28th February 2011
Kevin Williamson doesn’t like unions much.
Nearly 90 percent of government employees in the United States are employed at the state and local level. A very large number of them, many millions, belong to public-sector unions. State and local bureaucrats are much more likely to be unionized than federal bureaucrats — more than twice as likely, in fact; 19 percent of federal workers are unionized, but 30 percent of state workers are, and 43 percent of local workers. These are very high levels of unionization across the board — only 8 percent of private-sector workers are union members — but much, much higher at the state and local level. That is significant because, contra Polman, McCartin, and the bulk of the Democratic commentariat, these unions do not influence public policy mainly through engaging in collective bargaining. They influence it by determining the outcome of elections.
Many union critics in the past few days have referenced Stanford professor Terry M. Moe’s fascinating paper “Political Control and the Power of the Agent,” published by the Journal of Law, Economics, and Organization in 2005, citing a single extraordinary fact: In the elections Professor Moe studied, union support was as valuable as incumbency in determining winners. That fact is, in and of itself, sobering: Incumbency is generally the most powerful factor in elections — short of a major scandal or similar political catastrophe, incumbents most often are relatively secure in their reelections. The fact that union support turns out to be not only as powerful a factor but, in fact, a slightly more powerful factor in the most significant contests demands a reevaluation of our fundamental thinking about who is really in charge of our state and local governments.