Spatial structure and labor productivity: Evidence from prefectures in China

Abstract: http://journals.sagepub.com/doi/full/10.1177/0042098018770077#abstract

Scholars have long recognized the importance of the relationship between a region’s economy and its spatial structure. However, there is a lack of valid and reliable evidence on the performance of different spatial structures. There are at present no conclusive answers regarding the question of whether mono- or polycentricity spurs greater economic competitiveness. While proponents of monocentric spatial structure base their argument on the advantages of agglomeration, the promoters of polycentricity believe that a polycentric region can take advantage of agglomeration economies without incurring the same costs as its monocentric counterpart, and borrow size from each other through effective and dense regional networks. Both mono- and polycentricity can have positive or negative effects on regional economic performance depending on which one can provide a better balance between agglomeration advantages and disadvantages. However, the idea of a balanced polycentric urban system has gained widespread currency in planning and territorial development strategies, though this has not been confirmed by systematic empirical studies. Regions in developing countries with a rapid rate of urbanization, like China, are attempting to promote polycentric development. And there is consensus that China is a useful setting for studying the effects of spatial structure on labor productivity.

This paper contributes to the literature by providing new empirical evidence based on Chinese data to inform this unresolved debate. We show robust evidence that spatial structure correlates with labor productivity. More specifically, we show that monocentric spatial structure at the prefectural level in China can contribute to economic efficiency. In addition, this paper sheds light on how the spatial organization of regions affects their economic performance, i.e., the economies of agglomeration in China still play a dominant role. Hence, a monocentric prefecture, often with a much larger central city after controlling for regional size, has advantages in agglomeration economy.

Endogeneity poses a problem for our study and all others in this area. Basic approaches to exploring the causality of spatial structure and economic efficiency may lead to biased estimates because of the potential reverse causality. Likewise, another endogeneity issue may arise from missing variables as we cannot rule out the possibility of the existence of some unmeasured factors which influence both local outcomes and spatial structure. Aiming to navigate around these issues, we deploy a historical instrumental variable (i.e. spatial structure in 1953) which comes from the first census conducted in China. Moreover, in working with sub-national Chinese data, the accuracy and reliability of data demands special attention. Numerous published studies on China have their analyses distorted by misinterpreting China’s population counts. The most widely reported population data for provinces, prefectures, and counties are de jure measures of where people are legally registered to live under the hukou system, rather than measures of where they actually live. In order to avoid this kind of error, we stick to the population census which is the only data resource in China enumerating population on a resident basis.

This paper offers reliable and consistent evidence for the role morphological monocentricity at prefectural scale plays in enhancing productivity, and it also attempts to reveal why prefectures in China with a more monocentric structure perform better in terms of labor productivity. This is a step towards filling the empirical gap in the scholarly debate on the economic performance of monocentric and polycentric spatial structure. Our findings should be taken as an important warning for urban planners and policy makers. Developing small cities to foster prefecture wide polycentricity may come at a price for region-wide efficiency.

Still our results need to be interpreted with a certain dose of caution. Although we have followed the classic measurements of morphological mono- and polycentricity (Pareto and HHI), these measurements still cannot fully capture the full nature of mono- and polycentricity. Second, it is worth noting that our finding is valid only at prefectural scale in China. Given that the economic performance of urban structure is sensitive to geographical scales, our finding cannot generally invalidate the utility of polycentric forms particularly at other spatial levels.

Read the paper on Urban Studies - OnlineFirst here