Tuesday, February 22, 2011


There is currently serious conflict in several states, including Wisconsin, Ohio, and Indiana over union bargaining rights for state employees. The claim of the Republican administration in each of these states is that union power needs to be curtailed in order to get the state budget under control.

I have some personal experience with labor unions, having actually been a one-time union member (I can't say "card-carrying," as I don't remember having been issued a membership card). When I was 19, I worked for a couple of summers for Winchester Canada, which had a unionized plant in Cobourg, Ontario. Winchester appears to still be in business, and of course they make firearms. An ancient Winchester product was what Chuck Connors used in "The Rifleman." When I was 8 years old my friends and I thought that Chuck Connors was very cool.

Working for Winchester was quite weird on many dimensions, including the fact that I had absolutely no interest in guns - a lack of interest that continues to this day. I am allergic to firarms. I see one and run the other way.

An interesting feature of working life at Winchester Canada was that the union was clearly stifling innovation. What I saw certainly conformed to the picture of unions that comes out of Parente and Prescott's "Barriers to Riches," which cites labor unions and trade barriers, among other things, as factors that lower TFP (total factor productivity) and the standard of living in a country. At the Winchester plant, every activity was governed by a set of workplace rules, written up in the union contract. Management and production workers thought of each other as adversaries. This was not a conducive environment for thinking up new ways of organizing production or designing new products. It seemed clear that you could take away the union, create more flexibility, increase productivity, pay everyone more, and bring about a Pareto improvement.

Sometimes I get into arguments with my family in restaurants about economic questions ("Dad, don't get so excited, people are looking."). I enjoy this, perhaps as much as, or more than, helping my sons solve math problems. When unions come up, I take the hardcore laissez faire route, just to be provocative. My youngest son will then tell me about what the union movement accomplished in terms of workplace safety. Maybe he has a point. We don't want to go too far in seeing externalities under every rock, but this certainly looks like one. Getting firms to invest efficiently in safety may require collective action. Maybe the government would not have taken appropriate action on workplace safety if not for the labor movement. In the Winchester plant, I was certainly pleased with whatever safety features were provided for me. My foreman liked to encourage safe practices by showing us his missing thumb, which had been lost while working on the same belt sander that I was required to operate.

One simple way to look at unions comes from Econ 101, where we just apply standard monopoly power arguments. Labor law gives workers the right to effectively act as a monopoly seller of labor. Result? The union drives up wages and extracts rent from the firm. But that argument goes only so far. As long as the firm faces competition, this has to discipline the union. Extract too much rent and you drive the firm out of business.

So what is going on in Wisconsin, Indiana, and Ohio? In general, union organization is not an easy thing in the United States, relative to what happens in other rich countries. Twenty two states, mainly in the south and in the middle of the country have right-to-work laws. In some states, state employees have much less power to form unions relative to what exists in the private sector. However, in Western Europe, unions tend to be relatively powerful. In Canada, labor law is much more conducive to union formation and power. For example, most (if not all) Canadian provinces do not allow the hiring of permanent replacement workers during a strike, and some will not permit the hiring of temporary replacement workers. Strikes of public service workers in Canada are infamous, from old-time disruption in the post office to more recent strikes involving garbage collectors and transit workers in Toronto. The difference in labor laws in Canada and the US is reflected in unionization rates. The US has a unionization rate of only 7% in the private sector, and 29% in the public sector. In Canada, the comparable statistics are 16% in the private sector and 71% in the public sector.

Now, if we believe Scott Walker, the Governor of Wisconsin, public spending in Canada should be wildly out of control. We know, of course, that government is doing much more redistribution in Canada than is the case generally in the United States. But in Canada actual expenditures of all levels of government on goods and services amounted to 21.2% of GDP in Canada in 2009, and 20.6% of GDP in the US. Not much difference there. Further, in spite of union power in the public sector, the Canadian federal government was able to turn around a deficit which had exceeded 5% of GDP in the mid-1990s. Before the recent recession, the Canadian federal government had been running surpluses for several years. We all know how that compares to recent US fiscal performance.

Is Scott Walker likely to save much money by picking on his public sector unions? That's very doubtful. He's certainly creating plenty of unproductive conflict. Is what he is doing politically smart? That's hard to tell. Picking a fight with unions in Madison, Wisconsin may not be the brightest idea. Anyone who has spent time in Madison (4 years for me) knows that there is a large reserve army of people who would enjoy nothing better than spending a couple of weeks camping out in the State Capitol building to bother a Republican Governor. This might play well in the rest of the state, however, where Madison is sometimes viewed as sin city.


  1. >An interesting feature of working life at Winchester Canada was that the union was clearly stifling innovation.

    Aha. So you think our Mercedes, BMW, Porsche are not very attractive innovative products. I mean the work force belongs to a union and the union even sits on the board of these companies.

  2. So you think our Mercedes, BMW, Porsche are not very attractive innovative products. I mean the work force belongs to a union and the union even sits on the board of these companies.

    Maybe if it weren't for the unions they would have invented flying cars by now, like in the Jetsons... ;)

  3. Are there any good numbers on how overpaid american public union workers are to canadian public union workers? As a Canadian it seems like our public unions are much more willing to work with government in times that need us to pull back on spending. I know in Ontario there was recently talk, and maybe it went through, of a wage freeze on public employees without much fuss from the unions. It might be media bias to report only on the striking unions, but it seems like American unions will strike for any small reason, like having to pay reasonable amounts into their pensions plans.

    Just some thoughts, I really know nothing about how unions in America are structured.

  4. Stephan,

    Yes, I can imagine organizational arrangements where there is something called a union, and something called management, and together they are collectively making decisions about how the firm should be run that are indeed efficient. From what I know about management science, as practiced by people in business schools, there appear to be fads in that area attached to particular business models, and for some reason the cases often relate to auto manufacturing. The idea basically seems to be: American model bad; Japanese and German models good.

  5. Ian,

    The Canada/US difference seems puzzling to me, which is in part why I mentioned it. It would be useful to know what the numbers are on strikes, but my impression is just the opposite of yours. The only strikes I hear about are the ones in Canada, particularly in the public sector. It's not clear we should focus just on strike activity anyway, as the power of unions is not all reflected in strikes: the threat of a strike can matter a lot without us every seeing the phenomenon in equilibrium. Ontario, for example, has very strong teachers' unions, and the result is that there appears to be a persistent excess supply of people who want to teach at the going wage. This all seems puzzling, as you would think the federal and provincial governments in Canada would be unable to control spending, which appears not to be the case. Has anyone tried to argue that arrangements like this could be efficient?

  6. @Stephen Williamson
    The car industry was just an example. In every mid-size and large German enterprise there's a union involved. It's called Mitbestimmung (participation) where elected union members serve on the board of companies. And wage bargaining is organized along industry sectors with binding results for all members. I think this model served Germany well because it prevents any single industry sector from the fallacy of composition. For one enterprise it might be beneficial to slash wages to gain some competitive advantage. But if all competitors do the same it simply lowers nominal demand in the economy.

    I think the problem in the US private sector is too few unions and not the other way round.

  7. Steve,

    I really like this topic too. Here is a nice paper I saw in a conference that explores the theoretical link between unionization and productivity: http://www.laef.ucsb.edu/pages/conferences/gad10/papers/bridgman.pdf

  8. @German Cubas
    Cool. One more of these papers elaborating pages over pages about the economy on Alpha Centauri. Here we go with SUPER realistic assumptions: Each household owns a representative portfolio of shares in all firms and own the capital. More productive vintages of the technology arrive exogenous and can be costless adopted by firms. And on an on and on … Assumptions 3.1 to 3.6 are only ridiculous. But for sure you will enter with this paper the competition in Harvard who can come up with the a lot of math masking the void.

  9. German,

    Thanks for the paper. Jim Schmitz has done some interesting work in this area too:



    Apparently you are willing to make some bold statements about how wonderful unions are, but you are not interested in science.

  10. @Stephen Williamson
    Stephen. I'm not an economist. Although I work with them. I'm a mathematician. And let me assure you. This is not science. This is Feynman's cargo cult par excellence.

    But let's not quarrel. I have learned a lot in the past years about your economists bio top and that there's no good argument available to stop you pseudo-scientist from going rogue.

    I hope you forgive me to post in the future some irritating comments on your blog. I find it interesting and as long as you can publish some Euler equations without interference of people who actually understand them you should be more than happy. Or?

  11. Stephan,

    We have something in common. In my younger days, I was a mathematician too. I once had a conversation with Bill Zame (http://www.econ.ucla.edu/zame/), who has a PhD in math, and later moved into economics. I told him I thought that with his math toolkit, he would have found economics very easy. He said that was not true, and that he had spent a lot of time "getting" economics. The logic does not necessarily come naturally to a mathematician, i.e. doing social science is a tricky business. Feynman actually got economics to some extent. I don't know where the quote is, but he understood incentive problems, that GDP is not some fixed pie that you can split up at will, and that redistributing income can change the size of the pie.