Case Study 1 — Why the Labor Demand Elasticity Determines the Minimum Wage Debate

The debate over the minimum wage is one of the most consequential — and most empirically contested — disputes in modern economics. At its core, the debate is not about whether the minimum wage causes some employment effects (almost everyone agrees there are some). It is about how big those effects are. And "how big" is exactly the question elasticity is designed to answer.

This case study walks through the debate as a question about labor demand elasticity. By the end, you should understand why two reasonable economists looking at the same labor market can come to different conclusions about a $15 federal minimum wage — and you should understand why the elasticity number matters far more than the political rhetoric usually admits.

The simple model

In the supply-and-demand model from Chapter 5, the labor market works like any other market. The "good" being traded is hours of labor. Workers (suppliers) offer labor in exchange for wages. Firms (buyers) hire labor in exchange for the work performed. The intersection of labor supply and labor demand sets the equilibrium wage and the equilibrium number of hours worked.

Now suppose the government imposes a minimum wage above the equilibrium. The simple model predicts:

• Workers offering labor at the new higher wage exceeds firms' demand for labor at that wage
• The result is a labor surplus — what we ordinarily call unemployment
• Some workers benefit (those who keep their jobs at the higher wage)
• Some workers lose (those who can't find jobs at the higher wage)
• The total number of jobs falls

This is the textbook prediction. It's been in economics textbooks for nearly a century. And it has been the dominant framing of the minimum wage debate in policy discussions.

The size of the predicted effect depends entirely on the elasticity of labor demand. If labor demand is highly elastic (say, an elasticity of −2 — a 10% wage increase causes a 20% reduction in employment), then a minimum wage increase causes large job losses. If labor demand is inelastic (say, an elasticity of −0.2 — a 10% wage increase causes only a 2% reduction in employment), then a minimum wage increase causes much smaller job losses, and the gains for workers who keep their jobs may exceed the losses for those who don't.

So the entire policy question turns on a number: what is the actual elasticity of labor demand for low-wage workers?

The Card-Krueger study

In 1994, David Card and Alan Krueger published a study that changed the minimum wage debate. They examined fast-food restaurant employment in New Jersey before and after a New Jersey state minimum wage increase, comparing it to fast-food employment in eastern Pennsylvania (which did not raise its minimum wage) over the same period.

The simple model predicts: New Jersey fast-food employment should fall relative to Pennsylvania.

What Card and Krueger found: New Jersey fast-food employment did not fall relative to Pennsylvania. If anything, it rose slightly. The implied elasticity was approximately zero — and possibly even slightly positive.

This was a stunning result. It contradicted the textbook prediction. It suggested that the labor demand elasticity for low-wage workers might be much smaller than the simple model assumed — or even that the standard model was wrong about the direction of the effect.

The Card-Krueger study provoked an enormous amount of follow-up research. Some of it confirmed their findings; some of it challenged them. Neumark and Wascher, in particular, replicated parts of the study using a different data source and found a small negative employment effect, contrary to Card and Krueger. The methodological dispute went on for years.

Over time, a rough consensus has emerged among labor economists: the labor demand elasticity for low-wage workers is small — much smaller than the simple model predicted — but probably not zero. The CBO's estimates put the elasticity in the range of −0.1 to −0.4, depending on the size of the wage increase and the labor market conditions.

Why is the elasticity so small?

The honest answer is that it's complicated, and we don't fully understand it. Several mechanisms have been proposed:

1. Monopsony power

The simple supply-and-demand model assumes a competitive labor market — many employers competing for workers, no single employer with wage-setting power. In reality, low-wage labor markets often have monopsony power: a small number of employers in a local market who can effectively set wages below the competitive level.

When monopsony power exists, the standard prediction reverses for moderate minimum wage increases. A binding minimum wage on a monopsony actually raises employment, not lowers it, because it forces the monopsony employer to behave more like a competitive employer. We will see this in much more depth in Chapter 21.

The empirical evidence for monopsony in low-wage labor markets is increasingly strong. Recent work has found that employer concentration is higher than economists assumed, that workers face significant frictions in moving between jobs, and that wage-setting power is real even in industries that look superficially competitive (like fast food).

2. Productivity adjustments

When wages rise, firms can sometimes get more productivity out of workers — through reduced turnover, more training, better selection of new hires, or simply because workers are happier and more motivated. These productivity gains offset some of the cost increase, so the firm doesn't have to cut as many jobs.

The empirical evidence is mixed on this — productivity gains exist but are usually modest. They can explain some but not all of the small elasticity.

3. Price pass-through

Firms facing higher wages can pass some of the cost onto consumers as higher prices. The size of the pass-through depends on the elasticity of consumer demand for the firm's products. For fast food, where demand is moderately elastic (lots of substitutes but also fairly time-sensitive purchases), pass-through is partial. The wage increase reduces firm profit and raises consumer prices, but it doesn't fully reduce employment.

4. Labor market frictions

The simple model assumes workers can move freely from one job to another, and firms can hire and fire freely. In reality, both of these are slow and costly. When a wage rises, firms don't immediately fire workers (firing is expensive and disruptive); they slow hiring and let attrition take its course. When a wage rises, workers don't immediately quit other jobs to come to this one; they search slowly. The labor market adjusts gradually rather than suddenly.

These frictions mean that the short-run labor demand elasticity is even smaller than the long-run one. The short-run effect of a minimum wage increase is small; the long-run effect (over years) may be larger.

What this means for the policy debate

The minimum wage debate has shifted significantly over the last 30 years. In the 1970s and 1980s, the standard textbook position was: the labor demand elasticity is large enough that meaningful minimum wage increases cause significant job losses. By the 2000s and 2010s, the standard position had shifted: the labor demand elasticity is smaller than once thought, and moderate minimum wage increases (say, from $7.25 to $12) probably cause small job losses that are outweighed by wage gains for workers who keep their jobs.

The shift was driven by empirical evidence — Card and Krueger and the literature that followed — not by ideological change. Economists who would have opposed minimum wage increases in 1985 now support modest ones. Economists who supported large minimum wage increases in 1985 are still cautious about very large ones (say, $20 or above), because the elasticity may not be small at very high wage floors.

The IGM Forum poll on minimum wage policy reflects the shift. In 2013, when economists were asked about a hypothetical $9 federal minimum wage, opinion was divided: about 35% agreed it would not significantly reduce employment, 24% disagreed, and the rest were uncertain or had no opinion. By 2019, when asked about a $15 federal minimum wage, the split was even more pronounced — economists were less confident about a higher wage floor because the elasticity might be larger at that level.

This is what honest policy disagreement looks like in economics. The empirical evidence has moved the consensus on small minimum wage increases. The empirical evidence has not yet settled the question for very large increases. Economists disagree because they weight the available evidence differently, not because they have different ideologies. Some emphasize the Card-Krueger framework and conclude that even $15 will have small effects. Others emphasize the possibility that elasticities rise at higher wage floors and worry about $20 or above. Both positions are defensible.

What the elasticity cannot tell us

Even with a precise estimate of the labor demand elasticity, the minimum wage debate would not be settled. The elasticity tells you the positive fact (how many jobs would be affected). It does not tell you the normative question (whether the trade-off is worth it).

If a $15 federal minimum wage causes a 2% reduction in low-wage employment but raises wages for the 98% of low-wage workers who keep their jobs by 30%, is that a good policy? The answer depends on: - How you weight the gains (more income for workers who keep jobs) against the losses (no income for workers who don't) - Whether the workers who lose jobs were the most vulnerable or the least - How you account for indirect effects (less unemployment insurance demand, more consumer spending) - What you think the purpose of minimum wage policy is — preventing exploitation, lifting people out of poverty, supporting living wages, etc.

These are normative questions, and elasticity cannot answer them. But elasticity is the tool that lets us ask them precisely. Without it, the debate would be even more confused than it already is.

Discussion questions

1. The Card-Krueger study found a near-zero employment effect for a moderate minimum wage increase. What would the equivalent study look like for a much larger minimum wage increase (say, doubling the wage)? Would you expect the same result?

2. Monopsony power is one of the leading explanations for the small empirical elasticity. What would be evidence for monopsony in a particular labor market? What would be evidence against?

3. The case study notes that the labor demand elasticity is probably larger in the long run than in the short run. What does this imply for the timing of any negative effects of a minimum wage increase? Should we expect to see job losses immediately, or after several years?

4. Some economists argue that the right way to fight low-wage exploitation is not minimum wage but the Earned Income Tax Credit (EITC) — a wage subsidy from the government rather than a price floor on wages. Use elasticity to evaluate the comparison. What are the trade-offs?

5. The IGM Forum poll showed that economists are less confident about large minimum wage increases than small ones. What does this tell you about the structure of economic debate? Is "small effects for moderate increases, larger effects for large increases" a coherent position, or is it a way of avoiding hard choices?