Geoforum 55 (2014) 53–65

Contents lists available at ScienceDirect

Geoforum

journal homepage: www.elsevier .com/locate /geoforum

Going green or going away: Environmental regulation, economicgeography and firms’ strategies in China’s pollution-intensive industries

http://dx.doi.org/10.1016/j.geoforum.2014.05.0040016-7185/� 2014 Elsevier Ltd. All rights reserved.

⇑ Corresponding author.E-mail addresses: zhus@email.unc.edu (S. Zhu), hecanfei@urban.pku.edu.cn

(C. He).

Shengjun Zhu a, Canfei He b,⇑, Ying Liu b

a Department of Geography, University of North Carolina, Chapel Hill, NC 27599-3220, USAb College of Urban and Environmental Sciences, Peking University, Beijing 100871, China

a r t i c l e i n f o a b s t r a c t

Article history:Received 5 October 2013Received in revised form 15 May 2014

Keywords:Firm strategiesEnvironmental regulationPorter hypothesisPollution haven hypothesisFirm characteristics

The high-growth, resource- and pollution-intensive industrialization model that China has pursued hascaused severe environmental pollution and deterioration, particularly in a number of clusters in thecoastal regions of East and Southeast China, where the Reform and Opening-up policies first started.The lack of uptake of environmental norms/values, deficit of regulatory enforcement of environmentalpolicies, and insufficient institutional capacity have been compounding factors. As environmentalstandards were raised by China’s central government, the enforcement of environmental regulationhas been compromised more in inland China than in coastal regions, due to China’s ‘‘decentralizedgovernance structure’’ and regional disparity in terms of both economic development and environmentalpollution. This paper therefore argues that rising environmental regulations, as well as firm characteris-tics, regional hub effect and political environment, have all been particularly important in forcing China’spollution-intensive enterprises to restructure their production, through innovation, upgrading, geograph-ical relocation, outsourcing and plant closure, especially in China’s coastal regions. It contributes to recentstudies by developing a heuristic analytical framework that aims to be sensitive to the impacts of envi-ronmental regulation, political environment and regional hub effect over firm restructuring, but whichdoes so by stressing these impacts are simultaneously inflected by the nature and attributes of firms.The empirical analysis suggests a roughly inverted ‘‘U’’-shaped relationship between firm relocationtendency and firm size (or firm capability), resulting from complex interactions between politicalenvironment, regional hub effect and environmental regulation.

� 2014 Elsevier Ltd. All rights reserved.

Introduction

The real effects of environmental regulations over firm (re)loca-tion behavior, and competitive performance of industry and regionhave been the subject of heated debates since the outset of theenvironmental movement in the early 1970s (Jeppesen et al.,2002; Testa et al., 2011). One argument is that firms intend tolocate their business activities in countries or regions whereenvironmental regulations are relatively lax (Costantini andCrespi, 2008; Mulatu et al., 2010; Testa et al., 2011). Opponentsof stringent environmental regulation claim that higher environ-mental standards impose additional economic costs by distortingthe spatial pattern of economic development—inducing someregions or countries to be at a competitive disadvantage whencompeting for new investments and jobs (Jeppesen et al., 2002).

This statement purports that change in environmental regulationleads to a relocation of dirty goods production from countrieswith stringent environmental regulation to those with laxenvironmental regulation, resulting in the so-called ‘‘pollutionhavens’’ (Copeland and Taylor, 2004; Dean et al., 2009). This isthe commonly studied ‘‘pollution haven hypothesis (PHH)’’.

On the contrary, Porter and van der Linde (1995) have arguedthat costs for compliance with environmental regulations will beoffset by cost reductions resulting from technological innovationstimulated by the regulations. This argument is also known as‘‘Porter hypothesis (PH)’’. The origins of the PH can be traced backto the seminal work of Schumpeter (1947), who has underscoredthe importance of creative response of economies in adapting tochanges in external conditions, and to the induced-innovationhypothesis formulated by Hicks (1932), who pointed out thatchanges in the relative price of production factors act as a stimulusto technological change and efficiency improvement. In thisrespect, an industry’s competitiveness can be enhanced byproperly designed environmental regulations to encourage the

Table 1Pollution intensity coefficient� of China’s industries. Source: compiled by authors.

Industry Pollutionintensity

Industry Pollutionintensity

Non-ferrous metal mining & dressing 0.466 Leather, furs, down and related products 0.032Electricity and heating production and supply 0.417 Rubber products 0.019Papermaking and paper products 0.416 Fuel gas production and supply 0.019Smelting & pressing of ferrous metals 0.263 Metal products 0.014Water production and supply 0.245 Petroleum and natural gas extraction 0.014Chemical fiber 0.110 Clothes, shoes and hat manufacture 0.008Coal mining & dressing 0.104 Instruments, meters, cultural and office machinery manufacture 0.005Ferrous metal mining & dressing 0.100 Special equipment manufacturing 0.005Raw chemical material & chemical products 0.100 Tobacco products processing 0.004Plastic products 0.094 Beverage production 0.004Nonmetal mineral products 0.092 Ordinary machinery manufacturing 0.004Textile industry 0.088 Communications equipment, computer and other electronic equipment

manufacturing0.004

Non-metal ores mining & dressing 0.064 Transport equipment manufacturing 0.003Smelting & pressing of non-ferrous metals 0.064 Printing and record medium reproduction 0.003Food production 0.058 Furniture manufacturing 0.003Petroleum processing, coking and nuclear fuel

processing0.057 Cultural, educational and sports articles production 0.001

Medical and pharmaceutical products 0.049 Electric machines and apparatuses manufacturing 0.000Agricultural and sideline foods processing 0.044 Average pollution intensity 0.085

Note 1: �Pollution intensity coefficient of China’s 35 industries is calculated in three steps:(1) Pi,1 (waste water pollution intensity), Pi,2 (waste gas pollution intensity), and Pi,3 (waste residues pollution intensity) are defined as the amount of waste water (gas orresidues) that are discharged to produce one million Yuan of output in industry i.(2) The underlying formula is used to standardize Pi,1, Pi,2 and Pi,3:

Pai ¼

Pai �minðPa

1 ;Pa2 ;...;P

a35 Þ

maxðPa1 ;P

a2 ;...;P

a35Þ�minðPa

1 ;Pa2 ;...;P

a35Þ

, a = 1,2 or 3.

(3) Finally, pollution intensity coefficient of industry i, Pi , is measured as:

Pi ¼P1

i þP2i þP3

i3 .

Note 2: If the pollution intensity coefficient of one industry is higher than average (0.085), it is considered as pollution-intensive (in bold).

1 General Office of the State Council, May 25 2007, ‘‘Report on the First NationwidePollution Source Survey.’’ Retrieved on February 23 2014 from: http://www.gov.cn/zwgk/2007-05/25/content_626141.htm.

54 S. Zhu et al. / Geoforum 55 (2014) 53–65

application of new, innovative, and clean technologies (Hamamoto,2006; Kumar and Managi, 2009; Porter and van der Linde, 1995).Although PH is primarily focusing on the relationship betweenenvironmental regulation and innovation/competitiveness, thisargument can also be applied to portray the relationship betweenenvironmental regulation and firm (re)location as firms’ locationbehavior is highly related to innovation, competitiveness as wellas production efficiency (Leiter et al., 2011; Testa et al., 2011).

The gist of PH for our present purposes is that increase in envi-ronmental standards can actually improve competitiveness, offsetcompliance costs and encourage firm upgrading, whereas PHHcontends firms intend to relocate their business activities fromcountries (or regions) with stringent environmental regulation toones where environmental standards are relatively low, in orderto avert compliance costs. At the present time, while efforts toexplain the effect of environmental regulation over firm restructur-ing (or competitiveness) have attracted increasing scholarly atten-tion, a great deal of attention is directed towards analyzing therelationship between environmental regulation and firm restruc-turing based either on PHH (Dean et al., 2009; Spatareanu, 2007;Tole and Koop, 2010) or on PH (Costantini and Crespi, 2008;Hamamoto, 2006; Kumar and Managi, 2009; Murty and Kumar,2003). Most extant studies have taken a ‘‘black-or-white’’ attitude,and sought to verify one hypothesis while implicitly neglecting, ifnot completely negating, the other. The present paper will contrib-ute to this emerging polemic and argue that PH and PHH—in mostinstances—co-exist, by focusing specifically on the restructuring ofthe pollution-intensive firms in Shangyu, Zhejiang province inChina. By ‘‘pollution-intensive firms’’, we are referring to firms inindustries such as non-ferrous metal mining & dressing, electricityand heating production and supply, papermaking and paper prod-ucts, smelting & pressing of ferrous metals, water production andsupply, chemical fiber, coal mining & dressing, ferrous metal min-ing & dressing, raw chemical material & chemical products, plastic

products, nonmetal mineral products, and textile industry(Table 1), which have also been regarded as pollution-intensiveindustries in the First Nationwide Pollution Source Survey (2007)launched by China’s State Council.1

China has received constant attention for both its rapidly grow-ing economy and the serious environmental degradation that hasoccurred since the Reform and Opening-up policies of 1970s (Heet al., 2012). The high-growth, resource- and pollution-intensiveindustrialization model that China pursued has caused severe envi-ronmental pollution and deterioration (Chan and Yao, 2008; Heet al., 2012), particularly in a number of clusters in the coastalregions of East and Southeast China where the Reform and Open-ing-up policies first started (He et al., 2008; Wang, 2010; Weiet al., 2007; Wen, 2004). The lack of uptake of environmentalnorms/values, as well as implementation deficit of environmentalregulations and policies, and the lack of institutional capacity havebeen compounding factors. With an increasing number of pollutionincidences, such as the low visibility days in Beijing in 2013 causedby thick fog and haze, reported by the media, much attention hasbeen directed towards reducing pollution and to promoting cleantechnology as well as industrial upgrading. A series of laws, regu-lations, and standards such as the Comprehensive Work Plan of Sav-ing Energy and Diminishing Pollution, have therefore been issued,resulting in an increasing level of environmental stringency.

Pollution emissions vary across the Chinese regions as thecoastal regions have a relatively longer period of unbridled eco-nomic development (Tang et al., 2010) (Fig. 1), and the strengthof the enforcement of the environmental regulations varies acrossregions as well (Zhang and Fu, 2008). First, the variation ofenvironmental regulation could be attributed to the intention of

Fig. 1. Spatial distribution of output in pollution-intensive industries by province in China in 2010 (see Table 1 for definition of pollution-intensive industries in China).Source: Compiled by authors, from the China Industry Economy Statistical Yearbook 2011.

S. Zhu et al. / Geoforum 55 (2014) 53–65 55

China’s central government who saw the increasing number of pol-lution incidences as not only an urgent signal that coastal regionsshould shed low-cost, pollution- and resource-intensive produc-tion, but also an opportunity to relocate such production fromthe higher-cost coastal regions to release space and resources forhigher-value, clean industries while simultaneously encouragingeconomic development in less developed inland China (Yang,2012; Zhu and Pickles, 2014). For instance, in 2007, China’s Minis-try of Commerce and China Customs promulgated the List ofRestricted Commodities in Processing Trade, differentiating betweenallowed low-cost, labor-intensive processes inland and those thatare now restricted in coastal regions. Importantly for our purposes,pollution-intensive products made up most of the restricted low-cost process and products. Second, the Chinese government hasgranted much autonomy and responsibility in both environmentalpolicy and economic development to local authorities since the1980s (Jahiel, 1997, 1998), resulting in the so-called ‘‘decentralizedauthoritarianism’’ in terms of the enforcement of environmentalpolicies (He et al., 2012; Lieberthal and Lampton, 1992).Specifically, local states in coastal regions, where environmentaldegradation is much more severe and environmental awarenessamongst stakeholders is comparatively high, might enforce theenvironmental laws and regulations more genuinely, while localadministrations in less developed inland China with relativelylower levels of pollution are more likely to privilege industrialdevelopment at the cost of environment (Wang and Wheeler,2000, 2005). As a result, the enforcement of environmental regula-tion might be compromised more in inland China than in coastalregions. Such decentralized governance structure in China andthe variation of environmental stringency across regions it gener-ates therefore offer an ideal case to examine the impact of differen-tial environmental regulations and their enforcement over firmstrategies.

In this research, we focus on the pollution-intensive firms inShangyu, Zhejiang province of China and examine their restructur-ing strategies. This paper documents the diversity of trajectories inwhich different kinds of firms are attempting to deal with increas-ing environmental stringency. A tentative analytical framework isdeveloped in the second section to analyze firm strategies underthe influence of environmental regulation. After an overview of

the development of Shangyu’s pollution-intensive industries insection three, section four seeks to examine various adaptationsShangyu’s pollution-intensive firms are undergoing. Section fiveanalyzes the restructuring processes based on our tentative analyt-ical framework. The sixth section concludes the paper.

A heuristic analytical framework

Pollution haven hypothesis, Porter hypothesis and firm characteristics

Currently, there is much uncertainty in the literature aboutwhat role environmental regulations play in firm restructuringdecisions. The empirical evidence from testing the relevance oftwo contrasting views—PHH and PH—has been at best mixed andinconclusive (Dam and Scholtens, 2008; Jeppesen et al., 2002;Spatareanu, 2007; Tole and Koop, 2010; Zhang and Fu, 2008). Inthis section, we shall elaborate the first reason. Although the PHand its predecessor, the induced-innovation hypothesis formulatedby Hicks (1932), offer a useful analytical framework based onwhich the interactions between environmental regulation andthe dynamics of innovation and technology can be nicely exam-ined, both of them underestimate, if not neglect, the analysis ofthe actual availability of technological know-how and of the mech-anisms underlying the introduction of technological innovations(Antonelli, 2003; Keller, 2004). In other words, the introductionof properly designed environmental regulation may well catalyzetechnological innovation and industrial upgrading as it affectsmarket condition by opening up new profit opportunities, how-ever, this causality only holds water when firms (or innovationssystems) are equipped with adequate scientific, technical and tech-nological knowledge so that they are able to creatively respond tochanges in external constraints (i.e., environmental regulation)(Altman, 2001; Costantini and Crespi, 2008).

Throughout most of the PH and PHH literature there is a tacitassumption that all firms are relatively homogeneous (BenKheder and Zugravu, 2012; Leiter et al., 2011; Mulatu et al.,2010), and they do not merit attention in their own right. Giventhe fact that firms are actually heterogeneous with differentattributes and capabilities and firm strategies are normally made

56 S. Zhu et al. / Geoforum 55 (2014) 53–65

based on firm attributes, it may be inappropriate to examine thearticulation between environmental regulation and firm strategies(e.g., innovation and relocation) only through meso- or macro-levelanalysis without paying special attention to micro- or firm-levelfeatures. While changes in external environment (e.g., environ-mental regulations and their enforcement) undoubtedly shapefirms’ competitiveness, innovation and (re)location, we argue thatthe impact of changes in external environment is in turn shaped byfirm characteristics and capabilities (Lin et al., 2011; Myles Shaverand Flyer, 2000; Wang et al., 2010). To understand the relationshipbetween the stringency of environmental regulation and itsenforcement and firms’ strategies, more attention should be paidto whether firms are equipped with adequate technological knowl-edge and upgrading capability and how this capability interactedwith external environment (changes in environmental regulationsand their enforcement) to produce a positive (or negative) influ-ence on innovation (Jo and Lee, 2012; Teece, 2007). Specifically,in the PH, the importance of changes in environmental regulationshould not be overemphasized at the expense of the firm charac-teristics. Although firm attributes have been taken into accountin some studies, they are at most treated as control variables(Dam and Scholtens, 2008; Jeppesen et al., 2002; Testa et al.,2011). This paper contributes to this debate by developing a heu-ristic analytical framework centered on firm characteristics. Basedon this framework, we argue that with diverse financial and tech-nological capabilities, the co-existence of various types of firmscould play a fundamental role in the innovative performance ofthe region and therefore result in a more nuanced situation wherePH and PHH co-exist. In addition, whether firms act in ways thatare predicted by the PH or PHH is dependent on the firm-specificcharacteristics.

Regional hub effect and political environment

The second reason why empirical studies addressing the rele-vance of the PHH and PH have remained largely inconclusive isthat these studies have been troubled by the fact that researchersin some instances mistake a pollution haven effect (and Portereffect) for the PHH (and PH) (Taylor, 2004). Pollution haven effects(and Porter effects) emerge if differential environmental regula-tions affect the (re)location decision of pollution-intensive firms.Such effect, if at work, is just one factor among many that deter-mine firm strategies (Ben Kheder and Zugravu, 2012; Leiter et al.,2011; Mulatu et al., 2010). The PHH and PH, however, contend thatenvironmental regulation is the most important determinant forfirm’s competitiveness, innovation and (re)location decision, or atleast more important than other potential determinants. In a word,pollution haven effects and Porter effects are only one of severaldeterminants for firms’ (re)location decision and therefore, inrecent literature, much more attention has been drawn towardsweighing the relative strength of other determinants. The presentpaper seeks to present a way of undertaking such an analysis, spe-cifically by paying attention to another two types of determinants:regional hub effect and political environment.

Regional hub effectIn economic geography much has been written about the robust

growth of certain industrial districts, clusters or regional innova-tion systems (Piore and Sabel, 1984; Saxenian, 1994; Storper,1997). Industrial clustering has been seen to be important becauseit deepens social networks and inter-firm linkages among proxi-mate firms, especially if those linkages involve complicated formsof interaction and trust (Barnes and Gertler, 1999; Storper, 1997).Specialized labor markets, vertical and horizontal linkages andexchanges of tacit knowledge are especially dense, efficient andvibrant when it is possible for agents to have face-to-face interaction.

The competitiveness of a firm partly hinges on its embeddedness intolocal clusters made up by heterogeneous and homogenous firms(Granovetter, 1985; Porter, 1990). Interpretations on geographicalclustering of industries thus have tended to focus on the sources ofcompetitiveness driven by local economic agents and their verticaland horizontal relationships which generate collective efficiencyand knowledge spillovers (Marshall, 1920 [1890]), institutional thick-ness (Amin and Thrift, 1994), embeddedness (Granovetter, 1985) andtraded and untraded interdependencies (Scott, 1988; Storper, 1997).The importance of such ‘‘clusters’’ does not lie very much in thephysical space per se, but rather in the formal and informal inter-firmrelationships as well as the regional innovative milieux that areimportant for firm’s competitiveness and regional development.Here, we use ‘‘regional hub’’ to capture the idea of a ‘‘cluster’’ (orindustrial district, regional innovation system) that is not merely a‘‘container’’ in which favorable location factors may happen toexist or not, but rather as a milieu in which a wide range of actorsare incessantly interacting with one another through formal andinformal, horizontal and vertical linkages.

Political environmentIt is increasingly acknowledged that developing economies

need to embed private initiative in a framework of public actionthat inflects industrial restructuring, relocation and technologicaldynamism beyond what private governance would generate ontheir own (Bair and Dussel Peters, 2006; Dussel Peters, 2008). Thisrecognition is now particularly widely perceived in those countrieswhere market-oriented reforms were taken the farthest and thedisappointment about the outcomes caused by market failures(e.g., environmental degradation) is correspondingly the greatest.In China, the social and environmental consequences of low-value,pollution-intensive production have become increasingly serious,forcing the central government to become more active in regulat-ing the trajectories and geographies of change in the industry(Wheeler, 1999). Additionally, China wants to become an innova-tion-led economy, with the state aggressively trying to move itsindustries upward along the value chain through investments inR&D, advanced and cleaner manufacturing, with an emphasis ondomestic innovation (Hu and Jefferson, 2008; Zhu and Pickles,2014). Since 2005, China has also created a policy apparatus todesign and implement energy-efficiency throughout the economy(Zhou et al., 2010). Laws have been passed, regulations promul-gated, new programs created and implemented to supportenergy-efficiency investments and incentivize energy-intensityreductions. One important offshoot of the effort is that it also tendsto reduce pollution.

The central government’s environment values have been shift-ing from a pro-growth to a pro-environment orientation, with asimultaneous rise in environmental awareness of the public(Francesch-Huidobro et al., 2012; Wheeler, 1999). However, moreeffective environmental protection cannot be achieved merely bystronger efforts from the central government and heightenedpublic and media awareness, because the enforcement of nationalenvironmental policies and regulations is primarily the responsi-bility of environmental protection bureau (EPB) at the local level.Although local EPB is under the functional supervision of the corre-sponding agency at a higher administrative level, as a part of localgovernment, it is financed by the local government (Francesch-Huidobro et al., 2012; Lo and Tang, 2006). The enforcement of envi-ronmental regulations is therefore, in most instances, determinedby the local EPB’s institutional capacity and, in particular, localgovernment’s political support (Francesch-Huidobro et al., 2012).Local governments do not always share the concerns that motivatecentral government policies and they have, at times, respondeddifferently, resulting in increasing environmental standards andemphasis on upgrading through investments in innovation and

S. Zhu et al. / Geoforum 55 (2014) 53–65 57

cleaner technology at the national level, and segmented enforce-ment of environmental laws and variegated environmental strin-gencies at the local level (Zhan et al., 2013).

Finally, driven by the so-called ‘‘grasp the large, let go thesmall’’ policy, governments at all levels tend to compete for enter-prises above a certain size that have potentials not only to upgradethrough investments in innovation and cleaner technology but alsoto create a considerable amount of employment, whereas smallfirms, especially unprofitable, energy- and pollution-intensiveones, have been sometimes disregarded by both national and localadministrations (Dasgupta et al., 1997; Sutherland, 2003).

We formulate a heuristic analytical framework that aims to besensitive to the impacts of environmental regulation, political envi-ronment and regional hub effects over firms’ strategies, but whichdoes so by stressing these impacts are simultaneously inflected byfirm attributes and capabilities. In Fig. 2, we use the transition fromsolid arrow to dashed arrow to indicate such an ‘‘inflection’’ effectcaused by firm characteristics.

Research design and site characteristics

Fieldwork carried out during 2013 included open-endedinterviews with two groups of agents in Shangyu, China (Fig. 3).A total of thirty-six face-to-face interviews were conducted withthirty-two firms. Each interview was with a senior manager andwas accompanied by a shop floor visit to see the plant in operation.In addition, six interviews were conducted with representativesfrom the Administrative Committee of Hangzhou Bay ShangyuIndustrial Area, Shangyu Economic and Trade Bureau, and ShangyuEnvironmental Protection Bureau. Each interview lasted at leasttwo hours, and those with key interviewees such as representa-tives from local government officials and several leading entrepre-neurs each took half a day. When possible, informants wereinterviewed more than once. Local government was first contactedand interviewed to provide an understanding of the generalstructure of Shangyu’s pollution-intensive industries. Afterward,interviews with local entrepreneurs were organized with theguidance of the local government. Interview findings were supple-mented with secondary information collected from sector-specificpublications, company reports, and websites. Due to confidential-ity agreements with the companies, the identities of firms andinterviewees remain with the authors.

The traditional Shangyu model is a system of production cen-tered on private enterprises, and embedded in dense, historicallyrooted local institutions (Shangyu Government, 2013; Yuan,2010). Production and employment geographically concentrate in

Fig. 2. A heuristic analytical framework for firm strategies in pollution-intensiveindustries. Source: compiled by authors.

a number of industrial clusters. In 2008, 67.4% of Shangyu’senterprises (around 8,400 enterprises) located in various industrialclusters, directly employing around 179,600 workers and complet-ing industrial output value of 71.4 billion Yuan, which respectivelyaccounted for 75.0% and 73.5% of the city’s total (Yuan, 2010).Important clusters specializing in one or several industriesinclude: copper pipe manufacturing cluster in Tangpu and FenghuiTown, children’s wear cluster in Tangpu Town, machinery andequipment manufacturing cluster in Caogejiang Sub-district,lighting manufacturing cluster in Lihai Town, bearing manufacturingcluster in Shangpu Town, and fine chemicals manufacturing cluster inHangzhou Bay Shangyu Industrial Park, etc. (Shangyu Government,2013; Yuan, 2010).

Since the initiation of Reform and Opening-Up Policies of 1970s,Shangyu has achieved dramatic economic growth. In the early2000s, this cluster-based, private-sector-dominated industrializa-tion model started to be challenged by the increasing level of envi-ronmental deterioration in China’s coastal region since mostfactories focused on pollution-intensive production. Decades ofunbridled economic development have resulted in not only tre-mendous increases in energy consumption, environment pollutionand degradation, but also growing social pressure and awarenessfor higher-quality environment and cleaner production, forcinglocal government to become more active in regulating economicdevelopment and adjust the trajectories of industry. In 2010,Shangyu Government issued the Plan on Compulsorily RestructuringPollution-intensive Industries in Shangyu, also known as the ‘‘three-year plan’’, requiring 10 pollution-intensive industries and 140enterprises to be annually evaluated and monitored by Shangyu’sEconomic and Trade Bureau, Environmental Protection Bureau,Finance and Local Taxation Bureau and Water ConservancyBureau.2 The last ‘‘three-year plan’’ released in 2007 has compelled51 enterprises to green their production, and forcefully shut down36 pollution-intensive projects. On May 27, 2013, Shangyu Govern-ment renewed the Plan, indicating its resolution to continue opti-mizing and upgrading its industrial structure for another threeyears (Shangyu Government, 2013). This paper therefore focuseson this round of industrial restructuring triggered by increasingsocial awareness and rising environmental standards in Shangyu inparticular and in China’s coastal regions in general.

Different firms, differ strategies

In this section we turn to case studies of firm’s diverse strate-gies. The firms we discuss are actual firms active in industrialrestructuring in distinct ways (Table 2). They are also ideal typicalfirms that represent the wider dynamics in the regional cluster. Weinterrogate each and use the cases to examine firm restructuringthrough our tentative analytical framework. It is certainly not asurprise that different firms in different industries choose differentrestructuring strategies, and that context and contingency arecritical in shaping these differences. Our goal is not to argue thatfirm strategies differ, but to emphasize that we need to understandthe ways in which firms are making restructuring decisions inways that articulate with a wider range of factors such as regionalhub effects, environmental regulation and political environment.

Firm A: going green in situ

Firm A was originally made up by three wholly-ownedsubsidiaries:

2 Environmental Protection in China, May 9, 2013, ‘‘Restructuring and Adjustmentof Pollution-intensive Industries in Shangyu, Zhejiang.’’ Retrieved on September 16,2013 from: http://www.epuncn.com/news/chengshi/74985.htm.

Fig. 3. Location of Shangyu, China. Source: compiled by authors.

58 S. Zhu et al. / Geoforum 55 (2014) 53–65

(1) Established in 1993, subsidiary A1 is a thermal power plant,and its main products include thermal energy for heat (80%)and electric power (20%). As the only power station in theShangyu Economic Development Zone that has adoptedthe combined heat and power (CHP) systems (also knownas cogeneration systems),3 it serves the entire Shangyu Eco-nomic Development Zone.

(2) A2 was established in 2002, manufacturing mixedyarns, high-end knitted and crocheted fabrics. It is highlydomestic-oriented, with Yangtze River Delta as its mainmarket.

(3) A3 is a paper products plant. It manufactures 0.15 billionsquare meters of high-strength corrugated paper annuallyand primarily targets markets in neighboring cities.

All three subsidiaries are in pollution-intensive industries. Asenvironment regulations increased in Shangyu, firm A, in 2007,invested 15 million to start its fourth subsidiary A4—a wastewatertreatment plant—designed to remove biological and chemicalwaste products from waste water for the entire firm.

The average cost of water disposal is often 5% of the total. It isrequired to remove or at least reduce pollutants in waste waterto a certain degree. If not, local government will heavily fine usor even shut down our factories until relevant problems areproperly solved. It takes ten thousand [Yuan] to run the waterdisposal facility for one day, excluding expenses on labor.However, the entire firm generates 3,000–5,000 tons of wastewater every day. Due to economies of scale, the average cost isway lower than that of small firms.

[General Manager of Firm A, translated from Chinese]

3 CHP is to simultaneously generate two types of energy—heat and electricity—from one fuel source. By capturing and using the waste heat, CHP offers tremendousefficiency and thus both cost savings and environmental benefits.

In addition, firm A has already established facilities for wastegas primary treatment, and invested another 7 million Yuan inthe secondary treatment facility. It has also replaced coal with nat-ural gas as its main fuel source, in order to increase energyefficiency, reduce air pollution, and to lower greenhouse gas emis-sions. As a key enterprise in Shangyu, firm A’s efforts at diminish-ing pollution and saving energy has been generously supported bythe local government, in the form of tax credits and subsidies. Forinstance, the latter will send monetary assistance that reimburses50–70% of the former’s expenditure on any upgrading and innova-tions made to reduce pollution.

Firm B: relocating to industrial parks

Firm B, established in 1996, specializes in printing and dyeing. Ithas formed an annual production capacity of 5,000 tons of reactivedyes and 5,000 tons of Para Base Ester, and over 10,000 tons of allkinds of other printing and dyeing auxiliary and common auxiliary.It procures raw materials primarily from Shanghai and Zhejiang,and Shanghai Baosteel is its largest supplier. Although it has theright to import and export, its principal market is Yangtze RiverDelta and its largest customer is Shaoxing Textile City.4

In response to the rising environmental standards, it invested10 million Yuan to build up its own ‘‘three-waste’’ treatment facil-ities. In 2002, another 0.45 billion Yuan was spent to relocate partof its production into the Hangzhou Bay Shangyu Industrial Park(Fig. 3). The centralized waste water treatment system in theindustrial park collects waste water from all factories within theindustrial park and is capable of processing 0.1 million tons ofwaste water every day. In doing so, cost per unit has beendecreased dramatically. In addition, this industrial park is locatedin an unpopulated, remote area, causing little social discontent.

4 Shaoxing Textile City is one of the largest textile and apparel markets in China.

Table 2Baseline characteristics of six pollution-intensive firms in Shangyu.

Firm A B C D E F

Year established 2002 1996 2002 2001 2002 1976Number of

workers300 200 73 27 35 30

Annual sale (unit:Yuan)

200 million 100 million 27 million 10 million 5 million 5 million

Main productsand mainactivities

Thermal power; textileprinting and dyeing; paper

Printing and dyeingauxiliary and commonauxiliary

Washingagents

Recycling ofplastics

Textiledyeing

Textile printing & dyeing

Ownership Private-owned Private-owned Private-owned Private-owned Private-owned

Town & village-owned before1997; private-owned after 1997

Firm strategy Going green in situ Relocating to industrialpark

Relocating topollutionhaven

Outsourcing topollution haven

Relying onlarge firms

Waiting and dying

S. Zhu et al. / Geoforum 55 (2014) 53–65 59

Due to such relocation, our transportation cost has increased by10 Yuan per ton. We also have to face a substantial logisticalcost and managerial cost increase. However, the cost of waterdisposal has declined drastically. Plus, we are able to obtainland at lower prices and take advantage of tax credits, as areward granted by the local government for our compliancewith environmental regulations. Overall, such adjustment isnecessary and beneficial in the long-term.

[General Manager of Firm B, translated from Chinese]

Finally, firm B is planning to relocate the rest of its productionand headquarters into the industrial park too.

In the short-run, environmental regulation tends to squeezefirms’ profit margins. However, in the long-run, more stringentenvironmental regulation will force low-cost, low-end smallfirms to exit first. Then, the rest could have bigger marketsand start to compete on quality rather than price.

[General Manager of Firm B, translated from Chinese]

Firm C: relocating to ‘‘pollution havens’’

Firm C, established in 2002, is specialized in the manufacture ofdetergent, penetrant, levelling agent, dispersant, emulsifier, andsoftener. It has formed an annual production capacity of 50,000tons of dyeing and finishing auxiliary. Its raw materials are pro-cured domestically, especially from suppliers in the same province.Its customers are highly concentrated in Yangtze River Delta, threeof which account for around 60% of its total output. Like firm B, it isalso a chemical enterprise; unlike firm B, it targets the niche oflow- and medium-end products. The majority of its output is madeup by low-value, low-tech primary products while medium- andhigh-value synthetic products only account for 20%. The entry-bar-rier is low for such low-cost production, and the market has beencrowded by a large number of homogeneous medium- andsmall-sized firms that compete on price, resulting in low profitmargins and a large degree of product similarity. By focusing oncost-sensitive production, firm C is particularly vulnerable to theincreasingly stringent environmental regulation. Additionally, itis located close to residential areas, therefore has to deal with lar-ger social pressures and discontents.

In 2010, it invested 12 million Yuan on R&D and importing clea-ner and more advanced technologies, and sought to imitate firm B’sstrategy and to relocate its pollution-intensive production intoindustrial parks.

If we can relocate into industrial parks which are far away fromresidential areas, we do not need to deal with social discontentsand our compliance cost can be drastically lowered. However,recently, industrial parks started to reject relocating firms

specialized in certain pollution-intensive industries. Forinstance, Hangzhou Bay Shangyu Industrial Park refuses chem-ical firms like us. In addition, medium- and small-sized firmsare also unwanted. For instance, Daoxu Industrial Park onlywelcomes large, key firms.

[General Manager of Firm C, translated from Chinese]

In 2005, the Administrative Committee of Hangzhou Bay Shang-yu Industrial Area hired eight environmental experts, who wouldevaluate and interview firms that apply to relocate into the Indus-trial Park. Nearly a thousand applicant firms have been rejected,due in large part to their pollution-intensity and size. As a med-ium-sized, pollution-intensive enterprise, firm C has to find newpollution havens, as its general manager argued:

Two of my friends already relocated their factories to northernJiangsu where environmental regulation is comparatively lax.This relocation lowered their environmental costs. In addition,land is cheap over there and it is easier to enter local industrialparks. Local supporting facilities are not bad. Northern Jiangsuis not far away from Yangtze River Delta [in which our custom-ers and suppliers concentrate]. Since relocating to Shangyu’sindustrial parks is impossible, northern Jiangsu appears to bea nice alternative.

[General Manager of Firm C, translated from Chinese]

Firm D: outsourcing to ‘‘pollution havens’’

Firm D, a recycled plastic products manufacturer established in2001, is located in Tangpu Town. The company recovers scrap orwaste plastic and reprocesses the material into PO, PP, PE andPVC recycled plastic granules. 80% of its scrap or waste plastic isgathered from local factories (e.g., umbrella factories in Shangyu),and the rest is reclaimed in neighboring cities (e.g., Yuyao). Shang-yu and Yuyao are also its two largest markets, accounting for 70%and 30% of its total output respectively. Recycled plastic is pro-duced by the reclamation and reprocessing of waste plasticthrough thermal depolymerization, plastic pyrolysis, and/or heatcompression. The term ‘‘recycled’’ does not make the recycled plas-tic industry environment- and eco-friendly; instead, the specificrecycling process is highly pollution-intensive and results in largeamounts of air and water pollutants. There were around 20 recy-cled plastic firms in Shangyu, three fourths of which have alreadybeen forcefully shut down by the local authority since 2012. As myinterviewees disclosed, some entrepreneurs re-opened their facto-ries in locations where environmental regulations are relativelylax, such as Anhui, northern Jiangsu and western Shandong.

Recycled plastic industry is overall profitable and is still aseller’s market. In 2012, the price of finished products increased

60 S. Zhu et al. / Geoforum 55 (2014) 53–65

6–7%, while the cost of raw materials declined 10%. Environ-mental regulation is a double-edged sword. On the one hand,a lot of firms have closed or relocated because of the increaseof environmental standards, resulting in a larger market forthe rest. On the other hand, it leads to substantial cost increase.In addition, because the waste water and air treatment facilitiestake up some of our land and labor, we have to downsize ourproductive facilities.

[General Manager of Firm D, translated from Chinese]

In order to retain production capacity, firm D outsources itsmost pollution-intensive—scrap or waste plastic shredding andwashing—to one factory located in another city where environ-mental regulation is lax. Although such a sourcing strategy hasincreased its transportation cost by 30,000 Yuan per month, itenables firm D to raise its productive capacity and to avert theincreasing level of environmental regulation in Shangyu. However,this strategy is at best a temporary resolution.

We do not plan to relocate completely at this stage, because therest of our production in Shangyu is less pollution-intensive.However, if the environmental regulations keep increasing, wemay relocate completely to Anhui, northern Jiangsu and wes-tern Shandong, like some of my friends did.

[General Manager of Firm D, translated from Chinese]

Firm E: relying on large firms

Firm E offers printing, dyeing and processing for all types of fab-rics including cotton, polyester, and poly-cotton. Firm E andanother 30 small firms (each has less than 50 employees) are allassociated enterprises of one large textile firm-Zhangxing-locatedin the industrial park of Fenghui Town in Shangyu. Both of FirmE’s downstream and upstream linkages unfold locally. On the onehand, it is heavily dependent on local suppliers for raw materials(e.g., printing and dyeing auxiliaries and dyes). On the other hand,it concentrates on basic, low-value and low-end ‘‘processing withsupplied materials’’, exclusively supplying local clients, three ofwhich make up 80% of its total output. As a small, pollution-intensive, cost-sensitive firm, it has been increasingly pressuredby environmental regulation.

Last year, local government established a monitoring stationexactly beside our factory. Moreover, according to the newenvironmental standards, we have to replace our current equip-ment and technologies with more advanced, cleaner and effi-cient ones. We are also required to replace coal with naturalgas as the main fuel source. This alone will cost us half millionYuan.

[General Manager of Firm E, translated from Chinese]

Resources required to comply with environmental regulationslimit the amount of resources available for other investments. Assuggested by our interviewee, all factories in Shangyu need to con-nect their sewage system to the city’s sewage system so that thelocal government is able to monitor water pollution and transferwaste water to a central treatment plant. However, Firm E is onekilometer away from the nearest sewer lines, and it will take atleast two million Yuan to make such connection.

We do not want to relocate to other cities. First, our suppliersand clients are both local. Second, long-distance relocation toanother place where political, cultural and societal contextsare all different is unwise and we may have to deal with variousrisks and uncertainties. Moreover, as a small firm, it is hard notonly to get support from Shangyu’s government, but also todraw any attention and win any financial aid from local

governments in regions we are relocating to. Instead of gettingconnected to the city’s sewage system, we have partiallyrelocated to the industrial park of Fenghui Town.

[General Manager of Firm E, translated from Chinese]

As indicated above, industrial parks only welcome largeenterprises, and Firm E has passed this threshold by becoming anassociated enterprise of one large textile firm—Zhangxing—locatedin the industrial park of Fenghui Town in Shangyu.

Firm F: waiting and dying

Like firm E, firm F also specializes in printing, dyeing and pro-cessing for all types of fabrics including cotton, polyester, andpoly-cotton, and both of its downstream and upstream linkagesunfold locally as well. It is supplied exclusively by a local printingand dyeing auxiliaries and dyes firm—Baosheng—and 80% of firmF’s clients are concentrated in Shangyu. Emerging out of a town-and-village enterprise, it underwent a process of privatization in1997, which advanced its productive capacity. By focusing onlow-cost, pollution-intensive production (e.g., processing withsupplied materials), firm F is particularly vulnerable to the increas-ing environmental regulation. As a result, its annual sales haveplummeted from 20 million Yuan to 5 million in recent years. Asits general manager argued:

Increasing environmental standards are eroding the profitmargin drastically. Although we have connected our sewagesystem to the city’s sewage system, the volume of waste waterwe can discharge is strictly controlled and monitored. This haslimited our production capacity dramatically, in particular inrush seasons.

[General Manager of Firm F, translated from Chinese]

The new ‘‘three-year plan’’ has raised the environmentalstandards to another level, with respect not only to how muchwaste water firms could discharge but also to how clean firms’technology should be. Firm F, as a small firm with limited financialand technological capability, appears to reach an impasse.

We are literally dying. We are unable to deal with the stringentenvironmental regulation, because we cannot afford those cleantechnologies. Relocation is not an option for us either, since ourcustomers and suppliers are all located in Shangyu. In addition,it is impossible for small firms to obtain any tax credit andfavorable policies from local governments, even if we relocateto inland China. Instead, we may have to deal with unfamiliarlegal, political and social environment as well as theaccompanying risks and uncertainties.

[General Manager of Firm F, translated from Chinese]

Going green or going away

Based on case studies, we formulate a two-dimensionaldiagram that depicts the impacts of environmental regulation,political environment and regional hub effect over firm relocationtendency (and/or capability), but which does so by stressing theseimpacts are simultaneously inflected by firm characteristics(Fig. 4). The Y axis represents firms’ relocation tendency (and/orcapability) and the X axis is firm size (from small to large), as aproxy for firm capability. Three lines in Fig. 4 portray therelationships between firm relocation tendency and firm size, interms of environmental regulation, regional hub effect and politicalenvironment respectively.

First, we argue that PHH and PH co-exist and whether firms actin ways that are suggested by the PHH or PH is largely determined

Fig. 4. Firm characteristics and firm relocation tendency/capability. Source: com-piled by authors.

S. Zhu et al. / Geoforum 55 (2014) 53–65 61

by firm-specific characteristics. Firm attributes therefore shouldnever be underestimated or relegated to secondary or control vari-ables that either have negligible impacts over firms’ innovation,competitiveness and (re)location decisions or are unable to funda-mentally alter the impact of environmental regulation as well asregional hub effect and political environment.

A firm’s characteristics, such as size, financial and technologicalcapability, and nature of production, therefore have direct influ-ences not only on its innovation-related strategies, but also onthe way in which and the extent to which its competitivenessand (re)location decision are affected by changes in environmentalregulation. Importantly for our purposes, the introduction of prop-erly designed environmental regulation may encourage and posi-tively stimulate large and mature pollution-intensive firms toinnovate and upgrade to high-value, clean production (PH), sincethey have a comprehensive resource base and adequate technolog-ical knowledge to bring about innovation. By contrast, small firmswith limited financial and technological capability usually faceresource constraints in the process of innovation; therefore, asenvironmental regulation increases, they tend to relocate, at leasttheir pollution-intensive production, to regions where environ-mental standards are relatively low, i.e., the so-called ‘‘pollutionhaven’’ (PHH). In addition, a larger amount of output also enablesbig firms to capture the economies of scale, potentially lower theaverage compliance cost, and alleviate the negative impacts ofenvironmental regulation, while small firms’ profit margins mightbe eroded so drastically that some of them have to downsize pro-duction or close altogether. In a word, the effects of firm character-istics are inter-connected with those of the environmentalregulation, resulting in a more nuanced situation where PH andPHH co-exist and the extent to which they matter depends on firmattributes. As environmental regulation increases, the smaller afirm is, the more likely it is to relocate, the less likely and capableit is to innovate (Fig. 4).

However, the above analysis solely based on environmentalregulation and firm attributes leaves many open questions. Forexample, large firms with a comprehensive resource base andadequate technological knowledge are capable to stay in locationswhere environmental regulation is relatively stringent and toinnovate does not mean they have to do so. While the power ofinnovation and upgrading to generate economic value and rewardpioneers with exceptional profits is a deeply-held belief of entre-preneurs, it is also widely acknowledged that innovation alwayscomes with risks and uncertainties. If relocation can bring cost-saving whereas innovation is risky and uncertain, why do not largefirms relocate, as environmental standards increase? To clarify this

question, the articulation between political environment and firmattributes are explored to offer a second dimension of explanation.

The awareness of the political environment is of central impor-tance particularly in the context of China where the state remains aheavy-handed actor in regulating industrial activities (Wei, 2000,2001). In the past decade, the national government has shown astronger commitment to improving environmental conditions inChina by enacting many new environmental laws, increasinginvestment in green technologies, and encouraging energy-effi-cient, innovation-led production (Lo and Tang, 2006). Meanwhile,the ‘‘decentralized governance structure’’ has also empoweredlocal governments to get involved in shaping the regional economyas planners, developers and policy-makers, and some of them havebecome heavy-handed actors that are ever more convinced of theimportance of their ‘‘steering’’ role (He et al., 2008; Wei, 2001;Wei et al., 2007). Specifically, local states in coastal regions, whereenvironmental degradation is much more severe and environmen-tal awareness amongst stakeholders is comparatively high, mightenforce environmental laws and regulations more genuinely andpay more attention to innovation-led, green production (Wangand Wheeler, 1996, 2000, 2005), while local administrations in lessdeveloped inland China with relatively lower levels of pollutiontend to hold stronger pro-growth, and relatively weaker pro-environment values (Van Rooij and Lo, 2010).

This ‘‘decentralized governance structure’’ gives rise to regionalprotectionism and inter-regional competition. For many, competi-tion between local governments led to a ‘‘race to the bottom’’ withpoor environmental standards and weak enforcement of nationallaws (PHH). For others, China is simultaneously engaged in a ‘‘raceto the top,’’ with some enterprises aggressively trying to upgradethrough investments in innovation and cleaner technology (PH).Importantly for our purposes, local governments in the less-developed inland regions regard the variation of environmentalstringency across regions as an opportunity to attract investmentand boost economic development. They lobby firms and offer lowland rent and other favorable policies (Zhu and Pickles, 2014). Theresult of these practices is increasing inter-regional competition fornew investments, with local governments in coastal provinces see-ing aggressive relocation to other provinces as weakening theirown plans for local economic development (Li and Fung ResearchCentre, 2008). Consequently, they too have become increasinglyactive in encouraging pollution-intensive enterprises to switch tocleaner technology and to upgrade in situ. However, driven bythe so-called ‘‘grasp the large, let go the small’’ policy, govern-ments at all levels tend to compete for enterprises above a certainsize that have potential not only to upgrade through investmentsin innovation and cleaner technology but also to create a consider-able amount of employment, whereas small firms, especiallyenergy- and pollution-intensive ones, have sometimes been disre-garded by regional administrations in both coastal and inlandregions (Sutherland, 2003). Additionally, small firms relocatingfrom coastal regions to inland China often encounter regional pro-tectionism as indigenous firms tend to crowd out new-comers,especially small, immature ones. In other words, regional adminis-trations in coastal regions have been encouraging and, to someextent, forcing large, key enterprises to switch to cleaner technol-ogy and to upgrade in situ; it is unlikely for small, immature firmsto obtain favorable policies even if they relocate to inland China;medium-sized, mature firms therefore might be most capable torelocate because local governments in coastal regions do not placeas many limits on their ways as on large, key enterprises and theyare often welcomed by the local governments in inland provinces(Fig. 4).

Small, immature firms are also limited by their high reliance onregional hubs. Firm-specific capability affects the net benefits firmscan receive from industrial clustering and further influences their

62 S. Zhu et al. / Geoforum 55 (2014) 53–65

location choice (Myles Shaver and Flyer, 2000). Large, mature firmswhose suppliers and clients spread out over a large area are oftenbetter able to absorb distant and even unrelated external knowl-edge and to adjust to unfamiliar cultural, political and economicenvironments than their small, immature counterparts (Teece,2007). High levels of technological and financial capabilitiesempower large firms to easily absorb uncertainties and risks ofrelocation. Small, immature firms, by contrast, need to imitateneighboring successful predecessors in the same cluster beforethey become full-fledged. In addition, most of small firms’ suppli-ers and customers are locally based, which further attaches smallfirms to local industrial clusters. The ‘‘incubator theory’’ alsoimplicitly resonates with this statement while arguing that highlycentralized locations, as ‘‘incubators’’, attract a disproportionatenumber of new firms, due in large part to advantages derived notonly from availability of sufficient skilled labor, inputs and otherrelated services but also from rapid communication possibilitieswith customers and suppliers (Hoover and Vernon, 1959). In thissense, in terms of regional hub effect, the smaller a firm is, themore it is dependent on local clusters in which it is embedding,the less likely it is to relocate (Fig. 4).

Three lines in Fig. 4 are thereafter synthesized into one inverted‘‘U’’-shaped line as shown in Fig. 5, based on which the two-dimen-sional diagram can be roughly divided into three regions:

(1) Region III: firms that fall into in this category are large, keyenterprises that have a comprehensive resource base andadequate technological knowledge to bring about innova-tion. Their large amount of output also enables them to cap-ture the economies of scale, potentially lower the averagecompliance cost per unit, and alleviate the negative impactsof environmental regulation (see firm A). Since they areregarded by local governments as crucial to employmentand investment, regional administrations in coastal regionshave been encouraging and, to some extent, forcing large,key enterprises to switch to cleaner technology and toupgrade in situ. This has created invisible obstacles on theirways to relocation and lowered their relocation capability,on the one hand, and offered them favorable policies, taxcredits and easy entry into local industrial parks (see firmA and B), on the other hand. Although they might be mostcapable to relocate to places (e.g., inland China) with com-paratively lax environmental regulation but different, unfa-miliar political, cultural and economic environments (alsosee Fig. 4), the regional hub effect is often dwarfed by theeffects of the other two dimensions. As environmental stan-dards increase, firms in Region III tend to innovate andupgrade in situ, and therefore we name Region III as ‘‘Porter

Fig. 5. Three regions of firm relocation. Source: compiled by authors.

Region’’ to indicate, in this region, increase in environmentalstandards can actually improve competitiveness, offset com-pliance costs and encourage firm upgrading.

(2) Region I: small firms with limited financial and technologicalcapability usually face resource constraints in the process ofinnovation; therefore, as environmental regulationincreases, they tend to relocate, at least their pollution-intensive production, to regions where environmental stan-dards are relatively low. Due to the ‘‘grasp the large, let gothe small’’ policy, it is almost impossible for them to obtainsubstantial favorable policies and financial aid from localgovernments in coastal regions to support innovation andupgrading. However, they are highly dependent on regionalhub effect and local industrial clusters, due not only to theirlocally-based social networks but also to their immaturityand low capability to deal with the uncertainties and risksof relocation. It is also unlikely for small, immature firmsto obtain favorable policies even if they relocate to inlandChina. As a result, neither relocation nor upgrading in situis feasible. It may be rational to close down operations thatare both pollution-intensive and render low profits, as sug-gested by the general manager of firm F. Firms that fall intothis category either have attached to and relied on largefirms (see firm E), or have been waiting and dying (firm F),or have already closed altogether.

(3) Region II: firms in this region are medium-sized enterprisesthat have a less comprehensive resource base and less ade-quate technological knowledge than large, key enterprises.For them, resources required to comply with environmentalregulations limit the amount of resources available for otherinvestments. With less favorable policies and tax creditsgranted by local governments in coastal regions, medium-sized firms’ innovation and upgrading to cleaner, greenertechnology could be risky and cause deficits that jeopardizethe entire firm. On the other hand, medium-sized firms havemore far-flung social network and higher capability to adjustto unfamiliar environment, than small firms. They are alsowelcomed by local governments in inland China. Conse-quently, firms in Region II tend to relocate their businessactivities (see firm C), or at least outsource part of their pol-lution-intensive production (see firm D), to regions whereenvironmental regulations are relatively lax. Medium-sizedfirms therefore appear to behave in ways that are suggestedby the PHH, making Region II a ‘‘Pollution Haven Region’’. Atthe interface between Region II and III, in some instances,firms may be relocating or outsourcing to new pollutionhavens, and simultaneously investing in innovation andupgrading as well (see firm C), resulting in a mixed situationwhere ‘‘Porter Region’’ and ‘‘Pollution Haven Region’’overlap.

Our two-dimensional diagram therefore reveals the mechanismunderlying the divergences of Shangyu’s pollution-intensive firms’strategies (Table 3).

Conclusion

This article formulates a heuristic analytical framework thatsynthesizes environmental regulation, regional hub effect, politicalenvironment, firm characteristics and firm restructuring in aninteractional way. The usefulness of this framework is illustratedin our case study and subsequent analysis of how firm restructur-ing has been co-shaped by various factors in ways that are muchmore nuanced than reflected by either PHH or PH alone. In drawingfrom a single case study, we realize the potential limitations of our

Table 3Comparison of pollution-intensive firms in Shangyu. Source: compiled by authors.

Firm A Firm B Firm C Firm D Firm E Firm F

Firm Size Large SmallFinancial and technological capability Strong WeakCompliance costs per unit Small BigGeographical distribution of customers and

suppliersNationally- or regionally-based Locally-based

Reliance on local clusters Low HighSupports from local governments in coastal

regionsHigh Low

Supports from local governments in inlandChina

High Low

Relocation Strategies Norelocation

Intra-regionalrelocation*

Inter-regionalrelocation

Inter-regionalrelocation**

Intra-regionalrelocation*

Norelocation

* Intra-regional relocation to industrial parks in Shangyu.** Inter-regional outsourcing or subcontracting.

S. Zhu et al. / Geoforum 55 (2014) 53–65 63

analysis. However, the case of Shangyu provides a number ofinsights to the interpretation of firm restructuring, and our find-ings echo with other recent studies (Leiter et al., 2011; Wang andLin, 2013). The high-growth, resource- and pollution-intensiveindustrialization model that China pursued has caused severe envi-ronmental pollution and deterioration, particularly in a number ofclusters in the coastal regions of East and Southeast China wherethe Reform and Opening-up policies first started. The lack ofuptake of environmental norms/values, as well as implementationdeficit of environmental regulations and policies, and the lack ofinstitutional capacity have been compounding factors. Since theearly 2000s, coastal factories have increasingly had to confront dif-ficulties generated not only by the increasing environmental costsof this regionally concentrated pollution-intensive, low efficiencygrowth model, and also by the growing social pressures againstpollution. As environmental standards were raised by China’s cen-tral government, the enforcement of environmental regulationhave been compromised more in inland China than in coastalregions, due to the ‘‘decentralized governance structure’’ andregional disparity in terms of both economic development andenvironmental pollution. This paper therefore argues that risingenvironmental regulations, especially in China’s coastal regions,have been particularly important in forcing China’s pollution-intensive enterprises to restructure their production, throughinnovation, upgrading, geographical relocation, outsourcing andplant closure.

Presently, while efforts to explain firm restructuring and envi-ronmental regulation have attracted increasing scholarly attention,most of attention is directed towards analyzing firm restructuringand environmental regulation based either on PHH or on PH aloneand extant studies have fallen short in disclosing the full picturewhere a variety of interrelated and interacting factors all haveimpacts over the articulation between environmental regulationand firm restructuring. We therefore argue that fixation oneither PHH or PH blinds us to the complex mechanism underlyingfirm restructuring. First, firms are embedded in their originallocations, to different extents, and benefit from local vertical andhorizontal linkages they have cultivated over time in local clusters.Second, incorporating state institutional actions is of centralimportance particularly in the context of China where the stateremains a heavy-handed actor in regulating industrial activities.Furthermore, not only has China’s central government been heavilyinvolved in local economic development, the ‘‘decentralizedgovernance structure’’ has also empowered local governments toparticipate directly in the development process as planners, devel-opers and policy-makers, far beyond providing just public goods.

In addition, this paper contributes to recent studies by develop-ing a heuristic analytical framework that aims to be sensitive to the

impacts of environmental regulation, political environment andregional hub effect over firm restructuring, but which does so bystressing these impacts are simultaneously inflected by firmattributes. Small firms with limited financial and technologicalcapability usually face resource constraints in the process ofinnovation; therefore, as environmental regulation increases (e.g.,in Shangyu), they tend to relocate or outsource their pollution-intensive production, to regions where environmental standardsare relatively low (e.g., inland China). However, their relocationcapability has been not only weakened by their heavy relianceon regional environment, but also inhibited by insufficientgovernment support they are able to obtain. In contrast, large,key enterprises that have a comprehensive resource base and ade-quate technological knowledge to bring about innovation havebeen encouraged and forced by local governments in coastalregions to switch to cleaner technology and to upgrade in situ.Finally, medium-sized firms in between are most likely to act inways suggested by the PHH. In total, it suggests a roughly inverted‘‘U’’-shaped relationship between firm relocation tendency andfirm size (or capability), resulting from complex interactionsbetween political environment, regional hub effect and environ-mental regulation.

This paper has two limitations. First, it only examines pollution-intensive industries, therefore the generality of the insights gainedbased on our analytical framework needs to be discussed. Second,it is impossible to incorporate all dimensions that may haveimpacts over firm restructuring into our heuristic analytical frame-work. One important dimension omitted in this research is theprice of input factors based on the ‘‘factor endowment hypothesis’’,which emphasizes that abundance in resources improves the pro-duction possibilities of firms (Copeland and Taylor, 2004). Firmstend to locate in or relocate to regions with abundant resourcesthat they need, or regions where specific resources are compara-tively cheap. These two issues can be partly solved throughmodifying our analytical framework and thinking environmentalregulation as one specific type of input factor. Increase in environ-mental standards leading to changes in compliance costs couldtherefore treated as a certain type of factor endowment change.The reason why we only focus on environmental regulation andpay less attention to other factors (e.g., labor) is that the latterhas relatively negligible impacts, compared with environmentalcosts, over firms’ strategies in pollution-intensive industries. Thisheuristic framework can thus be flexible, if we change thedimension of ‘‘environmental regulation’’ to ‘‘input factors’’ in gen-eral and, in some stances, to ‘‘labor cost’’ if we intend to analyzefirms’ strategies in the labor-intensive industries. In this sense, thisheuristic framework is not designed to pre-determine specificways to examine firm restructuring, but to adapt to both different

64 S. Zhu et al. / Geoforum 55 (2014) 53–65

industries and different contexts and to be flexible and reflexive toits analytical limits.

Acknowledgements

We are indebted to the entrepreneurs and government officialsin Shangyu’s pollution-intensive industries who generously sharedtheir time and stories. Special thanks also go to two anonymousreviewers for their constructive suggestions and comments.This work was supported by the National Natural ScienceFoundation of China (No 41271130). All translations have beendone by the authors. The authors are responsible for all errorsand interpretations.

References

Altman, M., 2001. When green isn’t mean: economic theory and the heuristics of theimpact of environmental regulations on competitiveness and opportunity cost.Ecol. Econ. 36 (1), 31–44.

Amin, A., Thrift, N., 1994. Globalization, Institutions, and Regional Development inEurope. Oxford University Press, Oxford.

Antonelli, C., 2003. The Economics of Innovation, New Technologies, and StructuralChange. Routledge, London.

Bair, J., Dussel Peters, E., 2006. Global commodity chains and endogenous growth:export dynamism and development in Mexico and Honduras. World Dev. 34 (2),203–221.

Barnes, T., Gertler, M., 1999. The New Industrial Geography: Regions. Regulationand Institutions, Routledge, London.

Ben Kheder, S., Zugravu, N., 2012. Environmental regulation and French firmslocation abroad: an economic geography model in an international comparativestudy. Ecol. Econ. 77, 48–61.

Chan, C.K., Yao, X., 2008. Air pollution in mega cities in China. Atmos. Environ. 42(1), 1–42.

Copeland, B.R., Taylor, M.S., 2004. Trade, growth, and the environment. J. Econ.Literature 42 (1), 7–71.

Costantini, V., Crespi, F., 2008. Environmental regulation and the export dynamics ofenergy technologies. Ecol. Econ. 66 (2–3), 447–460.

Dam, L., Scholtens, B., 2008. Environmental regulation and MNEs location: does CSRmatter? Ecol. Econ. 67 (1), 55–65.

Dasgupta, S., Wheeler, D., Huq, M., 1997. Bending the rules: discretionary pollutioncontrol in China, Policy Research Working Paper Series. World Bank, New York.

Dean, J.M., Lovely, M.E., Wang, H., 2009. Are foreign investors attracted to weakenvironmental regulations? evaluating the evidence from China. J. Dev. Econ. 90(1), 1–13.

Dussel Peters, E., 2008. GCCs and development: a conceptual and empirical review.Compet. Change 12 (1), 11–27.

Francesch-Huidobro, M., Lo, C.W.H., Tang, S.Y., 2012. The local environmentalregulatory regime in China: changes in pro-environment orientation,institutional capacity, and external political support in Guangzhou. Environ.Plan. A 44 (10), 2493–2511.

Granovetter, M., 1985. Economic action and social structure: the problem ofembeddedness. Am. J. Sociol. 91 (3), 481–510.

Hamamoto, M., 2006. Environmental regulation and the productivity of Japanesemanufacturing industries. Resour. Energy Econ. 28 (4), 299–312.

He, C., Pan, F., Yan, Y., 2012. Is economic transition harmful to China’s urbanenvironment? evidence from industrial air pollution in Chinese cities. UrbanStud. 49 (8), 1767–1790.

He, C., Wei, Y.H.D., Xie, X., 2008. Globalization, institutional change, and industriallocation: economic transition and industrial concentration in China. Reg. Stud.42 (7), 923–945.

Hicks, J.R., 1932. The Theory of Wages. Macmillan, London.Hoover, E.M., Vernon, R., 1959. Anatomy of a Metropolis: The Changing Distribution

of People and Jobs within New York Metropolitan Region. Harvard UniversityPress, Cambridge, Massachusetts.

Hu, A.G., Jefferson, G.H., 2008. Science and technology in China. In: Brandt, L.,Rawski, T. (Eds.), China’s Great Economic Transformation. Cambridge UniversityPress, New York.

Jahiel, A.R., 1997. The contradictory impact of reform on environmental protectionin China. China Quart. 149, 81–103.

Jahiel, A.R., 1998. The organization of environmental protection in China. ChinaQuart. 156, 757–787.

Jeppesen, T., List, J.A., Folmer, H., 2002. Environmental regulations and new plantlocation decisions: evidence from a meta-analysis. J. Regional Sci. 42 (1), 19–49.

Jo, Y., Lee, C.-Y., 2012. Technological capability, agglomeration economies and firmlocation choice. Regional Stud., 1–16.

Keller, W., 2004. Internnational technology diffusion. J. Econ. Literature 42 (3),752–782.

Kumar, S., Managi, S., 2009. Win–win opportunities and environmental regulation:test of the porter hypothesis. In: Managi, S., Kumar, S. (Eds.), The Economicsof Sustainable Development: The Case of India. Springer, US, New York,pp. 157–166.

Leiter, A.M., Parolini, A., Winner, H., 2011. Environmental regulation andinvestment: evidence from European industry data. Ecol. Econ. 70 (4),759–770.

Li & Fung Research Centre, 2008. China’s Industry Relocation and Upgrading Trends:Implications for Sourcing Business, China Distribution & Trading. Li & FungResearch Centre, Hong Kong.

Lieberthal, K., Lampton, D., 1992. Bureaucracy, Politics, and Decision Making inPost-Mao China. University of California Press, Berkeley.

Lin, G.C.S., Wang, C.C., Zhou, Y., Sun, Y., Wei, Y.H.D., 2011. Placing technologicalinnovation in globalising China: production linkage, knowledge exchange andinnovative performance of the ICT industry in a developing economy. UrbanStud. 48 (14), 2999–3018.

Lo, C.W.-H., Tang, S.-Y., 2006. Institutional reform, economic changes, and localenvironmental management in China: the case of Guangdong province. Environ.Polit. 15 (2), 190–210.

Marshall, A., 1920. Principles of Economics, eight ed. Macmillan, London.Mulatu, A., Gerlagh, R., Rigby, D., Wossink, A., 2010. Environmental regulation and

industry location in Europe. Environ. Resource Econ. 45 (4), 459–479.Murty, M.N., Kumar, S., 2003. Win–win opportunities and environmental

regulation: testing of porter hypothesis for Indian manufacturing industries. J.Environ. Manage. 67 (2), 139–144.

Myles Shaver, J., Flyer, F., 2000. Agglomeration economies, firm heterogeneity, andforeign direct investment in the United States. Strateg. Manag. J. 21 (12),1175–1193.

Piore, M., Sabel, C., 1984. The Second Industrial Divide. Basic Books, New York.Porter, M.E., 1990. The Competitive Advantage of Nations. Free Press, New York.Porter, M.E., van der Linde, C., 1995. Toward a new conception of the environment-

competitiveness relationship. J. Econ. Perspect. 9 (4), 97–118.Saxenian, A., 1994. Regional Advantage: Culture and Competition in Silicon Valley

and Route 128. Harvard University Press, Cambridge.Schumpeter, J.A., 1947. The creative response in economic history. J. Econ. Hist. 7

(02), 149–159.Scott, A., 1988. Metropolis: From the Division of Labor to Urban Form. University of

California Press, Berkeley.Shangyu Government, 2013. Economic Development in Shangyu. Shangyu

Government, Shangyu.Spatareanu, M., 2007. Searching for pollution havens: the impact of environmental

regulations on foreign direct investment. J. Environ. Dev. 16 (2), 161–182.Storper, M., 1997. The Regional World: Territorial Development in a Global

Economy. Guilford Press, New York.Sutherland, D., 2003. China’s Large Enterprises and the Challenge of Late

Industrialism. Routledge Curzon, London.Tang, S.-Y., Lo, C.W.-H., Fryxell, G.E., 2010. Governance reform, external support, and

environmental regulation enforcement in rural China: the case of Guangdongprovince. J. Environ. Manage. 91 (10), 2008–2018.

Taylor, M.S., 2004. Unbundling the pollution haven hypothesis. Adv. Econ. Anal.Policy 4 (2). Article 8.

Teece, D.J., 2007. Explicating dynamic capabilities: the nature andmicrofoundations of (sustainable) enterprise performance. Strateg. Manag. J.28 (13), 1319–1350.

Testa, F., Iraldo, F., Frey, M., 2011. The effect of environmental regulation on firms’competitive performance: the case of the building & construction sector insome EU regions. J. Environ. Manage. 92 (9), 2136–2144.

Tole, L., Koop, G., 2010. Do environmental regulations affect the location decisionsof multinational gold mining firms? J. Econ. Geogr..

Van Rooij, B., Lo, C.W.-H., 2010. Fragile convergence: understanding variation in theenforcement of China’s industrial pollution law. Law Policy 32 (1), 14–37.

Wang, C., Lin, G.C.S., Li, G., 2010. Industrial clustering and technological innovationin China: new evidence from the ICT industry in Shenzhen. Environ. Plan. A 42(8), 1987–2010.

Wang, C.C., Lin, G.C.S., 2013. Dynamics of innovation in a globalizing China: regionalenvironment, inter-firm relations and firm attributes. J. Econ. Geogr. 13 (3),397–418.

Wang, H., Wheeler, D., 1996. Pricing industrial pollution in China: an econometricanalysis of the levy system, Policy Research Working Paper Series. World Bank,New York.

Wang, H., Wheeler, D., 2000. Endogenous enforcement and effectiveness of China’spollution levy system, Policy Research Working Paper Series. World Bank, NewYork.

Wang, H., Wheeler, D., 2005. Financial incentives and endogenous enforcement inChina’s pollution levy system. J. Environ. Econ. Manage. 49 (1), 174–196.

Wang, J., 2010. Beyond Industrial Clusters: Theorectical Research on China’sIndustrial Clusters. Science Press, Beijing.

Wei, Y.H.D., 2000. Regional Development in China: States, Globalization, andInequality. Routledge, London.

Wei, Y.H.D., 2001. Decentralization, marketization, and globalization: the tripleprocesses underlying regional development in China. Asian Geogr. 20 (1–2),7–23.

Wei, Y.H.D., Li, W., Wang, C., 2007. restructuring industrial districts, scaling upregional development: a study of the Wenzhou model, China. Econ. Geogr. 83(4), 421–444.

Wen, M., 2004. Relocation and agglomeration of Chinese industry. J. Dev. Econ. 73(1), 329–347.

Wheeler, D., 1999. Racing to the bottom? foreign investment and air pollution indeveloping countries, Policy Research Working Paper Series. World Bank, NewYork.

S. Zhu et al. / Geoforum 55 (2014) 53–65 65

Yang, C., 2012. Restructuring the export-oriented Industrialization in the Pearl RiverDelta, China: institutional evolution and emerging tension. Appl. Geogr. 32 (1),143–157.

Yuan, Z., 2010. Industrial clusters in Shangyu. Distr. Econ. 8, 38–39.Zhan, X., Lo, C.W.-H., Tang, S.-Y., 2013. Contextual changes and environmental

policy implementation: a longitudinal study of street-level bureaucrats inGuangzhou, China. J. Pub. Admin. Res. Theory.

Zhang, J., Fu, X., 2008. FDI and environmental regulations in China. J. Asia PacificEcon. 13 (3), 332–353.

Zhou, N., Levine, M.D., Price, L., 2010. Overview of current energy-efficiency policiesin China. Energy Policy 38 (11), 6439–6452.

Zhu, S., Pickles, J., 2014. Bring in, go up, go west, go out: upgrading, regionalisationand delocalisation in China’s apparel production networks. J. Contemp. Asia 44(1), 36–63.