Research

We use population-wide administrative data from Denmark and an event-study design spanning nearly two decades to examine the impact of having a parent with dementia on adult children’s labor market, physical health, and mental health outcomes. We find no meaningful effects on labor supply, earnings, or physical health care use. In contrast, mental health care use increases substantially, driven by daughters, beginning five years before a parent’s dementia related death, peaks around the time of death, and converges to the counterfactual trend over seven years. Results suggest that robust long-term care policy can largely insulate adult children economically from parental dementia, but mental health spillovers persist, spurring nearly a decade of elevated use.

How much should program impacts vary across places when cash transfers are implemented at national scale? We study this question using a nationwide experiment embedded in a humanitarian cash transfer program for Syrian refugees in Lebanon that randomizes households to alternative district budget allocation rules. While allocation rules shift resources toward demographically distinct households, average improvements in poverty outcomes are similar across households prioritized by these rules. In contrast, program impacts vary sharply across locations, with geographic heterogeneity dominating differences generated by targeting. Qualitative evidence links this variation to locally binding constraints, including housing obligations, accumulated debt, and institutional frictions.

Background: This paper investigates the long-term effects of the measles, mumps, and rubella (MMR) vaccine on healthcare, education, and economic outcomes using a novel dataset from the Copenhagen School Health Records Register. Methods: To address potential endogeneity, we use within-sibling variation in vaccination status induced by different periods of vaccine availability in Denmark. Results: Our findings reveal that, prior to the establishment of herd immunity, vaccinated cohorts experienced substantial protection against hospitalizations related to MMR. During the same period, we also observe modest improvements in educational outcomes and positive—though statistically insignificant—estimates for labor market outcomes among siblings with discordant vaccination status. We find no impact of vaccination on hospitalizations, education, or economic outcomes for cohorts born after herd immunity was established, a period when everyone benefited from herd protection regardless of individual vaccination status. Conclusions: MMR vaccination, before herd immunity, reduced hospitalization due to Measles, Mumps, and Rubella. The impact on later outcomes, such as education, income, and employment lacks statistical precision at conventional levels. Evidence suggest positive self-selection in vaccination among families with high socioeconomic status.

Background: The Copenhagen School Health Record Register (CSHRR) includes health information from school examinations and is now updated with information on measles, mumps and rubella vaccines for the cohorts born from 1977 to 1994. Aim: The aim of this study is to provide a comprehensive description and validation the newly digitised vaccine information in the CSHRR. Methods: We describe the data collection and the newly digitalised information in the CSHRR. We investigate the extent to which the full CSHRR population is representative of Copenhagen and the entire Danish population. Furthermore, we explore how the registry information on vaccination uptake based on reimbursement data matches the vaccine information obtained from CSHRR for the period during which both data are available. Results: The CSHRR population matches closely the complete population of all schoolchildren in Copenhagen, and information on vaccine uptake in CSHRR matches with vaccine registry data for later cohorts. However, a sizable proportion of the immigrant children in the CSHRR have missing information on vaccination. Removing children who have had no additional immunisations enhances data quality. Conclusions: The CSHRR covers a large share of the Danish population and includes detailed vaccine information. By linking the data to other registry data, the updated CSHRR is valuable resource for future research.

We evaluate the impact of a nationwide public health intervention on deaths from sudden infant death syndrome (SIDS), using population data from Denmark in a regression discontinuity research design. The information campaign—implemented primarily through a universal nurse home visiting program—reduced infant mortality by 17.2 percent and saved between 11.6-13.5 lives over 10,000 births. The estimated effect sizes are 11-14 times larger among low birthweight and preterm infants relative to the overall population. Improvement in infant mortality is concentrated among those with low socio-economic status and with limited access to health information, thereby reducing health inequities at birth.

We study two yearlong unconditional cash-based assistance programs for Syrian refugees in Lebanon, and show that they improve consumption, child well-being, food security, and reduce livelihood coping. Despite high transfer values, we find no evidence of lasting effects at six months after either program ends. Households spend transfers on basic needs, and take children out of work and re-enroll them in school. Beneficiaries increase cash savings and their stock of durable goods, but liquidate and spend these assets during or soon after the beneficiary period. The results are similar for longer assistance periods and across transfer modalities.

The short-lived effects of unconditional cash transfers to refugees, Journal of Development Economics, Volume 160, 2023, 102942, ISSN 0304-3878, https://doi.org/10.1016/j.jdeveco.2022.102942.

Using a strict, age-specific lockdown order for adults aged 65 and older in Turkey, we examine the mental health consequences of an extended period of tight mobility restrictions on senior adults. Adopting a regression discontinuity design, we find that the curfew-induced decline in mobility substantially worsened mental health outcomes, including somatic and nonsomatic symptoms of mental distress (approximately 0.2 standard deviation). Exploring potential channels, we document an increase in social and physical isolation, with no evidence of robust changes in labor market outcomes or intrahousehold conflict for this subpopulation.

Twenty-six states require that women seeking an abortion wait between 18 and 72 h after receipt of counseling before the abortion can be completed. Thirteen states require that the counseling be given in person necessitating at least two visits to the provider. In April of 2015, Arkansas increased the waiting period for an abortion from 24 to 48 h and more significantly, required that women receive the counseling in person. The two-visit requirement, unnecessary from a medical standpoint, substantially increases the cost in terms of travel, lost work, child care, and exposure to stigma and harassment upon accessing abortion facilities. We use a regression discontinuity design to analyze the immediate effect of Arkansas’s 2015 mandatory waiting period (MWP) law on abortion rates. We use de-identified, individual-level data from the Arkansas Department of Health (DOH) on all abortions performed in Arkansas from 2000 to 2020. Our study is the first to use monthly individual data and stratify analyses by race/ethnicity, age, parity, marital status and schooling. Abortion rates fell 17 percent among all women, but 22 percent among white non-Hispanics and 14 percent among black non-Hispanics immediately after the law went into effect. We show that the law decreased abortion rates the most among unmarried adults with children. If Roe v. Wade is overturned, abortion will become illegal in Arkansas. Given the decline in abortion rates associated the MWP’s two-visit requirement, abortion rates will likely fall further as travel distance to the nearest legal provider increases.

Another day, another visit: Impact of Arkansas’ mandatory waiting period for women seeking an abortion by demographic groups, Journal of Public Economics, Volume 213, 2022, 104715, ISSN 0047-2727, https://doi.org/10.1016/j.jpubeco.2022.104715.

We develop and assess the performance of an econometric prediction model that relies on administrative data held by international agencies to target over $380 million annually in unconditional cash transfers to Syrian refugees in Lebanon. Standard metrics of prediction accuracy suggest targeting using administrative data is comparable to a short-form Proxy Means Test, which requires a survey of the entire target population. We show that small differences in accuracy across approaches are largely attributable to a few data fields. These results are robust to a blind validation test performed on a random sample collected after the model derivation, as well as the type of estimator used for prediction. We discuss relative costs, which are likely to feature prominently when alternative approaches are considered in practice.

Targeting humanitarian aid using administrative data: Model design and validation, Journal of Development Economics, Volume 148, 2021, 102564, ISSN 0304-3878, https://doi.org/10.1016/j.jdeveco.2020.102564.

We study how individual political preferences changed in response to the influx of over 3.5 million Syrian refugees to Turkey during 2012–2016. Using a difference-in-differences research design, we compare the political outcomes in geographic areas with high versus low intensities of refugee presence before and after the beginning of the Syrian Civil War. To address the endogeneity of refugees’ location choices, we adopt an instrumental variables approach that relies on (1) historical dispersion of Arabic speakers in Turkish provinces and (2) driving distances between Turkish and Syrian residential areas to predict the flows of refugees across Turkish provinces during the study period. We find strong polarization in attitudes towards refugees between the supporters and opponents of the ruling Justice and Development Party (AKP). However, regression analyses of monthly survey data suggest that the massive inflow of refugees induced only a modest net drop in support for the AKP. Refugee inflows did not have a significant impact on election outcomes during the study period.

We study the impact of the sudden arrival of more than three million Syrian refugees in Turkey on Turkish businesses. Our empirical methodology exploits the annual variation in refugee outflows from Syria and geographic concentrations of Arabic-speaking communities in Turkey, which date back to the rupture of the Ottoman Empire after World War I. Using yearly firm census data from 2006 to 2015, we document that refugee inflows induced a positive impact on the intensive and extensive margins of firm production. The effects were stronger for smaller firms and for firms operating in the construction and hospitality sectors. Increased informal activity by firms drove most of these changes.

Blessing or burden? Impacts of refugees on businesses and the informal economy, Journal of Development Economics, Volume 146, 2020, 102490, ISSN 0304-3878, https://doi.org/10.1016/j.jdeveco.2020.102490.

Objectives. To determine whether the 2014 Affordable Care Act Medicaid expansion affected well-being in the low-income and general adult US populations. Methods. We obtained data from adults aged 18 to 64 years in the nationally representative Gallup-Sharecare Well-Being Index from 2010 to 2016 (n = 1 674 953). We used a difference-in-differences analysis to compare access to and difficulty affording health care and subjective well-being outcomes (happiness, sadness, worry, stress, and life satisfaction) before and after Medicaid expansion in states that did and did not expand Medicaid. Results. Access to health care increased, and difficulty affording health care declined following the Medicaid expansion. Medicaid expansion was not associated with changes to emotional states or life satisfaction over the study period in either the low-income population who newly gained health insurance or in the general adult population as a spillover effect of the policy change. Conclusions. Although the public health benefits of the Medicaid expansion are increasingly apparent, improved population well-being does not appear to be among them. Public Health Implications. Subjective well-being indicators may not be informative enough to evaluate the public health impact of expanded health insurance.

Between July 2005 and July 2007, the Oregon Supplemental Nutrition Program for Women, Infants and Children program conducted the largest randomized field experiment (RFE) ever in the United States to assess the effectiveness of a low-cost peer counseling intervention to promote exclusive breastfeeding. We undertook a within-study comparison of the intervention using unique administrative data between July 2005 and July 2010. We found no difference between experimental and nonexperimental estimates but failed to determine correspondence based on more stringent criteria. We show that tests for nonconsent bias in the benchmark RFE might provide an important signal as to confounding in the nonexperimental estimates.

Randomized experiments provide unbiased estimates of treatment effects, but are costly and time consuming. We demonstrate how a randomized experiment can be leveraged to measure selection bias by conducting a subsequent observational study that is identical in every way except that subjects choose their treatment—a quasi-doubly randomized preference trial (quasi-DRPT). Researchers first strive to think of and measure all possible confounders and then determine how well these confounders as controls can reduce or eliminate selection bias. We use a quasi-DRPT to study the effect of class time on student performance in an undergraduate introductory microeconomics course at a large public university, illustrating its required design elements: experimental and choice arms conducted in the same setting with identical interventions and measurements, and all confounders measured prospectively to treatment assignment or choice. Quasi-DRPTs augment randomized experiments in real-world settings where participants choose their treatments.

Objectives. To describe demographic characteristics and abortion timing of minors in Arkansas who obtained an abortion through a judicial bypass, and to contrast the frequency of judicial bypass in other states in the United States.

Methods.We used individual-level data on all abortions to minors (n = 2624) performed in Arkansas from 2005 to 2014 and aggregated data from 10 additional states on abortions obtained through a judicial bypass. We characterized correlates of the judicial bypass and its association with the timing of abortion. Results. Minors in Arkansas who used the courts were more likely to be 17 years of age, nonresidents of Arkansas, Hispanic, and with fewer previous pregnancies than their younger, resident, non-Hispanic White, and multigravida counterparts; 7.7% of abortions to minors were obtained via a bypass in 11 states, but only 2% if Texas and Arkansas are excluded.

Conclusions. The demographics of minors who obtain an abortion through a judicial bypass differ significantly from those who have parental consent, and there is widespread variation in the proportion of bypass cases across states.

Couples in Turkey exhibit son preference through son-biased differential stopping behavior that does not cause a sex ratio imbalance in the population. Demand for sons leads to lower ratios of boys to girls in larger families but higher ratios in smaller families. Girls are born earlier than their male siblings, and son-biased fertility behavior is persistent in response to decline in fertility over time and across households with parents from different backgrounds. Parents use contraceptive methods to halt fertility following a male birth. The sibling sex composition is associated with gender disparities in health. Among third- or later-born children, female infant mortality is 1.5 percentage points lower if the previous sibling is male. The female survival advantage, however, disappears if the previous sibling is female. Having an older female sibling shifts the gender gap in infant mortality rate by 2 percentage points in favor of males. The improvement in infant mortality is strongest in favor of males who have no older male siblings.

Little experimental evidence exists on the causal impact of class time on academic performance when students have access to extensive course material online. We randomized 725 college students into traditional twice-per-week and compressed once-per-week lecture formats in introductory microeconomics. Students in the traditional format scored 3.2 out of 100 points higher (0.21 standard deviations) on the midterm than those in the compressed format but a statistically insignificant 1.6 points higher (0.11 standard deviations) on the final. There were no differences in non-cognitive outcomes. Students in the middle tercile of predicted test scores performed worst in the compressed format relative to those in the traditional format but there was little difference in test scores by format in the top tercile of predicted performance. While the compressed format offers clear savings in classroom space and professors’ time, these savings come at some cost to student performance.

We randomized 1948 pregnant women who intended to breastfeed to receive either standard WIC breastfeeding support or standard support plus telephone peer counseling. We estimated the effect of peer counseling on any and exclusive breastfeeding at 1, 3, and 6 months. At 3 months postpartum, 50.5% of women in the treatment group were still breastfeeding versus 44.1% in the control (P = .01). The effect was driven entirely by an improvement in exclusive breastfeeding, where rates in the treatment group at 3 months (27.3%) exceeded those in the control group (20.2%) by 7.1 percentage points (P < .01). We found no effect of peer counseling at 1 or 6 months. Telephone peer counseling improved exclusive breastfeeding at 3 months postpartum among WIC participants.