It’s an absolute certainty that anytime a portion of the market performs well, a new fund is created for it. It’s like clockwork.
There will always be a market for this stuff because the past returns look great and people love a good history of returns (no matter how short). Of course, I could argue it’s a contrarian indicator to stay away because the good returns were already made.
Now, ETFs make it even easier for the industry to mass produce these fad funds to take advantage of the latest hype. Case in point is the latest ETF creation for the FANGs – Facebook, Amazon, Netflix, Google, along with a few similar high performers.
All these new funds do is add to the confusion in fund land.
The rise of index funds and ETFs made the process of deciding what funds to own somewhat easier. The low cost and the average returns are a suitable choice for most investors. But because it’s become the suitable choice, the industry can’t make enough of them.
Two years ago there were 49 different S&P 500 index funds. I have no idea how many exist today, but that’s 47 more than necessary (one ETF and one index fund are all that’s needed). And that doesn’t include the hundreds of actively managed funds that just track the index, but some of that is changing.
Right on schedule, the next reinvention is driving an abundance of smart beta or factor-based index funds. It’s not much different than the old active – somewhat higher costs with the same promise of higher returns.
But that’s the point really. When the industry can’t make money one way, it pivots.
However, the latest pivot should be a reason for you to pause:
A more full-throated critique came recently from Antti Suhonen of Aalto University in Finland. In a recent Journal of Portfolio Management paper titled “Quantifying Backtest Overfitting in Alternative Beta Strategies,” Suhonen reviewed the “daily returns of 215 alternative beta strategies across five asset classes, eleven identifiable strategy groups, and fifteen sponsor investment banks.”
The results, he wrote, “strongly support the existing literature on selection and publication biases and backtest overfitting.” In other words, an index designed based on past results didn’t result in the same sort of outperformance once the strategy was deployed in the real world.
Michael Batnick summed this up perfectly: “The worst ten-year period for any backtest is the next ten years.”
The reality is there are so many reported “factors” it’s hard to separate what actually performs well from those that just test well. There also seems to be a lot of confusion around whether a backtest proves a new “factor” or was just leeching off one or more of the time-tested factors we already know exist – value, momentum, etc.
Is it a false positive or the real deal?
It would seem that the old fund disclaimer is due for a makeover:
Past Backtested performance may not be indicative of future real results.
- For Active Funds, It’s Time to Adapt (or Die) – Morningstar
- Passive Investing Might Not Be Great for Growth – N. Smith
- Every Great Investment Hurts – M. Housel
- Why Simple Beats Complex – A Wealth of Common Sense
- The Dark Side of Globalization: An Update on Country Risk! – Musings on Markets
- What Is a “Good” Decision? – Psychology Today
- How Economics Became a Religion – Guardian
- Capitalism the Apple Way vs. Capitalism the Google Way – The Atlantic
- AI and ‘Enormous Data’ Could Make Tech Giants Harder to Topple – Wired
- Zero — Invented or Discovered? – Farnam Street