Performance Perspectives Blog

No, no, no, no, no!!!!!

by | Jan 27, 2011

Is my title strong enough? Perhaps not, but it’s all I have right now. It is in response to an online article I stumbled upon that strongly encourages investors to ask for, believe it or not, time-weighted returns! The article, by Doug May and titled “Five rules for measuring portfolio performance,” offers such advice as:

  • encouraging investors to “Ask if your advisor can provide a ‘time-weighted’ portfolio return calculation. By using a time-weighted return, fluctuations that have nothing to do with portfolio picks can be eliminated from the evaluation.” Because advisors often don’t manage the money, what possible benefit is there to eliminate the cash flows from the evaluation? And even when advisors do manage the client’s money, there is still a good reason to show money-weighting, as clients also want to know how their portfolio or they are doing, which time-weighting is unable to provide.
  • to justify this, he offers an example: “if an investor decides to add money to his or her account at the market bottom, this additional money invested at the optimal moment will skew portfolio returns upward. Portfolio investments could be lagging the market, but the timing of the additional money might result in above-average returns.” “Skew portfolio returns?” No, we want the impact of the flows to be shown so that the investor can see exactly how they are doing.
  • “By using a ‘time-weighted’ measurement, which is the industry standard for periods longer than three months, distortions from contributions and withdrawals will be reduced or eliminated.” I confess that this is a new one: the industry standard for periods longer than three months? I am not familiar with that standard. In fact, there are no standards on what to report to clients; there are only standards on what to show a prospective client.

I am quite sure that Mr. May is a very nice and intelligent person; he is a CFA charterholder which reflects a great deal of investment knowledge. However, he is relying on an incorrect view regarding the purpose of time-weighting. Let us not forget that this form of measurement was introduced in the mid 1960s as a way to evaluate how the manager performed. I have, as my loyal readers know, opined on this topic many times. I think that my newsletter discussion  last June summarized my arguments quite well.

It is way too easy for individuals to recommend time-weighting; it’s more challenging to step up and point out this measure’s shortcomings and the value of money-weighting. But we must!

Free Subscription!

The Journal of Performance Measurement

The Performance Measurement Resource.

Click to Subscribe