Market disintegration as a pre-cursor to the Great Divergence

Daniel Bernhofen, Markus Eberhardt, Jianan Li , Morgan Stephen

22 June 2016



China is in the midst of perhaps the single most significant transition in its remarkable economic journey – the leap from a labour-intensive to a genuinely knowledge-based economy. Its ability to bring about this transformation will shape the country’s broader trajectory for decades to come.

History offers no clear-cut lessons, but there are certainly instances of China standing at a similarly momentous crossroads. The reforms of the late 1970s represent the most obvious example of effective change; the failure to industrialise at the start of the 19th century provides maybe the most striking case of something altogether less successful (Brandt et al. 2014).

The circumstances surrounding the Great Divergence, as it is known, when Europe pulled ahead of China and the rest of the world, remain a source of lively academic debate (see Vries 2015 for a recent survey). The issue is notably controversial from a Chinese perspective. How could it be that China, which had produced some of the defining inventions of the early modern age, failed to develop in tandem with Western Europe in general and Britain, the cradle of the Industrial Revolution, in particular?

A long-held view maintained that Western Europe was characterised by integrated markets, which had taken root because of state-supported property rights institutions, while in China, despite the unified political system created by a dynastic empire, no unified market had emerged (see Li 2000 for a discussion of this early literature). In recent years the prevailing view has been that early modern China’s economy was not only on a par with that of Western Europe but closer to the neoclassical ideal of a market economy (Pomeranz 2000). A major weakness of this ‘revisionist’ argument is poor and fragmentary data to underpin it. Seminal quantitative work by Shiue and Keller (2007) supports the revisionist view that market integration in Southern China was comparable to Western Europe on the eve of the Industrial Revolution. These authors employ cointegration analysis to investigate grain price behaviour in Qing China during the 18th century and compare their results to those from grain prices in European markets. The continuing rise in prominence of their findings has helped transform what was once a revisionist view into a now conventional view that with regard to market integration “all was well and good with China in the eighteenth century” (Sng 2014, p. 108), thus ruling out this factor as an explanation for the Great Divergence.

Our recent work has come to challenge this conventional wisdom (Bernhofen et al. 2016a, 2016b). We find a significant decline in Chinese market integration during the 18th century, in contrast to stable levels of integration from our analysis of European grain price data.

We use the same Chinese data as Shiue and Keller (2007), albeit over an expanded time horizon and supplemented with a sample of Northern Chinese prefectures. We moreover draw on the full monthly data series for grain prices in China (not a subsample of selected months) and seek out European grain price data at the same frequency for comparison.1 This allows us to examine how market integration changed over our sample period by specifying a rolling window of analysis.

In a short research note, we adopt identical methods to Shiue and Keller (2007) and focus on the Southern Chinese sample originally investigated by these authors (Bernhofen et al. 2016a). Our rolling window analysis suggests that Chinese markets disintegrated over time with the most advanced regions in the Yangtze Delta showing the most significant drop between the mid-18th and early 19th centuries.

There are a number of caveats attached to the Engle-Granger cointegration methodology applied in this and Shiue and Keller’s (2007) analysis. Prime amongst these are the requirement of non-stationary price series, as well as the empirical setup of pairwise price analysis more generally, assuming cross-section independence across market pairs. These potential shortcomings are addressed in our recent CEPR Discussion Paper (Bernhofen et al. 2016b). We build on a theoretical framework for grain price behaviour which postulates that local grain prices are jointly determined by factors such as local and global weather shocks and by the market’s relative position in the existing trading network.2 Accordingly, the investigation of price convergence between any two markets can be misleading unless account is taken of the ‘outside options’ available to consumers and traders, as well as the relative attractiveness of these opportunities. Furthermore, bilateral price movements in reaction to common shocks may make price pairs appear to co-move, although the cause for this co-movement is entirely unconnected to spatial arbitrage (e.g. widespread flooding or locust plague).

Figure 1 depicts the time paths of market integration in Chinese regions and European economies using the half-life (in months) as the metric for the degree of market integration. Although China’s level of market integration was still comparable to Europe’s in the 1740s, a very substantial gap had opened up by the end of the 18th century.  Market disintegration had already set in during the ‘Golden Age’, the reign of Emperor Qianlong, whether we focus on the country as a whole, the economically most advanced region of the Lower Yangtze region, or other parts of the empire.

Figure 1 Grain price divergence in Chinese regions and European economies

Notes: We plot the half-lives implied by our analysis of grain price convergence in Chinese regions and European Economies. The half-life measures the number of time periods (here in months) until half the effect of a shock has dissipated. We omit confidence intervals to aid illustration. The longer the half-life, the slower is price convergence to the equilibrium. Results for Belgium, England, France, and North China are derived from wheat prices, those for Southern Chinese regions are for rice prices. Results are derived from rolling windows of 20 (Chinese and English monthly data) and 10 (French and Belgian monthly data) years’ length.
Source: Bernhofen et al. (2016b).

The decline in market integration we uncover is substantial, but is it perhaps too big to be credible? Collapses of market integration such as that we describe are typically only found in times of widespread natural or man-made disaster, whereas the 18th century was by Chinese standards a period of relative calm. However, reviewing the discussion of the significant difficulties associated with transporting low value-to-weight goods like grain over vast distances (Evans 1984, Eastman 1988, Cheung 2008, Kim 2008), one should perhaps be more surprised that high levels of market integration had ever been achieved in early modern China.3

Our empirical evidence of a remarkable decline in Chinese market integration during the 18th century is compatible with a range of historical narratives which suggest four intertwined factors that potentially explain a decline in market integration (detailed references can be found in the Bernhofen et al. 2016b).

  • First, population pressure on arable land, especially in grain surplus interior provinces, which led to a decline in the grain available for trade between these and the advanced regions on the Eastern Seaboard;
  • Second, environmental degradation that affected farming and transport, primarily stemming from the inherently instable water control systems which were in an adversarial relationship with the environment;
  • Third, technological factors, namely the absence of significant advances in transport technology or infrastructure and agricultural technology;
  • And lastly, a decline in the capacity of the Qing state to invest in development and promote further market integration, in part driven by fiscal weakening and in part by ‘grain protectionism’ among local officials whose paramount concern was to ‘nourish the people’ to avoid civil strife.

In the early 1970s, Mark Elvin sought to explain the Great Divergence using a concept that he called ‘the high-level equilibrium trap’. The basic implication of this construct was that pre-industrial China, rather than somehow lacking the attributes that propelled Western Europe into uncharted economic territory, was a victim of its own success. Elvin proposed that China’s pre-industrial economy had reached a stage at which supply and demand were so finely balanced and labour so cheap that investments in capital-intensive technology to enhance efficiency would not be profitable (Elvin 1973). In contrast, the constraints (land and labour) and imbalances that beset the likes of Britain gave every incentive to invest and innovate in new technologies. The result: China stalled and Europe moved on ahead.

While our study cannot directly speak to this argument related to the Great Divergence, we can suggest that China’s early modern economic success –  coupled with administrative and political stasis – as characterised by Elvin’s ‘high equilibrium trap’, helps to explain our findings of market disintegration. However, a rigorous investigation evaluating the explanatory significance of these potential determinants of market disintegration is left for future work.       


Andrade, P. and M. Zachariadis (2016), “Global versus local shocks in micro price dynamics.” Journal of International Economics, 98: 78-92.

Beck, G. W., K. Hubrich and M. Marcellino (2015), “On the Importance of Sectoral and Regional Shocks for Price-Setting.” Journal of Applied Econometrics, forthcoming.

Berg, K. A. and N. C. Mark (2015), “Third-country effects on the exchange rate.” Journal of International Economics, 96(2): 227-43.

Bernhofen, D., M. Eberhardt, J. Li and S. Morgan (2016a), “The Evolution of Markets in China and Europe on the Eve of Industrialisation.” Unpublished mimeo, May 2016, available online at

Bernhofen, D., M. Eberhardt, J. Li and S. Morgan (2016b), “Assessing Market (Dis)Integration in Early Modern China and Europe.” CEPR Discussion Paper 11288, May 2016.

Brandt, L., D. Ma, and T. G. Rawski (2014), “From Divergence to Convergence: Reevaluating the History behind China's Economic Boom.” Journal of Economic Literature 52(1): 45-123.

Cheung, S.-W. (2008), The Price of Rice: Market Integration in Eighteenth-Century China. Bellingham: Western Washington University Press.

Eastman, L. E. (1988), Family, Fields and Ancestors: Constancy and Change in China’s Social and Economic History, 1550-1949. Oxford University Press.

Elvin, M. (1973), The pattern of the Chinese past: A social and economic interpretation. Stanford University Press.

Elvin, M. (2004), The Retreat of the Elephants: An Environmental History of China. Yale University Press.

Evans, L. (1984), “Junks, Rice, and Empire: Civil Logistics and the Mandate from Heaven.” Historical Reflections, 11(3): 271-313.

Head, K. and T. Mayer (2014), “Gravity equations: Workhorse, toolkit, and cookbook.” In: Gita Gopinath, Elhanan Helpman, and Kenneth Rogoff (eds) Handbook of International Economics, Vol. 4. Amsterdam: Elsevier North-Holland.

Kim, N. (2008), “Transport im China der spaeten Kaiserzeit, 1500-1900: Eine Bestandsaufnahme.” In: Stieferle, Rolf Peter (Ed) Transportgeschichte. Berlin.

Li, L. M. (2000), “Integration and Disintegration in North China’s Grain Markets, 1738-1911.” Journal of Economic History, 60(3): 665-99.

Pomeranz, K. (2000), The Great Divergence: China, Europe, and the Making of the Modern World Economy. Princeton University Press.

Shiue, C. H., and W. Keller (2007), “Markets in China and Europe on the Eve of the Industrial Revolution.” American Economic Review, 97(4): 1189-1216.

Sng, T H. (2014), “Size and Dynastic Decline: The Principal-Agent Problem in Late Imperial China 1700-1850.” Explorations in Economic History, 54: 107-27.

Vries, P. (2015), State, Economy and the Great Divergence: Great Britain and China, 1680s-1850s. London: Bloomsbury.


[1] We use monthly rice and wheat prices in up to 221 Chinese prefectures (1740-1820), covering all but one of the provinces of Qing China proper. Our European data from England (1770-1820), Belgium (1765-1794) and France (1800-1872) is also at monthly intervals, while additional annual data is from Germany (1700-1800) and a cross-European sample of markets (1700-1820).

[2] Our conceptual framework incorporates a mechanism akin to multilateral resistance in the gravity model of bilateral trade flows (Head and Mayer, 2014, for a recent survey), the third country effect in the analysis of exchange rate movements (Berg and Mark, 2015), and the distinction of global and local shocks in recent work on (micro) price dynamics (Beck, Hubrich and Marcellino, 2015; Andrade and Zachariadis, 2016).

[3] The ‘cost of rice to transport rice’ in land transport zones amounted to 6-7% a day (Evans, 1984: 286), indicating that “self-sufficiency was of necessity the dominant economic reality” (ibid, 296) in these (Northern) parts of the country. In Southern China, the notion that the maladministration of rivers and canals, and ensuing decline in inland waterway navigability had already set in during the Qianlong era is also widely documented (inter alia Li, 2000; Elvin, 2004).



Topics:  Economic history

Tags:  Great Divergence, China, market disintegration, grain prices

Professor, School of International Service, American University, Washington, DC

Assistant Professor, School of Economics, University of Nottingham; CEPR Research Affiliate

Postdoctoral Researcher, School of Economics, University of Xiamen

Professor of Chinese Economic History, University of Nottingham; Dean of Social Sciences, University of Nottingham Ningbo China campus