Stop Screening Funds This Way

“Whenever I’m about to do something, I think, ‘Would an idiot do that?’ And if they would, I do not do that thing.”—Dwight Schrute, The Office.

Sometimes, I think I’m too subtle. I don’t want to oversell a study, so I’ll write that the data suggests this or that this measure tends not to help in fund selection. So, if you read from beginning to end, there should be a clear takeaway, but not everyone does that.

I’ve looked at the predictive power of different time periods and found that longer periods are somewhat useful and shorter ones are not. There are other fundamental data points more useful than performance, but if you are using performance, longer term over the manager’s tenure is definitely better.

And yet, I have a friend, let’s call him Bo, who often incorporates a year-to-date return screen into a search for the best funds in a Morningstar Category. Bo clearly doesn’t follow the advice of noted investor Dwight Schrute. Only an idiot would use year-to-date performance. Short-term results are random—they don’t reflect a manager’s skill. Taken to extremes when returns are really dramatic, you are not just introducing noise but actually doing harm because markets tend to snap back in a way that corrects big moves. In other words, the fund’s holdings get overpriced, and you’re essentially buying high when you screen on year-to-date performance.

My colleague Jeff Ptak looked at the very rare club of funds that put up 100% returns in a single calendar year, and the results after that spectacular year were downright ugly. Only a handful even managed to have positive returns over the ensuing three-year periods.

Interestingly, Jeff also found that five-year before-fee returns were surprisingly good as predictors. As Jeff notes, this is something worthy of further study, as most studies find only slight predictive value.

There’s another thing that you should avoid when screening on funds: Ranking by yield. Moving up in yield takes higher risk, and getting one of the highest yields in a category generally means taking on a lot of risk. Aiming for an above-average yield in a low-cost fund is fine, but watch out for the more extreme version.

One way funds boost yield is by using leverage. That’s pretty simple, but leverage has costs, and it raises risks when markets sell off.

In fixed income, higher yield generally comes from more credit risk, more interest rate risk, more liquidity risk, or a combination of these risks.

A skilled manager with a good analyst team can find the risks worth taking. However, a less skilled manager can produce a big yield simply by taking on a whole lot of risk. And in bond funds, risks might only show up every few years.

In equities, companies with decent yields tend to be slower growers but have solid cash flows to support their dividends. However, the highest yields imply that they aren’t sustainable. Investors dump shares of stocks on shaky footing that might run out of cash. That makes the stock’s yield higher, but it might soon have to cut dividends or, in extreme cases, go belly up.

Among exchange-traded funds, we’ve spotted a few mechanical indexes and strategic-beta funds that weight stocks by yield, and it means they’ve got a glass jaw. SPDR S&P International Dividend ETF DWX does this. Other funds will limit themselves to just the 50 highest-yielding stocks and equal-weight them. That’s almost as bad. Global X SuperDividend US ETF DIV and Global X SuperDividend ETF SDIV do this.

Conclusion

I understand why people screen on short-term returns and big yields, because those are the results they want. But this is just looking in the rearview mirror. You don’t get to capture those past results. You may be buying at the top.