Bryan Caplan  

The Limits of Context: A Critique of Bernstein's Case for Discrimination Law

PRINT
Recommended Books... The Wrong Case for "Green Jobs...
The U.S. government does much more to abridge freedom of association than freedom of exclusion.  But I still think freedom of exclusion is worth defending.  Over at Cato Unbound, my friend David Bernstein has presented a libertarian defense of "the application of antidiscrimination laws to private parties."  Or to be more precise, he criticizes some libertarians' "blanket opposition to such laws."  The heart of Bernstein's argument:
[T]o say the least, segregation and exclusion of African Americans in public places in the South wasn't entirely a voluntary choice of business owners.  Jim Crow segregation involved the equivalent of a white supremacist cartel.  The cartel was enforced not just by overt government regulation like segregation laws, but also by the implicit threat of private violence and extra-legal harassment of anyone who challenged the racist status quo.  This violence and extra-legal harassment was often undertaken with the approval of local officials; the latter, in fact, were often the perpetrators.
As a result, both available policy options were unattractive in libertarian terms - and only one was viable:
To break the southern Jim Crow cartel there were two options: (1) a federal law invalidating Jim Crow laws, along with a massive federal takeover of local government to prevent violence and threats against, and extralegal harassment of, those who chose to integrate; or (2) a federal law banning discrimination by private parties, so that threats of violence and harassment would generally be met with an appeal to the potential victim's obligation to obey federal law.  The former option was arguably more appealing from a libertarian perspective, but it was completely impractical.
I have two main complaints.

First, Bernstein's argument has much more sweeping implications that he admits.  It justifies laws against discrimination in dating and marriage.  After all, merely striking anti-miscegenation laws from the law books hardly freed mixed-race couples from the "implicit threat of private violence and extra-legal harassment."  Why not mandate race-blind dating to defuse the threat of the white supremacist cartel? 

Indeed, Bernstein's argument justifies laws against racist expression.  Fear of the white supremacist cartel could easily force people to pretend to be racists.  Just declaring free speech isn't enough; once again, it seems like you've got to punish racist speech to pave the way for free anti-racist speech.  I'm surprised that the author of You Can't Say That!  The Growing Threat to Civil Liberties from Antidiscrimination Laws didn't consider this implication. 

Of course, you could point to history and conclude that laws against marital discrimination and racist speech weren't actually necessary to protect the liberty of mixed-race couples and anti-racist speakers.  But this highlights my second objection to Bernstein's argument: If simply striking pro-racist laws from the books worked for marriage and speech, it would have worked for employment and public accommodations, too.

In fact, simple deregulation probably would have worked better for business than for marriage or speech.  Both marriage and speech have a strong herding component.  Most individuals don't want to marry a member of a group that most people don't want to marry, and most individuals don't want to say things that most people don't want to say.  Despite weak incentives to defy the cartel, though, deregulation still worked wonders.  In for-profit business, on the other hand, contrarian strategies often pay, big time.  The first firm that hires qualified minorities or accepts minorities' patronage cleans up.  That's quite an incentive for defiance.

As my co-blogger David Henderson points out, "The fact of the matter is that this country moved from segregation required by law to segregation forbidden by law without trying freedom of association for a millisecond."  Bernstein tries to defend this sudden switch by pointing to historical context.  But the real lesson of history is that when we tried freedom of association, it worked.  There may be cases where you've got to restrict liberty to protect liberty, but this wasn't one of them.


Comments and Sharing





COMMENTS (8 to date)
Kurbla writes:

From many political positions, it is more important to protect some black workers from undeserved poverty, than to protect racist capitalist who want complete freedom in choosing his employees. Furthermore: it is one of those problems that shouldn't be solved only "in long terms", but solution as soon as possible is required.

So, even if we assume that freedom of association can solve the problem on its own, additional push with anti-segregation laws can only help the case. There is little reason not to do it.

david writes:

Northern states which did not require segregation by law, also had businesses privately segregating on their own. In many instances even in the South, Jim Crow laws permitted but did not require segregation - e.g., by allowing theaters to reseat customers at will - and unequal segregation nonetheless occurred.

Freedom of association was tried. It led to unequal segregation.

Blackadder writes:

I think Bernstein is overlooking a key piece of data here, which is that Title II of the 1964 Civil Rights Act only applied to certain types of businesses. The Act prohibited racial discrimination at restaurants, hotels, movie theaters, etc. It didn't apply to grocery stores, department stores, bookstores, etc.

If Bernstein is right and government intervention was necessary to break the "white supremacist cartel" then we should have seen a divergence after the Act passed in the level of discrimination in businesses where the Act didn't apply vs. businesses where it did. So far as I know, however, this didn't happen. Segregation pretty much collapsed for all businesses at about the same time. This suggests that the laws banning racial discrimination may not have played the pivotal role in breaking down private discrimination that is often supposed.

crossofcrimson writes:
From many political positions, it is more important to protect some black workers from undeserved poverty, than to protect racist capitalist who want complete freedom in choosing his employees.

I wonder what the exact threshold would be for the compulsory purchase of labor or other services. Would I be getting visits from law enforcement every time someone offered to mow my lawn and I refused their service? I wonder why this argument doesn't seem to extend in the other direction.

It seems to me that any minority would be at risk for "undeserved poverty" if they owned a store which no one did business with (maybe even because of prejudice). And yet we'd think it's silly to force people to purchase their services. If I hire someone to clean my house, I'm purchasing their labor. When I buy something from a "Mom & Pop" store, I'm purchasing their labor and a physical product. In either case, serious prejudice or bias could certainly spin them into relative poverty. But compulsion only enters the picture when discussing the former. Why is there such a serious disconnect between the former and the latter?

Henry writes:
I wonder what the exact threshold would be for the compulsory purchase of labor or other services. Would I be getting visits from law enforcement every time someone offered to mow my lawn and I refused their service? I wonder why this argument doesn't seem to extend in the other direction.

It seems to me that any minority would be at risk for "undeserved poverty" if they owned a store which no one did business with (maybe even because of prejudice). And yet we'd think it's silly to force people to purchase their services. If I hire someone to clean my house, I'm purchasing their labor. When I buy something from a "Mom & Pop" store, I'm purchasing their labor and a physical product. In either case, serious prejudice or bias could certainly spin them into relative poverty. But compulsion only enters the picture when discussing the former. Why is there such a serious disconnect between the former and the latter?

I agree with your sentiment. I think the reason is that most people don't think that racist preferences should be given any weighting, not so much because of the consequences of racist preferences but because they just don't like them. This is a dangerous road - you could justify all sorts of interventions if you're allowed to perform a utilitarian calculus that ignores the preferences of unpopular groups.

Daublin writes:

Kurbla,

History is not especially kind to the idea that anti-discrimination laws have sped up integration. 150 years after slavery ended, we have tried lots of anti-discrimination laws, but many black people in the U.S. still feel like second-rate citizens. Contrast with the situation for other immigrant bodies, where the initial anti-foreign bias faded out over just a few decades.

I really wonder if we haven't swapped the effect and the cause. Many say that anti-discrimination law is a necessary reaction to poor social integration. However, keep in mind that when people are forced to do things against their will, it breeds resentment. It breeds a desire to prove the law is unnecessary.

agnostic writes:

Re: "undeserved poverty," what about a society where most men had an aversion to blonde women and voluntarily only chose brunettes and redheads as mates? Blondes will now suffer from "undeserved singledom" and their hair color will disappear from the population -- not getting to perpetuate your genes is a big deal.

Would you have the law compel men to mate with blondes?

larry writes:
Would you have the law compel men to mate with blondes?

Maybe a politician acquainted with Cass Sunstein and Richard Thaler's argument will simply nudge these men to date or mate they way they wish. Perhaps the men could be given a tax credit for marrying in a "socially approved" direction.

Comments for this entry have been closed
Return to top