International Sociology http://iss.sagepub.com

Political-Economic Integration, Industrial Pollution and Human Health: A Panel Study of Less-Developed Countries, 19802000 Andrew K. Jorgenson International Sociology 2009; 24; 115 DOI: 10.1177/0268580908099156 The online version of this article can be found at: http://iss.sagepub.com/cgi/content/abstract/24/1/115

Published by: http://www.sagepublications.com

On behalf of:

International Sociological Association

Additional services and information for International Sociology can be found at: Email Alerts: http://iss.sagepub.com/cgi/alerts Subscriptions: http://iss.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.co.uk/journalsPermissions.nav Citations http://iss.sagepub.com/cgi/content/refs/24/1/115

Downloaded from http://iss.sagepub.com at NORTH CAROLINA STATE UNIV on January 13, 2009

Political-Economic Integration, Industrial Pollution and Human Health A Panel Study of Less-Developed Countries, 1980–2000 Andrew K. Jorgenson North Carolina State University

abstract: Bridging multiple areas of sociology, the author tests hypotheses derived from foreign investment dependence theory, ecologically unequal exchange theory and world society theory in analyses of industrial organic water pollution in less-developed countries, 1980–2000. Using panel data from a variety of sources, the author proposes three hypotheses: (1) industrial organic water pollution intensity is positively associated with foreign investment in manufacturing; (2) industrial organic water pollution intensity is positively associated with overall export intensity; and (3) industrial organic water pollution intensity is negatively associated with the presence of environmental international non-governmental organizations. Further analyses investigate the impact of organic water pollution on infant mortality rates in less-developed countries. In general, findings for Prais–Winsten regression analyses with panelcorrected standard errors and generalized least squares panel regression analyses of less-developed countries confirm the tested hypotheses, and indicate that industrial organic water pollution intensity does indeed contribute to infant mortality, net of the effects of economic development, fertility rates, health expenditures and other relevant factors. keywords: environmental sociology ✦ globalization ✦ infant mortality ✦ politicaleconomy ✦ water pollution

Introduction A growing body of social scientific research bridges the areas of global political-economy and environmental sociology to assess how different

International Sociology ✦ January 2009 ✦ Vol. 24(1): 115–143 © International Sociological Association SAGE (Los Angeles, London, New Delhi, Singapore and Washington DC) DOI: 10.1177/0268580908099156

115

Downloaded from http://iss.sagepub.com at NORTH CAROLINA STATE UNIV on January 13, 2009

International Sociology Vol. 24 No. 1

forms of structural integration impact greenhouse gas emissions (e.g. Jorgenson et al., 2007; Roberts et al., 2003), forest degradation (e.g. Jorgenson, 2006c; Shandra, 2007b), the ecological footprints of nations (e.g. Rice, 2007) and other environmental outcomes (e.g. Torras, 2005; York, 2007). The current study contributes to this surging area of integrated literatures as well as the sociology of health by focusing on an often overlooked form of environmental degradation with potentially devastating human health and ecological consequences: organic water pollution.1 More specifically, two rigorous forms of panel regression analyses are employed to test hypotheses concerning the effects of foreign investment, international trade and the presence of environmental international non-governmental organizations on industrial organic water pollution intensity in less-developed countries, between 1980 and 2000. The tested hypotheses are derived from multiple theoretical traditions, including foreign investment dependence theory, ecologically unequal exchange theory and world society theory. Further analyses evaluate the impact of industrial organic water pollution on infant mortality rates in lessdeveloped countries, net of the effects of economic development, health expenditures, fertility rates and other relevant factors. In general, findings confirm the tested hypotheses, and also provide empirical evidence of the potential health consequences of organic water pollution, which further emphasizes the importance in studying how large-scale structural factors impact the environment. In the following section I describe how different industrial/ manufacturing processes contribute to organic water pollution, and I discuss the potential human health consequences associated with exposure to organic water pollutants. This is followed by a literature review in which I summarize relevant points of theorization and prior research, which lead to the formalization of the hypotheses tested in the subsequent analyses. Following the literature review and presentation of hypotheses, the analyses section describes (1) the samples and panel regression methods employed in the study and (2) the dependent and independent variables included in the reported models. Next, I present and summarize the findings for the analyses, with a particular focus on the three hypotheses tested as well as the impact of organic water pollution on infant mortality rates in less-developed countries. I conclude by discussing the key results of the study as well as their theoretical relevance, and make a call for other social scientists to utilize their methodological tools and analytical skills to investigate how different political-economic factors contribute to forms of environmental degradation, particularly forms that have potentially serious ecological and human well-being costs. 116

Downloaded from http://iss.sagepub.com at NORTH CAROLINA STATE UNIV on January 13, 2009

Jorgenson Industrial Pollution and Human Health

Industrial Organic Water Pollution and Human Health Organic water pollutants from manufacturing can be traced to a variety of productive activities, including steelmaking, pulp and paper manufacturing, food processing, the processing of industrial chemicals and textile production (Eckenfelder, 2000; WRI, 2005). Water pollution from steel production results from (1) the leaching of electric arch furnace dust into ground water and (2) the dispersal of water used to cool coke after it has finished baking (Andres and Irabien, 1994). Some of the most common organic pollutants in effluents from pulp and paper manufacturing are lost cellulose fiber, starch and hemi-cellulose and carbohydrate (Stanley, 1996). Food processing plants with inadequate waste treatment facilities are a source of disease causing organisms, bacteria, viruses and parasites (Manahan, 2005), and clothing manufacturing involves a variety of dyes and fixers, some of which become waste. These forms of waste, which contain pathogenic organisms, make their way into lakes, rivers, canals and other untreated bodies of water, many of which are used by human populations for different subsistence purposes, especially by those living in higher poverty conditions (Frey, 2003; WRI, 2005). Different synthetic and natural organic chemicals are also used to produce pesticides, plastics, paint pigments, medicines and other commonly used products (Manahan, 2005). If these sorts of chemicals are not handled carefully or disposed of properly, they are more likely to contaminate public water sources. Overall, the use of organic chemicals in different productive activities results in waste products that often end up as water pollutants, and many of these organic materials are highly toxic and capable of remaining in the environment for long periods of time. This is most common with organic chemicals that are highly resistant to decomposition and biodegradation (Eckenfelder, 2000). Like many other forms of human-caused environmental degradation (e.g. greenhouse gas emissions, deforestation), in recent decades industrial organic water pollution has increased dramatically in many lessdeveloped countries (World Bank, 2007; WRI, 2005). While organic water pollutants can cause great harm to local and regional aquatic ecosystems (e.g. Miller, 2000), here I am most concerned with their potential human health consequences. The human use of polluted water is associated with serious health outcomes, including birth defects, spontaneous abortion, various types of cancer and death (Cadbury, 2000; McGinn, 2000). A large proportion of all diseases and deaths are caused by the consumption of polluted water, and water-borne diseases are perhaps the largest category of communicable diseases contributing to infant mortality in less-developed

117

Downloaded from http://iss.sagepub.com at NORTH CAROLINA STATE UNIV on January 13, 2009

International Sociology Vol. 24 No. 1

countries (WRI, 2005). In general, infants’ immune systems are less developed than adults, and thus less capable of resisting or battling disease-causing organisms found in polluted water. Moreover, organic pollutants can accumulate in the fatty tissue of mammals, including humans. This process is often referred to as a form of ‘bioaccumulation’ or ‘biomagnification’ (Czub and McLachlan, 2004), and organic pollutants are known to accumulate in the fatty tissue of women, which can be passed on to infants and young children through breastfeeding (e.g. Burns et al., 2003). These human health consequences clearly point to the importance in studying the social structural causes of industrial organic water pollution.

Literature Review and Hypotheses Tested in the Current Study Foreign Investment and the Environment During recent decades, many less-developed countries experienced a deepening of foreign debt, which resulted in austerity measures developed by global governance institutions (e.g. Stiglitz, 2002). These austerity measures, especially structural adjustment programs, encourage governments of indebted countries to eradicate structural barriers that deter foreign investment. Attracting foreign capital is often considered a means of stimulating economic growth to assist in debt repayment while increasing the overall quality of life for domestic populations (e.g. OECD, 1999). Thus, in an effort to attract transnational corporations and foreign investment, many less-developed countries have attempted to create and maintain more favorable business conditions, including tax reductions and relaxed labor laws as well as exemptions to environmental laws and regulations designed to protect the natural environment from activities in both the secondary and primary sectors (e.g. Jorgenson, 2007; Leonard, 1988). The threat of capital flight (real or perceived) is an additional incentive for less-developed countries to offer regulatory concessions to foreign-based firms (Wallerstein, 2005), and many less-developed countries are also less likely to ratify international environmental treaties, some of which deal explicitly with productive activities that are of direct relevance for transnational firms (Roberts and Parks, 2007). Globalization scholars consider these deregulation processes to be part of an overall form of political globalization (Chase-Dunn and Jorgenson, 2007; Chase-Dunn et al., 2000), or what Phil McMichael (2004) refers to as the ‘globalization project’. Partly resulting from these emergent political-institutional dynamics,2 the relative presence of foreign investment stocks within lessdeveloped countries increased almost sevenfold during recent decades from 4 percent of their overall GDP in 1980 to approximately 28 percent in 2000 (OECD, 2001; United Nations, 1992, 1994, 1996, 2000, 2003). 118

Downloaded from http://iss.sagepub.com at NORTH CAROLINA STATE UNIV on January 13, 2009

Jorgenson Industrial Pollution and Human Health

A growing body of macrosociologists and international relations scholars attempt to advance an ‘ecostructural’ framing of foreign investment dependence theory (e.g. Grant et al., 2002; Jorgenson, 2003; Jorgenson et al., 2007), and the latter has a long history in comparative sociology and related social science traditions3 (e.g. Bornschier and Chase-Dunn, 1985; Chase-Dunn, 1975). With the aforementioned political-institutional dynamics and conditions in mind, it is argued by ecostructural investment dependence theorists that a large proportion of foreign investment in less-developed countries finances highly polluting and ecologically inefficient manufacturing processes and facilities, much of which are outsourced from developed countries (e.g. Grimes and Kentor, 2003; Jorgenson, 2007). Recent cross-national studies of greenhouse gas emissions as well as other forms of air and water pollution support these assertions (e.g. Jorgenson, 2006a, 2006b, 2007; Kentor and Grimes, 2006). Moreover, some political-economic sociologists and international relations scholars would argue that the environmental impacts of foreign investment dependence are indicative of the relative ‘North/South’ externalizing of environmental costs that are likely to take place in transnationally organized and globally distributed production (e.g. Jorgenson, 2003; Redclift and Sage, 1998; Rice, 2007). The outsourcing of environmentally harmful production processes by transnational firms, primarily those headquartered in more-developed countries, is also illustrative of the ‘Netherlands Fallacy’ analogy in environmental sociology and ecology, which points to the invalid assumption that the consumption-based and production-based environmental impacts of the Netherlands population (and other affluent societies) stay within their geopolitical borders (e.g. Ehrlich and Holdren, 1971; York and Rosa, 2003). Drawing from this growing body of research and theorization on the potential environmental impacts of foreign investment, in the subsequent analyses I test the hypothesis (H1) that within less-developed countries, industrial organic water pollution intensity (emissions per 1000 workers) is positively associated with foreign investment in manufacturing.

Environmental Impacts of International Trade Many sociologists and other social scientists (e.g. Hornborg et al., 2007; Jorgenson and Kick, 2006; Lofdahl, 2002) assert that the recent upswing in the globalization of trade (Chase-Dunn et al., 2000) is positively associated with different forms of environmental degradation, particularly in less-developed countries. While the world economy experienced an overall increase in the globalization of trade in recent decades, the extent of this increase is even more pronounced for less-developed countries. For example, the level of export intensity (exports as percentage GDP) for all less-developed countries combined almost doubled from 1980 119

Downloaded from http://iss.sagepub.com at NORTH CAROLINA STATE UNIV on January 13, 2009

International Sociology Vol. 24 No. 1

(16.04 percent) to 2000 (28.77 percent) (World Bank, 2007).4 A common explanation for the association between increased environmental degradation and the globalization of trade, which partly corresponds with theorization concerning the environmental impacts of foreign investment dependence, is that in order to be relatively competitive in the world economy, international trade creates added pressures for less-developed countries to lower environmental standards for export-oriented production (e.g. Schofer and Hironaka, 2005). This is even more pronounced for highly indebted countries since structural adjustment programs often involve the promotion of export-oriented production as a means of stimulating revenue to put towards debt repayment (e.g. McMichael, 2004). Moreover, the theory of ecologically unequal exchange (e.g. Hornborg et al., 2007; Jorgenson, 2006c; Rice, 2007; Roberts and Parks, 2007) asserts that the globalization of trade, which is intimately tied to the broadening and deepening of global commodity production, helps to create conditions that further enable the populations of more-developed countries to partially externalize their environmental costs to less-developed countries. Put differently, the populations of more-affluent, higher-consuming countries are positioned advantageously in the contemporary world economy, and thus more likely to secure and maintain favorable terms of trade allowing for greater access to the natural resources and sink capacity of areas within less-developed countries. This greater access facilitates the externalization of environmental and human well-being costs of production, consumption and resource extraction to less-developed countries, which contributes to heightened resource depletion and pollution within their borders. Prior cross-national studies of greenhouse gas emissions (e.g. Jorgenson, 2007; Schofer and Hironaka, 2005), forest degradation (e.g. Kick et al., 1996) and the ecological footprints of nations (e.g. Jorgenson, 2009; Jorgenson and Burns, 2007; Rice, 2007) lend support to arguments concerning the environmentally destructive aspects of international trade for less-developed countries. To further assess the validity of these theoretical articulations, in this study I test the hypothesis (H2) that within less-developed countries, industrial organic water pollution intensity is positively associated with export intensity (exports as percentage GDP).

World Society Theory and EINGOs According to Smith and Wiest (2005), the total number of international non-governmental organizations (transnationally organized citizen groups) grew from fewer than 1000 in the 1950s to almost 20,000 by the year 1999. Along with their growth in overall numbers, the relative presence of such organizations in both developed and less-developed countries has also increased dramatically in recent decades (Smith and Wiest, 120

Downloaded from http://iss.sagepub.com at NORTH CAROLINA STATE UNIV on January 13, 2009

Jorgenson Industrial Pollution and Human Health

2005; Wiest and Smith, 2007). Some scholars, drawing from the world society approach in macrosociology (e.g. Boli and Thomas, 1999; Meyer et al., 1997), assert that the presence of environmental international nongovernmental organizations (EINGOs) is likely to influence the use of more environmentally friendly forms of production, extraction and consumption (e.g. Frank et al., 2000; Shandra, 2007a). EINGOs help to constitute, diffuse and reinforce world cultural norms while also intervening in global political processes to help frame the language of environmental treaties and policies. These forms of constitution, diffusion and reinforcement are constructed through global cultural and associational processes (Meyer et al., 1997: 144–5). Environmental treaties and related policies can also create conditions in which EINGOs are able to effectively target governments to fulfill their own environmental policy commitments and environmental stewardship roles. What is more, EINGOs can change the practices of governments and multilateral institutions through direct lobbying to enact trade restrictions and more rigorous production regulations (Shandra, 2007b). It is also quite common for EINGOs to facilitate and support domestic social movements, related environmental protection efforts and conservation programs and provide technical assistance to domestic organizations and local communities (Bradshaw and Schafer, 2001; Frank, 1999; Frank et al., 2000). EINGOs can also pursue the development and promotion of environmental problem-solving initiatives at a more subnational level, which can serve as substitutes for government action (e.g. Ndegwa, 1996; Schofer and Hironaka, 2005). These sorts of activities often involve the creation of technical guidelines and drafting codes of conduct (Boli and Thomas, 1999; Shandra, 2007a). As discussed by Newell (2000) and Shandra (2007b), the Forestry Stewardship Council’s voluntary accreditation program serves as a clear example. Primary sector corporations that participate in the accreditation program must agree to and follow a series of explicit forest management principles. The Forestry Stewardship logo implies that the corresponding wood product originates from an independently certified forest that meets the principles of the EINGO in question. It is quite possible that these sorts of programs could be effective in the secondary sector (i.e. manufacturing) as well as other primary sector activities (e.g. mining, agriculture). Thus, governments and private sector enterprises can be directly and indirectly influenced by EINGOs to increase protection of the natural environment, and recent cross-national investigations of deforestation and carbon dioxide emissions support these assertions (Schofer and Hironaka, 2005; Shandra, 2007a, 2007b). Building on world society theorization and prior empirical research summarized in the preceding discussion, I test the following hypothesis (H3) in the current study: within less-developed countries, industrial organic 121

Downloaded from http://iss.sagepub.com at NORTH CAROLINA STATE UNIV on January 13, 2009

International Sociology Vol. 24 No. 1

water pollution intensity is negatively associated with the presence of EINGOs.

The Analyses Samples and Methods Used in the Analyses To (1) test the study’s three hypotheses and (2) investigate the effect of industrial organic water pollution on infant mortality rates, I analyze panel data ranging in years from 1980 to 2000. Consistent with most research in this tradition, I restrict the analyses to less-developed countries. Lessdeveloped countries are defined as those not falling into the top quartile of the World Bank’s (2007) country-level income quartile classification. To maximize the use of available data, I allow the actual sample sizes to vary among all tested models for both the water pollution and infant mortality analyses. Due to substantial differences in the availability of panel data for the two dependent variables and all independent variables of interest, the overall sample sizes as well as mean number of observations per country differ substantially for the tested models. The sample sizes in the tested models range from 83 to 337, with the mean number of observations per country ranging from 3 to 11.233. The Appendix lists the countries included in the different series of reported analyses. Because of my commitment to utilize all available data, which explains the rather unbalanced nature of the study’s panel dataset, I use two different methods suitable for the analyses of cross-national panel data (Beck and Katz, 1995; Halaby, 2004). For models with an average of at least 10 observations per country, I employ cross-sectional time-series Prais–Winsten (PW) regression with panel-corrected standard errors (PCSE), allowing for disturbances that are heteroskedastic and contemporaneously correlated across panels (Beck and Katz, 1995). The models that employ PW regression with PCSE are those that test the first two hypotheses. In all reported PW regression analyses I correct for first-order autocorrelation (AR(1)), treating the AR(1) process as common to all panels. In the PW regression models I also include dummy-coded variables for each unit (country) and for each period (year). This approach controls for the potentially unobserved heterogeneity that is cross-sectionally invariant within years (period effects) and temporally invariant within countries (e.g. geography, natural resource endowments). Overall, this type of model is robust against omitted control variables, more closely approximates experimental conditions and allows for more rigorous testing of hypotheses (Greene, 2000; Wooldridge, 2002). For models with fewer than an average of 10 observations per country, I employ generalized least squares (GLS) random effects (RE) panel regression with robust standard errors5 (e.g. Frees, 2004; Hsiao, 2003). The 122

Downloaded from http://iss.sagepub.com at NORTH CAROLINA STATE UNIV on January 13, 2009

Jorgenson Industrial Pollution and Human Health

models that employ GLS RE panel regression are the analyses of water pollution that test Hypothesis 3 as well as all reported analyses of infant mortality rates. The RE model treats country-specific intercepts as a random component of the error term. In analyses where the time dimension is relatively smaller, such as three to seven time points, a RE modeling approach is preferable to the common ordinary least squares (OLS) fixedeffects (FE) approach or PW regression with PCSE (with fixed effects) because fewer degrees of freedom are necessary to account for the subjectspecific (e.g. country-specific fixed-effects) parameters (Frees, 2004: 78). Also, results of the Hausman test statistic (all non-significant) indicate that GLS RE regression is more appropriate than OLS FE regression for all the reported models where the average number of observations per country is fewer than 10. I also include dummy-coded variables (unreported) for each year in all GLS RE models, which controls for potentially unobserved heterogeneity that is cross-sectionally invariant within years (i.e. period effects).

The Dependent Variables The first dependent variable, which I conceptualize as industrial organic water pollution intensity, is a measure of the overall amount of industrial organic water pollutants per 1000 workers. More specifically, these data measure the amount of industrial organic water pollutant emissions in kilograms per day divided by the amount of industrial workers in thousands. These data, which are logged (ln) to minimize skewness, are obtained from the World Bank (2006, 2007). Organic water pollutants are measured by biochemical oxygen demand (BOD), which refers to the amount of oxygen that bacteria in water will consume in breaking down waste. This is a standard water-treatment test for the presence of organic pollutants. In particular, the measures include water pollutants from manufacturing activities as defined by the two-digit divisions of the International Standard Industrial Classification revision, which consists of organic water pollutants from the manufacturing of primary metals, paper and pulp, chemicals, food and beverages, stone, ceramics, glass, textiles, wood and manufactured goods included in the two divisions of classification labeled as ‘other’ manufactured goods (divisions 38 and 39). Industrial organic water pollution intensity is also the primary independent variable of interest in the analyses of infant mortality. The second dependent variable is infant mortality rate, which measures the number of infants who die before reaching one year of age, per 1000 live births in a given year. These data are gathered from the World Bank (2006, 2007). The main sources of mortality data are direct or indirect estimates based on sample surveys or censuses and vital registration systems. According to the World Bank (2006), a ‘complete’ vital registration system, 123

Downloaded from http://iss.sagepub.com at NORTH CAROLINA STATE UNIV on January 13, 2009

International Sociology Vol. 24 No. 1

which involves a system covering at least 90 percent of the population, is the best source of age-specific mortality data. However, complete vital registration systems are fairly uncommon in less-developed countries. Thus, for many less-developed countries, estimates must be derived from sample surveys or by applying ‘indirect’ estimation techniques to census, survey or registration data.

Independent Variables Used for Hypothesis Testing in the Water Pollution Analyses To test Hypothesis 1, I employ the following measure: accumulated stocks of secondary sector foreign direct investment as percentage of total GDP. These data are logged (ln) to minimize skewness. Stocks as percentage of total GDP are the most commonly used measure of foreign investment dependence in macro-comparative research6 (e.g. Alderson and Nielsen, 1999; Jorgenson, 2006a; Jorgenson et al., 2007; Kentor, 2001). Foreign direct investment stocks data are obtained from the United Nations’ World Investment Directories (1992, 1994, 1996, 2000, 2003) and the OECD’s International Direct Investment Statistics Yearbook (2001). Total GDP data are measured in 1995 US dollars (World Bank, 2006). The measures of secondary sector foreign direct investment stocks consist of investment in the following manufacturing activities: food and beverages, tobacco, textiles and clothing, leather, wood and wood products, publishing and printing, coke, petroleum products, nuclear fuel, chemicals and chemical products, rubber and plastic products, non-metallic mineral products, metal and metal products, machinery and equipment, electrical and electronic equipment, precision instruments, motor vehicles and other transport equipment, other manufacturing and recycling. Exports as percentage of total GDP, which control for the extent to which a country is integrated into the international trading system, are used to test Hypothesis 2 in the water pollution analyses. These data, which I log to minimize skewness (ln), are obtained from the World Bank (2007). Prior research on environmental outcomes treats this measure as an indicator of export intensity (e.g. Jorgenson and Burns, 2007; Rice, 2007). While I would prefer to use measures that include only exports in manufactured goods, to the best of my knowledge, adequate panel data for manufacturing exports as percentage total GDP are currently unavailable. Thus, I employ a less nuanced, but commonly used measure of exports. I use two EINGO measures to test Hypothesis 3 in the water pollution analyses. The first measure is EINGO presence, which represents the number of EINGOs that report having members in a given country. The second measure is EINGO intensity. To calculate the intensity measure, I divide the total number of EINGOs in a nation by its population in millions (World Bank, 2007). The INGO data were first gathered by Smith 124

Downloaded from http://iss.sagepub.com at NORTH CAROLINA STATE UNIV on January 13, 2009

Jorgenson Industrial Pollution and Human Health

(2004) and Smith and Wiest (2005), who collected information from The Yearbook of International Associations and coded those INGOs explicitly focusing on the environment. I include both EINGO measures for two reasons. First, there is no theoretical specification for using one over the other. Second, EINGO presence is only moderately correlated with population size.7

Additional Independent Variables Included in the Water Pollution Analyses Gross domestic investment as percentage of total GDP represents the level of domestic investment in fixed assets plus net changes in inventory levels. I obtain these data from the World Bank (2006). Controlling for domestic investment allows for a more rigorous assessment of the effect of foreign investment on organic water pollution intensity (H1). Of substantive relevance, some scholars suggest that domestically controlled manufacturing is likely to be less environmentally harmful than foreigncontrolled manufacturing (e.g. Jorgenson, 2006a). Partly through effective pressures by local organizations and communities, domestic investors and firms are more likely than transnational firms and foreign capital to invest in ‘greener’ methods of production (Young, 1997). I would prefer measures of domestic investment for only the manufacturing sector. However, those types of data were unavailable at the time of this study. Since the measures of organic water pollution deal explicitly with manufacturing sector activities, in the water pollution analyses I control for manufacturing as percentage of total GDP, which quantifies the extent to which a domestic economy is manufacturing based. These data are gathered from the World Bank (2006). Additional Independent Variables Included in the Water Pollution and Infant Mortality Analyses GDP per capita is included as a control for level of economic development. These data, which I gather from the World Bank (2000, 2005), are measured in 1995 US dollars and logged (ln) to minimize skewness. Scholars working in the world-systems tradition and similar politicaleconomic approaches (e.g. Roberts et al., 2003) assert that middle-developed or ‘semi-peripheral’ countries tend to have enough manufacturingoriented technologies to compete on the world market but not enough technologically advanced infrastructure to do so as ‘eco-efficiently’. Lessdeveloped or ‘peripheral’ countries tend to have relatively less manufacturing technologies and capital-intensive production. Thus, when excluding the most-developed countries (see definition of the countries presented earlier), one might expect level of development to be positively associated with industrial organic water pollution intensity (see also Jorgenson, 125

Downloaded from http://iss.sagepub.com at NORTH CAROLINA STATE UNIV on January 13, 2009

International Sociology Vol. 24 No. 1

2007). Turning to infant mortality, prior research consistently shows a negative association between infant mortality rates and level of development in less-developed countries (e.g. Chung and Muntaner, 2006; Moore et al., 2006; Ram, 2006; Shen and Williamson, 2001). Level of democracy/autocracy are used as a measure of democratization. These data, which I gather from the World Resources Institute (WRI, 2005), are indexed on a scale from −10 to +10. A score of +10 indicates a strongly democratic state; a score of −10 indicates a strongly autocratic state. A fully democratic government has three essential elements: fully competitive political participation, institutionalized constraints on executive power and guarantee of civil liberties to all citizens in their daily lives and in political participation. A fully autocratic system sharply restricts or suppresses competitive political participation. The chief executives are chosen by an elite group and exercise power with few institutionalized constraints. The WRI obtains these data from the Polity IV Project.8 Both ecological modernization theory and political modernization theory assert that democratization can lead to environmental reforms and more sustainable production processes because it creates conditions in which concerned groups and organizations influence policy development and behavior (e.g. Ehrhardt-Martinez et al., 2002; Mol, 2001). Prior research also finds a negative association between infant mortality rates and democratization in less-developed countries (e.g. Shandra et al., 2004). Secondary education is included as a measure of human capital. More specifically, these data, which I gather from the World Bank (2006), quantify percent gross secondary school enrollment. Gross enrollment ratio is the ratio of total enrollment, regardless of age, to the population of the age group that officially corresponds to the level of education shown. According to the World Bank (2006), secondary education completes the provision of basic education that began at the primary level, and aims at laying the foundations for lifelong learning and human development, by offering more subject- or skill-oriented instruction using more specialized teachers. Prior research links this form of human capital to lower levels of environmental degradation (e.g. Torras, 2005) and lower infant mortality rates (e.g. Frey and Field, 2000; Moore et al., 2006; Ram, 2006; Shen and Williamson, 2001).

Additional Independent Variables Included in the Infant Mortality Analyses To conduct a more rigorous assessment of the effect of industrial organic water pollution on infant mortality in less-developed countries, I control for fertility rates, which is consistently shown in prior research to be a key contributor to infant and child mortality rates (e.g. Bongaarts and Watkins, 1996; Brady et al., 2007; Heuveline, 2001). More generally, as fertility increases, so does child mortality and infant mortality, since more 126

Downloaded from http://iss.sagepub.com at NORTH CAROLINA STATE UNIV on January 13, 2009

Jorgenson Industrial Pollution and Human Health Table 1 Descriptive Statistics

Industrial organic water pollutants (ln) Infant mortality rate Secondary sector FDI stocks as % GDP (ln) Exports as % GDP (ln) Domestic investment as % GDP EINGO presence EINGO intensity (ln) Manufacturing as % GDP GDP per capita (ln) Fertility rate Democratization Secondary education Health expenditures as % GDP Health expenditures per capita (ln)

N

Mean

527

5.301

253 337

SD

Skewness

Min.

Max.

.215

.275

4.700

5.940

59.821 1.354

31.275 .653

.481 .216

8.300 .070

144.000 3.640

522

2.906

.608

−.485

.770

4.560

522

21.153

6.874

.333

.002

43.920

83 83 516 513 376 527 462 99

15.024 .473 17.456 6.745 4.280 1.266 41.468 5.001

10.251 .476 7.244 1.005 1.511 6.495 18.973 2.243

1.018 1.731 .574 .182 .588 −.225 .215 .530

1.000 .010 .800 4.840 1.890 −9.000 3.300 .990

49.000 2.270 40.720 8.940 8.500 10.000 98.600 11.480

97

4.098

1.199

−.064

1.160

6.690

fertility means more chances for mortality, all else being equal. The measures of fertility rates, which we obtain from the World Bank (2007), specifically represent the number of children that would be born to a woman if she were to live to the end of her childbearing years and bear children in accordance with current age-specific fertility rates. I also include two measures of health expenditures in the infant mortality analyses. Indeed, prior research links health expenditures to lower mortality rates (e.g. Chung and Muntaner, 2006; Shandra et al., 2004). Both measures are gathered from the World Bank (2006). The first measure is health expenditures as percentage of total GDP. Health expenditures are the sum of public and private health expenditures, and include the provision of health services (preventive and curative), family planning activities, nutrition activities and emergency aid designated for health. The second health measure is health expenditures per capita. These data, which we log (ln) to minimize skewness, are the average amount of total health expenditures per person, measured in US dollars. Table 1 presents the descriptive statistics and Table 2 provides the pairwise bivariate correlations for all variables included in the reported analyses. 127

Downloaded from http://iss.sagepub.com at NORTH CAROLINA STATE UNIV on January 13, 2009

Downloaded from http://iss.sagepub.com at NORTH CAROLINA STATE UNIV on January 13, 2009

Industrial organic water pollutants (ln) Infant mortality rate Secondary sector FDI stocks as % GDP (ln) Exports as % GDP (ln) Domestic investment as % GDP EINGO presence EINGO intensity Manufacturing as % GDP GDP per capita (ln) Fertility rate Democratization Secondary education Health expenditures as % GDP Health expenditures per capita (ln)

.242 .071

.216 −.422

−.161 .347 −.381

.057 .345 .089 −.232 .375

.247

2. 3.

4. 5.

6. 7. 8.

9. 10. 11. 12. 13.

14.

1.

1.

Table 2 Pairwise Bivariate Correlations

−.658

−.608 .788 −.404 −.740 −.352

−.303 −.298 −.531

−.380 −.481

−.424

2.

.286

.388 −.230 −.034 .191 .118

.116 .323 .171

.474 .152

3.

.123

.211 −.103 .158 .165 −.080

.053 .405 −.068

.259

4.

.147

.292 −.505 .092 .449 −.079

.168 −.078 .481

5.

.264

.186 −.482 .323 .473 .043

.209 .177

6.

.315

.184 −.151 .328 .122 .448

−.128

7.

.421

.440 −.588 .149 .518 .129

8.

.906

−.537 .445 .572 .418

9.

−.567

−.443 −.758 −.274

10.

.492

.421 .429

11.

.504

.206

12.

.703

13.

Jorgenson Industrial Pollution and Human Health

Findings Table 3 presents the results of the PW regression with PCSE, which test the first two hypotheses. I report findings for four models, all of which include secondary sector foreign investment stocks as percent total GDP, domestic investment as percent total GDP and exports as percent total GDP. Models 1 and 2 each include a fourth predictor: manufacturing as percent total GDP in Model 1 and GDP per capita in Model 2. The latter is also included in Models 3 and 4. I include democratization and secondary education as additional predictors in Model 3, and Model 4 consists of predictors with statistically significant effects in any of the preceding three models. In all reported regression analyses (Tables 3–5), I provide unstandardized regression coefficients, which I flag for statistical significance. I also provide standardized regression coefficients and standard errors as well as R2 values, sample sizes and mean number of observations per country for each tested model. The relatively high R2 values for the PW regression analyses are not surprising since all four models include both period-specific and nation-specific intercepts. The results presented in Table 3 indicate that secondary sector foreign investment stocks as percent GDP positively affect industrial organic water pollution intensity, and the positive effect is statistically significant across all four models.9 Thus, the PW regression with PCSE analyses confirms the first hypothesis, and provides support for ecostructural theorization concerning the environmental impacts of foreign investment dependence and the outsourcing of environmental degradation through the transnational organization of production (e.g. Jorgenson, 2006b; Jorgenson et al., 2007). Turning to the second hypothesis, findings suggest that export intensity positively affects industrial organic water pollution intensity as well. However, the positive effect of exports as percent GDP is non-significant in Model 3. Compared to the first hypothesis, the PW regression analyses of export intensity are less conclusive, but the results do indeed provide moderate confirmation of Hypothesis 2. These findings also lend some support to the assertions of ecologically unequal exchange theorists (e.g. Hornborg et al., 2007; Rice, 2007) as well as related political-economic arguments concerning the possible environmental impacts of the globalization of trade, particularly in the context of export intensity (e.g. Schofer and Hironaka, 2005). The effect of domestic investment as percent GDP is negative in all four models, while the effect of manufacturing as percent GDP is non-significant. The former corresponds with assertions made by Young (1997) and Jorgenson (2006a), who both suggest that for a variety of reasons, domestically owned firms are more likely than transnational firms to invest in environmentally friendly forms of manufacturing. The non-significant effect of the latter, combined with the divergent effects of foreign investment and domestic 129

Downloaded from http://iss.sagepub.com at NORTH CAROLINA STATE UNIV on January 13, 2009

International Sociology Vol. 24 No. 1 Table 3 Results from Cross-Sectional Time-Series Prais–Winsten Regression of Industrial Organic Water Pollutants per 1000 Workers with Panel-Corrected Standard Errors Correcting for AR(1) Disturbances

Secondary sector FDI stocks as % GDP (ln) Exports as % GDP (ln)

Domestic investment as % GDP Manufacturing as % GDP

Model 1

Model 2

Model 3

Model 4

.074*** [.225] (.019) .069*** [.194] (.018) −.003** [−.087] (.001) .001 [.009] (.002)

.060*** [.181] (.021) .079*** [.226] (.019) −.010*** [−.318] (.002)

.085*** [.258] (.020) .023 [.065] (.018) −.012*** [−.369] (.002)

.083*** [.254] (.019) .033** [.094] (.019) −.012*** [−.371] (.002)

.039*** [.186] (.012)

.030** [.142] (.014) .001 [.034] (.002) −.003*** [−.256] (.001) .997 299 10.679

.038*** [.181] (.014)

GDP per capita (ln)

Democratization

Secondary education

R2 N Mean number of observations per country

.997 297 11.000

.996 337 11.233

−.003*** [−.268] (.001) .997 299 10.679

Notes: *p <.075, **p <.05, ***p <.01 (one-tailed tests); standardized coefficients are in square brackets; standard errors are in parentheses. The models include nation-specific and period-specific intercepts that are not shown. Dependent variable is logged (ln) to minimize skewness.

investment, underscores the importance in considering the environmental impacts of the organization of production as well as the scale of production. Organic water pollution intensity is positively associated with per capita GDP, the effect of democratization is non-significant and the effect of secondary education on this form of industrial pollution is negative. While they are not the focus of the current study, the statistically significant effects of economic development and education levels are quite consistent with prior research in this general tradition (e.g. Roberts et al., 2003; Torras, 2005). Controlling for 130

Downloaded from http://iss.sagepub.com at NORTH CAROLINA STATE UNIV on January 13, 2009

Jorgenson Industrial Pollution and Human Health Table 4 Results from Generalized Least Squares Random Effects Panel Regression of Industrial Organic Water Pollutants per 1000 Workers

EINGO presence

EINGO intensity (ln)

Manufacturing as % GDP

Model 1

Model 2

Model 3

Model 4

−.002*** [−.132] (.001) .078 [.212] (.047) −.005 [−.194] (.004)

−.003*** [−.157] (.001) .074 [.200] (.050)

−.003*** [−.163] (.001) .003 [.008] (.048)

−.002*** [−.138] (.001)

.053** [.193] (.029) .030 [.108] (.035) −.008*** [−.263] (.002) .243 83 3

.066*** [.243] (.028)

GDP per capita (ln)

−.009 [−.055] (.034)

Secondary sector FDI stocks as % GDP (ln) Exports as % GDP (ln)

Domestic investment as % GDP R2 N Mean number of observations per country

.161 83 3

.090 83 3

−.008*** [−.256] (.002) .256 83 3

Notes: *p <.075, **p <.05, ***p <.01 (one − tailed tests); standardized coefficients are in square brackets; robust standard errors are in parentheses. The models include period-specific intercepts that are not shown. Dependent variable is logged (ln) to minimize skewness.

democratization appears to suppress the effect of export intensity, but its nonsignificant effect on industrial water pollution contrasts with arguments made by some political and ecological modernization theorists (e.g. EhrhardtMartinez et al., 2002), who suggest that higher levels of democratization lead to the implementation of more environmentally friendly forms of manufacturing since it creates conditions in which concerned groups can influence policy development and organizational behavior. Table 4 reports the findings for the GLS RE analyses of industrial organic water pollution, which test the third hypothesis. It warrants 131

Downloaded from http://iss.sagepub.com at NORTH CAROLINA STATE UNIV on January 13, 2009

International Sociology Vol. 24 No. 1

noting that the sample sizes and mean number of observations per country are both dramatically smaller than for the analyses reported in Table 3. These differences are a function of the relatively limited availability of EINGO panel data for less-developed countries. Results of four tested models are reported. Model 1 consists of both EINGO measures as well as manufacturing as percent GDP, and Model 2 includes both EINGO measures as well as per capita GDP. Model 3 consists of both EINGO measures as well as the measures of foreign investment, export intensity and domestic investment. Model 4 includes only predictors found to have statistically significant effects in any of the three preceding models. Findings indicate that EINGO presence negatively affects industrial organic water pollution intensity, but the effect of EINGO intensity is nonsignificant. As I discuss earlier, there is no theoretical justification for using one of these two measures over the other. Thus, it appears – at least in the context of industrial water pollution – that the number of EINGOs present in a country is of great relevance. While these particular analyses could be considered more exploratory since two EINGO measures are used, I argue that their results do generally confirm the current study’s third hypothesis that industrial organic water pollution intensity is negatively associated with EINGO presence in less-developed countries. These findings also correspond with prior studies of deforestation and greenhouse gas emissions (e.g. Schofer and Hironaka, 2005; Shandra, 2007a) as well as world society theorization (e.g. Frank et al., 2000; Meyer et al., 1997). The effects of manufacturing as percent GDP and level of development are both non-significant, with the latter contrasting the PW regression analyses reported in Table 3. The effect of export intensity is also non-significant in the RE analysis, while the effect of secondary sector foreign investment as percent GDP is positive, and the effect of domestic investment on water pollution intensity is negative. Table 5 presents the results of the GLS RE analyses, which assess the effect of industrial organic water pollution on infant mortality rates in less-developed countries. I report results of five tested models, all of which include industrial organic water pollution intensity. Model 1 also includes per capita GDP, and Model 2 also consists of fertility rate. Model 3 includes the water pollution variable as well as per capita GDP, fertility rate, democratization and secondary education. Model 4 consists of industrial organic water pollution intensity, per capita GDP and both health expenditure measures. Model 5 is slightly reduced to organic water pollution intensity, per capita GDP and the health expenditure measure found to be statistically significant in Model 4. The large differences in sample sizes and mean number of observations per country between Models 1–3 and Models 4–5 are a function of the relatively limited availability of health expenditures panel data for less-developed countries. 132

Downloaded from http://iss.sagepub.com at NORTH CAROLINA STATE UNIV on January 13, 2009

Jorgenson Industrial Pollution and Human Health Table 5 Results from Generalized Least Squares Random Effects Panel Regression of Infant Mortality Rates

Organic water pollutants per 1000 workers (ln) GDP per capita (ln)

Model 1

Model 2

Model 3

Model 4

Model 5

26.991*** [.186] (9.646)

17.158*** [.120] (6.498)

12.854** [.088] (7.129)

14.349** [.100] (7.933)

13.101** [.090] (7.996)

−21.571*** [−.693] (2.956)

−5.436** [−.175] (2.411) 15.561*** [.751] (.954)

−2.546 [−.082] (2.720) 13.983*** [.675] (1.274) −.267* [−.055] (.167) −.220** [−.134] (.108)

−16.027*** [−.515] (4.035)

−17.247*** [−.554] (2.926)

−1.709** [−.120] (1.036) .065 [.002] (3.086) .335 97 3.7

−1.719*** [−.121] (.664)

Fertility rate

Democratization

Secondary education Health expenditures as % GDP Health expenditures per capita (ln) R2 N Mean number of observations per country

.343 253 7.2

.595 218 6.2

.615 195 5.7

.346 99 3.7

Notes: *p <.075, **p <.05, ***p <.01 (one-tailed tests); standardized coefficients are in square brackets; robust standard errors are in parentheses. The models include period-specific intercepts that are not shown.

Prior to discussing the relationship between water pollution and infant mortality, I briefly summarize the effects of the statistical controls. With the exception of fertility rate, per capita health expenditures and per capita GDP in Model 3, the effects of all statistical controls are negative and statistically significant, which corresponds with prior cross-national research on infant mortality and similar health-related outcomes (e.g. Chung and Muntaner, 2006; Frey and Field, 2000; Ram, 2006; Shandra et al., 2004). The positive effect of fertility rate is also quite consistent with related studies (e.g. Brady et al., 2007), and the non-significant effect of per capita GDP in 133

Downloaded from http://iss.sagepub.com at NORTH CAROLINA STATE UNIV on January 13, 2009

International Sociology Vol. 24 No. 1

Model 3 is likely a function of multicollinearity, since this model includes multiple controls that are all at least moderately correlated with one another. However, these findings are not the focus of the analyses. Turning to the variable of interest, the effect of industrial organic water pollution intensity is positive and statistically significant across all tested models.10 While these findings are consistent with policy statements made by various international organizations (e.g. WRI, 2005) as well as research in the health sciences (e.g. McGinn, 2000), they also highlight the importance in studying how social structural factors contribute to water pollution and other forms of environmental degradation, particularly in less-developed countries. This research shows that (1) particular forms of economic integration (foreign investment dependence, export intensity) contribute to the intensity of industrial organic water pollution in less-developed countries, while (2) other forms of political-structural integration (EINGO presence) have the opposite effect, and (3) this particular form of pollution is clearly detrimental to the health of infants in less-developed countries.11

Conclusion Bridging the areas of global political-economy, environmental sociology and the sociology of health, this study contributes to the mounting social scientific literature concerning the environmental impacts of different forms of political-economic integration as well as the potential health consequences of human-caused environmental degradation. More specifically, the PW regression with PCSE and GLS RE panel regression analyses of less-developed countries confirm the hypotheses that industrial organic water pollution intensity is positively associated with foreign investment dependence in manufacturing, and conversely, industrial organic water pollution intensity is negatively associated with the presence of EINGOs. Results also provide moderate confirmation of the hypothesis that industrial organic water pollution intensity is positively associated with export intensity in less-developed countries. Besides illustrating the divergent environmental impacts of various political-economic forms of structural integration, these findings lend support to the ecostructural framing of foreign investment dependence theory, recent advances in world society theory and the theory of ecologically unequal exchange. Further analyses indicate that infant mortality rates are positively associated with industrial organic water pollution intensity in less-developed countries, net of the effects of economic development, health expenditures, fertility rates and other factors. Considering their empirically valid, human health consequences, understanding the social structural causes of water pollution and other forms of pollution is indeed critical. Moreover, with the recent upswings in the globalization of foreign investment and trade as well 134

Downloaded from http://iss.sagepub.com at NORTH CAROLINA STATE UNIV on January 13, 2009

Jorgenson Industrial Pollution and Human Health

as the increasing presence of international non-governmental organizations, particularly in less-developed countries, theorization and research concerning their potential environmental impacts are perhaps more important now than in prior decades. It is my hope that this research will help to encourage other social scientists to utilize their analytical toolbox and methodological abilities to pay closer attention to these sorts of human–environment relationships. Without doubt, methodologically rigorous and theoretically integrated social scientific research on such topics could certainly assist global governance institutions, civil society groups, domestic political institutions and other relevant organizations in the formation, promotion and implementation of more effective policies and practices to help reduce environmental degradation and concomitant human suffering. In this vein, the results of the current study suggest that the often assumed economic benefits of (1) attracting foreign investment, particularly in the secondary sector, and (2) the promotion of export-oriented production both need to be weighed against their potential environmental costs and related human health consequences. Conversely, the findings indicate that transnationally organized citizen groups can make a difference. Considering their increasing presence within less-developed countries as well as the growing intensification of linkages among such groups (Reese et al., forthcoming), the negative statistical association between organic water pollution and EINGO presence supports the optimistic notion embraced by many transnational and global civil society groups that another world is indeed possible. The next steps in this particular research agenda are as follows. First, I plan to conduct similar forms of inquiry for other environmental outcomes, which will allow for comparing the extent to which the impacts of political-economic integration vary for and among different types of degradation. Using newly available panel data and rigorous modeling techniques, I will pay particular attention to environmental outcomes that – like organic water pollution – are known to contribute directly to human health problems. More importantly, drawing from the results of the current study, in future investigations I will also consider the extent to which the environmental impacts of economic integration vary by the level of EINGO presence as well as by other structural factors, including levels of state strength. These sorts of ‘contextual analyses’, which involve the modeling of interaction effects, will allow for assessing whether or not the relative presence of transnationally organized citizen groups and strength of states can mitigate the potential environmental harms posed by exportoriented production and foreign investment dependence. Like the current study, in these future investigations I will conduct structural analyses of the relationships between the investigated forms of environmental degradation and particular human health outcomes, such as infant mortality and child mortality. 135

Downloaded from http://iss.sagepub.com at NORTH CAROLINA STATE UNIV on January 13, 2009

International Sociology Vol. 24 No. 1 Appendix: Countries Included in the Analyses H1

H2

H3

x x

x x

x x

x x x x x x x x x

x x x x x x x x x

x x x x x x x x x

x x x x x x x x

x x x x x x x x

x x x x x x x x

x x x x x

x x x x x

x x x x x

x x x x

x x x x

x x x x

x x

x x

x x

IMR Argentina Bangladesh Benin Brazil Cameroon China Colombia Costa Rica Dominican Republic Ecuador El Salvador Ghana Haiti Honduras India Indonesia Kenya Malaysia Mexico Morocco Nepal Nicaragua Nigeria Pakistan Panama Peru Philippines Rwanda Senegal Sri Lanka Thailand Turkey Uganda Venezuela Zimbabwe

x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x

Note: x denotes if a country is included in any of the models that test the corresponding hypothesis (H1, H2, H3) or any of the tested models in the infant mortality analysis (IMR).

Notes The author thanks the editor and anonymous reviewers for helpful comments on an earlier draft of this article. An earlier version was presented at the 15th Conference of the Society for Human Ecology, Rio de Janeiro, Brazil.

136

Downloaded from http://iss.sagepub.com at NORTH CAROLINA STATE UNIV on January 13, 2009

Jorgenson Industrial Pollution and Human Health 1. Exceptions include two recent empirical investigations conducted by Jorgenson (2006b, 2007). Unlike the current study, these two analyses focus explicitly on the effects of secondary sector foreign direct investment on total levels of industrial organic water pollution and total anthropogenic carbon dioxide emissions (both ‘scale’ outcomes) (Jorgenson, 2007), and growth in industrial water pollution intensity in less-developed countries relative to developed countries (Jorgenson, 2006b). The latter study also uses less rigorous methods that do not correct for heterogeneity bias. Moreover, neither study tests additional hypotheses derived from other areas of macrostructural theorization (e.g. world society theory, ecologically unequal exchange theory), nor do they include rigorous panel analyses of the potential human health impacts of industrial organic water pollution for less-developed countries. 2. The focus here – particularly as it pertains to the first hypothesis – is the potential effects of secondary sector foreign investment on industrial organic water pollution intensity, not the determinants of foreign direct investment. Thus, I greatly limit the discussion of the latter. For examples of thorough yet very different descriptions and empirical analyses of the determinants of foreign investment presence, see Asiedu (2002), Chakrabarti (2001) and Shandra et al. (2003). I thank an anonymous reviewer for suggesting I make this important clarification. 3. The long-standing theory of foreign investment dependence asserts that the accumulated stocks of foreign investment generally make a less-developed country more vulnerable to different transnational and global politicaleconomic conditions, which often leads to a variety of negative consequences for domestic populations, including suppressed economic development (e.g. Kentor, 1998), increased domestic income inequality (e.g. Alderson and Nielsen, 1999), overurbanization (e.g. Bradshaw, 1987) and problems with food security (e.g. Jenkins and Scanlan, 2001). 4. For these estimates of export intensity, less-developed countries are those in which their 2005 GDP per capita was 10,725 (constant US dollars) or less (World Bank, 2007). 5. While I take a relatively conservative approach by employing PW regression with PCSE and GLS RE regression for the reported analyses, elsewhere I test all models with both methods as well as OLS FE panel regression. The substantive findings for all analyses are very similar to those reported in the study and are available from the author upon request. 6. For a brief critique of foreign direct investment data, see Herkenrath and Bornschier (2003). 7. For the current study’s dataset, total population (natural log) is correlated with EINGO presence at .291. 8. Available online at: www.bsos.umd.edu/cidcm/inscr/polity/index.htm 9. Elsewhere I also control for foreign direct investment rates and foreign investment flows as percent GDP. The effects of both on water pollution are nonsignificant, and including them does not suppress the positive effect of foreign investment stocks as percent GDP. 10. Elsewhere I include a variety of additional statistical controls in the analyses of infant mortality rates, including measures of international trade (both

137

Downloaded from http://iss.sagepub.com at NORTH CAROLINA STATE UNIV on January 13, 2009

International Sociology Vol. 24 No. 1 imports and exports measures), foreign investment, domestic income inequality, state strength, foreign debt and foreign aid. The effects of all additional controls are non-significant, and their inclusion does not suppress the positive effect of industrial organic water pollution intensity on infant mortality rates in less-developed countries. 11. I am not suggesting that other forms of environmental degradation do not contribute to infant mortality and other health-related outcomes. Indeed, I consider investigations of other pollutants and human health in crossnational context to be a critical area for future research. However, considering (1) the inclusion of a variety of statistical controls in the reported analyses, (2) the inclusion of additional controls in unreported analyses, as discussed in note 9, and (3) the use of OLS fixed effects and PW regression with fixed effects in additional analyses as described in note 5, I am confident in the validity of the reported statistical relationship between infant mortality rates and industrial organic water pollution intensity within lessdeveloped countries.

References Alderson, A. and Nielsen, F. (1999) ‘Income Inequality, Development, and Dependence: A Reconsideration’, American Sociological Review 64: 606–31. Andres, A. and Irabien, J. A. (1994) ‘The Influence of Binder/Waste Ratio on Leaching Characteristics of Solidified/Stabilized Steel Foundry Dust’, Environmental Technology 15: 343–51. Asiedu, E. (2002) ‘The Determinants of Foreign Direct Investment to Developing Countries: Is Africa Different?’, World Development 30: 107–19. Beck, N. and Katz, J. (1995) ‘What to Do (and Not Do) with Time-Series CrossSection Data’, American Political Science Review 89: 634–47. Boli, J. and Thomas, G. M., eds (1999) Constructing World Culture: International Nongovernmental Organizations since 1875. Stanford, CA: Stanford University Press. Bongaarts, J. and Watkins, S. (1996) ‘Social Interactions and Contemporary Fertility Transitions’, Population and Development Review 22: 639–82. Bornschier, V. and Chase-Dunn, C. (1985) Transnational Corporations and Underdevelopment. New York: Praeger. Bradshaw, Y. (1987) ‘Urbanization and Underdevelopment: A Global Study of Modernization, Urban Bias, and Economic Dependency’, American Sociological Review 52: 224–39. Bradshaw, Y. and Schafer, M. (2001) ‘Urbanization and Development: The Emergence of International Non-Governmental Organizations amid Declining States’, Sociological Perspectives 43: 97–116. Brady, D., Kaya, Y. and Beckfield, J. (2007) ‘Reassessing the Effect of Economic Growth on Well-Being in Less-Developed Countries, 1980–2003’, Studies in Comparative International Development 42: 1–35. Burns, T., Kentor, J. and Jorgenson, A. (2003) ‘Trade Dependence, Pollution, and Infant Mortality in Less Developed Countries’, in W. A. Dunaway (ed.) Crises

138

Downloaded from http://iss.sagepub.com at NORTH CAROLINA STATE UNIV on January 13, 2009

Jorgenson Industrial Pollution and Human Health and Resistance in the 21st Century World-System, pp. 14–28. Westport, CT: Praeger. Cadbury, D. (2000) ‘The Chemical Juggernaut’, People and the Planet 6: 18–21. Chakrabarti, A. (2001) ‘The Determinants of Foreign Direct Investment: Sensitivity Analyses of Cross-Country Regressions’, Kyklos 54: 89–114. Chase-Dunn, C. (1975) ‘The Effects of International Economic Dependence on Development and Inequality: A Cross-National Study’, American Sociological Review 40: 720–38. Chase-Dunn, C. and Jorgenson, A. (2007) ‘Trajectories of Trade and Investment Globalization’, in I. Rossi (ed.) Frontiers of Globalization Research: Theoretical and Methodological Approaches, pp. 165–85. New York: Springer. Chase-Dunn, C., Kawano, Y. and Brewer, B. (2000) ‘Trade Globalization since 1795: Waves of Integration in the World-System’, American Sociological Review 65: 77–95. Chung, H. and Muntaner, C. (2006) ‘Political and Welfare State Determinants of Infant and Child Health Indicators: An Analysis of Wealthy Countries’, Social Science and Medicine 63: 829–42. Czub, G. and McLachlan, M. S. (2004) ‘Bioaccumulation Potential of Persistent Organic Chemicals in Humans’, Environmental Science and Technology 38: 2406–12. Eckenfelder, W. (2000) Industrial Water Pollution Control. Boston, MA: McGrawHill. Ehrhardt-Martinez, K., Crenshaw, E. and Jenkins, J. C. (2002) ‘Deforestation and the Environmental Kuznets Curve: A Cross-National Investigation of Intervening Mechanisms’, Social Science Quarterly 83: 226–43. Ehrlich, P. and Holdren, J. (1971) ‘Impacts of Population Growth’, Science 171: 1212–17. Frank, D. (1999) ‘The Social Bases of Environmental Treaty Ratification’, Sociological Inquiry 69: 523–50. Frank, D., Hironaka, A. and Schofer, E. (2000) ‘The Nation-State and the Natural Environment over the Twentieth Century’, American Sociological Review 65: 96–116. Frees, E. (2004) Longitudinal and Panel Data: Analysis and Applications in the Social Sciences. New York: Cambridge University Press. Frey, R. S. (2003) ‘The Transfer of Core-Based Hazardous Production Processes to the Export Processing Zones of the Periphery; The Maquiladora Centers of Northern Mexico’, Journal of World-Systems Research 9: 317–56. Frey, R. S. and Field, C. (2000) ‘The Determinants of Infant Mortality in Less Developed Countries: A Cross-National Test of Five Theories’, Social Indicators Research 52: 215–34. Grant, D., Jones, A. and Bergesen, A. (2002) ‘Organizational Size and Pollution: The Case of the US Chemical Industry’, American Sociological Review 67: 389–407. Greene, W. (2000) Econometric Analysis, 4th edn. Upper Saddle, NJ: Prentice Hall. Grimes, P. and Kentor, J. (2003) ‘Exporting the Greenhouse: Foreign Capital Penetration and CO2 Emissions 1980–1996’, Journal of World-Systems Research 9: 261–75.

139

Downloaded from http://iss.sagepub.com at NORTH CAROLINA STATE UNIV on January 13, 2009

International Sociology Vol. 24 No. 1 Halaby, C. (2004) ‘Panel Models in Sociological Research: Theory into Practice’, Annual Review of Sociology 30: 507–44. Herkenrath, M. and Bornschier, V. (2003) ‘Transnational Corporations in World Development: Still the Same Harmful Effects in an Increasingly Globalized World Economy?’, Journal of World-Systems Research 9: 105–39. Heuveline, P. (2001) ‘Demographic Pressure, Economic Development, and Social Engineering: An Assessment of Fertility Declines in the Second Half of the Twentieth Century’, Population Research Policy Review 20: 265–96. Hornborg, A., McNeill, J. R. and Martinez-Alier, J., eds (2007) Rethinking Environmental History: World-System History and Global Environmental Change. Lanham, MD: Alta Mira Press. Hsiao, C. (2003) Analysis of Panel Data, 2nd edn. New York: Cambridge University Press. Jenkins, J. C. and Scanlan, S. (2001) ‘Food Security in Less-Developed Countries, 1970–1990’, American Sociological Review 66: 714–44. Jorgenson, A. K. (2003) ‘Consumption and Environmental Degradation: A CrossNational Analysis of the Ecological Footprint’, Social Problems 50: 374–94. Jorgenson, A. K. (2006a) ‘Global Warming and the Neglected Greenhouse Gas: A Cross-National Study of the Social Causes of Methane Emissions Intensity, 1995’, Social Forces 84: 1777–96. Jorgenson, A. K. (2006b) ‘The Transnational Organization of Production and Environmental Degradation: A Cross-National Study of the Effects of Foreign Capital Penetration on Organic Water Pollution Intensity, 1980–1995’, Social Science Quarterly 87: 711–30. Jorgenson, A. K. (2006c) ‘Unequal Ecological Exchange and Environmental Degradation: A Theoretical Proposition and Cross-National Study of Deforestation, 1990–2000’, Rural Sociology 71: 685–712. Jorgenson, A. K. (2007) ‘Does Foreign Investment Harm the Air We Breathe and the Water We Drink? A Cross-National Study of Carbon Dioxide Emissions and Organic Water Pollution in Less-Developed Countries, 1975–2000’, Organization and Environment 20: 137–56. Jorgenson, A. K. (2009) ‘The Sociology of Unequal Exchange in Ecological Context: A Panel Study of Lower Income Countries, 1975–2000’, Sociological Forum 24(1). Jorgenson, A. K. and Burns, T. (2007) ‘The Political-Economic Causes of Change in the Ecological Footprints of Nations, 1991–2001: A Quantitative Investigation’, Social Science Research 36: 834–53. Jorgenson, A. K. and Kick, E., eds (2006) Globalization and the Environment. Leiden: Brill. Jorgenson, A. K., Dick, C. and Mahutga, M. (2007) ‘Foreign Investment Dependence and the Environment: An Ecostructural Approach’, Social Problems 54: 371–94. Kentor, J. (1998) ‘The Long Term Effects of Foreign Investment Dependence on Economic Growth 1940–1990’, American Journal of Sociology 103: 1024–46. Kentor, J. (2001) ‘The Long Term Effects of Globalization on Income Inequality, Population Growth, and Economic Development’, Social Problems 48: 435–55. Kentor, J. and Grimes, P. (2006) ‘Foreign Investment Dependence and the Environment: A Global Perspective’, in A. K. Jorgenson and E. Kick (eds) Globalization and the Environment, pp. 61–78. Leiden: Brill.

140

Downloaded from http://iss.sagepub.com at NORTH CAROLINA STATE UNIV on January 13, 2009

Jorgenson Industrial Pollution and Human Health Kick, E., Burns, T., Davis, B., Murray, D. and Murray, D. (1996) ‘Impacts of Domestic Population Dynamics and Foreign Wood Trade on Deforestation: A World-System Perspective’, Journal of Developing Societies 12: 68–87. Leonard, J. (1988) Pollution and the Struggle for the World Product: Multinational Corporations, the Environment, and International Comparative Advantage. Cambridge, MA: Harvard University Press. Lofdahl, C. (2002) Environmental Impacts of Globalization and Trade. Cambridge, MA: MIT Press. McGinn, A. (2000) ‘POPs Culture’, Worldwatch 13: 26–36. McMichael, P. (2004) Development and Social Change: A Global Perspective, 3rd edn. Thousand Oaks, CA: Pine Forge Press. Manahan, S. (2005) Environmental Chemistry. Boca Raton, FL: CRC Press. Meyer, J., Boli, J., Thomas, G. and Ramirez, F. (1997) ‘World Society and the Nation-State’, American Journal of Sociology 103: 144–81. Miller, G. T. (2000) Living in the Environment: Principles, Connections, and Solutions. Pacific Grove, CA: Brooks and Cole. Mol, A. (2001) Globalization and Environmental Reform: The Ecological Modernization of the Global Economy. Cambridge, MA: MIT Press. Moore, S., Teixeira, A. and Shiell, A. (2006) ‘The Health of Nations in a Global Context: Trade, Global Stratification, and Infant Mortality Rates’, Social Science and Medicine 63: 165–78. Newell, P. (2000) ‘Environmental NGOs and Globalization: The Governance of TNCs’, in R. Cohen and S. Rai (eds) Global Social Movements, pp. 117–34. London: Athlone Press. Ndegwa, S. (1996) The Two Faces of Civil Society: NGOs and Politics in Africa. New York: Kumarian Press. OECD (Organization for Economic Cooperation and Development) (1999) Foreign Direct Investment and the Environment. Paris: OECD Publications. OECD (Organization for Economic Cooperation and Development) (2001) International Direct Investment Statistics Yearbook 1980–2000. Paris: OECD Publications. Ram, R. (2006) ‘Further Examination of the Cross-Country Association between Income Inequality and Population Health’, Social Science and Medicine 62: 779–91. Redclift, M. and Sage, C. (1998) ‘Global Environmental Change and Global Inequality: North/South Perspectives’, International Sociology 13: 499–516. Reese, E., Herkenrath, M., Chase-Dunn, C., Giem, R., Guttierrez, E., Kim, L. and Petit, C. (forthcoming) ‘North–South Contradictions and Convergences at the World Social Forum’, in R. Reuveny and W. Thompson (eds) North and South in the World Political Economy. Oxford: Blackwell. Rice, J. (2007) ‘Ecological Unequal Exchange: International Trade and Uneven Utilization of Environmental Space in the World System’, Social Forces 85: 1–24. Roberts, T. and Parks, B. (2007) A Climate of Injustice: Global Inequality, North–South Politics, and Climate Policy. Cambridge, MA: MIT Press. Roberts, T., Grimes, P. and Manale, J. (2003) ‘Social Roots of Global Environmental Change: A World-Systems Analysis of Carbon Dioxide Emissions’, Journal of World-Systems Research 9: 277–315.

141

Downloaded from http://iss.sagepub.com at NORTH CAROLINA STATE UNIV on January 13, 2009

International Sociology Vol. 24 No. 1 Schofer, E. and Hironaka, A. (2005) ‘The Effects of World Society on Environmental Outcomes’, Social Forces 84: 25–47. Shandra, J. (2007a) ‘The World Polity and Deforestation: A Cross-National Analysis’, International Journal of Comparative Sociology 48: 5–28. Shandra, J. (2007b) ‘International Nongovernmental Organizations and Deforestation: Good, Bad, or Irrelevant?’, Social Science Quarterly 88: 665–89. Shandra, J., Ross, R. and London, B. (2003) ‘Global Capitalism and the Flow of Foreign Direct Investment to Non-Core Nations, 1980–1996: A Quantitative, Cross-National Analysis’, International Journal of Comparative Sociology 44: 199–238. Shandra, J., Nobles, J., London, B. and Williamson, J. (2004) ‘Dependency, Democracy, and Infant Mortality: A Quantitative, Cross-National Analysis of Less Developed Countries’, Social Science and Medicine 59: 321–33. Shen, C. E. and Williamson, J. (2001) ‘Accounting for Cross-National Differences in Infant Mortality Decline (1965–1991) among Less Developed Countries: Effects of Women’s Status, Economic Dependency, and State Strength’, Social Indicators Research 53: 257–88. Smith, J. (2004) ‘Exploring Connections between Globalization and Political Mobilization’, Journal of World-Systems Research 10: 255–85. Smith, J. and Wiest, D. (2005) ‘The Uneven Geography of Global Civil Society: National and Global Influences on Transnational Association’, Social Forces 84: 632–52. Stanley, A. (1996) The Environmental Consequences of Pulp and Paper Manufacture. London: Friends of the Earth. Stiglitz, J. (2002) Globalization and its Discontents. New York: W. W. Norton. Torras, M. (2005) ‘Income and Power Inequality as Determinants of Environmental and Health Outcomes: Some Findings’, Social Science Quarterly 86: 1354–76. United Nations (1992, 1994, 1996, 2000, 2003) World Investment Directory, Vols 1–8. New York: United Nations. Wallerstein, I. (2005) ‘After Developmentalism and Globalization, What?’, Social Forces 83: 1263–78. Wiest, D. and Smith, J. (2007) ‘Explaining Participation in Regional Transnational Social Movement Organizations’, International Journal of Comparative Sociology 48: 137–66. Wooldridge, J. (2002) Econometric Analysis of Cross Section and Panel Data. Cambridge, MA: MIT Press. World Bank (2000, 2005, 2006, 2007) World Development Indicators (CD-ROM). Washington, DC: World Bank. WRI (World Resources Institute) (2005) Earth Trends Data CD-ROM: The Wealth of the Poor. Washington, DC: WRI. York, R. (2007) ‘Structural Influences on Energy Production in South and East Asia, 1971–2002’, Sociological Forum 22: 532–54. York, R. and Rosa, E. (2003) ‘Key Challenges to Ecological Modernization Theory’, Organization and Environment 16: 273–88. Young, O. (1997) Global Governance: Drawing Insights from the Environmental Experience. Cambridge, MA: MIT Press.

142

Downloaded from http://iss.sagepub.com at NORTH CAROLINA STATE UNIV on January 13, 2009

Jorgenson Industrial Pollution and Human Health Biographical Note: Andrew Jorgenson is an Assistant Professor of Sociology at North Carolina State University. He is currently investigating the environmental and social impacts of the transnational organization of production, the structure of international trade and international non-governmental organizations. His recent articles appear in Social Forces, Social Problems and other scholarly journals. He is co-editor of Globalization and the Environment (Brill Academic Press, 2006) and co-editor of the Journal of World-Systems Research. Address: Department of Sociology and Anthropology, North Carolina State University, Campus Box 8107, Raleigh, North Carolina 27695-8107, USA. [email: [email protected]]

143

Downloaded from http://iss.sagepub.com at NORTH CAROLINA STATE UNIV on January 13, 2009