Bryan Caplan  

Why Encourage Discrimination? The Case of Mandatory National Origin Labels

PRINT
Political Mood Cycles... The Pelican Brief and the Peop...
You've probably noticed that imports are labeled by national origin.  This is usually required.  The most obvious effect of such regulation is to slightly disadvantage foreign producers by raising their cost of production.  But the only slightly less obvious effect is to reduce consumers' cost of discrimination.  If you have to do your own homework to discover products' national origin, you'll probably accept your ignorance and decide based on price and apparent quality.  If regulation imposes national origin labels, you might play favorites instead.

Once you start throwing around the charge of "discrimination," however, it's easy to see that there are two distinct kinds of national origin discrimination that consumers might commit. 

One is standard taste-based discrimination - tipping your personal scales against products from countries you don't like.  When I was a kid, crusty American war veterans kept the fire of World War II glowing by refusing to buy goods from wicked Japan. 

The other, however, is statistical discrimination - rationally using national origin as a signal of product quality.  Plenty of people still aren't fond of the Germans, but still eagerly buy products with the "Made in Germany" label because they correctly identify German manufacturing with quality.

Given Americans' near-religious objections to both taste-based and statistical discrimination, it's puzzling that the law doesn't merely allow, but actively facilitates them.  Most Americans, I suspect, would defend these laws with arguments that would make them cringe in other contexts.  E.g. "Consumers have a right not to support countries of which they don't approve," or "Knowing what country a product comes from helps consumers make a wise choice." 

Challenge: Is anyone willing to often even a semi-plausible economic argument in defense of mandatory national origin labels?


Comments and Sharing





COMMENTS (27 to date)
Abhi S. writes:

I think the interesting question is how powerful an effect does taste-based discrimination produce? For example, I wonder, by what proportion did sales for goods labeled as originating from predominantly Islamic countries fall after 9/11? How does the distribution of these taste-based discrimination effects mirror the distribution of soft power throughout the world and its countries?

Jacob writes:

My argument is two-pronged: information has costs and benefits, but I will address them in reverse order.

First, the benefits. Information economics often suggests that a lack of information can easily make for a less efficient market. Many people would probably prefer not to purchase a watch, for example, if they had access to certain information, e.g. that the watch was made in Apartheid South Africa (ASA). But if that information is not readily available or too costly to obtain, the result will be adverse selection, which I will argue is inefficient.

This lack of information is inefficient because while there is no such thing as demand for ignorance, there is plenty of supply. To continue with the South African watch example, it would be wrong to conclude that because consumers would rather buy a watch of unspecified origin than one labelled "made in ASA," that consumers demand ignorance, or that they derive some utility from their ignorance. They WOULD rather know, but they have no choice but NOT to know, because there is no incentive for white South African watch manufacturers to label their products accurately, and in fact there might be a disincentive for supplying such information. Hence, due to a lack of information, supply does not meet demand in the free market for watches made in ASA. This S-D mismatch is by definition inefficient.

You might believe that ignorance is bliss, which would be fine for you, but I prefer markets with complete information. I think that most people would agree with me. So that's my first prong.

My second prong deals with the costs. Complete information would of course be nice (especially for economists who like to model based on the assumption of complete information [*cough* ARNOLD KLING *cough*]), but in reality information has a cost. What is the cost associated with the labeling of national origins? It is probably very low. What does it cost, really, to inform consumers that their watch is being used to finance segregation in SA--the ink required to print the information on the box? The time it takes the designer to get the label on there? The marginal increase in law enforcement required to enforce the law? These costs seem pretty neglible to me.

There is of course the cost to white South African watch producers--but that is like the cost of banning leaded paint to sellers of leaded paint. The benefit to consumers outweighs those costs, as long as we believe that (informed) markets are efficient. Which (mostly) we do.

If private markets tended towards complete information, there would be no role for the government to mandate labels of national origin. But as my Apartheid-watch example tries to show, that is not the case. There is plenty of room for governments to require labels of national origin if the benefits outweigh the costs, which I think is likely. What is the benefit of knowing whether your watch is financing segregation, or your diamonds fueling a civil war, or your olive oil being made by Israeli settlers, or your (possibly leaded) toys being made in China? The marginal gains from this kind of information are real, and the costs are... petty.

So there's my argument. I hope you like it.

PS. I think a much better argument could be made against requiring certain types of financial disclosure and accounting standards. But that is for another post...


Matt Flipago writes:

I agree that Jacobs case is a possible position, although all information like that is possible to find, so it's really only a matter of cost for efficiency, so consumers can efficiently perform boycotts of certain nations that they disagree with, or the opposite, such as buying Haitian products. And I think a fair amount of people don't trust foreigners, or they do not trust foreign safety standards(China). Coupled with attitude of hating foreigners, believing Americans deserve those jobs, or believing manufacturing sites not from the developed world are sustained through slavery. Isn't it possible that lack of competition for low skilled workers, could help out low skilled workers more then giving them money directly?

Still I don't see many people giving good reasons why I should pay for someone else to get information they were too lazy or unwilling to pay for. How do we know if it's worth it, when everything is taken into account?

George J. Georganas writes:

Assume racist preferences in a consumer. Labelling makes it easier for him ("or her" deliberately not included) to choose products produced by his favourite race of producers, while making no one worse off.
"... Americans' near-religious objections to both taste-based and statistical discrimination" is not, in my experience, a realistic "given". It is an empirical matter to decide whether there are enough racists around to justify economically the small extra cost of labelling. I would think there are more than enough to meet the Kaldor (I think we called it thus) criterion.

Doc Merlin writes:

Americans aren't against statistical discrimination in general, they are against statistical discrimination based on race for specific races.

Leo writes:

Maybe this is a useful way of promoting free trade policies because consumers will realise how much of their produce comes from other countries and also realise that the ones with "made in USA" often cost more. This might induce consumers and voters to be less sympathetic to protectionist and anti-free trade policies which would obviously have economic benefits.

Amaturus writes:

When I studied abroad in Switzerland, I noticed that both restaurants and the student cafeteria would advertise where the meat in dishes came from. I don't think it consciously changed my choices, but I did find it interesting that so much chicken came from Brazil.

Pedro writes:

Jacob: You forgot about the incentives facing the competitors of South African watch producers. If consumers have a preference for the products of free labor, competitors have an incentive to let consumers know where their watches come from.

SI writes:

Because people infer national quality from product quality, mandatory labelling encourages nation-states to enforce quality standards [on exports].

Such standards of quality are somewhat ambiguous in terms of economic benefits: while they may remove from the market low-quality items in demand, they do also reduce the customer's transaction cost (gathering information on quality, checking whether a given product is compatible with another etc.).

It can, at least, be argued that some cheap-product producers are performing a kind of rent-seeking when they produce extremely low-quality products that resemble normal products. Imagine a fairly cheap, quality watch, perhaps at $30, that breaks down after 5 years. Then some foreign company copies your product, but their shoddy quality makes it break down in 2 years. They sell it for $25, and customers must research for 20 minutes to discover the difference. Then you would be out of business unless people value their time at less than $15/hr.

This kind of rent stems from that transaction/information cost.

Steve Sailer writes:

We should have more detailed labeling laws describing value added by nationality. Why get tricked into paying for, say, Japanese or Italian workmanship when it's mostly cheap Chinese labor?

Jody writes:

Ignoring cost of labeling, national origin allows better preference maximizing by consumers (for those that care; for those that don't care, they're indifferent, again ignoring labeling costs).

We don't care that a consumer prefers a red widget to a green widget, so why should we care that a consumer prefers country A to country B?

Or to turn the question around, what would be a semi-plausible economic argument in defense of mandatory widget color labels? Now apply the same argument to national origin labels.

Blackadder writes:

Leonard has the best argument, though of course it is less economic than political.

floccina writes:

Additionally those labels can be quite deceiving because many things have parts and inputs from many countries.

Lori writes:

It sounds to me like your agenda is to score points for neoclassicalism by claiming you can have your information asymmetry, and your prices incorporating all information, too. Operating on less information feels disempowering, whether it really is or not. Any number of movements are seeking transparency, and want supply chain information in general to be non-proprietary, or at least effectively leaked to the public, or voluntarily disclosed.

Then again, maybe you're just trying to be snarky by accusing consumers of discriminatory intent. Is this to criticize consumers as discriminatory or to defend discrimination as OK? I don't subscribe to the ideology that every private sector entity is an 'individual,' so I'm OK with different standards applying to products, businesses and actual individuals (human beings). The latter shouldn't be discriminated against, and I see plenty of apologetics on behalf of discrimination against people on this rather rabid blog.

Hyena writes:

I offer three:

(1) Different nations have different regulatory schemes and technical standards which all products will be influenced by.

(2) Identifying the country of origin allows a consumer to make judgments about the distance a product has traveled (important lately in food).

(3) Part of the product is sometimes its origin, a connection to a group of people and a place. Likewise, firms may wish to suggest their origin to cash in on this; mandatory labeling circumvents this fraud.

Ben writes:

Looks like I got here too late. I think Hyena's third suggestion is very believable, particularly given the prominence of cultural symbols in marketing (think Cyrillic characters on cheap American vodka, or HaraJuku logos on T-shirts from Bangladesh).

Claire writes:

How can the consumer know standards w/ no access to data when there are varying producers? Origin labels are as vital as ingredients labels. Informed choice is a part of discerning when it comes to produce. Some people don't care so long as it's cheap but others seek out ethical producers for quality!

Don writes:

I'll give you one you've maybe never thought of. I work as a lawyer in a consumer products business. Hyena's #1 is important: it tells consumers what product safety standards (for example) have been applied to a product. But an even better one is: you can use it as evidence of a passing off or related lawsuit against a counterfeiter.

The company I work for manufactures all of its goods in the USA. These goods are often counterfeited in China and imported. Importers of the counterfeit product, playing ignorant and hoping we'll fall for it, ask how they were supposed to know that the products they bought were counterfeit. At which point I can point to the origin labeling notice and ask them whether they really believed they were buying authentic goods marked "Made in USA" but imported from China. This is usually good enough to shut down that argument, and we can use that to leverage them into ceasing import of counterfeit product. In my mind, that's a win.

Ted writes:

Uh, the reason is mostly legal and has nothing much to do with discrimination.

They are useful in the enactment of countervailing duties, anti-dumping provisions, preferential treatment based on bilateral trade agreements etc. Also, under the endlessly complicated WTO agreements, ROOs are mostly used for light-touch protectionism and / or preferential treatment to certain countries.

BZ writes:

I also want to throw in with Hyena's third suggestion.

National Origin labels are how I found out that English Muffins are made in California, Hagen Daaz in New Jersey, and Dixie cups in Connecticut (think about it).

Hyena writes:

So... do I get a prize?

Matt Flipago writes:

LOL, SI did the math wrong in support of labeling laws. A watch that last 5 years for 30 dollars, is 60 dollars in 10 years, the other watch is 125 dollars in 10 years, that is 65 extra dollars, not to mention you will be without a watch as numerous points. if you only work 20 minutes, which seems like an extremely long time on average researching the difference, which means you would only research on the internet at a wage of 185 dollars an hour, if it took 4 minutes, your approaching 1000 dollars an hour. SI, this is not mocking you it's just the number 15 dollars an hour and 185 are not in the same order of magnitude.

Robert Johnson writes:

Isn't information asymmetry a potential cause of market failure? Shouldn't greater transparency help markets to clear? Seems like there's a textbook example having to do with lemons in the market for used cars...

Hyena suggests why the information is important.

Lori writes:

Robert Johnson: I think our host is advancing the argument that equilibrium-seeking is an infinitely more powerful 'force of nature' than information asymmetry. If he can score an additional zinger like boycott=racism, that's the icing on the cake. I don't buy it, but that seems to me to be what the economists mean by 'strong efficiency.' Another agenda I sense is the centrality of self interest. In other words, my guess would be that our host would say a consumer taking anything other than the self-interested consideration of 'bang for the buck' into consideration is at best a fool, and getting less then their money's worth. It stands to reason by that logic that information about anything other than product price or performance is irrelevant. What rational consumer would even be curious about such irrelevant things? Likewise investors would be fools to stray from going for the maximum 'buck for the buck,' or ROI, hence the irrationality and utter futility of the politically correct mutual funds we've all heard of.

In short, I suspect he's ridiculing the hunger many of us have for a different world being possible.

David C writes:

I know that in guitar buying, certain countries have better reputations than others. Some companies will have manufacturers in multiple countries. An American Fender is worth significantly more than a Mexican Fender. The older Made in Japan BOSS pedals are worth more than their modern Taiwanese counterparts. This effect is even more pronounced in Japan where American guitars have skyrocketed in value. Japanese guitar manufacturers have an easier time selling their products in America than they do in Japan.

This is an extreme example. If the law was eliminated, I'm guessing a lot of guitar buyers would demand proper labeling from the manufacturers with factories in multiple countries.

Musicians are probably guided primarily by marketing and word of mouth and hearsay. Most striking is the large market for overpriced guitar cables despite their offering no discernible value.

Statistical discrimination is exactly this plausible argument for mandatory labeling.

And if people want to pay extra for opportunities to discriminate, why not sell such opportunities?

It's good for both bigots (who pay extra but get what they want), and non-bigots (who buy discriminated product more cheaply).

Lori writes:

It's not fair to assume bigotry is the motive for locational preferences among consumers. I believe it's much more often a desire to exercise some 'moral authority' over the business community via like union labels and 'bugs,' or over the politics of nations, via country of origin. Whether this is a rational decision another story, of course. I classify it as understandable but not particualrly rational. If I buy bananas when the Costa Rican bananas are in but not when the Columbian or Honduran ones are, it is not because I'm bigoted against Columbians or Hondurans. It's because on my 'mental map of the world,' Costa Rica is a blue state and Honduras and Colombia are red states. Applying the same logic to the map of the USA, I choose Lowe's over Home Deepoe, Tegrat over Walmart etc. This is all of course an exercise in futility, but like voting, the illusion of moral authority draws one in, egged on by the sense of powerlessness that comes from existing day-to-day under a buyers' market for labor, a polity in which money talks (but only in large amounts) and a the suggestion box at work is a paper shredder in disguise...

I've forgotten which of Burnham's Machiavellians put the notion of 'moral authority' in my vocabulary.

Comments for this entry have been closed
Return to top