Abstract: Cost-control interventions that target physicians’ clinical discretion are common in healthcare, but evidence on their efficacy is scarce; in particular for “soft” policies when liability is unlikely to be enforced by the regulator. We study the effectiveness of preferred drug policies (minimum prescription quotas of specific “preferred” drugs) in altering physicians practice styles within the high volume drug class of HMG-CoA-reductase inhibitors (statins) in the German statutory health insurance system. Using a nationally representative panel of ambulatory care physicians between 2011 and 2014, we exploit the decentralized institutional setting to estimate physician responses to variation in preferred drug policies across regional physician associations over time in a generalized difference-in-differences design. Results show that although the cost-control mechanism increases average policy adherence, this effect is mainly driven by physicians with initially high use rates of preferred drugs. We argue that such misdirection may limit the policy’s usefulness in reducing inappropriate practice variation among healthcare providers.
Abstract: Fertility has a strong biological component generally ignored by economists. Using the UK Biobank, we analyze the extent to which genes, proxied by polygenic scores, and the environment, proxied by early exposure to the contraceptive pill diffusion, affect age at first sexual intercourse, age at first birth, completed family size, and childlessness. Both genes and environment exert substantial influences on all outcomes. The anticipation of sexual debut and the postponement of motherhood led by the diffusion of the pill are magnified by gene-environment interactions, while the decline in family size and the rise in childlessness associated with female emancipation are attenuated by gene-environment effects. The nature-nurture interplay becomes stronger in more egalitarian environments that empower women, allowing genes to express themselves more fully. These conclusions are confirmed by heterogeneous effects across the distributions of genetic susceptibilities and exposure to environmental risks, sister fixed effects models, mother-daughter comparisons, and counterfactual simulations.
Abstract: We study the effects of maternal age on infant health. Age at birth has been increasing for the past several decades in many countries, and correlations show that health at birth is worse for children born to older mothers. In order to identify causal effects, we exploit school entry cutoffs and the empirical finding that women who are older for their cohort in school tend to give birth later. In Spain, children born in December start school a year earlier than those born the following January, despite being essentially the same age. We show that as a result, January-born women finish school later and are (several months) older when they marry and when they have their first child. We find no effect on educational attainment. We then compare the health at birth of the children of women born in January versus the previous December, using administrative, population-level data, and following a regression discontinuity design. We find small and insignificant effects on average weight at birth, but the children of January-born mothers are more likely to have very low birthweight. We interpret our results as suggestive of a causal effect of maternal age on infant health, concentrated in the left tail of the birthweight distribution, with older mothers more likely to give birth to (very) premature babies.