The occupational structure of late Imperial China, 1736-1898 Cheng Yang Today’s enormous global disparities in national average income levels are overwhelmingly a product of differential rates of economic growth over the last 200 years. China was the world’s most developed economy at the beginning of the last millennium, with the flow of technological knowledge overwhelmingly moving west. Following the British Industrial Revolution, Europe, North America and Japan surged ahead and fundamentally diverged from most of the rest of the world. Historians have hence long debated the timing of this “Great Divergence.” Since its inception, the Great Divergence debate has created tremendous interest in comparative global economic history and directed both academic and public attention toward linking modern-day global disparities with our past. However, the discussion still faces serious empirical limitations, particularly on the Chinese side. Faced with this empirical challenge, many scholars have nonetheless made impressive progress on both primary sources and methodology in the past two decades. By producing new estimates of Chinese occupational structure using a challenging, underused but abundant historical source (刑科题本the Xingke Tiben, a collection of trial papers for all homicides in China, 1736-1898), this PhD thesis aims to improve this empirical basis. Although the source contains rich occupational data, it is noted that the resulting dataset cannot be considered as a random sample of all work carried out in the economy due to the source’s inherent biases. Assessments of the source’s representativeness and a series of rectifications of the different biases were thus necessary and carried out in this thesis. The comparison with other sources indicated that the biases in the source are systematic, consistent, and modest. These comparisons suggest that the secondary-sector share is most reliable, while the primary sector is somewhat under-represented and the tertiary sector slightly over-represented. Meanwhile, the secondary-sector labour force has clearly gendered characteristics, with the female’s secondary-sector employment share being much higher than the male’s. This study thus follows a gendered approach, although the current male estimates are much more robust than the female estimates due to the relatively small female sample size. Further, to take full advantage of the source’s spatial coverage and temporal consistency, the thesis attempts regional and long-run analysis. Although the sample is created through random sampling, in the cases of smaller areas or sparsely populated prefectures or the post-1860 periods when the total XKTB available dropped greatly, the sample size can be very small and does not have the strength of statistical significance. What I present in these cases are thus preliminary findings and a possible interpretation. They are hypotheses. The research presented in the thesis makes a contribution to Chinese economic and social history by improving its empirical basis, and, through this, further hopes to contribute to global comparative economic history. Following a concise introduction, chapter one of this thesis provides a discussion of current literature, illustrating the critical need for improving its empirical basis. Chapter two unveils the primary source, demonstrates its quality, richness and challenges as a source for occupational data, and summarises the XKTB dataset created in this thesis. Chapter three provides the methodological core for the full utilisation of the source, as well as addressing potential biases. Using this new evidence, chapter four to seven presents and discusses the new estimates for the occupational structure of eighteenth- and nineteenth-century China and its regions of disparate trajectories, including but not limited to, her economically-advanced region of the Lower Yangtze. Chapter seven summarises the key findings, main contributions and future directions of this research. Preliminary analysis using this new source suggests that both China and the Lower Yangtze lagged well behind England by 1736, and most probably even before 1700, with higher primary-sector shares, considerably lower secondary-sector shares, and higher tertiary-sector shares – the latter explained by the presence of substantial excess labour. Thereafter, it seems that China probably did not undergo major sustained structural change until the twentieth century, as the occupational structures in 1761-70, 1821-30 and 1881-90 are remarkably similar, although we cannot rule out fluctuations in 1771-1820 and 1831-80. Indeed, the timing of the Great Divergence between China and England might need to be pushed back earlier than 1700. Further, these estimates might prove useful later as a new factor to be considered alongside other factors such as man to land ratio, wage level or sectoral productivity in the broader discussion about the diverse paths of development in the modern-era global context, although these evidence in the thesis could not do this justice to them at the current moment.