Last week, Seth Ackerman wrote a Jacobin blog post in which he gave us a snarky attack on the record of “left neoliberalism” in the United Kingdom. Basically, he showed that while New Labour managed to reduce poverty somewhat with cash transfer programs, the progress was meager and could not be sustained. Since the programs were financed out of a series of asset bubbles, the UK has seen poverty go back up again with the recent crisis.

I don’t have much quarrel with this account, but I’m not sure it can bear the weight of the argument that Seth wants to put on it. He suggests that the UK experience is a refutation of the general strategy of progressive neoliberalism, which Freddie DeBoer felicitously dubbed “globalize-grow-give”:

First, you embrace the standard globalization model of reduced or eliminated tariff walls, large free trade agreements such as NAFTA or CAFTA, deregulation, and general trade liberalization. This encourages international trade and the exporting of jobs from highly-regulated, fairly well compensated, high worker standard of living places like the United States to the cheap labor, low regulation, low worker standard of living places like China or Indonesia. This spurs international economic growth in both the exporting and importing countries. Here at home, higher growth results in higher tax revenues which can then be redistributed from those at the top of the income distribution (who have benefited from the globalized trade regime) to those at the bottom of the income distribution (who have been hurt by the globalized trade regime that undercuts their wages and exports their jobs).

I think that if you want to really criticize this view, you need to look beyond the UK, which is neither a very generous nor a particularly well-designed welfare state. As it happens, my day job involves analyzing cross-national income data, so I’m going to perpetrate some social science on y’all.

The way I read the “globalize-grow-give” critique, you can extract an empirical claim about how the income distribution should look in a G-G-G economy. The distribution of income before taxes and transfers will become increasingly unequal due to deregulation and globalization, but the distribution after taxes and transfers are accounted for will not become vastly more unequal because government is compensating for the inequality in the private market.

To test this, I did some simple calculations, following other researchers who have done similar things. Using data from the Luxembourg Income Study, I calculated the Gini coefficient, a standard measure of inequality, for several different countries. I calculated two different Ginis:

• The Gini of market income. Market income is defined here as income from wages, pensions, self-employment and property. This is income before any taxes or transfers are accounted for.
• The Gini of disposable income. This is the income that people actually have to spend, after taxes are deducted and any transfers are added in. (For more details about the variables, see the postscript).

Unfortunately, the difficulty of harmonizing cross-national data means that the numbers I have access to are a bit out of date — specifically, they end before the current crisis period. I still think we can learn something useful from them, however. The way G-G-G neoliberalism is supposed to work, the Gini of market income should go up but the Gini of disposable income should not — or at least should rise more slowly. We can think of the difference between market income inequality and disposable income inequality as a rough measure of the amount of redistribution done by the state.

So here’s what things look like in the UK:

This figure basically supports Seth’s argument. Market income inequality has gone way up in the last few decades, but disposable income inequality has gone up by a lot as well. The state is doing a bit more redistribution than it used to, but not enough to make up for the rise in private-market inequality. If you look at the United States, the situation is even worse, as the state has done essentially nothing to counter rising inequality in market income:

The question, though, is whether it has to be like this. Let’s put the UK alongside another rich European economy, Germany:

Here we see something very interesting. Before you take taxes and transfers into account, the rise in inequality in Germany looks very similar to what happened in the UK — indeed, the two countries converge to almost the same value by 2005. But disposable income inequality has stayed flat in Germany, because the German state has used taxes and transfers to counteract rising inequality.

Every good social democrat loves the Nordic model, so let’s finish off with a look at Sweden:

Here the story is a bit different — both market income and disposable income inequality have remained pretty flat, although both have risen a bit. The important thing to note here is that even in the most socialist of welfare states, market income inequality is very high, nearly as high as it is in the UK or US. The fact that Sweden is one of the least unequal countries on earth has to do almost entirely with taxes and transfers.

So what can we conclude from all this? Let me be clear that don’t think this is a knock-down argument in favor of “globalize-grow-give” as a political model. But I think the best argument against the G-G-G model is not that it’s economically impossible or dependent on asset bubbles. Rather, I’d point us back to the political arguments enumerated by me, Henry Farrell, and Cosma Shalizi among others. What makes Sweden and Germany different is not that their economies are different from those in the US and UK (although they are), but that they have different political environments, featuring things like a hegemonic Social Democratic party in Sweden and a strong labor movement in Germany.

So if left-neoliberalism is to be a workable political agenda rather than the motto of useful idiots for the “globalize-grow-keep” agenda of the right-wing neoliberals, it has to either make its peace with the sources of working-class power that currently exist, or else come up with workable models of what might replace them.