The Morningstar Mirage: What those fund ratings really mean

[OldShooter]

Interesting WSJ article this morning. Big analysis that says that Morningstar stars are not at all predictive. That's no real news to anyone who has been paying attention to the S&P Manager Persistance report cards, but certainly it confirms what S&P has been reporting for years.

The article is behind the WSJ paywall, but you might be able to google yourself to a copy you can read.

 

[Totoro]

Isn't this also what Morningstar itself also said? The star rating was in my mind always an indication of past performance, nothing more.

I'd be more interested in the correlation between their fair value assessments (paid subscription) and subsequent stock returns.

 

[audreyh1]

Yep star rating is not predictive. M* has great information, but I think when I bought a highly rated fund half the time I was buying something that was already past better days.

 

[OldShooter]

Isn't this also what Morningstar itself also said? The star rating was in my mind always an indication of past performance, nothing more. ...

You're right, although I think they have not exactly been shouting it from the rooftops. In the article they say that they have never claimed that the stars were predictive except maybe a little bit. But I'll bet that at least half of the investing public doesn't understand this. Morningstar is in kind of a conflict of interest. If they emphasize the truth too much they lose revenue.

The other thing people don't realize is that in any category, no matter how abysmal the performance has been, the best 5% of funds get five stars. 5% get one star, and the other three steps are allocated similarly. So it might be that a one-star fund in a hot category outperformed a 5-star fund in a losing category.

I'd be more interested in the correlation between their fair value assessments (paid subscription) and subsequent stock returns.

Not me. Analysts are just a bunch of chattering monkeys. Random chance determines which monkey looks like a genius on any given day.

No one who can actually make accurate predictions is going to be putting on a suit and going to work at Morningstar or any other investment firm.

 

[OldShooter]

Yep star rating is not predictive. M* has great information, but I think when I bought a highly rated fund half the time I was buying something that was already past better days.

Actually, probably not. The idea of "better days" and trends implies that significant momentum effects exist. The S&P Manager Persistence reports pretty much conclude that fund performance is random. Hence, the star rating is completely irrelevant as a predictor.