When incentives work too well

I spent part of the weekend at a Durham University psychology open day with my daughter.

Quite apart from being reminded that university open days require a level of parental stamina not usually mentioned in the prospectus, it was a useful reminder that, in a reward context, incentives are not just about money. They are about behaviour.

In the reward world, we spend a lot of time on quantum, grant levels, vesting periods, performance conditions, dilution, accounting cost and tax treatment. All of that matters. But none of it answers the more basic question.

What is this plan actually causing people to notice, believe and do?

Same behaviour, different message

One behavioural science example from the open day was nicely simple.

How do you persuade drivers to switch off their engines when level crossing barriers are down?

You could frame the message as personal financial loss: You throw away your money by leaving your engine on. As environmental harm: Warning - pollution. Or as social responsibility: Think of the children.

The behavioural objective is the same. Switch the engine off. But the psychological lever is different. Money. Pollution. Children. Self-interest. Civic duty. Identity. Guilt. Social norms.

The point is not that one message is always better than another. The point is that the message changes the behaviour because it changes what the person pays attention to.

That is exactly what reward plans do.

Reward is behavioural architecture

A bonus plan is not just a calculation. An LTIP is not just a vesting schedule. A sales commission plan is not just a percentage. A MIP is not just a waterfall.

Each one is a piece of behavioural architecture. It tells people what the organisation values, what it wants them to optimise, what risks it wants them to take and what trade-offs it is prepared to tolerate.

Sometimes the message is clear. Sometimes it is not.

A sales commission plan that rewards revenue may achieve revenue. It may also cause discounting, weak margin discipline and poor customer selection. A bonus plan intended to drive performance may instead drive internal positioning if people believe the outcome is too discretionary or politically allocated. A MIP may be described as ownership, but if participants do not understand the hurdle, the capital structure, leaver risk or exit dependency, it may not feel like ownership. It may feel like a complicated promise, subject to rules they do not really understand, redeemable at some point in the future that may never come. Read more here.

None of that means the plans are badly designed. It does mean that legal and economic design is only part of the exercise.

The incentive may work too well

One uncomfortable and perhaps obvious truth is that incentives often do exactly what they are designed to do.

Morgan Housel (my favorite author again) puts it neatly: “Incentives: the most powerful force in the world. When the incentives are crazy, the behavior is crazy.” His broader point is that people can justify and defend almost anything if the incentive structure points them in the wrong direction.

Charlie Munger (one of the world's most famous investors) made a similar point through the Federal Express example. FedEx reportedly struggled to get its night shift to move packages quickly enough while workers were paid by the hour. The answer was not another speech, another policy or more management intervention. It was changing the incentive. Pay by the shift, not by the hour, and the behaviour changed.

There is a familiar Soviet planning story, often told in the context of Stalin-era tractor quotas, that makes the same point in more absurd form. Production went through the roof because that is what the system rewarded. Whether the tractors were needed, ordered, useful or economically rational is a different question. The target was hit. The behaviour was created. The result may still be nonsense.

Reward plans can do a version of the same thing.

If you pay for revenue, do not be surprised if people chase revenue. If you pay for EBITDA, do not be surprised if investment gets deferred. If you pay for individual origination, do not be surprised if collaboration becomes harder. If you pay for fewer reported safety incidents, do not be surprised if reporting becomes less attractive. If you pay people simply to stay, do not be surprised if they stay without becoming more committed.

The question is not whether a metric can be measured. The question is whether it is a good proxy for the behaviour the business actually wants. That is a harder question. It is also the one that matters.

Framing is not decoration

Retention awards are a good example.

There are situations where a retention award is rational. A business may need key people to stay through a transaction, restructuring or critical delivery period.

But the framing matters.

One version says: you are critical to the next phase of this business. Another says: we think you may leave, so here is some money.

Same cash. Very different psychological signal. The first version reinforces value and commitment. The second confirms the person’s suspicion that the door is open - and hands them a reason to start thinking about what is on the other side of it.

Communication cannot be an afterthought. If people do not understand what they have, how it works, what has to happen for value to emerge and what could cause value to be lost, much of the incentive effect is diluted. A plan does not motivate people merely because it exists.

There is also a governance point here. Organisations say they want candour, challenge and early warning. But if the practical incentive is to keep bad news away from senior people, bad news will travel slowly. That is not a cultural mystery. It is incentive design.

The practical implication

The practical implication is uncomfortable. Reward design cannot be treated only as a legal, tax or financial exercise. The behavioural question - what will this plan cause people to notice, believe and do - has to be in the room from the start. Not as a communications afterthought. Not as an engagement initiative. But as a design input.

That means being honest about what the chosen metric actually incentivises, not what it is intended to incentivise. It means asking whether participants have a genuine line of sight or whether the plan is asking people to be motivated by something they cannot meaningfully influence. It means thinking about the framing before the paperwork, not after it.

A company can spend a great deal of money on incentives and still get very little behavioural return. Worse, it can pay for behaviour it did not really want.

That is why design matters. The framing matters. The perceived fairness matters.

In remuneration, the question is not simply what people are paid. It is what the reward structure causes them to notice, believe and do next.

At Burges Salmon, we help companies design, review and communicate incentive arrangements that work commercially, legally and behaviourally. That includes bonus plans, LTIPs, all-employee share plans, management incentive plans, retention arrangements and transaction-related awards.