Environ Dev Sustain (2011) 13:845–862 DOI 10.1007/s10668-011-9294-8
Do environmental management systems improve environmental performance? Empirical evidence from Italian companies T. Daddi • M. Magistrelli • M. Frey • F. Iraldo
Received: 24 October 2010 / Accepted: 8 March 2011 / Published online: 27 March 2011 Springer Science+Business Media B.V. 2011
Abstract This paper reports the results of a research carried out on environmental performance in companies with an environmental management system registered under the EMAS regulation. By applying an empirical approach, the article analyses the trend of the environmental performance of a sample of 64 Italian companies pertaining to 6 different industrial sectors, which have achieved EMAS registration for at least three years. The trends of the environmental indicators included in the environmental statements, officially validated by a third party and traceable to a period of time before and after the registration, were collected and evaluated. In particular, the influence of the EMAS registration on the improvement or the worsening of the environmental performance was analyzed and thus the ability of this instrument to pursue continuous improvement, a founding principle of environmental management certification schemes. Data obtained indicate that in some cases, EMAS leads to an actual improvement of environmental performance, even in the short period. Keywords Environmental performance EMAS Environmental statement Principle of continuous improvement
1 Introduction In recent years, the issue of ‘‘continuous improvement,’’ a founding principle of voluntary certification schemes (such as EMAS and ISO 14001), has been the subject of strong Readers should send their comments on this paper to
[email protected] within 3 months of publication of this issue. T. Daddi (&) M. Magistrelli M. Frey F. Iraldo Sant’Anna School of Advanced Studies, Piazza Martiri della Liberta` 33, 56127 Pisa, Italy e-mail:
[email protected] M. Frey F. Iraldo IEFE—Institute for Environmental and Energy Policy and Economics, Bocconi University, via Guglielmo Roentgen, 1, 20136 Milan, Italy
123
846
T. Daddi et al.
debate among scholars in the area of environmental management systems. How much this principle is then translated into real improvements in performance of organizations that have adopted an environmental management system has to be empirically assessed, in order to confirm the effectiveness and validity of the environmental voluntary instruments adopted by policy makers, who devote increasing financial support to these policies (Arimura et al. 2008). EMAS-registered companies are suited to carry out research related to the analysis of environmental performance, as the regulation requires the annual publication of the environmental statement, a record which shows the main data and environmental indicators of the organization, data verified and validated by an independent third party. This paper reports the results of a study based on the analysis of data in these records, in a sample of 64 Italian industrial companies pertaining to 6 different productive sectors, in order to verify the trend of their performance by comparing it during the periods before and after EMAS registration. Although it is based on empirical evidence, it must be stressed that the ‘‘benefits’’ of environmental management systems should not be restricted to the environmental performance of the organization that adopts them (Biondi et al. 2000). Along with tangible benefits, such as improved environmental performance, in evaluating the effectiveness of the instrument, those intangible benefits which are difficult to assess should also be taken into account. For example, the entrepreneurs’ increased environmental awareness, the employees’ participation, and involvement in effective environmental management practices, the greater cooperation with those agencies in charge of permits and controls, the increased ability to handle environmental emergencies, the increased ability to manage compliance with environmental regulations etc. (Steger 2000; Jaeger 1998and Khalid et al. 2004), all of which indirectly contribute to environmental improvement in a wider sense.
2 EMS and environmental performance The difficulty in assessing the link between EMAS and environmental performance stems from a number of methodological challenges (Dahlstro¨m et al. 2003). Firstly, performance improvement can be operationalized in very different ways, for example as absolute reductions of emissions or improved eco-efficiency; as short-term or long-term improvement; an upwards performance trend or one which is better than that of similar organizations; and so on. In practice, organizations will usually see improvement on some indicators and worsening on others. Secondly, the environmental performance of companies can be characterized by a strong inherent variability, e.g., due to short- and medium-term changes in capacity utilization, raw material prices, product characteristics (Welch et al. 2002). Shifts in performance may also be the outcome of larger business decisions (e.g., outsourcing or relocation of resource-intensive production steps, plant modernization) or external pressures (e.g., environmental legislation, media reporting) (Al-Tuwaijri et al. 2004). Finally, most studies relied on perceptions by environmental managers on the ability of EMAS or ISO 14001 to improve environmental performance (and not on primary data referred to environmental parameters, such as emissions or resource consumption). Moreover, although environmental statements provide reliable quantitative data on performance of EMAS organizations, there are a number of problems with the availability and comparability of those data: lack of harmonization (indicators, measurement units), different reporting levels (process, site, firm, group), lack of time series data, insufficient information on products, processes, and output, etc.
123
Do environmental management systems improve environmental performance?
847
These difficulties, and the range of approaches to address them, explain why studies have come to different results. Although the literature does not provide a simple answer to the question of EMAS effectiveness, a number of recent studies have produced interesting insights. However, some studies (e.g., UNI/ASU 1997) adopt a generally more optimistic tone than others (e.g., FEU 1998; Wagner 2002). In general, surveys found that a majority of respondents reported a moderate level of environmental effectiveness stemming from EMS adoption, although a considerable variability between companies was also observed (Steinle and Baumast 1997; Kuisma et al. 2001; Andrews et al. 2003). The Research Centre on European Environmental Law (FEU) found that EMAS implementation brings an improvement in regulatory compliance (FEU 1998) in their survey of 27 German EMAS-registered companies and analysis of 200 environmental statements. A survey of German EMAS companies found that the adoption of the management system has had a positive impact in a range of areas (especially waste generation, resource use and water consumption), but was unable to quantify the magnitude of improvement (UNI/ASU 1997); using the same list of environmental aspects in a survey of French EMAS sites, Schucht (2000) obtained similar results (reduction of liquid effluents and water pollution is reported as another important effect in the French case). Steger (2000) concluded that EMSs (including EMAS) support legal compliance, but pointed out that it is difficult to determine the actual environmental effects of better compliance because non-compliance is often concerned with formal infringements rather than material breaches. He found, however, little evidence that EMS is a strong autonomous driver for performance improvement. In their econometric analysis of the data obtained through a survey of 2000 European companies, Johnstone et al. (2004) found that EMSs (including both ISO 14001 and EMAS) played ‘‘a distinct role in encouraging firms to undertake measures to improve their environmental performance in a number of areas.’’ The impact of EMS was particularly important in the generation of waste water and air emissions and in the reduction of environmental impacts from accidents. Robust quantitative evidence about the longer-term impact of EMAS on performance does not yet exist because of a shortage of time series data. However, researchers have tried to assess whether EMAS helps promote environmental innovations. Rennings et al. (2003) found in a survey of 1277 EMAS-certified German facilities and 12 in-depth case studies that environmental managers consider the implementation of EMAS as a substantial contribution to the introduction of environmental innovations, especially organizational ones. In a subsequent survey of production managers in 588 German facilities, this time evaluating different Integrated Product Policy initiatives, Rennings et al. (2004) found a weak but significant positive influence of ISO 14001 and EMAS on environmental product innovations. Hertin et al. (2004) performed regressions and times series analysis on European industrial companies and production sites with different EMS policies. Their main finding was that the link between a company’s EMS and environmental performance (measured with eco-efficiency indicators) is weak and ambiguous: companies with a formal EMS performed better on a number of indicators, but worse on several others, and only a small number of correlations were statistically significant. Hertin et al. (2008) reporting the outputs of a research project named MEPI based on data of 274 companies and 400 production sites in six manufacturing sectors in six EU countries confirm the weak link between EMS and environmental performances. Analyzing a sample of 306 German manufacturing firms, Wagner (2002) found no significant differences in energy efficiency between firms with and without EMS (EMAS and ISO 14001), neither for the year 2001 nor for the period 1991–2001.
123
848
T. Daddi et al.
Barla (2007) highlighted how the plants certified ISO14001 improve their biological oxygen demand of water releases while not improve other parameters of the same environmental aspect, and Anton et al. (2004) also found that ‘‘the adoption of a more comprehensive EMS has a significant impact in terms of reduction in the intensity of toxic releases’’ and pointed out that the importance of these measures tends to especially visible on companies with initially poor environmental records. Researchers of the University of North Carolina concluded from a survey of US facilities that have implemented EMSs that ‘‘results suggest strongly that the introduction of an EMS does make an observable difference to a facility’s environmental performance’’ (NDEMS 2000); Potoski and Prakash (2005) also found some evidence that ISO 140001 certified industrial companies in the US reduced toxic emissions faster than non-certified facilities. Finally, a recent study carried out by Iraldo et al. (2009), based on a sample of 100 organizations interviewed within the scope of the EU project EVER (Evaluation of EMAS for its Revision), investigated whether or not an EMS implemented within the EMAS regulation has an effect on firms’ performance. The econometric analysis shows a positive impact of well-designed environmental management system on environmental performance and as a consequence on technical and organizational innovations.
3 Hypothesis and methods The goal of this research is to investigate whether the companies that have achieved EMAS registration have consequently improved their environmental performance, already in the short term, by implementing the environmental management system. In order to evaluate whether EMAS registration leads to actual improvements in environmental performance, we made use of data reported by registered companies on their EMAS environmental statement, the document periodically published and updated and available to the public. In particular, we compared some significant environmental indicators, regarding water consumption, waste production, and energy consumption, in three different periods of time: 2 years before the registration, the year of registration, and 2 years later. The selection of the four indicators depended on the availability and comparability of the data included in the environmental statements selected. In fact, in many cases, companies operating in the same sector use different kinds of indicators, which do not allow a direct and meaningful comparison. With the publication of the new EMAS Regulation (Reg. CE 1221/2009—so-called EMAS III), this problem should be overcome thanks to the fixation of some homogeneous key performance indicators (KPIs) that every organization will have to elaborate and report in the statement, regardless of its specific sector or branch. The research method we used presents both advantages and disadvantages. The main operational problems are the collection of environmental statements from organizations, especially of the oldest ones, and the full availability of environmental data in these publications. On the other side, indicators based on objective data enabled us to assess the real contribution of EMAS registration to environmental improvement, without resorting to subjective measures built through questionnaires or interviews. We used a sample of Italian EMAS-registered organizations, chosen among the firms that achieved the first three-year-renewal of EMAS registration within March 2008. The restriction was motivated because this method requires some data concerning the second year after registration, and these pieces of information are published in the environmental statement only after the first renewal. After this selection, and considering deleted
123
Do environmental management systems improve environmental performance?
849
companies and voluntary-missed renewals, the target population of the Italian EMAS Register fell to 219 organizations. We classified these organizations according to their NACE code, and we found 33 categories. We considered only categories that contained at least ten organizations, in order to guarantee significant industry-related findings. We also excluded organizations with NACE code 84.11 and 84.12 (Public administration), because most of their contribution to environmental improvement is related to indirect aspects, rather than to direct performance in resource consumption or pollution emission. After this second filter, the population involved in the research reduced to 156 firms, belonging to six industrial sectors (Table 1). The study population consisted in 156 registered organizations. To determine the sampling error, a level of reliability or probability of 95.5% was considered, in accordance with the normal law of probability. The sample size was related to sample reliability, to the margin of error in calculation of answers, and to parameters p and pq, as follows: n¼
Z 2 pð1 pÞ e2
where it results that e¼
Z ðpð1 pÞÞ1=2 n1=2
being n¼
nm 1 nm =N
where nm is the sample size (number of environmental statements analyzed: 64), N the population size (156), Z the number of standard deviations above and below the average (1.96), p the distribution parameter (considered as p = 0.5, maximum variance), and e is the error. In accordance with this expression, a sampling error of 2.55% was obtained for a level of sample reliability of 95.5%. This error was considered acceptable. For each organization included in the sample, we collected and analyzed the following figures from their validated Statements: amount of production (or raw material processed) expressed in an appropriate unit of measurement (depending on the sector), water consumption expressed in m3, electric energy consumption expressed in kWh, total energy
Table 1 Population and composition of the survey sample NACE code and industrial sector
EMAS population
Sample
% of the population
10—Manufacture of food products and beverages
49
19
31.41
20—Manufacture of chemicals and chemical products
24
8
15.38
23—Manufacture of non-metallic mineral products
12
7
7.69
35—Production and distribution of electricity, gas, and steam
29
10
18.59
38.1, 38.2—Collection, treatment, and disposal of waste
26
13
16.67
38.3 Recovery of materials
16
7
10.26
156
64
Total
123
850
T. Daddi et al.
consumption expressed in TOE (Tons of Oil Equivalent), and waste generation expressed in kg. Unfortunately, other important measures about further environmental aspects, like emissions to air and releases to water, had been discarded due to lack of data. In the following step, for each company, we generated four time series of indicators per unit of production, with the aim of allowing comparison along time. It is interesting to note that the publications in environmental statements of three of the four indicators we used (Water consumption, Electric energy consumption, Waste production) are now compulsory according to the new EMAS III Regulation (see Annex IV, Reg. CE 1221/2009). Since the goal of our analysis was to understand the contribution of EMAS registration to environmental improvement, we elaborated a measure that made it possible to perform aggregations and comparisons between firms. Then, we calculated the percentage variation (i.e., improvement or worsening of a given KPI) for each indicator between the first time (two years before EMAS registration) and the second one (the year of EMAS achievement), between the second and the third (2 years after EMAS registration), and between the first and the third (i.e., the whole period considered) (Table 2). At the end of data processing, we had a database composed of 12 percentage variations (3 variations times 4 environmental indicators) for each of the 64 organizations of our sample.
4 Results and discussion Based on the methodology and sample presented earlier, the data obtained from the environmental statements were processed and analyzed in order to highlight trends in the reporting periods. To this end, the percentage variations between the environmental indicators of the various years under consideration were classified according to five categories depending on the percentage difference in the reference period (Table 3). Through these distinctions (details of which are presented in the tables below), the companies that have achieved the best performance were identified. 4.1 Overall results Some interesting results refer to the trends of the environmental indicators of the whole sample in the pre-registration period, emerging from the analysis of the performance of the
Table 2 Environmental indicators considered and reference periods for trend analysis Environmental indicator
Pre-registration Post-registration Aggregate trend (from N - 2 to Na) (from N to N ? 2) (from N - 2 to N ? 2)
Water consumption (m3/total production)
±…%
±…%
±…%
Electric energy consumption (kWh/total production)
±…%
±…%
±…%
Total energy consumption (tons of oil equivalent/total production)
±…%
±…%
±…%
Waste production (kg/total production)
±…%
±…%
±…%
a
N, year of achieving of EMAS registration
123
Do environmental management systems improve environmental performance? Table 3 Classification of the different percentages herein analyzed
High worsening
851
Environmental indicator increase more than 50%
Worsening
From ?50% to ?2.5%
Stable
From ?2.5% to -2.5%
Improvement
From -2.5% to -25%
High improvement
Less than -25%
Not available
Not available data
Table 4 Pre-registration trend (from N - 2 to N) Water consumption High worsening Worsening Stable Improvement
Electric energy
Total Energy
Waste
Number
%
6
10
5
15
36
14
19
18
24
19
80
31
9
9
10
4
32
13
14
20
22
17
73
28
High improvement
9
6
3
5
23
9
Not available
7
1
0
4
12
5
values recorded 2 years before the EMAS registration (N - 2) until the year in which this goal was achieved (N) (Table 4). Most of the companies during the period prior to the EMAS registration recorded a trend of worsening in performance. The production of waste is the environmental aspect that worsens in the majority of cases, with a worsening in 34 companies. Few companies in the sample show a remarkable improvement (‘‘high improvement class’’), although nearly 30% had an improvement of indicators from 2.5–25% (‘‘improvement’’ class). Table 5 tries to capture the effect that the introduction of the registered EMAS environmental management system has on the analyzed environmental indicators in the short term. Generally speaking, we can say that the EMAS registration has produced an improvement of environmental performance indicators in more companies than the previous period. Considering both the percentage of ‘‘improvement’’ and ‘‘high improvement’’ classes, the increase is about 45%. On the other hand, we can easily note that the total number of EMAS companies that are worsening their performance is not significantly different from the period before the registration. Moreover, the situation 2 years after obtaining the registration is perfectly symmetrical between the cases of worsening and improvement. Although the period may be considered a little too short to assess the effects Table 5 Trend indicators over the period of post-registration (from N to N ? 2) Water consumption High worsening Worsening Stable
Electric energy
Total energy
Waste
Number
%
9
9
3
13
34
13
14
22
24
21
81
32
7
7
8
3
25
10
25
20
23
13
81
31
High improvement
8
5
6
14
33
13
Not available
1
1
0
0
2
1
Improvement
123
852
T. Daddi et al.
Table 6 Aggregate trend (from N - 2 to N ? 2) Water consumption
Electric energy
High worsening
10
10
7
19
46
18
Worsening
14
26
22
19
81
32
Stable
Total Energy
Waste
Number
%
2
5
7
4
18
7
Improvement
20
18
23
7
68
26
High improvement
11
4
5
11
31
12
7
1
0
4
12
5
Not available
of EMAS on performance, Table 5 shows that the introduction of the environmental management system has in some cases improved the performance of indicators, but it apparently rejects the hypothesis that continuous improvement is pursued by companies more effectively or intensively, once they become EMAS registered. This indication is confirmed by Table 6, which shows the comprehensive results for the entire period. The analysis of the whole data from the two periods under study (before and after registration) shows that the higher number of environmental indicators that improved in the 2 years’ post-registration was not significant in absolute terms, since they were not able to offset the worsening occurred in the previous 2 years. In other words, 2 years after EMAS registration, the environmental performance indicators concerning the four aspects analyzed have worse values than the 2 years before the adoption of EMAS in 50% of cases, while performance has improved in only 39% of the analyzed cases. This general result is quite meaningful, since it is not influenced by ‘‘external’’ contingencies, such as the year in which the sampled companies obtained registration, the regulation according to which registration has been released (EMAS I or EMAS II), or the accredited verifier that validated the statement. Nevertheless, we cannot take this general result as the only and undisputable outcome of the analysis, since the overall indicators on the whole sample are not able to capture some impact- and sector-specific effects of EMAS Registration, which are essential to properly judge the effectiveness of this management (and policy) tool. 4.2 Results by environmental aspect This section reports the data relating to the trend of the indicators broken down by environmental aspect. The table below reports the results for ‘‘water consumption,’’ showing how the values for this indicator varied in the analyzed period (please taking into account that in the figures the values are rounded up or down and this could cause in some cases the sum of percentages achieve the rate of 101% instead of 100%) (Fig. 1). As stated in the previous paragraph, the time period which is more useful to appreciate the impact of EMAS on environmental performance is the one from the date of the registration (N) to the two following years (N ? 2), which is the only period that is entirely covered by an effective environmental management system, fully ‘‘in place’’ to pursue continuous improvement objectives. During this period, there is an actual improvement of the indicators, with an increase in the ‘‘improvement’’ class from 22 to 39% over the previous period. Adding up to this data, the ‘‘high improvement’’ class, the percentage goes from 36 to 52% in the same period. In this case, therefore, one can say that the number of
123
Do environmental management systems improve environmental performance?
853
Fig. 1 Evolution of environmental indicators for ‘‘water consumption’’
companies that started to improve their performance, thanks to the EMAS registration, is considerable. During this period, however, there is also an increase in the ‘‘high worsening’’ class that raises by about 50% of the value recorded in the pre-registration period. A possible justification for the worsening can be seen in the fact that during its implementation, EMAS involves a thorough verification of the legal compliance of the production process and of the site in which the company is located. The need to start monitoring water consumption in many of the sampled companies has been caused by the EMAS process itself. As it is clear by the reading of the validated statement, before the final verification, these companies did not have all the sources of water abstraction (e.g., wells) fully authorized, or they were not subject to regular monitoring of drained water levels, so that ‘‘water consumption’’ showed an increase in the indicators as a consequence of the need to monitor a previously unauthorized water source and of the connected availability of new data on the abstracted water (i.e., of the full implementation of the environmental management system). The two following tables report the data on electricity consumption and the overall consumption of energy (Figs. 2, 3).
Fig. 2 Evolution of environmental indicators for ‘‘consumption of electricity’’
123
854
T. Daddi et al.
Fig. 3 Evolution of environmental indicators for ‘‘overall consumption of energy’’ expressed in TOE (Tons of Oil Equivalent)
The trend of the indicators of electricity and of the overall energy consumption shows an essentially constant flow, demonstrating how EMAS registration did not affect this environmental aspect in the short term. One slight improvement is evident in the ‘‘high improvement’’ class, and it refers to total consumption of energy that goes from 5% in the pre-registration period to 9% in the post-registration period. The insufficient influence of the environmental management system on this aspect is due to some reasons. First, energy is not only a relevant environmental aspect in running a business but it is also strongly related to production costs, which are crucial for the economic and financial performance of a company. Therefore, as it is reported in many of the sampled environmental statements, the EMAS, companies had in the past, already adopted those technical and management measures that reduce energy consumption before being registered (thus reducing operating costs consequently). The second reason is due to the Italian environmental legislation. In this context, there is a legislative decree that provides for the appointment of a figure called ‘‘Energy Manager’’ for companies that exceed 10.000 TOE of energy consumption, with the specific task of taking action and implementing interventions to reduce energy consumption and planning targets for improving this aspect, whether an EMS is adopted or not. Last but no least, the analyzed environmental statements report that the energy-related measures adopted by the sampled companies in the most recent years have been mostly focused on self-production and use of renewables, the increase of which does not affect the two indicators selected for our analysis. For the reasons set out above, it often happens that, especially in large companies, the energy aspect shows little improvement with the introduction of EMAS. The picture below shows the results of the trend of indicators for ‘‘waste’’ (Fig. 4). The indicator concerning waste production shows a peculiar trend. In the time period prior to EMAS registration, the majority of the indicators report a negative trend. During the first 2 years of EMAS registration, a large number of indicators are still worsening, although the ‘‘high worsening’’ class reduces its percentage. During the same period, there is a sharp increase in the number of indicators that show a considerable improvement, going from 8 to 22% of the sample. We have contrasting data: on the one hand, in the period from N to N ? 2, there is an improvement of some indicators that can be attributed to EMAS registration, on the other hand, the ‘‘high worsening’’ value is constantly over 20% and seems to suggest that there is no evident benefit of the registration on this environmental aspect. Moreover, in the whole period from N - 2 to N ? 2, it appears that
123
Do environmental management systems improve environmental performance?
855
Fig. 4 Evolution of environmental indicators for ‘‘waste’’
the improvement due to EMAS registration fails to offset the substantial worsening occurred in the previous period. 4.3 Result by industry: the food sector An analysis of the data broken down by sector enables to understand whether the impact of EMAS registration on the companies’ ability to improve environmental performance is to any extent correlated with the specificities of the production cycle and the (average) dimension of the companies. In the following, we present some insights into two of the sectors involved in the study. The sector with identification code NACE 10 (Manufacture of food products and beverages) is represented by 19 companies and is the most numerous of the sample (as it represents the majority of the population covered by the study: 49 registered organizations out of 156). They are mostly small businesses, and more in detail, many of them are dairies located in Emilia–Romagna, which participated to some of the early publicly funded pilot projects on the EMAS registration carried out in Italy. In Table 7, the four investigated environmental aspects have been aggregated by combining the corresponding indicators with each of the 19 companies considered (this is the reason because we have a total of 76 cases for each period equal to the sample of 19 observed for the 4 indicators considered).
Table 7 Overall trends of the indicators for the food industry for all environmental aspects Pre-registration trend (from N - 2 to N) High worsening Worsening Stable Improvement
Post-registration trend (from N to N ? 2)
Aggregate trend (from N - 2 to N ? 2)
9
12%
7
9%
15
20%
31
41%
31
41%
32
42%
8
11%
10
13%
5
7%
22
29%
25
33%
17
22%
High improvement
4
5%
3
4%
5
7%
Not available
2
3%
0
0%
2
3%
123
856
T. Daddi et al.
The majority of the indicators analyzed presented a worsening both in the two subperiods and in the whole period considered. Yet, even in this very negative case, by observing the change from pre-registration to post-registration, we can notice a slight improvement. In particular, there is a decrease in the percentage of the ‘‘high worsening’’ class and an increase in the ‘‘improvement’’ class by four percentage points. Looking at some of the environmental issues, we see how water consumption, one of the most important environmental aspects of the food-processing industry, shows a marked improvement in the post-registration period (Fig. 5). Comparing the trend of the indicators between periods N - 2/N and N/N ? 2, we notice that the sum of the 2 classes indicating the worsening vary from 53 to 26%, while the classes indicating the improvement shift from 21 to 58%. This is a clear signal of the need to focus the planned improvement actions on the most significant environmental aspects, as it is explicitly requested by EMAS. The same cannot be said for the indicator that shows the trend of total energy consumption of food businesses in the sample (Fig. 6). In this case, the effect of the adoption of EMAS is anomalous. In the 2 years immediately following the registration, the indicators related to energy consumption worsen. Some of the consideration previously made is valid: energy-related investments were carried out by many sampled companies long before (and independently from) registration; therefore, in some cases, there was less room for improvement when the environmental objectives and targets had to be set within the EMAS process. Moreover, energy is a peculiar cost for companies because, as practitioners are well aware of, it does not proportionally decrease in cases of production downturns. Consequently, fixed consumption shares (e.g., for heating or cooling or to maintain plant efficiency) usually have a strong negative impact on an environmental indicator (i.e., the ratio between energy consumption and value of production) when there is a decrease in production. This happened for some of the food companies analyzed in our work. Although not so marked as in the case of water consumption, even in the case of waste generation (shown by Fig. 7), by comparing the two periods—before and after registration—we notice a decrease in the ‘‘high worsening’’ class and a proportional increase in the ‘‘high improvement’’ class. But we also have to emphasize that the cumulative datum referring to the two ‘‘worsening’’ classes is higher in the second period than in the first one.
Fig. 5 Trend of the environmental indicator related to water consumption in the food sector
123
Do environmental management systems improve environmental performance?
857
Fig. 6 Trend of the indicator for ‘‘overall consumption of energy’’ expressed in TOE (tons of oil equivalent) in the food sector
Fig. 7 Trend of the environmental indicator related to the production of waste in the food sector
The production of waste is certainly an area where, in the very short term, the adoption of an environmental management system can have both a positive effect (e.g., by immediately promoting and organizing the reuse and recycling of waste materials within the company) and a negative effect on an environmental indicator. The last effect is paradoxical, but it has been often detected in the EMAS-validated statements of our sample: the waste-related performance registered by the indicator worsened because, thanks to the implementation of a management systems, the company increased its capacity to identify, classify, and manage waste materials and products (therefore, the indicator inevitably shows a greater quantity of ‘‘waste produced’’). 4.4 Results by productive sector: the chemical industry In the chemical industry (NACE code No 20—Manufacture of Chemicals and Chemical Products), 8 environmental declarations out of a population of 24 were analyzed. This subsample is characterized by medium to large companies, with highly significant environmental aspects (Table 8).
123
858
T. Daddi et al.
Table 8 Overall trends of the indicators for the chemical industry for all environmental aspects Pre-registration trend (from N - 2 to N)
Post-registration trend (from N to N ? 2)
Aggregate Trend (from N - 2 to N ? 2)
High worsening
3
9%
1
3%
3
9%
Worsening
9
28%
8
25%
7
22%
Stable
6
19%
2
6%
2
6%
10
31%
16
50%
15
47%
High improvement
4
13%
5
16%
5
16%
Not available
0
0%
0
0%
0
0%
Improvement
In the chemical industry, the improvement following EMAS registration is particularly notable. The sum of the indicators that shows an improvement (in the both ‘‘improvement’’ and ‘‘high improvement’’ classes) goes from 44 to 66% considering the periods N - 2 to N and N to N ? 2. Similarly, there is a decrease in the indicators of the ‘‘high worsening’’ and ‘‘worsening’’ classes from 37%, in the period prior to registration, to 28% 2 years after obtaining EMAS. Even analyzing the classes individually, we notice an increase in the percentage of the ‘‘high improvement’’ class and a greater decrease in the ‘‘high worsening’’ class. Data for the water consumption indicator are fully consistent with the effect emphasized by the general indicators for this sector (Fig. 8). Despite the fact that chemical companies, since the early 70’s, have been among the pioneers of environment-oriented technological innovation, when in later years, they approach EMAS registration they kept on investing on environmental improvements. In the years following the EMAS registration, in fact, we reckoned a general increase in the improvement of the indicators in the sample, also associated with a reduction in the worsening companies. Our empirical evidence points out an improving trend also on energy consumption, although in a less marked way than in water use (Fig. 9). As in all the other investigated sectors, for the chemical sector as well, the impact of EMAS on total energy consumption shows only a slight improvement, due to the dynamics already described earlier. It has to be noted that many chemical companies of the sample
Fig. 8 Trend of the environmental indicator of water consumption in the chemical sector
123
Do environmental management systems improve environmental performance?
859
Fig. 9 Trend of the indicator for ‘‘overall consumption of energy’’ expressed in TOE (tons of oil equivalent) in the chemical sector
Fig. 10 Trend of the environmental indicator related to the production of waste in the chemical sector
were able to slightly improve energy consumption even after EMAS registration, but this is mostly the outcome of a more effective energy management, rather than of relevant technological innovation or structural intervention on the plant (Fig. 10). Finally, in the case of the waste production indicator, the results show only a relative improvement in the period following EMAS registration. Even if the number of companies that are able to improve their performance does not increase after EMAS registration, there is anyway a reduction in the ‘‘high worsening’’ class with respect to ‘‘worsening’’ and an increase in the number of indicators that shows a ‘‘high improvement’’ with respect to those that report only an ‘‘improvement.’’
5 Conclusions The outcome of our analysis should be assessed according to the empirical evidence on which it is grounded. First of all, we emphasize that the indicators, being based on the data contained in the environmental statements of EMAS-registered companies (and therefore on information that has been validated by an accredited third-party verifier), should be considered as particularly reliable.
123
860
T. Daddi et al.
Moreover, the sample of EMAS statements has been selected so to guarantee representativeness and avoid biases that may be linked to the year in which EMAS registration has been achieved, the regulation taken as a reference or to the approach followed by the accredited verifier. Nevertheless, the reader should bear in mind that our point of view is just one of the angles from which EMAS influence on performance can be assessed. As demonstrated by literature, in fact, there can be positive effects of EMAS beyond the direct influence on environmental performance (e.g., better legal compliance, improved management capabilities, and so on). Actually, in many cases achieving a higher level of monitoring, control and management of an environmental aspect have the unexpected consequence of worsening the corresponding indicator (e.g., previously unconsidered waste categories or nonmonitored water sources). Having said that, we can draw some general conclusions from the results of our analysis. The trend in the reported indicators shows that the application of EMAS in most cases leads to relatively positive effects on performance concerning the investigated environmental aspects. Comparing the results, validated in the environmental statements, for the pre-registration period (from N - 2 to N) with those for the post-registration period (from N to N ? 2), it generally emerges that in most cases, EMAS increases the (prob)ability for the registered company to obtain improvements (or high improvements) and/or it lessens the probability to and the risk of worsen (more or less significantly) the environmental performance. This holds true also for most of the breakdown analyses by sector and by environmental aspect, carried out within our study. Among the aspects on which we focused, water consumption and waste production appear to be more positively affected by the introduction of an environmental management system. There are relevant exceptions, instead, concerning energy consumption. Despite the overall (and relatively) positive effect of EMAS on environmental performance detected for the companies in our sample, some less optimistic considerations should be proposed. First of all, often the absolute value for the indicators shows that worsening (usually in the pre-registration 2-year period) is greater than improving (post-registration), thus preventing an ‘‘offset’’ of the negative environmental performance within 2 years from registration. Second, we should acknowledge that for most companies, EMAS registration does not represent a ‘‘turning point’’ neither in their environmental strategies nor (consequently) in their performance. In many EMAS statements analyzed in our sample, there is a sort of ‘‘continuity’’ (be it a positive or a negative trend) from the pre-registration to the postregistration period. The strongest (and most beneficial) effects of EMAS can be seen in terms of shifting from ‘‘improvement’’ to ‘‘high improvement’’ or from ‘‘high worsening’’ to just ‘‘worsening’’ or to a ‘‘stable’’ performance. This proves that EMAS is not a revolutionary tool for most companies, but a ‘‘step’’ of a coherent process of gradual improvement in environmental management. Moreover, we should highlight the particular economic period of the research. In some of analyzed cases, the economic downturn could have made worse the environmental indicators. Third, we should conclude with a positive note. The largest part of the literature on EMAS supports the idea that the introduction of an environmental management system cannot have effects in the short term; on the opposite, a company should wait till a medium-long run to perceive the first significant improvements. This hypothesis is not confirmed by the Italian case study described in this article, where the concept of
123
Do environmental management systems improve environmental performance?
861
‘‘continuous improvement’’ starts to be applied not only in the medium to long term but also to the years immediately following EMAS registration. Obviously, for many companies in our sample, this improvement was achieved by grasping the so-called low hanging fruits, i.e., by implementing those measures or interventions that could yield a significant improvement with moderate managerial, organizational, or economic efforts. If and how the trends we identified can be considered as a general pattern toward a real and fruitful ‘‘continuous improvement’’ of environmental performance, remains to be demonstrated. Many opportunities for further research in this field are left open, including the possibility to repeat this study on the same sample of companies in the near future, to verify whether the ‘‘trend inversion’’ obtained with the introduction of EMAS is really able to consolidate and even strengthen over time.
References Al-Tuwaijri, S., Christensen, T., & Hughes, K. (2004). The relations among environmental disclosure, environmental performance, and economic performance: A simultaneous equations approach. Accounting, Organizations and Society, 29(5–6), 447–471. Andrews, R.N.L., Amaral, D., Keiner, S., Darnall, N., Gallagher, D.R., Edwards, J.D., 2003. Environmental management systems: Do they improve performance? Project Final Report for US EPA, Office of Water and Office of Policy. Economics and Innovation. Available form: http://www.cota.vt.edu/. Accessed March 2009. Anton, W. R. Q., Deltas, G., & Khanna, M. (2004). Incentives for environmental self-regulation and implications for environmental performance. Journal of environmental economics and management, 48(1), 632–654. Arimura, T., Hibiki, A., & Katayama, H. (2008). Is a voluntary approach an effective environmental policy instrument? A case for environmental management systems. Journal of Environmental Economics and Management, 55(3), 281–295. Barla, P. (2007). ISO certification and environmental performance in Quebec’s pulp and paper industry. Journal of environmental economics and management, 53(3), 291–306. Biondi, V., Frey, M., & Iraldo, F. (2000). Environmental management systems and SMEs: Barriers, opportunities and constraints. Greener Management International, 29, 55–69. Dahlstro¨m, K., Howes, C., Leinster, P., & Skea, J. (2003). Environmental management systems and company performance: Assessing the case for extending risk-based regulation. European Environment, 13(4), 187–203. FEU. (1998). Vorlaufige Untersuchungsergebnisse und Handlungsempfehlungenzum Forschungsprojekt ‘Evaluierung von Umweltmanagementsystemen zur Vorbereitung der 1998 vorgesehenen Ueberpruefung des gemeinschaftlichen Oeko-Audit-Systems’. Frankfurt. Hertin, J., Berkhout, F., Wagner, M., Tyteca, D., (2004). Are ‘soft’ policy instruments effective? The link between environmental management systems and the environmental performance of companies. SPRU Electronic Working Papers Series. Available from: http://www.sussex.ac.uk/spru/research/sewps. Accessed Feb 2009. Hertin, J., Berkhout, F., Wagner, M., & Tyteca, D. (2008). Are EMS environmentally effective? The link between environmental management systems and environmental performance in European companies. Journal of Environmental Planning and Management, 51(2), 259–283. Iraldo, F., Testa, F., & Frey, M. (2009). Is an environmental management system able to influence environmental and competitive performance? The case of the eco-management and audit scheme (EMAS) in the European union. Journal of Cleaner Production, 17(16), 1444–1452. Jaeger, T. (1998). Umweltschutz, umweltmanagement und umweltberatung—ergebnisse einer befragung in kleinen und mittleren unternehmen. ISO Bericht 55. Koln: ISO Institut. Johnstone, N., Scapecchi, P., Ytterhus, B., & Wolff, R. (2004). The firm, environmental management and environmental measures: Lessons from a survey of european manufacturing firms. Journal of Environmental Planning and Management, 47(5), 685–707. Khalid, A., Babakri, K. A., Bennett, R. A., Rao, S., & Franchetti, M. (2004). Recycling performance of firms before and after adoption of the ISO 14001 standard. Journal of Cleaner Production, 12(6), 633–637.
123
862
T. Daddi et al.
Kuisma, M., Lovio, R., & Niskanen, S. (2001). Hypotheses on the impact of environmental management systems in industry. Helsinki: Ministry of the Environment. NDEMS Project report (2000). The Effects of Environmental Management Systems on the Environmental and Economic Performance of Facilities. Available from: http://ndems.cas.unc.edu/document/PR2.pdf. Accessed March 2010. Potoski, M., & Prakash, A. (2005). Covenants with weak swords: ISO 14001 and facilities’ environmental performance. Journal of policy analysis and management, 24(4), 745–769. Rennings, K., Ziegler, A., Ankele, K., Hoffman, S., Nill, J., (2003). The influence of the EU environmental management management and audit scheme on environmental innovations and competitiveness in Germany: An Analysis on the basis of case studies and a large-scale survey. ZEW Discussion Paper n. 03-14. Available from: http://ideas.repec.org/s/zbw/zewdip.html. Accessed Feb 2010. Rennings, K., Ziegler, A., Rehfeld, K., (2004). Integrated product policy and environmental product innovations: An empirical analysis’’. ZEW Discussion Paper n. 04-71. Available from: http://ideas. repec.org/s/zbw/zewdip.html. Accessed Feb 2010. Schucht, S., (2000). The IMPLEMENTATION OF THE ENVIRONMENTAL MANAGEMent and EcoAudit Scheme (EMAS) regulation in France. Available from: http://www.cerna.ensmp.fr/Documents/ 2000-B-2.pdf. Accessed Dec 2009. Steger, U. (2000). Umweltmanagementsysteme–erfahrungen und perspektiven. Zeitschrift fu¨r umweltpolitik und umweltrecht, 4, 467–506. Steinle, C., & Baumast, A. (1997). Oeko-Audit: Problemstand und Empfehlungen fuer eine erfolgreiche Praxis (Vol. 1, pp. 11–64). Osnabrueck: Universitaet Hannover. UNI/ASU. (1997). O¨ko-Audit in der mittelsta¨ndischen Praxis–Evaluierung und Ansa¨tze fu¨r eine Effizienzsteigerung von Umweltmanagementsystemen in der Praxis. Bonn: Unternehmerinstitut e.V., Arbeitsgemeinschaft selbsta¨ndiger Unternehme. Wagner, M., (2002). The Relationship between environmental and economic performance of firms and the influence of ISO 14001 and EMAS: An empirical analysis and implications for government policy. Paper presented at the 5th Environmental Management Accounting Network Europe (EMAN-Europe) Annual Conference, University of Gloucestershire, Cheltenham, UK, 11–12. Welch, E., Mori, Y., & Aoyagi-Usui, M. (2002). Voluntary adoption of ISO 14001 in Japan: Mechanisms, stages and effects. Business Strategy and the Environment, 11(1), 43–62.
123
