生産性ネタをもう一丁。以下はDiego A. Comin(ダートマス大)、Javier Quintana Gonzalez(スペイン銀)、Tom G. Schmitz(ボッコーニ大)、Antonella Trigari(同)による表題のNBER論文(原題は「Measuring TFP: The Role of Profits, Adjustment Costs, and Capacity Utilization」、ungated版が掲載されている著者の一人のHP)の要旨。

Standard methods for estimating total factor productivity (TFP) growth assume that economic profits are zero and adjustment costs are negligible. Moreover, following the seminal contribution of Basu, Fernald and Kimball (2006), they use changes in hours per worker as a proxy for unobserved changes in capacity utilization. In this paper, we propose a new estimation method that accounts for non-zero profits, structurally estimates adjustment costs, and relies on a utilization proxy from firm surveys. We then compute industry-level and aggregate TFP growth rates for the United States and five European countries, for the period 1995-2016. In the United States, our results suggest that the recent slowdown of TFP growth was more gradual than previously thought. In Europe, we find that TFP was essentially flat during the Great Recession, while standard methods suggest a substantial decrease. These differences are driven by profits in the United States, and by profits and our new utilization proxy in Europe.


Positive profits create a wedge between output elasticities and factor shares. We use industry-level profit shares estimated by Gutierrez (2018) to compute this wedge, and find that in most countries and industries, the zero-profit assumption underestimates the output elasticity of labour and materials, and overestimates the output elasticity of capital. This is important, as capital tends to grow faster than other inputs in the long run, and is less volatile over the business cycle. Thus, the zero-profit assumption underestimates TFP growth in the long run, and overestimates its volatility and cyclicality.
The second new element in our paper regards adjustment costs. Adjustment costs for capital and employment are important in many business cycle models, and constitute the leading explanation for why firms change their level of capacity utilization over time. Furthermore, adjustment costs matter for TFP measurement, as they change the effective growth rate of capital or labour inputs in periods with large changes in investment or hiring (e.g., during the recovery from a deep recession, or in the early years of a new industry). Nevertheless, Solow and BFK assume that adjustment costs are either inexistent or negligible.


*2:cf. これ