Abstract
This research examines the connection between political donations, lobbying, levels of production, and state enforcement efforts in the coal industry. The authors draw on treadmill of production theory as developed by Schnaiberg to situate the analysis within green criminology. Specifically, based on treadmill of production theory the authors hypothesize that coal companies are more likely to increase political donations and lobbying efforts prior to the conclusion of any state enforcement effort (administrative, civil, or criminal violation). In addition, and consistent with treadmill of production theory, the authors hypothesize that the probability of environmental enforcement will be lower for coal companies that are more embedded in the treadmill of production and higher for companies less embedded in the treadmill. To test these hypotheses, a case-crossover design is used that allows for comparisons within companies by looking at treadmill-related characteristics at the time of the violation and at randomly chosen points in time before and after that violation. The authors discovered that while lobbying efforts and level of embeddedness in the treadmill were unrelated to state enforcement, political donations significantly increase for companies just prior to the conclusion of an enforcement event (odds ratio = 6.36). It is also discovered that corporate restructuring is related to environmental enforcement. The article concludes by offering insights into alternative analysis and uses of treadmill of production theory as it relates to green criminology.
Keywords
Introduction
This article expands the scope of green criminology by drawing on treadmill of production (ToP) theory to examine the issue of environmental enforcement among coal companies in the United States. Recently, there have been calls by green criminologists to better integrate ToP theory into the study of environmental crime (Greife & Stretesky, in press; Lynch & Stretesky, in press). Green criminology has become a rapidly expanding area of study within criminology, promoting various green criminological frameworks. For instance, White’s (2008, p. 50) approach to green criminology has emphasized “environmental justice, with a special focus on human rights and social equity; and ecological justice, with a special focus on the biosphere generally and the rights of non-human as well as human” (see also, Gibbs, Gore, McGarrell, & Rivers, 2010). Green criminology has also advanced theories on the nature and causes of crime in relation to political economic theory and class relations (Lynch, 1990). Lynch (1990) argued that capitalism contributes to various types of environmental crime and deviance that are unevenly distributed according to race and class (see also Stretesky, 2008). Although Lynch’s theoretical approach to green criminology is unique, it is yet to be developed to its full potential. One way to advance the notion of a green political economy approach to environmental crime and deviance is to draw on the established ToP theory (Schnaiberg, 1980; see also, Gould, Pellow, & Schnaiberg, 2004, 2008). ToP theory is similar to green criminology in that both propose that the political economic conditions of class conflict help explain the emergence and distribution of environmental problems. While green criminology has lacked the breadth of development found in ToP theory, it has not been extensively used to examine the enforcement of environmental laws (for exceptions, see Lynch, Stretesky, & Burns, 2004a, 2004b; Stretesky & Lynch, 2011).
We begin this analysis by briefly examining ToP theory as it relates to coal production. Next, we describe how ToP theory can be used to examine environmental enforcement in the coal industry. Our research examines the coal industry because it is a major driving force behind manufacturing production. For example, today the United States produces nearly 1,100 million short tons of coal annually (U.S. Energy Information Administration [EIA], 2011a), and ranks second in the world to China in terms of coal production with a 16% world share. During the years 2000-2010, coal was the largest primary energy production source in the United States and in 2010 alone, coal was responsible for generating 22 of the 75 quadrillion British thermal units of energy manufactured in the United States (U.S. EIA, 2011b). Beyond manufacturing, coal production is related to class interests as demonstrated by its association with a long history of class conflict and violence in the United States (Corbin, 1981). The coal industry also has a strong association with political institutions in the United States. For instance, the Center for Responsive Politics (2011) identified the coal industry as the largest donor of campaign contributions to politicians within the mining industry during the 2011-to-2012 campaign cycle. Moreover, the New York Times recently reported that one politician who led opposition against a bill to enhance Environmental Protection Agency (EPA) regulations that might have harmed the coal industry, profited by receiving “more than $187,000 in mining related donations, including coal, more than any other federal candidate in this election cycle” (Quinones, 2011).
The coal industry has changed drastically over the years with respect to its productive and extractive methods. With those changes came intensified environmental destruction that facilitates coal extraction along with economic and policy revisions (e.g., changes to the Clean Water Act, see Lynch, Burns, & Stretesky, 2010) needed to sustain high coal production demands driven by the desire for low-cost energy, which serves the interest of the coal industry in minimizing the costs of coal extraction. Today, strip mining is a common form of coal extraction that requires less labor than traditional underground mining, and contributes to environmental destruction through the use of chemical explosives, valley fills, and high pollution consequences (Stretesky & Lynch, 2011). Strip mining technology has allowed for the increase of coal production while at the same time reducing the coal extraction work force. According to the U.S. Mine Safety and Health Administration (2011), the number of miners decreased from 219,661 in 1978 to 110,052 in 2008. This decrease has occurred along with the decline of the unionized work force in the coal industry (Mayer, 2004). Given the nature of coal production and its relationship to class conflict, politics, and energy production in the United States, we argue that the study of coal enforcement is ripe for analysis from the perspective of a political economy of environmental crime (i.e., green criminology) situated in ToP theory. To be sure we are not the first criminologists to study this issue of mining regulation. For example, Shover, Clelland, and Lynxwiler (1986) study the creation of surface mining regulation and oversight as an outcome of conflict. We are, however, the first to look at how individual corporations react to enforcement of that regulation that may limit production.
The Treadmill of Coal Production
ToP theory originates in Schnaiberg’s (1980) work, The Environment: From Surplus to Scarcity. Here, Schnaiberg reviews how changes in production accelerate environmental degradation, or the erosion of the natural environment and the destruction of ecosystems. In the case of coal production, environmental degradation occurs when mining operations contaminate ground water through the chemicals used in mining practices and when strip-mining operations such as mountaintop removal mining alter the landscape which can lead to loss of species.
ToP offers an important and alternative theoretical perspective to ecological modernization theory (Mol, 1995) as well as to the efficiency approaches taken by industry (Zhong, Li, Li, & Ge, 2011). Ecological modernization theory optimistically suggests that technological advances and an environmental ethic are fostered by civil society and implemented in social regulation (Mol & Spaargaren, 2000). Unfortunately, the promise that technology and environmental regulations will reduce the environmental degradation associated with coal has not yet been realized (Bell & York, 2010; for an empirical example see, Fleishman, Alexander, Bretschneider, and Popp, 2009). Although it might be argued that “clean coal” (i.e., the technology that can make coal emissions less harmful) is reducing environmental degradation, it should also be pointed out that the United States burns more coal today than in the past and therefore technological gains that promote environmental protection have been offset—at least to some extent—by increased levels in coal extraction. In short, coal production has expanded rapidly in the United States and is at an all time high (Bell & York, 2010). Methods of coal extraction have become increasingly more destructive for the environment as the industry continues to increase the amount of coal mined through mountaintop removal and other forms of strip mining (U.S. EIA, 2011a). Mountaintop removal mining has negative impacts on water quality and is associated with damage to aquatic life when mined sites are compared with unmined sites (Pond, Passmore, Borsuk, Reynolds, & Rose, 2008). Moreover, raw coal must still be processed, which produces significant amounts of coal sludge that is stored in impoundments and composed of various minerals and chemicals that are used to clean coal (Bell & York, 2010). Sometimes these impoundments leak and can cause significant environmental damage (Chatterjee, 2009; Wigginton, Mitchell, Evansc, & McSpirit, 2008). Thus, the production of clean coal has done little to reduce environmental damage or lower coal-associated carbon emissions (Ball, 2009). In short, coal mining is environmentally destructive.
ToP suggests that changes in manufacturing production enhance the forms of ecological degradation that have been occurring since the mid-1940s (Schnaiberg, 1980). Thus, increasing levels of pollution and natural resource extraction are related to Western capital that was invested in chemical- and energy-intensive technology. Chemical-intensive technology such as strip mining increases production while reducing labor costs (Bell & York, 2010). Coal strip mining is more chemically intensive as it increases production through chemical explosives and processing. Thus, fewer miners are needed to produce coal today than in the past. As Gould et al. (2008) point out, the effect of chemical technology is to simultaneously displace workers and set the stage for competition between firms. Thus, production-increases become the standard against which coal extraction is evaluated. Capital investment in chemical technology is therefore the critical link between environmentally destructive production methods and increasing natural resource depletion. In short, firms must maximize profit, reduce their work force, and expand production or risk being viewed as an unattractive capital investment in the financial markets. To make a profit and prevail over competitors, companies constantly find technologies that increase production and reduce human labor costs (see also Rifkin, 1995).
ToP theory also addresses state constraints on coal companies accomplished through the implementation of more stringent environmental laws. In response to these constraints, coal executives argue that regulations will force them to lay off workers (and/or reduce worker benefits) and will reduce their contributions to the tax base (Schnaiberg, 1980). For instance, one coal company executive recently stated,
As always, [our company] is committed to complying fully with EPA regulations. We have spent the last two months identifying all possible options to meet the requirements of this new rule, and we are launching a significant investment program to reduce emissions across our facilities. However, meeting this unrealistic deadline also forces us to take steps that will idle facilities and result in the loss of jobs. (Luminant, 2011)
ToP argues that the threat to labor posed by environmental regulations often pits labor against environmental groups who push for environmental regulation. This contradiction is clearly reflected in a campaign by coal miners unions and the industry, where, for example, on Interstate 64 in West Virginia a billboard erected by the organization Friends of Coal states, “Don’t let EPA bureaucrats take away our coal jobs” (Malloy & Roddy, 2010). In short, companies argue that slowing the rate of production through regulation will harm workers and the state (Montrie, 2003). It is for precisely this reason that organized labor and the state often support increased production at the expense of the environment (Gould et al., 2008).
This constant push to produce more is what leads to natural resource depletion (also called environmental disorganization) and drives this “treadmill of production.” In terms of the coal industry, these changes are not so transparent. Gould et al. (2008) point out that the visibility of the relationship between production and environmental disorganization may lessen over time as corporations continue to suggest that production is “environmentally friendly.” This process is similar to the idea of “greenwashing” (Greer & Bruno, 1996) or the representation that ecological destruction is being reduced when in fact it is increasing (see also, Brisman, 2008). Moreover, Gould et al. (2008) have argued that although green technology may sometimes decrease pollution per unit produced, it does not necessarily decrease environmental pollution since the number of units produced increases.
Another important aspect of ToP is the observation that cost-based considerations force production to shift or relocate, for example, from the “West” or “North” to the “South” (or from more developed to less developed nations). For instance, some less developed countries in the world have seen a significant increase in the extraction and use coal power in order to provide products to the West (Stretesky & Lynch, 2009). Thus, while the United States leads the world in consumption, its citizens do not immediately suffer all the environmental impacts associated with worldwide coal extraction processes that facilitate expanded consumption.
Important to the area of green criminology are ToP observations that environmental regulations and laws also give the appearance of halting or severely punishing those responsible for causing ecological disruption (Burns, Lynch, & Stretesky, 2008; Lynch et al., 2004a, 2004b). These laws are supposed to be neutral and independent rules that protect the environment and public health. This observation by treadmill theorists has significant implications for green criminology. For instance, green criminologists are concerned with both the construction and enforcement of environmental laws and the way that enforcement restricts or facilitates environmental destruction (Katz, 2010; Mares, 2010; Walters, 2010). In the case of coal extraction, laws that allow federal applications for permit exceptions related to, for example, the negative impacts of activities such as mountaintop mining, can be used to circumvent state enforcement actions (e.g., related to Clean Water Action, Section 404, see Hough & Robertson, 2009; Davidson, 2009). 1 Yet, at the same time, some states—coal-producing states—argue that the federal regulations are too stringent and negatively affect them economically.
In sum, ToP provides a mechanism for understanding how the production and regulation of coal that has become part of the larger ToP facilitates environmental destruction. These same issues are of interest to green criminology, though little empirical research has been undertaken to date on these issues within green criminology.
As noted, the assumptions of ToP theory sit well with the perspective taken on environmental destruction within green criminology generally (Beirne & South, 2007). In particular, the focus on political economy emphasizes the original formulation of green criminology as conceptualized by Lynch (1990). Likewise, this view can also be used to direct attention to issue of environmental justice (Stretesky & Lynch, 1999). Some of the effects of coal mining, for instance, are localized, such as the use of chemicals to clean coal, which have destructive effects on local water sources and soils, as well as the environmental degradation associated with strip mining (Stretesky & Lynch, 2011). These impacts are experienced by local residents who tend to be poor, especially in the Appalachian region where surface and mountaintop removal mining techniques are widely used. Research has demonstrated, for instance, that residents living near mountaintop removal sites suffer from increased incidents of cancer (Hendryx, Wolfe, Lou, & Webb, 2011), birth defects (Ahern et al., 2011), cardiovascular disease (Esch & Hendryx, 2011; Hendryx, 2009), respiratory and kidney diseases (Hendryx, 2009), and overall general poor health quality of life (Zullig & Hendrxy, 2011). Thus, the effects of the political economic organization of coal mining not only affect local environmental quality (Bernhardt & Palmer, 2011), but the health of residents.
Coal Crime Hypotheses
Schnaiberg (1980) has pointed out that the state (and by extension its enforcement agencies) is the best mechanism for reducing the damage caused by production. At the same time, ToP suggests that the state’s duty and interests are shaped by its intimate connection to capital and a desire for increased tax revenue (see also, Boyce, 2002; Chambliss, 2001; O’Connor, 1973). These observations lead us to draw directly on ToP theory in order to examine three hypotheses. Each hypothesis examines how variations in the corporate characteristics of coal companies located in the treadmill come into play when they are targets of enforcement actions by the state and when they are not targets of enforcement action by the state.
According to ToP, treadmill institutions directly oppose constraints on production. For instance, companies in the coal industry advertise their pursuit of “clean coal” technology while mobilizing significant resources against enforcement that may constrain their production and expansion. Gould et al. (2008) specifically suggest that treadmill corporations will use their economic power to shape the political (and thus, criminal justice) landscape. We extend their arguments to suggest that this includes attempts to reject criminal labels. This can be accomplished by shaping the political and enforcement landscape.
One way to shape the enforcement landscape is to effectively counteract legislation and enforcement practices that may limit production (see also Shover et al., 1986). Gould, Schnaiberg, & Weinberg (1996) suggest that resistance to legislation is often carried out through lobbying and donation practices that ironically reflect alliances between corporations and the state which have repercussions for environmental protection. The focus on the power of corporations to influence environmental protection through direct and indirect donations and political lobbying lead us to the following two hypotheses:
Hypothesis 1: The threat of enforcement against coal companies is associated with an increase in political donations by those companies.
Hypothesis 2: The threat of environmental enforcement against coal companies is associated with an increase in lobbying efforts by those companies.
Thus, we see environmental enforcement as an organizational outcome that is related to power struggles between civil society, state agencies, producers, and labor (Gould et al. 1996). Our hypotheses reflect the application of enforcement that is likely to be countered by coal companies. Although such hypotheses are clearly consistent with green criminology and ToP theory, they are yet to be empirically examined in the case of criminal violations in the coal industry.
The political economy of production and consumption may also influence the level of enforcement brought to bear on those organizations highly embedded in the ToP. The economic power derived from providing energy generates resources that coal companies can use to influence legislation. Moreover, because energy production drives the economy, energy production itself may be translated into political power in general, influencing the status of energy producers and industries. That is, the state, other treadmill institutions, and labor are well aware that constraining energy production may reduce economic development. Thus, in the case of coal companies, the state may be more likely to enforce violations against a company during periods of declining production, when they are less embedded in (important to) the treadmill of coal production. In short, the regulation of coal companies that produce the most coal relative to total coal production would have a significant impact on other producers since strict enforcement could increase the costs of coal production. Strict regulation could produce harmful economic outcomes. These observations lead to the following hypothesis:
Hypothesis 3: Of the coal corporations that have received an environmental violation, they are more likely to receive a violation of environmental laws during periods when they produce less coal relative to overall coal production.
Together, these three hypotheses allow us to examine how changes in coal company characteristics are related to state enforcement in the coal industry.
Data and Method
To test our hypotheses regarding the associations of political donations, lobbying activity, and coal production and environmental violations in the coal industry we used a bidirectional case-crossover design (Janes, Sheppard, & Thomas, 2005). This allows us to examine changes within companies in relation to environmental crime and is useful for studying the prediction of a discrete event over a predetermined amount of time. The event year (when the environmental violation occurred) was compared with a year before the event and a year after the event. In our case, we took a company with an environmental violation, and compared it during the year with a violation to a randomly selected year before the violation and to a year randomly selected after the violation. These two randomly selected years served as controls to compare with the differences between the violation year and two other points in time. The analysis uses a fixed-effects model that compared within-company variations before, during, and after the violation occurred. Between company differences are not modeled.
Our sample contains all coal companies that received an administrative, civil, or criminal violation from the EPA during the period 1991-2010, and donated money to federal campaigns during the study period. Although our study period covered the years 1990-2011, to use the bidirectional case-crossover method, we needed at least 1 year before and after the violation, therefore the violation years needed to be truncated by 1 year on each end of the study period to enable control selection.
It should be noted that the total number of coal-producing companies in the United States during the 20-year study period was relatively small, and the number of companies that received environmental violations was smaller. When we added the requirement that companies in the sample also needed to have contributed to federal campaigns, the total number of unique coal companies = 19. 2 Therefore, our sample size for the case-crossover study is n = 57, which consists of 19 violation years and 38 control years (19 before years and 19 after years). We recognized that this is a small sample; however, it represents the total number of all coal companies that committed violations and contributed to campaigns during the time period examined.
Dependent Variable
The dependent variable was a dichotomous indicator of whether it was a case year or a control year. Therefore, if it was a case year, an environmental violation occurred (coded “1”) and if it was a control year, no environmental violation occurred (coded “0”). The data on administrative and civil violations were gathered from the U.S. EPA (2010) and the criminal violations were found through a search of all criminal cases using Lexis-Nexus. 3
Political Donations, Lobbying, and Production
To determine if coal companies attempted to mitigate the role of the state from slowing down the ToP in the coal industry, we created four variables. The first two variables involved direct federal campaign contributions to candidates made by each company, its employees, and political action committees. These data were gathered for the years 1990-2011 from the website of the Center for Responsive Politics, which provides data for all federal campaign contributions (more than $200) by year. Two variables were therefore created from the donations data.
The first donations variable was the average political donations for the 3 years before the crime was reported or before the control year. It was computed by adding up the donations for the 3 years prior to the violation (or control) and then divided by 3 to get the 3-year average. That value was then divided by 100,000 so the odds ratios could be interpreted in $100,000s. We label this variable “Donations, 1-3 year lag.” The second donations variable was the average of federal political donations for the 4 to 6 years before the violation (or control) and was calculated in the same manner as the 3-year lag donations variable. We label this variable “Donations, 4-6 year lag.” As noted, we hypothesize that the 1-3 year and 4-6 year donations average variables are positively related to environmental violations. During the time leading up to a violation, we expected that a company would increase its amount of donations to offset the negative effects of the violations. We examined donations prior to the violation because that is the point in time where a company is likely to try to use its economic power to influence enforcement and/or protect its reputation. Thus, our assumption is that companies know about potential environmental violations and try to influence policy makers before their violations are adjudicated.
We examine the Donations, 1-3 year lag and the Donations, 4-6 year lag to see if there is a weaker relationship between donations and environmental violations for the Donations, 4-6 year lag variable than the Donations, 1-3 year lag variable. We expect that this might be the case because the Donations 4-6 year variable is less proximate to environmental violations than the Donations, 1-3 year lag.
Although it is difficult to accurately measure the relationship between corporations and government officials, we used a dichotomous variable (coded as yes = 1 and no = 0) to indicate whether or not the coal companies employed lobbyists and lobbying firms or had a political action committee during the case and control years. We obtained this information from the 1990-2011 editions of the publication Washington Representatives (Columbia Books, 1990-2010). As noted, we hypothesized that the association between lobbying and environmental violations should be positive, as we expected that companies with violations attempt to insulate themselves from future violations. Employing lobbyists can also help gain access to influential government insiders who may be perceived as influencing the violations decision-making processes.
Our final independent variable was used to measure the embeddedness of a coal company within the ToP. We hypothesized that during years that coal companies produced less coal, the more likely the company would be to receive an environmental violation. The logic of the ToP suggests that more embedded actors are less likely to face enforcement for their production-related behavior. This lack of enforcement may aid in maintaining the treadmill function. Therefore, companies that account for large percentages of the annual total coal output of the United States would be less likely to be penalized, according to treadmill logic. We tested this hypothesis with a coal production variable. That variable represents the yearly annual coal output of a company as a ratio of overall U.S. coal output for that year. The production data were gathered from U.S. EIA (2011a).
Control Variables
Previous research (Boies & Prechel, 2002) suggested that companies that go through periods of uncertainty often resort to restructuring, which allows companies to insulate themselves from negative effects of uncertainty. Receiving an environmental violation is one form of uncertainty and corporations who have incurred violations may restructure to distance themselves from the event. Corporate restructuring involves buying, selling, or reorganizing subsidiaries. This activity was measured using a dichotomous variable (coded as restructure = 1 and no restructure = 0) and labeled “Corporate Restructure” that indicated whether the company (or a major part of it) was sold within a few years of the violation (or the nonviolation control years).
It is also possible that commission of an environmental violation was a function of the size of the company. We created a principle component factor of two indicators of company size, annual company sales and number of employees in the company (labeled “Company Size”). These data were collected from Dun and Bradstreet’s (2010) “Million Dollar Databases” for the years 1990-2011. Previous research on corporate crime (Grant, Jones, & Bergesen, 2002) suggests that larger companies commit crime more often, so we would expect this relationship to be positive. 4
Analysis and Results
As noted, we address three hypotheses in this analysis. First, we seek to determine if campaign donations increase significantly prior to violations. Second, we examine whether coal companies are more likely to employ lobbyists during years when they violate environmental laws than in years when they do not violate environmental laws. Third, we seek to determine whether the amount of coal production is negatively associated with receiving an environmental violation. Tables 1 to 4 report the results of multivariate models used to examine these three hypotheses. We used fixed-effects logistic regression to estimate the odds ratios comparing violation years with control years.
Bidirectional Case-Crossover Design: Effects of Corporate Political Donations (1-3 and 4-6 Year Lags) on Reported Crime in the Coal Production Industry, 1990-2011
p < .10. **p < .05 (one-tailed).
Bidirectional Case-Crossover Design: Effects of Corporate Lobbying on Reported Crime in the Coal Production Industry, 1990-2011
p < .10. **p < .05 (one-tailed).
Bidirectional Case-Crossover Design: Effects of Company Coal Production as a Percentage of Total U.S. Coal Production on Reported Crime in the Coal Production Industry, 1990-2011
p < .10. **p < .05 (one-tailed).
Bidirectional Case-Crossover Design: Effects of Corporate Political Donations (1-3 Year Lag) and Corporate Restructuring on Reported Crime in the Coal Production Industry, 1990-2011
p < .10. **p < .05 (one-tailed).
Earlier we acknowledged that our sample was small for various reasons. However, our study examined the entire population of environmental violations in the coal industry committed by companies who donated money to federal campaigns during the study period; therefore, nothing can be done to increase our sample size. We realized that the confidence intervals were wide due to the small number of cases and produce large standard errors. These outcomes are related to the difficulty in studying a rare event such as state-corporate crime, where the data can also be difficult to gather. The statistical power of our analysis was low because of the small number of cases. We would note, however in this case that statistically significant findings would provide strong support for our hypotheses, as it is more difficult to find a sufficiently strong relationship in a small sample than in a large sample.
Table 1 contains models of the effects of political donations on environmental violations by coal companies, controlling for corporate restructuring and company size. In all three models, the average donations over the 3 years prior to the violation were significant (p < .05), whereas the donations for the 4-6 year period prior to the donations entered into the second and third equations were not statistically significant. In the case of the 1-3 year lag the odds of a violation increased by a factor of 6.25 (Table 1) with each $100,000 donation suggesting that donations increase prior to an environmental violation. These findings support our first hypothesis that environmental enforcement against coal companies is associated with an increase in political donations by those companies prior to the violation. Coal companies significantly increased their donations during the years immediately preceding the adjudication of the environmental violation. Corporate restructuring was also significantly and positively associated (odds ratio = 3.18; p < .10) with receiving an environmental violation, whereas company size did not have an effect.
Table 2 contains equations for the prediction of environmental violations in the coal industry using lobbying activity, controlling for restructuring, and company size. An increased use of lobbying was not related to receiving an environmental violation. In fact, the pseudo R2 values were very close to 0, indicating a very poor fitting model. Corporate restructuring remained significant in these models, whereas company size did not. Table 3 examines the third hypothesis that company production was negatively associated with receiving an environmental violation. The equations do not support this hypothesis, rather it appears that company production and receiving an environmental violation are not associated at all, as the odds ratios are 0.99 in each model. Similar to the previous tables, corporate restructuring was significant, whereas company size was not.
Finally, Table 4 provides models of environmental violations by coal companies using the significant predictors from Tables 1 and 2. When company size was not controlled, 1-3-year lagged donations were significant (p < .10) and corporate restructuring was significant (p < .05). Interestingly, when we controlled for company size, the association between donations and environmental violations became stronger (p < .05), whereas the association between restructuring and violations weakened (p < .10) but remained statistically significant. Again, odds ratios for political donations in Table 4 were between 5.24 and 6.36 (p < .10).
Discussion and Conclusion
This article examined enforcement against coal companies as a function of three hypotheses that are consistent with the ToP theory. Specifically, we examine the association between coal company political campaign contributions, corporate lobbying, and relative contribution to coal production, and environmental enforcement within companies. As the ToP theory observes and as we hypothesized (i.e., Hypothesis 1) there is a positive and significant association between political donations and enforcement. This finding is consistent with prior research on campaign contributions in the case of contracts (Hogan, Long, Lynch, & Stretesky, 2006) and with research on mining enforcement regulation in general (Shover et al., 1986). In addition, it can be plausibly argued from existing research that companies that have been treated unfavorably may try to alter their political treatment by increasing political donations (Long, Hogan, Stretesky, & Lynch, 2007). Again, such findings are consistent with ToP theory and corporate strategies used to shape regulation and enforcement (Shover et al., 1986).
It is important to note that the relationship between campaign contributions and favorable treatment is a widely accepted assumption with significant empirical support when examining donations across companies (see, e.g., Hogan et al., 2006; Hogan, Long, & Stretesky, 2010; Long et al., 2007). One of the issues raised in the current research was whether a relationship could be extended to explain the pattern of political campaign contributions within companies at different points in time. This is a much more conservative examination of donations, but one that fits well with the ToP theory. We found support for this relationship: Companies appear to expand donations when enforcement actions are being initiated and prior to enforcement resolution. In exploring the relationship between political campaign contributions and environmental enforcement records against coal companies, we have added to the empirical studies of corporate crime. Importantly, we argued that corporations would use campaign contributions to influence environmental social control and corporate perceptions as suggested by ToP. The connection we suggested between the use of political campaign contributions and environmental social control appear important and support an extension of ToP theory into green criminology. Buttel (2004) noted that the ToP has been underutilized in the 21st century; we argue that one way to rectify this is to encourage the development of treadmill within green criminology.
This research also explored two other hypotheses that are not often examined in the corporate crime literature by looking at lobbying and embeddedness in the coal production treadmill. For example, our results do not indicate support for the hypothesized relationship between environmental enforcement and lobbying (Hypothesis 2). That is lobbying does not appear to change in response to or matter as much as changes in political donations when companies suspect that their legal violations will result in an enforcement action. Other research that examines the influence of donations and lobbying across companies seems to support this finding (e.g., Hogan et al., 2006). We are unsure why this finding results. We note that perhaps variations within companies are not likely to produce such relationships since corporations have relatively stable numbers of lobbyists over their life course, or perhaps because lobbying efforts only change significantly in relation to corporate restructuring or in efforts to affect legislation rather than enforcement. Likewise, it is also possible that lobbying is carried out more effectively through trade associations representing an industry rather than by individual corporations. Thus, the use of lobbyists does not vary within companies to the same extent as the use of political donations. This relative lack of within-company variation with respect to the use of lobbyists may have produced an unusually conservative test of lobbying effects.
ToP theory has also been used to describe normalized expectations associated with the nature of productive and consumptive economic system structures under capitalism in the United States. In that view, exploitation of the environment is seen as a normal dimension of the political economy and that of the production and consumption cycles. That is to say the profit-oriented goals of capitalism are tied to the need to extend the exploitation of nature in order to produce commodities for consumption to increase profit. Based on the general propositions contained in ToP theory, we argued that since the exploitation of nature is a normal aspect of American capitalism, a weak social control network used to protect the environment from exploitation would be produced. We expected to see this relationship realized at the company level (i.e., within-company behavior studied over time) when examining the relationship between relative production and environmental enforcement (i.e., Hypothesis 3). This was not the case. We hypothesized that ToP theory could be expanded to the analysis of coal production within companies and suggest a number of reasons why this relationship was not evident in our data. Each of the possibilities noted below is open to empirical investigation and reflects an interpretation of the coal sector’s location in the U.S. ToP.
First, based on ToP theory, it is entirely plausible that the factors that influence the level of coal production by a particular company are diffused throughout the structural network of the ToP and, therefore, do not appear with significant force in any particular manifestation of treadmill power relations at the company level. That is to say, that treadmill factors are so ubiquitous that they cannot be measured at the company level. Indeed, in the United States coal companies rank among the most powerful corporate actors. The coal industry in its entirety sponsors as many political action committees as the banking industry, the field of accounting, alcoholic beverage producers, and general manufacturing. Thus, individual fluctuations in production are much less important to enforcement practices than the total level of production in the coal sector. Such an explanation is consistent with ToP theory, but is yet to be empirically verified.
Second, with respect to the coal industry, enforcement may be more significantly influenced by coal-related rule making as influenced by political donations. For example, in September, 2011, the House of Representatives passed the Transparency in Regulatory Analysis of Impacts on the Nation (known as the TRAIN Bill and H.R. 2401). TRAIN limited the EPA’s ability to enforce existing regulations by requiring additional analyses of the effects of soot and mercury emissions (two important concerns of coal-fired power plants). 5 Data reported by the Federal Election Commission (http://www.fec.gov) shows that coal and electric companies donated more than $475,000 to the Energy and Commerce Committee during 2011. One representative, Fred Upton, received more than $96,000 (http://www.fec.gov). In short, coal campaign contributions may be the mechanism through which production matters. Such an observation that contributions mediate the relationship between company production and enforcement is only partially supported by our analysis because production itself, while related to donations, is not related to enforcement. However, the severe restrictions placed on our analysis by the conservative nature of the research design may have placed too severe limits on the analysis.
Further research could also use an alternative approach for testing Hypothesis 3. Our study was a within-company analysis of corporations who had both donated money to political campaigns and received an environmental violation. A different approach for testing how the embeddedness (as defined by overall coal production) of a coal company affects whether they receive an environmental violation, would be to examine all coal companies, and model whether or not the proportion of overall annual U.S. coal production by any company was related to receiving a violation. This approach, while outside the purview of this article, would provide further insight into whether embeddedness in the ToP insulates companies from the law.
Whereas the ToP focuses on the relationships between the state, industry, and labor, our empirical analyses did not focus on labor and the relationship between violations and workers’ rights. However, previous research has examined the relationship between environmental crime and corporate culture in the form of worker’s rights. For instance, Stretesky and Lynch (2011) looked at the relationship between environmental violations and a corporate culture factor that included a measure of mine safety and health violations. They discovered no relationship between environmental violations and the corporate culture factor. This finding, however, is yet to be replicated in other studies and the notion that poor and dangerous working conditions might be related to production practices using ToP theory needs to be better developed in the literature. In short, it is reasonable to think that increases in production might be associated with worker health and safety under certain conditions.
It is also important to note that the relationship between enforcement and production is complex, which complicates empirical ToP studies such as this one. For example, in these data coal companies did not lose their permits as a result of enforcement actions taken against the companies. Thus, in simplistic terms, potential enforcement by the state did not threaten to directly stop treadmill practices among this group of coal producers. However, there are a variety of indirect mechanisms by which enforcement would reduce production or the accumulation of company profits (i.e., see Foster, 2005, who notes that the ToP is also a treadmill of accumulation). Instead, we argue that criminal and civil enforcement may affect companies in three important and indirect ways that help explain why donations and crime may be correlated. First, large fines that are leveled against a company can affect the accumulation of wealth that may be used to expand production. In an effort to achieve deterrence, EPA fines are structured to remove all economic benefit that is gained from noncompliance and illegal activity (Garlow & Ryan, 1994). Thus, successful enforcement slows a company’s rate of accumulation.
Second, operation costs are associated with regulation and firms and politicians often argue that reducing these regulation costs will lead to more economic growth (Harrington, 1988). For instance, Harrington (1988) has noted that inspections and oversight create costs for firms that are separate from the costs associated with penalties for violating environmental laws. As state regulatory budgets become strained, regulators often target specific firms, based on previous behavior, for regulatory oversight (Helland, 1998). Thus, when a company violates the law and is identified as a “bad” actor additional costly environmental oversight is likely to be the result. This added oversight can, by its very nature, reduce corporate accumulation and potentially affect production by identifying violations that allow for increases in production, but which may have gone undetected. Helland (1998) observes that firms have an incentive to try to minimize targeting costs (i.e., those additional regulatory costs associated with increased inspections that result from being labeled as a noncomplying actor) and would like to move out of the bad firm category into the good firm category. One way to avoid targeting, then, may be to demonstrate willingness to self-police so as to reduce the need for additional oversight (Helland, 1998). Another method, as we argue here, may be to increase political donations. Political donations might be considered a form of government greenwashing that may serve to alter the corporate image among politicians. Donations may even help to indirectly influence the political regulatory landscape by lowering oversight costs through the implementation of policies that demonstrate good firm behavior (Schnaiberg, 1980). Third, enforcement activities may also reduce production if fixing the violation means that companies need to slow production to comply with environmental regulation.
As noted, the production of coal has increased considerably over the past 20 years. However, in the case of violating companies that we study, coal production is essentially stable at the source of the violation and does not trend with industry production. 6 This suggests that enforcement could place a mild constraint on these producers even while overall industry production increases remain largely unaltered. Thus, enforcement clearly does not directly halt the ToP as the coal industry continues to produce more and do so with more environmentally destructive practices and fewer people. The fact that production is flat for the companies in this study is at least consistent with evidence that enforcement is likely to have placed at least some constraints, even if indirect, on production. Moreover, as noted in the economic literature, the treadmill of accumulation is also likely to be affected as targeting costs increase, unless, of course, those costs can be minimized through various corporate strategies such as self-policing and political donations.
The ToP is quite complex, making empirical testing of parts of the treadmill difficult. For example, our data examine federal enforcement practices. Thus, the results in this study are not generalizable to state enforcement actions or to citizen suits that are often filed when state and regulatory agencies fail to act. It should be noted that in all major coal-producing states, the states have primacy of enforcement of the Clean Water Act, 7 leading many enforcement actions to occur at the state level, which we have not analyzed in this study. An important question this raises is, when does the EPA pursue environmental violations at the federal level? Mintz (1995) points out that a number of factors may determine whether a case is taken forward for enforcement. These factors include the strength of the case, the offender’s prior criminal history, the seriousness of the case, resources of the state agency to handle the cases, and Department of Justice resources at the time of the violation. Together these factors suggest that more serious violations and those that are politically relevant are likely to be pursued by the EPA for enforcement. It is unclear how our bias in federal enforcement might affect our results. If political advocacy on the part of coal industry can keep a case out of the hands of federal prosecutors then our findings may be conservative. In short, we may have excluded important political outcomes for those offenders who are not punished because they were able to use their political power to alter enforcement or even perceptions about the harm that their actions may have caused. In these cases state enforcement may be more likely. Whether a similar relationship between donations and enforcement would emerge at the state level is in need of additional study. However, it is likely that firms donate money to state politicians in order to influence enforcement. In the case of citizen suits companies may donate to politicians prior to their violations. Thus, while we might expect similar patters to emerge between all types of violations it is yet to be empirically examined and warrants further examination.
We believe that this analysis of the interaction between the state and the coal industry demonstrates that ToP is operating efficiently in the coal industry. Despite the limitations inherent in our analysis, we have provided some insight into ToP theory as it relates to crime and deviance in the coal industry. We hope that such an analysis encourages future research into the area of ToP and green criminology.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.