“Raising the minimum wage reduces employment.”
Introductory economics predicts that a price floor on labour reduces the quantity demanded. Since Card and Krueger's 1994 New Jersey study, a large empirical literature has found the effects of moderate increases on employment to be small or undetectable, and economists' views have shifted accordingly.
The live dispute is about size and limits. Nearly everyone agrees a high enough minimum wage would cost jobs; nearly everyone agrees small increases have small effects. What remains contested is where the threshold lies (often framed relative to the local median wage), whether hours cut rather than jobs, and how to read evidence when increases spread to high-wage cities like Seattle.
Any increase in the legal wage floor. Effects plausibly depend on the size of the increase relative to local median wages (the "bite"), which ranges from about 35% to over 60% across countries and regions.
Fewer people employed, or fewer total hours worked, than without the increase. Effects on hiring, hours and future job growth are often distinguished.
When Seattle raised its minimum wage to $13, low-wage workers' hours fell enough that their total earnings declined on average. A wage floor binds on hours even when headcount looks unchanged.
Jardim and colleagues used unusually detailed Washington State administrative data on hours: the rise to $13 reduced hours in low-wage jobs by around 9%, lowering low-wage payroll; experienced workers gained while new entrants lost.
Jardim et al. (2022). Minimum-Wage Increases and Low-Wage Employment: Evidence from Seattle. AEJ: Economic Policy, 14(2), 263–314.
Limitation
A competing Berkeley study of Seattle's food-service sector found no job losses. Seattle's booming labour market makes the counterfactual hard to construct.
The federal increases of 2007–2009 reduced employment among the low-wage workers directly affected, particularly young and less-educated workers, and slowed their wage growth over the following years.
Clemens and Wither compared workers in states bound by the federal increase with similar workers in states with higher existing minimums and found significant declines in employment and in subsequent income growth among those directly affected.
Clemens & Wither (2019). The minimum wage and the Great Recession. Journal of Public Economics, 170, 53–67.
Limitation
The increases coincided with the Great Recession, which hit bound states differently, complicating identification.
Even sympathetic estimates predict job losses from large increases. The CBO's analysis of a $15 federal minimum found a median of over a million jobs lost alongside large income gains for most low-wage workers.
CBO estimated that raising the federal minimum to $15 by 2025 would reduce employment by 1.3 million workers in the median estimate (with a range of nearly zero to 3.7 million) while raising pay for 17 million and lifting 1.3 million out of poverty.
Congressional Budget Office (2019). The Effects on Employment and Family Income of Increasing the Federal Minimum Wage.
Limitation
The range includes zero, and the CBO later revised its methods. A forecast is not evidence of past effects.
When New Jersey raised its minimum wage, fast-food employment did not fall compared with neighbouring Pennsylvania. This natural experiment overturned the assumption that small increases inevitably cost jobs.
Card and Krueger surveyed 410 fast-food restaurants before and after New Jersey's 1992 increase from $4.25 to $5.05 and found no reduction in employment relative to eastern Pennsylvania.
Card & Krueger (1994). Minimum Wages and Employment: A Case Study of the Fast-Food Industry in New Jersey and Pennsylvania. American Economic Review, 84(4), 772–793.
Limitation
Neumark and Wascher re-analysed with payroll records and found employment declines; Card and Krueger's reply using BLS data upheld their result. One sector and one increase.
Across 138 state-level increases, the number of low-wage jobs did not fall: jobs below the new minimum disappeared and were replaced by jobs just above it.
Cengiz, Dube, Lindner and Zipperer's "bunching" method compared the number of jobs below and just above new minimums over five years and found the overall number of low-wage jobs essentially unchanged, for minimums up to about 59% of the median wage.
Cengiz, Dube, Lindner & Zipperer (2019). The Effect of Minimum Wages on Low-Wage Jobs. Quarterly Journal of Economics, 134(3), 1405–1454.
Limitation
Their result holds for increases within the historical US range. It does not tell us about minimums well above 60% of the local median.
Low-wage employers have wage-setting power because workers can't costlessly switch jobs. In such "monopsony" markets, a minimum wage can raise both pay and employment, and the UK's national minimum wage rose sharply without significant job losses.
Dube's 2019 review for the UK Treasury of international evidence concluded that minimum wages had raised low earnings with little effect on employment, and the UK's minimum, now among the highest relative to the median in the OECD, has seen record employment.
Dube (2019). Impacts of minimum wages: review of the international evidence. HM Treasury; Manning (2003). Monopsony in Motion. Princeton University Press.
Limitation
UK employers adjusted through hours, prices and reduced pay progression above the floor. Effects may appear only with long lags.
A sufficiently high minimum wage relative to local wages would reduce employment.
Moderate increases within the historical US range have had small employment effects in most studies.
Minimum wage increases raise the earnings of most affected workers who keep their jobs.