This paper will be presented at the 2017 Northern Finance Association Conference being held in Halifax, Nova Scotia from September 15-17. Canadian Investment Review is proud to be a media sponsor for this event for the third year in a row.
Hedge funds manage around three trillion US dollars currently and their capital is coming largely from sophisticated investors. Since these investors are also by and large institutional, therefore less prone to behavioural biases specific to individuals, hedge funds flows are often perceived as “smart money” by financial economists and practitioners alike.
However, anecdotal evidence about investment decisions into hedge funds suggests that investors attribute large importance to qualitative data about the funds in which they plan to invest. Talk referring to “looking into the hedge fund manager’s eyes” or about personal relationships all invites decision making that may ultimately be subject to various behavioural biases.
Consistent with this evidence, in our paper “Hedge Funds Flows and Name Gravitas,” we produce evidence that investors chase funds whose names are correlated to a combination of words relating to economics and geopolitics or that convey power. Such words associate with influence, authority, know-how and good judgement — qualities to whom we refer as gravitas.
The notion of gravitas originates from the literature on leadership, where it has been used to signal desirable qualities of those in charge. In our context, a name with gravitas similarly conveys that the fund “knows what it’s doing,” that it is connected politically as well as able and apt to understand and exploit economic opportunities through its investments.
What is gravitas?
In order to uncover the popular patterns in hedge funds names, we use a psychological dictionary, the Harvard Inquirer, which has been used by authors looking to uncover semantic content in financial or news reports. We use the technique of principal components from linear algebra in order to uncover main “themes,” or word combinations that are popular in hedge fund names and discover that gravitas is the most popular such theme.
The precise definition of gravitas is that it consists of a linear combination, in nearly equal proportions, of words from politics, from economics, of names of nations, other geopolitical names and of words conveying power.
Name gravitas and flows
To begin with, given the leadership literature, it may make sense for an investor to chase gravitas in the hedge fund managers’ personality. However, this is not the effect we document. We in fact explicitly control for this possibility, comparing side-by-side funds that either have the same manager, or belong to the same hedge fund company, but have different names. We present evidence that it is name gravitas, rather than a personality trait of the manager, that investors are chasing.
We also find that the size of the name-chasing effect is economically significant. Averaging across various regression models, we obtain that adding one more word with gravitas to the name will increase the flow into the average fund by $227,120 every year. Similarly, we find that an increase by one standard deviation in name gravitas attracts $759,967 more in flows to the average fund every year.
Finally, this relationship between name gravitas and flows remains after we control for a large set of variables that the literature on hedge fund flows (or, more generally, fund flows) deems as important.
“Cool” names and fund performance
The interesting question is whether name gravitas is positively correlated with fund performance. Such a correlation should however be surprising, given we established that it is not the fund manager, but the name gravitas that investors are chasing. If indeed name chasing is irrational, then, as the better funds should avoid irrational investors (because these investors may act irrationally in other respects, for example, they may withdraw their funds when the fund needs capital the most), we should expect that the better funds have less name gravitas (as having more gravitas attracts the wrong investor type). Consequently, we should in turn expect that the funds with more name gravitas underperform.
Indeed, we find evidence that funds with more name gravitas consistently underperform those with the lowest name gravitas. For example, funds whose names are positively associated with gravitas have annualized alphas that are 0.97% (or 0.73%, when an alternative factor model is used) lower than those of the funds with negative gravitas exposures. Similarly, the annual Sharpe ratios of the high name gravitas funds are 0.18 lower, their average annualized returns are 0.82% lower, the maximum drawdowns are 5.06% higher, and their volatilities are 0.57% higher than those of the funds whose names are negatively associated with gravitas. These differences are also statistically significant for alphas and Sharpe ratios. Although these differences are not statistically significant in the case of returns, for example, their sign is consistent with the assertion that high name gravitas funds underperform.
Finally, despite the fact that funds with gravitas receive more flows, we document that these funds have a higher propensity to fail than those with less name gravitas. For example, probit estimations suggest that the probability of attrition of the funds with the highest gravitas measure in our sample is 5.38% higher than that of the funds whose name has no gravitas. This is hardly surprising, as the high gravitas funds attract investors who may flee when the fund needs them most.
Consistent with these funds attracting less sophisticated investors, we also document that high name gravitas funds use more liquid strategies (and thus have shorter lockups and fewer restrictions), charge higher management fees (so they do not to rely on excess performance to get paid) and report to fewer hedge fund databases (that is, they do less indirect advertising work). This evidence seems to indicate that underperforming, less sophisticated funds name themselves with more gravitas as a marketing ploy to attract more investors.
Duped forever?
Having documented that some investors chase hedge fund names does not mean that these investors never learn about the true ability of their funds. Indeed, we find evidence that investors learn and as they do, adjust their irrational reaction to names. More specifically, we show that flows’ response to name gravitas declines with the fund’s size as well as with the age of the fund. The gravitas effect is estimated to disappear as the average fund gets older than 9 years of age (the average fund in our sample is only half as old) or when a fund reaches a size that is so large that none of the funds in our sample achieved it yet. It therefore seems that the gravitas effect is resilient — even though investors learn, fund names still matter.
Furthermore, we additionally find evidence that flows’ sensitivity to gravitas declined in the latter half of our sample, consistent with the hypothesis that investors became more familiar with hedge funds as an investment vehicle, and also with the idea that the means to perform due diligence and to estimate the quality of hedge funds have improved recently.
We also find that investors’ response to gravitas in fund name is conditional of that investor’s level of sophistication. For example, we find evidence that as the hedge funds minimum investment requirements are higher – consistent with the clientele of that fund being wealthier and potentially more sophisticated – flows respond to gravitas in a less sensitive manner. Furthermore, we also show that flows’ sensitivity to gravitas is lower for funds whose investors are qualified purchasers.9 To estimate just what type of investor might be impervious to name gravitas, we can use the regression analysis in the main paper Joenväärä and Tiu (2017) in order to calculate the minimum investment at which an investor will not respond to name gravitas. We uncover this number to be around U.S. $50 million – an amount that is possible, but highly implausible given the current industry practices (by comparison, the median fund in our sample is half as large in total assets under management). It is puzzling that even after we control for learning and investor sophistication the flows’ response to gravitas does not completely disappear.
Conclusions
Hedge fund managers and their clients are perceived to be the most sophisticated investors. One might expect that this sophistication makes both less susceptible to irrational behaviour. While some authors document that hedge fund managers themselves may be influenced by behavioural factors, a fact that is perhaps unsurprising given that the hedge funds industry has rather low barriers of entry, our paper is the first to document the presence of a bias in the decision making process of (sophisticated!) hedge fund investors.
Although more recently in the sample our results are weaker (but still significant) — which is likely a testament to the recent quality improvements in hedge funds due diligence and in understanding their investment strategies — it is puzzling that the gravitas effect is resilient to investor learning or controls for investors sophistication. This implies that more could be done in terms of how hedge funds investment decisions are made or in terms of defining which investors are appropriate for hedge funds.
Download a full version of this paper with footnotes and references included.
This paper will be presented at the 2017 Northern Finance Association Conference being held in Halifax, Nova Scotia from September 15-17. Canadian Investment Review pleased to be a media sponsor for this event for the third year in a row.