This is a preview of subscription content, log in via an institution.
Notes
- 1.
For all monetary variables, we adjusted prices for inflation by using the Consumer Price Index to take into consideration the effects of commodity prices. Further, we obtained the per capita amount of senior welfare by dividing the expenditure by the size of population aged 65 years and over and the per capita amount of child welfare by dividing the expenditure by the size of population aged 15 years and below.
- 2.
The base amount of income assistance in each area is decided by the Minister of Health, Labour and Welfare; the municipalities are not allowed to set the amount on their own. However, because the base amount is sometimes revised by a municipality at different times, the expenditure per person for income assistance (obtained by dividing the total amount of expenditure for livelihood assistance by the total population of the prefecture) can change. At the same time, the decision as to whether income assistance for those who are in poverty and who have applied for such assistance should be provided is primarily left up to the municipalities or welfare offices. Therefore, it is possible that the expenditure per person for income assistance changes depending on the payment rate of income assistance. If the operating organization sets the payment rate higher, then the expenditure per person for income assistance should increase, whereas if the rate is set lower, then it should decrease. The effect of the payment rate itself is controlled for separately as the ratio of recipients of income assistance, as discussed later.
- 3.
According to the police data discussed earlier, of 15,802 suicide victims whose reasons and motive have been clarified in the year 2010, 7,020 people died by suicide due to depression and 2,637 people due to schizophrenia or other mental disorders.
- 4.
In terms of the annual income average of decile groups of all households, we used the quinquennial data from 1979 onward. We calculated the values for 2005 and 2006 based on the trends from 1999 and 2004. For other variables, we used the quinquennial data from 1980 onward. We used values calculated based on the trends from 2000 to 2005 as the data for 2006.
- 5.
We must be careful when interpreting the variable that may accompany sudden fluctuations in circumstance, such as household income and the number of single-female-parent households. To see what effects the interpolated data had on the estimated results, we first re-estimated the model after removing the interpolated variables from the model. Next, we re-estimated the model by removing the data for 2005 and 2006. Lastly, for the overall unemployment rate, we re-estimated the model by using the overall unemployment rates in 10 areas calculated by the Labor Force Survey and aggregating the other data by area (1983–2006). Because we were able to obtain results similar to those presented in Chap. 3 for all the additional analyses, we believe that the analysis results presented in this chapter are robust.
- 6.
The data were provided by the Statistical Information Institute for Consulting and Analysis as data for purchase. We referred to the National Economic Accounting published by the Cabinet Office for income of prefectural residents prior to 1996.
- 7.
Because the amount of public investments is not an exogenous variable, it is difficult to discuss the causal relationship in the estimation using the fixed effect model. We also performed an additional analysis using the GMM estimator to estimate the dynamic panel models by Arellano and Bond (1991), but we obtained results similar to those presented in Tables 2 and 3. Upon doing so, we inserted the suicide rate and the amount of public investments from a quarter before as well as all the variables included in the estimated equation (except for the amount of public investments) into the right side of the equation and used the data from two quarters before as manipulated variables. In this estimation, we only used the data from 1984 to 2006, because there are some missing values in the suicide rates of the year 1981 in some prefectures. The estimation was made using the xtabond2 of Stata through the one-step method.
References
Arellano, M., & Bond, S. (1991). Some tests of specification for panel data: monte carlo evidence and an application to employment equations. Review of Economic Studies, 58, 277–297.
Kondo H. (2008). Syakaishihonseibi ni okeru seijikeizaigakuteki sokumen. Financial Revue 2008, 68-92.
Levin A., Chien-Fu L., & Chia-Shang J. C. (2002). Unit Root Tests in Panel Data: Asymptotic and Finite-Sample Properties. Journal of Econometrics, 108, 1–24.
The Statistics Bureau, Ministry of Internal Affairs and Communications, the Government of Japan. (Various years). System of Social and Demographic Statistics of Japan, the basic data for prefectures (1975 to 2008).
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Copyright information
© 2017 Springer Nature Singapore Pte Ltd.
About this chapter
Cite this chapter
Sawada, Y., Ueda, M., Matsubayashi, T. (2017). Economic and Welfare Policies and Suicide. In: Economic Analysis of Suicide Prevention. Economy and Social Inclusion. Springer, Singapore. https://doi.org/10.1007/978-981-10-1500-7_8
Download citation
DOI: https://doi.org/10.1007/978-981-10-1500-7_8
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-10-1499-4
Online ISBN: 978-981-10-1500-7
eBook Packages: Economics and FinanceEconomics and Finance (R0)