Project details : 2018

Taku Suzuki (Teikyo University), Ma Xinxin (IER), Ichiro Iwasaki (IER)

In the framework of the project, we had a kick-off meeting at Kyoto University in April 2018, and then carried out literature search, coding of empirical results reported in collected studies, and their systematic review/meta-analysis each for research objectives of the project members. The research results have been presented at the MAER-NET Colloquium in Melbourne (October 2018), the International Workshop of Advanced Study on Transition Economics and Comparative Economics in Kyoto (November 2018), the Invited Seminar of Modern Economics of Fudan University in Shanghai (December 2018), and the Tokyo Workshop on Comparative Political Economy of Emerging Markets (March 2019). We also publish an edited book, a book chapter, 3 journals articles, and a discussion paper as the final outcomes from the project.

Andreas P. Kyriacou (Universitat de Girona), Tomohito Okabe (IER)

In the context of that project entitled "Institutional Differences in Political Budget Cycle". the three members of the research team have empirically examined the existence of a political budget cycle (PBC) in a panel of 70 developed and developing countries over the period 1996 to 2016. We found that elections (especially programmed elections which are not subject to political manipulation) increase (reduce) budget deficits (surpluses). This result is robust to the inclusion of potentially confounding variables including lagged values of the dependent variable, economic development, economic growth, government quality, age and level of democracy.

Importantly, we began to explore the possibility that the impact of election on the PBC is determined by the level of economic development. Our initial findings indicate that elections do not have an influence on the fiscal balance in ‘rich’ countries, while in ‘poor’ countries elections lead to an increase in the public deficit (or reduction of surplus). The analysis reveals that the results are not driven by democratic experience (as in Brender and Drazen 2005) or government quality (as in Shi and Svensson 2006).

We now hypothesize that this result is driven by different discount rates of voters in each setting. In poor countries, scarcity means that voters have high discount rates in the sense that they value public rents in the short term to fiscal sustainability in the medium to long-term (see, similarly, Charron and Lapuente 2010, for this time-preference approach). This leads politicians to adopt expansionary policies during election periods. This approach is interesting and potentially productive since it does not require that voters be inexperienced with democracy. Nor does it require that politicians be corrupt. They simply respond to voter-short-termism, by adopting expansionary policies in election periods.

In the context of the current research project we are now exploring this hypothesis fully. This requires further robustness checks through the introduction of additional control variables such as constitutional checks and balances, transparency and fiscal decentralization (De Hann and Klomp 2013; Eslava 2011) as well as the employment of alternative dependent variables in the guise of disaggregated public expenditure and revenue variables (e.g. compensation of public sector employees).

Toshiaki Watanabe (IER), Hiroki Yamamoto (The University of Tokyo)

In this project, we investigate descriptions on the economic condition from "Economy Watchers Survey" using the machine learning method and obtain consumers' inflation expectations from news articles in the "Nikkei Shinbun." The research outcome is reported in a discussion paper and is presented at the 26th RCAST Macroeconomic Workshop and the 27th Symposium of the Society for Nonlinear Dynamics and Econometrics.

Ryo Kambayashi (IER), John Tang (Australian National University), Kaori Sato (Kokushikan University), Yukiko Asai (University of Chicago)

This project investigates the following three research agendas related to human capital accumulation, job search, job transitions and promotions in the Japanese labor market: (1) Relationship between wages around job transitions and firm-specific human capital, (2) Costs of employment and labor demand, (3) Female labor force participation post World War II. We held a research seminar in November 2018 and published a working paper this year.

Yutaka Arimoto (IER), Yukichi Mano (Hitotsubashi University)

To examine the determinants of low quality fertilizers in East Africa, we collected fertilizer samples from markets along the Kigali-Kampala-Nairobi highway from January to February 2019, sent them to ICRAF, Nairobi for the content testing, and also had interviews to retailers and wholesalers of agricultural inputs to obtain the information on their input suppliers and other shop attributes. We have been analyzing the data to identify the causes of low quality fertilizers floating around markets in East Africa.

Mari Tanaka (Hitotsubashi University), Chiaki Moriguchi (IER)

We compiled data from the Ministry of Education annual reports, Higher School student registers, and the Personnel Inquiry Records of 1939. Using these data and taking advantage of repeated bidirectional reforms observed in 1898-1930, we analyzed the short- and long-run impacts of admission reforms for higher education in Japan. The results show statistically significant changes not only in application behaviors and admission outcomes, but also in the geographical distribution of elites forty years after the reforms. Our research is the first to quantify the long-run effects of admission reforms on human capital formation. To provide further empirical analysis, we also digitized a part of the Japanese Who's Who published in 1952 using this research grant.

Masahiko Tsutsumi (IER), Ryoichi Nanba (Chubu Region Institute for Social and Economic Research)

We prepared the dataset by using "National Accounts of Japan" (Cabinet Office) and "National Survey of Family Income and Expenditure" (Ministry of Internal Affairs and Communications) for a generational accounting model in Japan. Then, we estimated the key parameters related to the distributional structure of benefit and burden among inter- and intra-generations, and ran policy simulations. In March, having discussions by members, we summarized the intermediate report, which was presented at the Joint Usage and Research Center Programs Seminar, Hitotsubashi University.

Osamu Saito (Hitotsubashi University), Leigh Shaw-Taylor (University of Cambridge), Tokihiko Settsu (Musashi University)

Throughout the academic year, Yamamoto worked on labor supply behavior of English women, Saito on an international comparison of occupational structures including Japan and Britain, Shaw-Taylor on occupational structures and female labour force participation rates in Britain, and Settsu on labor productivity in Japan. The results were presented at various international conferences. In March, Shaw-Taylor was invited to Japan. While in Hitotsubashi, he finalised the format that will be used for occupational data for 18 countries, covering half the world population in 1900. He also worked on the estimation of labour force participation rates by region in England and Wales. During his stay, we, four, met together at Hitotsubashi to share our findings. We also held a research seminar at Osaka University in March.

Nobuyuki Kitamura (IER), Takashi Unayama (IER), Yukiko Ito (Japan Center for Economic Research), Masayuki Kawagoe (Nihon University), Masayuki Kobayashi (The University of Tokyo), Masato Takara (ARISE analytics), Rieko Nagamachi (Otemon Gakuin Univesity), Saeko Maeda (Japan Center for Economic Research), Megumi Niimura (Teikyo Heisei University), Maki Yokoyama (The University of Tokyo), Bruce Chapman (The Australian National University), Shiro Armstrong (The Australian National University), Lorraine Dearden (University College London)

Using Japanese Labor Force Surveys and panel surveys of JHPS and KHPS, as well as Basic Structure of Wage Surveys, we estimated income distribution of university graduates by income bracket, thereby estimating budget cost of the government for different income contingent schemes. We newly included time dependency of income in the estimation. The paper was accepted in the special issue of Economics of Education Review. As further research, we are now estimating return on university education and change in the wage structure and the effect on fertility.　

Kazuhiro Motohashi (The University of Tokyo), Kyoji Fukao (IER)

The project described two promising sub-projects, "Growth and survival of innovative startup firms in Japan: (technology) competition, uncertainty and technological opportunities" and "Hiring or collaborating: the role of scientists in firm performance in Japan". The first project required access to official survey data (census data) held at NISTEP and RIETI and the plan was to collaborate with researchers in Japan who have direct access to the data at NISTEP or RIETI. Unfortunately, it proved not possible to get the commitment of NISTEP/RIETI researchers for this project. My plan is not to postpone this project to my sabbatical year at HIAS, and apply for access to the data myself.

The second project involved publicly available patent data and data processing work by the research group of prof. Motohashi. Here access to the data was not restricted and this allowed me to involve expert researchers in Belgium (Sam Arts) and the Netherlands (Paul Hunermund) in the research team. During my research stay at Hitotsubashi University, funded by the Joint Usage and Research Centre grant, I visited NISTEP frequently and discussed data processing issues and collaboration. The data itself were only ready and transferred with a considerable delay (fall/winter 2018), which hampered the implementation of the planning for the project. The information from Scopus on scientist-inventor profiles and academic publications could not be released due to a restrictive and expiring license to this data source. Furthermore, the initial analysis of the patent data revealed a number of inconsistencies and data issues that required discussion and adjustment of research plans. Finally, we were able to produce an analysis of the trends in patenting behavior in Japan: by universities, firms in collaboration with universities, firms in collaboration with university scientists, and firms with former scientists hired as inventors. We adjusted the direction of research and conducted research at the patent level. We performed preliminary analysis of the quality (forward citations) of firm patents as a function of collaboration with universities or scientists, and hired former scientists. Results are summarized in an extended abstract.

During the research activities, we expanded the research team to include Kenta Ikeuchi (RIETI) given his expertise related to the Japanese patent data, as well as Sam Arts (KU Leuven) and Paul Hunermund (Maastricht University) given their expertise in research on industry-science linkages.

The abstract and trends were presented at the research meeting in February 2019, including the new team members. One of the outcomes of the discussion was that the database is in need of quality improvement as not all scientists are currently identified on Japanese patents. This however, requires programming related to addresses and names in Japanese and is difficult to implement by the research team in Belgium and the Netherlands. Work to improve scientist identification is currently ongoing in Japan under the supervision of Kazuhiro Motohashi.

The lack of access to the original work on the characteristics of Japanese scientists based on Scopus (due to licensing issues) can be remedied by using the latest version of Microsoft Academic Graph, a public access database on scientists and publications. This implies redoing name-matching and disambiguation exercises, in a major additional database construction exercise. We are currently exploring whether we can follow this route, as we think that the performance effects of inventive efforts of firms in collaboration with (former) university scientist crucially depend on the characteristics of these scientists (research strength, applies vs basic research). If this is feasible, our plan is to complete a full working paper in the fall of 2019. We will submit all output from the project to the JUC as required.

Yoko Ibuka (Keio University), Noriko Inakura (Osaka Sangyo University), Chiaki Moriguchi (IER), Naohito Abe (IER)

In this project, we estimated prefecture-level price indexes for calorie and major nutrients in Japan from 1975 and 2015. For this purpose, we merged food expenditure data from "Household Survey" (which cover about 270 food items) with price data from "Retail Price Survey" and nutritional data from "Japanese Food Composition Database." Using the prefecture-level panel data, we also examined the relationships between nutrient prices and child obesity and found that the relative price of carbohydrate to dietary fiber is negatively associated with children's BMI.

Ekaterina Hertog (University of Oxford), Kamila Kolpashnikova（National Taipei University), Kiyomi Shirakawa (IER), Ryota Chiba (IER)

During 2018-19 thanks to joint usage and research center programs funding we have analysed gender differences in the participating types and social contexts of domestic work and care work in Japan between 1981 and 2016. We confirmed that the format of 1981 and 1986 data is different from the later surveys so now we constrain our comparative analysis to 1991-2016-time frame. We compared Japanese time use trends with a number of OECD countries, such as UK, Germany, and also South Korea and we now believe that putting Japanese time use trends into a broader international perspective will be more informative and will enable us to locate Japan more precisely within the varieties of welfare capitalism. Towards that aim Dr. Hertog has started converting anonymized data into internationally agreed MTUS (multinational time use survey) format. Finally, Professor Shirakawa has taken a lead examining the impacts of modern technology on domestic and care work and value of unpaid work in Japan.

Yessica C. Y. Chung (National Ping Tung Uiversity of Science & Technology), Yukinobu Kitamura (IER), Takeshi Maru (Bunri University of Hospitality), Atsushi Chitose (Tokyo University of Agriculture and Technology)

We have analyzed the impact of promotion policy of agricultural industrialization on LPLC (local production and local consumption). Private corporations and JAs had low rate of LPLC, and thus the impact on land conservation is weaker than the case individual farmers engaging in industrialization. Second, an analysis of the impact of internet use on agribusiness performance showed that internet use significantly increases sales (per labor) and the proportion of sales direct to consumers.

Takashi Unayama (IER), Naomi Kodama (Nihon University), Takahisa Dejima (Sophia University), Shinsuke Ito (Chuo University), Mariko Murata (Statistical Information Institute for Consulting and Analysis), Keiichi Sato (Senshu University), Hiroaki Matsuura (Shoin University)

This research explores the relationship between residents' consumption and household assets as well as the relationship between asset ownership and employment using residential areadata (individual data) from the National Survey of Family Income and Expenditure. This research also attempts the construction of a microsimulation model for population projection based on microdata of "Comprehensive Survey of Living Conditions" which precisely reflects changes of demographic characteristics for individuals and households. Partial results of this research were presented at the workshop ‘Actual Situation of the Social Economy Based on Microdata’ held at Hitotsubashi University in March 2019, and will be published as a discussion paper by the Institute of Economic Research, Hitotsubashi University and in other publications.

Yutaka Arimoto (IER), Ralandison Tsilavo (Kyoto University)

A daily trade survey for rice traders in Alaotra was conducted in June-October, Madagascar, and a baseline survey on demographic and business information was conducted in December-February. In March, we visited the site and conducted a qualitative follow-up survey, finding the importance of the intermediaries connecting urban merchants and local merchants and farmers. In March, I made a research presentation on the rice market in Madagascar at a symposium in Taiwan.

Reiko Goto (IER), Kotaro Suzumura (Hitotsubashi University), Naoki Yoshihara (The University of Massachusetts)

This research project reexamined various impossibility theorems of social choice theory in the context of practical use and desirable social systems. As a result, we have several possibility theorems as follows: (i) desirable voting rules escaping from Arrow's impossibility theorem; (ii) classifying desirable class of well-being indices that respect dominance principle and weaker version of individual preference axioms; (iii) formulating Amartya Sen's capability approach and investigating its theoretical properties of equality of capabilities by formal economic model analysis. In addition, the project had monthly workshop and succeeded to organize the research group for promoting natural and social scientific study on human-being's moral judgments and normative intuitions.

Renée B. Adams (University of Oxford), Tokuo Iwaisako (IER), Konari Uchida (Kyushu University), Yung-Ling Chi (National Chung-Hsing University)

The impacts of recent institutional reforms on the corporate governance of Japanese firms are investigated using Japanese micro data and by the international comparison using the data from developed economies. We held the conference in conjunction with the Asian Finance Association's annual meeting in June 2018. We also presented our researches in the organized session at the WEAI meeting in March 2019. These events allow us to disseminate the results of this project internationally.

Yuhei Miyauchi (Stanford University), Jie Bai (Harvard University)

Using the research funds including the support from IER, we have already finish a pilot RCT study, which helps us to (i) organize and refine the recruitment procedure for treatment/control firms, (ii) design the contents of information intervention, and (iii) organize and refine the contents and the procedure of follow-up survey. Based on these results, we are planning to implement a full-scare RCT under a collaborative work with RIETI in this summer.

Nugyen Duc Thanh (Vietnam Institute for Economic and Policy Research), Nguyen Thanh Tung (Vietnam Institute for Economic and Policy Research), Futoshi Yamauchi (International Food Policy Research Institute)

At the inception of the research project titled "Firm and Farm Size Puzzles in Developing Countries", I have conducted the empirical research for Vietnam and Indonesia using the micro panel datasets.
As originally planned, I collaborated with the Vietnam Institute for Economic and Policy Research (VEPR) and constructed a merged data based on Vietnam's firm census, provincial competitiveness index (PCI) data, and administrative data on the district-level minimum wages. We applied difference-in-difference model to estimate the effect of the 2011 minimum wage hike on firm size, profit, employment, and entry and exit behaviors under heterogeneous labor and product market competition. At the end-February in 2019, we conducted field interviews on firm's perceptions on the labor and land regulations covering formal and informal enterprises in two northern provinces (Hanoi, Bac Ninh) and three southern provinces (Ho Chi Minh city, Ba Ria-Vung Tau, and Binh Duong). We have been developing theoretical and empirical analyses to prepare the first draft working paper for the upcoming research seminar (scheduled in July 2019). We continue to explore the research collaboration with IFPRI on the farm size issue using Vietnam's Household Living Standard Survey (VHLSS) panel data.
I finalized two working papers: (i) a paper that estimate the impact of disaster aid after the 2004 Indian ocean tsunami on employments and income dynamics of microenterprises in Aceh province, Indonesia; and (ii) another paper that estimates the impact of factor market allocation on firm size in sub-Saharan Africa. I presented the papers in seven local and international seminars and am finalizing them for journal submissions. I intend to extend African firm analysis to more specific empirical studies in collaboration with development economists working in the related field.

Takashi Unayama (IER), Ryo Kambayashi (IER), Masayoshi Hayashi (The University of Tokyo), Shun-ichiro Bessho (The University of Tokyo), Kazufumi Yugami (Kobe University), Koyo Miyoshi (Aichi-Gakuin University)

We considered the potential research plan to analysis wage structuWe considered the potential research plan to analysis wage structure about employee with disabilities using microdata from "Statistical Survey of Actual Status for Salary in the Private Sector". Since the survey has never used in an academic research, it was quite important to verify differences from other statistics. We contacted the National Tax Agency for data use. As it was difficult to use data in the fiscal year, we proceeded with the procedure for data acquisition in the next fiscal year.re about employee with disabilities using microdata from "Statistical Survey of Actual Status for Salary in the Private Sector". Since the survey has never used in an academic research, it was quite important to verify differences from other statistics. We contacted the National Tax Agency for data use. As it was difficult to use data in the fiscal year, we proceeded with the procedure for data acquisition in the next fiscal year.

Takashi Unayama (IER), Shinpei Sano (Chiba University), Ryosuke Okazawa (Osaka City University)

First, to gain individual data of "School basic survey" (Ministry of Education, Culture, Sports, Science and Technology-Japan), we made an application for use after prior consultation with the department in charge. Next, we surveyed previous literature and its methods. After the data arrived at the end of 2018, we worked on replicating the totaled values of the main items that have been published. In addition, while understanding the feature of this data, the task for creating panel data became clear. Therefore, we discussed how to overcome it.

Abu Perves Shonchoy (Florida International University), Akib Khan (IDinsight), Takashi Kurosaki (IER)

1. Focus Group Discussions
A total of four Focus Group Discussions were held on site with the residents of the community. The Research Team along with the Field Manager conducted multiple sessions (October and December 2018) with the mothers who have at least one child under the age of five. The aim was to improve on the design of our Growth Charts. Since these charts will be hanged inside the households during the entire period of our intervention, it is important to ensure that the caregivers are able to understand the content.

2. Community Census
A community census was deployed in September 2018 with the aim of locating households which meet our eligibility criteria of having at least one child between the age of 6- and 24-months. The study site was extended towards the North-West in order to recruit additional households in our study which could give us the required sample yield of eligible households. The census was resumed in November 2018 to survey the remaining sub blocks in the previously demarcated area of Gulshan-e-Sikandarabad, Karachi and five additional sub blocks in the extended area.

3. Food Preparation Survey
A food preparation survey was conducted in January 2019 with the aim of understanding the already existing child nutrition practices in the target community. 20 women were interviewed separately with infants in the age bracket of 6-24 months. The survey through 24HR child feeding recall and other questions investigated the breastfeeding practices, number of meals and snacks consumed by the infant, actual food items consumed by the child during past 24 hours and hygiene conditions in the household. The survey results showed that many mothers are giving table food to the infant because of time constraint. If they are cooking separately, it is something very quick to make. The children are eating unhealthy snacks and there is little diversity in food items consumed. Mothers are cooking fresh food every day with little storage of left-over food.

4. Dry Run
The Research Team conducted a Dry Run in March 2019. The purpose was to understand the feasibility of the intervention plan, test intervention material and finalize logistics for the final Intervention phase which shall commence after the Baseline Survey. The Dry Run was deployed in twelve households which were divided into groups of three and each group had children in the age bracket of 6 and 8 months, 9 to 11 months and 12 to 24 months respectively. The results show that the caregivers were not very welcoming to the idea of inviting caregivers with heavy equipment to measure their child's weight and height. Moreover, the mothers were keen to listen to the nutrition counseling, however, they had a short attention span which is why we need to shorten the counseling manual. The GroMoTo tool was very well-received and well-understood. The mothers were able to understand how healthy food can help their child follow a healthy growth trajectory and they seemed to understand the difference between the red and green zone on the height-for-age chart.

Iichiro Uesugi (IER), Chihiro Shimizu (Nihon University)

This project made significant progress by the support of Joint Usage and Research Center. First, we purchased the data set that describes the land use in Tokyo for 1986, 1991, 1992, 1996, 1997, 2001, 2002. 2006, 2007, 2011, and 2012. This data set, which is based on the Tokyo Metropolitan Land Use Survey, includes the variables necessary for our economic analysis. Specifically, the data set includes information about land use and building size for each square-kilometer grid. The variables include the building floor area, the size of building footprint, the number of buildings, and the Herfindahl-Hirschman Index (HHI) that are disaggregated by major land use and building structure.

Second, we combined this data set with the information about property transactions, Chika-Koji land price, urban redevelopment projects, Population Census, Economic Census, and Establishment and Enterprise Census. We constructed a set of grid-level variables and analyzed the time-series and cross-sectional variations in land use.

Third, we wrote a discussion paper entitled, "Real Estate Bubble and Burst, Public Involvement, and Land-Use Efficiency," and presented it in a workshop we organized. The workshop, 2019 Workshop on resource allocation and productivity, was composed of three paper presentations. Two other presentations are: "Long-Run Renewal of REIT Property Portfolio through Strategic Divestment" by Masatomo Suzuki (Univ. of Tokyo) and "Reallocation of Tangible Assets and Productivity" by Kaoru Hosono (Gakushuin University). The workshop was extremely productive with very active discussions among participants. Many constructive comments and suggestions at the workshop helped us advance this research project. The participants including those from foreign institutions were also pleased to learn more about resource allocation issues.

Our analytical results are summarized as follows. The Japanese economy experienced in the late 1980s the so-called Bubble Period that is characterized by high asset prices and large consumption. The subsequent stagnant economy is often termed the "lost two decades." We hypothesize that active real estate transactions during and after the Bubble Period made an impact on aggregate productivity by changing the allocation of land for various uses. Active transactions during a real estate boom can potentially increase productivity through the optimized land use but may decrease productivity if the optimization process is slow.

We compile 25 years of geographical information system (GIS) data on built properties, land use, population, and businesses for Tokyo's 23 wards between 1986 and 2011. We study how changes in land ownership and land use affected local economies and aggregate productivity. First, we develop the theoretical framework to analyze how land use can be associated with aggregate productivity. Second, we study how the characteristics of transactions changed during and after the Bubble Period. Sales by individuals to firms were active during the Bubble Period despite large cross-sectional variation. The most active markets include Shinjuku. Sales by firms to individuals became more active after the period, particularly after 1997. At the same time, foreclosure sales due to personal bankruptcy increased throughout the sample period until 2001. Third, we analyze the progress of quasi-public urban redevelopment projects. During the Bubble Period, a relatively large number of projects started. These projects were subsequently completed in the late 1990s and the early 2000s. Fourth, we analyze the change in land use, population, and establishments. We find the progress of land ownership concentration during the Bubble Period. We also find a secular upward trend in the number of stories and floor area throughout the sample period.

We further analyze a particular area around Shinjuku Tomihisa Cho to understand the specific change in land use. This case study reveals that active transactions during the Bubble Period can result in prolonged stagnation in economic activities before these properties are fully redeveloped to the highest and best use (HBU). Specifically, the transacted properties became vacant for years before structures were demolished to vacant lots. Many lots remained vacant for additional 5 to 10 years. The form and process of redevelopment also vary from a purely private redevelopment to a quasi-public urban redevelopment project to eminent domain for a major road construction. In the next step, we will analyze whether public involvement accelerated or decelerated redevelopments.