Funds from this grant provide two years of administrative and operational support for the business and economics desk of the Conversation U.S., an experimental new platform in journalism.  Based on similar efforts already successfully launched in the United Kingdom and Australia, the Conversation U.S. is an experiment in direct journalism, providing a platform where researchers and experts write public-facing news and analysis pieces themselves, published under their own bylines, which are then edited in collaboration with experienced journalists for clarity and objectivity.  Content produced by the Conversation is released to the public free of intellectual property restrictions and it encourages other sites—even “competitor” news sites—to reuse and repurpose its content.  The project represents a promising response to changing economic prospects in the news industry and facilitates more direct communication between researchers and the public.

The ratio of men to women earning undergraduate degrees in economics is three to one.  This grant funds a project by Harvard economist Claudia Goldin to try to understand why.   Over the next four years, Goldin will pursue three interconnected activities.  The first is analysis of administrative data to better understand gender differences in selecting courses and majors.  The second is convening researchers to coordinate similar research efforts by other scholars and to design interventions aimed at testing hypotheses suggested by the data.  And the third is to run a kind of “randomized controlled trial” by matching baseline economics departments with those that receive financial incentives to use an intervention.  These experiments will provide evidence about whether and to what extent departmental interventions can effectively raise the number of women who major in economics, and may suggest ways to effectively address gender imbalances in other fields.

Funds from this grant support a series of surveys, tests, and experiments by Princeton behavioral psychologist Eldar Shafir that examine scarcity and its implications for the social and behavioral sciences.  Findings to date suggest that how closely people’s behavior complies with standard economic models of rationality depends interestingly on the constraints they face when making decisions.  Shafir has found that those who are poor (or put into an experimental situation of scarcity) often act more like the rational “homo economicus” posited by normative economic theorists.  In contrast, those who are rich (or who are put into an experimental situation of plenty) often exhibit curious biases and behavioral anomalies that deviate from what standard economic models predict.  Abundance, Shafir’s research suggests, makes inconsistency and irrationality more affordable.  The findings stand in stark contrast to the widespread belief that those in poverty make poor economic decisions.  The truth may be exactly the reverse. This grant will fund the continuation and expansion of Shafir’s research over the next three years, allowing deeper investigation of what factors explain behavioral deviation from traditional economic models and the implication for the design and implementation of policy interventions. 

Behavioral economics should have many implications for government policy.  With this motivation, the Prime Minister’s Office in the U.K. established a Behavioral Insights Team (BIT) in 2010 to help bring insights from behavioral economics to the design and evaluation of government policy. The BIT’s successes in a wide variety of areas—from tax collection to energy efficiency to organ donation—have inspired other countries to launch similar initiatives, including Australia, the Netherlands, Israel, and the United States. The academics and government officials leading such efforts have much to gain by comparing notes with one another.  This grant funds a joint effort by David Halpern, head of the BIT in the U.K., and Max Bazerman, head of the Behavioral Insights Group (BIG) at Harvard, to organize a series of meetings, international conferences, and advanced courses that will bring together researchers from all over the globe to exchange the latest insights on the intersection of behavioral research and public policy.

In contrast with academics in other fields, economists and other social scientists have been slow to adopt new open source programming languages, instead sticking with expensive proprietary applications like Matlab and STATA when doing modeling or running complex analyses on data.  Because such programs cannot be used without a license, their popularity hamstrings reproducibility, hampers archiving, and hinders reuse of research that employs them. This grant funds efforts by Nobel Prize–winner Tom Sargent of New York University and computational economist John Stachurski of Australian National University to speed the adoption of Python—a compact, powerful open source programming and computation platform—among economists and social scientists.  Funds from this grant will bring Stachurski to NYU for a year to work with Sargent on expanding and promoting the usefulness of Python to economists everywhere.  They will develop free Python modules and teaching materials, publicize the capabilities of the new iPython notebook, give presentations, publish a textbook, and further develop the materials and resources freely available on their website, quant-econ.net.

This grant funds a project led by Nicholas Bloom (Stanford University) and Steven Davis (University of Chicago) to examine the relationship between uncertainty in the business community and economic performance.  Partnering with the U.S. Census Bureau, Bloom and Davis will survey the nearly 45,000 U.S. business establishments in the 2016 Annual Survey of Manufacturing, asking respondents to provide forecasts about the coming year, including expected demand for products, prices, cost increases, employment, planned investment, and both industry- and economy-wide performance outcomes.  The resulting will be a powerful new dataset that will provide the first direct measure of uncertainty in the business community.  In a related effort, Bloom and Davis will partner with the Federal Reserve Bank of Atlanta to survey a smaller sample of 1,000 businesses on a monthly basis, providing a complementary dataset that will be able to measure how business uncertainty changes over time and in response to new information.   Bloom and Davis plan to use these datasets to construct new measures of economic uncertainty and address a variety of questions, including the impact of uncertainty on business and aggregate economic performance; whether firm-level uncertainty reduces investment, hiring, and R&D; whether firm forecasts of business conditions and outcomes exhibit cognitive biases, and if so, whether these biases vary by firm age, size, performance, management experience, or external conditions.