Click to enlarge
Chart
Source: Moneybeat

 

I have been in the midst of a big research project that has led to me looking askance at the claims and long term returns of hedge funds.

It began with the research I did for Romancing Alpha, Forsaking Beta, and has led to other interesting places. But as we have learned, some things are not as they appear.

The chart above is a case in point

This is one of those return charts that looks impressive when you first see it, but once you delve deeper you learnt hat it is actively misleading.

It is a time weighted return series, and as such, shows the annual returns of managers regardless of assets under management (AUM).A more honest display would be an asset weighted return series.

Here is why: Imagine a manager who is up 10% every year for 9 years, then down 30% in year 10.  You might think he is creating a lot of Alpha and net net has created a lot of wealth.

But having done some research, I now know quite a few things I did not last year:

1. The Hedge fund industry has swollen, from barely $100B to well over $2T from 1997 to 1012;

2. Hedge fund managers ability to create Alpha typically is inverse to their size;

3. The distribution of Alpha is nonGaussian (not a smooth bell curve)

4.  In 2008, Hedge funds lost all of the profits they had previously made (and then some). Add in 2&20% and hedgies are a poor bet for most investors.

(Note: We haven’t even touched upon all of the errors that are on the Hedge fund index — survivorship bias, backfill problem, self reporting issues, etc.).

Which brings us back to our theoretical manager 10% for years with a 30% loss in year 10: Imagine he is a John Paulson like manager, who did really well until he hit the wall.

His best years were when the firm was small — up 10% when its a $50-$150 million fund. His gains are low millions, even 10s of millions. Then after a few big rounds of publicity, the funds he manages swell to several billion dollars. Down 30% wipes out all of the gains of the prior decade — and then some — in asset weighted series.

But in a time-wighted series, he looks pretty good.  Its an epic fail, a colossally misleading returns .

Which is of course why so many managers love them!

 

 

Source:
Europe Charting Hedge Funds’ Long Term Gains
Harriet Agnew
Money Beat June 12, 2013  
http://blogs.wsj.com/moneybeat/2013/06/12/charting-hedge-funds-long-term-gains/

 

Category: Data Analysis, Hedge Funds, Investing

Please use the comments to demonstrate your own ignorance, unfamiliarity with empirical data and lack of respect for scientific knowledge. Be sure to create straw men and argue against things I have neither said nor implied. If you could repeat previously discredited memes or steer the conversation into irrelevant, off topic discussions, it would be appreciated. Lastly, kindly forgo all civility in your discourse . . . you are, after all, anonymous.

18 Responses to “Bad Chart: Hedge Funds’ Long Term Gains”

  1. BenE says:

    That’s great insight. I was already a pessimist on hedge funds but I hadn’t thought of the bias from asset weighting vs time-weighting.

  2. NoKidding says:

    HFRI and S&P500 are hard to distinguish. S&P was about 700 in 1996 and 1600 today, about 2.3x so it must be the top one. In case other like me are slow thinkers.

  3. farmera1 says:

    You could of had a piece of the action when Buffett bet a cool million with “a new york hedge fund” that over a ten year period an index fund of the S and P 500 would beat the hedge fund. Turns out Buffett chose Vanguard S and P 500 index fund and the dudes from Protege chose to average the return of five hedge funds. Buffet is winning, you know expenses and all that. Big hill to climb when the smart boys usually take a cool, 20% and 2%. We do know for sure who is getting rich for poor returns, it’s those smart guys running the hedge funds.

    Five years into the ten year bet, Buffett is well ahead.

    http://www.businessinsider.com/buffett-still-winning-hedge-fund-bet-2013-1
    http://www.cnbc.com/id/100405142

    I’m thinking of going to Vegas and starting a business called hedged gambling. Big “investors” come in and give me their money. I place the bets on the roulette table, I keep 20% of all the profits and charge a minimal 2% administration fee. I think there is a market for my new idea.

  4. rbtrage says:

    It’s strange that dividends are not included for the S&P return; 92.54% is just the price return without yield.
    I did a quick ROR check on Bloomberg. With Dividends, S&P total return was +157.06.
    Makes me wonder if income was left out of the other indices too and why?

  5. constantnormal says:

    A great observation, backed (as usual) by actual data … and even better, I see no reason why it should not apply to mutual funds and ETFs as well. We see split-adjusted returns on individual stocks, which compensates for the changes in numbers of shares outstanding, but is there anyone who actually tracks funds (of all types) using an asset-weighted metric?

    One would think that Morningstar or some similar ranking entity would leap at the chance to do so … Forbes, Fortune, or Money would also seem to be likely candidates to adopt this metric …

    BR, have you patented this? Don’t tell me that it can’t be done, there are too many counter-examples of “unpatentable” things receiving patents … at least slap a name on it and brand it …

  6. Just to be fair, if you asset weight the hedgies, you need to asset weight the stocks, bonds, etc. How would it change the results if we were to asset-weight the S&P 500 returns as well? Ditto for the bond market, etc.

    Particularly for the bond market – we know that total credit outstanding hits its highs when yields are low, because that’s WHY the yields are low. Too much credit is out there chasing too little borrowing.

    Also we already know the answer for the bond market – when yields are near zero, forward returns on bonds mathematically cannot be high.

    So I think asset-weighting the returns will change everything for a lot of those curves. A lot of “past returns” will not be predictive of “future performance” once asset levels have been inflated.

  7. Manofsteel11 says:

    If one moves beyond index-focused observations about the industry averages, then a deeper analysis is due.
    Some investors move in and out of HF based on size and performance, with many refraining from going into large ‘multi-strategy’ black boxes altogether.
    Some own equity as active partners in the funds and benefit from research/insights while also harvesting fees.
    Some have friends who offer them managed accounts with greater control, better terms and closer oversights.
    Some actually know niche spaces where few parties operate and where alpha still exists.
    Some use HF as diversification to huge existing exposures, knowing full well that it would cost them more to try and create similar diversification with ‘cheaper talent’.
    Some know specific markets better than any non-specializing investor with limited infrastructure could ever know.
    I could go on, but I think you get the point. Like any other investment vehicle, HF have advantages and disadvantages, and results often depend on the investor rather than on industry statistics.

  8. 873450 says:

    It’s all about whose pockets the gains land in.

    And now for just $1,0000 anyone is an eligible victim.

  9. SilentK says:

    I understand where you are coming from. But the same can be said about the Vanguard S&P 500 index fund. I’m sure the assets in 2008 right before the crash were a lot bigger than in 1996. So why invest in index funds at all if they are going to lose more money in aggregate in one year than the previous 10? You can make this argument for virtually any asset class.

    You need to think of it in terms of a single investor putting one dollar in the fund at the beginning of the return period. What does he care if the fund grows and then loses more money in aggregate in one year than the previous years? All he cares about is his one dollar. So if his one dollar compounds at 10% for 10 years then loses 30% in a year he is still up. Its the suckers who came in late which are in the hole.

  10. mt3209 says:

    I like this analysis. It reminds me of the current debt bubble and the 30+ year decline in interest rates. Small moves up today wipe out more money then was created all along the way.

  11. Jaimin says:

    Quite Good work. And I thought HF only started to under perform after 2008 crisis. You think it gives different results if they take 2002-2007 and thats where most money started to go with after having established 5 year track record etc..

    In terms of Asset weighted returns, I remember reading in In one of the books by Jack Schwager that evaluate the money manager on how much absolute $$ he has made over period – most managers do well around $50-250 mn and as they move up it then becomes very difficult to provide great returns (without additional risk)

  12. Zephos says:

    Thankyou Barry. Sums up the HF industry these days, too many fish in the pond.

  13. [...] Dumb hedge fund performance chart using time-weighted return series when clearly this is not how it works in investing.  (TBP) [...]

  14. Livermore Shimervore says:

    “So why invest in index funds at all if they are going to lose more money in aggregate in one year than the previous 10? You can make this argument for virtually any asset class.”

    Most hedge fund investors who buy and hold thru the downturns will come up negative and pay high fees.
    Most S&P 500 index investors who buy and hold thru the downturns will come up positive and pay low fees.

    Yet inexplicably wealthy money continues to chase the former. I guess passive investments are not chic enough. Losing money is the new black.

  15. thedoubledown says:

    Barry,

    There is a hedge fund index which is asset-weighted it is the DowJones Credit Suisse Index http://www.hedgeindex.com which is the old Credit Suisse Tremont HF Index started in 1994. For the 1997-2012 period it is up 241% and the S&P 500 total return is up 157%. It is the most reliable HF index but does have a lot of biases that you mention.

    David