Why investors should avoid hedge funds

At the beginning of January, Simon Lack published The Hedge Fund Mirage: The Illusion of Big Money and Why It’s Too Good to Be True. It basically does for hedge funds what the Kauffman Foundation did for private equity, and shows that while fund managers can do extremely well for themselves, fund investors would be better off avoiding the asset class entirely.

Markopolos eyes a fortune from BNY whistleblowing

The book was well received (see reviews by Jonathan Ford and Jonathan Davis, in the FT, and James Pressley, for Bloomberg), and even Andrew Baker, the chief executive of the Alternative Investment Management Association, said that “much of the book is sensible advice for investors in hedge funds about the need to be well-educated about the industry”. Baker did quibble with the message of the book, saying that it revealed “more about investor behaviour than manager performance”. But if managers really cared about their investors, they would surely take that message to heart, rather than simply keep on trying to maximize their own performance and pay. Baker concluded with figures showing capital flowing back into hedge funds, and concluded not that those investors were being foolish, but rather that “many investors consider hedge funds not a mirage, but an oasis.”

Fast forward to this week, however, and now Baker is singing a very different tune:

Many of us in the industry looked at the arguments in the book with initial interest, and then growing scepticism. Many of the most sensational claims appeared not to be backed up by any figures. Where there were figures, the methodology was slapdash or flawed, and further undermined by simple errors. We, in the Alternative Investment Management Association, began to wonder if The Hedge Fund Mirage was itself an illusion.

Baker and the AIMA have in fact now published a 24-page paper entitled “Methodological, mathematical and factual errors in ‘The Hedge Fund Mirage’”, seeking to debunk the book. But a close reading of the paper reveals that there’s much less there than meets the eye.

In fact, the AIMA paper has convinced me of the deep truth of Lack’s book in a way that the book itself never could. Reading a book, it’s often very hard to judge just how reliable the author is, or how cherry-picked the data might be. But if a high-profile hedge-fund industry association spends months putting forward a point-by-point rebuttal, and that rebuttal is utterly underwhelming, then at that point you have to believe that the book has pretty much got things right.

AIMA kicks off by responding to the first sentence of Lack’s book: “If all the money that’s ever been invested in hedge funds had been put in Treasury bills instead, the results would have been twice as good”. They come out fighting, saying that “extraordinary as it may seem, nowhere in the subsequent 174 pages is this central claim supported by clear evidence”. And of course they say that the claim is not in fact true:

What our own calculations, using the same core data and time period as the book, in fact show is that investors allocated $1.24 trillion into hedge funds between 1998 and 2010 and have $1.78 trillion to show for it, a 44% return, while the same amount invested in T-bills over the same period would have produced $1.52 trillion, a 23% return. So to paraphrase the book’s opening line, if all the money that had been invested in hedge funds had been put in T-bills, the results would only have been half as good.

Read on a little bit, however, and it turns out that AIMA can — and did — replicate Lack’s results. The group publishes two tables (Table 1 and Table 4): the first one, which attempts to reproduce Lack’s methodology, concludes that between 1998 and 2010, hedge fund investors lost $6 billion, while T-bill investors made $309 billion. That’s even worse than what Lack says! The second table, using AIMA’s own methodology, does indeed show hedge funds outperforming T-bills between 1998 and 2010. But, it’s worth noting that even the AIMA-methodology table shows T-bills outperforming hedge funds between 1998 and 2009.

After having replicated Lack’s results, AIMA then attempts to explain that there are “five fundamental flaws” needed in order to obtain them. But all those “flaws” seem like very sensible assumptions to me.

Firstly, AIMA takes much umbrage at the fact that Lack is looking at the returns to investors, rather than at internal hedge-fund returns. In the jargon, Lack’s figures are “dollar-weighted”, which means that he looks at what happens to actual dollars as they are actually invested in hedge funds. AIMA’s preferred figures are “time-weighted”, which is much less helpful. If a tiny unheard-of fund has a couple of spectacular years, attracts lots of money, and then sees only mediocre performance, then the dollar-weighted returns will be low while the time-weighted returns will be high.

AIMA’s argument is that it’s not up to the fund manager when external investors choose to invest their money in a fund, and that therefore if you’re measuring the performance of fund managers, you should use a time-weighted series rather than a dollar-weighted series. That’s reasonable. But Lack isn’t trying to measure the performance of fund managers, here. He has no interest in trying to work out how good John Paulson is, for instance. Instead, he’s attempting to measure the performance of hedge funds, in aggregate, and is trying to work out whether investing in hedge funds is a good idea. And for that purpose, a dollar-weighted series is absolutely the right one to use.

This is actually AIMA’s fourth beef as well: they don’t like the fact that Lack is using the HFRX Global Hedge Fund Index to measure hedge-fund returns, as opposed to other indices which show higher returns. But what they don’t say is that there’s a reason HFRX returns are lower than other indices — which is that HFRX is dollar-weighted and the other ones aren’t.

The next two flaws that AIMA complains about surround Lack’s idea of what exactly is meant by “profit”, from the point of view of a hedge-fund investor. Obviously, it doesn’t include fees paid to the fund manager. But according to Lack, the sensible measure of profit to use is “profits in excess of treasury bills, the riskless alternative”. After all, if you can take no risk and make more money than a hedge fund, it’s a trivially easy decision to do just that.

AIMA doesn’t like this:

This is mistaken because it replaces real dollars going to investors (hedge fund returns figures are post- fees) and replaces them with an arbitrary and flawed definition of what is “real”. For example, if an investor earned $10 from hedge funds when he could have earned $2 from Treasuries, the “real” money he received was still $10, not $8 ($10-$2).

I’m with Lack on this one: hedge funds should not be taking credit for risk-free returns that an investor could have gotten by investing in Treasury bills. The value added by hedge funds is excess return, and T-bills are a very good proxy for the risk-free rate.

AIMA also dislikes the way that Lack corrects for survivorship and backfill bias in published hedge-fund returns. These are huge, well-known problems: survivor bias is the way that funds stop reporting their returns — and eventually fizzle out entirely — when they perform badly, while backfill bias is the way that funds often only get included in indices once they have a string of decent returns, which then get added in to the index retroactively. Put the two together, and you find that hedge-fund indices are generally overstated by about three percentage points; it’s entirely reasonable to account for that when working out the actual returns to investors.

But here’s the thing: Lack only includes survivor and backfill bias in his calculations once in his book, in the table on page 64. All of the other tables and calculations exclude it, and they still show utterly pathetic returns to investors, compared to enormous profits for fund managers. Now to be fair to AIMA, those utterly pathetic returns are still positive. So for instance, here’s one of Lack’s main sum-it-all-up charts:

The main thing that you’re looking at here is the final line. If you look at the money that investors made by investing in hedge funds, it comes to $70 billion — a number substantially smaller than the $379 billion that the hedge funds managed to skim off in fees. Now the $70 billion is profit over and above the risk-free rate of return on Treasury bills. But it’s not the end of the story. Because funds-of-funds were very popular for most of this period, investors also paid some $61 billion to them. Which left them with the grand total of $9 billion in profits, compared to the $440 billion that the hedge-fund industry took in fees.

And even that overstates the amount of money that hedge-fund investors wound up with. There are lots of other fees, too, going to hedge fund consultants, family offices, and the like. On top of that, the fees in this table are actually understated, by some unknowable amount. I’ll let Lack explain, from his book:

Estimating fees on the industry as if it’s one enormous hedge fund does include one simplification, in that it excludes any netting of positive with negative results. To use a simple example, if an investor’s portfolio included two hedge funds whose results cancelled out (one manager was +10 percent while the other was −10 percent) the investor’s total return would be 0 percent and for our purposes here we’ll assume that no incentive fee was paid on the 0 percent return. However, in reality the profitable manager would still charge an incentive fee. It’s not possible with the available data to break down the returns to that level of detail, so the fee estimates derived are understated, in that it’s assumed incentive fees are charged only on the indus try’s aggregate profits, whereas in fact all the profitable managers would have charged incentive fees with no offset from the losing managers.

On top of that, Lack assumes that all hedge funds were below their high-water mark for the entire post-2008 period, and charged only management fees with no performance fees at all. It’s a very conservative assumption: of course some hedge funds were charging a performance fee as well as their management fees. So total fees are surely higher than Lack calculates here.

In any event, while AIMA tries to reverse-engineer Lack’s comparison of hedge-fund returns to T-bill returns by looking at his chapter on fees, that’s not actually what Lack was doing when he wrote his opening sentence. Instead, if you look at page 7 of his book, he explains that over the time period in question, T-bills returned 3% annualized, while hedge fund investments returned 2.1%. That calculation isn’t based on fees at all: it’s simply based on assets-under-management figures from BarclayHedge. It’s hardly surprising that there’s a very wide range of reported returns and assets for hedge funds: the industry is secretive and there’s no central clearinghouse for such figures. So no one can know for sure what total investor returns for the asset class have been. But there’s certainly no reason to believe that higher figures are more reliable than lower figures, as AIMA seems to do.

AIMA concentrates very much on Lack’s fee calculations, which doesn’t serve it well. Its complaints on that front are decidedly niggling: the worry, for example, about the fact that the fees that hedge funds charge should not be considered to be profits. But AIMA doesn’t actually look at the biggest costs of hedge funds — things like rent, or Bloomberg terminals — to see whether such things are taken out of headline returns, or whether they’re paid entirely by the fund manager. Maybe because they wouldn’t like the results.

AIMA also complains that average fees in the industry “are closer to 1.75% and 17.5%” than they are to 2-and-20. I’d take that complaint a bit more seriously if they weren’t at the same time trying to make the opposite point, that Lack’s survivor-bias and backfill-bias calculations should be offset by the fact that “some of the largest and most successful hedge funds choose not to report to databases for a variety of reasons”. Those large and successful hedge funds, of course, are the ones most likely to be charging more than 2-and-20. (Renaissance, for instance, charges 5-and-44.)

AIMA’s complaints aren’t only about fees. It also tries to say that Lack isn’t being realistic about how hedge funds are used in reality: they’re not designed to replace stocks and bonds, they say. Indeed, they say, “a hedge fund (a market-neutral fund, in particular) is not designed as a stand-alone investment but as a diversifier for an equity portfolio”. As such, the proper thing to look at is not hedge-fund returns versus various combinations of stocks and bonds, but rather various combinations of stocks and bonds, on the one hand, versus various combinations of stocks-and-bonds-and-hedge-funds, on the other. They then trot out familiar charts showing that hedge funds give you lovely diversification benefits, even if they underperform the stock and/or bond markets.

This argument is directly addressed on page 159 of Lack’s book. He points out that diversification benefits are generally calculated using the most generous hedge-fund index that can be found, using returns going back as far as possible — since hedge-fund returns in the 1990s, when the hedge fund industry was much smaller than it is today, were significantly higher than what we’ve seen of late. On top of that, the returns used are the returns of the average fund, not of the average investor. If small funds outperform large funds, on average — and they do — then average investor returns will be significantly lower than average fund returns. AIMA’s pretty chart showing a portfolio with hedge funds outperforming a portfolio without hedge funds — that’s not based on actual returns to actual investors, it’s all based on theoretical returns to theoretical investors.

Of course, some hedge-fund investors do manage to do quite well. That’s a statistical inevitability. But we’re not talking about the art of picking outperforming hedge funds, here, we’re talking about whether it makes sense, in general, to invest in hedge funds at all. Which is why AIMA’s epilogue is so annoying:

Consider the following quotes:

“Some of the most talented investors in history run hedge funds.”

“There are plenty of investors that are happy with their hedge fund investments.”

Those are not taken from an AIMA document, but from ‘The Hedge Fund Mirage’. The book, at times, is prepared to acknowledge that certain investors in hedge funds have done very well from their investments – a conclusion that is seemingly at odds with the claims made elsewhere in the book.

This is just silly. Of course there’s no conflict between saying that some hedge fund managers have done very well, and even that some hedge fund investors have done very well, but that in aggregate the managers have done much better than the investors, to the point at which the investors would have been better off not investing in hedge funds at all.

Indeed, if AIMA has been reduced to this rather pathetic game of “gotcha”, that says to me that there really is no robust refutation of Lack’s book. There are huge risks involved in investing in any hedge fund: they’re illiquid investments, and can blow up through bad luck or bad faith or old-fashioned bad investing at any time. In order to compensate for those risks, investors should be getting substantial excess returns. They’re not. So they should stay away.