A new study released in October 2021 revealed the sneaky behavior of mutual fund managers in selecting their comparative benchmarks. The SEC requires mutual funds to illustrate how fund performance compares to an index for 1, 5 & 10 year periods. The purpose is to help investors measure the value added from the wisdom of the fund managers. Importantly, benchmark comparison is the single most critical factor driving decisions when investors choose a fund, particularly among unsophisticated investors. But the SEC rule only requires that the benchmark selected be “appropriate,” which gives fund managers wide latitude to exploit and maneuver around that interpretation. The SEC also does not restrict a fund’s ability to change benchmarks and add new benchmarks at will, with fund managers having discretion as to why. The study makes it clear that managers favor themselves rather than objectivity when sculpting perception.
If it’s not bad enough that the vast majority of fund managers underperform their benchmark, you might be shocked to learn that many also engage in behavior to hide this fact. Kevin Mullaly and Andrea Rossi prepared the study, which focused on actively-managed domestic stock funds. It showed that fund managers backdate comparative benchmarks to erase history that might shed a bad light on the fund. The study found that when swapping benchmarks, the new one has at least -2.21% poorer performance than the original in 55% of all cases. The probability that this is a random occurrence is 0.004. Likewise, when funds remove an index, they tend to delete one with at least a 1.53% higher return than the one they replaced it with in 54% of all occurrences.
Funds with poorer relative performance and those facing redemptions or “outflows” are also more likely to engage in this activity. This behavior from fund management is also more prevalent in broker-sold funds, rather than no-load or institutional funds — the study linked this to less sophisticated investors who are less likely to recognize and call out such behavior.
One would think that the industry would be concerned with reputational risk and choose to impose comparative standards more closely associated with similarity, not skew. There are two industry standard statistics to objectively measure “similarity.” One is called r-squared (R2), which measures how correlated each change in an asset’s price is to a benchmark. A benchmark with a higher r-squared is a better match for the asset. The second is called Tracking Error, which in its simplest form compares the dollar value difference over a certain time period of a portfolio and a benchmark. The SEC could just about end the shenanigans with a requirement that “appropriate” means the benchmark with the highest r-squared, along with prohibiting backdating of results: you chose it, you eat it. This would be critical to restoring honor to the system since 82% of index selections by fund companies for comparison DO NOT currently use the index with the highest r-squared.
After reading the study, I felt disappointed in an industry I thought had more class. But it is just another reason why my firm opts out of mutual fund investing in general and chooses the indexes themselves or securities that mimic the indexes, which lowers costs and lowers taxes, and gives investors a purer form of participation.
Please see IMPORTANT DISCLOSURE information.