Economic Analysis Series No.201THE ECONOMIC ANALYSIS

The Relationship between Female Employment and Firm Performance Taking Biases in Reporting into Account

By Masaya NISHIHATA and Isamu YAMAMOTO

Abstract

This paper investigates the relationship between female employment and firm performance taking possible sample selection bias in the reporting of information on the number of female regular employees and managers into account. Using panel data of listed firms spanning the period from 2010 to 2015, we find no evidence that a higher share of females among regular employees is associated with higher firm performance as measured by firms’ return on assets (ROA). We also find no significant relationship between the share of females among managers and firm performance. On the other hand, we find that the larger the share of women that are managers among all female employees, the higher firms’ productivity as measured by their total factor productivity (TFP) tends to be. In particular, we find that TFP is higher when the share of women that are managers among all female employees is 15-20%. These effects are larger when we address the sample selection bias, suggesting that using data only on firms that report information on female employees may underestimate their impact on productivity. These findings are not necessarily consistent with the implication of Becker’s (1971) discrimination hypothesis and imply that the increase in female employment may have affected firm performance through an increase in labor productivity in recent years.

JEL Classification Codes: J71, J31, L25
Keywords: female employment, gender discrimination, sample selection bias

The Changing Relationship Between Investment and Tobin's q:
Empirical Analysis of Listed Manufacturing Firms in Japan

By Takayuki ISHIKAWA

Abstract

Many empirical studies on the investment function employ Tobin's q model and consider financial constraints by adding certain variables in an ad hoc manner. Using a panel of financial statement data for Japanese listed manufacturing firms spanning the period from 1975 to 2016, this study estimates the investment function using Tobin’s marginal q. The results show that the relationship between the investment ratio and the marginal q in Japan is not fixed but differs depending on the period. Specifically, the three key results were as follows: (1) the relationship changed around 1997; (2) this structural change in the relationship in 1997 has persisted until today; and (3) if this structural change is not taken into account, the role of financial variables may be overestimated. These findings imply that, first, this structural change can explain why investment has not been growing significantly despite the economic recovery in Japan. Second, they imply that the corporate analyses and policy evaluations in previous studies, due to the failure to recognize the structural change, may have been incorrect. Why the relationship between the investment ratio and the marginal q changed and how this structural change occurred remains an issue to be explored.

JEL Classification Codes: E22, E60, G31
Keywords: marginal q, financial constraint, investment, secular stagnation.

How Do Perceived Returns to Education Affect Schooling Decisions and Urban-Rural Gaps in Educational Attainment?
Empirical Analysis Using Regional Differences in College Attendance Rates among High-School Graduates in Japan

By Ryosuke MORIYASU

Abstract

This paper analyzes the impact of college wage premiums as perceived by students themselves on their decision to pursue higher education. The analysis is based on individual data on Japanese high-school seniors from the “Longitudinal Survey on High School Students’ Career Paths” conducted by the University of Tokyo between 2005 and 2011.

Furthermore, regarding the factors that cause difference between the returns perceived by students in urban and rural areas, this paper focuses on urban-rural differences in the availability of information and routes of acquiring information on colleges.

The key findings are as follows. First, Gary Becker’s human capital theory suggests that the higher the actual wage premium, the more strongly do students desire to pursue higher education. However, the current study shows that it is students' subjective view of the wage premium enjoyed by college graduates (perceived returns) that increased the desire to go on to college.

Second, such perceived returns were statistically significantly lower in rural than in urban areas. This suggests that regional differences in the subjective return on higher education are the reason for urban-rural gaps in educational attainment. However, it should be noted that much of this effect is caused by the negative effect of the socioeconomic status of students’ family, such as the annual household income and educational attainment of parents.

Third, in urban areas, there were various information channels for students that led to higher perceived returns to college education in statistically significant ways. On the other hand, there were no significant routes of acquiring information for students in rural areas other than pamphlets issued by colleges.

JEL Classification Codes: D84, J24, I21, I24
Keywords: perceived returns, information, wage premium, urban-rural gaps in educational attainment, multilevel analysis

Cost Effect of Mergers: Public Hospitals in Japan

By Daiya ISOGAWA, Hiroshi OHASHI and Sahoko FURUTA

Abstract

This paper evaluates the efficiency effects of mergers of public hospitals in Japan, using panel data and controlling for the possible endogeneity of mergers. The estimation results show that hospital mergers reduced hospitals’ operating costs by 20%. Labor costs for hospital staff and material costs play the largest role in this efficiency improvement. The mergers reduced the number of doctors and administrative staff, which contributed to the reduction in labor costs. On the other hand, the number of nurses and technical staff, both of whom are often regarded as substitutes for doctors, increased. At the same time, the average salaries of doctors working at the merged public hospitals increased. Taken together, these results suggest that experienced doctors remained at the merged hospitals, while less experienced ones were replaced by other hospital staff. The estimation results also indicate that the average size of hospitals shrunk upon being merged.

JEL Classification Codes: G34, I18, L11
Keywords: Hospital mergers, efficiency effects, panel data

ESRI International Conference 2019
"AI, Robotics and Labor Market"

Economic and Social Research Institute