Randy Komisar named this in The Monk and the Riddle in 2000, watching founders describe a plan with two stages: do the thing you do not want to do, in order to reach the point where you can do the thing you do. Stage one is instrumental and finite. Stage two is the life.

The observation is that stage two rarely arrives, and that the reason is structural rather than a failure of nerve.

Why the second stage recedes

The trigger moves with the circumstances. The number that would be enough was set when the income was lower, and it rises with it. So does the standard of living it has to support, which is the hedonic treadmill applied to a threshold rather than to a feeling. Nothing is wrong with the arithmetic at any point; the finish line is simply being recalculated on current conditions.

The reference group changes. Success moves you into a room where the people are further along, and enough is defined against whoever is nearby. See relative deprivation.

The commitments accumulate. The house, the school fees, the standard the family is now used to. Each was affordable when taken on and each raises the floor, which is what golden handcuffs describes from the inside.

The identity sets. After long enough, stage one is not a means any more; it is who you are, and leaving it is not the completion of a plan but a loss of self-description. This is the hardest of the four to see from inside and the most decisive.

The part that is not a mistake

It is worth resisting the version of this that treats all deferral as self-deception, because most deferral is correct.

Training for a profession, saving for a business, building something that pays later — all of these require accepting a worse present for a better future, and the alternative is not a better life but a life with no capital of any kind in it. Career capital is the argument for stage one and it is a good argument.

The failure is specific: a trigger condition that moves. A plan with a fixed, checkable, written condition is a plan. A plan whose condition is when I have enough or once things are stable is not, because those are recalculated continuously against a shifting baseline.

Making the trigger hold

Three things, and the first does most of the work.

Write the number down, with a date, before you are near it. A condition set in advance is the only kind that can be met, because a condition assessed in the moment is assessed by someone with a new reference point.

Name what stage two actually is. Most deferred plans have a vague second stage, and a vague thing cannot compete with a concrete stage one that has a salary attached. If stage two cannot be described in a sentence, the deferral is not toward anything.

Take some of it now. Not as a compromise but as a test. A plan that cannot accommodate any of stage two in the present is usually a plan whose stage two would not survive contact — and the small version tells you that cheaply, while there is still time to change the plan rather than discovering it at the end.

Once stage one has run long enough, the years already spent become an argument for continuing: it would be a waste to stop now. That is sunk cost in one of its most expensive forms, because what it is protecting is not money but a story about what the time was for.

The question that cuts through it is the one regret minimization asks, and it is unusually clean here: not whether stage one was worth it — it may well have been — but whether, from here, another five years of it is what you would choose if the previous ones had not happened.