2006 - Amazon Web Services
Environmental RTDI Programme 2000 – 2006
Sustainable Development – Environmental Technology, Dynamic Environmental Capabilities and Competitiveness
(2004-SD-T1-v)
Final Report
Prepared for the Environmental Protection Agency
by
Centre of Innovation and Structural Change
National University of Ireland, Galway
Authors:
Dr. Rachel Hilliard, Professor Donald Goldstein and Valerie Parker
ENVIRONMENTAL PROTECTION AGENCY
An Ghníomhaireacht um Chaomhú Comhshaoil
PO Box 3000, Johnstown Castle, County Wexford, Ireland
Telephone: +353 53 916 0600 Fax +353 53 916 0699
Email: info@epa.ie Website:
© Environmental Protection Agency 2008
ACKNOWLEDGEMENTS
This report is published as part of the Environmental Research Technological Development and Innovation Programme under the Productive Sector Operational Programme 2000–2006. The programme is financed by the Irish Government under the National Development Plan 2000–2006. It is administered on behalf of the Department of the Environment, Heritage and Local Government by the Environmental Protection Agency which has the statutory function of co-ordinating and promoting environmental research.
DISCLAIMER
Although every effort has been made to ensure the accuracy of the material contained in this publication, complete accuracy cannot be guaranteed. Neither the Environmental Protection Agency nor the author(s) accept any responsibility whatsoever for loss or damage occasioned or claimed to have been occasioned, in part or in full, as a consequence of any person acting, or refraining from acting, as a result of a matter contained in this publication. All or part of this publication may be reproduced without further permission, provided the source is acknowledged.
The EPA ERTDI Programme addresses the need for research in Ireland to inform policymakers and other stakeholders on a range of questions in relation to environmental protection. These reports are intended as contributions to the necessary debate on the protection of the environment.
ENVIRONMENTAL RTDI PROGRAMME 2000–2006
Published by the Environmental Protection Agency, Ireland
PRINTED ON RECYCLED PAPER
ISBN: 1-84095- to come
Price: €
Details of Project Partners
Dr. Rachel Hilliard
Department of Management
J.E. Cairnes School of Business & Economics
National University of Ireland, Galway
Ireland
Tel: +353 91 492932
E-mail: rachel.hilliard@nuigalway.ie
Professor Donald Goldstein
Department of Economics
Allegheny College
Meadville, Pennsylvania, U.S.A.
Tel: +1 814 332 3340
E-mail: dgoldste@allegheny.edu
Valerie Parker
Centre for Innovation & Structural Change
J.E. Cairnes School of Business & Economics
National University of Ireland, Galway
Ireland
Tel: +353 91 492817
E-mail: valerie.parker@nuigalway.ie
Table of Contents
Acknowledgements & Disclaimer ii
Details of Project Partners iii
List of Tables vi
List of Figures ix
Executive Summary x
Chapter 1. Introduction 1
1.1 Overview 1
1.2 Project Statement 1
1.3 Rationale 1
1.4 Research questions 2
1.5 Specific objectives 3
1.6 Context 3
1.7 Target Sectors 4
1.8 Data Construction 10
1.9 Mail-out survey 11
1.10 Case studies 11
Chapter 2. Literature Review 12
2.1 Overview 12
2.2 Organisational Capabilities 14
2.3 Measurement Issues 16
2.4 Modelling issues 21
Chapter 3. Measuring Environmental Performance, Practice,
and Capabilities 28
3.1 Introduction 28
3.2 Measuring environmental performance 28
3.3 Measuring environmental practice 34
3.4 Measuring organisational capabilities 40
Chapter 4. Testing Environmental Performance, Practice,
and Capabilities 45
4.1 Introduction 45
4.2 Contemporaneous determinants of environmental
Performance 45
4.3 Mediating effects of organisational capabilities 62
4.4 Conclusions 68
Chapter 5. Economic Performance in Relation to Environmental
Performance and Practice 71
5.1 Introduction 71
5.2 Measuring Economic Performance 71
5.3 Economic Performance and Environmental Impact 72
5.4 Economic Performance and Organisational Response
to IPC Licensing 74
5.5 Economic Performance and Organisational Capabilities 78
5.6 Conclusions 82
Chapter 6. Case and Survey Data Analysis 84
6.1 Introduction 84
6.2 Data collection 84
6.3 Findings on Dynamic Capability in Environmental 86
Management and Cleaner Technology
6.4 Capability in environmental management
and cleaner technology 101
6.5 Other factors impacting on environmental
performance and cleaner technology adoption 109
6.6 Environmental regulation and competitiveness 112
6.7 The role of the regulator 114
6.8 Conclusions 119
Chapter 7. Policy Implications 121
7.1 Program Efficacy 121
7.2 Opportunities for Standardisation 122
7.3 Enforcement and Assistance 123
References 124
Acronyms 133
Appendices 134
A1. Sample Facilities by Sector 134
A2. Environmental Performance of Case Study Companies 139
List of Tables
Table 1.1 Companies in the Final Sample 10
Table 1.2 Survey Response Rate 11
Table 3.1 Sector Averages: Disaggregated Emissions Performance 31
Table 3.2 Sector Averages: Waste Performance 31
Table 3.3 Sector Averages: Resource Performance 32
Table 3.4 Sector Averages: Management Practice 36
Table 3.5 Sector Averages: Technology Practice 40
Table 3.6 Sector Averages: Dynamic Capability 44
Table 4.1 Environmental impact vs organisational practice 50
Table 4.2 Environmental impact vs management categories 51
Table 4.3 Aggregate environmental impact vs technology by approaches 52
Table 4.4 Aggregate environmental impact vs technology by stages 53
Table 4.5 Early and late Sub-period Averages:
Aggregate environmental impact vs organisational practices 54
Table 4.6 Aggregate environmental impact
vs lagged technology composite 55
Table 4.7 One-year changes in environmental impact
vs lagged technology composite and categories 56
Table 4.8 Environmental impact vs organisational practice by sector 57
Table 4.9 Environmental impact vs management categories by sector 58
Table 4.10 Environmental impact vs technology approaches by sector 60
Table 4.11 Environmental impact vs technology stages by sector 61
Table 4.12 Key emissions vs static (learning by doing) capability 62
Table 4.13 Key emissions vs practice-static capability interactions 65
Table 4.14 Key emissions vs practice, static capability 66
Table 4.15 Key emissions vs practice, static capability –
For low, high internal integration DC segments (EPA data) 68
Table 5.1 Environmental impact vs operating efficiency 72
Table 5.2 Environmental impact vs operating efficiency by sector 73
Table 5.3 Changes in environmental impact vs operating efficiency 74
by sector
Table 5.4 Environmental practice vs operating efficiency 75
Table 5.5 Operating efficiency vs technology practice by approaches 77
Table 5.6 Operating efficiency vs technology practice
by production stages 78
Table 5.7 Operating efficiency vs practice-static capability interactions 79
Table 5.8 Operating efficiency vs practice, static capability 80
Table 5.9 Operating efficiency vs practice-static capability interaction
variables, low vs high dynamic capability (DC) segments 81
Table 5.10 Operating efficiency vs practice and static capability,
low vs high dynamic capability (DC) segments 82
Table 6.1 Survey Response Rate 85
Table 6.2 Interview Response Rate 86
Table 6.3 Performance, Practices and Capability in 87
Case Company PAINT1
Table 6.4 Performance, Practices and Capability in 89
Case Company METAL5
Table 6.5 Performance, Practices and Capability in 90
Case Company PAINT4
Table 6.6 Performance, Practices and Capability in 91
Case Company PAINT2
Table 6.7 Performance, Practices and Capability in 92
Case Company METAL1
Table 6.8 Performance, Practices and Capability in 93
Case Company METAL3
Table 6.9 Performance, Practices and Capability in 94
Case Company METAL6
Table 6.10 Performance, Practices and Capability in 95
Case Company METAL7
Table 6.11 Performance, Practices and Capability in 96
Case Company METAL8
Table 6.12 Performance, Practices and Capability in 97
Case Company PAINT3
Table 6.13 Performance, Practices and Capability in 98
Case Company PAINT5
Table 6.14 Performance, Practices and Capability in 99
Case Company METAL2
Table 6.15 Performance, Practices and Capability in 100
Case Company METAL4
Table 6.16 Use of teams 106
Table 6.17 Use of teams and interdepartmental cooperation for 106
environmental projects
Table 6.18 Conditional Associations: Environmental impact vs 108
technology practice categories
Table 6.19 Level of Competition in the Industry 113
Table 6.20 Competitive Strength 114
Table 6.21 Overall Impact of Environmental Management 114
Table 7.1 Environmental performance vs time 121
List of Figures
Figure 3.1 Environmental Performance Variables Construction 33
Figure 3.2 Cross-Sector Technology Practice Matrix 38
Figure 3.3 Practice and Static Capability Variables Construction 42
Figure 4.1 Key emissions vs composite technology practice 46
Figure 4.2 Key emissions vs composite management practice 47
Figure 4.3 Total waste vs composite technology practice 47
Figure 4.4 Total waste vs composite management practice 48
Figure 4.5 Combined resource use vs composite technology practice 48
Figure 4.6 Combined resource use vs composite management practice 49
Executive Summary
The project was funded during 2005-2008, through the Centre for Innovation and Structural Change (CISC) at National University of Ireland, Galway. The goal was to develop new analytical tools for modelling the environmental and economic performance of companies in Irish industry; test those tools using statistical, survey, and case study analysis of companies across industry sectors; and infer policy measures with potential for encouraging sustainable business development. A key focus was whether organisational ‘static and dynamic environmental capabilities’ could help explain differential companies adaptation to licensing. Companies studied are those covered by EPA’s Integrated Pollution Control (IPC) licensing programme, beginning in 1996 near its inception and through its end in 2004. Specific goals included:
To develop an integrated database on environmental performance, management and technology, and financial performance, among companies across industrial sectors. Sectors were chosen for number of facilities, preponderance of companies with their own accounts, lack of intra-firm trade that could bias financial data, and range of environmental aspects and approaches: metal fabricating, paint and ink manufacturing, and wood products and preservation. Financial data are from the Companies Records Office; environmental data come primarily from license applications, Annual Environmental Reports, and correspondence on file with EPA.
To develop indicators of:
Environmental performance. Three sets were created: an index of ‘key emissions’, frequently reported pollutants important in each sector; for waste, total tonnes, percent hazardous, and percent disposed; and in resource use, electricity, fuel, water, and a composite indicator. Mass data is normalised by employment for comparability over time and companies, and the indicators are expressed as ratios with sector averages to permit cross-sector comparability.
Environmental management. Hundreds of projects were scored and used to create annual measures of procedural, planning, and training related management practice, and a composite.
Environmental technology. Hundreds more projects were used to create annual technology indicators categorised both by pollution prevention approach and stage in the production process. Each project is scored according to how widespread its use in the facility and sector-specific criteria on how clean the technology, with comparability ensured by uniform approach and stage categories for all three sectors.
To develop indicators of organisational capabilities that may complement the effect of management and technology practices on environmental performance. An indicator of ‘static capability’ was developed based on ‘learning by doing’ research, as a function of experience with particular kinds of practice measured by number of projects and elapsed time. ‘Dynamic capability’ was operationalized using activities involving information search and processing, internally within the company and externally from outside sources. EPA data is supplemented with responses to a mailed-out survey.
To develop statistical models of the relationships between environmental performance, economic performance, environmental technology and management practices, and organisational capabilities. Nonparametric partial correlation is used, appropriate to a small sample with many extreme values and non-normal distributions, and to control for interrelationships among variables. Higher levels of environmental practice are associated with reduced key emissions. For waste and resource use, reverse causality characterises the relationship with technology practice: heavier impacts stimulated greater practice. Static capability generally plays a mediating role in technology, and shows the importance of experience gained prior to IPC licensing; little support for the role of dynamic capability is found. Finally, economic performance measured as operating efficiency is enhanced by environmental technology practice. There is little evidence that environmental effort impedes economic results.
To develop a more nuanced understanding, through qualitative research, of the processes and dynamics at work within the broad statistical contours. Case interviews reveal the importance of higher order management practice, including integrative processes of planning, wide searches for information, cross-functional problem solving, and team-based activity. Doing this internally rather than via consultants facilitates more change, and combining these organisational activities with technology upgrades is also important. Several expressed the view that EPA could be more helpful with a shift toward advice and assistance rather than (in their view) simply enforcement.
To develop recommendations on effective policy interventions in the regulation of the environmental impact of industrial activity. The efficacy of the IPC licensing system is supported by downward trends in the environmental impact indicators over the time period studied. That, and the role of early experience in generating accumulation of capabilities, suggest the importance of getting facilities into the programme as quickly as possible. Effectiveness could be enhanced by greater standardisation of monitoring and reporting requirements across similar facilities, lessening the burden on EPA staff, increasing perceived fairness among regulated companies, and facilitating outcomes assessment. Finally, developing EPA’s role in providing expert assistance to licensees would strengthen the dissemination of best practices while improving relations with key stakeholders.
CHAPTER 1: Introduction
1.1 Overview
The Dynamic Environmental Capabilities research project was conducted over three years from 2005-2008 and funded by the Environmental Protection Agency (EPA) of Ireland, as part of its ERTDI Programme 2000 – 2006, under Sub-Measure 2: Sustainable Development. Project funding was provided to, and organisational support received from, the Centre for Innovation and Structural Change (CISC) at National University of Ireland, Galway. CISC was established in 2002, following a substantial award in research funding under Cycle 3 of the Programme for Research in Third-Level Institutions (PRTLI) under the Irish Government’s National Development Plan. The objective of CISC is to meet the analytical and policy challenges posed by change which emerges from knowledge-intensive innovation.
1.2 Project Statement
The goal of the project was to develop new analytical tools for modelling the environmental and economic performance of companies in Irish industry; test those tools using statistical, survey, and case study analysis of companies across industry sectors; and infer policy measures with potential for encouraging business development along a sustainable path. The project also examined why some companies in similar sectors and of similar size adapt to licensing more successfully than others, and the existence and importance of internal competencies, or ‘static and dynamic environmental capabilities’ is explored. Companies studied are those covered by the EPA’s Integrated Pollution Control (IPC) licensing programme.
1.3 Rationale
Better environmental performance by individual companies is fundamentally a matter of implementing sustainable development. This is true, firstly, because in its business decision making context, cleaner production is embedded within a broader set of framework conditions. Tax and regulatory policy, capital markets, consumer demand, public expectations, and an array of inter-organisational networks all affect managerial decisions regarding cleaner technologies and practices. Because these influences and others shape the economic trajectory created by the sum of private business choices, cleaner production and policies to encourage it must be understood in sustainable development terms.
Secondly, companies’ environmental performance is bound up with sustainable development, because how cleaner production is achieved will determine the severity of the trade-off between society’s economic and environmental goals, or whether they can be harmonised. The capabilities that companies exhibit for recognising economically promising ways of reducing environmental impacts, and for creating new ways, appear to differ widely in ways that are little understood. Better analytical tools for modelling this process are required in order to determine the kinds of framework condition changes that would be most consistent with a sustainable development path.
The project aimed to capture the multidimensional nature of sustainable development. It took an innovative approach to developing new data sets and new indicators. The project contributes to the development of Irish sustainable development policy by developing and testing a set of analytical tools for understanding environmental and economic performance of industry, how they are linked and how they can be improved. The research thus aimed to foster the integration of the concerns of environmental and industrial/innovation policies.
This project also aimed to explore important issues in current research within the broad evolutionary literature on the economics of management and the company. Key questions include the relationships between automatic and purposeful activity in routines, between path-dependence and change management in response to external stimuli, and between technological and managerial capabilities. (See, for example, Zollo and Winter 2002.) In the present context, answering these questions may help to illuminate the corporate-level mechanisms and processes suggested by the induced innovation hypotheses regarding environmental regulation.
1.4 Research questions
What are the determinants of cleaner production? What factors underlie significant variation in the extent to which companies have reduced environmental impacts post-IPC licensing? The literature suggests a key role for adoption of cleaner technologies, although considerable variability remains in how effectively given technologies are utilised (MEPI, 2001). Previous research suggests that dynamic capabilities will mediate the ability to take-up cleaner technologies effectively (Hilliard, 2002).
What is the relationship between environmental performance and competitiveness? What factors underlie the combination of environmental performance that has significantly improved post-IPC licensing, and economic performance that has done likewise? Are the two connected? Research results on the economic effects of cleaner production at the company and broader levels are mixed, in part because of the failure of researchers to develop the kind of detailed indicators that are created in this project.
What policy implications arise from application of these analytical tools? EU efforts aimed at encouraging the diffusion and efficacy of environmental technology usage have increasingly focused on “horizontal” measures, which affect the framework conditions within which business decision makers operate rather than specifying particular technologies that must be employed (Joint Research Centre, 2004). By studying uptake of cleaner technologies along with organisational changes in the context of environmental and economic performance, horizontal policy measures can be proposed with a greater likelihood of contributing to sustainable development.
1.5 Specific objectives
▪ To collect data and develop an integrated database on the environmental performance, environmental management, environmental technology and financial performance of companies in different industrial sectors.
▪ To develop indicators of environmental performance, environmental management, and environmental technology.
▪ To develop indicators of organisational capabilities that may mediate the ability of industry to meet higher environmental standards without compromising competitiveness, including the dynamic capability for learning and adaptation.
▪ To develop a statistical model to explore the relationships between environmental performance, competitive, environmental technology take-up and management, and organisational capabilities.
▪ To develop a more nuanced understanding, through qualitative research, of the processes and dynamics at work within the broad statistical contours.
▪ To develop recommendations on the most effective role and nature of policy interventions in the regulation of the environmental impact of industrial activity.
1.6 Context
Ireland’s early environmental regulation programme has created favourable conditions for this research. The Environmental Protection Agency (Licensing) Regulations of 1994 phased in requirements that companies in Irish industry receive Integrated Pollution Control (IPC) licenses. Licensing was based on the use of “best available technologies not entailing excessive costs” (BATNEEC) and, determining the specific BATNEEC for each sector and then for each licensee, it set emissions limits (Clinch and Kerins, 2002). The licence is also designed to induce the development of environmental management systems, including the formalisation of procedures and the development of environmental data collection for target setting, an aspect Cunningham (2000) describes as being the key to driving continuous improvement and innovation. The programme’s combination of best technologies, environmental management and emissions limits, and its focus on pollution prevention rather than end-of-pipe controls, made it a relatively demanding but sophisticated regulatory approach (Cunningham, 2000).
Initial research on the response of licensed companies suggests that the IPC licence is effective in stimulating the development of desired behaviour. Duffy et al (2003) found, in a survey of licensees, a widespread use of indicators and wide variety of types of indicator. They identified an average of six indicators per company across 48 categories of indicator. There is also evidence of significant early improvement in environmental performance. The rate of recycling increased from 31% to 51% between 1995 and 1998 (EPA, 2000) Clinch and Kerins (2002) used reported data from the first 18 months of licensing to look at changes implemented by licensees. They found that in 120 licensed companies that made up the population of IPC licensees at that time there was for example an average of 77% drop in BOD load, a 7% drop in VOC emissions, a 68% drop in emissions of class 1 organic substances, a 92% drop in emissions of phosphorous, a 50% drop in emissions on nitrogen.
Companies’ licensing applications and Annual Environmental Reports (AERs) under the IPC programme represent a rich source of data on the organisational and technological aspects of environmental decision-making (EPA, 1997). The project combines this EPA data with business data from the Companies Registration Office (CRO) for integrated modelling of environmental and economic performance and their interaction.
1.7 Target Sectors
1.7.1 Introduction
We determined from the outset to study the sources of environmental and economic performance in the context of particular industry sectors. In so doing, we are following best practice in the large research literature around these issues (see for example MEPI, 2001).
The basic theory is that variations in company practices that might affect their environmental or economic performance, and how these practices translate into outcomes, are often highly industry sector-specific. Companies in different sectors have different ranges of technology, environmental concerns, and supply chain and market demand considerations related to environmentally relevant process and product decisions. It is possible to define sectors so that we can examine potential member and decide whether that company is engaged in the same kind of process as others in the group, selling the same kind of product in the same kind of markets, and facing the same kind of environmental problems. The key is whether theoretically defensible criteria can be developed for defining “the same kind” in each sector – criteria that are economically, technically, and environmentally meaningful.
On the other hand, conclusions from a study that is too narrowly defined around homogeneous sectors may lack generalisability. We are looking for systematic principles underlying economically viable “greening,” and to be systematic in a useful way those principles should be inferred from a sample representing a significant degree of variation. Thus, we also want sectors that display a useful range of characteristics.
With these considerations in mind, we have selected the following target sectors:
▪ Metal fabricating;
▪ Paint and ink manufacturing; and
▪ Wood sawmilling and preservation.
We engaged in a process of consultation with the available literatures and with experts in the field, and established that these three sectors offer the best match between general considerations and the particularities of what is found in the EPA’s population of licensed companies. They give us a range of sophistication in production processes, from relatively simple (wood) to complex (paint). Environmental concerns across the sectors involve different media (air emissions from VOCs in metal and paint, and diverse water emissions for all three) and levels of hazard. Most are industries where niche demand for more environmentally friendly products has appeared in some markets (paint and wood). Of key practical importance, the EPA’s files contain enough members of each sector to permit meaningful comparison, basic statistical analysis, and anonymity for the companies. Our initial plan was to focus on only two sectors; however we did not find sufficient numbers to create a robust sample from only two, and therefore decided to widen the sample to include three licensed sectors.
There are two other possible problems, given our project goal of integrating environmental with financial data, which these sectors allow us to minimise. One is the potential for a mismatch between company and facility level information. EPA licensing is by single facility, whereas financial records are generally for the company as a whole. Therefore we need to avoid sectors with many licensed facilities of companies that have many other non-licensed facilities, which severely reduce the relevance of comparing environmental and financial results. The slaughter sector proved to be largely comprised of a small number of multi-facility companies (AIBP, Dawn Meats, Glanbia and Kepac) all with consolidated accounts, making it impossible to determine turnover and profit for separate facilities. Therefore, the slaughter sector was eventually dropped from the final sample. The remaining three sectors either contain mostly facilities associated with a business unit that trades and submits financial reports on its own.
The second related problem arises when a facility is a subsidiary of a multinational company and engages in intra-firm trade. Multinationals whose units purchase inputs from and/or sell outputs to subsidiaries in other countries tend to set the prices for these intra-firm trades such that the total tax bill is minimised across countries. Thus financial results for any given subsidiary, even if reported separately in its country, are difficult to interpret reliably. We have chosen sectors that contain viable numbers of companies that engage only in arms-length input and output transactions, and thus do not suffer from this ‘transfer pricing’ problem.
In defining our three sectors and determining which of the EPA’s licensed facilities should be assigned to each, we start with the logic of the European NACE categories (Nomenclature génerale des Activités économiques dans les Communautés Européennes). Like SIC codes in the U.S., NACE codes are product-based. They group companies making like products, in terms of the input and output markets and technical processes involved. (A useful web reference is University of Strathclyde, 2005). NACE codes for EPA-licensed companies were obtained from the Central Statistics Office of Ireland (CSO).
NACE assigns four digits to each company, with increasing degrees of specificity and similarity as one proceeds from the second to the third and then the fourth digit. Thus the highest degree of sector coherence would be provided, theoretically, by including only companies with the same four digit NACE codes. We have started there, but have looked at three and even in a few cases two digit levels as well. There is inevitably some ambiguity in assigning NACE codes to individual companies, and our goals and hence criteria are not identical with those of the national statisticians. Sometimes it makes sense to use other available information to decide whether a company ought to be grouped with others with which it shares only the first two or three NACE digits.
One source of such detailed information is the EPA’s licensing classes. The EPA has grouped licensees into classes according to key environmental impacts. There is considerable overlap between NACE and EPA categories in terms of companies’ inclusion, although the amount of correspondence differs greatly across categories. Being in the same EPA class may be an important clue to ‘sufficient’ similarity for two companies with only two or three NACE digits in common, given the goals of our project. For EPA decision makers, this sampling design will offer considerably more cross-class variation than if sectors were defined primarily along licensing class lines, thus increasing generalisability of results in terms of the diverse environmental impacts of concern to the Agency. Definitions, rationales, and categorisation issues for each sector are given below. Listings of the company facilities for each sector are provided in Appendix 1, and a table providing information about the final sample is provided after the sector summaries.
1.7.2 Metal fabricating
1.7.2.1 Coverage: Definitional issues
A large number of companies in the EPA’s licensing programme are engaged in fabrication of metal parts. Products whose manufacture involves similar processes include the following: enclosures or cabinets for electronics components; containers and tanks; structural steel and builders hardware; and radiators and heating panels. The U.S. EPA (1995) suggests categorising the major process steps as fabrication per se, surface preparation, and finishing. Production may entail some combination of the following processes: shaping (forging or pressing); cutting (drill or punch, milling machine, lathe, or grinder); welding; degreasing and cleaning; and plating, painting, or other coating. Shared environmental concerns include handling of oils (cutting, cold forming or machine lubricating) and waste metal; Volatile Organic Compound (VOC) emissions from degreasers, cleaners and surface coatings; release of heavy metals; and noise. Producers engaged in these activities have access to a common range of potential environmental impact-reducing technologies. These include improved metal and oil waste segregation and recycling; non-oil lubricants in stamping or forming operations; clinching rather than welding processes; low-VOC or non-solvent cleaning and degreasing agents; and water-borne, high-solid, or powder based coatings (see IDEM, 2004).
The two major questions in terms of the sector’s boundary which emerged were whether to include electroplating facilities and foundries. Electroplating is a kind of surface finishing in which a coating of one metal is placed upon another by means of electro-deposition. This activity has a major point of intersection with metal fabricating that employs other surface finishing processes: surface preparation via cleaning and degreasing. Because degreasing has been a key environmental concern in this and related industries, this offers an argument in favour of including electroplating in our sector definition. At the same time, other (probably more critical) environmental issues for electroplaters are not significantly shared by other fabricating processes as described above: release of heavy metals (and frequently cyanide), intrinsic to the plating process; and large-scale water usage, with water-borne emissions as a major problem. Thus we have decided to exclude electroplaters per se.
Foundries engage in the casting of pieces by melting metal ingots and transferring the molten material to moulds or forms. Significant waste streams include solid metallic wastes, including toxic heavy metals, and lighter organics emissions to air from furnaces. Casting is typically followed by some rough-finishing processes (e.g. cutting off) to prepare the metal for further processing. In addition, some facilities may engage in further operations such as heat treating, de-scaling, and coating; however, these activities are generally considered in the literature to be part of metal fabrication rather than casting (see for example U.S. EPA 1998). We follow the literature by excluding foundries from the sector to be studied.
While electroplating and casting in and of themselves will not be within our definition of metal fabricating, we decided to include facilities that do either in addition to significant amounts of other, more widely shared fabricating activities. We hope that this will allow us to compare them across companies along those dimensions held in common.
1.7.2.2 Coverage: Categories
The basic NACE codes for the kinds of metal fabrication in which we are interested are the following, for manufacture of:
2811. Structures and parts of structures
2812. Builder’s carpentry and joinery…
2821. Tanks, reservoirs, and containers…
2822. …Radiators and boilers
Another code that appears for a few included companies is 2840, ‘Forging, pressing, stamping and roll forming….’
A number of EPA licensed activity classes appear among these companies. Most common is 3.9, ‘Boiler-making’; also assigned to several of the facilities we have included is 12.2, ‘Coating materials’, which reflects the fact that many metal fabrication companies use significant quantities of solvent-based coatings on their products.
1.7.2.3 Sector sample.
The EPA has 29 facilities that fall within the range of metal fabricating activity to be included in this sector. This number excludes 11 that engage exclusively or predominantly in electroplating. A further seven are excluded from statistical analysis due to lack of data, and one company is excluded as their processes were deemed too different following examination of the data. Therefore the final number of metal fabrication facilities utilised for statistical analysis is 21.
1.7.3 Paint and ink manufacturing
1.7.3.1 Coverage: Definitional issues
Manufacture of paints, inks, and related coatings (hereafter referred to simply as ‘paints’ unless otherwise noted) involve very similar processes. With this sector, there appear to be no major definitional ambiguities. The EPA’s population of licensed facilities contains a sufficient number in this sector for inclusion in the study, and the nature of the processes (manufacturing, with a chemical dimension) provides a good complement to the other sectors chosen.
Paints can be either solvent or water based. The great bulk of environmental impact from paints comes from their use, not their production (see National Centre for Manufacturing Sciences, 2004). When the substance is applied, its volatile components escape quickly into the atmosphere, and for solvent based paints this equates to VOC emissions. Thus their production and sales mix between water and solvent based product can be considered as a major determinant (indirectly) of the environmental impact of paint manufacturers. In addition, there may be toxic solids in specialty paints, which may persist after application.
As for the production process itself, impacts and technologies for their reduction include the following: enclosure (or lack thereof) of storage, transfer, and mixing equipment; use and disposal of wash water and/or solvents for equipment cleaning, including separation and recovery where possible; and generation and handling of waste product (EPA, 1996).
1.7.3.2 Coverage: Categories
For this sector there is a great deal of overlap between the NACE and EPA classifications. The single NACE code into which facilities should fall is 2430, ‘Manufacture of paints, varnishes and similar coatings, printing ink and mastics,’ although inevitably there is some apparent misclassification which we have attempted to correct. Most of these paint and related manufacturers are found in EPA class 5.7, ‘The manufacture of paints, varnishes, resins, inks…’.
1.7.3.3 Sector sample
This is the target sector with the smallest number of sample facilities. The number to be included in the study is 18, with nine IPC-licensed companies being excluded as their processes are too different from the core sample to allow for comparison. A further five are excluded from statistical analysis due to lack of sufficient data. Therefore the final number of paints companies utilised for statistical analysis is 13.
1.7.4 Wood sawmilling and preservation
1.7.4.1 Coverage: Definitional issues
An important component within the forest products industries is the sawmilling and pressure treatment of wood. The EPA’s licensing programme covers numerous such facilities. These operations generally receive raw logs or rough cut wood and mill it to shape and size, then pressure treat with chemicals for water resistance. Typical products are for the building, agricultural, and home DIY trades: construction lumber, building frames and roof trusses, poles and posts, and fencing.
Traditional pressure treatment has utilised toxic substances like creosote or arsenic-based compounds. In recent years, alternatives have become available and are increasingly being substituted, and the extent of this switch can be an important element in producers’ environmental performance. Environmental impacts include treated wood entering the solid waste stream, and treatment chemicals entering ground and surface waters.
The major definitional question for this sector has to do with facilities making composite wood-based products, such as plywood or various kinds of particle, fibre board or veneer products. These products involve breaking down the basic structure of the wood, as well as the use of resins and other bonding agents. Because the environmental impacts from these processes tend to differ from those arising from sawmilling and pressure treatment, and because the companies in composites production tend to be much larger and often multinational, we have excluded the small number of composites producers from the sector for the purposes of this study.
1.7.4.2 Coverage: Categories
The basic NACE codes for this sector are:
2010 Sawmilling, planing and impregnation of wood
2030 Manufacture of builders carpentry and joinery
The corresponding EPA licensed activity class is 8.3, ‘Treatment of wood with preservatives.’
1.7.4.3 Sector sample
There are 31 licensed facilities in the sector. The sample includes a major segment of the Irish sawmilling and wood treatment industry, covering the five major mills that account for 75% of industry output and four of the six that account for an additional 21% (COFORD, 2004). Four companies are excluded as they are involved in composites production, and one for furniture production. A further five are excluded from statistical analysis due to lack of sufficient data, and one company is excluded as it manufactures rather than uses the treatment chemicals and is therefore too different from other companies in the sample. Therefore the final sample of wood facilities utilised for statistical analysis is 25.
Table 1.1 Companies in the final sample
|Sector |No. in original sample |No. excluded during study |Final no. used for statistical analysis |
|Metals |29 |8 |21 |
|Paints |18 |5 |13 |
|Wood |31 |6 |25 |
|Slaughter |33 |33 |0 |
|Total |110 |51 |59 |
1.8 Data Construction
Data construction involves gathering raw information and using it to construct variables for analysis. Both steps – what to look for, and how to put it together – are guided by the research questions that the project seeks to answer (see below). We are interested in what distinguishes the companies that were able to meet the challenges of the Irish EPA’s IPC licensing effectively in environmental terms and competitively in economic terms. [In 2004, IPPC licensing, Integrated Pollution Prevention & Control, superseded the earlier Integrated Pollution Control licensing programme (see EPA 2005, European Union 2005)].
This central research interest suggests that we test the relationships among the following:
▪ Companies’ environmental practices;
▪ The environmental performance associated with those responses;
▪ The economic performance coincident with the above; and
▪ Organisational capabilities that we hypothesise may act as complements, by mediating the relationship between the environmental responses and the environmental and economic outcomes.
The following describes data sources and construction protocols for variables representing each of the above kinds of phenomena.
Sources of information include the following:
▪ Licensed companies’ records and reports lodged with the EPA – initial licensing applications, Annual Environmental Reports (AERs), and correspondence;
▪ These companies’ financial records collected by the Companies Registration Office (CRO);
▪ A mail-out survey of sample companies that collected additional information about the processes and sources of environmental decision making; and
▪ Semi-structured one-to-one interviews with a selected sample of case study companies.
Companies initially included in the study were in four industry sectors: metal fabricating, paint and ink manufacturing, sawmilling and wood preservation, and the slaughter of livestock. However, the slaughter sector was eventually dropped from the final sample as there was insufficient company-level financial data to enable analysis of the environmental-economic interface.
1.9 Mail-out survey
A survey questionnaire was mailed out to all 78 sample companies, including companies that had ceased production during the study years, 1997-2004 (17 in total). Its goal was to elicit additional information about companies’ sources and processes of decision making regarding environmental management. Returns were received from 21 respondents (response rate = 27%, or 34% when excluding companies that had ceased production), as per Table 1.2 below.
Table 1.2 Response Rate
|Sector |
| |Metals |Paints |Woods |
|pH performance |0.634 |0.289 |No data |
|Carbon to air |59.0 |39.1 | |
|COD |8.04 |2.47 | |
|Suspended solids |0.582 |0.473 | |
|Zinc to water |0.119 | | |
|Notes: |
|1. Averages are computed across all company-years. |
|2. All are in mass amounts, kg/year, normalised per employee. |
|3. pH is expressed as deviation from 7.5 (absolute value). |
|4. Missing values reflect insufficient data (fewer than three companies). |
|5. Extreme values are excluded using interquartile range method. |
There is a great deal of variability in averaged individual emissions across sectors. Care must be taken in assuming performance differences based on this variation. What might be considered large or small amounts of each sub-variable is likely to differ considerably by sector. The methodology used here should prevent that from distorting results, as well as being applicable even across sectors with very different lists of key emissions.
3.2.2 Waste
The basic waste categories are generic across sectors, and increasingly standardised internationally via the Pollution Emissions Register (PER) reports. Waste is classified in the IPC facility documents as either ‘hazardous’ or ‘non-hazardous’ depending on the severity of its potential impacts; and its ultimate handling is classified as involving ‘recovery’ via some kind of treatment and reuse, versus ‘disposal’ to the environment (e.g., via incineration or land filling). Combinations of these disaggregated variables have been used to create three environmental performance variables in waste, each expressed as ratios with their respective sector averages: Total waste (normalised by employment), percentage of total waste that is disposed, and percentage of total waste that is hazardous (no normalisation required for the latter two). Again, these final facility variables expressed as ratios with their sector averages must average to 1.0 across each sector, and we provide a look at the underlying sector waste differences by showing the sector averages themselves in Table 3.2.
|Table 3.2 Sector Averages: Waste Performance |
| |Metals |Paints |Woods |
|Total waste |7.97 |6.05 |173.19 |
|Percent hazardous |14.8% |23.1% |5.7% |
|Percent disposed |32.8% |47.1% |40.3% |
|Notes: |
|Averages are computed across all company-years. |
|Total waste is in tonnes per employee. |
|Extreme values have been excluded using interquartile ranges. |
While it is possible that the difference between the wood products sector and the other two in total waste reflects real differences in production processes and/or environmental performance, it may arises from reporting errors or inconsistencies in the way at least some of the woods facilities report the data. Six facilities report total waste amounts orders of magnitude greater than the other ten with waste data. It is common for otherwise-wasted wood byproducts to be collected and used as kiln fuel onsite, and examining company records suggests that some may treat this as ‘waste’ while others do not. On the other hand, the relatively low hazardous waste percentage in woods, a process traditionally reliant on toxic preservatives, suggests the progress made by these facilities in substituting more benign compounds, a suggestion that seems to be borne out in the statistical tests in Chapter 4.
3.2.3 Resource usage
The EPA asks licensed facilities to report the annual use of electricity, primary fuels, and water in the AERs. We construct variables for each. Water is given in m3; electricity end-use in MWh; liquid and gaseous fuels in litres or m3; and coal in tonnes. We convert all primary fuels to MWh (Carbon Trust, 2008).[6] For statistical analysis we use variables for water, "electricity end-use," and "primary fuels," where the latter sums MWh for each facility across its liquid, gaseous, and/or solid sources. What the energy variables measure is (the inverse of) energy efficiency; we assume that usage in MWh provides a reasonable proxy for environmental impact (Carbon Trust, 2008).[7] All resource variables are, again, normalised relative to employment for comparability purposes, and expressed as ratios to the relevant sector averages for cross-sector analysis.
Like for waste, above, we present the sector averages themselves in Table 3.3.
|Table 3.3 Sector Averages: Resource Performance |
| |Metals |Paints |Woods |
|Electricity end-use |12.36 |10.02 |42.32 |
|(MWh/employee) | | | |
|Primary fuel |8.91 |21.14 |32.65 |
|(MWh/employee) | | | |
|Water |78.76 |78.49 |22.11 |
|(m3/employee) | | | |
|Notes: |
|Averages are computed across all company-years. |
|Total waste is in tonnes per employee. |
|Extreme values have been excluded using interquartile ranges. |
In Figure 31, we summarise graphically the sequence of steps in creating the environmental performance variables described above.
Figure 3.1 Environmental Performance Variables Construction
[pic]
3.2.4 Pollution non-compliance
While company AERs contain direct environmental performance data on emissions, waste, and resource use, the correspondence in EPA files contains indirect data: notifications of regulatory non-compliance involving actual environmental impacts. We distinguish between these notifications and non-compliances that are procedural in nature (having to do with record keeping, reporting, etc), which will be treated as ‘management practices’ below.
EPA licensing requires companies to remain within ‘emission limit values’ (ELVs) for specified pollutants, and to refrain from significant environmental damage in general. When reports or site inspections show a facility to be out of compliance in either of these respects, a notification of non-compliance may be issued. Analysis of hundreds of such notifications shows that the EPA indicates the degree of severity it attaches to non-compliances by using particular words and phrases to signal companies in a consistent manner how much latitude remains before they can expect legal action to be initiated. The degree of severity is a function of both the nature of the infraction and its history – how responsive the company has been in addressing the problem.
The non-compliances based environmental performance variable is thus NC_POL, with each year’s value equal to the sum of pollution non-compliances weighted by their degrees of severity as follows:
4 * (# with prosecution) + 3 * (# with threat of legal action) + 2 * (# with threat of further enforcement) + 1 * (# with no threat)
The sector averages for this variable are as follows: metals, 2.03; paints, 1.89; and woods, 2.01. It thus appears that no major differences exist across these three sectors in the extent to which EPA have found them out of compliance with respect to environmental performance limits, specific or implied.
3.3 Measuring environmental practice
Environmental performance is directly affected by company practices. We distinguish between practices involving technology and those characterised by organisational systems or activities. We refer to the latter as ‘management practices’. There are three kinds of management practices that might affect environmental performance, by influencing the company’s ability to identify and act upon factors that can affect its environmental impacts: planning, training, and procedural. We develop measures of each by identifying and scoring discrete reported activities or projects of the appropriate type.
3.3.1 Management
3.3.1.1 Planning
This variable relates not to ‘planning’ qua orderly execution of pre-determined activities, but rather to processing of and/or search for information in the course of evaluating possible courses of action. We use information from the AERs’ Environmental Management Plans (EMPs) and from the correspondence files to construct a variable to capture this function as applied to planning actions related to environmental performance. Because they inherently involve evaluation of future actions, we treat pollution prevention and energy audits as planning projects. We score reported planning projects based on the degree to which concrete goals or targets are specified; relevant data or information is used to factor past experience systematically into decision making; and there is evidence of follow through. For each facility-year, the value of the management planning variable is the sum of the year’s projects, each scored 1-3:
3 (specific target + use of data + follow through); 2 (target + (data OR follow through)); 1 (target OR data OR follow through).
3.3.1.2 Training
By disseminating information about environmental impacts, technologies, and/or management systems, employee training programmes may affect companies’ environmental performance. We score training programmes according to their concreteness and the extent to which they appear to drive changes in employee behaviour. The scale runs up to five because, unlike for planning projects, training can generate spill-over effects over time and through the organisation. For each facility-year, the value of the management training variable is the sum of the year’s projects, each scored 1-5, with points given as follows:
1 (baseline for a reported programme) + 0-2 (extent to which driving change) + 0-2 (degree of concreteness/specificity).
3.3.1.3 Procedures
Sample companies must track, record, and report regulated activities and outcomes. Such procedural activities may affect environmental performance by providing information on which impact-reducing steps can be based and evaluated. The timeliness and completeness with which EPA monitoring, record keeping, and reporting requirements are met in the company’s AER can be quantified and combined into a measure of procedural management practices (see EPA, ‘AER guidance notes’). In scoring, we consider summary emissions data in EPA format; PER data in EU-defined format; waste data along hazardous-nonhazardous and disposed-recovered dimensions; an Environmental Management Plan (EMP) for ongoing and future pollution reduction; and resource usage data on electricity, fuel, and water. One component of each company’s annual value for the management procedural variable is an AER score between 0 and 11:
1 (if turned in) + 0-2 (no-fair-good summary emissions data) + 0-2 (no-fair-good PER data) + 0-2 (no-fair-good waste data) + 0-2 (no-fair-good EMP) + 0-2 (no-fair-good resource usage data).
Another source of procedural data is EPA non-compliance notifications of a procedural (rather than pollution-oriented) nature. These notifications use a fairly precise set of phrases to indicate the degree of severity assigned to each non-compliance by the regulatory agency. These phrases are used to create a severity-weighted sum of the year’s procedural non-compliances, in assigning a value to this component of each company’s annual value for the management procedural variable:
-4*(# with prosecution); -3*(# with threat of legal action); -2*(# with threat of further enforcement); -1*(# with no threat).
The facility-year value for the management procedural variable is the sum of these AER and procedural non-compliance scores. There is some potential for double counting, as missing or incomplete AERs can generate notices of procedural noncompliance. But EPA inspectors exercise judgment in choosing how to deal with this kind of problem, and thus a noncompliance notification for inadequate AERs provides additional information beyond the AER deficiencies themselves.
Table 3.4 shows sector averages for the management practice variables described above, in addition to a combined variable formed from the sum of the three. By two of three individual indicators and their composite, the paint and ink manufacturing facilities appear by these measures to exhibit a higher level of management practice. This impression is strengthened by some of the statistical results in Chapter 4.
|Table 3.4 Sector Averages: Management Practice |
| |Metals |Paints |Woods |
|Procedures |-3.24 |-1.20 |-1.68 |
|Planning |3.72 |3.20 |2.45 |
|Training & Development |1.28 |1.85 |0.51 |
|Composite (Sum) |2.28 |5.00 |2.54 |
3.3.2 Technology
Environmental performance is also affected by technology practices or choices. The license applications, AERs, and correspondence files contain information about what we refer to as technology ‘projects’: changes in the specific inputs, processes, and/or equipment by which outputs are created. There are two main challenges in transforming technology projects into appropriate practice variables: defining the variables for cross-sector analysis while capturing sector-specific characteristics; and representing the ongoing effects of prior years’ projects. We consider each in turn.
3.3.2.1 Cross-sector technology matrix
For each sector, we create a matrix within which technology projects are located. One dimension of the matrix categorises projects according to a standard classification of pollution-prevention approaches (see US EPA, 1995). The other dimension of the matrix breaks down each sector’s production process into major stages, according to available technical sources on that sector. This matrix makes it possible to test whether technological changes at particular points in the production process, or using particular pollution prevention approaches, are more or less important in improving environmental performance.
The key feature of the technology matrix is that the stage-of-production dimension is defined using sector-specific criteria, but within a generalised schema common to all sectors. That, and the fact that the other dimension uses pollution-prevention categories also common across sectors, allows us to score technology projects using sector-specific criteria but compare them across sectors in analysing the data.
The pollution-prevention dimension of the matrix uses the following categories:
▪ Raw materials – substitution with less polluting inputs, elimination
▪ Closing the loop – segregation and on- or off-site reuse of waste, product, and/or by-product
▪ Equipment changes – modification, replacement
▪ Process changes – not elsewhere counted
The stages-of-production dimension uses these general categories: product design, preparation, basic production, finish work, housekeeping/other. These five general stages are specified as follows in locating each sector’s projects:
Metal fabricating. Given the sample’s exclusion of facilities whose primary activities are casting (foundries) or electroplating, the stages are (US EPA, 1995; IDEM, 2004):
▪ Product design – especially choice of the finish coating with respect to environmentally relevant characteristics
▪ Metal shaping – cutting, grinding, forming, etc.
▪ Surface preparation – cleaning, degreasing, etc.
▪ Finish coating – application of painting, plating, etc.
▪ Housekeeping/other – storage, cleaning, bunding, waste handling, packaging, etc.
Paints and inks. Typically pigments, base media, and other materials are obtained from suppliers and then prepared and blended at the facility (ERI, 2004; P2Rx, 2005):
▪ Formulation – choice of base, pigments
▪ Dry milling and mixing – drying raw materials combined prior to wet processing
▪ Wet milling and mixing – further grinding and blending with wet materials
▪ Filtering and filling – final product preparation for shipping
▪ Housekeeping/other – storage, cleaning, bunding, waste handling, etc.
Wood sawmilling and preservation. Rough logs are transformed into construction, utilities, or landscaping products (Environment Canada, 2002; US EPA, 1995):
▪ Product design – choice of pressure treatment chemical; also, sourcing lumber from sustainably managed forests
▪ Conditioning and cutting – debarking, pre-drying, sawing
▪ Treatment – impregnation of cut wood with weather-proofing chemicals
▪ Storage and drip – drying and storage of treated wood
▪ Housekeeping/other – storage, cleaning, bunding, waste handling, packaging, etc.
Each project is assigned to the appropriate one among the 20 cells in the technology matrix for that facility-year – for example, a raw materials substitution in the finishing stage of production. The projects are scored on a scale of 1-5, depending on their nature and scope:
1 = End-of-pipe, small scale (e.g. bunding, over-ground pipes and tanks, dust filters)
2 = End-of-pipe, medium to large scale (e.g. waste water treatment plant (WWTP), incinerator)
3 = Clean technology, less fundamental to production process and/or small scale (e.g. recycling paper, pallets etc; inventory control)
4 = Clean technology, medium role in process and/or scale (e.g. RM substitution; heat recovery unit)
5 = Clean technology, more fundamental to production process and/or large scale (e.g. solvent distillation and re-use plant-wide; product development/re-design)
Clean technology projects are those judged to prevent or reduce environmental impacts (emissions, waste, and resource use) at the source; end-of-pipe, in contrast, entails controlling a given impact once created (Christie and Rolfe, 1995). Scale refers to the extent of the project’s effect relative to that portion of the facility’s activity to which it could apply.
Figure 3.2 shows the technology matrix:
Figure 3.2 Cross-Sector Technology Practice Matrix
| |Product Design |Preparation |Middle Stage |Finish Work |Housekeeping & Other|
|Raw Materials | | | | | |
|Equipment Change | | | | | |
|Loop Closing | | | | | |
|Other Process | | | | | |
For example, in 1998 metals facility 1 switched from a solvent based paint to a non-solvent powder coating for most of its finished products. This is a raw materials change, and is coded at the product design stage. The project is assigned a score of 4 – clean technology, applied to most but not all products – and this is added to the total in the top-most left cell of the matrix for that facility-year.
All project scores for each facility-year, in each matrix cell in Figure 3.2, are added together. These disaggregated cells can be combined as desired to create the corresponding technology practice variables. In the empirical work reported below, we have aggregated facility-year cell totals across production stages, and alternatively across pollution prevention approaches. For example, we test the effectiveness of loop closing projects at all production stages, or of projects at the preparation stage across all pollution prevention approaches. The algorithm for turning these project matrix cells into technology practice variables has to do with impact over time, to which we now turn.
3.3.2.2 Ongoing effects of prior years’ projects
We recognise that technology projects affect performance cumulatively over time. Projects once implemented potentially affect performance in subsequent years as well. But these effects decrease over time, as equipment depreciates, and as the fit between projects and the surrounding production systems in which they are embedded becomes less precise due to changes elsewhere. A large literature suggests that technology investments do not affect performance fully in the year of their implementation, and that once fully operational the ‘efficiency schedule’ of investment entails an approximately ten percent annual rate of decay in impact (Doms, 1992).
While this literature deals primarily with fixed investment, in our data equipment projects represent less than half (about 40%) of the total. It is likely that there is less persistence in the effects of non-fixed technology projects. Therefore, we transform the summed projects from the technology matrix cells into technology practice variables assuming five year project lifetimes. Each new project’s score enters the variable at half its value in its first year, full value the second, then 75, 50, and 25 percent of the original value in project years three, four, and five. Let us designate the summed projects from a particular technology matrix cell or cells as ‘PROJS it’—for example, all equipment-related projects, or all loop-closing projects for company i in any year t in the panel (t≤8). The corresponding technology practice variable TECHit is defined as follows:
[pic]
The first term is the current year’s (t) projects, and the second gives the decreasingly weighted projects from the prior year (t-1) back to the fourth year prior; projects from years further back, their five-year lifetimes having expired, are dropped. TECHit reflects the cumulative influence of the active technology stock, with the most recent projects (excepting the current years) weighted heaviest.
The technology matrix approach as introduced here is designed so that technology practice variables are scored using sector specific criteria, but the same set of variables is shared sample-wide for cross-sectoral analysis. The aim, as with the environmental performance and management practice variables discussed earlier, is to facilitate both inference of cross-sector dynamics and exploration of distinctions among the relationships at the sector level. Table 3.5 shows the sector averages, including for a composite variable given by the sum of approaches or stages (either delivers the same total).
|Table 3.5 Sector Averages: Technology Practice |
| | |Metals |Paints |Woods |
| |Raw materials substitution |3.50 |2.40 |1.67 |
|Approaches | | | | |
| |Closing the loop |2.41 |4.33 |2.59 |
| |Equipment investment |4.75 |6.20 |5.55 |
| |Process change NEC |2.60 |2.05 |3.04 |
| |Product design |0.46 |2.06 |1.59 |
| | | | | |
|Stages | | | | |
| |Preparation |2.87 |1.59 |3.62 |
| |Basic production |2.06 |3.70 |3.99 |
| |Finish work |4.05 |0.64 |1.42 |
| |Housekeeping/other |4.03 |6.99 |2.17 |
| |Composite (Sum) |13.26 |14.98 |12.85 |
Table 3.4 displays some interesting patterns. Like in management, the paints sector has the highest total for technology practices. Equipment investment is the most heavily used pollution prevention approach across the sectors, and the paints facilities’ composite advantage is maintained in equipment. (Results reported in Chapter 4 suggest that technology practice often is driven by environmental performance, with impact problems stimulating investment, rather than the more technology-reduced impacts association one might expect. The only sector in which the latter pattern seems to dominate is paints – interesting, given its higher average practice values for the management and technology composites and equipment investment in particular.)
4. Measuring organisational capabilities
As noted in Chapter 2, researchers have explored whether the efficacy of companies’ management and technology practices may be affected by underlying organisational capabilities. Strategies used by past researchers in operationalising the concept of capability have included asking for managers’ own perceptions of organisational capability relative to their competition (Christmann 2000); defining capability as a statistical residual, a portion of performance unaccounted for by measured explanatory variables (Dutta et al. 2005); and inferring capability from observable concomitant activities or characteristics (Sharma and Vredenburg 1998). Our approach is most closely related to this last one.
We distinguish between capabilities that complement the efficacy of particular kinds of practices, versus specialised capabilities enabling companies to learn new ways of doing things when the competitive environment changes. The first type we refer to as ‘static capabilities.’
Static capabilities
We hypothesise that companies build static capabilities through accumulated experience, or learning-by-doing (LBD). To match this idea with a measurable proxy, we begin with Piscitello’s approach (2004) of finding an “experiential measure” (777) for a given firm based on “the technology it employs” (759). We quantify this by adapting the learning curve literature and using our annual technology variables, extending it to management variables as well.
The literature on LBD in manufacturing suggests that learning occurs through experience, measured as cumulative production with a technology or output, and evidenced as decreasing unit labour time (Argote and Epple, 1990). Rather than cumulative production, we use the passage of time and the amount of practice, following implementation of particular kinds of projects, to proxy for experience.[8] We also extend standard LBD usage by considering not only technology, but also management practices, as experience might increase the efficacy of management practices that affect performance. The LBD literature suggests that learning occurs with respect to experience with particular kinds of technology (Klenow 1998). This idea seems consistent with the capabilities approach, and we adapt it by using our technology and management practices variables to isolate specific kinds of experience: with raw materials change, say, or planning for alternative courses of action. Finally, rather than unit labour time, the variable that is thought to be enhanced by LBD in the present context is environmental performance. We do not attempt to estimate the parameters of this experience-performance relationship empirically, but construct a static capability variable on the assumption that LBD is taking place. Empirical results are then used to test the validity of the assumption.
The construction posits two contributors to ‘experience’ with a particular kind of technology or management practice: time since the appearance of the first project of that type, and the number of projects implemented. Consider, for example, a company that has been in the dataset from 1998 onward. Suppose its first project in, say, raw materials substitution occurs in 1999, and by 2003 it has implemented two additional raw materials substitution projects. The 2003 value for the corresponding static LBD capability variable equals 7: 4 years post-initial project (the company enters 2003 with the benefit of relevant experience in 1999-2002) plus 3 total projects (through 2002). More practice, measured by number of projects and passage of time, equals by construction more learning and hence greater capability. Greater capability, in turn, is hypothesised to affect environmental performance in a manner complementary with the technology and management practice variables themselves.
This simple proxy for experience addresses two problems in empirical operationalisation of organisational capabilities. One is the difficulty in defining capabilities non-tautologically with respect to the performance they are thought to enhance. Our method does not depend on observation of successful outcomes to (falsely) infer the presence of an unobserved characteristic theorised as leading to the outcome; LBD as we define it is measured independently of environmental performance.
Secondly, while static capability so defined is inevitably correlated with the related (technology or management) practice, we believe that it is not excessively so. This is because the static capability construction is a function of the number of corresponding practice projects but not the values or scores of those projects. Overall, the correlation of practice and corresponding LBD variables is approximately 30%, and we feel justified in entering both as ‘independent’ variables in Chapter 4’s empirical models explaining environmental performance. The data can then play its appointed role in falsifying the hypothesised relationships or not, as reported in the results in Chapter 4.
Like the management and technology practice variables, the static capability variables begin with reported facility activities as classified into projects. But the project scoring and variable constructions for practices and capabilities are different and relatively independent from one another. Figure 3.3 summarises graphically the steps involved in constructing the practice and static capability variables.
Figure 3.3 Practice and Static Capability Variables Construction
[pic]
1. Dynamic capabilities
A key finding of the LBD literature is that significant disparities exist among companies in the pace and strength of organisational learning. Our theoretical framework suggests that differential dynamic capabilities may be at work. We want to measure how companies locate, process, and utilise the information involved in creating knowledge and capability. These processes occur through organisational integration (Grant, 1996): flow and processing of information both internally within the company, and externally between the company and sources in its environment. We create dynamic capability variables for each. As in the case of static LBD capabilities, we have defined dynamic capability independently of the performance outcomes it is thought to enhance, thus avoiding the tautological trap of inferring capability from performance.
Before defining these variables, we note that while theoretical considerations led us to specify static LBD capabilities as evolving during our sample period, it is not clear whether dynamic capability should be expected to change over this time frame. As noted in section 2, Winter (2003) suggests we are dealing with a first-order change capability, whose special function is to facilitate modification of static ones. This function suggests a higher order capability whose makeup is more or less fixed over the time period during which it acts upon changing static capabilities. In addition, the following discussion suggests that our information on dynamic capability in sample companies does not consistently correspond to particular years. Thus for reasons both practical and theoretical, we define dynamic capabilities as fixed characteristics that do not vary with time, and incorporate them in the empirical tests accordingly.
Internal integration might occur through the management training and planning practices introduced above. There we were concerned with the direct effect each year’s practice might have on performance. Here we consider the indirect role that training and planning might play in facilitating the company’s ability to search for and usefully integrate new information. This kind of internal integration might also occur through management work practices like cross functionality and team production, which appear (although infrequently) in facilities’ EPA files. When reference to relevant training, planning, or work practice activity appears, we score it according to the concreteness of its goals and the extent to which it is driving change. The sum of scores for training, planning, and work practice activities in a given year becomes a facility-year value for internal dynamic capability, and these annual values are then averaged across years for the facility’s EPA-based internal dynamic capability variable. All sample companies have data for this variable.
Because information on internal integration-related practices in the EPA files is infrequent, appearing as a by-product of what the EPA requires rather than as its focus, we have supplemented it with results from our mailed-out survey questionnaire. We combine survey responses relevant to internal integration into a single variable, encompassing number of key company personnel involved in environmental management; percent of workforce receiving environmental training; frequency of team problem solving; and frequency of interdepartmental cooperation. All of these activities can contribute to the flow of information within the company: the search for, and identification and processing of, information that would facilitate learning and the creation of new static capabilities.
In contrast to the EPA data, which is contained in annual reports and other dated documents, we have not been able to verify that the survey data reliably distinguishes the specific years within the panel period that reported practices have occurred. Because of this and the theoretical considerations discussed above, we construct a single, time-invariant survey-based measure of internal dynamic capability for each facility. Both the EPA-based measure (all companies) and the survey-based one (respondent subset) are used in the empirical tests reported in Chapter 4.
External integration is dynamic capability operating through knowledge-creating information flows linking the company and its outside environment. The variable is constructed from survey data only, because the EPA files do not contain information that is relevant here. External dynamic capability is constructed from scored survey responses on the number of key outside parties (customers, vendors, and others) involved in environmental management; integration of vendors with the sample company’s own staff in managing new environmental technology; number of information sources on environmental issues; participation in ongoing stakeholder initiatives with community, NGO, or governmental bodies; memberships in professional associations; and number and longevity of formal certifications (ISO 9000, ISO 14000, and EMAS). Again, we define external integration as a time-invariant characteristic of the company.
Table 3.6 shows the sector averages for the three dynamic capability variables described above. It should be noted that the EPA-based and survey-based internal integration measures are not directly comparable, as they incorporate different components and facilities samples. Across sectors, we can see that the wood products facilities lag the others in all three variables. Metals and paints appear to differ mainly with respect to the survey-based internal integration variable. It should be kept in mind that only 16 companies responded to the survey, and we have not attempted to test the representativeness of that group. The EPA-based internal integration variable, in contrast, incorporates data from the full sample of 59 IPC licensed companies in these three sectors. For that variable, as for the management and technology practice variables tabulated above, the paints facilities show the highest average.
|Table 3.6 Sector Averages: Dynamic Capability |
| |Metals |Paints |Woods |
|Internal integration – EPA data |5.59 |6.18 |3.14 |
|Internal integration – survey data |20.75 |17.22 |14.73 |
|External integration – survey data |8.08 |8.79 |5.80 |
Chapter 4 – Testing Environmental Performance, Practice, and Capabilities
4.1 Introduction
In this chapter, we statistically test hypotheses arising from our basic research questions. Section 4.2 utilises the full EPA-based data set to examine whether various measures of annual environmental performance are significantly affected by practices, both management and technology, implemented year to year. In section 4.3, we test whether the practice-performance relationship is mediated by organisational capabilities: first, static capabilities that are learned over time, again making use of the full EPA data set; and then dynamic capability to be a good learner, employing both a limited empirical definition of dynamic capability using the complete EPA-based data set, and a fuller definition based on the subsample of facilities for which we also have mail-out survey results.
4.2 Contemporaneous determinants of environmental performance
Chapter 3 introduced common rubrics for defining major dimensions of environmental performance, and the management and technology practices that might affect performance. In these rubrics the variables are scored according to sector specific criteria and these scores compared across sectors. As discussed in the prior chapter, we characterise the organisational actions over which decision makers have some control as ‘projects,’ and from project scores construct variables on management and technology practices, with the latter defined to accommodate lingering effects of earlier years’ technology projects. The environmental performance they hope to improve is measured as emissions, waste, and resource use, with each normalised by employment to account for scale, and expressed as a ratio with its sector average to permit comparability across sectors. In this section, these variables based on facility level data from the Irish EPA are used to model the determinants of manufacturers’ environmental performance.
4.2.1 Cross-sectoral relationships
First we consider the practice-performance relationship in the full sample of all three sectors combined.
4.2.1.1 Basic results
We begin by examining the relationship between highly aggregated practice and performance variables. The management and technology categories are combined to give a composite variable for each. On the environmental impact side, emissions have already been aggregated for each facility according to what is considered ‘key’ by industry sector; as noted in Chapter 3, wood products facilities do not report emissions in normalisable mass amounts, so the statistical results for emissions utilise only metals and paints facilities. Total facility waste in tons is used; and resource usage in electricity, fuel, and water have been combined. The scatter plots beginning on the next page provide a first step in getting a feel for these relationships. For key emissions, the relationships with the practice variables (Figures 4.1 and 4.2) seem to be generally negative, but with a great deal of dispersion. The variance of emissions is much greater at lower levels of technology practice, certainly, and probably management practice as well: At the higher levels of practice, the range of corresponding emissions values shrinks and concentrates at a generally lower level.
The pictured relationships in total waste (Figures 4.3 and 4.4) look very different. With technology, it is possible that the overall association is positive, and the degree of waste dispersion seems smaller at lower levels of technology. On the other hand, there may be a concavity in the shape: At very low and very high levels of technology practice, waste seems to be low; but at the middle technology levels, waste can be high, low, or in between. This feature turns out to figure prominently in the statistical analysis to come. With respect to management, there may also be some concavity, with quite a few very high waste readings corresponding to middle of the range levels of management practice. None of the handful of points with extremely low management scores – indicating severe reporting noncompliances – have either very high or very low waste levels. Resource usage exhibits a pattern with technology practice similar to that of waste performance, with an even more clear overall upward (positive) slope and less dispersed waste values at very low technology. With management practice, resource use seems to show a highly dispersed, generally upward trend.
Figure 4.1 Key emissions vs composite technology practice
[pic]
Figure 4.2 Key emissions vs composite management practice
[pic]
Figure 4.3 Total waste vs composite technology practice
[pic]
Figure 4.4 Total waste vs composite management practice
[pic]
Figure 4.5 Combined resource use vs composite technology practice
[pic]
Figure 4.6 Combined resource use vs composite management practice
[pic]
We now begin testing the direction and strength of these relationships statistically, using Spearman’s rank-order correlations.[9] Facilities’ management and technology practice values are themselves highly correlated (Spearman’s correlation of .351, significant at 1%); companies strong in one tend to be strong in the other. Therefore, in looking at the correlations between each kind of practice and performance, we use ‘partial correlation’ to control for the effects of the other practice – for example, the Spearman’s correlation between emissions and management practice, controlling for (holding constant, or removing) the effect of technology. We also control for the year, in all tests, since many of the variables exhibit time trends that may or may not be related to the relationships of interest. Our expectation is the following:
Hypothesis: Both management and technology practices will be negatively correlated with environmental impacts. Higher levels of practice will be associated with lower impacts.
Table 4.1 shows the most aggregated statistical associations.[10] Both the management and technology variables are, as expected, negatively correlated with emissions: At this aggregated level, both kinds of organisational activities aimed at improving this dimension of environmental performance are associated with reduced impacts, at standard levels of statistical significance. Combined management practice may be weakly correlated with reduced waste and resource use as well, although not at an accepted level of statistical significance. While the hypothesised relationship is negative, we use a two-tailed significance standard to incorporate the possibility of unanticipated positive relationships as well.
|Table 4.1 Full-sample Partial Correlations: |
|Environmental impact vs organisational practice |
|(Probability values in parentheses) |
| | | |Combined resource use |
| |Key emissions |Total waste |N=76 |
| |N=105 |N=125 | |
|Technology (all categories, |-.196** |.233*** |.419*** |
|controlling for management & year) |(.047) |(.009) |(.000) |
|Management (all categories, |-.179* |-.126 |-.045 |
|controlling for technology & year) |(.071) |(.166) |(.745) |
|***Significant at 1% level; **5%; *10% (two-tailed). |
|Based on Spearman’s rho. |
An example is the positive correlation between technology and both total waste and combined resource use. Disaggregating resource usage shows this correlation to hold across electricity (5% significance level), primary fuels (1% level), and water (5% level). A possible explanation for this unexpected result is that there is a reverse causality at work, with facilities generating high waste levels and/or resource usage undertaking technology investments intended to reduce them. Although there is a persistence effect built into the technology variables, a sufficient lag in efficacy of these investments could complicate inferences about the direction of causality. One way of exploring this is to separate the earlier and later periods in the panel. Another is to disaggregate the practice measures to see if certain kinds of technology projects are driving the unexpected positive relationship. We return to these considerations below.
We first turn to a finer-grained analysis by beginning to disaggregate the explanatory variables, starting with management practices and their partial correlations with environmental performance. As described above, we can separate management practice into projects and activities involving procedures (record keeping and reporting), planning (evaluation of alternatives), and training (staff development related to environmental impact management). As in Table 4.1, the results in Table 4.2 are to be read column-wise; for example, the partial correlations between emissions and management procedures, controlling for planning and training, and so on. In addition, given the management-technology correlation noted above, it is important to control for one when disaggregating the other. To economise on degrees of freedom, in each partial correlation of a disaggregated management category we control for other management disaggregates singly and for the aggregate technology variable, and vice versa.
|Table 4.2 Full-sample Partial Correlations: |
|Environmental impact vs management categories+ |
|(Probability values in parentheses) |
| |Key |Total |Combined resource use |
| |emissions |waste |N=76 |
| |N=105 |N=125 | |
|Procedure (controlling for planning, |-.232** |-.177* |.058 |
|training) |(.020) |(.053) |(.629) |
|Planning (controlling for procedure, |-.228** |-.110 |-.149 |
|training) |(.022) |(.231) |(.210) |
|Training (controlling for procedure, |.257*** |.185** |.054 |
|planning) |(.010) |(.042) |(.655) |
|***Significant at 1% level; **5%; *10% (two-tailed). |
|+Each partial correlation also controls for year and aggregate technology. |
|Based on Spearman’s rho. |
Table 4.1’s negative relationship between emissions and combined management categories is shown in Table 4.2 to be driven by procedural and planning related management activities. Indeed, training related management practice shows an unexpected positive partial correlation with emissions, and this positive relationship with training extends to total waste as well. It is possible for training, like in the case of technology vs waste and resource use, that a kind of reverse causality is in effect, with companies facing certain kinds of environmental challenges undertaking higher levels of the (intended) impact-reducing practice in question.
Table 4.3 presents corresponding partial correlations between aggregate environmental impact measures and technology, disaggregated by approach to pollution prevention: raw material substitution; closing the loop by reusing (formerly) waste input or output; equipment investment; and other process change. Table 4.4 gives the aggregate environmental impact-disaggregated technology correlations, with technology categorised according to stage in the production process: product design, preparation, basic processing, finish work, and housekeeping/other.[11]
Emissions are strongly, negatively associated with two of the technology approaches in Table 4.3, both having to do with process-related approaches, and with two of the technology stages in Table 4.4; but there is a positive correlation with equipment investment when breaking down technology by approach and with basic processing when disaggregating by stage of production. As for waste, we can now see that its unexpected positive correlation with technology at the aggregate level (Table 4.1) appears to be driven by a rather strong positive correlation with equipment investments (Table 4.3). Resource usage, on the other hand, shows a broad positive association with most technology categories along the approach dimension and two categories along the stage dimension.
|Table 4.3 Full-sample Partial Correlations: |
|Aggregate environmental impact vs technology by approaches+ |
|(Probability values in parentheses) |
| |Key |Total |Combined resource use |
| |emissions |waste |N=76 |
| |N=105 |N=125 | |
|Raw materials (controlling for others)|.037 |-.069 |.230* |
| |(.713) |(.453) |(.054) |
|Closing loop (controlling for others) |-.260*** |.068 |.075 |
| |(.009) |(.459) |(.535) |
|Equipment |.199** |.338*** |.259** |
|(controlling for others) |(.047) |(.000) |(.029) |
|Process, NEC |-.357*** |-.155* |.227* |
|(controlling for others) |(.000) |(.090) |(.057) |
|***Significant at 1% level; **5%; *10% (two-tailed). |
|+Each partial correlation also controls for year and aggregate management. |
|Based on Spearman’s rho. |
Looking at patterns by technology category, we see that the equipment investment approach is positively correlated with environmental impacts of all three kinds. In addition, practices at the basic processing stage show positive correlations with two of the environmental impact measures; so do the miscellaneous practices under the housekeeping stage, although the third impact measure there (key emissions) shows a negative correlation.
For key emissions, then, and for some of the waste correlations, the data are consistent thus far with the idea that certain kinds of IPC licensing-related management and technology choices have been successful in reducing environmental impacts. These results come from considering simultaneously both the cross sectional and time series dimensions of the data for facilities in the full, three-sector sample. Refining this preliminary conclusion will require at least exploring the statistical relationships at a single-sector level, and bringing to bear insights from individual facility interviews that have been conducted.
On the other hand, there are the unexpected associations noted above, and apparent patterns in these results thus far may be consistent with a reverse-causality scenario involving the relationship between environmental problems and technology investment. Facilities with elevated waste, for example, may attempt to address the problem by means of equipment changes (Table 4.3), especially at the finishing production stage (Table 4.4).[12] If managers are focused on resource use as well as emissions and waste in their dealings with the regulators, then this logic could also apply to the resources-technology categories correlations.
|Table 4.4 Full-sample Partial Correlations: |
|Aggregate environmental impact vs technology by stages |
|(Probability values in parentheses) |
| |Key emissions |Total |Combined resource use |
| |(N=105) |waste |(N=76) |
| | |(N=125) | |
|Product design (controlling for |.014 |-.101 |.105 |
|others+) |(.891) |(.275) |(.386) |
|Preparation |-.302*** |-.033 |.014 |
|(controlling for others+) |(.002) |(.720) |(.910) |
|Basic processing |.167* |.105 |.410*** |
|(controlling for others+) |(.099) |(.257) |(.000) |
|Finish work |-.004 |.223** |.177 |
|(controlling for others+) |(.970) |(.015) |(.143) |
|Housekeeping/other |-.232** |.195** |.301** |
|(controlling for others+) |(.021) |(.034) |(.011) |
|***Significant at 1% level; **5%; *10% (two-tailed). |
|+Each partial correlation also controls for year and aggregate management. |
|Based on Spearman’s rho. |
4.2.1.2 Reverse causality?
The appearance of positive associations between environmental impacts and actions intended to reduce impacts is an important issue in terms of regulatory efficacy, and makes exploring the reverse-causality scenario worthwhile. Here, the panel structure of the data allows us to investigate whether the relationships of interest are driven by variation across companies in the cross section, changes over time, or both. We can begin by seeing whether early investments correlate with reduced (or at least not heightened) environmental impacts later on.
We define the years 1996-2000 as ‘early,’ and 2001-2004 as ‘late.’ All impacts and management and technology projects within these sub-periods are averaged, giving each facility a single early value and a single late value for each impact, management, and technology category. In technology, we average the project scores rather than the practice variables, because the latter as constructed incorporate persistence effects of past technology projects, and this makes it impossible to cleanly separate the time periods. Table 4.5 shows the cross-sectional partial correlations between aggregate impacts and activities.
The numbers of observations are severely limited, because only those containing data for all the relevant impact and activity categories in the given sub-period are used. Most of these by-period matchings fail to show statistically significant correlations. The results for waste and technology are
|Table 4.5 Partial Correlations, Early and late Sub-period Averages: |
|Aggregate environmental impact vs organisational practices |
|(Probability values, observations in parentheses) |
| | |Impacts | |
| | |Key emissions |Total |Combined resource | |
| | | |waste |use | |
|Practic|Management+ |-.198 |-.086 |-.156 | |
|es | |(.447, N=18) |(.734, N=19) |(.610, N=14) |Early practices – |
| | | | | |Early impacts |
| |Tech projects++ |.094 |-.008 |.093 | |
| | |(.720, N=18) |(.974, N=19) |(.762, N=14) | |
| |Management+ |-.288 |.032 |.271 | |
| | |(.218, N=21) |(.875, N=27) |(.310, N=17) |Early practices – |
| | | | | |Late impacts |
| |Tech projects++ |-.364 |.127 |.302 | |
| | |(.115, N=21) |(.537, N=27) |(.256, N=17) | |
| |Management+ |-.161 |-.134 |.375 | |
| | |(.433, N=27) |(.424, N=39) |(.103, N=21) |Late practices – |
| | | | | |Late impacts |
| |Tech projects++ |-.148 |.294* |.345 | |
| | |(.470, N=27) |(.073, N=39) |(.136, N=21) | |
|* Significant at 10% level (two-tailed). |
|+ Each management partial controls for aggregate technology projects. |
|++ Each technology projects partial controls for aggregate management. |
|‘Early’ = 1996-2000; ‘late’ = 2001-2004. |
|Based on Spearman’s rho. |
at least not inconsistent with the reverse-causality story, with the late impact – late technology correlation positive (bottom panel) as it is in the full panel reported earlier, while the relationship disappears when considering early practice and late impact (middle panel).
We can more fully use the time series dimension of the data to examine the relationship between environmental impacts and successively shorter lags in the organisational practices. Here we focus on technology[13] and its relationships with waste and resource usage, because it becomes increasingly clear that this is where the possibility of the reverse-causality scenario is strongest. Table 4.6 shows the relevant partial correlations.
|Table 4.6 Partial Correlations: |
|Aggregate environmental impact vs lagged technology composite |
|(Probability values, observations in parentheses) |
| | |Waste |Resource use |
| | |(total) |(combined) |
|Technology lagged | |.198 |.154 |
|four years+ | |(.123, N=64) |(.364, N=39) |
|Technology lagged | |.165 |.125 |
|three years+ | |(.139, N=84) |(.387, N=52) |
|Technology lagged | |.128 |.259* |
|two years+ | |(.204, N=102) |(.048, N=61) |
|Technology lagged | |.141 |.384*** |
|one year+ | |(.128, N=120) |(.001, N=71) |
|Technology | |.233*** |.419*** |
|without lag+ | |(.009, N=125) |(.000, N=76) |
|***Significant at 1% level; **5%; *10% (two-tailed). |
|+Controls for year and aggregate management. |
|Based on Spearman’s rho. |
The final row in Table 4.6, technology vs environmental impacts with no lag, simply reproduces the results from Table 4.1. As we correlate resource impacts with practices further and further back in time, the strength of the positive association progressively recedes. This pattern suggests that something like the reverse-causality scenario may be going on: problems currently and in the recent past may stimulate heightened levels of technology investment. On the other hand, the pattern in the technology-waste correlations is not so clear, although the strong positive link is only present in the current period.
The correlations in Table 4.6 do not appear to offer evidence (negative correlations) that technology investments pay off later in reduced impacts. Here it is possible that the tests are confounding levels and changes: the high-investment facilities would still be those starting with very high environmental impact levels, and perhaps even later improvements are not sufficient to reverse this strongly enough to create negative correlations between early investment and later impacts.
We can briefly explore this final aspect of the reverse-causality scenario by looking at the correlation between technology practice and environmental impact changes, rather than levels as we have done thus far. Table 4.7 gives the relevant results, for percent change in the impact variables from one year prior. Recall that the annual technology variables incorporate diminishing persistence effects from prior years’ investments. Thus we are asking whether various dimensions of technology practice have the cumulative effect of creating reduced impacts over time. Again, we limit attention to the total waste and resource usage impacts for which the positive correlations appeared.
In Table 4.7 we have shaded the cells corresponding to the unexpected, statistically significant positive correlations reported in earlier tables. Generally speaking, those rather strong positive correlations between technology and environmental impact levels, reported earlier, have here become statistically insignificant negative correlations when considering instead environmental impact percent changes. While looking at waste and resource changes rather than levels eliminates the greater practice – greater impact association, like the progressive lags in Table 4.6 it does not suggest that higher levels of technology practice produced impact reductions year-on-year.
|Table 4.7 Partial Correlations: One-year changes in environmental impact |
|vs lagged technology composite and categories |
|(Two-tailed probability values in parentheses) |
| |% Change in Total Waste, |% Change in Combined |
| |N=87 |Resource Use, N=49 |
|Technology |-.117 |-.126 |
|(all categories, combined+) |(.286) |(.397) |
|Raw materials |.024 |-.066 |
|(controlling for other approaches+) |(.834) |(.668) |
|Closing loop |-.097 |.191 |
|(controlling for other approaches+) |(.386) |(.215) |
|Equipment |-.172 |-.162 |
|(controlling for other approaches+) |(.122) |(.293) |
|Process, NEC |.070 |-.005 |
|(controlling for other approaches+) |(.533) |(.975) |
|Product design |-.012 |.036 |
|(controlling for other stages+) |(.916) |(.820) |
|Preparation |.002 |-.091 |
|(controlling for other stages+) |(.988) |(.561) |
|Basic processing |.002 |-.156 |
|(controlling for other stages+) |(.986) |(.316) |
|Finish work |.030 |-.134 |
|(controlling for other stages+) |(.792) |(.390) |
|Housekeeping/other |-.138 |-.218 |
|(controlling for other stages+) |(.221) |(.160) |
|+Each partial correlation also controls for year and aggregate management (lagged one year). |
|Based on Spearman’s rho. |
Nevertheless, these explorations are at least not inconsistent with the idea that the unexpected positive correlations indicated high levels of high levels of waste and resource usage that then stimulated technology changes in response, rather than technology changes resulting in perversely elevated levels of environmental impact. Along with the results reported in Tables 4.5 and 4.6, they suggest that this reverse causality scenario may have been at work.
4.2.2 Sector-specific relationships
In this section we examine whether the preceding results, based on dynamics that are common across the three industry sectors studied, differ by sector. In Table 4.8 we look at the broad correlations reported for the full sample in Table 4.1, between environmental impact types and combined technology and management practices, over the entire panel period. (No results are reported for wood products emissions, because these facilities did not report emissions in mass amounts that could be normalised by employment, as reported in Chapter 3.)
|Table 4.8 Partial Correlations+ By Sector: |
|Environmental impact vs organisational practice |
|(Probability values, observations in parentheses) |
|Sectors |Practices |Impacts |
| | |Key |Total |Combined |
| | |emissions |waste |resource use |
| |Technology |-.193 |.147 |.455** |
|Metal fabrication |(all categories) |(.142, N=61) |(.275, N=59) |(.013, N=31) |
| |Management |-.055 |.051 |-.149 |
| |(all categories) |(.681, N=61) |(.704, N=59) |(.442, N=31) |
| |Technology |-.182 |.044 |.125 |
|Paint & ink |(all categories) |(.249, N=44) |(.804, N=36) |(.589, N=23) |
|manufacturing | | | | |
| |Management |-.378** |-.115 |.228 |
| |(all categories) |(.014, N=44) |(.516, N=36) |(.320, N=23) |
| |Technology |No data |.404** |.510** |
|Wood products & |(all categories) | |(.033, N=30) |(.021, N=22) |
|treatment | | | | |
| |Management | |-.240 |-.053 |
| |(all categories) | |(.219, N=30) |(.823, N=22) |
|***Significant at 1% level; **5%; *10% (two-tailed). |
|Based on Spearman’s rho. |
|+Technology partials control for management, and vice versa; both control for year. |
The sectoral breakout reveals differences regarding the puzzle of the positive technology-waste and resource usage correlations in the full-sample results, tentatively attributed in the preceding analysis to a reverse-causality scenario. This positive association shows up most consistently in the wood products sector. The metal fabricating sector also exhibits the pattern found in the full sample results, for technology vs resource use. But the paint and ink sector does not display this positive correlation. On the other hand, the negative full-sample association between key emissions and combined management practices (reported in Table 4.1) is shown in Table 4.8 to be driven by the paint and ink facilities.
A finer-grained look is provided in Table 4.9, which reports partial correlations for individual management practice categories by sector. Here as for the full-sample results earlier, each partial controls for the other management categories as well as combined technology practice.
|Table 4.9 Partial Correlations+ By Sector: |
|Environmental impact vs management categories |
|(Probability values, observations in parentheses) |
|Sectors |Management Practices+ |Impacts |
| | |Key |Total |Combined resource use |
| | |emissions |waste | |
| |Procedure |-.094 |-.043 |-.058 |
|Metal fabrication | |(.487, N=61) |(.754, N=59) |(.773, N=31) |
| |Planning |-.099 |-.037 |-.416** |
| | |(.464, N=61) |(.789, N=59) |(.031, N=31) |
| |Training |.255* |.223 |.155 |
| | |(.056, N=61) |(.101, N=59) |(.440, N=31) |
| |Procedure |-.432*** |.069 |.309 |
|Paint & ink | |(.005, N=44) |(.709, N=36) |(.198, N=23) |
|manufacturing | | | | |
| |Planning |-.421** |-.097 |.332 |
| | |(.007, N=44) |(.917, N=36) |(.164, N=23) |
| |Training |.205 |-.303* |.041 |
| | |(.205, N=44) |(.092, N=36) |(.868, N=23) |
| |Procedure |No data |-.341* |-.100 |
|Wood products & | | |(.088, N=30) |(.694, N=22) |
|treatment | | | | |
| |Planning | |.068 |.214 |
| | | |(.740, N=30) |(.393, N=22) |
| |Training | |.317 |-.062 |
| | | |(.115, N=30) |(.808, N=22) |
|***Significant at 1% level; **5%; *10% (two-tailed). |
|Based on Spearman’s rho. |
|+Each management partial controls for year, other management categories, and aggregate technology. |
Disaggregating by sector and management category shows procedural management associated with lower impacts in emissions for the metals sector and in waste for wood products. Planning is associated with lower impacts in resource use for metals and in emissions for paint and ink. On the other hand, training and development correlates positively with emissions levels in the metals sector, and (perhaps) positively just below standard significance levels with total waste in metals and wood products; but for the paint and ink facilities, training is associated with lower waste levels. In general, it appears that management practices more consistently exhibit intended outcomes in the paint and ink sector than in the other two.
We turn finally to disaggregated technology practices at the sector level, first by pollution prevention approach and then by stage in the production process. Table 4.10 shows the partial correlations between impact types and technology approaches, controlling also for aggregate management practice. The cross-sectoral positive correlation between equipment investment and environmental impact is shown here to be driven by facilities in the metals and wood products sectors. Beyond that, no clear patterns stand out. Paint and ink facilities using raw material substitution – which most frequently indicates substituting water for solvent based inputs – have reduced waste. Metals facilities with lots of miscellaneous process changes have done likewise with emissions. But the paint sector exhibits an association between higher waste and more loop closing projects – generally representing reuse of (formerly) waste materials and/or product in subsequent rounds of production. Reverse causality may be at work, but the numbers of observations are insufficient to explore this using techniques like those reported above for the more aggregated data.
Table 4.11 reports the impact correlations for technology organised according to stage in the production process. In paint and ink, the positive association between waste and preparation-stage practice (pigment grinding and mixing) seems to correspond to the positive waste-loop closing correlation in the technology approaches results from Table 4.10. On the other hand, paint facilities exhibited reduced resource usage with greater practice at the same preparation stage; it is possible that addressing the waste problems there had the side effect of permitting reduced water and/or energy use. In metals, greater practice at the preparation stage (mostly cleaning and degreasing) is associated with lower emissions, possibly from reduced solvent usage. There is a link between greater practice at the basic processing stage and higher resource impacts in metals (cutting and shaping) and paints (blending of wet and dry inputs). In the woods sector, strong positive associations characterise impacts and finishing stage practice (post-pressure treatment drip and drying). For all of the above positive correlations, the reverse causality scenario is possible but cannot be explored statistically.
Taking Tables 4.10 and 4.11 together, the results may suggest (albeit not strongly) that technological efforts focused at more fundamental changes, involving the composition of the product itself and/or of the materials employed, provide a more consistent pattern of environmental impact reduction. This result would be consistent with a stylised fact reported frequently in the literature, that ‘cleaner technologies’ – those aimed at reducing impacts at the source rather than cleaning them up at the ‘end of the pipe’ – are most promising.
These tables also provide additional detail regarding the continuing puzzle of positive correlations between technology practice and environmental impact. They suggest that these positive associations tend to appear mostly in practices involving tangible investment in the core production processes across all sectors. Again, these sectoral results themselves leave open whether the positive associations indicate unintended and unwanted consequences, a reverse-causality sequence, or some combination of the two.
|Table 4.10 Partial Correlations+ By Sector: |
|Environmental impact vs technology approaches |
|(Probability values, observations in parentheses) |
|Sectors |Tech. practices by |Impacts |
| |pollution prevention | |
| |approach | |
| | |Key |Total |Combined resource use |
| | |emissions |waste | |
| |Raw materials |.063 |.031 |.205 |
| |substitution |(.646, N=61) |(.832, N=59) |(.316, N=31) |
| | | | | |
|Metal fabrication | | | | |
| |Closing the |-.114 |-.163 |.152 |
| |loop |(.401, N=61) |(.238, N=59) |(.460, N=31) |
| |Equipment |.184 |.302** |.234 |
| |change |(.175, N=61) |(.027, N=59) |(.249, N=31) |
| |Process change, |-.376*** |.114 |.257 |
| |NEC |(.004, N=61) |(.414, N=59) |(.205, N=31) |
| |Raw materials |-.123 |-.408** |-.219 |
| |substitution |(.455, N=44) |(.023, N=36) |(.381, N=23) |
| | | | | |
|Paint & ink | | | | |
|manufacturing | | | | |
| |Closing the |-.194 |.420** |.125 |
| |loop |(.237, N=44) |(.019, N=36) |(.620, N=23) |
| |Equipment |-.040 |.109 |-.031 |
| |change |(.808, N=44) |(.558, N=36) |(.903, N=23) |
| |Process change, |-.031 |-.225 |.243 |
| |NEC |(.851, N=44) |(.223, N=36) |(.331, N=23) |
| |Raw materials |No data |-.081 |.182 |
| |substitution | |(.701, N=30) |(.485, N=22) |
|Wood products & | | | | |
|treatment | | | | |
| |Closing the | |-.141 |-.094 |
| |loop | |(.501, N=30) |(.720, N=22) |
| |Equipment | |.670*** |.423* |
| |change | |(.000, N=30) |(.091, N=22) |
| |Process change, | |-.148 |-.038 |
| |NEC | |(.480, N=30) |(.886, N=22) |
|***Significant at 1% level; **5%; *10% (two-tailed). |
|Based on Spearman’s rho. |
|+Each partial correlation controls for year, other technology approaches, and aggregate management. |
|Table 4.11 Partial Correlations+ By Sector: |
|Environmental impact vs technology stages |
|(Probability values, observations in parentheses) |
|Sectors |Technology practices |Impacts |
| |by stage in production | |
| |process | |
| | |Key |Total |Combined resource use |
| | |emissions |waste | |
| |Product design |-.120 |-.031 |-.261 |
| | |(.384, N=61) |(.854, N=59) |(.208, N=31) |
| | | | | |
|Metal fabrication | | | | |
| |Preparation |-.470*** |.027 |.137 |
| | |(.000, N=61) |(.847, N=59) |(.515, N=31) |
| |Basic processing |.080 |.039 |.434** |
| | |(.562, N=61) |(.780, N=59) |(.030, N=31) |
| |Finish work |.055 |.130 |-.013 |
| | |(.689, N=61) |(.354, N=59) |(.951, N=31) |
| |Housekeeping/other |-.104 |-.043 |.655*** |
| | |(.451, N=61) |(.759, N=59) |(.000, N=31) |
| |Product design |-.045 |-.064 |-.212 |
| | |(.790, N=44) |(.736, N=36) |(.414, N=23) |
| | | | | |
|Paint & ink | | | | |
|manufacturing | | | | |
| |Preparation |.168 |.383** |-.518** |
| | |(.314, N=44) |(.037, N=36) |(.033, N=23) |
| |Basic processing |-.014 |-.147 |.672*** |
| | |(.935, N=44) |(.438, N=36) |(.003, N=23) |
| |Finish work |-.052 |-.016 |.215 |
| | |(.758, N=44) |(.932, N=36) |(.407, N=23) |
| |Housekeeping/other |-.387** |-.055 |.356 |
| | |(.016, N=44) |(.772, N=36) |(.161, N=23) |
| |Product design |No data |-.071 |.410 |
| | | |(.741, N=30) |(.115, N=22) |
| | | | | |
|Wood products & | | | | |
|treatment | | | | |
| |Preparation | |.159 |-.255 |
| | | |(.459, N=30) |(.340, N=22) |
| |Basic processing | |.085 |.254 |
| | | |(.694, N=30) |(.343, N=22) |
| |Finish work | |.571*** |.483* |
| | | |(.004, N=30) |(.058, N=22) |
| |Housekeeping/other | |.174 |-.188 |
| | | |(.416, N=30) |(.486, N=22) |
|***Significant at 1% level; **5%; *10% (two-tailed); based on Spearman’s rho. |
|+Each partial correlation controls for year, other technology stages, and aggregate management. |
4.2.3 Discussion
The data appear to suggest that the environmental effects of organisational choices in management and technology, made at these facilities in the context of IPC licensing, are highly context specific. At the level of aggregated sectors and management and technology practices, higher levels of practice are linked with improved performance with respect to the emissions that are key for these facilities. At the sectoral level, this holds at accepted significance standards only for management practices in the paint and ink sector. Paints facilities exhibit this strong relationship with reduced emissions for greater managerial effort in both procedures and planning. The puzzling association of greater technology practice with more waste and resource impact seems likely to reflect a reverse-causality process, whereby environmental impact problems stimulate increased activity aimed at reducing those impacts; this seems especially true for equipment investments in the metals and woods sectors and practice in the core production stages across sectors. (It is noteworthy that equipment investment is not associated with reduced impact in any environmental category, sector, or time structure.) Finally, there may be indications that changing the composition of products and/or materials is better able to reduce environmental impacts than other technology changes that take those as given.
In these results, no account has been taken of the possible role of underlying organisational capabilities in mediating the efficacy of the practice-performance relationship. We turn now to statistical results on this possibility.
4.3 Mediating effects of organisational capabilities
In results reported in the previous section, while many practices are associated with better performance in terms of lower emissions, as expected, a surprising finding is that environmental performance in waste and resource usage often show positive correlations with practice, especially in technology. We explored there the possibility that high levels of waste or resource use stimulate companies to invest more heavily in technology changes intended to improve performance, a ‘reverse causality scenario.’ Because here we want to test the potential for organisational capability to act as a complement with practice in improving environmental performance, we will avoid these complications by focusing on the key emissions dimension of performance in this section. Given the lack of usable emissions data in the wood products sector, this limits consideration to the metals and paints sectors.
Our goals are to shed light on the following questions:
First, does static capability accumulate through practice within the time period of our panel? We would like to know if the kinds of purposeful activities implemented by sample companies generate a learning by doing (LBD) effect, as differential levels of experience across companies and over time improve performance and make given practices more effective in improving environmental performance. The maximum time in the panel is nine years, so we will be testing whether static capability and this mediating effect on the practice-performance relationship can develop that quickly.[14]
Hypothesis 1: Static capability will be negatively associated with emissions, controlling for the effect of the corresponding kind of practice.
Hypothesis 2: Practice will be less negatively correlated with emissions when controlling for the effect of the corresponding static capability. I.e., we hypothesise that some of the apparent impact of practice reflects a complementarity effect due to static capability.
Next, does dynamic capability facilitate adaptation to the heightened environmental standards ushered in by IPC licensing? Dynamic capability should improve learning, and companies that are better learners should be more successful in implementing management and technology changes that improve environmental performance. Thus we want to test whether companies with stronger dynamic capabilities responded more effectively to the demands of IPC licensing by building new static capabilities for particular kinds of environmental impact reduction.
Hypothesis 3: The relationships suggested in Hypothesis 1 will be stronger among the higher dynamic capability companies.
4.3.1 Learning by doing: Static capabilities
In Chapter 3, we defined the static capability variables as functions of the time following first implementation of a given type of project, and the number of like projects subsequently implemented. We start with Hypothesis 1 by examining the direct effect of static LBS capability on emissions performance. We know that technology and management practices are correlated with each other, and the corresponding capabilities are also (Spearman’s rho for combined technology capabilities and combined management capabilities is .787, significant at the 1% level.) In addition, we theorise that capabilities and practices are related. Thus, in testing the effect of each capability on emissions, we control for the impacts of the other capability and both practices, along with the year. Table 4.12 gives the results.
This test attempts to address the first research question posed above – whether static capabilities emerge through experience (learning by doing) during the IPC period covered by the panel. That question is animated in part by concern within the research literature with the relative importance of capability inherited from the past vs that which is subject to purposeful direction in the present. As defined in the above tests, the LBD static capability variables incorporate only. We have access to information about both experience tied to projects undertaken during the IPC years and pre-IPC activities, the latter being contained in companies’ license application files. To use that information to
|Table 4.12 Partial Correlations: |
|Key emissions vs static (learning by doing) capability |
|(Probability values in parentheses; N=105) |
| |Static capability |Static capability |
| |based on IPC period projects only |based on pre-IPC and IPC period |
| | |projects |
|Technology |-.074 |-.200** |
|static capability |(.462) |(.045) |
|Management |-.003 |.085 |
|static capability |(.974) |(.398) |
|**Significant at 5% level. |
|Each partial controls for the effects of year, the other capability, and both practices on emissions. |
|Partial correlations based on Spearman’s rho. |
address the issue of prior expertise, we have computed the LBD static capability variables alternately to ignore and to allow pre-IPC projects to feed into the accumulation of experience as measured in those variables. The results for these alternative versions appear column-wise in Table 4.12. Neither technology nor management capability calculated without pre-IPC experience affects emissions performance. But the data are consistent with Hypothesis 1 for technology LBD capability incorporating both IPC and pre-IPC experience.
Our basic strategy in testing Hypothesis 2 is to model static organisational capabilities as complements to the effect of direct practices upon performance. This approach has been applied to environmental impact-reduction by Christmann (2000) and to the efficacy of information technology investment by Brynjolfsson and Hitt (2000). We begin by specifying the complementarity in terms of a multiplicative interaction variable. For a given company and year, we multiply the value of the technology or management practice variable of interest by the value of the corresponding learning-by-doing (LBD) static capability variable. The expected correlation between emissions and these interaction variables is negative: Given some level of practice, we expect a higher level of LBD static capability to reduce emissions; and given some level of LBD static capability, we expect a higher level of practice to reduce emissions.
The two kinds of practice and capability tested here, with results reported in Table 4.13, are again management combined and technology combined, with capability computed with and without pre-IPC experience. All four partial correlations between emissions and practice-LBD capability interactions are negative as expected, although weakly for all but the technology interaction using the inclusive capability measure. These results are not inconsistent with Hypothesis 2 on the mediating role of capabilities with respect to the performance-practice relationship.
|Table 4.13 Partial Correlations: |
|Key emissions vs practice-static capability (LBD) interactions |
|(Probability values, number of cases in parentheses) |
| |Static capability |Static capability |
| |based on IPC period projects only |based on pre-IPC and IPC period |
| | |projects |
|Technology interaction+ |-.194* |-.234** |
|(Practice ( Static capability) |(.051, N=105) |(.018, N=105) |
|Management interaction++ |-.178* |-.180* |
|(Practice ( Static capability) |(.066, N=110) |(.068, N=106) |
|**Significant at 5% level; *10% (two-tailed). |
|+ Controls for year, and management practice and static capability. |
|++ Controls for year, and technology practice and static capability. |
|Partial correlations based on Spearman’s rho. |
Above we have tested jointly for the explanatory significance of both the practice and the related static capability accumulated through experience with that kind of practice. It is also important to explore the complementarity hypothesis by pulling apart the interaction and examining the respective roles of each, practice and capability. We approach this by looking in Table 4.14 at the partial correlations between emissions and both management and technology practice and static capability. Due to the cross-correlations, we control again each time for the effects of the other three variables. Hypothesis 2 suggests that the effect of practice on performance is mediated by static capability. In a partial correlation framework, this implies that some of the apparent effect of current practice on performance in fact reflects the role of an underlying capability. Therefore, for comparison, we reproduce in Table 4.14 the relevant numbers from Table 4.1 from the previous section, showing the practice-performance relationships with capabilities not included in the model. As noted earlier, we are focusing on emissions to avoid potential reverse-causality complications.
In moving from Table 4.1 to the expanded model, we expect to see the partial of practice reduced.[15] Meanwhile, by Hypothesis 1 we expect to see a negative partial correlation of static capability with emissions performance. However, because this model merely expands that reported in Table 4.13, where we saw the emissions-capability correlations while controlling for practices, we know that only one of four LBD capability measures meets that expectation. As before, the first column of new results in Table 4.14 shows the set of partial correlations for which LBD static capability is computed using only projects from the IPC years making up the panel, while the second column displays the partials based on LBD static capability incorporating pre-licensing projects as well.
|Table 4.14 Partial Correlations: |
|Key emissions vs practice, static capability |
|(Probability values in parentheses; N=126 unless otherwise indicated) |
| |Table 4.1 correlations using |Static capability |Static capability |
| |practice only |based on IPC |based on pre-IPC and |
| |(N=105) |period projects only |IPC period projects |
|Technology |-.196** |-.095 |-.070 |
|practice |(.047) |(.298) |(.483) |
|Technology | |-.041 |-.211** |
|static capability | |(.657) |(.033) |
|Management |-.179* |-.213** |-.210** |
|practice |(.071) |(.018) |(.021) |
|Management | |-.100 |.062 |
|static capability | |(.274) |(.495) |
|*Significant at 10% level. |
|Each partial controls for the effects of year and the other three variables on emissions. |
|Partial correlations based on Spearman’s rho. |
Looking at the first column of new results in Table 4.14, we find (in comparison with the numbers from Table 4.1) little support for Hypothesis 2 in explaining emissions performance when considering LBD static capability excluding pre-license period projects. Management practices still appear significantly correlated with reduced emissions, and management capability does not play a statistically significant role; and while the technology practice correlation is reduced, its static capability is also not significantly correlated with emissions. This changes for technology, although not for management, in the final column: Allowing its capability variable to capture prior experience permits the expected mediation effect to appear, with the role of practice independently (controlling for capability) reduced more sharply while LBD capability displays a significantly negative association with emissions performance.
Tables 4.13 and 4.14 shed some light, then, on Hypothesis 2: Static capability defined in terms of learning from experience acts as a complement with related technology practices in reducing emissions, when we account for prior experience alongside IPC experience. The apparent management practice-capability capability as modeled via interaction in Table 4.13 is apparently driven strictly by the direct effect of management practice on emissions. These results also suggest that LBD capability may take time to evolve. In what follows, as we turn to testing the role of dynamic capabilities in the development of static ones, we therefore focus on the LBD static capability measures that include pre-IPC licensing experience.
4.3.2 Dynamic capabilities
Hypothesis 3 asks whether dynamic capabilities, via their role in creating new static capabilities, helped determine which companies adapted best to the new regulatory regime. In Chapter 3 we suggested that dynamic capability (DC) involves identifying and processing new information both internally within the company and from its external environment. We defined an internal integration DC variable using EPA information on evaluative planning, workforce training, and cross functional and team work practices. This DC measure gives data coverage that is extensive across companies and years, comparable to that employed in the preceding tests of static capability. But it is not intensive, giving only fleeting glimpses of the relevant activities, because that is not what EPA’s reporting requirements set out to capture. Thus we will also look at a more intensively defined internal integration DC measure for the 16 companies that responded to our survey, based on number of key personnel involved in environmental management; percent of workforce receiving environmental training; frequency of team problem solving; and frequency of interdepartmental cooperation.
In addition, we have a survey based measure of external integration DC, incorporating number of outside parties (customers, vendors, and others) involved in environmental management; integration of vendors with own staff in managing new environmental technology; number of information sources on environmental issues; participation in ongoing stakeholder initiatives with community, NGO, or governmental bodies; memberships in professional associations; and number and longevity of formal certifications.
Our modelling strategy here is to divide the sample into halves: the higher and lower DC companies so defined. Then we re-run the complementarity tests from the preceding subsection, separately now for the higher and lower DC segments – using in turn each of the DC measures defined above. We expect that learning by doing will play a stronger role among the higher DC companies: A decrease in the relative performance-determining role of practice when static capability is partialed out, and a significantly negative association with emissions performance for static capability itself. Table 4.15 gives the results for the first set of partial correlations, segmented into those facilities with above and below median values for the EPA-based internal integration DC measure.
We are interested in how these numbers compare to the unsegmented static capability results from Table 4.14, all relative to Table 4.1’s practice-performance correlations with no capabilities represented. The results in Table 4.15 are mixed. First, we note (and will return to) the asymmetry in the number of observations for the two segments: The low DC half of the facilities according to the EPA based measure report far less data. For management, in the higher-DC segment Table 4.14’s unsegmented result continues, even more strongly: no complementarity but a very strong practice-performance link, independent of static capability. In the lower-DC segment, completely against expectation, a powerful complementarity effect appears: management practice’s association with lower emissions disappears (actually, is reversed), and there is a strong correlation between static management capability and lower emissions. While these results are interesting and will be discussed in the chapter conclusion, they do not conform to the complementarity hypothesis.
|Table 4.15 Partial Correlations: |
|Key emissions vs practice & static capability (pre-IPC experience included) – |
|For low, high internal integration DC segments (EPA data) |
|(Probability values in parentheses) |
| |Table 4.1, practice |Table 4.14, pre-IPC |Below-median EPA_INT |Median & above EPA_INT |
| |only |experience included |(N=19) |(N=86) |
| |(N=105) |(N=126) | | |
|Management |-.179* |-.210** |.340 |-.276** |
|practice |(.071) |(.021) |(.215) |(.012) |
|Management | |.062 |-.607** |.181 |
|static capability | |(.495) |(.016) |.103 |
|Technology |-.196** |-.070 |-.676** |-.087 |
|practice |(.047) |(.483) |(.006) |(.436) |
|Technology | |-.211** |.288 |-.221** |
|static capability | |(.033) |(.298) |(.046) |
|*Significant at 10% level. |
|Each partial controls for the effects of year and the other three variables on emissions. |
|Partial correlations based on Spearman’s rho. |
But the results for technology in Table 4.15 offer some support for Hypothesis 3. As predicted, there is no complementarity in the lower-DC segment; practice shows a strong association with improved performance independent of capability, which is not significantly correlated with performance. In the higher DC segment, the predicted complementarity relationships appear, with the independent effect of practice reduced and a significant association between static capability and lower emissions.
We now turn to the survey based internal and external integration DC segmentations. Unfortunately, here the numbers of observations are extremely low, and even where sufficient data exists for partial correlations to be calculated (all except for below-median external integration DC), there are no statistically significant associations between emissions and any practices or static capabilities. What does continue is the uneven pattern of data reporting: Segmenting by survey based internal integration DC, there are 10 valid observations from the 8 below-median facilities, and 15 from the 8 above-median facilities; and segmenting by survey based external integration DC, there are 3 valid observations from the 8 below-median facilities, and 22 from the 8 above-median facilities.
4.4 Conclusions
Taken together, our findings in section 4.2 suggest that companies under the IPC licensing program undertook management and technology changes, measured here as ‘practices,’ that in many instances had the desired effect of reducing environmental impacts. This is particularly true for impacts in the form of key emissions and – at the aggregate level – for both management and technology practices. Disaggregating management practices shows it to be procedural and planning-related management practice that is associated with reduced emissions; on the other hand, higher levels of training practice are unexpectedly associated with increased environmental impacts. Disaggregating technology practices alternately by each dimension of the technology matrix (see Chapter 3) reveals that loop closing and other process change approaches, and preparation and housekeeping stages, drive the association between greater practice and reduced emissions.
Unexpectedly, aggregated technology practice is positively correlated with the level of both waste and resource usage. Disaggregating shows that technology practice using the equipment investment approach and at the core production stage are both associated with higher emission levels. Similarly, we find many positive correlations between disaggregated technology practice and both waste and resource usage, again most consistently with respect to equipment investment at the the middle, basic processing stage. Using a variety of approaches, we explored the possibility that this unexpected relationship arises from a reverse causality scenario, with heavy environmental impacts stimulating higher levels of technology investment in core production areas. While the data cannot confirm this scenario, it is mostly consistent with it.
Breaking out the practice-performance results by sector sheds additional light on some of the above. At the level of aggregated practices, the positive technology-impact, reverse-causality phenomenon shows up most strongly in the wood products sector (across both waste and resource use impacts; we lack emissions data for these facilities), and not at all in paint and ink manufacturing. Perhaps related, in the paints sector (but not the others) aggregate management practice is associated with reduced emissions. In the disaggregated results the paints sector exhibits significant correlations between reduced emissions and management procedures and planning. Together with the sector averages reported in Chapter 3, these patterns may suggest that paint and ink facilities tended to be more consistently successful than those in metals and woods in deploying managerial and technological practices to reduce environmental impacts.
What do our results tell us about the nature, evolution, and role of organisational capabilities? Our results are generally consistent with the idea that not just each year’s technology practices are crucial to understanding performance (we tested this only for emissions), but also underlying learned-by-doing static technological capabilities. But this relationship is not revealed in the data with regard to management capability. In addition to affecting emissions directly, accumulated static technological capability also appears to complement ongoing technology practices in their effect on emissions performance. It appears that these relationships are somewhat stronger when pre-IPC period projects are accounted for in the static capability measures, suggesting that the experiential process of capability accumulation takes considerable time.
The results are weaker when we attempt to bring dynamic capability (DC) into the picture. We sought to explore whether dynamically capable companies were better learners by re-testing for the practice-static capability complementarity in low and high dynamic capability segments. Segmenting the sample using three DC measures – EPA data-based internal integration and survey based internal and external integration – revealed that companies below the median by all three report less data and hence offer fewer observations for testing. We infer from this that companies with higher levels of self-reported activity in systematically searching for and processing new information did better in meeting some basic demands of the (then) new IPC regulatory regime: tracking environmental impacts and reporting them along with practices thought to be relevant in reducing those impacts. This offers some indirect corroboration of our initial working hypothesis in this research, that dynamic capability would be important to adaptation and change.
On the other hand, the data provide only very limited evidence for our hypothesis that higher-DC facilities would exhibit stronger learning of new static capabilities. This evidence appears when measuring DC by the more plentiful EPA data on internal integration, and not surprisingly applies only to static capability in technology. But the survey based DC measures permit only very small numbers of observations to be used in the corresponding segmented models, and even where sufficient data exists for partial correlations be calculated, no statistically significant associations appear.
Chapter 5 – Economic Performance in Relation to Environmental Performance and Practice
5.1 Introduction
In this chapter, we examine the relationship between the environmental efforts and outcomes experienced by IPC licensees and their economic performance. Ultimately, we would like to know if the 1990s change in regulatory regime affected companies’ economic competitiveness, one way or the other. As noted in Chapter 2, many business leaders and scholars have argued that improving environmental quality must come at an economic cost, with a loss in efficiency among regulated companies. On the other hand, it is possible that companies that meet licensing requirements more effectively will reduce the regulatory cost burden, or even turn environmental performance into a cost advantage. It is also possible that top performing companies may translate environmental enhancements into a sales revenue advantage, tapping changes in demand for some kinds of products that have created a ‘green’ market segment.
2. Measuring Economic Performance
Because competitiveness involves a company’s standing relative to similar companies and others in the market, a full-fledged investigation is beyond the scope of this study. What we have done is to use Company Records Office (CRO) data to construct a measure of ‘operating efficiency.’ Operating efficiency is here defined as the ratio of annual earnings before interest, taxes, depreciation and amortisation to assets, where start and end of year assets are averaged to get a better representation of the scale of operation over the year the revenue and cost flows were generated.
For company i and year t,
Operating Efficiencyit = (Operating Profit + Depreciation + Other Amortisation) / Average Total Assets,
where ‘operating profit’ appears in the CRO Profit & Loss accounts with depreciation already subtracted, but before interest and taxes; and ‘average total assets’ is over years t and t-1. This is a measure of operating income generated by the assets; it is not affected by companies’ differences in capital structure, choice of depreciation schedules, or one-off charges. We can thus abstract from those factors and explore the relationships among economic performance and environmental practices and outcomes.
Unfortunately, many companies in the sample fall below CRO’s size threshold for required reporting of financial results. Only 29 have sufficient data to permit calculation of operating efficiency in any year, representing 115 facility-years in total. Their distribution across sectors is not too dissimilar to the sector proportions in the full sample: 12 in metal fabricating, 6 in paint and ink manufacturing, and 11 in wood products and preservation. Hence, the numbers of observations available for analysis are balanced but limited.
3. Economic Performance and Environmental Impact
A starting point is to examine the statistical relationship between operating efficiency and the measures of environmental impact discussed in Chapters 3 and 4 (‘environmental performance’). Table 5.1 presents the correlations, which are nonparametric for the reasons discussed in Chapter 4.
|Table 5.1 Nonparametric Correlations: |
|Environmental impact vs operating efficiency |
|(Probability values and number of observations in parentheses) |
| |Key emissions[16] | |Combined resource use |
| | |Total waste | |
|Average facility operating efficiency |.061 |-.025 |.123 |
|(Simple correlations, cross section only) |(.830, N=15) |(.900, N=27) |(.627, N=18) |
|Annual facility operating efficiency |.101 |-.076 |.228* |
|(Partial correlations, controlling for year) |(.446, N=57) |(.475, N=89) |(.073, N=61) |
|Based on Spearman’s rho. |
|*Significant at the 10% level (two-tailed). |
The first row of results looks solely across companies, with each facility’s economic and environmental variables averaged over its years in the panel. In the second row, the full panel data set is used, pooling the cross section over companies with the time series over years for each facility. The pooled results are partial correlations between operating efficiency and each environmental measure in turn, controlling for (holding constant) the effect of year. This is important, because both profitability and the environmental variables trend weakly downward over time, for reasons which probably include factors unrelated to the profitability-environmental impact relationship.
Table 5.1 offers no evidence that higher economic performance accompanies better environmental performance per se, which would imply statistically significant negative correlations. The only statistically significant correlation is positive: Higher operating efficiency is weakly associated with heavier resource usage. These results are not consistent with the notion that ‘it pays to be green,’ and in the case of resource use fit with the idea that it will be costly to reduce impacts.
Nevertheless, it is possible that the simple specifications in Table 5.1 mask some relationships of concern. For example, both environmental impacts and profitability may differ by sector, for reasons that go beyond relative facility performance and have to do with basic characteristics of products, processes, and markets; these characteristics can interact with the relationship between impact and profitability at the firm level, but a simple a cross-sectoral sample can confound the relationships. In addition, whether impacts are high or low, IPC licensing and other forces are pushing companies to reduce them, and we would like to know how economic performance is affected by changes in – not just levels of – environmental performance. We take these possibilities one at a time.
Table 5.2 breaks out the correlations from Table 5.1 by sector.
|Table 5.2 Partial Correlations By Sector: |
|Environmental impact vs operating efficiency (controlling for year) |
|(Probability values, observations in parentheses) |
|Sectors |Impacts |
| |Key |Total |Combined |
| |emissions |waste |resource use |
|Metal |.302* |.104 |.338* |
|fabrication |(.055, N=42) |(.502, N=45) |(.068, N=31) |
|Paint & ink |-.375 |-.491** |-.401 |
|manufacturing |(.138, N=18) |(.033, N=20) |(.251, N=11) |
|Wood products |No data |.008 |-.335 |
|& treatment | |(.968, N=27) |(.138, N=22) |
|***Significant at 1% level; **5%; *10% (two-tailed). |
|Based on Spearman’s rho. |
What we see here is that the link between higher profitability and heavier resource use is driven by facilities in the metal fabricating sector. Indeed, this sector also shows that positive association in emissions as well. All the partial correlations for the paints sector and one of two for woods are negative, though only one is statistically significant: in paints, lower levels of waste are associated with higher profitability. These sectoral differences are interesting and potentially significant from a regulatory perspective.
We now switch from levels to changes in environmental impact, testing the association between operating efficiency and change in impact from the previous year. Table 5.3 shows the results, for the full sample and the sectors separately. For the sectors combined, there is a significant association between year-on-year reduced waste and higher profitability. In the sectoral breakouts, where in many of the cells the numbers get very small, the full-sample relationship in waste is shown to be driven by wood products facilities. In addition, looking at one-year impact changes we do not find the higher impact-higher profitability association that appeared for metals with respect to impact levels. It may be that many of the metal fabricating facilities are mitigating environmental problems without systematically (across the facilities) affecting profitability one way or the other.
|Table 5.3 Partial Correlations: |
|Changes in environmental impact vs operating efficiency by sector (controlling for year) |
|(Probability values, observations in parentheses) |
|Sectors |One-year changes in Impacts |
| |Change in key |Change in total |Change in combined |
| |emissions |waste |resource use |
|Full |.140 |-.215* |-.118 |
|sample |(.352, N=47 |(.076, N=70) |(.423, N=49) |
|Metal |.153 |-.082 |-.226 |
|fabrication |(.403, N=33) |(.636, N=37) |(.311, N=23) |
|Paint & ink |.217 |-.069 |-.468 |
|manufacturing |(.477, N=14) |(.824, N=14) |(.290, N=8) |
|Wood products |No data |-.574** |.369 |
|& treatment | |(.013, N=19) |(.145, N=18) |
|***Significant at 1% level; **5%; *10% (two-tailed). |
|Based on Spearman’s rho. |
Taken in combination, the results shown in Tables 5.1-5.3 suggest that, for these IPC-licensed Irish manufacturers, whether environmental performance exacts an economic toll depends: on whether we look at levels of or changes in environmental impact, and on the industry sector. Adding these refinements moves us progressively away from Table 5.1’s suggestion that green is costly. It may be, but only in the metals sector, and only with respect to impact levels and not impact improvements. Low and reduced waste, in particular, tends to be associated with higher profitability. However, none of these correlations tell us anything directly about companies’ greening efforts themselves, and what effect those efforts may have had on operating efficiency. We explore this question next.
4. Economic Performance and Organisational Response to IPC Licensing
We start by considering the relationships between economic efficiency and companies’ IPC-related management and technology practices. The practice measures are defined in Chapter 3: management practices in procedures, planning, and training; technology practices distinguished either by pollution prevention approach (materials substitution, loop closing, equipment investment, other process change) or production stage (product design, preparation, basic production, finish work, housekeeping); and for each a composite summed across the categories. We look first at the composite management and technology practice variables.
The relationships between operating efficiency and the composite practice measures are shown in Table 5.4.
|Table 5.4 Partial Correlations: |
|Environmental practice vs operating efficiency |
|(Probability values in parentheses) |
| |Management composite |Technology composite |Technology |
| | | |lagged one year |
|Annual facility operating efficiency |-.086 |.227** | |
|(Controlling for year and the |(.439) |(.039) | |
|other practice) |N=85 |N=85 | |
|Annual facility operating efficiency |-.035 | |.261** |
|(Controlling for year and the |(.759) | |(.021) |
|other practice) |N=80 | |N=80 |
|**Significant at 5% level (two-tailed). |
|Based on Spearman’s rho. |
Each partial correlation in Table 5.4 controls for the effect of time and the other practice variable, which is important due to the positive correlation between management and technology practices. While there is no significant statistical association between economic efficiency and management practice, the link with technology practice is positive and statistically significant: For facilities and in years with high levels of technology practice, operating efficiency is high (controlling for the effects of management and year). It is possible that the direction of causation runs the other way, with stronger economic results permitting more technology expenditure. We control for this by lagging technology practice one year, since last year’s technology expenditures cannot be directly attributable to this year’s cash flow. The final column in Table 5.4 shows that the association is, if anything, a bit stronger for lagged technology practice.
This result does not necessarily support the contention that it pays to work at being green. We uncovered in Chapter 4 the possibility that more serious environmental impact problems stimulate increased technology practice, rather than the causal arrow running from practice to environmental performance. This linkage appeared most consistently for equipment-related practice, for impacts in resource use, and for the woods and metals sectors. But in Tables 5.1 and 5.2, above, we find that (in the metals sector, at least) higher profitability can be associated with higher levels of resource use. If these two dynamics are strong enough, they may confound the interpretation of Table 5.4 by creating the spurious appearance of a causal link between higher levels of technology practice and profitability. Both may, in fact, be showing the effect of their relationships with environmental impacts.
A simple way to address this possibility is to add a control for environmental performance, to hold it constant in looking at the environmental technology-economic efficiency relationship. If the apparent association between technology practice and profitability is being affected by the relationship of each with impacts, then this should remove that effect and allow the relationship of interest here (if any) to show through. We have repeated the calculations shown in Table 5.4 three times, each time controlling for (partialling out) a different environmental impact: key emissions, total waste, and resource uses. The results are mostly unchanged; to conserve space we do not show them here (the full results are available upon request). As in Table 5.4, management practice exhibits no significant correlations with operating efficiency in any iteration. Of the six technology practice correlations with operating efficiency (three lagged and three not), five remain significantly positive; the coefficient for lagged technology with the resource use control is positive but not statistically significant. It thus appears unlikely that Table 5.4’s positive link between profitability and composite environmental technology practice is spurious and reflects instead the connection between each and environmental impacts.
Further exploration of this issue is possible – for example, disaggregating by sectors and by kinds of technology practice in addition to by impact type as above. But the number of scenarios to be explored becomes very large, and we leave that extension for future research.
On the other hand, the relative effectiveness of specific kinds of management and technology practices is a matter that is central to this study, and here we would like to get a finer-grained look at the aggregate technology-profitability relationships shown in Table 5.4. Chapter 4’s analysis identifies specific kinds of environmental practices that are ‘best’ in the sense of most strongly associated with better environmental performance, and others that appear to have been undertaken defensively in response to poor environmental performance. It will help us understand the relationships among practices and both environmental and economic performance to see how those specific practices are related to efficiency. We know that there are various associations between environmental impacts and both efficiency and technology practices, and thus it is appropriate to continue controlling for impacts here, as we began in the paragraphs above. (Since management practices are shown above to be uncorrelated with efficiency, we do not control for them in what follows.)
We begin by disaggregating technology practice to the level of pollution prevention approaches. Table 5.5 reports the results, controlling for each environmental impact in turn, and focusing on unlagged technology since the discussion above suggests (and we have confirmed) that the lags do not affect the outcome.
The results in Table 5.5 suggest that the positive relationship between technology practice and operating efficiency may be driven in terms of approaches by closing the loop to capture and re-use product and/or materials that had previously become waste. On the other hand, substitution of more environmentally benign raw materials may be costly. Given that the woods facilities drop out when controlling by key emissions (see footnote), these results suggest that this effect may concentrate in that sector, where switching to less toxic preservatives has been the critical substitution. Equipment investment, which Chapter 4’s results suggest is often defensive in nature, is not significantly related to operating efficiency.
|Table 5.5 Partial Correlations: |
|Operating efficiency vs technology practice by approaches |
|(Probability values in parentheses; N=57) |
|Technology approach |Environmental impact controlled for: |
| |Key emissions[17] |Total waste |Resource use |
| |(N=59) |(N=92) |(N=64) |
|Raw materials (controlling for year, |.096 |-.226** |-.282** |
|impact, and other approaches) |(.490) |(.035) |(.030) |
|Closing loop (controlling for year, |.336** |.188* |-.042 |
|impact, and other approaches) |(.013) |(.081) |(.753) |
|Equipment (controlling for year, |.078 |.144 |.110 |
|impact, and other approaches) |(.574) |(.184) |(.405) |
|Process, NEC (controlling for year, |.003 |.001 |.155 |
|impact, and other approaches) |(.980) |(.991) |(.242) |
|*Significant at 10% level (two-tailed). |
|Based on Spearman’s rho. |
Next we disaggregate technology practice by stage in the production process. Table 5.6 gives the results. In terms of production stages, the positive association between economic efficiency and aggregate technology practice reflects rather strong positive correlations at the core (basic) production stage, along with housekeeping-type practices for the metals and paints facilities (since the woods sector drops out when controlling for emissions). The latter suggests the economic potential of the range of practices capture here, such as recycling, spill prevention, and the like. On the other hand, technology practice in product design is associated with reduced profitability controlling for waste or resource use; this corresponds to Table 5.5’s finding for materials substitution and again is likely driven, we believe, by preservatives change in the woods facilities that are excluded in the emissions controls. (Adopting less toxic preservatives is coded as a materials substitution approach at the design stage.) Finally, in two of three cells practice at the preparation stage is apparently costly, showing a negative correlation with operating efficiency; this result is interesting in light of Chapter 4’s finding that practice at that stage is strongly associated with better environmental performance.
|Table 5.6 Partial Correlations: |
|Operating efficiency vs technology practice by production stages |
|(Probability values in parentheses; N=57) |
|Technology stage |Environmental impact controlled for: |
| |Key emissions[18] |Total waste |Resource use |
| |(N=59) |(N=92) |(N=64) |
|Product design (controlling for year, |-.116 |-.385*** |-.503*** |
|impact, and other stages) |(.408) |(.000) |(.000) |
|Preparatory work (controlling for year, |-.399*** |-.306*** |.014 |
|impact, and other stages) |(.003) |(.004) |(.920) |
|Basic production (controlling for year, |.302** |.510*** |.172 |
|impact, and other stages) |(.028) |(.000) |(.196) |
|Finish work (controlling for year, |.059 |-.193* |-.075 |
|impact, and other stages) |(.677) |(.076) |(.575) |
|Housekeeping/other (controlling for |.489*** |.127 |-.065 |
|year, impact, and other stages) |(.000) |(.243) |(.628) |
|***Significant at 1% level; **5%; *10% (two-tailed). |
|Based on Spearman’s rho. |
5. Economic Performance and Organisational Capabilities
5.5.1 Static capability
One of the core questions in this research, investigated for environmental performance in Chapter 4, is whether the ability to do certain kinds of things well in a given competitive environment – a ‘static’ organisational capability – can complement the practice-performance relationship by making particular practices more effective. We found there some evidence that as experience accumulates with more projects and the passage of time using them, sample facilities did create static capabilities in technology but not management; the underlying level of capability helps explain some of the apparent effect of a given technology practice on environmental performance. An important question is whether a similar effect might extend to the economic performance associated with practices undertaken in response to IPC licensing. We narrow the question, as above, and in this section look at whether static capabilities as measured contribute to these companies’ ability to implement environmental practices in ways that enhance operating efficiency.
Our approach, as in Chapter 4, will be two-pronged: first, to proxy the complementary effect using a multiplicative practice-capability interaction term; and second, to use partial correlation to control for the respective effects of practice and static capability on efficiency. Table 5.7 shows the results for the first. Again, due to the strong management-technology cross correlations, we use appropriate controls for each. Finally, since the static capability measure computed to include the effects of pre-IPC licensing experience proved in Chapter 4 to be somewhat more potent, that is the version we use in Chapter 5.
|Table 5.7 Partial Correlations: |
|Operating efficiency vs practice-static capability interactions |
|(Probability values, number of cases in parentheses) |
| |Correlation with operating efficiency |
|Management interaction+ |.008 |
|(Practice ( Static capability) |(.943, N=92) |
|Technology interaction++ |.361*** |
|(Practice ( Static capability) |(.001, N=84) |
|***Significant at 1% level (two-tailed). |
|+ Controls for year, and management practice and static capability. |
|++ Controls for year, and technology practice and static capability. |
|Partial correlations based on Spearman’s rho. |
Table 5.7 shows that the management interaction variable is uncorrelated with operating efficiency. The technology interaction, however, is strongly, positively correlated: Given some level of static technology capability, increased practice is associated with increased efficiency, and given some level of practice, greater capability is associated with higher efficiency. These findings dovetail with those in Table 5.4; here the effect of technology practice on operating efficiency is amplified when its corresponding static capability is taken into account in this form.
We now move to the second approach to testing the complementary role of capabilities, pulling them apart from practice to examine the independent impact of each on operating efficiency in a partial correlation setting. We want to compare these results directly with those in Tables 5.4, where the association between environmental practice and efficiency were tested without incorporating a role for static capabilities. The first column in Table 5.8 reproduces the numbers from Table 5.4 for easy comparison.
Neither of the management variables is significantly correlated with operating efficiency here, consistent with the practice-only result in the first column. Technology practice, on the other hand, is a bit more strongly correlated with increased efficiency; this is not what is predicted by the partial correlation approach to capturing complementarity, which would suggest that with the effect of the underlying complement (capability) now controlled or held constant, the measured impact of the practice variable would actually decline. At the same time, the results here show that when measured independently, each controlling for the other, both technology practice and static capability are rather strongly, positively associated with economic performance. Each seems to play a role, and higher levels of both are linked with greater operating efficiency.
|Table 5.8 Partial Correlations: |
|Operating efficiency vs practice, static capability |
|(Probability values in parentheses; N=126) |
| |Table 5.4 results: |Static capability |
| |Controlling for other |based on pre-IPC and IPC period |
| |practice and year (N=85) |projects (N=84) |
|Technology |.227** |.280** |
|practice |(.039) |(.012) |
|Technology | |.252** |
|static capability | |(.024) |
|Management |-.086 |-.100 |
|practice |(.439) |(.380) |
|Management | |-.150 |
|static capability | |(.184) |
|**Significant at 5% level; *10% (two-tailed). |
|Each partial controls for the effects of year and the other three variables on efficiency. |
|Partial correlations based on Spearman’s rho. |
5.5.2 Dynamic capability
We are also interested in whether companies with specialised organisational capabilities for change – dynamic capabilities – developed stronger static capabilities during the IPC licensing period. Possible sources of this enhanced learning-by-doing effect, as described in Chapter 3 and tested for environmental performance in Chapter 4, are integration with sources and uses of new information, both internally and externally with respect to organisational boundaries. This sub-section tests whether our data are consistent with dynamic capability with respect to economic performance.
For each dynamic capability measure, we segment the facilities into higher and lower dynamic capability groups. The tests then proceed by replicating in each segment the two steps in 5.5.1, using the static capability-practice interaction term and then separating the two. Table 5.9 gives the results for the interaction terms. We omit the versions with controls for key emissions, because as noted above they do not materially affect the results.
As in the unsegmented results reported in Table 5.7, the management practice-static capability interaction variable shows no significant correlations with operating efficiency in any of the dynamic capability segments. This changes for technology. Using both of the internal integration dynamic capability measures, the technology practice-static capability interaction is correlated with efficiency in the high but not the low dynamic capability segments – as predicted. (The correlation for the high external integration group is also quite high, but given the small number of observations it does not meet statistical significant The efficiency-increasing interaction
|Table 5.9 Partial Correlations: |
|Operating efficiency vs practice-static capability interaction variables, |
|low vs high dynamic capability (DC) segments |
|(Probability values in parentheses) |
| |Internal integration DC, EPA|Internal integration |External integration |
| |data |DC, survey data |DC, survey data |
|Technology interaction |.191 |-.018 |.087 |
|(Practice ( Static capability) |(.351) |(.952) |(.869) |
|Low DC segment |N=29 |N=17 |N=9 |
|Technology interaction |.348** |.745* |.329 |
|(Practice ( Static capability) |(.011) |(.089) |(.250) |
|High DC segment |N=55 |N=9 |N=17 |
|Management interaction |.111 |-.066 |.117 |
|(Practice ( Static capability) |(.538) |(.823) |(.825) |
|Low DC segment |N=36 |N=17 |N=9 |
|Management interaction |.075 |-.467 |-.202 |
|(Practice ( Static capability) |(.597) |(.350) |(.488) |
|High DC segment |N=55 |N=9 |N=17 |
|**Significant at 5% level; *10% (two-tailed). |
|Each partial controls for the effects of year and the other practice and static capability. |
|Partial correlations based on Spearman’s rho. |
between technology practice and static capabilities that were acquired during the IPC licensing period, shown in Table 5.7, appears here to be driven by those facilities with greater within-firm informational search and processing capability.
We now separate the practice and static capability components. Table 5.10 replicates the results of Table 5.8, but broken into low and high dynamic capability groups of facilities for each of the three measures. It is difficult to see any coherent patterns in the numbers shown in Table 5.10. There are a few very large correlations, but they neither are consistent with our dynamic capabilities hypothesis nor suggest any particular alternative explanations. In the survey-based measure columns, the numbers of observations are very small, and we are asking more of this data in terms of control variables than we did in the interactions results in Table 5.9. We can only conclude that these results are not supportive of the notion that companies’ learning of new static capabilities varied by dynamic capability level.
|Table 5.10 Partial Correlations: |
|Operating efficiency vs practice and static capability, |
|low vs high dynamic capability (DC) segments |
|(Probability values in parentheses) |
| | |Internal integration DC, EPA |Internal integration DC, |External integration |
| | |data |survey data |DC, survey data |
|Low DC |Technology |.260 |-.026 |.888** |
|segment |practice |(.210) |(.934) |(.044) |
| | |N=29 |N=17 |9 |
| |Technology |.239 |-.111 |-.913** |
| |static capability |(.249) |(.718) |(.030) |
| | |N=29 |N=17 |N=9 |
|High DC |Technology |.336** |.731 |.365 |
|segment |practice |(.016) |(.160) |(.219) |
| | |N=55 |N=9 |N=17 |
| |Technology |-.036 |.447 |.122 |
| |static capability |(.802) |(.450) |(.691) |
| | |N=55 |N=9 |N=17 |
|Low DC |Management |.009 |.399 |-.811* |
|segment |practice |(.965) |(.177) |(.096) |
| | |N=29 |N=17 |N=9 |
| |Management |-.364* |-.458 |.904** |
| |static capability |(.074) |(.116) |(.035) |
| | |N=29 |N=17 |N=9 |
|High DC |Management |-.033 |-.671 |.007 |
|segment |practice |(.819) |(.215) |(.981) |
| | |N=55 |N=9 |N=17 |
| |Management |.197 |-.028 |-.041 |
| |static capability |(.166) |(.964) |(.895) |
| | |N=55 |N=9 |N=17 |
|**Significant at 5% level; *10% (two-tailed). |
|Each partial controls for the effects of year and the other three variables on emissions. |
|Partial correlations based on Spearman’s rho. |
5.6 Conclusions
Given the limited number of companies whose financial data permitted calculation of operating efficiency, this examination of the relationship between IPC responses and economic performance has produced some interesting results. The relationship between performance in the environmental and economic realms depends on the sector and the environmental impact measure: In metal fabricating, there is an association between higher levels of impact and of profitability in two of the three impact measures; but this is not true of the other two sectors, where for paint and ink facilities better performance environmentally and economically tend to go hand in hand. This more sanguine note continues if rather than environmental impact levels we look at year-to-year changes. Efficiency is not correlated along any dimension with increases in impacts, while for waste generation in particular it is associated with reductions in impact.
In addition, facilities’ environmental technology practices are in many instances associated with higher levels of operating efficiency. Disaggregating technology practice indicates that the association may be driven by technology practices in loop-closing approaches and in the core production and housekeeping process stages. On the other hand, there are also practices that correlate with reduced profitability. Raw material substitution and product design have this effect, which is likely to be concentrated in the wood products sector where, apparently, the switch-over to less toxic preservatives has been costly.
It appears that static capabilities related to environmental technology practice emerged during the panel period and complemented the positive effect of aggregate-level technology practice on economic performance. Modelling this as a multiplicative practice-capability interaction variable, and also separating the two through partial correlation, both produce results consistent with the role of static capabilities. The evidence is far weaker for dynamic capabilities. Dividing sample facilities into low and high dynamic capability groups and replicating the interaction variable tests does suggest that the higher dynamic capability companies exhibit stronger static technology capability effects, i.e., were better learners. But repeating this for the separate practices and static capabilities partial correlations provides no support for the dynamic capabilities hypothesis.
On balance, the results reported here mimic the broad findings in the research literature that whether it pays to be green depends on many factors, including industry sector and what dimension of environmental performance is of concern. The methodologies presented in this study for measuring and testing the variables of interest allow us to identify at a fairly fine-grained level what those contingencies are for this group of facilities. Our understanding can be further nuanced by bringing to bear results from mail-out survey responses and in-depth case interview with a smaller number of companies, to which we now turn.
Chapter 6 – Case and Survey Data Analysis
6.1 Introduction
Case study and survey research was used to complement the larger scale statistical analysis. The postal questionnaire was modelled on research questions and hypotheses stated in Chapter 1 as a means of gathering richer company-specific data in addition to the information gathered from the EPA files and publicly held financial records. Its goal was to elicit additional information about companies’ sources and processes of decision-making regarding environmental management. Quantitative analysis identifies systematic statistical associations among sets of variables as defined and measured. It can suggest whether the data as gathered and modelled is ‘inconsistent with’ particular explanations of how and why things happen; however, statistical analysis does not provide such explanations. To assist in understanding the implications of the quantitative analysis, case studies of thirteen companies were undertaken.
The cases provide the best available look at the role of dynamic capability, because (as noted in the preceding chapters) we have little statistical data in this area. The issues it entails, moreover, are well suited to qualitative and contextual examination. This analysis allows us to draw out lessons about what makes for successful environmental management within companies, and how this may be fostered by companies themselves, as well as supported by the regulator.
6.2 Data Collection
6.2.1 Postal Survey Data Collection
The survey was directed towards the individual within the company who acted as supervisor of the company’s IPC license and key contact with the EPA. A special effort was made to counteract non-response and/or inaccurate or incorrect data being collected. With this in mind, the layout, format and presentation of the questions was simple and consistent, and clear instructions were given on how long it would take to complete, and how to respond to the questions.
The questionnaire was divided into three distinct sections, namely:
▪ Management, which focussed on managerial and strategic issues arising from environmental pressures;
▪ Technology, which posed questions on the technological implications of on-site environmental improvements; and
▪ Competitiveness, which explored the impact of the environment and IPC licensing on a company’s ability to compete successfully in the market.
A large proportion of the questions were open response questions; specifically we asked whether certain aspects of management, technology and competitiveness within the company had changed during the years the company had held its IPC license. This was a deliberate attempt to encourage a greater degree of accuracy and information from respondents.
Several steps were taken to boost the response rate to the questionnaire. Personalised covering letters to the appropriate individual explained why the research was important and the significance of the contribution they were making to the research project by filling out the questionnaire. Also, given the level of information available on each company prior to them being approached, each questionnaire was tailored to the company’s specific sector. Following a lag period of one month, companies which had not responded were then contacted directly by phone, which resulted in the returning of five additional questionnaires.
The questionnaire was mailed out to all companies in the three sector samples (78 in total), regardless of whether the company had ceased trading prior to the survey date of July 2006. This was done to eliminate any bias against companies that had closed down, and also in the hope that the survey might be returned by a former employee but, as might be expected, we did not receive any responses from companies that had ceased trading. Table 6.2 details the response rate from the sample (18 responses applicable to 21 sample companies[19]).
Table 6.1 Survey Response Rate
|Sector |No. replied to|Response rate from |Response rate excluding|Response rate from |Response rate from those used|
| |survey |full sample |closed companies |those used in stat. |in stat. analysis excluding |
| | | | |analysis |closed companies |
|Metals |7 |24% |33% |29% |35% |
|Paints |5 |28% |50% |38% |50% |
|Woods |9 |29% |30% |20% |21% |
|TOTAL |21 |27% |34% |27% |31% |
3 Case Study Data Collection
Case findings were analysed to illuminate the model of environmental and economic performance suggested by the statistical analysis. The purpose of the case studies is to draw out the relationships between (i) dynamic capability, that is, processes for driving and implementing continuous environmental improvement, (ii) technology and managerial practices and (iii) performance outcomes, both environmental and economic. Companies for case study were chosen for variation in industry sector, and in results (some that exhibit the combination of high performance and explanatory variables identified in the statistical results, and some that do not); and for willingness of management to cooperate with researchers. Interviews and study of relevant company materials were conducted for five paint companies (50% of companies still operating) and eight metal companies (38% of companies still operating) in May-June 2007. No woods companies were interviewed. For further details of the exact variable values and sector rankings for the case companies please see Appendix 2. For details of the measures applied to the performance of these companies, refer back to chapter 3.
Table 6.2 Interview Response Rate
|Sector |Number interviewed |Response rate from full sample |Response rate from those used in statistical |
| | |N = 78 |analysis |
| | | |(excluding closed) |
| | | |N = 51 |
|Paint |5 |28% |50% |
|Metal |8 |28% |47% |
|Wood |0 |0% |0% |
|TOTAL |13 |17% |25% |
6.3 Findings on Dynamic Capability in Environmental Management and Cleaner Technology
1 Introduction
One-to-one interviews were conducted with senior managers responsible for environmental management in thirteen sample companies. We will refer to these managers by the generic term Environmental Manager (EM), although in practice it is very rare for a company to have a dedicated EM. Usually these managers hold other technical responsibilities, such as operations, production, health and safety, quality assurance. In the interviews we asked managers about key cleaner technology projects, identified from the company’s AER reports, and then probed further to explore the management processes that supported the identification and implementation of that project. We also discussed the company’s experiences in managing its environmental responsibilities.
2 The cases
The tables below present a summary of the findings from each case, allowing an overview of the particular factors at work in each individual company underpinning environmental performance, management and cleaner technology adoption. The tables synthesise the findings from our analysis of data held by the EPA with the interview data. This is followed by a thematic discussion of the findings.
Table 6.3 Performance, Practices and Capability in Case Company PAINT1
|Strong environmental performance |Strong routinised approach to continuous environmental improvement |
|( Very low emissions/waste, relative to sector |( Structured planning – many projects to investigate possible clean technologies |
|( Zero pollution non-compliances |( Cross-functional teams to develop/implement projects |
|( Positive relationship with Agency; licence seen as providing structure and supporting company ethos of |( Regular meetings to pursue improvement projects |
|transparency |( Goals and targets to direct improvements |
|( Impact of licence is managed, and so not constraining |( Learning from outside experts |
|( High levels of technology and management practices; best in sector | |
| |This is one of the largest companies in this sector, and has the benefit of a corporate parent and access|
|Cleaner technology projects: |to corporate resources. However, the strength of this company’s environmental management cannot be |
|Paint1 had a high volume of wastewater that was proving difficult for their WWTP to handle. They faced |attributed solely to these advantages, and much of the good practice evidenced in this company is |
|the possibility of non-compliances or a substantial investment in new waste treatment equipment. Instead |developed on site. |
|they chose to investigate ways to reduce the volume of wastewater washings generated by cleaning tanks |‘We have a policy of involving all the relevant people in projects from the shop floor to managers |
|between batches of paint production. The solution involved developing three different new procedures, |because we have found in the past you can’t get anything implemented, or you can’t get people on the shop|
|each of which contributed to reduced wash volumes. (i) Water used for washing out tanks is not treated |floor to buy into projects if they are not involved in them themselves. And we have found that it is an |
|but sent to a holding tank from where it can be reused in another batch of paint production. This process|excellent way of generating ideas and getting people to implement their own ideas so it is imperative |
|change reduces the use of virgin water and reduces the volumes going for treatment. It is not a complete |that you involve everybody in projects like that.’ |
|solution though, as the amount of wash water that can be added without affecting product quality is | |
|limited. |Improvement projects like this one are identified and developed at regular Safety, Health and Environment|
|(ii) Water washings from the production of dark paint cannot be reused in this way: a specialist recovery|committee meetings held every two months. The company has both a shop-floor and a management committee, |
|company has been identified to handle these wastes. |to maximise involvement and problem-solving, providing both a bottom-up, operational consideration of |
|(iii) Reuse the wash wastes as an input into another company’s production; wastewater is taken by a local|issues, as well as a more strategic view. Both committees actively look for environmental improvement |
|concrete manufacturer and used to colour their products. They helped the concrete company secure a county|projects, reflected in the large number of planning projects (25 in total) Paint1 initiated from the time|
|council licence for the use of the waste washings and installed the necessary infrastructure for them. |of licensing. They also have a formal system that allows all employees to flag up any potential hazards |
| |that they have identified: ‘it heightens people’s perceptions and attention to detail and in doing so it |
| |reduces the potential of either injury or loss of containment etc. And we have trained everybody and this|
| |training is also included in our induction.’ |
| |In achieving their goal of managing the load on the WWTP and reducing water use, the company drew |
| |extensively on internal technical knowledge to generate possible approaches. They also made good use of |
| |external advice. The project involved a team of six people including department heads from manufacturing,|
| |laboratory, and site resources and also the Clean Technology Centre at Cork Institute of Technology. And |
| |in implementing the solution they were able to use their knowledge of the regulatory process, as well as |
| |providing technical support, to help with the integration of an alternative use for their waste. |
| |The company has also learnt that developing its own capacity is important: they used to receive outside |
| |help in completing their AERs, and struggled to manage this when the help was no longer available. |
| |The company has a positive perception of IPC licensing, despite finding the administrative requirements |
| |challenging at times. This may because of the environmental sensitivity of their site, for which they |
| |have a strong feeling of stewardship; ‘‘the integrity of that river would be uppermost in our minds all |
| |the time.’ Environmental management is accepted very much as a part of doing business: ‘we as a company |
| |understand that without an IPC licence, we wouldn’t be able to manufacture on this site.’ |
Table 6.4 Performance, Practices and Capability in Case Company METAL5
|Strong Environmental Performance |Moderately routinised approach to continuous environmental improvement |
|( Very low emissions and waste, relative to sector |( Company uses cross functional teams; wide involvement seen as key to developing green reputation |
|( BATNEEC compliant at time of licensing |( Deliberate efforts to access external information |
|( Very low pollution non-compliances |( Measurement of impacts used to drive projects |
|( EM seen as key element in meeting customer demands and also being cost-competitive |( Licence used to drive change |
|Moderate level of technology projects | |
|Weaker management practices |Metal5 is a subsidiary of an international company. Environmental management has a strongly corporate |
| |emphasis, and major projects include a company-wide initiative to build a green brand, with a focus on |
|Clean technology projects: |environmental products and a reputation for environmental performance. |
|Metal5 set an ambitious target of eliminating chemical waste and eliminating disposal to landfill waste. |They have an established process for problem-solving, using cross-functional teams. Metal5 regularly made|
|This resulted in 20 tonnes per annum or 70 per cent of chemical waste being reused back into other |use of teams and inter-departmental cooperation on environmental projects. The company used teams from |
|products rather than disposed off. Recycling alternatives for all the main landfill wastes were |different departments and across sites (chemical, mechanical, environmental, H&S, marketing etc.) to work|
|identified. The project was given a high profile, and the recycling drive is displayed prominently around|on new product design and production technologies modifications projects. They meet every two months to |
|the site to increase awareness and engage staff participation. |discuss progress on the new product/project. Pressure from customers to produce more environmentally |
|The company is also working on developing environmentally efficient products, produced in an |friendly products had lead to more teamwork to achieve this. The company emphasises involvement across |
|environmentally responsible manner to develop a green brand reputation. |the whole staff, as well as top management involvement. |
| | |
Table 6.5 Performance, Practices and Capability in Case Company PAINT4
|Fair Environmental Performance |Emerging evidence of routinised approach to continuous environmental improvement |
|( Average emissions and waste, relative to sector (inherent to nature of production process) |( Planning ‘happens organically’; team work rarely used |
|( Waste non-compliances; prosecution |( Some forward investigation of projects |
|( Demanding relationship with Agency |( Extensive use made of external advice/information to compensate for limited internal resources |
|( Environmental management not a company priority; cost a significant driver of environmental actions |( Struggle to get commitment to environmental management at highest level |
|Low levels of technology and management practices | |
| |Paint4 has a production mix that is relatively low in its environmental impacts and hence ‘we don’t have |
| |a very onerous or strenuous licence … it isn’t that difficult conforming.’ Despite this, or perhaps |
|Cleaner technology projects: |because of it, environmental management historically had a low profile with management, leading to |
|Paint4 also carried out investigation into ways of reducing the volume of wastewater from the washing of |neglect of procedures and prosecution for waste non-compliances. |
|production equipment. With a smaller volume of business than Paint1 there is a longer production cycle |The company has started to give greater priority to environmental issues, albeit ‘we were starting at a |
|between batches of a given colour, so resuing the washings back nto paint production would require the |very low base to be honest with you.’ They are addressing environmental performance on a number of |
|segregation and storing of the wastewater for long periods; the lack of scale made this approach |fronts, moving from resolving the source of the non-compliance to changing procedures to ensure this is |
|impractical. The company developed a product that reused wash waste, which they were able to manufacture |not repeated. |
|and sell for a lower price because of the savings on disposal costs. However, this product is subject to |Initiatives include: accrediting their EMS with ISO14001; conducting an energy audit and tracking energy |
|greater colour variations between batches than standard products and is not popular for that reason; it |consumption per litre of output to identify trends and problem areas; working to instil environmental |
|does not sell in high enough volumes to dispose of all the wastewater from washings. |awareness with the workforce; changing processes with suppliers. |
|The company also made changes to reduce disposal of contaminated packaging from raw materials. They |Although the company does not have formal processes for driving continuous environmental improvement, ‘I |
|negotiated with one supplier to start reusing drums used to deliver materials, and where possible have |suppose it grew organically rather than somebody sitting down … it just sort of happened and now it is in|
|switched to reusable integrated bulk containers. This has reduced waste to landfill by 75 per cent, from |place, it seems to be working very well.’ |
|two skips per week to one skip per fortnight, saving on labour, waste charges, and administration. |This confidence can be seen in the company’s use of external help. Having used consultants to prepare for|
|Keeping the site clear of empty packaging also has a soft benefit of reinforcing waste awareness amongst |ISO14001 accreditation ‘if we were doing it again I wouldn’t bother because they write a very generic |
|employees. |type of policy and manual … You end up spending as much time yourself going through it basically to |
| |understand it yourself and to make sense and to personalise it for the company.’ |
| |In Paint4 there are the signs of developing environmental management and cleaner technology capabilities,|
| |and a growing willingness to extend environmental improvement. |
Table 6.6 Performance, Practices and Capability in Case Company PAINT2
|Weak Environmental Performance |Commitment to continuous environmental improvement |
|( High emissions, waste and resource use relative to sector |( Active exploration of improvement projects |
|( Zero pollution non-compliances |( Creative problem solving – cooperation with other companies |
|( Demanding relationship with Agency |( Expertise in environment used to build business relationships. |
|( Negative impact of environmental compliance on competitiveness; job losses | |
|High levels of technology practices; good on management practices |Paint2 is a small company that has a full-time environmental manger because of the severe environmental |
| |issues that have been inherited on a new plant site. From his knowledge of the environmental issues |
|Cleaner technology projects: |around manufacturing printing inks, the environmental manager has been able to advise customers on |
|The environmental manager (EM) has worked with printing companies assisting on the switch from |managing the environmental impact of using printing inks in their processes.The company has been creative|
|solvent-based to water-based printing inks, helping with trials by providing technical advice. |about using their environmental manager as a problem-solving resource to build relationships with |
|The EM has also used his experience of environmental treatment technology to help a manufacturing |customers and retain business: ‘trying to use the environmental angle to our advantage and to customers’ |
|customer resolves its waste water problems, sourcing a novel WWTP that allows for the recycling of waste |advantage as well. It’s a win-win we think.’ |
|water back into the process. | |
|Another customer is a printing company that had disposal problem with cleaning cloths contaminated with | |
|solvent-based printing inks; the cloths had to be treated as hazardous waste. The EM identified a | |
|cleaning technology, located a laundry company prepared to take on the technology, sourced sustainable, | |
|fair-trade cleaning cloths and advised the laundry on their regulatory requirements. The outcome is that | |
|the cleaning cloths are now reused. Furthermore, the laundry has been able to offer the service to other | |
|printing companies, further reducing the total amount of hazardous waste being produced, as well as | |
|expanding the scope of their services by moving into the cleaning of other contaminated materials for | |
|printing companies. This one intervention has introduced cleaner technology and cost savings to a range | |
|of companies. | |
|The EM has also worked with both customers and suppliers to make changes to the way raw materials and | |
|products are packaged, introducing reusable containers to save on disposal costs. Printing inks are now | |
|delivered in drums with a protective liner, allowing the drums to be reused, being refilled with same ink| |
|and clearly labelled. The customer had to agree to the changes, accepting that the reused drums would not| |
|compromise quality, as well as having a process to return the empty drums. Paint2 had to invest in a new | |
|technology to change the filling process. | |
Table 6.7 Performance, Practices and Capability in Case Company METAL1
|Fair Environmental Performance |Moderately routinised approach to continuous environmental improvement |
|( Average emissions, waste, resource use, relative to sector |( Policy of cross functional teams |
|( not BATNEEC compliant at time of licensing |( Regular environmental improvement meetings |
|( High level of pollution non-compliances, prosecution |( Accessing of external information |
|( Relationship with Agency was adversarial, now cooperative | |
|High level of technology projects; 6th in sector; driven by need to meet IPC emission levels |Metal1 has had a ten-year struggle to reach compliance with BATNEEC and the ELVs set in its licence. The |
|Strong procedures, weaker on planning |company has a unique production mix and a sensitive local environment, discharging into a river. |
| |Finding a technical solution to their particular wastes has been a long process of trial and error, |
|Cleaner technology projects: |requiring development of strong internal technical capabilities in order to evaluate and customise |
|Metal1 faced severe difficulties in meeting their ELVs and in finding technical solutions to reduce |available chemical and equipment based options for reducing and treating their waste stream. |
|emissions. Like Paint1 they combined range of technologies. They combined some waste separation equipment|The company began researching the problem ahead of being licensed, in anticipation of the requirements. A|
|with chemical changes to the process and off-site recovery. This involved installing new plant, but the |cross-functional project team was put together. As well as an extensive search of technology suppliers, |
|saving on materials and waste disposal costs delivered a payback within three years. The company searched|the company also identified a similar company in the US and arranged an exchange of technical information|
|extensively for these solutions, using suppliers and external consultants, but in the end in-house |to their mutual benefit. |
|expertise was most important in developing a solution for their own processes. |A wide range of possible solutions was explored, with the company implementing a number of equipment and |
| |process changes. ‘We use consultants but most technological solutions come from within the company.’ |
| |Now the company sets annual goals for environmental improvement, at the time of interview they were |
| |pursuing energy, waste and water reductions. These are progressed through monthly meetings of the |
| |environmental team. The company also has a programme of environmental training to involve the wider |
| |workforce in these goals. |
| |Metal1 shows strong dynamic environmental capability. Its problem-solving routines are established, and |
| |the company pursues continuous environmental improvement. The difficulty in achieving compliance led to |
| |prosecution, but the company defended itself successfully on the basis of its extensive efforts to |
| |resolve the problem; it has the capacity to manage the implications of its adverse situation. |
Table 6.8 Performance, Practices and Capability in Case Company METAL3
|Poor Environmental Performance |Emerging evidence of routinised approach to continuous environmental improvement |
|( High emissions and waste; moderate resource use, relative to sector |( Use of cross functional teams |
|( not BATNEEC compliant at time of licensing |( Reliance on external consultants |
|( High level of pollution non-compliances | |
|( Agency not seen as supportive |Where possible Metal3 make deliberate efforts to coordinate environmental improvement with general |
|( Used licence to drive upgrading |equipment upgrading. so that they work in the same direction. Their view of IPC licensing is that it is a|
|Moderate level of technology projects |mechanism for upgrading both technology and procedures. |
|Strong planning, weak procedures |They struggled initially with their licence, incurring non-compliances in both emissions and procedures, |
| |but have now reached a state of compliance. |
|Cleaner technology projects: |The company has put effort and resources into environmental management, under the responsibility of the |
|They have designed-in cleaner technology in new plant, with closed loop processes achieving a high level |operations manager, with a full-time health, safety and environment manager. |
|of environmental performance. |They carry out environmental training of all staff. |
|They have also optimised existing processes to maximise raw material use, saving money on both materials |They make deliberate efforts to identify where efficiency can be achieved, in waste streams and |
|and disposal costs. |manufacturing processes. |
|They have also developed a process for treating a spent raw material so that it can be reused by another | |
|industry; currently they can only use this for some of their waste, as the capital costs to scale up to | |
|treating the entire waste stream are prohibitive. | |
|They also invested in a furnace technology, which allows them to recover metals from waste, reusing the | |
|metals and reducing the remaining waste stream by 70 per cent. | |
Table 6.9 Performance, Practices and Capability in Case Company METAL6
|Poor Environmental Performance |Emerging evidence of routinised approach to continuous environmental improvement |
|( High emissions, waste and resource use, relative to sector |( Staff involvement in some projects |
|( BATNEEC compliant at time of licensing | |
|( High level of pollution non-compliances; prosecution |Metal6 is another company that struggled with their environmental emissions and subsequent prosecution: |
|( Positive relationship with regulator - advice |‘Things were seriously disastrous, physically things were disastrous, administration-wise things were |
|( Used licence to drive upgrading |absolutely disastrous.’ Metal6 deliberately combines environmental improvement with general efficiency |
|Low level of technology projects |efforts. During their IPPC licence review ‘when we started the application we not alone looked at licence|
|Low level of management practices |stuff, we were saying “Why is it that we are doing that anyway? Could we not do it slightly better than |
| |that?” … So it has been very positive.’ Similarly, a review of costings targetted resource efficiency; |
|Cleaner technology projects: |‘it was less of an environmental push, more of an economic push, but it obviously had an environmental |
|Metal6 is another company that consciously plans to run efficiency and environmental improvements |kickback then as well, in that we were reducing all the time the amount of materials we were actually |
|together. A competitiveness driven review of producing and costs led them to a programme of efficiency |consuming.’ |
|improvements that reduced raw material use. A subsequent IPPC licence review was used as opportunity to |They make use of data on environmental emissions to drive internal improvement, and find this such an |
|review procedures and optimise processes. The company was also in crisis at this time, having been |important aspect of their environmental control that they go beyond their licence requirements in some of|
|prosecuted for pollution non-compliances. Like Metal1 they tackled these problems with a range of |their monitoring. Data on metals emissions in their waste-water, identified as part of the IPPC review, |
|technologies. Some very simple but effective changes were made as part of this overhaul. |led them to pursue process changes to reduce the waste. The environmental manager identified this a key |
|A change in piping was all that was required to eliminate a bunded sump area; this removed 1000 litres of|benefit of not engaging consultants: |
|sludge from the WWTP and considerably simplified maintenance. |‘If you get an outside firm of consultants in to do the application … you just get that information and |
|Moving to a different type of cleaner resulted in dramatically improved performance from pumps, extending|give it to them … and you are not actually looking at it yourself … you closed the loop more when you are|
|their lifespan from two months to over three years and saving on maintenance costs, equipment cost as |doing it yourself … I think that is maybe how we got a bit more out of it.’ |
|well as disposal costs. |The company accesses external information systematically, ‘we worked the triangle really of the chemical |
|A bioremediation system was installed to handle oil-contaminated wastes from the cleaning process. Again |suppliers the customer and ourselves.’ They built a team of internal management and technical advisors |
|there was a quick payback, with savings in the prolonged life of the cleaning solution, reduced |from suppliers to triall technological alternatives. |
|specialist disposal costs and improved quality of the cleaning process. |They also participated in a local environmental cooperation network project, and also retain links with |
|A major project undertaken by the company was to develop an alternative for a particularly toxic chroming|clean technology experts in the local third-level institute. |
|process that they wanted to eliminate, mindful of the RoHS and WEEE directives, involving a three-year |They are now categorised by the EPA as a low-risk company, with reduced auditing. Their most recent audit|
|programme of investigation. |report commented that ‘the site staff are to be commended.’ |
Table 6.10 Performance, Practices and Capability in Case Company METAL7
|Fair Environmental Performance |Routinised approach to continuous environmental improvement |
|( High emissions; low resource use and waste, relative to sector |( Use of cross functional teams |
|( not BATNEEC compliant at time of licensing |( Regular meetings of EM and resource use committee |
|( Minor pollution non-compliances |( Commitment to self-development of projects, using external advice |
|( Agency not seen as supportive | |
|( Struggles to reconcile EM and competitiveness |Metal7 is another company that is struggling to bring down a key emission, in this case emissions of VOCs|
|Moderate level of technology projects |from solvents. The company is working to reduce its solvent usage by 50 per cent, again pursuing this |
|Strong management practices |ambitious goal by a number of technical routes. Working on the solvent management plan is an ad hoc cross|
| |functional technical team. |
|Cleaner technology projects: |Metal7 stresses the value of working on solutions in-house, as a way of gaining the maximum benefit from |
|Metal7 is following a solvent management plan with the goal of reducing their VOCs by 50 per cent over |new knowledge and expertise and developing new skills, ‘rather than pay someone consistently to have to |
|two years. They have been moving to water based products across the range of coatings that they offer. |come in and teach you how to do the same thing all over again.’ |
|One process they are changing is to move to water based primers; as their most solvent based product this|They have environmental and energy committees that meet every two months, and use measurement data to set|
|one change will reduce their annual usage of VOCs by 25 tonnes per annum. |targets and monitor progress. For example, in reducing landfill the company identified the largest area |
|They are also looking at reuse of some solvent based cleaning materials; the solvent is less effective |of opportunity for improvement, in this case paper, and made increased efforts to use electronic |
|after recycling, so the company has identified an alternative reuse for it as a paint thinner. |alternatives. |
|Prompted by the measuring of waste for IPC reporting, the company has increased focus on waste to |They participated in the EPA’s grant programme for cleaner technology (CGPP) using it as initial support |
|landfill, reducing landfill waste by 60 per cent through raising general awareness of the need for |for a long running project to move to water based primers. This has involved a team from the company |
|segregation and recycling. |working with suppliers in developing and trialling the new primers. Introducing the new primer will also |
| |involve changing coating procedures, retraining staff in application techniques and investment in new |
| |equipment. |
Table 6.11 Performance, Practices and Capability in Case Company METAL8
|Strong Environmental Performance |Moderately routinised approach to continuous environmental improvement |
|( Very low emissions and waste, relative to sector |( Policy of cross functional teams |
|( BATNEEC compliant at time of licensing |( Regular environmental improvement meetings |
|( Zero pollution non-compliances |( Accessing of external information/advice |
|( Positive relationship with Agency – seen as helpful | |
|( EM is cost burden, managed to achieve compliance and control cost |Metal8 have been licensed relatively recently and much of their activity has involved resolving |
|High level of technology projects; best in sector |compliance issues identified as part of their licence, bringing bunding and wastewater treatment up to |
|Weaker management practices |BATNEEC standards and putting monitoring in place. They struggle to find the resources for environmental |
| |management and to meet their commitments, finding the administration a burden. |
| |However, Metal8 came into IPC licensing with strong resource efficiency, which reflects its emphasis on |
|Cleaner technology projects: |lean production. Like other companies we interviewed, they too have an explicit aim to work environmental|
|Their most significant cleaner technology project is the development of an alternative process for one of|and general efficiency initiatives in the same direction, although they find the continuous improvement |
|their products. The original process was described as ‘messy’, involving high temperatures and the |requirements of their IPC licence to be onerous. |
|generation of hazardous waste. The new processs substituted water quenching for oil quenching, operating |They use established problem-solving processes for environmental improvements. There is a tradition of |
|at much lower temperatures, reducing cost of energy and materials, as well as hazardous waste disposal |team based working w |
|and improving worker conditions in the production area. If implemented there is a short payback of one |hich they use to widen the participation and experience available for improvement projects. The |
|year. |development of the project involved an internal cross-departmental team, and the support of the Clean |
| |Technology Centre as external advisors funded through the EPA’s Cleaner Greener Production Programme |
| |(CGPP). |
| |Bi-monthly meetings are used to identify and progress projects for cost reduction, environmental impacts |
| |as well as health and safety issues. |
| |Where environmental efforts can be integrated into broader site management the company can use |
| |established routines though they are yet to achieve this for specific environmental management processes.|
Table 6.12 Performance, Practices and Capability in Case Company PAINT3
|Good Environmental Performance |Limited evidence of routinised approach to continuous environmental improvement |
|( Low emissions, waste, resource use, relative to sector |( Some use of cross functional teams |
|(inherent to nature of production process) |( Some forward investigation of projects |
|( Zero pollution non-compliances |( Extensive use made of external advice/information to compensate for limited internal resources |
|( Positive relationship with the Agency | |
|( Impact of environmental compliance not proactively managed; cost is main driver of environmental |There is no environmental team, the environmental manager has sole responsibility for environmental |
|actions |matters, and gives about two hours per week to this, reflected in this quote: ‘We got a non-compliance, I|
|Low levels of technology and management practices |think. It only came today, so I haven’t had a chance to read it.’ |
| |This company is somewhat protected by the inherent nature of its production mix which means it has an |
|Cleaner technology projects: |undemanding licence. However, despite this, the lack of a capacity to develop the required routines and |
|The environmental manager identified waste as a major cost within the company and their priority in terms|capabilities to adequately manage the company’s environmental performance has resulted in problems for |
|of environmental focus. A waste audit had been commissioned from the company’s waste disposal agents the |the company. Following a fire, procedures were found to be deficient and production closed by the HSA. |
|previous year, but nothing had been followed up: no targets identified, no plan put in place. |The company also found itself undergoing a premature IPPC licence review after they expanded operations |
|‘I actually asked them could they do a waste audit because I wanted some way of keeping track of waste |without informing the EPA. These events threaten to have a significant impact on the company’s ability to|
|because we would have quite a lot of different streams in house, from I don’t know. I haven’t even |operate. |
|counted the number of waste streams. | |
|The only savings in waste that were identified had been achieved by the purchasing manager negotiating | |
|disposal prices. | |
|Although in Paint3, both the company and the environmental manager are strong technically, and the | |
|company has engaged in extensive reformulation of products to develop water-based alternatives, these | |
|capabilities are not used to address environmental impact. ‘The main driving force would be having | |
|products that customers want to buy’ whereas the IPPC licence ‘wouldn’t have a huge impact.’ | |
Table 6.13 Performance, Practices and Capability in Case Company PAINT5
|Weak Environmental Performance |Limited evidence of routinised approach to continuous environmental improvement |
|( High emissions, relative to sector |( Environmental management responsibility rests with one person |
|( Adversarial relationship with Agency; time-consuming |( No use of teams; no procedures driving new projects |
|( Managing environmental impact is costly; compliance is main driver | |
|Average levels of technology practices; weaker on management practices |Paint5 is another company where there is little evidence of what we would call dynamic environmental |
| |capability: there is no use of teams, no process of identifying environmental improvement projects, no |
|Cleaner technology projects: |integration of environmental management across the company. |
|The company has pursued clean technology, notably control of fugitive emissions through equipment |The managing director has found the IPC licence to be extremely onerous and the company is resistant to |
|changes, but provides no information on these projects in their annual reports to the EPA. |the value of the licence, but takes the environmental obligations very seriously; ‘the biggest challenge |
| |has been to get the people in the business … to realise that compliance with the licence, and I mean 100 |
| |per cent compliance, is really important.’ |
| |The focus is on control and compliance. Environmental initiatives are directed towards gaining internal |
| |control of licence requirements such as monitoring, testing and procedural compliance. Where many |
| |companies contract these specialist tasks out, the view of this company is that ‘you cannot get the same |
| |internal commitment using consultants.’ |
Table 6.14 Performance, Practices and Capability in Case Company METAL2
|Fair Environmental Performance |Limited evidence of routinised approach to continuous environmental improvement |
|( Average emissions and waste; low resource use, relative to sector |( Company culture resisis change |
|( not BATNEEC compliant at time of licensing |( No use of cross functional teams |
|( Some pollution non-compliances |( Limited accessing of external information |
|Low level of technology projects; last in sector |Metal2 has been censured by the EPA for poorly defined environmental management responsibilities and |
|Weak management practices |failure to address the requirements of their licence. They have on-going pollution non-compliances for |
| |the limits set on water emissions in their licence. |
|Cleaner technology projects: |The environmental role in this company is undertaken as part of the quality assurance function, a role |
|The company carried out an energy audit, but ‘the report cost us money, but there haven’t been any |that is oriented towards the documentation and maintenance of rigorously defined technical procedures and|
|changes made as a result of the survey, not yet anyway, although the report is there.’ |compliance with external accreditation standards. |
|They have been successful in implementing waste separation and recycling over a three-year period, with |The company resisted the conditions of their IPC licence strongly, and their approach to managing their |
|some resistance to change from the workforce. |environmental impact and licence obligations is focussed around compliance with monitoring and reporting |
| |requirements. They discharge to a river, and are investigating getting connected to the sewer as the |
| |least disruptive way of managing their emissions to water. |
| |They do not have an environmental team or any cross functional cooperation on addressing environmental |
| |matters, they do not actively look for projects, and there is no evidence of a problem-solving capacity |
| |directed towards environmental improvement or management. |
Table 6.15 Performance, Practices and Capability in Case Company METAL4
|Weak Environmental Performance |Limited evidence of routinised approach to continuous environmental improvement |
|( High emissions, relative to sector |( No identification of key projects |
|( Many procedural non-compliances, reporting not completed |( No use of cross functional teams |
|( Managing environmental impact is costly; compliance is main driver |( Limited accessing of external information |
|Weak technology practices; weak management practices | |
| |Metal4 has ongoing problems with compliance with its emissions to water but has made little headway in |
|Cleaner technology projects: |pursuing a solution or in establishing environmental management systems for testing and reporting. At the|
|At the time of interview the company had been recently taken over and the future of the site was |time of interview the company had been recently taken over and the future of the site was uncertain. |
|uncertain. The EM’s approach was ‘non-strategic, just keep the thing going’ and minimum input’; no |Environmental responsibilities are part of the role of an external consultant looking after general |
|projects were being carried out. |administration, and his goal is the minimum need to reach compliance. One avenue being pursued is to |
| |reduce the physical floor area of the operation to fall outside the remit of IPC. |
| |The company is under some pressure from the EPA because of the longstanding non-compliance, including |
| |frequent site visits from EPA inspectors. |
| |The manager’s view is that the EPA’s actions were reasonable but that managing the requirements of the |
| |licence is a ‘hindrance’ and a diversion of effort from what is more important. The manager is strongly |
| |of the view that environmental management is of no benefit to his company, although he sees the wider |
| |benefit to society. Even initiatives such as energy audits are only undertaken to satisfy the EPA; ‘we |
| |have to go through the pretence of doing it anyway, so that’s what the companies do.’ |
6.4 Capability in environmental management and cleaner technology
6.4.1 Capability for Environmental Management
IPC licensing is unusual in being explicitly designed to require the development of an environmental management system. Companies are required to demonstrate the introduction of a system of procedures, as well as measures and target setting. One of the advantages to companies of having an IPC licence is that it provides a lever for the environmental manager to gain commitment from top management and from staff to prioritise environmental improvement and formalise environmental management.
Paint2: ‘The people who hold the main purse strings are spending on environmental stuff because they have to’ but without legislation ‘they wouldn’t bother’
Paint4: ‘From my point of view a great benefit is to be able to go to senior management and say I need to get this site cleaned up or we need to take a closer look at our waste.’ Previously it was difficult to ‘get the point across that we actually had to go and do it.’ ‘There is more awareness through the general population [of staff] … changing the culture.’
Paint3: ‘It gives me a great opportunity because we just got the new licence to bring it to everyone’s attention and to make people responsible’
Metal5: Without IPC ‘there would be absolutely no importance whatsoever.’
Metal8: ‘It made us more environmentally aware, that would probably be the biggest thing about it. And we do maintain that regardless of the workload that’s involved. It is keeping us aware, so that any projects we engage in, we are aware of the environmental effect of it’
Paint1: ‘What it did was it gave us a structure, an environmental structure.’
In the survey we asked how influential environmental issues are on discussions about company strategy. Eight respondents (38%) said environmental issues are ‘very’ influential, twelve (57%) said they are ‘somewhat’ influential and one (5%) replied that they have a ‘minor’ influence. Respondents were also asked whether progress and problems in addressing the requirements of IPC licensing entered into discussions about company strategy: twelve (57%) said ‘regularly’ while the remaining nine (43%) replied ‘sometimes’. Six respondents (29%) also stated that the influence of the environment had changed during the time the company has had its IPC license. The IPC license and environmental issues are an integral part of discussions on company strategy and over time, these issues have become more important and influential.
6.4.1.1 Benefits of good environmental management
As well as appreciating the IPC process for the focus it gives to environmental issues within the company, interviewed companies described a range of other benefits from having environmental management in place, and very few companies maintained there was no benefit. Companies reported being able to use their environmental management as a marketing tool, with advantages for selling to larger corporate clients, qualifying for tendering processes and differentiating from competitors. Some companies have used IPC compliance as an opportunity to drive technology upgrading and process improvements. And there are many companies that have achieved cost savings through projects to improve resource usage and eliminate wastes. There have also been important softer benefits, such as managing site risk and improving worker conditions.
6.4.1.2 Impact of poor environmental management
A less obvious benefit to companies of ensuring they have good environmental management is avoiding the consequences of weak environmental management procedures. Companies with poor environmental control end up in situations where resolving the problem absorbs large amounts of financial resources and management attention. In the companies that we visited that were facing these kinds of problems there was often a full-time or near full-time environmental manager, other environmental support staff, as well as high involvement from top management. One company we spoke to had the attitude that compliance was not a major issue and environmental matters were given no more than two hours a week as part of the environmental officer’s other responsibilities. This company expanded operations without considering the licence implications and ended up being required to undergo an extensive licence review process. Companies that fail to resolve environmental management issues find themselves under increasing scrutiny from the EPA, with increased site visits and monitoring requirements.
Furthermore inadequate environmental management can end up acting as a severe limit on the company’s ability to operate. Inability to resolve high emissions can force companies to cut back on manufacturing. One company had a situation where a solvent based product exploded, resulting in a prohibition order preventing solvent based manufacture until procedures to prevent reoccurrence were put in place.
6.4.1.3 Turning around poor environmental management
A good example of the negative and positive sides of environmental control is Metal6, a company that was prosecuted in 2003, followed by a licence review, and had to undertake extensive efforts to improve environmental management. They reviewed all their processes, reorganised site procedures, eliminated some raw materials and changed processes. They improved their environmental control, their working conditions and their efficiency. Now the company is in a position where they are considered to be a low risk site and have reduced auditing by the EPA.
6.4.2 Cleaner technology
The case profiles show the range of technologies pursued by companies in managing their environmental impacts. Companies are implementing product redesign, process changes, raw material substitutions and housekeeping efforts. Often projects are carried out at the margins of the manufacturing process, avoiding recongiguring of the core processes. Paint2: ‘anything like that that is doable, that is easy, that doesn’t cause any hassle, because otherwise …’ An exception is reduction in solvent use, which has been extremely challenging for both sectors we looked at, requiring changes to equipment, materials and processes, involving extensive cooperation with suppliers, as well as education of customers.
The majority of major projects identified by companies in the case study interviews were projects that achieved an economic return and were not very disruptive to the main manufacturing process; examples include looking at washwater, packaging reduction, changes to cleaning processes. They are however projects that require implementation and integration, not just drop-in solutions. Even simple projects such as reducing packaging waste (Paint2, Paint4) involves negotiating with suppliers to make changes, with new processes and equipment. Changing cleaning processes involves trials of new materials and changes to internal routines.
Benefits to these projects come from a number of sources: companies save on materials costs, on waste disposal costs, on maintenace costs, on labour costs, on better quality of results and on enhanced worker conditions.
A number of companies had identified where a waste stream could become a useful input into another product or industry, providing a closed loop solution. Paint1, Metal3, Paint4, Metal5 and Paint2 were companies who had identified reuse projects. These projects often require integrating processes across the companies involved, as well as securing permissions for the transport of waste. Reusing waste streams in this way is of particular economic and environmental benefit in Ireland, where the limited scale of industrial activity means that specialised waste often has to be shipped abroad.
Most of the companies that we spoke to were engaged in energy reduction programmes. Often this was as simple as changing the lighting: one company moved from fluorescent to high-halogen lighting at a 95 per cent saving on lighting costs. Collecting information on energy usage can be important in directing efforts. Paint4 tracks energy usage per litre of product produced, which allowed them to identify when their usage increased disproportionately; this was followed by an energy audit which involved putting meters on individual appliances which identified compressors and air conditioning as the most fruitful area for savings.
6.4.2.1 Reverse Causality
In the quantitative analysis we identified an unexpected association whereby companies with higher resource use and waste undertook higher levels of technology projects. Reverse causality is the idea that, instead of a high number of projects producing strong environmental performance and lower emissions, in these cases poor performance has triggered the adoption of projects to respond to the problem and reduce waste and resource use.
This leverage factor discussed above may explain some of what we are seeing here. A number of companies that we interviewed identified an environmental crisis brought on by prosecuting or threatened prosecution by the EPA for long-standing and unresolved pollution. Paint4 and Metal6 for example were both prosecuted for wate management non-compliances. This crisis then enabled to environmental manager to secure top management commitment to environmental management, and an increase in resources to invest in remedial projects and systems. These companies often adopted a multi-pronged approach, with improvement projects in a number of areas, or a number of technologies to deal with one problematic area. We also interviewed many companies where the high costs of energy resources and higher waste disposal charges had led the identification of these as areas to be targetted for reductions. These efforts often involve a series of related projects: audits to identify key areas, followed by upgrading of equipment (for energy efficiency) and the introduction of new procedures (for waste seperation and energy use reduction).
6.4.3 Dynamic capability for continuous environmental improvement
A company that provides a good example of how strong dynamic capability, that is, strong processes for continuous environmental improvement, can translate into a programme of technology projects and a strong environmental performance, as well as supporting the integration of environmental management with broader company goals is Paint1.
Looking at companies that do not have these problem-solving capabilities brings home even more strongly the importance of dynamic capability for environmental management. The negative impact of a lack of dynamic capability can be seen in Paint5, which struggles to manage the impact of environmental compliance on the business. Environmental compliance remains the responsibility of one person who finds that one of his biggest challenges was getting people in the company to realise the importance of compliance. Some employees believed that it was more important to improve productivity regardless of compliance. The company has an adversarial relationship with the regulator, and a significant amount of time is spent in disputing licence requirements.
6.4.3.1 Identifying areas for environmental improvement
Developing a programme of environmental technology projects starts with deliberate efforts to plan for such projects, Successful companies put in place activities to identify significant environmental impacts and then search for solutions. Two key processes are (i) collecting environmentally salient data and using it to make informed decisions about environmental improvement and (ii) planning in regular times for considering environmental improvements.
Metal8, with the highest level of technology projects in its sector, holds regular meetings of the environmental improvement group, every two months.
Paint1 holds committee meetings (management and SHE committees) every two months to discuss options to continually improve the company’s performance.
In Metal5 the project team meet on a regular basis, every two months, to discuss progress in green product development.
Paint4 and Metal6 are companies that are still developing their environmental capability. They are both systematic in the collection of data on environmental impacts to direct their improvement efforts.
Metal7 is another company that collects and uses data to ensure improvement efforts are targetted on the areas of biggest potential return, meeting every two months to identify and progress projects.
6.4.3.2 Implementing solutions
Implementing projects in areas that are new to the company requires processes to develop learning. Companies that are successful in developing cleaner technologies make sure that they involve a wide range of expertise in these projects, at the planning and the implementation stage.
At Paint1, there was evidence of cross-functionality between departments, with operations, SHE, site resources and quality assurance working together on the same projects. During the case study interview, the interviewee stated that the company had a policy of including all relevant people from the shop floor up in the projects they conducted, as they had learned that it was very difficult to get buy-in without involving staff. They also saw that it was an excellent way of generating new ideas through brain-storming and of involving staff by encouraging them to implement their own ideas. Projects that required capital expenditure were passed onto management for approval; otherwise projects were implemented on the shop floor.
Metal5 also regularly made use of teams and inter-departmental cooperation on environmental projects. Team members included R+D, technical, production, EHS, marketing, quality and finance staff. During interview, the interviewee said that teams were frequently used for product design and production technologies modifications. The company used teams from different departments and across sites (chemical, mechanical, environmental, H&S, marketing etc.) to work on new projects. They met every two months to discuss progress on the new product/project. Pressure from customers to produce more environmentally friendly products had lead to more teamwork to achieve this.
Cross functionality between departments and teamwork was mentioned several times in Metal1’s AERs, specifically when dealing with IPC compliance, costs, energy reduction and waste reduction. The company established a project team to focus specifically on lowering their emissions to water (an on-going problem); the team included the manufacturing manager, the metal finishing supervisor, the head of metal finishing and a project manager.
The survey data supported the common but not universal use of teams in industry. It also show that companies are more likely to use teams for general rather than environmental projects.
Table 6.16 Use of teams
| |Uses teams for environmental projects |Total |
| |Regularly |Sometimes |Rarely |Regularl|
| | | | |y |
|Uses teams for general projects |Regularly |3 |4 |1 |
As with use of teams, companies are more likely to use inter-departmental cooperation on general rather than environmental projects. This may suggest that environmental projects are not as integrated into the organisational fabric of the company as other projects are which can inhibit the longer-term success of environmental improvement efforts. In the survey, responding companies were also asked if their use of teams and inter-departmental cooperation had changed during the years the company had held its IPC license. The majority did not reply; however those who did (24%) gave a number of varied explanations as to how it had changed: one company had established an interdepartmental environmental steering committee, another said employees were now more involved in environmental matters. A large Metals manufacturer stated that teams were very solid at the beginning of the licensing period, they then waned in importance, but now they are again very important, because of recent pressure from customers to produce more environmentally friendly products. Another company (a small Woods company) states that they have only four staff and the manager makes all the decisions; therefore teams or interdepartmental cooperation are not utilised.
Table 6.17 Use of teams and interdepartmental cooperation for environmental projects
| |Interdepartmental cooperation on environmental projects |Total |
| |Regularly |Sometimes |Rarely |Regularl|
| | | | |y |
|Uses teams for environmental projects |Regularly |4 |0 |0 |
6.4.3.3 Building competence
Successful companies have processes or policies to capture learning. This can be capturing internal knowledge, by using cross functional teams, and by building on prior learning by maintaining an environmental improvement group. It can also mean taking internal control of areas of new expertise, rather than subcontracting them to consultants. This is more costly/disruptive in the short term, but has long run dividends in increasing the capcity of the company. These processes can allow companies to pull together necessary resources for environmental management projects, without needing a full-time environmental manager.
Paint2 had to employ environmental consultants to conduct modelling on water migration through the site; they found that the consultants were useful, but were looking to make a profit and therefore needed to be closely supervised.
Paint4 used consultants to help achieve ISO 14001 certification; however, if they were to do it again, the interviewee believed that they would do it entirely in-house. He felt that the consultants wrote a very generic policy and manual, and he ended up spending significant time to understand and personalise it for the company.
Metal6 conducted a licence review in 2005-2006. The interviewee thought that the process was of great benefit to the company, as they did it internally rather than using external consultants. It gave them the opportunity to closely examine their processes (many of which had been in place for years), think about more efficient ways of doing things and ultimately the company made several positive changes as a result.
Metal7 also saw the benefit of not using external consultants. The interviewee stated that the company did not use consultants as they preferred to work on solutions themselves and therefore learned new skills and kept new information and expertise in-house; thus the company benefited as a result. The company had no R+D department and sourced information from the internet, trade journals, research reports, suppliers
Another route for capturing knowledge is for the environmental team to work with external experts to learn from them. Impartial advice was greatly appreciated by the case companies, who accessed sources such as Entreprise Ireland, IBEC, CTC, and UK trade organisations. This type of advice is different to outsourcing responsibility for a project to consultants, and also different to purchasing an off-the-shelf solution from a supplier. Both specialist consultants and suppliers are an imrortant source of technical solutions for companies, but the most successful companies are those that plan such projects with a view to developing internal competence as part of implementation.
6.4.3.4 The role of perception in managing environmental responsibilities
Companies that manage environmental compliance successfully are those that see the licence as an opportunity for change within the company. There are many examples of companies that combine licensing with upgrading of technology, with the development of superior products and with cost reduction programmes. Having a positive approach in the way the licence is viewed is the starting point for these processes, not the outcome. Paint1, Metal3, Metal6 and Metal5 are examples of companies that are happy to go beyond compliance to give them a buffer of performance that they are comfortable with and to support strategic goals.
Companies that have the perception that the licence is an unfair imposition, and who stay in a stance of resistance are companies that retain a compliance focussed mindset and do not see opportunities to work environmental management in a way that reinforces broader strategic goals. This further reinforces their perception of the extreme demands of environmental compliance as their reactive approach often results in the operation of the licence being more binding.
6.4.3.5 Evidence from the survey data
Survey responses can be used to explore some of the relationships tested in the larger statistical sample based on EPA data. In Chapter 4, using limited information on many facilities, we examined the determinants of effective environmental impact-reducing practice using measures of organisational capability. Here, the survey gives us more capability-related information for a small number of facilities. We have coded as 0-1 companies’ responses to several survey questions that might provide clues to sources of dynamic capability, the capacity to change practice by finding and/or processing new information. We have also coded these facilities’ numerical values for the environmental impact and combined technology practice variables into 3 categories (0-1-2). For each value of the survey variables, we then compute the Gamma statistic to measure the association between impact and technology practice. Table 6.18 shows the results.
Table 6.18 Conditional Associations: Environmental impact vs technology practice
categories (Cross-tabulation layered by level of survey variable)
|Survey variable |Environmental impact |
|(Section, question number in parentheses) | |
| |Key emissions |Total waste |Resource usage|
|Environmental strategy (Compet, 3) |0 |x |x |x |
| (0 = compliance; 1 = cost reduction and/or sales position) |1 |nc |Neg |nc |
|Key players integration (Mgt, 1) |0 |nc |x |Pos |
| (0 = management or EHS only; 1 = includes others) |1 |x |x |x |
|Technical expertise (Tech, 1) |0 |Pos |x |Pos |
| (0 = not…; 1 = utilised in implementing environ. projects) |1 |x |x |x |
|Professional memberships (Mgt, 5) |0 |nc |x |Pos |
| (0 = no employees in environ. associations; 1 = some) |1 |x |Neg |x |
|Impacts and technology are ranked and categorised by lower, middle, and upper thirds. |
|Associations are Gamma statistic: ‘Pos’ or ‘neg’ = significant at least at the 10% level; ‘x’ = not significant. |
|‘NC’ = no calculation; one test variable has no variation, or fewer than 3 valid cases. |
The observation numbers in the table’s cells are very small (in no case more than nine), and the Gamma statistics are insignificant or cannot be calculated for most. However, the computed associations are interesting. In two cases, higher levels of technology practice are associated with lower impacts (in total waste); both are for the higher value of the corresponding survey variable: When environmental strategy goes beyond compliance, and when employees have external professional memberships related to environmental aspects. There are four cases where higher levels of technology practice are associated with greater impacts (one in emissions and three in resource use); all are for the lower value of the corresponding survey variable: When only management or dedicated environmental staff are key in relevant decision making, when technical expertise is not cited as a factor in project implementation, and when no employees have external environmental professional affiliations. The numbers are far too few to be definitive, but they are not inconsistent with the idea that dynamic capability building activities enhance learning and performance.
6.5 Other factors impacting on environmental performance and cleaner technology adoption.
Environmental regulation is not the only factor affecting the environmental performance of companies. They have to reconcile a range of external factors that both work against or in favour of improving the environmental impact of their activity.
6.5.1 The influence of market demand.
Product advances often provide an opportunity for designing out environmentally harmful substances, such as solvents, but customers are often slow to embrace new products.
Paint1: ‘It is amazing that the water based products are far more durable than the solvent based products. And it has also reduced the use of solvents as well. But the water based technology has been slow [to take off], customers have been slow to move over as well.’
Paint3: ‘Very often the end customer doesn’t know the difference … so you have to educate customers as well.’
Water based products can have knock-on effects for customers using them in their own products, for example requiring new testing, the installation of drying equipment or expertise in a new process.
Paint3:‘People are very reluctant to move from solvent based paints because it is what they know, especially smaller customers.’ ’80 per cent [of business] would be smaller customers but they don’t care.’
Paint4: ‘The general perception was that they weren’t familiar with it [low solvent paint product] and therefore they weren’t prepared to take a chance on it.’ ‘We got an award from Enterprise Ireland for bringing it out but Irish customers weren’t ready for it.’
Metal4: ‘to change anything in their process is a chore so they’d rather not do it’
Poor customer response can inhibit further investment in these technologies.
Paint1: ‘Well the demand would drive the technical side of it if the demand was greater but it’s not.’
Customers can also act as an obstacle to environmental improvements, requiring the retention of processes and products with a greater environmental impact, despite the availability of technically superior alternatives.
Paint4: ‘You know, we’d like to change. If we could do everything here water-based it would be a lot easier, but you have to offer a complete range.’
Metal6: ‘The customers at the end of the day look at their invoices and they honestly don’t care whether a company is 100 per cent compliant with the EPA or whether they are opening a tap and sending it down to the local river … so the customers are never, and I mean never, going to drive the environmental issues in a company’
This can lead licensed companies to feel frustrated that they are caught in a triangle of constraint between the requirements of their customer base, the technical solutions available from suppliers and the demands of regulation on the other hand.
Metal7: ‘the EPA seem to be putting pressure on the applicator. Whereas those making a decision regarding what is used is either the client on one side of him or the paint manufacturer that is supplying the product into the country.’
There are also companies that see a competitive advantage in being able to offer environmental solutions to their customers. One company, a business-to-business supplier, has provided environmental expertise to its customers, companies that also have to manage the environmental impact of their operations. This has helped the licensed company to retain business.
And another company believed its experiences in reducing solvents to meet IPC demands would put it ahead of the field when its competitors come under a new EU directive. Similarly, a metal coatings company is investing significant resources in environmental projects with the strategic view that responding to demand for energy efficient products from end customers will get them ahead of their competition.
Paint2: ‘Trying to use the environmental angle to our advantage and to our customer’s advantage as well. It’s a win-win situation … It will definitely give you a commercial advantage now.’
6.5.2 The influence of standards
EU directives are accepted by companies as something they have to anticipate and respond to. They appear to be more acceptable to companies than IPC, as they operate across all companies, licensed or not, and across all markets – the pain of adjustment is felt by all their competitors too. It also means that their international suppliers are also responding to the directives and providing support and inputs to help them meet new requirements.
Paint3: ‘our suppliers would make us aware of whatever new pieces of legislation were coming down the line, so we would re-formulate around that. We changed from aromatic white spirits to de-aromatised white spirits … because it meant we didn’t have to have a dead fish and a dead tree symbol on our paints.’
Paint4: ‘The VOC directive … all of our paints have to be compliant with that legislation and that really is a driving force for product redevelopment within the industry.’
Paint3: ‘we will try and use that [EU directives] because they [customers] will be under the same laws as we are, so we try and highlight … the advantage of having a water based system or less hazardous materials.’
Metal5: ‘Environmental issues and increasing competitors out there and the demands coming in from big customers … The company’s forced to go that way [but] it is using this in the exact same way as it would have used fire regulations coming in mid nineties or being given a protocol for the amount of CFCs being used in foams and getting ahead of the competition.’
Whereas when companies are subject to meeting an environmental emission limit value under IPC, there may often be no technical solutions easily available from suppliers, if most of the ssuppliers’ customer base is not being regulated on this.
Metal7: ‘At the moment we are one company in our entire industry trying to convince the manufacturer, the international companies of the world, to manufacture a water-based product for one single user.’
But the other side of product standards is that they can act as a constraint on the development of cleaner technologies. Companies that need to meet technical specifications or a high level of quality for their products can struggle as they aim to close the loop by reusing and recycling waste materials. For example, a number of paint companies that had solvent recovery plants have moved away from recovery and now ship the waste to handling agents. This is a cost saving, as they receive payment for the waste solvent; it reduces activity on the site; and it also eliminates the quality problems experienced when reusing solvents in their own production. Similarly, one of the metal fabrication companies was recycling chemical wastes back into products but ran into difficulties in being able to guarantee the exact specification of the product for fire certification, and the initiative had to be dropped.
6.5.3 The influence of cost
Environmental regulation aimed at moving companies towards clean technology is described as ‘win-win’ because being environmentally conscious with resources and wastes is usually also cost saving. For the companies we spoke to the win-win often works the other way round – that projects driven by cost considerations have an environmental bonus. This is true of production costs, and also as market prices for energy resources and waste disposal rise. The reverse can also happen; a number of companies in the paints and printing inks sector have discontinued on-site solovent reycling for cost reasons.
Metal5: ‘It was less of an environmental push, more of an economic push, but it obviously had an environmental kickback then as well, in that we were reducing all the time the amount of materials we were actually consuming.’
Metal5: ‘Generally efficiency projects are nearly always driven by production.’
Paint3: ‘Because we are spending so much money on waste disposal that would probably be our number one priority’
Paint4: ‘We have an energy audit going on at the moment, not for the EPA or anything, just purely for ourselves from an efficiency point of view… That’s usually the driving force in any industry, costs, cost saving.’
Paint2: ‘It is money going out through the stack
Metal8: ‘There is always a push to keep costs down. And bringing the environment into it is of course a great benefit. But what we have to say is cost is a predominant factor.’
6.5.4 The influence of the production mix
Many of the paints companies are moving out of solvent based products and also moving out of milling/blending in favour of new production technologies such as in-can tinting. These processes are often inherently less polluting but the driver is not environmental. Similarly, many metals companies are discontinuing processes, either because of EU directives, lack of market demand, or by a loss of competitiveness with production in emerging economies. Their production mix is moving to a less environmentally adverse profile, but shaped by other economic and regulatory factors.
Paint1: ‘Not really no [when asked if the technology adoption was environmentally motivated] It’s something that we wanted to do ourselves and move on. You’ve got to move on with the technology, but it had a double benefit.”
Environmental improvements can be driven by other factors besides environmental management and regulation. In our case study research we found a number of companies who actively take advantage of this by coordinating environmental management efforts with general efficiency and upgrading.
6.6 Environmental regulation and competitiveness
While the research did not seek to establish the cost of environmental complaince, some companies indicated the significant impact on their business, and especially on their relative competitiveness. This is supported by the research of Clinch and Kerins (2002) who estimate environmental costs in the surface coatings sector, for the three years between 1997 and 1999, at approximately €200,000 per company, including almost €45,000 for admiinstrative costs.
There were a number of cases of companies, among the companies we interviewed, reducing production in a direct response to being subject to an IPC licensing. One company was moving out of solvent based licensing in a drive to reduce output to a level where an IPC licence would no longer be a requirement; buying in product had seen employment drop from 60 to 20 employees. A company struggling to get emissions below the ELV have outsourced part of their production to China. Another company was forced to subcontract part of an order to a nearby (unlicensed) competitor when a customer refused to accept water-based coatings; to have met the demand for a solvent-based coating in-house would have caused them to exceed their ELVs. An owner-manager that we interviewed said that the demands of an IPC licence were making him consider whether he wanted to continue in business.
Companies reported operating on a margin of 10 per cent, which means that even to support EPA fees for licensing and monitoring of €15,000 requires additional sales of €150,000. One company estimates they have had to increase turnover by 30 per cent to support environmental costs, while another company put their costs at around 2 per cent of turnover, something they felt affected their ability to price competitively. The business significance of the cost impact of environmental regulation can be seen from the approximately 39 per cent of our survey companies whose competitive strategy is to be a low cost producer. The 44 per cent who rely on the quality and uniqueness of their product should be in a better position to absorb increased costs. Of companies in these sectors licensed under IPC from 1997 to 2004, about half of the paints companies (eight out of eighteen companies) and a third of the metals companies (eight out of 29) were no longer operating, which may be a reflection of broader competition in those sectors, and gives some indication of the intensity of competitive pressure faced by companies. This is supported by our survey research, where over 80 per cent of companies indicate that the level of competition in their industry has increased.
Table 6.19 Level of Competition in the Industry
| |Frequency |Percent |Valid Percent |
|Increased |17 |81.0 |81.0 |
|Stayed the same |1 |4.8 |4.8 |
|Decreased |3 |14.3 |14.3 |
|Total |21 |100.0 |100.0 |
We also asked companies to rate their competitive strength relative to the rest of their industry over time (1998-2004), with a score of five being most competitive and one being least competitive. Most respondents believed their competitiveness had increased during the IPC licensing period, although not in the paints sector; it is possible that their perceived loss of competitiveness reflects at least in part factors beyond those examined in this research.
Table 6.20 Competitive Strength
|Competitive strength |Metals |Paints |Woods |Total |
| |N = 7 |N = 5 |N = 9 |N = 21 |
|2004 |4.29 |2.80 |3.67 |3.67 |
|2001 |4.33 |2.60 |3.78 |3.65 |
|1998 |2.86 |3.60 |3.11 |3.14 |
|Increased |6 |0 |6 |12 |
|Decreased |1 |3 |1 |5 |
|No change |0 |2 |2 |4 |
More encouragingly, when asked directly, companies were approximately evenly split between those who see environmental management in the IPC program as having increased or decreased their economic competitiveness.
Table 6.21 Overall Impact of Environmental Management
| |Frequency |Percent |Valid Percent |
|Increases competitiveness |8 |38.1 |40.0 |
|Decreases competitiveness |9 |42.9 |45.0 |
|Neither |3 |14.3 |15.0 |
|Total |20 |95.2 |100.0 |
|Missing |1 |4.8 | |
|Total |21 |100.0 | |
The economic impact of IPC licensing on companies is non-trivial. Clinch and Kerins (2002) note that the dynamism of the Irish economy since IPC was introduced in 1994 has softened the impact of the environmental regulation. While there is broad societal agreement that the costs of pollution must be borne by the polluter, it is also important to keep in mind the societal benefits of economic activity and therefore to seek to achieve our desired level of environmental protection at the least economic impact.
6.7 The role of the regulator.
The purpose of this project was to investigate environmental problem solving in IPC licensed companies. Inevitably, companies wished to express their views on their experience of being licensed. There were companies that had very strong negative opinions of the way that licensing is operated and its impact on their operations; there were other companies that had high praise for the EPA and saw their relationships with the regulator as very positive. There is also the caveat that people with a strongly negative view of the EPA may have been motivated to participate in our research as a way of airing these opinions. We report some of these comments here for what can be learnt about improving the ability of companies to engage in continuous environmental improvement.
6.7.1 Perceptions of fairness in environmental regulation
As we reported above, there are companies that accept their IPC licence, that are proactive in managing their licence requirements, mitigating the disruption on the business and taking advantage of opportunities. There are other companies that continue to resist their licence, taking an adversarial stance with the EPA, moving to compliance with great effort and reluctance. Many companies, in both groups, referred to a lack of fairness in the way the IPC licence is administered. They are sensitive to the competitiveness implications of a licence, especially if it falls unevenly across a sector of competitors. If a company’s attitude to being licensed can be influenced by the regulatory stance, and if this then translates into a greater willingness for companies to accept not resist their environmental responsibilities, then there will be benefits to the regulator, industy and society.
6.7.1.1 A level playing field
Companies generally accept the EPA’s role in protecting the environment. However, where the burden of compliance falls unevenly, there is a perception of unfairness and unreasonableness. They feel aggrieved when other companies are not licensed, or licensed more leniently. This has to be set against the EPA’s goal that the licence is determined by the specific impact of the activity on its specific environment.
Companies feel unfairly disadvantaged where their competitors are not also subject to licensing.
Metal7: ‘We see it as something we want to continue but we would like everybody else to be in the same boat with us. One company’s performance environmentally is not going to make as much of an impact as say an entire industry.’ ‘We are the only company that has an IPC licence in the industry and we have been reminding the EPA of this fact for a number of years and nothing has changed. So we end up having to use, in most cases, more expensive clean technologies, more expensive clean products such as water based paints, whereas most of our competitors can continue to use the solvent based products … It has put us at a financial disadvantage in tendering for work and it has not given us any benefits.’
Metal2: ‘The EPA should really level the playing field’
Paint5: ‘My issue with them is the number of unlicensed companies out there… I think it is highly unjust that we have nearly a dozen competitors out there that do not have licences.’
They also feel unfairly treated when different categories of polluters are treated differently:
Metal1: ‘We found it very difficult to accept that the only people to pollute the river and cause a fish kill were the county council and they were never prosecuted.’
Small companies in particular felt the unreasonableness of their requirements, relative to the size of their operations.
Metal4: ‘They [multinational company in same industry] have the same burden there that I have, albeit at a larger volume. Their report still has twenty sections, same as mine … that doesn’t strike me as very helpful.’
Paint2: ‘I’ve looked at different people’s licences and for the first twenty pages they are all the same… A massive company and we are the same.’
6.7.1.2 Fairness in the administration of the licence
Companies frequently expressed frustration where licence conditions or enforcement appears unreasonable. This if often where the reporting requirements are unnecessarily onerous or repetitive. “When you consider the amount of our emissions, I think that was a sledge-hammer to crack a nut.’ ‘Nothing is sharp or to the point’ said one EM, referring to a self assessment exercise which came with 66 pages of guidance notes. Another aspect of this is where the EPA changes policy or introduces new policies without communicating the change. Similarly, when they they do not understand why they are asked to monitor a particular impact, companies start to lose commitment. Where the licensing is aligned to the way companies do business, it is seen as being legitimate. One company struggled with ELVs for their solvent emissions, which they found ‘not in line with the practicalities of the operation.’ Now they have agreed a solvent management plan with their inspector, reducing solvent usage by half, effectively to meet the same ELVs, which they find ‘tough but it relates more to what we are actually doing.’
The EPA’s approach can also lead to resistance rather than promote active engagement with IPC. While clearly the EPA has to act on pollution non-compliances and has enforcement as a central mission, the confrontational approach should be balanced with longer term opbjectives to encourage real engagement with environmental continuous improvement by companies.
‘Enforcement became the issue. The pressure was huge … We were treated like children in school. They were bullies and their behaviour was totally unacceptable.’
‘The EPA is a law unto itself’
‘It is always the heavy hand’
‘They’ve an extremely confrontational approach. they are pained to say that anything is good.’
‘All of the people in our company, all of them at this stage, have a lack of respect for the Agency’
Another aspect of fairness, related to a perception of unprofessionalism, is that the EPA enforce reporting guidelines on industry but do not themselves respond in a timely manner. The agency is seen as being quick to look for things, slow to get back, often leading to hold ups in the implementation of projects.
‘The information was all one way. We were sending reports to the EPA and getting no response. The EPA did not give back information.’
‘I mean this is May, and we’ve put in this [report] in January and … We don’t have our reports in on time, they are very quick to tell you.’
6.7.2 Support for environmental management
Companies express frustration when they are making efforts to meet their licence conditions or develop technology projects but they cannot get information from the people they identify as being the experts, the EPA. Again, there is a perception of unreasonableness that they are being pushed to make changes, but without much guidance. Views on this aspect were divided, and there was evidence from companies with very constructive relationships
6.7.2.1 Advice on meeting IPC requirements
Companies face technical challenges in meeting their IPC licence conditions. Often they are developing procedures for previously unfamilair activities, and are aware of the importance of achieving compliance. They see the regulator as having the competence in this area, of understanding the licence conditions and having experience with other licensed companies. They therefore find it unreasonable and inefficient when the EPA adopts a stance of requiring companies to develop their own procedures and technologies without guidance.
Paint5: ‘The other day a guy came in [from the EPA to carry out monitoring]. And he said “You haven’t set a maximum, you haven’t set a warning limit and an action limit.” And I said “Ok, throw me out what you think” “We can’t do that, it is up to you to do that” And I buy a book [from the EPA] … I cannot make head nor tail of it.’
Paint2: ‘And we need you to tell us whether we are going in the right direction. Initially they said “Just do it, and we’ll tell you.” And I said “No, we won’t. Tell me before I do it.”’
Metal4: ‘‘If you weren’t aware that an option is there, they are not going to tell you, and that is what is missing.’ ’To comply has been difficult because you can’t get a straight answer as to what you have to do.’
Paint3: ‘At one stage our inspector didn’t like the way we did the PER so last year we got our consultant to do the PER and he liked that even less, so this year I went back to the original one’
Paint5: ‘They are secretive, they don’t want to give out any information’
There are also companies that had a more positive and constructive relationship with their inspector, and value being able to get advice before pursuing environmental projects:
Paint1: ‘We don’t hide anything from them … and they know that if we have a problem they will get a call from us’
Metal3: ‘They have been [recently] a lot more open in relation to dealing with things. I think they are probably getting a bit smarter than they have been.’ ‘What was a huge problem [for the EPA] a few years ago was a perfectly sensible solution now.’
Metal1: ‘We want to work on a partnership basis with the EPA.’
Metal6: Their inspector was ‘fantastic and at the end of the phone literally all the time’ with advice on explaining the IPPC licence, advising on what was needed and, most importantly with limted resources, helping set priorities.
Metal8: ‘so I had correspondence with my inspector to say can I go down this route before I commit myself. And he’s given me guidance on that and we’ll have to get information for him. But we have that relationship with him … I would see that as teamwork in its own way. It is in the EPA’s interest to help us … I think they are there to help me.’
Metal4: ‘I’ve found them very helpful, very straightforward, very easy-going, in terms of advice to complete their paperwork but after that they don’t appear to have the technical capability to deal with specific problems’
6.7.2.2 Source of advice on best practice in cleaner technology
Some companies also expressed surprise that the EPA does not function as a source of learning, disseminating knowledge of best practice from the experiences of other IPC companies or technologies developed elsewhere. There were a number of examples of companies (see Paint2, Metal3) that have implemented very effective off the shelf technologies that are widely available but not adopted in Ireland.
Paint2: ‘Simple things, you know, that were there, that someone has done before’
Paint2: ‘If they [EPA] did help, they could learn and then make it fit the purpose … The idea would be much more cooperative, much more practical, if we were getting free help. … And then to police it and see it go through … That would mean so much.’
Metal4: ‘How am I going to drop heavy metals out of the wash-water? [Our inspector] she hadn’t a clue. Her boss, he didn’t know.’
Metal4 ‘If a small company could interact with the EPA in a practical manner where they got advice as well as knuckles being rapped, then that would be fair.’
6.8 Conclusions
6.8.1 Lessons for companies
6.8.1.1 The role of perception
The company culture with respect to change is a significant factor. Companies that perceive their environmental obligations as an integral part of their management priorities are better able to manage regulatory requirements and take opportunities to minimise the impact.
6.8.1.2 Management processes
Teamwork and inter-departmental cooperation maximises the expertise brought to bear on a problem or project, as well as increasing broad commitment to the project, increasing its chances of successful implementation. It is also key to make continuous environmental improvement into a routinised management task, with established processes for data collection and regular planning and decision-making.
6.8.1.3 Developing learning
In learning about new approaches, the companies often access external sources of help and advice. Some employed professional environmental consultants, while others asked suppliers and customers for input. There can be a significant difference in the way in which external help is used. Some companies effectively outsource tasks to external specialists, without internalising any knowledge about the project. capability is developed where companies retain internal control of the project, but commit to learning what is required with the help of external experts.
6.8.2 Lessons for the regulator
6.8.2.1 Protecting the legitimacy of the regulator
As can be expected, the case study companies had very diverse relationships with the EPA. Some found the EPA helpful and supportive; however the majority of companies interviewed found the EPA inflexible, uninformative and unhelpful. A few of the companies had what could be described as very adversarial dealings with the EPA, while others had a very positive and open relationship. The legitimacy of the EPA is undermined when companies do not respect the agency and their inspectors. If companies start to feel a disconnection between their efforts and any real environmental benefit, this can lead to poor engagement with the IPC process. IPC licensing has the aim of compliance with environmental standards, but it also aims to be developmental, to encourage the establishment of environmental management processes and routines of for continuous environmental improvement. It is this more fundamental aspect of the licensing that has the potential to reconcile environmental protection with economic impact. It is also the aspect of the licensing that is most vulnerable to company perception and attitude.
6.8.2.2 Supporting continuous environmental improvement
Companies need to acquire knowledge outside their expertise to implement environmental technologies. There is a lot of frustration that the EPA cannot be more of a resource and support to companies. Interviewees point to other forms of regulation and audit, such as SHA and quality auditors, where the relationship is one of partnership as well as enforcement.
Views on the role of the EPA in providing support and guidance to companies in making changes to meet their environmental obligations are mixed. There are companies that have very constructive relationships with the EPA. This suggests that there are differences between individual inspectors, and perhaps provides an opportunity for the EPA to adopt this stance across all inspecting teams.
There were a number of examples where companies were able to develop creative solutions to waste problems by cooperating with another company who could reuse a waste stream in their production. There is an opportunity here for the EPA to foster such arrangements, by recognising their benefits in their enforcement of the licence, and also by acting as a facilitator or broker.
There might also be an opportunity for the EPA to increase the sector-specific experience of their inspectors by encouraging sectoral specialisation and information exchange, even without moving from the current regional organisational structure. If the EPA are able to develop as a resource, they may find being knowledgeable about the industry they are regulating improves the effectiveness of the licence, as well as improving environmental performance.
6.8.2.3 Reconciling environmental protection with economic development
We are sensitive to the difficult balancing act that the EPA has to perform in satisfying all of its stakeholders. Strict enforcement is being used to meet the expectation of society that the EPA will regulate and reduce the environmental impact of industrial activity. But the EPA also has an interest in society’s more subtle goal: to balance environmental protection with economic development, which is also in the interests of society. As one of the environmental managers we talked to said ‘the only way this company can be 100 per cent environmentally friendly is to close…other than that you have to balance economics with environmental performance.’
Chapter 7 – Policy Implications
7.1 Program Efficacy
Perhaps the broadest policy relevant finding from this research is the diverse evidence that the IPC licensing program worked. While we discuss specific accomplishments, it is important to note at the start that during the 1996-2004 period studied, these facilities’ overall environmental performance seems to have improved. The environmental impact measures devised here trend gently downward with time, as indicated in the correlations reported in Table 7.1. (The result for combined resource use is driven by electricity and water reductions.) While the correlations are small, and hover around the edge of statistical significance; in addition, they refer to environmental impact variables that are normalised by employment, and decreasing per worker impact could be swamped by growth effects. Nevertheless, their regulatory significance is that a program like IPC licensing can induce companies to move toward cleaner production.
|Table 7.1 Simple Correlations: |
|Environmental performance vs time |
|(Probability values and observations in parentheses) |
| |Correlation with year |
|Key |-.119 |
|emissions |(.133, N=160) |
|Total |-.144* |
|waste |(.063, N=167) |
|Combined resource |-.189* |
|use (electricity, fuel, water) |(.058, N=101) |
|*Significant at 10% level (two-tailed). |
|Based on Spearman’s rho. |
Gradual improvement overall in environmental performance is consistent with the evidence that emerges here that the management and technology practices employed by these facilities, as they responded to the IPC licensing requirements, were associated with specific impact reductions. Lower levels of key emissions, in particular, were associated with increased management and technology practice at an aggregated level, and with disaggregated types of both that are pinpointed in the results reported in Chapter 4. The results for waste and resource usage are more complex, given the apparent dynamic of facilities with more serious environmental impact problems then undertaking elevated levels of practice in response. While we do not find specific evidence that particular practices ultimately reduced particular impacts, nevertheless those impacts were reduced over time, as noted above.
We also find evidence that as companies accumulated experience with IPC-related technology practices, they created new organisational capabilities that complemented the direct effects of those practices on performance. Companies learned by doing, and this helped them reduce environmental impacts in line with the policy aims of the IPC licensing program. The evidence suggests that early experience weighs heavily in this process. The implication for policy is that early involvement pays dividends, and delay (in coverage or in strong engagement with the program) may be costly. It may be worthwhile seeking means by which early movers can be induced to share environmental experience with relative newcomers.
While the limited statistical results we could assemble on specifically dynamic, change-oriented capabilities did not provide evidence of their importance, the case interviews did. Companies that actively sought and integrated new information on environmental management and technology reported higher levels of performance, especially if these processes were self-organised and not farmed out to a consultant. Impartial outside information sources were reported to be at a premium. Some interviewees argued that EPA could offer a valuable service in this regard by providing more informational and technical assistance, a point to which we return below.
Finally, both the statistical results from Chapter 5 and the survey responses in Chapter 7 suggest that on balance, the enhanced environmental effort and outcomes noted above did not come at the cost of reduced economic performance. Operating efficiency was actually positively related to facilities’ levels of environmental technology practice. It appears that IPC regulation did in many instances stimulate the kinds of cost-reducing ‘innovation offsets’ that Michael Porter (1995) and others have claimed. From the public’s point of view, EPA should be seen as having encouraged environmental quality without on balance creating an economic burden.
Nevertheless, the course of our research suggests, with some corroboration from the case interviews, that IPC program design could have been more efficient in certain respects. It is to this point and its implications for IPPC licensing and further program development that we now turn.
7.2 Opportunities for Standardisation
The IPC licensing process is based on facility-specific requirements for monitoring, reporting, and limiting particular environmental impacts. The specificity includes what emissions are covered; where, how, and when they are to be monitored; what levels are to be permitted; and how they must be addressed. The approach is in broad principle consistent with the widely shared understanding that environmental impact is highly context-specific by nature. But we will argue in this section that in practice, the nature and degree of specificity imposes unnecessary burdens on both EPA and licensed companies, ultimately compromising the effectiveness of the program.
Even within the same licensing class, into which facilities are grouped according to their most important environmental impacts, covered emissions and related requirements vary widely company by company. This degree of customisation means that EPA staff must design, monitor, and enforce as many regulatory packages as there are licensed facilities. Despite what seem to these researchers to be high levels of effort by skilled and dedicated staff members, the quality of the self-reported data in the files is inconsistent at best. We suggest that a reason may be the difficulty of tracking and ensuring compliance with such large numbers of highly detailed but disparate license regimens. Smaller numbers of indicators, more standard across relevant groups of licensed facilities, would likely result in more time for inspectors to ensure reporting compliance and, hence, better quality data.
While of course the policy goal is not to enhance convenience for researchers, nevertheless the difficulties encountered during this study in assembling comparable indicators from a myriad of distinct company reporting regimes may be clues to an additional problem. It is important that the results of regulatory programmes be capable of objective assessment. And to the degree that common requirements are set for comparable groups of companies, assessment is facilitated. Here the agency’s company-specific approach creates a hurdle. This barrier is further heightened by the divergence between EPA’s licensing classes and standard industrial sector classifications. Policy-relevant research such as the EU’s MEPI study typically organise findings and compare performance by standard industry classifications such as NACE, and if cross-national program comparison is a policy goal here, then more standardisation may be desirable.
A final related issue has to do with the licensed companies’ perceptions regarding the fairness of the program. It is to be expected that licensees may complain about having to meet regulatory requirements. But it is possible that sometimes the level of monitoring and reporting detail entailed in these facility-specific licenses gives rise to more difficulty and occasional resentment than is warranted by the results. In addition, differences in regulatory treatment of similar companies has been observed to engender additional resentment. A perception that the playing field is not level is not conducive to willing compliance and active engagement.
7.3 Enforcement and Assistance
The foregoing issue leads to a set of observations based on the survey responses and, especially, the case interviews. The perception among some companies that they were subjected to arbitrarily harsher treatment than their competitors was a subset of a more generalised view that EPA should offer assistance as well as enforcement. We have suggested above that, to the extent there is any merit in this view – and we certainly recognise the tendency to bias in such views – it may be partly traceable to the policy of facility-specific requirements. But it is possible that there are opportunities for a more assistance-oriented role for the agency.
Two kinds of possible assistance were raised in case interviews. One is to take as a major task the dissemination of information about best practices to licensees. The state of knowledge about the sources of environmental impact and the options available for reducing them is constantly changing. The most common source of relevant information for these companies is through vendors and consultants, and while these may be important conduits in some cases, in general the company representatives interviewed expressed dissatisfaction and even suspicion regarding the quality of information conveyed. Several voiced the wish that EPA help fill this gap. It may be that dedicated programmes and staff in this area would be both useful to licensees and conducive to more uniformly positive company-agency relations.
Cross company learning would also be facilitated by a move to smaller numbers of indicators, more standard across relevant groups of licensed facilities. Companies would have more in common, allowing them to share best practice. The EPA licensing team might gain more common experience across the group of licensed companies, allowing them to act as a transfer of knowledge.
Another request we heard was for EPA to assist licensed companies by serving as in effect a broker, helping match companies having marketable waste streams with those that could benefit from access to them. We do not know the extent of the possibility for such a role, if any, but the idea is indicative of the frequently heard desire for the agency to find ways to augment its perceived enforcement focus with others more oriented to partnership.
Like any environmental regulator, EPA must address the often divergent needs and perspectives of many stakeholders. Ultimately, its responsibility lies in fulfilling the public trust for protection of Ireland’s environmental assets. The research findings reported throughout this study support the proposition that the public’s interest has been well served by the IPC licensing program. We hope that the suggestions raised in this concluding chapter may provide some small input as EPA moves forward in its work.
References
Alpay, E., Buccola, S. and Kerkvliet, J. 2002. ‘Productivity growth and environmental regulation in Mexican and U.S. food manufacturing’. American Journal of Agricultural Economics, 84 (4): 887-901.
Argote, L. and Epple, D. 1990. ‘Learning Curves in Manufacturing’. Science, 247 (4945): 920-924.
Bansal, P. 2005. ‘Evolving sustainably: A longitudinal study of corporate sustainable development’. Strategic Management Journal, 26 (3): 197-218.
Becker, M. 2005. ‘A framework for applying organizational routines in empirical research’. Industrial and Corporate Change, 14 (5): 817-846.
Bresnahan, T., Brynjolfsson, E. and Hitt, L. 2002. ‘Information technology, workplace organization, and the demand for skilled labour: Firm-level evidence’. Quarterly Journal of Economics, 117 (1): 339-376.
Brown, S.J. and Warner, J.B. 1980. ‘Measuring security price performance’. Journal of Financial Economics, 8 (3): 205-258.
Brynjolfsson, E. and Hitt, L. 1995. ‘Information technology as a factor of production: The role of differences among companies’. Economics of Innovation and New Technology, 3 (4): 183-200.
Brynjolfsson, E. and Hitt, L. 2000. ‘Beyond computation: Information technology, organizational transformation and business performance’. Journal of Economic Perspectives, 14 (4): 23-48.
Brynjolfsson, E. and Hitt, L. 2003. ‘Computing productivity: Firm-level evidence’. Review of Economics and Statistics, 85 (4): 793-808.
Chandler, A. 1992. ‘Organisation Capabilities and the Economic History of the Industrial Enterprise’. Journal of Economic Perspectives, 6 (3): 79-100.
Christensen, J. F., 1996. ‘Analysing the technology base of the firm’. In Foss, N.J. and Knudsen, C. eds. Towards a Competence Theory of the Firm. London: Routledge.
Christmann, P. 2000. ‘Effects of “Best Practices” of Environmental Management on Cost Advantage: The Role of Complementary Assets’. Academy of Management Journal, 43 (4): 663-682.
Christie, I. and Rolfe, H. 1995. Cleaner Production in Industry. London: PSI Publishing.
Clinch, J.P. and Kerins, D. 2002. Assessing the efficiency of Integrated Pollution Control Licensing: Environmental studies research series working papers 2002. ESRS 02/10, University College Dublin.
Cohen, M.D., Burkhart, R., Dosi, G., Egidi, M., Marengo, L., Warglien, M. and Winter, S.G. 1996. ‘Routines and other recurring patterns of organisations: contemporary research issues’. Industrial and Corporate Change, 5 (3): 653-698.
Costello, N. 1996. ‘Learning and routines in high-tech SMEs: analyzing rich case study material’. Journal of Economic Issues, 30 (2): 591.
COFORD (National Council of Forest Research and Development). 2004. COFORD Connects: Processing/Products. No. 4.
Cunningham, D., 2000. IPC, BAT and voluntary agreements. Journal of Hazardous Materials 78: 105–121.
Derwall, J., Guenster, N., Bauer, R. and Koedijk, K. 2005. ‘The eco-efficiency premium puzzle’. Financial Analysts Journal, 61 (2): 51-64.
Dewulf, J. and H. Van Langenhove. 2005. ‘Integrating industrial ecology principles into a set of environmental sustainability indicators for technology assessment’. Resources, Conservation and Recycling, 43: 419-432.
Doms, M.E., 1992. ‘Estimating Capital Efficiency Schedules within Production Functions’. Centre for Economic Studies, U.S. Bureau of the Census, CES 92-94.
Dosi, G., Nelson, R.R. and Winter, S.G. 2000. ‘Introduction’. The Nature and Dynamics of Organisational Capabilities. Oxford: OUP.
Duffy, N., McCarthy, C., Zoehrer, M., 2003. Environmental benchmarking for IPC
industries: Synthesis Report. ISBN: 1-84095-103-6. EPA, Wexford.
Dutta, S., Narasimhan, O. and Rajiv, S. 2005. ‘Conceptualizing and measuring capabilities: Methodological and empirical application’. Strategic Management Journal, 26 (3): 277-285.
Environment Canada. 2002. ‘Wood Preservation’ in The Green Lane.
ec.gc.ca/toxics/wood-bois/pubs/assesmentsection4_e.html#_Toc504929728
Environmental Protection Agency (Ireland). 1996. BATNEEC Guidance Note For The Chemical Sector.
Environmental Protection Agency (Ireland). 1997. IPC Guidance Note for Annual Environmental Report. Ardcavan: EPA
Environmental Protection Agency (Ireland). 1998. Report on IPC Licensing and Control 1997. Ardcavan: EPA
Environmental Protection Agency (Ireland). 2005. Integrated Pollution Prevention and Control (IPPC) Licensing in Ireland.
epa.ie/Licensing/IPPCLicensing/#d.en.216
Environmental Protection Agency (US). 1995. Profile of the Fabricated Metal Products Industry. Sector Notebook Project, September.
Environmental Protection Agency (U.S.). 1995. Profile of the Lumber and Wood Products Industry. Sector Notebooks series.
Environmental Protection Agency (U.S.). 1998. Profile of the Metal Casting Industry. Sector Notebooks series.
ERI: Environmental Road-mapping Initiative. 2004. Paints and Coatings: Impacts, Risks and Regulations. National Centre for Manufacturing Sciences.
European Union. 2005. The IPPC Directive.
europa.eu.int/comm/environment/ippc/
Figueiredo, P.N. 2003. ‘Learning, capability accumulation and companies’ differences: evidence from latecomer steel’. Industrial and Corporate Change, 12 (3): 607-643.
Fijal, T. 2007. ‘An environmental assessment method for cleaner production technologies’. Journal of Cleaner Production 15: 914-919.
Filbeck, G. and Gorman, R.F. 2004. ‘The relationship between the environmental and financial performance of public utilities’. Environmental and Resource Economics, 29 (2): 137-157.
Gabel, H.L. and Sinclair-Desgagné, B. 1997. ‘The Firm, Its Routines, and the Environment’. INSEAD Working Papers Series. Fontainebleu, France: INSEAD Centre for the Management of Environmental Resources.
Gavetti, G. and Levinthal, D. 2000. ‘Looking Forward and Looking Backward: Cognitive and Experiential Search’. Administrative Science Quarterly, 45 (1): 113-137.
GEMI. 1998. Measuring Environmental Performance: A Primer and Survey of Metrics in Use. Washington, D.C., Global Environmental Management Initiative.
Goldstein, D. 2002. ‘Theoretical Perspectives on Strategic Environmental Management’. Journal of
Evolutionary Economics, 12 (5): 495-524.
Goldstein, D. and Hilliard, R. 2008. ‘The Drivers of Environmental Performance: A European Methodology and Preliminary Results’. Working paper.
Goldstein, D., Hilliard, R. and Parker, V. 2005. Sector Selection. Galway, Ireland, CISC Environmental Capabilities Project, October.
Grant, R.M. 1996. ‘Prospering in Dynamically-competitive Environments: Organizational Capability as Knowledge Integration’. Organization Science, 7 (4): 375-387.
Hart, S. and Ahuja. G. 1996. ‘Does it pay to be green? An empirical examination of the relationship between emission reduction and firm performance’. Business Strategy and the Environment, 5 (1): 30-37.
Helfat, C.E. 1997. ‘Know-how and asset complementarity and dynamic capability accumulation: The case of R&D’. Strategic Management Journal, 18 (5): 339-360.
Helfat, C.E. and Peteraf, M.A. 2003. ‘The dynamic resource-based view: capability lifecycles’. Strategic Management Journal, 24 (10): 997-1010.
Hempell, T. 2002. ‘What’s spurious, what’s real? Measuring the productivity impacts of ICT at the firm-level’. Discussion Paper No. 02-42, Centre for European Economic Research (ZEW), Mannheim.
Hilliard, R.M. 2001. ‘An analysis of industry response to the challenges of environmental regulation: An organisational capabilities approach’. Paper given at the European Association for Evolutionary Political Economy Conference, Siena, November.
Hilliard, R.M. 2002. ‘The role of organisational capabilities in cleaner technology adoption: an analysis of the response of the pharmaceutical manufacturing sector in Ireland to IPPC licensing regulations’. Paper given at the European Roundtable on Cleaner Production, Cork, October.
Hilliard, R.M. 2004. ‘Conflicting Views: Neoclassical, Porterian and Evolutionary Approaches to the
Analysis of the Environmental Regulation of Industrial Activity’. Journal of Economic Issues, 38 (2): 509-517.
Hilliard, R.M. and Jacobson, D. 2003. ‘Organisational capabilities and environmental regulation: the case of pharmaceutical companies in Irish regions’. In O’Leary, E. ed. Irish Regional Development: A New Agenda. Dublin: The Liffey Press.
Hutchinson, J. 2008. Personal communication.
Indiana Department of Environmental Management (IDEM). 2004. Sector Specific Technology Transfer Guide: Metal Products Industry, March.
idem/oppta/p2/assessments/
ISO. 1996. ISO 14001: Environmental management systems – Specification with guidance for use. Geneva: International Organisation for Standardization.
ISO. 1999. ISO 14031: Environmental management – Environmental performance evaluation –
Guidelines. Geneva: International Organisation for Standardization.
Jaffe A., Newell, R. and Stavins, R. 2000. ‘Technological Change and the Environment’. Working Paper Series, John F. Kennedy School of Government, Harvard University.
Jaggi, B. and Freedman, M. 1992. ‘An examination of the impact of pollution performance on economic and market performance: Pulp and paper companies’. Journal of Business Finance and Accounting, 19 (5): 697-713.
Johnson, S.D. 1996. ‘Environmental performance evaluation: Prioritizing environmental performance objectives’. Corporate Environmental Strategy, 4 (1): 17-28.
Joint Research Centre. 2004. Promoting Environmental Technologies: Sectoral Analyses, Barriers and Measures. Brussels: European Commission.
Karavanas, A., A. Chaloulakou, and N. Spyrellis. 2009. ‘Evaluation of the implementation of best available techniques in IPPC context: an environmental performance indicators approach’. Journal of Cleaner Production, 17 (4): 480-486.
King, A. and Lenox, M. 2001. ‘Does it really pay to be green? An empirical study of firm environmental and financial performance’. Journal of Industrial Ecology, 5 (1): 105-116.
King, A. and Lenox, M. 2002. ‘Exploring the locus of profitable pollution reduction’. Management Science, 48 (2): 289-300.
Klassen, R. and McLaughlin, C. 1996. ‘The impact of environmental management on firm performance’. Management Science, 42 (8): 1199-1214.
Klassen, R. and Whybark, D. 1999. ‘The impact of environmental technologies on manufacturing performance’. Academy of Management Journal, 42 (6): 599-615.
Klenow, P.J., 1998. ‘Learning Curves and the Cyclical Behavior of Manufacturing Industries’. Review of Economic Dynamics, 1 (2): 531-550.
Kogut, B. and Zander, U. 1992. ‘Knowledge of the firm, combinative capabilities, and the replication of technology’. Organisation Science, 3 (3): 383-397.
Konar, S. and Cohen, M. 2001. ‘Does the market value environmental performance?’ Review of Economics and Statistics, 83 (2): 281-289.
Loasby, B.J., 1998. ‘The organisation of capabilities’. Journal of Economic Behaviour and Organisation, 35 (2): 139-160.
MEPI. 2001. Measuring the Environmental Performance of Industry: Final Report. Brussels: European Commission.
Nelson, R. and Winter, S.G. 1982. An Evolutionary Theory of Economic Change. Cambridge, MA: Belknap Press.
Nehrt, C. 1996. ‘Timing and intensity effects of environmental investments’. Strategic Management Journal, 17 (7): 535-547
O’Sullivan, M.A. 2000. Contests for Corporate Control: Corporate Governance and Economic Performance in the United States and Germany. Oxford: Oxford University Press.
Palmer, J. 2008. Personal communication.
Palmer, K., Oates, W. and Portney, P. 1995. ‘Tightening Environmental Standards: The Benefit-Cost or the No-Cost Paradigm?’ Journal of Economic Perspectives, 9 (4): 119-132.
Penrose, E. 1959. The Theory of Growth of the Firm. Oxford: Basil Blackwell.
Porter, M. 1991. ‘America's Green Strategy’. Scientific American, 264 (4): 168.
Porter, M. and van der Linde, C. 1995. ‘Toward a new conception of the environment competitiveness relationship’. Journal of Economic Perspectives, 9 (4): 97-118.
P2Rx. 2005. The Paint and Coating Manufacturing Topic Hub.
Russo, M. and Fouts, P. 1997. ‘A resource-based perspective on corporate environmental performance and profitability’. Academy of Management Journal, 40 (3): 534-560.
Schaltegger, S. and Burritt, R. 2001. ‘Eco-efficiency in corporate budgeting’. Environmental Management and Health, 12 (2): 158-174.
Sharma, S. and Vredenburg, H. 1998. ‘Proactive corporate environmental strategy and the development of competitively valuable organizational capabilities’. Strategic Management Journal, 19 (8): 729-753.
Shiller, R.J. 2003. ‘From efficient markets theory to behavioural finance’. Journal of Economic Perspectives, 17 (1): 83-104.
Siegel, S. and Castellan, N.J. 1988. Nonparametric Statistics for the Behavioral Sciences (2nd Ed.). New York: McGraw-Hill.
Solow, R.M. 1957. ‘Technical Change and the Aggregate Production Function’. Review of Economics and Statistics, 39: 312-320.
Teece, D.J. and Pisano, G. 1994. ‘The dynamic capabilities of companies: an introduction’. Industrial and
Corporate Change, 3 (3): 537-556.
Teece, D.J., Pisano, G. and Schuen, A. 1997. ‘Dynamic capabilities and strategic management’. Strategic Management Journal, 18 (7): 509-530.
Teece, D.J., Rumelt, R. Dosi, G. and Winter, S.G. 1994. ‘Understanding Corporate Coherence: Theory and Evidence’. Journal of Economic Behaviour and Organization, 23 (1): 1-30.
Tyteca, D. 1999. ‘Sustainability indicators at the firm level: Pollution and resource efficiency as a necessary condition toward sustainability’. Journal of Industrial Ecology, 2 (4): 61-77.
University of Strathclyde. 2005. Industrial and Economic Activities Classification Schemes.
Wagner, M. 2001. A review of empirical studies concerning the relationship between environmental and economic performance. Lueneburg, Germany: Centre for Sustainability Management.
Wagner, M., Schaltegger, S. and Wehrmeyer, W. 2001. ‘The relationship between the environmental and economic performance of companies: What does theory propose and what does empirical evidence tell us?’ Greener Management International, 34 (Summer), 95-109.
Walley, N. and Whitehead, B. 1994. ‘It's Not Easy Being Green’. Harvard Business Review, 72 (3): 46-52.
Wätzold, F., Büstman, A., Eames, M., Lulofs, K., and Schuck, S. 2001. ‘EMAS and Regulatory Relief in Europe: Lessons from National Experience’. European Environment, 11 (1): 37-48.
Winter, S.G. 2003 ‘Understanding Dynamic Capabilities’. Strategic Management Journal, 24 (10): 991-995.
World Business Council. 2000. Measuring Eco-Efficiency – A Guide to Reporting Company Performance. London: World Business Council for Sustainable Development.
Young, C.W. and Rikhardsson, P.M. 1996. ‘Environmental performance indicators for business’. Eco-Management and Auditing, 3, 113-125
Zollo, M. and Winter, S.G. 2002. ‘Deliberate learning and the evolution of dynamic capabilities’.
Organization Science, 13 (3): 339-351.
Acronyms
AER Annual Environmental Report
BAT Best Available Technologies
BATNEEC Best Available Technologies Not Entailing Excessive Costs
CISC Centre for Innovation and Structural Change
COD Chemical Oxygen Demand
CRO Companies Registration Office
CSO Central Statistics Office
EBP Environmental Best Practice
ELV Emission Limit Value
EMP Environmental Management Programme
EMS Environmental Management System
EPA Environmental Protection Agency
ERTDI Environmental Research, Technological Development and Innovation
EU European Union
IPC Integrated Pollution Control
IPPC Integrated Pollution Prevention and Control
LBD Learning-by-doing
NACE Nomenclature génerale des Activités économiques dans les Communautés Européennes
NUIG National University of Ireland, Galway
PER Pollution Emissions Register
PRTLI Programme for Research in Third Level Institutions
ROHS Restriction of Hazardous Substances
SIC Standard Industrial Classification
VOC Volatile Organic Compound
Appendix 1
Sample Facilities by Sector
(Facilities removed from sample are at end.)
METAL FABRICATING
Sample = 29
Sample used for statistical analysis = 21
|Name |Lic. No. | |Name |Lic. No. |
|APW Enclosures (Dublin) Limited* |485 | |Johnson Manufacturing Limited |675 |
|APW Galway (Ballybrit) Limited** |133 | |Kells Stainless Limited |475 |
|APW Galway (Ballybrit) Limited** |557 | |Kingspan Building Products Limited |65 |
|APW Galway (Deerpark) Limited** |619 | |Loredo Limited |281 |
|Atlas Aluminium Limited**** |436 | |Maysteel Teoranta |624 |
|Basta Limited |269 | |Ossian Limited |272 |
|Brewery, Chemical & Dairy Engineering Limited |283 | |Pierce Engineering Limited |300 |
|Byrne-Mech Limited |369 | |Radley Engineering Limited*** |314 |
|C-Fab Limited* |114 | |Runtalrad Limited*** |33 |
|Containers & Pressure Vessels Limited*** |289 | |Sapphire Engineering Limited |380 |
|Crown Packaging Ireland Limited |98 | |SIAC Butlers Steel Limited |518 |
|Fort Wayne Metals Ireland Limited |646 | |True Temper Limited |615 |
|Galco Steel Limited |632 | |Van Leer Ireland Limited*** |107 |
|Grant Engineering Limited incorporating Azcroft |294 | |Veha Radiators Limited*** |663 |
|Limited | | | | |
|HDS Energy Limited*** |286 | |W.I. Limited |293 |
|Irish Pioneer Works (Fabricators) Limited |407 | |Waterford Metal Industries Limited |385 |
* C-Fab (114) moved to a new site and changed its name to APW Enclosures (485) in 2000 resulting in a reviewed license. Therefore C-Fab and APW Enclosures are treated as the same facility.
** APW Galway (133, 557 and 619) moved sites and changes names several times during the licensing period, resulting in several license reviews. Therefore APW Galway (133, 557 and 619) is treated as one facility for this study.
*** Containers & Pressure Vessels (289), HDS Energy (286), Radley (314), Runtalrad (33), Van Leer (107) and Veha (663) are all excluded from statistical analysis due to lack of data.
**** Atlas (436) is excluded from statistical analysis as it became apparent during the gathering of data that their processes were too different from other companies in the sample to be comparable.
PAINT AND INK MANUFACTURING
Sample = 18
Sample used for statistical analysis = 13
|Name |Lic. No. | |Name |Lic. No. |
|BASF Printing Systems Ireland Limited** |228 | |I. B. C. Limited |231 |
|Circle Paints Manufacturing Ireland Limited** |245 | |International Coatings Limited |122 |
|Circle Syntalux Limited* |524 | |INX International Ink Company Limited** |252 |
|Coates of Ireland Limited t/a Coates |241 | |Manders Coatings & Inks Ireland Limited |250 |
|Lorrilleux ( also know as Sun Coates) | | | | |
|Crown Berger (Ireland) Limited |248 | |Packaging Inks & Coatings |253 |
|Devcon Limited |260 | |Shamrock Aluminium Limited |249 |
|Dulux Paints Ireland Limited |218 | |Sun Chemical Inks Limited** |230 |
|Eastman Chemical Luxemburg S.A.R.L. |548 | |Syntheses Limited* |216 |
|FSW Coatings Limited |244 | |Trimite Truecoat Limited |239 |
|General Paints Limited |229 | |Valspar Ireland Limited |209 |
|Global Switch Property (Dublin) Limited |109 | | | |
* Syntheses (216) and Circle Syntalux (524) are treated as the same facility. Syntheses (216) moved to a new site in 2000 resulting in a new license (524). Circle Paints Limited took over the license and changed the name to Circle Syntalux in 2004.
** BASF (228), Circle Paints (245), INX (252) and Sun (230) are excluded from statistical analysis due to lack of data.
WOOD SAWMILLING AND PRESERVATION
Sample = 31
Sample used for statistical analysis = 25
|Name |Lic. No. | |Name |Lic. No. |
|A.S. Richardson & Company Limited |333 | |McCool’s Sawmills Limited* |318 |
|Adhmaid Cill Na Martra Teoranta |334 | |Murray Timber (Ballon) Limited |556 |
|Ballyfin Sawmills Ltd. |478 | |Murray Timber Products (Galway) Limited* |692 |
|Broderick Manufacturing Limited T/A Kilrush |424 | |P.D.M. Limited |325 |
|Trading*** | | | | |
|CCM Limited T/A Kenn Truss |346 | |Palfab Limited |338 |
|Coillte Teoranta Dundrum |367 | |Pat Regan Newbrook Limited |335 |
|Coolrain Sawmills Limited |323 | |Protim Abrasives* |326 |
|Doherty Brothers Timber Company Limited |354 | |Shannonside Building Supplies Limited |319 |
|Earrai Coillte Chonnacht Teoranta |355 | |Spaight Timber Preservatives Limited* |331 |
|Gem Manufacturing Company Limited |351 | |Superwarm Homes Limited, Fermoy (AKA Barry |328 |
| | | |and Sons) | |
|Glenfarne Wood Products Limited |625 | |Superwarm Homes Limited, Louth* |368 |
|Glennon Bros. Cork Limited |344 | |T. & J. Standish (Roscrea) Limited |320, 706 |
|Glennon Bros. Timber Limited |327 | |Waterford Joinery Limited |350 |
|Grainger Sawmills Limited |594, 691 | |Woodfab Timber Limited |358 |
|IJM Timber Engineering Limited |363 | |Woodford Timber Products Limited* |377 |
|Laois Sawmills Limited |322 | |Woodroe Limited |349 |
* McCool’s (318), Murray Timber (692), Spaight (331) and Superwarm (368) are excluded from statistical analysis due to lack of data.
** Protim Abrasives (326) is excluded because it makes rather than uses the preservatives and is therefore too different from other companies in the sample.
*** Broderick (424) is excluded because…?
SLAUGHTERING OF LIVESTOCK
Sample = 33
|Name |Lic. No. | |Name |Lic. No. |
|AIBP Limited T/A AIBP Bandon |188 | |Galtee Food Products Limited |174 |
|AIBP Limited t/a AIBP Cahir |204 | |Glanbia Fresh Pork Limited |180 |
|AIBP Limited T/A AIBP Nenagh |184 | |Glanbia Fresh Pork Limited |181 |
|AIBP Limited t/a AIBP Rathkeale |191 | |Glanbia Fresh Pork Limited |182 |
|AIBP T/A AIBP Clones |190 | |Henry Denny & Sons Ltd |161 |
|AIBP T/A AIBP Dublin |189 | |Irish Country Meats(Sheepmeat) Ltd |177 |
|AIBP T/A AIBP Dundalk |185 | |Kepak Athleague |168 |
|AIBP t/a AIBP Waterford |205 | |Kepak Clonee |167 |
|Ashbourne Meats |194 | |Kepak Cork |595 |
|Cabglove Limited |172 | |Kepak Hacketstown |166 |
|Charleville Foods |173 | |Kildare Chilling Company |170 |
|Dawn Country Meats Limited |178 | |Liffey Meats (Cavan) Limited |169 |
|Dawn Country Meats Ltd., t/a Western |48 | |M. J. Bergin & Sons Limited |192 |
|Proteins* | | | | |
|Dawn Meats (Exports) Limited |179 | |McCarren & Company Ltd. |171 |
|Dawn Meats (Midleton) Ltd. |176 | |Meadow Meats Limited |183 |
|Donegal Meat Processors |187 | |Queally Pig Slaughtering Limited |175 |
|Fair Oak Foods (Clonmel) Limited |165 | |Slaney Foods Limited , Slaney Foods |193 |
| | | |International Ltd. and Slaney Proteins | |
* Western Proteins (48) is excluded as their main process is rendering.
FACILITIES REMOVED FROM SAMPLE
|Name |Lic. No. |Sector |Reason for removal |
|Cavanagh Foundry Limited |479 |Metal |Casting, not fabrication |
|Computer Plating Specialists Limited |278 |Metal |Electroplating |
|Galco (Cork) Limited |391 |Metal |Electroplating |
|Galco (Dublin Limited |284 |Metal |Electroplating |
|Galvotech (International) Ltd |292 |Metal |Electroplating |
|Hitech Plating Limited |276 |Metal |Electroplating |
|Hitech Plating Limited |434 |Metal |Electroplating |
|Hitech Plating Limited |568 |Metal |Electroplating |
|Irish Finishing Technologies Limited |384 |Metal |Electroplating |
|Sperrin Galvinisers Ltd |658 |Metal |Electroplating |
|Waterford Plating Company Limited |280 |Metal |Electroplating |
|Colfix (Dublin) Ltd. |80 |Paint |Asphalt, not paint, production |
|G. Bruss GmbH Dichtungstechnik |465 |Paint | |
|Henniges Elastomers Ireland GmbH |243 |Paint |Injection moulding of rubber parts |
|Irish Rubber Components Ltd |641 |Paint | |
|LPD (Ireland) Ltd/ Weathercreate Coatings Ltd |257 |Paint | |
|MC-Building Chemicals Mller and Partners. |464 |Paint |Manufacturing elastomeric construction |
| | | |sealants |
|C & N Oils Limited |43 |Slaughter |Rendering |
|Carton Group Ltd. |49 |Slaughter |Rendering |
|Castlemahon Food Products |46 |Slaughter |Rendering |
|College Proteins Limited |597 |Slaughter |Rendering |
|Dublin Products Ltd. |41 |Slaughter |Rendering |
|Marrow Meats Limited |364 |Slaughter |Rendering |
|McCarren & Company Ltd. |171 |Slaughter |Rendering |
|Meadow Meats Limited |183 |Slaughter |Rendering |
|Monaghan Poultry Products Limited |362 |Slaughter |Rendering |
|Monery By-Products (2000) Limited |591 |Slaughter |Rendering |
|Munster Proteins Limited t/a Waterford Proteins |586 |Slaughter |Rendering |
|Munster Proteins Ltd. |637 |Slaughter |Rendering |
|National By-Products |565 |Slaughter |Rendering |
|Premier Proteins (2000) Limited |592 |Slaughter |Rendering |
|United Fish Industries Limited |416 |Slaughter |The manufacture of fish-meal and fish-oil. |
|Doherty Brothers Timber Company Limited |354 |Wood |Distribution only, not production |
|Finsa Forest Products Limited |22 |Wood |Composites production |
|Masonite Ireland |21 |Wood |Composites production |
|Smartply Europe Limited |1 |Wood |Composites production |
|Weyerhauser Europe Limited |593 |Wood |Composites production |
| | | | |
| | | | |
| | | | |
Appendix 2: Environmental Performance of Case Study Companies
Table A2.1 Environmental Performance – Paint case study companies
| |Paint1 |Paint2 |Paint3 |Paint4 |Paint5 |
|Environmental performance variables |
|BATNEEC at time of licensing |Y |N |Y |Y |N |
|Key emissions, flow* | |+110% |-100% |-32% |-37% |
|Sector ranking | |10 |1 |5 |4 |
|Key emissions, mass* |-89% |+67% |-100% |-32% |+27% |
|Sector ranking |2 |11 |1 |5 |10 |
|Total waste, weight* |+42% |-51% |-55% |+64% | |
|Sector ranking |9 |5 |4 |10 | |
|% Hazardous waste |-62% |+80% |-92% |-14% |-33% |
|Sector ranking |3 |11 |2 |8 |6 |
|% Disposed waste |-76% |-23% |+29% |+7% |+21% |
|Sector ranking |1 |4 |10 |6 |7 |
|Waste composite** |-32% |+2% |-51% |+24% | |
|Sector ranking |3 |7 |1 |8 | |
|Electrical usage* |+18% |-17% |-22% |-44% | |
|Sector ranking |6 |4 |3 |1 | |
|Fuel usage* | |+54% |-75% |-20% | |
|Sector ranking | |6 |1 |4 | |
|Water usage* |+123% |-89% |-87% |-18% | |
|Sector ranking |8 |2 |3 |5 | |
|Resource composite** | |-17% |-62% |-27% | |
|Sector ranking | |4 |2 |3 | |
|Pollution non-compliance*** |0.00 |0.00 |0.00 |0.00 |0.00 |
|Sector ranking |1 |1 |1 |1 |1 |
|Legal actions |N |N |N |Y‡ |N |
| | | | | | |
|Management variables | | | | | |
|Planning |5.90 |1.00 |2.83 |4.14 |3.00 |
|Sector ranking |1 |13 |7 |4 |6 |
|Training & development |2.25 |5.50 |1.80 |1.14 |0.75 |
|Sector ranking |5 |1 |6 |9 |11 |
|Procedures |10.71 |5.67 |-1.43 |-3.00 |-3.60 |
|Sector ranking |1 |3 |7 |9 |11 |
|Composite |20.00 |12.33 |2.50 |2.29 |2.20 |
|Sector ranking |1 |3 |8 |9 |10 |
| | | | | | |
|Technology approach variables | | | | | |
|Raw materials |4.53 |2.88 |2.75 |1.07 |5.13 |
|Sector ranking |2 |6 |8 |9 |1 |
|Closing loop |13.38 |5.83 |1.56 |6.43 |4.38 |
|Sector ranking |1 |3 |10 |2 |7 |
|Equipment |15.06 |8.00 |2.63 |2.50 |5.66 |
|Sector ranking |1 |4 |10 |11 |6 |
|Process |3.63 |2.63 |2.13 |0.89 |1.88 |
|Sector ranking |2 |3 |6 |12 |8 |
|Technology stage variables | | | | | |
|Product development |3.41 |2.88 |2.75 |2.14 |5.13 |
|Sector ranking |4 |5 |6 |7 |1 |
|Preparation |6.06 |3.75 |1.88 |0.36 |1.88 |
|Sector ranking |1 |2 |5 |9 |5 |
|Core production |6.28 |1.25 |1.56 |0.00 |3.72 |
|Sector ranking |4 |11 |10 |12 |6 |
|Finish work |0.00 |0.00 |1.50 |1.07 |0.00 |
|Sector ranking |7 |7 |2 |6 |7 |
|Housekeeping/other |20.84 |11.46 |1.38 |7.32 |6.31 |
|Sector ranking |1 |4 |13 |6 |7 |
|Composite |36.59 |19.33 |9.06 |10.89 |17.03 |
|Sector ranking |1 |3 |10 |9 |5 |
*Figures normalised against employment, expressed as % above (+) or below (-) sector average.
**Sum of the three waste/resource categories, expressed as % above (+) or below (-) sector average.
*** Average pollution non-compliance score per year. ‡ Procedural
Table A2.2 Environmental Performance – Metal case study companies
| |
|BATNEEC at IPC |
|Planning |
|Raw materials |
|Product development |0.00 |0.00 |0.00 |
-----------------------
[1] The development of the concept may also be seen as a response to criticism that the automaticity implied in Nelson and Winter’s concept of routines means that the evolutionary economics theory of the firm is as deterministic as the neoclassical theory of the firm (O’Sullivan, 2000).
[2] There is a potential bias in normalising this way, as technology change over time might increase productivity. If output per worker rises, then all else equal, so will mass emissions. With mass emissions in the numerator and employment in the denominator, then, the measure may be biased upward as technology changes over time. But our hypothesis is that changing technology will reduce (normalised) emissions. Thus the measure tilts the scales against accepting our hypothesis.
[3] Data integrity is guarded at this point by removing extreme normalised values, and then requiring that sector averages in each variable contain data from at least three companies. Extreme values arise mostly from erratic numbers self-reported in the AERs, and we have attempted to pre-exclude those that seem clearly to reflect measurement error. Remaining outliers are screened using ‘outer fence’ values derived from inter-quartile range analysis.
[4] We divide each year’s facility emission by the sector average for that emission across all sampled years (1997-2004), not the sector average for that year alone. We do this because sector averages for specific emissions often shows a distinct time trend; but we want the company’s emission index to vary with its own performance without confusion with company performance relative to its sector. E.g., if the facility’s carbon emissions to air are declining, we want that to appear in the data and be associated with its contemporaneous practices, even if others in its sector reduced carbon emissions as well. Using annual facility emissions as ratios with an all-period sector average gets around this problem, while still allowing us to compare sectors with different key emissions.
[5] The wood sawmilling facilities were dropped from the Key Emissions calculation. Their most significant environmental impact is from the chemicals used in treating the lumber for preservation. In addition, the sawmilling stage emits large quantities of sawdust, which may create waterborne and airborne solids. Frequently reported in the EPA’s records for these facilities are pH, COD, suspended solids, copper, chromium, and arsenic to water. However, rather than mass amounts these facilities report mostly flow emissions concentrations (e.g., mg/m3), which are difficult to compare across facilities. Thus Key Emissions could not be calculated for the sector.
[6] Carbon Trust's conversion factors, in kWh/m3 except as noted, are: natural gas, 10.9; diesel oil, 10,900; kerosene, 10,300; LPG, 7100; fuel oil, 11,900; and coal, 7472 kWh/tonne.
[7] One could be more precise about at least one environmental impact, by estimating tons of CO2 per facility from the electricity and fuel totals in MWh. In addition, a total energy efficiency variable could combine end-use electricity and primary fuels. Both would require adjusting purchased electricity for national electricity-generation primary fuel mix and average transmission losses, and are beyond the scope of the present study.
[8] Solow (1957) suggests that the passage of time builds useful experience when increasing know-how in the broader environment is available to the firm.
[9] We have chosen nonparametric statistical techniques for the following reasons. First, scatter plots of the data show that it does not conform even approximately to the usual assumption of a normal distribution. Related, many of the variables exhibit numerous extreme values, which can seriously bias parametric estimates. (We have attempted to distinguish between measurement or recording errors, to be corrected or excluded, and potentially legitimate values, of which we retain all but the most extreme as indicated by interquartile ranges.) Finally, it may be premature to attribute meaningfully uniform intervals to the values arising from the data construction methods described in this study, and hence we are more comfortable with analytical techniques based on rank-ordering.
[10] The different numbers of observations in the columns of Table 4.1 reflect the partial correlation calculation process, which begins with a set of simple correlations using only observations for which there is data on all three (here) variables of concern – e.g., in column one, emissions, technology, and management.
[11] Because the approaches and stages variables are constructed from the same technology project scores, merely grouped differently, both sum to the same technology aggregates shown in Tables 1 and 2.
[12] Managers may also turn to staff training programmes as suggested by Table 4.2.
[13] It is possible to return to using the constructed technology variable in Table 4.6, not project scores as in Table 4.5, because the lag structure of the correlations excludes projects later than the lag year.
[14] We are also, unavoidably, testing jointly the appropriateness of our learning by doing-based model of static capability and its significance so defined in mediating the practice-performance relationship. If standard statistical tests show ‘significance,’ then assuming we have defined capability appropriately, we have learned it is important in this setting. If standard significance tests fail, then either our hypothesis about learning by doing is wrong, or we have proxied it wrong, or both.
[15] See for example Siegel and Castellan (1988). An alternative might be to segment the sample and re-run the partial correlation tests for emissions and practices, separately for companies whose static capability averaged over the years is low vs high. The problem is that for static capability (unlike dynamic capability in the next sub-section), we look a priori for change over time, and in fact have time series data across companies with which to t*–´= Š ‹ « km•¥¦–
—
¢
ö÷>èì*,bh?v‘©ÄÆÚèest for this.
[16] As noted in Chapter 3, ‘Key Emissions’ includes only metals and paints facilities; emissions data from woods facilities cannot be properly normalised and has been excluded.
[17] As noted in Chapter 3, ‘Key Emissions’ includes only metals and paints facilities; emissions data from woods facilities cannot be properly normalised and has been excluded.
[18] As noted in Chapter 3, ‘Key Emissions’ includes only metals and paints facilities; emissions data from woods facilities cannot be properly normalised and has been excluded.
[19] In the case of two wood companies, the environmental manager who returned the questionnaire was the representative for other sister companies in the sample, which did not return the questionnaire. We presumed that this person was involved in setting policy at all associated facilities. In these cases, the responses of the questionnaire were repeated to reflect the other company(s) represented by that manager. Therefore, although 18 questionnaires were returned in total, the results are applicable to 21 companies in the sample.
................
................
In order to avoid copyright disputes, this page is only a partial summary.
To fulfill the demand for quickly locating and searching documents.
It is intelligent file search solution for home and business.