A successful effort to unionize a workplace apparently reduces the market value of affected publicly-traded firms, even if there is no immediate change in their operating performance. In Long-Run Impacts of Unions on Firms: New Evidence From Financial Markets, 1961-1999 (NBER Working Paper 14709), co-authors David Lee and Alexandre Mas estimate that the average effect of a union win at a workplace is to decrease the market value of the affected business by at least $40,500 (in $1998) per worker eligible to vote, based on monthly stock prices for 24 months before and after a vote to unionize. Their simulations suggest that a policy-induced doubling of unionization in the United States would “lead to a 4.3 percent decrease in the equity value of all firms at risk of unionization.”
The decrease in equity value associated with unionization begins at the time the union wins its election and continues for about 15 months afterward. Calculations of the effects of a union victory suggest that it produces large negative returns of 10 to 14 percent. The authors also find that the effects are quite variable, depending on the degree of support for the union. When unions win elections with a bare majority, there is almost no union effect. But when unions win by a large margin, the effect can be as large as 25 to 40 percent.
papers.nber.org
corner.nationalreview.com
Unions, Cont'd [Jonah Goldberg]
Paul Kersey responds to the pro-union emailer in this post:
Jonah, about that pro-union email:
1. Your correspondent is right that rising labor costs create an incentive for the use of more capital, but that isn't the only factor in play here, because unions affect more than just wages — work and seniority rules can make capital investments less valuable. I have a copy of the Ford/UAW contract in 1941 here; it's 24 pages long. The 2007 agreement went on for 1200 pages. The notorious jobs bank that kept employees on the payroll for years while they weren't working was in part a response to automation. Unions can subvert the use of new technology too and it can be in their interests (at least in the short term) to do so.
2. Comparisons to Europe are interesting, but there's something of an apples-to-oranges factor here, you're dealing with more extensive welfare systems and different labor laws. It's worth noting that Ireland, which according to your correspondent had higher unionization rates in manufacturing and higher productivity growth between 1980 and 2005, also had a big drop in unionization, at least overall (I couldn't find specific numbers for Irish manufacturing) from 1980 to 2003 — with union membership falling from a high of 57.1 percent of the workforce in 1980 to 35.3 in 2003. Rising productivity and falling union density — even if starting from a very high point — means that the role of unions in terms of boosting Irish productivity is at best unclear.
3. Getting back to the US, our research on right-to-work laws found that RTW states had lower per-unit labor costs, meaning that whatever capital improvements employers had in non-RTW states were not enough to overcome wage and benefit costs imposed by unions.
4. If you want to compare the performance of unionized versus non-unionized companies in the US, why limit yourself to Costco versus Wal Mart? Anyone want to talk about the auto industry?
5. Your union correspondent is right about one thing — there's nothing in economic theory that says that unions have to be a drag on an economy. In my humble opinion unions the real problem is a clunky federal law (The National Labor Relations Act) that divides the workforce into rigid bargaining units and then says that either everyone in the unit is represented by a union or nobody is. I can't think of any other situation in the US where a competent adult can have a representative chosen for him by others without some way to opt out as an individual and pick his own rep or negotiate for himself. The practical effect of the law has been to make unions largely unaccountable to the men and women they represent. If the lines of accountability were clearer, it would probably be easier for unions to win worker support on the merits and union officials wouldn't need to push for things like card-check and binding arbitration. I don't think that's what we wanted as a society, and I don't think union officials themselves realize just how out of whack the incentives they work under are. But that's the law we have, and it's a big part of the reason why unions in the US have the sort of negative effects, and poor reputation, that they do.
corner.nationalreview.com
Unions & Productivity [Jonah Goldberg]
Last night's post from my union-political-director guy (in response to this) elicited much consternation and vexation among Corner readers. The key problem, as many noted, is the claim that unions improve productivity by making labor more expensive. The UPD-G made this point in both of his posts, and it has bothered me from the get-go. If I understand it right, the argument goes like this: Unions make labor more expensive, so employers try to avoid hiring expensive workers and instead invest in productivity boosting equipment and the like. I kept thinking I misunderstood this point, because it seemed to me that pro-union forces wouldn't want to admit that hiring fewer Americans and automating industry is why unions are "good for the economy." This, it seems to me, is an argument for why unions are good for union members, which I never really disputed. Anyway, my Productivity Guy finally caught the bat-signal and chimed in. Here's his take:
Jonah - I saw the message from your union political director guy, and thought it was time for your productivity guy to chime in...
He notes (correctly) that per unit labor costs and productivity aren't the same thing - but "productivity" and his narrow definition of "productivity" aren't the same thing either. He defines productivity as output per unit of labor hours, but this is labor productivity, which is only a partial productivity measure. A more comprehensive measure of productivity is total factor productivity (TFP), which is output produced per total resource inputs (labor and capital). He's claiming that higher union wages increase productivity, because they force firms to use more capital or better technology (broadly speaking, to substitute capital for labor), which in turn benefits the entire economy. Let's investigate this scenario....
Suppose a firm has two options it can choose to produce all-purpose widgets - a labor intensive option, which uses a relatively large number of workers and few 'machines,' and a capital intensive option, which substitutes machinery for workers. Both options are available to the firm, and if both labor and capital markets are unconstrained (i.e. no unions) and the firm selects the labor intensive option, it has done so because it is more profitable than the capital intensive technology. This is the same thing as saying that the labor intensive technology is more productive overall (i.e. leads to lower overall costs, and doesn't impact revenues assuming the same, given number of widgets are produced in either case). TFP is therefore higher, but since the firm used relatively more labor, labor productivity will be lower compared with the other option (increase use of an input and its marginal productivity declines). Now let's investigate the union political guy scenario, where wages are higher - if wages get high enough, this could in fact lead the company to select the capital intensive technology. This uses fewer workers and therefore raises labor productivity. But the TFP of this technology must necessarily be lower, because if it wasn't it would have been selected in the initial, unconstrained case.
You can also look at it another way...the basic economics of unionization are that unions set wages that are above unconstrained, market levels, which necessarily leads to a lower quantity of labor being employed. If firms are going to hire relatively fewer workers, which ones are they going to keep? The more productive ones. If any firm lined up its workers from least to most productive and fired the bottom half, the productivity of its workforce will necessarily rise - but that doesn't mean it's a good idea, or that it will make the company more productive or profitable overall.
I confess I don't know the literature on unionization and productivity very well, but if this is the best they can do it seems like the case is closed. Artificially raising the price of labor is not a path to higher productivity or prosperity. It's also well known that focusing on partial productivity measures (like labor productivity) can be misleading, and TFP is a more reliable metric.
PS A personal note - my father was an electrician and an IBEW member his whole life. I'm not necessarily anti-union, particularly in the early days of industrialization when they helped to establish safer and more humane conditions in many workplaces. But the case for the merits of unions shouldn't rest on far-fetched arguments that they're forcing companies to become more productive.
corner.nationalreview.com |
