Published papers

This chapter measures the industry origins of the sustained energy productivity improvement (EPI) in postwar Japan from 1955 to 2019 by explicitly considering changes in energy quality, such as the progress of electrification, and changes in industrial structure. The golden age of energy productivity according to its gross measure, which is a simple economy-wide indicator, was seen in Period-II (post-oil crisis: 1973–1990) when the annual growth rate of EPI accelerated to 2.7% from 1.3% in the preceding Period-I (high economic growth: 1955–1973). However, the adjusted EPI peaked in Period-I when energy quality and industrial structure changes are controlled (1.5%). It then decelerated to 1.4% in Period-II and almost zero in Period-III (post-bubble economy: 1990–2008). In Period-IV (post-global financial crisis: 2008–2019), the adjusted EPI recovered to 0.5%, largely because of changes in composition at the product level. Under certain assumptions, the baseline estimate of future gross EPI is expected to be 1.0% per annum, much lower than the policy target of 2.5% set by the Japanese government in 2021.

In the post-war economy of Japan, the high-pressure economy as defined by Okun and Yellen maintained its real exchange rate relatively low and kept the economy mildly inflationary. Not only did it encourage short-term employment but it also fostered a long-term productivity trend as measured by its total factor productivity (TFP). International pressure for a stronger yen after the Plaza Accord, coupled with the restrictive monetary policy of the Bank of Japan, created a large exchange rate gap and accordingly, the extreme Japan-U.S. output price gap, referred to as the price level index (PLI), in 1995. It further caused the squeeze of domestic wages and led to the deflationary pressure that plagued the Japanese economy for decades. In our opinion, the exchange rate also depressed domestic capital accumulation caused by the internationally low nominal rate of return on capital and that situation caused the TFP stagnation in Japan.

The paper defines and computes alternative measures of output, input and Total Factor Productivity (TFP) for the production sector of the Chinese economy. The current System of National Accounts (SNA) input measure that corresponds to Gross Domestic Product (GDP) does not measure the income generated by the production sector since it includes depreciation and excludes capital gains and losses on assets used in the production sector. The paper suggests an accounting framework that addresses these problems with the existing SNA gross income measure and implements a net income measure using the Augmented Productivity Database (APDB) developed by Asian Productivity Organization and Keio University for China over the years 1970-2020. Real gross and real net income generated by the Chinese production sector is decomposed into explanatory factors including TFP growth using the framework suggested by Jorgenson and Diewert and Morrison. TFP growth is further decomposed into technical progress and inefficiency components using the nonparametric approach developed by Diewert and Fox. The APDB has estimates for the price and quantity of agricultural, industrial, commercial and residential land used in China. The paper argues that changes in land use should be treated in the same manner as inventory change and added to the alternative output measures. It turns out that Jorgensonian user costs for land are frequently negative. The problems associated with negative user costs are discussed in the paper.

This paper estimates the asset service lives (ASL), the rates of depreciation, and the declining balance rates (DBR) based on a finely-defined classification of assets, which distinguishes 369 asset types, using data on ASL and aged-prices of disposed assets collected in the Survey on Capital Expenditures and Disposals from 2006 to 2014 in Japan. This survey collected 1,111,913 data of disposed assets from business accounts of private corporations, of which 936,861 are defined as retired assets acquired new and are used to estimate the survival profiles, and of which 84,291 are defined as sold assets for continued uses with positive prices are used to estimate the age-price profiles. The Japan’s rates of geometric depreciation estimated in this study are broadly similar to the estimates at Statistics Canada (Baldwin, Liu, and Tanguay, 2015), but considerably higher than those used in the U.S. BEA (2013), e.g. 7.5%, 7.3%, and 3.2%, respectively, for industrial building construction, 7.6%, 6.7%, and 2.5% for office buildings, 29.1%, 27.9%, and 19.3% for automobiles, 24.4%, 24.9%, and 11.2% for telecommunication equipment, and 17.9%, 17.2%, and 11.0% for industrial machinery. The sources in these gaps are due to the shorter ASL and the higher DBR in Japan and Canada, compared to the U.S. The exceptional asset is computers. Japan’s rate of depreciation for computers is estimated as 29.6%, compared to 43.1% in Canada and 33.9% in the U.S. The gap is originated from much longer ASL observed in Japan (8.0 years on average), compared to Canada (4.9 years).

What can we learn from the experiences of sustained energy productivity improvement (EPI) in the economic growth and structural change of the postwar Japanese economy to guide ap-propriate energy and environmental policies? This paper aims to derive implications for policy by discussing the structural factors contained within the macro-observed EPI gross indicators, es-pecially the factors behind the acceleration of the EPI since the late 2000s. The measurements in this paper show that there has been no policy-driven acceleration in the sustained realization of the EPI in the postwar Japanese economy, but rather a significant diminution in the speed of improvement over the past half-century until the late 2000s. The acceleration of the EPI in recent years has been accompanied by a shift of energy-intensive goods to overseas production, and domestic industries that are required to curb consumption have been forced to lower their capital and labor productivity. The evaluation, including price changes, shows that the real unit energy cost facing the Japanese economy rose sharply in 2021–22, increasing its vulnerability to energy price hikes. In an energy transition that will take decades, it is necessary to ensure economic efficiency during the transition period without obsessing over poor domestic emission reductions.

While carbon neutrality toward 2050 may provide growth opportunities for Japanese companies with advantages in energy-saving and other technologies, the impact of the increased cost burden on the competitiveness of the Japanese economy must be carefully monitored during the long-term transition process. In a Japanese economy that has had to face relatively high energy prices for a century and boasts a highly energy-efficient industrial structure, the costs of additional energy savings have already multiplied significantly. In addition, after more than a decade of policy support by the Feed-in Tariff (FIT) program, the expansion of renewable electricity has significantly increased the price of electricity in Japan, and the burden on households and industry has grown to several times what was assumed when the FIT program was introduced in 2012. It is increasingly important to observe the impact of energy price hikes on Japanese industry and households, as well as the cost burdens required for the energy transition, in near-real time. This paper introduces the development of monthly estimates of the Real Unit Energy Cost (RUEC) and its components, an indicator that comprehensively assesses the resilience of the domestic economy to energy price hikes, and reports preliminary figures through January 2023, the most recent month available.

The paper constructs estimates of China’s Total Factor Productivity (TFP) growth over the period 1970-2020 using primarily the Asian Productivity Organization (APO) data base augmented by additional data developed by Koji Nomura at Keio University. China's TFP growth will be measured using the following methods: (i) the exact index number method developed by Jorgenson (and his coworkers) and Diewert and Morrison and (ii) a nonparametric method developed by Diewert and Fox. Economy wide estimates of rates of return on assets will also be calculated. Estimates for quality adjusted labour developed by Nomura will be used as well as estimates for missing inputs such as Land. Finally, instead of using GDP as the output concept, alternative estimates of TFP growth using net output as the output concept will be provided. The alternative measure is useful in explaining Chinese real income growth.

This paper develops a wholesale and retail service production database of Japan (WRJ) at a level subdivided into 5,740 categories for wholesale and 6,888 categories for retail, in order to re-examine the price-quantity breakdown of production values of wholesale and retail services and to revise the productivity account for commercial industry. According to our examinations based on WRJ, the current estimate on the service price of commerce in the Japanese system of national accounts (JSNA) is underestimated for 1980–2005. And conversely, it is highly likely that it is overestimated for 2005–17, in which all WRJ estimates based on different assumptions declined and the KLEMS aggregate input price in commercial industry also declined, whereas only the current JSNA estimate is increasing. It suggests that during this period, the growth rate of total factor productivity in commerce industry is underestimated by 0.3% per annum.

Measuring benchmark production systems plays a crucial role in determining the accuracy of GDP statistics. The purpose of this paper is to find a type of relationship between main activities (JSIC 395 activities) and secondary products (SDS 900 products), based on the newly designed Survey on Diversification of Sales (SDS) in Japan. Cataloging the production structure of detailed secondary products by main activity enables the examination of the accuracy of the benchmark production system. According to our examination, 14.9 trillion yen (about 7.7 trillion yen in terms of GDP) of the domestic production value of secondary products may be missing in the current supply table in the 2015 benchmark year.

本稿は、日本経済における実質単位エネルギーコスト（RUEC）とその日米格差の測定によって、エネルギー価格高騰による経済的な耐性を評価することを目的としている。日米比較によれば、日米RUEC格差は米国のシェール革命以降に急速に拡大しており、近年では米国に比して60%上回る水準にまで拡大している。それは戦後のピークに達する水準であり、エネルギー価格変動に対する日本経済の脆弱性が相対的に高まっていることを示している。日米RUEC格差の拡大要因は、エネルギー生産性格差の縮小と実質エネルギー価格差の拡大の両面によっている。近年、米国のエネルギー生産性の改善率は高く、日本の優位性はこの20年間で半減している。 日本経済における長期的な傾向をみれば、RUECが上昇へと転じた転換点は1990年代半ばである。オイルショック後に強化されたエネルギー価格高騰に対する耐性は、この20年間に再び脆弱化している。第一の要因は、おもに賃金率低下によって誘導されたデフレ型の実質エネルギー価格の上昇である。1990年代半ば以降、二度のオイルショックを含む期間（1973–1995年）に比して、名目エネルギー価格の上昇率は1/3ほどに縮小されたが、実質エネルギー価格の上昇率としては年率1.0%から1.5%へとむしろ高まっている。デフレ型のRUEC上昇は、エネルギー価格高騰に対する耐性の脆弱化がエネルギー多消費産業のみではなく、より広範な産業活動へ及ぶことを意味している。第二の要因は、エネルギー生産性改善における低迷である。実質エネルギー価格上昇が逓増傾向にありながらも、安価に利用可能な省エネ技術は限られたものとなり、エネルギー生産性の改善率は逓減している。 長期化するデフレ経済は、エネルギー価格高騰に対する脆弱性も増大させてきた。日本経済におけるエネルギー価格高騰への耐性強化のためには、再エネ推進のためのコスト拡大の抑制、原子力や石炭利用を含めた適切なエネルギーミックスの実現とともに、労働生産性の改善を通じた賃金上昇や高付加価値化など、デフレ経済からの脱却のための経済政策が求められる。

産業連関表（基本表）における卸・小売マージンは、日本の国民経済計算体系（JSNA）での商業生産額のベンチマーク推計値となるとともに、商品別マージン率はコモディティフローへと組み込まれ、GDP統計の精度に対して直接・間接的に大きな影響を与えている。野村・宮川・新井（2017）では産業連関表における商業マージン額について、その推計における諸課題を検討し、とくにはじめての「経済センサス–活動調査」に基づく2011年表において大幅な過小推計にあることを指摘している。本稿では、その再検討とともに基本分類レベルでのマージン表の改訂値を求め、産業連関体系のリバランスを通じてGDPにあたえる時系列的な影響の評価をおこなう。本稿での推計結果によれば、商業マージンの過小推計バイアスによるGDPとしての欠落は1995年では4.8兆円、2000年5.1兆円、2005年6.6兆円であり、2011年では25.6兆円と評価される。そうした改訂による名目GDP成長率としての影響は、2000年代半ばまでは年率0.1%ポイント未満と軽微であるが、2005–2011年では年率0.6%ポイントほど（年平均マイナス1.07%成長からマイナス0.44%成長へ）の上方改訂となる。現行のGDP統計に内在する過小推計バイアスは、かなり蓋然性が高いと考えられる諸仮定のもとで導かれたものであり、2011年基準JSNAにおけるGDP統計の再検討が求められよう。

本稿は1885–2016年における日本経済の長期エネルギー生産性の推移を概観しながら、とくに第II期：高度経済成長期（1955–73年）、第III期：オイルショック後（1973–90年）、第IV期：ポストバブル経済期（1990–2008年）、そして第V期：ポスト世界金融危機（2008–16年）の4期間において、電力化の進行などエネルギー品質の変化や、産業構造の変化を明示的に考慮に入れたうえでのエネルギー生産性を測定し、その産業起因を分析することを目的としている。マクロレベルでの見かけ上の生産性改善の黄金期は第III期にみられるものの、エネルギー品質や産業構造の変化を統御すれば、生産性改善のピークはむしろ第II期であり、エネルギー多消費産業が相対的に拡大しながらもエネルギー生産性の改善を伴うものであったことが示される。第II期の年率2.0%から、第III期には年率1.5%、低成長となる第IV期では年率0.1%へと減速するが、第V期においては再び年率1.6%への回復している。しかしその要因分析によれば、それは必ずしも持続的な要因ではない。一定の仮定に基づけば将来におけるエネルギー生産性改善は年率0.2–0.6%と予測される。

This paper provides new benchmark estimates of industry-level price differentials between Japan and the U.S. for 2011 based on a bilateral price accounting model anchored to the Japan-US input-output tables. We apply the model to translate available demand-side data on purchaser’s price PPPs for final uses (e.g. the Eurostat-OECD PPPs) and intermediate uses (e.g. the METI survey) to unmeasured producer’s price PPPs for industry output. These PPPs allow us to produce price level indexes at the industry level, which we use to assess price competitiveness between Japan and the U.S. Under the nominal exchange rate of 110.6 yen per dollar as of the beginning of July 2018, we estimate that producers in Japan have a pricing advantage in 66 of 106 industries in the manufacturing sector, and in 24 of 50 industries in the service sector. We conclude that price competitiveness of Japanese service industries has considerably improved in the more recent time period. However, Japanese producers have a significant price disadvantage in comparison to their U.S. counterparts in electricity and gas supply, and most of the agricultural producing industries.

我が国におけるストック統計の不備、特に社会資本ストック統計の不備という現状に対し、国民経済計算（SNA）と整合的なかたちで公共財としての社会資本形成の推計をおこなった。また、同時に民間・公的別固定資本マトリックスの1955年から1992年にわたる時系列推計のフレームワークを提示した。