Toward the great divergence: Economic growth in the Yangzi Delta, 1393–1953

The dissertation was completed in 2023 at the University of Oxford, under the supervision of Professor Stephen Broadberry. This work was partly funded by Designated Studentship Scholarship of the University of Oxford and PhD Bursary Scheme of Economic History Society.

The full dissertation can be accessed through the Library of the University of Oxford.

This dissertation examines the economic growth process of the Yangzi Delta, one of China's most developed sub-regions, from 1393 to 1953. For the first time in the literature, an annual-frequency GDP series is constructed using an output-based approach. The new GDP data shed light on how the Yangzi Delta's economy evolved over seven centuries.

From the perspective of Chinese economic history, the nine Taihu Lake-centred prefectures studied in this dissertation had a GDP per capita well above the national average for a long time (~170%). However, before the mid-19th century, the local economy appeared to be trapped in what Mark Elvin (1973) refers to as the ‘high-level equilibrium trap.’ While the GDP per capita exhibited a Malthusian-style growth pattern, numerous technological and managerial innovations allowed it to remain at a high level.

In the context of global economic history, this dissertation incorporates the Yangzi Delta's directly calculated GDP into global GDP comparisons for the first time. Since the late 14th century, the Yangzi Delta maintained a GDP per capita level comparable to that of the most developed European economies, including those of Italy, the Netherlands, and Britain. However, by the late 17th century, this neck-and-neck growth pattern came to an end. From that point onward, the leading European economies surpassed the Yangzi Delta in terms of GDP per capita, and the gap widened irreversibly. In short, the Great Divergence is believed to have begun at the end of the 17th century.

This dissertation is divided into two sections: Measuring Economic Growth (Chapters 2–4) and Analysing Economic Growth (Chapters 5 and 6). To measure economic growth in the Yangzi Delta, I first examine the region's population and cultivated land in Chapter 2. Although many macro-level population and land-use figures exist in Chinese historical records, as Ho (1959) pointed out, these figures cannot be used directly. The primary focus of Chapter 2 is to reconstruct data on population and cultivated land while considering institutional factors. From an institutional perspective, I delve into the traditions and practices of China's statistical system, explaining when the available data can be considered reliable and when it cannot. By carefully assessing the boundaries of the historical data, I am ultimately able to construct a complete set of benchmark data on population and cultivated land in the Yangzi Delta. In the long run, decreasing cultivated land per capita was the dominant characteristic of the region's demographic and land-use patterns.

Chapter 3 examines the historical agricultural economy of the Yangzi Delta, with the primary objective of constructing a long-term agricultural output dataset. This chapter begins by calculating grain yields per unit of land. I collect nearly 3000 samples from the Yangzi Delta and surrounding areas to establish a long-term grain yield trend series. To capture short-term annual fluctuations, I gather over 20,000 descriptive records from local gazetteers and convert these textual records into ranked series using standardised methods from historical climate studies. The integration of long-term and short-term data allows me to generate an annual-frequency grain yield dataset. Next, I separately examined crop types, cultivated area, and non-crop sectors, combined with grain yields, and ultimately derived an annual-frequency agricultural value-added series.

Chapter 4 focuses on the industrial and service sectors, beginning with the central question: how did the Yangzi Delta's economic structure evolve over time? This question arises primarily because agricultural output per capita declined over time. If non-agricultural output per capita did not increase significantly, GDP per capita in the Yangzi Delta would have inevitably declined in the long run. In other words, sustaining economic growth in the region would have required substantial structural shifts towards non-agricultural sectors.

First, I analyze qualitative evidence. Most of the Yangzi Delta's non-agricultural sectors did not appear to have expanded significantly faster than the population. Indeed, the urbanisation rate—often used as an indirect measure of economic structural transformation—suggests that urban population growth did not outpace overall population growth. If these industries only grew at a rate proportional to the population, non-agricultural output per capita would have remained stagnant at best. When agricultural output per capita declined while non-agricultural output remained unchanged, structural changes in the economy inevitably took place, leading to what appears to be a process of proto-industrialisation. However, this transformation was insufficient to prevent a further decline in GDP per capita, as non-agricultural growth did not significantly exceed population growth over the long run. Next, based on the overall trend of non-agricultural sector growth derived from qualitative evidence, I conducted a detailed estimation of industrial and service sector output using the output-based approach.

The second section, Analyzing Economic Growth, begins in Chapter 5with an investigation of the Yangzi Delta's economic trajectory over several centuries. Using the newly constructed GDP dataset, I explore several key macroeconomic questions: (1) Was the Yangzi Delta the most developed region in China? (2) What were the long-term trends and turning points in the Yangzi Delta's economic growth process? (3) Did the Yangzi Delta experience a Malthusian trap? (4) Can Boserupian or Smithian growth models help explain the region's economic growth patterns?

Chapter 6 places the Yangzi Delta's economic growth process in a global context, with a particular focus on addressing the Great Divergence Debate. Based on directly estimated GDP data for the Yangzi Delta, I conclude that the Great Divergence occurred at the end of the 17th century when compared to the most developed European economies. I also conduct robustness tests to verify the dating of the Great Divergence and explore several counterfactual scenarios, considering what might have happened if the divergence had not occurred. All available evidence suggests that the Great Divergence took place at the end of the 17th century.

Beyond the Great Divergence debate, I compare the Yangzi Delta's economy with that of most of Europe and argue that the greatest economic challenge for the Yangzi Delta was not merely the divergence itself but rather the region's failure to capitalise on the broader wave of economic growth that swept across Eurasia during the 19th century. In fact, apart from England, the first long-term and sustained per capita GDP growth in most economies around the world almost all began in the 19th century. In other words, although the Great Divergence in the 17th century caused the Yangzi region to lose its position as the global economic leader, as long as it could take off alongside other economies in the 19th century, it would not be considered too late (such as Japan). However, the per capita GDP of the Yangzi region showed no improvement in the 19th century and was ultimately left behind by the wave of global growth. This failure directly contributed to the Yangzi Delta lagging far behind even the poorest European economies by the early 20th century.