Mutual Fund Fees...Where To Begin??

John Bogle, the grand daddy of mutual fund indexing wrote a very nice Op-Ed piece the other day.  If you know who he is, there's really nothing new in the piece: Management fees directly impact (negatively) the performance of mutual funds, and even Morningstar now confesses that fees better predict performance than their own star system.

Today, Neil Hennessy of the Hennessy family of mutual funds writes in response that Bogle is wrong.  There's a shocker.  The guy who founded the only not-for-profit mutual fund company says fees are bad and the guy running the for profit says he's wrong.  That's a good starting point, but there's more.

Hennessy's first point of insight is that rhetorical device known as the "irrelevant analogy":

We have all learned the lesson that cheaper is not always better. Would you choose the doctor with the cheapest rates and highest number of patients, who then has a larger base to spread costs among?

(Frankly, the medical analogy is pretty good for investing: You're buying a product you have no ability to evaluate.  I've faced many blank stares from friends telling me about their "great doctor" and I ask them for data to prove their doctor is great...but that's a different topic...)

Next Hennessy turns to a laundry list of other expenses investors face.  He's right.  You pay all kinds of expenses you shouldn't.  He misses half of them.  The problem is: Bogle isn't wrong!  Low expenses predict performance.

I'd bet low management fees predict low "other expenses".  What do I mean?  You can bet Vanguard's board fees are lower than Hennessy's.  I'd also bet Vanguard's liability insurance costs less.  Ditto lawyers.  Why?  Vanguard manages mutual funds for the benefit of mutual fund shareholders, not mutual fund managers.  Seems pretty logical that if you manage the fund to benefit the shareholders, then your expenses of covering your own conflicted backside drop.

Hennessy's last argument stands tall as one of my favorites.  Only performance matters, and reported performance takes out fees, so who cares?  Again, he misses the point: Performance isn't predictable, but fees are.