The paper also explores the argument that a good manager may beat bad managers over time and all that is required is a longer historical performance record. “For manager skill to have a proper opportunity to assert itself there has to have a reasonable length of time and three-to-five years is nowhere near long enough,” Schofield says. In fact, the paper shows that based on numbers alone it would take 157 years to have 75 per cent confidence that the good manager, in the paper’s example, will beat all the bad managers. Given the fact it is common practice to filter managers using performance, Schofield believes at least the limitations of doing that need to be more transparent, and further, that academic studies to assess managers according to process are needed.

“Given the fact an initial focus on numbers is almost unavoidable, people doing that analysis need to be aware of the limitations of that analysis, and the role of luck,” he says. Ironically, for a firm whose corridors are filled with mathematicians, INTECH’s view is a focus on more qualitative rather than pure quantitative analysis for manager selection would be beneficial. “Investors should look at what needs to happen to deliver the alpha.”