April 14th, 2026
Misaligned Boundaries: Border Manipulation in Qing Prefectures
(Mengru Wang, NUS Economics, Economic History)
Abstract not available yet.
April 21st, 2026
When Does Overseas Education Build State Capacity
(Nuo Lei, NUS Economics, Development, Public & Political Economy)
Abstract not available yet.
May 5th, 2026
The Evaluation of the Impact of Generative Artificial Intelligence on Innovation in Biopharmaceutical Enterprises
(Yuan Tian, Center for Enterprise Growth and National Economic Security Research, Tsinghua University, Innovation & Growth)
This study investigates whether generative artificial intelligence (AI) has materially improved firm-level innovation in the biopharmaceutical industry. Using Chinese A-share listed biopharmaceutical manufacturing firms from 2017 to 2024, we treat the recent wave of generative AI and AI for Science, represented by AlphaFold2 and ChatGPT, as an exogenous technological shock and employ a difference-in-differences design to identify its causal effect on corporate innovation output. Firms ranked in the top 10% of industry AI penetration in 2021 are defined as the treatment group, while the remaining firms serve as the control group. Innovation output is measured by the logarithm of one plus the number of invention patent applications.
The results show that generative AI significantly increases innovation output among firms with stronger pre-existing AI capabilities. After controlling for firm fixed effects, year fixed effects, and firm-level characteristics, the interaction term remains significantly positive, suggesting that the post-shock innovation performance of high-AI-exposure firms improves substantially relative to that of other firms. Further analyses indicate that generative AI not only raises patent quantity but also enhances R&D efficiency by improving the conversion of R&D inputs into innovation outputs, thereby providing evidence against the “strategic innovation” explanation. In addition, the effect remains robust after excluding firms located in first-tier cities, implying that generative AI may exhibit technological spillover and inclusive effects across regions.
May 12th, 2026
Submissions Related to Editors’ Research Are More Likely to Succeed: Evidence from Economics Journals
(En Qi Teo, Toulouse School of Economics, Others)
Examining the editorial process at four leading economics journals, we find that editors were less likely to desk reject manuscripts related to their own research interests. This result is robust to controls for research trends, journal missions and time-varying journal characteristics. We use information on latent manuscript quality revealed by referee reports to ascertain the mechanism---whether editorial favoritism or strategic submissions by authors. Structural estimates point to the latter. Consistent with this interpretation, more productive scholars were disproportionately likely to submit related manuscripts. Finally, editors' research interests did not affect the downstream decision to invite revisions.