Chasing last year's best fund is a reliable way to lag

Everything below about performance chasing comes from what actually happens rather than from what is supposed to.

Persistence is weak

Studies tracking top-quartile funds into subsequent periods consistently find limited persistence. Some of what looks like skill is sector or style exposure that happened to be in favour. When that style falls out of favour, the same fund appears in the bottom quartile.

The direction of this result is consistent across studies while its magnitude is disputed, partly because the answer depends heavily on how peer groups are defined and how closed funds are handled.

Buying high by construction

A fund appears at the top of a table after it has already performed, which means you buy after the gain. If the driver was a style cycle, you are buying near the point of maximum enthusiasm.

Put simply, this is the mechanism behind the behaviour gap rather than bad luck. Money follows the tables with a lag, so the bulk of it arrives after the performance that attracted it, and the average holder therefore experiences less than the fund itself reported.

Switching has direct costs

Dealing costs, spreads, time out of the market during a transfer and potential tax in a taxable account all apply. Frequent switching accumulates these without any compensating advantage. Each switch also restarts the psychological clock, making the next disappointment arrive sooner.

Transfers between providers can leave the money uninvested for days or weeks, which is a risk running in both directions and one nobody is compensated for taking.

What would justify a change

A change in the fund's mandate, a substantial rise in charges, a change in the index tracked, or the fund no longer fitting your plan. Underperformance against a peer over a short period is not on that list. Writing the criteria down in advance prevents post-hoc justification.

For a tracker, persistent drift from the index it follows belongs on the list too, and it shows up as tracking difference over several years rather than in any single period.

Boredom is the actual enemy

A well-constructed portfolio is uneventful, and uneventful feels like inattention. The urge to act is strongest when nothing needs doing.

Directing that energy at contribution rate rather than holdings is where it produces a return. If the urge persists, a small ring-fenced portion kept separate does less harm than repeatedly rearranging the main portfolio, provided the amount is one whose loss would change no plan you have.

Adjust the size of it until it is something you would actually do tired.

Where the table itself misleads

League tables rank funds inside a category, and a fund can reach the top of one by drifting outside it, which is a change of exposure rather than evidence of skill. A one-year table is dominated by whichever style was in favour, and the same table over ten years is dominated by which funds survived long enough to appear on it.

Currency movements can lift or drop a fund several places without anything happening to the underlying holdings, particularly where your own currency differs from the one the fund reports in. The column with any established relationship to what happens next is the charge, which is usually the smallest print on the page.

The takeaway

Write your sell criteria in advance. Nothing on that list will be "it had a bad year".

The version you keep doing is the version that works.