The incidence of unions

To some extent higher union wages translate into higher prices for consumer goods.  Over a five year time horizon I’ll guess at 50 percent pass through, adding that most of these goods are bought by other laborers.  Just to be flippant, for each dollar gained by a union member, I’ll guess that labor market "outsiders" lose 50 cents.

Notice we haven’t even counted negative effects on the rate of future economic growth, or for that matter costs to employers.

We already don’t have workers, viewed as a class, coming out ahead.

I would be curious to hear the numbers assumed by those who wish to encourage labor markets by law.  I would be curious to hear how much they think, over say a ten-year time horizon, wages deviate from labor productivity.

Inquiring minds wish to know.


Since we're in the land of more or less making things up here rather than using real statistics or measures I wonder what you'd say to Kevin Drum's point that the biggest expansion in US history was also the time with very high union membership. Apparently it's just an article of faith to you that unions are bad (so it's okay to just make stuff up) but how do you explain Drum's point? I suppose you might try a faith-based approach and swear that the economic expansion of the 50's and 60's would have been even bigger w/o unions but that would really be off in the land of religion.

Matt said:

"Since we're in the land of more or less making things up here rather than using real statistics or measures"

Real stats or not, last time I checked labor prices are still a component of final prices. If a firm is forced to raise wages because they are forced to deal with the union then they are going to have to raise prices. Not only are workers producers but they are also consumers. As a result, a nominal wage increase due to union bargaining can be eaten away at due to increase in the price level. What seems so made up or faulty with this logic? If every hamburger seller in a market faces the same increase in ground beef prices, then everyone will simply raise prices. If all companies face high labor prices due to union activity then they will all simply raise their prices.

Ah, but you are only considering nominal prices. If unions can raise the real price of labor relative to capital, they will encourage substitution away from labor towards capital. This should raise total costs by less than the increase in labor costs. Then, depending on how competitive the market is, the price charged by the producer will either remain static (in the competitive case) or increase by less than the increase in labor costs.

very good point Tom. BTW, anyone have a figure on what percentage of government employees are unionized?

I suppose this means that with 100% unionization, you will have 50% increase in net gross income for workers (presumably redistributed from profits). So tell me, why are workers getting a bad deal here?

So raising wages of workers is always bad? I don't see anything that makes an union negotiated wage increase any different from any other wage increase.

By your logic, any wage increase just impoverishes everyone else in the country. Therefore for the good of the people, workers should be paid nothing and the savings should be passed on to the executivess who deserve more pay for all their productivity increasing innovations.

Thank god, the median wage hasn't increased in this country in the past 30 years. Just think of the dystopia that might result if the majority of Americans saw any of the benefits of economic growth.

Higher union wages and dampened growth are the price society has to pay for keeping business owners' greed in check.

I have no empirical evidence for this, but I think that despite the dwindling rate of unionized workers, the specter of unionization keeps companies from treating their workers poorly.

†with 100% unionization, you will have 50% increase in net gross income for workers (presumably redistributed from profits). So tell me, why are workers getting a bad deal here?†

The labour share of net national income (exkl. depreciation of capital) is around 75% in the US. (By the way, this is the same or slightly higher than Scandinavia with 90% collective agreement share.)

When you have a 50% increase of that will give labour a 113% share of US GDP.

Sounds like a good deal for workers! Free Lunches for everyone!

The liberaltarians impressive us more and more every day.

Interesting problem.

Let's give instant U.S. citizenship and a plane ticket here to anyone who gradutes with an econ degree at any university in the world and see what happens to American economists' wages.

I think, judging from the marching morons commenting here, that somebody told their union members that unions are under attack.

Unions are for the unions, not for societal benefits at large.

By the way, Spencer, the whole point behind a union is to prevent an employer from controlling costs by shedding workers. That is why a gov't minimum wage is not like a union minimum wage.

Gee -- I always though the purpose of a union was to maximize the economic wellbeing of its members.

It is amazing the things you learn from libertarian that no one else in the world knows.

It is true. Higher labor costs mean higher product costs. The same goes for material.

Assuming that labor is a product subject to the supply and demand curve, the question is whether having a union artificially restricts demand to raise prices or whether the union just equalizes the market forces, minimizing the capitalists' ability to manipulate the market and depress prices.

Compare a labor market with no unions to a stock market with no regulations. In that case, company insiders (like the capitalists) would have an advantage over investors (like the workers) due to insider information. Regulation equalizes the market. Yes, there is a cost. Of course.

On the other hand, if we had comprehensive worker protections written into law and regulation, we would not need unions as much, if at all. But, that, too comes at a cost.

The question is not whether there is a cost, but who should bear the cost and whether the absence of that cost is morally justified.

Finally. When there is a vibrant middle class, the economy benefits. The middle class buys the stuff the capitalists make. Unions help ensure a vibrant middle class. Therefore, unions are a net benefit. Professor Cowen has not addressed this concept.

I cannot say what the deviations in wages verses productivity would be over a longer time span. Labor unions do best when there is a fairly inelastic demand curve for the labor which the labor union provides. In other words, will the demand for the good or service which the labor union provides labor for be around tomorrow?

It is for these reasons that unions usually oppose things like free trade agreements, the free flow of capital, competitive pressure from non-union labor, and usually cluster around parts of the economy like automobile manufacture, airlines, and government bureaucracies. Unions also tend to like barriers to entry to their respective job fields. If unions can achieve their aims, usually through the political process, then they can reap gains for their members.

I would imagine that the numbers for deviation between productivity and wages would be correlated with the factors which I have pointed out above. I would especially like to see what the situation is like with workers in government bureaucracies.

Comments for this post are closed