Strike Lengths: Correcting for Prestrike Announcements and the Ratio of Bargaining Size to Firm Size

Anticipating the duration of a labor strike can be vital for both sides of the dispute, as well as outside observers. The
methods of a pair of studies using Canadian data are surveyed to analyze labor strikes in the United States from 1992 to 2008.
Corrections are made for strikes with predetermined lengths (“one-day” strikes and the like), whose durations are more a function
of the prior announcements than of other factors, such as number of employees striking and macroeconomic conditions. Strikes
are found to be generally shorter when the striking unit represents a larger portion of the firm’s total workers, a proxy
for its bargaining power. This ratio provides a better understanding of the strike dynamics (including expected length) than
do sheer bargaining unit size or sheer firm size.