Corporate sustainability reporting guidelines;
What is the trigger?
Master Thesis IE&B
Author: Geert Jan Grimberg
Student Nr: 1158147
Supervisor: Prof. Dr. van Ees
ABSTRACT
In this article I examined the creation and adoption of corporate sustainability reporting (CSR) guidelines across 65 countries worldwide. CSR is the process of communicating the social and environmental effects of organisations’ economic actions to particular interest groups within society and to society at large. CSR guidelines can be either mandatory or voluntary and provide a framework for companies to report on their environmental, social and economic impacts. Social contract theory is the generally accepted theory to explain the relation between companies and the society in which they operate. Social contract theory assumes the bounds of justice to be at the core of sustainability reporting.
Legitimacy theory extends upon that by assuming demands from within society at large to influence corporate behaviour (i.e. companies respond to demands from within society). According to accountability theory, companies have to comply with CSR guidelines in a country, because law requires them to do so. This also is an extension upon social contract theory.
Williamson’s institutional framework provides insight in the extent to which institutional boundaries, in the broadest sense, influence corporate behaviour. Williamson defined four levels of analysis in his ‘new institutional economics model’, of which I focus on level 1 and 2. Level 1 represents informal institutions, the norms, customs, mores and traditions of a society. The second level in the
framework presented by Willliamson represents the institutional environment and incorporates formal rules.
A country’s institutional boundaries, represented by culture and the institutional environment of a country, might determine the creation and adoption of national CSR guidelines. Next to these country specific forces, I incorporate economic integration, in order to account for globalisation effects.
Both country specific and internationalisation factors explain the diffusion of CSR guidelines; although institutional and cultural factors prevail over economic integration.
A countries’ legal system is to a minor extent of importance, although a civil-law system tends to have a more pronounced influence on the creation and adoption of CSR guidelines than a common-law system. The rule of law in a country,
representing the enforceability of rules, is of importance and has an effect on corporate compliance. Of the cultural variables proposed by Hofstede, individualism has a positive and significant and power distance a negative and significant
influence on the availability and number of CSR guidelines in a country.
INTRODUCTION
The expression corporate social responsibility has become part of the business language over the last decades. There is increasing attention for companies to behave socially responsible and perhaps even greater attention for those who do not behave responsible. A well-known example is the boycott of oil company Shell’s products in 1997. This was caused by Shell’s intentions to sink its floating oil installation (the Brent Spar) to the bottom of the Atlantic Ocean (Brent Spar’s long saga, BBC News).
Ten years before the Brent Spar saga, the World Commission of Environment published the Brundtland report, named after the chairman of the commission. The commission stated that sustainable development is “meeting the needs of the present without compromising the ability of future generations to meet their own needs”. This is a rather general statement, which goes for everyone within society. The Organisation for Economic Cooperation and Development states that corporate responsibility involves the search for an effective “fit” between businesses and the societies in which they operate. They state that the notion of "fit" “recognises the mutual dependence of business and society. A business sector cannot prosper if the society in which it operates is failing and a failing business sector inevitably detracts from general well-being. Corporate responsibility refers to the actions taken by businesses to nurture and enhance this symbiotic relationship with society.” (What is corporate responsibility?, OECD)
Increasing pressures on the business sector to act in the interest of the societies, in which they operate, sowed the seeds for an extension in corporate reporting. An infinite amount of expressions like corporate responsibility-, sustainability-, environmental-, social-, corporate citizenship- and triple bottom-line reporting all encompass the same idea; the communication of not only financial information, but moreover the societal and environmental impacts of the reporting entity (i.e. company). Gray et al. (1996) define corporate sustainability reporting (CSR) as: “The process of communicating the social and environmental effects of
organisations’ economic actions to particular interest groups within society and to society at large”.
number of companies publishing sustainability reports (or publishing social and environmental information in the annual report) can be noticed.
Over the years various CSR guidelines, providing a reporting framework for
organisations, have been established as well. The Global Reporting Initiative (GRI) is one of the most well-known issuers of voluntary CSR guidelines. The GRI was founded by amongst others the United Nations in order to establish a voluntary worldwide framework for sustainability reporting. Apart from the GRI, countries have started to develop national CSR guidelines, either mandatory or voluntary. The Danish government, in 1995, was the first to legislate annual environmental reporting for approximately 2000 Danish firms. The required content of the mandatory annual environmental report was threefold. The law required general company information in the first place. Secondly qualitative information of the environmental impact and thirdly quantitative information on the use of polluting substances in production processes, products and wastes as well as emissions to air, water and soil (Rikhardsson, 1996). Recently the British government was about to take a major step forwards in mandatory CSR reporting as well. The mandatory CSR framework would have required companies to include environmental, social and ethical information in an annual Operating and Financial Review (OFR). However, the government decided to abandon the projected reporting requirements (Brown to scrap reporting rule, Financial Times, November 28, 2005 and www.trucost.com, December 8, 2005).
CSR has been the subject of academic research since the beginning of the nineties. Prior research, especially within the field of accountancy, examined the relation between corporate characteristics and levels of corporate social disclosure (e.g. O’Donovan, 2002 and Gray, Javad, Power, Sinclair, 2001).
organisations and their stakeholders. Social contract theory states that
“organisations agree to operate within certain bounds imposed by society in order to enjoy continued access to product and resource markets” (Campbell et al. 2003).
The creation and adoption of national CSR guidelines, such as the Danish mandatory environmental reporting scheme, has not been examined in CSR literature so far. Within corporate governance literature, on the other hand, researchers have examined the diffusion of corporate governance codes worldwide and the factors triggering the diffusion of these codes. Aguilera and Cuervo-Cazurra (2004) state that, “the diffusion of organisational practices is to be seen as an innovation in the social system”. They found that institutional (endogenous) and internationalisation (exogenous) forces provide support for the creation and adoption of corporate governance codes. Although corporate governance codes and guidelines regarding CSR are not similar, they share the aim to increase corporate transparency. Therefore in this research I will analyse the following problem statement:
“Which factors determine the creation and adoption of national corporate sustainability reporting guidelines?”
THEORY AND HYPOTHESES
Introduction
Elkington (1999) states that “sustainability accounting referred to as triple bottom line accounting, aims to report on an organisation’s economic, environmental and social performance”. The expression tripple bottom line comprises the three elements: people, planet and profit. Corporate sustainability reporting (CSR) guidelines provide a common reporting framework for companies, in order to enhance, amongst others, the accessibility, transparency, comparability and
accountability of their social (people), environmental (planet) and economic (profit) performance. The most common set of guidelines, issued by the Global Reportig Initiative have been established through extensive stakeholder dialogue. In other words, these guidelines have been established on a mere practical set of
recommendations, best practices and opinion, instead of on a theoretical foundation.
The inclusion of environmental, social and economic information is the cornerstone of CSR, although differences in content between various guidelines might exist. The key question in CSR is what to report on within each of the three dimensions
(people, planet, profit). Lamberton (2005) states that because “sustainability is a multi dimensional concept it is not directly measurable and therefore requires a set of indicators to enable the performance towards its multiple objectives to be
assessed”.
The GRI and various national CSR guidelines draw upon the three-dimensional definition of sustainability and provide both quantitative and qualitative indicators assessing the organisation’s performance towards sustainable development. The GRI distinguishes qualitative and quantitative indicators. Qualitative indicators enable the assessment of organisational policies and programs in these cases where quantifying performance is impossible. This mainly concerns information on the social performance of a company. It is hard to quantify non-discrimination policies for example.
Environmental indicators concern a broad range of topics, including: materials, energy, water, biodiversity, emissions, waste and transport.
Social indicators, which are more often qualitative than economic and environmental indicators, concern labour practices and decent work, human rights, society and product responsibility.
Voluntary reporting guidelines, like the GRI guidelines described above, provide a CSR framework against which companies are able, but not enforced to report. Voluntary reporting guidelines tend to reduce information asymmetries between a company and its stakeholders (Brammer, S. and Pavelin, S., 2004). The voluntary nature of these guidelines does not imply that companies are not bounded by law. Although reporting is voluntary under a voluntary reporting scheme, voluntary reporting does take effect in a legal context. Public disclosure of negative information on for example environmental impacts might trigger compensation claims or prosecutions (Globalt ansvar, 2003).
On the other hand some critical remarks surround the voluntary nature of
guidelines, suggesting that it is hard to attain transparency through a stakeholder dialogue (O’Dwyer, 2005). Although voluntary corporate sustainability reporting guidelines, established through stakeholder dialogue, will result in the accounts produced, the relative power relations between corporations and some stakeholders mean that corporations may inevitably come to control the dialogue and reporting agenda (Owen et al., 2001; O’Dwyer, 2005).
Mandatory reporting guidelines, on the other hand, must be complied with by law, no matter what. According to the literature, theorists argue that legislating corporate sustainability reporting is beneficial to voluntary guidelines since it increases corporate transparency (O’Dwyer, 2005).
Larrinaga (2002) states that legislation is beneficial since it could:
Improve sustainability reporting by providing a mandatory reporting scheme for every company, which closes the gap between reporting and non-reporting companies.
Require companies to report on ‘bad’ news as well, instead of trying to conceal it.
Theoretical background
The theoretical concepts, explaining the creation and adoption of mandatory and voluntary CSR guidelines, adopted within this research stem from social contract theory. Social contract theory starts from the assumption that companies have a social contract with society in order to conduct their operations within the bounds of justice (Mathews, 1987). As society expects companies to act in congruence with the society at large, it also expects companies to report on its social and
environmental performance.
Legitimacy theory extends upon social contract theory and “entails behaviour where companies respond to demands of divergent interest groups, and act to legitimise their actions” (Tilt, 1994). In other words, legitimacy goes further than social contract theory by incorporating the response of companies to demands from within society. Social contract theory merely describes the existence of a relationship between the company and the society in which it operates, but not the consequence of this relationship.
Suchman defines legitimacy (1995) as “a generalised perception or assumption that the actions of any entity are desirable, proper, or appropriate within some socially constructed system of norms, values, beliefs and definitions”. Legitimacy theory does not entail that CSR is bounded by law. Within legitimacy theory, two broad stances have been identified; strategic and institutional legitimacy (Suchman, 1995).
Strategic legitimacy concerns a single-firm, managerial point of view, focussing on the extent to which companies influence the public opinion and seek for public support (Suchman, 1995; Wilmshurst and Frost, 2000). Criticism regarding CSR, pointing at companies trying to conceal their misbehaviour by voluntarily publishing biased sustainability information (Larrinaga et al., 2002), must be seen in the light of strategic legitimacy. Strategic legitimacy theory explains voluntary corporate sustainability reporting, irrespective of the availability of CSR guidelines. In other words, companies themselves determine sustainability disclosure instead of the society in which they operate.
Institutional legitimacy, as opposed to strategic legitimacy, stresses “the
part of a dynamic system within which institutions shape corporate behaviour. The creation and adoption of CSR guidelines, according to institutional legitimacy theory, can be seen as a dynamic institutional process with the objective to increase
corporate transparency and avoid companies from strategically legitimising their environmental and/or social performance.
Accountability theory, which also extends upon social contract theory, considers compliance with the law to be at the core of CSR (Gray, 1991, 1992; Gray et al. 1988 as quoted in Tilt, 1994). In other words, accountability extends upon both social contract theory and legitimacy theory, because it takes the law as the starting point for corporate compliance. Social contract theory assumes the bounds of justice to be at the core of sustainability reporting. Legitimacy theory assumes demands from within society at large to influence corporate behaviour. According to
accountability theory, companies have to comply with CSR guidelines in a country, because law requires them to do so.
Social contract theory, extended by legitimacy and accountability theory, provides a theoretical background for the creation and adoption of CSR guidelines. National CSR guidelines are the result of a dynamic societal process, in which institutional boundaries influence corporate behaviour. In the next paragraphs, I will elaborate upon these institutional boundaries and their influence on the creation and adoption of national CSR guidelines.
Institutional theory
Jeurissen (2004), states that “half of the social responsibility of corporations is determined by the surrounding society, which imposes institutional conditions within which companies operate”. Quantifying social responsibility in percentages, like Jeurissen does, might be questionnable. However, the general idea that institutions conditions determine the boundaries in which companies operate is elaborated upon within this paragraph. North (1991, p. 97) states that: Institutions are humanly devised constraints that structure political, economic and social action. These institutional conditions or constraints can be either “informal (sanctions, taboos, customs, traditions and codes of conduct) or formal (constitutions, laws and property rights)” (North, 1991).
upper level 1, which exists of norms, customs, mores and traditions (Williamson 2000, p. 596) represents informal institutions, which is in line with the division proposed by North. Licht (et al, 2005) assume that the informal level can be seen as culture, as it represents the “common knowledge” of a society. The second level in the framework presented by Willliamson represents the institutional environment and incorporates formal rules. In the third level Williamson introduces governance structures at the level of the firm, which focus on transaction cost economics. The last and fourth level comprises resource allocations. This level represents Neo-classical economics, in which marginal analyses are conducted.
The first and second levels of Williamson’s framework provide insight in the way in which a country’s culture and its institutional environment shape in part the behaviour of companies. I will first discuss the second level, which resembles the institutional environment of a country, represented by formal rules. Thereafter I will elaborate upon the role of national culture.
Institutional environment. Two broad legal systems exist: the common-law and civil-law legal system (La Porta et al. 1998, p. 1131-1132; Jaggi and Low, 2000, p.499-500). Civil-law legal systems are also referred to as code-law regimes and comprise French, German and Scandinavian legal regimes. The main differences between common and civil-law legal regimes are that common-law is first made by judges and then incorporated into legislature, whereas civil-law regimes depend upon statutes and comprehensive codes and the opinions of legal scholars. In other words, common-law is merely a regime of practices, where law originates in court, on a case basis. Civil-law, on the other hand, is formed by theorists, which put into practice ‘abstract’ rules. According to a Swedish research on the role of legal regimes in corporate citizenship, there is a lack of cases (litigation) within civil law legal regimes. They conclude that within civil-law legal regimes there is less emphasis on litigation and judicial pronouncement as a strategy for resolving difficult public policy dilemmas (Globalt ansvar, 2003). Another distinction between civil and common law lies in the financial risk of going to court. In the US for example each party pays its own cost, whereas in Europe the ‘loser’ has to pay all the costs, including those of the ‘winner’.
court are relatively high, ‘society’ has to rely on corporate information, instead of on litigation. CSR guidelines tend to overcome the information asymmetries between a company and ‘society’. Hence, I would argue that CSR guidelines arise in those legal regimes where less jurisprudence exists and cost of going to court are relatively high (i.e. civil law legal regimes). I therefore propose that:
H1a: CSR guidelines are more likely to be developed in countries with a civil-law regime than in countries with common-law legal regimes.
Accountability theory, as mentioned before, considers compliance with the law to be the core of CSR (Gray, 1991, 1992, and Gray et al., 1988; quoted in Tilt, 1994). La Porta et al. (1996) introduced the concept of rule of law in order to assess the quality of the law enforcement in a country. Corporate compliance with formal rules is more likely to happen in a country where the quality of the law enforcement is higher. I therefore propose that:
H1b: Mandatory corporate sustainability reporting guidelines are more likely to be developed in countries where the enforceability of rules is stronger.
National Culture. Newman and Nollen (1996, p. 754) present a definition of national culture, which they adopt from Beck and Moore (1985) and Hofstede
Incorporating national culture in quantitative research, like this research, seems to be rather difficult, since it is hard to quantify the concept of national culture. Hofstede provides a conceptual model for the quantitative analysis of national culture. He distinguished five dimensions, which all contain aspects of national culture; individualism, power distance, masculinity, uncertainty avoidance and long-term orientation. Individualism measures the extent to which countries reinforce individual or collective achievement and interpersonal relationships. Power distance is the extent to which people accept an unequal, hierarchical distribution of power (Hope, 2003) Masculinity stands for a preference for achievement, heroism, assertiveness and material success (Hope, 2003). Uncertainty avoidance (UA) measures the extent to which people feel uncomfortable with ambiguity and an uncertain future (Hope, 2003). Hofstede describes long-term orientation as characterised by persistence, ordering relationships by status and observing this order, thrift, and having a sense of shame, whereas short-term orientation is characterised by personal steadiness and stability, protecting your "face”, respect for tradition and reciprocation of greetings, favours, and gifts.
Hofstede quantified these five dimensions by interviewing IBM managers within over 50 countries between 1967 and 1973. Although the dimension long-term orientation might provide insight in the creation and adoption of CSR guidelines, there is a limited availability of data. Therefore I will not elaborate on this dimension, nor include it in the research.
The Hofstede model is surrounded by criticism. Critics point at the fact that he did not use a national representative sample, but used employee data from a large multinational (Licht et al, 2004, p. 238). In other words, critics do not believe IBM managers in a country to represent the national culture of the respective country. Another point of criticism is the fact that data obtained at the beginning of the seventies might be outdated. However, the Hofstede data are still being used frequently since there is no substitute.
In accounting literature the influence of Hofstede’s cultural dimensions on financial disclosure has been investigated, especially the influence of the dimensions
individualism and uncertainty avoidance (Hope, 2003, p. 221). I will elaborate upon these two dimensions and the influence on the creation and adoption of CSR
Jaggi and Low (2000) state that in individualistic societies, individuals care for themselves, and the environment is more competitive and less secretive, suggesting a positive association with financial disclosure. The assumption that individualistic societies are less secretive, and that companies are likely to disclose more
information, might apply to sustainability information as well. In collectivistic countries there might be a tendency to be more secretive with respect to sustainability information.
Strategic legitimacy theory states that companies, which voluntarily disclose sustainability information without the existence of a CSR framework, tend to
strategically legitimise their actions. In individualistic countries it is more likely that companies disclose sustainability information, irrespective of the existence of CSR guidelines, because the overall level of transparency in these societies seems to be higher. In order to prevent companies from strategically legitimising their actions, CSR guidelines have to be developed in order to attain corporate compliance with these guidelines.
I therefore assume that:
H2a: Corporate sustainability reporting guidelines are more likely to be developed in individualistic than in collectivistic societies.
And subsequently, I assume that:
H2b: Individualism will have a more pronounced influence on the probability that mandatory guidelines are available than voluntary guidelines.
I hypothesise that:
H3: The higher the level of power distance, the less likely it is that CSR guidelines have been issued in a country.
Internationalisation
So far, the creation and adoption of CSR guidelines has been placed in institutional and cultural theory, implying that only national forces foster the diffusion of CSR guidelines. However, CSR guidelines might also be created and adopted, according to practices in other countries. Large multinational enterprises are coping with information demands by a large group of stakeholders worldwide, instead of only home-country demands (Newson, M and Deegan, C. 2002, p. 186). For these companies, this implies an extension of social contract theory from a national to a global level. In other words, large multinational companies have to legitimate their actions not only to their home-country’s stakeholders, but to a large globally dispersed group of stakeholders. It would be ignorant, not to take into account internationalisation effects for the diffusion of CSR guidelines. Aguilera and Cuervo-Cazurra (2004, p. 416-417) argue that practices are spreading faster across the globe due to globalisation effects. Therefore, I will test the following hypothesis for the diffusion of CSR guidelines:
H3 Overview of hypotheses
The following scheme (Figure 1) represents the hypotheses I will investigate in this research.
Figure 1. Determinants of corporate sustainability reporting guidelines
METHODOLOGY
Census and data collection
The census in this study comprises 65 countries, all around the world (see Table 1, appendix). The countries were selected based on the availability of data on both the cultural dimensions provided by Hofstede and the institutional dimensions proposed by La Porta, Lopez-de-Silanes, Shleifer, and Vishny (LLSV). Hofstede data are available for 52 countries in the census. LLSV data on legal regime and rule of law are available for 49 countries in the census. The census represents a broad range of countries, ranging from industrialised countries, to developing and third world countries. The extensiveness of the census is however pre-determined by the availability of explanatory data.
For all 65 countries, I investigated whether corporate sustainability guidelines have been established between 1995 and 2005. This time frame has been chosen
considering that Denmark, in 1995, was the first country to provide a framework for corporate sustainability reporting (Green Accounts Act). The main information sources have been the KPMG International survey of corporate sustainability reporting 2005 and the ACCA “Towards Transparency report 2004”. By the end of 2005, 29 countries had issued at least one CSR guidelines. Table 2 includes all countries which had issued a CSR guideline between 1995 and 2005. I did not encounter the availability of CSR guidelines in Argentina, Brazil, Chile, China, Colombia, Ecuador, Egypt, Ethiopia, Georgia, Ghana, Hong Kong, India, Indonesia, Israel, Jordan, Kenya, Macedonia, Malaysia, Mexico, Namibia, Nepal, New Zealand, Nigeria, Pakistan, Peru, Philippines, Russia, Singapore, South Korea, Sri Lanka, Switzerland, Taiwan, Thailand, Turkey, Uganda, United States, Uruguay, Venezuela and Zimbabwe.
guidelines, it might result in findings which would stimulate further research on a broader range of guidelines. Moreover, reporting on CSR performance in the form of a corporate report, instead of in a database, is beneficial for the general audience, since it increases the accessibility and the transparency of CSR information. Sustainability reporting guidelines have been issued for the corporate world, but also for governments, pension funds and non-governmental organisations.
Guidelines which aim it is to increase the transparency on sustainability matters of governments (either local or national), non-profit organisations, pension funds and issuers of financial products are not included in the sample; the database only consists of corporate sustainability reporting guidelines.
It has to be mentioned that the European Modernisation Directive, which has been adopted by all member states of the European Union, is included for each country separatel. This directive requires listed companies in all member states to report on environmental performance and employee matters from 2005 onwards. However, each member country first had to ratify the Modernisation Directive into national legislation.
Dependent variables
The aim of this research is to find which factors contribute to the creation and adoption of corporate sustainability guidelines in countries. The dependent variable is corporate sustainability guidelines, which is measured in four ways: guideline availability, mandatory guideline availability, voluntary guideline availability and number of guidelines.
One of the main limitations of this research is the supposed homogeneity (i.e. no difference in content) of the guidelines, whereas in practice guidelines can be heterogeneous. The French New Economic Regulations reporting framework, for example, is generally believed to be the most extensive of all guidelines. The nature of the guideline (i.e. voluntary or mandatory) is part of the research. In order to account for the difference in the nature of the guidelines, I have classified guidelines as; no guideline at all, voluntary and/or mandatory.
have issued no guideline at all. Guideline availability is a dichotomous variable, having the value of 1 in case a country has established a guideline and 0 if not.
Mandatory guideline availability. In order to account to some extent for the heterogeneity of the guidelines, I constructed another dichotomous variable, which takes the value of 0 in case no guideline has been established and the value 1 in case mandatory CSR guidelines are available.
Voluntary guideline availability. This dependent variable takes the value of 0 in case no guideline at all has been issued. It takes the value of 1 in case voluntary reporting guidelines do exist.
Number of guidelines. In order to account for the differences in the number of guidelines between countries, I included the dependent variable number of guidelines. This variable is a count variable, taking the value of zero in case no guidelines have been established and increases by one, for each guideline which is established. The creation and adoption of CSR guidelines is a dynamic, rather than a static process. New CSR guidelines might replace existing ones, or new CSR
guidelines might cover topics of analysis which had not been covered before. Denmark for example, was the first to legislate environmental reporting, leaving social reporting uncovered in the first place. The inclusion of this dependent variable allows for the notion that CSR guidelines evolve over time.
Independent variables
The independent variables for globalisation, institutions and culture are described below, respectively.
Economic integration. Economic integration is used as an independent variable to account for the international diffusion of corporate sustainability reporting
guidelines. In this research economic integration is the sum of imports and exports divided by the country’s GDP. This measure has been used previously in the
have taken the average of the years for which information was available. For Taiwan no data on their exports and imports were available for the period 1995-2005.
Culture. Hofstede, as mentioned in the theoretical framework of this research, distinguished five dimensions of national culture; individualism, masculinity, power distance, uncertainty avoidance and long-term orientation. Although the way in which Hofstede obtained his data has been criticised, his dataset is still one of the most frequently used datasets for the quantification of culture in accounting and other social science research (Hope, 2003). I obtained the data on the four cultural dimensions from Hofstede’s website of (http://www.geert-hofstede.com).The availability of data on long-term orientation is limited. Hofstede has quantified this construct for only 28 countries in the census. Therefore I have not included long-term orientation in this research.
Individualism – collectivism. Individualism measures the extent to which countries reinforce individual or collective achievement and interpersonal
relationships. A high score on individualism indicates that the individual is the focal point within a society (Hofstede), whereas a low score indicates that a country is established upon strong and coherent collectives.
Masculinity – femininity. This dimension stands for a preference for achievement, heroism, assertiveness and material success (Hope, 2003). A high score indicates that the country experiences a high degree of gender differentiation, and the male is the dominant figure in the society (Hofstede).
Power distance. Power distance is the extent to which people accept an unequal, hierarchical distribution of power (Hope, 2003). A high score on power distance implies that inequalities within a society are allowed to grow. In countries with low scores, opportunities are equal for everyone within the society and hardly any hierarchical constraints exist (Hofstede)
Civil-law legal family. As mentioned in the theoretical overview, two general legal families are identified; the civil law and the common law legal regimes. This variable will have the value of 1, in case a country has a civil law legal regime and 0 in case it has a common law legal regime. I have obtained national classifications from the study by LLSV (1996).
Rule of law. In order to measure the strength of a legal system, by means of the enforceability of rules, I include the independent variable rule of law. Rule of law, as an independent variable, has been used in a study conducted by La Porta, Lopez-de-Silanes, Shleifer, and Vishny (LLSV). Rule of law is measured on a continuous scale from 0 to 10, with 0 for countries with a low tradition for law and order and 10 for countries with a strong tradition of law and order and thus high levels of
enforceability of rules. LLSV obtained their data from the International Country Risk Guide. This research focuses on the period 1995-2005. However, due to monetary constraints, it is not possible to obtain the rule of law data for these years.
Therefore, I use the average rule of law data from 1982 until 1995, which were used by LLSV (1997). I assume the rule of law of a country not to have changed drastically within a decade of time and therefore believe the data by LLSV to
resemble the rule of law for 49 countries in my census, for which data on rule of law are available.
Method of analysis
The aim of this research is to find whether institutions, culture and
internationalisation have an effect on the creation and adoption of CSR guidelines. The dependent variables guideline availability, mandatory guideline availability and voluntary guideline availability are dichotomous variables. A dichotomous variable is a categorical variable that categorises data into two groups, often being coded 0 and 1. For linear regression analysis the dependent variable is required to be
it estimates whether the independent variables have a positive or negative influence on the likelihood that the independent variable takes the value of 1 (i.e. guideline available, mandatory guideline available and voluntary guideline available). For each of the three dichotomous dependent variables, I will analyse the relationship:
Guideline availability, Mandatory guidelines availability, Voluntary guideline
availability = β 0 + β 1*Civil-law legal regime + β 2*Rule of law + β 3*Power distance + β 4*Individualism + β 5*Masculinity + β 6*Uncertainty avoidance + β 7*Economic integration + constant
The count variable number of guidelines is not an interval nor a ratio variable, but a discrete variable. The number of guidelines can only be 0, 1, 2, 3 etc., but never any value in between, such as 2.3. In order to analyse the relationship between the independent variables and the count variable number of guidelines, a Poisson regression can be used (Aguilera and Cuervo-Cazurra, 2004). A Poisson regression estimates the goodness of fit of the model in which number of guidelines is the dependent variable and is influenced by the various independent variables. It also provides information on the influence of the independent variables, either positive or negative, on the probability that a certain number of guidelines have been issued. The independent variable legal family is included as an integer, in order to account for its binary nature.
For the count variable, number of guidelines, I will analyse the relationship:
Number of guidelines = β 0 + β 1*Civil-law legal regime + β 2*Rule of law + β 3*Power distance + β 4*Individualism + β 5*Masculinity + β 6*Uncertainty avoidance + β 7*Economic integration + constant
Predicted value Hypothesis Independent variable Logistic regression Poisson regression H1a:
CSR guidelines are more likely to be developed in countries with a civil-law regime than in countries with
common-law legal regimes.
Civil-law legal regime
β 1 > 1 β 1 > 0
H1b:
Mandatory corporate sustainability reporting guidelines are more likely to be developed in countries where the enforceability of rules is stronger.
Rule of law β 2 > 1 β 2 > 0
H2a:
Corporate sustainability reporting guidelines are more likely to be developed in individualistic than in collectivistic societies.
Individualism Β4 > 1 Β4 > 0
H2b:
Individualism will have a more pronounced influence on the
probability that mandatory guidelines are available than voluntary guidelines
Individualism Β4 > 1 Β4 > 0
H3:
The higher the level of power distance, the less likely it is that CSR guidelines have been issued in a country.
Power distance
Β3 < 1 Β3 < 0
H4:
Corporate sustainability reporting guidelines are more likely to be created and adopted in countries that are more integrated in the world economy than in countries which are less integrated in the global economy.
Economic integration
Limitations
This research is a first exploration in the analysis of the creation and adoption of corporate sustainability guidelines. In case this research brings up interesting results, the challenge for further research might to be to overcome the potential limitations of this research.
The supposed homogeneity of the CSR guidelines is one of the limitations of this research. Although I have tried to overcome this homogeneity in part by allowing for the differences between mandatory and voluntary guidelines, the content of the guidelines is not taken into account.
Next to the homogeneity assumption, the absence of accurate explanatory data, in some cases, is a limitation to this research.
The data on Rule of law do not represent the period of my research, but the average of a decade earlier. The cultural data presented by Hofstede are often surrounded by criticism, as mentioned, before, suggesting that the data are outdated and do not resemble national differences. Although I am fully aware of these criticisms, to date there is hardly any substitute for cultural data. The results must thus be interpreted with caution. It is often argued that corporate sustainability reporting is a cross-border practice by nature. Social and environmental performance cross borders, and therefore companies also report on issues which take place in
numerous countries all around the world. The idea of social contract theory adopted within this research, including national culture, and institutions, might therefore not cover the entire picture.
RESULTS AND ANALYSIS
Table 2 shows the countries in which CSR guidelines are available, the issuers of the guidelines and their nature, either mandatory or voluntary. The majority of CSR guidelines arise in Europe (66 out of 70). This might imply a bias in the census; since I only have taken account those CSR guidelines aiming at the reporting of CSR performance in the form of a report. It might be that European countries favour publishing information in the form of a report, instead of in a for example publicly available database.
National governments seem to have taken the lead in the process of corporate sustainability reporting, with 63 out of the 70 guidelines being issued by a
government. The results thus clearly indicate that governments are the main actors in the process of establishing guidelines for CSR. I argued that guidelines arise in a society in order to overcome legal boundaries. The finding that governments issue the majority of all guidelines causes suspicion regarding social contract theory. It might however be argued that governments are influenced by pressure groups in the field of corporate sustainability reporting. Tilt (1994) argues that pressure groups act on behalf of society, influencing corporate sustainability disclosure by lobbying either a company directly, or through government bodies (indirectly). This can in fact point to the presence of ‘structuration dynamics’, as argued by
institutional theory (Suchman, 1995). In other words, an institutional system is not static, but represents a dynamic system which evolves over time through societal pressures.
Table 3 provides the frequencies, means and standard deviations of all dependent and independent variables. The tables 4 and 5 show the correlations between the dependent and independent variables and among the independent variables, respectively. When considering the institutional dependent variables, a strong and positive correlation between rule of law and all four dependent variables can be seen, which is statistically significant on the 1% level in every case. There is a small, but positive correlation between civil-law legal family and the dependent variables, which is only significant for number of guidelines (0,27, p<0,05, Table 4). Of the cultural independent variables, individualism exhibits the strongest positive correlation to all dependent variables, which is significant on the 1% level. This might provide preliminary evidence for Hypothesis 3a, which states that
available in a country. A negative and significant relation (p<0,01) exists between all four dependent variables and Power Distance.
Power distance shows a negative and strong correlation with rule of law (-0,574, p<0,01) and an even stronger negative correlation with individualism (-0,694, p<0,01). There is a strong and positive correlation between individualism and rule of law (0,678, p<0,01). Caution has to be taken with these three variables when interpreting the results of the regressions. The results of the correlations provide evidence to some extent, that institutional settings, cultural factors and economic integration are related to the availability and number of CSR guidelines in a country. Table 6 shows the results of the logistic regression of guideline availability on
institutional, cultural and economic integration predictors. Table 7 presents the results of the Poisson regression on the number of guidelines as a dependent
variable. Model 1 shows the logistic regression for all seven explanatory variables on guideline availability. The models 2-4, include only one of the three predictors, individualism, rule of law and power distance a time, because of their strong and significant correlation (Table 3b). Tables 8 and 9 provide the results for the logistic regressions of the dependent variables mandatory guideline availability and
voluntary guideline availability, respectively.
Rule of law has a positive influence on the probability that mandatory CSR
guidelines are available, as proposed in H1b, although statistical significance lacks in those models wherein both rule of law and individualism are included. This might imply multicollinearity between the independent variables rule of law and
individualism. In case individualism and power distance are excluded, rule of law has a positive and significant influence (p<0,01) on the dependent variables, which is strongest in the case of mandatory guideline availability as the dependent
variable (Table 8, model 3, Exp(B) = 3,063, p<0,01). Hypothesis 1b is accepted, based upon the regressions and the results of the correlations.
Individualism, as hypothesised (H2a), has a positive influence on the probability that CSR guidelines are developed in a society. For each regression in which individualism is included, I found a positive and significant effect, implying that the level of individualism of a society tends to be decisive for the establishment of CSR guidelines. The positive influence of individualism is strongest, as hypothesised (H2b), for mandatory guideline availability as a dependent variable and in the case rule of law and power distance are excluded (Table 8, model 2, Exp(B) = 1,441, p< 0,01.)
Power distance, which correlates strongly negative (p<0,01) with individualism and has a negative, significant (p<0,01) effect on the dependent variables, as
hypothesised (H3), in case rule of law and individualism are excluded. According to Zarzeski, high power distance societies have fostered business practices that
discourage extensive information sharing (quoted in: Hope, 2003). The environment of secrecy is strengthened by the fact that society accepts the unequal distribution of power. Power distance and individualism, according to the correlation results, seem to influence CSR guidelines in opposite directions. High levels of individualism resemble low levels of power distance and vice versa. Both constructs implicitly characterise a country based on transparency and/or closeness. Relatively open and transparent societies thus provide better grounds for the establishment of CSR guidelines, which is in line with legitimacy theory.
Uncertainty avoidance has an unpredicted positive influence on the likelihood that CSR guidelines are available, although the relation is only significant in models without power distance and rule of law. Uncertainty avoidance measures the extent to which people in a society are uncomfortable with an uncertain future (Hofstede). Brammer and Pavelin (2004) argued that CSR guidelines reduce information
the positive, but not in all cases significant, influence of uncertainty avoidance on CSR guidelines.
Masculinity has a relatively weak negative influence on the dependent variables, which is only significant on the 10% level for mandatory guidelines. It is therefore hard to provide evidence for a significant relation between masculinity and CSR guidelines.
Economic integration (H4) seems to positively influence the availability of CSR guidelines, as expected, although the strength of the effect is relatively small compared to individualism and rule of law. The Poisson regressions however, do not show strong positive and significant effects. The positive effect of economic
integration is largest in the case of mandatory guidelines (Table 8, model 2, Exp(B) = 1,102, p<0,05).
DISCUSSION AND CONCLUSIONS
In this research I examined the creation and adoption corporate sustainability reporting guidelines in 65 countries. The main question throughout this research has been: Which factors determine the adoption and creation of Corporate Sustainability Reporting guidelines?
I have taken into account the voluntary and mandatory nature of the guidelines, and I assumed the content of the various CSG guidelines to be homogenous. Therefore, it might be interesting for future research, especially since my results, yet provide interesting results, to examine more thoroughly the differences in content of the various sustainability reporting guidelines.
The results provide insight in the development of national guidelines regarding CSR and the factors which contribute to the creation and adoption of these guidelines. The results clearly show that CSR guidelines are not yet a worldwide diffused practice. European countries seem to have taken the lead with respect to these specific organisational practices. These findings are supported by Kolk (2002), who analysed the international state of affairs in corporate sustainability reporting on a firm level in 2002. She stated that European and Japanese governments are
involved in developments regarding CSR, whereas this involvement is lacking in the case of United States and South-Korean governments.
The lack of government involvement in the case of the United States, might be explained in part because of the nature of the American legal system, which is characterised as adversarial legalism, introduced by Kagan (as quoted in Kolk, 2002). Adversarial legalism is defined as “policymaking, policy implementation, and dispute resolution by means of lawyer-dominated litigation” (Kagan, 1991). In other words, in the United States there is a relatively high reliance on litigation in the case of revolving public policy dilemmas, compared to other Western countries. Kagan states that “Western European polities care about justice, environmental regulation, and preventing professional or government malpractice. In some areas, such as workers’ rights and land use regulation, many Western European polities have “more law” that the United States” (Kagan, 1991). These observations, made by Kagan, might provide an explanation for the absence of corporate sustainability guidelines in the United States.
corporate governance, Anglo-Saxon countries, dominated by the United States and the United Kingdom, have taken the lead. I conclude that for CSR guidelines it might be the other way around, since European countries seem to dominate.
However, as mentioned before, the selection of only those CSR guidelines aiming at the inclusion of sustainability information in the form of a report might have biased the results in favour of European countries. The conclusion that Europe dominates in the process of the creation and adoption of CSR guidelines needs to be scrutinised. The results and selection of data in this research do not provide an unbiased conclusion.
The results indicate that institutional and cultural and internationalisation account, to varying extents, for the diffusion of CSR guidelines. Institutional and cultural factors seem to have a more pronounced influence than economic integration, suggesting that the availability of CSR guidelines in a country depends more on country-specific circumstances than on internationalisation effects.
The positive and significant influence of individualism provides evidence for
institutional legitimacy theory. Countries with high levels of individualism are more likely to develop CSR guidelines in order to prevent companies from only reporting on sustainability matters solely to legitimise their actions. This finding is supported by the fact that, according to an overview of corporate sustainability reporting worldwide by ACCA (Towards transparency 2004) the first companies started to report on sustainability matters yet in 1990, whereas I found that the first guideline was established five years later in Denmark (Green Accounts Act, 1995). This implies a sequence of events in which companies take the lead in order to be restricted by guidelines and so on.
The unpredicted positive influence of uncertainty avoidance might indicate that guidelines are established in order to reduce information asymmetries between stakeholders and companies.
The distinction between civil-law and common-law, based upon the assumption that civil-law countries tend to have less litigation and costs of going to court are higher provides the expected results, although not significant in every case. The
Social contract theory, extended by legitimacy and accountability theories, provides a theoretical framework for the analysis of CSR guidelines, which is supported by the results.
The overall conclusion results in the notion that CSR guidelines are not yet a worldwide diffused practice, but incrementally diffuse over time and space. The worldwide adoption and establishment of CSR guidelines, is not expected to happen rapidly, because national characteristics seem to dominate effects of
FURTHER RESEARCH
This article falls short in identifying differences in content between the various CSR guidelines. I have only taken into account the voluntary and mandatory nature of the guidelines, suggesting too much homogeneity between the guidelines.
Therefore, it might be interesting for future research, especially since my research yet provides interesting results, to examine more thoroughly the differences between the various corporate sustainability reporting guidelines.
A point of interest resulting from my research is the huge involvement of
governments in the establishment of CSR guidelines. I argued that governments are prone to influence by pressure groups and act as a societal ‘ gateway’ to establish CSR guidelines. It might be interesting to investigate the role of pressure groups in this context and the way in which governments react to societal pressures.
The differences between my research and the study on corporate governance codes by Aguilera and Cuervo-Cazurra (2004), which in part led me to conduct this
REFERENCES
ACCA, Towards transparency: progress on corporate sustainability reporting 2004,
http://www.accaglobal.com/sustainability/
Adams, C.A. 2004. The ethical, social and environmental reporting-performance portrayal gap, Accounting, Auditing & Accountability Journal, Vol. 17, No. 5.
Aguilera, R.V. and Cuervo-Cazurra, A. 2004. Codes of good governance worldwide; what is the trigger? Organization studies, 25(3)
Brammer, S. and Pavelin, S. 2004. Voluntary social disclosures by large UK companies, Business ethics: a European review, Vol. 13, Nrs. 2/3, p. 86-88.
Campbell, D. Craven, S. And Shrives, P. 2003. Voluntary social reporting in three FTSE sectors: a comment on perception and legitimacy. Accounting, Auditing & Accountability Journal, Vol. 16, Nr. 4, p. 558-561.
Globalt ansvar (Swedish partnership for global responsibility), 2003. Legal issues in corporate citizenship, p. 7-21.
Gray, R., Javad, M., Power, D.M. and Sinclair, C.D. 2001. Social and environmental disclosure and corporate characteristics, Journal of Business Finance and Accounting, 28(3) & (4).
Gray, R., Owen, D. and Adams, C., Accounting and accountability; changes and challenges in corporate social and environmental reporting, Londen: Prentice Hall (1996)
Haniffa, R.M. and Cooke, T.E. 2005. The impact of culture and governance on corporate social reporting, Journal of Accounting and Public Policy, 24, p. 398.
Hummels, H. 1998. Organizing ethics: a stakeholder debate, Journal of Business Ethics, Vol. 17 No.13, p. 1406-1408.
Jaggi, B. and Low, P.J. 2000. Impact of culture, market forces and legal system on financial disclosures, the International Journal of Accounting, Vol 35, No. 4, p. 496-501.
Jeurissen, R. 2004. Institutional conditions of corporate citizenship, Journal of Business Ethics, 53, p. 94.
Johnson, S., La Porta, R., Lopez-de-Silanes, F and Shleifer, A. 2000. Tunneling, American Economic Review, Vol. 90, No. 2, p. 496-501.
Kagan, R.A. 1991, Adversarial Legalism and the American Government, Journal of public policy analysis and management, Vol. 10, No. 3, p. 369-372.
Kolk, A. 2002. Maatschappelijke verslaggeving: de internationale situatie, Maandblad voor Accountancy en Bedrijfseconomie, p.510-511
KMPG. 2005. International survey of corporate sustainability reporting.
van der Laan Smith, J., Adhikari ,A and Tondkar, R.H. 2005. Exploring differences in social disclosures internationally: A stakeholder perspective, Journal of
Accounting and Public Policy, 24.
La Porta, R. Lopez-de-Silanes, F, Shleifer, A. And Vishny, R.W. 1997. Legal determinants of external finance, the Journal of Finance, Vol. 52, No. 3.
Lamberton, G. 2005. Sustainability accounting – a brief history and conceptual framework, Accounting forum, 29
Licht, A. N., Goldschmidt, C. and Schwartz, S. 2005. Culture, law and corporate governance, International Review of Law and Economics, 25.
Mathews, M.R. 1987. The Implementation of the Interpretive Paradigm: Philosophical Bases Underlying Social Accounting, Accounting Forum, Vol. 10 No. 3
Newman, K.L. and Nollen, S.D. 1996. Culture and congruence; the fit between management practices and national culture, Journal of International Business Studies,
Newson, M. and Deegan, C. 2002. Global expectations and their association with corporate social disclosure practices in Australia, Singapore and South Korea, the International Journal of Accounting, 37.
North, Douglas C. 1991. Institutions, the Journal of Economic Perspectives, Vol 5, No. 1, , p. 97.
O’Donovan, G. 2002. Environmental disclosures in the annual report, Accounting, Auditing & Accountability Journal, Vol. 15 No. 3.
Rikhardsson, P.M., 1996, Developments in Danish environmental reporting, Business Strategy and Environment, Vol. 5, 262-272
Suchman, M.C. 1995. Managing legitimacy: Strategic and institutional approaches, Academy of management review, Vol. 20, No. 3.
Tilt, C.A. 1994. The influence of external pressure groups on corporate social disclosure, Accounting, Auditing & Accountability Journal, Vol. 7 No. 4, p. 49-59.
Williamson, O.E. 2000. The new institutional economics; tacking stock, looking ahead, Journal of Economic Literature, Vol. 38, No. 3.
Reporting, A test of legitimacy theory, Accounting, Auditing & Accountability Journal, Vol. 13 No. 1, pp. 11-12
Internet resources:
BBC News
http://news.bbc.co.uk/1/hi/sci/tech/218527.stm
Hofstede, G.; Cultural dimensions resources:
http://www.geert-hofstede.com/geert_hofstede_resources.shtml
Non-Financial Reporting and Corporate Social Responsibility (NFRCSR) Resource Centre
http://www.nfrcsr.org/
Organisation for Economic Cooperation and Development
http://www.oecd.org
San Francisco State University
APPENDICES
Table 1, 65 countries included in census in alphabetical order
Argentina France Mexico South Korea
Australia Georgia Namibia Spain
Austria Germany Nepal Sri Lanka
Belgium Ghana Netherlands Sweden
Brazil Greece New Zealand Switzerland
Canada Hong Kong Nigeria Taiwan
Chile Hungary Norway Thailand
China India Pakistan Turkey
Colombia Indonesia Peru Uganda
Cyprus Ireland Philippines United Kingdom
Czech Republic Israel Poland United States
Denmark Italy Portugal Uruguay
Ecuador Japan Russia Venezuela
Egypt Jordan Singapore Zimbabwe
Estonia Kenya Slovakia
Table 2. Number of CSR guidelines, issuing actors and nature in those countries in which CSR guidelines have been issued.
Table 3. Descriptive statistics
N Mean Standard deviation 1. Number of guidelines 65 1,08 1,47 2. Guideline availability 65 0,42 0,497 3. Mandatory guideline availability 65 0,37 0,486 4. Voluntary guideline availability 65 0,40 0,494 5. Civil-law legal family 65 0,52 0,503
6. Rule of law 49 6,85 2,63 7. Power distance 52 56,0 20,97 8. Individualism 52 45,94 25,17 9. Masculinity 52 51,12 18,33 10. Uncertainty avoidance 52 63,98 23,04 11. Economic integration 65 75,16 47,18
Table 4. Correlations between dependent and independent variables Number of
guidelines Guideline available Mandatory available Voluntary available
1. Civil-law legal family 0,27* 0,117 0,191 0,117
2. Rule of law 0,638** 0,619** 0,646** 0,619** 3. Power distance -0,492** -0,436** -0,520** -0,436** 4. Individualism 0,665** 0,654** 0,669** 0,654** 5. Masculinity -0,276* -0,027 -0,147 -0,027 6. Uncertainty avoidance -0,036 0,050 0,077 0,050 7. Economic integration 0,198 0,278* 0,227 0,278*
Table 5. Correlations among independent variables
1. 2. 3. 4. 5. 6.
1. Civil-law legal family 1
2. Rule of law 0,112 1 3. Power distance -0,008 -0,574** 1 4. Individualism -0,095 0,678** -0,694** 1 5. Masculinity -0,196 -0,12 0,115 0,042 1 6. Uncertainty avoidance 0,490** -0,233 0,082 -0,15 0,144 1 7. Economic integration -0,109 0,152 -0,014 0,086 -0,059 -0,205
Note numbers in bold are significant, ** and * represent statistical significance at 1 and 5 percent, respectively
Table 6. Logistic regressions of institutions, culture and economic integration on guideline availability
Model 1 Model 2 Model 3 Model 4 Predicted Exp(B) Exp (B) Exp (B) Exp (B) 1. Civil-law legal family + 4,958 4,873 1,089 1,911
2. Rule of law + 1,303 2,203*** 3. Power distance - 1,040 0,946*** 4. Individualism + 1,141** 1,127*** 5. Masculinity ? 0,986 0,989 0,995 1,007 6. Uncertainty avoidance ? 1,063 1,061* 1,022 1,014 7. Economic integration + 1,030* 1,033** 1,007 1,015* 8. Constant 0,00** 0,00*** 0,00** 2,235 N 45 51 45 51 Chi-square 30,532*** 35,694*** 22,198*** 15,948
Predicted percentage correct 84,4% 82,4% 77,8% 74,5%
Note: Exp(B) or eB, refers to the coefficient of the independent variable An Exp(B) > 1, has a positive
influence, whereas an Exp(B) < 1, has a negative influence
Table 7. Poisson regressions of institutions, culture and economic integration on number of guidelines
Model 1 Model 2 Model 3 Model 4
Predicted Coefficient (t-value) Coefficient (t-value) Coefficient (t-value) Coefficient (t-value) 1. Civil-law legal family + 0,876*
(1,900) 0,917** (2,341) 0,535 (1,267) 0,717** (2,118) 2. Rule of law + 0,159 (1,108) 0,479*** (4,871) 3. Power distance - -0,0058 (-0.504) -0,029*** (-4,350) 4. Individualism + 0,034** (2,684) 0,047*** (5,771) 5. Masculinity ? -0,007 (-0,920) -0,008 (-1,094) -0,007 (-1,022) -0,008 (-1,224) 6. Uncertainty avoidance ? 0,008 (0,796) 0,007 (0,939) 0,004 (0,518) 0,005 (0,674) 7. Economic integration + 0,003 (0,858) 0,004 (1,166) 0,002 (0,485) 0,006** (2,148) 8. Constant -3.847*** (-3,189) -3.540*** (-4.171) -4,169*** (-3,827) 0,7304 (1,222) Degrees of freedom 37 45 39 45 N 45 51 45 51 R2 estimation 0,621 0,608 0,533 0,342
Note: for each independent variable coefficients are given with their respective t-values in parenthesis. R-squared estimation is measured as: 1 – (residual deviance / null deviance)
Table 8. Logistic regressions of institutions, culture and economic integration on mandatory guideline availability
Model 1 Model 2 Model 3 Model 4 Predicted Exp(B) Exp (B) Exp (B) Exp (B) 1. Civil-law legal family + 59,681 89,841** 1,473 1,997
2. Rule of law + 1,361 3,063*** 3. Power distance - 1,017 0,922*** 4. Individualism + 1,404** 1,441** 5. Masculinity ? 0,914* 0,913* 0,967 0,989 6. Uncertainty avoidance ? 1,233* 1,244** 1,035 1,030 7. Economic integration + 1,096* 1,102** 1,006 1,014* 8. Constant 0,00* 0,00** 0,00** 5,898 N 45 51 45 51 Chi-square 42,866*** 50,306*** 29,527*** 22,146***
Predicted percentage correct 95,6% 94,1% 80,0% 78,4%
Note: Exp(B) or eB, refers to the coefficient of the independent variable An Exp(B) > 1, has a positive
influence, whereas an Exp(B) < 1, has a negative influence
Numbers in bold are significant and ***, ** and * represent statistical significance at 1, 5 and 10 percent, respectively
Table 9. Logistic regressions of institutions, culture and economic integration on voluntary guideline availability
Model 1 Model 2 Model 3 Model 4 Predicted Exp(B) Exp (B) Exp (B) Exp (B) 1. Civil-law legal family + 9,032 7,063 2,078 2,559
2. Rule of law + 1,186 1,981*** 3. Power distance - 1,029 0,949*** 4. Individualism + 1,125* 1,118*** 5. Masculinity ? 0,992 0,993 0,998 1,008 6. Uncertainty avoidance ? 1,053 1,058* 1,015 1,015 7. Economic integration + 1,030* 1,032** 1,007 1,015* 8. Constant 0,00** 0,00*** 0,001** 0,657 N 45 51 45 51 Chi-square 27,097*** 33,183*** 19,517*** 15,476***
Predicted percentage correct 84,4% 82,4% 77,8% 74,5%
Note: Exp(B) or eB, refers to the coefficient of the independent variable An Exp(B) > 1, has a positive
influence, whereas an Exp(B) < 1, has a negative influence
