How Much Did The Liberty Shipbuilders Forget?

This paper produces new estimates of the rate of organizational forgetting in the well-known case study of US wartime ship production. I show that estimation is easily colored by problems of unobserved product heterogeneity and sensitivity to specification of the learning curve. Using data recently constructed from primary sources at the National Archives, I produce new estimates of organizational forgetting at the rate of no more than 4 percent, and possibly less than zero percent, per month. These are much smaller rates than previously reported. However, the paper also stresses the fact that our ability to obtain reliable estimates of rates of organizational forgetting is extremely limited.