|Author Name||TOKUI Joji (Faculty Fellow, RIETI) / MIZUTA Takeshi (formerly Hitotsubashi University)|
|Creation Date/NO.||March 2022 22-J-014|
|Research Project||Refinement and Analysis of the Regional-Level Japan Industrial Productivity Database: Analysis of Regional Industrial Linkages and Productivity|
|Download / Links|
From the evidence of the R-JIP database we know that the main sectors driving recent productivity differentials among regions in Japan are not manufacturing but service sectors. At the same time, it is well known that in many service sectors locational convenience is crucial for productivity, and this characteristic of each region is supposed to be reflected in its land price. But land input as a factor of production is not usually accounted in the KLEMS-type productivity factorization. Thus, the calculated difference of total factor productivity in service sectors among regions may actually include the contribution of land input, which is expected to vary with the relative convenience of each location. In this paper we estimate land input of each industry in each prefecture in accordance with the R-JIP database 2021 and conduct a new analysis of productivity differentials among regions in Japan taking land input as well as labor and capital into account. To estimate the stock value of land we start from prefectural total value of commercial and industrial land, which is surveyed by the Ministry of Internal Affairs and Communications for the purpose of assessing real estate tax revenue of each municipality, and we correct the gap between market land price and government valuation for tax purposes. Then prefectural total value of land is multiplied by the industrial ratio of capital stock from the R-JIP database, and land to capital stock ratio from the Statistics of Corporations by Industry, and within manufacturing it is divided by land input from the Census of Manufacturers to get the value of land used by each industry. We then convert the estimated stock value of land into the value of land service input by applying the concept of user cost.