Thread by @ChrisMasterjohn

A famous study done in 2001 and 2002 in Chicago and Boston randomly applied white-sounding and black-sounding names to 4,890 fake job applications and found that white-sounding names got 50% more callbacks (10.06% vs 6.7%).

nber.org/papers/w9873

A more recent study in 2016 did the same thing with 15,000 job applications in New Jersey and New York City, before and after bans were put in place asking about criminal records.

academic.oup.com/qje/article-ab…

Overall whites got 22% more callbacks than blacks (12.9% vs 10.5%). While differences in cities & methodology might account for differences from the 2001-2002 study, I suspect that this at least in part reflects a substantial decline in racial discrimination over those 15 years.

Unfortunately, this paper found that "ban-the-box" prohibitions about asking if an applicant has a criminal record led to a major increase in racial discrimination.

Before "ban-the-box," whites got 7.6% more callbacks than blacks. After "ban-the-box," whites got 36% more callbacks than blacks.

This indicate that the "ban-the-box" laws increased racial discrimination against blacks by 4.7-fold.

If we assume that progress accounts for the decline from 50% to 7.6% over 15 years, then racial discrimination declined 2.8 percentage points per year, and "ban-the-box" erased ten full years of progress in the decline of racial discrimination.

This can easily be explained by a substitution of group discrimination for individual discrimination, when employers lost an important part of their ability to filter out people with a criminal record.

This is, in @ThomasSowell's language, a substitution of type 1b discrimination (group characteristics) for type 1a discrimination (individual characteristics). independent.org/publications/t…

While it is important for those with a criminal record to have access to jobs, this should not come at the expense of black people without criminal records.

Increased racial discrimination will increase the racial disparities in crime and poverty, causing a vicious cycle, which will ultimately hurt, not help, blacks who do have criminal records.

On the one hand, I believe this shows that efforts to reduce racial discrimination have been successful over the last two decades.

On the other hand, as much as I support criminal justice reform, oppose police brutality, oppose imprisoning nonviolent offenders, and would like to see many of the laws that put nonviolent offenders in jail abolished...

...this suggests that interfering with an employer's ability to filter by criminal record causes enormous regression of the progress against racial discrimination.

I believe the solution to this is to maximize employers' ability to sort people according to individual characteristics that matter to them, including criminal record...

...while also raising awareness among employers of unconscious racial bias and educating them about how it can hurt their business.

/endthread