The adoption of environmental management practices in a transition economy.

Henriques, Irene ; Sadorsky, Perry

INTRODUCTION

Environmental management in transition economies is difficult because economic and institutional change creates significant risks and uncertainty for strategic decisions (Peng, 2000) and this is particularly true for the Central and East European (CEE) countries. Economic reforms in the CEE countries were primarily designed to increase the efficiency of state-owned enterprises and make products more competitive at an international level (Filatotchev, 2005). Economic reforms were made more difficult by political change, the break-up of the former Soviet Union and the collapse of the East European trading bloc (Uhlenbruck, Meyer, and Hitt, 2003; Filatotchev, 2005).

While all aspects of the managerial decision-making process are affected when an economy goes through transition, in this paper our interest is to investigate the decision to adopt environmental management practices in a transition economy--namely, Hungary. According to the Organisation for Economic Co-operation and Development (OECD) economic survey of Hungary (OECD, 2005, p. 29), 'though good progress is being made in environmental outcomes, the Hungarian authorities recognise that the overall strategy and co-ordination in policy needs strengthening and are working on a sustainable development plan'. This has created considerable uncertainty for businesses. Businesses must grapple with the uncertainty associated with the effectiveness of alternative facility-level environmental management practices to address the myriad of external environmental pressures (regulatory, community, foreign buyers/owners, etc).

A few studies examine the factors influencing a company's decision to undertake environmental management practices in developed economies (eg, Henriques and Sadorsky, 1996; Nakamura, Takahashi and Vertinsky, 2001; Anton, Deltas and Khanna, 2004). In general, these papers find that environmental management practices in developed countries are influenced by a number of different stakeholders (such as consumers, investors and suppliers), the company's export orientation, its size and environmental regulation.

The factors affecting a facility's decision to undertake environmental management practices in transition or emerging economies, whose environmental regulations are usually not as firmly established as in developed economies, have yet to be fully explored. One interesting exception is the paper by Dasgupta et al. (2000). Dasgupta et al. (2000) examine environmental compliance in Mexico and find that environmental management efforts, as measured by an environmental practice score and environmental training, have a strong impact on environmental compliance. Moreover, they find that plant size, multi-division status, post-secondary education, formal regulation and public trading of the firm's stock each contributes to increased environmental efforts. In a related paper, Bluffstone and Sterner (2006, this symposium) use a sample of firms from Bulgaria, Hungary, Lithuania, Poland, Romania and the Slovak Republic to examine the relationship between various factors, like privatisation, export orientation, public pressure and environmental regulation, on environmental behaviour. They find that, in general, these factors contribute to the adoption of environmental techniques.

In this paper, we use a 2003 OECD survey of manufacturing facilities from Hungary to study the role of a host of environmental stakeholder pressures (regulatory, community, investor, managerial), export orientation, size, foreign ownership and head office influence on a facility's response to adopt specific environmental management practices. Eight environmental management practices are examined. Individual adoption equations are used to assess the factors that lead facilities to adopt a given environmental practice; however, such a methodology does not take into account the fact that the majority of facilities that undertake an environmental management system (EMS) tend not to undertake just a single environmental practice but a set of practices. To allow for this possibility and to account for the comprehensiveness of the EMS, we take the sum of the number of environmental practices adopted (Anton et al., 2004). Together, these adoption equations shed light not only on the factors affecting the decision to adopt a single practice but also the factors that led facilities to adopt a more comprehensive EMS.

The paper is organised as follows. The next section summarises the factors that are likely to affect a facility's decision to adopt environmental management practices. Section 3 introduces the data and the methodology employed. The econometric estimates are presented in the penultimate section, and the last section concludes.

WHY ADOPT AN ENVIRONMENTAL MANAGEMENT PRACTICE?

Environmental management practices are defined as a host of managerial innovations that emphasises an organisation's commitment to improving the natural environment via such practices as, the formulation of a written environmental policy, the collection and reporting of environmental data, environmental awareness training of employees and linking employee compensation to environmental performance.

A written environmental policy is an organisation-wide pledge for responsible environmental management, which is made public and stipulates the organisation's general philosophy for environmental improvement (Darnall and Edwards, 2006). The collection and reporting of environmental data are the practices that translate an organisation's environmental policy (written or not) into action. The setting of environmental performance indicators/goals allows an organisation to monitor and evaluate its environmental performance across time. To realise these environmental goals, an organisation may also choose to train employees how to manage environmental issues. Some organisations also choose to include an environmental incentive to an employee's compensation by linking the employee's compensation to the environmental performance of the organisation (Gabel and Sinclair-Desagne, 1993).

The collection and reporting of environmental data are undertaken via audits, benchmarking and environmental public reports. An audit is the collection and compilation of a facility's environmental emissions data by its employees (an internal audit) or a third party Can external audit). Such audits allow the organisation to periodically identify discrepancies within the business organisation (Netherwood, 1998). External audits are required if a facility chooses to obtain environmental certification (eg, EMAS or ISO 14001 certification). Environmental benchmarking is a structured process by which a firm or facility seeks to identify and replicate best practices to enhance its environmental performance. According to Vorhies and Morgan (2005), benchmarking is one of the most popular management tools in the world and is becoming a primary instrument in firms' total quality management (TQM), knowledge management and process improvement efforts. An environmental public report is a publicly available document prepared by the company specifying its environmental record.

Under situations where managers have incomplete information as to the outcome of their efforts, a great deal of experimentation takes place, suggesting that facility managers need specific kinds of information to respond effectively to the environmental demands placed on them (Dasgupta et al., 2000). The latter suggests that not all facilities may respond equally to the same impetus. In fact, facilities are affected by a host of stakeholders who may have the ability to affect a facility's decision-making process when information is incomplete.

Stakeholder Pressures

A great deal of environmental pressures emerges from an organisation's stakeholders (eg, Freeman, 1984; Mitchell et at., 1997). In general, a company faces a daunting array of potential environmental risks connected with various pressure groups that, if not addressed, may adversely affect a company's bottom line (Henriques and Sadorsky, 1996). Consequently, the more pressure a facility is under to take into account the environmental impact of its actions, the more likely it will adopt environmental management practices. It is, therefore, necessary to assess the potential risks associated with each pressure both from outside the firm and within the firm.

External stakeholders include consumers, regulators, community groups and environmental groups. (1) The potential risks associated with consumers (2) include the boycotting of the company's product that has a direct impact on the company's bottom line. The potential risks associated with regulators are (a) regulatory changes that render current company processes or product impacts unacceptable; (b) non-compliance penalties; (c) the risk that the company's product will be eliminated, substituted or phased out; and (d) the risk that the company's raw inputs may be banned or restricted. Environmental and community groups can also exert significant influence via their ability to influence the legislative system and consumer buying patterns via third-party suits and lobbying. Together, all these factors pose a significant risk to the organisation that, if ignored, can be extremely costly.

Internal stakeholders include shareholders/investors, management and employees. From an environmental context, shareholders/investors become concerned when (a) environmental fines begin lowering profits; (b) environmental goals are not met; and (c) environmental concerns are such that the company is unable to attract new capital or new investors. In fact, according to Whiteley and Czaban (1998), Hungarian firms working to raise substantial funds for new equipment find it difficult to obtain bank support, making technical solutions to environmental problems an even greater challenge. The potential risks that management faces include increased inability to deal with the environmental issues and dismissal if the issues in question affect the facility's bottom line or reputation. Employees can also pose risks when the lack of training or awareness leads them to perceive that top management is not committed to their well-being. In this case, better employees may seek to quit due to the reputational cost of working for a bad polluter or employees may decide to strike over concerns of internal pollution.

In general, the greater the perceived pressures exerted by these stakeholders to address environmental issues, the more likely the facility will adopt environmental management practices.

Export Orientation

Another important factor influencing a firm's decision-making process is its export orientation. The more export oriented the facility, the higher the benefits it may accrue from the more visible actions taken to protect the environment. According to Nakamura et al. (2001), this may occur because foreign customers tend to be less able to monitor the performance of the facility or firm--as a result, more visible signs of environmental commitment such as having an EMS or a certified EMS (eg, ISO 14001 or EMAS) may legitimatise their reason for doing business with the facility.

In the case of economies in transition, increased trade with western countries can contribute to the adoption of environmental management practices and increased environmental performance for firms (Andonova, 2003). It is often the case that quality standards are higher in developed economies and high-quality standards can only be met with up-to-date technology. Alternatively, export orientation may have little or no impact on a facility's decision to adopt environmental management practices if the goods being produced for export are not subject to high-quality standards. In her study of CEE firms, Andonova (2003) finds that export-oriented firms adopt clean technologies faster than non-export-oriented firms, but that export orientation has little impact on a firms decision to conduct environmental audits or take significant steps towards ISO 14001 certification. Hence, we are uncertain as to the relationship between export orientation and environmental management practices.

Financial Position of the Facility

Hungarian firms working to raise substantial funds for new equipment find it difficult to obtain bank support (Whiteley and Czaban, 1998). The latter suggests that liquidity constraints may be serious impediments to firms that wish to undertake the investments needed to improve environmental performance (Earnhart and Lizal, 2006). In this case, successful financial facility-level performance may provide the facility with the internal funds necessary for it to invest in the equipment, personnel and EMSs needed to meet their environmental goals. Earnhart and Lizal (2006), in their examination of Czech firms from 1993 to 1998, find evidence that successful financial performance improves future environmental performance. The latter suggests that greater financial performance provides firms facing liquidity constraints the resources necessary to undertake their environmental initiatives. We predict that the greater the facility's financial performance, the greater its ability to implement environmental management practices.

Head Office Perspective on Environmental Issues

Environmental issues are long-term in nature and usually require an organisation to take a long-term perspective when assessing the costs and benefits associated with any type of investment. Here a company's head office decision to provide a facility with an environmental R&D budget is a proxy for this long-term perspective. (3) Consequently, we predict that manufacturing facilities that are given environmental R&D budgets are more likely to undertake environmental management practices.

Foreign Ownership

The impact of foreign ownership on a facility's decision to undertake environmental management practices is more complex. According to Nakamura et al. (2001), foreign owners may, on the one hand, be less willing to contribute to the social well-being of the country in which the facility is located and, as a result, less inclined to invest in environmental protection above the level of required regulation. On the other hand, foreign owners may increase environmental protection practices to secure goodwill from the regulatory authorities of the host country so as to prevent discrimination or increase their legitimacy in the eyes of these authorities. The extent to which foreign ownership obliges a facility to undertake environmental management practices is, therefore, an empirical question.

Managerial Experience with Other Management Systems

For those facilities that have already obtained ISO 9001 registration and/or follow TQM system principles, the implementation of an EMS requires less investment in money and time because it is very similar to ISO 9001 and the principles of TQM. ISO 9001, for example, has several elements that are useful for the implementation of an EMS--management structure, review meetings, documentation and record procedures, internal audits and procedure for corrective action. (4) These elements help build organisational capabilities to implement an organisation-wide management system with employee empowerment and cross-functional coordination (Barney, 1991). Consequently, if a facility has a TQM system, it is more likely to adopt environmental management practices relative to those that do not.

Facility Size

The impact of facility size is proxied by the number of employees. Larger facilities tend to possess the skills, both human and capital, that can facilitate their ability to commit to environmental practices. Consequently, the impact of facility size on the implementation of environmental management practices is predicted to be positive.

Regulatory Enforcement: Inspection Frequency

A facilitys own experience with regulatory enforcement and monitoring are critical factors (Dasgupta et al., 2000) in influencing it's decision to undertake environmental management practices. Some empirical studies (eg, Magat and Viscusi, 1990; Andonova, 2003) have found that both regulatory environmental inspections and the threat of inspections induce firms to comply with environmental regulations. We predict that the larger the number of inspections a facility receives (past), the more likely the facility will adopt environmental management practices so as to signal their environmental commitment to the enforcement agency.

METHODS AND DATA

The data for this present paper are taken from a large OECD industrial survey. (5) This survey was undertaken to collect the data necessary to study facility-level environmental management practices in the manufacturing sector. Hungary is the only transition economy included in the database and the focus of this paper.

The initial Hungarian sample comprised 1530 manufacturing firms with production facilities with at least 50 employees (Kerekes et al., 2004). These firms were sent a survey in May 2003. Given that the organisational unit under study was the facility, firms that had many production facilities were asked to answer the questionnaire with reference to the facility at which they were located or with which they were most familiar. During this period, several follow-up telephone calls were also conducted to prompt responses. In total, 466 facilities responded giving us a response rate of 30.5%. In order to verify that companies with EMSs were not the only respondents, the responses to one of the many questions regarding environmental management practices, namely, whether the company had implemented an EMS was monitored. No significant bias in the pattern of responses was observed. In fact, of the 466 respondents, 60.2 % had not implemented an EMS, 27.9 % had implemented an EMS and 11.9% were in the process of implementing an EMS. (6) A complete data set with no missing values for the response variables or any of the explanatory variables was available for 182 facilities. No significant bias in the pattern of responses was observed in the sub-sample (182 respondents).

Our dependent variables consist of eight environmental management practices described in the section Why Adopt An Environmental Management Practice, as well as the total number of environmental practices a facility undertakes (count). The individual environmental management practices are dummy variables depicting whether the facility (1) has a written environmental policy; (2) employs environmental criteria in the evaluation and/or compensation of employees; (3) has environmental training programmes (4) performs external audits; (5) performs internal audits; (6) benchmarks environmental performance; (7) provides a public environmental report; and (8) uses environmental performance indicators/goals. Individual adoption equations, however, do not take into account the fact that the majority of facilities that undertake an environmental system do not undertake just a single environmental practice but a set of practices. To allow for this possibility and to account for the comprehensiveness of the EMS, we take the sum of the number (count) of environmental practices adopted (Anton et al., 2004).

The independent variables used in each model include the scope of a facility's market as a proxy for export orientation, the natural logarithm of the number of full-time employees in the facility as a proxy for facility size, the facility's overall business performance over the past 3 years, whether the facility has an environmental R&D budget, the influence of commercial buyers, the influence of management, the influence of community groups, the influence of public authorities, the influence of shareholders/investors, whether the facility has implemented a quality management system, whether the head office is located in a foreign country as a proxy for foreign ownership, the number of times facility has been inspected by environmental authorities (total over 3 years) and industry dummies to control for industry differences (the chemical industry is the omitted category). For convenience, variable names and a brief description of the data and the construction of the variables are provided in Table 1. Data on market scope were collected using a four-point cardinal scale while data on influence of buyers, influence of management, influence of community, influence of public authorities and influence of investors were originally collected using a three-point Likert scale. To avoid possible bias in the regression coefficients from using data coded on a scale with few points (or data coded on a cardinal scale), each of these variables were re-coded as a 0, 1 dummy variable using the coding information provided in Table 1.

There is considerable variation in the frequency of environmental practices. Slightly more than half of the facilities had a written environmental policy (59.3%), internal audit (57.7%) or performance indicators (52.2%) while only 18.7% of the facilities used environmental criteria in the evaluation of employees (denoted in Table 2 as an employee evaluation). There is also considerable correlation (correlation coefficient greater than 0.5) between the environmental practice measures (Table 2).

The decision to adopt a specific environmental practice is a binary decision. A discrete response model is employed and empirically tested using binary probit estimation. Binary probit models are estimated using maximum-likelihood techniques and quasi-maximum-likelihood standard errors are computed for the coefficient standard errors. Likelihood ratio statistics are used to test joint hypotheses on the estimated coefficients.

Insofar as the total number of practices adopted by a facility is concerned, our measure, count, ranges from a minimum of zero to a maximum of eight. Count data can be modelled using a Poisson distribution, which restricts the mean value equal to the variance, or an alternative distribution, like a negative binomial distribution, which includes separate parameters for the mean and variance (Wooldridge, 2002). A regression-based test of the Poisson restriction (P=0.66) is not rejected at conventional levels. We, therefore, present results from quasi-maximum likelihood estimation of the Poisson distribution.

EMPIRICAL RESULTS

Empirical Results on Individual Environmental Management Practices Table 3 reports regression results on the factors, the construction of which are discussed in the preceding section, which impact environmental management practices. Although no one variable is statistically significant across all eight environmental management practice equations, the signs of those coefficients, which are statistically significant, are mostly as expected.

Facility size has a positive and significant impact on a facility's decision to adopt an external audit and adopt environmental performance-based indicators. The marginal effects for the facility size variable in the external audit and performance-based indicators equations range from 12.6% to 14.7%. These results indicate that increasing the facility size by one unit increases the probability of a facility adopting an external audit by 12.6% while increasing facility size by one unit increases the probability of a facility adopting environmental performance indicators by 14.7%.

Facility business performance has a positive and significant impact on a facility's decision to adopt a written environmental policy. A one-unit increase in facility business performance increases the probability of adopting an environmental written policy by 7.6%.

As predicted a priori, having an R&D budget is an important determinant to adopting environmental practices. The coefficient on this variable has a positive and statistically significant impact on six environmental management practices (external audit, internal audit, performance indicators, public report, employee training and written policy). The marginal effect of having an R&D budget ranges from 34.7% to 51.8%. Having an R&D budget is a particularly important driver in the adoption of an employee training programme. Having an environmental R&D budget increases the probability of a facility adopting an employee training programme by 51.8%. Hence, a long-term perspective on environmental issues is a critical determinant.

As expected, facilities that view the influence of management employees as very important are more likely to adopt environmental practices (benchmarking, external audit, internal audit, performance indicators and written policy). The marginal effect for the influence of management employees ranges from 15.3% to 23.8%.

Consistent with our expectations, facilities that have a TQM system are more likely to adopt environmental management practices for benchmarking, external audits, performance indicators, employee training, and a written environmental policy. The latter supports Nakamura et al's. (2001) suggestion that participation in a TQM reduces the information search and learning costs involved in implementing environmental practices. Having a TQM system is a particularly important driver in the adoption of a written policy. Facilities with a TQM are 53.8% more likely to adopt a written policy.

As expected, facilities that have their head office located in a foreign country are more likely to adopt environmental criteria in the evaluation/ compensation of employees, conduct internal audits, adopt performance indicators, undertake employee training, and have a written environmental policy. Marginal effects range from 5.3% to 26.9%.

The number of times a facility is inspected positively impacts the probability that a facility will adopt environmental management practices for benchmarking and public environmental reporting. One additional inspection over a 3-year period increases the probability of benchmarking (public environmental report) by 2.5% (1.7%).

A facility's market scope has a positive and statistically significant impact on two of the environmental management practices (performance indicators, employee training), but a negative and statistically significant impact on employee evaluation/compensation. Marginal effects for the market scope variable range from -6.0% to 18.9%. Market scope has a statistically insignificant impact on environmental practices for external audit and internal audit, which is consistent with what Andonova (2003) finds.

By contrast, the influence of public authorities has no statistically significant impact on a facility's decision to adopt any environmental management practice. In other words, plants that report greater public authority pressure do not exhibit greater environmental management efforts than their counterparts. Two possible explanations may account for the latter.

The first is that facilities may know that the government's first priority is the economy as opposed to the environment. Kosztolanyi (1999) and Lynch (2000) both suggest that the Hungarian government's need to improve upon its environmental record had more to do with European Union accession rather than a genuine interest in protecting the natural environment. The second is the complicated structure of environmental protection in Hungary that makes it difficult to know how and to whom to respond. Lynch (2000) notes that the national authority for environmental issues in Hungary is divided into several authorities. For example, standards for indoor air quality are set by a different agency than standards for outdoor air quality (O'Toole and Hanf, 1998; Lynch, 2000).

Surprisingly, investors have no statistically significant influence on whether a facility adopts environmental management practices. Such a result is quite plausible if investors believe that Hungarian facilities have already adopted the values, practices and systems of the parent company as Danis and Parkhe (2002) found in their study of 17 Hungarian-Western international cooperative ventures.

The influence of buyers and the influence of community groups each have no statistically significant impact on a facility's decision to adopt environmental management practices. The influence of community groups is often not very strong in transition economies. According to Lynch (2000), public support for the environment in the CEE countries, which may have actually been greater under Communist rule, has been declining as the realities of post-Communist life have set in. In general, the concern for the environment is not a top priority by either communities or public authorities in Hungary, unless it is pushed as a mandate by the European Union.

Empirical Results on the Count of Environmental Management Practices

In practice, the majority (72 %) of facilities adopts two or more environmental management practices (Figure 1). Consequently, it makes sense to empirically model the number (count) of environmental management practices. These results are shown in the last two columns of Table 3. Notice that the model fit, as measured by the R-squared value, is highest for the environmental practices equation that uses count as a dependent variable, suggesting a more robust fit to the data then is provided by any one of the individual practices equations. The statistically significant coefficients are of the expected sign. In particular, facility size, having an environmental R&D budget, viewing the influence of management employees as important, having a TQM system, and being international each have a positive and statistically significant impact on the number of environmental management practices adopted by a facility. The marginal effects for these variables range from 0.46 to 2.34 and measure the contribution to the number of environmental practices adopted from an increase in one of the explanatory variables. Having a TQM system, for example, is particularly important because it increases the number of environmental practices by 2.

Variables for the influence of buyers, influence of community, influence of public authorities and influence of investors have little impact on the number of environmental management practices adopted by a facility. This seems reasonable since none of these variables show strong explanatory power in any of the individual environmental practice equations.

Perhaps, a bit surprising is the fact that inspection frequency, which does show significant explanatory power in the environmental practice equations for benchmarking and public environmental report, has no statistically significant impact on the number of environmental management practices. One plausible explanation for this result is that past practice in Hungary was one in which the existing fine system had very low fine rates (Morris, et al. 1999; Lynch, 2000). Inspections, therefore, lead to enforcement actions consisting of very low fines and/or expected fines that, in turn, result in a low incentive to adopt environmental management practices.

CONCLUSION

While models of environmental management have been proposed and tested using data from developed economies, less work has been done for transition economies. In this paper, we use data from manufacturing facilities in Hungary to study the impact that environmental stakeholder pressures (regulatory, community, investor, managerial), export orientation, size, foreign ownership and head office influence have on a facility's decision to adopt specific environmental management practices. Eight environmental management practices are examined both individually and jointly. The model fit from the joint environmental management practices equation is higher than the model fit from any of the individual environmental management practices equations. The findings in this paper are helpful in gaining a better understanding of the factors that increase the likelihood of adopting environmental management practices in a transition economy.

For most of the models that we examine, having an environmental R&D budget has a positive and significant impact on adopting environmental management practices. This is consistent with our prior expectation and the existing literature. For the individual practices equations, the marginal effect can be as high as 51.8% (in the case of employee training). Facilities that view the influence of management workers as very important are also more likely to adopt a specific environmental management practice. This illustrates the importance of environmental leadership at the management level.

As expected, a quality management system is an important driver in the count equation as well as in five of the individual practices equations. In fact, having a TQM system increases the number of environmental practices by 2, the largest of any of the explanatory variables studied. This supports the proposition that a quality management system tends to reduce the rather steep learning curve associated with the introduction of EMS (Nakamura et al., 2001). Facility association with international firms also increases the likelihood of adopting environmental management practices (by as much as 53.8% in the case of adopting a written policy).

Unlike in the case of developed countries, our empirical results for Hungary find that the influence of buyers, influence of community, influence of public authorities and influence of investors have little impact on the number of environmental management practices adopted by a facility. Whether the influence of these stakeholders on the adoption of environmental practices strengthens as Hungary moves towards becoming a developed economy remains to be seen.

Acknowledgements

We thank Dietrich Earnhart and Lubomir Lizal for their very useful comments and suggestions. The data upon which this study is based are the exclusive property of the OECD. The views contained in this paper are those of the author(s) and may not reflect those of the OECD. The financial support of Environment Canada and the OECD is gratefully acknowledged.

REFERENCES

Andonova, L. 2003: Openness and the environment in Central and Eastern Europe: Can trade and foreign investment stimulate better environmental management in enterprises? Journal of Environment and Development 12 (2): 177-204.

Anton, WR, Deltas, G and Khanna, M. 2004: Incentives for environmental self-regulation and implications for environmental performance. Journal of Environmental Economics and Management 48(1): 632-654.

Barney, JB. 1991: Firm resources and sustained competitive advantage. Journal of Management 17(1): 99-120.

Bluffstone, RA and Sterner, T. 2006: Explaining environmental management in Central and Eastern Europe? Comparative Economic Studies this symposium.

Danis, WM and Parkhe, A. 2002: Hungarian-Western partnerships: A grounded theoretical model of integration processes and outcomes. Journal of International Business Studies 33 (3): 423-455.

Darnall, N and Edwards Jr, D. 2006: Predicting the cost of environmental management system adoption: The role of capabilities, resources and ownership structure. Strategic Management Journal 27(4): 301-320.

Dasgupta, S, Hettige, H and Wheeler, D. 2000: What improves environmental compliance? Evidence from Mexican industry. Journal of Environmental Economics and Management 39(1): 39-66.

Earnhart, D and Lizal, L. 2006: Effects of ownership and financial performance on corporate environmental performance. Comparative Economic Studies (forthcoming).

Filatotchev, I. 2005: Corporate governance, exporting and performance of firms in transition economies. Academy of Management Best Papers Proceedings, Hawaii, 2005.

Freeman, RE. 1984: Strategic Management: A Stakeholder Approach. Pitman/Ballinger (Harper Collins): Boston.

Gabel, HL and Sinclair-Desagne, B. 1993: Managerial incentives and environmental compliance. Journal of Environmental Economics and Management 24(3): 229-240.

Henriques, I and Sadorsky, P. 1996: The determinants of an environmentally responsive firm: An empirical approach. Journal of Environmental Economics and Management 30(3): 381-395.

Kerekes, S, Harangozo, G, Nemeth, P and Nemcsicsne Zsoka, A. 2004: Environmental policy tools and firm-level management practices in Hungary OECD Environment Directorate (accessed on 23 September 2005), www.oecd.org/dataoecd/26/0/31686250.pdf.

Kosztolanyi, G. 1999: Where there's muck there's brass: Hungary and the environment. Central Europe Review 1 (12); September (accessed on 23 September 2005), www.ce-review.org/99/12/ csardas12.html.

Lynch, D. 2000: Closing the deception gap: Accession to the European union and environmental standards in East Central Europe. Journal of Environment and Development 9(4): 426-437.

Magat, W and Viscusi, WK. 1990: Effectiveness of the EPA's regulatory enforcement: The case of industrial effluent standards. Journal of Law and Economics 33 (2): 331-360.

Mitchell, RK, Agle, BR and Wood, DJ. 1997: Toward a theory of stakeholder identification and salience: Defining the principle of who and what really counts. Academy of Management Review 22(4): 853-886.

Morris, GE, Revesz, T, Zalai, E and Fucsko, J. 1999: Integrating environmental taxes on local air pollutants with fiscal reform in Hungary: Simulations with a computable general equilibrium model. Environment and Development Economics 4(4): 537-564.

Nakamura, M, Takahashi, Tand Vertinsky, I. 2001: Why Japanese firms choose to certify: A study of managerial responses to environmental issues. Journal of Environmental Economics and Management 42(1): 23-52.

Netherwood, A. 1998: Environmental management systems. In: Corporate Environmental Management, Vol. 1. Earthscan: London. pp. 35-58.

OECD. 2005: OECD Economic Surveys: Hungary. Organisation for Economic Co-operation and Development: Paris, France.

O'Toole Jr, L and Hanf, K. 1998: Hungary: Political transformation and environmental challenge. Environmental Politics 7: 93-112.

Peng, M. 2000: Business Strategies in Transition Economies. Sage: London.

Uhlenbruck, K, Meyer, K and Hitt, MA. 2003: Organizational transformation in transition economies: Resource-based and organizational learning perspectives. Journal of Management Studies 40(2): 257-282.

Vorhies, DW and Morgan, NA. 2005: Benchmarking marketing capabilities for sustainable competitive advantage. Journal of Marketing 69 (1): 80-94.

Whiteley, R and Czaban, L. 1998: Institutional transformation and enterprise change in an emergent capitalist economy: The case of Hungary. Organization Studies 19(2): 259-280.

Wooldridge, JM. 2002: Econometric Analysis of Cross Section and Panel Data. The MIT Press, Cambridge MA.

World Markets Research Centre. 2005: WMRC Country Report: Hungary (Advanced Country Analysis and Forecast) (Accessed on 16 September 2005), www.WorldMarketsResearchCentrecom.

(1) According to World Markets Research Centre (2005, p. 100), environmental groups are not well organised in Hungary. Consequently, we do not include this pressure in our empirical analysis.

(2) Note that these can be commercial buyers or end-consumers depending on whether an organisation is upstream or downstream in the production process, respectively.

(3) Whether a facility has an environmental R&D budget is not a measure of the level of investments, but rather a dummy variable specifying whether head office gave the facility an environmental R&D budget allocation.

(4) Note that a facility need not have a TQM system to adopt any environmental management practice or system. The adoption of a TQM system simply makes the implementation of an environmental management practice or system (ie, set of practices) less costly given that employees have some familiarity with such processes.

(5) Seven countries were surveyed including Canada, the United States, Germany, Hungary, France, Japan and Norway. The authors were part of the OECD research team that conducted the survey.

(6) Please see Kerekes et al. (2004) for an overview of the Hungarian sample and descriptive statistics.

IRENE HENRIQUES & PERRY SADORSKY

Schulich School of Business, York University, 4700 Keele Street, Toronto, Ontario, Canada M3J 1P3. E-mails: [email protected] or [email protected]

Table 1: Description of variables used in the analyses
of environmental management practices equations

 Expected
 sign on
Variable Name Description coefficient

Dependent variables

 Benchmark Benchmarked environmental NA
 performance (O=no; 1=yes)

 External audit Implemented external NA
 environmental audits
 (O=no; 1=yes)

 Employee evaluation Implemented environmental NA
 criteria in the evaluation
 and/or compensation of
 employees (O=no; 1=yes)

 Performance Implemented environmental NA
 indicators performance indicators/goals
 (O=no; 1=yes)

 Internal audit Implemented internal NA
 environmental audits
 (O=no; 1=yes)

 Public report Implemented a public NA
 environmental report
 (O=no; 1=yes)

 Employee training Implemented environmental NA
 training programs (O=no; 1=yes)

 Written policy Implemented a written NA
 environmental policy
 (O=no; 1=yes)

 Count Sum of the EMS practices NA
 variables (maximum=8,
 minimum=0)

Explanatory variables

 Market scope Scope of facility's market ?
 (0=local; O=national;
 1=regional; 1=global); proxy
 for export orientation

 Facility size Natural logarithm of the number +
 of full time employees in a
 facility

 Facility business Assessment of overall business +
 performance performance over past 3 years
 (1=revenue has been so low as
 to produce large losses;
 2=revenue has been insufficient
 to cover costs; 3=revenue has
 allowed us to break even;
 4=revenue has been sufficient
 to make a small profit;
 5=revenue has been well in
 excess of costs); proxy for
 financial position of facility

 Facility has an Does facility have an +
 R&D budget environmental R&D budget?
 (O=no; 1=yes); proxy for
 head office perspective
 on environmental issues

 Influence of buyers Influence of commercial buyers +
 (0=not important; O=moderately
 important; 1=very important);
 proxy for stakeholder pressure
 --buyers

 Influence of Influence of management +
 management employees (0=not important;
 0=moderately important;
 1=very important); proxy
 for stakeholder pressure--
 management

 Influence of Influence of community groups +
 community (O=not important; 0=moderately
 important; 1=very important);
 proxy for stakeholder pressure
 --community

 Influence of Influence of public authorities +
 public authorities (O=not important; O=moderately
 important; 1=very important);
 proxy for stakeholder pressure
 --public authorities

 Influence of Influence of shareholders/ +
 investors investors (0=not important;
 0=moderately important; 1=very
 important); proxy for
 stakeholder pressure--investors

 Facility has a Does facility have a TOM system +
 TQM system (O=no; 1=yes); proxy for
 managerial experience with
 other management systems

 Firm is international Head office located in foreign ?
 country? (O=no; 1=yes); proxy
 for foreign ownership

 Inspection frequency Number of times facility has +
 been inspected (total over 3
 years); proxy for regulatory
 enforcement

 Industry dummies Omitted category is chemical ?
 products.

Table 2: Correlations and descriptive statistics

Variable Benchmark External Employee
 audit evaluation

Benchmark 1.000
External audit 0.517 1.000
Employee evaluation 0.207 0.173 1.000
Performance indicators 0.498 0.609 0.261
Internal audit 0.394 0.690 0.239
Public report 0.346 0.186 -0.018
Employee training 0.436 0.582 0.293
Written policy 0.347 0.528 0.138
Count 0.693 0.798 0.390
Market scope 0.159 0.199 -0.067
Facility size 0.277 0.309 0.112
Facility business 0.089 0.217 0.093
performance
Facility has an R&D 0.197 0.215 0.001
budget
Influence of buyers 0.031 0.088 0.039
Influence of management 0.220 0.178 0.143
Influence of community 0.102 0.079 0.145
Influence of public 0.093 -0.044 0.016
authorities
Influence of investors 0.131 0.085 -0.015
Facility has a TOM 0.262 0.267 0.002
system
Firm is international 0.103 0.164 0.116
Inspection frequency 0.230 0.183 0.142
Mean 0.379 0.467 0.187
Maximum 1.000 1.000 1.000
Minimum 0.000 0.000 0.000
Std. Dev. 0.487 0.500 0.391
Observations 182 182 182

Variable Performance Internal Public
 indicators audit report

Benchmark
External audit
Employee evaluation
Performance indicators 1.000
Internal audit 0.561 1.000
Public report 0.276 0.237 1.000
Employee training 0.518 0.549 0.230
Written policy 0.529 0.491 0.295
Count 0.788 0.772 0.481
Market scope 0.239 0.097 0.028
Facility size 0.346 0.206 0.164
Facility business 0.114 0.115 0.014
performance
Facility has an R&D 0.219 0.227 0.240
budget
Influence of buyers 0.110 0.065 0.052
Influence of management 0.287 0.175 0.141
Influence of community 0.127 0.007 0.026
Influence of public 0.034 0.053 0.066
authorities
Influence of investors 0.183 0.123 0.165
Facility has a TOM 0.250 0.237 0.161
system
Firm is international 0.244 0.178 0.063
Inspection frequency 0.175 0.141 0.177
Mean 0.522 0.577 0.489
Maximum 1.000 1.000 1.000
Minimum 0.000 0.000 0.000
Std. Dev. 0.501 0.495 0.501
Observations 182 182 182

Variable Employee Written Count
 training policy

Benchmark
External audit
Employee evaluation
Performance indicators
Internal audit
Public report
Employee training 1.000
Written policy 0.541 1.000
Count 0.768 0.721 1.000
Market scope 0.188 0.170 0.193
Facility size 0.260 0.260 0.359
Facility business 0.198 0.239 0.199
performance
Facility has an R&D 0.240 0.217 0.292
budget
Influence of buyers 0.096 0.061 0.100
Influence of management 0.186 0.193 0.280
Influence of community 0.170 0.058 0.128
Influence of public 0.005 0.068 0.053
authorities
Influence of investors 0.119 0.075 0.163
Facility has a TOM 0.285 0.410 0.351
system
Firm is international 0.186 0.234 0.237
Inspection frequency 0.165 0.156 0.251
Mean 0.489 0.593 3.703
Maximum 1.000 1.000 8.000
Minimum 0.000 0.000 0.000
Std. Dev. 0.501 0.493 2.647
Observations 182 182 182

Variable Market Facility Facility
 scope size business
 performance

Benchmark
External audit
Employee evaluation
Performance indicators
Internal audit
Public report
Employee training
Written policy
Count
Market scope 1.000
Facility size 0.128 1.000
Facility business -0.044 0.121 1.000
performance
Facility has an R&D 0.039 0.252 0.031
budget
Influence of buyers 0.192 0.100 0.204
Influence of management -0.023 0.104 0.106
Influence of community -0.036 0.102 0.081
Influence of public -0.119 0.044 0.047
authorities
Influence of investors 0.001 0.157 0.199
Facility has a TOM 0.157 0.238 0.186
system
Firm is international 0.195 0.141 0.169
Inspection frequency 0.005 0.400 0.119
Mean 0.626 5.777 3.703
Maximum 1.000 9.048 5.000
Minimum 0.000 3.332 1.000
Std. Dev. 0.485 0.981 1.082
Observations 182 182 182

Variable Facility Influence Influence
 has an R&D of buyers of
 budget management

Benchmark
External audit
Employee evaluation
Performance indicators
Internal audit
Public report
Employee training
Written policy
Count
Market scope
Facility size
Facility business
performance
Facility has an R&D 1.000
budget
Influence of buyers -0.017 1.000
Influence of management 0.095 0.249 1.000
Influence of community 0.091 0.145 0.276
Influence of public 0.082 -0.011 0.090
authorities
Influence of investors 0.246 0.119 0.379
Facility has a TOM 0.130 0.089 0.035
system
Firm is international -0.104 0.174 0.110
Inspection frequency 0.383 -0.100 -0.019
Mean 0.088 0.401 0.412
Maximum 1.000 1.000 1.000
Minimum 0.000 0.000 0.000
Std. Dev. 0.284 0.491 0.494
Observations 182 182 182

Variable Influence Influence Influence
 of of public of
 community authorities investors

Benchmark
External audit
Employee evaluation
Performance indicators
Internal audit
Public report
Employee training
Written policy
Count
Market scope
Facility size
Facility business
performance
Facility has an R&D
budget
Influence of buyers
Influence of management
Influence of community 1.000
Influence of public 0.123 1.000
authorities
Influence of investors 0.192 0.177 1.000
Facility has a TOM 0.039 0.114 0.030
system
Firm is international -0.083 -0.035 0.041
Inspection frequency 0.142 0.070 0.146
Mean 0.302 0.841 0.368
Maximum 1.000 1.000 1.000
Minimum 0.000 0.000 0.000
Std. Dev. 0.460 0.367 0.484
Observations 182 182 182

Variable Facility Firm Inspection
 has a is frequency
 TOM system international

Benchmark
External audit
Employee evaluation
Performance indicators
Internal audit
Public report
Employee training
Written policy
Count
Market scope
Facility size
Facility business
performance
Facility has an R&D
budget
Influence of buyers
Influence of management
Influence of community
Influence of public
authorities
Influence of investors
Facility has a TOM 1.000
system
Firm is international 0.014 1.000
Inspection frequency 0.091 -0.057 1.000
Mean 0.852 0.275 5.423
Maximum 1.000 1.000 100.000
Minimum 0.000 0.000 0.000
Std. Dev. 0.356 0.448 10.986
Observations 182 182 182

Table 3: Probit and Poisson regression (count)
results for environmental management practices

 Dependent variable

 Benchmark External audit

Explanatory variable Coef. M.E. Coef. M.E.

Constant -2.293 ** -0.814 -2.991 *** -1.169
Market scope 0.227 0.079 0.366 0.141
Facility size 0.116 0.041 0.323 ** 0.126
Facility business -0.145 -0.051 0.145 0.056
 performance
Facility has an 0.554 0.212 0.974 * 0.367
 R&D budget
Influence of buyers -0.263 -0.092 -0.177 -0.069
Influence of management 0.557 ** 0.200 0.530 ** 0.206
Influence of community 0.127 0.046 0.072 0.028
Influence of public 0.356 0.118 -0.234 -0.093
 authorities
Influence of investors -0.108 -0.038 -0.267 -0.103
Facility has a TQM 0.943 ** 0.267 0.691 * 0.246
 system
Firm is international 0.286 0.104 0.292 0.115
Inspection frequency 0.071 *** 0.025 0.026 0.010
Food -0.081 -0.028 -0.894 ** -0.309
Textiles -0.562 -0.172 -0.301 -0.113
Wood -8.991 *** -0.382 -8.954 *** -0.490
Paper 0.886 0.342 0.310 0.123
Non-metal -1.238 ** -0.297 -0.718 -0.248
Metal -0.376 -0.122 -1.170 *** -0.361
Machine 0.332 0.122 -0.285 -0.109
Transportation 1.697 *** 0.582 -0.450 -0.164
R-squared (a) 0.277 0.267
Log likelihood -87.318 -92.231
LR statistic (a) 66.926 *** 67.052 ***
Percent correctly 75.27 76.92
 predicted
Number of observations 182 182

 Dependent variable

 Employee evaluation Internal audit

Explanatory variable Coef. M.E. Coef. M.E.

Constant -1.079 -0.112 -1.095 -0.420
Market scope -0.510 ** -0.060 -0.052 -0.020
Facility size 0.132 0.014 0.145 0.056
Facility business 0.022 0.002 -0.011 -0.004
 performance
Facility has an -0.358 -0.029 1.303 *** 0.364
 R&D budget
Influence of buyers 0.164 0.018 -0.002 -0.001
Influence of management 0.380 0.042 0.406 * 0.153
Influence of community 0.295 0.034 -0.248 -0.096
Influence of public 0.046 0.005 0.105 0.041
 authorities
Influence of investors -0.470 -0.045 -0.084 -0.032
Facility has a TQM -0.489 -0.067 0.457 0.180
 system
Firm is international 0.433 * 0.053 0.521 ** 0.190
Inspection frequency 0.014 0.001 0.014 0.006
Food -0.860 ** -0.060 -0.424 -0.166
Textiles -7.994 *** -0.140 -1.053 ** -0.396
Wood -7.837 *** -0.069 -0.442 -0.175
Paper 0.924 0.180 0.701 0.230
Non-metal -0.633 -0.043 -1.008 ** -0.382
Metal -0.114 -0.011 -0.360 -0.142
Machine -0.431 -0.038 0.024 0.009
Transportation 0.043 0.005 -0.334 -0.132
R-squared (a) 0.184 0.192
Log likelihood -71.561 -100.234
LR statistic (a) 32.169 ** 47.514 ***
Percent correctly 82.97 71.98
 predicted
Number of observations 182 182

 Dependent variable

 Performance indicators Public report

Explanatory variable Coef. M.E. Coef. M.E.

Constant -3.102 *** -1.231 -0.122 -0.049
Market scope 0.411 * 0.163 -0.285 -0.113
Facility size 0.371 ** 0.147 0.075 0.030
Facility business -0.037 -0.014 -0.165 -0.066
 performance
Facility has an 1.095 ** 0.362 0.979 ** 0.347
 R&D budget
Influence of buyers -0.027 -0.011 0.181 0.072
Influence of management 0.614 ** 0.238 0.196 0.078
Influence of community 0.242 0.095 -0.190 -0.076
Influence of public -0.044 -0.017 0.173 0.069
 authorities
Influence of investors 0.042 0.016 0.234 0.093
Facility has a TQM 0.659 ** 0.256 0.280 0.111
 system
Firm is international 0.673 ** 0.255 0.136 0.054
Inspection frequency 0.022 0.009 0.043 * 0.017
Food -0.630 * -0.246 -0.936 *** -0.347
Textiles -1.671 ** -0.517 -0.764 * -0.284
Wood 0.962 0.319 -1.510 *** -0.446
Paper 0.397 0.151 0.050 0.020
Non-metal -0.495 -0.194 -1.023 ** -0.361
Metal -0.853 ** -0.322 -0.706 * -0.266
Machine 0.003 0.001 -0.077 -0.031
Transportation 0.480 0.180 0.053 0.021
R-squared (a) 0.329 0.164
Log likelihood -84.504 -105.483
LR statistic (a) 82.946 *** 41.252 ***
Percent correctly 78.02 68.68
 predicted
Number of observations 182 182

 Dependent variable

 Employee training Written policy

Explanatory variable Coef. M.E. Coef. M.E.

Constant -2.333 *** -0.922 -2.944 *** -1.122
Market scope 0.489 ** 0.189 0.294 0.113
Facility size 0.195 0.077 0.171 0.065
Facility business 0.086 0.034 0.200 * 0.076
 performance
Facility has an 1.603 *** 0.518 1.267 ** 0.352
 R&D budget
Influence of buyers -0.135 -0.053 -0.224 -0.086
Influence of management 0.359 0.142 0.525 ** 0.195
Influence of community 0.393 0.156 0.021 0.008
Influence of public -0.300 -0.119 0.062 0.024
 authorities
Influence of investors -0.183 -0.072 -0.316 -0.122
Facility has a TQM 0.667 * 0.244 1.510 *** 0.538
 system
Firm is international 0.533 ** 0.210 0.770 ** 0.269
Inspection frequency 0.014 0.005 0.018 0.007
Food -0.182 -0.071 -0.614 -0.240
Textiles -1.020 * -0.336 -1.082 ** -0.407
Wood -9.244 *** -0.519 -0.068 -0.026
Paper 1.195 * 0.416 -0.180 -0.070
Non-metal -0.072 -0.028 -0.350 -0.137
Metal -0.387 -0.147 -0.686 -0.268
Machine -0.282 -0.110 -0.461 -0.179
Transportation -0.128 -0.050 -0.694 -0.271
R-squared (a) 0.251 0.303
Log likelihood -94.480 -85.710
LR statistic (a) 63.257 *** 74.496 ***
Percent correctly 69.78 74.18
 predicted
Number of observations 182 182

 Dependent variable

 Count

Explanatory variable Coef. M.E.

Constant -0.177
Market scope 0.075 0.279
Facility size 0.123 ** 0.457
Facility business 0.019 0.070
 performance
Facility has an 0.327 *** 1.209
 R&D budget
Influence of buyers -0.040 -0.146
Influence of management 0.316 *** 1.169
Influence of community -0.007 -0.024
Influence of public 0.062 0.229
 authorities
Influence of investors -0.066 -0.245
Facility has a TQM 0.632 *** 2.339
 system
Firm is international 0.261 *** 0.965
Inspection frequency 0.002 0.006
Food -0.376 ** -1.394
Textiles -0.873 ** -3.231
Wood -0.615 *** -2.277
Paper 0.149 0.553
Non-metal -0.513 ** -1.901
Metal -0.436 ** -1.614
Machine -0.107 -0.396
Transportation -0.118 -0.435
R-squared (a) 0.426
Log likelihood -379.174
LR statistic (a) 156.317 ***
Percent correctly NA
 predicted
Number of observations 182

Quasi-maximum-likelihood standard errors used in computing
P-values. *** P < 0.01, ** P < 0.05, * P < 0.10.
Marginal effects (M.E.) shown beside coefficient estimates.

(a) McFadden R-squared values reported for the individual
practices equations.

(b) Likelihood ratio test of a model against a model that
only includes a constant.