Arnold Kling  

Unionization's Decline, a Human Capital Story

PRINT
The Forgotten Men... Oil Economics...

Jane Galt asks,


Why hasn't labour successfully colonised the non-manufacturing world, outside of the public sector?

I think that the answer might start with Gary Becker's distinction between specific human capital and generic human capital. Specific human capital is capital that is built up in a particular company, so that you cannot take it with you. Generic human capital is general knowledge, which is portable. Or, as Michael Piore put it when he taught my graduate school course in labor economics, "Specific human capital means knowing where the men's room is located. Generic human capital means being able to read the sign on the door."

In manufacturing, workers develop specific human capital. As someone who actually worked in a factory for a couple of summers, I can attest to this. You learn to operate the particular machinery in the plant, but that knowledge is of no value in a different plant.

In the service sector, skills are often transferable. You may have a license (to be a teacher, a nurse, or what have you) that makes you transferable. Or you may have a skill set (sales, general management, computer programming) that is transferable.

With specific human capital, there is mutual bargaining power. The company values your experience, but your opportunity cost is low, so they could try to keep your pay low and exploit you. So a union helps you out.

With generic human capital, you do not need bargaining muscle. If you are way underpaid, you simply take another job. So a union helps less.

So, my explanation for declining unionism is that it is part of a larger trend, which is toward increased importance of generic human capital and reduced relative importance of specific human capital. That trend in turn reflects changes in technology and the work force.

Along these lines, Tyler Cowen asks why unions are so strong in Hollywood. I think that this may be another instance where people have low opportunity cost relative to their marginal product. The proverbial part-time actor, part-time waiter.

For Discussion. What does the economics literature say about the reasons for the decline of unionism? Has a similar decline taken place in other countries as well?


Comments and Sharing


CATEGORIES: Labor Market



TRACKBACKS (1 to date)
TrackBack URL: http://econlog.econlib.org/mt/mt-tb.cgi/190
The author at Tim Worstall in a related article titled Carnival of the Capitalists writes:
    COTC is up for this week at Ashish’s Niti. All the regular stars are there, two of my favourites are David Tufte dealing with the ’flu vaccine shortage then over supply (he starts with [Tracked on January 31, 2005 1:11 PM]
COMMENTS (6 to date)
Bill Stepp writes:

As a former factory worker during a couple summer vacations, I must dissent from the view that the ability to operate a machine is a nontransferable skill. Surely a lathe operator or forklift driver can do the same job in more than one company. Skilled machine operators move from factory to factory all the time, frequently making lateral job changes involving the same skill sets, although sometimes moving into management and I would guess even sales on occasion.
Also, this idea won't win me any Brownie points, at least with Chicagoites, but I don't believe in the concept of "human capital." It's part of the Chicago conspiracy, going back to Knight, to turn the trinity of productive factors--land, labor, and capital-- into one catchall factor called capital.
Ever wonder why you don't find "human capital" in the works of Mises, Hayek, and Rothbard? It's beacuse they saw analytical distinctions between the factors (which I think are brought out most clearly in Rothbard's "Man, Economy, and State"), which make their trifurcation useful and even necessary.
For example, workers (labor) are paid wages, which aren't capitalized, unlike the rents earned by owners of capital goods, which are. Workers aren't owned by anyone, save for slaves, who literally are human capital. Capital goods can be bought and sold, unlike workers in a system of free labor.
What about a professional athlete (or other skilled worker), who signs a long-term contract? Can't his contract be bought out, and so doesn't this make the return from his work a return on capital, not a wage?
No, he's still a worker, just like a truck driver. It's just that he's far more productive than a driver, so his discounted marginal value product is much higher.
What about the owner(s) of a sports franchise, or rather, what about the value of the franchise? That's determined just like the value of a stock--by its free cash flow discounted to the value of the firm.

Randy writes:

It seems to me that labor unions fail for the same reason that all socialist programs fail - free market forces eventually overwhelm them. The reason such programs often appear to be successful is that it takes time for market forces to adapt. It can take years for even small programs like union locals - and generations for large programs like social security - but the free market will eventually adapt. Socialism only works within a free market, and then only temporarily.

Lawrance George Lux writes:

The general Worldwide trend remains the decline of Unions. Arnold's comment on generic human capital is valid, but there are other contributory reasons: the transference of power to Corporate structure through aggregation of financial assets, the actual differentiation of human skills, and finally the greatest; the inability of Unions to negotiate with the immediacy and specialized skill as can individual labor and Management. lgl

In my blog, OpinionMeister, I link to this article and comment on it. My comments were as follows:

The article does not get into the question of why unions have been so successful among local, state and federal government workers, but I think this same analysis fits. How many skills that a government civil-service employee picks up at work could possibly be transferable to the private sector.

superdestroyer writes:

Unions have expanded at the government level because all governments are basically right to work states. A federal employee does not need to join a union to work for the federal government. Also, public employee unions are much more limited in what they can negotiate. Wages and benefits are set indepenent of unions most of the time. Public unions negotiate on items such as creating barriers to entry or working conditions. Thus, government workers have less reason to oppose unionization and governments have fewer economic reasons to oppose unionization.

David Foster writes:

This analysis is just plain wrong. As Bill points out, plenty of working-level factory jobs are highly transferrable: in fact, one of the primary characteristics of mass production is the ability to train people for assembly jobs in a matter of a couple of weeks. You say you learn to operate the machinery in a particular plant but that knowledge is of no value in another plan: this is strange since totally custom machinery is rare: everybody buys their machine tools, robots, etc from the same companies.

Comments for this entry have been closed
Return to top