Every Monday, the National Bureau of Economic Research, a nonprofit organization made up of some of North America’s most respected economists, releases its latest batch of working papers. The papers aren’t peer-reviewed, so their conclusions are preliminary (and occasionally flat-out wrong). But they offer an early peek into some of the research that will shape economic thinking in the years ahead. Here are a few of this week’s most interesting papers:
Authors: Benjamin N. York, Susanna Loeb
What they found: An experimental text-message-based program for parents of preschoolers led to more involved parents and higher literacy scores in children.
Why it matters: Numerous studies have shown the importance of good parenting on childhood development, and that wealthier parents spend more time, on average, on activities that help their children learn. But programs aimed at improving parenting can be expensive and time-consuming. As an alternative, the authors of this paper tested a simple, inexpensive technique: They sent text messages with parenting tips and facts to parents of preschoolers in San Francisco. A separate group of parents received “placebo” texts about kindergarten enrollment, vaccination requirements or similar administrative topics. Parents who received the substantive texts reported spending more time telling stories, pointing out words that rhyme, playing games and engaging in other learning-oriented activities. Teachers reported that those same parents asked more questions about their children’s learning. And the children themselves scored better on literacy tests.
Key quote: “Perhaps the most compelling implication of this study’s findings is that text messaging represents an extremely viable strategy for promoting parental involvement. The vast majority of American adults have cell phones, nearly all cell phone owners already send and receive texts, and texting rates are particularly high in black and Hispanic populations. Moreover, virtually all text messages are opened (by comparison, the e-mail open rate in education is about 36 percent). Last year, we spent less than one dollar per family to send text messages, and fixed program expenses such as content development costs trend towards zero as the program scales.”
Data they used: Data from a randomized controlled trial among parents in the San Francisco Unified School District.
Title: “Gray Matters: Fetal Pollution Exposure and Human Capital Formation”
Authors: Prashant Bharadwaj, Joshua Graff Zivin, Matthew Gibson, Christopher A. Neilson
What they found: Exposure to air pollution in utero had significant negative effects on fourth-grade test scores for children in Chile.
Why it matters: Prenatal health can have long-term effects on health and intelligence, and air pollution can have significant negative impacts on fetal development. So public health researchers have theorized that exposure to pollution before birth could have permanent consequences for both individuals and the economy. But it has been difficult to link the two directly; more polluted areas tend to be poorer, for example, and problems take years or even decades to develop. In this paper, the authors use a large dataset from Santiago, Chile, that tracks births, environmental conditions and subsequent test scores. They use comparisons between siblings to help control for issues unrelated to pollution, such as income. They find that exposure to high levels of carbon monoxide while in utero leads to significantly lower math and reading scores in fourth grade.
Key quote: “Our empirical results are also of direct importance for policy makers. Carbon monoxide is directly regulated throughout the developed and an increasing share of the developing world. Nearly all of these regulations are based on the benefits associated with reductions in pollution-related health, mortality and hospitalizations. Our results suggest that such an approach underestimates regulatory benefits. … It completely ignores the human capital effects, which have been largely invisible, but may well rival the more dramatic health effects in magnitude since they affect a much broader swath of the population.”
Data they used: Birth and environmental data from the Health Ministry of the government of Chile; educational data from SIMCE, “a national standardized test administered in all schools in Chile.”
Title: “Does Conflict of Interest Lead to Biased Coverage? Evidence from Movie Reviews”
Authors: Stefano DellaVigna, Johannes Hermle
What they found: Movie reviewers aren’t biased toward films produced by their employers’ parent companies.
Why it matters: Media outlets — including FiveThirtyEight — are often owned by conglomerates, which means that publications often review movies released by their own corporate siblings. In this paper, the authors look at whether those potential conflicts of interest influence reviews – whether, for example, the Wall Street Journal tends to give more positive reviews to films released by 20th Century Fox, which was until recently owned by the same parent company. They find no evidence of bias, either in which films the publications choose to review or in the reviews themselves. They also find no evidence of bias in the review-aggregation site Rotten Tomatoes during the period it was owned by News Corp.
Key quote: “This paper provides some evidence on how media outlets navigate the tradeoff between professional journalism and revenue maximization for the owners. In this case, professionalism and reputation concerns appear to trump possible short-term revenue gains.”
Data they used: More than half a million reviews aggregated by Rotten Tomatoes and Metacritic.