← All posts

September 14, 2026 · 4 min read

The losses that start before the layoff

A woman works alone at a laptop in a wide open-plan office at the end of the day, the desks and chairs around her cleared and empty, her coat over the back of her seat.

There is a moment most people treat as the beginning. The meeting invitation with no agenda. The letter. The badge that stops working on a Tuesday. Everything before it was the job, and everything after it is the search.

The economics does not agree about where the line falls.

The decline starts before the letter does

In 1993, Louis Jacobson, Robert LaLonde and Daniel Sullivan matched administrative earnings records against information about the firms people worked for, and followed workers who separated from firms in distress. High-tenure workers lost about 25 percent of their annual earnings, and kept losing it for years. That figure gets quoted often.

The finding underneath it gets quoted less. The losses begin mounting before the separation. Not after the letter. Before it.

Which makes sense the moment you say it out loud. A company in trouble stops giving raises before it stops giving jobs. Headcount freezes. The ambitious project gets deferred to next year, and then next year is not offered. Discretionary budget goes first, and discretionary budget is what funds the work that makes a career legible: the migration, the pilot, the thing you would have led. By the time anyone hands you a letter, the year of evidence you would have built has already not been built.

The same study found something else worth sitting with: the losses were large even for people who found new jobs at similar firms. That is not a story about landing badly. It is a story about what you were able to carry with you.

Then the market decides how much it costs

Steven Davis and Till von Wachter looked at the same kind of event across different economic weather, using Social Security earnings records from 1974 to 2008. Men displaced in mass layoffs lose about 1.4 years of pre-displacement earnings when national unemployment is below 6 percent, and about 2.8 years when it is above 8 percent. Both figures are present-value, discounted over twenty years.

Same worker. Same skills. Same fifteen years of work behind them. Twice the loss, decided by the month they happened to land in.

Put the two studies together and the shape is uncomfortable. The first says the erosion starts before you are told. The second says the size of the bill is set by conditions nobody consults you about. Between them they account for most of what a career transition actually costs, and neither one is a thing you control.

What is left is what you have ready

This is not an argument for pessimism, and it is certainly not an argument for keeping a résumé permanently updated. Nobody does that, and the people who claim to are usually updating a document rather than keeping a record.

It is an argument about when the raw material exists.

The details that make a strong case, the scale you operated at, the decision that was actually yours, the number the work moved and where that number came from, are available for a short window and then they are not. They are easiest to write down in the week the work ends. They are hardest to reconstruct in the month you need them, which is the month you are also managing everything else a job loss brings with it.

The two studies say the pressure arrives before the announcement and the cost is set afterward. The window in between is narrow and it closes quietly. So the useful question is not am I ready to look for a job. Almost nobody is, and it is not a fair question. It is smaller than that:

Is the work I finished this quarter written down anywhere it will survive?

Not formatted. Not tailored. Written down. What it was, how big, what changed, and how you know.

Five minutes against two point eight years

That asymmetry is the whole case. A few minutes at the end of a project, against a loss measured in years of earnings, most of which is already in motion before anybody says anything to you.

You cannot control whether your employer is quietly in trouble. You cannot control the unemployment rate the month you go looking. You can control whether the last three years of your work still exist in a form you can use, and that is decided long before it matters, in ordinary weeks when nothing is wrong.

This is the habit Standing Proof exists to hold: the record built while the work is still fresh, so the search starts from evidence instead of memory. Start with something you finished recently, or read the method first.

Start before you need it.

The evidence you capture calmly today is the proof you won’t have to reconstruct desperately later. Five minutes, whenever something worth remembering happens.

Capture my first experience, free