David R. Henderson  

What You Mean "We", Scott Alexander?

PRINT
Re-regulation in the UK... Good for Jared Bernstein...

My co-blogger Bryan Caplan, as well as many other bloggers and friends, have recommended Scott Alexander's (that's not his real name) blog, slatestarcodex. I have had it on my feed for a number of months now and, like them, find it refreshing and informative.

Which is not to say that I think he understands economics as well as economists do. Why should he? It's not his area of expertise. On July 22, he wrote:

We don't trust the free market to necessarily preserve racial equality - that's what anti-discrimination laws are for. We don't trust the free market to necessarily preserve worker safety - that's what OSHA and related regulations are for. We don't even trust the free market to necessarily preserve fire safety - that's why federal inspectors have to come in every so often to make sure you're not secretly plotting to let your employees fry.

Now I don't trust the free market to preserve racial equality either. You can't preserve what has never existed. But what I do trust the free market to do is to undercut racial discrimination--by making those who discriminate pay for discrimination even if a government can't catch them. There's a large economics literature on this. And I do trust free markets more than I trust government. After all, governments in the South, in South Africa, and in Nazi Germany, to name three, had pro-discrimination laws. Here's what Linda Gorman writes in "Discrimination" in The Concise Encyclopedia of Economics:
Many people believe that only government intervention prevents rampant discrimination in the private sector. Economic theory predicts the opposite: market mechanisms impose inescapable penalties on profits whenever for-profit enterprises discriminate against individuals on any basis other than productivity. Though bigoted managers may hold sway for a time, in the long run the profit penalty makes profit-seeking enterprises tenacious champions of fair treatment.

And worker safety? Read W. Kip Viscusi's piece "Job Safety" in the Concise Encyclopedia. Here's the opening paragraph:

Many people believe that employers do not care about workplace safety. If the government were not regulating job safety, they contend, workplaces would be unsafe. In fact, employers have many incentives to make workplaces safe. Since the time of Adam Smith, economists have observed that workers demand "compensating differentials" (i.e., wage premiums) for the risks they face. The extra pay for job hazards, in effect, establishes the price employers must pay for an unsafe workplace. Wage premiums paid to U.S. workers for risking injury are huge; they amount to about $245 billion annually (in 2004 dollars), more than 2 percent of the gross domestic product and 5 percent of total wages paid. These wage premiums give firms an incentive to invest in job safety because an employer who makes the workplace safer can reduce the wages he pays.

Bot the Gorman and the Viscusi articles are worth reading it total.


Comments and Sharing


CATEGORIES: Labor Market , Regulation




COMMENTS (25 to date)
E. Harding writes:

I pointed out to Scott that insurance company inspectors are often a far stronger incentives for workplace safety than OSHA ones.
"Now I don't trust the free market to preserve racial equality either. You can't preserve what has never existed."
-I find that statement brilliant.

Michael Strong writes:

Yes, it is pity that while Scott Alexander is brilliant in many respects, he remains attached to many progressive myths regarding markets. Thanks for calling attention to a few of them.

Handle writes:

Scott's 'we' clearly means his target audience who share those concerns as their center-mass views, and he is using the rhetorical technique of trying to encourage this audience to extend the principled rationale they use in those other cases when important values are at stake to the realm of civil tolerance for the holders of eccentric, heterodox, or contrarian opinions.

Many Libertarians, for example, have publicly abandoned the doctrinaire libertarian position of absolutist free-association in all matters, and have embraced certain anti-discrimination rules against private actors which are policed by and with all the coercive powers of the State.

I don't know if that's your view, but the usual rationale of those libertarians who support punitively curtailed association rights is that otherwise it might be possible for certain racial minorities to be unfairly and completely shut out of entire sectors, with non-governmental social forces causing society to get stuck in some kind of self-sustaining bad equilibrium.

And Scott is making the point that, especially for anyone who embraces that kind of logic, they should now apply it to another major and related social problem to which he is attempting to alert them.

The bottom line is that the evolution of modern modes of internet discourse, and coordinated politically-motivated action and aggressive abuse online, has created a new situation that can completely excommunicate the holders of an ever wider swath of opinion from the very possibility of holding any kind of prominent position in the public or private sector. Whatever one's opinion of the merits of the case, many people are coming to realize that the kind of mob-action that forced Brendan Eich out of his job is indicative of a real, disturbing and growing problem. And one that existing legal policies cannot mitigate.

The opinion-policing mob is empowered as never before, and the first amendment isn't enough. Most Libertarians, for example, support the immunity that tenure provides to vested academics and federal judges, precisely because they understand that, civil right or not, certain views simply could not be aired or decisions made without some kind of insulation from negative social consequences - and specifically - the continuation of one's salaried employment.

So, we can quibble with all his examples. But I would encourage you to engage with and address the overall theme and concern he's expressed in the piece, which seems to me to be a genuinely concerning phenomenon in our time, and one for which Libertarians have, as of yet and as far as I can tell, offered no good and effective solution.

It is of course understandable for any Libertarian to be reluctant to grant the state yet another class to which it can apply and expand its anti-discrimination role. However, if the doctrine is to be compromised for anti-discrimination reasons, then perhaps the most justifiable application of these powers (and certainly within the self-interest of heterodox Libertarian scholars themselves) would be to ensure that contrarians in the ordinary public - not just scholars and judges - are granted the equivalent protections of 'tenure' and are free to express their views and honest sentiments and openly support political causes without the chilling fear of losing their livelihoods for reasons of manufactured 'public outrage' and unrelated to their performance on the job.

pgbh writes:

It's basically a Hansonian "we", i.e., "this is how the majority of people think." It's not necessarily an endorsement of a particular point of view.

Unlearning writes:

It always bothers me when libertarians use 'economics' as a shorthand for 'my politics', for two reasons:

(1) Mainstream economics does not agree with libertarian politics. Economic theory includes a lot of market failures which imply the market won't automatically reward any sort of positive behaviour. In fact, most economists seem to support anti-discrimination laws:

http://econfaculty.gmu.edu/klein/PdfPapers/Klein-Stern%20AJES%202007.pdf

(Table 1)

One example in the literature which speaks specifically to discrimination laws is Akerlof's Identity Economics.

(2) Critics of economics see this and believe you. They then attack economists as being free market. Economists retort that they are not (see link above). Another level of jadedness is added to a debate which really doesn't need any more jadedness.

The fact is that economics can be used to support a wide variety of political positions. In the 30s the basic model was used to defend central planning (and Stiglitz attacked Socialism using his criticisms of this basic model). There's economics which supports anti-discrimination laws and worker safety regulations and there's economics which doesn't. At best the view that economics is automatically anti-government is dated.

I really wish people would stop misusing economics as a way to try and add credibility to their political views.

Yaakov writes:

The examples brought by Scott are around because after many years of trial and error, they were found as the best ways to convince the public to allow government intervention. To combat that, we need to address not only whether the free market could do a fair job, and not only bring examples of extreme government failure, but wee need to present arguments why the current government policies achieve worse outcomes than free market outcomes.

That seems to be a tough problem, as most of the public has never dealt with government safety regulations* and has very little tolerance of people who have views they do not like.

* Several years ago there was a fire in a supermarket in Israel and four workers were burnt to death. The set-up was, at least in Israeli standards, a perfect example of a fire hazard that should never have existed (a second floor wooden balcony with only one exit loaded with flammable materials). The place was completely authorized by the fire department, probably based on bribes. The regulations are so strict, that people do anything to get around them. Schools are exempt from fire safety certification by the fire department (they instead have more lenient rules), because it is practically impossible to get such certification.

John Becker writes:

Scott is absolutely right that we, or at least most people, don't trust the government to do these things and that those organizations exist because of that distrust. Most people do not trust markets and I believe personally that about 90% or more of the population could not comprehend the economic way of thinking not matter how well it was explained to them. The interesting part about this and the biggest obstacle in the world to human freedom is that most people seem innately hardwired to distrust commerce across all times and cultures.

Rich Berger writes:

My understanding is that modern day non-discrimination laws started in the 1960's with prohibition of race-based discrimination. These laws have been extended in various areas (disability, gender, etc.) since then on the assumption that all discrimination is wrong and the only way to reduce if not eliminate it is through such laws.

Even assuming that these laws achieved their intended results, what have been the side effects? We now have an environment with a hyper-sensitivity to any imagined discrimination can result in lawsuits. The fight over SSM commerce and religious values is just starting. More pernicious still is the heightened sense that racism is pervasive and the key obstacle to black progress. What could be worse for a human being to give him a ready made excuse that he cannot succeed because people are against him? People may be against you because of the color of your skin or the way you talk or they may simply be indifferent to you. Do you have to be loved to succeed? Isn't your own power to shape your life far more important than whether others actively support you?

History is what it is and we cannot rerun it to see if a different approach would have been better. If nondiscrimination laws had not been enacted, we might still have white-only clubs, or (God forbid) men-only clubs, but many other organizations which would gladly be open to everyone. I think open-mindedness would have been much broadly based and firmly founded if it had been allowed to grow under freedom rather than compulsion. The urge to stamp out "discrimination" has trampled the right of free association and will soon be attacking the freedom of religion.

Relative to David's post, my understanding is that the Jim Crow laws in the south were an attempt to stop private businesses from serving all customers, rather than just whites. Likewise minimum wage and union mandates were used against black workers. And, of course, slavery itself required government enforcement.

David R. Henderson writes:

@unlearning,
Your whole comment is based on a misunderstanding. My point in the post is that economic analysis shows that people pay a price for discriminating. I did not say, as you infer, that most economists are against anti-discrimination laws.

Sieben writes:

The discussion over race is entirely unprincipled. Libertarians should stand up loudly for people's right to discriminate. Not apologize and say "well, there would be less discrimination on the market...".

What if they were right? What if libertarianism would lead to a white-male patriarchy? Would you abandon your ideals and call for compromise?

Nick writes:

@David R. Henderson

There is also an economic literature of markets in which people rationally discriminate against people of certain skin colors. See, e.g., here for an overview.

I took Unlearning's point to simply be that economic theory is not some sort of smackdown in favor of the thesis that free markets punish discriminatory behavior, so when you say '[I don't] think he understands economics as well as economists do,' you are perhaps giving the mistaken impression that 'markets punish racially discriminatory behavior' is a conclusion endorsed by the economics profession, as opposed to one of several explanatory models.

Philo writes:

@ Handle:

You put into the mouths of "many libertarians" an argument based on the fact that "it might be possible for certain racial minorities to be unfairly and completely shut out of entire sectors, with non-governmental social forces causing society to get stuck in some kind of self-sustaining bad equilibrium," without government enforcement of anti-discrimination laws. But whatever public policy is adopted, "it might be possible" for something bad to happen (indeed, something bad is *bound to happen*!)--aren't there bad things that might happen if government is charged with enforcing anti-discrimination laws? A *decent argument* for or against a policy would have to provide estimates of *magnitude* (how bad?) and *probability* (how likely?) for that policy *and for its most plausible alternatives*. The argument of your "many libertarians" does not rise even to the level of *decent*.

MikeDC writes:

The discrimination point works much better than the workplace safety point because the compensating wage differential is mostly based on the supply of labor. If you've got an impoverished and hungry subsistence or near subsistence level population, you don't have much have to worry about their safety. You just replace them if they get hurt or die.

So I take that as a fairly legitimate moral limitation. Maybe one of the few where a free market doesnt push in the right direction.

On the other hand, discrimination is always costly in a free market because it's ultimately a matter of the buyer demanding to purchase from a constrained supply of goods.

Anonymous writes:

@Philo:

I responded to a comment on similar lines on Scott Alexander's blog, but I'll repeat it here. This is a good argument against lots of government interventions, but I have doubts that it works very well for anti-discrimination laws. The problem is that, while something like the FDA can and does make bad decisions that everyone agrees are bad, it's hard to think of comparable outcomes of anti-discrimination laws that most people will consider as bad - particularly in the eyes of anyone who leans left. The underlying assumption is that discrimination is always wrong, that no rational employer would ever discriminate. So any libertarian who wants to remain respectable in the eyes of people other than far-right extremists is stuck arguing, not just that markets will weed out irrational discrimination, but that markets will weed out ALL discrimination.

In light of this, it is difficult to convincingly argue that anti-discrimination laws might ever be harmful - because they serve to outlaw something that nobody is willing to defend. You can easily point to bad outcomes of the FDA or of minimum wage laws - situations where unapproved drugs or low-wage jobs are a good thing, banning them is a bad thing. But nobody is prepared to defend situations where discrimination is a good thing.

R Richard Schweitzer writes:

E Harding brings up a reminder;

Manufacturers Mutual Insurance Company.

A fascinating study in cooperative competition.

Andy Hallman writes:

As a few others have commented, it seemed Scott was using "we" the way Robin Hanson uses it to mean "things my target audience believes" and not necessarily "things I believe." Perhaps Scott will see this post and clarify his position in the comments.

David R. Henderson writes:

@Andy Hallman,
As a few others have commented, it seemed Scott was using "we" the way Robin Hanson uses it to mean "things my target audience believes" and not necessarily "things I believe." Perhaps Scott will see this post and clarify his position in the comments.
Perhaps. And, if so, that makes the title of this post even more appropriate.

David R. Henderson writes:

@MikeDC,
The discrimination point works much better than the workplace safety point because the compensating wage differential is mostly based on the supply of labor. If you've got an impoverished and hungry subsistence or near subsistence level population, you don't have much have to worry about their safety. You just replace them if they get hurt or die.
But the employer will worry about it to the extent the employees do. Which means that legislation requiring more safety than workers demand, when workers know the risks, will hurt workers.
This might not be obvious and so you have motivated me to do a blog post on it soon.

MikeDC writes:

@ David R. Henderson,
I look forward to reading it. I'll confess I'm still skeptical because I think in the case of a big pool of subsistence level employees, it's going to be difficult for the employee to show they "care". Of course he cares about death or injury, but if the labor market is sufficiently large that people have no other choices then it seems an across the board legislation to improve safety might be beneficial.

Oddly, the employer sometimes "cares" more than the employee if you consider examples of slavery. Slaves, by definition, couldn't demand better conditions, but in the US at least, they were still costly laborers. The price of a slave in the US was much higher than in some other places (Cuba or Haiti). With American slavery, slave owners showed some concern for their slaves because ultimately the slave was a costly and valuable resource. If I understand correctly, mortality rates among slaves were much higher in countries with greater supplies though. So I think the problem comes from the supply of labor being so large that the equilibrium wage (whether it accrues to the worker or not) is to low for anyone to "care".

My response as a libertarian wouldn't be to fight this fight, but to argue that societies seem to almost never reach this point without capricious government regulation of labor and property. Such as slavery. That is, if people had basic freedom and stability, they would quickly rise above subsistence level and demand better working conditions (or more pay for dangerous conditions).

john hare writes:

I heard about an unintended consequence of anti discrimination recently. A black man and a white man apparently equally qualified applied for a state government job. The white was hired because he could be fired without consequences if he failed to work out while firing the black 'might' result in problems if he didn't work out.

So the black guy doesn't get a chance because it's too hard to correct a possible error with all the current laws and incentives.

Sooner or later people have to be told that you don't build successful countries, companies, teams, or even families with victims. Winners are not victim worshipers.

Anonymous writes:

Hell, I'll reiterate my other reply to that post over here too, why not.

I think the reason I found that phrasing mildly irritating is that it implies that the laws that exist do so for a good reason, as though some sensible person has sat down and determined what the best laws are, and more or less everyone has agreed. In reality, the laws that exist do so because of the political process - they were popular enough at some point with some group or other that it was politically advantageous to enact them, and there hasn't been a strong enough resistance to them since then to prevent inertia from keeping them in place.

This does not at all justify phrasing such as "We have this law because we expect it to produce this good outcome, or prevent that bad outcome". Not only is there no 'we', as different groups of people will have different thoughts on whether the law is good or bad, but there is no reason to expect the correct group to either be larger than the incorrect group, or to have gotten its way and had its view of the law implemented. You might personally like the law, might think it is justified for certain reasons, but that is an entirely separate question to the one of why we have the law.

"We have this law because..." is patronizing appeal to popularity; "We have this law, and I think that's a good thing because..." is the honest version.

David R. Henderson writes:

@Anonymous,
"We have this law because..." is patronizing appeal to popularity; "We have this law, and I think that's a good thing because..." is the honest version.
Really good point. Thanks.

Roger Sweeny writes:

But the employer will worry about it to the extent the employees do. Which means that legislation requiring more safety than workers demand, when workers know the risks, will hurt workers.

Thomas Schelling had a good article on that, reprinted in Choice and Consequence (1985).

marathon man writes:

@ anonymous

In light of this, it is difficult to convincingly argue that anti-discrimination laws might ever be harmful - because they serve to outlaw something that nobody is willing to defend.
Ah, the unseen! Ask any employment lawyer or CFO how much the average company pays to prevent, defend, negotiate and settle anti-discrimination claims which are often filed with little to no merit. Consider what the average EEOC investigation costs in overhead or outside counsel spend. How many unproductive employees are retained for fear of a Title VII claim?

Anti-discrimination laws are no less a drag on employment levels, productivity and profitability than any other government barnacle – OSHA, minimum wage, ADA, FSLA, NLRB, ICE, Sarbanes Oxley, etc.

Anonymous writes:

@marathon man

I do wonder how many of those employment lawyers or CFOs are willing to publicly state their opposition to anti-discrimination laws. I'd expect that the few who would, would couch them in ten or twenty justifications that no-it's-not-that-we-think-discrimination-is-okay-it's-just-that-laws-against-it-have-negative-consequences-honestly-I'm-not-sexist-honest. This is unlike with, say, argument against the FDA, where part of the argument is not just that the law will be misused, but that the thing the law intends to ban is in fact really not bad at all, that people very often want unlicensed drugs for perfectly good reasons.

It can be difficult to argue in defence of the thing that is banned in other cases like this, such as explaining why working below the minimum wage might very often be a perfectly okay choice. But these cases are nothing like as hard as standing up and seriously arguing that it's often rational and fine for employers to discriminate based on something like gender. I'm not entirely comfortable doing it myself, and I'm here posting anonymously on an blog run by libertarian-leaning economists. Someone who wants to make this argument in real life had better be a top-class debater and be prepared to spend the rest of their life defending their position.

Comments for this entry have been closed
Return to top