Sustainability performance of the insurance industry
In this paper it is examined in what extent insurance companies are incorporating sustainability in their business activities. At first glance, the role of insurance companies appears to be small. However, this role can be large because of several specific characteristics of the insurance industry. The insurance industry monitors, selects, screens and advices companies. The conclusions from this analysis are that social and ethical aspects of sustainability are integrated in the business activities of insurance companies. However, environmental aspects are less well integrated in the business activities of insurance companies.
Sustainability performance of the insurance industry
University of Groningen
Author: Renate Meekenkamp Student number: 1257811
Emailadress: renatemeekenkamp@hotmail.com
Faculty: Faculty of Economics and Business Specialization: Finance
Table of contents
1. Introduction 4
2. The insurance industry 7
3. Methodology 12
3.1 Selection of financial conglomerates and insurance companies 12
3.2 Methodology 13 3.3 Criteria 16 4. Results 25 4.1 Regional analysis 25 4.2 National analysis 28 4.3 Individual analysis 32 4.4 Group analysis 33 4.5 Additional analyses 44 4.6 Factor analysis 47
4.7 Summary of the results 49
5. Conclusion 53
Reference list 56
Appendixes:
I List of the insurance companies 60
II Supplementary description of the sustainability criteria 64 III T-test, Kolmogorov-Smirnov test and ANOVA test 67
IV Sustainability criteria 69
V Results national level per criterion 76
VI Ranking of insurance companies 78
VII Results Kolmogorov-Smirnov test 81
VIII Results Spearman rank correlation test 83
1. Introduction
Sustainability and climate change are ‘hot’ debates, and many researchers, companies and governments are involved in this debate. One of the most important topics of the G8 top in 2007 was climate change, and the emission of CO2. At first sight, it appears that the financial
sector has a small role in this debate. It is a relatively environmental friendly industry and its direct impact on the environment, through waste, and energy usage, is small. However, the indirect impact of the financial industry is very important in this debate.1 The financial industry has an influence on other industries, because this industry not only collects money and makes investments, but also selects, screens, monitors, and advices companies on their business activities (Scholtens (2006)). Through this influence the financial industry can stimulate sustainability. One of the sectors of the financial industry is the insurance industry. The insurance industry is the world’s largest economic sector and almost every consumer and business is, in some way, involved with this industry (Mills and Lecomte (2006)). The insurance industry is dynamic and has already begun the stimulation of sustainability through the development of sustainable and climate ‘friendly’ products, and within the insurance industry there exists a growing awareness that business and good environmental management go hand in hand (Mills and Lecomte (2006)).
In this paper the focus is on the sustainable development of insurance companies. One of the most cited definitions of sustainable development is: ‘development that meets the needs of the present without compromising the ability of future generations to meet their own needs’ (WCED, 1987, p. 46). In general there are three types of sustainability, namely ecological sustainability, social sustainability and economic sustainability. Ecological sustainability sees sustainability as activities that do not lead to deterioration in the physical environment. Social sustainability implies that the strengths of the social systems are maintained and that vulnerability to unexpected shocks is limited. Economic sustainability relates to the maximum productivity possible without affecting the stock of productive assets. Economic efficiency then guards the optimal usage of scarce productive resources and the maximization of utility through consumption.2
1 Mills and Lecomte (2006) 2
The aim of this paper is to develop an overview of the sustainability performance of insurance companies. The analysis of Bijl (2007) on the sustainability performance of banks is used as a starting point for this paper. Bijl extended the analysis of Jeucken (2001). In her analysis she incorporated criteria regarding social sustainability as well, and made a comparison between her results and the results of Jeucken. Both Jeucken and Bijl find that banks in Europe are leading in sustainability.3 According to Jeucken, differences in culture, legislation and market forces can be an explanation for this. In Europe the focus is more on the environment, while in North-America and the Pacific community involvement is perceived to be more important. The European governments also use more often market instruments to stimulate companies in the development of more environmentally oriented products.4 The abovementioned two papers focus only on one sector of the financial industry, namely banks. Result from other papers about the financial industry and sustainability is that this industry is incorporating sustainability in its innovations and developments (Jeucken (1998), Bouma, et al. (2001)). However, in these papers perspective is missing (Scholtens (2006)). Only companies or company divisions which are already performing well or which are expected to perform well are analysed.
This paper analyses the sustainability performance of the insurance industry, and adds new insights to the existing literature on sustainability within the financial industry. Both large and small insurance companies are included in the analysis to obtain an accurate perspective. In this paper the research question is:
In what extent are insurance companies incorporating sustainability in their business activities?
Sustainability performance in this analysis is not only related to the environment, but also incorporates social and ethical aspects. To assess the sustainability performance of insurance companies I will asses the companies on 23 sustainability criteria. These criteria reflect environmental, social and ethical aspects of sustainability.
In total 153 insurance companies are included. Information is assembled from sustainability and annual reports of the companies or from official websites. The sustainability performances will be analysed on four levels, namely on the regional, national, individual and group level. The three regions in this analysis are Europe, North-America and Japan. The
3 Bijl (2007), p. 78 4
performance on the national level will be tested with the two-sided t-test to determine whether the national performance significantly differs from the average performance. For the individual analysis the insurance companies are ranked according to their performance. The insurance companies are divided into four groups, namely financial conglomerates, life insurers, general insurers and mixed insurers, for the group analysis. Performances will be compared between the groups, and I use the ANOVA test to determine whether the average performance between the groups significantly differ. Besides these four analyses, I examine if there is a relation between the size of the company and the sustainability performance, and compare the inclusion in sustainability indices with the sustainability performance.
The results from these analyses are that many insurance companies have not fully integrated sustainability in their business activities. Usually the social aspects of sustainability are integrated in the activities of insurance companies. Insurance companies are active in the community and in sponsoring. However, the environmental aspects of sustainability are less integrated and many insurance companies have not integrated environmental sustainable activities. I find a significant positive relation between company size and performance, and between the inclusion in sustainability indices and performance.
2 The insurance industry
This chapter starts with a brief description of the insurance industry, followed by an overview of the current literature on sustainability and the financial industry. From the literature I am able to derive several hypotheses about the sustainability performance of the insurance industry.
Insurance companies are private companies which take over certain risks of insurees or enable them to build up capital against the payment of a premium (Verbond van Verzekeraars (2002)). Within the context of the insurance policy the insurance company will compensate the losses incurred by the insuree. Insurance companies have an important role within society. Society would not function without the presence of insurance companies taking over certain risks. Insurance companies need to be reliable, persistent and they require the trust of their insurees. Only insurance companies which offer certainty, manage the premiums carefully and run their business socially responsibly will succeed.5
In Scholtens (2006) an overview of the requirements for insurability of an event are given, these requirements are drafted by Berliner (1982) and Swiss Re (2005). Before an insurance company offers to compensate certain risks, the company will assess whether the risk is controllable and quantifiable, and whether it is related to one or more future insecure events or events of which it is only uncertain when it will take place. For insurance companies it has to be acceptable to take over the risk; average loss should be moderate and the loss frequency high. Information asymmetry can not be excessive (information asymmetry refers to the situation in which the insuree knows more about the risks than the insurer). If these requirements are satisfied, the insurance company has to determine the context and premium of a policy, and the insurance sector has to have sufficient capacity to bear the risks. Finally, it should be made certain that the insurance policy does not undermine the norms and values in a society, and coverage of the risks should be legitimate. In table 1 an overview of the requirements for insurability of an event are summarized. Differences in contexts and premiums can be explained by differences in cost structures, damage statistics and mortality tables.6
5 Gedragscode Verzekeraars, Verbond van Verzekeraars (2002) 6
Table 1: Requirements for insurability7
Category Criterion Characteristic
1 actuarial risk - uncertainty measurable 2 loss from events independent 3 maximum loss controllable 4 average loss moderate 5 loss frequency high
6 information asymmetry not excessive
7 market insurance premium affordable 8 cover limit acceptable 9 sector capacity sufficient
10 Society public policy consistent with coverage 11 legal framework allows the coverage
One of the most prominent sustainability issues of the insurance industry is climate change. Until now it is not clear what the consequences of climate change are on the weather, and what the cultural, social and economical developments will be (Scholtens (2006)). However, it is perceived that climate change will have a major impact on the insurance industry, and the financial industry in general.8 The effects of climate change are likely to become manifest through changes in the spatial distribution, frequencies, and intensities of ordinary and catastrophic weather events.9 Climate change affects the requirements of insurability (see table 1). Risks and uncertainties are less easy to measure, variability of loss events will heighten, and the insurance sector may have insufficient capacity to cover certain risks (IPCC (2001)). This may lead to uninsurability of certain risks. There are several possibilities for the insurance industry to overcome uninsurability of certain risks (Scholtens (2006)).
The insurance industry also has to adapt to the new risks and opportunities caused by climate change. Mills, et al. (2005) offer several recommendations for the insurance industry to deal with the risks and opportunities. The insurance industry has to improve loss data collection and actuarial analyses, and needs to analyze the negative and positive implications of climate change on their business, investments and customers. Furthermore, the insurance industry should promote and support sustainable activities (for example, advanced building codes) and participate in climate research.10 Park (2005) argues that the insurance sector has acquired
7 Table from Scholtens (2006) 8 Mills and Lecomte (2006)
9 IPCC report ‘Climate Change 2001: Impacts, Adaptation, and Vulnerability’ 10
much knowledge from managing the risks posed by natural disasters and he argues that this knowledge should be better channelled to improve human security and reduce ecological vulnerability linked to global climate change. According to Mills and Lecomte (2006) to adaptive capacity of the insurance industry is large, and this sector has dealt successfully with other issues in the past.
differences are probably driven by differences in culture, law and market forces (Jeucken (2001)), I expect to find similar results for insurance companies. From the results of these papers I can derive the first two hypotheses.
Hypothesis 1: Insurance companies perform better on the social aspects of sustainability than on the environmental aspects.
Hypothesis 2: Insurance companies in Europe offer more often sustainable financial products than insurance companies in North-America and Japan.
Jeucken (2001) and Bijl (2007) find in their analyses that European banks perform better on sustainability than banks in the other two regions. Explanations for this result may lie in differences in the culture, law, and market forces between the regions (Jeucken (2001)). In Europe greater emphasis is on the environment, whereas in North-America and the Pacific the emphasis is on community involvement. The European governments use more often market instruments to stimulate companies in the development of more environmentally oriented products.11 I expect European insurance companies to be leading in sustainability as well, because of differences in culture, law and market forces.
Hypothesis 3: European insurance companies perform better on the sustainability criteria than insurance companies in the other two regions.
A final hypothesis that I want to test in this paper relates to the size of insurance companies. In this analysis both large and small insurance companies are included, and this makes it possible to examine whether there is a relationship between the size of insurance companies and sustainability performance. A study by Kolk, et al. (2001) assumes that larger companies have better integrated sustainability in their business activities, since they have a greater social responsibility and face more pressure from stakeholders to be transparent about their activities. Larger companies are also being monitored more by the public. Therefore, they have to justify their actions more often. Sarmento, et al. (2007) analyse the environmental strategies for a sustainable environment of potential polluting companies in Portugal. They find that large environmental efforts on innovation are positively related with company size.
11
Small enterprises have less environmental innovation and lower employees’ environmental training. I expect that for insurance companies sustainability performance also is positively related to size. Larger insurance companies have a greater social responsibility and are being monitored more by the public. This leads to the final hypothesis.
3. Methodology
In this chapter I will discuss the data and methodology of this paper. I will start with an elaboration of the selection of the insurance companies. Then the different analyses are explained. Finally, the sustainability criteria are described.
3.1 Selection of insurance companies and financial conglomerates
In this analysis large and small insurance companies are included. Inclusion of both types gives an accurate overview of the sustainability performance of the insurance industry. The inclusion of insurance companies with a different company size makes it also possible to examine whether there is a relationship between company size and sustainability performance. Four indicators for company size are employed, namely total assets, number of employees, net premiums collected and shareholders’ equity. Insurance companies are selected on whether they provide financial statements or an annual report on their website in English or Dutch. If a company does not provide a financial statement or annual report on their website, or the report is in another language, the company is excluded.
I used a report by the EU to select the financial conglomerates. 12 In this report a selection is made of the financial conglomerates with their headquarter in Europe. Several conglomerates where excluded, as they only have a website in their own language. For North-America, Canada, and Japan financial conglomerates are chosen on the basis of the information provided on the company’s website, and by a list on financial conglomerates of Wikipedia.13 To determine the insurance companies for Japan I used the membership lists of The General Insurance Association of Japan, and The Life Insurance Association of Japan.14
For the regional analysis the insurance companies are separated into three geographical regions and separation is according to the location of their headquarter. These three regions are Europe, North-America, and Japan. The selected insurance companies are divided in four groups for the group analysis, namely in financial conglomerates (FCs), life insurance companies, general insurance companies, and mixed insurance companies. FCs are large financial companies which are active in different financial sectors (banking, insurance, etc.), and usually across borders. Life insurance companies only offer life insurance products, and general insurance companies offer products for non-life insurance (health and casualty
12http://ec.europa.eu/internal_market/financial-conglomerates/docs/20060424_conglomerates_bycountry_en.pdf 13 http://en.wikipedia.org/wiki/List_of_Investment_Banks
14 http://www.sonpo.or.jp/e/about_us/members.html
insurance products). The insurance companies which offer both types of products, life and non-life, are included as a mixed insurance company. Information on the websites of these companies, regarding the products they offer, is used to determine the type of insurance company.
An overview of the number of companies is given in table 2, and a division is made between the different groups and regions. In total 153 companies are included. For Japan only 14 insurance companies are included in this analysis, this is mainly due to the fact that many Japanese insurers do not publish a report in English. An extended overview of the included companies and their size characteristics can be found in Appendix I.
Table 2: Overview of number of FCs and insurance companies included in the analysis
FCs Life General Mixed Total
Europe 54 11 6 17 88
North-America 7 8 12 24 51
Japan 3 3 5 3 14
Total 64 22 23 44 153
3.2 Methodology
In this paper the sustainability performance of insurance companies is analyzed on different levels. First, the performance of insurers in the three geographical regions will be compared with each other. Then I will analyze the sustainability performances on a national level. Finally, the performances per group will be evaluated. These three analyses will give an accurate overview of the sustainability performance of insurance companies. It is then possible to identify per region, country and group those companies that are performing well and those that need to improve their sustainability performance.
The analyses are based on the performance of the insurance companies on 23 sustainability criteria. These sustainability criteria can be divided in four categories (Bijl (2007)), namely
1. sustainability reporting, behavioural codes, and environmental care systems 2. environmental sustainability in practice
3. social-economical activities 4. governance codes
These criteria will be described in the next section of this chapter. The division of the criteria in these categories is optimal as it highlights the different aspects of sustainability. Insurance companies show that they are aware of the importance of sustainability, that they want to integrate sustainability into their business conduct, and want to inform their stakeholders by publishing a sustainability report, signing behavioural codes, and/or by certifying their environmental management systems.15
Environmental sustainability in practice and social-economical activities are methods through which companies show their commitment by adopting actual sustainable activities.16 Criteria in these categories show how sustainability is integrated in the business activities, and in what extent companies are transparent about their performance and goals. These criteria examine if a company has internal systems, and offers sustainable products. The second category focuses on environmental aspects of sustainability and the third on social aspects of sustainability. Sustainability is not only related to the environment, but also business activities focusing on social aspects are important.
Corporate governance relates to the ‘ways in which the suppliers of finance to corporations assure themselves of getting a return on their investment’ (Shleifer and Vishny (1997)). It is preoccupied with the ways in which a corporation’s insiders can credibly commit to return funds to outside investors and can thereby attract external financing (Tirole (2005)). Good corporate governance ensures transparency, fairness and accountability (OECD (2002)) and by building confidence and trust the corporation has access to external finance and can make reliable commitments to creditors, employees and shareholders. Good corporate governance principles are important for the creation of sustainable long-term value for all stakeholders of a company.17 Corporate governance is related to another aspect of sustainability, ethics, and
15 Bijl (2007) 16 Bijl (2007) 17
for that reason I include two criteria related to corporate governance in the analysis. In this analysis only two criteria are included, because the main focus in this paper is on the environmental and social aspects of sustainability.
The sustainability performance will be analyzed on a regional level first. To perform this analysis the average performance of the companies from the three regions will be compared. This comparison will be done for all 23 criteria. Then I analyze the sustainability performance on a national level. The performance of companies from a country is compared with the average performance of all companies. A two-sided t-test examines whether a performance is significantly different.18 The two-sided t-test compares the values of observations with the average and standard deviation. A critical value indicates whether an observation is significantly higher or lower than average. This test is only valid if the values of the observations are normally distributed. The one-sided Kolmogorov-Smirnov test is used to examine whether the observations are normally distributed.19 With this test the cumulative relative frequency, and the cumulative probability of these observations is constructed (as though the data series would be normally distributed). Subsequently, the absolute distance between these two frequencies is derived. Then the largest absolute distance is compared to the critical value. The null hypothesis of a normal distribution is rejected when this distance is larger than the critical value.20 The performances also are compared on the group level. To test whether the performances significantly differ between the groups I use a one-factor Analysis of Variance (ANOVA).21 ANOVA is a statistical technique which compares the average performance of the groups to determine if they significantly differ, details of this technique are included in Appendix III.
Besides these analyses three additional tests are included to expand the analysis of the sustainability performances. First, I will examine whether there is a relation between the company size and sustainability performance. I expect a positive relationship between these two variables (hypothesis 4), and I use the Spearman rank correlation test to determine this relationship. This test is a technique which tests the direction and strength of a relationship between two variables.22 Second, I will compare the performances of the companies with their inclusion in sustainability indices. In the analysis of Bijl (2007) the sustainability
18
See Appendix III
19 See Appendix III 20 Voorhoeve (2002), p. 5-6 21 See Appendix III 22
performances of banks also are compared with the inclusion in sustainability indices. Again the Spearman rank correlation determines if there is a significant correlation. Worldwide there are several sustainability indices and only companies that satisfy specific sustainability criteria are included. So, insurance companies which are included are rewarded for their performance on sustainability. I will elaborate on these indices in the next section of this chapter. This test will verify if my results are corresponding to the inclusion of companies in sustainability indices, given that companies that are included in sustainability indices have to comply with certain sustainability standards and rules. The Spearman rank correlation test determines whether there is a relation. Third, I will examine the correlations between the performances on the four categories and the 23 criteria. Correlations are determined by the Spearman rank correlation test. It will be interesting to determine whether the performance on the corporate governance principles is correlated with the performances of the other categories.
3.3 Criteria
The sustainability performance of FCs and insurance companies is analyzed according to 23 sustainability criteria. These criteria emphasize environmental, social and corporate governance aspects. In table 3 an overview of these criteria is given and the sustainability indices also are included in this table. The criteria and the sustainability indices are briefly described in this section.23 The table also shows how these criteria are measured and what source is used to gather information. The most important source is the sustainability reports of the FCs and insurance companies. When this report is not available, the environmental or social report is used. Ultimately, the financial or annual report is used if none of the abovementioned reports are available. The latest version is always used. I also use information provided on the official websites of the companies, and information from official websites of the criteria. I assume that a company does not follow a behavioural code nor has, for example, an environmental management system if this is not stated in their report or on their webpage.
The following criteria are included in this analysis,
- Behavioural codes, sustainability reporting and environmental care systems
23
o Sustainability reports
o Environmental statements and behavioural codes
o Standards for and certification of environmental management systems - Environmental sustainability in practice
o Policy: environmental and supply chain management
o Transparency on internal and external environmental data and goals for the near future
o Environmental risk analysis, sector-exclusion, World Bank and OECD principles
o Financial products with a sustainability element - Social-economical activities
o Community involvement
o Sponsoring
o Work conditions for employees - Corporate Governance codes
o Company ethics (Code of Conduct)
o Compliance with a Code
Sustainability reporting, behavioural codes and environmental care systems.
Through an annual sustainability report, adoption of behavioural codes, or implementation of environmental care systems a company shows that it is committed to sustainability and willing to take social, ethical and environmental issues into account. The company understands the importance of the environmental and social aspects of sustainability.
Sustainability reporting
Currently, stakeholders are not solely interested in the financial performance of companies and therefore companies have to go further. Stakeholders are also interested in activities related to sustainability, ethics, and society.24 Sustainability or Corporate Social Responsibility (CSR) reports are a way in which companies can communicate to their stakeholders how they are engaged in sustainability activities. These reports contain information on the position of the company towards the environment, society and its employees.
24
Table 3: Overview of the sustainability criteria, scale and source
Nr. Criterion Scale Source
1 Sustainability reporting Year of first published report Company website
2 ICC Business Charter Adoption of guidelines (yes/no)
Website and sustainability reports
3 UNEP FI Adoption of guidelines (yes/no)
www.unepfi.org, and www.unpri.org
4 Equator Principles Adoption of guidelines (yes/no) www.equator-principles.com 5 Global compact Adoption of guidelines (yes/no) www.unglobalcompact.org
6 Who cares wins Participation in publication (yes/no)
Who cares wins statement, www.unglobalcompact.org
7
Environmental management
systems (certified) ISO14001, EMAS
Website and sustainability reports
8 Environmental policy Policy designed by insurer (yes/no)
Website and sustainability reports
9 Supply chain management Policy designed by insurer (yes/no)
Website and sustainability reports
10
Transparency environmental
performance Quantitative/qualitative
Website and sustainability reports
11
Transparency environmental
goals Quantitative/qualitative
Website and sustainability reports
12 Environmental risk analysis Usage of risk analysis (yes/no)
Website and sustainability reports
13 Sector exclusion Usage of sector exclusion (yes/no)
Website and sustainability reports
14 World Bank guidelines Adoption of guidelines (yes/no)
Website and sustainability reports
15 OECD guidelines Adoption of guidelines (yes/no)
Website and sustainability reports
16 Sustainable financial products Supply of these products (yes/no)
Website and sustainability reports
17 Community involvement
Involvement through foundation or voluntary work (yes/no)
Website and sustainability reports
18 Sponsoring Sponsoring activities (yes/no)
Website and sustainability reports
19 Working conditions
Training, equal career opportunities (yes/no)
Website and sustainability reports
20 Company ethics
Code of Conduct, Code of Ethics (yes/no)
Website and sustainability reports
21 Compliance Compliance with a Code (yes/no)
Website and sustainability reports
22
Dow Jones Sustainability
Group Index Inclusion in DJSGI DJSI World Constituent Data
23 FTSE4Good Inclusion in FTSE4Good (yes/no)
FTSE4Good Global Index Constituent Data
24 Ethibel ESI Excellence Europe
Inclusion of European insurers in
Ethibel (yes/no) Ethibel constituent Data
25 Domini Social Index DSI
Inclusion of North-American insurers in
DSI (yes/no) DSI constituent Data
26 Carbon Climate Index
Inclusion of insurers in the Carbon
The provision of sustainability reports is a good criterion in this analysis to measure sustainability performance, as it indicates that a company is involved in aspects of sustainability and wants to communicate this to their stakeholders (Bijl (2007)). These reports can focus specifically on environmental aspects of sustainability, but also on social aspects.
Environmental Statements and Behavioural Codes
Through environmental statements and behavioural codes a company states that it is actively and concretely willing to integrate sustainability in her business conduct, and complies with the associated rules of a statement or code (Bijl (2007)). Therefore, these criteria are a good measurement of sustainability performance. In this subsection I will describe the statements and codes used in this analysis to examine sustainability performance.
One of the criteria is the International Chamber of Commerce (ICC) Business Charter for Sustainable Development. The ICC developed this Charter in response to the World Commission on Environment and Development report, and this Charter sets out 16 principles for environmental management. Environmentally relevant aspects of health, safety and product stewardship are covered by this Charter. This criterion solely measures environmental aspects of sustainability.25
Secondly, the United Nations Environment Programme statement by Financial Institutions (UNEP FI) on the Environment and Sustainable Development is used as a criterion. The UNEP FI is a global partnership between the UNEP and the private financial sector. FCs and insurance companies that have signed this statement show that they want to achieve sustainability. Both environmental and social aspects are involved in this criterion.
Another criterion is the Equator Principles and whether a FC or insurance company has signed these. The Equator Principles are a financial industry benchmark for determining, assessing, and managing social and environmental risk in project financing. Companies which have adopted these principles have to ensure that the projects they finance are developed in a manner that is socially responsible and reflects sound environmental management practices.26 This criterion also concerns both environmental and social aspects.
The final code is the Global Compact of the UN and the related ‘Who Cares Wins’ statement. The Global Compact is an international program which joins companies, UN agencies, labour and civil society to support universal environmental and social principles. The Global Compact contains ten principles in the areas of human rights, labour, the environment and
25 www.bsdglobal.com 26
anti-corruption. The ‘Who Cares Wins’ is a report with which the financial sector states that social, environmental and governance practices are of extreme importance to the functioning of the financial markets.27 Both environmental and social aspects of sustainability are incorporated in these criteria.
Standards for and certification of environmental management systems
An environmental management system is preferably a part of the normal management system and focuses specifically on the control and improvement of environmental performances. Through an environmental management system the focus is continually on the position of the environment within the business conduct. Two important starting points are central, namely the compliance with legislation and rules and the control of environmental risks. The second starting point is the aim for a permanent improvement of environmental performances.28 There are two possibilities for certification of environmental management systems, namely certification by the ISO 14001 and/or by the EMAS standard. These criteria are solely involved with environmental aspects of sustainability.
Environmental sustainability in practice
Environmental sustainability in practice can be tested by several criteria. Insurance companies can use an environmental and supply chain management to examine the sustainability performances of their suppliers (Bijl (2007)). Transparency of environmental performances and goals indicate in what extent insurance companies are active in environmental sustainability. Environmental sustainability is also expressed by the environmental risk analyses of insurance companies, sector exclusions, and whether they follow certain guidelines.
Environmental and supply chain management
An environmental policy is developed by FCs and insurance companies when they want their business activities to be environmentally responsible. In this policy it is reflected how these companies decrease their impact on the environment, and how they manage environmental risks (Bijl (2007)). With the environmental policy a company goes along with society’s demand to become more transparent regarding environmental aspects of their business activities. FCs and insurance companies use their supply chain management as a guideline for
27 www.globalcompact.org 28
their relations with suppliers (Bijl (2007)). Companies that use a supply chain management expect that their suppliers have similar practices regarding sustainability. Some FCs or insurance companies exclude suppliers, if these suppliers do not obey the same environmental and/or social practices as them. The environmental policy is only concerned with the environmental aspects of sustainability, but the supply chain management can also incorporate social and ethical aspects.
Transparency of environmental performances and goals
Information over environmental performances and goals is usually included in the sustainability reports of companies. As mentioned before, insurers realize that they have a social responsibility and that they can help to achieve a more sustainable situation. This development leads to more transparency on environmental performances and goals of insurance companies. A company can be transparent in two manners, namely in a quantitative or qualitative manner. Quantitative means that for example reductions or goals are stated in numbers or percentages, while in a qualitative manner no numbers are given but only a statement. Thus, transparency in a quantitative manner is more accurate.
The two criteria are incorporated in this analysis, because they indicate that an insurance company is working on its environmental performance and satisfies the need of stakeholders to report on this performance. Furthermore, when a company drafts environmental goals for the coming years it is committing to a certain performance. When a company is transparent in a quantitative manner this is more accurate and real than transparency in a qualitative manner.
Environmental risk analysis, sector-exclusion, World Bank- and OECD-principles FCs and insurance companies which perform environmental risk analyses want to examine the sustainability of the companies they invest in. When insurance companies exclude a specific sector with a higher environmental risk, they can reduce their own environmental risks and employ their influence to stimulate sustainability. Furthermore, FCs and insurance companies can create more awareness of the environmental impact of companies in other industries and possibly will be able to force them to change their policy towards a more sustainable one.
companies.29 Environmental risk analysis, exclusion and the principles are good criteria of sustainable performance by FCs and insurance companies, and they are only involved with environmental aspects. If an insurance company satisfies these criteria it indicates that they exert pressure on other companies to become more sustainable as well.
Financial products with a sustainability element
Insurance companies and FCs can offer insurance products which have a sustainability element. With these products they communicate to their costumers that they are committed to the environment and want to contribute to improve it. Insurance companies can offer various products with a sustainability aspect. For example, life insurance products of which the premiums are only invested in sustainable funds. These sustainable funds only contain companies and/or projects which do not deteriorate the environment or society. This criterion is mainly involved with environmental aspects of sustainability, and companies that satisfy this criterion are helping the development of sustainability.
Social-economical activities
In addition to environmental aspects social and economic aspects are also an important element of sustainability. This subsection discusses several criteria to analyze in what extent FCs and insurance companies incorporate social-economical aspects of sustainability in their business activities.
Community involvement
Community involvement by companies indicates that in some way a company is involved in, and wants to stimulate the (local) community. This involvement can be in different forms. Several FCs and insurance companies have created a foundation, this foundation then decides which projects are funded and which donations are made. These foundations can be oriented on the local community or more globally. Voluntary work is another possibility of community involvement. Usually employees of the company are involved in voluntary projects, for example teaching or building activities. Community involvement is an excellent criterion for the social aspects of sustainability performance, while it indicates that an insurance company or FC actively participates in the community and wants to improve it.
29
Sponsoring
Two types of sponsoring are support for a specific project by a FC or insurance company, or a partnership with an organization. Sponsoring is very closely related to community involvement, as both concern the (local) community. Different forms of projects for sponsoring exist, for example sponsoring of sport-related activities, art, music, environmental projects or health-related projects. Sponsoring shows that a company is concerned with the community and wants to help to develop the community. Hence, in addition to community involvement sponsoring is a good criteria for sustainability performance, and only social aspects of sustainability are involved.
Work conditions for employees
In this analysis the work conditions for employees focuses on development possibilities and equal career opportunities. Development possibilities include training, educational possibilities, etc. Equal career opportunities are an indicator of social aspects of sustainability. Not only are equal opportunities for men and women relevant, but also for persons with different ethical backgrounds. More companies are not only focused on equal opportunities but also on the possibilities and benefits of diversity.30 Working conditions for employees is an important element of a company’s social policy, and is oriented toward the internal social aspects of sustainability.
Corporate Governance Codes
This category reviews the way the FCs and insurance companies conduct relations between the supervisory board, managing board, and shareholders. The criteria included in this category measure in what extent companies want to create long-term sustainable value for their stakeholders.
Company ethics (Code of Conduct)
Company ethics are summarized in the company’s Code of Conduct or Code of Ethics, and they specify how a company resolves ethical issues. In a Code of Conduct the company’s social responsibilities, and its standards and values are stated. The Code of Conduct is a set of rules outlining the responsibilities of, or proper practices for, a company.31 There are differences between the existing Codes as regards to the required extent of the Code. A
30 Chapter 2, p. 31
company which has a Code of Conduct agrees to several aspects of social responsible business activities. Therefore, the Code of Conduct is a good criteria for social elements of the sustainability performance of the FCs and insurance companies.
Compliance with a Code
On the international and national level rules and principles regarding corporate governance have been made for companies. In every country different corporate governance codes exists. For example, in the United States the Sarbanes-Oxley act and the Tabaksblat Code in the Netherlands. The Sarbanes-Oxley act was created in response to several major corporate and accounting scandals (for example Enron), and these scandals reduced the public trust in accounting and reporting practices. The European Commission stated that every member state has to draft a code, and companies have to comply or explain why they do not comply. In Japan the Japan Corporate Governance Forum issued corporate governance principles. Provisions in the codes relate to compliance with and enforcement of the code, the management and supervisory board, financial reporting, and shareholders. Research has been done on the compliance with the codes, and these papers find a high level of compliance (Akkermans, et al. (2006), Werder, et al. (2005)). However, these authors note that compliance may only reflect symbolic compliance and not actual compliance.32 Compliance with a code reflects social aspects of sustainability.
Sustainable indices
The inclusion in sustainable indices is only used as a control device in this analysis. These sustainable indices are good measurements of sustainable performance, as these indices show which companies are leading in sustainability performance. Inclusion in sustainability indices is only possible if a companies complies with certain standards and rules. The five sustainability indices which will be used as a control device are the following33,
o Dow Jones Sustainability Group Index
o FSTE4Good
o Ethibel ESI Excellence Europe
o Domini Social Index
o Carbon Climate Index
32 Akkermans, et al. (2006), p. 17 33
4. Results
In this chapter the results of the analyses of the sustainability performances of insurance companies will be discussed. I will begin with the analysis on a regional level. The performances of the companies from Europe, North-America, and Japan are compared per criterion. Subsequently, the performances on a national level will be analyzed. The two-sided t-test determines whether a country is performing better or worse than average. The individual performances of the insurance companies are analyzed as well. The insurance companies are ranked according to their overall performance. Then the sustainability performances on the group level are discussed. The ANOVA test determines whether the performances between the groups are significantly different. In addition to these three analyses, I test if there is a relationship between company size and sustainability performance. Furthermore, a comparison is made between sustainability performance and inclusion in sustainability indices. Finally, the correlations between the performances per category and per criterion are examined. Results from these tests indicate whether performances per category and criterion are related to each other.
4.1. Regional analysis
In this section the sustainability performances of the companies are compared on a regional basis. Three regions are distinguished in this analysis, namely Europe, North-America and Japan. In table 4 the performance is given per region, and the overall performance is included in the table. The number of companies that satisfy a criterion also is given. Performance is measured as a percentage, and this percentage reflects the number of companies within a region that satisfy the criterion.
Sustainability reporting, behavioural codes and environmental care systems
Approximately 70% of the Japanese and European insurance companies publish a sustainability report or have a separate section on sustainability in their annual report. In North-America only 20% of the insurance companies publish a report, compared to Japan and Europe this number is low. Similar results are found in other papers on sustainability reporting of companies (i.e. Kolk (2002), and Scott (2003)). Relatively more companies follow the UNEP FI and Global compact principles compared to the other principles. Only European companies participate in the ‘Who Cares Wins’ project and have an EMAS certification. A reason for this can be that the ‘Who Cares Wins’ project and the EMAS certification are both European initiatives.
Table 4: Performances of the insurance companies on the sustainability criteria on a regional level
(performances are given in percentages and the number of companies that satisfy a criterion)
Europe North-America Japan Total
nr. criterion N % N % N % N %
Sustainability reporting, behavioural codes and environmental care systems
1 Sustainability reporting 58 67% 10 20% 10 71% 78 51%
2 ICC Business Charter 9 10% 2 4% 1 7% 12 8%
3 UNEP FI 33 38% 6 12% 6 43% 45 30%
4 Equator Principles 14 16% 4 8% 2 14% 20 13%
5 Global compact 27 31% 0 0% 6 43% 33 22%
6 Who cares wins 8 9% 0 0% 0 0% 8 5%
7 ISO 14001 28 32% 1 2% 8 57% 37 24%
8 EMAS 9 10% 0 0% 0 0% 9 6%
Environmental sustainability in practice
9 Environmental policy 52 60% 8 16% 8 57% 68 45%
10 Supply chain management 29 33% 3 6% 0 0% 32 21%
11 Transparency environmental performance 51 59% 7 14% 9 64% 67 44%
12 Transparency environmental goals 51 59% 7 14% 9 64% 67 44%
13 Environmental risk analysis 31 36% 3 6% 4 29% 38 25%
14 Sector exclusion 8 9% 2 4% 0 0% 10 7%
15 World Bank guidelines 16 18% 4 8% 3 21% 23 15%
16 OECD guidelines 7 8% 0 0% 0 0% 7 5%
17 Sustainable financial products 33 38% 5 10% 5 36% 43 28% Social-economical activities
18 Community involvement 67 77% 30 59% 14 100% 111 73%
19 Sponsoring 67 77% 31 61% 13 93% 111 73%
20 Education/training 62 71% 11 22% 5 36% 78 51%
21 Equal career opportunities 53 61% 20 39% 5 36% 78 51% Corporate governance
22 Company ethics 54 62% 25 49% 6 43% 85 56%
Environmental sustainability in practice
In this category performance per criterion differs (table 4). Approximately half of the insurance companies has an environmental policy and is transparent about their environmental performances and goals. While on the other criteria total performance is much lower. Again European and Japanese FCs and insurance companies perform better than North-American companies.
Only European and North-American insurance companies exclude sectors, and have a supply chain management. One-third of the European insurers have a supply chain management compared to 6% of the North-American and none of the Japanese insurers. Bijl (2007) finds that North-American banks are performing best on this criterion. An explanation for this difference in result can lie in the inclusion of smaller insurance companies in this analysis, whereas Bijl only included large banks. Almost 40% of the European and Japanese insurance companies offer sustainable financial products compared to 10% of the North-American insurance companies. North-American companies also are less transparent about their sustainability performances and goals compared to European and Japanese companies.
Social-economical activities
The performances of the insurance companies on these criteria are less diverse than on the previous two categories. None of the regions performs better on all criteria. The performances can be found in table 4. More than 70% of all insurance companies are active in the community and in sponsoring, and more than half of the insurance companies offer educational possibilities for their employees and equal career opportunities. The social-economical aspects of sustainability are the most integrated in the business activities of insurance companies.
Corporate governance
The performances on the two criteria in this category are very similar, approximately half of all companies has a Code of Conduct and complies with a Code. Relatively more European insurance companies have a Code of Conduct and comply with a code.
Less than half of the North-American and Japanese companies have a Code of Conduct. However, it is interesting to find that for these two criteria North-American companies perform a bit better than Japanese companies. Many of the smaller companies included in this analysis are not listed on a stock exchange and therefore they are not obliged to comply with a Code, and this can explain the lower percentage of companies that complies.
4.2. National analysis
The performances of the FCs and insurance companies also are compared on a national level. This analysis will give an overview on the country level and which companies are performing better or worse than average. In table 5 the performances of the individual criteria are aggregated into the four categories, and total performance is given (this is the average of the performances on the four categories). The performance per category is the average of the individual performances per criterion included in the category. This table includes also the associated p-values of these percentages. The p-value indicates whether the performance is significantly different than average or not significantly different than average. 34 The national performances per criterion can be found in Appendix V. I use a two-sided t-test, with a significance level of 5%, to determine whether the national performance significantly differs from the average overall performance. This test is valid since all data series are normally distributed according to the Kolmogorov-Smirnov test.35 In total companies from 20 countries are included in this analysis. The performances of Cyprus and Hungary are taken together, since only one company from Cyprus and one from Hungary is included in this analysis. Thus, the performances of 19 countries are compared with the unweighted average performance. Results from this analysis can give a mixed result, since for some countries relatively more small companies are included, and in general these companies perform less on the sustainability criteria (see results size analysis is section 4.5).
34 Critical values estimated with 17 degrees of freedom and a 5% significance level 35
Sustainability reporting, behavioural codes and environmental care systems
Total performance of this category is only 25% and this is low compared to the total performance of the other categories. Overall these criteria are not integrated in the business activities of many insurance companies, and this national overview indicates that many countries can make improvements to expand their national sustainability performance. In this category three countries perform better than average, namely France, Spain and Sweden. Performances between the countries differ. Spain has a total performance of 63% compared to 4% of the U.S. Seven countries perform worse than average, the U.S., Denmark and Cyprus/Hungary have a percentage below 10%. In these countries only a small number of insurance companies provide a sustainability report, adopted a code or has an environmental care system. The national performance per criterion (Appendix V) indicates that in this category a small number of companies have adopted behavioural codes or have an environmental care system (most countries perform worse than average on these criteria). In 6 countries all companies publish a sustainability report and approximately half of the countries perform better than average on this criterion. Of the 19 countries only in 5 countries one or more insurance companies have an EMAS certification. Only in 6 countries none of the insurance companies has an ISO 14001 certification, hence more companies have an ISO 14001 certification of their environmental management system. A reason for this result can be that EMAS is an European initiative, however in several European countries none of the insurance companies has this certification.
Environmental sustainability in practice
Table 5: National performances per category and total performance of the insurance companies on sustainability
(performance is given in percentage with the corresponding p-value)
Country nr. of companies Reports, etc. p-value Practice p-value Social p-value Governance p-value Total p-value Austria 2 13% 0,01* 22% 0,00* 88% 0,01** 75% 0,06 50% 0,08 Belgium 2 31% 0,05 44% 0,01** 75% 0,10 50% 0,02* 50% 0,08 Czech Republic 3 4% 0,00* 0% 0,00* 58% 0,04* 0% 0,00* 16% 0,02* Denmark 5 8% 0,00* 11% 0,00* 25% 0,00* 50% 0,01* 24% 0,03* Finland 2 13% 0,01* 39% 0,09 88% 0,01** 50% 0,01* 48% 0,10 France 7 48% 0,01** 62% 0,00** 100% 0,00** 93% 0,01** 76% 0,00** Germany 7 29% 0,10 30% 0,10 39% 0,00* 64% 0,09 41% 0,04* Ireland 2 19% 0,08 39% 0,09 100% 0,00** 100% 0,00** 65% 0,01** Italy 7 27% 0,12 33% 0,10 68% 0,09 100% 0,00** 57% 0,07 Netherlands 12 15% 0,01* 22% 0,00* 54% 0,03* 58% 0,10 37% 0,03* Norway 3 29% 0,10 52% 0,01** 83% 0,01** 83% 0,02** 62% 0,01** Spain 4 63% 0,00** 64% 0,00** 100% 0,00** 88% 0,01** 79% 0,00** Sweden 4 41% 0,01** 53% 0,01** 75% 0,10 63% 0,09 58% 0,03** Switzerland 5 25% 0,13 36% 0,09 65% 0,09 80% 0,03** 52% 0,07 UK 21 28% 0,11 40% 0,06 89% 0,01** 79% 0,04** 59% 0,05** US 44 4% 0,00* 7% 0,00* 42% 0,00* 41% 0,02* 24% 0,03* Canada 7 14% 0,01* 17% 0,00* 64% 0,09 36% 0,01* 33% 0,03* Japan 14 29% 0,10 30% 0,10 66% 0,09 32% 0,00* 39% 0,04* Other 2 6% 0,00* 39% 0,09 88% 0,01** 50% 0,01* 46% 0,09 Total 153 Unweighted average 25% 35% 72% 65% 49% SD 0,1528 0,1624 0,2171 0,2284 0,1592 Note: ** significantly better than average
one-third of the companies offer sustainable products. On this criterion only 8 countries have a performance better than average. The performance on the sector exclusion criteria is the lowest, and in many countries none of the insurance companies excludes sectors from their activities. This performance is disappointing, since insurance companies can exercise their influence to improve the sustainability activities of other industries through the exclusion of sectors. In the analysis of Bijl (2007) also a small number of banks exclude specific sectors.
Social-economical activities
In this category seven of the 19 countries have a sustainability performance above average, and only four have a performance worse than average (Denmark, Germany, the Netherlands, and the U.S.). The total performance of this category is 72%, which indicates that the social aspects of sustainability are well integrated in the business activities of insurance companies. In three countries all insurance companies satisfy the four criteria in this category, namely France, Ireland and Spain. France and Spain are performing better than average on all categories. However, Ireland only performs better than average on this category and the corporate governance category, and does not perform significantly different than average in the other two categories. In Ireland the social and ethical aspects of sustainability are better integrated than the environmental aspects. In this category performances also are not much dispersed, only the performance of Denmark is relatively low. In this category the average performance of the community involvement and sponsoring criteria are the highest (Appendix V). More than half of the countries have a performance better than average on these two criteria, and only the Czech Republic a low number of companies satisfy these criteria (20%). On the other two criteria in this category performances are more dispersed. In general a company that performs better on the training criterion also performs better on the equal career opportunities criterion.
Corporate governance
and Cyprus/Hungary, which perform well on the social aspects of sustainability also have a good performance in this category. The average performance is higher the compliance criterion than on the Code of Conduct criterion (Appendix). Insurance companies that are listed on a stock exchange are obliged to comply to a code and this can explain the better performance.
4.3. Individual analysis
From the national analysis it can be concluded that some countries are performing worse than average. However, in the national analysis insurance companies that have an excellent performance remain unobserved. The national performance may not always accurately reflect the performance in a country, because it is an average. In this section the individual performances are analysed to determine which companies are performing well, and to obtain an additional perspective besides the national analysis. In appendix VI a ranking is made of the total performances per insurance company. Total performance is given as the percentage of the number of criteria which are satisfied. The performance per category is also given for the insurance companies. Allianz is performing the best on the sustainability criteria, followed by ABN AMRO, BBVA, Barclays and HSBC. The best performing insurance companies usually satisfy all the social and governance criteria, and perform well on the environmental sustainability in practice criteria. Performance on the reports, etc. category is usually lower, and several of the insurance companies in the top 20 have a performance around the 50%. In the national analysis it was also found that insurance companies perform better on the other three categories. Differences between the sustainability performance of insurance companies are large.
only 2 of the 44 U.S. insurance companies have a total performance above 50% (Citigroup and J.P. Morgan Chase). Most U.S. insurance companies have a high rank number (a low performance), thus the national analysis of the U.S. gives an accurate view of this country. Differences in the performance within countries can perhaps partially be explained by company size, and in section 4.5. it is examined whether this relationship exists.
From this analysis it also becomes apparent that the 24 best performing insurance companies are all financial conglomerates, and the worst performing companies are mixed insurance companies. In the next section the performances between the four groups will be examined, and from this analysis it will become evident which group is performing best per criterion.
4.4. Group analysis
In this section the performances of the companies are analyzed on the group level. The insurance companies are divided in four groups, namely in financial conglomerates, life insurers, general insurers and mixed insurers. This division is based on the products offered by the insurance companies. A comparison will be made per criterion, and I use an one-factor Analysis of Variance (ANOVA) to determine whether the average performances significantly differ between the groups.36 ANOVA is a statistical technique which compares average performance groups to determine if they significantly differ. The tables included in this section contain the percentage and number of the companies that satisfy a criterion, and this is given per group on a regional level. The overall performance per group and the results from the ANOVA test, the F-statistic and probability, also is given.
Sustainability reporting, behavioural codes and environmental care systems Sustainability reporting
An environmental or social responsibility report is published by 77% of the FCs (table 6). Reporting differs between conglomerates, several conglomerates publish a separate environmental report, while others publish a general sustainability report which includes social and governance issues. From the results I can conclude that financial conglomerates in Europe were leading in sustainability reporting, since they began to publish these reports first (see appendix IV). Scott (2003) and Jeucken (2001) also find that European companies have been publishing sustainability reports longer. Relatively more FCs publish a sustainability report compared to insurance companies in the other three groups. A reason for this difference
36
can be that FCs are relatively larger, have a greater social responsibility and face more pressure from stakeholders to be transparent.37 However, the p-value of the ANOVA test is larger than 0.05, thus the average performances between the groups are not significantly different and I can not conclude that FC perform better on this criterion.
Table 6: Number and percentage of insurers that publish a sustainability report
(and F-statistic and probability of the ANOVA test) Sustainability
reporting Total FC Life General Mixed F-statistic: Prob: N % N % N % N % N % 1,57 0,271 Total 78 51% 49 77% 9 41% 5 22% 15 34% Europe 58 66% 41 76% 8 73% 2 33% 7 41% North-America 10 20% 5 71% 0 0% 0 0% 5 21% Japan 10 71% 3 100% 1 33% 3 60% 3 100% Behavioural Codes
From table 7 it becomes apparent that only FCs signed the ICC Business Charter for Sustainable Development, and of the FCs only 17% signed this charter. Compared to the other criteria in this category this is percentage is low. The UNEP FI is signed by insurance companies from all four groups, but a larger fraction of the FCs signed this declaration. In Japan all FCs signed this declaration. The performances between the four groups do significantly differ on these two criteria (p<0.05), thus FCs are performing better on this criterion than the other three groups.
The Equator Principles are only signed by companies from two groups, namely by FCs and mixed insurance companies. From the mixed insurance companies only one company, Manulife in Canada, adopted these principles. In the FCs group, relatively more Japanese FCs adopted the principles. The result from the ANOVA test indicates that the performances between the groups significantly differ (p<0.05). Between the groups are differences in the average performance.
Table 7: Percentage of insurers that signed the ICC BCSD
(and F-statistic and probability of ANOVA test)
ICC Business Charter Total FC Life General Mixed F-statistic: Prob: N % N % N % N % N % 18,61 0,001 Total 12 8% 11 17% 0 0% 0 0% 1 2% Europe 9 10% 8 15% 0 0% 0 0% 1 6% North-America 2 4% 2 29% 0 0% 0 0% 0 0% Japan 1 7% 1 33% 0 0% 0 0% 0 0% 37
