Abstracts of Selected Recent NBER Working Papers

WP 14580

Loretti Dobrescu, Laurence Kotlikoff, Alberto Motta

Why Aren't Developed Countries Saving?

National saving rates differ enormously across developed countries. But these differences obscure a common trend, namely a dramatic decline over time. France and Italy, for example, saved over 17 percent of national income in 1970, but less than 7 percent in 2006. Japan saved 30 percent in 1970, but only 8 percent in 2006. And the U.S. saved 9 percent in 1970, but only 2 percent in 2006. What explains these international and intertemporal differences? Is it demographics, government spending, productivity growth or preferences? Our answer is preferences. Developed societies are placing increasing weight on the welfare of those currently alive, particularly contemporaneous older generations. This conclusion emerges from estimating two models in which society makes consumption and labor supply decisions in light of uncertainty over future government spending, productivity, and social preferences. The two models differ in terms of the nature of preference uncertainty and the extent to which current society can control future societies' spending and labor supply decisions.

WP 14624

Kevin Milligan, Mark Stabile

Do Child Tax Benefits Affect the Wellbeing of Children? Evidence from Canadian Child Benefit Expansions

A vast literature has examined the impact of family income on the health and development outcomes of children. One channel through which increased income may operate is an improvement in a family's ability to provide food, shelter, clothing, books, and other expenditure-related inputs to a child's development. In addition to this channel, many scholars have investigated the relationship between income and the psychological wellbeing of the family. By reducing stress and conflict, more income helps to foster an environment more conducive to healthy child development. In this paper, we exploit changes in child benefits in Canada to study these questions. Importantly, our approach allows us to make stronger causal inferences than has been possible with the existing, mostly correlational, evidence. Using variation in child benefits across province, time, and family type, we study outcomes spanning test scores, mental health, physical health, and deprivation measures. The findings suggest that child benefit programs in Canada had significant positive effects on test scores, as has been featured in the existing literature. However, we also find that several measures of both child and maternal mental health and well-being show marked improvement with higher child benefits. We find strong and interesting differences in the effects of benefits by sex of the child: benefits have stronger effects on educational outcomes and physical health for boys, and on mental health outcomes for girls. Our findings also provide some support for the hypothesis that income transfers operate through measures of family emotional well-being.

WP 14634

Guy David, Sara Markowitz, Seth Richards

The Effects of Pharmaceutical Marketing and Promotion on Adverse Drug Events and Regulation

This paper analyzes the relationship between postmarketing promotional activity and reporting of adverse drug events by modeling the interaction between a welfare maximizing regulator (the FDA) and a profit maximizing firm. In our analysis demand is sensitive to both promotion and regulatory interventions. Promotion-driven market expansions enhance profitability yet may involve the risk that the drug would be prescribed inappropriately, leading to adverse regulatory actions against the firm. The model exposes the effects of the current regulatory system on consumer and producer welfare. Particularly, the emphasis on safety over benefits distorts the market allocation of drugs away from some of the most appropriate users. We then empirically test the relationship between drug promotion and reporting of adverse reactions using an innovative combination of commercial data on pharmaceutical promotion and FDA data on regulatory interventions and adverse drug reactions. We provide some evidence that increased levels of promotion and advertising lead to increased reporting of adverse medical events for certain conditions.

WP 14637

Angus Deaton, Jane Fortson, Roberta Tortora

Life (Evaluation), HIV/AIDS, and Death in Africa

We use data from the Gallup World Poll and from the Demographic and Health Surveys to investigate how subjective wellbeing (SWB) is affected by mortality in sub-Saharan Africa, including mortality from HIV/AIDS. The Gallup data provide direct evidence on Africans' own emotional and evaluative responses to high levels of infection and of mortality. By comparing the effect of mortality on SWB with the effect of income on SWB, we can attach monetary values to mortality to illuminate the often controversial question of how to value life in Africa. Large fractions of the respondents in the World Poll report the mortality of an immediate family member in the last twelve months, with malaria typically more important than AIDS, and deaths of women in childbirth more important than deaths from AIDS in many countries. A life evaluation measure (Cantril's ladder of life) is relatively insensitive to the deaths of immediate family, which suggests a low value of life. There are much larger effects on experiential measures, such as sadness and depression, which suggest much larger values of life. It is not clear whether either of these results is correct, yet our results demonstrate that experiential and evaluative measures are not the same thing, and that they cannot be used interchangeably as measures of "happiness" in welfare economics.

WP 14671

Phillip Levine, Diane Schanzenbach

The Impact of Children's Public Health Insurance Expansions on Educational Outcomes

This paper examines the impact of public health insurance expansions through both Medicaid and SCHIP on children's educational outcomes, measured by 4th and 8th grade reading and math test scores, available from the National Assessment of Educational Progress (NAEP). We use a triple difference estimation strategy, taking advantage of the cross-state variation over time and across ages in children's health insurance eligibility. Using this approach, we find that test scores in reading, but not math, increased for those children affected at birth by increased health insurance eligibility. A 50 percentage point increase in eligibility is found to increase reading test scores by 0.09 standard deviations. We also examine whether the improvements in educational outcomes can be at least partially attributed to improvements in health status itself. First, we provide further evidence that increases in eligibility are linked to improvements in health status at birth. Second, we show that better health status at birth (measured by rates of low birth-weight and infant mortality), is linked to improved educational outcomes. Although the methods used to support this last finding do not completely eliminate potentially confounding factors, we believe it is strongly suggestive that improving children's health will improve their classroom performance.

WP 14679

Nicole Maestas, Mathis Schroeder, Dana Goldman

Price Variation in Markets with Homogenous Goods: The Case of Medigap

Nearly 30 percent of Americans age 65 and older supplement their Medicare health insurance through the Medigap private insurance market. We show that prices for Medigap policies vary widely, despite the fact that all plans are standardized and even after controlling for firm heterogeneity. Economic theory suggests that heterogeneous consumer search costs can lead to a non-degenerate price distribution within a market for otherwise homogenous goods. Using a structural model of equilibrium search costs first posed by Carlson and McAfee (1983), we estimate average search costs to be $72. We argue that information problems arise from the complexity of the insurance product and lead individuals to rely on insurance agents who do not necessarily guide them to the lowest prices.

WP 14690

Angus Deaton

Instruments of Development: Randomization in the Tropics, and the Search for the Elusive Keys to Economic Development

There is currently much debate about the effectiveness of foreign aid and about what kind of projects can engender economic de-velopment. There is skepticism about the ability of econometric analysis to resolve these issues, or of development agencies to learn from their own experience. In response, there is movement in development economics towards the use of randomized controlled trials (RCTs) to accumulate credible knowledge of what works, without over-reliance on questionable theory or statistical methods. When RCTs are not possible, this movement advocates quasi-randomization through i nstrumental variable (IV) techniques or natural experiments. I argue that many of these applications are unlikely to recover quantities that are useful for policy or understanding: two key issues are the misunderstanding of exogeneity, and the handling of heterogeneity. I illustrate from the literature on aid and growth. Actual randomization faces similar problems as quasi-randomization, notwithstanding rhetoric to the contrary. I argue that experiments have no special ability to produce more credible knowledge than other methods, and that actual experiments are frequently subject to practical problems that undermine any claims to statistical or epistemic superiority. I illustrate using prominent experiments in development. As with IV methods, RCT-based evaluation of projects is unlikely to lead to scientific progress in the understanding of economic development. I welcome recent trends in development experimentation away from the evaluation of projects and towards the evaluation of theoretical mechanisms.

WP 14692

Helen Levy, David Weir

Take-Up of Medicare Part D: Results from the Health and Retirement Study

We analyze data from the Health and Retirement Study on senior citizens' take-up of Medicare Part D. Take-up among those without drug coverage in 2004 was high; about fifty to sixty percent of this group have Part D coverage in 2006. Only seven percent of senior citizens lack drug coverage in 2006 compared with 24 percent in 2004. We find little circumstantial evidence that Part D crowded out private coverage in the short run, since the persistence of employer coverage was only slightly lower in 2004 -- 2006 than it was in 2002 -- 2004. We find that demand for prescription drugs is the most important determinant of the decision to enroll in Part D among those with no prior coverage. Many of those who remained without coverage in 2006 reported that they do not use prescribed medicines, and the majority had relatively low out-of-pocket spending. Thus, for the most part, Medicare beneficiaries seem to have been able to make economically rational decisions about Part D enrollment despite the complexity of the program. We also find that Part D erased socioeconomic gradients in drug coverage among the elderly.


Annamaria Lusardi, Punam Anand Keller, Adam Keller

New Ways to Make People Save: A Social Marketing Approach

In this study, we use a social marketing approach to develop a planning aid to help new employees at a not-for-profit institution contribute to supplementary pensions. We employed different methods, such as surveys, focus groups and in-depth interviews, to "listen" to employees' needs and difficulties with saving. Moreover, we targeted specific groups that were less likely to save and contribute to supplementary pensions, such as women and low-income employees. The program we developed is not only effective but also inexpensive. While this program was implemented at a single institution, it is suitable to be applied to a variety of employers and demographic groups.

WP 14720

John Helliwell, Christopher Barrington-Leigh, Anthony Harris, Haifang Huang

International Evidence on the Social Context of Well-Being

This paper uses the first three waves of the Gallup World Poll to investigate differences across countries, cultures and regions in the factors linked to life satisfaction, paying special attention to the social context. Our principal findings are: First, using the larger pooled sample, we find that answers to the satisfaction with life and Cantril ladder questions provide consistent views of what constitutes a good life, with an average of the two measures providing a clearer picture than either measure on its own. Second, we find strong evidence for the importance of both income and social context variables in explaining within-country and international differences in well-being. For most specifications tested, the combined effects of a few measures of the social and institutional context are as large as those of income in explaining both international and intra-national differences in life satisfaction. Third, the very significant influences of both income and social factors permit the calculation of compensating differentials for social factors. We find very large income-equivalent values for key measures of the social context. Fourth, the international similarity of the estimated equations suggests that the large international differences in average life evaluations are not due to different approaches to the meaning of a good life, but to differing social, institutional, and economic life circumstances.



National Bureau of Economic Research, 1050 Massachusetts Ave., Cambridge, MA 02138; 617-868-3900; email:

Contact Us