Is it time for risk certification?

The question above might cause you to respond, "we already have certification programs for risk," and you would, of course, be correct. However, they're for a much broader view of risk than I have in mind. For example, the "FRM" covers much more than risk that those involved in investment management typically see.

During last month's Performance Measurement Forum meeting in Boston, I asked our members if having a certification for risk would be a good idea, and got very good response. The certification would cover:

• risk measures (e.g., standard deviation, beta, downside deviation, tracking error)
• risk-adjusted measures (e.g., Sharpe ratio, Treynor ratio, Information ratio, and Modigliani-Modigliani)
• risk-attribution (e.g., the work that Jose Menchero and Philippe Gregoire have done).

While the CIPM program includes risk, it doesn't to the extent that I'm suggesting.

Two names for the program have already been suggested: CRP (Certified Risk Professional) and CIRP (Certified Investment Risk Professional); others, perhaps more worthy, can be considered.

At this point we're merely floating the idea, to see what folks think.

There's no doubt that risk measurement's role has increased significantly. Many questions continue to surface as to what measures to use, how to calculate them, and how the risk team should be "married to" or "integrated with" the performance folks (or, for that matter, if they should BE the same folks).

Feel free to offer your thoughts.

An example of why I insist that those who comment identify themselves

Occasionally readers of this blog send me comments; and usually the writers identify themselves. My friend Steve Campisi has been the most frequent one to share his views, which often extend my thinking or offer alternative views, all of which are welcome. Contradictory thinking or opposing views are also invited.

But once in a while someone comments anonymously, and long ago I announced that these comments won't be posted. Very rarely, but on occasion, sarcastic remarks have been offered: it's a shame that these individuals won't identify themselves. And, on occasion the comment has value, but given my rule simply won't be posted. Today I received the following, regarding a post from February of this year. That post suggested that without knowing the rules, mistakes could be made and was inspired from something that had appeared on Facebook. This person offered the following:

I certainly wouldn't follow any of your 'Investment Performance' because your maths is flawed. The actual correct answer is -20.

The order of mathematical operands is:
Items in brackets or parentheses = ()
Other operands eg Indicies [sic] etc
Division = /
Multiplication = *
Addition = +
Subtraction = -
and if you are faced with multiples of the same operand then you must complete them from left to right.

Thus:
0+50*1-60-60*0+10=x
0 + 50*1 - 60 - 60*0 + 10 =x
0 + 50 - 60 - 0 + 10 =x
50 - 60 - 10 =x
-10 - 10 = -20

The mathematical problem that was posed was:

0 + 50 x 1 - 60 - 60 x 0 + 10 = ???

The writer is correct in the order, but somehow flipped a sign (note what I've shown in red above).

Given the writer's wording (the use of the word "maths") and notation (the use of the single quotation (') mark rather than the double (")), I'm guessing he or she is a "Brit," though I could, of course, be mistaken. The correct answer, as revealed in the blog post, is zero.

This anonymous person made a simple error, which we are all subject to. If they hold true to their threat, they will not see this post, which is fine, but at least I get to once again state that anonymous posts won't be shown ... oh, I guess I showed this one. Darn! Well, never again!!!

Say it aint so ...

I have been aware that there is no requirement for the Global Investment Performance Standards (GIPS(R)) to undergo a quinquennial revision; however, I believe there is an expectation for a review every five years in order to determine if a revision is in order. It appears that this may not be occurring this time.

It has been rumored that changes to the Standards will be introduced into guidance statements, which I find somewhat upsetting. I have opposed such acts, as they invariably lead to problems. Two cases in point:

1) The Significant Cash Flow guidance, which I worked on, introduced required disclosures, which were easy to overlook. Unfortunately, when the 2005 edition of the Standards was published, we overlooked these guidance statement-embedded required disclosures, making the job of the GIPS folks who took advantage of significant cash flows a bit cumbersome. It was just too easy to miss the required disclosures, which many firms did.

2) The Error Correction guidance introduced requirements which were not in the previously publicly reviewed version (I have previously offered my view that such major changes warranted the document going back out to the public). When GIPS 2010's draft was circulated, many of us saw for the first time these requirements; the public responded in opposition. And so, they were dropped from the Standards, but remained in the "GS," meaning they remained in effect. Much confusion resulted, which I think made the final hours of the 2010's birth overly challenging.

The intent of the GS is to offer guidance on the Standards, not to impart new rules. It's interesting to note that because of the Error Correction problem, guidance was needed on the guidance, which came in the form of a Q&A.

The 2010 edition was much more extensive than the 2005 version; perhaps one might suggest unnecessarily so. The Standards still warrant a five-year review and, if it's determined changes are needed, for it to be revised, not for them to be introduced via a backdoor approach (i.e., guidance statements). I do hope the rumor proves to be a false one, and that the GIPS Executive Committee will invest the time to revise the Standards and, if necessary, provide guidance on the rules, not impart new ones through guidance.

A GIPS Wish List

The Spaulding Group's April newsletter was just published, and it includes my "wish list" for GIPS(R) (Global Investment Performance Standards). Anticipating that some may be thinking of ways to improve the Standards, I thought sharing some of my ideas might be helpful.

Please feel free to chime in!

Performance Measurement Forum Meeting in Boston This Week

Our North American "Chapter" of the Performance Measurement Forum spring meeting is this week in Boston, the site of last week's horrific and shameless attack. The meeting is at the Marriott in the Back Bay, not far from the explosions.

A few years ago we held a meeting in New Orleans, after Katrina devastated the region, as a show of support for the local economy and residents. While there, several of us engaged in a community service project.

This week's session has been planned for months, so there was no response to the event.

No doubt there will be some discussions about what occurred, though performance and risk will be the principal topic; one that will surely offer fodder for this blog.

p.s., want to join the forum? The North American chapter is essentially closed to new members, but the European has a few openings. To learn more, please contact Patrick Fowler.

Dirty Harry as a role model for Type I and Type II errors

You're probably familiar with the statistical concept of type 1 and 2 errors. One way to look at it is that one addresses the case where you think you're right, but you're wrong; the when you think you're wrong, but you're actually right.

A thought occurred to me last night that the classic Clint Eastwood movie Dirty Harry shows this in two parts (warning, graphic content):

:


In one case, the criminal was thinking that Harry had fired only five bullets, but he was wrong, he had fired six, and his gun was empty. In the other case he thought he had fired six bullets, but he was wrong, too, as Harry had one bullet left. But, which was the worse of the two errors?

In evaluating Type I and II errors, it's helpful to investigate the impact of both errors, to determine which, if we had to, we'd prefer to make.

I'm finalizing a study on attribution, where, among other things, I've found cases where holdings-based attribution can cause effects to show the wrong sign. For example, we might see a positive allocation effect, when it should be negative, or a negative allocation effect when it should be positive. Both are errors, but they mean different things.

If we mistakenly report allocation that actually hurt performance as being positive (i.e., contributing to performance), then we're misrepresenting our skill to the client, telling them that we did something right, when we didn't.

If we report allocation as a negative, meaning it hurt performance, when in reality it was positive, contributing to the outcome, then we're hurting the manager. One might even suspect that a manager could be terminated for failing when they actually succeeded.

Again, both are errors, but which is worse? I guess it's a classic case of, it depends.

How did Bernie get away with it? What's the role of the verifier in identifying fraud?

While meeting with a London-based verification client, the subjects of fraud as it relates to GIPS(R) (Global Investment Performance Standards) and Bernie Madoff came up. They are related, in a way.

How did Bernie get away with it? To put it simply he (a) had a very solid reputation on Wall Street and was highly respected, thus any suggestion that he, of all people, would do anything improper seemed ludicrous and (b) he had a "closed system," meaning he did everything for his clients (brokerage, trading, portfolio management, custody); thus, there were no "checks and balances" on what he was reporting.

Over the past five years or so, we've learned of a few asset managers who claimed compliance with GIPS and had been verified, but who also committed fraud. "GIPS verification is not designed to detect fraud." But is that all that we can say? I think not.

If a verifier fails to detect fraud, is that acceptable, given that verification isn't designed for that purpose? I guess it depends on what was going on which perhaps they might have been expected to notice.

At one point in the movie "The Jerk," Steve Martin (Navin) plays a gas station attendant. If you saw the movie, you'll recall that he quickly established that he's not terribly bright. A customer comes in who, as I recall, is Hispanic; he gives Navin a credit card which was obviously stolen (since it had what appeared to be Jewish woman's name). Well, even Navin managed to realize that a crime was occurring.

The television show "Hogan's Heroes" often had Sgt Schultz proclaiming "I see nothing," when the prisoners were behaving in a manner that was beyond what would have normally been allowed.

I wouldn't favor verification being extended to the point where verifiers are expected to detect fraud. However, I think verifiers need to be "on guard" to the possibility that fraud might be committed.

Not all improper behavior is as obvious as Schultz and Navin were confronted with. "Would a reasonable person (or, perhaps more accurately, a reasonably qualified verifier) be expected to discover a problem in the course of their work?" That is what needs to be assessed, in the event fraud is later detected and reported.