The main objective of this research project was to develop a framework to examine the incentives motivating firms to adopt an environmental management system (EMS) and explain the observed diversity in the comprehensiveness of the EMSs adopted by a sample of Standard and Poor's (S&P) 500 firms. This research analyzed and identified the types of firms (based on their observable characteristics) more likely to feel external pressures from consumers, investors, a competitive market, and regulatory pressures from existing and anticipated mandatory regulations. It sought to quantify the extent to which differences in these pressures faced by firms explain the differences in the scope of EMSs adopted by them and the relative importance of various regulatory and market-based pressures. This analysis has implications for the design and targeting of policy initiatives towards firms less likely to be self-motivated to demonstrate strong environmental stewardship in response to public policy and stakeholder concerns. The framework examined the effectiveness of EMSs in achieving a reduction in toxic pollution intensity. In measuring the impact of EMS adoption on environmental performance, we incorporated the possibility that the factors that led to the adoption of EMSs also can have a direct impact on reducing emissions. Lastly, the implications of EMSs for the economic performance of firms also was examined.
We examined the extent to which potential "win-win" opportunities to reduce pollution and increase output exist and the possible costs of realizing these opportunities when pollution reduction is costly. Firms for whom these costs are low are considered to be more environmentally efficient. We then examined whether the extent of EMS adoption by firms is likely to lower the costs of achieving these opportunities. Finally, this research project examined both the direct effect (if any) of EMSs on firm profitability (through a social reputation effect) as well as any indirect effect it may have (by raising environmental efficiency and therefore competitiveness) on profitability.
There is a growing trend towards "business-led" initiatives by corporations to change corporate culture and management practices by incorporating environmental concerns in production decisions. These initiatives include the development of firm-structured EMSs, trade association programs emphasizing codes of environmental management, and international certification programs that set standards for environmental management such as the International Standards Organization. These EMSs represent an institutional change in the management of corporations and an internally motivated effort at environmental self-regulation by firms. The organizational changes that integrate the environmental concerns into production decisions enable firms to identify opportunities for pollution (waste) reduction, link employee compensation to improvements in environmental performance, implement plans to make continuous improvements in production methods and environmental performance, report progress to stakeholders and set aside funds for environmental accidents. It is important to recognize that firms have flexibility in the design of their EMS, and these systems can vary considerably across firms in the set of practices adopted.
EMSs may be motivated by a number of factors. These include a desire to discover “win-win” opportunities for pollution control (that reduce pollution and raise productivity or reduce costs for the firm). A reduction in waste generation could be accompanied by improvements in process efficiency of the firm and an increase in competitive advantage and value for shareholders. Other motivations for EMSs include the desire of a firm to differentiate its products and appeal to “green consumers” and investors to gain market share and lower the costs of raising capital from the market. By fostering decisionmaking based on full information about the processes, technologies, and resources needed to make environmental improvements, EMSs can enable firms to select pollution control strategies that are cost effective, thereby lowering the costs of compliance with existing or anticipated regulations. These benefits to a firm from adopting an EMS as well as the costs of an EMS, which include coordination, employee training, and audits, are likely to vary among firms. The comprehensiveness of EMSs, measured by the number of environmental practices adopted, is therefore likely to differ considerably across firms.
Data. This study uses both primary and secondary data. The data sources include environmental and financial data from the Toxic Release Inventory (TRI) and S&P's Compustat as well as data on firm-specific environmental management practices from the Investors Research Responsibility Center (IRRC). For the first part of the study examining the determinants of the comprehensiveness of the EMS adopted by firms, the dependent variable is the count of management practices adopted by a sample of S&P 500 firms in 1994 and 1995. The management practices included are designating personnel specialized in addressing environmental issues, having an environmental policy, setting corporation-wide internal standards, environmental auditing, setting aside funds to cover risks of future environmental liability costs, and buying insurance to cover liability or remediation costs of environmental incidents. Additionally, they include training and rewarding workers to find opportunities to prevent pollution, applying the philosophy of Total Quality Management (TQM) to making continuous efforts at improving performance across the firm’s activities and improving environmental performance and evaluating the environmental performance of potential suppliers, partners and clients, while making strategic business decisions and publishing environmental reports that benchmark a firm’s commitment and performance. Of the 500 firms included in the IRRC survey, only firms that responded to the survey in 1994 and 1995, reported to the TRI, and that had financial data available, are included in this study. This results in a sample of 176 firms for 1995 and 159 firms for 1994, with 156 firms having observations for both years.
Explanatory variables include various proxies for regulatory pressures faced by firms such as a threat of liabilities (proxied by the number of Superfund sites for which a firm is held liable), inspections, civil penalties, abatement expenditures, as well as several proxies for market pressures faced by firms. The latter include consumer pressure (proxied by the type of good produced, final or intermediate), public pressure (proxied by the toxic release intensity), investor pressure (proxied by the ratio of sales to assets), competitive pressure (proxied by the industry concentration ratio and the multinational status of the firm), and peer pressure (proxied by the number of environmental management practices adopted by other firms in the industry). All time-dependent explanatory variables and environmental performance and financial characteristics are measured with a 5-year lag (for the years 1989 and 1990) to avoid endogenous regressors because the adoption of some practices may have occurred prior to 1994 or 1995.
For the second part of the study, we examined the impact of EMS adoption on environmental performance and we could only use a subset of the sample created for the first part of the study due to missing observations for some of the explanatory variables. The dependent variable for this part of the study is the toxic release intensity of firms measured in 1994 and 1995. The time-dependent explanatory variables include the extent of EMS adoption (measured by the count of practices adopted) and other proxies for regulatory and market-based pressures that could directly influence the toxic release intensity of firms, as well as firm-specific characteristics. These variables now are measured contemporaneously. We used a pooled sample with a total of 313 observations: 164 firms for 1995 and 149 firms for 1994 with 146 firms with observations that were common across both years.
We selected those industries from the S&P 500 list that had at least eight firms for which complete data for 1994 and 1995 were available. This was necessitated by the fact that we were estimating environmental efficiency at the industry 2-digit Standard Industrial Classification (SIC) code level and needed to estimate firm-specific inefficiencies that were specific for each 2-digit SIC code and each year. This resulted in a pooled sample of 186 firms from eight industries, Food and Kindred Products (SIC 20), Paper and Allied Products (SIC 26), Chemicals (SIC 28), Petroleum Refining (SIC 29), Primary Metal Industries (SIC 33), Electric and Other Electrical Equipment (SIC 36), Transportation (SIC 37), and Instruments and Related Products (SIC 38).
Determinants of the Comprehensiveness of the EMS Adopted by Firms. The factors motivating firms to voluntarily adopt some or all of 13 management practices such as TQM, corporate-wide environmental standards, self-auditing, and participating in voluntary programs were analyzed. We focused on analyzing the observed diversity in the comprehensiveness of the EMS adopted by firms rather than the decision to adopt individual practices because there are synergistic relations between these practices and an individual practice by itself may achieve little. Two empirical models, Poisson and Ordered Probit, were developed to analyze the data.
We found that firms facing a stronger threat of liabilities and in industries that are more highly regulated, as reflected in higher pollution abatement cost expenditures per unit sales, were significantly more likely to seek innovative ways to improve their environmental management. The threat of liabilities, proxied by the number of Superfund sites for which a firm has been listed as potentially responsible, had a positive impact on the quality of EMS adopted, but at a diminishing rate. However, we also found that penalties for noncompliance with existing environmental statutes or inspections to ensure compliance with them were not significant motivators of EMS adoption. These results indicate that the incentives for the adoption of an EMS were not to improve compliance with existing regulations, but to reduce the threat of liabilities and the costs of compliance with those regulations. Firms that are selling final goods and are in closer contact with consumers are more likely to have a more comprehensive/higher quality EMS than firms in intermediate good industries. We also found that firms with larger onsite releases per unit sales were more likely to have a comprehensive EMS. Furthermore, firms that have a high capital-output ratio or a low sales-asset ratio, and thus are likely to be more affected by high costs of capital due to potentially negative investor and market reactions to performance, also are more likely to adopt a more comprehensive EMS. This indicates that the potential for adverse reactions from communities, shareholders, and environmental groups generated by public disclosure of TRI motivated firms to adopt proactive strategies to reduce pollution.
In contrast to the impact of onsite releases on EMS adoption, we found that firms with smaller offsite transfers per unit sales were more likely to adopt a comprehensive EMS. Firms with low offsite transfers per unit sales are those that already have decreased waste being sent for end-of-pipe abatement or that generate waste that is either unsuitable for offsite transfer or too costly to transfer offsite. Our results suggest that these firms are more likely to be seeking other innovative strategies for further reducing waste generation per unit output. We found weak evidence that firms in more concentrated industries were more likely to adopt a more comprehensive EMS. We also found that firms with a stronger multinational presence are more likely to adopt a comprehensive/higher quality EMS. As expected, we found that firms that are more innovative as indicated by their R&D expenditures and likely to have the knowledge and capability to find less costly solutions to their environmental problems were more likely to adopt a comprehensive EMS. Additionally, firms with older equipment were more likely to undertake environmental management, possibly because it was less costly for them to replace old equipment with newer less pollution-intensive equipment.
In addition, we also examined the impact of various incentives on the distribution of the count of environmental practices adopted by different types of firms using quantile regression methods. These regressions provide an estimate of the effect of various explanatory variables on each of the quantiles of the distribution of environmental management practices. We found that the effect of being a final goods producer is stronger for firms with a low propensity to adopt a more comprehensive EMS (conditional on their other characteristics). Among firms that have less comprehensive EMSs (conditional on their other characteristics), being a final good producer has a much larger impact on adoption than among firms that have a more comprehensive EMS (conditional on their other characteristics).
The Impact of EMS Adoption on Environmental Performance. We estimated the impact of EMS adoption and other factors on aggregate toxic releases per unit sales using instrumental variable methods that control for the endogeneity of the EMS adoption decision. We also examined whether EMS adoption has a differential impact on different types of pollutants or methods of disposal. We disaggregated total toxic releases into those emitted onsite, those transferred offsite, and hazardous air pollutants (HAPs). The set of instruments consisted of variables that are correlated with the comprehensiveness of the EMS adoption decision and had been used in the first part of the analysis described above.
The extent of EMS adoption has a statistically significant negative impact on total releases per unit sales. Our analysis also shows that the lagged level of this variable has a very strong and positive influence on current levels of toxic releases per unit sales. Firms with a lower sales-asset ratio have a higher ratio of toxic releases per unit sales. None of the other market pressure variables are statistically significant. The number of Superfund Sites has a positive and statistically significant (at the 10-% level) effect on toxic releases per unit sales, indicating that firms listed as potentially responsible parties (PRPs) for a larger cumulative number of Superfund sites were more pollution intensive. Although innovativeness of the firm has a negative effect on its releases per unit sales, the effect is not statistically significant. The effect of the EMS on firms with low-pollution intensity in the past is negative, but small and insignificant, while on firms with high-pollution intensity is large and significantly negative. With regards to the other variables, we found that market and regulatory pressures do not directly influence environmental performance. Rather, they play an indirect role in improving environmental performance by inducing the adoption of more comprehensive EMSs.
For firms that were more pollution intensive, we found that the adoption of a more comprehensive EMS has a significantly negative impact on their onsite releases per unit sales and on offsite transfers per unit sales, but an insignificant impact on their HAP per unit sales. Furthermore, the emissions reduction effect of an EMS is stronger for offsite releases than for onsite releases. This could indicate that firms were focusing their proactive efforts more towards reducing the use of pollution disposal methods that were costly and more highly regulated. Adoption of a more comprehensive EMS has no effect on any of the different types of releases per unit sales for firms that were less pollution intensive. For all firms, however, higher past-pollution intensity contributes to higher current-pollution intensity. Finally, we found that innovative firms, irrespective of their pollution intensity in the past, were making statistically significant reductions in their HAP/SALES ratio, while the reductions in all onsite releases are less pronounced.
Implications of Environmental Management for Economic Performance. We first determined the extent to which there exists productive inefficiency and environmental inefficiency among the sample firms that produce both priced or desirable outputs and undesirable outputs or pollutants. To capture a firm's performance over multiple dimensions (production of good and bad outputs) with a single measure, we used Data Envelopment Analysis. We measured productive efficiency using a hyperbolic measure of distance of each firm from the best-practice industry frontier; this distance measures the potential for simultaneous increase in outputs and contraction in inputs and undesirable outputs. Environmental efficiency is measured by the ratio of the hyperbolic measures of efficiency under the assumption of free disposability of pollution and under the assumption that pollution disposal is costly. High environmental efficiency indicates a relatively low opportunity cost (in terms of loss of desirable outputs) of reducing pollution. We found that the magnitude of the opportunities for win-win reductions in pollution and increase in output among our sample firms in the S&P 500 are small. We found that 72 percent of the sample firms are fully efficient, and for these firms, implying that there was at least one input or pollutant that could not be reduced any further without also reducing output. The percentage increase in industry output in 1995 that would result if all firms were to become fully efficient would range between 0.94 percent to 6 percent, while onsite releases could be reduced by 0.2 percent to 4 percent and offsite transfers could be reduced by 0.02 percent to 3 percent. Thus, the extent to which win-win gains in efficiency alone could reduce pollution is low. We also found that 80 percent of the sample had an environmental efficiency greater than 0.9, implying that the cost of increasing efficiency for these firms to its maximum level and that raising output and lowering pollution (when pollution reduction is costly) is negligible or small.
Our analysis shows that the comprehensiveness of the EMS adopted has a significantly positive impact on the level of environmental efficiency. However, benefits of EMS in raising environmental efficiency decrease as the pollution intensity of the firm increases. Additionally, firms facing a higher threat of liabilities for Superfund sites and undertaking greater R&D expenditures as well as firms facing greater potential adverse reaction to high toxics release intensity are more likely to be environmentally efficient. Capacity utilization and capital productivity also affect the environmental efficiency of firms. Being a final good producer also had a positive and statistically significant impact on the level of environmental efficiency of a firm. Age of assets was, however, not found to have a significant impact on efficiency, and R&D intensity was found to have a significant, but negative, impact on environmental efficiency.
Next, we examined the impact of EMS on a firm's profitability. We found that environmental performance variables such as the number of Superfund sites and lagged toxics release intensity have a significantly negative impact on profitability. Additionally, as expected, we found that firms with younger assets and a higher R&D intensity are more profitable. Furthermore, we found that, with the exception of two industries, Food and Kindred Products (SIC 20) and Primary Metals (SIC 33), the comprehensiveness of the EMS and the level of environmental efficiency have a significantly positive impact on profitability. A firm's profitability increases, but at a diminishing rate, as the comprehensiveness of the EMS increases. A firm's profitability increases, but at a diminishing rate, as the degree of EMS increases. Thus, there appears to be a net benefit to firms of adopting EMS and of internalizing their environmental externalities. The two industries for which EMS and environmental efficiency do not influence profitability represent two extremes in terms of their environmental performance in our sample. SIC 20 is on average a small polluter and SIC 33 is on average a very large polluter. This suggests that the market values adoption of an EMS by firms that are neither very small polluters or very large polluters and where EMS is likely to have an impact on environmental performance and costs of abatement. For very small polluters, investors may not care about pollution levels and efforts to improve environmental management. For very large polluters, EMS may not be considered "good enough" to improve environmental performance and other more fundamental changes in products and processes may be needed to convince investors that future environmental performance of such firms is likely to improve.
These results suggest that, in addition to environmental performance indicators such as toxic release intensity and the number of Superfund sites, proactive efforts by firms to improve their environmental performance (by adopting EMSs) as well as more efficient use of resources are rewarded by the market, at least in some industries. EMS adoption has both a direct impact on firm profitability, possibly by improving a firm’s reputation and reducing its environmental riskiness as well as an indirect effect by increasing a firm’s environmental efficiency.
These results show that EMS adoption can have social benefits by reducing the toxic release intensity of firms and by potentially raising environmental efficiency or lowering potential costs of pollution reduction among firms. EMS adoption also benefits firms by raising their expected profitability. However, the magnitude of these social and private benefits vary across firms. This suggests that policies seeking to promote EMS adoption among firms should not seek universal adoption; rather, they should focus on providing targeted incentives to encourage adoption particularly by firms that have large toxic release intensity. These policies should include a credible threat of stringent mandatory regulations as a backstop if self-regulation does not improve environmental performance. Additionally, the finding that firms are penalized by the market for having high toxics release intensity and for being at higher risk for environmental liabilities brings forth the importance of information disclosure programs that provide environmental information about firms to the public and thereby allow the market to provide incentives to firms to internalize their externalities.
Publications and Presentations:
|Khanna M, Anton WRQ. Corporate environmental management: regulatory and market-based pressures. Land Economics 2002;78(4):539-558. |
|Khanna M. Economic analysis of non-mandatory approaches to environmental regulation. Journal of Economic Surveys 2002;15(3):291-324. |
|Khanna M, Anton WRQ. What is driving corporate environmentalism: opportunity or threat? Corporate Environmental Strategy 2002b;9(4):409-417. |
public policy, pollution prevention, toxic pollution, S&P 500 firms. , Economic, Social, & Behavioral Science Research Program, RFA, Scientific Discipline, Ecology and Ecosystems, Economics & Decision Making, Economics and Business, Social Science, decision-making, alternative compensation, benefits assessment, business-led environmental management, cost benefit, cost/benefit analysis, decision analysis, economic incentives, economic research, ecosystem valuation, environmental policy, environmental values, policy analysis, pollution prevention, psychological attitudes, public policy, public values, source reduction, standards of value, toxic release inventory