The Shape of Economics: Semantic Structure, Knowledge Flows, and Specialization
A study of 50,000 economics papers using SPECTER2, a language model, reveals that economics is not fragmenting but its subfields are growing more distinct internally. Applied and empirical research is expanding, while theoretical and financial topics are declining, a trend linked to the credibility revolution. Fields that connect with distant areas grow faster, though individual interdisciplinary papers do not get more citations. The research offers a new data-driven view of how economics is evolving.
What it examines
This study uses advanced text analysis (SPECTER2 transformer model) to map 50,000 economics papers, exploring how the field is changing. It examines whether economics is becoming more specialized or fragmented, and investigates what drives the growth of new research areas within the discipline.
What it concludes
The research finds economics is specializing, not fragmenting. Fields that connect with distant areas grow faster, mainly through researcher movement, not individual papers. Applications include guiding research policy, fostering innovation, and mapping knowledge flows. Future work could explore other disciplines or deeper mechanisms behind interdisciplinary growth.
Evidence objects
A sweeping analysis of 50,000 economics papers using advanced language models reveals that, contrary to expectations, subfields are not fragmenting but growing more internally cohesive while maintaining steady distances from each other.
key_findings bullet 1 · key_findings · validation V0
The study finds a notable shift: applied and empirical research areas are expanding, while theoretical and financial topics are shrinkinga trend linked to economics 'credibility revolution' and changing research priorities.
key_findings bullet 2 · key_findings · validation V0
Fields whose researchers and citations span distant intellectual areas grow faster, suggesting innovation thrives on cross-field movement; however, individual interdisciplinary papers do not attract more citations, highlighting researcher mobility as key.
key_findings bullet 3 · key_findings · validation V0
This paper uniquely applies SPECTER2 transformer embeddings to analyze 50,000 economics papers, offering a novel, large-scale semantic mapping of research specialization and knowledge diffusion. Its innovative bibliometric methodology yields fresh insights into field homogeneity, researcher mobility, and interdisciplinary citation patterns, making it compelling and impactful for economics and related disciplines.
key_findings bullet 4 · key_findings · validation V0
Raw abstract and provenance
How is economics evolving as a discipline—are its subfields converging or fragmenting, and what drives the growth of new research areas?
Source row: 2006 · abstract type: snippet