Over-trading is one of the most commonly reviled sins in investing and is often cited as a reason that investors should avoid active fund management. It’s a crime of which I’ve often worried I might be guilty, so I’ve decided to have a direct look at it.

The reason over-trading is bad is that it results in additional costs from dealing fees, stamp duty and most relevantly for me the spread (as I like to trade in small-cap stocks). Over several trades these costs start to add up. Unless you’re getting a better return from what you buy than what you sell, then trading is going to cost you a significant sum of money.

To look at whether I’m over-trading my first port of call is to calculate my portfolio turnover. This is the total value of portfolio sales over the past year divided by the total value of the portfolio. I’ve worked out that my portfolio turnover is 125%. That means that I’ve bought and sold shares worth more than my whole portfolio over the last year or that my average holding period is just under 10 months. Gulp, I thought it would be high but that’s worse than I thought!

By way of contrast, most of the investors I follow have extremely low portfolio turnover and pride themselves on it. For example, Woodford income had about 15% turnover in 2015 and Fundsmith turnover is close to zero.

So why am I trading so much and how badly is it harming my returns?

Trading costs for me (selling and buying again) tend to be around 2-10% depending on the size of the spread. This means that my high turnover might be costing me between 2.5% to 12.5% of my portfolio returns a year. This is a lot especially when compounded over time. I need to see whether my trades are creating sufficient value to compensate for this.

All in all I’ve sold about 50 shares since this time last year, though some of these were because of acquisitions (Hydro International and Avesco) and some were because I was topslicing a holding when it had got too large (Avesco, Micro Focus, Boohoo (twice), Burford, Brainjuicer (now System 1) and On the Beach).

To judge my sell decisions, I’ve considered what return the share I sold would have made since then had I continued to hold. It’s important to account for the opportunity cost of a holding i.e. that by selling it you have the opportunity to invest the proceeds into something else. I’ve therefore compared the return of the sold shares against the return of the rest of my portfolio since the sale. I’ve looked at sales made before the end of 2016 so sufficient time has elapsed since the sale to be able to judge whether it was a mistake.

I made 29 sales between 20 April and the end of 2016, excluding the acquisitions and the topsliced shares. Only three of the shares I sold performed better than the rest of my portfolio after I sold them. Only four delivered a return within 10% of the portfolio. Most did considerably worse. That’s a better performance than I thought! It suggests my sell decisions have been pretty good and certainly I’m a lot better off having made these trades (taking account of the cost).

On the other hand, most of the shares that were topsliced went on to perform somewhat better than the rest of my portfolio after I sold them. Given the reason I topsliced was to preserve diversity and reduce risk, this isn’t necessarily a bad thing but it is costing me some return. Perhaps I might be better served by a more concentrated portfolio?

The fact remains that turnover is high and costing me. This isn’t because my sell decisions are bad but because some poor performing shares are being bought in the first place. This might be a somewhat inevitable feature of a strategy based on momentum that holds winners and sells losers. Nevertheless, one can always do better. I still feel I should aim to reduce portfolio turnover and the ways for me to do this is to refine the buy decisions or to topslice less and have a more concentrated portfolio.