Feed on
Posts
Comments

I’m writing final exams and hope to be back at this in due time. In the meantime, here was one particularly interesting working paper I had emailed to me today:

Discrimination and the Effects of Drug Testing on Black Employment by Abigail K. Wozniak  –  NBER Working Paper #20095 (LS PE)

Nearly half of U.S. employers test job applicants and workers for drugs.  A common assumption is that the rise of drug testing must have had negative consequences for black employment.  However, the rise of employer drug testing may have benefited African-Americans by enabling non-using blacks to prove their status to employers.  I use variation in the timing and nature of drug testing regulation to identify the impacts of testing on black hiring.  Black employment in the testing sector is suppressed in the absence of testing, a finding which is consistent with ex ante discrimination on the basis of drug use perceptions.  Adoption of pro-testing legislation increases black employment in the testing sector by 7-30% and relative wages by 1.4-13.0%, with the largest shifts among low skilled black men. Results further suggest that employers substitute white women for blacks in the absence of testing.

Lots to say here of course. In other research:

Is Smoking Inferior? Evidence from Variation in the Earned Income Tax Credit
by Donald S. Kenkel, Maximilian D. Schmeiser, Carly J. Urban  –  NBER Working Paper #20097

In this paper we estimate the causal income elasticity of smoking participation, cessation, and cigarette demand conditional upon participation.  Using an instrumental variables (IV) estimation
strategy we find that smoking appears to be a normal good among low-income adults:  higher instrumented income is associated with an increase in the number of cigarettes consumed and a decrease in smoking cessation.  The magnitude and direction of the changes in the income coefficients from our OLS to IV estimates are consistent with the hypothesis that correlational estimates between income and smoking related outcomes are biased by unobservable characteristics that differentiate higher income smokers from lower income smokers.

In case it is not clear from that last abstract, what this finding is telling you is that, the amount of cigarettes desired to be smoked INCREASES with income, that runs quite counter to the conventional wisdom. The reason for the conventional wisdom is that higher-income people have far lower smoking rates than lower-income people. What this paper tells us is that, not surprisingly, higher-income people are in a material way different than low-income people. This is not surprising to many of us. but of course is not polite to suggest in polite company. I would caution however that since this is a study using the EITC, we are not comparing the poorest people with the richest people in the income distribution.

Leave a Reply