Correlated Trading by Pension Fund Managers

Despite the common perception that institutional investors herd, it is difficult to identify the reasons for correlated trading. For example, managers might buy into or out of the same securities over some period due to correlated information, perhaps from analyzing the same indicator. Alternatively, a manager might infer private information from the prior trades of better-informed managers and trade in the same direction. Also, managers might disregard their own information and trade with the crowd due to the reputational risk of acting differently from other managers. Finally, managers might simply have correlated preferences over certain types of securities.

In a recent paper, I study correlated trading by Colombian pension fund managers in the presence of a peer-based underperformance penalty known as the Minimum Return Guarantee (MRG). The MRG resembles a reputational risk, in that the manager might be penalized for having lower returns than her peers. With the MRG, the risk is explicit as the manager will be penalized financially if returns are below the maximum allowed shortfall relative to the peer benchmark. The rationale for the MRG, which is a common piece of the regulation in defined contribution pension systems, is to discourage excessive risk taking by pension fund managers.

The Colombian case is ideal to study correlated trading by pension managers for at least two reasons. First, it is the archetypical country that adopted a Chilean-like pension reform based on individual retirement accounts, replicating much of Chile’s regulation, including the MRG. Characterizing the effects on the trading behavior of the peer-based performance penalty offers some interesting lessons for the many economies that also followed the Chilean example. Second, the Colombian government changed the MRG formula in June 2007, increasing the maximum allowed shortfall and thereby loosening the MRG. This policy change provides a natural experiment that allows me to measure the change in behavior associated with the change in the underperformance penalty, arguably holding constant other possible explanations for correlated trading.

The evidence indicates that in order to mitigate the risk of falling below the minimum return requirement, managers follow the portfolio of their peers by buying securities in which they are underexposed relative to the industry; a behavior that is more pronounced for managers underperforming vis-à-vis their peers. In the sample period, managers did not sell securities with overexposure relative to their peers. Since these pension funds were in an accumulation stage, managers were presumably able to reduce their participation in any security by holding their absolute position in that security constant.

The results also suggest that because of the peer-based performance penalty, a manager is likely to have a lower participation in assets with high volatility as well as assets with low liquidity. This happens for at least two reasons. First, even when the portfolio of a manager is close to her peers, small differences in holdings of highly volatile securities potentially increases the exposure to the peer-based performance penalty. Second, less liquid securities might be harder to rebalance as fund managers may find it difficult to either enter or exit these positions at their requested price, experience execution delays, or receive a price at execution significantly different from their requested one. These transactions costs would make pension fund managers more exposed to falling below the required minimum.

In Colombia, there are only four of these pension funds managers and the total assets under management are large relative to the size of the domestic equity and bond markets. Given these facts, asset prices are likely to be affected by the trades of these institutions and correlated trading might amplify shocks to the domestic capital markets. Additionally, market power combined with the MRG might trigger the equilibrium in which pension fund managers end up with portfolios that are overly weighted in securities with high liquidity and low volatility, which in most cases are securities which shorter maturities (see for example Opazo, Raddatz, and Schmukler (2014) who found this type of behavior for Chilean pension funds). This equilibrium is undesirable, not only because it undermines the ability of domestic capital markets to generate long term financing for firms, but because it is against the long term objective of the contributors to these pensions funds, namely, maximizing income at retirement.

The results in this study challenge the use of peer-based performance measurements for pension funds. Alternative performance measures based on risk-adjusted-returns should eliminate some of the pervasive incentives that lead to herding among these managers. The regulatory authority needs to focus on aligning the long term objectives of the fund contributors with the sometimes short term objectives of fund managers.  In this sense, performance fees and other compensation schemes based on returns are only desirable as long as they serve this purpose.

References

Opazo, Luis, Claudio Raddatz, and Sergio Schmukler, 2014. “Institutional Investors and Long-Term Investment: Evidence from Chile.” World Bank Policy Research Working Paper 5056.