Where the value investing strategy still works
Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.
The writer is founder and chief investment officer of Verdad Advisers
Trailing performance has not exactly been a selling point for value investment strategies in recent memory. As the last 15 years unfolded, a wave of bad performance has seen allocators shift remorselessly away from the strategy of investing primarily at undervalued companies to more growth-orientated approaches.
This has hollowed out the value investing community. As far as I can tell, the universe of value managers has shrunk to the aged whose records pre-2009 are strong enough to keep investors loyal, quants who trust the long-term data and cranks who are weirdly obsessed with natural resources stocks.
The recovery from the Covid pandemic now seems like a false dawn for the value community in the US, as artificial intelligence-mania swept the market in 2023 restoring the fortunes of the briefly embarrassed growth managers and punishing the briefly optimistic value managers.
The value “factor”, which measures the returns of going long cheap stocks and short expensive stocks, suffered a 50 per cent retreat in the US from 2009 to the depths of Covid in 2020, as figures from the data library of Dartmouth professor Kenneth French show. It then recovered sharply through the end of 2022 before falling off again, starting almost exactly when ChatGPT was released in November 2022. Since then the value factor is down 11 per cent in the US. This has not been a pretty ride for advocates of value investing.
But there’s a myopia to focusing so much on the recent experience of the US value investor community. This is because the story looks strikingly different internationally.
Value suffered a similar swoon as in the US, suffering a 30 per cent drawdown through the depths of Covid. But since March 2020, its performance in non-US developed markets has been a steady, almost linear rise. International value has even had strong performance during the ChatGPT era. The outperformance since March 2020 has been so good that the value factor now shows outperformance over three-year, five-year, 10-year and 15-year periods. The value factor is up 36 per cent in developed ex-US markets.
So, while many assumed value was dead, it turned out it had just got on an aircraft and headed for Europe and Japan.
Any arguments for the death of value — and the abandonment of it in the US — have to grapple with this strong performance internationally. If value was a statistical anomaly never worth betting on, why has it paid to bet on it internationally?
If something structural has shifted in the market to make value not work, why has that only affected the US? If quantitative investing is too simplistic and just buying the cheapest stocks is a dumb strategy, why have the idiots who practice this approach outperformed in Europe and Japan?
I would argue that the best interpretation of this data is to see the underperformance of value in the US as historically contingent, dependent on a specific set of events occurring. And that set of events has been major technological innovations first in cloud computing and most recently in AI.
These allow companies with the best technology to grow to global scale with near zero marginal costs, producing winners that can increase profits at a rate and a scale that seemed impossible in previous decades.
But history has shown that markets tend to adapt to technological shifts. Competitors learn the new technology and drive down prices. Businesses in other industries adopt the new technology and use it to improve their own operations.
We can imagine how this might work with AI: the initial spoils go to the first company to invent a workable AI system (say OpenAI), but over the next three to five years, competitors emerge that can produce the same technology. So the innovator’s pricing and margins drop. And then, maybe, a group of boring old-world companies with big customer service departments all adopt AI technology and dramatically reduce headcount and improve margins.
While the initial equity returns would flow to the innovator, the longer-term benefits might accrue to the customers of that technology. Economic theory would hold that at least customers need to derive a return on investment over and above what they spend on the new technology.
Even if the sun seems to have fallen on US value investors, the dawn has emerged internationally and, theory and evidence would suggest, the sun will rise again, perhaps sooner than expected, in the US as well.
Comments