Some of these laws addressed real problems — until the Norris–La Guardia Act, for instance, courts often issued injunctions to end strikes, essentially forcing people to work. The NLRA, however, was a mistake from top to bottom.
As it was originally enacted, the NLRA defined several “unfair labor practices” for businesses, but none for unions. (In 1947, the Taft-Hartley Act added some for unions.) Short of closing a plant entirely, management could no longer discourage workers — using such tools as hiring, firing, discipline, and promotion — from participating in union activities. The only significant exception, confirmed in a 1938 Supreme Court case, was that an employer could hire replacement workers during a strike, and was under no obligation to fire the replacements to make room for returning strikers. The strikers were still considered company employees, however, and had to be reinstated when positions opened up.
Further, the act set up the system through which unions are recognized: First, they have to get 30 percent of workers in the “bargaining unit” (typically, the workers at a given plant) to sign cards. Then, the NLRB supervises a secret-ballot election, and if the union wins, it has the right to represent the workers. (Alternatively, the union can get signed cards from 50 percent of workers and avoid an election, but only if the company voluntarily recognizes the union.) The company is required to negotiate in good faith with its union on a contract; if negotiations falter, the union may strike.
The logic here is simple and straightforward: We want workers to be paid more and treated better; therefore, we’ll arm workers with the weapons they need to gain concessions from employers. Unfortunately, it wasn’t until the 1950s that Milton Friedman proved the economic fallacy of this plan. Yes, unions often manage to get higher pay and better working conditions for their members. But in response, unionized businesses hire fewer workers. The workers who aren’t hired by union companies go to non-union companies — where their competition drives down wages — or remain unemployed. In other words, in the private sector at least, the gains of unionized workers aren’t gains for the working class as a whole; they’re gains by some workers at the expense of others. (In the public sector, which is not covered by the NLRA, higher wages simply come from taxpayers.)
And economics aside, the NLRA system involves a tremendous amount of coercion. Companies have to negotiate with unions — with the threat of a federally protected strike in the background — whenever their employees vote to make them. Businesses may not fire workers for union activities, including striking, regardless of what the relevant contracts say. And while the current case against Boeing involves a fresh issue — whether companies may factor in local labor laws and past strikes when deciding where to build new capacity — the NLRA has long been interpreted to mean that a business may not shut down a unionized plant and reopen it somewhere else (a “runaway shop”) in response to strikes or union demands. If a union loses an election, the NLRB may decide the employer engaged in “unfair labor practices” and force the company to bargain with the union anyway.