Polanski's statistical model differs from previous models in a couple of ways: She doesn't unrealistically assume that developing countries will run at full employment, as most World Bank studies do, for instance. And what her model finds is pretty revealing. Basically, any of the "plausible trade scenarios" that could emerge from the talks would only produce a one-time gain to the world of $40 to $60 billion. That's nothing, really. That would only amount to pennies a day for everyone in the world—if the gains were distributed evenly, and there were no negative effects to trade liberalization.

But the gains aren't distributed evenly. There are winners and losers. In Polanski's model, the United States, the EU, Japan, and China would all benefit greatly from either the Doha or the Hong Kong liberalization proposals—China would gain anywhere from 0.8 to 1.2 percent of GDP. Many poorer countries, by contrast, would suffer pretty heavily: Sub-Saharan African countries would see a 1 percent drop in income if a liberalization agreement was reached, while Bangladesh and many countries in East Africa would lose anywhere from 0.1 to 0.5 percent of GDP.

Poorer countries, as it turns out, would be hurt by the reduction in agricultural tariffs and subsidies now under discussion. Partly this is because, as I mentioned long ago, many developing countries are net food importers, and depend on existing farm subsidies to keep food cheap. For other countries, meanwhile, the reduction of tariffs would hurt subsistence farmers, who make up the bulk of production in many developing countries, and can't compete in a "free trade" world market. And in many countries, farmers driven off their land won't be able to find manufacturing jobs very easily.

That brings us to manufacturing. Trade liberalization for manufactured goods would create some benefits, and would increase the demand for unskilled labor in many poorer countries. But that increased demand almost certainly wouldn't increase wages, partly because those developing countries won't be running at full employment—most cities will see a surplus of agricultural workers kicked off their land (which means that it's slum time).

Plus, as pointed out before, these figures omit all the costs from these WTO talks. Many developing country governments, for instance, derive a significant portion of their income from tariffs. How will they make up those lost revenues once the tariffs are slashed? By cutting social services? Running deficits? Raising sales taxes? Devil in the details. And what about the cost of WTO intellectual property agreements, which can be, in many cases, written primarily for the benefit of corporations in the United States and Europe?

The utilitarian perspective on all this is interesting, meanwhile. By one count, China has 200 million desperately poor people living on less than $1 a day, and 600 million subsisting on less than $2. Some of those people will see their lives improved—albeit very marginally—by a Doha pact. But in the countries that would lose from Doha there are roughly 267 million dollar-a-day folks, and 486 million people living on less than $2 a day. So almost as many people become worse off as become better off. Which is to say again that "Trade, not aid," is a pretty weak strategy for alleviating world poverty. And, at the very least, the Doha negotiations need to be tweaked to protect some of the poorer countries from the negative effects here.