With the information and communications technology (ICT) revolution in the 1990s, it has became widely recognized that General Purpose Technologies need complementary intangible investments in addition to traditional tangible investments. As Basu et al.(2003) and Brynjolfsson, Rock and Syverson (2018) pointed out, these complementary expenditures cause a bias in the Solow residuals.
While it has been established that traditional tangible investments are accompanied by adjustment costs, Brynjolfsson, Rock and Syverson (2018) regard the adjustment costs as complementary intangible investments which are not measured in GDP statistics. They take estimates of the total stock of research and development (R&D) capital and the total stock of capitalized selling, general, and administrative (SG&A) expenses using the listed items in the financial statements, and find that the adjustment cost of R&D is highest among the categories of capital.
Following their problem consciousness, we estimate the investment adjutment costs of Japanese listed firms. But two different points emerge from our estimate: we disaggregate the tangibles along with R&D; based on the interpretation change of the investment adjustment costs, we inspect how the characterics of adjustment costs change from before the ICT revolution of 1997 to after it. We find that investments of [a] Building & Structure, [d] Tools, Furniture & Fixture and [f] R&D are accompanied by their own adjustment costs, which we interpert as complimentary intangible investments, and that their adjustment costs are larger in order. While the fact that the adjustment cost of R&D investmet is estimated largest is consistent with the previous literature, it is notable that these estimated adjustment costs are reduced from the year 1997 and before to the year 1998 and afterwards. These reductions imply that the Japanese economy has been lagging in catching up to the ICT revolution due to the fact that complementary intangible investments were insufficient.