Most Individual Stocks Don’t Outperform Cash?

A new academic paper was recently published with a confusing yet provocative title: Do Stocks Outperform Treasury Bills?. Of course they do… right? An excerpt from the abstract:

Most common stocks do not outperform Treasury Bills. Fifty eight percent of common stocks have holding period returns less than those on one-month Treasuries over their full lifetimes on CRSP. […]

But everyone knows stocks return more than cash. How does this work? Taken altogether, stocks outperform cash. But if you picked any individual company, your results can vary from total bankruptcy to extraordinary wealth. The paper found that if you pick an individual company and held it over its lifetime, it would be more likely than not to underperform a 4-week T-Bill (classified as a cash equivalent). You can use the T-Bill as an approximate tracker of inflation.

Wes Gray points out at Alpha Architect that this idea has been explored before. Here’s a chart of the distribution of total lifetime returns for individual U.S. stocks. The research is done by Blackstar Funds, via Mebane Faber at Ivy Portfolio.

captialdist

The U-shaped distribution shows that there are a lot of big losers and a lot of big winners. Actually some are huge winners. In the end, a small minority of stocks have been responsible for virtually all the market’s gains.

captialdist2

Here we see that out of the 26,000 stocks studied, these 10 stocks below have accounted for 1/6th of all the wealth ever created in the US stock market.

captialdist3

I should reiterate that these are lifetime returns, from when they appeared in the CRSP database until now or whenever they liquidated. Unless you bought these stocks essentially at IPO (or 1926 when the database starts), you probably didn’t get these returns. If you go out and buy a well-established company today, your distribution of returns will likely look different. You’d be less likely to go bankrupt but also less likely to make a 20,000% return.

If you take a step back, as Larry Swedroe points out, this means it is technically quite easy to outperform an index fund. You simply either (1) avoid investing in a few big losers or (2) invest extra in a few big winners. That’s it! Gotta be easy to filter out a few duds, right? Yet, the lack of outperformance on average by professional managers continues, and the managers that outperform can’t be predicted ahead of time. So you can keep looking for the needles in the haystack, or you can buy the whole haystack.

Comments

  1. John Walker says:

    Yes, picking individual stocks can be risky. This is interesting. There was one thing in this that blew me away. The 2 million percent return on Altria. One dollar at IPO (probably 100 year ago at least) is $2,000,000 today.

    • Yup, I think it’s a good reminder that these numbers are “lifetime” returns, in others words from IPO (or 1926 if IPO was earlier) to either now or whenever they liquidated. If you only limit yourself to large-cap stocks, your odds of getting a stock that beats inflation goes up to 70% because you are picking the stocks that you already know would be relatively successful. But you also reduce your possibility of a crazy high return.

  2. It is so interesting that the matter in which you design your study can influence your results. That way you can have the data to support any conclusion you want. If you look at randomness, you will get randomness.

    I find it ironic that somehow the editors of WSJ have been able to achieve market like returns with only 30 stocks.. And done that for over 100 years. I guess by focusing on quality blue chips, those leaders with strong competitive advantage, and minimizing turnover, you can do pretty well over time.

    • Yes, you can get market-like results with just 30 mega cap stocks, but I think the important take-away is that you won’t get market-beating returns. If you could just leave out a “bad” blue chip or buy a little extra of a “good” blue chip, why haven’t 10 managers just made a mutual fund that does that exact thing? It’d be easier that dealing with small stocks due to higher liquidity of those mega caps.

      • I agree that achieving market beating results is very hard. Even many index investors seem to have trouble meeting the market due to behavior, chasing returns and asset classes, and excessive trading. I think that patience may result in doing better than average over long periods of time. (like the original S&P 500 companies & their descendants from 1957 beat the S&P 500 index though 2003 or that Corporate Leaders Trust which bought 30 or so blue chips in 1935 and held on to them and descendants without selling )

        I just try to look at it from a position of seeing what has worked consistently, why it may have worked, and go from there. For example historically, quality has done well and so has value ( and dividends too vs non dividends). Of course, we both know that this may not work on a go forward basis. And apparently IPOs have not done very well for investors in aggregate, though a few outliers may tell you otherwise 😉

        Perhaps many mutual funds do poorly because their goal is to attract assets to make money for the fund management business, because of their high fees, high turnover etc. I reviewed a mutual fund a few months ago that essentially lost 100% of investor money over the past 50 years due to high costs and over trading.

        But I am not sure why mutual funds cannot do as well as the journalists from WSJ in selecting companies 😉 Actually, sometimes even the Dow Jones committee makes mistakes – like replacing IBM in 1939, which apparently was before it really prospered as it ushered the age of computing.

        Either way, I would be very interested in getting my hands on the data that the researchers used.

        PS A lot of people assume that I am trying to beat some average when I select my dividend stocks. I actually want to live off my dividend income and not worry too much about stock prices.

        • I actually like the idea of dividend investing, as long as you are being wise about it (as you are) and not picking solely based on yield. I differ from some many Bogleheads in that I do think there is something “different” about a company that can pay out a reliable dividend while still being able to support and grow their existing operations like General Mills, Johnson & Johnson, or Coca-cola.

  3. Really interesting analysis. Since they don’t appear to adjust the dollars in the lifetime wealth creation graph for inflation, though, wouldn’t that bias the results towards companies that have done well more recently?

  4. Jonathan, thank you, this was very interesting and a totally new understanding for me. Did not at all know that the distribution looks like this. As far as showing me something totally new this is probably my favorite post of yours in a couple years.

Speak Your Mind

*