Disparate outcomes for different races and sexes can result from several causes:

• “Animus”: This is a fancy word for “hatred.” It can also include looking down on a group. Unfortunately dividing the world into “us” and “them” is built deep into the human psyche, and can especially easily target racial and sex differences. See my sermon “Us and Them,” and if you are willing to read a long, but excellent book, Robert Sapolsky’s Behave: The Biology of Humans at Our Best and Worst. I see animus (or possibly statistical discrimination stereotypes) behind many policies that are stated in race-neutral terms but have a disparate impact, such as the long-time harsher treatment of crack cocaine abuse (associated in the minds of policy makers with Blacks) and powder cocaine abuse (associated in the minds of policy makers with Whites).

• Statistical Discrimination: This is people using race or sex as a cognitive shortcut rather than learning more about each individual. A crucial fact of the social sciences is that differences within any demographic group are much, much larger than differences in the average between groups. So even when there are differences in the averages (say, because of history), it is a fool’s game to judge by group averages when one has had any chance at all to learn more about an individual.

• The Long Shadow of the Past: Intergenerational transmission of wealth matters. Intergenerational transmission of health can also be crucial. If your grandmother was malnourished, that affects the womb your mother gestated in, and therefore the womb your mother developed, which you were gestated in. And there is intergenerational transmission of education: parents who have had more education are likely to better impress on their kids the value of education. Occasionally, a parent might emphasize the importance of education precisely because the parent wasn’t able to get much, but more often, if they have less education they will, for example, read less to their kids, which communicates less importance of books. Where there is intergenerational persistence, ancestors starting in an enslaved situation is a tough starting position.

• Policies That Keep the Poor Down that Often Had Racist Roots but Many of Which are Now Usually Thought of as Innocent. Such policies make the long shadow of the past much longer. It is these policies the best candidates for systemic racism, which I will define as structures that have a racist effect even when there is no animus or statistical discrimination. (If a society were in every way an opportunity society—and one that especially does right by the children of the poor, I would not classify what remained of the long shadow of the past as systemic racism.)

Talking about systemic racism is only helpful if one points to what systemic racism is made of. Focusing on policies that all too many people think of as innocent, here are four key components of systemic racism:

1. Anti-construction rules and attitudes that make it difficult for poor people to live near rich people. (See “Why Is Housing So Expensive?”). Residential segregation makes it harder for talented poor kids to get good mentoring about professional careers, to mention one of a host of problems with it. Residential segregation makes life much harder for the poor in other ways, too. For example, residential segregation is encourages police to concentrate their policing efforts on fighting crime in the rich areas that have more political clout. If the poor and the rich lived side by side, the rich would care about crime fighting helpful to the poor next door, because that would also be protecting their own necks.

2. Opposition to charter schools and to longer school days and school years. (See “Evidence on Charter Schools: 'Getting Beneath the Veil of Effective Schools: Evidence from New York City' by Will Dobbie and Roland Fryer” and “Magic Ingredient 1: More K-12 School”. What I find remarkable about the evidence for charter schools is that for the large fraction of charter schools that are oversubscribed and choose students by lottery, one has school by school randomized controlled trial evidence for the value of each charter school relative to the schools that those who lose the lottery for that school go to. There may be some bad charter schools, and some that, while good, are no better than excellent local public schools, but we know which ones they are! And we know that the rest are valuable to students.

3. Government connivance in allowing the deadly idea that sugar is OK to continue. (I highly recommend the hard-hitting book Good Energy, by Casey Means and Calley Means.) This affects almost everyone in our society, but hits the poor much worse than the rich, who can more easily get more accurate health information.

4. Occupational Licensing. The rest of this post is about occupational licensing. Also see “When the Government Says ‘You May Not Have a Job’”.

Alex Tabarrok is one of my favorite bloggers. He wrote a February 7, 2025 review in the Wall Street Journal for Rebecca Allensworth’s new book, The Licensing Racket. The picture at the top alludes to the fact that in many states, hair-braiding requires a government license. Let me quote some of his key points:

• Nearly a quarter of American workers now require a government license to work, compared with about 5% in the 1950s. Much of this increase is due to a “ratchet effect,” as professional groups organize and lobby legislatures to exclude competitors.

• Governments enact occupational-licensing laws but rarely handle regulation directly—there’s no Bureau of Hair Braiding. Instead, interpretation and enforcement are delegated to licensing boards, typically dominated by members of the profession. Occupational licensing is self-regulation. The outcome is predictable: Driven by self-interest, professional identity and culture, these boards consistently favor their own members over consumers.

• Enforcement efforts tend to protect turf more than consumers. Consumers care about bad service, not about who is licensed, so take a guess who complains about unlicensed practitioners? Licensed practitioners. According to Ms. Allensworth, it was these competitor-initiated cases, “not consumer complaints alleging fraud, predatory sales tactics, and graft,” where boards gave the stiffest penalties.

• the AMA and the boards limit the number of physicians with occupational licensing, artificially scarce residency slots and barriers preventing foreign physicians from practicing in the U.S. Yet when a physician is brought before a board for egregious misconduct, the AMA cites physician shortage as a reason for leniency. When it comes to disciplining bad actors, the mantra seems to be that “any physician is better than no physician,” but when it comes to allowing foreign-trained doctors to practice in the U.S., the claim suddenly becomes something like “patient safety requires American training.”

• Voluntary certification can effectively replace many occupational licenses. Consider computer security, one of the most critical fields for consumer safety. Instead of requiring occupational licenses, professionals in this field rely on certifications such as the CISSP (Certified Information Systems Security Professional) to demonstrate expertise and competence.