Barry Ritholtz has a quite compelling piece in the Washington Post - the title is self explanatory on the topic. Frankly it could be applied to any money manager for most of the items (excluding too big which would be an issue for hedge or mutual funds but not individual accounts), but I thought it was a useful list. Link to the longer piece here - but here are the bullet points. Interestingly with his last two points he mentions why performance (at least in the short term) is not one of his reasons.
While there are many reasons to hire a fund manager, there are just as many reasons to fire one. Here are my main criteria:
When they suffer from style drift: This happens quite often; a manager developed an expertise in a given area but is looking beyond that. Maybe they got bored. Maybe the new cow in the pasture caught the bull’s eye. Whatever it is, they are doing less and less of why you bought them in the first place. That’s your signal that it’s time to move on.
When they fight the dominant market trend: Bill Miller’s market-beating 15-year streak came to an end amid a value trap. He bought more and more of his favorite holdings — banks, GSEs and investment houses — right into the financial collapse. Doubling down again (and again) is not a valid investment strategy. Whatever advantages he had heading into 2008 disappeared.
When they seem to lose their edge: Whether it’s success or money or a loss of interest, managers sometimes lose the “fire in the belly.” Determining this is admittedly challenging. We often find out about some personal demons — divorce, alcohol, whatever — after the fact. Regardless, when whatever it was that made them a top stock picker starts to fade away, you should also.
When they become a closet indexer: (Mark's note - this is a major pet peeve of mine... so many 401k plans have 8 'large cap' funds whose holdings are all essentially a mimic of the S&P 500. Also why buy your 174th best idea?) When a fund owns 100, 150, 200 names, it effectively becomes a high-cost index. Even if they have the top performing stocks, it will be in such small quantities as to not move the needle. This is an easy fix — you replace them with a low-cost, passive index.
Notice that performance is not a factor in any of the points above. There are two reasons for that:
Process, not outcome: Investors ought to be focused on creating a reproducible methodology, regardless of luck or misfortune in any given quarter. Investing is a probabilistic exercise, and performance can slip for a quarter or two — even when the manager is doing everything right.
Mean reversion: The opposite of chasing performance (and buying high) is dumping a weak quarter (selling low) that then snaps back.