" Get your facts first, then you can distort them as you please ." - Mark Twain
" There will always be arguments against indexing. If passive funds never existed, traditional fund managers would be collecting an additional $40 billion in annual fees. I have never heard a sound argument against indexing advanced by active managers. Indeed, I can extend that lesson: I have never heard a sound argument against any investment practice coming from those who would profit if they are believed. " - John Rekenthaler
The greatest change in investment management over the past 40 years has been the growing impact of index investing. Since 1975, passive investing has increased from 0.01% of total assets under management to 28% in 2017. Some industry observers expect the figure to hit 50% by 2024 . This seismic shift between active and passive investing is driven by two simple facts. Index funds are far cheaper. And it is extraordinarily hard for an actively managed fund to outperform the markets.
I've written about this difficulty several times before, but a recent paper  by Hendrik Bessembinder of Arizona State University has taken a new research angle that deserves discussion . Here is the paper's abstract.
" Most common stocks do not (outperform Treasury bills). Slightly more than four out of every seven common stocks that have appeared in the CRSP (Center for Research in Security Prices) database since 1926 have lifetime buy-and-hold returns, inclusive of reinvested dividends, less than those on one-month Treasurys. When stated in terms of lifetime dollar wealth creation, the entire gain in the U.S. stock market since 1926 is attributable to the best-performing 4% of listed companies. "
It takes some time to process this paragraph. Since 1926 roughly 57% of all publicly traded stocks cumulatively earned returns less than a 30-day Treasury bill. This type of number goes against every core tenet we've heard most of our investing lives. " Buy stocks for the long run " or " a great way to lose your money over your lifetime is to only buy bonds ." And there are many more of these tidbits of wisdom. As usual there is some truth to these, but Bessembinder's research certainly shows an interesting counterpoint to these claims.
But the data gets more interesting as you dig deeper into the paper. Bessembinder finds that:
- 96% of all equities over the past 80 years have added nothing to the stock markets' total returns.
- More than half of the common stocks in the CRSP database delivered negative lifetime returns.
- The single most frequent outcome (when returns are rounded to the nearest 5%) observed for individual common stocks over their full lifetimes is a loss of 100%.
- A final kicker: When Bessembinder ran single stock models  only 27% of the time did the stock outperform a one-month Treasury bill.
Why this is important
Bessembinder's research was received with all the industry excitement that you might think. No investment manager (including this one) likes to read about how small your margin of error is and the long odds you face in picking successful investments. That is not to say outperformance isn't possible. My partner John Dorfman has proven this over the years with an outstanding record. But putting aside the issues raised by the findings (mostly related to ego) there are some critically important findings that investors should take to heart.
Shows how hard business success really is
Many investors have actually never been owners of a business. For those who have had the pleasure (or agony as the case may be) of creating, building and growing a company, these owners know this is awfully hard work. It is estimated that 90% of all startups close their doors within the first year. Finding a company that will continue to grow and create value over a 20-year time horizon is a difficult prospect. There are simply not too many prospects to be had in such a competitive marketplace. Bessembinder's research graphically demonstrates this.
Focus on the 4%
Being able to identify the 4% of stocks that will generate long-term gains is extraordinarily difficult. Just like those who said they wanted to know where they would die so they wouldn't go there, investors have to develop a process that gives them the best chance to find long-term compounding machines. Many exhibit similar attributes - prudent allocation of capital, deep competitive moat, product/services with a long-term strategy/need and outstanding management that have a flexible strategy but strong core values.
As always play not to lose
Once again it should be noted the most successful investors have been those who prudently invested with the idea of not losing any principal. Every investor will have a bad year or two. Such is the nature of investing. If you can avoid the truly catastrophic losses (such as those stocks who over their lifetime lost 100% of their value), the good years will take care of themselves.
It's not often when a paper such as Bessembinder's gets published. His research looks at stock returns in a truly novel way. His research makes it clear there are two major ways to outperform the market. You can take advantage of short-term price dislocation versus a company's value or use long-term compounding to achieve market outperformance. This type of investing requires strong behavioral attributes, tremendous patience and a deep knowledge of your investments. After finishing the paper, most investors will have a greater appreciation for long-term outperformance. Mark Twain said to get your facts first. I strongly encourage all investors to do that and start by reading Bessembinder's research. Any distortion after that is all up to you.
As always I look forward to your thoughts and comments.
Disclosure: I own no stocks cited in this article.
 "Asset Managers - Global: Passive Market Share to Overtake Active in the US No Later than 2024," Moody's Investor Services, Feb. 2, 2017
 "Do Stocks Outperform Treasury Bills?" Hendrik Bessembinder, Arizona State University, Aug. 22, 2017.
 I should add that there have been considerable questions about both his methodology and findings. For an alternate view, see Gary Carmell's response here.
 This model was based on holding a random stock for each month between 1926 and 2016. Each monthly return was then measured against the return on a one-month Treasury bill.Premium Members
This article first appeared on GuruFocus .