China's Regional Development

Review and Prospect

Ming Lu, Zhao Chen, Zhu Xiwei, Xu Xianxiang

• Publisher: Routledge
• Year: 2013
• Topic: Economics
• eBook ISBN:
  9781135016333
• Subtopic: Economic History
• Index: Economics

China is a large developing economy and it has been deeply involved in globalization since its economic reform and opening-up. Simultaneously, China has seen a significant change in the spatial distribution of economic resources, especially capital and labor. In the recent 10 years, economists have made significant progress in both theoretical and empirical studies on related topics.

The book provides an overview on the existing literature and current policy debates on what we have known and what we have misunderstood. This book includes an analytical framework of the New Economic Geography (NEG) with political economy to help us understand China's regional development issues. The book of 10 chapters is organized into four thematic sections. The first section is a theoretical discussion on the relationship between economic agglomeration and interregional balanced development. The second section is a political economy analysis on regional and urban-rural development. The third section provides a summary on empirical literatures concerning from market segmentation and institutional barriers to production factor mobility. The final section consists of four empirical chapters on the relationship between agglomeration and balance, which is the core of relevant policy debates.

The book argues that for China to achieve both efficiency and balance for regional development, China may need to reform its systems which constrain production factors mobility. This book is a valuable reference for readers who are interested in spatial economics and the Chinese economy, especially its regional and urban development.

Part I

Economic agglomeration and interregional development

1 Agglomeration and regional coordination

Xiwei Zhu and Yongliang Tao

1.1 Introduction

Spatial agglomeration of economic activity is a prominent fact with modern economic growth: (1) From a time dimension perspective, economic output grows continuously (Romer, 1986); (2) from a spatial dimension perspective, economic activity constantly agglomerates toward a few regions (Krugman, 1991a). Mainstream economists have studied the former for a long time, while research into the latter has only gradually formed and developed since the 1990s. As a matter of fact, spatial agglomeration of economic activities is a very widespread phenomenon in the real world, which can be observed at various levels. At the global level, in the 30 years since the 1980s, the aggregate economic output of NAFTA countries, 27 EU countries, and East Asian countries has always stayed above 70 percent of world output.¹ At the national level, on the one hand, the economic activity of developed nations such as the US and Japan shows a distinct trend toward spatial agglomeration. Take Japan as an example. In 2001, three megalopolis economic rims that occupy only 10.4 percent of Japan’s land mass accounted for 48.6 percent of Japan’s population, 66.2 percent of its GNP and 68.9 percent of its industry (Liu, 2006). On the other hand, according to our calculations, as the world’s largest developing country, the spatial agglomeration of economic activities in China has constantly increased in the 30 years since reform and opening up. Using the Herfindahl–Hirschman Index (henceforth, HHI) of mainland China’s interprovincial industrial output as an example, in 1978, the HHI of 31 provinces (excluding Taiwan, Hong Kong, and Macau; although this refers to provinces, directly administered municipalities and autonomous regions, the short version “province” will be used here) was 557.31; in 2009 it rose to 729.44, an increase of 30.89 percent.² In 2009, Jiangsu, Zhejiang, and Shanghai, which are located around the Yangtze Delta and occupy less than 3 percent of land mass, produced 22.14 percent of the industrial output; realized 40.87 percent of exports and 43.06 percent of imports; and generated 21.29 percent of GDP in China. These empirical facts indicate that economic agglomeration is a common phenomenon that occurs during economic development worldwide, and not only is this reflected at the global level, but also at the regional level within the world’s major economies.

However, when it comes to mainland China after reform and opening-up, accompanying spatial agglomeration of economic activities is the significant expansion of regional disparities. Especially since the early 1990s, the trend of increasing regional disparities is very evident and becomes an important reason for the increase in regional income disparity (see, e.g. Chen and Xu, 2004; Guan et al., 2006; Wan et al., 2007; Wan et al., 2005; Wang and Fan, 2004; Wei and Sun, 2004; Xu and Li, 2006).³ After entering the twenty-first century, in order to promote coordinated economic development among regions and decrease regional disparity, the central government successively implemented regional development strategies such as the “Grand Western Development Program”; “Revitalization of Northeast-China Old Industrial Bases” and “Rise of Central China”, etc., slowing down the expansion rate of regional disparity compared to the 1990s (Wei and Sun, 2004; Xu and Li, 2006). Furthermore, Liu et al. (2009) stated that, after 2004, regional disparity measured by the ratio between the per capita income of Eastern regions and Central and Western regions demonstrated a slightly downward slope.

With regard to the debate on economic agglomeration and regional coordination in China, we believe the first question that needs to be answered is: Is the degree of spatial agglomeration in China’s economy too high? Moreover, as far as China is concerned, is agglomeration of economic activities in coastal areas mainly responsible for increasing the per capita income gap between regions? What effects have a series of regional development strategies represented by the “Grand Western Development Program” had on reshaping China’s regional economic geography? Finally, is there a proper way to realize both economic growth and regional coordination? To this end, the next section of this chapter will systematically review New Economic Geography’s (NEGs) theoretical research on the spatial agglomeration of economic activities and regional coordinated development; section 1.3 will reply to, and comment on, the above questions in view of international experience, and proposes policy implications for the future development of China’s regional economy; the final section will summarize the main conclusions of this chapter. We will point out institutional barriers in China that cause inefficient allocation of production factors; these not only restrict the spatial agglomeration of economic activities, but also are not conducive to reducing regional disparities, and may even hinder China’s long-term economic growth.

1.2 Industrial agglomeration and regional disparity: a theoretical review

NEG’s theoretical roots can be traced back to research on agricultural locations by nineteenth-century classical economists. The Isolated State, published by Von Thünen in 1826, explores the agricultural location issues within an isolated state composed of a city and surrounding agricultural regions. It is regrettable that until the 1990s, research regarding economic geography had always been at the periphery of mainstream economics (Krugman, 1991b). Krugman believes that mainstream economics neglects the location decisions of economic activities due to a key technical problem: How to consider the market structure? Essentially, discussions on the location decisions of economic activities must deviate from the neoclassical paradigm, i.e., constant returns to scale and perfect competition. As long as economists lacked analytic instruments to rigorously portray increasing returns and imperfect competition, research on economic geography will be excluded from the mainstream. Fortunately, Dixit and Stiglitz (1977) used an ingenious simplification to successfully incorporate increasing returns and imperfectly competitive markets into the general equilibrium analytic framework, clearing the technical barrier for the emergence of NEG. Based on the D-S setup, Krugman (1991a,b) took economies of scale, monopolistic competition and transport costs as keystones to construct a theoretical model; he successfully explained the spatial agglomeration of economic activities, marking the birth of NEG. Differing from the traditional research on location decision, NEG puts more emphasis on describing the microeconomic basis of individual choice and market clearing: The location choice of the economic agent—the firm—relies on the relative strength of agglomeration forces and dispersion forces, thus making spatial agglomeration of economic activities endogenously determined in the model. This section will review the classic theories of NEG and comment on the issues of industrial agglomeration and regional coordination.

1.2.1 Re-thinking of the core-periphery model

The initial core-periphery model, established by Krugman, studies an economic system including two regions with symmetric endowments; two sectors (the monopolistically competitive manufacturing sector and the perfectly competitive agricultural sector), and two production factors (mobile skilled labor and immobile unskilled labor). A static core-periphery model does not have capital accumulation and the number of differentiated manufactured goods are constant. Geographic distribution of firms in equilibrium is determined by the relative strength of agglomeration and dispersion forces.

Agglomeration forces usually have the following two sources:

1. Home market effect: Due to the existence of transport costs,⁴ firms tend to locate in a region where market demand is large, and export products to the region where market demand is small, in order to lower transport costs. That is to say, in equilibrium, firms that locate in the region with the larger demand take up a larger proportion of total firms nationwide than the share the region’s demand does. In other words, the region with larger market demand has a magnification effect, which means the share of firms it attracts is a more than proportional share of the demand.

2. Price index effect: Due to the existence of transport costs, the consumer price index (CPI) is lower in the region with a high share of firms. Thus, under equal nominal wage rates in both regions, skilled labor in that region could obtain a higher real-wage rate, causing mobile skilled labor to shift towards it. In other words, the region with a high share of firms can pay a relatively low nominal wage rate to attract skilled labor, thereby obtaining relatively high profits; this further encourages firms to concentrate in the region with more firms. The former agglomeration force is called backward linkage, while the latter is called forward linkage.

Dispersion forces generally have the following two sources:

1. Market crowding effect: All other conditions being equal, on the one hand, the region with a high share of firms has tougher competition, causing firms in that region to face lower demand in the local market than firms in the other region; on the other hand, transport costs have a “protection effect,” which causes the impact of foreign demand on firms’ profits to be “discounted” compared with local demand. ⁵ Therefore, firms in the region with a high share of firms have low profits, encouraging firms to shift towards the region with fewer firms.

2. Immobile factors:⁶ Unskilled labor employed in agricultural sector is tied to land and becomes an immobile factor. But unskilled labor also consumes manufactured goods, therefore the larger the relative quantity of symmetrically distributed immobile factor compared to the mobile factor (skilled labor), the smaller the difference between the quantities demanded for manufactured goods in the two regions, and the more likely it is for firms to be equally dispersed.

The relative strength of agglomeration and dispersion forces, which are usually determined by transport costs, will determine the spatial equilibrium distribution of firms. Specifically, when transport costs in an economy are high, which means the dispersion forces in the system are larger than the agglomeration forces, the agricultural and manufacturing sectors maintain a steady symmetric distribution in the two regions. As transport costs fall (in other words, as regional integration increases), agglomeration and dispersion forces both decrease. However, dispersion forces are more sensitive to a decrease in transport costs than agglomeration forces, so that dispersion forces decrease at a faster rate. There exists a certain critical level for transport costs (let ϕ_B be known as the break point). If transport costs are lower than the break point (ϕ > ϕ_B), agglomeration forces are larger than dispersion forces (Figure 1.1), and the symmetric equilibrium becomes unstable. As the transport costs continue to decline, firms gradually move to one region and the number of firms in each region is no longer equal. Furthermore, there exists another critical level for transport costs (let ϕ_S be known as the sustain point).⁷ If transport costs are lower than the sustain point (ϕ > ϕ_S), all firms will be concentrated in one region and the other region will only be left with an agricultural sector. Such an equilibrium is stable and the economic system forms a core-periphery structure.

Figure 1.1 Dispersion forces, agglomeration forces, and transportation costs.

For welfare consideration, since the shipment of manufactured goods requires transport costs, the CPI of the core is lower than that of the periphery. Thus, consumers in the core (skilled and unskilled labor) enjoy greater well-being than those in the periphery (unskilled labor). When transport costs are very high, both regions maintain the initial symmetric distribution, with the welfare of skilled labor and unskilled labor being equal for both regions, though the welfare of skilled labor is higher than that of unskilled labor. As transport costs fall below the break point, the symmetric equilibrium becomes unstable. There are differences in the two regions’ CPI due to some firms shifting from one region to the other, which causes the well-being of unskilled labor in the two regions to become unequal. As transport costs fall below the sustain point, the two regions form a core-periphery structure, with all skilled labor agglomerated in the core. Now the welfare of consumers in the two regions successively is: the well-being of skilled labor living in the core is greater than that of unskilled labor in the core, which is greater than that of unskilled labor in the periphery. When transport costs decline to near zero, CPI of the core and the periphery tend to converge, and the welfare of the unskilled labor in both regions eventually equalizes.⁸ Thus, during the process where transport costs decline from infinite to zero, the real income differential of residents in the two regions displays an inverted U-shaped curve, initially increasing and then decreasing (Figure 1.2). The inverted U-shape of the real income gap between regions demonstrates the fact that during the process of government-facilitated regional economic integration, it is inevitable that “a few people get rich first,” and the income gap between the regions that got rich first and those left behind will constantly increase for a prolonged period of time. Only when economic integration deepens does the social welfare increase simultaneously with a continuous decrease in the relative well-being gap between regions. That is to say, regarding the government, regional integration policies need to be pushed faster in order to prevent the economy from lingering in the “pain period” for a long time; and regarding the public, enough confidence and patience is required in the government’s regional integration policies to narrow down regional income disparities, and quick results should not be expected.⁹ What also needs to be noted, is that attaining the above-mentioned “inverted-U” trend is dependent upon a sufficient mobility of factors. Once factor mobility is restricted, delay and difficulty in attaining the final convergence is inevitable.

Figure 1.2 Regional gap and transportation costs.

The original NEG model relied to a great extent on certain strong assumptions, such as two regions, two sectors, two factors, and zero transport costs for agricultural goods. The extension of the theory lies in gradually relaxing these assumptions, testing the robustness of the conclusions and improving the elaboration of the theory. Once these rigorous assumptions are relaxed, not only is redis...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Contents
• List of figures
• List of tables
• List of contributors
• Foreword
• PART I Economic agglomeration and interregional development
• PART II Political economy and analysis
• PART III Segmentation and integration
• PART IV Agglomeration and balance
• Postscript
• Index