Cherry-Picking Data to Make Active Managers Look Good

by Charles Rotblut | May 30, 2019

“Active managers have been much more successful over the past decade than is commonly realized,” asserts Neuberger Berman. The asset manager backs up this statement with return data from the last 20 years. It’s a bold statement intended to spark interest in actively managed funds, which the company offers. Whether you agree with the statement depends largely on how much you think the data adjustments are justified.

The study’s authors purposely did not look at the entire universe of mutual funds. In explaining their rationale, they write, “Instead of comparing the performance of an index against an entire universe of active strategies, it’s more appropriate to reframe the conversation to include only those managers in the top three quartiles of performance and thus eliminate a small cohort of poorly performing funds unlikely to attract significant investment flows.” Based on this, they conclude that 84% of active managers have beaten the S&P 500 index over the past 20 years.

Unlike the children from Garrison Keillor’s fictional Lake Wobegon, not all funds are above average. Some are flashes in the pan, doing well over a certain period of time before crashing and eventually disappearing from the fund universe. While Neuberger Berman included funds that were liquidated, by excluding the worst-performing quartile, it is possible that those funds were only included during their good years and excluded during their lousy years. Such a viewpoint assumes that an investor would have enough foresight to predict which funds would have maintained their outperformance AND got out of those same funds before their returns fell to the bottom quartile. Using the full universe of actively managed funds would have resulted in a worse batting average.

The report bears this out. The rolling 10-year period analysis shows large-cap blend funds ranking in the top three quartiles beating the S&P 500 66% of the time. When the all-funds universe is analyzed instead, the batting average falls to 33%—a big difference.

In addition to excluding certain funds when they did poorly, the analysis also used rolling periods as opposed to a straight 20-year holding period. We know some actively managed funds will outperform over a given time period. We also know that the best funds over, say, a five-year period are frequently not the best ones over the next five years. While investors may jump in and out of funds, it’s a big stretch to assume that the majority of investors—individual or institutional—successfully jump from one outperformer to another outperformer.

Then there are the returns of the last 10 years, which the report’s authors attempt to dismiss. They cite higher correlations within S&P 500 (meaning the returns of large-cap stocks have been more similar) as “depriving active managers of the opportunity to distinguish winners from losers through fundamental research.” Yet, the rationale for accepting the higher expense ratio of actively managed mutual funds is to benefit from the skill of their managers. While certain managers (e.g., value fund managers) can be given a pass because their types of strategies have simply been out of favor since 2007, the entire set of active managers should not be. These managers have been paid to beat the market.

Throughout the history of the markets, performance data has been cherry-picked, twisted and sometimes tortured to tell a story that someone selling an investment-related product wants you to hear. While there are many of us in the investment industry who show the data in open light, there are enough questionable claims to warrant maintaining a degree of skepticism. With Neuberger Berman, it is simply a case of cherry-picking data to put active mutual funds into a more positive light.

As far as actively managed mutual funds are concerned, there are some fund managers who have beaten the S&P 500 or another appropriate benchmark over an extended period. Are they easy to identify in advance? No. For those who want to try, a place to start is to look for funds with lower expense ratios, a systematic approach to investing and a truly active approach to investing. Comparatively smaller size in terms of assets managed relative to their peers is also a good trait to seek, since large amounts of investor inflows can wreck even good strategies. You will also need a good dose of patience because even good managers will incur periods of underperformance.

More on AAII.com
AAII Sentiment Survey

Two out of five individual investors are pessimistic about the short-term outlook for stocks. The latest AAII Sentiment Survey also shows a pullback in neutral sentiment and a very slight change in optimism.

Bullish sentiment, expectations that stock prices will rise over the next six months, increased 0.1 percentage points to 24.8%. Optimism was last lower on December 12, 2018 (20.9%). This is the 15th time this year that optimism is below its historical average of 38.5%.

Neutral sentiment, expectations that stock prices will stay essentially unchanged over the next six months, fell 4.1 percentage points to 35.1%. Even with the drop, neutral sentiment remains above its historical average of 31.0% for the 17th time in 18 weeks.

Bearish sentiment, expectations that stock prices will fall over the next six months, rebounded by 4.0 percentage points to 40.1%. Pessimism was last higher on January 2, 2019 (42.8%). This is the third consecutive week that bearish sentiment is above its historical average of 30.5%.

Pessimism is now at an unusually high level (more than one standard deviation above its historical average). The breakpoint between typical and unusually high readings is 39.8%. Optimism remains at an unusually low level, below the breakpoint of 28.1%. Historically, unusually low levels of bullish sentiment and unusually high levels of bearish sentiment have been followed by higher-than-median six-month returns for the S&P 500 index.

Many individual investors have been monitoring trade negotiations, particularly between the U.S. and China. We’ve also heard from AAII members who are concerned about a drop in stock prices occurring, and the recent weakness may be playing into their worries. Also having an influence are monetary policy, Washington politics (including President Trump), geopolitics, valuations, corporate earnings and the pace of economic growth.

This week’s special question asked AAII members what influence first-quarter earnings are having on their outlook for stock prices. Nearly two out of five respondents (38%) say either earnings have not affected their outlook or that other factors are having a bigger influence. Approximately 20% say any positive impact from earnings is being offset by the trade war. Seventeen percent of respondents think earnings are near a cyclical top or don’t support the current level of valuations. About 18% view first-quarter earnings as a positive for stocks.

Here is a sampling of the responses:

  • “I do not use earnings to determine my outlook. The economy has more going on than just earnings.”
  • “Any positive influence from earnings has been negated by the trade war.”
  • “First-quarter earnings and those for prior quarters are at a level where future earnings growth will be more difficult.”
  • “Earnings have generally been positive.”


This week’s Sentiment Survey results:

Bullish: 24.8%, up 0.1 points
Neutral: 35.1%, down 4.1 points
Bearish: 40.1%, up 4.0 points

Historical averages:

Bullish: 38.5%
Neutral: 31.0%
Bearish: 30.5%
Take the Sentiment Survey.

Discussion

Larry from NY posted over 7 years ago:

This is the second article I have recently seen from AAII warning about cherry picking data or being wary of how data is presented when trying to pick an equity or fund to invest in. I would like to point out that I think you are guilty of that too, at least when it comes to performance data for SSR. On the website you highlight that SSR since its inception has a total return of 275.2% while the benchmark has a total return of 257.6% during that same period. But the the question that comes to mind is this simply a case of a few good years at the beginning hiding subpar performance over the majority of the years. Now, to your credit, you do list the return for each year from 2002 to 2018. Taking that data, I then ask the question, if I had $10000 to invest at the beginning of each year from 2002 to 2018, would I have been better off at the end of 2018 investing in the SSR portfolio or the benchmark. The answer for the beginning of 2002 is SSR, but for every other year, beside the beginning of 2009 which is essentially a tie, I would have been better off investing in the benchmark. It was this analysis that led me to believe SSR was not a good option.


You need to log in as a registered AAII user before commenting.
Create an account

Log In