Last updated 2022-02-10.
Many investors, even professionals, are ignorant about obvious facts that they really should know. When I say a fact is “obvious”, what I mean is that you can easily observe it by looking at widely-available data using simple statistical tools.
A list of obvious but underappreciated facts:
Fact 1. A single large-cap stock is about 2x as volatile as the total stock market. A small-cap stock is about 4x as volatile. It’s common knowledge that individual stocks are risky, but most people don’t know how to quantify the risk, and I believe they tend to underestimate it. I wrote a whole essay about this because I think it’s the most important underappreciated investing fact.
Fact 2. A basket of 50 randomly-chosen stocks isn’t much more volatile than the total stock market. This is kind of the opposite of fact 1. According to my backtest,1 the total US stock market had an annual standard deviation of 17.3%, whereas a randomly-chosen selection of 50 stocks had a standard deviation of 20.2%.2 That’s higher, but not by much.
Baskets of 30 stocks had standard deviations of 21.7%; 10 stocks had 25.5%; 5 stocks had 31.9%.
(For more on this, see Elton & Gruber (1977), Risk Reduction and Portfolio Size: An Analytic Solution.)
My anecdote about this: When I was considering opening a donor-advised fund (DAF), I had phone calls with a few investment advisors who I thought might do a good job of managing it. I remember one advisor in particular who I liked at first because she said she was a fan of value and momentum investing. I asked her if she could invest my DAF in a concentrated portfolio of about 60 value stocks, like I describe in this essay. She declined, saying that a 60 stock portfolio would be far riskier than the broad stock market, and without any extra expected return to compensate. This is obviously false if you simply look at the data. A randomly-chosen collection of 60 stocks is a little riskier than the broad market, but not much riskier. And it’s true that there’s no reason to take on risk if there’s no benefit, but if you’re specifically investing in value and momentum strategies, then historically, concentration risk also brought higher return, as I showed previously.
Fact 3. Stocks can underperform bonds for 30+ years at a time. Investment manager Meb Faber wrote an article about this, inspired by two amusing Twitter polls:
What stretch of underperformance by stocks vs. bonds would you be willing to tolerate before selling your stock allocation?— Meb Faber (@MebFaber) February 6, 2020
Would you be willing to invest in an asset that historically outperforms bonds by a few percentage points per year, but has, multiple times, generated zero outperformance for stretches lasting over 30 years?— Meb Faber (@MebFaber) February 6, 2020
53% of investors said they would not tolerate 10 years of stocks underperforming bonds. What did they do from 2000 to 2010, when US stocks performed worse than US bonds? (Bonds have beaten stocks many times, but 2000 to 2010 is such a recent decade that you’d think people would remember it.) Fully 76% of respondents said they wouldn’t tolerate 30 years of underperformance. So either they don’t invest in stocks at all (unlikely), or they don’t know that stocks have in fact underperformed bonds for 30 years on more than one occasion.
Fact 4. Bonds are pretty risky after adjusting for inflation. US short-term T-bills—supposedly the safest investment around—have experienced a historical drawdown of –49% after inflation. On another fun Twitter poll by Meb Faber, only 36% of respondents guessed that the drawdown was worse than –45%. (Meb Faber’s Twitter is a great source of underappreciated investing facts.)
Fact 5. The last 3–5 years of performance (of a fund or asset class) tell you almost nothing about future performance. For instance, see Hoffstein (2016), Are 3-year track records meaningful?
And yet, 89% of managers would replace an investment strategy that has underperformed for 3 years. And when investment managers switch strategies after a period of poor performance, the new strategy tends to perform worse than the old one would have3.
Fact 6. Buybacks are financially equivalent to dividends. Many investors like dividends but dislike buybacks. But these are just two different ways of returning capital to shareholders. In an efficient market, if a company buys back 2% of its shares and then you sell 2% of your holdings, that’s exactly the same as if the company paid a 2% dividend. Conversely, if the company pays you a 2% dividend and you use that money to buy more shares, that’s exactly the same as if the company did a 2% buyback. These scenarios are slightly different for tax purposes, but in financial terms, they’re identical.
Sort-of-obvious but controversial facts
These facts in this section are less obviously true, either because they come from theoretical models that don’t necessarily hold in practice, or they rely on mixed empirical data. But I believe they’re still underappreciated, and it’s somewhat surprising just how underappreciated they are among investment professionals.
Fact 6. Actively-managed mutual funds are just as risky as index funds. I don’t know if this is a common misconception among ordinary investors, but I frequently hear investment professionals claim that actively-managed funds are less risky than index funds, or that they perform better in market drawdowns. The evidence on this is mixed, but generally doesn’t look promising for the actively-managed funds (e.g., see Vanguard (2018), specifically Myth #5).4
I could be wrong about this one, because the evidence isn’t great either way. But the mutual fund managers who confidently claim that they can reduce risk are definitely not standing on solid ground.
“It is difficult to get a man to understand something, when his salary depends on his not understanding it.” Most people know by now that active mutual funds don’t outperform index funds. So in an attempt to justify their existence, investment managers claim that they provide value by reducing risk. This is (probably) false, but it’s not as well-known that it’s false, so they think they can get away with it.
Maybe this fact isn’t exactly obvious—figuring it out required inventing new math. But it was proven over half a century ago, and the man who proved it went on to win a Nobel Prize in economics, in large part thanks to precisely this result. You’d think it would be more well-known by now.
A weaker version of this fact: if you want international exposure, buy a world stock market fund. Don’t buy a US market fund and then claim you’re getting international exposure because some US companies sell internationally. I’ve seen many investing professionals recommend the latter (including Jack Bogle). History shows that an international stock index is not perfectly correlated to a US index, so you can trivially get better diversification by buying internationally.
Fact 8. In theory, the best way to increase expected return is by using leverage, not by increasing the ratio of stocks to bonds. This isn’t an empirical fact, but it easily follows from CAPM, the standard theory of investment risk. CAPM makes some assumptions that don’t hold in practice. But under certain real-life conditions, it does indeed make sense to use leverage than rather than overweight risky assets. A few sophisticated investors, such as Bridgewater Associates, abide by this principle, but the vast majority of investors don’t.
(Bridgewater pioneered risk parity investing, where you allocate to each asset class such that each asset class exposes you to equal risk, and then add leverage. Risk parity is theoretically optimal when every asset class has the same risk-adjusted return.)
Fact 9. Gold is not a great inflation hedge.6 Many people claim that gold can hedge inflation, and there are reasonable theoretical reasons to expect it to. But gold is very volatile, and it frequently performs poorly even in times of high inflation. It might still make a good addition to a portfolio, but it doesn’t provide the straightforward inflation protection that a lot of people believe it does.
Fact 10. Art has not outperformed the stock market. Art sellers like to claim that art is a great investment (e.g., 1, 2), and they cite studies showing the returns to classic art. These results are due to selection bias78. You can only measure an artwork’s change in value if it gets sold once and then sold again later. And art only tends to get re-sold when its value goes up, so your sample excludes most of the artwork that depreciated over time.
Obvious facts that are becoming more well-known
20 or 30 years ago, most people didn’t know these facts. but fortunately, they’ve become more widely understood in recent years.
Fact 9. High fees are really bad.
Fact 10. Actively-managed funds generally can’t outperform index funds.
Many people know these facts, and the market for low-fee index funds is growing rapidly. Even so, about 2/3 of investment dollars still reside in expensive actively-managed mutual funds, so we have a long way to go.
Some more facts
AQR has written three papers about investing facts and fictions. Most of AQR’s facts are obvious in the sense that you can easily verify them using empirical data. They’re somewhat less obvious than the facts above in that they’re about more specific types of investments, rather than the general behavior of markets.
Their three papers (PDF warning):
- Fact, Fiction and Momentum Investing (2014)
- Fact, Fiction and Value Investing (2015)
- Fact, Fiction and the Size Effect (2018)
Why do so many investment professionals not know these obvious facts?
I’m not sure. I can see two common themes:
- Investment professionals believe things because they can make more money by believing them, not because they’re true.
- Investment professionals don’t look at empirical data, so regardless of how obvious an empirical fact might be, they will never notice it.
One possible explanation is that when people think they know something, they don’t feel the need to look into it. For example, if you “know” that bonds are safe after adjusting for inflation, why would you look up their historical performance? Even so, that would mean financial professionals don’t make a habit of looking at simple data that’s relevant to their work, which seems bad. Or, like, you’d think this information would propagate. Somebody tweets about an obvious but little-known fact; their followers realize how wrong they were, and tell their colleagues about it; those colleagues tell their other colleagues; and pretty soon, everybody knows. (I must admit that I didn’t know bonds had experienced a –49% drawdown until I read Meb Faber’s tweet, so I’m also guilty of not looking at obvious data sometimes.)
(I do think professional credentials provide a positive signal, at least. In my experience, CFAs are more likely to know these obvious facts.)
Maybe experts in most fields don’t know obvious facts. I don’t know much about most fields, so I can’t say. The two areas that I know a lot about are computer science and investing (and I know more about computer science because I studied it in college). In my experience, there aren’t a lot of obvious computer science misconceptions among programmers/computer scientists (unless I also believe all the common misconceptions).
Maybe investment advisors aren’t the right reference class for “experts”. Maybe the experts are the people publishing academic papers on investing.9 I’d guess that academics in finance know obvious facts more often than investment advisors do.
Sometimes, people don’t know important facts, but they behave correctly anyway. I frequently hear people touting advice such as, “pick an investment allocation and stick with it no matter what.” You might come up with this advice if you knew that stocks can underperform bonds for 30 years at a time, and that 3–5 years of historical performance tells you almost nothing about future performance. And many people follow this advice even without knowing those facts. If they end up doing the right thing, it doesn’t really matter if they know why they’re doing it. On the other hand, almost everyone behaves incorrectly in other ways—overweighting their home country stock, or holding on to concentrated investments for too long.
(This reminds me of Statistical Literacy Among Doctors Now Lower Than Chance: doctors perform depressingly poorly on the Bayes mammogram problem, but when they get a positive test result in real life, they still do the correct thing—rather than jumping to conclusions, they order more tests.)
My analysis took about two hours, which is long enough that someone wouldn’t do it on a whim, but short enough that I still think it’s fair to describe the result as “obvious”. ↩
- Each year from 1973 to 2013, choose 50 stocks uniformly at random.
- Simulate a portfolio that buys the chosen stocks, weighted by market cap.
- Calculate the standard deviation over the portfolio’s annual total return.
- Repeat 100 times and take the arithmetic mean of all the standard deviations.
The standard error was 0.17 percentage points. ↩
Stewart, Neumann, Knittel & Heisler (2009). Absence of Value: An Analysis of Investment Allocation Decisions by Institutional Plan Sponsors. ↩
Ptak (2018), Will Active Stock Funds Save Your Bacon in a Downturn? found that actively-managed funds did tend to outperform in drawdowns, but not by enough (in the author’s opinion) to make up for their overall bad performance.
What I’d really like to see is how much alpha mutual funds have in up or down markets, but I haven’t found any research on that.
Erb, Harvey & Viskanta (2020). Gold, the Golden Constant, COVID-19, ‘Massive Passives’ and Déjà Vu. ↩
Korteweg, Kräussl & Verwijmeren (2013). Does It Pay to Invest in Art? A Selection-Corrected Returns Perspective. ↩
Disclosure: The linked article did not convince me that classic art returns are due to selection bias. Rather, I was confident a priori that the apparent high returns came from selection bias, and this was the first empirical result I found that supported my prior hypothesis. ↩
Successful day traders and quants are experts in some sense, but their work doesn’t actually have much in common with the way most people invest. ↩