How long is the long run?
Small-cap investors everywhere and value investors in the US (though not in Europe or the UK) have been suffering persistent underperformance for quite a while now. The mantra of value and small-cap fund managers anywhere is that ‘in the long run’, value stocks will outperform and small-cap stocks outperform large caps. But how long is the long run?
Over the last (almost) 100 years, since 1926, US small-cap stocks have outperformed large-cap stocks by 2.85% per year. So, in the long run, they did outperform. But, as Edward McQuarrie emphasises, these long-term averages disguise extremely long periods of underperformance that can easily last a decade or two.
Just look at the split of US small-cap vs. large-cap performance by 20-year periods below. While small-cap stocks outperformed for every 20 years between the 1930s and 1970s, they underperformed significantly in the last 20 years of the 20th century and again since 2018. Indeed, if you had invested in US small caps over the last 40 years, you would barely have managed to keep pace with large-cap stocks.
Performance difference between US small caps and large caps
Source: McQuarrie (2025).
This is why I am always sceptical when I see people quote the Fama-French data of factor performance since 1926. None of us has been an investor since 1926, and even for endowments and pension funds, 100 years is an awful long time to wait until your thesis comes true.
It reminds me of a quote by a computer science professor of mine (admittedly, it was a quote from the 1990s, but bear with me): “In practice, it doesn’t matter if a computer can never solve a problem or if you have to wait five minutes in front of the screen”.
Similarly, if an investor has to wait ten years or more for a factor like value or size to work, it might as well never ever work because by that time, the vast majority of investors are no longer invested in it.
And this is the key message of McQuarry’s note. Just like in his superb exposé on long-term performance differences between stocks and bonds (see here, here, and here), he shows how long ‘the long run’ can be in practice. To paraphrase Keynes: Factors can stop working for longer than you can stay solvent.
Indeed, one fun chart in his new paper is that academic results themselves can be subject to false identification. Take a look at the chart below that shows the performance of US large-cap stocks with high valuation vs. US small-cap value. The initial sample Fama and French used to identify size and value as outperforming factors coincides with a relatively unique period of the last 100 years, when expensive large-cap stocks in the US underperformed significantly. In the previous 35 years, since the publication of their study, expensive large-cap stocks in the US have almost continuously outperformed small-cap value stocks over any ten-year investment horizon you choose.
Which brings me to a provocative question: If we didn’t know anything about factor investing today and we only looked at this for the first time, would we still conclude that value stocks outperform in the long run? Would we conclude that small-cap stocks outperform in the long run? What about large, expensive stocks vs. small value stocks?
McQuarrie provides his answers in the note. So, check it out if you are interested, but before you do, think about it yourself first. Given the data shown here, what is your conclusion?
Rolling 10-year performance of large overvalued stocks in the US vs. small-cap value stocks
Source: McQuarrie (2025).
In my view, value is about asymmetry in information. In markets where information is difficult to collect, value persists. In US equities, there’s almost no value left because of the rapid transmission of information.
Really good post. I suppose you could argue you get paid for the risk/mispricing because the timing and magnitude of the payoff is so irregular and unpredictable as to exhaust patience.