Monday, April 25, 2011

MorningStar Measures Adverse Timing

That investors have different average returns than the assets they invest in, is the least appreciated return bias. Ilia Dichev (2005) looked at a variety of indices, and equity net inflows, and found this adjustment reduces returns by 1.3% for NYSE/AMEX for the period 1926-2002, 5.3% for Nasdaq from 1973-2002, and 1.5% for 19 major international stock exchanges from 1973-2004. These are big numbers, and almost universally an ignored by those studying the equity 'risk' premium. You can be sure that if this adjustment could be made consistent with some risk story it would be taken seriously, but what we see is influenced by what we believe and so we don't see this effect. To be precise, we don't see it as 'interesting' or anything other than random noise, because it does not fit into any big paradigm, and killing the equity risk premium would destroy the primary empirical point for asset pricing theory.

Morningstar is now measuring this for fund categories, which is very helpful. They note that due to adverse timing (!), recent attempts by investors to 'chase' market returns have hurt them:

In 2010, the average domestic fund earned a return of 18.7% compared with 16.7% for the average fund investor, making for a gap of 200 basis points. For the trailing three years, that gap was 128 basis points. For the past five years, it was 98 basis points, and for the past 10, it was 47 basis points.

Interestingly, their studies all show a large bias that gets smaller as one goes backward over time. I'm skeptical adverse timing is primarily a recent issue because the 2001 internet bubble, as John Cochrane emphasizes, is the poster-child for adverse timing. Perhaps there's a straightforward bias to their metric, because it seems strange that it attenuates for every asset class at the same rate.

Further, as Antti Ilmanen notes in his new book Expected Returns, 'momentum' in strategies is statistically significant, it is there, and suggests a 'Mathew Effect' for allocating capital. I wonder how the adverse timing, which implies average investors are buying more prior to declines than prior to increases, relates the momentum in asset strategy returns.

hat tip: Tadas Viskanta

No comments: