All Commentary
Wednesday, March 2, 2011

Guest blogging: Unions

The recent conflicts in Wisconsin and other states over unions have a generated much heat but little light. That is because most discussion asks the wrong question and ignores the economics of unions.

The right question is not whether unions are good or bad. A union is a voluntary association of employees that attempts to increase the compensation of its members by bargaining collectively with an employer. No one should object to such an association, nor to its desire to improve its members’ welfare. Unions per se are neither good nor bad.

The right question, instead, is whether policy should attempt to give unions more power than they would have under laissez faire. To answer that question, consider the economic effects of such policies.

Without help from policy, unions might still form and attempt to bargain collectively, and some employers might agree because they find collective bargaining easier than negotiating compensation with every employee. The ability of these “free market” unions to increase wages, however, would be modest: if their demands implied compensation much above the individually negotiated levels, employers would just refuse to bargain with unions.

But this outcome – in which unions have little effect on wages and benefits – is exactly what economic efficiency demands. Absent policy protection, employers will hire additional labor when the value of the extra output this labor can produce exceeds the compensation the labor would require to voluntarily supply this effort. Any policy that raise labor’s cost above this level means that employers will hire fewer people and produce an inefficiently low level of output, or produce in an inefficient manner by substituting relatively expensive capital for what could be relatively cheap labor.

Given this economic reality, why does anyone support policies that protect unions?

Presumably, to redistribute income to union members. Protections for unions indeed have this effect, but this kind of redistribution makes no sense.

Union protections do not help the truly poor; they raise the wage of those who would earn a moderate wage without union protections. Indeed, by encouraging firms to hire fewer people, unions mean increased income for some but a zero income for others, thus exacerbating poverty. In extreme cases, union protections drive businesses overseas in search of cheaper labor, meaning large reductions in employment.

Union protections also raise the cost of goods and services in unionized sectors by increasing the cost of production. This means that everyone, including the poor, face higher prices for cars, houses, transportation, and so on. In the public arena, union protections mean higher costs of government services, which adversely affects everyone but especially the poor.

The bottom line is that government policy should not protect unions in any way. This means no requirements for employers to bargain collectively, no rules that force all employees to be covered by union contracts, and no impediments to hiring non-union labor. Similarly, it means no restrictions on firing strikers, or closing plants, or hiring replacement workers.

Jeffrey Miron is Senior Lecturer and Director of Undergraduate Studies at Harvard University and Senior Fellow at the Cato Institute. Miron blogs at and is the author of Libertarianism, from A to Z.

  • Jeffrey Miron is a Senior Lecturer and Director of Undergraduate Studies in the Department of Economics at Harvard University, as well as a senior fellow at the Cato Institute.