Why did we invest in one of the AQR funds?

Elena Chirkova

Nov 27, 2024

In early November, our GEIST fund made investments in one of the funds of the AQR group of the legendary investor Cliff Asness. Asness mainly specializes in stocks and is a staunch supporter of value investing. At the same time, AQR stands for Applied Quantitative Research. There is no contradiction here; quantitative analysis can and should be applied to the value concept. The uniqueness of the fund lies in the fact that it has no manager. The fund is managed by a computer program. We are encountering this for the first time. Usually, even if a fund trades according to some formula with a robot, it has a manager. But here it doesn't. Perhaps to emphasize that the algorithm makes decisions 100%. 

The idea of the fund stems from a well-known study by AQR employees, in which they discovered over a long series of statistical data that it is most profitable to buy "quality" and "low volatility." Here, "quality" and "volatility" are understood as two of the four now-standard factors that can be used to break down stock returns — these are "value" (here value means cheap stocks), "quality," "volatility," and "growth," or "momentum." As modeling showed, taking volatility into account, it is most profitable to buy quality stocks. Fast-growing ones can grow very quickly for a while, and then fall just as quickly and lose to quality ones. Why is it important to maximize the "return/volatility" ratio? Because the lower the volatility, the more leverage can be taken, thereby further increasing returns. (Technically, leverage is taken by acquiring contracts with leading banks like swaps, which require minimal reservation.) 

The results of this study were published in the famous article “Buffett’s Alpha” by authors Andrea Frazzini, David Kabiller, and Lasse Heje Pedersen, which I read immediately after its release in 2012 when I was studying Buffett's strategy. The name of the investment guru is in the title of the article for a reason. AQR believes that Buffett's genius lies in the fact that he was the first to guess to invest in quality low-volatility stocks without any complex research. This is how he earned. In my opinion, several other factors and features of the strategy contributed to his beating the market, but that's not important. 

The fund's portfolio is highly diversified. The algorithm analyzes almost all stocks of developed markets with a few exceptions. For example, biotech company stocks, which depend on the success of one or two studies, and low-liquidity stocks are excluded. The rest are ranked based on numerous factors from most promising to least promising. Approximately 80% of the stocks that the model deemed the best are bought long, and short positions are opened on the remaining ones. In addition, the model assesses the probabilities of market correction and its potential depth. If a large and rapid correction — over 20% per month — becomes likely, the algorithm begins to reduce market exposure. 

The model on historical data since 1995 showed a return of 13.9% per annum. The fund should theoretically earn less due to expenses and fees. However, in practice, since 2012 it has earned 14.8% per annum on average, which is related to the fact that the market itself grew better on average during this period than since 1995. We fully share AQR's investment idea, and therefore allocated the maximum amount for us to the fund — 4% of the fund's net assets.