Benjamin Powell – a fine economist who earned his PhD in Economics at George Mason University and who now serves on the faculty at San Jose State University – wrote, with his student Dave Skarbek, this excellent op-ed  on sweatshops in developing countries. It appears in today’s Christian Science Monitor.
Here’s the crux of Powell’s and Skarbek’s findings (a fuller version of which will be published in the Journal of Labor Research ):
We examined the apparel industry in 10 Asian and Latin American countries often accused of having sweatshops and then we looked at 43 specific accusations of unfair wages in 11 countries in the same regions. Our findings may seem surprising. Not only were sweatshops superior to the dire alternatives economists usually mentioned [such as working on subsistence farms], but they often provided a better-than-average standard of living for their workers.
The apparel industry, which is often accused of unsafe working conditions and poor wages, actually pays its foreign workers well enough for them to rise above the poverty in their countries. While more than half of the population in most of the countries we studied lived on less than $2 per day, in 90 percent of the countries, working a 10-hour day in the apparel industry would lift a worker above – often far above – that standard. For example, in Honduras, the site of the infamous Kathy Lee Gifford sweatshop scandal, the average apparel worker earns $13.10 per day, yet 44 percent of the country’s population lives on less than $2 per day.
Powell’s and Skarbek’s lesson is straightforward and important. But it’s a lesson too often ignored by "activists" who would rather pose and prance as moral crusaders than analyze situations in ways that might actually help people. The lesson is summarized by what I call "The Economist’s Question: "As compared to what?"
In and of itself, situation A is neither good nor bad; it is good or bad only in comparison with it’s real alternatives. This lesson is a hard one, perhaps — it’s certainly an unromantic one — but it’s indispensable for sound analysis.