Can LLMs Credibly Transform the Creation of Panel Data from Diverse Historical Tables?

TL;DR

This paper demonstrates that multimodal LLMs can effectively digitize complex historical tables, drastically reducing costs and errors, and producing data comparable to human validation for economic analysis.

Contribution

It introduces a validated LLM-based pipeline for digitizing historical data, significantly lowering costs and errors compared to traditional methods.

Findings

Error rate reduced from 40% to 0.3%

Data matches gold standard with 98.6% R^2

Enables new economic analyses with historical data

Abstract

Multimodal LLMs offer a watershed change for the digitization of historical tables, enabling low-cost processing centered on domain expertise rather than technical skills. We rigorously validate an LLM-based pipeline on a new panel of historical county-level vehicle registrations. This pipeline is 100 times less expensive than outsourcing, reduces critical parsing errors from 40% to 0.3%, and matches human-validated gold standard data with an $R^2$ of 98.6%. Analyses of growth and persistence in vehicle adoption are statistically indistinguishable whether using LLM or gold standard data. LLM-based digitization unlocks complex historical tables, enabling new economic analyses and broader researcher participation.