Last Thursday and Friday, I attended the Financial Research Association’s Symposium on Performance and Risk Measurement in Boston. There were a lot of informative sessions and I took almost ten pages of notes, so I should have enough material for a bunch of posts.
Unfortunately, I wasn’t able to travel on Wednesday evening, so by the time I arrived at the hotel around 10:15, I had missed the first session, which was given by Holly Miller from StoneHouse Consulting. I’ll try and speak to her this week and see if she can give me a summary or if she has a blog posting I can link to.
The next two sessions reviewed the upcoming changes to the GIPS standards (called the GIPS 2010 Exposure Draft, which you can find here) and were hosted by Karyn Vincent, president and founder of Vincent Performance Services. I first met Karyn at the FRA Tech & Ops Conference in Miami back in February when I was lucky enough to find myself sitting next to her at lunch. I found her to be extremely knowledgeable about all aspects of performance measurement, so I was looking forward to these two sessions.
There was also a roundtable to provide some pros and cons for each proposed change and encourage discussion. The panelists were:
- Todd Juillerat, Global Head of Performance, State Street Global Advisors
- Jason Millard, Beacon Verification Services
- Steve Riordan, President, Riordan Consulting
- Mike Savage, Director of Performance Distribution, Sungard
These sessions provoked some lively comments and discussions between the panel and the audience. Karyn went through many of the proposed changes to GIPS for 2010 and gave the rationale that the committee used to come up with each change. I won’t go over every single change, since you can easily do that for yourself on the GIPS website. I’ll just review some of the highlights:
4.A.29 Firms will be required to disclose the 3 year annualized ex-post standard deviation of the composite and benchmark.
This change provoked the most responses from the audience. While everyone agrees that performance without risk is meaningless, not everyone believes that standard deviation is an effective measure of risk. Committee members from European countries were pushing for multiple risk measurements to be used, while US members wanted less, so presenting standard deviation was the compromise. The EU members don’t feel that this change goes far enough in bringing risk into the disclosure process. The committee was just trying to get risk in there some way, according to Karyn.
Jason said that he “hates this one” for a number of reasons, one of which is because the time period is arbitrary. Also, the use of standard deviation assumes you have a normal distribution, which often isn’t the case for most portfolios. Todd countered by saying that a compromise is, by definition, a sub-optimal solution, so people on both sides will take issue with this change.
4.A.5 FIRMS MUST disclose the presence, use, and extent of leverage, derivatives and/or short positions, if material, including a description of the frequency of use and characteristics of the instruments sufficient to identify risks.
The key point here, noted Todd, is that this only applies to material positions only and “material” is not defined by the standard. Each firm must make their own definition of “material” when evaluating their holdings in light of this change, Karyn stated. Mike was concerned about the time and effort required for firms to document all of their leveraged positions along with descriptions of the use. “Where do you stop?” He questioned how much time and effort will be required to create and maintain this documentation. Firms don’t have to track the percentage of short positions, only an answer of “yes” or “no” that they’re used is sufficient, Karyn clarified.
3.A.1 All actual discretionary PORTFOLIOS MUST be included in at least one COMPOSITE.
The change here is the deletion of the words “fee-paying” before discretionary, such that now all portfolios, both fee-paying and non-fee-paying, must be included in at least one composite. Non-fee-paying portfolios would include proprietary assets. This sparked some discussion since principals might be more aggressive with their own funds and including them in composites with customer portfolios might skew the performance. However, it was pointed out the assumption that all assets in the same composite are managed in a similar manner. Therefore, any proprietary portfolios that are managed differently, should be put into their own, separate composite.
Section III – Verification, The GIPS standards RECOMMEND that FIRMS be verified. 1. VERIFICATION MUST be performed by a qualified independent third-party.
This proposed change also requires firms to disclose whether or not they have been verified. While the verification requirement is voluntary, some smaller firms may cite the cost of hiring a third-party verifier as a road block. Cost shouldn’t be used as a basis for non-verification, said Mike. Also, in the current environment of heightened regulatory scrutiny, any firms that claim GIPS compliance will be double-checked by regulators.
Karyn put out a call for feedback from all firms on the GIPS 2010 Exposure Draft. The public comment period will last until 1 July 2009.