We often get questions about a problem all asset managers face: a client requests you to sell some of his/her/their securities, to free up cash that they plan to withdraw. Between the time that the cash is created and it's finally taken by the client, it's sitting in the account, impacting the portfolio's return. What options does the manager have? There are a few:
- You can leave the cash there and accept the impact. When reporting to the client, you can explain that the cash diluted the return and that you would have invested it if you could have, but at the client's request, it was created for them.
- You can move the cash out immediately into a "temporary account." This account would be on YOUR system (i.e., the custodian doesn't know about it), and would segregate the cash you raised from the account and any other cash that may be there (i.e., cash you can invest).
- You can move the cash into a different cash "bucket" (or pseudo security), which you would flag as being "unmanaged," so that it isn't included in the returns.
Once the cash is created it's nondiscretionary, is it not? When it was invested it was under your control, but once you create it for the client, you cannot invest it, and so it should arguably be segregated. By moving it into a temporary account, you accomplish this. Also, if you can move it into a different security category and simply give it a different name ("non-discretionary cash") which you've flagged as unmanaged, your return calculation will exclude it.
Some of our clients raise cash for clients who say they'll take it immediately, only to find that the cash is still there weeks (or months) after it was raised. It's unfair for the cash to impact the return when the manager can't invest it. It's just like any other unmanaged or nondiscretionary asset: it should be excluded from the return. Have some other thoughts? Please share them.
A recent WSJ headline referenced one major bank's problems as regulators challenge its risk management practices. This made me think of comments I usually make during our firm's Fundamentals of Performance Measurement course, when I touch on risk management: having a risk policy can be risky!
This comment is based on a consulting assignment we had several years ago, that involved a plan sponsor's challenge of one of its managers risk management policies. They felt that the manager had failed to live up to their documented control procedures, and we were asked to conduct an analysis to discover what we could regarding this matter.
Of course, a policy has to be sound and as complete as possible. The point here is that having one on paper but not in reality can be hugely problematic. Not having one is also a problem, as more and more clients expect to see some risk management in place. I recall a NYC asset manager client who contacted me one day about establishing one: their largest client had asked what their policy was, and the reality was that they didn't do any risk measurement or management.
Policies, procedures, and controls cannot reside only on paper; they must be exercised and validated occasionally, to ensure they are performing as expected, and from time to time enhanced, too.
This has been quite a year. We've had our ups, and our downs, just like all years. Overall, The Spaulding Group had a great year. Our verification practice continues to grow, and our GIPS(R) verification clients now number more than 100. Our PMAR conferences both had record attendance, and our training numbers were on the rise. Our new software certification service has been well received, and we will shortly award certifications to two of our clients. We are grateful for those who have hired us to serve them, as well as our the many professionals we have the pleasure of working with. On a personal note, I am thankful for the team we've assembled.
A dear friend, who I met almost 50 years ago when, as young men, we were members of the Order of DeMolay, and who passed away 6+ years ago, told me how he always liked it when Chanukah and Christmas overlapped. While the holidays have nothing in common other than their proximity, I agreed that having them occur at roughly the same time was especially nice. This year they are separated quite a bit, and so our Jewish brethren have already exchanged their holiday gifts.
Our pastor's letter in this weekend's church bulletin referenced two events which have taken away from the holiday season for many of us: Hurricane Sandy, which smacked our region a few weeks back, and those horrific killings which struck a nearby Connecticut school 10 days ago.
Yesterday was the last Sunday of Advent, and just like the prior Sunday, the four candles in our Advent wreath were changed to black, as a sign of mourning for those who perished in Newtown. The song we began the mass with brought tears to my eyes, and the tears appeared again yesterday, as I reflected upon those candles. Our hearts continue to ache for those who perished and the loved ones they left behind.
Our family will celebrate Christmas tomorrow, though tonight we will go to church with our sons, our daughter-in-law, and our grandsons. We will also exchange gifts this evening, since this year our grandchildren (and their parents) will be "at the other grandparents" for Christmas day.
Despite the difficult times, we will try to focus on the happiness that the holiday brings. I am blessed that my wife of 40 years and I will share it with family.
We hope our Jewish friends and colleagues had an enjoyable Chanukah, and wish our Christian friends and colleagues a happy and blessed Christmas.
The Spaulding Group's December newsletter deals with the use of spreadsheets, and their all too frequent use as "systems." I point out, as I've done countless times elsewhere, that they are:
- time-consuming
- error prone
- cumbersome
- not databases
and so should be avoided, if possible.
Earlier this year I was asked to develop an attribution model for a client that employs option writing along side its long stock positions. It's a global firm, so the model has to also be sensitive to currency movement, though since the client doesn't employ a hedging strategy, naive multi-currency attribution works.
I completed the assignment, and delivered the spreadsheet. Because of its complexity, the firm will most likely not employ it, which I can fully understand.
I do see the exercise as a way to prototype a model, so as to gain confidence in the approach, formulas, assumptions, etc. From such an effort a programmed system can be built.
This exercise will also likely result in an article for The Journal of Performance Measurement(R), to explain the model as well as to further support the use of spreadsheets to prototype models, but not to turn into "systems."
When I teach our fundamentals of performance or attribution classes, I often mention how we adopt ideas from other segments of society. Measuring performance, for example, is found in many areas, with sports perhaps being the most thorough (with baseball leading the way).
Attribution, too, is found in many places. Police departments conduct attribution when attempting to ascertain the cause(s) of an accident, and fire departments do the same when investigating a fire.
The recent horrific massacre at the Sandy Hook Elementary School is causing a rash of attribution analysis, as the contributing causes are being analyzed. How much did mental illness play a part, should gun control laws be examined, does violence on television lead to violence in real life, etc.? And just as performance attribution is controversial, so are many of these discussions.
I'm conducting a study which demonstrates how some approaches to performance attribution or flawed, and can lead to incorrect conclusions (an article summarizing preliminary results will be published shortly). This, in spite of the fact that we're talking numerical attribution. In the case of analyzing a crime, much of the work will be subjective, which can be influenced by various factors, including political and public pressure. To avoid flaws in the results, objectivity must be complete and the evaluation unbiased.
David Kopel wrote an interesting piece in Monday's WSJ titled "Guns, Mental Illness, and Newton." He points out how we've seen a significant increase in random mass shootings over the past 30 years. Some are quick to "rush to judgement," while many admit they "haven't a clue." President Obama wants action taken to make these events an impossibility, which is admirable and something we all would support, though how realistic this goal is is yet to be determined.
My heart continues to ache for those who died and the loved ones they left behind. The attribution analysis has begun, and hopefully action can be taken to reduce the risks of similar attacks taking place.
Almost every day I receive questions from clients and others about various aspects of performance and risk. And for the most part I am able to respond, either with my opinion or an answer based on fact.
What occurred yesterday in Newtown, CT leaves me, and no doubt you, too, with no answers. How anyone could commit such a heinous crime baffles us to no end. How anyone can kill small, frail, innocent children, as well as innocent adults, is something that none of us can give a good reply to, other than perhaps madness. This young man was set on a journey of destruction; his motives or the basis for such action will probably never be known. Man's inhumanity to man is still shocking to me. But it's difficult to think of anything that can compare to this horror.
One of my friends expressed being somewhat at loss as to how to pray for the parents who lost their young children; I suggested that to pray that they be given peace and comfort is an okay thing to do. They will forever, probably every single day, be reminded of their loss. Nothing I can think of can compare to such a loss, especially at this time of year, as most of these young folks were awaiting the arrival of Santa in less than two weeks, and others just wrapping up the festival of lights, a time that no doubt brought them much joy.
Two weeks from yesterday is the Feast of the Holy Innocents, when we remember those victims of Herod's pursuit of the newborn Christ child. Those youngsters who died yesterday are holy innocents, too.
My heart aches as I think of what occurred. As a parent and grandparent, perhaps this event is more hurtful, though I don't believe we need to have any children to know how horrible this crime was.
I pray for the parents, family, and friends who suffered the loss of a child yesterday. And, I pray for the victims; it is my hope that they are at peace, resting in the arms of a loving God.
At this month's Fall (Autumn) Performance Measurement Forum meeting, which was held in one of my favorite cities (San Francisco), one member posed a question about ancillary systems (i.e., those systems that sit aside the packaged or programmed systems that most asset managers use, including manual and spreadsheet). The reality is that EVERY FIRM just about has at least one spreadsheet-based system to support their performance measurement needs. This discussion raised a question about "best practices" for such systems.
Interestingly, this past weekend's WSJ had an article ("Ex-Trader's Gambit Bites Goldman") which discusses how a former Goldman and Morgan Stanley trader "using a manual system typically for trades done over the counter or those handled by a floor broker, entered 60 false 'sell' trades to make it look as if he was reducing his position." Does this not speak to the need for some form of "best practices"?
At our Spring Forum meeting (which we expect to be held in Boston) we will address this subject. Here are a few quick thoughts (you're invited to chime in with your own):
- Avoid having manual or spreadsheet based systems when packaged software can accomplish what you require
- KNOW what systems exist within your organization. Construct an inventory of all systems, who built them, who's responsible for their maintenance, who uses them, and what they're used for
- Know what data is used by these systems and what data comes out of them, to be used by other systems
- Identify the controls that are in place to ensure integrity of the information
- Put in place a process by which such systems must go through a testing process before they can be implemented...
- ...also, for changes!
- Identify systems that are used as an alternative to programmed/packaged systems
Give me time, and I'll come up with a few more, but this is at least a start.
As you might expect, this is one of the areas we look at when we conduct our operational reviews. It's not uncommon to find spreadsheet based systems that pose huge threats to the integrity of the firm's data.
.png)
