Do Private Household Transfers to the Elderly Respond to Public Pension Benefits?

What is it about?

In this paper, we examine the impact of China’s New Rural Pension Scheme (NRPS) on inter vivos private transfers (from children to parents), particularly among adults in the age group of 60 years and above. The NRPS program was introduced in 2009 in response to rising demographic and old-age poverty concerns in China in the last decade (Liu and Sun 2016; Holzman, Robalino and Takayama 2009: 111–18). In 2007, approximately 11 percent of China’s population was in the age group of 60 years and above, making up 21 percent of the world’s elderly population (UN 2007). Similar to other developing countries, the Chinese government faced additional pressing challenges: large rural and informal agricultural populations, high internal migration flows (Sabates-Wheeler and Koettl 2010), and weak institutions (Musalem and Ortiz 2011). The new program, a defined-contribution pension program, was made available to all rural residents who were 16 years of age or older. In this study, we examine the program’s impacts on inter vivos transfers using various quasi-experimental techniques. We use a large longitudinal dataset from which we focus on a cohort of individuals whose data on program participation and inter vivos transfers are rich. The eligibility of an individual for the program was based on his/her contributions and age. Our empirical approach to estimate the effects of the program relies on within-country variation in program implementation due to the staggered implementation of the policy across communities. We use this staggered implementation as a source of identifying variation, to detect the impact of this program on pension participation between those individuals living in newly integrated communities between 2011 and 2013 and those individuals who were not offered similar program benefits. Using data from 2009 to 2013 from the China Health and Retirement Longitudinal Study (CHARLS) and the China Health and Nutrition Survey (CHNS), we find a statistically significant impact of the availability of NRPS benefits on lowering the probability of the incidence of receiving a transfer from adult children by 7.4 percentage points and a lowering of the actual amount of transfers by 10.4 percent. We do not detect statistically significant evidence that the benefits’ availability impacted transfers sent to children. The small amount of evidence provided by previous studies in developing countries supports our findings that elderly pension benefits tend to crowd-out private transfers received. In addition, we find that the receipt of benefits had a significant effect on the intensive and extensive margins of private transfers. More specifically, NRPS benefit recipients were 10.4 percentage points (55.1 percent) less likely to receive transfers. And the receipt of benefits was associated with a decline in transfers of about 18.8 percent of the average transfer amount. Similarly, we did not find that the receipt of benefits affected private transfers sent to children. We investigate the magnitude of the crowd-out effect and we find a crowd-out effect of −0.03 (imprecisely estimated ), suggesting that family members do offset their inter vivos transfers to elderly parents in response to pension benefits, but that substitution is extremely small in magnitude and much smaller than previously estimated using data from middle- and high-income countries.

Featured Image

Photo by Ewan Yap on Unsplash

Photo by Ewan Yap on Unsplash

Why is it important?

How family members respond to public transfers can play an important role in assessing the efficiency and welfare impact of re-distributive policies on the welfare of the elderly. Of particular concern is the possibility that public transfers induce a reduction in private family transfers, thereby dampening the re-distributive effect of public transfers and resulting in an overall reduction of available savings, and thus welfare. This scenario is exceptionally relevant to developing countries and rapidly growing economies, such as China, where such crowd-out effects can affect millions of elderly people, many of whom live in poverty. China is a particularly compelling setting to examine inter-generational transfers because of the country’s high age dependency ratio, with a large elderly cohort and relatively small working-age population, due to the only recently relaxed one-child policy (World Bank 2017). The World Bank (2001) estimates that expanding formal safety nets and public transfer programs in developing countries will likely displace private transfers by 20 to 91 percent. Previous empirical studies that examine the magnitude of crowd-out effects on inter vivos private transfers provide mixed evidence or some support for positive small crowd-out effects (Lucas and Stark 1985; Cox 1987; Cox and Rank 1992; Cox and Jakubson 1995; McGarry and Schoeni 1995; Altonji et al. 1997; Cox et al. 1998), although these studies are largely based on the experience of high-income countries. Evidence on the existence and magnitude of crowd-out effects in emerging and middle-income economies, such as China, is extremely limited. We find small program effects on subsequent inter vivos transfers and these small effects imply that the introduction of pecuniary pension benefits can have additional welfare-improving consequences, contrary to the results based on previous empirical studies from other middle-income countries. Our findings have important welfare implications for the poorest households, in particular. One of the objectives of the NRPS program was to ensure the well-being of the aging population and to mitigate the risk of old-age poverty in rural areas, where the poverty rate is already very high. Although conceptually the welfare effects can be ambiguous ex ante, any empirical behavioral response and the tendency to further crowd-out familial transfers to the elderly may be of high concern in a program for which one of the objectives is to address old-age poverty. Yet, among the poorest households in our sample, we find no evidence to suggest that there is significant crowding-out of familial transfers to elderly parents as a result of increased NRPS benefits, and the estimated effect size in that sub-sample is smaller than the one based on the whole sample. Thus, among this demographic, any ex ante concerns regarding disincentives for family transfers appear to be secondary, although understanding the mechanisms behind this interplay between poverty status and strength of the inter vivos transfers merits further research. The findings reported in this paper are important, because we show that pension reform in the world’s largest country could produce pecuniary benefits for program beneficiaries without spurring a negative behavioral response from household members, as suggested by studies conducted in middle- or high-income countries. This study has concentrated on estimating the behavioral response related to inter vivos transfers to beneficiaries of the NRPS program, launched in part to mitigate the risk of old-age poverty. However, this large public program also may have had additional broader impacts on the Chinese population, its economy, and on income inequality. How the NRPS program influenced these outcomes should be the subject of future work.