Is There a Libertarian Case Against Free Trade?
For some time now Mises.org has come out squarely in favor of free trade, and has opposed the new attacks on trade and "globalization" (notably from Paul Craig Roberts). I myself have written numerous pieces along these lines, and have thus received perhaps hundreds of email comments on the topic. One type of argument is typical anti-capitalist rhetoric; the profit motive causes companies to exploit workers, degrade the environment, and rip off consumers, and only benevolent government intervention can check the evil market. Naturally I don't find such emails very persuasive.
However, there is a growing number of emails falling into a distinct class. Many people tell me that they are (or were) genuine supporters of free trade, but that the standard laissez-faire defenses seem hollow in the present situation. These people think that the warnings of Paul Craig Roberts (PCR) and others are generally correct, even if their actual expositions perhaps contain unwarranted generalizations and other holes that a hyperanalytical critic can attack. I write the present article in the hope of allaying the fears of such friends of liberty and to reassure them that their cherished principles are still valid.
PRINCIPLED VERSUS EMPIRICAL DEFENSES
On one issue I totally sympathize with PCR and his fans: Too many apologists for free trade, outsourcing, etc. make purely a priori, theoretical cases while the skeptics point to reams of statistics (describing low savings, drop in living standards, mushrooming trade deficits, etc.). The result of course is that the orthodox defenders appear to have their heads in the sand, to be saying "everything is well" without even knowing the actual situation.
In one sense this apriorism is perfectly valid: To the extent that opposition to free trade, outsourcing, etc. necessarily implies coercive government policies, then it really is the case that a believer in natural rights could make a principled argument in favor of free trade, the consequences be damned. (By analogy, someone could state categorically that it is always wrong to torture an innocent child, regardless of the circumstances.)
However, I have seen many commentators (particularly in blogs) go further than this, and argue that every single change in global trade patterns benefits every single participant so long as the participation is voluntary, and hence (the argument continues) critics such as PCR are simply wrong when they say that, for example, increased capital mobility is making American workers poorer.
This is an invalid argument, unfortunately, and I urge free traders to be more careful in their statements. Although I definitely agree that empirically the vast majority of Americans are richer because of globalization, that is not something we can deduce from first principles. Yes, both parties expect to gain from every voluntary transaction. However, from that insight we cannot conclude that no one is ever made poorer by a voluntary transaction in which others engage.
For a trivial example, the owners of a small grocery store might truly be poorer, even in the long run, if a Wal-Mart comes to town. In the same way, it is theoretically possible that a huge jump in productivity among Indian and Asian professionals could leave Americans poorer per capita. To argue the contrary view (which I happen to endorse) requires more than simple statements of principles.
"WE DON'T LIVE IN A LIBERTARIAN UTOPIA."
A very frequent objection runs like this: Economic models proving the benefits of free trade assume a world of laissez-faire nations. But in reality, we have competitors such as China who manipulate their currency, force their people to work for slave wages, etc.
This type of argument entirely misconceives the case for free trade. The logic of comparative advantage is unilateral; it applies regardless of whether other countries reciprocate. Let's switch examples to make this clearer: I would argue that cutting the income tax would make Americans richer. This argument does not rely on my assuming that every other nation in the world also has low income tax rates or is simultaneously cutting their income tax rates. If the reader can understand this point when it comes to taxes on Americans' income, hopefully it shouldn't be too hard to grasp it when it comes to taxes on Americans' purchases of foreign goods. (As far as the specific practices that China uses to allegedly make the United States poorer, see this article.)
LOSS OF MANUFACTURING
Another typical worry is that the Invisible Hand mechanism is shortsighted when it allows the United States to fritter away its manufacturing base. Yes, it may be "profitable" to outsource jobs to India and Asia in order for our fat consumers to get cheap TVs and round-the-clock customer service hotlines when these same TVs won't stop showing subtitles in Spanish, but what happens if and when we go to war with China? At that point a workforce that is only trained to cut lawns, serve burgers, and design websites will be definitely suboptimal, to say the least.
This is a big topic, and we all know the standard disclaimer about statistics, but nonetheless I think some perspective is needed. First, in this context the issue isn't manufacturing employment but rather manufacturing output. (If we can make twice as many tanks per worker, the fact that we lay off half the workers doesn't mean we're suddenly vulnerable to invasion. Manufacturing productivity grew annually by 4.8% between 2000 and 2003.) Yes, there was a relatively large drop in output starting in 2001 (see chart 1 here), where the single biggest drop (from the previous year) exceeded 5%. However, this recessionary slump followed an almost decade-long boom; from 1992 to 2000, total manufacturing output rose by 55%, and in some sectors (such as industrial and electrical machinery) it more than doubled. So when people point to the drop in manufacturing employment since, say, the 1950s, don't for one second believe that our economy is producing less stuff than it did fifty years ago.
Naturally the skeptics can quibble with these figures; looking at the sources it is obvious that some of these people (such as the White House!) have an incentive to paint a rosy scenario. Yet even if one could convincingly demonstrate that manufacturing capacity was in a permanent (not just a temporary) slide, that wouldn't be enough to cast aside belief in laissez-faire. This is because it is not at all clear that the recent trends in manufacturing are due to outsourcing, free trade, etc.
For example, the recent slump in manufacturing is largely due to a drop in domestic demand because of the recession. Also, as far as the slump can be related to trade, it's not that cheap imports and outsourcing are destroying US jobs, so much as that a drop in foreign demand is hurting US exporters: one estimate says that from 2000 to 2003 the US export sector lost 742,000 manufacturing jobs. This type of data is certainly inconsistent with the standard story, which would have it that American workers who used to produce for American consumers are being replaced by cheaper foreign labor.
Of course one could counter-reply that US export firms are also using foreign labor to cater to their foreign customers. My point is simply that the drop in manufacturing figures by itself is inconclusive, and could just be a reflection of the worldwide recession. But I believe there is a quite simple reason to reject the theory that the huge hit in manufacturing is due to outsourcing: even economists critical of the government's figures estimate that 406,000 jobs were outsourced in 2004. That's certainly not a number to sneeze at, but it's a gross figure, not a net one. (In other words, it doesn't subtract all of the thousands of jobs that were insourced in 2004.) However one wants to juggle the figures, there's no way that net outsourcing can explain the huge loss of jobs in manufacturing.
Before closing this section, I urge traditional believers in laissez-faire to remember that the market economy is a dynamic order. It is misleading to look at "jobs lost" due to some factor or other and ignore the jobs created by the same underlying processes. For example, I am quite sure that over 100 million US jobs have been destroyed by machinery during the 20th century, in the sense that one could count up every single job that was eradicated by the introduction of a particular labor-saving device. Naturally this doesn't mean that currently the vast majority of Americans are unable to find work, and that those who do must sell their labor hours for pennies. Yet arguments comparable to this lie behind much of the anti-globalization hysteria.
HOW TO GENERATE A CRISIS: THE HI-TECH SECTOR
One of the standard defenses of globalization goes like this: Yes, in the new worldwide economy it is more efficient to have low-wage foreigners produce cars and other manufactured goods, while highly skilled US workers concentrate on things such as software and biotechnology. This is progress in the same way that Third World countries hope to shift output away from agriculture and into manufacturing.
The response to this "naïve" optimism is that outsourcing is destroying even the high tech jobs. For example, PCR goes so far as to claim that not a single US export industry is experiencing job growth, and that this is due to free trade and outsourcing. I thought this to be unlikely — surely software engineering was growing! — so I went to the Bureau of Labor Statistics (BLS) and looked at the category called "software applications." (The category of computer jobs was broad so I picked a narrower category.) The data went back to 1999, and the latest (non-forecast) available were 2004, so those are the years I picked, and lo and behold I found that during this period total employment rose from 287,600 to 425,890 (with mean wages rising as well).
However, I was a bit worried that this figure might be misleading, since I had seen so many other pessimistic summaries. For example, this piece by the Economic Policy Institute (EPI) claims that between 2000 and 2004, "software jobs" were down by over 100,000. Now there are two interesting things here: First, one of their cited sources is the BLS itself — so even if you think the BLS data are, well, BS, that can't explain away the numbers for the EPI study. Now when I look at the BLS figures for 2000 and 2004, it's not at all obvious which categories they included and which they excluded in order to get their figure. For example, if you just look at "Computer Software Engineers: Applications" and "Computer Software Engineers: Systems Software" (the only two categories with "software" in the title), then there is a gain of 104,660 jobs in this period. Okay, maybe these authors were using a broader base. But if you include everything that has the word "Computer" in its description, then there is still a gain of 132,440 jobs from 2000 to 2004.
Now there are some particular sectors in the computer industry that took a hit in this period; I don't deny that. But my point is that in order to portray the entire computer (or more narrowly, software) industry as shedding jobs, you would have to very carefully pick your categories to include the ones that had losses and exclude the big gainers (at least if you're going to use BLS data).
There is a second problem I have with many of these pessimistic figures, of which the EPI piece is a good example. Why do all of them show the losses since 2000? After all, it's not as if outsourcing and "free trade" (which is in quotation marks since we do not have free trade) turned on in 2000 and were nonexistent before then. I suspect that the reason the alarmists go back to 2000 (rather than, say, 1999 even though the BLS has this data on the same page) is that 2000 numbers capture the status of the high tech sectors at the height of the dot-com bubble.
Finally on this point, let's step back a minute and consider the big picture. Do the critics of globalization really want to claim that the invention of the Internet has made it harder for Americans to get jobs related to computers?
RHETORIC AND PUBLIC CHOICE
The last topic I'll address is the argument (made by several emailers) that PCR and others oppose whatever it is that's happening right now, and they choose to call it "free trade" even though, admittedly, that's not such a good label.
To this I respond that it's not merely misleading, but 100% counterproductive, to oppose New World Order, backroom multinational corporation/NGO/Third World slave labor camp deals by saying one has "second thoughts on free trade." Yes, the true friend of free markets and liberty opposes thousand-page regulatory documents like NAFTA and GATT. If a multinational corporation cuts a deal with a petty tyrant and gets villagers to make sneakers at gunpoint, that's not "free trade" by any stretch.
Yet what is served by calling these things "free trade" and then claiming to oppose free trade? Not only does that confuse everyone, but it plays right into the hands of the very people engineering these vast alleged conspiracies against the American middle class. If PCR and others succeed in whipping up the public against "Benedict Arnold" CEOs and shareholders who place "profits above people," are we really going to trust the masses to elect politicians who will enact regulations that exactly correspond to PCR's ideal vision? (I ignore for the moment that PCR hasn't yet offered any description of what his alternative to "free trade" is.)
As with the hysteria over global warming, nuclear power, and every other innovation that changes our culture, the trends in globalization have led to fearful predictions. Upon closer scrutiny, most of these warnings fall apart. Yet even if the situation were dire, it still would not follow that placing tariffs or restrictions on capital mobility would make Americans richer. Until the critics can come up with at least one actual recommendation that is (a) a deviation from laissez-faire and (b) will make Americans richer, I see no reason to reject the presumption of liberty. As is so often the case, not only is freedom just, it is also pragmatic.
 I emailed the EPI about this in February but haven't received an answer so far.