So what can we actually do about sweatshops?

JDN 2457489

(The topic of this post was chosen by a vote of my Patreons.) There seem to be two major camps on most political issues: One camp says “This is not a problem, stop worrying about it.” The other says “This is a huge problem, it must be fixed right away, and here’s the easy solution.” Typically neither of these things is true, and the correct answer is actually “This is a huge problem, well worth fixing—but we need to do a lot of work to figure out exactly how.”

Sweatshop labor is a very good example of this phenomenon.

Camp A is represented here by the American Enterprise Institute, which even goes as far as to defend child labor on the grounds that “we used to do it before”. (Note that we also used to do slavery before. Also protectionism, but of course AEI doesn’t think that was good. Who needs logical consistency when you have ideological purity?) The College Conservative uses ECON 101 to defend sweatshops, perhaps not realizing that economics courses continue past ECON 101.

Camp B is represented here by Buycott, telling us to buy “made in the USA” products and boycott all companies that use sweatshops. Other commonly listed strategies include buying used clothes (I mean, there may be some ecological benefits to this, but clearly not all clothes can be used clothes) and “buy union-made” which is next to impossible for most products. Also in this camp is LaborVoices, a Silicon Valley tech company that seems convinced they can somehow solve the problem of sweatshops by means of smartphone apps, because apparently Silicon Valley people believe that smartphones are magical and not, say, one type of product that performs services similar to many other pre-existing products but somewhat more efficiently. (This would also explain how Uber can say with a straight face that they are “revolutionary” when all they actually do is mediate unlicensed taxi services, and Airbnb is “innovative” because it makes it slightly more convenient to rent out rooms in your home.)

Of course I am in that third camp, people who realize that sweatshops—and exploitative labor practices in general—are a serious problem, but a very complex and challenging one that does not have any easy, obvious solutions.

One thing we absolutely cannot do is return to protectionism or get American consumers to only buy from American companies (a sort of “soft protectionism” by social construction). This would not only be inefficient for us—it would be devastating for people in Third World countries. Sweatshops typically provide substantially better living conditions than the alternatives available to their workers.

Yet this does not mean that sweatshops are morally acceptable or should simply be left alone, contrary to the assertions of many economists—most famously Benjamin Powell. Anyone who doubts this must immediately read “Wrongful Beneficence” by Chris Meyers; the mere fact that an act benefits someone –or even everyone—does not prove that the act was morally acceptable. If someone is starving to death and you offer them bread in exchange for doing whatever you want them to do for the next year, you are benefiting them, surely—but what you are doing is morally wrong. And this is basically what sweatshops are; they provide survival in exchange for exploitation.

It can be remarkably difficult to even tell which companies are using sweatshops—and this is by design. While in response to public pressure corporations often try to create the image of improving their labor standards, they seem quite averse to actually improving labor standards, and even more averse to establishing systems of enforcement to make those labor standards followed consistently. Almost no sweatshops are directly owned by the retailers whose products they make; instead there is a chain of outsourced vendors and distributors, a chain that creates diffusion of responsibility and plausible deniability. When international labor organizations do get the chance to investigate the labor conditions of factories operated by multinational corporations, they invariably find that regulations are more honored in the breach than the observance.

So, what would a long-run solution to sweatshops look like? In a word: Development. The only sustainable solution to oppressive labor conditions is a world where everyone is healthy enough, educated enough, and provided with enough resources that their productivity is at a First World level; furthermore it is a world where workers have enough bargaining power that they are actually paid according to that productivity. (The US has lately been finding out what happens if you do the former but not the latter—the result is that you generate an enormous amount of wealth, but it all ends up in the hands of the top 0.1%. Yet it is quite possible to do the latter, as Denmark has figured out, #ScandinaviaIsBetter.)

To achieve this, we need more factories in Third World countries, not fewer—more investment, not less. We need to buy more of China’s exports, hire more factory workers in Bangladesh.

But it’s not enough to provide incentives to build factories—we must also provide incentives to give workers at those factories more bargaining power.

To see how we can pull this off, I offer a case study of a (qualified) success: Nike.

In the 1990s, Nike’s subcontractors had some of the worst labor conditions in the shoe industry. Today, they actually have some of the best. How did that happen?

It began with people noticing a problem—activists and investigative journalists documented the abuses in Nike’s factories. They drew public attention, which undermined Nike’s efforts at mass advertising (which was basically their entire business model—their shoes aren’t actually especially good). They tried to clean up their image with obviously biased reports, which triggered a backlash. Finally Nike decides to actually do something about the problem, and actually becomes a founding member of the Fair Labor Association. They establish new labor standards, and they audit regularly to ensure that those standards are being complied with. Today they publish an annual corporate social responsibility report that actually appears to be quite transparent and accurate, showing both the substantial improvements that have been made and the remaining problems. Activist campaigns turned Nike around almost completely.

In short, consumer pressure led to private regulation. Many development economists are increasingly convinced that this is what we need—we must put pressure on corporations to regulate themselves.

The pressure is a key part of this process; Willem Buiter wasn’t wrong when he quipped that “self-regulation stands in relation to regulation the way self-importance stands in relation to importance and self-righteousness to righteousness.” For any regulation to work, it must have an enforcement mechanism; for private regulation to work, that enforcement mechanism comes from the consumers.

Yet even this is not enough, because there are too many incentives for corporations to lie and cheat if they only have to be responsive to consumers. It’s unreasonable to expect every consumer to take the time—let alone have the expertise—to perform extensive research on the supply chain of every corporation they buy a product from. I also think it’s unreasonable to expect most people to engage in community organizing or shareholder activism as Green America suggests, though it certainly wouldn’t hurt if some did. But there are just too many corporations to keep track of! Like it or not, we live in a globalized capitalist economy where you almost certainly buy from a hundred different corporations over the course of a year.

Instead we need governments to step up—and the obvious choice is the government of the United States, which remains the world’s economic and military hegemon. We should be pressuring our legislators to make new regulations on international trade that will raise labor standards around the globe.

Note that this undermines the most basic argument corporations use against improving their labor standards: “If we raise wages, we won’t be able to compete.” Not if we force everyone to raise wages, around the globe. “If it’s cheaper to build a factory in Indonesia, why shouldn’t we?” It won’t be cheaper, unless Indonesia actually has a real comparative advantage in producing that product. You won’t be able to artificially hold down your expenses by exploiting your workers—you’ll have to actually be more efficient in order to be more profitable, which is how capitalism is supposed to work.

There’s another argument we often hear that is more legitimate, which is that raising wages would also force corporations to raise prices. But as I discussed in a previous post on this subject, the amount by which prices would need to rise is remarkably small, and nowhere near large enough to justify panic about dangerous global inflation. Paying 10% or even 20% more for our products is well worth it to reduce the corruption and exploitation that abuses millions of people—a remarkable number of them children—around the globe. Also, it doesn’t take a mathematical savant to realize that if increasing wages by a factor of 10 only increases prices by 20%, workers will in fact be better off.

Where would all that extra money come from? Now we come to the real reason why corporations don’t want to raise their labor standards: It would come from profits. Right now profits are extraordinarily large, much larger than they have any right to be in a fair market. It was recently estimated that 74% of billionaire wealth comes from economic rent—that is to say, from deception, exploitation, and market manipulation, rather than actual productivity. (There’s a lot of uncertainty in this estimate; the true figure is probably somewhere between 50% and 90%—it’s almost certainly a majority, and could be the vast majority.) In fact, I really shouldn’t say “money”, which we can just print; what we really want to know is where the extra wealth would come from to give that money value. But by paying workers more, improving their standard of living, and creating more consumer demand, we would in fact dramatically increase the amount of real wealth in the world.

So, we need regulations to improve global labor standards. But we must first be clear: What should these regulations say?

First, we must rule out protectionist regulations that would give unfair advantages to companies that produce locally. These would only result in economic inefficiency at best, and trade wars throwing millions back into poverty at worst. (Some advantage makes sense to internalize the externalities of shipping, but really that should be created by a carbon tax, not by trade tariffs. It’s a lot more expensive and carbon-intensive to ship from Detroit to LA than from Detroit to Windsor, but the latter is the “international” trade.)

Second, we should not naively assume that every country should have the same minimum wage. (I am similarly skeptical of Hillary Clinton’s proposal to include people with severe mental or physical disabilities in the US federal minimum wage; I too am concerned about people with disabilities being exploited, but the fact is many people with severe disabilities really aren’t as productive, and it makes sense for wages to reflect that.) If we’re going to have minimum wages at all—basic income and wage subsidies both make a good deal more sense than a hard price floor; see also my earlier post on minimum wage—they should reflect the productivity and prices of the region. I applaud California and New York for adopting $15 minimum wages, but I’d be a bit skeptical of doing the same in Mississippi, and adamantly opposed to doing so in Bangladesh.

It may not even be reasonable to expect all countries to have the same safety standards; workers who are less skilled and in more dire poverty may rationally be willing to accept more risk to remain employed, rather than laid off because their employer could not afford to meet safety standards and still pay them a sufficient wage. For some safety standards this is ridiculous; making sufficiently many exits with doors that swing outward and maintaining smoke detectors are not expensive things to do. (And yet factories in Bangladesh often fail to meet such basic requirements, which kills hundreds of workers each year.) But other safety standards may be justifiably relaxed; OSHA compliance in the US costs about $70 billion per year, about $200 per person, which many countries simply couldn’t afford. (On the other hand, OSHA saves thousands of lives, does not increase unemployment, and may actually benefit employers when compared with the high cost of private injury lawsuits.) We should have expert economists perform careful cost-benefit analyses of proposed safety regulations to determine which ones are cost-effective at protecting workers and which ones are too expensive to be viable.

While we’re at it, these regulations should include environmental standards, or a global carbon tax that’s used to fund climate change mitigation efforts around the world. Here there isn’t much excuse for not being strict; pollution and environmental degradation harms the poor the most. Yes, we do need to consider the benefits of production that is polluting; but we have plenty of profit incentives for that already. Right now the balance is clearly tipped far too much in favor of more pollution than the optimum rather than less. Even relatively heavy-handed policies like total bans on offshore drilling and mountaintop removal might be in order; in general I’d prefer to tax rather than ban, but these activities are so enormously damaging that if the choice is between a ban and doing nothing, I’ll take the ban. (I’m less convinced of this with regard to fracking; yes, earthquakes and polluted groundwater are bad—but are they Saudi Arabia bad? Because buying more oil from Saudi Arabia is our leading alternative.)

It should go without saying (but unfortunately it doesn’t seem to) that our regulations must include an absolute zero-tolerance policy for forced labor. If we find out that a company is employing forced labor, they should have to not only free every single enslaved worker, but pay each one a million dollars (PPP 2005 chained CPI of course). If they can’t do that and they go bankrupt, good riddance; remind me to play them the world’s saddest song on the world’s tiniest violin. Of course, first we need to find out, which brings me to the most important point.

Above all, these regulations must be enforced. We could start with enforceable multilateral trade agreements, where tariff reductions are tied to human rights and labor standards. This is something the President of the United States could do, right now, as an addendum to the Trans-Pacific Partnership. (What he should have done is made the TPP contingent on this, but it’s too late for that.) Future trade agreements should include these as a matter of course.If countries want to reap the benefits of free trade, they must be held accountable for sharing those benefits equitably with their people.

But ultimately we should not depend upon multilateral agreements between nations—we need truly international standards with global enforcement. We should empower the International Labor Organization to enact sanctions and inspections (right now it mostly enacts suggestions which are promptly and dutifully ignored), and possibly even to arrest executives for trial at the International Criminal Court. We should double if not triple or quadruple their funding—and if member nations will not pay this voluntarily, we should make them—the United Nations should be empowered to collect taxes in support of global development, which should be progressive with per-capita GDP. Coercion, you say? National sovereignty, you say? Millions of starving little girls is my reply.

Right now, the ability of multinational corporations to move between countries to find the ones that let them pay the least have created a race to the floor; it’s time for us to raise that floor.

What can you yourself do, assuming you’re not a head of state? (If you are, I’m honored. Also, any openings on your staff?) Well, you can vote—and you can use that vote to put pressure on your legislators to support these kinds of polices. There are also some other direct actions you can take that I discussed in a previous post; but mainly what we need is policy. Consumer pressure and philanthropy are good, and by all means, don’t stop; but to really achieve global justice we will need nothing short of global governance.

We do not benefit from economic injustice.

JDN 2457461

Recently I think I figured out why so many middle-class White Americans express so much guilt about global injustice: A lot of people seem to think that we actually benefit from it. Thus, they feel caught between a rock and a hard place; conquering injustice would mean undermining their own already precarious standard of living, while leaving it in place is unconscionable.

The compromise, is apparently to feel really, really guilty about it, constantly tell people to “check their privilege” in this bizarre form of trendy autoflagellation, and then… never really get around to doing anything about the injustice.

(I guess that’s better than the conservative interpretation, which seems to be that since we benefit from this, we should keep doing it, and make sure we elect big, strong leaders who will make that happen.)

So let me tell you in no uncertain words: You do not benefit from this.

If anyone does—and as I’ll get to in a moment, that is not even necessarily true—then it is the billionaires who own the multinational corporations that orchestrate these abuses. Billionaires and billionaires only stand to gain from the exploitation of workers in the US, China, and everywhere else.

How do I know this with such certainty? Allow me to explain.

First of all, it is a common perception that prices of goods would be unattainably high if they were not produced on the backs of sweatshop workers. This perception is mistaken. The primary effect of the exploitation is simply to raise the profits of the corporation; there is a secondary effect of raising the price a moderate amount; and even this would be overwhelmed by the long-run dynamic effect of the increased consumer spending if workers were paid fairly.

Let’s take an iPad, for example. The price of iPads varies around the world in a combination of purchasing power parity and outright price discrimination; but the top model almost never sells for less than $500. The raw material expenditure involved in producing one is about $370—and the labor expenditure? Just $11. Not $110; $11. If it had been $110, the price could still be kept under $500 and turn a profit; it would simply be much smaller. That is, even if prices are really so elastic that Americans would refuse to buy an iPad at any more than $500 that would still mean Apple could still afford to raise the wages they pay (or rather, their subcontractors pay) workers by an order of magnitude. A worker who currently works 50 hours a week for $10 per day could now make $10 per hour. And the price would not have to change; Apple would simply lose profit, which is why they don’t do this. In the absence of pressure to the contrary, corporations will do whatever they can to maximize profits.

Now, in fact, the price probably would go up, because Apple fans are among the most inelastic technology consumers in the world. But suppose it went up to $600, which would mean a 1:1 absorption of these higher labor expenditures into price. Does that really sound like “Americans could never afford this”? A few people right on the edge might decide they couldn’t buy it at that price, but it wouldn’t be very many—indeed, like any well-managed monopoly, Apple knows to stop raising the price at the point where they start losing more revenue than they gain.

Similarly, half the price of an iPhone is pure profit for Apple, and only 2% goes into labor. Once again, wages could be raised by an order of magnitude and the price would not need to change.

Apple is a particularly obvious example, but it’s quite simple to see why exploitative labor cannot be the source of improved economic efficiency. Paying workers less does not make them do better work. Treating people more harshly does not improve their performance. Quite the opposite: People work much harder when they are treated well. In addition, at the levels of income we’re talking about, small improvements in wages would result in substantial improvements in worker health, further improving performance. Finally, substitution effect dominates income effect at low incomes. At very high incomes, income effect can dominate substitution effect, so higher wages might result in less work—but it is precisely when we’re talking about poor people that it makes the least sense to say they would work less if you paid them more and treated them better.

At most, paying higher wages can redistribute existing wealth, if we assume that the total amount of wealth does not increase. So it’s theoretically possible that paying higher wages to sweatshop workers would result in them getting some of the stuff that we currently have (essentially by a price mechanism where the things we want get more expensive, but our own wages don’t go up). But in fact our wages are most likely too low as well—wages in the US have become unlinked from productivity, around the time of Reagan—so there’s reason to think that a more just system would improve our standard of living also. Where would all the extra wealth come from? Well, there’s an awful lot of room at the top.

The top 1% in the US own 35% of net wealth, about as much as the bottom 95%. The 400 billionaires of the Forbes list have more wealth than the entire African-American population combined. (We’re double-counting Oprah—but that’s it, she’s the only African-American billionaire in the US.) So even assuming that the total amount of wealth remains constant (which is too conservative, as I’ll get to in a moment), improving global labor standards wouldn’t need to pull any wealth from the middle class; it could get plenty just from the top 0.01%.

In surveys, most Americans are willing to pay more for goods in order to improve labor standards—and the amounts that people are willing to pay, while they may seem small (on the order of 10% to 20% more), are in fact clearly enough that they could substantially increase the wages of sweatshop workers. The biggest problem is that corporations are so good at covering their tracks that it’s difficult to know whether you are really supporting higher labor standards. The multiple layers of international subcontractors make things even more complicated; the people who directly decide the wages are not the people who ultimately profit from them, because subcontractors are competitive while the multinationals that control them are monopsonists.

But for now I’m not going to deal with the thorny question of how we can actually regulate multinational corporations to stop them from using sweatshops. Right now, I just really want to get everyone on the same page and be absolutely clear about cui bono. If there is a benefit at all, it’s not going to you and me.

Why do I keep saying “if”? As so many people will ask me: “Isn’t it obvious that if one person gets less money, someone else must get more?” If you’ve been following my blog at all, you know that the answer is no.

On a single transaction, with everything else held constant, that is true. But we’re not talking about a single transaction. We’re talking about a system of global markets. Indeed, we’re not really talking about money at all; we’re talking about wealth.

By paying their workers so little that those workers can barely survive, corporations are making it impossible for those workers to go out and buy things of their own. Since the costs of higher wages are concentrated in one corporation while the benefits of higher wages are spread out across society, there is a Tragedy of the Commons where each corporation acting in its own self-interest undermines the consumer base that would have benefited all corporations (not to mention people who don’t own corporations). It does depend on some parameters we haven’t measured very precisely, but under a wide range of plausible values, it works out that literally everyone is worse off under this system than they would have been under a system of fair wages.

This is not simply theoretical. We have empirical data about what happened when companies (in the US at least) stopped using an even more extreme form of labor exploitation: slavery.

Because we were on the classical gold standard, GDP growth in the US in the 19th century was extremely erratic, jumping up and down as high as 10 lp and as low as -5 lp. But if you try to smooth out this roller-coaster business cycle, you can see that our growth rate did not appear tobe slowed by the ending of slavery:

US_GDP_growth_1800s

 

Looking at the level of real per capita GDP (on a log scale) shows a continuous growth trend as if nothing had changed at all:

US_GDP_per_capita_1800s

In fact, if you average the growth rates (in log points, averaging makes sense) from 1800 to 1860 as antebellum and from 1865 to 1900 as postbellum, you find that the antebellum growth rate averaged 1.04 lp, while the postbellum growth rate averaged 1.77 lp. Over a period of 50 years, that’s the difference between growing by a factor of 1.7 and growing by a factor of 2.4. Of course, there were a lot of other factors involved besides the end of slavery—but at the very least it seems clear that ending slavery did not reduce economic growth, which it would have if slavery were actually an efficient economic system.

This is a different question from whether slaveowners were irrational in continuing to own slaves. Purely on the basis of individual profit, it was most likely rational to own slaves. But the broader effects on the economic system as a whole were strongly negative. I think that part of why the debate on whether slavery is economically inefficient has never been settled is a confusion between these two questions. One side says “Slavery damaged overall economic growth.” The other says “But owning slaves produced a rate of return for investors as high as manufacturing!” Yeah, those… aren’t answering the same question. They are in fact probably both true. Something can be highly profitable for individuals while still being tremendously damaging to society.

I don’t mean to imply that sweatshops are as bad as slavery; they are not. (Though there is still slavery in the world, and some sweatshops tread a fine line.) What I’m saying is that showing that sweatshops are profitable (no doubt there) or even that they are better than most of the alternatives for their workers (probably true in most cases) does not show that they are economically efficient. Sweatshops are beneficent exploitationthey make workers better off, but in an obviously unjust way. And they only make workers better off compared to the current alternatives; if they were replaced with industries paying fair wages, workers would obviously be much better off still.

And my point is, so would we. While the prices of goods would increase slightly in the short run, in the long run the increased consumer spending by people in Third World countries—which soon would cease to be Third World countries, as happened in Korea and Japan—would result in additional trade with us that would raise our standard of living, not lower it. The only people it is even plausible to think would be harmed are the billionaires who own our multinational corporations; and yet even they might stand to benefit from the improved efficiency of the global economy.

No, you do not benefit from sweatshops. So stop feeling guilty, stop worrying so much about “checking your privilege”—and let’s get out there and do something about it.