Pure technology gaps and production predictability

• Productivity lag between average and frontier machine is an aggregate technology gap.
• Aggregate technology gaps are decomposed into trend and mean-reverting component.
• Pure technology gaps are the mean-reverting component.
• Vector error correction model is used to verify their predictive power for production.

An average machine lags in terms of productivity and technological advancement behind a cutting-edge machine. This lag was first defined by Cummins and Violante (2002) as the technology gap. Using the vector error correction model, I show that the technology gap is cointegrated with human capital factors, and then decompose it into a long-run trend and a transitory mean-reverting component, which I term as the pure technology gap. I show that the pure technology gap has a predictive power for the aggregate production. Intuitively, a high pure technology gap acts as an economic shock that increases production in the long term due to a higher future productivity level.