The Number of Large Strikes in the US is at the Highest Level in Decades

Halfway through the year, the number of large strikes in the United States is at the highest level in decades.

n the 1980s there was a phrase used to describe the increasing unwillingness of US labor unions to go on strike. In the disastrous 1981 air traffic controllers’ strike, over ten thousand members of the Professional Air Traffic Controllers Association (PATCO) were fired by President Reagan. That was seen as the beginning of a rougher era of union-busting and increasingly adversarial labor relations, and large strikes in the US fell in the following years. The phrase “PATCO Syndrome” came to describe the fear of striking and losing.

In reality, US labor relations have never been cordial and large strikes have been declining since the 1970s. The Bureau of Labor Statistics (BLS) releases a report early every year of the number of large work stoppages the previous year. BLS defines large work stoppages as “involving 1,000 or more workers lasting one full shift or longer.” In counting work stoppages, it includes strikes and lockouts (where the employer refuses to let the union members work), but the list is usually mostly strikes. In 2017, BLS released this stunning chart summarizing seventy years of collecting this data.

Clearly, the number of large strikes has plummeted. For the thirty-year period starting in the mid-1940s, the average number of annual strikes was over three hundred per year. That means on almost every day in that era a large strike would start. The average in the most recent decade has been about fourteen per year. Continuing this trend, the  BLS’s most recent report listed only seven large strikes for 2017, the second lowest on record after 2009.

There is another record of strikes maintained by the Federal Mediation and Conciliation Service (FMCS). It tracks work stoppages of all sizes related to contract bargaining in the private sector, though it also seems to include some public sector strikes as well. A full analysis of this data needs to be done, and it’s not clear to me at this point how complete it is.