Home | Asset Allocation | Most Popular Mutual Funds | Advisor Commentaries | Subscribe | About Us | About the Data | Archives | Advertise
 


Evaluating Active Managers:
The Role of Belief Systems

John R. Minahan
June 3, 2008

Go to page Previous, 1, 2, 4, Next     Email Article   Display as PDF

Sadly, most traditional managers I interview are unable to credibly explain how they trade before their perspective gets into prices.  Many managers appear to not even think in these terms, but instead focus on exploiting past patterns (e.g. "low price-book always works in the long-run") or buying “Mom and apple pie” stocks (e.g. "we only buy the highest qualities companies, the companies that made America great").  There is nothing necessarily wrong with attempting to exploit past patterns or with buying quality companies, but unless a manager can explain why their perspective is not reflected in the prices of what they buy and sell, there is no reason to expect they will outperform in the future. 5 6

Evaluating belief management

The key questions I have regarding a manager's belief management are:

  • How does the manager self-evaluate their investment process?  What evidence does the manager look at to do this?
  • Does the firm’s culture support or retard greater awareness of potentially unconscious beliefs?

I will discuss each in turn:

How does the manager self-evaluate?

The fundamental problem of performance evaluation is that returns have a substantial random element.  This makes it difficult for an evaluator to conclude that good performance derives from successful execution of a sound investment strategy.  This is obviously true for an external evaluator such as me, but an internal evaluator faces the same problem. 

Some managers don’t put much thought into self-evaluation.  If performance relative to a benchmark or peer universe is good, they stop there.  If performance is poor they may put more effort into analysis so as to craft a favorable spin on the situation.  These are not the managers I want to hire. 

I prefer to hire a manager who is driven to make honest and careful assessments of their investment process in good times and bad, and who understands that good performance can occur for reasons other than successful execution of the investment strategy.  That is, I prefer a manager with a proactive belief system with respect to self-evaluation of their investment process.  There are two things such a manager can do to partially mitigate the fundamental problem of performance evaluation:

  • Get the benchmark right.7  This ought to go without saying.  Yet the overwhelming majority of managers I review use a benchmark that is not meaningfully connected to either the investment process or a reasonably specified passive alternative.  The simplest example of this is a manager that selects securities from a universe materially different than the index to which the manager is compared.  Externally, they may have no choice about their benchmark.  Internally, they can use whatever they find most useful.  If they are not careful about this choice that tells me they are not serious about self-assessment.
  • Supplement performance data with non-performance indicators of process success.  A manager with a proactive belief system wants to know if their process is working, and recognizes that good performance (even relative to a good benchmark) is not sufficient to come to that conclusion.  Therefore, these managers seek additional indicators that the process is working.  For example:
    • A manager who predicts earnings can track the accuracy of these predictions.
    • A manager who predicts bond upgrades and downgrades can track those predictions.
    • A manager who develops a fundamental view of the future can confirm whether the future actually plays out as predicted.

The key is that a manager recognizes there is a wide variety of data that can bear on self-evaluation, and proactively seeks out and uses the data available.  If a manager does not do this, it seems fair to conclude that it is not important to the manager to improve over time.

5 It is important to note that not all managers fail this test.  Some managers understand that trading ahead of price is the central issue, and are very good at explaining how the opportunity arises and how they exploit it.  I have had somewhat better experience with alternatives managers than traditional managers in this regard, but in both camps there are managers who do this very well.

6 Even when a manager can offer a plausible reason for superior performance, there remains the question as to whether there is any evidence that the plausible reason is the actual reason.  It can be very difficult to generate such evidence, but as an evaluator I want to see that the manager has tried.  Any manager that is seriously attempting to outperform realizes that he has as much at stake as anyone in understanding whether good performance actually derived from the manager's strategy or just good fortune. 

7 Benchmarks are essential.  As Waring and Siegel (2006) have argued, the universal goal of all active management is to add value over a benchmark.  “Benchmark-free” investing is more a matter of not being clear about what the benchmark should be than it is an actual style of investing. 

 

Go to page Previous, 1, 2, 4, Next

Display article as PDF for printing.

Would you like to send this article to a friend?

Remember, if you have a question or comment, send it to .


Contact Us
Website by the Boston Web Company