How can CSR identity be evaluated? A pilot study using a Fuzzy Expert System

Journal of Cleaner Production 141 (2017) 1000e1010

Contents lists available at ScienceDirect

Journal of Cleaner Production journal homepage: www.elsevier.com/locate/jclepro

How can CSR identity be evaluated? A pilot study using a Fuzzy Expert System Andrea Venturelli a, *, Fabio Caputo a, Rossella Leopizzi a, Giovanni Mastroleo a, Chiara Mio b a b

Department of Management, Economics, Mathematics and Statistics, University of Salento, Lecce, Italy Department of Management, Ca’ Foscari University of Venice, Venice, Italy

a r t i c l e i n f o

a b s t r a c t

Article history: Received 17 February 2016 Received in revised form 20 September 2016 Accepted 21 September 2016 Available online 22 September 2016

The aim of this study is to propose a method to evaluate the Corporate Social Responsibility (CSR) identity of a firm. Using this method, based on a fuzzy expert system (FES), it is possible to generate a comprehensive rating for the assessment of the sustainability of a firm. Up to now, measurement has been hampered by a lack of clarity in theoretical frameworks and empirical methods for the Corporate Social Responsibility construct. The algorithm of the Fuzzy Expert System aggregates multicriteria evaluations of a problem. The assessments of behavior and the resulting decisions are represented in blocks of rules, drawn up by an inference engine in fuzzy logic. The Fuzzy Expert System unites the ability of an expert system to simulate the decision-making process with the uncertainty typical of human reasoning, present in fuzzy logic. Despite the spread of Corporate Social Responsibility practices among firms, there is not a commonly accepted method of measuring sustainability. Moreover, although Environmental social governance (ESG) rating agencies provide Corporate Social Responsibility ratings, their methods have certain weaknesses. Considering the growing importance of socially responsible financial markets, this topic could be of vital importance for decision-makers in the management of their investments, by remedying deficiencies in methods used by sustainability rating organizations. The outcome of the application is a system designed to measure the CSR identity of a firm. On the management side, the possibility to identify the determinants of the different Corporate Social Responsibility intermediate indicators making up the final Corporate Social Responsibility index would allow CSR-compliant managers to use this information for decision-making purposes. © 2016 Elsevier Ltd. All rights reserved.

Keywords: CSR identity Fuzzy Expert System CSR assessment CSR score Performance measurement

1. Introduction Nowadays it is increasingly important for companies to implement Corporate Social Responsibility (CSR) practices. At the same time, the significance given and the attention paid by financial markets to socially responsible investments e that is, in CSRcompliant firms e is growing, too. Investors are demanding ever more accurate information. Despite the spread of Corporate Social

* Corresponding author. E-mail addresses: [email protected] (A. Venturelli), fabio.caputo@ unisalento.it (F. Caputo), [email protected] (R. Leopizzi), giovanni. [email protected] (G. Mastroleo), [email protected] (C. Mio). http://dx.doi.org/10.1016/j.jclepro.2016.09.172 0959-6526/© 2016 Elsevier Ltd. All rights reserved.

Responsibility practices among firms, however, there is no commonly accepted method of measuring sustainability. Moreover, although Environmental Social Governance (ESG) rating agencies provide Corporate Social Responsibility evaluations (Chelli and Gendron, 2013; Escrig et al., 2010; Finch, 2004), their methods have certain weaknesses. Sometimes, higher scores for one domain may conceal very low scores in another domain (Escrig-Olmedo et al., 2014). Indeed, a major criticism of these rating agencies is the lack of transparency in their methods (Stubbs and Rogers, 2013). The aim of this study is to propose a method to measure the CSR identity (Otubanjo, 2013) of a firm. Using this method, based on a fuzzy logic expert system, it is possible to generate a comprehensive rating for the assessment of the sustainability of a firm. In fact,

A. Venturelli et al. / Journal of Cleaner Production 141 (2017) 1000e1010

to ensure that a firm is CSR-compliant, it is important to express Corporate Social Responsibility principles in terms of measurable variables. Up to now, measurement has been hampered by a lack of clarity in theoretical frameworks and empirical methods for the Corporate Social Responsibility construct. This paper combines both qualitative and quantitative methods. Qualitative data is collected from existing literature, Corporate Social Responsibility rating agency indicators and the responses to a semi-structured interview questionnaire by the Corporate Social Responsibility managers of a sample of selected listed companies. The qualitative data, interpreted through the researchers' theoretical lens, becomes the input into a model formalized through a fuzzy expert system (FES), which aims to evaluate the CSR identity of the selected firms. A Fuzzy Expert System has been chosen as it can merge the ability of an expert system to simulate the decisionmaking process with the vagueness typical of human reasoning, while still providing a numerical value as a response (Zadeh, 1965). Its features allow us to formalize the decision-making process related to the evaluation of the CSR identity by handling qualitative and quantitative variables and exploring the cognitive mechanisms underlying this process. The paper contributes to the existing literature on Corporate Social Responsibility research by employing a fuzzy logic method that has rarely been used in this field (Escrig-Olmedo et al., 2014). It combines the intuition and experience of experts (management view) with the formal rigor of a logic system (measurement view). The main findings of the paper have implications from both a theoretical and an empirical point of view. As regards theory, this study contributes to the broadening of the research community's understanding of the measurement of CSR identity. On the management side, on the other hand, being able to identify the factors influencing the different Corporate Social Responsibility intermediate indicators that make up the final Corporate Social Responsibility index provides CSR-compliant managers with more information to use for decision-making purposes. Considering the growing importance of socially responsible financial markets, this topic may prove to be of vital importance, above all for decision-makers in the management of their investments, by remedying the deficiencies in methods used by sustainability rating organizations. We also think that the fuzzy logic method is an appropriate way of assessing sustainability because of the “fuzzy” nature of sustainability (Chan et al., 2003; ~ oz, Phillis and Andriantiatsaholiniaina, 2001; Rivera and Mun 2009). The remainder of this paper is organized as follows: Section 2 briefly discusses the literature informing the method and thus provides a context within which the research framework is developed; Section 3 illustrates the research framework lying behind the choice of the Fuzzy Expert System method; Section 4 provides a detailed illustration of the Fuzzy Expert System model developed specifically for the research; Section 5 analyzes the main findings of the paper; and Section 6 presents the main conclusions. 2. Literature review Scholars have proposed a large number of theories regarding Corporate Social Responsibilities, for which the reader is referred to , 2004). a systematic review of the literature (Garriga and Mele Moreover, the same concept of Corporate Social Responsibility has taken on very different meanings at different times and in different places. Time and place are therefore key factors when it comes to defining the parameters of Social Responsibility: in fact, the influence of different legal, political and social systems may be considerable (Mio and Venturelli, 2013; Pava, 2008). In this paper it is not possible to provide a single definition of Corporate Social

1001

Responsibility; some authors (Matten and Moon, 2008) have highlighted the heterogeneous nature of the various definitions found in literature. The numerous definitions of Corporate Social Responsibility that have emerged are well summarized in a work by Dahlsrud (2008), to which the reader is referred. In particular, in this paper, we talk about the “CSR identity” of a firm, or in other words, the cohesive union between the Corporate Social Responsibility and the Corporate Identity of a firm (Otubanjo, 2013). CSR identity means, first of all, “what the firm does for stakeholders”, or in other words the “responsibilities of the firm in relation to its action and duties towards employees, customers, suppliers, media, investors, government and local community” (Otubanjo, 2013). It therefore involves economic, legal, ethical and philanthropic responsibilities. Firms have adopted Corporate Social Responsibility practices in different ways and for different reasons (Fray, 2007): compliance with environmental laws and regulations, health and safety of employees as well as of users of products and services, education, research and training, protection of indigenous people and culture, habitat protection, and adoption of environmentally friendly technologies (Kouikoglou and Phillis, 2011). Against this background, it is difficult to evaluate whether a firm is CSR-compliant or not, and it is also difficult to compare the Corporate Social Responsibility compliance of one firm with another's (Krajnc and Glavi c, 2005). Despite the spread of Corporate Social Responsibility practices among firms and of analyst organizations (Corporate Social Responsibility rating agencies, Environmental Social Governance information provider agencies or sustainability indices), there is no commonly accepted method of measuring sustainability (Pena, 1977; Rebai et al., 2016; Salvati and Zitti, 2009; Zarzosa, 1996). This deficiency is probably due to the fact that it is difficult to express CSR identity in a synthetic index, as to create such an index requires the selection and standardization of the variables, the building of the indicators, and the choice of a method for aggregation and weighting (Saveanu et al., 2014). Several papers dealing with the issue of Corporate Social Responsibility measurement have aimed to produce a Corporate Social Responsibility index (Atkinson, 2000; Chatterji and Levine, 2006; Cherchye and Kuosmanen, 2006), but a major limit of the methods used (Jiang et al., 2016; Martinez et al., 2005; Munda et al., 1994) has been the possible offsetting effect of scores (Delmas and Doctori Blass, 2010; Windolph, 2011), when, for example, higher scores in one domain may hide very low scores in another domain (Escrig-Olmedo et al., 2014; Zhao et al., 2016). This limit could be overcome by using a fuzzy logic expert system model. The potential of this method for generating synthetic measurement of sustainability has already been described (Calabrese et al., 2013; Phillis and Davis, 2009; Rajak and Vinodh, 2015). Up to now, very few studies have used a Fuzzy Expert System model to generate a Corporate Social Responsibility index. EscrigOlmedo et al. (2014), for example, have applied fuzzy logic to the Corporate Social Responsibility information supplied by Accountability Rating for 2008. The major limit of that work, however, is the fact that the developed sustainability rating used inputs previously defined by AR2008. Against this background, this paper overcomes the above-mentioned limit by proposing a model that combines variables already discussed in existing literature with variables used by rating agencies and variables indicated by Corporate Social Responsibility managers of a set of selected listed companies. The following section, to which the reader is referred, describes the process of selection of the different input variables. Our starting points are the scorecard measurement approach and detail from the advanced scorecard methods that look for quantitative and qualitative indicators able to measure CSR identity, but the Corporate Social Responsibility indicators are integrated in a Fuzzy Expert System model that produces a Corporate Social

1002

A. Venturelli et al. / Journal of Cleaner Production 141 (2017) 1000e1010

Responsibility index from the aggregation of three main components: human capital (HumanCap), strategy (Strategy) and reporting (ESGrep). This model discards the traditional triple bottom line approach to sustainability (economic, social and environmental components) (Elkington, 1997) used by all existing assessment systems and indices (Graafland et al., 2004; Kouikoglou and Phillis, 2011) and by the majority of previous works (Costa and Menichini, 2013; Govindan et al., 2013; Shengtian et al., 2010). In contrast, it is based on the idea (Porter and Kramer, 2006) that sustainability should be fully integrated into the strategic business model (Engert et al., 2016; Mio et al., 2015). According to this approach, which involves a synthesis of variables discussed in existing literature, variables used by rating agencies and variables indicated by Corporate Social Responsibility managers of a selected set of companies, Corporate Social Responsibility performance areas are:  Strategy e for example, risks and crisis management, customer relationship management, supply chain management;  Human Capital, which includes ownership, governance and human resources;  Reporting, namely environmental, social, governance and, finally, integrated reporting. Each of these areas consists of the traditional economic, social and environmental dimensions, but this approach is based on a transversal vision and application of Corporate Social Responsibility in a company, thanks to which Corporate Social Responsibility is integrated into all the business processes, from planning to reporting. The originality of this approach comes, first of all, from the fact that the reporting is considered to be an integrated process more than a simple output. Moreover, the reporting is strictly linked to strategy, since it measures the long term impact of the business model on human capital and on the other types of capital, expressions of value creation. This approach is in line with what emerges from the reading of the International Integrated Report Framework, in which strategy, capitals and reporting are fundamental elements of the business model. In fact, this work is collocated in the series of studies on instrumental theories; specifically, the approach adopted is “strategies for achieving competitive advantages” (Husted and Hallen, 2000), whereby the integration of Corporate Social Responsibility into systems of company governance and management becomes a crucial factor in the creation of value in the long and medium term (Crane et al., 2014; Graafland, 2016; Porter and Kramer, 2006). Our main research hypothesis in the paper is that a Fuzzy Expert System model is able to combine measurement (the Fuzzy Expert System has the formal rigor of a logic system) and management perspectives (the Fuzzy Expert System combines the intuition and experience of the experts) by developing a Corporate Social Responsibility index. The Fuzzy Expert System technique is able to provide a ‘numerical’ measurement of CSR identity while still taking into consideration the fuzzy nature of Corporate Social Responsibility (Kouikoglou and Phillis, 2011; Phillis and Davis, 2009; ~ oz, 2009). The following sections illustrate the Rivera and Mun research framework (Section 3) and the features of the Fuzzy Expert System model specifically developed for this paper (Section 4). 3. Research framework and method This section aims to illustrate the research framework and method employed in pursuit of the research objective. The paper

makes complementary use of qualitative (multiple case studies and semi-structured interviews) and quantitative methods (Fuzzy Expert System model). Whereas the previous section analyzes the theoretical perspectives that drove our research, this section focuses on the methods chosen to address our research hypothesis. The research uses both qualitative and quantitative methods. First of all, we describe the choice of the different variables as inputs for the Fuzzy Expert System model. We have combined variables already discussed in existing literature with variables used by rating agencies and variables suggested by Corporate Social Responsibility managers of a set of selected listed companies and interpreted by the researchers using an interpretivist approach.1 The CSR managers chosen belong to Italian companies considered trendsetters in the field of Corporate Social Responsibility. In particular, they were involved in the engagement of the National Institute of Accountants (CNDCEC) in the International Integrated Reporting Council (IIRC) Pilot Program. Seven Italian companies participated in the project. On the basis of the qualitative data, the researchers acted as experts (content experts) and, together with an expert on Fuzzy Expert System models (a method expert), worked on developing an ad hoc Fuzzy Expert System quantitative model. Most works in existing literature that deal with the assessment of Corporate Social Responsibility have made use of indicators related to environmental, social and economic performance, in particular, those provided by Global Reporting Initiative (Calabrese et al., 2016; Kouikoglou and Phillis, 2011; Widiarto Sutantoputra, 2009). Some authors (Kouikoglou and Phillis, 2011) have selected specific environmental indicators, such as air quality and climate protection, land integrity and water availability, together with social aspects, economic welfare, safety at work and employee training. Others (Munoz et al., 2008) have chosen, together with an economic performance index, indicators regarding the sustainability report and stakeholder orientation. Another work (Salmi Mohd and Reast, 2012) proposed a multidimensional construct based on eight dimensions: process, policy, values, environment, personnel, profit, people and politics. Specifically, process is about long-term activity towards stakeholders; policy is about compliance with regulations; values is about the firm's reputation and identity; environment is about activity to protect the environment; personnel is about the relationship with employees; profit is about Corporate Social Responsibility investments and return; people is about all stakeholders; politics is about the relationship with institutions. Girerdpotin et al. (2014) have identified a model based on three main independent dimensions related to: business stakeholders (employees, customers and suppliers), societal stakeholders (environment and society), financial stakeholders (stockholders and debt holders). The major Corporate Social Responsibility categories, as perceived by the managers of 102 United States companies (Gee and Norton, 2012), are: community, customers, ethics, employees, legal issues, society and philanthropy. The managers interviewed reported the importance of Corporate Social Responsibility integration with the strategy of the firm. To achieve this, they suggested a Corporate Social Responsibility audit to evaluate how effectively the organization achieves its Corporate Social Responsibility goals and implements its activities. Remisova and Baciova (2012) provided a set of indicators for measuring Corporate Social Responsibility towards one selected stakeholder: employees. In the matter of the variables chosen by rating agencies (Kinderyte, 2008;

1 According to the theory of interpretivism, sociological phenomena are not to be simply observed but must also be interpreted by the researcher. This means that there is no one absolute reality, but rather that different possibilities are generated by the perspective adopted to interpret the facts (Crotty, 1998; Ryan et al., 2002).

A. Venturelli et al. / Journal of Cleaner Production 141 (2017) 1000e1010

 pez et al., 2007; Scha €fer, 2005), there have been several interLo national indices and assessment systems such as the Dow Jones Sustainability Index, the FTSE4GOOD Index and Standard Ethics. As pointed out in the previous section, the majority of the existing kely and international assessment systems (Singh et al., 2009; Sze Knirsch, 2005; Timperley, 2008) and indices adopt the triple bottom line theory. The Dow Jones Sustainability Index was the first sustainability index in the world. It was created by Dow Jones, STOXX and SAM in 1999. The three dimensions based on the triple bottom line theory (economic, social and environment) comprise sub-criteria:  economic dimension: corporate governance, risks and crisis management, compliance/bribery and industry specific criteria, supply chain management;  environmental dimension: environmental performance, environmental reporting and other industry-specific criteria;  social dimension: social reporting, labor practice indicators and human rights, human capital development, talent attraction and retention, occupational health and safety, stakeholder engagement, and industry specific criteria. The FTSE4Good index, initiated by the London Stock Exchange and Financial Times in 2001, promotes ethical investments covering 100 businesses across England, Europe and America. The criteria developed for the index are categorized into policy, management and reporting. The sub-criteria are: the protection of human rights, supply chain and employment standards, and anti-bribery measures. Standard Ethics Ratings are based on the institutional guidelines of the EU, OECD and UN. The more than 200 indicators are classified into three macro areas (competition, ownership and management). Against this background, we propose a model that, as discussed above, combines variables in existing literature with variables used by rating agencies and variables suggested by the Corporate Social Responsibility managers of a set of selected listed companies. Some of the knowledge needed to design and build the system components (variables and elements for their evaluation, blocks of rules and weights for aggregation) is ‘pulled’ by the experts using various investigation techniques. For the purposes of this paper we have used a Focus Group with partially structured discussions. Specifically, we employed an NGT (Nominal Group Technique) approach as our meeting procedure. Nominal Group Technique involves a group of experts holding discussions chaired by a moderator (Duggan and Thachenkary, 2004; Dunham, 2006), with the aim of reducing to a minimum distortions due to personal interaction during the discussion. The Fuzzy Expert System model relies heavily on the knowledge and competence of the expert focus group. Although this method involves a high level of subjectivity and is therefore subject to measurement error, it is worth remembering that it addresses some of the inherent limitations of the inferential statistical methods which are based on the direct observation of phenomena (Bertin, 2005; Krueger and Casey, 2009; Morgan, 1993) such as the need to produce information quickly, to create a shared set of terms and concepts for the different parties involved in a communication process and to involve all these parties in the evaluation process. As with all methods, the expert focus group approach has its weaknesses, including factors such as individual stereotypes, which lead to the creation of a cognitive filter in the analysis (even by experts) of social phenomena; the relational dynamics, of a psychological and social nature, which affect communication between the different social protagonists; the selection processes and the representativeness of the group of experts referred to; the definition of the level of reliability of the analyses carried out. However,

1003

we decided to employ the expert focus group method because of its usefulness in research, such as ours, where success is strictly determined by the way in which the inputs are fuzzified and the blocks of fuzzy rules are constructed; only researchers who have an in-depth knowledge of a particular phenomenon can assign reliable values and construct reliable rule blocks. To overcome any possible measurement errors, we applied several corrective measures to counterbalance the use of the expert focus group. Corporate Social Responsibility managers were interviewed twice, in order to better focus on some of the insights they offered. Interviews were recorded and transcribed and, where necessary, a further shorter interview was scheduled in order to clarify or obtain more details about certain aspects of the previous interview. The interview transcripts were checked by interviewees for potential inconsistencies or mistakes. We also collected internal documents in order to support interviewees' statements and to make the analysis more detailed. In addition, inter-rater reliability was assessed: this involved data being independently coded and the coding compared for agreements (Armstrong et al., 1997). We also established a rule to be applied in the event of a disagreement (one main referent chosen in the group); and finally, we corroborated our model by testing its reliability against a series of real cases. The multiple case studies method was chosen e in preference to the single case study approach e as the most suitable for testing the reliability of the model adequately. In particular, the model was tested on three listed companies particularly involved in Corporate Social Responsibility issues. The companies selected operate in various business sectors which are catalogued with the method previously used by the London Stock Exchange, the Industry Classification Benchmark (ICB) developed by Dow Jones and FTSE. Table 1 summarizes the company profiles for the case studies chosen. The qualitative data was collected through semi-structured interviews, as outlined above, with the Corporate Social Responsibility managers of the three companies chosen. 4. The fuzzy logic method The analysis model is formalized through a Fuzzy Expert System. The algorithm of the Fuzzy Expert System aggregates multicriteria evaluations of a problem. The assessments of behavior and the resulting decisions are represented in blocks of rules, drawn up by an inference engine in fuzzy logic. The Fuzzy Expert System combines the ability of an expert system to simulate the decisionmaking process with the uncertainty typical of human reasoning, which is present in fuzzy logic (Facchinetti et al., 2001; Magni et al., 2006; Marchi et al., 2014). The model was used because it offers a number of advantages: firstly, it improves the description of the benefits of being CSR-compliant and increases the ease of understanding and implementation of the system, but it also provides a numerical value as a response, although not all the data are quantitative; in addition, the model makes it possible to manage a large number of inputs, chosen, as we have already mentioned, from among some of those used by rating agencies and some proposed in existing literature (Marquez and Fombrun, 2005; Morimoto et al., 2005; Shengtian et al., 2010), and through the use of intermediate variables provides a clear representation of the Table 1 Company profiles. Company

Industry

Employees

Total sales (mil.V)

A B C

industrials consumer services financials

14,828 54,408 78,333

5083 3,93 82,282

1004

A. Venturelli et al. / Journal of Cleaner Production 141 (2017) 1000e1010

 analysis of the final output.

Fig. 1. Final blocks of CSR identity.

assessment structure, making the vision of the entire system. The design and formalization of evaluations requires the intervention of an expert group, whose knowledge of the problem is crucial, especially in the construction of the blocks of rules. The proposed method is able to satisfy all these requirements: to formalize and automate the decision-making process concerning the evaluation of the CSR identity, and to manage the simultaneous use of qualitative and quantitative variables and analysis of mechanisms of the cognitive process, reducing the distortions of the real decision-making context. The implementation of the system has been divided into the following phases:  initial meeting with experts to define the inputs and conditions for the aggregation of intermediate variables;  layout of the evaluation model;  definition of the range of variables and their relative weights for the aggregation;  establishment and checking of the blocks of rules;  trial processing and optimization;  comparison with the cases of certain base evaluation;

The design of a Fuzzy Expert System is the first and most important step of the study. The main structure of the model is based on the aggregation of three components: human capital (HumanCap), strategy (Strategy) and reporting (ESGrep: Environmental, Social and Governance reporting). This model discards the traditional triple bottom line approach to sustainability (economic, social and environmental components), used by all existing assessment systems and indices in favour of an approach in which sustainability is fully integrated into the strategic business model. For this reason, these three dimensions have been chosen (see Fig. 1). The structure of the model is modular: the assessment is in stages which can be broken down as far as necessary. From the final output (CSR), through variables and blocks of intermediate rules (i.e. ESGrep), it is possible to trace back the evaluation process, noting the intermediate results, to the input variables. 4.1. Focus on three dimensions Once the inputs and intermediate variables had been identified, the layout of the three dimensions was established with the help of experts. For ease of reading it is recommended that the inputs be envisaged in the position to the left of each block (intermediate variable); the reader should then proceed through the subsequent aggregations to the last output intermediate variable (HumanCap, Strategy, ESGrep). The layouts of the three models are shown below (see Figs. 2e4).

Fig. 2. Human capital dimension.

A. Venturelli et al. / Journal of Cleaner Production 141 (2017) 1000e1010

1005

Fig. 3. Strategy dimension.

Fig. 4. Environmental, Social and Governance reporting dimension.

Fig. 5. ESG reporting - example of intermediate variable.

The aggregation of the inputs was carried out taking account of the relative weightings suggested by the experts for each individual block that produces an intermediate variable; for example, to obtain the intermediate variable “Quality of Report” in the Environmental Social Governance reporting dimension, the suggestion was to aggregate the intermediate variable “Stakeholder Engagement” with the other two inputs in accordance with the relative weightings in the subsequent block (see Figs. 5 and 6).

The rules obtained by respecting this weighting for the IF part in the “Quality of Report” block were subsequently checked by the experts for possible discrepancies in the consequent (THEN) part, resulting in the block in the table below. The block is made up of twenty rules: the five possible choices for the StkEngag combinate variable multiplied by two and by two again for the dummy variables MaterialMatrix and CostSav_AR. This construction was applied to all the aggregations differing from those with an equal

1006

A. Venturelli et al. / Journal of Cleaner Production 141 (2017) 1000e1010

Fig. 6. Rule block for the intermediate variable “Quality of report”.

weighting distribution. The next section analyzes the results of our Fuzzy Expert System model. 5. Results and discussion The output from the system created makes it possible to classify the values for the CSR identity of a firm. In this work, the model created (see Table 2) was applied, using a multiple case studies approach, to three companies, as described in section 3. The final results produced by the system are displayed in Table 3. The main result, measuring the CSR identity of the three companies chosen is the figure 92.86 out of 100 given for the first

company, 59.79 for the second and 79.58 for the third. These are positive results, above all for the first company, even when taking into account the fact that these three companies were chosen from among those particularly involved in Corporate Social Responsibility issues. The final figure is the result of the aggregation of the values for Human Capital (80 for the first, 60.64 for the second and 60 for the third), ESG reporting (100 for the first, 58.33 for the second and 83.33 for the third) and Strategy (75.19 for the first, 43.17 for the second and 58.45 for the third) systems. Examination of the values ascribable to these variables reveals the following (Tables 4 and 5). The Environmental Social Governance reporting variable derives from the aggregation of the three intermediate variables: (Quality of Report, Report Evaluation and Stakeholder

Table 2 Project description.

Input variables Output variables Intermediate variables Rule blocks Rules Membership functions

Human capital

ESG reporting

Strategy

28 1 12 13 1663 161

9 1 3 4 452 60

33 1 11 12 1922 176

Table 3 Final results. CSR: Input and output description

Variable

A

B

C

Human capital Environmental social governance reporting Strategy Final evaluation

HumCap ESGrep Strategy CSR

80,00 100,00 75,19 92,86

60,64 58,33 43,17 59,79

60,00 83,33 58,45 79,58

A. Venturelli et al. / Journal of Cleaner Production 141 (2017) 1000e1010

1007

Table 4 Human capital input. Human capital: Input description

Variable

% Increment of average hours of training per year % Of non-executive directors required to buy shares % Of directors with 4 or fewer other mandates Formalized skill mapping and developing process for employees Equal remuneration female/male % Females in management positions % Females on total workforce % Grievance resolution Health and safety programs % Of Independent directors on the board % Injury rate Number of meetings attended in percentage Number of committees % Of non-executive directors on the board % Occupational decease rate Number of other mandates of the board directors Performance assessment of board of directors Respect for the rights of minority shareholders Employee satisfaction survey % Participation in shareholders' meetings Presence of sustainability committee Presence of syndicate agreements Formalized skill mapping and developing process for top management Transparency of the annual median employee compensation Transparency of corporate governance policy Transparency of senior management remuneration % Of women on board and committee % Workforce based on minority, culture or similar

DeltaTraining DirBuyShares DirOtherM Employees EqualRemun FemaleMngt FemaleTotW GrievResol HandSProg IndipDir InjuryRate Meeting NCommittees NonExecDir OccDesRate OtherMandat Performance RightsOfMin SatisfSurvey SharMeet Sustainability SyndAgree TopMngt TranspAMEC TranspCGP TranspSMR Woman Workforce

Engagement). Examination of the values ascribable to these variables reveals the following (Tables 6 and 7). Finally, the Strategy variable derives from the aggregation of the eleven intermediate variables: (Codes of Conduct, Customer Relationship Management, Customer Satisfaction, Green, Management, Measurer to Manage Sustainability Risks, Operations, Operational Coefficiency, Reduction, Risks and Crisis Management, Supply Chain Management). Examination of the values ascribable to these variables reveals the following (Tables 8 and 9). The analysis of the three different dimensions shows that the overall results are strongly influenced by the Environmental Social Governance reporting dimensions. This is particularly true for A and C, which present the best results, if compared with B. Furthermore, the quality of the process of stakeholder engagement (Greenwood, 2007; Manetti, 2011), the publication of an integrated report (Rowbottom and Locke, 2016) and the presence of a GRI-assurance (Guidry and Patten, 2010) are considered the most significant variables. According to literature (Ayuso et al., 2011) the model shows that knowledge sourced from engagement with internal and external stakeholders contributes to a firm's sustainable innovation

A 13 0 80 yes Executive level 9 27 0 yes 55 38 95 3 86,7 0 3 medium Yes No 78 Presence Yes Yes High High High 28 27

B

C

0 0 92 yes no 10 61 0 yes 66 3 90 4 92,3 0 2 low Yes Yes 68,7 Absence No Yes Low High High 23 61

6 0 64 yes no 24 49 0 yes 75 9 90 5 90,9 0 1 low Yes Yes 47 Presence No Yes Medium_low High High 36 49

orientation, connected to a high value of CSR identity. In the same way, the results seem to confirm that CSR identity is strengthened by the positive association between the publication of an integrated report and the assured CSR report (Sierra-García et al., 2015). Currently, these results support the idea of reporting as an integrated process more than as a simple output, as described in Section 2. In particular, the conclusions regarding the reporting strictly linked to strategy are in line with what emerges from the reading of the International Integrated Report Framework and with the series of studies on instrumental theories (Husted and Hallen, 2000), whereby the integration of Corporate Social Responsibility into systems of company governance and management becomes a crucial factor in the creation of value in the long and medium term (Crane et al., 2014; Graafland, 2016; Porter and Kramer, 2006). 6. Conclusions This paper takes as its starting point the lack of methods available to assess the CSR identity of a firm. With the aim of filling this gap, we propose a model using a

Table 5 Human capital output. Intermediate and output variable description

Variable

A

B

C

Board effectiveness Board structure Board transparency Committees within the board Diversity/Non discrimination Employee satisfaction Governance Human capital development Health and safety Human resources Non discrimination Ownership Human Capital

B_Effectiveness B_Structure B_Transparency Committees Diversity EmplSatisf Governance HCDevelop HealthSafety HumRes NonDiscrim Ownership HumCap

60,00 98,09 100,00 100,00 79,17 0,00 87,50 80,00 33,33 53,70 96,67 100,00 80,00

60,00 81,17 66,67 33,33 100,00 33,33 71,02 80,00 100,00 66,67 60,00 43,00 60,64

60,00 100,00 66,67 100,00 99,17 33,33 75,00 100,00 100,00 77,41 59,33 33,33 60,00

1008

A. Venturelli et al. / Journal of Cleaner Production 141 (2017) 1000e1010

Table 6 ESG reporting input. ESGreporting: Input description

Variable

A

B

C

Assurance Category of stakeholders (%) Cost savings on additional revenue Stakeholder engaged Interactive section 2.0 Level of engagement Materiality matrix Type of engagement Type of report

Assurance CategoryStk CostSav_AR EngagedStk InterSect2 LevelEngag MaterialMatrix TypeEngag TypeRep

External and GRI 100 Calculated and reported 0 Other Results reported Asserted Other Integrated

External 75 Calculated and reported 0 Download manager None Yes Exchange info Sustainability

External and GRI 100 No 0 Interactive analysis Not reported outside Yes Dialogues through web Integrated

Table 7 ESG reporting output. Intermediate and output variable description

Variable

A

B

C

Quality of report Report evaluation Stakeholder engagement

QualityRep RepEval StkEngag

75,00 100,00 66,67

37,50 50,00 16,66

25,00 60,00 33,33

Environmental Social Governance reporting

ESGrep

100,00

58,33

83,33

Table 8 Strategy input. Strategy: Input description

Variable

A

B

C

% Satisfied clients % Reduction Co2 emissions Codes of Conduct areas Effectiveness of codes of conduct Collaborative initiative with suppliers Complaints management Contingency plans for the key social/environmental risks identified Formal process to identify critical suppliers Customer satisfaction measurement Timing of ecological targets % Reduction of energy consumption % New suppliers screened using environmental criteria Environmental management system ESG factors integrated in the suppliers selection decision % Reduction of direct and indirect gas emissions Incentives for suppliers % Materials that are recycled input materials % Renewable materials used Report on suppliers policy Risk response strategy/policy approved Presence of Risk Management Sensitivity analysis for energy price % Self-generated electricity - heating - cooling - steam Social management system Standard suppliers - issue covered Standard suppliers - quality of management system Sustainability policy Formal process to identify sustainability risks in supply chain Audit/third party assessment Risk analysis tools used Impact from transportation policy % Reduction of waste generation % Reduction of water used

ClientsSatisf Co2emissions CodCondAre CodCondEff CollabInitiatS ComplMng ContingePlan CriticalSupp CustomSatisf EcoTargets EnergyCons EnvCriteria EnvMngtSys ESGfactors GasEmissions IncentivesS RecycInputs RenewUsed RepSpolicy RespStrategy RiskMngt SAnalysEP SelfGenerat SocMngtSys SS_IssCov SS_QualMS SustainPol SustRisksSC ThirdParty Tools TranspPolicy WasteGener WaterUsed

71 20 8 4 Yes No Yes Yes yes Pluriannual 20 0 Occurred outside Yes 0 Economic 0 0 Yes Yes Yes Yes 0 Occurred outside 5 4 4 Yes Yes 3 2 0 0

0 0 5 2 Yes No No Yes yes Pluriannual 8 0 Occurred outside Yes 0 No 0 0 Yes Yes Yes No 0 No 1 5 3 No Yes 2 0 12 3

0 20 7 4 Yes No No Yes yes Annual 25 0 Occurred inside Yes 20 No 5 9 Yes Yes Yes Yes 7 Occurred outside 2 3 2 Yes No 4 1 0 15

Fuzzy Expert System approach, which serves to combine the intuition and the experience of the experts who supply a knowledge base with the formal rigor of a logic system. The main findings of the paper have both theoretical and empirical implications. From a theoretical point of view, this study contributes to broadening the understanding of the alternative measurement of CSR identity. On the management side, the possibility of identifying the determinants of the different Corporate Social Responsibility intermediate indicators making up the final

Corporate Social Responsibility score can assist CSR-compliant managers in decision-making. Indeed, by establishing the positioning of a company with respect to its competitors or with respect to itself in other periods, this model allows comparisons and, as a consequence, improvements in its own behavior to be made. Unlike the assessment given by the rating agencies, this model allows for the breakdown of the final output into the three dimensions and the relative variables. In this way, it is possible to determine the single variables that have most influenced the final

A. Venturelli et al. / Journal of Cleaner Production 141 (2017) 1000e1010

1009

Table 9 Strategy output. Intermediate and output variable description

Variable

A

B

C

Codes of Conduct Customer relationship Management Customer satisfaction Green Management Measurer to manage sustainability risks Operations Operational coefficiency Reduction Risks & crisis management Supply chain management Strategy

CodesConduct CustomRelM CustomSat Green Management MngSustRisks Operations OperCoeff Reduction RisksCrisMngt SupplChainMngt Strategy

65,91 35,35 80,47 14,29 62,37 84,58 90,91 50,00 44,44 80,95 79,64 75,19

43,52 0,00 33,33 0,00 36,82 69,44 54,55 60,00 66,67 42,85 50,00 43,17

59,62 0,00 33,33 16,19 39,87 36,00 72,73 50,00 77,78 52,38 69,83 58,45

output. This is particularly important for managers, in order to understand the single elements that need to be improved. The main limitation of the research derives from the fact that only firms particularly active in terms of Corporate Social Responsibility were able to provide and evaluate all the input variables of the model. Even though the peculiarity of the model is that it can evaluate those companies particularly involved in CSR, which have a CSR manager/function and are large enough to be able to make CSR investments, in the near future all companies will find it expedient to be compliant towards CSR in accordance with the International Integrated Report Framework. Against this background, the next phase of this work could be the application of the model to a larger number of companies, also in order to compare the results of this model with the evaluation of Corporate Social Responsibility compliance made in the same cases by rating agencies, as already discussed in Section 3. Acknowledgements We would like to thank the CSR managers of Italian listed companies involved in the analysis. References Armstrong, D., Gosling, A., Weinman, J., Marteau, T., 1997. The place of inter-rater reliability in qualitative research: an empirical study. Sociology 31 (3), 597e606. Atkinson, G., 2000. Measuring corporate sustainability. J. Environ. Plan. Manag. 43 (2), 235e252.   ~ o, M., 2011. Does Ayuso, S., Angel Rodríguez, M., García-Castro, R., Angel Arin stakeholder engagement promote sustainable innovation orientation? Ind. Manag. Data Syst. 111 (9), 1399e1417. Bertin, G., 2005. Tecniche basate sulle conoscenze degli esperti. In: Bernardi, L. (Ed.), Percorsi di ricerca sociale, vol. 1. Carocci, Rome, pp. 216e236. Calabrese, A., Costa, R., Menichini, T., Rosati, F., 2013. Does corporate social responsibility hit the Mark? A stakeholder oriented methodology for CSR assessment. Knowl. Process Manag. 20 (2), 77e89. Calabrese, A., Costa, R., Levialdi, N., Menichini, T., 2016. A fuzzy analytic hierarchy process method to support materiality assessment in sustainability reporting. J. Clean. Prod. 121, 248e264. Chan, F.T.S., Qi, H.J., Chan, H.K., Lau, H.C.W., Ip, R.W.L., 2003. A conceptual model of performance measurement for supply chains. Manag. Decis. 41 (7), 635e642. Chatterji, A., Levine, D., 2006. Breaking down the wall of codes: evaluating nonfinancial performance measurement. Calif. Manag. Rev. 48 (2), 29e51. Chelli, M., Gendron, Y., 2013. Sustainability ratings and de disciplinary power of the ideology of numbers. J. Bus. Ethics 112, 187e203. Cherchye, L., Kuosmanen, T., 2006. Benchmarking sustainable development: a synthetic meta-index approach. In: McGillivray, M., Clarke, M. (Eds.), Understing Human Well-being. United Nations University Press, Tokyo, pp. 139e169. Costa, R., Menichini, T., 2013. A multidimensional approach for CSR assessment: the importance of the stakeholder perception. Expert Syst. Appl. 40, 150e161. Crane, A., Palazzo, G., Spence, L.J., Matten, D., 2014. Contesting the value of ”creating shared value. Calif. Manag. Rev. 56 (2), 130e153. Crotty, M., 1998. The Foundations of Social Research: Meaning and Perspective in the Research Process. Sage. Dahlsrud, A., 2008. How Corporate Social Responsibility is defined: an analysis of 37

definitions. Corp. Soc. Responsib. Environ. Manag. 15, 1e13. Delmas, M., Doctori Blass, V., 2010. Measuring corporate environmental performance: the trade-offs of sustainability ratings. Bus. Strategy Environ. 19 (4), 245e260. Duggan, E.W., Thachenkary, C.S., 2004. Integrating nominal group technique and joint application development for improved systems requirements determination. Inf. Manag. 41 (4), 399e411. Dunham, R., 2006. Nominal Group Technique: a User's Guide. University of Wisconsin, Madison, WI. Elkington, J., 1997. Cannibals with Forks: the Triple Bottom Line of 21st Century Business. Capstone. Engert, S., Rauter, R., Baumgartner, R.J., 2016. Exploring the integration of corporate sustainability into strategic management: a literature review. J. Clean. Prod. 112, 2833e2850. ndez, A., 2010. Socially responsible investing: susEscrig, E., Muneoz, M.J., Ferna tainability indices, ESG rating and information provider agencies. Int. J. Sustain. Econ. 2 (4), 442e461. Escrig-Olmedo, E., Munoz-torres, M.J., Fernandez-izquierdo, M.A., Rivera-lirio, J.M., 2014. Lights and shadows on sustainability rating scoring. Rev. Manag. Sci. 8 (4), 559e574. Facchinetti, G., Magni, C.A., Mastroleo, G., Vignola, M., 2001. Valuing strategic investment with a fuzzy expert system: an Italian case. Proceedings of IFSA/ NAFIPS 2001, Vancouver, Canada, 25e28 July. Finch, N., 2004. International Rating Agencies and Sustainability. Macquarie Graduate School of Management. Project No. 952. Fray, A.M., 2007. Ethical behaviour and social responsibility in organizations: process and evaluation. Manag. Decis. 45 (1), 76e88. , D., 2004. Corporate social responsibility theories: mapping the Garriga, E., Mele territory. J. Bus. Ethics 53 (1e2), 51e71. Gee, M., Norton, S., 2012. Corporate social responsibility: strategic and managerial implications. Journal of leadership. Account. Ethics 10 (3), 37e43. Girerd-potin, I., Jimenez-garces, S., Louvet, P., 2014. Which dimensions of social responsibility concern financial investors? J. Bus. Ethics 121 (4), 559e576. Govindan, K., Khodaverdi, R., Jafarian, A., 2013. A fuzzy multi criteria approach for measuring sustainability performance of a supplier based on triple bottom line approach. J. Clean. Prod. 47, 345e354. Graafland, J.J., Eijffinger, S.C.W., Smid, H., 2004. Benchmarking of corporate social responsibility: methodological problems and robustness. J. Bus. Ethics 53 (1e2), 137e152. Graafland, J.J., 2016. Price competition, short-termism and environmental performance. J. Clean. Prod. 116, 125e134. Greenwood, M., 2007. Stakeholder engagement: beyond the myth of corporate responsibility. J. Bus. Ethics 74 (4), 315e327. Guidry, R.P., Patten, D.M., 2010. Market reactions to the first-time issuance of corporate sustainability reports: evidence that quality matters. Sustainability Accounting. Manag. Policy J. 1 (1), 33e50. Husted, B.W., Hallen, D.B., 2000. Is it ethical to use ethics as strategy? J. Bus. Ethics 27 (1e2), 21e32. Jiang, P., Xu, B., Geng, Y., Dong, W., Chen, Y., Xue, B., 2016. Assessing the environmental sustainability with a co-benefits approach: a study of industrial sector in Baoshan District in Shanghai. J. Clean. Prod. 114, 114e123. Kinderyte, L., 2008. Analysis and comparison of methodologies for corporate sustainability assessment. Environ. Res. Eng. Manag. 46 (4), 66e75. Kouikoglou, V.S., Phillis, Y.A., 2011. Application of a fuzzy hierarchical model to the assessment of corporate social and environmental sustainability. Corp. Soc. Responsib. Environ. Manag. 18, 209e219. Krajnc, D., Glavi c, P., 2005. How to compare companies on relevant dimensions of sustainability. Ecol. Econ. 55 (4), 551e563. Krueger, R.A., Casey, M.A., 2009. Focus Groups: a Practical Guide for Applied Research, fourth ed. Sage Publication, Thousand Oaks, CA. pez, M.V., Garcia, A., Rodriguez, L., 2007. Sustainable development and corporate Lo performance: a study based on the Dow Jones sustainability index. J. Bus. Ethics 75 (3), 285e300. Magni, C.A., Malagoli, S., Mastroleo, G., 2006. An alternative approach to firms'

1010

A. Venturelli et al. / Journal of Cleaner Production 141 (2017) 1000e1010

evaluation: expert systems and fuzzy logic. Int. J. Inf. Technol. Decis. Mak. 5 (1), 195e225. Manetti, G., 2011. The quality of stakeholder engagement in sustainability reporting: empirical evidence and critical points. Corp. Soc. Responsib. Environ. Manag. 18 (2), 110e122. Marchi, G., Vignola, M., Facchinetti, G., Mastroleo, G., 2014. International market selection for small firms: a fuzzy-based decision process. Eur. J. Mark. 48 (11/ 12), 2198e2212. Marquez, A., Fombrun, C., 2005. Measuring Corporate social responsibility. Corp. Reput. Rev. 7 (4), 304e308. Martinez, F., Dominguez, M., Fernandez, P., 2005. El analisis evolvente de datos en la construccion de indicadores sinteticos: una aplicacion a las provincias espanolas. Estud. Econ. Apl. 23 (3), 753e771. Matten, D., Moon, J., 2008. “Implicit” and “explicit” CSR: a conceptual framework for a comparative understanding of corporate social responsibility. Acad. Manag. Rev. 33 (2), 404e424. Mio, C., Venturelli, A., 2013. Non-financial information about sustainable development and environmental policy in the annual reports of listed companies: evidence from Italy and the UK. Corp. Soc. Responsib. Environ. Manag. 20 (6), 340e358. Mio, C., Venturelli, A., Leopizzi, R., 2015. Management by objectives and corporate social responsibility disclosure. First results from Italy. Account. Audit. Account. J. 28 (3), 325e364. Morimoto, R., Ash, J., Hope, C., 2005. Corporate social responsibility audit: from theory to practice. J. Bus. Ethics 62 (4), 315e325. Morgan, D.L., 1993. Successful Focus Groups: Advancing the State of the Arst. Sage Publication, Newbury Park, CA. Munda, G., Nijkamp, P., Rietveld, P., 1994. Fuzzy multigroup conflict resolution for environmental management. In: Weiss, J. (Ed.), The Economics of Project Appraisal and the Environment. Edward Elgar, Cheltenham, pp. 161e183. Munoz, M.J., Rivera, J.M., Moneva, J.M., 2008. Evaluating sustainability in organizations with a fuzzy logic approach. Ind. Manag. Data Syst. 108 (6), 829e841. Otubanjo, O., 2013. Theorising the interconnectivity between corporate social responsibility (CSR) and corporate identity. J. Manag. Sustain. 3 (1), 74e94. Pava, M., 2008. Why corporations should not abandon social responsibility. J. Bus. Ethics 83 (4), 805e812. Pena, J.B., 1977. Problemas de la medicion del bienestar y conceptosafines. Una Aplicacion Al Caso Espanol. INE, Madrid. Phillis, Y.A., Andriantiatsaholiniaina, L.A., 2001. Sustainability: an ill-defined concept and its assessment using fuzzy logic. Ecol. Econ. 37 (3), 435e456. Phillis, Y.A., Davis, B.J., 2009. Assessment of corporate sustainability via fuzzy logic. J. Intell. Robot. Syst. 55 (1), 3e20. Porter, M.E., Kramer, M.R., 2006. Strategy and society: the link between competitive advantage and corporate social responsibility. Harv. Bus. Rev. 84 (12), 78e92. Rajak, S., Vinodh, S., 2015. Application of fuzzy logic for social sustainability performance evaluation: a case study of an Indian automotive component manufacturing organization. J. Clean. Prod. 108, 1184e1192. Rebai, S., Mohamed, N.A., Dhafer, S., 2016. A multi-attribute utility model for

generating a sustainability index in the banking sector. J. Clean. Prod. 113, 835e849. Remisova, A., Baciova, Z., 2012. Measuring corporate social responsibility towards employees. J. East Eur. Manag. Stud. 17 (3), 273e291. ~ oz, M.J., 2009. The effectiveness of the public support policies for Rivera, J.M., Mun the European industry financing as a contribution to sustainable development. J. Bus. Ethics 94 (4), 489e515. Rowbottom, N., Locke, J., 2016. The emergence of IR. Account. Bus. Res. 46 (1), 83e115. Ryan, B., Scapens, R.W., Theobald, M., 2002. Research Method and Methodology in Finance and Accounting. Salmi Mohd, I., Reast, J., 2012. Measuring corporate social responsibility (CSR) with multidimensional scales: a caution on the risks of conceptual misspecification. In: European Conference on Management, Leadership & Governance. Salvati, L., Zitti, M., 2009. Substitutability and weighting of ecological and economic indicators: exploring the importance of various components of a synthetic index. Ecol. Econ. 68 (4), 1093e1099. Saveanu, T.G., Abrudan, M.M., Giurgiu, A., Mester, L.E., Bugnar, N., 2014. Measuring Corporate Social Responsibility practices of micro and small enterprises - pilot study. Proceedings of The 8^ International Management Conference Management challenges for sustainable development. Bucarest, Romania, 6e7 July. Sch€ afer, H., 2005. International corporate social responsibility rating systems: conceptual outline and empirical results. J. Corp. Citizsh. 20, 107e120. Shengtian, H., Weihui, F., Xiaosong, L., 2010. Achieving sustainability with a stakeholder-based CSR assessment model for FIEs in China. J. Int. Bus. Ethics 3 (1), 41e55. Sierra-García, L., Zorio-Grima, A., García-Benau, M.A., 2015. Stakeholder engagement, corporate social responsibility and integrated reporting: an exploratory study. Corp. Soc. Responsib. Environ. Manag. 22 (5), 286e304. Singh, R.K., Murty, H.R., Gupta, S.K., Dikshit, A.K., 2009. An overview of sustainability assessment methodologies. Ecol. Indic. 9, 189e212. Stubbs, W., Rogers, P., 2013. Lifting the veil on environmental-social-governance rating methods. Soc. Responsib. J. 9 (4), 622e640. kely, F., Knirsch, M., 2005. Responsible leadership and corporate social reSze sponsibility: metrics for sustainable performance. Eur. Manag. J. 23 (6), 628e647. Timperley, S., 2008. Corporate Social Responsibility Indexes: Measure for Measure. Dissertation. The University of Waikato. Widiarto Sutantoputra, A., 2009. Social disclosure rating system for assessing firms' CSR reports. Corp. Commun. 14 (1), 34e48. Windolph, S.E., 2011. Assessing corporate sustainability through ratings: challenges and their causes. J. Environ. Sustain. 1 (1), 61e81. Zadeh, L.A., 1965. Fuzzy sets. Inf. Control 8 (3), 338e353. Zhao, X., Hwang, B., Gao, Y., 2016. A fuzzy synthetic evaluation approach for risk assessment: a case of Singapore's green projects. J. Clean. Prod. 115, 203e213. Zarzosa, P., 1996. Aproximacion a la medicion del bienestar social. Idoneidad del indicador sintetico Distancia P(2). Span. J. Econ. Finance 24 (68), 139e163.