Several weeks ago I finally did something I’d been putting off for years: I logged into my workplace pension, took stock of the mess it had become and rebalanced it properly.
My pension had been neglected for ages. I’d made some rushed changes over the phone a few years ago – no real strategy, just listed some funds to buy – then work, family and a host of other excuses meant I never got time to jump back in and properly sort it out.
Because of this I had more than one multi-asset fund doing the same job, tiny allocations that were too small to contribute meaningfully, a cautious absolute return fund inside an account with a decades-long time horizon and JPM Natural Resources, which had quietly grown to nearly 30% of the total pot, up around 65% in the year to end of February.
That number matters, because it explains why fixing the problem was harder than it should have been. What made it difficult to act has a name: recency bias.
JPM Natural Resources had done well as commodities had a strong 2025, driven by the surge in gold and silver, energy price pressures, supply constraints and metals demand tied to the energy transition. Recency bias means it is easy to overemphasise recent events and can lead you to think these conditions will continue.
It felt wrong to sell it. But the only question that mattered was whether a 30% allocation to commodities is appropriate for a long-term pension. Forced to ask myself plainly, the answer was clearly no.
So, I trimmed the commodities fund, sold the multi-asset funds and dropped the absolute return strategy. I moved to a more considered allocation: index trackers as the core, overweights to Europe, Japan and emerging markets, some thematic exposure and a meaningful weighting to commodities (but much less than before).
Then, on 28 February, US and Israeli forces launched strikes on Iran. Brent crude surged and European natural gas prices spiked. The markets I’d shifted into sold off. In a few days, the carefully considered rebalance looked, on paper, like it could turn into an expensive mistake.
This introduced a whole new cognitive bias for me to wrestle with. Outcome bias is the tendency to judge the quality of a decision by what happened afterwards rather than by the reasoning at the time.
The strikes on Iran were not likely when I rebalanced. The fact that an unpredictable geopolitical shock subsequently moved markets in an unhelpful direction says nothing about whether the underlying decision was sound. A sound process will sometimes produce poor short-term results, especially when the short term includes a war that few saw coming.
So how do you hold firm when circumstances seem to be arguing against you?
The most useful thing is to return to your original thesis and test whether it has actually changed.
My reasoning was that a 30% accidental allocation to commodities was too large for a long-horizon pension and that Europe, Japan and emerging markets offered better value than US equities. The Iran conflict hasn’t altered any of that.
Energy prices are up sharply, but geopolitical oil shocks tend to fade faster than markets initially price in. If the thesis hasn’t changed, the conclusion shouldn’t either.
Closely related is the discipline of separating process from outcome.
Knowing only what I knew at the time, was this a reasonable decision? If the answer is yes, a bad short-term outcome is not evidence of a mistake. It is evidence that markets are unpredictable in the short term, which is not a new finding.
It also helps to distinguish between volatility and structural change.
Short-term price moves in response to a geopolitical shock are not the same thing as a permanent shift in the investment landscape. What would actually warrant revisiting the rebalance is a sustained closure of the Strait of Hormuz pushing oil structurally above $100, or a prolonged conflict permanently repricing risk across Asian and European markets. Neither has happened. Until it does, short-term price moves are noise.
Finally, and most importantly in this context: time horizon.
A workplace pension with decades to run is almost uniquely well-suited to ignoring short-term volatility. The discomfort of the past week is itself a reminder of why the rebalance was necessary. A 30% concentration in any single sector is a meaningful risk in a portfolio you cannot afford to get badly wrong.
Recency bias made it hard to act and outcome bias is now trying to convince me I was wrong to. The response to both is the same: go back to first principles, ask whether the thesis has changed and, if it hasn’t, stay the course.
What would make me reconsider?
A prolonged conflict materially disrupting global energy supply, persistent inflation forcing central banks to abandon rate cuts or clear evidence that the structural case for European and emerging market equities has deteriorated.
Those are the things worth watching. A bad week is not on the list.
Gary Jackson is head of editorial at FE fundinfo. The views expressed above should not be taken as investment advice.