Investigating the role of Strategic Corporate Social Responsibility in the adoption of sustainability oriented innovation Nicolas Poussing LISER – CREM 3, Avenue de la Fonte 4364 Esch-sur-Alzette Luxembourg [email protected]

Abstract This contribution explores the role of Corporate Social Responsibility (CSR) on the adoption on Sustainability Oriented Innovation. To analyse this relationship we adopt an empirical approach. We use a survey carried on in Luxembourg in 2008 on firms’ CSR practices jointly with the Community Innovation Survey 2010. With a dichotomous model (logit), we estimate the effect of different CSR strategies. We contribute to the literature by taking into account two types of CSR strategies: strategic CSR versus responsive CSR. The results shown that CSR has a positive impact on the adoption of Sustainability Oriented Innovation. Keywords: Strategic CSR, Sustainability oriented innovation, empirical approach, Community Innovation Survey.

1.

Introduction

In the global warming context, an increasing number of firms are taking into account their impacts on the environment (Porter and Reinhardt, 2007) and numerous academic papers focus on environmental innovation like the contribution of Rennings (2000). Over the past decade, environmental issue remains an important subject but is more often associated with social concerns. The concept of sustainable development has emerged (Faucheux et al., 2010) and sustainable innovations and sustainability oriented innovation become major subjects in the literature (Ketata et al., 2014). A literature review shows that many articles focus on firms’ environmental practices while sustainable innovation is more challenging than other types of innovation activities by introducing different layers of complexity (Hall and Vredenburg, 2003). In particular, a little attention is given to what drives firms to adopt sustainable innovation (Gilley et al., 2000; Paramanathan et al., 2004). Because more and more firms declare to adopt Corporate Social Responsibility, in other words, integrate social and environmental concerns to their business operations and their interactions with stakeholder on a voluntary basis (Commission of the European Communities, 2001), we hypothesise that Corporate Social

Responsibility could be a driver of sustainable innovation or sustainability oriented innovation. In order to test this hypothesise, we are going to take into account two types of CSR policies in accordance with Porter and Kramer (2006): strategic CSR versus responsive CSR. We hypothesise that a higher level of maturity of CSR policy (strategic CSR) drives organisations to adopt sustainable innovation. The question that arises is: Does Corporate Social Responsibility drive Sustainability innovation and do different CSR strategies have the same effect on the adoption of sustainable innovation? The main objective of this paper is to contribute to this literature by investigating the drivers of sustainable innovation, in particular sustainability oriented innovation. The value added of this contribution is threefold. We will investigate the role of strategic CSR. We will provide empirical evidence from a sample of more than 280 firms and we will be able to provide managerial and policy implications in that it finds under which conditions a firm addresses sustainable issue through innovation. This paper is structured as follows. The concepts used, such sustainable terms and strategic Corporate Social Responsibility, and the corresponding hypothesis will be described in section 2. An overview of the empirical strategy will be presented in section 3 (data, variables, method). Section 4 will summarise the findings. Section 5 concludes by presenting the policy implications, the limitations, and the future prospects.

2.

Concepts and hypothesis

In recent years, the importance of sustainable development stimulates research and many terms emerge in this field. A review made by Glavič and Lukman (2007) provides definitions of different concepts which include the words ‘sustainable’ or ‘sustainability’ (sustainable production, sustainable consumption, sustainable policy, sustainable development, sustainability policy). Glavič and Lukman (2007) notice that all these concepts refer to environmental protection, societal welfare and economic performance. The definition of sustainable development illustrates this fact. The Bruntland’s commission defines Sustainable development as “development that meets the needs of the present without compromising the ability of the future generations to meet their own needs” (WCED, 1987). Simultaneously, the importance of innovation in the corporate commitment to sustainable development is uncovered (Hart and Milstein, 2003; Hockert and Morsing, 2008) and an increasing body of research address social innovation, ecoinnovation, sustainable innovation and sustainability oriented innovation. Social innovations are “Innovative activities and services that are motivated by the goal of meeting a social need and that are predominantly developed and diffused through organizations whose primary purposes are social” (Mulgan, 2007). For OECD1, social 1

http://www.oecd.org/cfe/leed/Forum-Social-Innovations.htm

innovation "can concern conceptual, process or product change, organisational change and changes in financing, and can deal with new relationships with stakeholders and territories”. Social innovation seeks new answers to social problems by: identifying and delivering new services that improve the quality of life of individuals and communities; identifying and implementing new labour market integration processes, new competencies, new jobs, and new forms of participation, as diverse elements that each contribute to improving the position of individuals in the workforce (OECD, 2011). Eco-innovation is related to environmental issue and is associated to the resolution of environmental deterioration and degradation (Aghion et al., 2013; Veugelers, 2012; Ghisetti and Quatraro, 2014). For Schiederig et al. (2012), the concepts of green innovation, ecological innovation and environmental innovation are quite similar. Eco-innovation consist in the “production, assimilation or exploitation of product, production process, service or management or business methods that is novel to the organization (developing or adopting it) and which results, throughout its life cycle, in a reduction of environmental risk, pollution and other negative impacts of resources use (including energy use) compared to relevant alternatives” (Kemp and Foxon, 2007). The Community Innovation Survey (CIS) which addresses ecoinnovation in 2008 adopts the same definition and specifies that “the environmental benefits of an innovation can occur during the production of a good or service, or during the after sales use of a good or service by the end user”2. With sustainability innovation, we don’t consider one specific dimension. Social and environmental issues are addressed simultaneously. Sustainable innovation is broader than environmental innovation because sustainable innovation concept includes a social dimension. To date, various definitions are available. McElroy (2004) proposes three different definitions: “sustainability of innovation artifacts relative to meeting financial or business goals… sustainability of innovation artifacts relative to meeting social and/or environmental goals… and sustainability of innovation processes relative to the validity of their outcomes and their internal authenticity”. Knot (2003) focuses on the ability to support financial success or business growth over a period of time. It is generally accepted that sustainability innovation is related to innovations which contribute to the triple bottom line concept: economic, ecological and social benefits (Yoon and Tello, 2009; Wheeler and Elkington 2001). To resume, sustainability innovation refers to the definition of innovation gives by Roger (1995) - “an idea, practice or object that is perceived as new to an individual or another unit of adoption” - and the definition of sustainable development. Sustainable innovation takes into account both the ecological and the social dimension of innovation activities (Ketata at al., 2015).

2

http://ec.europa.eu/eurostat/documents/203647/203701/CIS_Survey_form_2008.pdf/e06a4c117535-4003-8e00-143228e1b308

In line with the concept of sustainable innovation, a growing body of literature in the field of sustainability-oriented innovation (SOI) emerges. Hansen and Goße-Dunker, (2012) underline that, in comparison with sustainable innovation, SOI consider the risk associate with the social and environmental dimensions (Paech et al., 2007). For Hansen et al. (2009), the market success and non-economic sustainability of SOI are uncertain. To illustrate this argument, Hansen et al. (2009) mentioned the research of Kölsch and Saling (2008) and Rennings and Zwick (2002) relative to the negative societal impacts of bio-fuel. With SOI, sustainability is a direction; a goal of the firm linked to a risk (Wagner and Llerena, 2008). “The concept of SOI expresses only an individual declaration of intent. A priori, the direction of the actual effects of an innovation to sustainable development is unknown” (Hansen et al., 2009, p. 687). If technology-push and a market-pull models explain the adoption of technological innovation, these models could be inappropriate to identify the determinant of sustainable innovation, in particular because innovations which take into account environmental concerns differ fundamentally from other types of innovation (Kemp and Soete, 1992). To promote environmental practices, taxes, regulation and incentives remain three important determinants (Acemogulu et al. 2012; Aghion et al. 2009, Veugelers 2012). In addition to these external factors, Demirel and Kesidou (2011) find a positive influence of internal firms’ behaviours which are voluntary implemented. As a consequence, in our point of view, Corporate Social Responsibility, assimilated by Antonioli and Mazzanti (2009) to voluntary measures, becomes a major determinant of the adoption of innovation when innovations pursue an environmental objective. This hypothesis is strongly supported by a significant body of literature which analyzes the relation between CSR and innovation and try to better understand why, as mentioned by the European Commission (2011), CSR drives innovation. Most of these studies show a positive effect of CSR on innovation. Mcwilliams and Siegel (2001) underlined the fact that CSR generates technological innovation. Nidumolu et al. (2009) consider CSR as a one of the key drivers of innovation. For Bocquet et al. (2013), CSR strategies lead to technological innovation. For Hart (1995), Jaffe and Palmer (1997), Surroca et al. (2010) and Renning and Rammer (2011), the implementation of environmental practices in the CSR context has an effect on innovation. Moreover, Poussing and Le Bas (2013) and Bohas et al. (2014) shown that Corporate Social Responsibility has a positive impact on business practices in favour of the environment. Poussing and Le Bas (2013) adopt an empirical approach with micro-data at the firms level to shown that CSR plays a positive role in the adoption of environmental innovation. In the same vein, the work by Bohas et al. (2014) underline the positive impact of CSR on the adoption of green IT. Because in essence environmental practices are a part of Sustainable Innovation or sustainability oriented innovation, CSR could be a determinant of this kind of innovation. This effect should be reinforcing by the fact that social issues concern

both CSR and sustainable Innovation (sustainable Innovation or sustainability oriented innovation). In line with this framework, we could formulate a first proposition: Proposition 1: CSR has a positive impact on the adoption of sustainability oriented innovation by firms. CSR can take many forms (Brammer et al., 2007). CSR practices are related the social, environmental and economic dimensions, so called the triple bottom line principle (Elkington, 1997). CSR is not characterized by a single activity, but by a set of very different activities (Lindgreen et al., 2008). CSR activities could be described along a continuum of actions between do nothing to do much (Carroll, 1979). In consequence, CSR is measured in different ways (Wolfe and Aupperle, 1991). Some measures come from firms’ publication, other measures from case studies, survey, reputation indices or perceptional scales (Waddock and Graves, 1997). The conceptualization of Carroll (1979) includes four dimensions: economic responsibilities, legal responsibilities, ethical responsibilities and discretionary responsibilities. Economic responsibilities are related to the obligation for businesses to make profit and produce services and goods. Legal responsibilities refer to the respect of the law. Ethical responsibilities expects that organizations adopt moral rules. Discretionary responsibilities refer to voluntary and charitable activities. For each of the dimensions of Carroll’s conception, Maignan and Ferrell (2000) develop measures. They also elaborate a typology of measure in three categories: expert evaluations, single- and multiple-issue indicators, and surveys of managers (Maignan and Ferrell, 2000). Other researchers proposed to distinguish two types of CSR practices: environmental practices and social practices (Baden et al., 2009; Fernando, 2010). It is also possible to investigate the reason why firms deploy CSR practices. In line with this point of view, Sethi (1979) propose a typology in four categories: reactive, defensive, responsive, and proactive. Other researches distinguish only two types of CSR initiative: proactive vs reactive (Groza et al., 2011; Du el al., 2007). Reactive strategy permits to protect the image after irresponsible actions occur. Proactive CSR consist in deploying CSR practices to prevent irresponsible actions. In most cases, the effects of CSR are not clear cut because the effect depends on which CSR practices are taking into account (Lankoski, 2009). When Brammer and Milligton (2008) or Barcos et al. (2013) analyse the link between CSR and firm performance, they show that some CSR practices have a positive impact on firm performance while other are not. In recent years, growing literature considers CSR as value-driven (Porter and Kramer, 2006; Vilanova et al., 2009). Some authors distinguish two kinds of CSR: CSR driven by pure altruism versus strategic CSR which is profitable (Lyon and Maxwell, 2008; Baron, 2001). For Porter and Kramer (2006), CSR creates a competitive advantage when firms integrate CSR into their strategic visions. Porter and Kramer (2006) distinguish strategic CSR, which is part of the business strategy

and ties in with the highest level of commitment, to responsive CSR which is a limited level of commitment in the firms. Burke and Logsdon (1996, 497) propose differentiating strategic CSR from responsive CSR through five strategy dimensions: (1) centrality (the ‘closeness of fit to the firm’s mission and objectives’); (2) proactivity (the ‘degree to which the programme is planned in anticipation of emerging social trends and in the absence of crisis’); (3) voluntarism (‘the scope for discretionary decision-making and the lack of externally imposed compliance requirements’); (4) visibility (‘observable, recognizable credit by internal and/or external stakeholders for the firm’); (5) specificity (the ‘ability to capture private benefits by the firm’)3. Because strategic Corporate Social Responsibility and sustainability oriented innovation contribute to the improvement of firms’ performance (Porter and Kramer, 2006; Maletič et al., 2015), we could hypothesis that the firms are going to implement both. Strategic Corporate Social Responsibility and sustainability oriented innovation are positively linked. Strategic Corporate Social Responsibility which is a more general strategy could be a driver of sustainable oriented innovation. According to these considerations we could improve our first proposition by formulating a more detailed one: To the best of our knowledge, no previous research investigates the role of Strategic Corporate Social Responsibility in the adoption of sustainable oriented innovation. To complete the strategic CSR framework, we could improve our first proposition by formulating a more detailed one: Proposition 2: strategic CSR has a positive impact on the adoption of sustainability oriented innovation.

3. Empirical strategy 3.1. The data To assess the effect of strategic CSR on the adoption of sustainability oriented innovation, we used two Luxembourgish data sets. The first data set comes from a survey of CSR practices by firms. The second data set comes from the Community Innovation Survey (CIS 2010) which is a commonly used data source for measuring private sector innovations (Veugelers, 2012). The CSR survey was conducted by Luxembourg Institute of Socio-Economic Research (LISER)4 in 2008. This survey included firms, with 10 employees and more, belonging to all economic sectors. This survey gives details about the CSR activities of firms in 2008. Among a population of 3.296 firms, we built a sample of Husted and Allen (2007) propose measures for each dimension of Burke and Logsdon’s conceptual framework (1996). 4 Luxembourg Institute of Socio-Economic Research, formerly CEPS/INSTEAD, http://www.liser.lu 3

2.511 firms. With a questionnaire in French, German and English, we obtain 1.114 responses. The survey provides details about CSR activities of the firms; in particular on the implementation of their CSR activities: the existence of a CSR department, allocation of a CSR budget, definition of measurable objectives, creation of a reporting system, training of the staff, etc5. The Community Innovation Survey was conducted by LISER in 2012, on behalf of STATEC (the National Statistics Institute of Luxembourg) with financial support from the European Commission (EUROSTAT). The target population of the CIS 2010 is the total population of enterprises, with 10 employees or more, in NACE Rev. 2 (these sections include most market activities). From a sample of 958 firms, we obtain 652 responses with face-to-face interviews. Many of the CIS questions have been used in prior versions of the survey6. The survey describes firms’ innovation behaviour in terms of product, process or organizational innovation for the period 2008-2010. In CIS 2010 a specific part of the survey is dedicated to innovation objectives. The ten questions introduced in this specific part of the questionnaire allow us to know if the firms implement sustainability or non sustainability oriented innovation. These two surveys followed exactly the same methodology for the sampling process: a stratified random sample of firms from the national database of companies located in Luxembourg, available from STATEC. In consequence, using the identification number for the firms, we merge the two data sets and solve an important limitation of the CIS data: the impossibility to assess path dependency (CSR survey gives a description of the CSR practices in 2008 and CIS covers the period 2008-2010). Our final dataset contain 286 firms. To obtain representative results of the studied population, we use a weighting system based on the sampling probability and the rate of response. In our sample, the proportion of large firms (with 250 employees and more) represents 4.0% of the sample. The proportion of industrial firms is 12.3%. Around one firm in four (27.0%) adopts CSR practices. Concerning innovation practices, 44.6% of firms implement a product innovation; 39.1% a process innovation and 55.8% implement an organizational innovation.

5

Appendix 1 provides the questionnaire items of the CSR survey used in our paper. The harmonized survey questionnaire is available (Last access: July 2015) at: http://ec.europa.eu/eurostat/documents/203647/203701/CIS_Survey_form_2010.pdf/b9f2c70e-0c464f82-abeb-c7661f1f2166 6

3.2. Sustainability oriented innovation variables In our database, a question concern the importance gives by the firms for ten different objectives for their activities to develop product or process innovations during the three years 2008 to 2010. Among these objectives, sustainability oriented innovations are captured by three items: two of them are relative to environmental issue: Reduce material and energy costs per unit output (ENERGY), Reduce environmental impacts (IMPACT) and one concern social issue: Improve health or safety of your employees (HEALTH). The other objectives are more market oriented innovation: Increase range of goods or services (LARGE), Replace outdated products or processes (REPLACE), Enter new markets or increase market share (MARKET), Improve quality of goods or services (QUALITY), Improve flexibility for producing goods or services (FLEXIBILITY), Increase capacity for producing goods or services (CAPACITY), Reduce labour costs per unit output (COST). The analysis of the importance given to each objective (high or medium importance) shows that the proportion of innovative firms is greater for non sustainability oriented innovations than for sustainability oriented innovations (cf. Figure 1). Approximately one quarter of firms innovate and consider important the objective to improve health or safety of your employees (26.6%) or to reduce environmental impacts (25.5%). Adopt an innovation with the objective to reduce material and energy costs per unit output are considered important for 20.6% firms. Figure 1: Objectives linked to product or process innovations during the three years 2008 to 2010 (% of firms) 60,0 50,0

40,0 30,0

50,7 44,7

41,6 36,7

32,5

29,0

26,6

25,5 20,6

20,0 10,0 0,0

We notice that 56.0% of innovative firms follow more than one objective at the same time. They follow more frequently (9.4%) six objectives.

It is important to see (cf. Table 1) that the firms never follow only sustainable objectives: Sustainable objectives or sustainability oriented innovation are always associated with non sustainability oriented innovation (32.9%). At the opposite, 24.1% follow only non sustainability oriented innovation. Table 1. Distribution of enterprises by type of innovation’s objectives (number, percentage in brackets)

NO Follow non sustainability oriented innovation TOTAL

YES

Follow sustainability oriented innovation NO YES 123 0 (43.0) (0.0) 69 94 (24.1) (32.9) 192 94 (67.1) (32.9)

TOTAL 123 (43.0) 163 (57.0) 286 (100.0)

Source: Community Innovation Survey 2010 and CSR 2008 survey (Luxembourg)

At this stage of our analysis, it seems relevant to analyse the probability to pursue only non sustainability oriented innovation (NO_SUSTAIN) and to analyse the probability to pursue both non sustainability oriented innovation and sustainability oriented innovation (SUSTAIN_NO_SUSTAIN).

3.3. CSR variables Strategic and responsive CSR (Porter and Kramer, 2006) are our main independent variables. To identify these two types of CSR, we use the method of Bocquet et al. (2013). With a cluster analysis, they differentiate firms according to their CSR policy (strategic versus responsive). They use questions, available in the CSR survey, about the implementation of CSR policies according to the five strategy dimensions mentioned by Burke and Logsdon (1996). The first dimension, ‘centrality’, is taking into account with two items: a document exists that describes the firm’s values and whether the firm communicates about its CSR commitment on the Web or in a report. For the second dimension, ‘proactivity’, Bocquet et al. (2013) examine the existence of a CSR action plan and the existence of an agenda. One item measure ‘Voluntarism’: the identification by the firm of its stakeholders. ‘Visibility’ is captured through the existence of a communication plan. Three items linked to value creation for the firm measure ‘Specificity’: the capacity to attract clients, the capacity to improve the firm’s image and the level of differentiation from the competition. With the items presented above, Bocquet et al. (2013) conduct a principal component analysis (PCA). The PCA identifies the uncorrelated factors which best

summarise the information contained in the theoretical dimensions. Next, a nonhierarchical cluster analysis determines the final number of clusters. Among our population of 286 firms, 8.7% adopt a strategic CSR policy (STRATEGIC), whereas 24.8% have a responsive one (RESPONSIVE). We find that the proportion of sustainability oriented innovation firms is higher among firms with a strategic CSR policy (see Table 2): indeed, 40.0% of firms with strategic CSR have adopted sustainability oriented innovation compared with 26.8% of firms with responsive CSR7. A large proportion (65.8%) of firms without CSR policy doesn’t adopt sustainability oriented innovation. Table 2. Firms CSR and sustainability oriented innovation (number, percentage in brackets) Firms without a CSR policy Firms without sustainability oriented innovation Firms with sustainability oriented innovation Total

Firms with a CSR policy

Total

Strategic CSR

Responsive CSR

125 (65.8)

15 (60.0)

52 (73.2)

192 (67.1)

65 (34.2) 190 (100.0)

10 (40.0) 25 (100.0)

19 (26.8) 71 (100.0)

94 (32.9) 286 (100.0)

Source: Community Innovation Survey 2010 and CSR 2008 survey (Luxembourg)

3.4. Other control variables Due to the small number of observations in our final sample, we are unable to introduce all the independent variables took into account in prior empirical research like Nguyen Van et al. (2004); Tidd et al. (2005), Wagner and Schaltegger (2004) or Ziegler and Rennings (2004). To identify the main factors we are going to include as control variables in our empirical model, we referred to the evolutionary framework. More specifically, because firm capabilities play a major role in innovative performance (Teece and Pisano, 1994), we capture it by taking into account the presence of employees with a higher education degree (dummy variable EMPHI). The speed in which products and services become old-fashioned (dummy variable PRODPER) measures technological opportunities which is another important innovation driver (Dosi, 1997). To take into account the effect of competitive intensity on firm innovation, we included in our model a dummy variable (MARCONC), which takes the value 1 when 7

We obtain similar results with the weighted population.

the competition of the market in which the firm is operating in is very intense, and otherwise a value of 0. As usual in previous research, the size, the sector of activity and belonging to group are taking into account. We follow the Commission Regulation N°1450/2004 of the European Parliament and of the Council concerning the production and development of Community statistics on innovation to introduce the size of the firms with three dummy variables: SMALL, from 10 to 49 employees; MEDIUM, from 50 to 249 employees; and LARGE, more than 249 employees. According to Wagner (2010), the innovation performance of the firms is linked to their size. Because the resources of large firms are bigger then small firms, the latter are less innovative, except in high-technology sectors (Cohen, 1995). With the dummy variable INDUS, we distinguish two sectors of activities: industry versus services (Gallego-Alvarez et al., 2011; Husted and Allen, 2007). We introduce the dummy variable GROUP that indicates whether the firm belongs to a group because Mairesse and Mohnen (2010) show that belonging to a group has an effect on the expenditures of RD. We also take into account three independent variables to assess the impact of the implementation of different type of innovation on the probability to follow sustainability oriented innovation. The dummy variable PRODUCT is equal to 1 when the firm implements a product innovation; PROCESS is equal to 1 when the firm implements a process innovation and ORGA is equal to 1 when the firm implements an organizational innovation. Appendix 2 lists descriptive statistics regarding the set of variables introduced into our econometric analysis.

3.5. Method In order to analyse the effect of strategic CSR on the adoption of sustainability oriented innovation, we adopt multivariate models. Because our dependent variable is binary (e.g. equal 1 if the firm adopted non sustainability oriented innovation and 0 if not), dichotomous models, such as logit and probit, are appropriate. Appeared in the 1920s (Thurstone, 1927), more generally used since the work on innovation diffusion by Davies (1979), they are currently very widespread (Hoetker, 2007). Considering that the probit and logit models give similar results (Davidson and MacKinnon, 1984; Morimune, 1979) we perform logit model. In this model, the decision to adopt sustainability oriented innovation or not is defined by yi, where yi = 1 when the company adopted this practice and yi = 0 when it did not. The probability of adoption of sustainability oriented innovation is conditional upon a series of exogenous variables: Prob(yi = 1) = F(β’xi) (1)

where F( ) indicates a cumulative distribution function, xi the explanatory variables and β the vector of the parameters to be estimated.

4. Results and findings With our estimations, we would like to know if strategic CSR conduct to adopt sustainability oriented innovation. To test this idea, we build up two logit models. Each model considers different innovative practices. In the first model (Model 1) we analyze the probability for a firm to adopt only innovation without any sustainable objective. Model 2 focuses on the probability for a firm to adopt innovation with both sustainability objectives and non sustainability objectives. In each model, we introduce the same independent variables. The estimation results are set out in Table 3. To take into account the goodness of fit of the models, we calculated the percentage of concordance and the Cox and Snell pseudo R square. The results are rather similar and suggest that the models appear to fit the data quite well. Table 3. The determinants of sustainability oriented innovation (Logit model)

CSR_STRA CSR_RESPONS NO_CSR PRODUCT PROCESS ORGA EMPHI MARCONC PRODPER SMALL MEDIUM

MODEL 1 NO_SUSTAIN Coefficient, Odds ratio Standard error in Estimates parentheses. -1.2372*** 0.290 (0.2833) 0.0502 1.052 (0.1502) REF. / 1.5949*** 4.928 (0.1221) 0.2021 1.224 (0.1247) -0.4604*** 0.631 (0.1208) 1.0314*** 2.805 (0.1975) 0.5311*** 1.701 (0.1255) -0.3837* 0.681 (0.1988) -0.8060*** 0.447 (0.1171) REF. /

MODEL 2 SUSTAIN_NO_SUSTAIN Coefficient, Odds ratio Standard error in Estimates parentheses. 0.54948*** 1.732 (0.1767) -0.1582 0.854 (0.14884) REF. / 1.8007*** 6.054 (0.1098) 1.8220*** 6.184 (0.1134) 0.5099*** 1.665 (0.1166) -0.1333 0.875 (0.1582) -0.2588** 0.772 (0.1198) -0.2193 0.803 (0.2114) 0.3964*** 1.486 (0.1117) REF. /

LARGE INDUS GROUP Intercept Nb. of Obs. -2 Log L % Concordant R Square

MODEL 1 NO_SUSTAIN Coefficient, Odds ratio Standard error in Estimates parentheses. -0.8050*** 0.447 (0.2946) 0.5698*** 1.768 (0.1633) 0.0337 1.034 (0.1293) -2.6964*** / (0.2110) 286

MODEL 2 SUSTAIN_NO_SUSTAIN Coefficient, Odds ratio Standard error in Estimates parentheses. -0.1800 0.835 (0.2977) -0.2985* 0.742 (0.1616) -0.0287 0.972 (0.1252) -2.4714*** / (0.1706) 286

2322.250 76.8

2483.423 84.9

0.8264

0.9763

* Coef. significant at the threshold of 10%, ** 5%, *** 1%. Source: Community Innovation Survey 2010 and CSR Survey (Luxembourg)

Regarding the effect of CSR strategy, the estimations give crucial results. The coefficients estimated from Model 1 tell us that adopt strategic CSR has a negative impact on the probability to be an innovative firm which follow only non sustainability objectives. When the firms introduce sustainability objectives in their innovative practices, the results from Model 2 show that strategic CSR has a positive effect on the probability to implement this kind of innovation. We could appreciate the effect of strategic CSR with the odds ratios; which is the ratio of the expected number of time that an event will occur to the expected number of times it will not occur (Allison, 1999). In Model 1, the odds ratio of 0.290 tells us that the predicted odds to innovate without any sustainability objectives for firms with strategic CSR are 0.29 times the odds for non strategic CSR firms. In other words, the odds ratio to innovative without any sustainability objectives for CSR strategic firms are 71% less than the odds for non strategic CSR firms. In Model 2, the odds ratio of 1.732 tells us that the odds ratio to innovative with sustainability objectives for CSR strategic firms are 73% higher than the odds for non strategic CSR firms. These results underline the positive effect of strategic CSR on the adoption of sustainability oriented innovation. Our findings show that responsive CSR has no impact on both the probability to innovate with sustainability objectives and the probability to innovate without sustainability objectives. With regard to the impact of innovation behavior of firms, the two models indicate that the firms which innovate in product have a higher probability to innovate with sustainable objectives (odds ratio = 6.054) than to only innovate without any sustainability objectives (odds ratio =4.928). The effect is more settled when the firm

is a process innovator or an organizational innovator. We observe that a process innovator has a higher probability to adopt sustainability oriented innovation and no impact on the probability to adopt innovation without any sustainable objective. Organizational innovator has a negative effect on the probability to innovate without any sustainable objective on the one hand and on the other has significant and positive effect on the implementation of sustainability oriented innovation. In most cases, the signs of coefficients related to control variables clearly show opposite effects, or at least different effects, when innovative firms follow sustainable objectives or not. Amazingly, the variable measuring firm technological capacity (EMPHI) has a positive impact on the probability of undertaking innovation without sustainable objectives and has no significant effect on the probability to adopt sustainability oriented innovation. The sign of coefficients of MARCONC is different in MODEL 1 and MODEL 2, meaning that operate in a market where the competition is very intense increase the propensity to adopt innovation without any sustainable objectives and impact negatively the propensity to adopt both innovation without any sustainable objectives and sustainability oriented innovation. The coefficient related to the speed in which products and services become old-fashioned (PRODPER) is not significant in MODEL 2 and significant at the threshold of 10% in Model 1. We could consider that this feature has no effect on the adoption of innovation adoption whatever the innovation’s objectives followed. With respect to size impacts, in comparison with the medium firms, the coefficient of the variable SMALL is negative in the MODEL 1 and positive in the MODEL 2. The small firms have a lower propensity to adopt innovation without sustainable objectives but they have a larger propensity to adopt sustainability oriented innovation. As a consequence, the small firms appear to be more sustainable friendly. Concerning large firms, we find a negative effect on the probability to innovate without any sustainable innovation but no effect on the adoption of sustainability oriented innovation. We cannot show evidence that large firms are more sustainable friendly. The variable INDUS has a significant positive coefficient in MODEL 1 and a negative and weakly significant coefficient in MODEL 2 meaning that industrial firms have a larger probability to implement innovation without sustainable objectives. Belonging to a group (GROUP) does not improve its probability To sum up, the decision to adopt sustainability oriented innovation is meanly driven by small technological innovative firms which operate in a market where the competition is not very intense and have implemented strategic CSR.

5.

Conclusions

To the best of our knowledge, no previous research investigates the role of Strategic Corporate Social Responsibility in the adoption of sustainable oriented innovation. Our contribution tries to fill this gap and complete the strategic CSR framework. With an empirical approach based on two survey carried on in Luxembourg, we show that strategic CSR is a determinant of Sustainability Oriented Innovation. Our results have implications in terms of public policy. We suggest that motivated firms to implement strategic CSR may contribute to increase the adoption of sustainable objectives through technological innovations The limitations of our research are summarized as follows. As we said in the methodological section, our sample is quite small in consequence we are unable to introduce all the independent variables took into account in prior empirical research. In addition, our analysis is limited to one country. A comparative analysis should be conducted. A comparison with French data could be the solution (the National Institute of Statistic INSEE collected CSR8 and CIS data). In addition further must take into account other divers of sustainable innovation (taxes, complying regulation, incentives to invest in green projects) because the drivers of sustainable innovation are interactive (Dinda, 2004).

8

http://www.insee.fr/fr/methodes/sources/pdf/questionnaire_unite_EnDD.pdf

Appendix 1: Questionnaire items from the Corporate Social Responsibility Survey used in the econometric model Is your company active in the field of Corporate Social Responsibility (CSR)?  Yes  No, but it is scheduled  No within less than 2 years  End of  End of questionnaire questionnaire Glossary: corporate social responsibility is the voluntary integration of companies' social and ecological considerations into their business operations and relations with their stakeholders. Being socially responsible means not only fully meeting the legal obligations applicable, but going still further, and investing "even more" in the human capital, the environment and relations with stakeholders (employees, customers, suppliers, non-governmental organisations, local authorities and shareholders). Where is your CSR policy described? (several replies possible) In your activity report  In a report dedicated to CSR  On your Web site  Nowhere  Other (give details): ______________________________ Do you have a document describing the values and priority concerns and/or motivations of your company in social and environmental terms?

 Yes

 No

Have you identified the stakeholders targeted by your CSR policy?

 Yes

 No

Before initiating your CSR policy, did you enter into contact with your stakeholders?

 Yes

 No

What are the three main effects you wish to achieve with your CSR policy? Attracting new employees Attracting investors Attracting new customers Improving the company's image Standing out from the competition Anticipating changes in legislation Reducing your costs Satisfying your stakeholders Reducing your impact on the environment

|__| |__| |__| |__| |__| |__| |__| |__| |__|

Increasing the well-being of your employees Other (give details): _______________________________

|__|

Before initiating your CSR policy, did you:

(several replies possible) Yes

Make a list of the actions already carried out within your company Make a list of the actions that could be envisaged within your company Study the actions carried out by other companies Collect information from specialised bodies Collect information from the public authorities Find out about existing CSR standards and labels Assess the costs of implementing CSR

□ □ □ □ □ □ □

No

□ □ □ □ □ □ □

Have you drawn up a schedule for the CSR actions you wish to carry out?

 Yes

 No

Have you drawn up any communication plans on your CSR commitments? In-house External

 Yes  Yes

 No  No

Appendix 2. Description and summary statistics of the variables Variables

Definition

NO_SUSTAIN

Firms with only non sustainability oriented innovation

SUSTAIN_NO_SUSTAIN CSR_STRA

Firms with both sustainability and non sustainability oriented innovation Firms with a strategic CSR profile

CSR_RESPONS

Firms with a responsive CSR profile

NO_CSR

Firms don’t adopt CSR

PRODUCT

Firm implements a product innovation that is a new or significantly improved product (good or service) Firm implements a process innovation that is a new or significantly improved process, organizational method or marketing method Firm implements an organizational method that is a new organizational method in their enterprise’s business practices (including knowledge management), workplace organization or external relations that has not been previously used by your enterprise The competition of the market in which the firm is operating in is very intense Products and services become rapidly old-fashioned

PROCESS

ORGA

MARCONC PRODPER EMPHI

SMALL

The firm has employees with higher education (who have either completed a master’s degree in a graduate school, or a university degree, or who hold a doctorate / PHD degree) Total number of employees is between 10 and 49

Means (standard deviation) 0.2412 (0.4285) 0.3286 (0.4705) 0.0874 (0.2829) 0.2482 (0.4327) 0.6643 (0.4730) 0.4545 (0.4988) 0.3671 (0.4828) 0.5734 (0.4954)

0.4160 (0.4937) 0.0804 (0.2724) 0.8496 (0.3580)

0.3496 (0.4776) MEDIUM Total number of employees is between 50 and 249 0.3951 (0.4897) LARGE Total number of employees is more than 249 0.2552 (0.4367) INDUS Belongs to the manufacturing sector 0.4755 (0.5002) GROUP Firm is part of a group 0.5349 (0.4996) Note: All variables are dummies and related to the period 2008–2010 (except CSR variables related to 2008). The main independent variables are in bold

Appendix 3. Correlation matrix Pearson Correlation Coefficients, N = 286 Prob > |r| under H0: Rho=0

NO_SUSTAIN

NO_SUSTAIN

SUSTAIN_NO_SUSTAIN

CSR_STRA

CSR_RESPONS

NO_CSR

100.000

-0.39456

-0.05878

-0.05920

0.08931