پیامدهای بهداشتی از عدم تأمین اجتماعی: شواهد از بحران حقوق بازنشستگی روسیه

We explore whether health is vulnerable to income shocks, focusing on a recent crisis in Russia during which many pensioners were not paid for an extended period of time. Using panel data spanning the crisis, we find that among affected pensioners, poverty rates doubled, and the intake of calories and protein and the use of health services and medications declined significantly. These pensioners were also 5 percent more likely to die in the two years following the crisis. However, households responded in ways that mitigated the impact of the crisis, replacing one-fifth of lost pension income through increased labor supply and asset sales.

Throughout the world, households face income risks and volatility, for example due to job loss and unemployment, or unanticipated health shocks. There has been a great deal of research on the ability to smooth consumption in the face of income fluctuations, and the mechanisms through which this is accomplished.1 An equally important concern is whether there are health consequences associated with imperfect consumption smoothing and incomplete insurance (see Strauss and Thomas, 1995, for a review). Understanding the health consequences is important not only because health is as important an indicator of well-being as consumption, but also because of the potential for ‘vicious cycles,’ wherein economic shocks leads to worse health, which in turn lead to diminished earnings capacity or even another income shock, such as job loss. Ironically, in many parts of the world, public social insurance programs themselves are increasingly becoming sources of volatility at the household level. For example, many countries face concerns over the stability of their pension systems, especially due to population aging, system financing or management. Further, there have recently been several cases of pension crises (for example, Argentina, China, Poland and Russia), resulting either in extended periods of non-payment of benefits, or the erosion of entitlements to below subsistence levels. This instability can be expected to directly affect the well-being of the elderly, who are typically highly dependent on pensions as their largest, or only, source of income. In this paper, we focus on the welfare and health consequences of the failure of the Russian pension system in 1996, and what strategies households employed to cope with the loss of income. Prior to 1996, Russia’s generous state pension system was fairly stable; pension fund revenues, collected primarily through payroll taxes, were broadly sufficient to cover outlays. However, poor economic performance, weakened tax enforcement and a wave of tax amnesties and exemptions leading up to the Presidential elections in 1996 resulted in a sharp decline in payroll tax collection, compromising the financial stability of the fund. As a result, approximately 14 million of the 39 million pensioners underwent a sudden, prolonged period during which they did not receive any payments (‘in arrears’). Our interest focuses around two major themes. The first is whether the loss of the pension adversely affected health. Given the great dependency of the elderly on the pension system, one might expect that a pension failure would lead to worsened health, especially because of the diminished ability to purchase the inputs into health (for example, nutrition, medication and the use of medical services). In this spirit, our work adds to the growing literature concerned with the impacts of socioeconomic status on health.2 An advantage over previous studies is that we exploit an exogenous income shock, so the direction of causality is clear, and directly examine the use of specific health inputs which could lead to changes in health outcomes. Other studies have found positive benefits of exogenous increases in income; for example, in a ‘reverse experiment’ of the present case, Case (2001) and Duflo (2000a) find that increases in pensions in South Africa lead to improvements in health. For Russia specifically, this study also provides an important link for understanding the dramatic decline in health that followed economic transition and the subsequent economic decline. Russia in the 1990s experienced one of the largest peacetime (non-epidemic) declines in health in human history (McKee, 1999; Brainerd and Varavikova, 2001). This paper fills an important gap, because while the health and economic crises have separately received considerable attention from researchers, the links between the two are less well explored. The second major theme we explore is the ability of households to respond to the loss of pension income. When an economic shock causes a decline in income, there are a variety of strategies that households can employ to maintain consumption, including changes in labor supply, adjustments to the stock of savings, credit and borrowing, or relying on private cash transfers from family and friends. The essence of the empirical strategy is to compare the outcome variables of interest (health and nutrition, labor supply, etc.) for households who were not paid their pensions with suitable ‘control’ or comparison groups of pensioner households that were. The strategy relies on two sources of variation in the incidence of pension arrears to identify the effects. The first is created by the regionally-decentralized structure of the pension system. Each of Russia’s 89 administrative regions or oblasts are independently responsible for collecting taxes and making payments to pensioners. Since no given oblast has an exact match between receipts and entitlements, the system relies on the redistribution of surplus revenues from regions with a high tax base relative to pension entitlements to those with a lower tax base. When payroll tax collection declined for both ‘debtor’ and ‘donor’ regions, the fiscally weak regions had lower revenues of their own and received reduced surpluses from the donors. Accordingly, while nearly all pensioners in wealthier regions continued to be paid, between one-third and one-half of pensioners in debtor regions experienced arrears. We also exploit variation within debtor regions created by program rules which establish priorities for which pensioners should be paid in the event of revenue shortfall. We use panel data from the Russian Longitudinal Monitoring Survey (RLMS), a nationally-representative survey of approximately 4000 households, covering the period before and after the pension crisis. 3 The data are well-suited to the present purposes, containing detailed information on income and expenditures. There is also a 24-hour food intake diary for all individuals, which when matched with food-conversion formulae, provide measures of consumption of specific nutrients such as calories and protein. We find that the pension crisis had a large impact on living standards, with income declining by one-third for arrears households, and poverty rates tripling to over 50 percent. There was also a significant decline in the purchase of inputs into health; daily intake of both calories and protein declined on average by 10 percent per person, and the use of medication and visits to doctors declined as well. Finally, men in arrears households were 5 percent more likely to die in the two years following the crisis, an increase in mortality risk comparable to being a smoker. These dramatic consequences arise despite the fact that households were able to respond in ways that mitigated the impacts of the shock. Households were able to replace on average approximately 20 percent of the lost pension income through greater working hours, asset sales, and borrowing. They also substantially reduced private transfers to other individuals. The remainder of the paper proceeds as follows. Section 2 discusses the Russian pension system and describes the crisis of 1996. We devote significant attention to the structure of the system and causes of the crisis, since they form an important part of the identification strategy for the empirical analysis. Section 3 examines the impact of pension arrears on living standards and poverty. Section 4 discusses the empirical strategy, identification issues and presents the results of the effects of the pension on health inputs and outcomes. Section 5 tests for evidence of coping mechanisms, while Section 6 concludes.

This paper has shown that Russia’s financially unstable pension system, characterized by irregular payments and adjustments in pension levels, can take a large toll on the financial and physical well-being of the elderly and their dependents. There are important implications for the numerous countries where pension systems are proving increasingly unstable. In cases where reform of the pension system become necessary, it will be important to pay attention to ensuring that households are able to cope during the transition phase, including efforts to identify and design alternate safety nets for the most vulnerable households. While there were significant adverse consequences of pension loss, households responded to the income shock in rational ways that are consistent with economic theory. Between increased hours of work and sale of assets, households replaced on average almost 20 percent of the lost pension income. In the absence of such adjustments, the impacts on health most likely would have been worse. However, households where the pensioner is not able to work, or where there are no other co-resident members who can work, and where there are few assets, are likely to be hurt more by the loss of an income source. In the longer run, households may be better able to cope because they have more time to identify potential responses and adjust, or they may be worse-off because some coping strategies soon become strained themselves (eg, assets and savings eliminated). Institutions which enable individuals to smooth consumption and self-insure, such as secure financial instruments, could prove valuable for assisting households in coping with economic shocks.