Executive Summary
Determining whether an active manager is having a positive impact is a difficult thing to measure, without a doubt. Yet before one can even begin to determine if a manager is delivering value, you must first consider what it takes to constitute "value" in the first place. How much does an active manager need to outperform, in order to be delivering value to the client, to be worth the fee that is paid to the manager? Yet for some reason, we scale we use to measure the cost is very different than how we measure (out)performance. Is there a double-standard here?
The inspiration for today's blog post draws again from a recent discussion thread hosted by Bob Veres on Financial-Planning.com about meeting the new "impossible" demand from clients to be aware and sensitive to the macro-economic environment, on top of everything else that we as planners must do. In a comment to the thread, Veres makes the point that he was surprised when looking at a "big" market timing database, that out of thousands of "market timing" services being tracked in 2008, only four of them saw the late 2008 debacle coming far enough in advance to avoid it; in other words, only four of the funds delivered positive returns, out of thousands being analyzed. As Veres states in response:
"...I sorted it to find out who had managed to avoid the alet 2008 debacle, and I was astonished--genuinely surprised--to discover that only fur out of thousands had accomplished this feat. And they weren't exactly world-beaters; the top performance for 2008, as best I can remember, was something like 1.6%.
...
Why was I astonished? Because the law of averages should have pushed SOMEBODY out of the pack. I was expecting the usual one or two outstanding performances in 2008, which you would expect from throwing dice a thousand times, but it didn't happen."
This sort of analysis of active management is not uncommon; a typical standard for evaluating an active manager is whether he/she is "smart enough" to see a problem like the 2008 financial fiasco coming, he should be able to avoid it and not lose money.
Yet consider what that really means; why is it that we will agonize over the last 7 or 10 or 20 basis points of expense ratios on a mutual fund or ETF, yet a manager who perhaps "merely" lost 10% in 2008 - a whopping 2,700 basis points of outperformance relative to the -37% total return on the S&P 500 - would be considered ineffective because the performance isn't positive. Is this a double-standard, or what?
After all, the manager who "just" lost 10% in 2008 could do nothing more than produce benchmark returns for the next 30 years, and would STILL have more money than the person who lost 37% in 2008 and then diligently focused on cost and saved 10 basis points of expenses per year for the following 3 decades. And conversely, if the market had been UP by 10% in 2008, we probably would find it easier to celebrate the manager than was up 37% with the same 2,700 basis points of outperformance. We are just blind-sided, it seems, by evaluating good relative performance in a declining market environment.
So the next time you're evaluating an active manager and whether the performance is any good or not, be certain to really consider what constitutes successful active management, and whether good relative outperformance (even if it's still a loss) might still be a homerun for the client; as far as I'm concerned, if 2,700 basis points of market timing outperformance would still be considered poor performance, I think we need a new definition of market timing! Or more directly, if 2,700 basis points of outperformance is weak performance because it wasn't an absolute positive return, but 10 basis points of expense ratio is so important to save, then maybe we need to look more carefully at outperformance and consider what we are comparing it to.
Of course, this is not to say that concerns about managing cost don't matter. They absolutely do. It's just astonishing the lengths that planners will go through to save a few basis points of cost, yet how little we value 100 basis points of outperformance, much less 1,000 or 2,700 basis points of excess return. It appears to me to be quite a double-standard!
So what do you think? How do you evaluate a manager's performance? Do you consider a -10% return in 2008 to be terrible, decent, or spectacular? Compared to what?
Good point about positive relative performance!
Sure, there’s a double standard, but I think the fault lies not in the Financial Planners or investors who are analyzing the funds, but in the objectives/expectations of the active managers themselves.
What Bob is referring to is “market timing” not simply active management. A fully invested active manger outperforming their benchmark by 2,700 bps would be spectacular, indeed. But what is a true market timer’s benchmark? A true market timer is one who will move 100% from one asset class to another (Large US to cash to Gold, etc.) How do you benchmark that? Using the S&P 500? That doesn’t fit. So, in a sense, the only true benchmark is absolute returns.
If we are talking about fully invested active management, like a typical actively managed open end fund, then the benchmarks make sense. Generally, they will be investing in a subset of stocks selected for outperformance relative to the entire set (such as Large Cap US).
The danger is the lack of persistence in active management outperformance, and the flow of funds in reaction to recent outperformance. Bob goes on to say that the prescient timers were not in the top 100 when sorted long term. So, how many active management/market timing investors are typically looking for consistent outperformance, yet clearly didn’t get it in the prescient timers identified by Bob. Active mgmt/timing investors will tend to “manage” and “time” their managers. How many of them stuck with the prescient few who got an additional 2,700 bps in 2008 as they underperformed for numerous years prior (as they must have in order to not appear in the top 100 of Bob’s long term screens)?