From the November 2012 issue of Investment Advisor • Subscribe!

The Art and Science of Manager Selection

In the end, you are judging a person’s character and abilities

Most of the time, investors are glad to see new money coming into a manager they’ve chosen because they’re pleased to have their choices confirmed by others. It’s even better if you know that the new money is smart money. That’s why I smiled this summer when I read that the Church of England and my company, Altegris, share the same pew, so to speak. Both of us have money with David Harding of Winton Capital Management. The Church of England picked Winton and two other hedge fund managers (Bridgewater and Blackstone) in 2011 to manage some of its $1.7 billion pension fund, doubling its allocation to hedge funds in the process.

By necessity, it also confirms that your manager selection process works when others follow in your footsteps and recognize the same talent. Still, it would be an interesting exercise to see how the Church of England selected Winton. Its superb track record as a commodity trading advisor (managed futures manager) is no secret, but its proprietary mathematical algorithms and advanced statistical analysis of historical market data are most definitely confidential. In other words, at the final stages of its manager selection process, the Church of England had to make a leap of faith.

Altegris began investing with Winton in the late 1990s. I remember how we selected the firm. We looked at Winton (which is David Harding’s middle name) from every conceivable angle—starting with having followed Harding’s career since 1997, when the Cambridge-educated theoretical physicist left the Man Group (where I began my career) to form his own commodities futures trading firm with $1.6 million in assets.

We watched Winton grow into a $29 billion AUM firm with more than 100 researchers whose areas of expertise range from operations research to statistics, climatology, actuarial science, extragalactic astrophysics and financial mathematics. We studied Winton’s computer models and its strategy of holding long and short positions in over 100 futures markets around the world. As with nearly all our managers, we insisted that Winton accord us full transparency so that we could monitor its trading activity and performance on a real-time basis. Even with transparency, however, the final decision came down to a combination of art and science because, in the end, you are judging a person’s character and abilities. If they were being frank, I think every other Winton investor would say the same thing.

Volumes have been written about the manager selection process, evidence that no one yet has successfully broken the code. That’s because manager selection, especially of the increasingly popular commodity trading advisors, is very difficult. Among other things, we conduct a “stress test” to see how a fund performs through a variety of market scenarios and extraordinary events. We want to make sure that a manager has not been guilty of improper data mining—that is, choosing only that portion of history that confirms a theory about investing. Models that managers rely on have to be truly and honestly researched. The buy and sell signals a manager points to as historical examples of how the model will work in the future have to be convincing to us from both a technical and “common sense” perspective. Above all, we want to avoid the bad version of curve-fitting—that is, forcing the data to conform to the model rather than the other way around.

Even when a manager is well-known to you, there are no shortcuts in the selection process if you decide to reinvest. For example, we have had a 10-year relationship with Capital Fund Management (CFM), a Paris-based commodity trading advisor as part of our hedge fund advisory business. When CFM made the decision to re-open in mid-2011, we approached the investment from a fresh perspective and re-validated our investment thesis, which was achieved by spending many hours reviewing CFM’s models with its key research personnel. We drew upon years of live data from CFM’s trading program to analyze its impact on our portfolio.

The two managers we have been discussing are both commodity trading advisors with proprietary algorithmic techniques, but the same rules apply to managers with more traditional or fundamental approaches. With fundamental managers, we drill deep to understand how they generate alpha. The selection process must involve exploring the manager’s special approach, conducting operational due diligence and typically insisting on full transparency—in short, all the tools available to us. But in the end, all of the tools do not permit us to avoid the final question—is the manager worthy of our trust?        

 

Reprints Discuss this story
This is where the comments go.