A previous version of this working paper was originally published in September 2025.
We develop and rigorously assess an LLM-based pipeline on a new panel of historical county-level vehicle registration tables from early 20th century U.S. state reports. Using human-transcribed gold standard data for evaluation, the pipeline achieves an exact cell match rate of 95.4 percent at approximately 50 times less expense than traditional outsourcing. The pipeline performs well at extracting table structure, where it reduces critical parsing errors from 61.4 percent to 0.35 percent, and in numerical transcription, where it exactly matches 96.7 percent of linked cells and achieves a mean absolute percentage error of 0.7 percent. The pipeline performs on par with human-based category alignment. We also assess pipeline performance in situ with two case studies that analyze the growth and persistence of historical vehicle adoption using common regression models. The significance and sign of effects are identical whether using LLM or gold standard data for all eight models tested, and the coefficient of interest is statistically indistinguishable in six of the eight models.
View the Full Working Paper