In the midst of the Northern Hemisphere’s “winter blues” back in February 2007, we told you about a research paper showing that hedge funds also suffer from a form of “seasonal affective disorder.” Only theirs was in the summer, not the winter (maybe a result of the old adage “sell in May and go away“?) This year seems to fit the pattern. After a promising start, hedge funds hit a soft spot beginning in May (although July looks to have been okay).
But what are the characteristics that put a fund at risk of experiencing the hedge fund equivalent of “S.A.D.”? A new paper by Mafalda Ribeiro and C. Machado-Santos of the Universidade Lusofona do Porto (ULP) in Portugal compares the seasonality of two separate hedge fund indices: the CS/Tremont hedge fund indices, which are asset-weighted, and the Edhec “indices of indices”, broad indicators of hedge fund performance that include both asset-weighted and fund-weighted indices.
As you can see from the Edhec indices on the chart below from the paper, hedge fund managers seem to enjoy the beach or cottage come summer (click chart enlarge):
Long biased strategies such as emerging markets seem to suffer the most during the dog days of August while equity market neutral (sort of) holds its own. This suggests that an equity beta may be responsible for part of this phenomenon (equities also tend to suffer in the summer and early fall).
Those who see hedge funds as glorified market makers may see this as further evidence that the lack of liquidity that accompanies the Northern Hemisphere summer also lies as the root of this phenomenon. (This study, for example, links poor summer equity returns to low liquidity and higher bid-ask spreads.)
Ribeiro and Machado-Santos also examine monthly results for the asset-weighted CS/Tremont indices (chart below from paper, click to enlarge).
Regular readers may recall our observation in the days after the Madoff story broke that this asset-weighted index was due to take such a significant hit from the mega-billions in Madoff feeders that its very usefulness could be called into question. (Witness the November EMN performance above). Ultimately, CS created a Madoff-free version of its index and the authors of this paper also examined that index too. But the fact remains that the CS/Tremont indices are heavily weighted in big hedge funds.
So by examining the differences between the Edhec indices and the CS/Tremont indices, we can likely glean some useful information about the seasonality of large hedge funds. For starters, assuming hedge funds are able to undertake some kind of “returns management”, large long short equity funds may realize more of their potential returns in December than smaller ones (compare Decembers’ red bars in the charts above). Also, it would appear that large global macro managers may have a harder time in August and September than smaller ones (green bars).
A casual observation of these two charts also suggests that, in general, larger hedge funds seem to experience higher highs and lower lows than the broader industry. This makes intuitive sense since the CS/Tremont indices are less diversified. Note also that larger market neutral funds may perform better in the summer months than smaller ones – thus avoiding the ravaging effects of dried up summertime liquidity. There are further differences between these charts that may or may not be instructive. But you can play “find the difference between these pictures” on your own.
The paper finds that the “seasonality coefficient” of 5 of the 7 strategies is highest in December (using the Edhec indices). But for 2 of them – Fixed Income Arb and Convertible Arb – other months took first prize (April and January respectively).
But when you examine the results from the CS/Tremont (asset-weighted) indices, only 3 of 7 peak in December. Market Neutral, Event Driven and Convert Arb experience their highest seasonality coefficient in the winter (March, January and January respectively). And fixed income was inconclusive .
So do some types of large hedge funds do better in January than in December? And is this in any way driven by “returns management” (a close cousin of a CEO’s “earning management”)? Perhaps. But as Ribeiro and Machado-Santos write, you don’t need to know the reasons in order to “buy low and sell high” hedge funds at different points during the year.