Short-term update: Shared on Abnormal Returns and Zero Hedge (where our last single article cleared 0.1 million reads). Will be published this week by a BloombergView columnist, as well as be used for a popular University of Cambridge talk next week. Also a reminder that Statistical Ideas now on twitter (@salilstatistics).
One would think that active managers would eventually outperform somewhere after the negative press that ensued a year ago. And now that 2015 performance data has been properly audited and tabulated, we can see what the new results are. We normally don't delve into the same topic twice, but as with some matters there is ample public curiosity to see what might have changed. This is one of those times where is it is critical to revisit our numerous warnings about the skill of active managers in outperforming their benchmarks, particularly now. Over the past year, through today, a number of interesting folks (too many to list here but will do so on social media) at the tip-top of the investment community and in journalism have taken a key interest in the Indomitable benchmarks article. In it we showed that in all 17 mutually-exclusive risky segments of the U.S. markets, active managers underperformed their benchmarks. The gap was so wild that it was probabilistically a once in a multi-thousand year event due to chance alone. In other words, they were specifically bad for your money, can't blame it on luck, and the government has listened. And the article was one of the most popular ever for this site. Today we take a unique look at the 2015 data recently released by S&P Indices. We scan not just U.S. fund managers, but also examine fund managers in all available countries globally (for stocks, bonds, and real estate). The results will only surprise those who are just now heeding attention.
Now for housekeeping we should note that only regions with 10-years of performance data were included, to capture pre-TARP/ZIRP, a full economic cycle, and to not muddle with statistics over the short-run that more arguably are the result of luck. Canada was the only exception to the rule, with only 5-years of found data, but their performance was so similar to the U.S. that it was worth showing with this disclaimer. What we notice from this exercise above is that active managers have underperformed (for reasons noted in these most-read articles from Pensions&Investments or from CFA) over a long-run not just in U.S. equities, but also across different assets around the world. This includes once-successful investment vehicles such as Berkshire Hathaway, and Sequoia Fund.
Sure there were 2 winners in the 15 mutually-exclusive funds above (or only 2 winners out of 18 if we look at value versus growth in the U.S. as described above), while last year's rolling 10-year returns showed 0 winners out of 17 funds. That was the point of this exercise to see what range in underperformance exists across a variety of instruments and cultures (some argue that overseas fund managers may be more quantitative, or disciplined, or dabble in more inefficient markets, etc.) There is little good of course to report, though not everything appears immensely cruel. We should note that the average underperformance on these 12 funds above is 1.2% (with a standard deviation of 1.2% as well), and the 1-year U.S. equity returns for just 2015 was -1.5% (versus a benchmark of +1.0%). So it is clear that the active fund manager results, in developed-market assets around the globe, have continued to underperform by a reasonable amount. And for U.S. equities, the dire contrast on how severely fund managers were underperforming showed through with yet another year where nearly all fund categories underperformed their benchmarks. Enough of an underperformance across most risk segments that it substantiates last year's results in Indomitable benchmarks were not the result of luck.
One would think that active managers would eventually outperform somewhere after the negative press that ensued a year ago. And now that 2015 performance data has been properly audited and tabulated, we can see what the new results are. We normally don't delve into the same topic twice, but as with some matters there is ample public curiosity to see what might have changed. This is one of those times where is it is critical to revisit our numerous warnings about the skill of active managers in outperforming their benchmarks, particularly now. Over the past year, through today, a number of interesting folks (too many to list here but will do so on social media) at the tip-top of the investment community and in journalism have taken a key interest in the Indomitable benchmarks article. In it we showed that in all 17 mutually-exclusive risky segments of the U.S. markets, active managers underperformed their benchmarks. The gap was so wild that it was probabilistically a once in a multi-thousand year event due to chance alone. In other words, they were specifically bad for your money, can't blame it on luck, and the government has listened. And the article was one of the most popular ever for this site. Today we take a unique look at the 2015 data recently released by S&P Indices. We scan not just U.S. fund managers, but also examine fund managers in all available countries globally (for stocks, bonds, and real estate). The results will only surprise those who are just now heeding attention.
And what a gratification it is to see a couple sole regions where fund managers outperformed their benchmark for an asset class! We highlight in green, Italy's and U.K.'s equity markets. That's two winner out of the dozen investment segments we looked at. Were all your eggs in Italy or U.K. over the past decade? Of course they were, since you are the miraculous exception to the rule (consistently above-average just like all money-men think of themselves). The results are even worse than they appear in the chart since similar to last year, the U.S. equity market can be decomposed into value versus growth and in all 6 of these mutually-exclusive U.S. investment categories the stock index beat the fund managers! For more on how well professionals have performed in predicting the markets, take a look at this important and viral cover story in the Wall Street Journal's MW.
Now for housekeeping we should note that only regions with 10-years of performance data were included, to capture pre-TARP/ZIRP, a full economic cycle, and to not muddle with statistics over the short-run that more arguably are the result of luck. Canada was the only exception to the rule, with only 5-years of found data, but their performance was so similar to the U.S. that it was worth showing with this disclaimer. What we notice from this exercise above is that active managers have underperformed (for reasons noted in these most-read articles from Pensions&Investments or from CFA) over a long-run not just in U.S. equities, but also across different assets around the world. This includes once-successful investment vehicles such as Berkshire Hathaway, and Sequoia Fund.
Sure there were 2 winners in the 15 mutually-exclusive funds above (or only 2 winners out of 18 if we look at value versus growth in the U.S. as described above), while last year's rolling 10-year returns showed 0 winners out of 17 funds. That was the point of this exercise to see what range in underperformance exists across a variety of instruments and cultures (some argue that overseas fund managers may be more quantitative, or disciplined, or dabble in more inefficient markets, etc.) There is little good of course to report, though not everything appears immensely cruel. We should note that the average underperformance on these 12 funds above is 1.2% (with a standard deviation of 1.2% as well), and the 1-year U.S. equity returns for just 2015 was -1.5% (versus a benchmark of +1.0%). So it is clear that the active fund manager results, in developed-market assets around the globe, have continued to underperform by a reasonable amount. And for U.S. equities, the dire contrast on how severely fund managers were underperforming showed through with yet another year where nearly all fund categories underperformed their benchmarks. Enough of an underperformance across most risk segments that it substantiates last year's results in Indomitable benchmarks were not the result of luck.
As an adviser for over 15 years now, I could never bring myself to tolerate index as an acceptable investment outcome. The key problem with most active managers is they display no skill. If you re-sort the data removing managers who display no skill, and compare index with managers who consistently display skill the outcome is vastly different.
ReplyDeleteThanks much for the comment Jonathan Bonnett. This article enjoyed nearly 30k reads today! Here are some articles (2 from The New York Times & 2 from my blog; the latter analysis Larry Summers told me was "scandalously great") that you might find illuminating relative to your suggestion that we ALWAYS only look at the small minority of "winners": http://www.nytimes.com/2014/04/06/business/the-oracle-of-omaha-lately-looking-a-bit-ordinary.html http://www.nytimes.com/2014/07/27/your-money/heads-or-tails-either-way-you-might-beat-a-stock-picker.html http://statisticalideas.blogspot.com/2014/07/the-ephemeral-success.html http://statisticalideas.blogspot.com/2015/09/harvard-education-state-school-returns_25.html
DeleteConsistently displaying skill when the competition is so fierce is hard. S&P has shown through their Persistence Scorecard that the high performance of many skilled managers does not persist. Their January 2016 report shows only 23% of all domestic funds remain in the first quartile. 27% fell to the bottom quartile and 10% were merged or liquidated.
ReplyDeleteHi Paul, great to see you after some time. I address some similar topics in my response to my friend Jon who commented above. Feel free to also share these comments on twitter (@salilstatistics), and on both Abnormal Returns, Zero Hedge, and later on Bloomberg as well. http://abnormalreturns.com/2016/04/12/tuesday-links-guaranteeing-mediocrity/ http://www.zerohedge.com/news/2016-04-12/active-managers-defeated-globally-again
DeleteSPIVA contains many mistakes. In the last edition, the 2015 performance of S&P500 is overstated by 3bps (or 2% of the indice perf) !
ReplyDeleteFor their comparison on ex-US equities indices, they use mostly non tradable/ costly to replicate indices; your results are mostly due to a bias in their numbers, hence the rare event probability
Thanks much for writing again Arnaud. We are not in a position to judge your perception that the S&P results are biased in a direction you are uncomfortable with. We do often hear these types of counter-arguments though (usually inaccurate comments from readers) on this blog or on social media. It doesn't help matters when such counter-arguments are not referencing a reliable primary research data that could serve the basis of a new future analysis. Please e-mail a proper table of performance results for which we can substitute SPIVA for in any asset class (most recent interest is in hedge funds): mehta@post.harvard.edu
DeleteMorningstar got a good research as well on the subject, comparing passive funds and active. Not surprisingly, results are much more balanced. ( it depends on fee, not style of management)
ReplyDeletehttp://corporate.morningstar.com/US/documents/ResearchPapers/MorningstarActive-PassiveBarometerJune2015.pdf
What makes me uncomfortable on top of the obvious indice mistake, is not the direction, but just the fact it makes little sense to compare performance including costs/tax/transactions fees etc,to indices that are not investable.
ReplyDeleteAs per SPIVA 2010 second page, “It is not possible to invest direct in an index. Indices are statiscal composites and their returns do not include payment of any sales charges or fees an investor would pay to purchase the securities they represent. Such costs would lower performance” (this full warning has been removed from newer SPIVA by the way)
Those costs compound over 10 years and give an obvious edge to indices. It would be more interesting to compare to tracker ETFs, so those costs are taken into account. Most of S&P indices ex-US equities don’t have ETF though, so you may use any other provider.
For high Yield, I used ETFs JNK, HYG, PHB, detail is here:
https://www.linkedin.com/pulse/truth-behind-active-vs-passive-debate-arnaud-lagarde-frm
For emerging markets, I used ETFs EEM and VWO, detail is here
https://www.linkedin.com/pulse/active-management-regularly-outperforms-passive-spiva-lagarde-frm
I used 10 Reit ETFs here:
https://www.linkedin.com/pulse/so-87-us-stock-funds-underperforming-how-many-etfs-lagarde-frm
I’m comfortable to be proven that active underperform passive funds.
What I’m uncomfortable with, is marketing from an index provider, which won’t release its underlying data, has a misleading methodology and some basic errors, be taken at face value