Globalization and Technology Cannot Explain Union Trends

Evan Soltas has been on a tear about unions, which has also generated a huge wave of responses in blogs and elsewhere (decent summary of the action here). It's clear that he is not a fan of them in net, but it's also clear that he has only a limited understanding of what they do.

For instance, in his initial take, he did not mention anything about their political role. He trumpets the importance of full employment while somehow never asking which political actors are meant to push the federal government to make that happen. Capital certainly won't: it is often benefitted by a buffer stock of unemployed people and high interest rates.

This is just one example, but it gets at the core of the problem with Soltas' approach here. He endeavoured to answer the union question with wonk-economics methods, when the union question is one of political economy.

This political economy blindspot is present both in his assessment of the value of unions and his assessment of what drives unionization trends. In Soltas' initial take, union decline is driven purely by economic fundamentals:

Economic change, particularly the rise of global competition, killed them. 

But unionization trends differ substantially across countries, all of which are confronted with global economic forces. The evolution of economic forces is an attractive theory, but it cannot account for cross-country differences.

Rates of union coverage (how many people are covered by a union agreement) and union membership (how many people belong to a union) vastly differ across developed countries (2007 figures):

Among these countries, union coverage spans from 13.3 percent in the United States to 99 percent in Austria. Five countries have greater than 90 percent union coverage. Another eight countries have more than 60 percent union coverage.

Part of the confusion (Americans in particular) seem to have with the union discussion is that they do not recognize the distinction between union coverage and membership. Countries serious about unions and collective bargaining can build those things into their labor market such that overt membership becomes basically unnecessary. Using membership figures to track the influence and penetration of unions in those countries is insufficient, particularly if you are trying to track the viability of unions in the face of global economic changes.

When it comes to trends, union coverage has declined overall among these countries, but that does not tell us much. Grouping the countries by their economic systems in the way comparativists typically do tells the real story (change from 1980-2007):

As best as I can tell, all of these countries are on the globe, yet somehow the fundamental economic forces of globalization and technology have affected them in very different ways. Again, this points to the dominant role a country's political economy plays in how its labor market institutions are structured.

These cross-country differences pose a considerable challenge to the economic-determinist theory of union decline. Soltas might come back and say that there is at least a political-economic-determinist theory of union decline for America because of the kind of country it is (liberal market), but this is a much different point.

Even this point, it should be noted, runs into other problems though:

Our liberal market neighbor to the north has somehow taken a very different path than us, which again suggests that economic fundamentals do not require the kind of union decline America has seen (this graph uses membership instead of coverage, but in Canada and the U.S., which do not have industry-wide bargaining, the two are essentially the same thing).