The Minimum Wage Debate
Why Raising the Minimum Wage Didn't Cause the Disaster Everyone Predicted
For decades, the most repeated claim in labour economics was that raising the minimum wage kills jobs — and then a natural experiment in the real world quietly dismantled it.
The Idea
The textbook version of the minimum wage debate is almost too clean: set a price floor above the market rate, and employers hire fewer workers. Supply and demand. Case closed. The problem is that real labour markets are not commodity markets, and workers are not interchangeable units of production. The more revealing concept is monopsony — the idea that in many local labour markets, employers hold significant power over workers, not the other way around. When there is effectively one dominant employer in a town, or when switching jobs carries serious costs (retraining, relocation, loss of seniority), wages can sit below what a genuinely competitive market would produce. In that situation, a minimum wage doesn't destroy jobs — it corrects a market failure. This reframing, developed seriously by economists David Card and Alan Krueger in the early 1990s, shifted the entire debate. Their insight wasn't that minimum wages are always harmless — it's that the harm depends heavily on context: how high the floor is set, how tight the labour market already is, and how much bargaining power workers actually hold. The honest answer to 'does the minimum wage kill jobs?' turns out to be: sometimes, a little, in specific circumstances — and sometimes, not at all, or even the opposite. That nuance is uncomfortable for both sides of the political argument, which is probably why it took so long to land.
In the World
In 1992, New Jersey raised its minimum wage while neighbouring Pennsylvania left its unchanged. Two economists, David Card and Alan Krueger, saw a rare opportunity: a near-perfect natural experiment. They surveyed fast food restaurants on both sides of the border before and after the change, expecting to find the predicted job losses in New Jersey. They didn't. Employment in New Jersey's fast food sector actually rose slightly relative to Pennsylvania's. The paper they published caused a genuine crisis in labour economics — not because everyone immediately agreed, but because it forced the field to take empirical evidence more seriously than elegant theory. Card went on to win the Nobel Prize in Economics in 2021, partly for this work. The Nobel committee's citation was notably pointed: it recognised that "many of the key questions in the social sciences can only be answered using natural experiments" — a rebuke to economists who had been deriving policy conclusions from models alone. More recently, Seattle's phased increases to a much higher minimum wage — significantly above the federal level — produced similarly mixed and contested results. Some studies found modest negative effects on hours worked; others found net income gains for low-wage workers overall. The honest takeaway from Seattle wasn't a clean verdict either way; it was that context, implementation speed, and local economic conditions all matter enormously. Which is, of course, a much harder thing to fit on a campaign poster.
Why It Matters
Most of us will, at some point, have a view on the minimum wage — whether voting, arguing at a dinner table, or running a business. The default mental model most people carry is the simple supply-and-demand picture, and it leads to confident conclusions that the evidence doesn't fully support. Understanding that labour markets can be distorted by employer power — and that a wage floor can sometimes correct rather than worsen that distortion — doesn't tell you what the minimum wage should be. But it does change the quality of the question you're asking. Instead of 'is a minimum wage good or bad?', you start asking 'under what conditions, set at what level, in which kind of labour market?' That shift from binary to conditional thinking is genuinely useful beyond this one policy debate. It's a template for approaching any question where the answer depends heavily on circumstances that ideologues on both sides prefer to ignore. Knowing that the textbook answer was wrong — and why — is a small but real kind of intellectual freedom.
A Question to Ponder
If the effect of a minimum wage depends so heavily on local conditions and employer power, who should really be setting it — national governments, regional authorities, or someone else entirely?
Get a new one of these every morning.
Start learning with Thinkable