U.S. Edition
Search Ticker
Matt Hougan

Morningstar Gets ETFs All Wrong

Related ETFs: FXI | PGJ | GXC | FNI | SPY
Share:

That point was hammered home yesterday when Morningstar published its list of "The Worst ETFs of 2007." That's a fun topic. With more than 200 ETFs launched last year, there are more than a few you could pick on.

But the ones Morningstar chooses make no sense at all. In fact, they showcase all the reasons why Morningstar gets ETFs wrong.

Take the very first funds on Morningstar's list: the SPDR S&P China ETF (GXC) and the First Trust ISE ChIndia ETF (FNI). "These markets have been smoking in recent years," writes Jeffrey Ptak of Morningstar. "Not surprisingly, they look rich to our eyes."

That may be true, but it's not the point. ETFs are best used in long-term asset allocation strategies, not for chasing hot trends. The way to evaluate ETFs is to choose the best one within each asset category, not to try to game the market by choosing the "right" asset categories.

In this case, the SPDR S&P China ETF (GXC) was actually one of the best ETFs launched last year. Compared with other China ETFs, such as the popular iShares FTSE/Xinhua 25 ETF (FXI), GXC offers a more diversified portfolio and broader exposure to the Chinese market. For starters, it has 150 components compared with FXI's 25. FXI also has more than 25% of its portfolio in its three largest names, while GXC takes a more diversified approach to the market. The third China ETF—the PowerShares Golden Dragon Halter USX China ETF (PGJ)—is too strongly underweight the Financials sector for my liking.

For investors looking to add China to their portfolios, GXC may actually be the best choice. 

Will China go up or down in 2008? I have no idea. But that doesn't make GXC a "bad" ETF, any more than the S&P 500 falling next year would make SPY a "bad" ETF.

It is the same basic flaw that dooms the Morningstar STAR rating system: a focus on short-term performance rather than long-term, sound portfolio construction. That may work for active funds (although the data suggest otherwise), but it certainly doesn't work for ETFs.