In response to my piece on the “opportunity wage” in which I cited research showing that increasing the minimum wage does not reduce teen unemployment, I received the following comment:
“Not that I imagine you care but the so-called state-of-the-art Allegretto/Dube/Reich study you and the NELP use as support for your claim has been shown to be methodologically flawed by Meer/West, Murphy, and Neumark/Salas/Wascher. Additionally, studies by Clemens/Wither, Baskaya/Rubinstein, Liu, Hyclak/Regmi, Powell, and Totty all suggest that increases in the min wage have negative employment effects on unskilled workers.
“You can call it lots of things but ‘economic nonsense’ is not one of them.”
The economic nonsense I was accusing Representative Kenneth Mendonca (Republican, District 72, Portsmouth) of had to do with his scare tactic of saying that machines are taking low wage worker jobs. Machines are coming for many jobs of course, but Mendonca’s claim that increases in the minimum wage may “accelerate the schedule” of bringing these machines on-line is nonsense.
As for Do Minimum Wages Really Reduce Teen Employment? being methodologically flawed, I reached out to the authors of the paper to find out how they have dealt with such criticisms.
Professor Michael Reich of UC Berkeley, one of the authors, responded by saying he is familiar with every paper mentioned. He referred me to a short piece by Ben Zipperer explaining the logic of the’s study and the methodological flaws of some of the papers cited by the commenter. “…some studies find negative effects of the minimum wage on teen employment because they fail to control for other economic factors that independently reduced employment around the time of a minimum wage increase,” writes Zipperer, “After controlling for these factors, we demonstrate that the large, negative effect on teen employment disappears.”
Reich, in response to my query, responded, “All [the studies cited] except Totty share the same weakness that we exposed in Dube, Lester and Reich 2010 and in Allegretto, Dube and Reich 2011: A failure to recognize that the states with higher minimum wages are not a random set of states.
“States that raised their minimum wages have had slower unskilled employment growth for reasons that have nothing to do with minimum wages. Failure to control for this omission leads to negative bias– for example, to finding negative employment effects years before a minimum wage is enacted. The bias disappears, though, when we add controls for nearby areas. Generally speaking, the papers on the commenter’s list ignore our critique; therefore they do not refute our findings, but rather reinforce them.”
Reich notes that “Totty’s paper uses a different statistical approach. His results actually support our claim.”
“In response to our critiques,” continued Reich, “Neumark, Salas and Wascher 2014 claimed that nearby areas do not make good control groups. We show (ADRZ: Allegretto,Dube, Reich and Zipperer 2014 and 2017) that they are wrong. We document in ADRZ the continued presence of negative bias specifically for the articles on the commenter’s list. ADRZ has been circulating for several and appears as the lead article in the May 2017 issue of ILR Review, the leading scholarly journal in this debate.
“Finally, the commenter’s list curiously does not include subsequent studies that do not find job loss, including Belman and Wolfson 2014, 2016; Dube, Lester and Reich 2016; Allegretto, Dube, Reich and Zipperer 2017; Allegretto and Reich 2017; Dube 2017; and Zipperer 2017.”
Studies continue to show that increasing the minimum wage is good for workers and has no impact on jobs.