La Ferrara, Chong & Duryea (2008), “Soap Operas and Fertility: Evidence from Brazil”.

Question 1

📘 Flashcard 1: What is the main research question of the study?

Answer 1

The study asks whether exposure to television—specifically soap operas (novelas) produced by Brazil’s Rede Globo—causally affected fertility decisions in Brazil between 1970 and 1991. It explores how media content, rather than formal policies or education, may have shaped women’s preferences for smaller families.

Question 2

🧪 Flashcard 2: What empirical method does the study use?
A:
The study uses a quasi-experimental difference-in-differences (DiD) strategy, leveraging the staggered geographic rollout of the Rede Globo TV signal across municipalities (AMCs) in Brazil. They exploit this as a natural experiment to estimate the causal effect of TV access (specifically Globo’s soap operas) on fertility rates, controlling for time and area fixed effects, and time-varying covariates.

Question 3

👩‍👧‍👧 Flashcard 3: What were the main effects of Globo’s novelas on fertility?
A:
Exposure to Globo is associated with significantly lower fertility, particularly:

A 2.7% decrease in average number of live births in treated areas.
A 0.6 percentage point reduction in the annual probability of giving birth per woman (approx. 6% of the mean).
Larger effects for older women (25–44), consistent with the idea that exposure affects stopping behavior rather than delaying first births.
Stronger effects among lower-income and less-educated women, who are more likely to rely on TV rather than print media.

Question 4

📊 Flashcard 4: What data were used and how was fertility measured?
A:
Two datasets:

Aggregate AMC-level Census data (1970, 1980, 1991): Measured average number of live births for women aged 15–49.
Individual-level 1991 Census retrospective panel: Reconstructed annual fertility histories from 1979–1991 using ages of children in the household.
TV access was inferred from GIS-based signal coverage maps matched to Globo transmitter data over time.

Question 5

🧩 Flashcard 5: How does the study critically assess its own identification strategy?
A:
Several robustness checks are employed:

Placebo tests using future Globo entry show no pre-trends.
Selection checks find no correlation between Globo rollout and prior fertility trends or demographic shifts.
Falsification tests using another network (SBT) found no effect on fertility, suggesting the influence is specific to Globo’s novela content, not just TV access.
The study also acknowledges limitations: novelas were not randomly assigned, and content influence cannot be cleanly separated from mere exposure to TV.

Question 6

🎥 Flashcard 6: What evidence supports the idea that content—not just TV exposure—mattered?
A:
The study shows:

Most novela heroines were childless or had only one child, contrasting with societal norms.
Children in Globo-covered areas were more likely to be named after novela characters, indicating cultural absorption.
Exposure to SBT (which broadcast foreign shows) had no effect on fertility, reinforcing the argument that locally resonant narratives drive the results.

Question 7

💡 Flashcard 7: What are the broader implications for policy and development?
A:
The study suggests that media content can powerfully shape social norms and behaviors, especially where literacy is low. Policymakers might:

Use soap operas or serialized media to influence behaviors in areas like public health, education, or gender norms.
Recognize media as a low-cost, scalable vehicle for norm-shifting interventions in developing contexts.
This shifts attention from just delivering information to crafting culturally resonant narratives that can influence demand-side behavior.