Oops!

A few days ago I announced the arrival of the June newsletter, and a couple attentive readers immediately made me aware of a typo. In the "logic" section on page 3 you'll find "And the inverse holds: when the manager does control the cash flows, money-weighting is inappropriate." ERROR! It should, of course, read "time-weighting is inappropriate."

In the original draft I had it correct, but when I reread it I misread it and thus changed it...my goof! Sorry for any confusion.

An argument for simplicity

In today's Wall Street Journal, Terry Teachout provides a very interesting series of observations on complexity, as it relates to writing and music: "Too Complicated for Words." He cites a paper by Fred Lerdahl, which speaks directly on complexity and music: "Cognitive Constraints on Compositional Systems."  I confess that I haven't yet read Lerdahl's piece, but I did fine Teachout's article quite good.

He quotes Lerdahl who wrote "Much contemporary music pursues complicatedness [sic] as compensation for a lack of complexity." Might a similar statement be tossed at some of the models we run into in the world of performance, attribution and risk? Do some authors or developers of models make things more complicated than necessary?

At times I believe this is the case, as I've heard lectures on certain models which made them quite difficult to comprehend. However, after investing the time to fully grasp them, I concluded that the lecturer might be accused of engaging in some obfuscation, to what ends one can only guess at.

I often spend a lot of time learning a model, so that I can attempt to present it in a more simplified manner, as I'd prefer that others understand it. We should not be like James Joyce; consider the following conversation between Joyce and H.G. Wells, as reported by Teachout:

• Wells: "You have turned your back on common men, on their elementary needs and their restricted time and intelligence.
• Joyce: "The demand that I make on my reader is that he should devote his whole life to reading my works." 

As one who has attempted to make it through Ulysses, I can attest to the challenge Joyce provides the reader. We should strive in our profession to simplify, whenever possible. 

GIPS Verifiers Linkedin Group!

We launched a new group on Linkedin: GIPS Verifiers.

This group is for individuals and firms engaged in GIPS(R) (Global Investment Performance Standards) verification: to share ideas, thoughts, issues, interpretations, problems, etc. It will allow verifiers the opportunity to initiate discussions with other verifiers, to gain assistance, address problems, etc.

All (verifiers) are welcome! We already have 12 members! And, we're global! So, if you're a verifier, please join us.

Large cash flows

An article I wrote for the CFA/CIPM newsletter was recently published on large cash flows. Recall that the Global Investment Performance Standards (GIPS(R)) now require firms to revalue portfolios for large flows. The article shows that this change may not be as trivial as you might think.

The real challenge occurs when you have two or more flows in a month. Let's say that your definition of large is 10%, and that you begin the month with $100,000. On the 5th your client deposits $50,000, which is clearly large, so you revalue and the portfolio is now worth $155,000. Then, on the 12th the client adds $15,000: is this large? Well, if we compare it with $100,000 it is, but shouldn't we compare it with the most recent valuation.

Or, let's say you begin with $100,000 and the client withdraws $50,000 on the 5th, which results in the portfolio being valued at $53,000. On the 20th the client adds $8,000...is this large? Not in comparison to $100,000, but definitely when we compare it to $53,000!

My recommendation: compare versus your most recent valuation date.

A client recently asked "can we net the flows to determine if we need to revalue," and I would say "yes!" This is a bit more complicated, but if you started with $100,000 and on the 5th they added $5,000 and then on the 6th $6,000, revaluation can be justified. You should document how you handle this in your policies and procedures.

Behold, I have prepared my case. I know that I am in the right.

The title for this post comes from the book of Job (chapter 13, verse 18). It occurred to me that this is quite applicable when we consider the debate on money- vs. time-weighting. I know that some of you are saying "no, not again!!!," but it's a topic which I believe continues to be an important one. We have definitely seen movement to adopting money-weighting by many firms, especially in the brokerage community, but opportunities exist for further employment of this approach.

As I explained at PMAR Europe earlier this month, the argument for money-weighting can be seen from three perspectives:

1. Documentation: there are many, many articles and books which through the years have shown how money-weighting is quite appropriate. Peter Dietz, the "father," if you will, of time-weighting, never abandoned money-weighting: he saw how it had applicability. Time-weighting was to be used to judge the manager. The Bank Administration Institute, too, in their 1968 standards, saw room for money-weighting, and the ICAA, in their 1971 standards, saw how money-weighting could be applied at the sub-portfolio level.
2. The numbers: all one needs to do is look at a few examples of positive returns when clients lose money and other nonsensical scenarios to show how money-weighting is a justified return to include along with time-weighting. And examples abound as to how money-weighting is superior and justified at the subportfolio level.
3. Logic: like it or not, the justification for time-weighting is based on the desire to eliminate the impact of cash flows that aren't controlled by the manager. And the inverse holds: when the manager does control the cash flows, time-weighting is inappropriate. To argue to the contrary simply obviates the first premise, and therefore says that there is no link whatsoever to cash flow controls as the basis for time-weighting, meaning that Peter Dietz, the BAI, ICAA, AIMR, GIPS(R) (Global Investment Performance Standards), and others are wrong!

What more is needed to convince the industry?

p.s., I expounded on this topic further in our June newsletter.

The Kramer effect

I taught a class yesterday and, as I typically do, when I got to the topic of peer groups I mentioned one of its disadvantages: questionable makeup of the universe. That is, does the manager belong here?

And, as I typically do, I cited an episode of Seinfeld to make my point: the episode where Kramer is telling Jerry and Elaine about his Karate class, and how he's the #1 student. What he fails to explain is that the other students are all small children.

This raises the question of the appropriateness of the peer group. Perhaps management assignment to peer groups isn't always so obvious when in error, but there is often a concern as to the appropriateness of placement. Thus, one of the many shortcomings of peer group benchmarks.

Liquidity risk

Bernd Fischer of Investment Data Services delivered a terrific presentation this week at the European meeting of the Performance Measurement Forum in London. Liquidity risk has been a rather difficult concept for me to get my arms around; granted, I haven't spent any time really reflecting on it, but conceptually it has been a challenge. Bernd's approach is a very thorough and reasonable one, that will provide asset managers and their clients with great insights into the changes taking place in their portfolio relative to liquidity, which can result from changes in daily trading volume, credit downgrades, and other factors.

There probably would be little interest in this subject had it not been for the huge downturn two years ago, which stemmed to some extent to the absence in liquidity for several asset types. We learned that a few asset managers have developed their own models to address this risk and it's great to learn that vendors are, too. We also learned that Statpro has a product, though we have no details on its capabilities.

I will go into this topic in a bit more detail in this month's newsletter. In addition, Bernd will be writing an article for The Journal of Performance Measurement(R) on this topic.

CIPM updates...

As noted to the right of this blog post, John Simpson, CIPM will host a webinar on June 25th at 12:00 noon (EST) to discuss the changes that have occurred to the Certificate in Investment Performance Measurement (CIPM) program.

John and I were "pioneers" as we were part of the initial group who took this important exam. And we have conducted training classes since the beginning. As the material has been updated, many who have already achieved certification might find it of value to discover what these changes have been.

For further information, please contact Patrick Fowler.

Dynamic valuation tests

I recently penned an article for a CFA Institute newsletter, which will appear later this week. In advance of that, I want to share some of the ideas from the article.

As you are probably aware, effective 1 January 2010, GIPS(R) (Global Investment Performance Standards) compliant firms must now revalue for large cash flows. I believe this is a lot more challenging that one would think at first glance. Let's take the case of an asset manager that uses monthly Modified Dietz and revalues for flows greater than 10%.

Consider the following:

What if there's a portfolio that begins the month with $100,000 and the client adds $20,000: clearly this is a flow that is greater than 10%, and so you revalue (let’s say that the initial $100,000 has grown to $101,000, so the portfolio value (with the flow) is now $121,000). A week later (still in the same month) a client gives the manager an additional $15,000...is this a large flow? If we compare it with the initial $100,000 we would say it is: but if we compare it to the most recent valuation ($121,000) it’s small, yes? I would argue that you shouldn't revalue for this second flow

Or, let’s say we begin with $100,000 and the client withdraws $20,000 (a large flow), and let’s say that the portfolio after the flow is valued at $82,000. And now the client adds $9,500: is this a large flow? No, if we compare it to the initial $100,000, but yes if we compare it to the most recent valuation. And I would argue that they should revalue.

My article addresses this in greater detail, and as soon as it's available I'll point you to it.

2,000 & counting

It wasn't too long ago that I reported that we had reached 1,000 unique visitors to our blog; well, we now have had 2,000! Thank you for visiting. And thanks to all who occasionally post comments, as these are helpful in providing various perspectives.

Let's hear it again ... why do you get examinations done?

During last week's presentation to the CFA Society of Philadelphia on the latest changes to the Global Investment Performance Standards (GIPS(R) 2010), I expressed my dislike for GIPS Examinations. I had a lengthy conversation with representatives from one firm that have them done to all of their "marketed" composites, a not unusual undertaking. When I asked "why" they do this, they seemed somewhat incredulous to being asked such a question, for surely the answer must be obvious. I think they thought this was a very odd question to ask, since it's so common in the States for examinations to be done. I went on to discuss this topic with the entire group during my presentation.

It is quite likely that firms spend 50% or more of their total verification cost on examinations, so understanding their true benefit is crucial. The truth is that they have little to do with GIPS; they test whether you're "cooking your books." Surely you know if you're doing this, so you don't need an outside party to tell you.

Firms become compliant primarily because of marketing benefits.
Firms get verified primarily because of marketing benefits.
And firms get examined primarily because ... perhaps it's because either (a) they were previously getting Level II Verifications done when they claimed compliance with the AIMR-PPS(R) or (b) because their verifier talked them into it. Please don't interpret the last point as a "shot" at other verification firms: many verification firms believe they're doing their clients a good service by conducting examinations; we just happen to disagree. If you're going to spend tens of thousands of dollars annually on examinations, shouldn't you have a good reason for it?

Our posture with our clients is quite simple: first, we discourage examinations because our research shows that they're rarely asked about in RFPs. However, if all of a sudden a client has an RFP that does ask about them and they feel obligated to have one done, then we'll make ourselves immediately available to do them, even if it has to be done over a weekend.

Time-weighting vs. Money-weighting ... a never ceasing issue

Carl Bacon and I will be battling one another tomorrow at the second day of PMAR Europe I on the subject of time- vs. money-weighting. Carl and I enjoy these debates and have engaged in them several times.

This evening I was speaking with a client who mentioned that she has tried to get her management to accept money-weighting, but to no avail. It occurred to me that she needs to get past the labels: just show the numbers.

Use examples to demonstrate the difference without using labels. We'll discuss three:

Our first example is somewhat classic: we lose money but have a positive return; but, we can also see a negative return. The second shows where our gain/loss is zero but we show a huge return in one case and a zero return in the second. And the third shows us making money but having a negative return in one case and a positive in the other.

And so ask your client what return makes sense to them if we're reporting on how THEIR PORTFOLIO did; not how the manager performed. If the client offers anything but ROR #2 in all cases, we have a problem. And, of course, ROR #2 is MWRR (ROR #1 is TWRR): 'nough said.

The rules: de jure or de facto

I recently commented about the use of the glossary and samples as ways to communicate GIPS(R) (Global Investment Performance Standards) requirements.

I'm reading Exchange Rate Regimes: Choices and Consequences and in chapter 4 the authors distinguish between de jure and de facto exchange rate regimes: that is, a country may state what their regime is (de jure), but in reality do something different from it (de facto).

Let's briefly distinguish these terms, since they're not in common usage:

• de jure: by right; according to law
• de facto: in fact; in reality.

When we have a GIPS rules communicated either in a glossary entry or a sample, can we, in fact, think of them as being de jure requirements, or would it be better to think of them as de facto?

As a verifier, when we review a firm's policies and materials, and discover a problem, we cite chapter and verse in order to give credence to my position. Is it acceptable to cite an example? I don't believe so. Examples have, in my way of thinking, never been used to say "this is what we expect," but rather "this is how you might want to do it." And so, if the "requirement" to include references to risk in a composite description are de facto, then one would still expect a compliant firm to adhere to it, but not based on official doctrine. Does this make sense?

I believe I understand the GIPS Executive Committee's challenge. They believe risk needs to be in the composite description; however, given the negative responses to their proposal in the earlier disclosure draft and the acknowledged risks involved in such a requirement (see my earlier comments on this issue), what are they to do?  And so, rather than make this a de jure requirement, they imply that they expect to see it via an example, which then turns it into a de facto one. And by doing so, they reduce the risks that a de jure requirement might involve. One could argue that such a practice of including risk would be deemed "best practice." Gaining clarity and buy-in is the challenge that awaits us.

Note: both definitions come from www.Dictionary.com.

One year of blogging

I'm usually better at remembering anniversaries ... June 2 marked the one year anniversary of this blog. And 250+ entries later, with close to 2,000 visitors, we're still going strong.

I greatly appreciate the comments that have come in and continue to welcome your thoughts. We've had a few posts that have involved several folks and that have gone on for quite a bit. Such dialogue is great!

Please feel free to reach out and suggest topics that we should be addressing. Thanks!

The art of the apology

Peggy Noonan's WSJ column today discusses Detroit Tigers' pitcher Armando Galarraga's near perfect game which was only broken up by a bad call made by umpire Jim Joyce. There were two exceptional lessons to take from that event. First, Galarraga's demeanor, which was humble, understanding, and forgiving. And second, Joyce's humility in his sincere and heartfelt apology to Galarraga, immediately after realizing his mistake. That apology helped diffuse the situation.

Once when I was mayor of North Brunswick (NJ) I got a very nasty e-mail from a resident who complained about the very poor job that was done in removing the snow from his road. I responded with an apology, and I took responsibility for what had occurred (we had failed to contract with an outside party to supplement our public works department, and the snow was too large for our team to handle in an efficient manner). Well, I guess he was so shocked by my admission of guilt that he responded by apologizing for his nastiness; he also mentioned that his wife was mad at him because he had included her name in the signature line.

George W. Bush was often criticized for not apologizing. Others, too, seem to have a difficulty admitting when they're wrong. I don't believe I have this problem as I've apologized quite a bit for the many goofs I've made. When you're wrong, admit it...what's so hard about that? But for many, it is.

In various aspects of our industry I see mistakes made; often quite innocently, where the guilty party of course had the best intentions but simply made a mistake. I recall one official document (that's all I'll say in describing it) that had been prepared which was heartily dismissed by the public. Well, later during a revision when I commented about it the author was quick to tell me that they didn't want to discuss it; I wasn't being accusatory, I was simply pointing something out relative to it. Why couldn't that individual simply have said something like "well, I guess I was wrong in my expectations."? For some, apologies are apparently not seen as a positive acknowledgment of error but perhaps a sign of weakness; if anything, I'd say they're a sign of strength.

This post is more about life than performance measurement. And so perhaps it's fitting to end with one of my favorite quotes from Ogden Nash, which is intended as advice to a new bridegroom:

"To keep your marriage brimming, 

With love in the loving cup, 

Whenever you're wrong, admit it; 

Whenever you're right, shut up." 

Don't be afraid to acknowledge errors. As Galarraga  pointed out, "nobody's perfect." Don't pretend to be.

Disclosure paradigm

I have become a big fan of Malcom Gladwell and have made it through Outliers: The Story Of Success, The Tipping Point: How Little Things Can Make a Big Difference, and Blink. I'm now working on his most recent book, What the Dog Saw: And Other Adventures and came across a discussion which has some relevance to our industry.

Unlike his earlier books, which have common themes running throughout them, this recent one is a hodgepodge of his earlier writings, including a discussion on the Enron case. Enron created SPEs (special-purpose entities) to overcome difficulties in obtaining financing and presumably for other purposes. And although they were documented in their financial records, there were questions as to whether or not enough information was provided. Gladwell writes that "you can't blame Enron for covering up the existence of its side deals. It didn't; it disclosed them. The argument against the company, then, is more accurately that it didn't tell its investors enough about its SPEs." (emphasis in original) He then asks, "But what is enough?" Each SPE (and Enron had 3,000) had paperwork in excess of 1,000 pages; edited versions averaged 40 pages.

This discussion on disclosures can extend to GIPS(R) (Global Investment Performance Standards) and what compliant firms share. Recall that the GIPS Executive Committee had proposed to require firms to disclose details about their risks in composite descriptions, and many in the industry were concerned about how much might be necessary; and, if enough wasn't shown, would that then put their firm at risk?

The point at hand can be easily summarized by Gladwell: "You can try to make financial transactions understandable by simplifying them, in which case you run the risk of smoothing over some of their potential risks, or you can disclose every potential pitfall, in which case you'll make the disclosure so unwieldy that no one will be able to understand it."  Furthermore, "in an age of increasing financial complexity, the 'disclosure paradigm' - the idea that the more a company tells us about its business, the better off we are - has become an anachronism."

I'd say it's a good thing that the proposed risk disclosure requirement was dropped, because many firms feared this disclosure challenge: when do you have enough! The problem still remains, however, because (as noted in a recent newsletter) Jonathan Boersma, Executive Director of GIPS at the CFA Institute Centre for Financial Market Integrity and member of the GIPS Executive Committee, mentioned at last month's PMAR VIII conference that the EC "feels strongly that risk should be addressed in the composite description." The sample descriptions in Appendix C are apparently being used to imply this need. But, to put it simply, is a risk disclosure required or not? In a recent post I mentioned how a requirement found its way into the 2005 edition through the glossary, hardly where one would expect to look; are samples now a means to convey requirements? Hopefully, not.

Model fees ... and what does this mean?

I gave a talk today at the CFA Society of Philadelphia on GIPS(R) 2010 (Global Investment Performance Standards) and mentioned that effective January 1, 2010, compliant firms must disclose, when reporting net-of-fee returns, if model or actual fees are used. Earlier this year I mentioned that there is no definition of "model fees" in the glossary but that I suspect it means highest actual fee. Well, someone asked if it's possible to use "average fee."

While I suspect not, given the lack of clarity here I'm reluctant to give a definitive "no!" With no explanation as to what "model fee" means, some might suspect they have some leeway.

I will e-mail the GIPS help desk to try to gain some clarity.

In the middle of the day...

I'm completing a report for a client who asked me to review their performance measurement system and processing. They offer their clients the option of treating flows as middle-of-day events. I strongly oppose this method.

First, there's a reduction in accuracy: we value portfolios at the start and end of day, but not at the middle. Thus, we're not having an exact method (which we'd have with start- or end-of-day treatment) but an approximate method.

Second, I believe the reason firms choose this is simply because they can't decide whether to use start- or end-of-day.

The answer, I truly believe, is:

• Inflows: treat as start-of-day events
• Outflows: treat as end-of-day events

and don't bother with midday.

Expanding the role of a glossary...but no more!

What is a glossary? Well, www.dictionary.com defines it as

glos·sa·ry

/ˈglɒsəri, ˈglɔsə-/ Show Spelled[glos-uh-ree, glaw-suh-]

–noun,plural-ries.

1.a list of terms in a special subject, field, or area of usage, with accompanying definitions.

2.such a list at the back of a book, explaining or defining difficult or unusual words and expressions used in the text

The GIPS(R) (Global Investment Performance Standards) 2005 edition had an expanded role for glossaries: as a source of requirements. Take the expression Composite Definition: it's defined as "Detailed criteria that determine the allocation of portfolios to composites. Composite definitions must be documented in the firm's policies and procedures." (emphasis added) Who would think to look to the glossary as a place for requirements? To me, a glossary is a place to go if I find a word or expression for which I need clarity, not to find requirements.

Well, the GIPS 2010 edition has changed this; we now find "Detailed criteria that determine the assignment of portfolios to composites. Criteria may include investment mandate, style or strategy, asset class, the use of derivatives, leverage and/or hedging, targeted risk metrics, investment constraints or restrictions, and/or portfolio type (e.g., segregated or pooled, taxable versus tax exempt)." Not only has the "requirement" been removed, the expression has been nicely expanded.

Since the standards require firms to make composite definitions available upon request (see paragraph 3.A.4), you have to have them somewhere. Should they be in your P&:P? I would think this would be a good practice, but it's not a requirement. But since they are part of the process used to assign accounts to composites, I'd say they either should be there or at least referenced in the P&P.

Like a Rolling Stone...

My West Coast amigo, Juan Simpson, has posted something regarding the Rolling Stones. Now, the trick is to figure out what the Rolling Stones have to do w/performance measurement ... well, check it out!

http://cipmexamtipsandtricks.blogspot.com/2010/05/performance-professional-like-rolling.html

P&P Swap

Recall that we extended an invitation for firms to participate in a GIPS(R) (Global Investment Performance Standards) Policies & Procedures swap. Well, we've gotten some very positive responses to this idea, and decided to extend the deadline by two weeks until June 15.

If you wish to join in, please send me (DSpaulding@SpauldingGrp.com) your P&P (redacted or otherwise; we won't do any editing) by June 15.

Residuals when you don't expect to see them

Most folks in our industry are aware that if you employ a holdings-based model for attribution, you're subject to residuals creeping in. Residuals are the unexplained differences between the excess return (portfolio return minus benchmark return) and the sum of the attribution effects. Because holdings-based models don't capture intraperiod cash flows, they're subject to residuals. And so, our research confirms that most asset managers prefer to use a transaction-based model to be more accurate.

However, if you're using a monthly return formula (e.g., Modified Dietz), then you may still end up with residuals, because the transaction-based approach is capturing an exact return, while the Modified Dietz arrives at an approximation to the true, TWRR. Ideally, you should be using a daily return method so that you will eliminate the residual.

Note: you can still have what might be called a longitudinal or intertemporal (occurring across time) residual, unless you employ an appropriate linking method.

p.s.,  I believe I'm the first to use the term "intertemporal" to describe this form of a residual. I believe it's a handy way to distinguish between single-period residuals which can arise from using a holdings-based model, and residuals across time, when using an arithmetic approach.