Tuesday, August 02, 2005

Lump of Wages?

I'm going to have to take issue with Battlepanda's response to Matt Yglesias' criticism of Josh Marshall's plug for a Washington Post article about sweatshop activist Charles Kernaghan.

She writes:
Matt Yglesia[s] falls into the classic trap for tough-minded liberals who want to demonstrate that they have a good grasp of economics: the lump of wages fallacy. This fallacy is especially erroneous when applied to to sweatshop wages, yet it is frequently trotted out to attack those who want to institute the most minimal protections and standards for the workers in third-world countries.
I don't think this can just be dismissed as a fad argument considering it's also trotted out by leading trade economists. (As an aside, I'm curious about where "lump of wages" comes from.)
There are two sets of competitive forces at work here which we must keep separate. The competition between poor countries is an all-too-real race to the bottom and it would indeed be most imprudent for individual third-world states to impose tough labor standards as companies will simply take their factories elsewhere. But the second set of competitive forces, the one that Matt cites when he says "closing down the sweatship option would seem to just force everyone to stick with misery" is simply a chimera.
It's the oft-talked-about "race to the bottom" that's the real chimera here. Yes, individual countries don't want to establish new labor laws for fear of losing jobs, but those laws were never there in the first place. They're already at the bottom. With no labor or environmental laws pre-existing, there's not much governments can do to lower costs for Nike. But they can try and raise productivity, which ultimately makes workers better off. Like I said, there's nowhere to go but up, which is what we've seen happen in South Korea, Singapore, and so forth.
In a perfectly competitive market, if wages go up then demand goes down. But if that's what we have today, then we wouldn't have a stock market to obsess over as profits would be driven down to zero too.
Generally, that demand curves slope down is true of every standard market model, from perfect competition to monopoly. If wages go up, demand for labor is definitely not going up.
Furthermore, such a tiny sliver of the costs of making a pair of pants go towards the seamstresses (compared to shop rent, store clerk wages, management salaries and profit) that even doubling it would not cause one less pair of pants to be sewn in the third world.
The problem here is that NIKE doesn't think about things like this. Yes, it's a relatively small part of costs. But you have to look at the *revenue* that each third world worker produces. It's quite probable (certain) that each advertiser, store clerk, and so forth produce far more *revenue* for Nike than do sweatshop workers, even though they don't actually make any product. In general, workers get paid their marginal revenue product (no matter what the labor market structure, perfect competition or monopsony). There is no reason to think that this is any different in a poor country.

So suppose the agitation works and wages are artificially doubled in sweatshops. If no one gets laid off that implies that labor demand is vertical. This then implies that there are no substitutes for third world labor. But there are plenty of substitutes, in particular, first world labor. Though wages are a lot higher, the amount of product and revenue a first world worker produces is correspondingly higher. Not so strange that unions here argue for higher wages and labor standards there.
It would have a negligable effect on the volume of commerce to third-world countries because of their decisive advantage in low-cost labor, and jump-start the kind of positive snowball effect of development in 3rd world countries that pro-globalization people enthuse about.
As I said, third world labor is low-cost, but also low-productivity. The countries that have experienced the "positive snowball effect", notably the "Four Tigers," started with sweatshops, probably worse than those that exist today. But fairly quickly, productivity rose, wages rose, standards of living rose.

Commenter Matt Singer adds:
Yeah, that also ignores efficiency wage theory and a ton of economics research that solidly demonstrates that the labor market, well, isn't a market.
It doesn't, actually. The major labor market failures are usually based on information asymmetry and the efficiency wage (and maybe wage rigidities). There's virtually no concern about information asymmetry in unskilled labor markets and there's no conflict between what I've said and efficiency wage theory. In both, wages are determined by marginal revenue product. Indeed, that 10% premium over the going wage that foreign companies pay in third world countries may well be due to some sort of efficiency wage.

In short, I don't think Matt Yglesias has fallen into any trap.


Anonymous Battlepanda said...

"Lump of wages" is a play on the "lump of labor" fallacy so frequently ridiculed by main stream economists. The correlation is not one-to-one, I'm just trying to make a point: Economists have no trouble debunking the myth that there is only so much work in the world, yet they often seem to operate under the assumption that there is only so much wage in the world and that higher wages necessarily means fewer jobs in proportion so that the total wage payout is the same or lower. This is not so for a demand as inelastic as our insatiable need for third-world labor. You prove my point by saying "It's quite probable (certain) that each advertiser, store clerk, and so forth produce far more *revenue* for Nike than do sweatshop workers, even though they don't actually make any product." Yet without that pair of pants, all that advertizing, store space, clerks...cannot generate revenue. The fact that the cost of making the pants is but a tiny part Nike's total costs makes it likely that a rise in labor costs will be absorbed.

As for first world labor as a substitute for third world labor. Do you really think that if the hourly wage of textile workers in China went from $0.50 per hour to $1.00 per hour that textile jobs will flood back into America, where the federal minimum wage is $5.15 an hour? If so, you're just as naive as the unions who think we can keep manufacturing jobs in America forever.

By the way, in the future you might want to separate what you're responding to from your comments, because keeping everything in one big paragraph make it difficult to parse.

11:28 AM  
Blogger henry said...

Okay, I see your point about the "lump of wages" and you are absolutely right. There is no set total amount of wages that a corporation pays. Decisions are made on the margin. The argument really boils down to an empirical question: what is the elasticity of labor demand in third world countries?

I think that, in the short run, you are probably correct. Labor demand is inelastic. But in the long run, I think it will be fairly elastic. So Nike may not fire too many workers right away, but they'll build their next plant somewhere else.

With regard to first world labor, as I said, it's far more productive. This is an empirical question as well, but I don't think that third world labor has a huge advantage over first world labor *after you take productivity into account.*

(I did divide by post using blockquote tags, which look okay in my browser. I don't know if they are supported in every browser or if RSS can see them or what...)

11:31 PM  
Blogger henry said...

Here's some evidence that I don't know what I'm talking about: There was a paper in the Industrial and Labor Relations Review by Martin Rama (2001). The result was basically that the Indonesian's doubled their real minimum wage in the early 90's and employment decreased only 5%. Wages increased 10-15%. The paper also argues that this hit small firms much harder, so it was in some sense a transfer from workers at small firms to workers at large firms.

11:46 PM  

Post a Comment

<< Home