« Hong Kong | Main | Why isn't The Wall Street Journal free on-line? »
Agreeing on unions?
Ezra Klein has an interesting post on union elections, favoring greater unionization; Jane Galt is not persuaded. [Addendum: Mark Thoma has more.]
I propose a deal. I'll agree that unions, in the best natural experiments we have, boost wages by about 10 to 20 percent. On the other hand, will Ezra (and others) agree that unions are mostly detrimental to the rate of economic growth?
If so, the utilitarian evaluation will boil down to the choice of discount rate, keeping in mind that under the left-wing account the gains follow mostly from redistribution more than from wealth creation.
Admittedly the empirical literature on unions and economic growth is murky. But it does seem that in non-socialist societies, more unionization lowers the growth rate. (In socialist societies, it may be that economy-wide unions internalize some poliitical externalities and lead to better policy, a'la Mancur Olson.) These papers are easy enough to criticize, so I'll admit I am more convinced by simple theory here than by any empirics. Unionization raises the cost of many growth-enhancing business decisions, such as layoffs or implementing technical progress.
Union supporters? Do we have an epistemic deal about how you are willing to lower the growth rate? And can we pull your true discount rate from the Stern/global warming debates? (I recall Jane once writing that a zero discount rate would require her to revise everything she believed, but I think the opposite is sooner true.)
Oh, did I mention that the union wage premium, especially for private sector employees, has been declining and may be disappearing altogether?
And no, I will not be moved by sarcastic attempted reductios which pretend that enslaving the workers would raise the growth rate too. Figure out yourself what is wrong with that.
Addendum: Johan Richter asked for: "Your preferred policy towards unions," so I'm calling this #12 out of 50. My preferred policy is laissez-faire, noting I have never been a right-wing, union basher, Morgan Reynolds sort of guy. But I see encouraging unions through the law as a relatively short-sighted solution to the problems of labor and the desire to raise living standards.
Posted by Tyler Cowen on March 4, 2007 at 07:32 AM in Economics | Permalink
Comments
If so, the utilitarian evaluation will boil down to the choice of discount rate
Sir Nicholas Stern has got one you can borrow.
Posted by: Paul Zrimsek at Mar 4, 2007 8:26:48 AM
As I posted elsewhere:
We are in the era of globalization - the Flat Planet.
Labor at ALL levels is now or is going to be a competitive market.
We have already seen the movement of the jobs related to assembly in factories. We have seen businesses that rely on telecommunications (eg computer Help Desk) migrate to low cost providers. We are seeing engineering and computer programming become global professions.
I believe that the flow of money will also affect who owns what businesses and how much CEOs get paid.
but... back to the subject: Unions.
A Union in the USA is almost a joke if the industry is global. Unions need to go forth to the world and raise up the working conditions and wages of the 75-80% of the planet that is so much lower than the USA, Europe, Japan et al _OR_ they can simply unionize here and watch as the jobs go elsewhere.
A second issue I have with Unions is: entrenchment. Many unions seem to have been in place for so long their mission is no longer to right-a-Wrong which may have existed when the union was formed but rather to preserve the tenure and power of the Union itself.
Posted by: anon at Mar 4, 2007 9:13:13 AM
As a card carrying Union Member, personal experience leads me to believe that Unions are bad for everyone in the long run.
Unions discourage people from broadening their skills (only do your job or else!).
Unions encourage people to rely on seniority as opposed to effort to advance themselves.
How can this be good for anyone?
On the other hand, if unions performed a signaling function, they might have some use. In other words, if belonging to union indicated that you were a hard worker and knew your stuff, they would benefit everyone. It would save employers on screening costs and save employees effort in trying to prove their true worth to potential employers.
Posted by: The Chieftain of Seir at Mar 4, 2007 9:13:54 AM
I hope you'd agree, though, that growth _as such_ isn't something most people should care about. It's like (and usually isn't much different from) a move to a more Kaldor-Hicks efficient state. Why should those who don't benefit care about it? If crushing unions would lead to higher growth, but the majority of that gain would go to management, making workers less well off, why in the world would they think that's good? This sort of agregate utilitarianism just has no obvious noramtive force at all.
Posted by: Matt at Mar 4, 2007 9:49:21 AM
"But I see encouraging unions through the law as a relatively short-sighted solution to the problems of labor and the desire to raise living standards."
What are the better, non-union solutions?
Posted by: Brian at Mar 4, 2007 10:13:51 AM
Tyler,
I think you miss the point if you think that the main benefit of a union is wages. I spent years in union jobs in nuclear power plants (operations) and oil refineries (hf acid alkylation). While I was glad that I got reasonable wages, I cared equally about job security and safety. It was a comforting thought when the reactor had tripped and I was trying to bring it back on line that I wouldn't be fired if I wasn't fast enough in carrying out the reactivity calculations. In the refinery, I knew that the union would back me up if I felt that we needed to do more to safe out an acid pump for maintenance.
I know that professors are unused to the sensation of being employed at the bosses whim, but that is the case for most people.
Posted by: Robb Lutton at Mar 4, 2007 10:16:00 AM
Robb,
Didn't you know that Tyler's as much a mystic as a scientist?
Got to keep the faith, dude....
Posted by: Martin at Mar 4, 2007 11:10:45 AM
Echoing Robb, I generally think there's a good point that unions may be able to protect stockholders (and others) by monitoring managers. Generally, it's not in the interests of stockholders to have managers that act like feudal lords.
I think, however, that unions are just as likely to team with managers to screw stockholders, or that managers buy off the unions in exchange for the opportunity to screw stockholders. And unions often oppose productivity-boosting changes that lead to firings, and that's not so good. I think unions still seem locked into a silly 1930s mindset about labor relations.
I think unions would make themselves more useful and would focus on ways in which they improve workers conditions by helping make the company more profitable if they didn't have certain special legal protections. I also think we should re-eaxamine the Wagner Act and startallowing company unions. This would allow more stockholder-worker alliances.
Posted by: Keith at Mar 4, 2007 11:14:45 AM
The first unions were not only about wages, they were about safety. I remember seeing stories on how, before unions, electrical workers would routinely die because the companies refused to implement safety rules.
Unions also help raise the quality of life of union members. My granmother worked in the sweat shops of the Lower East Side. I would want that for no-one. Yes, that could not happen today because of government regulations, but, do you really think we would have the regulations had the unions not been there?
Unions also are essential to the rise of the United States. While it may be true that unions can slow productivity, that is just the supply side of the equation. By raising wages and ensuring job security, unions enable workers to be confident that they will continue to receive wages and will purchase more.
So, rather than being a net negative on the economy, it may be that unionization is a net gain for the economy and benefit the nation as a whole.
Posted by: Allan at Mar 4, 2007 11:27:56 AM
I'm hardly a huge union booster, but I am a member of one, and I do work at a power plant--I'm going to echo what's being said in the last few comments to a certain extent, though I think "essential" is a massive overstatement.
People don't bargain correctly for individual risk. I can give you pages of examples of people underestimating, or misestimating, their individual risk. When in a relatively dangerous environment, such as a factory or power plant, collective bargaining improves (as far as I can tell, at any rate) the ability of the bargaining unit to assess risk, because it allows reasonable application of statistics by reasonably well informed people. It's much easier to look at a 600 person bargaining unit working in a mine and say "you know, *someone* here is going to need a really good health plan within the next 3 years" than it is for an individual 25-year-old to accurately consider their chance of ending up in the hospital.
That being said, I think there are a lot of negatives. The only person who benefits from a seniority system is the less qualified person who happened to get in the door early. I personally think it's a huge mistake to reward those people, and the vast majority of union-negotiated contracts have seniority provisions. I don't need a union to make me confident of finding or keeping employment--that's *my* job, by continuing to improve my skills and knowledge. Placing the burden of keeping me employed on someone else is just asking for a rude awakening.
Posted by: Steve Sandvik at Mar 4, 2007 12:06:12 PM
sandvik above makes a good point regarding the assesment of risk. It seems to be an under-explored idea when discussing unions.
Posted by: sa at Mar 4, 2007 12:17:01 PM
The points about unionism, safety, and the inner dynamics of organization are well taken. I would ask you to consider yet another possibility: unions frame the economic choices facing workers in a way that
Think about the limited-liability corporation. Looking purely at the essence of this legal construct -- limited liability -- one might be deeply puzzled as to why limiting entrepreneurial liability is a good thing. The theory behind it implies that, given rational entrepreneurs and investors, when the government limits their legal and financial liability the entrepreneurs will take on inefficiently high levels of risk. But I don't think that is what happens in practice -- in practice, limited liability corporations are a phenomenal way of coordinating economic activity by focusing confused, boundedly rational agents on solving real problems, rather than anxiously avoiding disaster.
Does that make sense? So the theory of, say, limited liability, doesn't quite capture the difference between the social role of an investor who can stick to limited liability entities, and the social role of an investor who cannot. Similarly, the choices and life paths available to a teenager who is quick but not brilliant are quite different in contexts where he will probably be in a unionized job, and where he will probably have a series of non-unionized jobs. His incentives to learn a particular craft, trade or profession, his incentives to select certain jobs over others, his incentives to voice concerns about the way he is being managed, are all deeply important.
This is not a *convincing* defense of unions, but I hope it explains why we should not be too sure about the economic theory of unions. (I imagine you are referring to monopsony.) There are theoretical points on either side of the debate -- we need to make the empirics less murky, rather than entrusting ourselves to pure theory and a few regressions.
Posted by: interestring at Mar 4, 2007 3:10:18 PM
Thanks for the reply. Just a clarification. Laissez-faire would mean that an employer can fire workers for joining unions, allowing employment contracts that prohibits strikes etc?
Posted by: Johan Richter at Mar 4, 2007 4:49:39 PM
I sure hope that's what he means by laissez-faire.
Posted by: R. S. Porter at Mar 4, 2007 4:59:53 PM
Much of the anti-union arguments revolve around economics (e.g. expense, efficiency). Much of the pro-union arguments revolve around social justice (e.g. safety, equitability). There is no hope for common understanding if the sides are approaching this from disparate ground. Economics doesn't trump social justice and vice versa.
Corporations are merely fictitious entities given legal status by society. Viewing business and society strictly through economic terms is both unnatural and can lead to devastating social consequences. Unions are another fictitious entity given legal status to balance influence and goals of corporations.
Economists have a way to view this situation, if only they would use it. Instead of looking for equilibria which maximize profit or efficiency, seek equilibria which best balance competing concerns, both economic and social.
Posted by: squik at Mar 4, 2007 5:05:43 PM
If you see encouraging union through the law as a bad thing, would you also agree that discouraging unions through the law is a bad thing?
So the question is how do you make the law neutral?
Is the current proposal to require secret ballots neutral?
Seeing who favors it and who opposes it would suggest that it is not a neutral law.
Posted by: spencer at Mar 4, 2007 5:20:32 PM
Mr. Cowen is underestimating the effects of unionization. While periods of increasing and decreasing union membership show no correlation to overall economic growth, they do correlate to increasing and decreasing income equity. One should not ignore the fact that union membership peaked around 1980, about the time that middle and lower middle wage began to stagnate. Since then, union membership has declined and worker wages have not kept up with the growth of income for the upper percentiles. Americans may not like to think of themselves as union labor, but they were better off when they were.
Posted by: Mike at Mar 4, 2007 5:22:37 PM
Johan Richter raises an important point.
Without, at a minimum, job protection for organizers and union members it would be close to impossible to form a union. So a complete "laissez-faire" approach is in effect opposition to unions.
To say that one favors that approach, without qualifying it somehow, is not taking a neutral position.
Posted by: bernard Yomtov at Mar 4, 2007 6:35:19 PM
I'm torn about unions. I recall having some heated arguments with my dad, who was a member of a union. However, when the company he worked for went bankrupt after his death, the union did nothing to protect the widowers and widows - they only cared about dues paying members. Eventually, the widowers and widows went to court and won their case. Pensions were maintained. That event verified my opinion of unions.
Fast forward 20 years. The overwhelming majority of my working life has been in non-union employment, though about half of it was and is in a unionized environment. Currently, I work as a non-union professional for a municipal government. In about a month's time or so, I will be a unionized professional.
Due to massive personnel changes in senior management, half-hearted attempts at corporate reorganization, questions about secondments and new hires, a '40 hour' work week that varied from 35 hours to 44 hours per week depending on your position, lack of communication between non-union employees and senior management and a general distrust of senior management, the non-union employees decided to form an independent trade union.
It wasn't always like this. Their was an informal agreement between the association representing the non-union employees (a voluntary social association) and senior management, that the non-union employees received the same wage increases and benefits as the union employees. In return, we worked during strikes by union employees, worked longer hours and were willing to discuss with senior management any budgetary and labour issues. Flexibility was assumed.
Unions bad? Unions good? I'll let you know after we negotiate our first contract.
Posted by: Vincent Clement at Mar 4, 2007 8:51:56 PM
I hesitate to bring up this anecdote because I can't prove that it happened, but I was told the story by one of the people involved, with whom I had a relatively long working relation who was a member of the involved union but was [he had to be] but who didn't like everything they did.
At the time, ca. 1977, I was doing some contract programming for a major east coast electric utility. Utilities have energy control centers, which are places where the supervisory control gear is concentrated and the operators make changes to meet the needs of power generation and and transmission and distribution. I worked for this company: http://www.qeiinc.com/ .
It would be embarassing to say the least if the energy control center didn't work during a power outage, so they had backup generators that took a few minutes to get online -- and they had batteries, two rooms full of impressively large lead/acid batteries, each of which could keep the place running for about an hour with enough power left over to get the generators started. [I presume they're using more refined battery technologies now, but this was thirty years ago.]
Lead acid batteries don't last forever, even when they float, so once a month someone was supposed to take a dummy load [a fan-cooled resistance element] and bridge it across each cell in turn for a few minutes, and write down the voltage. This chore took about an hour or two. Things were designed so that if there was a power failure while the battery bank was being tested the batteries would still deliver.
One battery room was routinely "checked" by one particular worker -- adjacent to his lunch hour. Apparently he "checked" the batteries by writing down plausible benign voltages in the log book and taking an early lunch. After a while a supervisor noticed that one battery room was wearing out a cell every so often, as expected, but that the other never seemed to need any replacements. He [the supervisor] did the test himself and found out that all h*ll would have broken loose if there had been a power failure and the redundancy of having two battery rooms turned out to be needed because the other one wouldn't come online, say because some relay didn't do its job.
Naturally, the utility tried to fire the guy. The union hired expensive legal talent and succeeded in making it not happen. The question of why the union traded away raises the diligent employees could have had in return for contract clauses that made it possible to prevent an employee who was not doing his job and could have caused a fifteen minute glitch to turn into a nightmare for a major utility is left as an exercise for the reader.
-dk
Posted by: Dick King at Mar 4, 2007 11:07:29 PM
I apologize for the double posting. For some reason the last paragraph-and-a-half got chopped off my previous attempt to post this story. I hope I haven't exceeded some length limit!
I hesitate to bring up this anecdote because I can't prove that it happened, but I was told the story by one of the people involved, with whom I had a relatively long working relation who was a member of the involved union but was [he had to be] but who didn't like everything they did.
At the time, ca. 1977, I was doing some contract programming for a major east coast electric utility. Utilities have energy control centers, which are places where the supervisory control gear is concentrated and the operators make changes to meet the needs of power generation and and transmission and distribution. I worked for this company: http://www.qeiinc.com/ .
It would be embarassing to say the least if the energy control center didn't work during a power outage, so they had backup generators that took a few minutes to get online -- and they had batteries, two rooms full of impressively large lead/acid batteries, each of which could keep the place running for about an hour with enough power left over to get the generators started. [I presume they're using more refined battery technologies now, but this was thirty years ago.]
Lead acid batteries don't last forever, even when they float, so once a month someone was supposed to take a dummy load, a fan-cooled resistance element, and bridge it across each cell in turn for a few minutes and write down the voltage. This chore took about an hour or two.
One battery room was routinely "checked" by one particular worker -- adjacent to his lunch hour. Apparently he "checked" the batteries by writing down plausible benign voltages in the log book and taking an early lunch. After a while a supervisor noticed that one battery room was wearing out a cell every so often, but the other never seemed to need anything. He did the test himself and found out that all h*ll would have broken loose if there had been a power failure and the redundancy of having two battery rooms turned out to be needed because the other one wouldn't come online, say because some relay didn't do its job.
Naturally, the utility tried to fire the guy. The union hired expensive legal talent and succeeded in making it not happen. The question of why the union traded away raises the diligent employees could have had in return for contract clauses that made it possible to prevent an employee who was not doing his job and could have caused a fifteen minute glitch to turn into a nightmare for a major utility is left as an exercise for the reader.
-dk
Posted by: Dick King at Mar 4, 2007 11:10:21 PM
If providers of capital can group together to form companies, then its only fair that providers of labour be able to group together to form unions...
Posted by: Stephen at Mar 5, 2007 12:11:44 AM
The American debates about unions miss an important element: our law says that if a majority of workers want a union, then that union represents 100% of all workers. No worker is allowed to bargain separately with the employer. It's like if 51% of cookie buyers want Oreos, then none of us is allowed to purchase Chips Ahoy.
In practice, some workers want rigid work rules; others will accept some flexibility of job assignments for more cash; others would trade cash away for better benefits; others don't care so much about benefits, but want to be able to leave early on Wednesdays. Union contracts force workers into one-size-fits-all contracts.
In New Zealand, every worker has the right to select his/her own representative. Some join a union, others bargain for themselves, a few hire a representative to bargain for them. That system eliminates debates about "right to work" or card check vs. elections. It seems to me to be heavily rooted in the English right to contract.
Posted by: Bill Conerly at Mar 5, 2007 12:28:28 AM
A union within a laissez faire context is a cartel like any other. People within a free society should be free to form cartels. They should also be free to face the consequences of joining a cartel. The historical record in the United States shows that when businesses colude in order to fix prices somebody always crosses the line and lowers prices in order to gain advantage. Workers should be allowed the same privelage and owners should be allowed to hire whomever they want. Using the violence of the state to force owners to have to deal with union labor, and only union labor, is an acquiescence of special interest. Unions are only good for the members of the union. They are not good for the owners or the consumers who have to buy goods or services from a company that uses union labor. Bloviate all you like about the importance of unions. However, at the end of the day we are all consumers and labor prices are merely factors of final prices. I see unions as a form of cost shifting with negative effects on productivity.
Posted by: John Pertz at Mar 5, 2007 8:49:19 AM
John Pertz,
Should capital be allowed protection via incorporation?
Posted by: theCoach at Mar 5, 2007 9:34:15 AM
