|
|
|
The Effect of Schooling, Wages, Marriage, and Socio-Economic Circumstances on Fertility Behavior in Russia (Job Market Paper)
Abstract:
The total fertility rate in Russia has been falling over the past several decades from above the replacement fertility level in the early 1960’s (2.42 children) to significantly below the ‘safety zone’ in 2000 (1.20 children). The low fertility rate is accompanied by the highest death rate among all countries with at least moderate development. These trends are reflected in a projected 30% decline in the Russian population by 2053 (World Bank report 2005). Even though fertility issues have received much attention in the economic literature, the massive fertility reduction in the transition countries of Central and Eastern Europe is only now becoming a subject of micro-economic analysis. Using the Russia Longitudinal Monitoring Survey (1994-2006), I estimate a comprehensive model of female life-cycle behavior, which accounts explicitly for the interdependence of annual reproductive choices, educational, employment, and marriage decisions as well as earnings outcomes and controls for individual- and community-level heterogeneity. Modeling these fertility-related outcomes jointly allows for a correction for the potential endogeneity of education, employment, and marriage choices in the conception equation arising from the existence of unobserved individual or community characteristics shaping all fertility related choices. The results demonstrate the importance of correction for endogeneity. My findings also indicate that fertility policies should be directed away from supplementation of non-labor income, as it is observed now, and toward macro-stabilization efforts and the reconciliation of the incompatibility of the career demands of the new market system with the requirements of motherhood.
The Effect of the Social Security Earnings Test on the Earnings Distribution and Hours of Work (Working Paper)
Abstract:
In this paper, I analyze the effect of the Social Security earnings test on the labor supply of working Social Security beneficiaries, by focusing on the last two changes enacted in 1983 and 2000. These changes eliminated the earnings test entirely for individuals of age 70-71 in 1983 and of normal retirement age and older in 2000, so working beneficiaries after attaining the specified age in the law became entitled to the full amount of their Social Security benefits regardless of their wages and salaries earned. Statistical comparison of earnings distributions of the affected and unaffected groups before and after the change allows the response in earnings patterns to the alteration of the earnings test parameters rather than to other trends in labor supply. In this paper, I revisit Friedberg (2000) and extend her analysis to include women and the most recent law change enacted in 2000. Analysis of the earnings distributions, utilizing March Current Population Surveys on 62-75 year old individuals for periods of three years before and after the specified changes, indicates substantial and statistically significant clustering of working beneficiaries, both women and men, just below the exempt amount. In addition to graphical and statistical analysis of the raw data, I estimate a structural labor supply model by implementing a maximum likelihood method for piecewise-linear budget constraints. The estimation of the model yields rather large estimates of wage and income elasticities (0.371 and -0.436, respectively), which suggest high sensitivity of the hours supplied by workers to the changes in the parameters of the earnings test. Even though the earnings test is currently applicable only to the beneficiaries of age 62-66, popular attention to the earnings test will be reinforced as more baby boomers start reaching age 62, greatly increasing the number of workers affected by the earnings test. Also, the earnings test will grow more binding, as normal retirement age rises to 67.
The Study of the Effect of Child Subsidies on Reproductive Behavior in Russia (Working Paper)
Abstract:
Over the past several decades, Russia has witnessed a substantial decline in fertility rates, falling as low as 1.2 children per reproductive lifetime of an average woman in 2000. Low birth rates along with high death rates resulted in what many observers call a “demographic crisis” manifesting itself in negative population growth, decreasing life-expectancy, and a rapidly aging population. This situation raises concerns regarding the sustainability of Russian economic development. In this paper, I approach these demographic circumstances by analyzing the effect of fertility policies on reproductive behavior. Fertility-encouraging measures in Russia operate mainly through supplementation of non-labor income in the form of child subsidies. A one-time birth payment is the most substantial part of child-related allowances, followed by monthly payments for the first 18 months to allow mothers to stay at home. The real value of child allowances has fluctuated significantly across a twelve-year period starting from 1994, reflecting both law changes and high inflation. Preliminary analysis of the Russia Longitudinal Monitoring Survey (1994-2006) adds a new dimension to the study of the effectiveness of child subsidies since it indicates the presence of uncertainty in the reception of the allowances for eligible families. In particular, 66% of eligible families received child subsidies in 1994, falling to only 20% in 1998. By 2005, this proportion improves significantly and goes up to 80%. There is additional disparity in the proportion of eligible households actually receiving child-subsidies across nine major regions participating in the study. Therefore, the conventional approach to study the responsiveness of reproductive behavior to child payments through measuring sensitivity of such behavior with respect to non-labor income might not be appropriate given the uncertainty embedded in the realized subsidy amounts. This issue is addressed by explicitly modeling the probability of the reception of the child payments within the life-cycle fertility framework. In this paper, the timing of the first two conceptions is estimated jointly with other fertility altering decisions such as schooling, working, and marital status, by applying a discrete factor random effects method with controls for unobserved individual and community permanent characteristics.
|
|
|
|
|
