Here’s a paradox I’ve been thinking about lately…
On one hand, “beating the market” consistently is one of the world's most challenging puzzles. On the other, there are hundreds of academic papers that claim to have solved that puzzle.
Study these papers, and you’ll find dozens of strategies that outperform the market over time.
What’s more, savvy exchange-traded fund (ETF) providers have translated these strategies into funds available to investors with the click of a mouse.
These smart beta ETFs bet on factors like momentum, insider buying or the Dividend Aristocrats to beat the market. Each of these strategies is backed by research conducted at the world’s leading investment firms and business schools.
Yet I’ve been disappointed by the real-world performance of smart beta ETFs. It seems that the instant a strategy is introduced through an ETF, it stops working.
Today, I want to explore the reasons behind this phenomenon… and what, if anything, the average investor can do about it.
The Replication Crisis
In his bombshell essay “Why Most Published Research Findings Are False,” Stanford Medicine professor John Ioannidis revealed that the results published in many medical research papers cannot be replicated by other researchers.
That means the latest study you read on vitamin C – and how researchers have “proven” it helps fight the common cold – is far more suspect than it first seems.
This “replication crisis” has upended the medical research applecart.
More relevant to us investors, this issue may have also infected academic research on investing. And researchers’ questionable conclusions may be costing average investors billions in lost profits…
Ioannidis’ financial counterpart is Campbell Harvey, a professor of finance at Duke University.
Harvey estimates that at least half of the 400 “market-beating” strategies identified in top financial journals over the years are worthless.
He challenges academics to take any so-called winning strategy and ask a different set of researchers to replicate it. And chances are about 50-50 that they can’t.
Even worse, Harvey argues that his fellow academics are in complete denial about the problem.
And he is unusually well qualified to make that judgment. As the former editor of The Journal of Finance, Harvey has written more than 150 papers on finance.
As the Financial Times recently explained, this is not like a child saying the emperor has no clothes. “Harvey’s escalating criticism… is more akin to the emperor regretfully proclaiming his own nudity.”
Lies, Damned Lies and Statistics
Mark Twain – who popularized the saying “Lies, damned lies and statistics” – would share Harvey’s skepticism.
Indulge me a quick academic point here.
In statistics, a p-value represents the probability that a finding is statistically significant – attributable to an actual factor and not pure chance. For example, it will show whether a particular drug works or whether value stocks outperform over time.
The problem is this: Researchers twist the data – blatantly or subconsciously. They may cherry-pick the metrics used or adjust the time period studied to obtain a statistically significant result.
We can blame “the system” for this problem.
Young finance professors can publish a paper with an eye-catching find in a prestigious journal – and they just might get tenure. As a result, investment strategies that look terrific on paper often flop in the real world.
What’s an Average Investor to Do?
I was impressed by the market-beating strategies offered by smart beta ETFs when they first hit the market. But I, too, have been disappointed by their real-world performance.
As an average investor, you have two choices.
First, you can stick to investing in a cheap ETF that tracks a major market index.
Most investors fail to beat the market over time. So by tracking an index, you’re guaranteed to do better than the average investor.
Second, ignore the promises offered by smart beta ETFs. Instead, shift your focus to investing in the very best companies.
Warren Buffett never calculated a p-value before investing in Coca-Cola, Apple or Costco.
You shouldn’t either.