Wednesday, November 30, 2011

A standard in name only?

One of our clients sent us a note recently, which stated that one of their clients told them there is a "standard to report alternative investments as net-of-fee, and to combine these with gross-of-fee returns." This "standard" apparently is not in writing, so perhaps the term was being used a bit loosely? Perhaps "a common practice" (although how common this is is open to debate) or "best practice" (though I would think it's not "best") might have been a better choice of words. In general I think "standards" should be written down somewhere; this is an example of one that isn't, though it's still worth discussing.

Mixing assets with net-of-fee returns with assets with gross-of-fee returns is never a good idea: what do you end up with? A hybrid, where you cannot make much sense out of what is being shown. If there is a "standard" (though again, it's probably a "best practice") it would be to not combine net-of-fee return assets with gross-of-fee, unless you have no choice. An example of how this used to be the case: in the early days of the AIMR-PPS(R), many U.S. firms included mutual funds in their composite, where the funds had net-of-fee returns (because at that time, firms were prohibited from "grossing-up" the fund returns) and separate accounts had gross-of-fee returns. The result was a composite gross-of-fee return that was lower than it would have been (had the fund been grossed-up). In 1996, the SEC issued a "no-action letter" which allowed firms to gross-up fund returns, so this is no longer a problem.

There are loads of things that mix well together, but net- and gross-of-fee returns aren't two of them.

Monday, November 28, 2011

Our performance numbers don't agree!

Last week I had a call from a client who said that the returns they calculate don't match those provided by the hedge fund. How can this be?

Hedge fund performance should be pretty easy to handle: once we have the monthly valuations, cash flows are typically restricted to being once a month (on the last or first day), so it's quite simple to do the math. And so, what might be going on?

Well, our client said that they recognize flows when they occur! This could be the problem, right?

In wanting to get the money to the hedge fund in time that it's available, many of their clients will wire funds a few days early, but the hedge fund pretty much (as I understand it) ignores the money; (i.e., just lets it sit outside the corpus of the fund), until they're ready to bring it in: as a partnership, they do partnership accounting, where they calculate an end-of-month NAV (net asset value), and then issue new shares based on this value and the money being deposited (or, in the case of withdrawals, reduce the number of shares). If our client treats a flow as an intra-month event, using either Modified Dietz or a daily measure (where they simply carry the fund value from the start of the month to every day in the month (since daily values won't usually be available)), differences can occur, yes? There can be other reasons for differences in returns, but this seems to be a likely candidate.

My advice: adopt the same method as the fund. For example, if a flow actually occurs on November 28, but the hedge fund won't recognize it until November 30, having the return reflect the flow on the 28th isn't going to help when trying to match up with the fund. The fund is not doing anything with this cash, so treat the flow as if it occurred on the 30th, just as the fund will. As with all of my posts, I welcome your thoughts.

Wednesday, November 23, 2011

"the sort of precision that gave one confidence"

My son, Douglas, gave me Rules of Civility by Amor Towles for my birthday. And while I don't often read novels, the review in the WSJ painted it as a book I should read (it likened it to The Great Gatsby, and after finishing it I can see why). Given that performance measurement is our passion, it's not surprising that I found inspiration within this text for a post.

In describing how a waiter poured three martinis, such that there was just enough to fill each glass with great exactitude (with not a drop too much, or too little), the narrator remarked how "It was the sort of precision that gave one confidence."

This line reminded me of something Roger Lowenstein wrote in his best seller When Genius Failed, about Long-Term Capital Management: "Long-Term did not merely concede the possibility of loss, it calculated the supposed odds of it occurring, and to precise mathematical degrees...The point was, Long-Term predicted the odds with precision." (emphasis in original) And no doubt, this precision gave LTCM's investors confidence in those who managed their money.

When a pilot informs the passengers that he expects to land at "about 5:17," the mere fact that he didn't round to the nearest five (i.e., 5:15) suggests a high degree of confidence, in spite of the qualifier, "about."

Precision implies accuracy and confidence, which isn't always valid. One should think about how far they want to go with the precision of their reporting, if the numbers are based on approximations, estimates, preliminary information, etc. Sometimes qualifying language is in order, to guard against confidence that might be misplaced.

Monday, November 21, 2011

Wondering about your error correction policy?

Here's an opportunity to gain some insights into the proper way to construct your policy. (If you claim compliance with the Global Investment Performance Standards (GIPS(R)), you are required to have a written policy!).

Join me on Monday, December 5, from 11 AM to 1 PM (EST), for our delayed November Webinar. We will go through the guidance statement, clarify some key points, and address:
  • Understanding the requirements
  • Common mistakes
  • Dealing with materiality
To register or for further information, please contact Jaime Puerschner or Patrick Fowler (732-881-5700).

Note that a modest site fee will be charged (and you can have as many people on the line as you'd like!), except for our verification clients and members of the Performance Measurement Forum (who can participate at no cost!).

Friday, November 18, 2011

Taking advantage of "senior moments"

When I was an undergrad math major at Temple University, roughly 40 years ago, a professor, who was teaching a Numbers Theory course, mentioned that there were three things mathematicians should always remember: the Pythagorean Theorem, Fermat's Last Equation, and he said he couldn't recall the third.

I often cite this episode when teaching our Fundamentals of Performance Measurement course, when I suggest that there are three things performance measurement professionals should know (in my scenario, I recall all three).

I thought of my professor's admonition when I heard about one of the Republican candidates for U.S. President, who said there were three departments he would eliminate, but failed to recall the third. His apparent stumbling didn't go over well. It seems that he hasn't learned the art of self-effacing humor, and turning moments like this into something funny.

Now that I am in my 60s (you're shocked to hear this, I know; people tell me all the time I don't look that old), I have the benefit of referring to episodes of forgetfulness as being "senior moments," something that always results in a laugh. And while my age probably has little to do with temporary memory lapses (given that everyone is subject to them), it's a line that fits well (at least for folks my age). Had the candidate made a similar statement, perhaps he would have gotten by without the negativity that has resulted.

Humor works, but only if the speaker is adept at pulling it off. This topic came up at a conference I was at this week; many of the attendees liked the humor I interjected during the session I moderated. Later, during dinner, I discussed this with a fellow dinner guest, and said how there is an art to telling a joke, and mentioned the joke about the fellow who walks into a bar, sits down and orders a beer. A few minutes later someone calls out "52," and everyone there began to laugh. A few minutes after that, someone called out "67," and the same thing occurred. Next, someone yelled "18," and the crowd again laughed loudly.

Turning to the bartender the fellow asked what was going on, since he didn't see any humor in numbers being called out. The bartender explained, "we have had the same neighborhood folks come in here for many years. And as you might expect, the same jokes kept getting told, over and over again. So, we decided to number them."

Hearing this, the man called out "44." Silence. Again, "44!" And again, silence; no reaction at all.

And so, he turned to the bartender and asked, "is there a joke numbered 44?" and was told "yes, there is."

"Well, how come when I yelled it out, no one laughed?" The bartender's response: "some people don't know how to tell a joke."

Wednesday, November 16, 2011

Understanding the why

While driving to Boston to moderate a panel on fixed income investing, my wife and I were listening to David Baldacci's latest book, Zero Day. Early in the book he explains why folks in the UK drive on the left-hand side of the road. It apparently dates  back to the "jousting days," when knights typically approached their opponent on the left hand side, since most were right handed, and this would be a better way to attack.

The reason Americans drive on the right-hand side is, as I understand it, is attributable to an American Indian custom of approaching oncoming riders on the right, as a way to show that they came in peace. Funny how both are apparently due to horse riding, but for completely opposite reasons.

Understanding "the why" behind why we do things can be quite helpful.

I was engaged in a conversation with another verifier last week at our European Performance Measurement Forum meeting in Budapest, Hungary. The question we addressed: "why do firms in Europe not have GIPS(R) (Global Investment Performance Standards) performance examinations done, while they are quite common in the States? The "why" here is pretty clear: it dates back to the days of the AIMR-PPS(R), when large accounting firms wouldn't do a "Level I" ("verification" in today's terms) verification, but would do a "Level II" (equivalent to today's examinations). And so, for many years lots of U.S. firms were only having Level II (examinations) done. When GIPS was introduced, these same verifiers dropped the prior restriction. As a result, the firms that perhaps would have avoided Level IIs had their verifiers offered Level Is continued as they had been.

Understanding "the why" might cause us to reconsider some of the things we do, though I wouldn't hold my breadth on anyone shifting the side of the road they drive on any time soon.

Tuesday, November 15, 2011

Real Estate - is it leverage or not?

The Global Investment Performance Standards (GIPS(R)) require compliant firms to make certain disclosures regarding the use and extent of leverage in their portfolios (See ¶ I.4.A.13).
Does this include real estate?

Answer: No!

Yes, real estate investing is a leveraged asset, but this isn't what the standards are referring to. Here we speak about things like options and margin accounts.

It therefore naturally follows that mortgage backed securities and asset backed securities are not leveraged assets, either! A verification client of ours was wrestling with this and fortunately called for clarity.
MBS and ABS in your portfolio? Fine; no need to disclose.

Saturday, November 12, 2011

Seeing risk as an opportunity

I'm sitting in the Lufthansa business class lounge, preparing to return home, after spending a few days in Budapest, Hungary, where we held our 51st meeting of The Performance Measurement Forum. The reading material includes the WSJ's European edition weekend edition, which doesn't "hold a candle" to the U.S. version, but at least provides me something to read. I was struck by a Dow Jones advertisement, with the heading "Risk is also an opportunity." One might think such a statement unnecessary, as one can't achieve rewards without taking some risk. Its opening sentence, "Protect your business from risk without hindering growth," does seem like a stretch, however. But perhaps I need to spend some time pursuing their website.

One of our presenters this week was Harry Kat, who has written for The Journal of Performance Measurement(R), as well as other publications. Harry's presentation on risk was interesting in its simplicity, and I'll have more to say about it in the coming days.

Back to the WSJ advertisement. Over the years I've collected a few such ads, as they can be quite interesting in how they portray risk. We rarely see anything addressing the flip side (performance, which some say, with risk are the two sides of the same coin). And yet risk continues to be quite a challenge, in understanding, measuring, and managing it. More to follow.

Thursday, November 10, 2011

Overselling time-weighting

Motivational speaker Zig Ziglar frequently interjects jokes into his presentations. One deals with a Roman Catholic girl who was asked by her Baptist boyfriend to marry him. This occurred many years ago, when such “mixed marriages” were often frowned upon. When the girl told her mother the news, her mother responded “Darling, you know you can’t marry a Baptist boy! We’re Catholics, and it just wouldn’t work out.” The girl replied: “I know, Mom, but we’re in love!”

The mother thought about it and offered a solution: “Why not take him to Church with you, and let him experience the beauty and power of a Catholic Mass. Perhaps you can convince him to convert.”

Well, the daughter did as instructed, and a few weeks later informed her mother that the boyfriend had agreed to become Catholic, so a date was set for the wedding.

But then a short time later, the girl came home crying, explaining that the wedding was off. “What happened?” asked her mother. “Did he decide not to become a Catholic?” “Worse,” explained the daughter. “I think I oversold him. He’s going to become a priest!”

Sadly, the investment industry has been oversold on time-weighting. Ironically, those who were doing the selling (e.g., Peter Dietz) never claimed that time-weighting was the universal way to measure performance, and often cited money-weighting as the preferred approach to evaluate performance from the client's perspective (that is, "how did my account do?," as opposed to "how did the manager do?").

I've been asked to write a brief article on this topic, and will provide a link to it shortly. Because of space limitations, I couldn't use Zig's joke in the article, so I decided to use it here; hope you liked it!

Wednesday, November 9, 2011

Who's the client? (revisited)

One of our GIPS(R) (Global Investment Performance Standards) verification clients posed the following question: "We have had the question come up here as to whether a potential limited partner in a partnership that we manage (managed in a similar manner to a mutual fund) is required to receive a GIPS compliant presentation?"

This raises the question as to who the client is?, for GIPS purposes. Our client cited a post I did more than a year ago.

As with a mutual fund, I would argue that for GIPS purposes, the "client" is the partnership, not the investors in the partnership (for a mutual fund, the fund is the "client"). The prospective shareholder/limited partner isn't investing in the composite's strategy; they're investing in a particular product, which no doubt comes with loads of disclosures that will give them all the information they require.

Might it be a good idea to include the appropriate composite presentation(s)? I would say, "yes," though only if doing so doesn't conflict with local regulators, who might have rules that make this inappropriate. But in my opinion, it's not required.

Monday, November 7, 2011

There seems to be a "lacuna" in the justification that was offered

I did not attend this year's annual GIPS(R) (Global Investment Performance Standards) conference in Chicago, though we sponsored it, and four of my colleagues were present. I learned that someone raised a question which I put forward here a few days ago, regarding the apparent introduction of a new rule, prohibiting the introduction of accounts within the month. An attendee apparently submitted a question on this item, and it was explained that it was believed that this was always the Standards' intent, or something to that effect.

Well, as this blog's title suggests, there seems to be a lacuna in this justification. And, if you're like me, you may find the term "lacuna" to be somewhat foreign to you. I learned of the word while reading Neal B. Freeman's piece in this past weekend's WSJ; the article marked the 50th anniversary of Bill Buckley's God and Man at Yale. And so, turning to one of my favorite websites, I discovered that it means "a gap or missing part, as in a manuscript, series, or logical argument." [emphasis added]

In this case, the missing part (or lacuna) seems to be the basis for this claim. Where in the Standards does one conclude that it was always the intent of the Standards? Surely there must be a paragraph that one can point us to. Absent that, my only conclusion must continue to be that this is a new rule, probably in response to some of my arguments challenging the aggregate method as a legitimate approach to measure a composite's average return.

p.s., While Freeman may not be quite the sesquipedalian (user of long words) that Buckley was, Bill would most likely still have been pleased with the article.

Saturday, November 5, 2011

Bold predictions about the future can be haunting

Dippin' Dots' claim of being the "ice cream of the future" appears to have been a bit aggressive, as it declared for bankruptcy.

Granted, this bold statement was merely their tag line, and perhaps some could foresee ice cream actually universally becoming like their product (I only tried it once, and that was enough for me), but it's still risky to be so confident about what lay ahead.

Perhaps this can serve as a metaphor for similar offerings of what tomorrow will bring.

Thursday, November 3, 2011

Periodicity for risk statistics (and other measures)

A client recently asked about the appropriateness of reporting volatility and tracking error, for one and three month time periods, based on daily values. In addition, whether the results should be annualized. He was curious about the "best practices" for reporting these values.

Volatility (presumably, standard deviation) requires at least 30 observations to have validity, and so a monthly statistic, based on the roughly 22 trading days in the month, wouldn't quite make it, but it's close; clearly the three-month period would. But, this begs a bigger question: the appropriateness of using daily values.

I would say that the "rule of thumb" is to only use monthly returns. And why is this? Because daily provide too much "noise." Just consider the past week's DJIA returns, where we've seen 100+ point movements up and down, allegedly in response to the economic crisis in Greece. Monthly returns smooth these gyrations out, into meaningful numbers. Quarterly would be arguably better, but to achieve the requisite 30 observations would require 7 1/2 years: a bit long for most folks.

As for annualizing the monthly or quarterly values, the "rule" is not to annualize returns for periods less than a year; I would think the same applies to risk stats. And, if your intent is to compare them to a monthly or quarterly return, why not have it represented in the same manner?

Disagree? Have different thoughts? Please chime in!

Tuesday, November 1, 2011

Best practice for composite construction

One of our GIPS(R) (Global Investment Performance Standards) verification clients called to ask what's the best practice for composite construction. Interesting question, I think.

We first touched on the meaning of "best practice," as it relates to the Standards, which is arguably an unknown, though I have concluded that it's what the Executive Committee thinks is best, which is fine by me; it's just important to know the context of the term. In the case of our conversation, though, I suspect the client meant what the industry might deem best, or best for ones clients or prospects.

If we turn briefly to the Standards, we know there are a host of recommended criteria available in the Composite Definition Guidance Statement. In reality, the standards afford the firm a great deal of latitude when it comes to construction. As is often stated, one can define their composites broadly, which allows more accounts to be included, with the likely result being more dispersion, or narrowly, meaning fewer accounts, with the anticipation of tighter dispersion.

Let's consider a brief example. Let's say that you have a U.S. large cap growth strategy, which you implement in two ways: with separate securities for your larger accounts, and mutual funds for your smaller ones. The "instruments used" can be a criteria for composite construction, and so you could have two composites: USLCG with separate securities and USLCG mutual funds. Or, you could combine them and have a single composite.

The advantages of a single composite are (a) more assets and (b) having to deal with just the single composite from a maintenance perspective; it also probably means wider dispersion. Going the dual composite route means tighter dispersion; it also means you'll have lower assets and have to work with two composites. Which is best practice? Well, if we're speaking from the standpoint of the firm it really depends on what they wish to accomplish. If you are disappointed that you'll have lower assets if you go with two composites, you can still show prospects the other composite, too, in order to demonstrate that your presence in that strategy is broader than one might conclude by looking at the single composite. From the prospect's perspective, to me it's clear that the dual composite approach makes more sense, because it means that they will see the one that aligns more accurately with their objectives and what you'll be doing for them. In the end, though, it's up to the firm. If you decide to have one large composite, then its description needs to indicate that mutual funds or (or possibly, "and/or") separate securities may be employed.

There are always trade-offs, and the firm gets to decide how they wish to deal with them. I was pleased that our client takes this so seriously, and wishes to do what is best for their prospects. It may mean more work for them, but it will allow them to better represent their success. There is more to be said on this topic, and we'll likely take it up again in the future. Your thoughts, ideas, and reactions are always invited.