Industrial Environmental Performance Metrics: Challenges and Opportunities (1999)

To help the reader understand the current use of environmental metrics within the four industries examined in this study, Chapter 2 describes a general metrics taxonomy, briefly reviews the characteristics that current literature and the committee's experience suggest are desirable in the formulation of useful metrics, and reviews the types of corporate decisions that are affected by environmental performance information. The chapter provides a general discussion only; more specific information and examples can be found in Appendixes A, B, and C.

Metrics developed for internal corporate use generally provide information on operations or management issues, but many also provide information useful to external stakeholders. Energy use, expressed either in absolute terms or on a per-unit-of-product basis, is an example of a measure of interest to both corporate managers and external stakeholders. The latter group (e.g., customers, regulators, investors, environmental groups) generally is interested in many internal corporate metrics and is concerned about the impacts of industry activities on the environment at the local, regional, and global levels. These concerns frequently relate to such issues as air quality, water quality, product recyclability, and regulatory compliance.

One widely used scheme of classifying metrics has been developed as part of the International Organization for Standardization (ISO) 14031 process (International Organization for Standardization, Annexes Testing Committee, 1996). This system groups metrics according to their utility in three areas:

• making operational decisions,
• making management decisions, and
• assessing the condition of the (external) environment.

Operational metrics generally measure potential environmental burden in terms of inputs and outputs of materials and energy (Box 2-1). Operational metrics can sometimes be thought of as describing a rough "mass balance" for an industrial activity. Ditz and Ranganathan (1997) developed a classification system that separates operational metrics into four subcategories.

• Materials Use—Quantities and types of materials used (useful for tracking resource inputs and distinguishing their composition and source).
• Energy Consumption—Quantities and types of energy used or generated (provides the analog to materials use; made even more useful when fuel types are differentiated).
• Nonproduct Output—Quantities and types of waste created before recycling, treatment, or disposal (most useful in distinguishing production efficiency from end-of-pipe control solutions).
• Pollutant Releases—Quantities and types of pollutants released to air, water, and land (can be differentiated according to whether hazardous or nonhazardous or of global or local concern).

In many cases the comparability of each of these measures can be greatly enhanced by normalizing them with respect to such parameters as number of products produced or sales revenue. Efforts have also been made to normalize according to a product's value to society. Such efforts have generally relied on sales price as a proxy for social value.

Environmental releases are, of course, an incomplete indicator of actual environmental impact. The impact of any contaminant depends on its biochemical or physical properties as well as on a number of local or global conditions. For example, a given amount of an organic contaminant can have quite different effects on water quality depending on the receiving water's volume, status (quiescent or turbulent), composition (fresh or saline), and initial condition (e.g., pH, concentration of dissolved O2). Therefore, while determining the mass of a release is critical to assessing environmental impact, this almost always provides a less than full accounting of the actual potential for environmental harm.

Operational metrics may also address product design, packaging, and transport. Given the conventional wisdom that 80 percent of a product's attributes are determined during the first 20 percent of the product development process, efforts to measure and improve environmental performance during the design phase can be particularly fruitful (DeLadurantey et al., 1996; Hoffman, 1997). Because

many companies outsource a considerable portion of their activities, some attention is also focusing on the environmental performance of suppliers who provide raw, unprocessed, or unassembled production materials and even complete subassemblies.

Operational metrics have, thus far, received the majority of industry's attention. This is due in part to their close connection to regulatory compliance. The fact that these metrics often address aspects of production that are both tangible and within a company's ability to control also should not be overlooked.

Whereas operational metrics provide an indication of the present state of a company's environmental performance, management metrics furnish information on steps being taken to influence operations. Management metrics describe such things as the allocation of funds and labor, implementation of environmental programs and new environmental policies, environment-related legal expenses, environmental remediation activities, and the status of environmental information systems (Box 2-2). Metrics like these are designed to inform management

and support decision making on the expenditure of time, money, and manpower required to maintain or improve a company's environmental performance. Information on how and where corporate resources are allocated can help to identify problem areas and opportunities for improvement. Management indicators, particularly those related to budgeting (i.e., for capital, operations, and maintenance), may also help companies assess the worth of previous environmental investments. Projects and programs that show a positive return on investment provide incentive for continued efforts to improve environmental performance.

In the final analysis, it is generally environmental condition metrics that are of greatest interest to industry and external stakeholders. Unfortunately, this is also the area in which the fewest number of robust metrics have been developed and implemented. Environmental condition metrics seek to provide information on the health of the environment and how it is changing. Ideally, these metrics would link specific industrial activities or emissions to environmental impacts. Establishing a causal relationship between pollutants and impacts, however, can be a complex undertaking. Nonetheless, there are a growing number of measures that are applied to environmental systems in an attempt to assess their relative health (Box 2-3).

Many of these measures have been derived from regulations or guidelines and are intended to maintain environmental standards that protect the health of humans and indigenous flora and fauna. Numerous measures of regulatory origin have been used to characterize water and air quality and soil and groundwater contamination in local or regional ecosystems. Environmental condition metrics have also been formulated for use at a global scale. Measures of atmospheric ozone and global mean temperature are examples of such indicators. These types of metrics are often proposed initially by the scientific community and later appropriated by governmental and nongovernmental organizations for use in assessing environmental health.

In recent years, increasing attention has been given to a broad range of metrics that seek to estimate the long-term sustainability of human activities. The definition of sustainability, particularly as might be applied within a corporate context, is still evolving, but some companies and researchers have begun to investigate how such principles might be practically implemented. This report explores some of these experimental efforts while also describing current practice in assessing environmental condition.

The committee believes that an essential precondition for developing useful environmental metrics is a well-defined set of environmental goals or objectives.

Once an objective has been established, an effective metric will influence behavior by measuring an organization's progress toward achieving that objective. While environmental goals may vary across an industry and over time, the vital attribute for any metric is that it shows a clear relationship to the desired objective. In order for a metric to be useful, it also must be possible to obtain reliable measurements and data; the information gathered must be relatively unambiguous and readily understandable; and it must be communicated internally to those decision makers capable of affecting change or, alternatively, to interested external stakeholders.

In addition to these general properties, industry, public stakeholders (e.g. local communities, citizen groups), and commercial stakeholders (e.g., lenders, investors, insurers) attach importance to certain specific attributes.

Some characteristics of metrics, especially those tied to operational and management decisions, have particular value within company walls. In 1996 the

Tellus Institute and World Resources Institute (WRI) conducted an informal survey of environment, health, and safety professionals to identify the characteristics industry finds most useful in an environmental metric (White and Zinkl, 1996, 1997). Survey respondents represented 33 mostly large (annual revenues in excess of $1 billion) corporations from several different industrial sectors (e.g., chemicals, electric, gas, sanitary services).

Survey respondents attached particular importance to two characteristics: the ability to verify metrics and the ability to reliably compare measurements over time (Figure 2–1). The importance of verifiability is due in part to traditional concerns over compliance. The ability to defend the validity of data or measurements is paramount in efforts to demonstrate compliance with permits or with state and federal regulations. The importance attached to comparability over time indicates industry's interest in tracking improvements in environmental performance. Without such comparisons it is impossible to determine whether actual progress has been made. The importance of this characteristic may increase in the future if more companies begin to link incentives (i.e., salary, bonuses, etc.) to environmental as well as financial performance. This same line of reasoning may also explain the relatively high value survey respondents attached to the comparability of metrics within a company. Such comparability provides consistency across different business units in an organization. It also improves a company's ability to compare the environmental impacts of disparate processes or products.

Most of the characteristics cited by survey respondents were of primary relevance within companies themselves, but some industry respondents showed substantial interest in metric attributes more useful for external application. For example, the ability to account for the scale of corporate operations through some form of normalization (e.g., per unit produced, per dollar of sales revenue) was thought to be an essential or helpful metric property by 92 percent of those polled.

Eighty-two and 76 percent of respondents, respectively, cited comparability across companies within an industrial sector and comparability across industrial sectors as being either essential or helpful metric attributes. Finally, and perhaps a bit surprisingly, 26 of the 33 individuals polled said that public reportability would be either an essential or a helpful metric attribute. The number of these characteristics and the degree to which each is incorporated into the development of new metrics will presumably vary depending on process and industry.

Finally, it should be noted that companies also desire metrics that can be used in building relationships with external stakeholders. These stakeholders include those with a direct financial interest in the organization or facility (e.g., lenders, insurers, investors) as well as the communities in which companies operate. An open exchange of information on environmental performance can be an important step in building trust with these stakeholders, particularly local

communities. Such public relations efforts can be important to avoiding negative media attention and more intensive regulatory scrutiny.

As considered by the committee, external stakeholders are individuals and groups that generally have little direct financial interest in an industrial operation or sector. Individuals, public interest organizations, and community groups tend to be concerned about environmental performance information that relates to potential threats to human health. However, these stakeholders also show evidence of becoming increasingly aware of long-term issues related to ecosystem health and sustainability (Delphi Group, 1998). Human health, ecosystem health, and sustainability are all related at some level. Current capability to scientifically link each of these to various environmental loads is limited, however.

Although much work remains to be done in the area of human toxicology, some clear connections between human health and exposure to hazardous substances have been established. Methods of assessing ecosystem health are somewhat less developed, but substantial progress has been made over the past 25 years in establishing standards for maintaining a robust local or regional environment. By comparison, public interest in sustainability is relatively new, and, accordingly, there is still considerable debate over how the concept should be defined, measured, and addressed.

Most stakeholders seem to accept the notion that compliance with federal and state environmental regulations protects public health. Nonetheless, hazardous emissions and on-site storage of hazardous materials, even when fully compliant with existing regulations, are of considerable concern to local communities. These concerns become much more acute when regulatory violations occur. Local communities also tend to react quickly and forcefully to nuisance problems such as odors, noise, and dust, whether or not these emissions pose a health risk.

Public concerns over ecosystem and human health are often comingled when water quality, air quality, and soil or groundwater contamination are at issue. Tracking the amount of persistent or bioaccumulative substances in the environment, even if within regulatory limits, will continue to be an area of industrial activity that receives considerable public scrutiny. Although the U.S. Environmental Protection Agency's (EPA's) Toxic Release Inventory (TRI) provides some useful data, the aggregate nature of the measurements and the lack of information regarding relative impacts often leave the public wanting for details. Some companies have tried to overcome these limitations (Imperial Chemical Industries Group, 1996; Wright et al., 1998), but no impact weighting system has yet proved widely acceptable.

The need for systems for ranking or weighting toxic releases is a major issue in the metrics field, as evidenced by the number of researchers and nongovernmental environmental organizations that have undertaken their own efforts to

disaggregate and rank pollutant emissions (usually based on TRI data). The EPA has also expended significant effort to devise a system for ranking pollutants (Hazard Ranking System) and provide guidelines on toxic thresholds (Ecotox Thresholds Software) (United States Environmental Protection Agency, 1999). The Environmental Defense Fund has recently established an online system that uses local maps to graphically link TRI emissions data to thousands of specific industrial sites throughout the United States. The website, called Scorecard, can disaggregate and weight emissions based on such characteristics as carcinogenicity. The website received over 1 million hits in its first 2 weeks, an indication of the public's considerable interest and demand for such information.

In addition to water, air, and soil quality, many public stakeholders express concern over the health of indigenous flora and fauna. In this context, measures of the number and type of various species being supported by an ecosystem have proved useful at both the local and the regional levels, while concern over endangered species often takes on global meaning. The public has also taken some interest in assessments of ''nuisance'' behavior caused by industrial activity occurring near or in an ecosystem (e.g., damaged aesthetic appeal, odors, noise, impaired visibility).

In recent years the public has begun to recognize the long-term environmental impacts of human activities and the need to assess these impacts. A number of investigations (including this one) are seeking to flesh out industry's role in measuring and adjusting its behavior in a manner that will improve sustainability. Measuring the use of renewable as opposed to depletable resources is one way some companies are trying to track more-sustainable activities. Likewise, some firms have begun to disaggregate and track energy sources (e.g., coal, oil, natural gas, solar) by carbon content or global warming potential to aid efforts to shift to less environmentally burdensome resources. Developing meaningful measures of sustainability is a major challenge for industry and society.

Stakeholders in the financial sector (e.g., banks, insurers, investment companies) are beginning to express some interest in corporate environmental performance. Many insurance and lending institutions have started to factor environmental information into their decisions regarding insurability, premiums, and loans. Much of this relatively new-found interest reflects the risks companies faced as a result of the 1980 Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA, "Superfund"). CERCLA makes current and past contributors to environmentally hazardous sites financially liable for cleanup. This liability can be transferred to those who buy a contaminated site, significantly increasing the financial risk in any type of property exchange. As a result, almost all real estate transactions now require an environmental impact assessment. Such assessments are also commonly required before insurance

companies will issue coverage. CERCLA's "strict" liability provision is largely indifferent to whether the offending actions were legal or not at the time they occurred. The law thus provides ample incentives for companies to reduce or eliminate activities (legal or otherwise) that may adversely affect the environment. Unfortunately, this law has had the unintended consequence of encouraging some industrial and commercial developers to seek contaminant-free "greenfields" rather than use existing developed areas.

Beyond the insurance and lending sectors, there are indications that the investment community is now beginning to take some interest in the environmental performance of companies. Investor attention to environmental issues is not widespread, although some research has posited a link between financial and environmental performance (Blumberg et al., 1996; Cohen et al., 1995; Hart and Ahuja, 1996). Still, many within the investment community remain unconvinced.

The United Nations Development Programme recently produced a working paper (Gentry and Fernandez, 1996) containing results from a survey of selected Fortune 500 chief financial officers (CFOs) and securities analysts. Respondents were asked to describe their use of environmental information in assessing the relative strength of companies (Box 2-4). Lack of confidence in quantitative measures that link environmental performance to financial performance was clear (question 1). However, in terms of qualitative evaluation, the environment appears to take on a somewhat larger role (question 2). The value of qualitative factors seems to outweigh that of strictly numerical data in the view of both the CFOs and analysts (question 3). Although it seems that environmental stewardship plays some role in assessing the economic health and investment potential of a company, the penalties for poor environmental performance appear much greater than the rewards for excellence (questions 4 and 5). This may be due to the absence of any agreed-upon measures for determining superior environmental performance. Regulatory compliance, on the other hand, is a firm lower bound by which corporate laggards can be judged.

Survey respondents' concern over compliance issues was clear when they were asked to rank environmental factors related to financial performance and management quality (questions 6 and 7). Corporate efforts to use environmental performance for competitive advantage received mixed reviews (question 8). While CFOs attach some value to ecoefficiency programs, analysts are either unconvinced or uninformed as to the cost-saving potential of such initiatives. It seems that if ecoefficiency activities are ever to receive attention from the financial community, communication of the rationale and potential benefits of such efforts must be improved (question 9). This underscores the importance of communicating environmental information to groups lacking extensive industrial experience. If metrics are to be widely used by external stakeholders, particularly those in the financial community, they will need to be both useful and comprehensible. The creation of transparent measures of performance may help overcome a number of actual and perceived barriers to industry incorporating envi-

ronmental factors into decision making (question 10). Although a small segment of the investment community does make use of environmental performance data, the large majority of investors gives such information little attention.

While it is apparent that a vast array of environmental parameters is being measured, tracked, and reported, little is known about the role this information plays in the day-to-day activities of companies. The value of a metric (or suite of metrics) relates directly to the number and nature of corporate decisions it influences.

Respondents of the joint Tellus Institute-WRI survey (White and Zinkl, 1997) were asked to describe and rank the types of decisions that were most affected by environmental performance metrics (Table 2-1). As might be expected, regulatory compliance topped the list: Over 60 percent cited this as the number one area of business decision making affected by metrics. Another 24 percent ranked compliance decisions second or third in importance. Strategic business planning ranked next in terms of importance, with 15 percent of those

polled placing it first and another 49 percent ranking it either second or third. Environmental metrics would therefore seem to have some impact on corporate investment decisions, but their utility for decision making in an array of other core business areas appears to be either limited or underutilized.

Given this, it is interesting to note how respondents ranked the overall importance of environmental criteria in corporate decision making (Figure 2-2). Sixty-seven percent "almost always" made use of environmental information when weighing decisions. When these responses are considered together with those given to the question of whether "adequate metrics" are currently available (Figure 2-3), it seems that a considerable gap exists between the type of environ-

Figure 2-2

Frequency of use of environmental criteria in corporate decision making.

SOURCE: White and Zinkl (1996).

Figure 2-3

Adequacy of metrics available to evaluate products or processes.

SOURCE: White and Zinkl (1997).

mental information required and that being supplied. It may simply be that metrics are used primarily for compliance because that is the area in which most metrics exist. There seems to be a considerable unmet demand for environmental information; substantial rewards might accrue to those able to collect and make use of it.

Blumberg, J., A. Korsvold, and G. Blum. 1996. Environmental Performance and Shareholder Value. Conches-Geneve, Switzerland: World Business Council for Sustainable Development.

Cohen, M., S. Fenn, and J. Naimon. 1995. Environmental and Financial Performance: Are They Related? Vanderbilt University, Owen School of Business. Nashville, Tenn.: Investor Responsibility Research Center.

DeLadurantey, C.E., R.J. Kainz, and M.H. Prokopyshen. 1996. Environment, health, and safety: A decision model for product development. Paper presented at the SAE International Congress and Exposition, Detroit, February 26–29.

Delphi Group. 1998. A Business Guide: Environmental Performance and Competitive Advantage. Ontario: Queen's Printer for Ontario.

Ditz, D., and J. Ranganathan. 1997. Measuring Up: Toward a Common Framework for Tracking Corporate Environmental Performance. Washington, D.C.: World Resources Institute.

Gentry, B.S., and L.O. Fernandez. 1996. Valuing the Environment: How Fortune 500 CFOs and Analysts Measure Corporate Performance . United Nations Development Programme (UNDP), Office of Development Studies, Working Paper Series. New York: UNDP.

Hart, S.A., and G. Ahuja. 1996. Does it pay to be green? An empirical examination of the relationship between emission reduction and firm performance. Business Strategy and the Environment 5:30–38.

Hoffman, W.F. 1997. A tiered approach to design for environment. Paper presented at the Conference on Clean Electronic Products and Technology, Institute for Electrical and Electronics Engineers, Edinburgh, Scotland.

Imperial Chemical Industries Group (ICIG). 1996. Annual Report on Safety, Health, and Environmental Performance. London: ICIG.

International Organization for Standardization, Annexes Testing Committee. 1996. ISO 14031 Environmental Performance Evaluation. Final ATC Report Edition. U.S. SubTag 4. New York: American National Standards Institute.

United States Environmental Protection Agency. 1999. Ecotox Threshold. Available online at http://www.epa.gov/oerrpage/superfnd/web/oerr/r19/ecotox/index.html. [January 7, 1999]

White, A., and D. Zinkl. 1996. Corporate Environmental Performance Indicators: A Benchmark Survey of Business Decision Makers. Boston: Tellus Institute.

White, A., and D. Zinkl. 1997. Green Metrics: A Status Report on Standardized Corporate Environmental Reporting. Boston: Tellus Institute.

Wright, M., D. Allen, R. Clift, and H. Sas. 1998. Measuring corporate environmental performance: The ICI environmental burden system. Paper presented at the NAE Workshop on Industrial Environmental Performance Metrics, January 28–29, Washington, D.C.