Why Emotional Bias Still Drives Most Capital Investment Decisions

Most companies would say their capex decisions are fact-based. Most of them are wrong. Not because anyone is acting in bad faith, but because a few well-known biases are built into how capital investment decisions typically get evaluated, and they’re hard to see from inside the process.

The first step toward removing them isn’t a new policy. It’s recognising that the flaw is probably already sitting somewhere in how requests get evaluated today.

Where the bias actually hides

Sunk cost protection. A project that’s already absorbed two years of budget and internal political capital is harder to kill than a new one, even when the numbers say it should be. The question “should we keep funding this” quietly turns into “how do we justify what we’ve already spent.”

The confident sponsor effect. A well-prepared, persuasive business case sponsor gets more benefit of the doubt than a stronger project pitched less smoothly. Over enough capex cycles, this systematically favors people who are good at pitching over projects that are good for the business.

Anchoring on last year’s budget. When this year’s capex envelope starts from last year’s number plus or minus a percentage, the strategic question of what the business actually needs this year never gets asked directly. The starting point does the deciding.

Loss aversion at the top. A project that risks a visible, attributable failure often loses out to a safer one with a mediocre return, even when the math favors the riskier option. Nobody gets blamed for the safe project that underperforms quietly.

None of these show up as a line item anywhere. They show up in which projects get funded and which get quietly deprioritised. These are patterns that are only visible looking back across many decisions, not from inside any single approval meeting.

Why this is a governance problem, not a people problem

The instinct is to treat this as an integrity issue, as if better people would make better decisions. That’s the wrong diagnosis. These biases show up in essentially every organisation, including ones with genuinely capable, well-intentioned finance teams, because they’re a feature of how humans evaluate uncertain trade-offs under pressure, not a character flaw specific to any one company.

Which means the fix isn’t a values statement. It’s structure: the same evaluation criteria applied to every request regardless of who’s asking, decisions documented at the time they’re made rather than reconstructed later, and a consistent process that doesn’t bend based on who’s in the room.

What “taking emotion out” actually looks like in practice

Standardising the inputs. Every capex request evaluated against the same financial criteria, the same strategic weighting, using the same assumptions, so a stronger presentation can’t substitute for a stronger project.

Separating the sponsor from the evaluation. The person requesting capital and the person analysing the request shouldn’t be pulled toward the same conclusion by the same incentives.

Keeping the record. When a decision is written down at the time it’s made, including what was approved, on what basis and by whom, sunk cost and hindsight both lose most of their power, because there’s a clear record of what was actually known when the call was made.

Building consistency into the process

None of this requires removing judgement from capital decisions. Good judgement is still the point. It requires making sure the same judgement gets applied consistently, request after request, instead of varying by who’s presenting or what happened last quarter. That consistency is what capex management is built to enforce: the same evaluation criteria and documented decision trail applied to every request, and approval workflows that route decisions the same way regardless of who’s asking.