The graph above shows the portion of hedge fund profits that go to the hedge funds, the fund-of-funds, and the investors, using total dollars from 1998-2010. This is from Simon Lack's presentation, which is derived from his book Hedge Fund Mirage. Notice 97% go to insiders, 3% to investors.
Like most strategies or asset classes, the money made on a percent basis is large because it basically grows until it can't work, at which point it loses more money than it made and then some more (see above graph, 2008 killed everything). It could be thought of as an endogenous reaction, because how else will you know at what point it stops working? Perhaps this is the essence of all financial cycles.
One good point by Lack was that fund managers inevitably argue they do not fit into clean categories; that they, unlike everyone else, is neither fish nor fowl. I suppose we all think we are unique, and don't like to be benchmarked, but on the other hand, everyone is riskier without a benchmark.
Aaron Brown said...These numbers are deeply flawed, as I pointed out a couple of months ago here
and in my Amazon review of the book
where you can read Mr. Lack's response.
I don't claim that hedge fund investors made a higher proportion of profits than Lack claims, just that neither Lack nor anyone else knows the answer.
Since few hedge funds are designed as stand-alone investments, the relevant question isn't how hedge fund profits are shared but, "Do investors who include hedge funds in diversified portfolios do better than those that do not?"
Here the evidence from large institutions is pretty clear that hedge funds help. On the other hand, some or possibly all of that advantage comes from skill in manager selection and ability to negotiate superior terms.
There's no evidence I know of that says an investor without selection skill or negotiating leverage is helped by hedge funds. It might be true, but my efficient markets bias leads me to suspect that investor is better off in low-cost, tax-efficient index funds.
I know Hedge Fund numbers are hard to get, so I appreciate the more sources sources in your Minyanville piece.
I read your piece, and you note Dichev and Yu find time weighted and dollar weighted HF returns are about the same, for the same fund, I presume? I see his SSRN paper shows that across all funds, he estimates 6.1 vs. 2.9% annualized returns, time vs. dollar weighting, which is large.
He then estimates another test with 5 and 10 year track records.
Then he tests portfolio-level dollar-weighted returns, and gets a 12.6 vs. 6.0 percent numbers (difference of 6.6%!).
So, I don't see where you get your inference that Dichev&Yu find dollar-weighted only lags time weighted returns when using an average across all firms (equal weighted).
But I agree with your general thought that there's nothing wrong with the seller making more than the buyer: that's true for Bill Gates, Henry Ford, John Bogle. That's a scale story, pretty boring, nothing scary.
You do hint at something I find very intriguing that I can't really formalize, which is, any passive investing of any kind should return a zero premium to a bank account, because there are too many incentives for all sorts of people to take a cut, and such an investor is too lazy or ignorant to discipline these suppliers.
If you want to preserve even more cash, keep in mind that dining places create most of their cash on the liquids, not the meals.
how can i make money
"I suppose we all think we are unique, and don't like to be benchmarked, but on the other hand, everyone is riskier without a benchmark."
I am really not sure that I agree with this, unless your benchmark is L+10% net.
If anything I think that benchmarking a strategy that should be idiosyncratic leads to problems. For example the most enthusiastic benchmarkers were cb arb funds in 2005/6.
On "leaving money in cash" or passive investing getting zero returns...I keep getting reminded of a story a German friend of mine who told me about how his grandfather went bankrupt during the great inflation days. He had made a broadly right call on the economy that the policies they followed would be disastrous, so sold his businesses and sat on cash!! So went bankrupt due to inflation. As Aaron points out in his book, you DO need to take a risk. "The biggest risk is life is not living". I am not justifying reckless investing, but pointing out that sometimes cash can be the worst investment...particularly at zero rates Need to be flexible enough to stay invested - provided you can be long or short depending on the environment.
The specific numbers may be hard to come buy but I find the very notion of performance fees to be just wrong.
You earn an asset management fee and often have your own capital (or should) in the fund you are managing. You do your job and get paid well to do it in the form of the asset management fee. Your "extra incentive" is that if you do well you can attract more assets.
The only time I'd say a performance fee makes sense is if the potential size of the investment is very very small so that your access to it is essentially worthy of a fee and that it's a sure bet. 99% of hedge funds don't fit this description and should simply not levy a performance fee on top of what they get paid to invest the money.
Not to pile on but it seems to me that the data for this would be so flawed as to not be useful. I also think the time-period being used here dominates the point being made more than the asset class specifics do.
There are all sorts of problems in charging 2/20 for lackluster performance (relative, absolute, however you want to measure it). But those issues are no greater than charging 50bps flat for even more lackluster performance. A long-only manager isn't accepting any greater downside than a HF manager if their returns are negative over a holding period.
I tend to default to the feeling that all of these "studies" really just point out that the average hedge fund isn't particularly good. Isn't that true about just about every asset class when viewed in isolation?
In the end I'm really just skeptical of the data though. I could construct a pretty stark counter-example to the above with high quality data... but it would be limited to the 100-200ish HFs I have exhaustive data for, which is hardly itself representative.
Post a Comment