Abstract
This paper examines the impact of idiosyncratic income shocks on household consumption, educational expenditure and fertility in Indonesia, and assesses whether the investment in human capital of children and fertility are used to smooth household consumption. Using four different kinds of self-reported economic hardships, our findings indicate that coping mechanisms are rather efficient for Indonesian households that perceive an economic hardship. Only in case of unemployment do we find a significant decrease in consumption spending and educational expenditure while fertility increases. These results indicate that households that perceive an unemployment shock use children as a means for smoothing consumption. Regarding the death of a household member or natural disaster we find that consumption per person even increases. These results are consistent with the argument that coping mechanisms even over-compensate the actual consumption loss due to an economic hardship. One important lesson from our findings is that different types of income shock may lead to different economic and demographic behavioral adjustments and therefore require specific targeted social insurance programs.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Public expenditure on social security and welfare includes compensation for loss of income to the sick and temporarily disabled, payments to the elderly, the permanently disabled, and the unemployed; family, maternity, and child allowances; and the cost of welfare services, such as care of the aged, the disabled, and children. It excludes expenditures on important safety net categories, including subsidies and public work program costs. The per capita GDP is in terms of the constant 1995 US dollars.
This analysis follows the study of Grimm (2009).
It is worth noting that the numbers in Table 1 are slightly different from those in Table 1 in Grimm (2009), which was also based on IFLS1, IFLS2 and IFLS3. Grimm (2009) reports that the share of households with a death of any member is 10% in 1993 and 1997 and 9% in 2000. The difference is likely to do with the way the final sample was constructed. The balanced panel used in Grimm (2009) has 6,303 households, whereas the sample size in this study is 5,138. The composition of the responses and their median costs in the two tables are similar to each other. One exception is the median yearly income of the deceased (and that of those who worked), which is zero in Grimm (2009) and 225 thousand rupiah here.
The ‘efficiency’ refers to a case where a household has the same level of consumption in each period.
We use log of consumption rather than consumption itself in the statistical analysis in order to minimize the influence of outliers.
Grimm (2009) included 13 province dummies and rural/urban areas as additional control variables.
It is possible that the consumption per person increases mainly due to having one less people to feed in a household after death. In order to check out this possibility, we ran the same regression with total household expenditure as a dependent variable. It is found that the total household expenditure also increases significantly after a member’s death.
In Grimm (2009) three different reactions as a consequence of death of a household member have been tested for their statistical significance. The results indicated that households indeed deplete assets and increase labor supply while no conclusive results were found for variations in transfers.
Considering that the conception takes place nine months before the actual birth, the births in the following period is used as a measure of conception of the current period.
Using the total educational expenditure rather than the per-capita term in column (2) of Table 11, we found qualitatively the same result. Another alternative is to use the educational expenditure per school child. However, since the educational expenditure is not asked separately for pre-school children, school children and adults in the IFLS, it is not possible to measure the per-school-child educational expenditure.
References
Alem, M., & Townsend, R. M. (2003). An evaluation of safety nets and financial institutions in crisis and growth. University of Chicago, Working Paper.
Cameron, L. A., & Worswick, C. (2001). Education expenditure response to crop loss in Indonesia: A gender bias. Economic Development and Cultural Change, 49(2), 351–363.
Cameron, L. A., & Worswick, C. (2003). The labor market as a smoothing device: Labor supply responses to crop loss. Review of Development Economics, 7(2), 327–341.
Chetty, R., & Looney, A. (2007). Income risk and the benefits of social insurance: Evidence from Indonesia and the United States. In T. Ito & A. Rose (Eds.), Fiscal policy and management in East Asia: NBER East Asia seminar on economics 16. Chicago: University of Chicago Press.
Duryea, S., Lam, D., & Levison, D. (2007). Effects of economic shocks on children’s employment and schooling in Brazil. Journal of Development Economics, 84, 188–214.
Foster, A., & Roy, N. (1997). The dynamics of education and fertility: Evidence from a family planning experiment. University of Pennsylvania, Working Paper.
Frankenberg, E., & Karoly L. (1995). The 1993 Indonesian Family Life Survey: Overview and field report. Santa Monica, CA: RAND.
Frankenberg, E., McKee, D., & Thomas, D. (2005). Health consequences of forest fires in Indonesia. Demography, 42(1), 109–129.
Frankenberg, E., & Thomas, D. (2000). The Indonesia Family Life Survey (IFLS): Study design and results from waves 1 and 2. Santa Monica, CA: RAND. DRU-2238/1-NIA/NICHD.
Grimm, M. (2009), Mortality and survivors’ consumption growth. Oxford Bulletin of Economics and Statistics (forthcoming).
Pitt, M., & Rosenzweig, M. (1986). Agricultural prices, food consumption and the health and the productivity of Indonesian farmers. In I. Singh, L. Squire, & J. Strauss (Eds.), Agricultural household models: Extensions, applications and policy (pp. 153–182). Baltimore: Johns Hopkins University Press.
Schultz, T. P. (1997). Demand for children in low income countries. In M. Rosenzweig & O. Stark (Eds.), Handbook of population and family economics (pp. 349–430). Amsterdam: Elsevier Science.
Skoufias, E. (2003). Economic crises and natural disasters: Coping strategies and policy implications. World Development, 31(7), 1087–1102.
Strauss, J., Beegle, K., Sikoki, B., Dwiyanto, A., Herawati, Y., & Witoelar, F. (2004). The third wave of the Indonesia Family Life Survey (IFLS): Overview and field report. WR-144/1-NIA/NICHD.
Townsend, R. M. (1994). Risk and insurance in village India. Econometrica, 62(3), 539–591.
Acknowledgements
The authors thank Steve Pudney, Arnstein Aassve, Fabrizia Mealli and participants at the International Conference on “Population and Development in Asia: Critical Issues for a Sustainable Future” in Phuket, Thailand in March 2006 for their helpful comments on an earlier draft of this paper. The project is funded by the Austrian Science Foundation (Contract No. P16903-605).
Author information
Authors and Affiliations
Corresponding author
Appendix
Appendix
The final sample in the analysis includes 5,136 households with no missing values for the relevant variables over all three waves (i.e. in 1993, 1997 and 2000). The summary statistics of pooled observations are presented in Table 13. All the expenditures are deflated by the CPI index by province by year so that they denote the values in Jakarta in 1993. The top and bottom 1% of the total consumption expenditure per person distribution and those of the change over two periods in expenditure distribution are trimmed in order to remove extreme values. The mean food and non-food expenditure per person is 44,386.2 rupiah, which is about 90% of the total expenditure. The mean per capita educational expenditure is 5,124.5 rupiah. The number of children born over the past five years is 0.43 assuming that those children were present at the time of survey. The average household size is 4.5 members, among whom about one-third is under age 15 and about one tenth is above age 59.
As discussed before, there are six variables indicating the experience of economic hardship over the past five years. In the pooled sample over three waves, the proportion of households that experienced a death or sickness of any member is 10% and 13%, respectively. About 12% of households experienced income loss due to crop loss, while only 2% of households were hit by natural disaster. The unemployment and income loss due to price fall were reported to be an economic shock by 4% and 7%, respectively. In terms of household head’s characteristics, 88% of them are Muslim, and 17% of them are female. On average, household head is 48 years old, and has 5.25 years of schooling. About 58% of households reside in rural area in the sample period. The bottom half of Table 13 shows the descriptive statistics of the differenced variables over two adjacent periods.
Rights and permissions
About this article
Cite this article
Kim, J., Prskawetz, A. External Shocks, Household Consumption and Fertility in Indonesia. Popul Res Policy Rev 29, 503–526 (2010). https://doi.org/10.1007/s11113-009-9157-2
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11113-009-9157-2