Population aging is widely expected to have detrimental effects on economic growth yet there is little empirical evidence about the magnitude of its effects. Many U.S. states have experienced substantial growth in the size of their older population during the last several decades. Maestas, Mullen, and Powell use the predictable component of this growth, which varies substantially across states, to estimate the impact of population aging on per capita output. They find that a 10% increase in the fraction of the population ages 60+ decreases per capita GDP by 5.5%. Two-thirds of the reduction is due to slower growth in labor productivity across the age distribution, while one-third arises from slower labor force growth.
This paper, by Fadlon and Neilsen, analyzes how people's health behaviors change when family members have adverse health events. It considers health events experienced by family members both within and across generations. The study finds that spouses and adult children immediately increase their health investments and improve their health behaviors in response to family shocks, and that these effects are both significant and persistent for at least several years. Exploiting the detailed nature of the data, the authors consider a variety of mechanisms influencing these behavioral changes, and highlight the salience of health risks as an important factor.
In addition to the conference paper, the research was distributed as NBER Working Paper w24042, which may be a more recent version.
Mommaerts examines whether informal care by family members explains the limited demand for long-term care insurance. Motivated by evidence that the availability of potential informal caregivers is correlated with lower insurance demand and that informal caregivers substitute for formal care, they estimates a dynamic model of long-term care decisions between an elderly parent and her adult child. The availability of informal care lowers the demand for insurance by 14 percentage points overall. An insurance policy that compensates informal care can generate substantial increases in insurance demand and family welfare, and decreases in Medicaid spending.
The health-wealth gradient, wherein the affluent are healthier than those with fewer financial resources, has been well-documented. However, the direction of this relationship is not as firmly established. Heiss, McFadden, Scarpati, Winter, and Wuppermann use plausibly exogenous changes in home prices during the recent housing crisis as a natural experiment for evaluating the effect of changes in wealth on health. Did health outcomes, such as chronic conditions, change due to large, rapid changes in home prices? Further, did patients curb their use of medical services like non-urgent hospitalizations, office visits, prescription drug use, and preventive care? The researchers focus on the effects among the American elderly population using a random sample of nine-million Medicare beneficiaries, finding effects on costs and use of health care services as well as morbidity. Beneficiaries respond to decreases in wealth by increasing their use of health care and selected preventive services, which goes in hand with increases in Medicare costs as well as detection of chronic conditions.
The welfare associated with public insurance is often difficult to quantify because the demand for coverage is unobserved and thus cannot be used to analyze welfare. However, in many settings, individuals can purchase private insurance to supplement public coverage. This paper outlines an approach to use data and variation from private complementary insurance to quantify welfare associated with counterfactuals related to compulsory public insurance. Cabral and Cullen then apply this approach using administrative data on disability insurance. Their findings suggests that public disability insurance generates substantial surplus for the sample population, and there may be gains to increasing the generosity of coverage.
In addition to the conference paper, the research was distributed as NBER Working Paper w22583, which may be a more recent version.
The rapid evolution into a 24h society challenges individuals’ ability to conciliate work schedules and biological needs. Epidemiological research suggests that social and biological time are increasingly drifting apart (“social jetlag”). This study uses a spatial regression discontinuity design to estimate the economic cost of the misalignment between social and biological rhythms arising at the border of a time-zone in the presence of relatively rigid social schedules (e.g., work and school schedules). Exploiting the discontinuity in the timing of natural light at a time-zone boundary, Mazzonna and Giuntella find that an extra hour of natural light in the evening reduces sleep duration by an average of 19 minutes and increases the likelihood of reporting insufficient sleep. Using data drawn from the Center for Disease Control and Prevention and the US Census, they find that the discontinuity in the timing of natural light has significant effects on health outcomes typically associated with circadian rhythms disruptions (e.g., obesity, diabetes, cardiovascular diseases, and breast cancer) and economic performance (per capita wages). They provide a lower bound estimate of the health care costs and productivity losses associated with these effects.
Inaccurate longevity expectations will lead to suboptimal life cycle planning with negative consequences for wellbeing in old age. Bago d'Uva, O'Donnell, and Van Doorslaer evaluate the accuracy of expectations by comparing probabilities of living to 75 reported in the Health and Retirement Study with the actual survival of the respondents to that age. On average, survival predictions are poor and downwardly biased. Men predict less accurately, but that is because they face greater uncertainty due to a higher mortality rate. Women underestimate their survival chances more than men. Predictions are least accurate and most noisy at the lowest levels of education and cognitive functioning. A simple model suggests that welfare would be higher if everyone made decisions on the basis of the base survival rate rather than relying on the individual-specific survival probabilities reported. Despite the predictions of the least educated being the least accurate, they are not unambiguously the least valuable for decisions.
This paper, by Lindeboom, Godard and Koning, analyzes the effects of health screening on disability insurance applications and labor market outcomes, based on a policy reform requiring more stringent health screening in the Netherlands – the so-called "Gatekeeper Protocol." The authors find that the more stringent screening required in selected regions had a significantly negative effect on disability insurance applications and inflows, particularly from those applying based on mental disorders. The treated and control cohorts were followed for eight years. Among those subject to stringent screening, the authors find elevated mortality risk in the pool of applicants, but not in the pool of non-applicants. This suggests that the policy improved the targeting of DI benefits to those in poorer health.
Measures of hospital quality are often constructed from risk-adjusted mortality, risk-adjusted spending, and patient readmissions rates, among other factors. Critics claim that this approach is limited because it does not account for confounders such as patient behavior and socioeconomic circumstance. In this study, Chandra and Staiger consider whether confounding factors invalidate this approach to quality measurement. They note that if quality measures are invalidated by confounding factors, then the closing of high versus low mortality hospitals shouldn't predict subsequent patient outcomes. Using a sample of several million Medicare patients hospitalized for 5 major conditions, they find that the closing of high versus low mortality hospitals does affect subsequent outcomes. The conclude, therefore, that hospital quality measures using risk-adjusted Medicare claims data are in fact highly validated measures of hospital performance.
Ageing in place policies encourage and facilitate that elderly postpone moving to a nursing home. These policies are considered a win-win: they keep public spending on long-term care for the elderly (LTC) in check and are in line with the preferences of the elderly. Moreover, they are assumed to have no effect on the health of the target population, but the absence of health effects has so far not been documented. Bakx, Wouterse, Van Doorslaer, and Wong evaluate the impact of a nursing home admission for the sub-population using Dutch administrative data from the period 2009-2013. We exploit the unique situation that in the Netherlands virtually all LTC is publicly financed and that an individual’s eligibility for these services is determined by a randomly assigned assessor who has substantial discretionary power. Using differences between assessors in the tendency to grant admission to a nursing home as a source of exogenous variation, we show that the impact of being eligible for nursing home care on mortality is zero on average, but with considerable effect heterogeneity. The costs of a nursing home admission are completely offset by lower spending on home care and medical care, which are in part driven by a sharp decrease in the probability of a hospital admission. These findings suggest that ageing in place policies are not a way to cut public spending. Moreover, if they are primarily a way of bringing the supply of LTC in line with preferences of the elderly for staying at home, this comes at the cost of an increase in health problems leading to hospital admissions.