Minimum wage

A minimum wage is a legal floor under the price of labour, and like any price floor set above the market-clearing level it produces a surplus. The surplus in this market is called unemployment. The law does not oblige anyone to hire, so what it establishes is not a wage but a prohibition: certain voluntary employment contracts become illegal, and the workers whose output is worth less than the floor are the ones excluded.

See also: Price controls

The mechanism

Murray N. Rothbard put it in the fewest words available. A minimum wage law is compulsory unemployment. The law says it is illegal, and therefore criminal, for anyone to hire anyone else below a stated rate. It provides no jobs; it only outlaws them, and outlawed jobs are the result.

The reason a floor bites at all is that wages are not set by the employer's generosity or by the worker's need. On the Austrian account an employer bids for a worker up to the discounted value of what that worker is expected to add to output, discounted because the employer pays now and sells later. Competition among employers pushes the bid toward that value, and competition among workers pushes it down toward it. A wage below the figure invites a rival to bid the worker away; a wage above it is a loss the employer cannot sustain.

It follows that a legislated floor cannot raise the value of anyone's output. It can only forbid the employment of those whose output falls short of it. Henry Hazlitt's formulation is that you cannot make a man worth a given amount by making it illegal for anyone to offer him less.

Margins other than the job count

Employment is the effect that gets measured, and it is not the only one or the largest. A firm facing a higher mandated wage has many ways to restore the relationship between what it pays and what it gets, and most of them do not show up as a lost job.

Training is cut first. On-the-job training is an investment the employer makes and partly recovers through wages below the trainee's eventual productivity. Raise the floor and that arrangement becomes illegal at exactly the bottom of the ladder, so the rung is removed. The worker gains a few cents now and loses the skills that would have raised his wage permanently.

Skilled labour is substituted for unskilled. Suppose a task can be done by three unskilled workers at $5 an hour or two skilled workers at $8. At current prices the firm uses unskilled labour and spends $15. Impose a $6 floor and the unskilled option costs $18 while the skilled one still costs $16, so the firm switches. The official data record one job lost and average wages rising to $8, which reads as a success, while the three least skilled workers are the ones now out of work.

Non-wage compensation is withdrawn. Scheduling flexibility, meals, uniforms, tolerance of inexperience and the willingness to keep a marginal worker through a slow quarter are all part of what an employer pays, and all are cheaper to cut than the legally fixed component.

Enforcement selects for lawbreakers. Where a price floor creates large gains from evasion, the firms with a comparative advantage in evading it do best, which means the regulation quietly advantages the least scrupulous employers over the ones who comply. This is the same structure as any other prohibition; see black market.

The Progressive-era case for the disemployment effect

The most striking thing in the history of minimum wage legislation is that its early advocates agreed it would cause unemployment among the least productive workers, and supported it for that reason.

Thomas C. Leonard documented the position across the Progressive-era economics profession. A wage floor was understood to draw a line below which the "unemployable" could not be hired, and excluding them from the labour market was presented as a social benefit, on the eugenic reasoning that competition from low-wage workers depressed the wages and the family formation of the racially preferred. Sidney and Beatrice Webb argued the point explicitly, and it was made in the American literature by economists of standing rather than by cranks.

The Davis-Bacon Act of 1931, which requires prevailing union wages on federally funded construction, has a legislative record of the same kind, with sponsors objecting to the employment of cheaper black labour brought in from the South.

The point is not that today's advocates hold those views. It is that the disemployment effect was not discovered by hostile economists afterwards; it was the design intent, and the modern position that the effect does not exist is the novel claim.

W. H. Hutt's The Economics of the Colour Bar traces the same logic through South African labour law, where white unions consistently supported minimum wage and equal-pay legislation for black workers. The mechanism is not subtle. A wage differential is what makes an employer pay for his prejudice; abolish the differential by law and discrimination becomes free, so more of it is bought.

Consistent with that, a study by William Even and David Macpherson found each 10 percent increase in a state or federal minimum wage associated with a 2.5 percent fall in employment among young white males without a high school diploma, 1.2 percent among Hispanic males and 6.5 percent among black males.

Puerto Rico, 1938

The clearest natural experiment came from applying one country's wage floor to another country's economy.

The Fair Labor Standards Act of 1938 set the first federal minimum at 25 cents an hour. The average American wage was then 62.7 cents, so the law raised pay for only about 300,000 workers out of a labour force above 54 million and passed almost unnoticed. It also applied to Puerto Rico, where much of the workforce earned three to four cents an hour.

Thomas Rustici's account is that the result was immediate and severe: cigar and cigarette manufacturing all but disappeared, and the needlework trade, which employed over 40,000 people in 1935, collapsed, with export values falling from more than $20 million in 1937 to barely $5 million by 1940. Congress amended the provisions after two years, cutting the applicable minimum to 12.5 cents and creating industry committees to set separate minima, but by then the industries were gone.

The experiment was then repeated. Amendments in 1974 and 1977 moved Puerto Rican minima up to mainland levels, reached by 1983. Alida Castillo-Freeman and Richard Freeman estimated that this reduced island employment by 8 to 10 percent, reallocated labour between industries, and shaped migration, with migrants to the mainland drawn disproportionately from those the minimum had put out of work.

Who lobbies for it

If a law reliably prices out the least skilled, the question is who wants it, and the public choice answer is that the beneficiaries are the substitutes for the workers excluded.

Unions are the clearest case. A union is an organisation restricting the supply of labour in an occupation, and non-union labour priced below the union rate is its most dangerous competitor. Organising every potential competitor is prohibitively expensive; legislating a floor under them costs the union nothing and is enforced by somebody else. Rustici's observation is that the unions lobbying hardest have consistently been those in apparel, textiles and agriculture, precisely where direct competition with non-union labour is greatest.

Firms paying above the floor have the parallel interest. A minimum wage imposes a competitor's cost structure on lower-cost rivals, which is the same service a tariff performs against foreign producers, with the advantage that supporting it can be described as social responsibility.

What the empirical dispute settles

For most of the twentieth century the disemployment effect was uncontroversial among economists. It became contested after studies by David Card and Alan Krueger in the early 1990s, most famously a comparison of fast food employment in New Jersey and Pennsylvania around a New Jersey increase, reported no employment loss.

The subsequent literature is large and does not converge. David Neumark and William Wascher's review of the post-1990 research concluded that the majority of credible studies still find negative employment effects concentrated among the least skilled.

The Austrian position on this dispute is not that the studies favouring the floor are fraudulent, and it does not depend on the count coming out either way. It is that a survey of one industry over eight months in two states cannot establish a general economic law, because everything else was moving at the same time and there is no controlled comparison available. The proposition that forbidding a transaction reduces the number of such transactions is not an empirical hypothesis awaiting a dataset; it follows from what a price floor is. If it did not hold, there would be no reason not to set the minimum at $100 an hour, and nobody proposes that.

The measurement dispute is about magnitude and about which margin absorbs the adjustment. It is worth having for that reason, and it is not the same question as whether the effect exists.

See also

• Unemployment
• Price controls
• Interventionism
• Labour
• Job training
• Black market
• Poverty
• Welfare state

Links

• Making Economic Sense by Murray N. Rothbard, ch. 36, "Outlawing Jobs"
• Economics in One Lesson by Henry Hazlitt, ch. 18
• The Economics of the Colour Bar by William H. Hutt, 1964
• The Hidden Costs of a Minimum Wage by Art Carden, July 2009
• Minimum Wage, Minimum Tax by Mark Thornton, November 1996
• The Negative Effects of Minimum Wage Laws (pdf) by Mark Wilson, June 2012
• Minimum Wages and Employment: A Review of Evidence From the New Minimum Wage Research (pdf) by David Neumark and William Wascher, November 2006
• Eugenics and Economics in the Progressive Era (pdf) by Thomas C. Leonard, Journal of Economic Perspectives, 2005
• Minimum Wages by Linda Gorman, from The Concise Encyclopedia of Economics
• Minimum wage at Wikipedia