Practical analysis for investment professionals
06 June 2016

Dumb Alpha: How to Build an Above Average Hedge Fund

I can’t make up my mind about hedge funds.

On the one hand, the fees strike me as excessive. Call me Scrooge if you like, but I do not want to pay anyone 2 plus 20 for the privilege of losing my money. And, as Simon Lack, CFA, pointed out in his book The Hedge Fund Mirage: The Illusion of Big Money and Why It’s Too Good to Be True, there is a lot of evidence that, after fees, the average hedge fund underperforms a traditional balanced stock/bond portfolio in up as well as down markets.

No wonder that in recent months, we have seen a barrage of critical news reports taking aim at hedge funds — even from renowned hedge fund managers like Steve Cohen. More and more institutional investors have started to pull their money out of the asset class after experiencing disappointing results.

On the other hand, whenever I listen to my colleague Bill Fung at Maple Financial Group, I start to believe in hedge funds again because there seem to be fund characteristics that allow you to select the better performing funds ex ante. Hedge funds, then, appear to be similar to all other forms of active management. If you can identify superior managers, you have a chance to reap great rewards in the form of superior performance or increased downside protection.

But if you invest in the “average” hedge fund, you pay a lot of fees for underperforming a traditional portfolio.

Subscribe Button

Aiming for the Average Hedge Fund

Of course, no pension fund or wealthy individual would ever admit that they want to invest in the “average” hedge fund. Everyone tries to find the few superior managers, even though as a group they inevitably end up with the aggregate market.

So put yourself in my shoes. I would be very happy owning a hedge fund with superior performance after fees, but I don’t think I have the expertise to select the best managers. Also, I am so stingy that it hurts me to pay fees in excess of 2% even if the fund manager is a genius. If only I could invest in the average hedge fund, but at much lower fees. With what I saved in fees, I would create superior hedge fund returns.

Selling Puts to Replicate the Average Hedge Fund

Luckily, Jakub W. JurekErik Stafford, and the good people at GMO have provided a dumb alpha idea on how to do exactly that. Jurek and Stafford have demonstrated that it is possible to replicate the typical hedge fund index performance by selling out-of-the-money puts and using a little bit of leverage. Ben Inker at GMO has gone one step further, and shown that one can just as well sell at-the-money puts and use no leverage.

The following chart illustrates my take on these ideas.


Dumb Alpha Hedge Fund Chart


The purple line shows the performance of the HFRI Weighted Composite Hedge Fund Index since late 1995. The green line was constructed by following a simple investment strategy. At the beginning of each month, I sold a one-month put option on the S&P 500 index that was about 1% out of the money. The exact rule varies a little bit, because I set the strike one tenth of the monthly volatility of the S&P 500. This way, the put option will be a little further out of the money if volatility is high, and a little bit closer to the current index reading if volatility is low. The money I would normally invest in hedge funds I simply put into a one-month Treasury bill.

After one month, this strategy has earned the interest on the Treasury bill and the premium on the sold put option. Of course, if markets go down, I will have to pay money on the put option that expired in the money, which reduces the return of the strategy. This simple procedure is repeated each month.

Tile for Geo-Economics

The Result? An Above Average Hedge Fund

The performance without any costs or fees is shown in the green line in the chart. The green line looks very similar to the purple line, especially at first, so it seems like this systematic put selling is able to qualitatively replicate the performance of hedge funds. In practice, one has to incur transaction costs, etc., but currently these costs tend to be very low. The fun fact is shown in the blue line. If one follows the simple put selling strategy and deducts 2% fees per year, one ends up pretty much exactly as the overall hedge fund index. Or, to put it another way, if one can systematically sell put options each month and invest money in Treasury bills for an annual cost of less than 2%, one can replicate the average hedge fund at much lower fees. The result is an investment strategy that performs better than the average hedge fund net of fees, and with lower fees overall.

It gets even better.

The similarities between the put selling strategy and the performance of the average hedge fund diverged in 2011. Since then, the average hedge fund has performed even worse than the put selling strategy after 2% costs. This is the very best dumb alpha can offer: a simple, low-cost investment method that outperforms more sophisticated and expensive strategies.

Sometimes one can have the cake and eat it too.

For more from Joachim Klement, CFA, don’t miss Risk Profiling and Tolerance: Insights for the Private Wealth Manager, from the CFA Institute Research Foundation, and sign up for his regular commentary at Klement on Investing.

If you liked this post, don’t forget to subscribe to the Enterprising Investor.


All posts are the opinion of the author. As such, they should not be construed as investment advice, nor do the opinions expressed necessarily reflect the views of CFA Institute or the author’s employer.

Image credit: ©iStockphoto.com/CSA-Archive


Professional Learning for CFA Institute Members

CFA Institute members are empowered to self-determine and self-report professional learning (PL) credits earned, including content on Enterprising Investor. Members can record credits easily using their online PL tracker.

About the Author(s)
Joachim Klement, CFA

Joachim Klement, CFA, offers regular commentary at Klement on Investing. Previously, he was CIO at Wellershoff & Partners Ltd., and before that, head of the UBS Wealth Management Strategic Research team and head of equity strategy for UBS Wealth Management. Klement studied mathematics and physics at the Swiss Federal Institute of Technology (ETH), Zurich, Switzerland, and Madrid, Spain, and graduated with a master’s degree in mathematics. In addition, he holds a master’s degree in economics and finance.

20 thoughts on “Dumb Alpha: How to Build an Above Average Hedge Fund”

  1. Martin Landry, CFA says:

    Sounds like you have a new hedge fund on your hands….

    1. Joachim Klement says:

      🙂 Do you have the seed money?

  2. Umed Saidov, CFA says:

    Joachim,

    Thanks for sharing your insight, which is quite interesting. I had a couple of questions:

    1. In the article you say: “I set the strike one tenth of the monthly volatility of the S&P 500.” Do I understand correctly that this rule maximizes returns, meaning changing the strike either way would lower returns?

    2. Do you think investing the put premiums into higher yielding debt (albeit more volatile) instruments could make sense, especially if we extend the trading period from one month to a quarter?

    3. Lastly, what was the source of data that you used to backtest your strategy?

    1. Joachim Klenent says:

      Hi Umed

      The level of the strike was not set to maximize returns but to match the performance of the HFRI index over time as closely as possible. In fact if you play around with strike levels and leverage you can easily create a strategy that performs better in hindsight. The question remains, however, if this is then a strategy that will perform better in the future or if it is just data mining.

      By the way I used Datastream data for all my backtests.

      In terms of “safe” securities you could theoretically go for corporate bonds but there are two reasons why I would not do that:
      1. The liquidity of corporate bonds is much lower than the liquidity of T-bills. Especially in a credit crunch like 2008 you risk being stuck with your bond positions when you need to buy them or liquidate them.
      2. Corporate bonds have a positive correlation with stocks, so when stocks go down credit spreads widen and you lose on your bonds as well. So they provide less protection than T-bills just when you need it most.
      All in all, the return is made with the pit premiums and not with the bond side and trying to increase returns on the bond side willintroduce risks that are correlated to stock market risks and defies the purpose of the safe asset in this case.
      Best
      Joachim

      1. Umed Saidov, CFA says:

        Joachim,

        Thank you for your detailed response. It clarifies a lot.

        One last question: I wonder how you came up with the rule: 1/10 x of monthly volatility of S&P 500.

        Was there any other reason to set the strike price at the level you did apart from matching ” the performance of the HFRI index over time as closely as possible. ” ? If not, some might consider that data mining.

        I wonder what the results could look like if we decompose the strike-setting rule further into something akin to:

        Put Strike = price level [x] with [35%] probability of being crossed within a month. Where [35%] is dictated by the investors’ risk/reward mix.

        1. Joachim Klement says:

          I agree with you. It is data mining to set the strike in such a way to come as close as possible to the HFRI.
          In practice a strike setting rule like the one you suggest would be more appropriate.

  3. I think one way of identifying above average asset managers is by finding the ones that specifically use CFS – Computer science, Finance and Statistics because I have found that these managers’ models are sufficiently robust and cross-validated at a sophisticated level to improve the odds for success. I can think of three of such managers – but I will only mention one. 🙂

    PCPpresentation.com.

    Cross-validation, stress testing, out-of-sample testing are all important questions to ask managers in order to rule out those managers that simply run over-fitted, over-optimized portfolios with lookback bias and whose models are based on past assumptions – which may or may not hold true in the future. A dangerous combination imho.

    https://www.hvst.com/posts/63417-how-everyone-can-prepare-for-a-more-unusual-investment-environment

  4. rjb says:

    Nice post. Quick follow ups though. First, what % of your capital is recommended to be dedicated to covering put selling (if the market drops, and your shares are called)? Second, what percent should be going to the one month T-Bills? Thanks in advance.

    1. Joachim Klement says:

      The model assumes that the 1-month T-Bills are used as collateral for the put options so no cash is needed. Of course in practice the T-Bills do not mature on the same day as the options so there might be some need for additional cash. That should however be rather low since I am selling one month put options and they are almost never more than 5% out of the money at maturity. My guess is that a 5% to 10% cash position plus the rest in T-Bills should suffice in practice but that is something one needs to check.

      1. @Alex,

        Please be careful before implementing a strategy such as this if you are aren’t sure about the details of selling theta when implied volatility is low. While I do believe in the strengths of this strategy that Joachim has mentioned and I’ve even done this strategy myself (thanks for writing a post on this strategy, Joachim!), it is not without its issues – as will all investment strategies (i.e. there is no holy grail).

        In fact, even Buffet used sold a ton of puts – back in 2008 or 2009 – at the lows of the market and when implied volatility was high. But if you’re not perfectly sure of the details of the strategy – what could go wrong, what could go right, then I would recommend paper trading the strategy on a forward basis or at a minimum using Statistics to help guide you at the link below.

        Here’s the key: You basically want to be able to stick with the strategy through its ups and downs on a long-term, multi-decade basis. If you don’t completely understand the loss profile of a strategy, then it becomes harder to stick with it during drawdown years or years with relative under-performance.

        That said, I’m actually wondering if I should sim this strategy myself b/c I could see how – in the future – if mean expected returns are lower and volatility higher – that could set up a situation where 60/40 trades only sideways for many years. In this scenario, it is within the realm of possibilities that Joachim’s strategy here could outperform 60/40 – or at least add value to an investor’s overall strategy. Please see my 100 equity curve sim’s of how 60/40 may perform in the future in the link above and good luck to all of us.

        https://www.hvst.com/posts/63829-here-s-an-easy-way-you-can-use-statistics-to-improve-your-returns

  5. Alex says:

    Hi. Could you clarify what do you when you assigned with underlying? Do you sell it right away by fixing losses and write another put uding the rest of the cash in your portfolio? How we should think about money in/outflow in your portfolio?
    Thanks

    1. Joachim Klement says:

      Hi Alex
      In my model I use S&P500 put options and they are cash settled. In case of a physical delivery of the underlying one would assume instant sake of the inderlying and investment of the proceeds in T-Bills.

  6. Garry says:

    Hi Joachim,

    A model based upon selling puts (leveraged or unleveraged) works until it doesn’t. You may recollect Victor Niederhoffer, the hedge fund manger who was carried out on stretcher (metaphorically) after selling leveraged puts on SPX.

    Question: Why not a bull put strategy for insurance?

  7. Dom says:

    I’m not fully on board with this. I’ve always seen the purpose of hedge funds as providing a relatively uncorrelated return stream. The free lunch of diversification and re-balancing are what add value.

    This strategy is definitely uncorrelated when equities rise any or fall sightly. However it is going to be nearly perfectly correlated when equities fall sharply. That’s exactly when you want an uncorrelated hedge in your portfolio.

  8. Doug says:

    Alas, these are counterfactual results. Had an investor actually done this, he would have impacted pricing of the options, particularly as his position got larger and larger in such a way as to bascially nullify the results, particularly the period of significant outperformance late.

    The problem is always that these returns aren’t dollar weighted and the impact of success – which is that you outgrow the money that can be realistically deployed in nearly every strategy – is never factored in (how can it be, really).

  9. richard says:

    Isn’t there already something like this? The CBOE S&P 500 PutWrite Index?

  10. richard says:

    Sorry, I hit the button too early…Isn’t this product similar to what you’re talking about? WisdomTree CBOE S&P 500 PutWrite Strategy Fund (PUTW)?
    “PUTW invests in one- and three-month Treasury Bills and sells or “writes” S&P 500 Index put options. The number of put options sold is chosen to ensure full collateralization, meaning the total value of the Treasury account must be equal to the maximum possible loss from the final settlement of the put options at expiration.”

  11. Gabe says:

    If one were to implement this strategy, lets say with $1,000,000. How would you derive how many contracts to sell short?

    My guess would be: If the S&P was trading at 2,200 you could buy ~454 shares, thus in this strategy you would have sold 45 contracts?

  12. Richard McGee says:

    Hi Joachim,

    The strategy of changing the moneyness in response to recent volatility is similar to the strategy of Moreira & Muir (2017), Volatility-managed portfolios, Journal of Finance.

    The issue with this type of volatility timing in the market is that most of the results come from avoiding a small number of bad events and historical performance relies on getting the timing right for those historical events (I have a breakdown of these timings in a draft/note here: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3832340).

    In your strategy the size of the periodic drawdown is going to be very sensitive to the moneyness of the puts sold when the music stops and a market crash happens.

    I found with the vol timing strategy, if historically you re-balance mid-month rather than at the end of the month, then the performance disappears. Is this because there is something special about end-of-month re-balancing that guarantees this timing or is it because the timing just happened to line up with a few negative market events historically?

    If it’s the latter there is a good chance the next crash might not be timed correctly, you will be writing closer to the money puts, and past performance will not reflect future returns!

Leave a Reply to Richard McGee Cancel reply

Your email address will not be published. Required fields are marked *



By continuing to use the site, you agree to the use of cookies. more information

The cookie settings on this website are set to "allow cookies" to give you the best browsing experience possible. If you continue to use this website without changing your cookie settings or you click "Accept" below then you are consenting to this.

Close