The Unintended Consequences of Collective Bargaining


Read the Full Analysis Here

This study analyzes the effect of unionization on economic growth on a state-by-state basis, and calculates the “deadweight loss” resulting from unionization. By raising the cost of labor, unions decrease the number of job opportunities in unionized industries. That, in turn, increases the supply of labor in the  nonunion sector, thereby driving down wages in those industries. The effect of this situation is to increase the natural rate of unemployment, thus imposing a deadweight loss of economic output on the economy.

Deadweight loss in this context means that unionization, by artificially increasing the price of a factor of production—labor—above the price that would be established in a free and competitive marketplace, comes at the cost of retarding economic output that would occur absent that artificial constraint on a free labor  market.

This assessment does not suggest that, in an ideal world, workers should be paid increasingly less to ensure further economic growth. Rather, increases in productivity—not artificial increases in labor prices—are the key to economic growth. 

The presence of deadweight losses arising from labor union activity can be shown in a formulation devised by labor economist Albert Rees (1953, 1963). Rees demonstrated the consequences of union wage-raising initiatives on levels of employment in both the union and nonunion sectors of the labor force.

The Rees formulation can be used to calculate the numerical value of deadweight losses from unionization if union density (the percentage of employees who are unionized), wage premiums associated with the presence of unions, and general elasticity of demand for labor are known. The elasticity of demand for labor measures how much the quantity of labor demanded by employers changes, given a change in the price of labor. Work done by Richard Vedder and Lowell Gallaway (1997) provides us the latest, best assessment of the elasticity of demand for labor.

Using this and other estimates, this study calculates the deadweight losses described by Rees as being associated with the presence of labor  unions for six different and select years during the period 1967 through 2000. On average, the results show a deadweight loss in workers’ wages of slightly less than a third of a percentage point.

Over a period of 50 years, the cumulative reduction in worker wages would be about 15 percent. Because wage payments are only a fraction, albeit a large one, of gross domestic product (GDP), the deadweight losses from unionization are a smaller fraction of that magnitude. However, over a long period, those small annual effects produce a substantial cumulative  loss of GDP—as much as a 10 to 12 percentage point loss over a half century.

It is worth noting that these figures are minimal estimates of the deadweight losses produced by labor unions. Rees’s analysis assumes a perfectly inelastic supply curve for labor, and elasticity could easily double the deadweight losses produced by unionization in America.

Deadweight loss contributes to interstate income differentials. To explore the extent of this phenomenon, the analysis defines a statistical model to explain the growth in real per capita income (RPCI) in states. The unionization rates and an additional five independent variables—manufacturing, income tax rates, RPCI in 1964, politics,  and college education equivalency—are included in the model to account for additional factors that are likely to affect the growth in income.

Most important for purposes of this report is the statistical significance (at the 5 percent level) of the regression coefficient for the average percent unionization variable. This measure indicates that every additional percentage point of average unionization in this time period reduced
the growth in RPCI by 1.73 percentage points. 

Knowing this relationship permits the calculation of the estimated effect of union-related deadweight losses on the growth in RPCI in each of the several states.

Two broad conclusions emerge from this document. First, the presence of labor unions that operate as bargaining agents in the process of collective bargaining has the potential to seriously inhibit economic growth in the several states and the District of Columbia. This conclusion suggests that the decision to officially encourage collective bargaining through public policy, which was the primary thrust of the National Labor Relations Act of 1935 (the Wagner Act), was rife with unintended negative consequences. 

The disparity in the relative incidence of unionization of the workforce in the United States leads to our second broad conclusion—that certain states, such as Michigan (which enacted a right to work law only in 2012), have suffered large amounts of foregone economic growth, while others, such as South  Carolina (which has had a right to work law for a long time), have been affected to a far lesser degree.

Those conclusions provide a strong case for viewing the passage of the Wagner Act in 1935 as a case of causing long-term economic trauma. However, state policy makers can mitigate some of the most damaging aspects of the Wagner Act by passing right to work laws.

Note: The study analyzes 1964-2011. Idaho, Oklahoma, Michigan and Indiana enacted right-to-work laws in 1986, 2001, 2012 and 2012, respectively.