宮川 努

Using the updated datasets on social infrastructure and regional productivity, we reexamine the contributions of social infrastructure on productivity improvements. Our study adds two new ideas to the previous studies on social infrastructure: 1) we consider the complementary role of intangibles to externalities of social infrastructure, and 2) we include a new measure -the spillover effects considering transaction volumes between two prefectures -—in our analysis. To conduct our study, we use three datasets: 1) the updated dataset of social infrastructure published by the Cabinet Office, Government of Japan, 2) the 2021 version of the Regional Japan Industrial Productivity Database, which measures intangibles at the prefectural level, and 3) the inter-regional Input-Output Tables, which are used for the measurement of our unique spillover effects. Using these datasets, we find that the ratio of intangibles to tangibles in the public sector in rural areas has not increased compared to those in the private sector and the public sector in the urban areas. In the estimation results based on the production function including social infrastructure and public intangibles, we do not find positive and significant externalities of social infrastructure. However, the coordination of social infrastructure with intangibles by industry contributes to the productivity improvements. In addition, we find positive and significant effects of social infrastructure in rural areas. We also find positive and significant contributions of social infrastructure for restorations from large earthquakes such as Great East-Japan Earthquake and Kumamoto Earthquake. Our results on intangibles and social infrastructure are consistent with our experiences during the COVID-19 pandemic, when the responses of the government to this pandemic were slow due to the lack of digitalization. Our study shows that we need to accumulate intangibles not only in the private sector but also in the public sector.

コロナ禍を経た日本で、デジタル化の遅れを取り戻す動きが盛んになっている。しかし、統計データの制約から IT 投資とそれに伴う人材育成や組織改革を組み合わせた包括的なデジタル化に関する定量的な分析は少ない。一方で従来型の投資ではないクラウド・サービスや生成AIなどの新たな情報サービスの利用形態も現れている。そこで本稿では、統計データの制約の下で、生産に寄与する IT 資産及び IT サービスの計測に関する新たなアプローチを試みる。またデジタル化を政府が推進する背景には、スピルオーヴァー効果の存在があるが、本稿では公的部門と情報サービス産業からのスピルオーヴァー効果に焦点をあてた推計を行う。推計の結果、これらの産業からのスピルオーヴァー効果が確認されたことから、もしスピルオーヴァー効果を重視するのであれば、政府の支援は情報サービス産業を中心にした方が望ましいと考えられる。

新たに作成された地域別・産業別生産性データベースと社会資本のデータを組み合わせて，社会資本の生産力効果を再検証した．我々の推計では，バブル崩壊後に全産業にわたって生産力効果が見られる．これはバブル崩壊後財政が悪化し，社会資本投資が抑制的に運営される中で，効率的な投資配分が行われてきたことを示唆している．推計結果から計測される社会資本の収益率は，総じて民間資本より高いが，これは非製造業において顕著である．また民間資本の収益率は地域差がほとんどないが，社会資本の収益率やトービンのＱは地域差が大きく，都市部以外の地域に社会資本が重点的に配分されたことを示している．This paper reexamines the effects of public capital on productivity growth at the regional level by using the Regional Japan Industrial Productivity Database. When we estimate a production function including public capital in TFP, we find positive and significant effects of public capital on productivity growth after the collapse of the bubble economy. Our results imply that public investment after 1991 was allocated more efficiently than before 1990, because total public expenditures have decreased since the mid 1990s. However, the dispersion of the estimated rate of return on public capital and its Tobin's Q has expanded since 1991. This implies that the government has allocated public capital to rural areas more heavily than to urban areas.