Buying stocks is easy – selling them is the difficult bit

A better selling process for most managers

We know that most fund managers fail to beat their underlying benchmarks consistently over the long run, partly because their fees are too high. Any number of studies prove this point. However, this isn't necessarily because managers are bad stock pickers other papers note that managers' "best ideas" tend to beat the market.

So what's the problem? An interesting new study might have the answer. Selling Fast and Buying Slow: Heuristics and Trading Performance of Institutional Investors, a 2018 paper from academics and researchers Klakow Akepanidtaworn, Rick Di Mascio, Alex Imas and Lawrence Schmidt, suggests asset managers are indeed generally good stock pickers. The authors looked at more than 750 portfolios between 2000 and 2016. On average, new buys beat both the market and a "counterfactual portfolio" where the researchers randomly added the money to an existing holding instead.

The problem arose when the managers decided to sell. Where a manager sold in a period around company news being released in other words, when there was a fundamental reason to sell their performance did not suffer. However, the same couldn't be said for other types of sale. Indeed, say the researchers, in most cases the average manager would have been better off by up to two percentage points a year by selling a stock completely at random.

So what's going on? The researchers conclude that it's all about attention or a lack of it. Managers do their research and consider all the angles when they are buying. But when it comes to selling, they take shortcuts (or use "heuristics", in the behavioural jargon). One reason for this is that the selling decision often comes about not because the fundamentals of the stock have deteriorated, but because the manager needs to raise money. And when a manager is trying to pick a share to sell on that basis, he or she tends to be drawn to those that have either lost a lot of money or gained a lot while sitting in their portfolio. In effect, when managers sell without thinking, they cut their winning stocks too early, or end up crystallising losses that they should have trimmed a lot earlier.

This has implications for your own selling process (in short, consider it as carefully as your buying process). But it also suggests a few tips if you are looking for a good fund manager. First, look for low portfolio turnover the less selling a manager does, the fewer mistakes he or she can make. Secondly, look for a relatively small number of holdings. If managers are good at buying, you want maximum exposure to their best ideas rather than a portfolio full of also-rans that are just there to make up the numbers.