From Baghdad to London: Lessons from one thousand years of urbanisation in Europe and the Arab world

Maarten Bosker, Eltjo Buringh, Jan Luiten van Zanden, 28 June 2008



Why did the Industrial Revolution begin in northwestern Europe? At the turning of the first millennium, Europe was a backward part of the world economy with low levels of urbanisation and income. But between 1000 and 1800, Europe surged from a backwater of the world economy to its most dynamic region. Understanding this development is a major challenge for economists and economic historians.

Institutions, but which?

The prime hypothesis explaining ‘the rise of Europe’ turns on institutional development. However, which type of institutions gave rise to the ‘European growth miracle’ is a subject of quite some debate.

Douglas North, for example, stresses that socio-political institutions constraining the predatory actions of the state mattered most,1 but this view has been challenged by, amongst others, Avner Greif who maintains that it was contracting institutions that mattered most. These facilitated economic exchange – such as developed by Italian merchants and cities – and this is what made the difference.2 Daron Acemoglu and Simon Johnson, making the same distinction between ‘vertical’ socio-political institutions (property rights) and ‘horizontal’ economic institutions (contracting institutions), present evidence for the vertical institutions mattering more when it comes to long-term economic growth.3

Contributing to this debate in a recent paper, we compare the long-term development of the urban systems of Europe and the Arab world between 800 and 1800.4 On the basis of a new and very large dataset of individual cities in Europe, North Africa and the Middle East, we are able to assess the importance of various factors that drove urban expansion. Consideration of factors such as geography, religion, and institutional provides some answers to the question why, during this millennium, the urban and economic centre of gravity moved from Iraq, or more generally the Arab world, to Europe and the shores of the Atlantic in particular.

More specifically, we are able to provide insights into the relevance of the economic institutions governing exchange on the one hand and socio-political institutions on the other in a way that explains Europe’s rise and eventual overtaking of the Arab world in terms of economic prosperity.5

Institutions governing economic exchange

We use the number and size of cities as our measure of economic performance and focus on the existence and development of positive feedbacks between cities. In particular, we study whether and to what extent cities profit from other cities in the sense that the presence of many, large, cities in a city’s neighbourhood seemed to spur its growth. A key part of our results concerns the way that these neighbourhood effects changes over time.

Our maintained hypothesis here is that a powerful neighbourhood effect signifies that the institutions governing exchange are efficient, so the growth of one city stimulates expansion in others. Our analysis shows some striking results.

From about 800 to 1200, the level of positive spillovers among Muslim cities was high – suggesting that the institutions governing exchange in the Arab world were efficient.

Around 800, the position of Arab cities was quite favourable. The Arab world enjoyed a highly integrated urban system reaching from Cordoba to Baghdad. Transaction costs were low as the region was politically united, shared a common language and a common (Islamic) legal system that included a number of institutions promoting exchange (such as the rule of using written contracts). There was also a highly efficient means of transportation between urban centres in the form of the caravan routes.6

During this same period, the neighbourhood effect did not exist in Europe. Europe did not have an integrated urban system, perhaps due to very high transport and transaction costs following the break-up of the Carolingian Empire around 900. Europe fragmented into a large number of political entities. Merchants spoke many different languages and a variety of legal regimes regulating exchange (Roman law in the south, customary law in the north) were in place.

As this period drew to a close, fundamental changes rocked both Europe and the Arab worlds – but with very different economic effects. Our empirical results for the 1000 to 1500 period show that Europe and the Arab world switched places in terms of their respective neighbourhood-effects.

Trading place: The year 1100

From 1100 onwards, Europe enjoyed an efficient urban system with positive feedbacks between cities (based much more on sea and river trade) in spite of the fact that it remained politically fragmented. In the Arab world, by contrast, the neighbourhood effects disappeared. There the break up of the Abbasid Caliphate was eventually followed by a new empire, the Ottoman Empire. To some extent, this took over the role of its predecessor – but without restoring the efficient system for economic exchange that was present during the Golden Age of Islam.

Interaction between Europe and the Arab world

Another interesting aspect of our findings is the importance of religion. Muslim cities interacted strongly positively with other Muslim cities, and Christian cities with Christian cities, but we find hardly any evidence of positive feedbacks between the two sets. This suggests that different institutions regulated exchange in these two worlds, and that exchange over religious borders was handicapped by much higher transaction costs, and oftentimes outright hostility, compared to exchange within each urban system.

Also the socio-political developments in Europe and the Arab world can help to explain the observed differences in evolution of the two urban systems.

Differences between the two urban systems

There were striking differences between the two urban systems that reveal interesting insights into the socio-political situation in the two regions. Cities in the Arab world were on average much larger than those in Europe, and the size of the “primate” city – the megapolis such as Baghdad, Damascus, Cairo or Istanbul – was much bigger; a fact that is indicative of a predatory state and low trade openness.7 Europe, on the other hand, developed a very dense urban system, with relatively small principle cities. Big cities in Europe were quite often located near the sea, being able to optimally profit from long-distance trade, whereas the largest cities in the Arab world were almost all inland.

European ‘producer cities’, Arab ‘consumer cities’

The sociologist Max Weber introduced a distinction between ‘consumer cities’ and ‘producer cities’. Using this classification, Arab cities were – much more than their European counterparts – consumer cities.8

The classical consumer city is a centre of government and military protection or occupation, which supplies services – administration, protection – in return for taxes, land rent and non-market transactions. Such cities are intimately linked to the state in which they are embedded. The flowering of the state and the expansion of its territory and population tend to produce urban growth, in particular that of the capital city.

In Europe cities are instead much closer to being producer cities. The primary basis of the producer city is the production and exchange of goods and commercial services with the city’s hinterland and other cities. The links that such cities have with the state are typically much weaker since the cities have their own economic bases. It is this aspect that accounts for the fact that Arab cities suffered heavily with the breakdown of the Abbasid Empire, while European cities continued to flourish despite political turmoil.

Between 1000 and 1300 Europe acquired an urban system dominated by typical producer cities, which prospered in spite of Europe’s political fragmentation. In fact, this fragmentation was strongly enhanced by the rise of independent communes – city-states, or cities with a large degree of local authority – which form the core of the political system of Europe’s urban belt stretching from Northern Italy to the Low Countries. Indeed, we still find this pattern in the so-called ‘Hot Banana’ – the industrial agglomeration that stretches from the southern UK to the Netherlands, through Germany and down to northern Italy.

Arab cities at this time were, by contrast, heavily influenced by strong, predatory states that could, and oftentimes did, impose a heavy tax or military burden on the cities in their realms. Under these predatory regimes it was typically only the capital city thrived, with this honour shifting from Baghdad to Damascus, Fez, Cairo and finally to Istanbul.

Why did Europe overtake the Arab World in the millennium between 800 and 1800?

Arab cities were part of the ‘predatory’ structure of the state. When the region was unified under the Abbasids, this worked well and the region experienced its ‘Golden Age of Islam’. Efficient institutions regulated exchange, allowing high levels of commercialisation and urbanisation. When state systems disintegrated, so did the urban system and the underlying commercial networks.

In Europe, after a period of disintegration, a different urban system more or less independent of ‘predatory’ states emerged. These managed to claim their own niche in the political economy of the period and developed increasingly effective ways of organising commercial exchange in spite of the fragmented political system.

It is this development in Europe of an economically well integrated urban system largely independent of large territorial states, spurred on by the effect of the Great Discoveries that can explain to a large extent why London, an economic backwater in 800, was able to overtake Baghdad, the formerly thriving capital of the Abbasid caliphate.


1 North, D.C. (1981). Structure and change in economic history. New York: Norton.

2 Greif, A. (2006). Institutions and the path to the modern economy: lessons from Medieval trade. Cambridge University Press.

3 Acemoglu, D., and S. Johnson (2005). “Unbundling institutions.” Journal of Political Economy 113, 949-995.

4 Bosker, M., E. Buringh, and J. Luiten van Zanden (2008). From Baghdad to London: The Dynamics of Urban Growth in Europe and the Arab World, 800-1800. CEPR DP 6833.

5 Timur Kuran argues that economic institutions were weaker in the Muslim world, while others argue that the European socio-political institutions were states fundamentally different – less predatory and more constrained. Kuran, T. (2003). “The Islamic commercial crisis: institutional roots of economic underdevelopment in the Middle East.” Journal of Economic History 63, 414-447.

6 See also Ronald Findlay’s and Kevin O’Rourke’s Vox column “Lessons from the history of trade and war” 10 March 2008.

7 Ades, A.F. and , E.L. Glaeser (1995). “Trade and circuses: explaining urban giants.” Quarterly Journal of Economics 110, 195-227.

8 Weber, M. (1958). The City. Translation and edited by Don Martindale and Gertrud Neuwirth. New York, The Free Press.

Topics: Development, Economic history
Tags: Arab world, economic history, Europe, institutions, urbanisation


This article touches on what

This article touches on what I think are the three deepest questions in all social science:

* Since 1550 or so, a scientific, commercial, and industrial revolution has occured. Why did it occur when and where it did?

* Why have some nations embraced those revolutions quickly and enthusiastically (the North Atlantic community), others enthusiastically but with considerable delay (Asian Tigers), unevenly (Latin America, India), and even not at all (sub-Saharan Africa)?

* Why did imperial China (Sung dynasty) and the Islamic caliphates get tantalizingly near to those revolutions, only to experience a sort of "failure of nerve"?

For years, my preferred tentative answers to these questions were the writings of Douglass North, and the 1986 monograph by Rosenberg and Birdzell. But I now suspect that their proposed explanations run aground on the recent experiences of China and India. While India is a common law jurisdiction, its legal system is tatty and inefficient. And Chinese growth post-Mao has taken place in what has largely been a business law vaccuum.

Maarten Bosker

Assistant Professor, Erasmus University Rotterdam

Postdoctoral researcher, Utrecht University

Professor in economic history at Utrecht University and senior researcher at the International Institute of Social History in Amsterdam