Recycling is good, right? I like to think of myself as a good person in a vaguely consequentialist system of ethics, so I’ll try to make myself feel better about having spent too much time arguing in the comments to this post on TLoOG by stitching my replies together into what I hope is a vaguely coherent blog post.
My thesis is that while sweatshops are bad, focusing on getting rid of sweatshops is usually even worse because less-bad alternatives don’t usually exist. It’s far better to acknowledge the circumstances that produce sweatshops (as Nicolas Kristof writes, sweatshops are a symptom of poverty, not a cause), and work to improve conditions and compensation and eventually build a cheery middle-class economy that can build cars in union shops and export its low-wage garment jobs to whatever hellholes are left, than it is simply to shut down sweatshops and leave their former employees with nothing more than all the alternatives that were worse than the sweatshop.
Here’s a hypothetical: Suppose that Phil Knight wants to establish a Nike sweatshop in Phnom Penh, staffed exclusively by kids who currently spend their days stalking barefoot through the junkyard in search of horribly toxic heavy metals and radioisotopes to sell for a pittance. He’s willing to invest $40M in this unusual and risky venture, but it would cost $50M to build a sweatshop with reasonable control and disposal of toluene, and he’s not sufficiently confident in the venture to risk $50M. You’ve been given the authority to tell Phil “yes” or “no”. What do you do?
Now let’s change the hypothetical slightly. Phil’s built his $40M sweatshop, it runs at a profit, and it employs thousands of Cambodian children who’re delighted to be soaking in toluene rather than lead, dioxin, and cesium. You visit the sweatshop as part of an Oxfam mission and are appalled to discover that the kids are showering in solvent at the end of every shift. You demand that Phil invest an extra $12M in his profitable sweatshop to protect workers from toluene exposure. Do you think Phil’s going to take his ball and go home, or try to keep the doors open?
The difference between the first scenario and the second is time, and consequently information. A marginal increase in safety and wages at t=0 might very well push an investor away from a sweatshop venture, whereas a larger investment two years down the road, when the investor knows that the venture is successful (or has simply fallen into the Sunk Cost Fallacy) might make very little difference to Mean Mr. Capital.
Marginal increases in worker safety and pay rate still cost money, which is going to affect how the employer sees their utility. For very marginal increases, probably no better option emerges — in fact, I’m guessing that this is how countries grow out of sweatshop economies (as parts of the PRC and India are doing now). And then, past a certain threshold, those sweatshop jobs cease being more profitable than sweatshop jobs elsewhere — imagine Korean sweatshop jobs moving to China, or Chinese sweatshop jobs moving to Vietnam, or (speculatively, I think) Vietnamese sweatshop jobs moving to Congo. Or imagine Ford moving production from Michigan to Mexico. Good for the Congolese and Mexicans who have more options, bad for the Vietnamese or Michiganders who expected their jobs to stick around in perpetuity. I think this process of working conditions getting incrementally better over time is a good thing, by the way, although not necessarily when the jobs disappear with only far worse alternatives to replace them:
“One German company bowed to popular pressure and laid off 50,000 child garment workers in Bangladesh. Some of you would have cheered on hearing it. But when Oxfam followed up, they found that thousands had turned to prostitution, crime, or starved to death.”
So when someone asks, rhetorically, “Can we agree that children for instance should not be engaged in certain labor, no matter how supposedly nimble their fingers?”… I have a hard time answering without snark. I’d rather that those Bangladeshi kids could quit the factory and go to nice charter schools with well-equipped playgrounds, and that their families would find other ways to make up the income, but that’s not on the menu. Yet. Given the options, I’d rather that the German clothing company had spent a bit more money on improving working conditions, particularly safety. And then a few years down the road, if the local economy had picked up a bit and the kids had options other than “prostitution, crime, and starvation”, then we can talk about child-labour laws.
I think it’s important to ask why these Bangladeshi kids are working in the first place, and my first hypothesis is “because their families are poor, and need the income”. Speaking to the Pakistani example (still from my Eric Crampton link):
In 1995, anti-sweatshop protesters led Nike, Reebok and others to close down soccer-ball and other garment manufacturing plants in Pakistan; mean family income dropped considerably; University of Colorado economist Keith Maskus found that many of the child labourers were later found begging or getting bought and sold in international prostitution rings.
Labour conditions are indicators of wealth, at least to a first approximation. Cambodian children who sort through toxic refuse at a garbage dump are cataclysmically poor; Vietnamese children who soak up vast amounts of toluene in Nike sweatshops are by comparison “only” in dire poverty. Chinese workers who occasionally jump off of buildings in Foxconn factory complexes are wildly rich compared to their grandparents who smelted pig iron in their backyards, but poor compared to the American workers assembling Corvettes in Bowling Green, Kentucky.
Improving working conditions in any of those places would be a great and noble thing to do (would the average Corvette buyer really care about a $500 increase in the dealer price? Chevrolet could use the extra money to cater gourmet lunches once a week), up to opportunity cost considerations (perhaps that money should go towards funding pensions instead). But I think questions about sweatshops lose a lot of their relevance if they’re considered independent of the broader context of wealth and poverty.
For example, that Vietnamese Nike factory provides a path, not out of poverty, but to a poverty less awful than picking through garbage soaked in dioxins. Improving safe-handling conditions for toluene at the factory would surely provide a path to still less awful poverty, if only because the workers would be healthier and able to work for longer. Shutting down the factory over toluene-exposure violations might cut off the least-bad option for its workers. Maybe it would scare the Reebok factory next door into improving worker safety. Is that a net win? It depends on what the ex-Nike employees are doing, and we’re back to handwavey arguments about summed utility functions.
Does this mean we should countenance egregious abuse of workers simply because the available alternatives are all worse? Of course not. But we should acknowledge the root condition (poverty) and the available alternatives when we decide how to respond to those abuses. I’m deeply skeptical of a “shut the bastards down” approach.