Racism is a complicated matter. Partisans of the left and partisans of the right are currently involved in a giant name-calling battle about whether or not the United States is a “racist nation”. Such a debate is truly empty. Different Americans think different things. The same Americans can be humanistic in one situation and racist in another. Everyone has the capacity to be more racist than anyone expects or less racist than anyone expects. The question is not whether any person or country is always racist or never racist. The question is what triggers the potential for racist activity that is always there - and what keeps that potential locked away, inactive and harmless.

If you want to be happy and upbeat about the prospects of containing and defeating racism, look at survey data on public attitudes. Hopkins and Washington of the Political Science Department of the University of Pennsylvania have published a recent analysis of racial attitudes among national samples of whites. The findings replicate for the present era the favorable results that have been reported for white racial attitudes in earlier periods. The presence among whites of negative attitudes towards blacks has been steadily declining. Prejudice is measured by asking people to rank whites on a scale from 0-100 on some aspect of social desirability and then asking then to rank blacks on the same scale. This is done on a number of different positive and negative attributes and the investigator measures the extent to which whites are evaluated more favorably.

Generally, these studies show white respondents rating whites more favorably than blacks – but the gaps are small. On a scale of 0 to 100 where 0 would indicate blacks and whites being rated equally and 100 would indicate total praise for whites and criticism for blacks – whites in 2008 had a prejudice indicator of 8.7. 8.7 is a lot closer to 0 than it is to 100. It does represent mild but persistent prejudice.

The situation had improved by 2018. In 2018, the prejudice indicator was 5.4. The amount of prejudice against blacks had dropped by nearly half. 5.4 is clearly not zero. But there was a marked trend for whites to show more racially egalitarian attitudes.

Ironically, most of decline occurred during the Trump years. Democrats in particular were emphatic about emphasizing their belief in racial equality. Much of the change in their attitudes seems to have been a negative reaction to the explicit racism of the Trump administration. In contrast, Republican prejudice scores increased during the Trump administration. Republican prejudice had been declining between 2008 and 2016 – but it rebounded strongly under Trump. Overall, however, the majority of the American population is not Republican. Among Americans as a whole, prejudice diminished with most of the reduction occurring in reaction to Trump.

Whether or not people say nice things about blacks on surveys has little to do with the actual presence or absence of racist behavior. People are totally capable of talking the talk without walking the walk. Police violence against blacks was utterly unchanged during this period. Having whites think nicer thoughts did little to protect blacks from actual violence in police interactions.

Grimmer news comes from studies of employment discrimination. Social scientists measure employment discrimination through the use of audit studies. You take two actors, a black and a white. You give them nearly identical vitas and qualifications. (You can’t make them exactly the same because that would be suspicious. But you can make them very evenly balanced. One person has a GPA of 3.6 vs. 3.7. In contrast, the 3.7 guy has three months less of work history.) You send the nearly identical job candidates in to apply for the same job. You see which of the two get interviews and which of the two get hired. Generally, the white candidates are much more likely to be interviewed and much more likely to be hired despite the fact that the black candidates that are not interviewed or hired are virtually identical.

A common variation on this theme is to send employers resumés of fictional job candidates with white or black sounding names. The white candidate might be named Brett, Alan or John. The black candidate might be named Jermaine, Jamal or Tyrone. The advantage of the resumé mailout system is that you can study a much larger pool of employers than would be possible by live visits from trained actors. Not surprisingly, the candidates with black names are less likely to be invited in for interviews.

Sanaz Mobasseri of the Boston University Business School has designed her own very nasty twist on resumé studies. She sent paired sets of white and black resumés to employers with low skill entry level jobs in Oakland, California. The resumés were also randomized for having a criminal record. Half of the white resumés had criminal records. Half of the black resumés had criminal records. Criminality was indicated by having job experience working in a state prison.

Employers will often not look at candidates that have extremely serious criminal records. They will however consider mild criminal records. The fictional criminals all had about eighteen months of experience in their prison job – indicating relatively short sentences.

Mobasseri added another twist to her study. She measured the prevalence of violent crime in neighborhoods near the place of employment. For every single job, she measured the number of violent crimes both very close to the place of employment (within 250 meters) to somewhat near to the place of employment (within a kilometer). She also looked at both short-term and middle-term crime patterns.

What did she find?

1) As is the case with other audit studies, black applicants were less likely to receive callbacks than white candidates. In the Oakland sample, black candidates were 11% less likely to get an interview.

2) Being in a high crime location lowered the willingness of employers to hire blacks. It did not lower the willingness of employers to hire whites or Hispanics. Only blacks received a penalty if the employer was concerned about crime.

3) Employers in high crime areas were more than twice as likely to interview a white with a criminal record as a black with a criminal record. Whites with criminal records were more likely to be given a break by the employer.

4) Having a criminal record significantly hurt white job candidates. Employers strongly favored white candidates without criminal records. But having a criminal record had only very small effects on black employment. Essentially all blacks were treated as if they had criminal records – whether or not they really did have criminal records. The hiring rates for blacks as a whole were nearly identical to the hiring rates for blacks with a criminal record. This suggests that employers viewed being black as synonymous with being a criminal.

*  *  *

This is not an encouraging study for people who are hoping that racial discrimination is going to go away any time soon. Note that the study was done in Oakland, California. California is in theory a liberal, tolerant place – although community activists in Oakland might challenge that characterization of Oakland. That said – if talk-the-good-talk Oakland has employers that treat black candidates as criminals, what does this potentially say about employers in Louisville, Kentucky, St. Louis, Missouri or Montgomery, Alabama?

I personally am much more concerned about racist behavior than I am about racist talk. Microaggressions are a real issue; macroaggressions are far worse. Shooting people or not giving them jobs are deeply consequential. America is showing little progress on either indicator.

Is America a racist nation? Many Americans try to do the right thing; they are not racist most of the time. The majority of whites know they are not supposed to be prejudiced – and try to contain their prejudice. They do not succeed 100% of the time, but the effort is there. More action needs to be taken against police violence. More action needs to be taken against employment discrimination.

Legal enforcement against either form of discrimination has been pretty minimal. Most equal opportunity laws exempt firms of 50 employees or fewer. A huge percentage of our economy is in firms that are 50 employees or fewer. In larger firms, actual prosecutions for racial discrimination are still rare. Juridical definitions of discrimination are quite strict. This makes it hard for plaintiffs to prove their case in court.

The good news is that many Fortune 500 companies self-police on questions of racial discrimination. These companies have human resources departments that are large, professionally trained, legally sophisticated, and, often, deeply committed to egalitarianism in the workplace. These human resource departments don’t completely eliminate issues of racial inequality in hiring, job titles or promotion. But they reduce the gross racial differentials. This provides real economic opportunity and upward mobility for minority employees.

Smaller and middle-sized companies are more problematic. Not all small or middle sized companies limit black access to jobs. But some have serious issues of discrimination. The population of small and middle-sized companies is sufficiently large that there will be a non-trivial number of bad actors in the pool.

Without legal sanctions against discrimination, employers are free to continue to give preference for white job applicants. As long as they do this, black incomes will continue to be lower than white incomes and black unemployment will continue to be higher than white unemployment. Economic differentials between the races will continue – and with that will come all of the tensions of strained race relations.

Does this mean America is racist?

You make the call.