Abstract
The number of personal bankruptcies has increased dramatically since 1990, and a growing number of filers are divorced. While previous research shows that divorce significantly increases the probability of bankruptcy, these studies assume divorce is exogenous. This study uses the Panel Study of Income Dynamics to investigate the relationship between divorce and bankruptcy. Single-equation probit results show that divorce significantly increases the probability of bankruptcy and bankruptcy significantly increases the probability of divorce. However, after controlling for endogeneity, the effect of divorce on bankruptcy and the effect of bankruptcy on divorce both fall by a significant amount and are statistically insignificant. The findings suggest that future research needs to more carefully model the role that financial distress plays within a marriage.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
One might question why gender is not included in the model. In the PSID, household heads are coded as the male for married households. A divorced household can be headed by either a male or female. The only time we observe a female-headed household in our sample is if that person is divorced. We never observe a female-headed household that is married. Thus, if we have a female-headed household in our sample, we know that she is divorced, and there is no independent variation between marital status and gender. In other words, gender is a proxy for marital status.
To calculate net financial benefit, we follow the same methodology as Fay et al. (2002) using state exemption levels and wealth and debt information included in the 1989 and 1994 PSID supplements. The data used to calculate state exemption levels can be found in Elias, Renauer, and Leonard (1995) and other editions from this series, which began in 1989.
Because most households respond to the PSID between March and June of the survey year, a household could file for bankruptcy in the summer of 1996 but not be classified as such in the 1996 wave. Consequently, the last year in our sample is 1995. Also, since we only have data on state exemption levels from 1989 on, our sample begins in 1989.
Using a series of repeated cross sections follows the existing bankruptcy research, which treats multiple observations of the same households as individual observations (Elul & Subramanian, 2002; Fay et al., 2002). Allison (1995) finds that this treatment of the data does not result in biased estimates or inflated test statistics.
The national filing rate from 1989 to 1995 for all households, regardless of marital status, was on average 0.83%. For all households in the PSID, the average filing rate was 0.38% suggesting that bankruptcy is underreported in the PSID. According to Hausman, Abrevaya, and Scott-Morton (1998), this underreporting of bankruptcy may lead to a slight downward bias in our estimated coefficients (see also Fay et al., 2002).
The percentage of households classified as divorced in our sample may seem low. Recall, however, that our sample includes married households and married households that eventually divorce. Also, the PSID usually follows only one spouse after a divorce. The PSID follows the person, male or female, that was selected to be part of the PSID study. For example, suppose the PSID selected a single woman to be part of the survey. If the woman subsequently marries, the PSID would survey both the woman and man. If she then divorces, the PSID would continue following the woman but not her ex-husband. If the PSID started with a married couple in the 1968 survey and the couple divorced, the PSID would follow both the man and woman. To account for this and other factors, the PSID uses sample weights to ensure that the sample is representative of the population as a whole.
In results not presented in this paper, we tested the robustness of our results to changing definitions of divorce. For example, we defined households as divorced if they divorced within the last year (i.e., divorced in 1991 and 1992 if they divorced in 1991). This decreases the number of divorced households by a third. For this reason, we opted to expand the number in our divorced sample to include those that divorced within the last two years. The results are robust to this change in the definition of divorce. Prior specifications of the model are available upon request.
Descriptive statistics are presented for household years rather than for the sample of unique households. There is little variation in means between the two samples.
The lagged bankruptcy rate should not affect whether a household is recently divorced unless there are shocks that drive both the divorce and bankruptcy processes. The results remain unchanged when the lagged bankruptcy rate is excluded from the model.
Like the lagged bankruptcy rate, the state divorce rate should not affect whether a household files for bankruptcy except to the extent that there are shocks that affect both divorce and bankruptcy.
It is not uncommon in the literature to find models that control for a change in income or a negative shock to income. For example, Fay et al. (2002) included a dummy variable that equaled one if the household’s income decreased. However, the coefficient on this variable was negative and insignificant. Other researchers have found similar results. In early analyses, we also used different specifications for income. However, the coefficients for these income variables were always statistically insignificant. For this reason, we did not include these specifications in the paper. The results are available upon request.
We also ran the results just using the actual wealth and debt data from the 1989 and 1994 PSID. The results are similar and the conclusions are identical. These results are not shown but available upon request.
According to Wooldridge (2002), large standard errors may also be caused by weak instruments in the first stage of estimation. To test the strength of our instruments, we conducted a joint test of the instruments in each equation and include the F-statistics from this test in Table A1. We find that the F-statistics are large with p-values less than 0.01, providing evidence that our instruments have sufficient explanatory power.
Because 19 states have no bankruptcy filers in our sample, we could not use state dummy variables with our full sample; living in one of those nineteen states perfectly predicts not filing for bankruptcy. In the results that include state dummy variables, we lose 1533 observations that lived in one of the 19 states without a bankruptcy filer.
We bootstrap the standard errors in the 2SLS, because of the sampling variation in the coefficients on the predicted values.
References
Allison, P. (1995). Survival analysis using the sas system, a practical guide Cary, NC: SAS Institute.
American , Bankruptcy Institute. (2004). U.S. Bankruptcy Fillings, 1980–2003. ABI World.
Attewell , P. (1999). The impact of family on job displacement and recovery. ANNALS, AAPSS, 562, 66–82.
Becker, G. S. (1973). A theory of marriage: Part I. Journal of Political Economy, 81(4), 813–846.
Becker, G. S., Landes, E. M., & Michael, R. T. (1977). An economic analysis of marital instability. Journal of Political Economy, 85(6), 1141–1188.
Black, S., & Morgan, D. (1999). Meet the new borrowers. Current Issues in Economics and Finance 5.
Caplovitz, D. (1974). Consumers in trouble: A study of debtors in default New York: Free Press.
Caplovitz, D. (1979). Making ends meet: How families cope with inflation and recession Beverly Hills, CA: Sage Press.
Del Boca, D. (2003). Mothers, fathers, and children after divorce: The role of institutions. Journal of Population Economics, 16(3), 399–422.
Del Boca, D., & Rocia, R. (2001). The effect of child-support policies on visitations and transfers. American Economic Review Papers and Proceedings, 91(2), 130–134.
Duncan, G., & Hoffman, S. (1985). A reconsideration of the economic consequences of marital dissolution. Demography, 22(4), 485–497.
Duncan, G., & Hoffman, S. (1988). What are the economic consequences of divorce? Demography, 25, 641–645.
Elias, S., Renauer, A., & Leonard, R. (1995). How to file for bankruptcy 5th ed. Berkeley, CA: Nolo Press.
Elul, R., & Subramanian, N. (2002). Forum-shopping and personal bankruptcy. Journal of Financial Services Research, 21(3), 233–255.
Fay, S., Hurst, E., & White, M. J. (2002). The household bankruptcy decision. American Economic Review, 92(3), 706–718.
Fisher, J. (2004). Marital status tenure and the decision to file for personal bankruptcy. Journal of Economics and Finance, 28(3), 348–360.
Fisher, J., & Angela, L. (2005). The ability of women to repay debt after divorce. Journal of Women, Politics, and Policy, 27(3–4), 161–168.
Hausman, J. (1983). Simultaneous equation models. In: Z. Griliches & M. Intriligator (Eds). Handbook of econometrics, Vol. I. Amsterdam: North-Holland.
Hausman, J. A., Abrevaya, J., & Scott-Morton, F. M. (1998). Misclassification of a dependent variable in a discrete-response setting. Journal of Econometrics, 87(2), 239–269.
Hoffman, S. D., & Duncan, G. J. (1995). The effect of incomes, wages, and afdc benefits on marital disruption. The Journal of Human Resources, 30(1), 19–41.
Johnson, W. R., & Skinner, J. (1986). Labor supply and marital separation. American Economic Review, 76(3), 455–469.
Lyons, A. C. (2003). How credit access has changed for divorced men and women: Evidence from the survey of consumer finances. University of Illinois, working paper.
Maddala, G. (1983). Limited-dependent and qualitative variables in econometrics New York: Cambridge University Press.
Moulton, B. R. (1990). An illustration of a pitfall in estimating the effects of aggregate variables on micro units. Review of Economics and Statistics, 72(2), 334–338.
Page, M., & Ann Stevens, H. (2004). The economic consequences of absent parents. Journal of Human Resources, 39(1), 80–107.
Peters, H. E. (1986). Marriage and divorce: Informational constraints and private contracting. American Economic Review, 76, 437–454.
Smock, P. (1993). The economic costs of marital disruption for young women over the past two decades. Demography, 30, 353–371.
Stavins, J. (2000). Credit card borrowing, delinquency, and personal bankruptcy. New England Economic Review, July/August, 15–30.
Sullivan, T. A., & Warren, E. (1999). Women in bankruptcy. American Bankruptcy Institute. http://www.abiworld.org/research/womenbank.html.
Sullivan, T. A., Warren, E., & Westbrook, J. L. (1995). Bankruptcy and the family. Marriage and Family Review, 21(3/4), 193–215.
Sullivan, T. A., Warren, E., & Westbrook, J. L. (2000). The fragile middle class New Haven: Yale University Press.
Wooldridge, J. M. (2002). Econometric analysis of cross section and panel data Cambridge, MA: MIT Press.
Van der Klaauw, W. (1996). Female labor supply and marital status decisions: A life-cycle model. Review of Economic Studies, 63(2), 199–235.
Zagorsky, J. (2005). How do marriage and divorce impact wealth? Journal of Sociology, 41(4), 406–424.
Author information
Authors and Affiliations
Corresponding author
Additional information
The authors would like to thank Julie Cullen and Kristin Kleinjans for helpful comments as well as participants of the 2004 MEA session on the “Economics of Marriage and Divorce.” All views expressed in this paper are those of the authors and do not reflect the views or policies of the Bureau of Labor Statistics (BLS) or the views of other BLS staff members.
Appendix
Appendix
Rights and permissions
About this article
Cite this article
Fisher, J.D., Lyons, A.C. Till Debt do us Part: A Model of Divorce and Personal Bankruptcy. Rev Econ Household 4, 35–52 (2006). https://doi.org/10.1007/s11150-005-6696-0
Received:
Accepted:
Issue Date:
DOI: https://doi.org/10.1007/s11150-005-6696-0