Read on Twitter
Thanks, @orgrem, for including me in the #TweetABlackEconomistsPaper thread. I'd like to talk about the work of three Black scholars that help us understand the many challenging economic issues facing us. Great idea by @femonomics! https://twitter.com/orgRem/status/1276541471949230080
First is Mel Stephens, a professor at U Michigan econ who has studied the effect of earnings shocks on household behavior, especially consumption.
As US jobless claims inch ever higher, an important question is: how will households respond to this loss of income? In fact, for many workers, these job losses may be permanent.
In "Worker displacement and the added worker effect," Prof. Stephens analyzes the impact of a husband's permanent job loss on their wive's labor market behavior. https://www.journals.uchicago.edu/doi/abs/10.1086/339615?journalCode=jole
He uses data from the PSID to estimate this effect. He finds that in response, wives' labor response can compensate for the long-term income loss by only 25%.
Moreover, the added worker effect is not even across the income distribution; the most significant effects are observed for households with high-earning men.
However, the paper's biggest takeaway is that most households never recover from this loss of income, even when a partner compensates with even more labor.
But the income effect is only one dimension of the dynamics unfolding in the household.
In "Job displacement, disability, and divorce" with Kerwin Charles, another influential Black economist and Dean of Yale SOM, they find evidence that job loss (of either spouse) due to layoffs increases the probability of divorce. https://www.journals.uchicago.edu/doi/abs/10.1086/381258
Mel's many other papers also provide insight into the long-term effects of job loss and displacement on factors such as long-term consumption ( https://www.mitpressjournals.org/doi/abs/10.1162/003465301750160018), and the impact of job-loss expectations on actual losses and changes in consumption ( https://www.mitpressjournals.org/doi/abs/10.1162/003465304323023796)
In addition to these important papers, so relevant to our current situation, Mel was influential to me and many of my PhD student peers at CMU when he was a junior prof there.
He helped us start student seminar series, which was responsible for so many students from @heinzcollege doing so well, and was always willing to have hallway conversations about econometrics and our papers.
Here is a link to Mel's website for his other research: https://lsa.umich.edu/econ/people/faculty/mstep.html
Now that I've started on this, I want to tweet about two other papers from friends that have influenced how I think about policy and methodology.
The first is a paper is by my friend Yaa Akosa Antwi, a health economist at Johns Hopkin's Carey Business School.
At a time where the #AffordableCareAct is being threatened again by the Trump administration, the big question is, what will the economic impact be?
In "Effects of Federal Policy to Insure Young Adults: Evidence from the 2010 Affordable Care Act's Dependent-Coverage Mandate", Yaa and her colleagues Asako Moriya and Kosali Simon address this question. https://www.aeaweb.org/articles?id=10.1257/pol.5.4.1
They find that the ACA dependent-coverage mandate increases coverage for young people and decreases in their labor-market flexibility.
This is important: As both economic as well as health coverage uncertainty increase, what will happen to young people in a time where labor market mobility and experimentation are vital for building their long-term earnings potential.
Here is a link to Yaa's website: https://carey.jhu.edu/faculty/faculty-directory/yaa-akosa-antwi-phd
Finally, I want to mention a working paper by my friend Edward McFowland at the University of Minnesota titled "Efficient Discovery of Heterogeneous Treatment Effects in Randomized Experiments via Anomalous Pattern Detection" https://arxiv.org/abs/1803.09159 .
As someone who has done many field experiments, a big question is finding a disciplined way of understanding heterogeneous treatment effects. Ed's paper provides a framework.
He reformulates heterogeneous treatment effects as a ML problem of anomaly detection. By defining a "normal" distribution of effect sizes, his subset scan algorithm then can find anomalous subgroups for whom the treatment effect is larger/smaller relative to the normal.
This integration of ML, experimentation, and econometrics is likely to be the future as experiments get larger, more complex, and researchers look for policy levers by understanding when treatments work and don't.
Ed's website is here: https://carlsonschool.umn.edu/faculty/edward-mcfowland-iii
