プロジェクト研究： 平成27年度採択課題一覧

本研究では、保育及び育児休業制度に代表される家族政策が、女性の就業と子供の発達に与える影響を評価する。女性の就業については、賃金、就業時間、雇用の安定性、職業訓練の機会といった「仕事の質」に与える影響や、その効果が長期的に持続するかどうかについて特に注目して分析を行う。子供の発達については、認知・身体能力に加え、健康面への政策効果を評価する。親と子供のかかわり方や、子供に対する人的資本投資にも着目することで、家族政策が子供に影響を与えるメカニズムも明らかにする。

我が国における先行研究では、因果関係と相関関係を明確に区別するような分析はあまりなされてこなかった。本研究では、固定効果法や操作変数法などを適用することで、政策の効果、つまり因果関係の識別を慎重に行い、より信頼性の高い結果の導出を試みる。

 Ekaterina Selezneva（Institute for East and Southeast European Studies）

Tatiyana Karabchuk（National Research University）、雲和広（一橋大学経済研究所）

This study aims at examining economic conditions of two genders in the labor market and their effects on the probability of giving birth. More precisely, the aim of this study is to investigate the effects of division of labor among households, gender wage gap, or gender segregation more broadly, on individual woman’s decision to give a birth.

Most of the gender studies in Russia and the Former Soviet States belong to the period after the collapse of the Soviet Union. Regardless the few existing studies, a relationship between gender pay gap in labor market and within individual households, and the effects of those on fertility is not yet investigated. Previous literature on demography and fertility has revealed that normative sense critically affect the probability to give birth. The current study addresses the question of the gender gap and fertility relationship both in Russia and the Former Soviet States.

近年、所得格差の拡大に伴って、日本における所得格差を拡大させる要因に関する分析が行われるようになった。これらの研究では、高齢化や若年者における所得格差拡大が全体の所得格差を拡大させる要因であることが指摘されている。所得格差拡大を軽減させるための政府の役割には所得再分配機能があるが、近年では所得税における所得再分配機能が低下していることが指摘されている。そのため、個票データを用いて所得税や社会保障の所得再分配機能について調べた論文も多く蓄積されている。

こうした所得格差や所得分配は、地域間で様相が異なる可能性がある。中国では地域間所得格差に関する研究が多く蓄積されており、例えばYang(1999)では、地方と都市部の所得格差が全体の格差の大半を占めており、セクター間の所得ギャップの拡大が所得格差を拡大させたこと、またその要因として都市部への補助金や投資などが挙げられている。他にも家計データを用いて地域間格差を分析した研究が蓄積されている。日本では、都道府県データを用いて都道府県間の経済格差を分析した研究があるが、個票データを用いた地域間比較の分析はほとんどない。このように地域間の所得格差に関する研究は日本ではほとんど行われていないが、実際には地域間格差が存在しており、日本の個票データで地域間格差の実態を明らかにすることは重要な課題だと考えられる。

加えて、ファミリー・フレンドリー施策などの育児支援が女性の就業行動に及ぼす影響についても検証したい。具体的には、女性のライフサイクルにおける、結婚、出産、子どもの就学などイベント前後における就業変化の在り様の推移を時代別に捕捉するだけでなく、イベント前における就業先におけるファミリー・フレンドリー関連の制度利用状況の違いが、就業行動にどのような影響を及ぼすか、特に先行研究事例が少ない、子どもの小学校入学時点における就業変化について検証する。

また、女性の就業状況の違いがもたらす、夫妻間における所得の配分の変化についても、考察する。妻が就業継続できることとなり、世帯員間における所得源泉の変化が夫妻間での消費行動、余暇・家事育児行動 (Browning, et al.2006)、夫妻間の所得・消費格差にどのような影響(Lise and Seitz 2011)を与えるのかを検証する。

学術研究のグローバル化と国際競争の進展を背景に、研究業績の重視や任期制・公募制の導入、研究者の女性比率の上昇等、日本の若手研究者の置かれた研究環境は近年大きく変化している。本研究プロジェクトでは、政府統計個票データを用いて、特に男女差や研究分野に着目しながら、大学院生や大学教員の採用・昇進・賃金等の実証分析を通じて研究者のキャリア形成や研究支援ニーズについて明らかにすることを研究目的とする。

米国では経済学分野の教育・研究に対する学術研究の蓄積が進展しているが、日本での研究は限定的である。本プロジェクトでは、文部科学省「学校基本調査」及び「学校教員統計調査」個票データを用いて、日本の高等教育機関・教員全体を分析対象とすることに特色がある。日本の大学における大学院生・教員の状況分析を通じて、研究者支援・育成政策提言の一助としたい。

林田実（北九州市立大学）、児玉直美（一橋大学経済研究所）、出島敬久（上智大学）、伊藤伸介（中央大学）、村田磨理子（（公財）統計情報研究開発センター）、池田欽一（北九州市立大学）、松浦広明（松蔭大学）

本研究の目的は，ワークライフバランスの進展が企業の生産性と雇用状況に与える影響，さらに社会保障政策との関連を定量化することである。具体的な分析対象としては，他の先進国とのワークライフバランスのあり方の差を踏まえ，日本の外資系企業(Inward FDI)や海外進出企業(Outward FDI) ，さらにはCSR(企業の社会的責任)に取り組む企業における女性の働き方を念頭においた多様な人材の活用(diversity)が挙げられ，年金などの社会保障制度等との関係も検討課題になる。

本研究においては，政府統計ミクロデータを用いて，まず労働需要側の分析として，企業の業績と財務状況の観点からワークライフバランスと賃金・雇用量の関連性について精密なモデル化を行うことによって，統計的・計量経済的な検討を行う。さらに労働供給側の分析として，ワークライフバランスと社会保障政策との関連性を追究するために，個人の就業行動や健康状態・介護状況が稼得所得や非勤労所得を含む可処分所得やライフスタイルに及ぼす影響を明らかにする。

それらによって，企業の雇用戦略と家計の就業行動，ワークライフバランスの進展との関係を総合的に明らかにすることが本研究の特色である。

Jean-Pascal Bassino (Ecole Normale Supérieure de Lyon)

1.A Overall presentation

The envisaged research includes two projects focusing on the relationship between indicators of

environmental sustainability and living standards in Japan.

During the last decade, the calculation of Genuine Savings (GS) has become a standard approach in the empirical assessment of (weak) sustainability. This approach pioneered by researchers of the World Bank (Hamilton and Clemens 1999) has been initially employed for developing countries. The calculation of GS is based on estimates of total national wealth (comprehensive wealth, CW), also calculated by the World Bank for a number of developing countries (World Bank 2006). Estimates of CW are obtained by aggregating the value of natural, human, and produced capital. GS is the net value (negative or positive) obtained by combining changes in stock values of the different components of CW over a given period of time (in general one, five, or ten years). The two main advantages of the GS approach are (i) to be based on straightforward and easily replicable estimation procedures, (ii) to convey to decision makers (who were initially national governments of developing countries, i.e. clients of the World Bank) a much clearer message than a simple adjustment of national accounts for the consumption of natural capital (e.g.Bartelmus 2009). GS has been also used recently for analyzing historical trends in the relationship between sustainability and living standards using country-level series for European countries and the United States, with times series going back to the pre-industrial revolution period, e.g. Greasley et al. (2014) for Britain and Lindmark and Acar (2013 for Sweden.

However, the GS approach suffers from a number of limitations:

(i) From a theoretical viewpoint, the relevant indicator of sustainable wellbeing is not a flow (GS) but a stock: the value of CW per capita (Dasgupta 2001, 2009). Population growth results in a “dilution” of CW, particularly its natural capital component (the human and physical capital components can be accumulated at a pace exceeding demographic growth; this is rarely the case for natural capital);

(ii) GS is convenient for providing a snapshot (or for a study covering a short time span) but it is not well suited for an investigation of long-term trends. Against this background, the empirical study by Arrow et al. (2012) estimating CW for China, Brazil, India, the United States, and Venezuela has a number of attractive features. This study covers a recent period, and only a short period of time (1995-2000), but it can be a source of inspiration for an economic analysis of the link between sustainability and living standards over a century or more. In particular, Arrow et al. (2012), which use per capita GDP as indicator of living standard, consider time as an asset and take into account TFP growth.

Some issues remain however unsettled or under discussion in the estimation of CW. In particular, Arrow et al. (2013) acknowledge that the standard procedure used for estimating the value of natural assets (and therefore CW and GS) ignores the value of fish stocks; the same remarks applies to the value of biodiversity. These types of assets are overlooked due to data limitations and difficulties in assessing the risk of loss of resilience. Furthermore, the calculation of GS and CW is undertaken at the country level. The implication is that the ownership, accumulation, and destruction of assets located in the rest of the world are overlooked. An additional limitation is that there is no clear consensus in the GS and CW literature on the treatment of intangible assets (excluding human capital; i.e. patents, know-how, institutions, and more broadly social assets). Hamilton and Liu (2014) follow a shortcut approach and regard intangible wealth as the ‘stock equivalent’ of total factor productivity. A final conceptual issue is related to the share of financial capital that is not backed by tangible assets and intangible assets (as defined above). Considering the magnitude and volatility of the stock of financial assets, and the intricate linkages between financial assets and natural assets, some clarification is required.

Below I describe the background and research objectives of the two projects (the list of references is at the end of section 2).

1.B - Project 1. Making the most of scarcity? Japanese natural assets since the 1870s

(Jean-Pascal Bassino, Kyoji Fukao and Osamu Saito)

Recent empirical work investigating the determinants of long run changes in living standards includes attempts by economic historian and environmental economists to estimate time series of the total value of natural assets for European countries and the Unites States as a component of total national wealth, i.e. CW and GS (e.g. McLaughin 2014). As a by-product, changes in stock value of natural capital per capita and in the composition of the stock of natural assets (including agricultural land) can be documented and analysed in a weak sustainability analytical framework. However, the literature dealing with historical trends in CW and GS does not take consider how institutional and technological changes, and the international economic integration has affected the access to and the management of natural assets by the different stakeholders in the 19th century, during the shift to modern economic growth, and in later decades. From that viewpoint, Japan appears as a particularly interesting case for at least four main reasons:

(i) Initial (i.e. pre-Meiji) conditions were characterized by an exceptionally efficient use, by the

international standards of the time, of very scarce natural resources especially in forestry, including silviculture (Totman 1989; Saito 2009, 2014). Available evidence suggests that dramatic changes in institutional arrangements introduced in early Meiji had a negative impact on for the sustainability of natural assets management, in particular in forestry; institutional changes introduced during WWII also had destructive consequences.

(ii) In spite of their scarcity, natural assets played a critical role in the initial of Japanese economic transformation, in the late Tokugawa and early Meiji. Silk, tea, and coal accounted for a large share of Japanese exports during that period, and the steady rise in land productivity of Japanese agriculture allowed feeding a growing population using exclusively domestic resources. The nutritional status of the population improved markedly although import of foodstuffs remained negligible until the 1920s.

(iii) Although initial conditions were characterized by extremely resource efficiency practices, the

adoption and adaptation of imported technologies resulted in a gradual transition toward of more resource intensive economic development trajectory. The evidence presented by Ayres (2008) and Warr et al. (2010) suggests that, as far as energy is concerned, these changes have been gradual.

(iv) The Japanese economy relied initially on domestic resources, but the total national stock of natural assets (i.e. owned by residents; and therefore distinct from the stock of domestic natural assets) increased rapidly in the late 19th and early 20th century. This applies not only to the use of natural assets in southern Sakhalin (Karafuto), Taiwan and Korea, but also to the ownership of or access to assets in foreign countries (and in high seas in the case of fish stocks). The estimation of the stock of Japanese natural assets should not be restricted to domestic natural capital but should also include natural assets owned or accessed by residents.

1.C - Project 2. Comprehensive wealth and living standards in Japan since the 1870s

(Jean-Pascal Bassino, Kyoji Fukao, Tokihiko Settsu, and Masanori Takashima)

Recent empirical work in environmental economic history using national accounting techniques has opened a promising path for the empirical analysis of the relationship between indicators of

environmental sustainability and living standards in the long run. However, the results obtained so far are inconclusive. Greasley et al. (2014) have assembled data for British capital back to 1765 for constructing net investment measures which are used as indicators of GS, along with two measures of future well-being: consumption per capita and real wages. They argue that the “results do not reject the postulated relationship between GS and future well-being, and show GS can be a forward looking indicator of future well-being for periods of up to 100 years”. However, in their study on Sweden covering the period 1850-2000, Lindmark and Acar (2013), find that GS may have been negative up until c. 1910.

However, the results presented in these two papers may be sensitive to the definition of the value of natural capital. As mentioned in section 1.A, the recent literature linking environmental and welfare economics (both theoretical and empirical) suggests that CW should be regarded as a better indicator of sustainable well-being than GS. An additional consideration is that both Greasley et al. (2014) and Lindmark and Acar (2013) use exclusively country level time series. Such a dataset does not seem suitable if the aim is to uncover a causal relationship. Since the time series constructed for Britain and Sweden do not exhibit much short term volatility, an alternative approach could be to estimate CW and GS for benchmark years at a regional level. An additional consideration is that the results for Britain and Sweden may have been affected by problems of coverage in the estimation of natural assets for the earlier period (as mentioned in section 1.A and 1.B in the discussion on the importance of fish stocks and biodiversity, in particular).