Sebastian Mallaby takes advantage of the woes of the Detroit Three to take a shot at unions, saying "given the object lesson from the collapse of the Big Three carmakers, government should think carefully before empowering labor unions further." All well enough, but his next sentence says "The growth of U.S. government need not be an economic disaster. Sweden and Denmark combine large public sectors with fast growth in GDP per capita," and as Matt Yglesias notes, Sweden has a union density of 78 percent, and Denmark has a union density of 80 percent, and Japan has a union density of 19 percent, and America has a union density of 12 percent, and so maybe this example doesn't show what Mallaby was hoping it would show. Incidentally, unions do a lot of things, and it would be nice if people like Mallaby would explain which things they think are relevant to Detroit's collapse. For instance: Unions use collective bargaining to secure comprehensive health benefits for their members. This means unionized workforces often have higher labor costs than non-union workforces. But it's weird to solve the competitive disadvantage of workers having health insurance by taking away their health insurance. We want people to have health insurance! Other countries, like Denmark and Sweden and even auto powerhouses like Japan and Germany, have solved this problem by instituting national health insurance. None have solved this problem by getting rid of unions. Unions also secure higher wages for their members, though no one I know of has tried to argue that the wage differential is an essential component here. Or maybe it's pensions. Or work rules that make it harder to fire employees. Or maybe it's management's adversarial relationship with unions that makes collaboration with the shopfloor difficult. Or maybe it's none of these things. But "unions" are a big concept, and not necessarily a useful one for this type of analysis. After all, Vegas's hotels are unionized, and they're doing fine. Japan and Sweden and Denmark have more unions than we do, and Mallaby is arguing that they are also better models. Maybe the problem isn't unions, but "car companies," and America should get rid of those, instead. Which is all to say that you have to figure out what differs across the models. Mallaby, and others taking up the blunt anti-union conclusion, aren't actually offering hypotheses in that direction, which suggests that what we're seeing here is more an effort to use Detroit's crisis to explain their views on unions than it is an effort to use unions to explain what's happened to Detroit.
