Trade Integration, Market Size, and Industrialization: Evidence from China's National Trunk Highway System

Abstract

Large scale transport infrastructure investments connect both large metropolitan centers of production as well as small peripheral regions. Are the resulting trade cost reductions a force for the diffusion of industrial and total economic activity to peripheral regions, or do they reinforce the concentration of production in space? This paper exploits China's National Trunk Highway System as a large scale natural experiment to contribute to our understanding of this question. The network was designed to connect provincial capitals and cities with an urban population above 500,000. As a side effect, a large number of small peripheral counties were connected to large metropolitan city regions. To address non-random route placements on the way between targeted city nodes, I propose an instrumental variable strategy based on the construction of least cost path spanning tree networks. The estimation results suggest that network connections led to a reduction in GDP growth among no ntargeted peripheral counties due to reduced industrial output growth. Additional estimation results present evidence that appears consistent with the existence of core-periphery effects of trade integration as found in increasing returns trade theory and economic geography.

An impact evaluation of investment in infrastructure: The case of a railway connection in Uzbekistan

“…A possible explanation could be that the decisions of businesses in the agricultural sector may have been affected by considerations regarding the connection by rail from the region in which the infrastructure was located, and its neighboring regions, to the central part of the country. A similar result is documented by Faber (2014) where the provision of the National Trunk Highway System network in the PRC led to reduced output growth among peripheral regions, rather than diffusing production in space. Table 9 shows the results of the estimation of the difference-in-difference coefficient for when the outcome variable is industrial output.…”

“…With the assumption of a common time path and the availability of pre-treatment and post-treatment data on outcome variables of interest, researchers can estimate the degree of departure from the counter-factual trajectory, which can be attributed to the provision of treatment, in this case some kind of infrastructure. In particular, the results of the impact evaluation of the People's Republic of China's National Trunk Highway System by Faber (2014) suggests that the network connections led to a reduction in GDP growth among peripheral counties, which were non-targeted or lay outside the network system. Similarly, Gonzalez-Navarro and Quintana-Domeque (2010) presented evidence on the impact of infrastructure on poverty reduction, where within 2 years of the infrastructure provision in the form of paved roads, households reacted with increased consumption of durable goods and the purchase of motor vehicles.…”

“…Although the body of literature covering middle-income countries has started to grow in recent years, particularly studies related to the People's Republic of China (Faber 2014;Wang and Wu 2012;Ward and Zheng 2013) and some East Asian countries (Yoshino and Nakahigashi 2004) mainly driven by their remarkable growth and improvement in conditions with regard to data dissemination, empirical literature examining either the role of infrastructure and its differential impact on economic outcomes in the context of Central Asian countries is as yet limited. Our paper attempts to shed light on the performance of infrastructure, focusing on the case of a railway connection in Uzbekistan.…”

“…Our third empirical context is based on empirical evidence obtained from the literature on transportation mode choice (Wang et al 2013) and connectivity (Faber 2014). The first group of authors analyzed interstate freight mode choices between truck and rail in Maryland, United States, and found that longer distances contribute positively to the use of rail as a means of transportation.…”

Geographic determinants of China's urbanization

“…Ongoing work focuses on the e ects of transportation infrastructure on city population decentralization (Baum-Snow et al, 2015) and on incomes (Banerjee et al, 2012;Faber, 2014).…”