How your wage is actually set

The market most people depend on is the least efficient one they will ever participate in: no public prices, enormous information asymmetry, and a counterparty who does this constantly while you do it rarely.

Part of how markets actually work.

Most people’s entire income comes through one market, and it is the worst-functioning market they will ever participate in.

Not rigged. Badly built — in ways that are structural, well documented, and mostly fixable from your side once you can see them.

What is wrong with it, mechanically

Run it through the four jobs from the overview.

Matching is slow and expensive. Finding a job takes weeks or months. So does filling one. Both sides invest heavily in a process that frequently ends in no match.

Price discovery is nearly absent. You do not know what your colleagues earn. You do not know what the employer budgeted. You do not know what the last person in the role was paid or what the rejected candidate asked for. In no other market would anyone accept negotiating a major transaction with none of this.

Liquidity is low in both directions. You cannot change jobs this week. They cannot replace you this week. Both sides are locked in, which is why dissatisfaction persists for years on both sides.

Settlement is the one part that works. You get paid, reliably, on time.

Three of four jobs done poorly. And unlike the stock market, nobody is being paid to fix it — there is no market maker for labour, because a human’s work cannot be held as inventory and resold.

Where the number actually comes from

Wages feel like they are set by what you are worth. They are mostly set by four other things.

A band decided before you appeared. Most employers have a salary range per role, set by market surveys and internal equity, usually before the role was posted. Much of the negotiation happens inside a band you cannot see and did not influence.

What replacing you would cost. This is the employer’s real constraint. Not the value you produce — the price of getting someone else to produce it. Those two numbers can differ enormously, and the gap is where profit lives. That is not exploitation; it is the wage channel working as described.

Your outside option, whether or not you use it. The single strongest determinant of pay is what someone else would pay you, because it sets the floor below which you leave. This is why changing employers so reliably raises pay more than staying: the market rate is repriced at hire and only partially updated afterwards.

When you last renegotiated. Internal raises are usually a percentage of a number set when you joined. An initial figure compounds for as long as you stay, which makes the starting number worth far more than it appears.

The asymmetry is the whole thing

The employer has done this hundreds of times. They know the band, the distribution of accepted offers, what candidates typically ask, and what happens when they say no.

You have done this perhaps five or ten times in your life. You do not know the band. You often do not know what the role is worth to within 30%.

That is not a negotiation between equals. It is a negotiation between a professional and an amateur, and the professional’s advantage is information, not toughness.

Which points at the fix. The advice to “negotiate harder” addresses the wrong variable. Getting the information is what closes the gap, and unlike confidence it is obtainable: published salary ranges where they are mandated, aggregated survey data, recruiters who see many offers, and people in the same role who will tell you if asked directly.

The most useful question is not “what should I ask for?” It is “what did people who took this role actually receive?”

The structural ceiling

Even played perfectly, this market has a limit that no amount of skill removes.

Your income is capped by hours. Skill raises the rate; nothing raises the hours in a week. And your pay is anchored to replacement cost, which is a market price you do not control and which technology can move sharply.

That is the site’s argument arriving in its most concrete form. When a task can be done by something an employer buys instead of someone they hire, replacement cost falls — and pay is anchored to replacement cost, not to the value produced. The work still gets done. It is paid for through a different channel.

None of which means abandoning the wage channel. For nearly everyone it is the main income and will be for a long time, and it funds everything else. It means knowing what it can and cannot do:

It can produce a surplus, reliably, with low variance — which is exactly the raw material stage 3 needs.

It cannot compound. There is no version where a wage grows exponentially on its own, because there is no ownership in it.

Practical consequences

Not advice so much as arithmetic.

Getting the information beats negotiating harder, because information is what the other side actually has.

The starting number compounds. Every future raise is a percentage of it. A one-time increase at hire is worth more than the same increase later.

Changing employers reprices you; staying does not. This is a structural feature of how internal raises are calculated, not a comment on loyalty.

Your outside option sets your floor whether or not you exercise it. Knowing what you could get elsewhere changes what you accept here, even in silence.

And it caps out. Optimise it, use the surplus, and understand that the ceiling is a property of the channel rather than of you.

Next in this series: how crypto markets work, where the market maker is replaced by an algorithm and the results are instructive.