Swedroe: Arbitrage Capital Increases Market Efficiency

Market efficiency can vary based on different factors.

The hypothesis of an efficient market is based on the concept that informed, rational traders would arbitrage away any temporary deviations from “correct” prices. Thus, price efficiency depends upon the actions of arbitrageurs and the availability of arbitrage capital.

When arbitrage capital is plentiful, anomalies should be quickly eliminated. However, if capital is scarce or there are sufficient frictions (such as trading costs, regulatory constraints and borrowing costs), while anomalies may shrink, they can still persist.

Effects Of Arbitrage Capital Availability

Ferhat Akbas, Will Armstrong, Sorin Sorescu and Avanidhar Subrahmanyam, authors of the paper “Capital Market Efficiency and Arbitrage Efficacy,” which was published in the April 2016 issue of the Journal of Financial and Quantitative Analysis, contribute to the literature on market efficiency by exploring the premise that the availability of arbitrage capital varies over time, which results in dynamic variation in the predictability of cross-sectional stock returns.

The authors begin by measuring return predictability using the ex-post return performance of a “quant” strategy designed to trade on five capital market anomalies documented within the academic literature.

Their study covered the period 1991 (the earliest available date for mutual fund flows) through 2009, and five anomalies: momentum, profitability, value, earnings and reversal. They noted: “Some of these anomalies earn large paper profits, and have persisted out-of-sample indicating that it is a challenge to attribute them to data mining. Further, it is difficult to come up with a risk-based story consistent with many of the anomalies documented in the literature. This suggests that cross-sectional anomalies may, at least in part, reflect temporary inefficiencies in market prices.”

To measure the amount of available arbitrage capital, Akbas, Armstrong, Sorescu and Subrahmanyam used flows to mutual funds whose trades mirrored those of their quant strategy (abbreviated as “quant funds”).

They identified these quant funds by regressing the returns of all mutual funds on the returns of a quant strategy and then selecting those with the highest return correlation. They write: “The flows to quant funds act as a proxy for flows to arbitrage strategies where arbitrage capital is used to target temporary pricing inefficiencies and move stock prices toward efficient benchmarks.”

Additionally, the authors noted: “Periods marked by high arbitrage flows are periods during which markets are more efficient. These periods are likely to see a correction of cross-sectional mispricing, resulting in lower returns to the quant strategy in the future. Conversely, any mispricing that is present at the beginning of periods with low arbitrage flows will likely persist throughout the period. Thus, periods marked by lower flows will be followed by periods with higher cross-sectional return predictability, which will manifest in the form of higher returns to the quant strategy.”

Research Findings

They selected mutual funds with loadings in the top 10% of all funds in terms of their exposure to the hedge fund strategy. These are the funds “most likely to trade on the type of cross-sectional price inefficiencies that are embedded in the construction algorithm of the quant strategy.”

First, the authors found that the quant strategy demonstrated strong performance over the full period they examined, from 1975 through 2009, as well as over four subperiods. They add: “This performance is particularly puzzling during the most recent decade, given the large number of funds that actively trade on these factors, as well as the vast amount of information (especially from academic research) available to fund managers about cross-sectional return predictability.”

The authors did find their results “support the notion that poor past performance and high volatility both lead to lower future quant flows.” This is consistent with previous research, which has found that most premiums tend to shrink after publication. Moreover, they add: “Investors are more likely to allocate money to quant strategies following strong market performance.” In other words, market efficiency is time-varying.

The authors concluded that “enhanced arbitrage flows to quant funds predict lower future profitability of the quant strategies, and thus greater capital market efficiency. This finding underscores the point that market efficiency is a dynamic concept, because markets become efficient owing to intervention by arbitrageurs, whose efficacy varies over time as the availability of arbitrage capital varies.”

Akbas, Armstrong, Sorescu and Subrahmanyam write: “Our results are robust to several alternative methodologies, including variations in the construction of the quant algorithm, the use of risk-adjusted quant returns, and the use of flows obtained from market-neutral hedge funds instead of mutual funds.”

The authors concluded: “The main empirical result in this paper—the negative relation between fund flows and future quant returns—is clearly consistent with the notion that an increased flow of arbitrage capital enhances cross-sectional market efficiency and vice versa—we call this a phenomenon induced by ‘limits-to-arbitrage.’”

They add that their results “provide a reasonable explanation for the persistence of cross-sectional return predictability, despite the increasing number of hedge funds that seek to trade based on the quant factors. Whenever stock prices are pushed away from equilibrium by exogenous forces, the presence of arbitrage capital is required in order to reestablish capital market efficiency. Absent such capital, predictability in the cross-section of stock returns can persist. Conversely, if arbitrage capital were to become freely available at all times and without rationing, the cross-sectional predictability would disappear. However, so long as the availability of arbitrage capital is time varying, the stock market is likely to exhibit time-varying predictability of returns in the cross-section.”

This paper contributes to the literature showing both that market efficiency is time-varying and the publication of anomalies doesn’t necessarily mean they will disappear, though they do have a tendency to shrink. Limits to arbitrage and capital flows impact the ability of arbitrageurs to play their role in market efficiency.

Larry Swedroe is the director of research for The BAM Alliance, a community of more than 140 independent registered investment advisors throughout the country.