In Defence of Liberty

Driven by data; ridden with liberty.

Minimum Wage and Maximum Noise

Employment is a transaction, where an employee sells their labour to an employer. If the transaction requires the validation of a third party, such as a government requiring specific terms, then the number of transactions will fall.

A common fallacy is to claim that a minimum wage has no negative effect on employment. For example, James Bloodworth at Left Foot Forward contends that “there is very little evidence to suggest that reducing or getting rid of the minimum wage would create jobs”. The author cites the 1994 study by David Card and Alan Krueger, which contrasted employment in the fast-food industry in New Jersey and Pennsylvania, following New Jersey’s minimum wage increase from $4.25 to $5.05 per hour. Their method was a telephone survey of selected fast-food restaurants in the two states, concluding that employment in New Jersey’s fast-food industry rose after the minimum wage hike.

The minimum wage is an emotional topic, which is why poor research with political amplification damages the discussion on its real effects. (Photo: citizenactionny)

The minimum wage is an emotional topic, which is why poor research with political amplification damages the discussion on its real effects. (Photo: citizenactionny)

There were anomalies in Card and Krueger’s data, noted by the Employment Policies Institute, such as the Wendy’s that had no full-time employees in February 1992, but 35 full-time employees in November 1992. In a 2000 paper, David Neumark and William Wascher performed the same study with administrative payroll records, and found that:

In contrast, a simple replication of CK’s differences-in-differences estimation using the payroll data indicates that the New Jersey minimum-wage increases led to a 3.9-percent to 4.0-percent decrease in fast-food employment in New Jersey relative to the Pennsylvania control group.

Living after Russian Roulette

The methodology of these ‘natural experiments’ suffers from the implicit limitation that it only records employment at firms alive both before and after the minimum wage increase; as Thomas Sowell pithily states: “reports based on surveying only survivors can create a false conclusion as interviewing people who have played Russian roulette.” Other studies attempt to distil a minimum wage increase from other state-wide or national effects. Unlike bad harvests, such policies are often rolled into a legislative bundle, drowning the minimum wage in maximum noise. Boisterous criticism of the Card-Krueger study is a faded memory, so contemporary citation grows common.

Similarly, Stella Creasy MP (Labour & Co-Operative, Walthamstow) claimed on Twitter:

The cited 2003 paper by Professor Stewart concludes that the 1999 minimum wage, and the subsequent up-ratings, had “no significant adverse effect” on “the probability of subsequent employment among those who wages would have had to be raised to comply with the new minimum”. However, this is not equivalent to suggesting it has “no effect on employment”. The UK’s minimum wage is porous, with exemptions for apprenticeships and internships, and lower price floors for different age categories. An employer may not reduce their staff rota due to this porous minimum wage, but may forestall hiring new employees. In Figure 1 of Professor Stewart’s paper, the employment rate reaches a plateau of around 74.5% after the October 2000 increase in the minimum wage.

The introduction of the minimum wage slightly slows the growth in the employment rate, before it stalls after the first increase. (Photo: Mark B. Stewart, University of Warwick)

The introduction of the minimum wage slightly slows the growth in the employment rate, before it stalls after the first increase. (Photo: Mark B. Stewart, University of Warwick)

The minimum wage is a trade-off, between higher employment and a higher wage floor for those in employment. Politicians and commentators often attempt to obfuscate its deleterious effects.

Advertisements
%d bloggers like this: