ESRI Discussion Paper Series No.346
Dissaving by the elderly in Japan: Empirical evidence from survey data

Keiko Murata
Professor, Graduate School of Social Sciences, Tokyo Metropolitan University, Japan
Visiting Fellow, Economic and Social Research Institute, Cabinet Office, Japan

Abstract

Using two micro datasets of household surveys, this study empirically examines the (dis)saving behavior of the elderly in Japan. Using the long-run dataset covering 20 years, the findings indicate that, on average, the elderly in Japan dissave, but the pace of dissaving of retired elderly appears to be excessively slow in light of the standard life cycle-permanent income hypothesis. The analysis suggests that one likely factor is the desire to leave a bequest. Both the saving rate and the pace of wealth decumulation show that retired households dissave more slowly if the head plans to leave a bequest. Retired elderly who intend to have savings for precautionary purposes are not found to dissave more slowly except for those who do not plan to leave a bequest to their children.


Structure of the whole text(PDF-Format 1 File)

  1. Full TextI will open in a new window(PDF-Format 379 KB)
  2. page1
    Abstract
  3. page2
    1. Introduction
  4. page4
    2. Data description
    1. page5
      2.1 Family Income and Expenditure Survey (FIES)
    2. page6
      2.2 Japanese Study of Aging and Retirement (JSTAR)
  5. page7
    3. Descriptive analysis based on the FIES
    1. page8
      3.1 Age consumption/income profiles
    2. page8
      3.2.Age-wealth profiles
    3. page9
      3.3 Do the elderly consume too little and save too much during retirement?
  6. page10
    4. Determinants of dissaving by elderly households
    1. page10
      4.1 Empirical methodology
    2. page12
      4.2 Regression results
    3. page13
      4.3 Discussion
  7. page15
    5. Concluding remarks and remaining issues
  8. page17
    References:
  9. page19
    Tables
    1. page19
      Table 1. Sample statistics
    2. page20
      Table 2. Basic statistics for regression
    3. page20
      Table 3. Determinants of saving rate of elderly households
    4. page21
      Table 4. Determinants of saving rate for elderly households
    5. page21
      Table 5(a). Consumption level of retired elderly
    6. page21
      Table 5(b). Consumption level of retired elderly: By planning to leave a bequest and precautionary saving
    7. page22
      Table 6. Determinants of the ratio of expected life-time wealth to expenditures of retired elderly
  10. page23
    Figures
    1. page23
      Figure 1. Age profiles of the saving rate, income, and consumption
    2. page23
      Figure 2. Income and consumption of aged households by employment status
    3. page24
      Figure 3. Age-wealth profiles by cohort (Median values)
    4. page24
      Figure 4. Elderly households’ consumption level
  11. page25
    Appendix
    1. page25
      Appendix 1. JSTAR data and missing-data imputation
    2. page25
      Appendix 2. Regression results using other specifications
      1. page26
        Appendix Table 1. Determinants of the saving rate of elderly households
      2. page27
        Appendix Table 2. Determinants of the saving behavior of elderly households
  • 1-6-1 Nagata-cho, Chiyoda-ku, Tokyo 100-8914, Japan.
    Tel: +81-3-5253-2111