The founding of new towns has been at the core of urban planning since the onset of civilisation. In recent times, policymakers have shown renewed interested in the creation of towns to channel regional economic growth. A prominent example is China, where a large-scale urban planning programme began in the 1980s to cope with the pressure of a growing urban population. The idea was to relocate hundreds of millions of rural inhabitants to live in purpose-built towns. Western media has branded these new towns as ‘ghost towns’, as ‘bridges to nowhere’, or as towns in search of populations. However, some have questioned this negative assessment of the planning policy, arguing that it is simply too soon to tell if the Chinese towns are a success or failure (Shepard 2015, van Hooijdonk 2018). The impact of new town foundations would be better understood with access to long-run historical data.
Why do towns thrive?
Historically there has always been a natural reason behind the emergence of towns in certain locations. Research has stressed the importance of either natural advantages, such as fertile soil or suitable trading sites (first-nature geography), or man-made advantages that foster higher local population densities (second-nature geography). Swedish Kings, just like Chinese town planners, decided to do things differently. They created towns in locations with no obvious geographic virtues, hoping that once they were built people would naturally migrate there. This guiding principle assumes that the advantages of a growing population are larger than the geographic fundamentals underpinning town locations. If rulers can coordinate investments on certain spots and make people move there, the towns will persist even if the location is suboptimal. How such towns have persisted, even after their initial advantage has become obsolete, has recently been investigated by a growing literature (Bleakley and Lin 2012, Michaels and Rauch 2016).
Swedish town foundations
In a recent paper, we assess the short-term and long-term impact of a town foundation policy that took place in Sweden between 1570 and 1810 (Cermeño and Enflo 2018). During this period Swedish monarchs founded new towns by granting rural parishes the monopoly rights to trade. As shown in Figure 1, 31 parishes became towns after the crown granted the city charters. While medieval towns only existed in southern and middle Sweden (where the soil quality was of sufficiently high standard to allow populations to live comfortably from their agricultural production; see Figure 2), the Crown founded towns throughout the Swedish territory.
Figure 1 Towns in Sweden, 1570–1810
Figure 2 Parish soil quality from FAO-GAEZ
Towns as a local market shock
Swedish mercantilist restrictions in the early modern period required that any trade had to take place within town walls, where a sales tax was levied. Importantly though, the existence of any traditional infrastructure of regional commerce was not a guiding principle, since the Kings wanted to direct local trade away from the informal rural markets and improve control within the town institutions (Sandberg 1996: 186). Other issues that the Crown considered included strategic military decisions involving the war against Denmark, and triggering the closure of several former medieval Danish foundations in the recently recaptured soil. We test the effect of a change in distance to town markets on extensive and intensive growth in the rural parishes, finding only significant and spatially decaying effects in terms of extensive growth (population and gross production), but no effect in terms of agricultural intensification (production per capita or yields). As a placebo test we use the case of the towns planned by Queen Christina (see Figure 1), who abdicated her throne before anyone could see them established.
Short-run and long-run impact of founded towns
By 1810 the founded towns were barely any bigger than ordinary rural parishes and the policy was considered a failure. This led Eli Heckscher (1963: 50) to argue that “privileges of township were conferred upon a great number of communities which were without any potentialities for urban economic development”. Our results suggest that the reason behind this apparent failure can be explained by the limited production capacity of the local hinterlands. Given the mercantilist trade regulations of the early modern period and the energy requirements for transportation, heating, and nutrition, the towns were organically constrained in their capacity to grow.
Throughout the course of the 19th century, mercantilist town privileges were gradually removed, becoming completely abolished by 1864. Population growth in founded towns took off simultaneously. Figure 3 shows the mean population in founded towns, placebo towns, and a set of counterfactual rural parishes selected for their similar pre-policy characteristics (population density, distance to coast and trading centres, and soil quality). While town foundations follow similar patterns to those of rural parishes as long as town privileges were in place, we find a sharp increase in town population around the mid-19th century. This remarkable take-off can neither be explained by geographical advantages, nor by pre-industrial agglomerations. We argue instead that a combination of sunk investments in transportation technologies (roads, railroads, ports) and non-transportation investments (townhalls, courthouses, banks etc.), along with future expectations, encouraged populations to settle in these locations once the Industrial Revolution had freed the organic constraints. Our results underline how the initial over-endowment of infrastructure helped drive populations to the founded towns up until 1900. Whereas only past population can significantly explain population by 2000. This result speaks to the recent literature about the long-run persistence of town foundations even after their initial advantage has become obsolete. Our paper adds a case of persistence that was created only after the initial advantage (local trading rights) had lost importance.
Figure 3 Mean population in founded towns, placebo (Christina) towns, and counterfactual parishes
We stress that credible coordination efforts based on investments may substitute for agglomeration economies and natural advantages when creating a new town, as long as any natural constraints to population growth have been removed. However, by demonstrating the initial failure of the town foundation policy, we underline the importance of overcoming any natural constraints to urban growth. Finally, and in relation to the debate about the Chinese ‘ghost towns’, our paper shows that ‘artificially’ founded towns may take time to thrive.
In light of our findings, and the examples of the Swedish kings, policymakers should consider how to signal to potential migrants and investors that they’ve committed to improving an area. Simultaneously, they must identify the potential constraints for growth in a post-industrial world. Finally, policymakers should be aware that their choices may trap populations in ‘sub-optimal’ locations for centuries.
Nevertheless, we note that the founded towns have persisted until today and appear neither better nor worse than their medieval counterparts in terms of long-run resilience. Thus, while kings may not be able to create towns that thrive, path-dependency forces certainly can.
Bairoch, P, J Batou and P Chevre (1988), The population of European cities: Data bank and short run summary of results: 800-1850, Geneva.
Bleakley, H and L Jeffrey (2012), "Portage and path dependence," Quarterly Journal of Economics 127(2): 587¬–644.
Cermeno, A and K Enflo (2018), “Can Kings create towns that thrive? The long-run implications of new town foundations,” CEPR Discussion Paper 13392.
Heckscher, E F (1963), An economic history of Sweden, Harvard University Press.
Michaels, G and F Rauch (2017), "Resetting the urban network: 117–2012," Economic Journal 128(608): 378–412.
Sandberg, R (1996), “Urban landownership in early modern Sweden,” in F-E Eliassen and G A Ersland (eds), Power, Profit, and Urban Land: Landownership in Medieval and Early Modern Northern European Towns, Scolar Press: 179-193.
Shepard, W (2015), Ghost cities of China: The story of cities without people in the world's most populated country, Zed Books Ltd.
Van Hoijdonk, R (2018), “The truth about China’s futuristic ghost cities”.