Although philisophically I agree with the goal of improving working conditions and wages, my personal experiences with unions have been mostly negative. As a result, my stance towards unionization is more nuanced than simple “pro” or “con”.
The one-sentance executive summary of my position is: “Unions are a net benefit when labor is a commodity, and a net cost when labor is a specialty.”
To make sense of this, I need to examine what is a union, and what is a commodity.
- A commodity is a kind of good in which all units are more or less interchangeable. Copper is a commodity – there’s no fundamental difference between copper mined in Australia and copper mined in the US.
- A union is most fundamentally a cartel of labor. That is, it is an agreement by workers not to undercut each other on the price of their labor.
I use the term “commodity labor” to mean jobs in which workers are more or less interchangeable. This may mean that the work is simple enough that anyone can do it – like for example, washing dishes. Or it may mean that the techniques involved are highly standardized and legislated, so that there is little difference between workers. An example of this would be electricians and plumbers.
Specialty labor, on the other hand, is where each worker has a unique skill set that is not easily replaced. This is true in almost any creative field; The one I am most familiar with is software engineering.
For commodity laborers, unions provide an important benefit of maintaining a decent wage. Normally, in an unrestricted free market, the suppliers of a commodity will underbid each other until the price of the commodity drops to just above its “replacement cost”, i.e. the cost to actually produce the commodity. For labor, the replacement cost is called “starvation wages” – the lowest possible that you can pay someone and still keep them alive and producing children. Before the development of unions, such working conditions were common.
Unions are at their most powerful when they have a means to prevent independent workers from undercutting the union price, when they in effect have a monopoly on labor. This is a classic prisoner’s dilemma – an unemployed individual might gain a temporary advantage by agreeing to reduced wages, but he is better off if no one else does the same.
In a specialty labor market, things are quite different. For one thing, workers aren’t competing with each other directly on price, since workers aren’t as interchangeable. If I need someone to work on Windows device drivers, I not only need to find someone who knows about that specific application domain, I need to decide how much I want to pay vs. the skill level I am likely to get – that is, I can get someone really good and pay a lot for it, or I can get someone not quite as good and pay less.
A programmer who is highly skilled at these tasks has a great deal of bargaining power and can command high wages. They have no need for a cartel to increase their bargaining power, and in fact such a cartel would merely act as a leveler. Even a mediocre engineer with a poor track record, who doesn’t have quite as much bargaining power, still has a great deal, and they can imagine that someday they might have more. As a result, they are unlikely to want to enter into a system where wages are negotiated collectively rather than individually.
Specialty labor markets are also where the negative effects of unions are most apparent, especially when the union is in a monopoly position. Unions can be a drag on innovation and creativity in a number of important ways, such as by requiring that promotions be based on seniority rather than pure merit. They also create barriers to entry (such as the exorbitant fees required for joining some unions) which might drive away impoverished but enterprising young workers.
A young, inexperienced worker with new ideas who is not weighed down by preconceived notions might fare better in a non-union environment, especially if they are in an industry in which individual creativity and enterprise are well rewarded.