Dumb Alpha: Don’t Just Do Something, Sit There!
“Past performance is no guarantee of future results” is a maxim every investor knows by heart. It has been repeated so often, no investor would base the selection of stocks or funds on past performance alone, right? Wrong.
There have been numerous studies that show investors are more inclined to buy equities that have gone up in price than those that have declined.
In the mutual fund world, individual fund investors tend to buy funds that have had good performance relative to their peers and to sell funds with bad absolute performance, according to a study by Zoran Ivkovic and Scott Weisbenner.
Don’t Just Sit There, Do Something!
But sophisticated investors couldn’t possibly be as “dumb” as individual investors, right? Wrong again.
Pension funds and other institutional investors have teams of fund selectors to help identify the best-in-class funds for every investment style or asset class. And these highly paid experts should certainly be able to select the better performing funds within each peer group.
The problem is that while expert fund selectors have more tools at their disposal and spend more time analyzing funds, they are also under pressure to justify their fees and salaries. It doesn’t sit too well with trustees if an adviser does all this work and then concludes that nothing needs to be changed — especially when a fund is underperforming.
A prolonged period of underperformance leads to increased pressure to not just sit there but to do something, and this pressure can eventually become the dominating factor driving decisions to fire and hire mutual fund managers.
Amit Goyal and Sunil Wahal documented this effect in institutional investor portfolios. Their major finding is so impressive that it is reproduced below.
In the years before fund managers are fired, they underperform their benchmarks, while newly hired manager outperform their benchmarks by wide margins. Once managers have been fired and new ones hired, however, this outperformance disappears. In fact, the fired managers on average tend to outperform newly hired managers in the three years after a change.
This is not to mock the investment consultant industry. I am guilty of these charges myself. In my previous incarnation as an investment consultant, I can remember several instances when performance pressure translated into an increased impetus to replace a good manager. And in another role as a portfolio manager, I came under tremendous pressure once performance was weak for more than a year. These experiences taught me a lot about career risk and frustrated me enough to write an academic paper on the topic that will soon be published in the Journal of Behavioral Finance.
Pick the Dogs Not the Stars
The above results indicate that, in practice, it might be better to pick past losers rather than past winners when selecting funds. If fund performance is mean reverting, then it might be a good strategy to select funds with several years of underperformance instead of funds with several years of outperformance.
The profitability of such a strategy was recently explored by Bradford Cornell, Jason Hsu, and David Nanigian. In what I call the “Dogs of Morningstar” strategy, they selected the funds in the commonly used Morningstar database that were in the bottom decile in terms of three-year performance and compared the performance of this group of funds to those in the top decile. In the subsequent three years, the funds in the bottom 10% outperformed the funds in the top 10% by more than 2%. Even the average funds outperformed the top decile funds.
There are several lessons investors can learn from this.
Cornell and his colleagues recommend putting more emphasis on fund selection criteria that have a proven track record of identifying superior funds ex ante. But there is also a dumb alpha technique here. Instead of being motivated by the rule “Don’t just sit there, do something,” investors might instead act based on the rule “Don’t just do something, sit there.”
Doing nothing and sticking to a selected fund for the long term — or just selecting funds at random and then sticking to them — can be a significant source of alpha in the long run. Of course, this requires something that many investors find very difficult to implement: patience and discipline.
For more from Joachim Klement, CFA, don’t miss 7 Mistakes Every Investor Makes (And How to Avoid Them) and Risk Profiling and Tolerance, and sign up for his Klement on Investing commentary.
If you liked this post, don’t forget to subscribe to the Enterprising Investor.
All posts are the opinion of the author. As such, they should not be construed as investment advice, nor do the opinions expressed necessarily reflect the views of CFA Institute or the author’s employer.
Image credit: ©iStockphoto.com/CSA-Archive
15 thoughts on “Dumb Alpha: Don’t Just Do Something, Sit There!”
Great article. You write, “Cornell and his colleagues recommend putting more emphasis on fund selection criteria that have a proven track record of identifying superior funds ex ante.”
Can you point me to articles, studies, etc. on such fund selection criteria with proven track record?
Cornell et al. mention some of these criteria at the end of their article. In general these criteria are things like fund size, fund age, management style, but it might also be active share and concentration of stocks. However, there is a lively debate these days if active share is really predictive.
Given these pressures, I’d expect some enterprising folks to build an incentive based fee structure in support of long term clients. Something a bit more sophisticated than short term trading fees. Maybe this is done in a different way, but I would think one might want to use an applied load or annual fee based on the time you commit your funds for? It sounds like a situation where there is a lot of value for the portfolio manager in a stable client base, and it might be best to encourage those with a day to day mindset to invest elsewhere? Granted, big inflows mean big dollars, and perhaps even a strong manager with solid results may not attract sufficient funds to keep the business going…. I freely admit to being largely ignorant about the pros and cons.
On the fund side there are already different fund share classes with different fee structures and loads which incentivize long-term investments. The problem in general is that the consulting side is incentivized by AuM and not by the duration of the relationship or the performance of their recommendations. So what is needed is an alignment of interests between consultants and asset owners – something that could be done with a different fee structure.
Indeed great article. You might like my opinion column earlier this year:
I know this dilemma all too well. Consultants, boards, trustees, etc. are not just clouded by emotion, their choices are often dictated by agency costs. They might know full and well that they are making inferior financial choices, but they make them anyway to create the appearance of doing something for clients, other board members, upcoming elections (for trustees), etc. I strongly suspect there might be a place for CFA Institute / the industry to create a curriculum and training and most importantly standards of conduct for institutional boards to hold them to a true fiduciary standard.
Thanks for your comments Ron. I completely agree there is a need that should be filled. The CFA Institute Research Foundation is currently updating its primer for trustees and this could be a first step.
Let’s take this up offline.
“In the years before fund managers are fired, they underperform their benchmarks, while newly hired manager outperform their benchmarks by wide margins. Once managers have been fired and new ones hired, however, this outperformance disappears. In fact, the fired managers on average tend to outperform newly hired managers in the three years after a change.” – taken directly from the article…Can you clarify this for me as I don’t understand. First it says newly hired managers outperform and then it says the old managers outperform? Am I missing something? Thank you.
No you aren’t missing anything:
* managers that underperform tend to get fired
* managers that outperform tend to get hired
* but if you track the performance of fired managers, you find that they tend to outperform their benchmarks
* and recently hired managers tend to underperform their benchmarks.
This would all be consistent with a large noise component to active returns. If these cycles of performance ups and downs coincide with manager reviews then you get the perfect recipe for destroying active returns
It’s interesting to know why there is an outperformance at fired managers. Could it be because the fund management team or investment style was changed? You often see that asset managers merge or change underperforming funds. Or is because the investment style is favourite again (like an outperformance of value stocks after a long time of underperformance). And could there be a survivorship bias? Because there are a lot of fired managers where the underperformance continues and finally they are out of business.
I think survivorship bias is not the issue here. The main driver – and I am speculating here – should be the fact that investment styles tend to go in and out of favor over time. and unfortunately, most factor exposures tend to have a three to seven year cycle, i.e. after three to five year of underperformance they start to outperform again and vice versa.
Very nice article.
I am confused about a couple of things and am eagerly waiting for your insights to learn.
1) You rightly mentioned that investment styles go out of favor leading to underperformance. For example, in the last quarter of 2016 and for the whole year, value outperformed growth. However, do we not use the proper benchmark for each fund based on investment style? Even though growth is out of favor, for a growth fund we are using a growth benchmark (and of similar market cap) and therefore I thought this argument does not hold that a particular fund underperformed ( a similar benchmark) because the style is out of favor? What are your thoughts on this?
2) As an investment consultant (in the learning mode), how should I evaluate superior managers (from the average ones) no matter how bad the near term (upto five years) performance has been for the superior manager. I know a lot qualitative factors come into the picture like manager tenure, investment process, sell discipline etc. But I am more interested to know how to evaluate the quality of sector and stock allocations that are likely to eventually pay off over a full cycle.
I am a regular reader of your blog.