are due to unknown and uncontrollable variables, which are usually forecasted as (i) stable, (ii) with the same tendency as years before, or (iii) based on future curves. However, history shows that those variables present a much higher volatility in the Latam than elsewhere.
The table below presents the volatility of key indicators in Brazil and USA.
With such volatile indicators, how come the local industry is still following the same strategy of developed western countries and has not yet adapted the model to the local reality? Why GPs are still disregarding extreme scenarios and not seeking longtail opportunities? We believe that this strategy or mentality of longtail return has not emerged in more developed regions because they are much more stable than Latam so GPs wouldn’t miss much by forecasting variables as stable, with the same tendency as years before or based on future curves. With that, the increment by adopting this strategy in the portfolio would be marginal. However, since the volatility of Latam is much higher, the increment in the performance by adopting a different investment strategy would be extremely important.
We believe most of the local GPs are still only mimicking the investment strategy of investment tycoons from developed western countries instead of creating their own strategies based on the local realities. In our view, the answer to why the Latam investment strategy has not been adapted to local reality is mainly due to risk aversion created by allocation constrains.
As oddly as this may sound, we do believe that the majority of GPs are more risk averse than they should be. We are constantly faced with investment opportunities that have good risk return profiles, but that present higher probabilities of bad returns. For example, if there is a 1% chance of an investment returning 1,000 x capital and a 99% chance of a full write off, in theory this investment should be made.
However, no Private Equity GP would take that risk if they could only invest in 5/10/15 opportunities per fund. GPs would prefer an investment with a worse mean return and a lower probability of losing capital, even though in the long run the more volatile option presents a better return profile. Why is that? Because they are more concerned with surviving and raising their next fund, so they seek less volatile investments.
The chart below is a simple exemplification of this idea. We are simulating a portfolio of investments with each one of them presenting a binary return profile. The investments have a 10% chance of returning 40 x and 90% chance of returning 0 x (with a mean return of 4 x, which we consider to be an interesting profile).
As shown in the chart, the average Buyout GP would hardly invest in a portfolio of such opportunities even with it presenting an interesting mean return. If GPs could make > 50 investments in the fund, they would be able to take this kind of risk and thus seek long tail returns while still avoiding major losses. That’s similar to what we see in Venture Capital that oddly enough has not been adopted within local Buyout managers.
One could conclude that those longtail returns are either extremely rare to find, so GPs shouldn’t waste energy trying to find them, or they present higher competition since all players in the market are supposedly looking for better risk return profiles.
We disagree with this belief. We argue that there is a structural condition that shies players away from these types of risk. GPs structured their business model to invest in only a specific number of deals and LPs requirements forced GPs into traditional templates. They are expected to do what has been traditionally done in the industry so far, so GPs do not have the flexibility to adapt their model. Therefore, GPs avoid those longtail opportunities because they are more risk averse than they should be. So not only those longtail opportunities present better risk return profile, but also offer lower competition.
In such an unstable region like Latam it is almost impossible to answer questions like: Will the middle class grow? Can world leading technologies be developed in Mexico? Will agribusiness continue to be the main force of the Brazilian economy? Will this GP be the clear leader in this market?
There are several scenarios for these questions, each with its own probabilities and its payoffs. To better navigate in this local reality, one should focus on opportunities and asset classes with different betas that presents each its longtail returns rather than few investments in less risky investments (are they really that less risky??). In other words, in such an unstable region like Latam we believe a better risk return profile of a portfolio can be easily achieved by a larger diversification rather than by finding specific investments that are less volatile. Embrace the local volatility and use it in your favor.
Throughout the last decade we have been increasingly using the strategy of embracing the local volatility by seeking longtail returns and increasing the number of deals in the portfolio. The chart below presents the return distribution of Spectra II (our oldest fund to adopt this strategy). It can be seen that we have been successful in reaching the longtail with 19% of our fund returning > 5 x. Part of the strategy obviously leads to losses and shielding ourselves from those unfortunate results would also shield ourselves from homeruns. Having said that, we have also been successful in losing money in a smaller part of the fund when compared to the industry; As can be seen below only 9% of our portfolio returned less than the amount invested (MOIC<1 x).
We think it’s about time Buyout in Latam as an industry evolve from mimicking international players to creating different strategies that fit more the local reality. We believe that seeking more longtail returns and increasing the number of deals are two interesting adaptations for the model. However, they are just the beginning of this evolution. Definitely there are several other adjustments and iterations that should improve the local industry. What do you think? What are the other adjustments that Local GPs should adapt? We would love to hear your thoughts.
