- Email: [email protected]
The effect of ISO 14001 certification on toxic emissions: an analysis of industrial facilities in the north of Spain
Journal of Cleaner Production 19 (2011) 1091e1095
Contents lists available at ScienceDirect
Journal of Cleaner Production journal homepage: www.elsevier.com/locate/jclepro
The effect of ISO 14001 certification on toxic emissions: an analysis of industrial facilities in the north of Spain Alberto Gomez*, Monica A. Rodriguez Business Administration Department, E.P.S. de Ingeniería, Campus Universitario de Gijón, 33203 Gijón, Asturias, Spain
a r t i c l e i n f o
a b s t r a c t
Article history: Received 11 March 2010 Received in revised form 24 January 2011 Accepted 26 January 2011 Available online 26 February 2011
During recent decades, the impact of industrial organisations on the environment has become clearly evident. It has also become more difficult to hide and more expensive. Globally, this has caused many enterprises to put all their efforts into seeking management instruments that allow them to reduce their negative impact on the environment as well as improving their economic efficiency. This environmental interest is either voluntary or forced by customers or by legal pressure. Environmental Management Systems (EMSs) are among the many tools that have appeared to fulfil that goal and that have drawn international researchers’ attention. The most popular system and the one most often used is the ISO 14001 standard. This paper examines empirically the influence that this certification exerts on the company’s pollutant emission policy. The analysis was carried out in four regions of Spain: Asturias, Cantabria, Galicia and Castilla-León and includes 126 industrial organisations. The goal of this paper is to investigate the behaviour of the companies according to their emissions policies once they have achieved the ISO 14001 certification. In the paper the Toxics Release Index of 56 certified companies is compared with the Index of 70 noncertified companies. Through a statistical analysis based on the Student’s t-test and the ManneWhitney U test, it was concluded that ISO 14001 does not represent an environmental proactivity signal clearly enough to result in a reduction of the company’s environmental polluting index. Ó 2011 Elsevier Ltd. All rights reserved.
Keywords: ISO 14001 Toxic emission Environmental management strategy Green supply chain management
1. Introduction It is widely accepted that the patent and ever-growing process of environmental degradation has not been constant over the years. On the contrary, it has been boosted by many years of uninterrupted development added to the massive consumption of energy and resources which has resulted in significant global risks such as global warming and ozone layer reduction. In 1972, therefore, the Conference of Stockholm was held, the first global United Nations conference on the environment. In it, it was accepted for the first time that it was not possible to maintain the growth achieved until then on a basis of unlimited exploitation of natural resources. At the same time, the United Nations Environment Program (UNEP) was created. This United Nations declaration was reaffirmed in 1992 with the celebration of the Earth Summit or Rio Conference, at which an attempt was made to adopt an approach to development that would ensure social and economic growth while protecting the
* Corresponding author. Tel.: þ34 985 18 21 06; fax: þ34 985 18 21 10. E-mail addresses: [email protected] (A. Gomez), [email protected] (M.A. Rodriguez). 0959-6526/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.jclepro.2011.01.012
environment. From that moment on, concern about environmental harm has grown more intensely (Nishitani, 2009). International regulations started to become important in successfully accessing markets, and, many companies have made environmental management a definitive priority. In response to the need to improve environmental quality (Bansal and Bogner, 2002), the International Organisation for Standardisation (ISO) in 1996 created a series of standards known as ISO 14000, comprising several advisory documents related to environmental management systems, environmental auditing, environmental performance evaluation, ecological labelling, lifecycle evaluation and product environmental aspects. This series was later revised in 2004, and although it is not the only way to efficiently implement an environmental management system, it does attempt to advise companies on developing better environmental policies. ISO 14001 is part of this series, and is the only standard specifically designed to meet the needs of an environmental management system. This standard has become required by customers and other interested parties, and as a result it has become a conditioning factor within the supply chain (Nawrocka and Parker, 2009). However, it is also true that there are many
1092
A. Gomez, M.A. Rodriguez / Journal of Cleaner Production 19 (2011) 1091e1095
companies still reluctant to adopt the standard, since they do not understand the benefits obtained from its application (Babakri et al., 2004). According to this standard, the environmental management system (EMS) offers the company a measurable result of its environmental performance, as it operates on a “Plan, Do, Check, Act” cycle basis. This means that it is put into practice as a continuous cycle of declaration, scheduling, execution, revision and improvement of the processes and actions carried out by the company (Perotto et al., 2008). In addition, industries are constantly facing environmental regulations that become ever stricter. For instance, in 1996 the EU adopted Directive 96/61/CE concerning integrated pollution prevention and control (IPPC Directive) by which industrial and agricultural facilities are required to obtain an environmental permit that guarantees compliance with a series of basic environmental requirements. This directive was incorporated into the Spanish legal system through the Law of Integrated Prevention and Control of Pollution, the aim of which is to prevent, and if that is impossible, to reduce and control pollution of the atmosphere, water and the soil by means of a system of integrated prevention and control. Whether due to compliance with current legislation or to market pressure, the business world’s commitment to the environment has become a variable of growing importance within an increasingly competitive market. Within this scenario, ISO 14001 offers various guidelines to design an EMS that enables the company to identify its own environmental policy (Murat, 2009), as well as to correct possible deficiencies. According to statistics published by the ISO (ISO, 2007), 154,571 certificates were issued during 2007 in 148 countries (29% in the services sector). This was an increase of 21% over 2006 when the total was 128,211 in 140 countries (27% in the services sector). Spain was ranked third with 13,825 certifications compared to 11,125 at the end of 2006. It is interesting to study the extent to which adopting the standard is related to an environmental transformation in productive operations and processes (González-Benito and González-Benito, 2008) and to technological modernisation. Several authors (Hart, 1995; Delmas and Toffel, 2003) have tried to formally study companies’ environmental strategic behaviour. They have observed that this behaviour describes a continuum that ranges from a passive or reactive strategy to the most advanced or proactive strategy. Companies showing a passive approach towards environmental management evaluate environmental aspects as cost factors, and therefore do not perceive the possible competitive advantages. Those with an active approach develop an environmental awareness only because of legislation. Proactive companies, on the other hand, weigh up environmental aspects in all decision-making processes, and see their obligation towards the environment as an opportunity to increase their share in markets for environmentally friendly products. If a reactive strategy is only intended to comply with the law, it will solve the environmental problem in the short run, but it will not be sufficient to improve the company’s competitive position. That is why, in response to the demands of their customers, employees, shareholders, the mass-media, etc. companies focus their actions on voluntary or proactive strategies for pollution prevention. In this sense, authors such as Porter (1991) and Porter and Van der Linde (1995) support the “Porter hypothesis”. This hypothesis supposes that in the relationship between proactive strategies and a company’s economy, it is possible to obtain a free-lunch, in which environmental quality targets are achieved. At the same time productive costs are reduced and productivity is improved. One usual practice within proactive strategies is to obtain ISO 14001 certification. It is in this that this paper finds its goal. The paper intends to investigate further the objective environmental consequences stemming from certification of a company’s EMS. Specifically, this
paper’s aim is to develop an empirical approach to the effect of ISO 14001 certification on the levels of environmental pollution, thus answering the question of whether this certification results in a reduction in the levels of polluting compounds released by the company or not. 2. State of the art Although nowadays the evident and costly impact industrial organisations exert on the environment is widely acknowledged, it is neither easy to find precedent papers that investigate the link between ISO 14001 certification and the company’s environmental performance nor papers that evaluate the time elapsed between the company’s certification and the moment it starts to reduce emissions. On the contrary, most papers tackle the ISO 14001 problem from the viewpoint of its proliferation in countries and industries (Delmas, 2000; Corbett and Russo; Kollman and Praskash, 2002; Corbett and Russo, 2001), its benefits (Tibor and Feldman, 1996), and the features and attributes of the organisations that tend to contribute to the improvement of the standard’s results. What is more, even today, ISO 14001 is still a widely challenged criterion, since it does not force the organisation to improve its environmental performance (Corbett and Kirsch, 2000). It can be seen that while some authors think that the adoption of the standard represents the possible obtaining of important advantages in environmental development, others argue that since it does not measure environmental development directly, it is not an appropriate tool for improving environmental sustainability. Few authors categorically conclude that ISO 14001 exerts a positive impact on pollution levels. Montabon et al. (2000) suggest that ISO 14001 may improve environmental performance as well as economic efficiency. Melnyk et al. (2002) also link the standard to a reduction in waste. Russo and Harrison (2001) have evaluated the impact of the standard on a sample of installations in the electronics sector and have concluded that the certification seems to lead to a significant decrease in the levels of toxic emissions. Whether this can be extrapolated to other industrial sectors has not been categorically answered. In another paper Russo (2007) points out that being one of the first installations to adopt the standard seems to be related to lower emission levels. Early certified installations achieved longlasting benefits. Additionally, the greater the experience the companies had with the standard, the lower the emissions. Being an early user of the standard and having greater experience are factors that together work to reduce emissions. Szymanski and Ikeda (2003) evaluated the time it takes reduction in emissions to take effect once certification has been achieved. They also analysed whether there were companies that tended to reduce their emissions faster. They found that more than half the companies analysed reduced their emissions during the first year but, since they did not use a control group, it is risky to state that the reduction was primarily due to certification. On the other hand, Andrews et al. (2003) found evidence that ISO 14001 does not affect the company’s environmental performance. King et al. (2005) focus their perspective on the reasons that cause a facility to seek certification. They found that ISO 14001 had a weak negative effect on the increase of the emissions. Since in their analysis few control variables were used, the authors themselves state that the study needed to be improved to yield more reliable results. 3. Sample selection The sample study was selected from companies belonging to different industrial sectors. They were all located in the regions of Asturias, Cantabria, Galicia and Castilla-León. It includes a sample
A. Gomez, M.A. Rodriguez / Journal of Cleaner Production 19 (2011) 1091e1095
made up of 126 companies, of which 56 have been ISO 14001 certified since 2001. The other 70 are not certified to the standard. These companies were taken from the Pollutant Release and Transfer Register (http://www.prtr-es.es/). This register was created by the Environment Ministry in order to make available information relating to the release of pollutant emissions into water, soil and the atmosphere. This register also made available waste transfers out of the locations of industrial complexes regulated by Law 16/2002 on the 1st of July, related to Integrated Pollution Prevention and Control (IPPC Law). This is a limited register since it only provides information from those industrial sectors regulated by the aforesaid law and also from large companies. The emission data is only stated if they are over the notification threshold established by Royal Decree 508/2007. These values are therefore not to be taken as limit-values. Data referring to atmospheric emissions was used in the analysis. The data was taken during the period 2001e2007, taking into account only installations with emission dispersal activity during the period. The data was measured for the amount of compounds released (kg/year) per industrial facility. Given the great amount of industrial agricultural and cattle facilities in the sample, these were removed from the study so as to avoid distortions in processing statistics. For the same reason, given the high toxic variability of the different compounds, the quantity of compound released was divided by its notification threshold. The amounts of all compounds released by each installation during one year were added together. In this way, a pollutant release index was obtained for every installation and year. Because of the bias of the data, extra finetuning was necessary; the value 1 was added to them and the result underwent a logarithmic transformation (Russo and Harrison, 2005; King and Lenox, 2000). Since the variable had been referenced making use of indexes, the mean values of the index were analysed per year. Toxics release index ¼ log[1 þ S(Ei/RQi)] Ei ¼ compound i emissions into air, water and soil. RQi ¼ notification-threshold values for compound i. i ¼ index representing every compound considered by the register. 4. Methodology
Fig. 1. Histogram for the certified companies sample.
distribution (they are even used when one of the groups has normal distribution and the other does not) and are related to ordinal or nominal quantitative data, high variances and small sample size (under 30 cases). Nevertheless, parametric tests can be applied to those cases that have a sample comprised of more than 30 cases, no matter the variable distributions. 4.1. Descriptive study of normality A descriptive study of normality was carried out by using SPSS 15.0 software in order to determine if the data in each sample came from a normal population. In a first verification, the histogram was obtained for each sample with unknown distribution. Then, the normal probability graphic was drawn from which the PeP and QeQ graphs were obtained. Since these methods are simply descriptive ones, it could happen that the data was not normal, although the verifications were reasonably well carried out. That is why it was necessary to continue the study with a statistical normality analysis by means of the KolmogoroveSmirnov test.
In order to carry out the analysis, two types of variables were considered: one quantitative, represented by the company’s toxics release index, the other of a categorical type, represented by the ISO 14001 certification. For the categorical variable a study group (certified companies) and a control group (non-certified companies) were defined. It was thereby possible to examine whether the fact of a company being certified according to the standard was somehow related to the degree of environmental pollution produced by the company. Since a different experimental situation is applied to each group, it is a question of a hypothesis contrast for two independent samples in which the following hypotheses were set out: H0: m certified companies ¼ m (m certified m non-certified ¼ 0) H1: m certified s m non-certified
non-certified
companies
In order to choose the appropriate type of contrast to apply, the kind of distribution in which the variables fit was examined, since, depending on this distribution, either parametric or non-parametric tests were to be used (Bartoszynski and NiewiadomskaBugaj, 2002). Parametric tests are best suited to the study of quantitative variables with normal distribution and equal variances. Non-parametric tests do not take into account the type of
1093
Fig. 2. Histogram for the non-certified companies sample.
1094
A. Gomez, M.A. Rodriguez / Journal of Cleaner Production 19 (2011) 1091e1095
Table 1 KolmogoroveSmirnov test for both groups.
Table 3 Descriptive statistics of the cases under study.
KolmogoroveSmirnov test for certified and non-certified installations
N
Certified companies emissions
Non-certified companies emissions
56
70
Normal parametersa,b
Mean Standard deviation
0.2452 0.17336
0.2121 0.18352
Most extreme differences
Absolute Positive Negative
0.129 0.129 0.112
0.169 0.169 0.160
Kolmogorove Smirnov Z
0.963
1.411
Asymptotic significance (bilateral)
0.312
0.037
a b
Index
CERTIF
N
Average range
Ranges total
Non-certified Certified Total
70 56 126
59.90 68.00
4193.00 3808.00
Once the fact that the “emissions” variable follows different distributions in both samples was verified, the study continued with a hypotheses contrast on the equality of means. Because of the number-of-cases criterion it would be suitable to perform a parametric contrast, but according to the type-of-distribution criterion, the most suitable would be a non-parametric contrast. Since they were both completely feasible, they both were performed so as to have them complement each another. 5.3. Parametric contrast
The contrast distribution is the normal. Worked out from the data.
In order to parametrically contrast the equality of means, a Student’s t-test was carried out under the following hypotheses:
4.2. Statistical study of normality To check the normality of the “pollutant index” variable in both samples, the following hypotheses were produced: H0: “pollutant index” variable is normally distributed H1: “pollutant index” variable is not normally distributed 5. Results and discussion 5.1. Descriptive study: histograms The histograms obtained for each sample are depicted and the results are presented in Figs. 1 and 2. As can be seen, they show curves that are not similar to a normal curve, which means it is not possible to state unequivocally that both samples come from normal populations. 5.2. Statistical study: KolmogoroveSmirnov test The results obtained by the KolmogoroveSmirnov test are shown in Table 1. The mean toxic release index for the certified group is 0.2452, with a standard deviation of 0.17336, while the mean toxic release index for the uncertified group is 0.2121, with a standard deviation of 0.18352. In this case, an asymptotic significance (bilateral) is obtained with a probability of 0.312 (p > 0.05) which leads us to reject hypothesis H1. It is not possible to reject the hypothesis that the “emissions” variable follows a normal distribution for any significance level lower than 31.2%. In short, in this case the variable index follows a normal distribution. On the other hand, in the group of non-certified facilities, the asymptotic significance has a probability of 0.037 (p < 0.05) which leads us to accept hypothesis H1. In this case, the variable does not follow a normal distribution.
Table 2 Summary of statistics for both groups (certified group and non-certified group). Group statistics
Index
Ranges
CERTIF
N
Mean
Standard deviation
Mean standard error
Certified Non-certified
56 70
0.2452 0.2121
0.17336 0.18352
0.02317 0.02193
H0: m certified companies ¼ m (m certified m non-certified ¼ 0) H1: m certified s m non-certified
non-certified
companies
Table 2 shows a brief summary of the cases processed by means of their descriptive statistics: N (sample size), mean, standard deviation and mean standard error. As can be seen in the following table, the variance homogeneity test or Levene’s test obtains a value “p ¼ 0.45” (p > 0.05) associated to the null hypothesis that the variances are homogeneous. Hence we assume that variances are homogeneous. Under this premise statistic t obtains a value of 1.029 (with 124 degrees of freedom) and the associated value “p” is 0.306 (p > 0.02). It can therefore be concluded that the differences between the emission levels of certified and non-certified installations are not significant (they are not statistically different for a significance level of 0.05 and as a consequence there is no link between the ISO 14001 certification and the pollutant emission index). The interpretation of the difference of means between both groups would set this difference within the population with high reliability, between 0.03051 and 0.09658, which allows us to understand whether the means are different between both groups. Since CI 95% contents zero, the non-difference of means in the “emissions” variable must be assumed within the population. 5.4. Non-parametric contrast The equality of means was contrasted from a non-parametric perspective by performing a ManneWhitney U test. The results lead to a conclusion that is similar to the one obtained in the previous comparison. First of all, a summary of the cases processed Table 4 Non-parametric contrast statistics. Contrast statisticsa Index ManneWhitney U Wilcoxon W Z Asymptotic significance (bilateral) a
Grouping variable: CERTIF.
1708.000 4193.000 1.238 0.216
A. Gomez, M.A. Rodriguez / Journal of Cleaner Production 19 (2011) 1091e1095
by means of their descriptive statistics is shown: sample size, mean, standard deviation and maximum and minimum values (Table 3). The information included inside the quantitative variable was processed to obtain different comparison statistics. The results are presented in Table 4. The p-value of 0.216 (p > 0.05) obtained for the asymptotic significance leads us to reject the alternative hypothesis “the mean of emissions is different in both groups”. Therefore, by accepting the null hypothesis we are concluding that there is no statistically significant association between the pollutant emission index and ISO 14001 certification. 6. Conclusions ISO 14001 certified and non-certified companies were analysed to investigate whether there is a relationship between certification and the toxics release rates of both groups. In the article, the Toxics Release Index of 56 certified companies are compared with that of 70 non-certified companies, with the help of the statistics program SPSS 15.0. Firstly, the normality of the Toxic Release Index was analyzed, with the conclusion that this index continues to follow the normal pattern of distribution in the case of certified companies, but in the case of non-certified companies this cannot be stated with assurance. To analyze the effect of certification, it was decided to compare the averages of both indexes with the help of a Student’s test and a ManneWhitney U test. Both tests concluded that these averages are the same, with a significance level of 0.05. The final conclusion is that there is no relationship that links the fact of a company being certified to its pollutant emission indexes. The empirical verification of the fact that the certification does not cause the pollutant emissions level to decrease significantly, suggests that the companies perceive the certification as a reactivetype investment rather than one that is proactive. In short, certification is considered to be a standard that must be adopted because of institutional pressure, but not as an investment that favours an improved environment. It could be said that the adoption of the standard is seen as a sign that tries to show the company as socially within the law without resulting in significant changes to the company’s management. In this sense, authors like Boiral (2001) warn that ISO 14001 may be used by companies as a marketing tool, something that is aggravated, according to Corbett and Kirsch (2000), by the fact that the certification postulates neither specific practices nor definite result requirements. Given the issue’s importance and topicality, research into environmental strategies adopted by companies and their responsibilities for environmental degradation needs to be more deeply investigated. It should nevertheless be noted that the conclusions obtained in this study are similar to the ones obtained by Barla (2007). He tried to verify whether the adoption of the standard produces a significant impact on the paper industry’s environmental performance. He found that certification does not lead to a reduction in suspended solid emissions. In his study, he concludes that facilities adopting the standard did not undergo any significant reduction in their emissions. The impact of the standard is highly variable: even though some facilities reduce their emissions after certification, the emissions levels of most facilities remain stable or even increase after certification. Further research is needed (according to production and studying the same industrial sector, for example) to determine whether potency-weighted air emissions alone are an
1095
adequate metric for determining whether certification results in reduced emissions. References Andrews, R.N.L., et al., 2003. Environmental management systems: do they improve performance? Final Report, 30 January 2003. Available from: http://ndems.cas. unc.edu (accessed September 2009). Babakri, K.A., et al., 2004. Recycling performance of firms before and after adoption of the ISO 14001 standard. Journal of Cleaner Production 12 (6), 633e637. Bansal, P., Bogner, W.C., 2002. Deciding on ISO 14001: economics, institutions, and context. Long Range Planning 35 (3), 269e290. Barla, P., 2007. ISO 14001 certification and environmental performance in Quebec’s pulp and paper industry. Journal of Environmental Economics and Management 53, 291e306. Bartoszynski, R., Niewiadomska-Bugaj, M., 2002. Probability and Statistical Inference. In: Wiley Series in Probability and Statistics. Wiley-Interscience. Boiral, O., 2001. ISO 14001 certification in multinational firms: the paradoxes of integration. Global Focus 13 (1), 79e94. Corbett, C.J., Russo, M., 2001. ISO 14001: irrelevant or invaluable, ISO Management Systems, (special issue: The impact of ISO 14001, December), pp. 23e9. Corbett C., Russo, M. ISO 14001: Irrelevant or Invaluable?, ISO management Systems. Corbett, C., Kirsch, D.A., 2000. ISO 14000: an agnostic’s report from the frontline. ISO 9000 þ IS0 14000 News March/April 2. Delmas, M., 2000. Barriers and Incentives to the Adoption of ISO 14001 in the United States. Duke Environmental Law and Policy Forum Fall. Delmas, M., Toffel, M.W., 2003. Institutional pressure and environmental management practices. In: Presented at the 11th International conference of the Greening of Industry Network. October, San Francisco. González-Benito, J., González-Benito, O., 2008. Operations management practices linked to the adoption of ISO 14001: an empirical analysis of Spanish manufacturers. International Journal of Production Economics 113, 60e73. Hart, S.L., 1995. A natural-resource-based view of the firm. Academy of Management Review 20 (4), 986e1014. ISO, 2007. The ISO Survey of Certification 2006. King, A., Lenox, M., 2000. Industry self-regulation without sanctions: the chemical industry’s Responsible Care Program. Academy of Management Journal 43, 698e716. King, A., et al., 2005. The strategic use of decentralized institutions: exploring certification with the ISO 14001 management standard. Academy of Management Journal 48, 1091e1106. Kollman, K., Praskash, A., 2002. EMS-based environmental regimes as club goods: examining variations in firm-level adoption of ISO 14001 and EMAS in UK, US, and Germany. Policy Sciences 35, 43e67. Melnyk, S.A., et al., 2002. Assessing the effectiveness of US voluntary environmental programmes: an empirical study. International Journal of Production Research 40, 1853e1878. Montabon, F., et al., Spring 2000. ISO 14000: assessing its perceived impact on corporate performance. The Journal of Supply Chain Management (4), 4e15. Murat, A., 2009. The benefits associated with ISO 14001 certification for construction firms: Turkish case. Journal of Cleaner Production 17, 559e569. Nawrocka, D., Parker, T., 2009. Finding the connection: environmental management systems and environmental performance. Journal of Cleaner Production 17, 601e607. Nishitani, K., 2009. An empirical study of the initial adoption of ISO 14001 in Japanese manufacturing firms. Ecological Economics 68, 669e679. Perotto, E., et al., 2008. Environmental performance, indicators and measurement uncertainty in EMS context: a case study. Journal of Cleaner Production 16, 517e530. Porter, M., 1991. America’s green strategy. Scientific American 264 (4), 168. Porter, M., Van der Linde, C., 1995. Green and competitive: ending the stalemate. Harvard Business Review 73 (5), 120e134. Russo, M.V., 2007. Explaining the impact of ISO 14001 on emission performance: a dynamic capabilities perspective on process and learning. Business Strategy and the Environment 18 (5), 307e319. Russo, M.V., Harrison, A., 2001Paper presented at the Ninth Greening of Industry Network Annual Conference, Bangkok, Thailand. An empirical study of the impact of ISO 14001 registration on emissions performance. Russo, M.V., Harrison, N.S., 2005. Organizational design and environmental performance: clues from the electronic industry. Academy of Management Journal 48, 582e593. Szymanski, M., Ikeda, S., 2003. ISO 14001 and reduction of toxic emissions. In: Proceedings of EcoDesign 2003: Third International Symposium on Environmentally Conscious Design and Inverse Manufacturing Tokyo, Japan. Tibor, T., Feldman, I., 1996. ISO 14000: A Guide to the New Environmental Management Standards. Irwin Professional Publishing, Chicago, Illinois.
Contents lists available at ScienceDirect
Journal of Cleaner Production journal homepage: www.elsevier.com/locate/jclepro
The effect of ISO 14001 certification on toxic emissions: an analysis of industrial facilities in the north of Spain Alberto Gomez*, Monica A. Rodriguez Business Administration Department, E.P.S. de Ingeniería, Campus Universitario de Gijón, 33203 Gijón, Asturias, Spain
a r t i c l e i n f o
a b s t r a c t
Article history: Received 11 March 2010 Received in revised form 24 January 2011 Accepted 26 January 2011 Available online 26 February 2011
During recent decades, the impact of industrial organisations on the environment has become clearly evident. It has also become more difficult to hide and more expensive. Globally, this has caused many enterprises to put all their efforts into seeking management instruments that allow them to reduce their negative impact on the environment as well as improving their economic efficiency. This environmental interest is either voluntary or forced by customers or by legal pressure. Environmental Management Systems (EMSs) are among the many tools that have appeared to fulfil that goal and that have drawn international researchers’ attention. The most popular system and the one most often used is the ISO 14001 standard. This paper examines empirically the influence that this certification exerts on the company’s pollutant emission policy. The analysis was carried out in four regions of Spain: Asturias, Cantabria, Galicia and Castilla-León and includes 126 industrial organisations. The goal of this paper is to investigate the behaviour of the companies according to their emissions policies once they have achieved the ISO 14001 certification. In the paper the Toxics Release Index of 56 certified companies is compared with the Index of 70 noncertified companies. Through a statistical analysis based on the Student’s t-test and the ManneWhitney U test, it was concluded that ISO 14001 does not represent an environmental proactivity signal clearly enough to result in a reduction of the company’s environmental polluting index. Ó 2011 Elsevier Ltd. All rights reserved.
Keywords: ISO 14001 Toxic emission Environmental management strategy Green supply chain management
1. Introduction It is widely accepted that the patent and ever-growing process of environmental degradation has not been constant over the years. On the contrary, it has been boosted by many years of uninterrupted development added to the massive consumption of energy and resources which has resulted in significant global risks such as global warming and ozone layer reduction. In 1972, therefore, the Conference of Stockholm was held, the first global United Nations conference on the environment. In it, it was accepted for the first time that it was not possible to maintain the growth achieved until then on a basis of unlimited exploitation of natural resources. At the same time, the United Nations Environment Program (UNEP) was created. This United Nations declaration was reaffirmed in 1992 with the celebration of the Earth Summit or Rio Conference, at which an attempt was made to adopt an approach to development that would ensure social and economic growth while protecting the
* Corresponding author. Tel.: þ34 985 18 21 06; fax: þ34 985 18 21 10. E-mail addresses: [email protected] (A. Gomez), [email protected] (M.A. Rodriguez). 0959-6526/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.jclepro.2011.01.012
environment. From that moment on, concern about environmental harm has grown more intensely (Nishitani, 2009). International regulations started to become important in successfully accessing markets, and, many companies have made environmental management a definitive priority. In response to the need to improve environmental quality (Bansal and Bogner, 2002), the International Organisation for Standardisation (ISO) in 1996 created a series of standards known as ISO 14000, comprising several advisory documents related to environmental management systems, environmental auditing, environmental performance evaluation, ecological labelling, lifecycle evaluation and product environmental aspects. This series was later revised in 2004, and although it is not the only way to efficiently implement an environmental management system, it does attempt to advise companies on developing better environmental policies. ISO 14001 is part of this series, and is the only standard specifically designed to meet the needs of an environmental management system. This standard has become required by customers and other interested parties, and as a result it has become a conditioning factor within the supply chain (Nawrocka and Parker, 2009). However, it is also true that there are many
1092
A. Gomez, M.A. Rodriguez / Journal of Cleaner Production 19 (2011) 1091e1095
companies still reluctant to adopt the standard, since they do not understand the benefits obtained from its application (Babakri et al., 2004). According to this standard, the environmental management system (EMS) offers the company a measurable result of its environmental performance, as it operates on a “Plan, Do, Check, Act” cycle basis. This means that it is put into practice as a continuous cycle of declaration, scheduling, execution, revision and improvement of the processes and actions carried out by the company (Perotto et al., 2008). In addition, industries are constantly facing environmental regulations that become ever stricter. For instance, in 1996 the EU adopted Directive 96/61/CE concerning integrated pollution prevention and control (IPPC Directive) by which industrial and agricultural facilities are required to obtain an environmental permit that guarantees compliance with a series of basic environmental requirements. This directive was incorporated into the Spanish legal system through the Law of Integrated Prevention and Control of Pollution, the aim of which is to prevent, and if that is impossible, to reduce and control pollution of the atmosphere, water and the soil by means of a system of integrated prevention and control. Whether due to compliance with current legislation or to market pressure, the business world’s commitment to the environment has become a variable of growing importance within an increasingly competitive market. Within this scenario, ISO 14001 offers various guidelines to design an EMS that enables the company to identify its own environmental policy (Murat, 2009), as well as to correct possible deficiencies. According to statistics published by the ISO (ISO, 2007), 154,571 certificates were issued during 2007 in 148 countries (29% in the services sector). This was an increase of 21% over 2006 when the total was 128,211 in 140 countries (27% in the services sector). Spain was ranked third with 13,825 certifications compared to 11,125 at the end of 2006. It is interesting to study the extent to which adopting the standard is related to an environmental transformation in productive operations and processes (González-Benito and González-Benito, 2008) and to technological modernisation. Several authors (Hart, 1995; Delmas and Toffel, 2003) have tried to formally study companies’ environmental strategic behaviour. They have observed that this behaviour describes a continuum that ranges from a passive or reactive strategy to the most advanced or proactive strategy. Companies showing a passive approach towards environmental management evaluate environmental aspects as cost factors, and therefore do not perceive the possible competitive advantages. Those with an active approach develop an environmental awareness only because of legislation. Proactive companies, on the other hand, weigh up environmental aspects in all decision-making processes, and see their obligation towards the environment as an opportunity to increase their share in markets for environmentally friendly products. If a reactive strategy is only intended to comply with the law, it will solve the environmental problem in the short run, but it will not be sufficient to improve the company’s competitive position. That is why, in response to the demands of their customers, employees, shareholders, the mass-media, etc. companies focus their actions on voluntary or proactive strategies for pollution prevention. In this sense, authors such as Porter (1991) and Porter and Van der Linde (1995) support the “Porter hypothesis”. This hypothesis supposes that in the relationship between proactive strategies and a company’s economy, it is possible to obtain a free-lunch, in which environmental quality targets are achieved. At the same time productive costs are reduced and productivity is improved. One usual practice within proactive strategies is to obtain ISO 14001 certification. It is in this that this paper finds its goal. The paper intends to investigate further the objective environmental consequences stemming from certification of a company’s EMS. Specifically, this
paper’s aim is to develop an empirical approach to the effect of ISO 14001 certification on the levels of environmental pollution, thus answering the question of whether this certification results in a reduction in the levels of polluting compounds released by the company or not. 2. State of the art Although nowadays the evident and costly impact industrial organisations exert on the environment is widely acknowledged, it is neither easy to find precedent papers that investigate the link between ISO 14001 certification and the company’s environmental performance nor papers that evaluate the time elapsed between the company’s certification and the moment it starts to reduce emissions. On the contrary, most papers tackle the ISO 14001 problem from the viewpoint of its proliferation in countries and industries (Delmas, 2000; Corbett and Russo; Kollman and Praskash, 2002; Corbett and Russo, 2001), its benefits (Tibor and Feldman, 1996), and the features and attributes of the organisations that tend to contribute to the improvement of the standard’s results. What is more, even today, ISO 14001 is still a widely challenged criterion, since it does not force the organisation to improve its environmental performance (Corbett and Kirsch, 2000). It can be seen that while some authors think that the adoption of the standard represents the possible obtaining of important advantages in environmental development, others argue that since it does not measure environmental development directly, it is not an appropriate tool for improving environmental sustainability. Few authors categorically conclude that ISO 14001 exerts a positive impact on pollution levels. Montabon et al. (2000) suggest that ISO 14001 may improve environmental performance as well as economic efficiency. Melnyk et al. (2002) also link the standard to a reduction in waste. Russo and Harrison (2001) have evaluated the impact of the standard on a sample of installations in the electronics sector and have concluded that the certification seems to lead to a significant decrease in the levels of toxic emissions. Whether this can be extrapolated to other industrial sectors has not been categorically answered. In another paper Russo (2007) points out that being one of the first installations to adopt the standard seems to be related to lower emission levels. Early certified installations achieved longlasting benefits. Additionally, the greater the experience the companies had with the standard, the lower the emissions. Being an early user of the standard and having greater experience are factors that together work to reduce emissions. Szymanski and Ikeda (2003) evaluated the time it takes reduction in emissions to take effect once certification has been achieved. They also analysed whether there were companies that tended to reduce their emissions faster. They found that more than half the companies analysed reduced their emissions during the first year but, since they did not use a control group, it is risky to state that the reduction was primarily due to certification. On the other hand, Andrews et al. (2003) found evidence that ISO 14001 does not affect the company’s environmental performance. King et al. (2005) focus their perspective on the reasons that cause a facility to seek certification. They found that ISO 14001 had a weak negative effect on the increase of the emissions. Since in their analysis few control variables were used, the authors themselves state that the study needed to be improved to yield more reliable results. 3. Sample selection The sample study was selected from companies belonging to different industrial sectors. They were all located in the regions of Asturias, Cantabria, Galicia and Castilla-León. It includes a sample
A. Gomez, M.A. Rodriguez / Journal of Cleaner Production 19 (2011) 1091e1095
made up of 126 companies, of which 56 have been ISO 14001 certified since 2001. The other 70 are not certified to the standard. These companies were taken from the Pollutant Release and Transfer Register (http://www.prtr-es.es/). This register was created by the Environment Ministry in order to make available information relating to the release of pollutant emissions into water, soil and the atmosphere. This register also made available waste transfers out of the locations of industrial complexes regulated by Law 16/2002 on the 1st of July, related to Integrated Pollution Prevention and Control (IPPC Law). This is a limited register since it only provides information from those industrial sectors regulated by the aforesaid law and also from large companies. The emission data is only stated if they are over the notification threshold established by Royal Decree 508/2007. These values are therefore not to be taken as limit-values. Data referring to atmospheric emissions was used in the analysis. The data was taken during the period 2001e2007, taking into account only installations with emission dispersal activity during the period. The data was measured for the amount of compounds released (kg/year) per industrial facility. Given the great amount of industrial agricultural and cattle facilities in the sample, these were removed from the study so as to avoid distortions in processing statistics. For the same reason, given the high toxic variability of the different compounds, the quantity of compound released was divided by its notification threshold. The amounts of all compounds released by each installation during one year were added together. In this way, a pollutant release index was obtained for every installation and year. Because of the bias of the data, extra finetuning was necessary; the value 1 was added to them and the result underwent a logarithmic transformation (Russo and Harrison, 2005; King and Lenox, 2000). Since the variable had been referenced making use of indexes, the mean values of the index were analysed per year. Toxics release index ¼ log[1 þ S(Ei/RQi)] Ei ¼ compound i emissions into air, water and soil. RQi ¼ notification-threshold values for compound i. i ¼ index representing every compound considered by the register. 4. Methodology
Fig. 1. Histogram for the certified companies sample.
distribution (they are even used when one of the groups has normal distribution and the other does not) and are related to ordinal or nominal quantitative data, high variances and small sample size (under 30 cases). Nevertheless, parametric tests can be applied to those cases that have a sample comprised of more than 30 cases, no matter the variable distributions. 4.1. Descriptive study of normality A descriptive study of normality was carried out by using SPSS 15.0 software in order to determine if the data in each sample came from a normal population. In a first verification, the histogram was obtained for each sample with unknown distribution. Then, the normal probability graphic was drawn from which the PeP and QeQ graphs were obtained. Since these methods are simply descriptive ones, it could happen that the data was not normal, although the verifications were reasonably well carried out. That is why it was necessary to continue the study with a statistical normality analysis by means of the KolmogoroveSmirnov test.
In order to carry out the analysis, two types of variables were considered: one quantitative, represented by the company’s toxics release index, the other of a categorical type, represented by the ISO 14001 certification. For the categorical variable a study group (certified companies) and a control group (non-certified companies) were defined. It was thereby possible to examine whether the fact of a company being certified according to the standard was somehow related to the degree of environmental pollution produced by the company. Since a different experimental situation is applied to each group, it is a question of a hypothesis contrast for two independent samples in which the following hypotheses were set out: H0: m certified companies ¼ m (m certified m non-certified ¼ 0) H1: m certified s m non-certified
non-certified
companies
In order to choose the appropriate type of contrast to apply, the kind of distribution in which the variables fit was examined, since, depending on this distribution, either parametric or non-parametric tests were to be used (Bartoszynski and NiewiadomskaBugaj, 2002). Parametric tests are best suited to the study of quantitative variables with normal distribution and equal variances. Non-parametric tests do not take into account the type of
1093
Fig. 2. Histogram for the non-certified companies sample.
1094
A. Gomez, M.A. Rodriguez / Journal of Cleaner Production 19 (2011) 1091e1095
Table 1 KolmogoroveSmirnov test for both groups.
Table 3 Descriptive statistics of the cases under study.
KolmogoroveSmirnov test for certified and non-certified installations
N
Certified companies emissions
Non-certified companies emissions
56
70
Normal parametersa,b
Mean Standard deviation
0.2452 0.17336
0.2121 0.18352
Most extreme differences
Absolute Positive Negative
0.129 0.129 0.112
0.169 0.169 0.160
Kolmogorove Smirnov Z
0.963
1.411
Asymptotic significance (bilateral)
0.312
0.037
a b
Index
CERTIF
N
Average range
Ranges total
Non-certified Certified Total
70 56 126
59.90 68.00
4193.00 3808.00
Once the fact that the “emissions” variable follows different distributions in both samples was verified, the study continued with a hypotheses contrast on the equality of means. Because of the number-of-cases criterion it would be suitable to perform a parametric contrast, but according to the type-of-distribution criterion, the most suitable would be a non-parametric contrast. Since they were both completely feasible, they both were performed so as to have them complement each another. 5.3. Parametric contrast
The contrast distribution is the normal. Worked out from the data.
In order to parametrically contrast the equality of means, a Student’s t-test was carried out under the following hypotheses:
4.2. Statistical study of normality To check the normality of the “pollutant index” variable in both samples, the following hypotheses were produced: H0: “pollutant index” variable is normally distributed H1: “pollutant index” variable is not normally distributed 5. Results and discussion 5.1. Descriptive study: histograms The histograms obtained for each sample are depicted and the results are presented in Figs. 1 and 2. As can be seen, they show curves that are not similar to a normal curve, which means it is not possible to state unequivocally that both samples come from normal populations. 5.2. Statistical study: KolmogoroveSmirnov test The results obtained by the KolmogoroveSmirnov test are shown in Table 1. The mean toxic release index for the certified group is 0.2452, with a standard deviation of 0.17336, while the mean toxic release index for the uncertified group is 0.2121, with a standard deviation of 0.18352. In this case, an asymptotic significance (bilateral) is obtained with a probability of 0.312 (p > 0.05) which leads us to reject hypothesis H1. It is not possible to reject the hypothesis that the “emissions” variable follows a normal distribution for any significance level lower than 31.2%. In short, in this case the variable index follows a normal distribution. On the other hand, in the group of non-certified facilities, the asymptotic significance has a probability of 0.037 (p < 0.05) which leads us to accept hypothesis H1. In this case, the variable does not follow a normal distribution.
Table 2 Summary of statistics for both groups (certified group and non-certified group). Group statistics
Index
Ranges
CERTIF
N
Mean
Standard deviation
Mean standard error
Certified Non-certified
56 70
0.2452 0.2121
0.17336 0.18352
0.02317 0.02193
H0: m certified companies ¼ m (m certified m non-certified ¼ 0) H1: m certified s m non-certified
non-certified
companies
Table 2 shows a brief summary of the cases processed by means of their descriptive statistics: N (sample size), mean, standard deviation and mean standard error. As can be seen in the following table, the variance homogeneity test or Levene’s test obtains a value “p ¼ 0.45” (p > 0.05) associated to the null hypothesis that the variances are homogeneous. Hence we assume that variances are homogeneous. Under this premise statistic t obtains a value of 1.029 (with 124 degrees of freedom) and the associated value “p” is 0.306 (p > 0.02). It can therefore be concluded that the differences between the emission levels of certified and non-certified installations are not significant (they are not statistically different for a significance level of 0.05 and as a consequence there is no link between the ISO 14001 certification and the pollutant emission index). The interpretation of the difference of means between both groups would set this difference within the population with high reliability, between 0.03051 and 0.09658, which allows us to understand whether the means are different between both groups. Since CI 95% contents zero, the non-difference of means in the “emissions” variable must be assumed within the population. 5.4. Non-parametric contrast The equality of means was contrasted from a non-parametric perspective by performing a ManneWhitney U test. The results lead to a conclusion that is similar to the one obtained in the previous comparison. First of all, a summary of the cases processed Table 4 Non-parametric contrast statistics. Contrast statisticsa Index ManneWhitney U Wilcoxon W Z Asymptotic significance (bilateral) a
Grouping variable: CERTIF.
1708.000 4193.000 1.238 0.216
A. Gomez, M.A. Rodriguez / Journal of Cleaner Production 19 (2011) 1091e1095
by means of their descriptive statistics is shown: sample size, mean, standard deviation and maximum and minimum values (Table 3). The information included inside the quantitative variable was processed to obtain different comparison statistics. The results are presented in Table 4. The p-value of 0.216 (p > 0.05) obtained for the asymptotic significance leads us to reject the alternative hypothesis “the mean of emissions is different in both groups”. Therefore, by accepting the null hypothesis we are concluding that there is no statistically significant association between the pollutant emission index and ISO 14001 certification. 6. Conclusions ISO 14001 certified and non-certified companies were analysed to investigate whether there is a relationship between certification and the toxics release rates of both groups. In the article, the Toxics Release Index of 56 certified companies are compared with that of 70 non-certified companies, with the help of the statistics program SPSS 15.0. Firstly, the normality of the Toxic Release Index was analyzed, with the conclusion that this index continues to follow the normal pattern of distribution in the case of certified companies, but in the case of non-certified companies this cannot be stated with assurance. To analyze the effect of certification, it was decided to compare the averages of both indexes with the help of a Student’s test and a ManneWhitney U test. Both tests concluded that these averages are the same, with a significance level of 0.05. The final conclusion is that there is no relationship that links the fact of a company being certified to its pollutant emission indexes. The empirical verification of the fact that the certification does not cause the pollutant emissions level to decrease significantly, suggests that the companies perceive the certification as a reactivetype investment rather than one that is proactive. In short, certification is considered to be a standard that must be adopted because of institutional pressure, but not as an investment that favours an improved environment. It could be said that the adoption of the standard is seen as a sign that tries to show the company as socially within the law without resulting in significant changes to the company’s management. In this sense, authors like Boiral (2001) warn that ISO 14001 may be used by companies as a marketing tool, something that is aggravated, according to Corbett and Kirsch (2000), by the fact that the certification postulates neither specific practices nor definite result requirements. Given the issue’s importance and topicality, research into environmental strategies adopted by companies and their responsibilities for environmental degradation needs to be more deeply investigated. It should nevertheless be noted that the conclusions obtained in this study are similar to the ones obtained by Barla (2007). He tried to verify whether the adoption of the standard produces a significant impact on the paper industry’s environmental performance. He found that certification does not lead to a reduction in suspended solid emissions. In his study, he concludes that facilities adopting the standard did not undergo any significant reduction in their emissions. The impact of the standard is highly variable: even though some facilities reduce their emissions after certification, the emissions levels of most facilities remain stable or even increase after certification. Further research is needed (according to production and studying the same industrial sector, for example) to determine whether potency-weighted air emissions alone are an
1095
adequate metric for determining whether certification results in reduced emissions. References Andrews, R.N.L., et al., 2003. Environmental management systems: do they improve performance? Final Report, 30 January 2003. Available from: http://ndems.cas. unc.edu (accessed September 2009). Babakri, K.A., et al., 2004. Recycling performance of firms before and after adoption of the ISO 14001 standard. Journal of Cleaner Production 12 (6), 633e637. Bansal, P., Bogner, W.C., 2002. Deciding on ISO 14001: economics, institutions, and context. Long Range Planning 35 (3), 269e290. Barla, P., 2007. ISO 14001 certification and environmental performance in Quebec’s pulp and paper industry. Journal of Environmental Economics and Management 53, 291e306. Bartoszynski, R., Niewiadomska-Bugaj, M., 2002. Probability and Statistical Inference. In: Wiley Series in Probability and Statistics. Wiley-Interscience. Boiral, O., 2001. ISO 14001 certification in multinational firms: the paradoxes of integration. Global Focus 13 (1), 79e94. Corbett, C.J., Russo, M., 2001. ISO 14001: irrelevant or invaluable, ISO Management Systems, (special issue: The impact of ISO 14001, December), pp. 23e9. Corbett C., Russo, M. ISO 14001: Irrelevant or Invaluable?, ISO management Systems. Corbett, C., Kirsch, D.A., 2000. ISO 14000: an agnostic’s report from the frontline. ISO 9000 þ IS0 14000 News March/April 2. Delmas, M., 2000. Barriers and Incentives to the Adoption of ISO 14001 in the United States. Duke Environmental Law and Policy Forum Fall. Delmas, M., Toffel, M.W., 2003. Institutional pressure and environmental management practices. In: Presented at the 11th International conference of the Greening of Industry Network. October, San Francisco. González-Benito, J., González-Benito, O., 2008. Operations management practices linked to the adoption of ISO 14001: an empirical analysis of Spanish manufacturers. International Journal of Production Economics 113, 60e73. Hart, S.L., 1995. A natural-resource-based view of the firm. Academy of Management Review 20 (4), 986e1014. ISO, 2007. The ISO Survey of Certification 2006. King, A., Lenox, M., 2000. Industry self-regulation without sanctions: the chemical industry’s Responsible Care Program. Academy of Management Journal 43, 698e716. King, A., et al., 2005. The strategic use of decentralized institutions: exploring certification with the ISO 14001 management standard. Academy of Management Journal 48, 1091e1106. Kollman, K., Praskash, A., 2002. EMS-based environmental regimes as club goods: examining variations in firm-level adoption of ISO 14001 and EMAS in UK, US, and Germany. Policy Sciences 35, 43e67. Melnyk, S.A., et al., 2002. Assessing the effectiveness of US voluntary environmental programmes: an empirical study. International Journal of Production Research 40, 1853e1878. Montabon, F., et al., Spring 2000. ISO 14000: assessing its perceived impact on corporate performance. The Journal of Supply Chain Management (4), 4e15. Murat, A., 2009. The benefits associated with ISO 14001 certification for construction firms: Turkish case. Journal of Cleaner Production 17, 559e569. Nawrocka, D., Parker, T., 2009. Finding the connection: environmental management systems and environmental performance. Journal of Cleaner Production 17, 601e607. Nishitani, K., 2009. An empirical study of the initial adoption of ISO 14001 in Japanese manufacturing firms. Ecological Economics 68, 669e679. Perotto, E., et al., 2008. Environmental performance, indicators and measurement uncertainty in EMS context: a case study. Journal of Cleaner Production 16, 517e530. Porter, M., 1991. America’s green strategy. Scientific American 264 (4), 168. Porter, M., Van der Linde, C., 1995. Green and competitive: ending the stalemate. Harvard Business Review 73 (5), 120e134. Russo, M.V., 2007. Explaining the impact of ISO 14001 on emission performance: a dynamic capabilities perspective on process and learning. Business Strategy and the Environment 18 (5), 307e319. Russo, M.V., Harrison, A., 2001Paper presented at the Ninth Greening of Industry Network Annual Conference, Bangkok, Thailand. An empirical study of the impact of ISO 14001 registration on emissions performance. Russo, M.V., Harrison, N.S., 2005. Organizational design and environmental performance: clues from the electronic industry. Academy of Management Journal 48, 582e593. Szymanski, M., Ikeda, S., 2003. ISO 14001 and reduction of toxic emissions. In: Proceedings of EcoDesign 2003: Third International Symposium on Environmentally Conscious Design and Inverse Manufacturing Tokyo, Japan. Tibor, T., Feldman, I., 1996. ISO 14000: A Guide to the New Environmental Management Standards. Irwin Professional Publishing, Chicago, Illinois.