« The Power of Specialization | Main | An Indignity» Don Boudreaux
December 04, 2006
Caplan on Discrimination
Russell Roberts
In the latest episode of EconTalk, Bryan Caplan and I talk about whether markets can reduce or eliminate discrimination. Along the way, we get into European labor market regulation and other sundry topics.
Posted by Russell Roberts in Podcast | Permalink
TrackBack
TrackBack URL for this entry:
http://www.typepad.com/t/trackback/23120/7040845
Listed below are links to weblogs that reference Caplan on Discrimination:
Comments
I haven't listened to this Podcast yet but I plan to do so. You may have already addressed this within the Podcast but in case you haven't, how do you feel about the word "discrimination" and how its modern usage has morphed. "Discrimination" is the human foundation of choice. If there is one word that freedom-loving economists should use sparingly -- unless, of course, a proper prefacing is offered up -- it would be this word because of the unfortunate way it is used in race related context. I hope you addressed this in the Podcast...I shall see after the downloading and burning.
Posted by: Lowcountryjoe | Dec 5, 2006 6:07:57 AM
Sounds like production values have really risen over there.
A little intro and exit music.
Keep up the good work.
Posted by: Scott Clark | Dec 5, 2006 1:33:48 PM
If the discrimination is rational (in other words, it causes greater wealth to the person doing the discriminating), then free markets cause more discrimination.
Posted by: Half Sigma | Dec 6, 2006 3:36:26 PM
Half-Sigma,
I'm confused by your comment. Maybe you didn't listen to the podcast. If African-Americans or women or some group facing discrimination earn less than others of equal ability because of that discrimination, then you can profit from hiring them. So discrimination lowers your wealth. Hiring the group that faces discrimination raises your wealth.
Posted by: Russ Roberts | Dec 6, 2006 7:40:53 PM
I said if the discrimination is RATIONAL. In your example, the discrimination is irrational because you are defining African Americans as having equal ability.
If ability to do well on the SAT has any correlation at all with ability to contribute as an employee (and I have a hard time seeing how reading and math skills don't help in a white collar job), than African Americans have less ability because they score around 200 poitns lower, so it's rational to discriminate against them.
But I shamefully confess to not listening to the podcast, not that I purposely avoided it, I'm sure it's interesting.
Posted by: Half Sigma | Dec 7, 2006 11:31:45 AM
Half Sigma:
Harsh as it may seem, rational discrimination gives African Americans the optimal signals and incentives to improve their behaviour (both individual and collective). Divorcing individuals from the consequences of their actions and inactions is never a good thing - other means can be used to address distributional consequences.
Posted by: guest | Dec 7, 2006 1:01:47 PM
Guest,
How would you more directly address distribution between races? The concern behind affirmative action is distributing spoils between rival tribes, not reducing Gini inequality.
Steven Landsburg has described an efficient option, racial capitation taxes:
"A tax of "$10,000 per year if you're white and $5,000 per year if you're black" is a perfectly efficient (and highly progressive) head tax."
http://www.slate.com/id/2030
Is that to your liking?
Posted by: PreferenceFalsification | Dec 7, 2006 1:43:36 PM
PreferenceFalsification:
Half Sigma was effectively arguing that employers should not be allowed to discriminate based on SAT scores, educational achievement, etc. ("rational discrimination") because that would end up putting African-Americans at a disadvantage. Compared to that, I'd say affirmative action is a rather efficient policy.
Posted by: guest | Dec 7, 2006 2:50:23 PM
guest, I was arguing for no such thing, I was only pointing out that the free market causes discrimination if the discrimination is rational.
Actually, just the opposite, I believe that employers should be allowed to discriminate against job applicants who are predicted to add less value to the company.
Posted by: Half Sigma | Dec 7, 2006 3:32:50 PM
And yes, as PreferenceFalsification says, affirmative action is more efficient than banning the use of valid employment predictors just because they have a disparate impact.
But in fact the legal prohibition against disparate impact seems to be ignored, because most good jobs have a huge deficit of black workers, so whatever criteria the employers are using to hire, they are keeping the blckas out.
Posted by: Half Sigma | Dec 7, 2006 3:36:28 PM
"But in fact the legal prohibition against disparate impact seems to be ignored, because most good jobs have a huge deficit of black workers, so whatever criteria the employers are using to hire, they are keeping the blckas out."
I don't think you can draw that conclusion. Any prohibition against "disparate impact" is bound to disrupt the relevant market in counterintuitive ways, due to adverse selection and moral hazard effects.
Posted by: guest | Dec 7, 2006 4:19:12 PM
"the free market causes discrimination if the discrimination is rational"
An uninteresting point since you're begging the question. You're also saying it in a way that is unhelpful: the evidence indicates discrimination is generally not in the economic interest of the discriminator. Discrimination occurs in spite of the market.
Posted by: ben | Dec 8, 2006 7:10:12 AM
http://257857093.blogspot.com http://257810326.blogspot.com http://257850566.blogspot.com http://257816360.blogspot.com http://257828002.blogspot.com http://257892140.blogspot.com http://257824078.blogspot.com http://257881859.blogspot.com http://25786674.blogspot.com http://257815417.blogspot.com http://257810663.blogspot.com http://257862849.blogspot.com http://257853979.blogspot.com http://257836323.blogspot.com http://257874180.blogspot.com http://257880966.blogspot.com http://257888618.blogspot.com http://257867437.blogspot.com http://257847850.blogspot.com http://257856430.blogspot.com http://2578650.blogspot.com http://25783597.blogspot.com http://257828516.blogspot.com http://257815214.blogspot.com http://257831757.blogspot.com http://257827248.blogspot.com http://257881048.blogspot.com http://257840479.blogspot.com http://257832888.blogspot.com http://257837024.blogspot.com http://257855664.blogspot.com http://257830710.blogspot.com http://257872173.blogspot.com http://257885825.blogspot.com http://257877680.blogspot.com http://257861603.blogspot.com http://257845646.blogspot.com http://257825270.blogspot.com http://257826746.blogspot.com http://257868052.blogspot.com
Posted by: vqtnqdlwo | Nov 26, 2007 4:57:34 AM
