One of my pet peeves is the way that insiders — whether corporate CEOs, hedge fund managers, or elected politicos — capture compensation (or credit) for normal cyclical gains they had little or nothing to do with.

This is the approach favored by the Crony Capitalists — those people pretending to be free market participants, and who merely pretend to be creating value. They are taking credit for structural successes that would have occurred with or without them. What they are actually doing is capturing value, not creating it — and then transferring it from its true owners (shareholders/investors) to themselves.

This is wrong; it is legalized theft.

If you want to see a good example of how CEOs transfer shareholder wealth to themselves, a good place to start is Roger Lowenstein’s 2004 book, Origins of the Crash: The Great Bubble and Its Undoing. The section on CEO compensation is astounding; these guys were essentially getting wildly overcompensated for being CEOs during a bull market. The prime example was the CEO of Heinz, who gave himself (with the tacit approval of his Board of Crony Directors) a $90 million bonus. And this was back in the early 1990s, when $90 million was real money.

Here is an idea for you corporate governance types: How about a compensation scheme based on genuine Alpha generation?

For corporate executives, this means their bonuses are based on something more than mere stock price irrespective of what the market and their sector is doing; I would suggest a combination of revenue gains (total sales) + total dollar profit growth (not a mere slashing of costs) + outperformance of stock relative to both the broad benchmark (S&P500/400/600 as appropriate) PLUS out-performance relative to their own sector.

In other words, stop paying excess bonuses for having the good fortune to be CEO during a bull market.

Which brings us to hedge funds. As we discussed over the weekend (A hedge fund for you and me? The best move is to take a pass), we discussed how much of the investing profits were captured by fund managers for themselves. It is a similar situation in that they are taking performance pay for Beta, not Alpha.

A better fee structure? Replace 2% + 20% current structure with a 1% + 33% of Alpha.

How would that work? Well, the 2% fee gets cut in half, for the simple reason that 2% fee on million dollars plus is excessive. But the real change is when it comes to the performance fee/bonus. That 20% of gains as of late has not been performance, its been only Beta. If their benchmark is up 20% and the manager is up 15%, there is precisely zero Alpha generated. So why should the manager get a bonus or a performance fee? They under-performed.

Instead, I propose a 33% of Alpha as a performance fee. The manager gets a bonus performance fee ONLY IF THEY CREATE EXCESS ALPHA OVER BETA — only on the percentage of gains over the benchmark.

Lets use a simple example: Two hedge fund managers run funds. Assume the market (their benchmark is the S&P500) is up 10%. Manager 1 (Fund ABC) is up 20%, while manager 2 (Fund XYZ) is up 10%. We will use a million dollar fund as a nice round number.

Fund ABC:  FUND +20%, SPX +10%

Fund ABC sees the manager handily out performing the market. The fee comparisons are below.

Example 1:

2 & 20:   $20,000 + $40,000
(Twenty thousand dollars is the two percent management fee; forty thousand is twenty percent of the two hundred thousand dollar gain. Half of which is Beta, half of which is truly Alpha.

1 & 33:  $10,000 + $33,000
(ten thousand dollars is the one percent management fee; thirty-three thousand is thirty percent of the one hundred thousand dollar Alpha gain.

In our example of the out-performing manager, 2&20 generates a 6% fee versus 4.3% fee for the 1&33 structure.

Example 2:
Fund XYZ: FUND +10%, SPX +20%

Fund XYZ sees the manager under perform the market. The fee comparison is:

2 & 20: $20,000 + $20,000
(Twenty thousand dollars is the two percent management fee; twenty thousand dollars is twenty percent of the hundred thousand dollar gain.

1 & 33: $10,000 + $0
(ten thousand dollars is the one percent management fee; zero dollars is thirty three percent of the gain above the markets.

In our example of the under-performing manager, 2&20 generates a 4% fee versus 1% fee for the 1&33 structure.

In both examples, managers are being wildly overpaid for Beta; in the other, they are receiving a bonus based in part on Alpha generation. Hence, their compensation is more aligned with the client’s.

Note that fund manager of ABC makes less under 1 +33 when they generate Alpha — $43k versus $60k. But look at the enormous savings that come from an underperforming manager: Instead of making $60k for partial Beta generation, he makes $10k — a quarter of the fee generation for partial Beta.

I don’t expect fund managers will rush out to embrace this model — unless the institutional players, trustees, and endowments demand it.

Category: Crony Capitalists, Hedge Funds, Investing, Really, really bad calls

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.

35 Responses to “Proposal for New Hedge Fund Fee Structure: 1% + 33% of Alpha”

  1. DaveWC says:

    In the first example it appears you’re using 33% of beta rather than alpha. It should be 33% of the gain over the markets which is $100k.

  2. constantnormal says:

    Does 33% of the alpha mean that when the account loses 30%, the fund manager kicks in a third of the losses (10%) as a rebate back to the customer? A poor-performance penalty, as it were …

    Alpha can be negative as well as positive, you know … and 33% (or 20%) of the gains plus 0% of the losses seems a bit one-sided …

    But if the fund manager is truly adding value, and not merely surfing the rising tide, this should not pose a problem …

    • Marc399 says:

      I agree, although I think a fixed fee schedule rather than a % of assets is better.

  3. NotAnEconomistButPlayOneOnTV says:

    Under the 1 & 33 in the first example, shouldn’t it be:

    $10,000 + $33,000

    Where $10,000 is 1% of $1,000,000 and $33,000 is 33% of ((20% -10%) x $1,000,000) where 20% is FUND gain and 10% is SPX gain

    So, FUND manager does worse under 1&33 in this scenario, hence, it is unlikely to change absent pension funds and other big money players demanding a different comp structure than that on offer.

  4. lmllet says:

    I understand where you are coming from but one of the reasons institutional investors invest in hedge funds is because mutual fund managers have become closet indexers and are obsessed by relative performance. They tend to be obsessed not to miss the next ‘melt-up’ but do not care that much about capital preservation. Declining by 40% when the market is down 50% is ‘fine’ as the manager did better than the market – or is it? For a benchmarked investment manager, ‘risk’ is defined as a deviation from the benchmark, irrespective of how the benchmark performs. For a hedge fund, risk should be defined as the probability of permanent capital loss.

    Hedge funds on the other hand focus on absolute performance; performance fees are ‘easy’ to gain in up market but the real value of hedge fund is (should be) to protect capital in down markets. Hence the high water marks for the payment of fees.
    Basing performance fees on relative performance would give the wrong incentive.

    • 1) If Absolute performance is their goal than they should get paid for that Alpha — NOT for Beta

      2) They may focus on Absolute performance, but they have stunk the joint up over 1, 2, 3, 5 and 10 years

      See this:

      The latest performance data (via the HFRX Global Hedge Fund Index) reveal that hedge funds haven’t fared well at all: They returned a mere 3.5 percent in 2012, while the S&P 500-stock index gained 16 percent. Over the past five years, and the hedge fund index lost 13.6 percent, while the indices added 8.6 percent. That’s as of the end of 2012; it has only gotten worse in 2013. Most hedge funds have fallen even further behind their benchmarks this year, gaining 5.4 percent vs. the market’s rally of 15.4 percent. As a source of comparison, the average mutual fund is up 14.8 percent

      -A hedge fund for you and me? The best move is to take a pass

      • lmllet says:

        my point is NOT to defend hedge funds’ performance – it is pretty clear that they have underdelivered. But with a high water mark a negative performance means no performance fees so investors don’t pay for beta. Hedge funds should be paid for overperforming cash over long period of time. It is a much better solution to pay performance fees quarterly based on trailing 5 year overperformance versus cash. But please do not re introduce indexing through the back door. This was the reason why hedge funds were introduced in the first place…………..

  5. FoF Guy says:

    I am a big fan of this website and your commentary.

    In this instance, though, I would suggest that you are ignoring the place risk should play in the equation.

    What if XYZ was market neutral? 10% would be heroic and they would certainly have earned the fee.

    It’s unfortunate that most commentary about hedge funds judges performance against the stock market and does not incorporate any discussion of risk. Let’s bring the concept of risk-adjusted returns into the discussion and I think we will better advance our thinking on the subject.

  6. perpetual_neophyte says:

    Barry – I’m off to a slow start after a long weekend, but doesn’t this article rely on some assumptions about beta that aren’t spelled out? In your second example (fund +10%, benchmark +20%), if the fund’s beta was 0.4 they would still have generated positive alpha.

  7. rs55 says:

    I have a more fundamental proposal: Why does the HF industry get to charge fees as a Percent of assets? Do corporate CEOs get paid as a Percent of their assets? Do Presidents of countries get paid as a Percent of their assets? Should surgeons charge fees as a percent of the patients assets?
    Its just a convenient convention – and explains the outlandish compensation. Of course its outlandish – Obama’s compensation would also be outrageous if he was paid as a percent of assets ( US assets roughly $100 Trillion).
    Large hedge funds should pay their ceos a salary and a performance bonus – like every other profession. Yes , then you can fine tune exactly what ‘performance” means. But of course the large slow moving herbivores who constitute the client base ( pension funds etc) are run by relatively underpowered corporate types who lack the self confidence or imagination to demand radical change.

  8. bfarzin says:

    Selection of the best benchmark may be challenging for some (many?) funds. What about a baseline return? Rather than SPX, look at some pre-defined level (say 5%?) and charge a fee if and only if that level is exceeded. So in down years, no money (even if the fund “out performed” some index) and in up years it requires exceeding that level. This could push the manager to be even more aligned with the investor.

    No question that 2&20 is a designed to make fund managers very wealthy at the expense of the investor community.

  9. streeteye says:

    There are some funds that are fixed income, market neutral, work globally but in a changing mix of countries, multi-asset class. Picking an index is not always easy.

    There’s a leverage issue / degree of market neutrality. It’s legit to try to make 100% of the S&P with 2/3 of the volatility. Should that guy’s index be 2/3 of S&P performance? Then it’s easy to cheat going high beta or higher exposure/leverage. Do you measure volatility ex post and derive alpha from that? More fair, but doesn’t work for private equity, thinly traded stocks where the marks aren’t meaningful/can be gamed.

    What happens to losses? If the index is down 30% and the fund is flat, does the manager get a 10% fee?

    Measuring alpha is not easy in the short run. In the long run, it’s a free market, if managers don’t deliver, one would think fees should self-correct.

    The 2/20 only makes sense for superstars. After HFs did well in the dot-com crash, demand for superstars created its own ‘supply’ which depressed returns since a lot of new entrants weren’t superstars.

  10. BenE says:

    Amen. The exact numbers can be negotiated but they should definitely be paid for alpha not beta. This criterion would go a long way into weeding out scammy hedge funds. Paying for beta is stupid since you can get it for almost free in a Vanguard etf.

  11. tgdc says:

    As a practical matter, this may be fine for highly traded equities but for things like credit default swaps how does one measure the “true” beta? The reality is that for many non-linear assets like this there is not just one beta… they can have one correlation with the market during “normal” times and a very different relationship to the market during “crisis” times. But the latter is the only time one really cares about their value while the former is the only data available for computing a beta.

  12. James Cameron says:

    > I have a more fundamental proposal: Why does the HF industry get to charge fees as a Percent of assets?

    Because people will pay it . . . and it would be well worth it if you were fortunate enough to have your money with one of the top funds, which there aren’t many of . . . the question is why, in the face of the performance data presented here and elsewhere, would someone pay these fees when their accounts are essentially being looted?

  13. beefcreek says:


    Interesting note and I agree. I think we should redo all kinds of metrics in the new normal economy. Another take on this is politicians taking credit where it is not due. As an example, current president is pushing that he is one of only five presidents to ever double the S&P 500 on his watch

    Or, good old Rick down in Texas claiming he rebuilt the economy there!

    How could you track those metrics…..

  14. VennData says:

    A fuel and his money are soon partied… with.

    If people want to have fun with hedge funds, let ‘em.

    If pension funds and endowment managers want to “invest” in hedge funds to justify THEIR salaries, they should prove they added value, which they can’t, so they don’t. Unless the board members of these funds are stupid, which is my theory, there must be an entire substrata of older folks who are on pension boards that have no idea that the fund they steer is doing crap, because they are financial idiots.

    Pension fund manager re-training should be Boehner’s top job re-training initiative: “Wash or would you like the wax too?”

    • You need to follow this logic a bit further and keep asking “qui bono?”. The pension funds that remain have their own “agency cost” issues… for the sake of simplicity let’s borrow the conservative dogma and make the language more colorful by replacing “pension beneficiaries” with “union workers”… then the trail of tears goes like this:

      (1) Public-sector union employees have entire life savings in pension system.
      (2) Pension fund’s “professional managers”, chasing yield, invest in exotic places including hedge funds.
      (3) “Winning” (surviving) Hedge funds siphon off 80-90% of positive performance via 2-and-20. Failing hedge funds liquidate, leave losses with pensions, and do not show up in statistics (?)
      (4) Pensions underperform benchmarks, and must plead with government for increased contributions to fund.
      (5) Taxpayers and/or union workers are forced to pony up to fund pension.
      (6) ?? The pension fund’s Professional Managers move on to work for Hedge funds and other big-finance scam institutions ??
      (7) ?? The government officials either move on to work for hedge funds and other big-finance scam institutions, or receive “campaign contributions” aka bribes from same…

      Winners: Hedge funds, pension managers, government officials on the take. The 1%
      Losers: Taxpayers, union workers and other pension beneficiaries, and all other recipients of government spending who are squeezed in budget battles driven by pension issues. The 99%

  15. [...] Time for a new way to pay hedge fund managers.  (Big Picture) [...]

  16. carchamp1 says:

    Barry, It’s not hard to figure out why I never see you on CNBC.

  17. Iamthe50percent says:

    I hope this isn’t a silly remark, but how about eliminating the fixed fee and splitting the alpha 50-50. I’m paying you to beat the market. If you don’t, you don’t get paid. If you do, we split the gains due to your performance and my capital 50-50? We might go even further and split the losses.

  18. JC in Va says:


    TFS Capital tried this and had some issues with execution with their Small Cap Mutual Fund and they had to get rid of it.

    Chuck Jaffe wrote an article about the challenges with execution and why they stopped.

    Here is some context:

    and here is Chuck’s article from 2010 and an excerpt…

    “The problem that TFS encountered, however, was that the fee structure lagged performance. Think of it this way: if you had a great 12 months last year, you get paid more this year, when results could be below-average. Thus, you are paying expenses based on past success or failure, unlike the hedge fund world where costs ride on current performance.

    Thus, TFS Small Cap is currently charging the max, 3%, because it gained 65.4% in 2009, compared to just 27.2% for the Russell 2000. While the fund was racking up that performance, however, it actually was charging the minimum expense rate, because it lost 38.4% in 2008 to lag the Russell by roughly two points.

    Eiben noted that the lag is a big part of the problem, because financial advisers wanted predictability and had a hard time explaining the structure and the timing disconnect to their clients. As the fund cancels the performance fee, it will set expenses at 1.75%; that’s above-average for equity funds, but easy to explain to the buying public.

    “It’s disappointing to us, because we think everyone should be able to understand fees that are based on results,” Eiben said. “But if people don’t understand the structure — and the rules don’t allow us to do something more simple and elegant — and they don’t buy the fund, then we have to go to a structure they will understand.”

    That’s a loss for fund investors, because it makes it less likely that other fund firms will take the performance-fee route in the future. Without a demand for change in fee structures, investors will be stuck with the payment setup they have now and fund managers will continue to bring in big paychecks, even when their performance is worthy of something less.”

  19. Ben says:

    Barry, the reason most hedge funds are underperforming the S&P is because they are running relatively low net exposure and have agreed to do so for their clients. So the problem is you are measuring them against the wrong yardstick. If you want to calculate their “alpha” you would have to calculate their net exposure and measure against that. Now whether this is a good long-term strategy is another question. But I think your take is unfair and not well thought out.

  20. “Crony Capitalists”

    the CatTag, by itself, is worth 4 points (this being Baseball Season, We’ll call it a ‘Grand Slam’)

    btw, BR, what’s up with that FusionIQ Active ETF ??

    you know, there is, still, a Large %, of the Market, that likes to ‘buy’em by the Box’..

  21. henry.bee says:

    Barry, this is well intended but you may be unintentionally advocating for relative performance, instead of absolute performance, which I believe is what you want the incentives to be based on.

    This is because the alpha/beta in a linear regression is only meaningful when the r-squared is high, otherwise alpha and beta are meaningless. And if you’re charging based on alpha then you’re implicitly saying that you want r-squared to be high (in order for the alpha to be meaningful). Your proposed incentive will end up rewarding the funds that have a low tracking error and a high information ratio the most, which are funds with the best relative performance.

    Consider a good absolute return fund that exhibits universally low correlation (low r-squared) to the publicly available benchmarks. This is what institutional investors want. And if you run a regression of such absolute return fund against any benchmark it’ll flat line, with alpha/beta/r-squared close to zero. This manager, having generated a highly desirable absolute return will be paid no incentive based on your proposal.

    Paying for alpha ends up rewarding relative performance. Exactly the opposite of your intention.

  22. No payment for pairing with the market, but a fat bonus for beating it. It’s an everything or nothing deal and it can encourage some risk taking that is not desired by investors perhaps.
    This was just my first idea, correct me if I’m wrong.

  23. leel says:

    as a hedge fund employee I AGREE with your concept of restructuring fees, BUT the execution is tricky.

    Are you investors paying for low-correlation, low volatility, or risk adjusted rewards?

    Finding a ‘perfect’ incentive is tough, since managers (and all people) focus too much on the rewards (and game the system), rather than the service provided. My scenarios below are not meant to be argumentative nor oppose your idea, but genuine considerations my colleagues & I have debated for 10+ years.
    If I return 20% in 1 year and the index returns 20% -my alpha is zero; BUT what if my drawdown is no greater than say 5%, whereas the index massively gyrates? Aren’t I adding value?

    What about in years, like 2008, where I’m -5% and the index is -40%…is an investor going to pay me 33% of 35%? Hence realizing significant losses? But the following year I lag the benchmark (+15% vs +28%)…so eventhough over 2 years I avoided a massive drawdown & made money over 2 years I don’t receive any incentive? Where as their mutual funds charged the a 5% commission to buy into the mutual fund, then charge 1+% in fees, then +additional bps in some jargony term which the consumer doesn’t realize is a fee paying for the mutual fund’s marketing & sales expenses?

    I’m not saying 2 & 20 is right, BUT there’s too much focus on hedge fund fees, not enough on excessive/disguised mutual fund fees.

    Excessive focus on fees ‘pushes’ investors into cheap ETFs…I met an investor in 2009 who told me “I don’t pay hedge fund fees, I only use ETFs…I don’t pay more than 35 bps” He was only down 40+% that year…So the investor saved himself 165 bps but cost himself somewhere between 2000 & 3500 bps….sounds like ‘pennywise, pound foolish to me’
    btw that investor was liquidating everything in his house as he proudly told me he sticks to low fee ETFs.

    The brokerage firms have told people they can invest on their own, but can also use a broker…individuals are overconfident of their skills. Their brokers are trained in sales, not risk management. Pick your poison: the arsenic or the arsenic???

    • carchamp1 says:

      It is not the “focus” on fees that is the problem. The problem is that managers do not provide enough value to justify the fees. The evidence of this is simply overwhelming.

      That said, you’re making thoughtful points. I agree that execution is a problem. I think the biggest one is the timeframe. One year is simply too short a time to judge the effectiveness of a manager. At a minimum, five years is needed. But, of course, a manager can’t wait five years to be paid. To me, using Barry’s 1 and 33 structure is on the money. Perhaps basing the alpha piece on a longer timeframe, like five years, would be better.

  24. TacomaHighlands says:

    Dang! Loved this!

  25. mbollinger says:

    Barry and others, I was looking for a discussion on hedge fund fees and I came across your article and think you are on the right track. I am putting together a fund myself and want to structure it so that I feel I am getting paid based upon the alpha generated. There are several complications that I believe need to be addressed in the structure you proposed and I believe the structure I am proposing addresses these. I would like to hear what you or other commentators think of the proposed fee structure.
    A complication I see in your proposed structure is how it handles sequential periods. For instance, if during one period the fund underperforms the benchmark, does this under-perforformance roll through to the next period so that in order to receive the incentive fee in the next period, the manager must dig himself out of the negative alpha generated from the prior period? I believe this needs to be part of a compensation scheme or else one could create a fund with huge beta so when the market is up the fund is way up and they get a large incentive compensation but when the market is down, they get no incentive compensation. In the long term there could be no alpha generated but the manager would have received a large fee due to the volatility of the fund.
    Another complication I see is that every period investors will be adding more money and redeeming money. How do you factor this into the compensation scheme?
    What I propose I call a rolling alpha. Every period the return on assets is measured and is compared to the return that would have been achieved if the same amount of assets had been invested in the benchmark. Either the fund will outperform or underperform the index. If the fund outperforms, then the incentive compensation fee will be applied to this outperformance. If the fund underperforms, this underperformance is calculated and rolled through to the next period. If in the next period the fund outperforms, the underperformance from the prior period is subtracted from the current period outperformance. This amount will be the rolling performance and if it is positive, then the incentive performance fee will be applied to this amount. If this amount is still negative, then no incentive fee is applied and the performance deficit will be rolled to the next period.
    I think this scheme aligns the investment manager very well with the investor as the incentive fee is only ever paid when the sum alpha is above.
    This may be a bit complicated so I think an example would be helpful.
    Assume that the incentive fee is 20%
    If a manager starts with $10,000 and in period 1, his fund returns 20% compared to 10% for a benchmark, the 20% fee will be applied to the $1,000 outperformance of his fund which will be $200. So the fund starts the second period with $11,800 and if the fund returns 10% compared to the benchmark that returns 20%, his fund will have underperformed by 10% or $1,180 and his fund will have $12,980 at the end of the period. The next period his fund performs at 20% compared to the benchmark of 0% or a 20% outperformance of $2,596. The prior period negative alpha is subtracted from the current period alpha ($2,596-$1,180=$1,416) and this is the amount that the incentive compensation fee of 20% is applied to. Therefore the incentive fee is $283.20.