Can Trade Bring Poverty?
Recent developments in the global economy—specifically, higher skilled foreign workers, lower costs of capital movement, and improved communications—have apparently undermined "the" case for free trade, one of the most bedrock conclusions of orthodox economics. In previous articles (1 and 2) I have restated the case in light of the specific concerns raised by economist Paul Craig Roberts. In the present essay, I want to focus on a convenient ambiguity in the new attack on trade, epitomized in a recent piece in Business Week.
Aaron Bernstein's story, "Shaking Up Trade Theory," starts out accurately enough: "Ever since Americans began fretting about globalization nearly three decades ago, economists have patiently explained why, on balance, it's a boon to the U.S." Bernstein goes on to say, "Now this long-held consensus is beginning to crack." If this were the extent of their complaints, the critics would have a point: it is theoretically possible for a country (particularly a small one) to lose from globalization, and admittedly this is something that most economists (including me) probably wouldn't have considered until the critics brought it up.
However, Bernstein (and just about everyone else who writes on this subject) conflates this correct proposition with the entirely different claim that somehow the trends of globalization undermine Ricardo's law of comparative advantage and hence the case for free trade. In Bernstein's words:
The central question [Nobel laureate Paul] Samuelson and others raise is whether unfettered trade is always still as good for the U.S. as they have long believed. Ever since British economist David Ricardo spelled out the theory of comparative advantage in the early 1800s, most economists have concluded that countries gain more than they lose when they trade with each other and specialize in what they do best. Today, however, advances in telecommunications such as broadband and the Internet have led to a new type of trade that doesn't fit neatly into the theory. (bold added)
Here Bernstein, and the skeptics he cites, have overstepped. The trends in globalization do not weaken the conclusions put in bold above. That is, Bernstein believes that the issues raised in his article at least suggest that perhaps countries don't "gain more than they lose when they trade with each other and specialize in what they do best." But as we shall see, this proposition remains untouched by the new wave of objections.
Comparative Advantage and Specialization: A Review
For the benefit of those readers who (mercifully) haven't taken an economics class in decades, let us quickly review Ricardo's argument. Recall the context: Adam Smith had made a persuasive case that citizens would benefit by importing goods from foreign countries whose workers were better at producing the commodities in question. Ricardo then took the argument a step further and demonstrated that an advanced country would benefit by trading even with a backward country inferior in producing every good.
This general proposition can be illustrated with a simple model of two countries and two goods. Thus, suppose that in a given day, a US worker can make either 100 DVDs or 10 radios. In contrast, suppose that in one day a Somali worker can make either 20 DVDs or 4 radios. The American worker is more productive in either industry than the African, and thus it seems as if the US could not possibly benefit from trade with Somalia.
But this logic is wrong, as Ricardo showed. For modern economists the true cost of a good is not the number of labor hours that went into it, but rather the value of the goods that could have been produced in its absence (the "opportunity cost"). In this important sense, it's actually cheaper for Somalis to make radios than for Americans: For every additional radio that the Somalis decide to produce, they have less manpower to devote to DVD production and hence sacrifice five DVDs; i.e. it costs Somalis five DVDs to produce one radio. In contrast, Americans must forfeit ten potential DVDs for every radio they produce; it costs twice as much (in terms of DVDs) to produce radios in America.
By the same token, it is cheaper (in terms of potential radios) to produce DVDs in the US than in Somalia. Thus, although the US has an absolute advantage in both industries, it only has the comparative (or relative) advantage in DVD production, while Somalia has the comparative advantage in radio production. Consequently, both countries can benefit if they specialize and trade, rather than producing everything domestically. For example, if Americans ship Somalis 8 DVDs in exchange for every radio that Somalis send to the US, both countries are obtaining the imported goods on better terms than they could achieve without trade. Specifically, the US now sacrifices 8 DVDs for every radio its citizens obtain, whereas above we computed the cost of a radio at 10 DVDs. And the Somalis now obtain 8 DVDs for every radio they sacrifice, whereas above we computed that the Somalis would only gain 5 DVDs for every forgone radio.
Finally, because we have seen the advantages to specialization and trade, we can easily see how any restrictions on such trade (whether tariffs, quotas, etc.) would hinder these gains. Yes, the US could certainly "create jobs" for radio manufacturers by blocking trade with Somalia, but the aggregate effect would be fewer radios and DVDs per capita for Americans. American consumers would lose more (in terms of commodities) than American radio producers would gain.
So how do the new trends in worldwide commerce influence the conclusions drawn above? They don't. Regardless of the changes of globalization, it is still the case that a given group of individuals can only benefit from gaining the ability to (a) obtain goods from people outside of the group, (b) sell their services to outsiders, and/or (c) rent their capital equipment to outsiders.
Take the case of India. Bernstein writes: "As skill levels improve . . . competition is coming on in the very products for which the U.S. has had a global advantage, such as software. If the new competition drives down prices too much . . . the entire U.S. economy could end up worse off."
Again, this is theoretically possible; it's the macro analog of the situation in which, say, Tiger Woods might be worse off if tomorrow there emerged an eighteen-year-old black golfer who never bogeyed. But does this possibility mean that Tiger Woods might do better to refrain from trading with his neighbors? Of course not. And in just the same fashion, even if Americans on net become poorer because of smarter Indians, this doesn't at all undermine the advantages of specialization and free trade. This is because the only way Americans on net can lose out is if other foreigners switch to hire Indians. And this development is not anything the American government can alter; e.g. the US government can't prohibit German or Japanese firms from hiring Indian workers. The only thing the US government can do is make it harder for American consumers or American firms to trade with foreigners, and such interference can only make Americans (on net) poorer.
"But this doesn't fit Ricardo's framework!"
The skeptics would be quick to pounce on my assertions. After all, how can I possibly claim that free trade still benefits Americans, when Ricardo's model didn't deal with today's enhanced labor and capital mobility?
My answer is that the principle of comparative advantage—as elucidated in the simplistic story of DVDs and radios—is still valid; when one understands why specialization and trade are good in the simple case, it's easy to see that they're good under more realistic circumstances. That's the reason introductory textbooks use the simple examples: they quickly illustrate the general principle, which the reader can then apply to the world.
I grant that this answer at first seems duplicitous. After all, if Ricardo's assumptions aren't upheld, how do we know his conclusion is still valid? But let's change the argument and see how it sounds. Suppose someone says, "Cultural and fashion trends now have people spending more time and money on their tans. This represents a gain to the producers of tanning lotion and tanning booths, but a loss to the victims of skin cancer and their immediate families. It is unclear how to handle this situation in terms of comparative advantage, and hence we can no longer be sure that free trade is beneficial."
How does the reader feel about this hypothetical argument? In my opinion it is quite weak, because it doesn't contain even a hint as to how the situation of tanning and skin cancer—which Ricardo admittedly did not address—has any relation to the benefits of trade.
I submit that the same is true of the alleged dangers of globalization. Yes, certain trends may harm particular people, but no one has even offered a hint as to how these people can mitigate the damage by restricting their own options for trade. Indeed, Paul Craig Roberts has written repeatedly that his concerns are not at all a call for tariffs or other forms of protection, and Bernstein writes that "[n]o one is advocating new trade barriers, which could be a cure that's worse than the disease." In other words, even the academic critics still agree that free trade is better than protectionism.
A Partial Proof
For those who want more than my intuitive assurances, I offer this proof to show that, when an American firm outsources jobs to foreign workers, it necessarily makes Americans richer on net. (Notice that I do not rely on any idiosyncratic assumptions of Austrian economics; this is straight mainstream theory.) Now this proof will not provide "the" case for free trade and unrestricted freedom of labor and capital migration, because it would be impossible for someone to devise an argument (in cost-benefit terms) that takes into account all future permutations in commerce. But it will nonetheless give a fairly general defense of outsourcing.
Let Hrepresent the high wages that American workers initially earn in a certain industry. Then suppose that improvements in telecommunications allow the American firms to fire all of these workers and outsource the jobs to foreigners, who work for wage F. The displaced Americans now must take up inferior jobs, earning the lower wage L.
Now then, what are the gains and losses to Americans from this shift? Well, each of the American workers—let us suppose there are N of them—now earn H-L less in wages. Thus the total loss to US workers from the outsourcing is (H-L)N. But on the other hand, the firms have lower labor costs, and hence their shareholders are richer by (H-F)N. Americans gain on net, therefore, if (H-F)N > (H-L)N, which is true whenever L>F. That is, we know that the outsourcing makes Americans wealthier on net whenever the foreign workers require a wage that is lower than the next-best wage that the American workers can earn. But this means that the outsourcing will always be efficient, because American firms will only outsource when American workers would rather work at the next-best job than suffer the necessary wage cuts to stay competitive with foreigners in the original industry. In other words, American firms will only outsource when L>F.
But doesn't this concede that American workers might lose at the expense of American capitalists? Why yes, it does. But I was under the impression that we were talking about net gains to Americans, and I believe American capitalists are part of this country. (Remember that this group consists not only of fat cats who inherit their money from daddy, but also of retired schoolteachers dependent on mutual funds.) Moreover, to the extent that there is healthy competition in the industry in question, the lower labor costs will quickly be passed on as lower prices to the consumers of the good. In other words, the higher returns to the shareholders of the outsourcing firms will only be temporary. Ultimately it will be the firms' customers whose gain outweighs the domestic workers' loss.
Before closing this section, I should point out that the failure to consider gains to capitalists is responsible for the notion that capital mobility can somehow make a country poorer. The idea is this: When capital was relatively immobile, the workers in countries such as the US were extremely productive because of the superior tools and equipment that augmented their output. However, as transportation costs plummet and foreign countries become more hospitable to investment, capital is being shipped abroad to the detriment of domestic workers. Therefore the new trends spell lower wages for Americans, making the US poorer on net.
Again, what this argument overlooks is the gain to US capitalists from the enhanced investment opportunities. To see this, suppose that a philanthropist wishes to bless the impoverished inhabitants of a small island in the Pacific. To this end, he arranges to have hundreds of tractors, computers, drill presses, and other expensive capital goods shipped to their country, and distributed equally among the natives. Yet one of his advisors, versed in the new literature on trade, points out that this generous donation may completely backfire: If the recipients of the gifts determine that they can obtain the best return by renting them out to foreigners—i.e. by putting them right back on the boat and sending them to other countries—then the islanders will not gain at all. Therefore, to ensure that the islanders do not foolishly forfeit the benefits of the capital goods, the advisor recommends that the philanthropist place an explosive device on each gift, which will explode the moment it leaves the borders of the island. This move will ensure that it is the islanders whose wages go up because of the capital equipment.
Of course this is crazy; you don't make people richer by placing arbitrary limitations on their capital goods. Specifically, the advisor's analysis overlooks the fact that the returns to the islanders in their capacity as owners of the donated equipment will be lower if they are forced to use the goods domestically. Yes, the wages of laboring islanders will be higher, but this gain will be more than offset by the lower rents earned by the island's capitalists. On net, the advisor's proposal would make the islanders poorer than they otherwise would have been. Enhanced capital mobility is a good thing, especially for capital-rich countries such as the US.
Americans account for a large fraction of world consumption, and hence it is empirically very likely that they will benefit tremendously from the combined effects of the various trends that have been dubbed "globalization." Nonetheless—as Samuelson and others have demonstrated—it is theoretically possible that a particular innovation or improved skill level among certain foreigners will make Americans poorer on net. Even so, this possibility in no way overturns the truism that everyone, including Americans, benefits from having more options at his or her disposal. Until the newest critics of trade give at least one plausible example of how a policy of unfree trade could make a country richer, I'm sticking with the orthodox conclusions.
To understand how it's at least possible for a country to lose from globalization, in my earlier article linked above, I asked the reader to imagine a country of 10,000 people who are experts at producing copy machines. With improved fax and email technology, it's possible that these people could end up poorer.
As with just about everything in the history of thought, the law traditionally associated with Ricardo was almost certainly known by earlier theorists. In this case, James Mill probably taught the principle to Ricardo.
Of course, if one wants to defend free trade on the grounds of natural law or other theories of justice, then a general defense can be given. In this case, the defender would say that even if tariffs or other forms of government intervention could make Americans richer on net, that wouldn't justify them. Alas, this type of appeal is not likely to persuade the critics of trade.