The argument to be made that increasing workers' pay increase the strength of unions is one that works mostly at the lower ends of the pay scale. It's a lot of work to be poor. When you're poor your time is wasted in dozens of ways during the day that you don't have to put up with when you're middle class or wealthy.

By giving workers more money, you give them more time that can be used to think about and act on union campaigns. It doesn't guarantee that the extra time will be used that way, but it may be.

I would argue that shouldn't be a disincentive to pay workers more, but then, I support many unions.

I haven't been answering because I actually am not sure what the theory is, but I think it is this.

Union workers tend to be higher-paid, and somewhat higher-skilled. You do get something for your extra money. So if you're required to pay higher wages, this encourages you to substitute toward union labor, so that you can at least get what you pay for. This means more jobs for union workers and consequently more power for the union leaders who get them the jobs.

Will - I can back up the better skills theory. As someone who has hired both union and non-union labor in the past, I can tell you I will always favor union electricians and wiring people.

You pay more per hour, but the quality of the work is far superior and they spend much less time on a job. Over the long term, I'd argue you pay less over the long haul using the better quality union labor since you don't have to fix as many problems later.

I don't know if you're comparing apples to apples, though, Barrett. While in general union labor might be more skilled than non-union labor, in my (admittedly limited) experience, properly licensed/certified non-union labor is at least as good in quality, while still usually cheaper than union labor.

Well, Sweth's claim aside,I think Barrett's observation backs up the idea. If you're forced to pay higher wages than you otherwise would, you try to extract as much value out of those wage payments as possible. A corollary of this is why high minimum wages tend to hurt very low-skilled workers. I might prefer to hire the low-skill worker at 4 or 5 dollars an hour, but if I have to pay 8, I will try to find somebody who is worth 8.

I will admit that the union guys we've used for wiring have usually been old timers from the AT&T/Ma Bell days who formed their own companies. My positive experiences may have more to do with the way the old AT&T trained its staff than the unonization of the labor.

This is a really strange line of argument, Will. Basically you're saying that the union workers, getting paid at rate x, shouldn't get paid more because it increases their political power. But then when their pay is increased to x+1, you say that their political power has increased because you are now paying them a rate that's appropriate for the services they are providing. That in itself seems kind of paradoxical, or at least sick.

But I don't even buy the premise that these people are getting more work because the prevailing market wage has increased. It might be in the case in the short run, when there isn't anybody else trained to do the job, but in the long run there will be entry at a lower wage rate, and the union will either have to return to their previous pay rate or they will lose market share to the competition.

If it's really true that unions really get more powerful when there are higher pay rates, I think the explanation has to be that they are more capable and more motivated to prevent people from defecting/entering. It's not so much that the money itself gets used for political purposes (or out-and-out intimidation of the competition), although I'm sure it does. But the higher wage rate relative to what would be the prevailing market rate means there's more to protect, so more gets spent to protect it.

I think the answer ultimately has to be that these folks are using a portion of those marginal dollars to protect that raise. I don't imagine the difference is that big, though, so I'm left wondering just how efficient unions are.

I thought Davis-B required basically a price floor for workers, regardless of whether they were union or not. Is that wrong? If that is what it does, then the idea is that it encourages governments to hire union labor rather non-union for all of the normal microeconomic reasons: setting a price floor encourages buyers to substitute toward high-quality versions of the product, which they wouldn't normally buy.

If Davis-B does something else, then ignore all of the previous comments.

If the union is really able to provide the level of quality they're providing at that lower wage rate, then raising the floor should mean someone else can provide a higher level of quality because there is more margin to work with. That's why you should get non-union entry in the long run. Your premise seems to be that union work is intrinsically better -- that's what I don't buy.

Ah. Alternatively one wouldn't even have to believe that union work was better, just that there are other political reasons that one would prefer to hire union work rather than non-union work if the wages are the same. (Namely the pressure unions can bring to bear on politicans who don't prefer unions.) The same effects would play out.