An Empire of Their Own?

A Longitudinal Dataset of Cotton Production and Industry in Republican China, 1919-1937

Abstract

This article introduces a new, hand-collected dataset documenting cotton production and industry in Republican China from 1919 to 1937. The dataset provides granular, county-level data on cotton acreage, yield, and the cultivation of both “native” and “foreign” (American upland) cotton varieties, offering a detailed perspective on the spatial and temporal dynamics of agricultural change during this critical period. This foundational data is augmented by three additional geospatial datasets documenting the locations and attributes of key economic and scientific nodes: (1) modern cotton mills, including their industrial capacity; (2) agricultural experiment stations; and (3) treaty ports and commercial centers. The dataset was constructed through a meticulous process of transcription, validation, error correction, and georeferencing, drawing from a variety of primary sources, including the publications of the Chinese Cotton Millowners Association and other contemporary industry reports. By detailing the dataset’s construction and acknowledging its limitations, this paper makes the case for its use as a unique tool to investigate the core mechanisms of agricultural modernization. Its multi-layered structure enables researchers to test hypotheses about biological innovation and institutional support, and to place China’s experience within the global history of cotton. Ultimately, it offers a granular case study of a nation striving to build its own industrial base, thereby complicating classic core-periphery models of development.