“The Children of the Missed Pill”, 2021-09-01 (; similar):
We assess the impact of exogenous variation in oral contraceptives prices—a year-long decline followed by a sharp increase due to a documented collusion case—on fertility decisions and newborns’ outcomes. Our empirical strategy follows an interrupted time-series design, which is implemented using multiple sources of administrative information. As prices skyrocketed (45% within a few weeks), the Pill’s consumption plunged, and weekly conceptions increased (3.2% after a few months).
We show large effects on the number of children born to unmarried mothers, to mothers in their early twenties, and to primiparae women. The incidence of low birth weight and fetal/infant deaths increased (declined) as the cost of birth control pills rose (fell). In addition, we document a disproportional increase in the weekly miscarriage and stillbirth rates. As children reached school age, we find lower school enrollment rates and higher participation in special education programs.
Our evidence suggests these “extra” conceptions were more likely to face adverse conditions during critical periods of development.
…This paper quantifies the Pill’s role in fertility and child outcomes using a sequence of events in which unexpected shocks affected the access to oral contraceptives. In particular, we exploit a well-established case of anticompetitive behavior in the pharmaceutical market, which—after a year-long price war between the 3 largest pharmaceutical retailers in Chile—triggered a sharp and unexpected increase in the prices of birth control pills.
The price war took place during 2007, and it effectively reduced the prices of medicines across the board. In particular, prices of oral contraceptives fell by 24% during that year. By the end of 2007, the 3 largest pharmacies agreed to end the price war and engaged in a collusion scheme in which they strategically increased the prices of 222 medicines. Oral contraceptives were included in this group, experiencing price increases ranging 30–100% in just a few weeks (45% on average in the first 3 weeks). We use daily information on prices and quantities sold in the country by the 3 companies from almost 40 million transactions to determine the date when the price changes for birth control pills took place. Using these data, we implement an interrupted time-series analysis (Bloom, 2003; 1988), which takes into account the seasonality of births, the general trends of fertility, as well as dynamics that arise because it takes time for the menstrual cycle to be fully regulated after discontinuing the Pill’s intake. We complement the pharmacies’ transaction data with administrative information from birth and death certificates collected 2005–3200816ya and administrative records on school enrollment 2013–32016. Our empirical strategy considers 2 different treatments: one stemming from a sustained and steady decline in prices (200717ya) and another one from a massive and sudden increase (first weeks of 2008).
View PDF: