Antonio Rodriguez

associate professor of economics @UCIrvine | international trade, labor markets, and some macro | 🇲🇽

[Antonio Rodriguez @4ntonioR]

What do diff-in-diff event-study designs really tell us about the employment effects of min wages?

My new paper with @NeumarkEcon tackles this using the stacked design of Cengiz-@arindube-@attilalindner-@benzipperer and the related LP-DiD design in Dube-Lindner’s HLE chapter. 🧵

[Jeffrey P. Clemens @jeffreypclemens]

@jondr44 @mattkahn1966 Some relevant prior research of interest by Monras:
https://t.co/jUys1ZwlAy

[Antonio Rodriguez @4ntonioR]

This creates yet another bias in CDLZ’s findings, also exacerbated by the negative pop-MW relationship. We document this bias clearly in Section 3.

It’s also important to note that the restaurant results are strongly negative, even when using varying population. 5/5

[Antonio Rodriguez @4ntonioR]

Great feedback @jondr44. Thank you!

While we have not looked at an extended pre-period for log population, we do show the population event-study paths in the paper. We show them for all the CDLZ events and for bundled-events samples. 1/5

[Jonathan Roth @jondr44]

A few thoughts on this interesting new minimum wage paper by @NeumarkEcon and @4ntonioR renalyzing the QJE paper by @arindube-@attilalindner-@benzipperer, which relates to one of my favorite topics: event-studies and parallel trends! 1/

[Antonio Rodriguez @4ntonioR]

In the log-population elasticity figure below, panel (d) shows no pretrends for the three bundled samples, at least from our view. Your figure should be close to the one in panel (b).

Another important clarification: CDLZ use emp/pop, not log(emp/pop). 4/5

[Antonio Rodriguez @4ntonioR]

@jeffreypclemens @pedrohcgs @jlaborecon @MichaelRStrain Yes, your paper is another important example of modern DiD finding negative effects. I should have been more careful. I meant the Callaway–Sant’Anna application as an early prominent example, not as the only one.

[Antonio Rodriguez @4ntonioR]

Worth noting: we are not the first to find disemployment effects of minimum wages using a modern DiD approach. In the application included in their massively influential 2021 paper, Callaway and @pedrohcgs find stat sig negative effects on teen employment that grow over time.

[Antonio Rodriguez @4ntonioR]

What do diff-in-diff event-study designs really tell us about the employment effects of min wages?

My new paper with @NeumarkEcon tackles this using the stacked design of Cengiz-@arindube-@attilalindner-@benzipperer and the related LP-DiD design in Dube-Lindner’s HLE chapter. 🧵

[Jeffrey P. Clemens @jeffreypclemens]

My forthcoming @jlaborecon paper with @MichaelRStrain also uses modern DiD estimators as supplements to the analyses within our partially precommitted analysis pan (which, incidentally, we developed prior to the modern DiD literature, but also made purposeful choices to sidestep the “staggered treatment adoption” issue):
https://t.co/NqAXsQQbWT

[Antonio Rodriguez @4ntonioR]

What do diff-in-diff event-study designs really tell us about the employment effects of min wages?

My new paper with @NeumarkEcon tackles this using the stacked design of Cengiz-@arindube-@attilalindner-@benzipperer and the related LP-DiD design in Dube-Lindner’s HLE chapter. 🧵

[Jeffrey P. Clemens @jeffreypclemens]

This is an under appreciated issue, and relates directly to a theoretical point made in a recent @JPubEcon paper by Gregory and Zierahn:
https://t.co/xkki9NeyqO

[Wojtek Kopczuk 🇵🇱🇺🇦 and 🇺🇲 @wwwojtekk]

My reading of the literature on employment effects of minimum wage is that nobody who works on it will ever be unemployed

[Antonio Rodriguez @4ntonioR]

@twittwoods Thanks for asking. We have a full paper on that:

[Antonio Rodriguez @4ntonioR]

My paper with Jha, Kala and @NeumarkEcon is now published at JUE. Min wages reduce restaurant jobs in most places, but bite less in larger cities. Ignoring city size overstates how much monopsony power softens these adverse employment effects. Open access:
https://t.co/Bt1Z2D0H97

[Antonio Rodriguez @4ntonioR]

What do diff-in-diff event-study designs really tell us about the employment effects of min wages?

My new paper with @NeumarkEcon tackles this using the stacked design of Cengiz-@arindube-@attilalindner-@benzipperer and the related LP-DiD design in Dube-Lindner’s HLE chapter. 🧵

[Antonio Rodriguez @4ntonioR]

To conclude, modern diff-in-diff methods yield the same old answer: minimum wages reduce jobs. The opposite finding in CDLZ and DL is the consequence of discretionary, narrow, and unsound researcher choices that favor that result.

Comments welcome. Link:

https://t.co/kb9XG90D5w

[Antonio Rodriguez @4ntonioR]

Why the difference? We show that CA and NY (1st and 4th most populous states) drive DL’s null results, pointing to heterogeneity rather than a zero MW effect.

Removing CA and NY, both weighted and unweighted estimates point to job losses, especially in the restaurant industry.