Civil society: Tracking 100 years of economic research beyond markets and states

with Samuel Bowles and Wendy Carlin 

Using topic modeling on the corpus of papers published in seven leading economics journals since 1900, we study the evolving emphasis in research on themes relating to the state, markets, and civil society, the latter referring to families, firms as organizations, other private organizations, neighborhoods, and identity groups. We document a shift between 1900 and 1970 away from research on state-related topics towards the market, even as the economic importance of the state was growing. This was followed by a substantial movement away from market topics towards topics related to civil society. We associate the first shift with the mathematical formalization of the Marshallian paradigm. The subsequent increased attention to civil society coincided with novel research questions and empirical methods including experiments and the use of large datasets. Since the middle of the last century advances in game theory and the economics of asymmetric information also facilitated the extension of economists’ research agendas to encompass themes central to economic behavior in civil society, including other-regarding preferences and social norms as well as strategic interactions not covered by complete contracts.(Economic Letters, 2024) 

CEPR coverage 

A Response: Game Theory in Economics

with Meghana Prasad

Adding to the ongoing debate on how best to understand game theory and its relevance to economics, a response to Atanu Sengupta and Abhijit Ghosh’s “Non-cooperative Game Theory and Pay-off” (EPW, 2017)

Economic Networks and the Dynamics of Wealth Inequality: A Longitudinal Cross-Cultural Investigation.

with Matthew Jackson, Tom Rutter et. al. 

The content of intermediate-level undergraduate textbooks represents a consensus in the discipline about what a student trained in economics should know. We use topic modeling to explore both this conceptual benchmark in leading textbooks and the content of economic research in journal publications over 115 years. Our mapping of content to 3-dimensional meta-topic spaces in microeconomics and macroeconomics reveals that the conceptual frameworks used in research have diverged over the last four decades from the benchmarks conveyed to majors through textbooks. We suggest that the origins of the divergences and the implications for economics education differ between microeconomics and macroeconomics. 

presented at NSF Networks Science Conference 2024

What do we think an economics student should know? A machine learning investigation of research and intermediate-level texts.

with Samuel Bowles, Wendy Carlin and Simon D. Halliday

The content of intermediate-level undergraduate textbooks represents a consensus in the discipline about what a student trained in economics should know. We use topic modeling to explore both this conceptual benchmark in leading textbooks and the content of economic research in journal publications over 115 years. Our mapping of content to 3-dimensional meta-topic spaces in microeconomics and macroeconomics reveals that the conceptual frameworks used in research have diverged over the last four decades from the benchmarks conveyed to majors through textbooks. We suggest that the origins of the divergences and the implications for economics education differ between microeconomics and macroeconomics.

LLMs for Agriculture in India: A field experiment with Brian Jabarian and Joshua Deutschmann in India and Kenya evaluating how LLM-based chatbots delivering weather forecasts affect farmers’ decisions, comprehension, and engagement.

Stigma, Communication, and AI: Studying whether AI can reduce social frictions in information exchange in socially stratified settings, and how human-AI vs. human-human interactions shape learning and mental model updating.

Human Judgment and Algorithmic Recommendations: Investigating how decision-makers interpret opaque AI input, and how image concerns and biased priors can lead to selective adoption and amplified discrimination.

Evaluating Opaque Algorithms: Exploring how users update beliefs when they can’t infer how a model works, and how this affects overall performance in joint human-AI decision-making.

Task Allocation and Labor Market Decisions among Couples: A field experiment with Adrian Blattner, Mariana Guido and Jason Weitze on the household dynamics of the motherhood penalty