The Analysis on Disparities of Fertility Rate of Japanese Municipalities

There are many disparities on fertility rate regarding the Japanese municipalities. In other words, while some municipalities have a high fertility rate, however, the fertility rate of others is very low. Disparity of fertility rate in municipalities is wider than in prefectures. In addition, as to the background of active discussion about vanishing rural regions in recent years, there is the fact that urban areas, including the Tokyo metropolitan area, have a low fertility rate, relatively. The purpose of this paper is to explore the factors underlying the regional disparity in the fertility rate and examine the relationship between the fertility rate and the population density. Although there are many complicated factors of disparity of fertility rate by regions, it could be said that the low fertility rate in urban areas is occurring due to compatibility with child rearing and female labor supply, resources for child rearing, and housing space, etc. Direct cost of having children and the possibility of compatibility with child rearing and female labor supply are affected by the location of residence. We treat population density as a proxy variable of these factors. Until now, there have not been many theoretical studies which examine the relationship between fertility rate and population density, and few empirical studies exist with respect to Japan. Therefore, premising that the optimal number of children in rural areas is more than in urban areas from the conclusion of our model under certain conditions, we measured the relationship between them in municipalities in Japan. As for total fertility rate by municipalities, we used the Specified Report of Vital Statistics prepared by the Ministry of Health, Labour and Welfare, which adopts Bayesian estimation to obtain stable data for small regions. Firstly, we examined whether population density or other socio - economic variables have affected the statistically significant effect on total fertility rate by municipalities using a sample of 1,890 which is from the above statistics. As a result, it was observed that municipalities with higher population density show a lower total fertility rate, and municipalities with a higher female labor participation rate have a higher total fertility rate. In addition, we obtained the result that a higher net immigration rate brought a higher fertility rate in municipalities. Furthermore, when we measured the relation between fertility rate and variables related to policies to improve low fertility such as capacity of day care centers or expenditure for child welfare using 1,742 samples for which data was available, we obtained positive coefficients that were statistically significant. It should be noted that it was difficult to explain the causal effect between them because we used cross - sectional data at one point Policy Research Institute, Ministry of Finance, Japan, Public Policy Review, Vol.14, No.1, February 2018 1 in time. Next, we measured changes of total fertility rate from 2005 to 2010. To compare total fertility rate in both years by municipalities, we should select municipalities whose administrative areas did not change during the period due to municipal mergers. Among 1,601 municipalities which satisfy the above criterion, total fertility rate has increased in 2010 compared to in 2005 at 1,187 municipalities; on the other hand, total fertility rate has decreased in 352 municipalities. Based on these data and municipalities, we regressed total fertility rate in 2010 on total fertility rate in 2005, and estimated the influence of population density to total fertility rate using panel data at two points in time. From these regression results, we obtained a negative coefficient of population density and positive coefficient of female labor participation rate in all cases. We also estimated how a rise in the population density affected the fertility rate by comparing municipalities whose fertility rate increased between 2005 and 2010 and those whose fertility rate did not rise. As a result, we discovered the fact that although the average total fertility rate in municipalities whose population density increased declined 0.091 points, the decline was 0.05 points if the average causal effect was taken into consideration.