Some trade-offs get refused rather than answered, and the refusal is doing work

The question people decline to answer

Ask what a particular safety improvement is worth in money and many people will refuse the question rather than answer it. The refusal is usually immediate and often accompanied by irritation at having been asked. This is not confusion about arithmetic. It is a judgement that the comparison itself is illegitimate, and that anyone willing to make it has revealed something about their character.

Researchers describe these as protected or sacred values, and the experimental picture is reasonably consistent: proposals to trade such a value against money produce not just refusal but moral disapproval of the proposer, and sometimes a stated willingness to accept a worse outcome rather than participate in the trade. Most of this evidence comes from stated responses to hypothetical scenarios, which is a genuine limitation, and behaviour in real settings with real costs is harder to study and less well documented.

The refusal is not irrational

It is easy to treat the refusal as a failure of clear thinking, and that reading misses what it accomplishes. A person who is known to refuse a certain trade cannot be approached with an offer, which removes an entire category of pressure from their life. The refusal functions as a commitment, and commitments work precisely by making some options unavailable.

There is also a signalling function that is not cynical. Willingness to price something is information about how you regard it, and in relationships of trust — professional, familial, political — being the sort of person who will not entertain certain comparisons is part of what makes the trust possible. Answering the question at all can damage something the answer was supposed to protect.

Declining to price it does not avoid the trade-off

What the refusal cannot do is make the trade-off disappear. Any finite budget for safety implies a rate at which money is being exchanged for risk, whether or not anybody states it. The rate exists in the arithmetic of what was funded and what was not. Refusing to name it does not set it to infinity; it sets it to whatever the unexamined pattern of spending happens to produce.

The predictable consequence is inconsistency. Without a stated rate, resources go where attention goes, which means towards vivid and recent risks and away from dull ones. Two measures with very different costs per unit of harm avoided sit side by side and nobody notices, because noticing would require the comparison that was ruled out at the start.

Where making it explicit helps

Explicit trade-offs are most valuable in aggregate decisions made repeatedly by institutions, where the alternative is not a principled refusal but an accidental and invisible rate. Stating a rate makes the pattern of decisions inspectable and arguable, which is the only route by which it can be improved.

They are least valuable, and most corrosive, in decisions about a specific identified person, and in relationships where the willingness to compare is itself the offence. The distinction is not really about magnitude. It is about whether the comparison is being made over a class of cases or against a particular individual who is present.

A route through, when you have to decide anyway

The move that resolves most of the tension is to separate the rate from the case. Decide the policy in advance, in the abstract, when no identified person is in front of you, and then apply it. That is uncomfortable in a different way, but the discomfort arrives in a setting where it can be reasoned about rather than in a setting where it deforms the decision.

The second move is to be honest about what is being protected. If the real objection is to being the person who states the number, say so — that is a legitimate consideration about role and relationship, and it can be handled by putting the decision somewhere else. If the objection is that the value is genuinely infinite, notice that no budget behaves as though anything is infinite, and that the claim will be contradicted by the next decision either way.

The version worth keeping

Sacred values are not a bug to be reasoned away. They carry commitments that are useful, they protect relationships that explicit pricing would damage, and a person with none of them is not obviously admirable. The defensible position is that they are appropriate for individual cases and for questions of character, and dangerous when they are used to avoid looking at a pattern of decisions that already embodies a rate.

The test is simple enough to apply. Ask whether the refusal is preventing a bad trade or preventing you from seeing the trades you are already making. Those are different situations, and only the first is protecting anything.