On the 'utilisation controversy': a rejoinder and some comments

The conclusions of Gahn and Gonzalez (2019) are weak for the following reasons: (i) The Federal Reserve Board (FRB) measure of utilisation is not appropriate for measuring long-run variations of utilisation because of the method and the purpose of its construction. Even if its difference with the measures of the average workweek of capital (AWW) was trivial, this would still be the case; if anything, it would show that the AWW is also an inappropriate measure. (ii) Gahn and Gonzalez choose to ignore the longest available estimate of the AWW produced by Foss, which has a clear long-run trend. (iii) Their econometric results are not robust to more suitable specifications of the unit root tests. Under these specifications, the tests overwhelmingly fail to reject the unit root hypothesis. (iv) Other estimates of the AWW, which were not included in Nikiforos (2016) confirm these conclusions. (v) For the comparison between the AWW series and the Federal Reserve series, they construct variables that are not meaningful, because they subtract series in different units. When the comparison is done correctly, the results confirm that the difference between the AWW series and the FRB series has a unit root. (vi) Stationary utilisation rate is not consistent with any theory for the determination of capacity utilisation. Even if demand did not play a role, there is no reason to expect that all the other factors that determine utilisation would change in a fashion that would keep utilisation constant.