Bridging Sustainability and Corporate Social Responsibility : Culture of Monitoring and Evaluation of CSR Initiatives in India

This study examines the continuum of sustainability and corporate social responsibility (CSR) policies, and analyzes broad patterns that have emerged with respect to monitoring and evaluation practices in the CSR programs of Indian companies under new CSR regulations. Under these regulations, the Indian firms are mandated to spend at least 2% of their profits on social and development sectors. We specifically analyze (i) how Indian companies have conceptualized the idea of sustainability in their annual sustainability reports, and how these ideas get reflected in their CSR policies, and (ii) the monitoring and evaluation practices in CSR interventions. The study uses both primary and secondary data sources, and employs text network analysis and narratives-based content analysis to analyze the data. We find that the conceptualization of sustainability is a largely rhetoric and customary exercise that does not take into account variations in firms’ businesses. This approach toward sustainability initiatives presents serious challenges to sustainability, including social sustainability. The study also finds that there is lack of ‘willingness’ and ‘readiness’ among Indian companies to measure and monitor the outcomes of CSR interventions, which is arguably one of the most robust ways to signal their commitment toward corporate sustainability. Although mandatory CSR spending is a recent phenomenon in India, our study establishes that it is only through the design of effective CSR policies that the best practices for Indian business community can emerge in the near future.


Introduction
Corporate social responsibility (CSR) is a "voluntary effort" to address social and environmental concerns integrated with business operations [1].Many scholars have highlighted its important role in promoting social development and welfare [2][3][4][5][6].Although not many estimations have been carried out to estimate the size of global CSR spending, a recent study conducted by the Varkey Foundation and UNESCO estimated that Fortune Global 500 companies' combined CSR budget in 2013 was 19.9 billion USD [UNESCO and Varkey Foundation, "Business Backs Education", January 2015.Available online: http://www.unesco.org/education/BBE-EPG-Report2015.pdf(assessed on 20 May 2018)].This investment is expected to grow several times over in the coming years.
The growing importance of CSR has prompted many countries to develop statutory frameworks to regulate it."By 2015, 98.4% of world top 64 economies have instituted a total of 366 policy instruments, among which 64.2% are mandatory" [7].International frameworks around the social responsibility initiatives of corporates demand a greater responsibility and accountability in their operations while dealing practical implications for companies in designing their CSR programs and monitoring and evaluation (M&E) frameworks.
Sustainability 2018, 10, x FOR PEER REVIEW 3 of 19 implications for companies in designing their CSR programs and monitoring and evaluation (M&E) frameworks.Figure 1 captures the objective of the study and the associated research questions that have been highlighted above.We earlier posited that the CSR initiatives of firms should be designed with the explicit intent to operationalize a firm-specific sustainability framework.This strategy is likely to help firms move toward their sustainability goal(s).The study focuses on two aspects of CSR programs: (i) the effectiveness of CSR program design, and (ii) M&E frameworks that firms adopt to track progress and measure impacts.These are highlighted (in blue) in Figure 1.Among the three possible sustainability outcomes of CSR initiatives (people, profit, and planet), we devote our efforts to understanding the social dimensions of sustainability that are germane to the problems of development in middle and low-income economies.
The rest of the paper is structured as follows.Section 2 explores the relevant literature on CSR, particularly zooming into sustainability and the prevalent monitoring and evaluation (M&E) practices of CSR initiatives.Section 3 discusses methodology, which includes details regarding data type, its collection, and methods to analyze the collected data.Section 4 presents data analysis, while section 5 discusses findings followed by discussion of research findings, limitations, and future directions of work are suggested in Section 6.

Literature Review
There is plenty of literature that empirically investigates the effectiveness of CSR initiatives and sustainability performance in improving companies' financial performance [14][15][16][17], brand equity, credit ratings [18,19], and customers' choice and perceptions [20].Another strand of literature investigates whether CSR has been successful in enhancing the social visibility and profile of firms [21,22].The literature also recognizes the role of CSR in promoting social development in low-income and developing countries [6,23,24], and at the same time provides ample discussion on the sustainability aspects of CSR initiatives where businesses are expected to serve the society in a way that they are able to generate a momentum toward sustainable development [25].While companies have been articulating their views on sustainability for the last three decades or more, embedding them within their operations remains a challenge [26].
Literature usually distinguishes among four major dimensions of sustainability, i.e., economic, environmental, social, and institutional [27], with social sustainability attracting the most attention as it is often assumed to be the 'weakest' sustainability component [28].As discussed in Ajmal et al. Figure 1 captures the objective of the study and the associated research questions that have been highlighted above.We earlier posited that the CSR initiatives of firms should be designed with the explicit intent to operationalize a firm-specific sustainability framework.This strategy is likely to help firms move toward their sustainability goal(s).The study focuses on two aspects of CSR programs: (i) the effectiveness of CSR program design, and (ii) M&E frameworks that firms adopt to track progress and measure impacts.These are highlighted (in blue) in Figure 1.Among the three possible sustainability outcomes of CSR initiatives (people, profit, and planet), we devote our efforts to understanding the social dimensions of sustainability that are germane to the problems of development in middle and low-income economies.
The rest of the paper is structured as follows.Section 2 explores the relevant literature on CSR, particularly zooming into sustainability and the prevalent monitoring and evaluation (M&E) practices of CSR initiatives.Section 3 discusses methodology, which includes details regarding data type, its collection, and methods to analyze the collected data.Section 4 presents data analysis, while Section 5 discusses findings followed by discussion of research findings, limitations, and future directions of work are suggested in Section 6.

Literature Review
There is plenty of literature that empirically investigates the effectiveness of CSR initiatives and sustainability performance in improving companies' financial performance [14][15][16][17], brand equity, credit ratings [18,19], and customers' choice and perceptions [20].Another strand of literature investigates whether CSR has been successful in enhancing the social visibility and profile of firms [21,22].The literature also recognizes the role of CSR in promoting social development in low-income and developing countries [6,23,24], and at the same time provides ample discussion on the sustainability aspects of CSR initiatives where businesses are expected to serve the society in a way that they are able to generate a momentum toward sustainable development [25].While companies have been articulating their views on sustainability for the last three decades or more, embedding them within their operations remains a challenge [26].
Literature usually distinguishes among four major dimensions of sustainability, i.e., economic, environmental, social, and institutional [27], with social sustainability attracting the most attention as it is often assumed to be the 'weakest' sustainability component [28].As discussed in Ajmal et al. (2018) [29], it is often even difficult to understand the information that companies report under the 'social dimension of sustainability'.The literature also highlights the pressure of global-level competition and internal and external stakeholders forcing the companies to adopt sustainability measures [30,31].
When companies are committed to ensuring the sustainability of their businesses, they should be able to monitor the progress and evaluate the impacts so that they are able to objectively ascertain whether they have fulfilled their sustainability commitment or not.This brings to the forefront the importance of having effective monitoring systems and the use of appropriate evaluation methodologies to estimate the impact of CSR interventions.However, there are significant gaps in assessing the readiness of CSR initiatives to monitor and evaluate their own impact, identify flaws in the design of CSR projects and implementation, and take remedial measures to ensure the effectiveness, impact, and sustainability of outcomes from CSR initiatives.Along the same lines, a 2012 United Nations Conference on Trade and Development (UNCTAD) report recognized that the major challenges with CSR programs are "scalability" and "sustainability" [32].The report pointed out that companies are facing an "auditing fatigue" with CSR interventions, as they have to go through multiple external auditing exercises that primarily focus on financial compliance, but lack internal assessment mechanisms of "real sphere" outcomes and impacts [32].
A study of CSR initiatives in Bangladesh recognized that CSR interventions have created positive impacts on targeted social groups; however, in order to replicate these models in other areas, the study emphasized that "there is a need for additional monitoring and critical evaluation" [33].Companies often lack the necessary capacity to monitor and evaluate their CSR interventions [34], while some may even hide information about their CSR interventions.A study of community health-related CSR activities implemented by Canadian companies in low-income countries reported that approximately 75% of the companies did not disclose pertinent information regarding their CSR programs on websites and in their reports [33].The remaining (25%) who provided information on their CSR activities did not provide any substantial information on the contribution of their interventions to the health sector in general or targeted groups in particular [35].One of the underlying factors that explains this information deficit is that companies operate several CSR programs simultaneously without chaining them into a coherent and effective CSR strategy [36].Without such an effective and clear strategy, companies may operate directionless CSR programs, ending up with unachieved outcomes and ineffective outreach to communities [37].
In her 2015 review of CSR practices of Australian mining companies, Bice (2015) [36] found that companies are required to do social impact assessments in order to comply with regulatory requirements.Bice (2015) [38] highlighted that both (that is, CSR and social impact assessment) share "foundational values", yet both practices remain isolated within the same company.She advocates bridging the gap between the two to generate appropriate knowledge for the company as to what exactly the needs of the local communities are, and how effectively they are being fulfilled [38].These findings are significant in light of the recent literature that recognized the institutionalized character of CSR within firms and society [39], and the potential of CSR to bring "positive social changes" [40].A study on Australian mining companies found that there are contradictions between the impact assessment of companies' operations and the investment in social development programs through CSR initiatives [41].Such contradictions can partly be attributed to the planning of CSR initiatives not having been necessarily based on actual community needs and gaps that could otherwise be highlighted by findings of impact evaluation (IE) of CSR initiatives in the past.Planning CSR initiatives is driven by "corporate headquarters-level policies" without connecting with on-ground situations [42].In such cases, CSR interventions do not yield expected outcomes.
Although it is not the case that companies do not engage in the evaluation practices of their CSR initiatives at all, the reliability of methods used, transparency in disclosing the results, and variability in replicating the outcomes of the studies remain areas of major concern [43].The difficulty in assessing the outcomes of CSR initiatives has been a critical issue for both academics and practitioners [44], triggering a quest for developing assessment frameworks to ascertain the impacts of CSR at both local and international levels [45].
To sum up, the existing literature has recognized the importance of CSR initiatives in addressing issues germane to social development while simultaneously pointing out the absence of effective M&E practices as part of firms' CSR policy and program design.The existing literature exhibits significant gaps in three important aspects: (a) the motivation, that is, whether corporate leadership wants to measure and assess the impact; (b) the program design, that is, conditioned on the willingness of the leadership to learn and establish outcomes of CSR initiatives, and whether the design of a CSR program is amenable to monitor the progress and evaluate the outcomes; and (c) the capacity, that is, whether companies have the internal capacity to implement M&E and impact evaluation (IE)-related tasks.This study tries to address these gaps by exploring the linkages between sustainability and CSR practices in a legislated CSR environment.

Methodology
Our study is based on primary and secondary data.For sustainability analysis, the study uses the Global Reporting Initiative's (GRI) compliant sustainability reports of 45Indian companies for the most recent year (2016)(2017).In order to study how Indian companies perceive CSR, we reviewed the CSR policies of 44 companies (the CSR policy of one company was either not available on public domain or not accessible).Our sample consists of large, small, and medium-sized private and public enterprises and companies working in diverse sectors such as chemicals, fertilizers, textile, energy, automobile, logistics, computers, electronics, etc.
We use text network analysis (TNA) to analyze sustainability reports as well as CSR policies, using InfraNodus, a text network visualization software (http://www.infranodus.com).TNA enables us to create a text network graph, identify the "structural properties" of the graph and "central concepts" emerging from the text, and visualize "pathways for the production of meaning within the text" [46].The method provides "a better insight into the hidden agendas present within a text and better understanding of its narrative structure" [46] (p.3).In a text network, words become the nodes, and connections between words become edges.While analyzing the text network, we focus on four distinct network measures: the density, average degree, modularity, and discourse polysingularity.These measures are briefly explained in Table 1.Apart from these measures, we also zoom into the concentration of the network and identification of the communities of words around which the text network is concentrated.

Density
The density of a network is the proportion of possible ties that are present [47].

Modularity
Modularity measures the density of the nodes that are more densely connected together inside a community of text than the rest of the network.A community is a sub-network within the text network [48].

Average Degree
Average degree is the ratio of the total number of edges (links between the words) and the total number of nodes (words).It highlights how many connections each word has to other words in the text [46].

Discourse Polysingularity
Discourse polysingularity indicates the nature of the discourse visualized in the graph.Polysingularity could be of following types: 'sparse'-a fragmented discourse; 'diversified'-a discourse with a high level of plurality; 'focused'-a discourse that prioritizes one or two topics; or 'biased'-pushing forward a specific agenda (Source: Nodus Labs).
In order to understand the implementation strategy of CSR initiatives, and especially for grasping M&E strategies and processes, we conducted semi-structured interviews with 15 CSR managers who hold positions of responsibility in the CSR division of the firms, and were in a position to share policy as well as implementation-related information.We promised interviewees not to disclose details of about their identity and affiliation.We designed an interview protocol that allowed the interviewer to further probe about different aspects and dimensions emerging from the discussion.The selection of respondents was carried out on the basis of convenient sampling, as ensuring the availability of CSR professionals for long interviews was found to be a difficult task.We ensured the representation of companies of different sizes (small, medium, and large) and different sectors (public and private); however, as the sample size is relatively small, equal representation was not feasible.
We use content analysis and narrative methods to analyze the primary data collected via interviews.The basic foundation of content analysis is multiple levels of coding and building further on the codes to generate themes.We coded the interview text and summarized the codes into manageable categories that enabled us in the identification of dominant themes emerging from the text.In order to convey the findings emerging from the interviews, we present statements made by the respondents in their original forms, except for some basic linguistic corrections.

How Do Companies Conceptualize Sustainability?
This section presents an analysis of sustainability reports of companies that follow (voluntarily) GRI standards of sustainability reporting.The GRI is a non-profit organization that was founded in 1997 and offers a globally accepted standard framework (and regular updates) for sustainability reporting that help corporations worldwide to "communicate their impacts on critical sustainability issues" (Global Reporting Initiative, https://www.globalreporting.org/Information/about-gri/Pages/default.aspx).GRI reports also enable other stakeholders to understand how companies perform with respect to their sustainability commitments of not harming the communities, the environment, and natural resources.While the current GRI standard (known as GRI-G4) requires companies to report on several categories, sub-categories, and aspects along with corresponding indicators; our analysis focuses on the companies' conceptualization of sustainability.Our review of sustainability reports highlight that some of these companies use distinct sustainability frameworks by defining "sustainability" as per their belief systems and organizational standpoint, while others simply follow the GRI framework without defining or providing concise views on sustainability.Table 2 provides a summative account of measures and observations drawing upon the TNA of relevant extracts from the sustainability reports.
The total number of nodes (or words) in the network indicates that the companies in our sample have used 150 distinct words while defining or conceptualizing their notion of sustainability.The text network density is 3.7%, which is indicative of a moderately connected network.Social network analysis literature acknowledges that as the size of the network increases, the density of the network decreases, even when the average number of ties per node does not change [47].Although the literature does not suggest a specific size or density scale for a connected network, a density of above 3% in larger networks (where the number of nodes is 150 or more) is indicative of a moderately connected network [49].Similarly, the average degree is 5.5, which implies that each word is on average connected with 5.5 other words, which represents significant network cohesiveness.Modularity helps in identifying dense communities in a text, whereby a community refers to a sub-network of words that are densely connected to each other.A modularity measure of above 0.4 indicates the presence of prominent communities within a text, whereas a low modularity score points towards the irrelevance of communities in the text network [50].The modularity value of the sustainability network was 0.30, which implies an absence of any significant communities in the network.This finding signals that there is absence of a dominant sustainability discourse at the firm level.The network is mostly concentrated around the topics displayed in Table 2, with a total of 21% of the nodes being organized around these topics.The text network (Figure 2) presents two distinct patterns The first is that the network consists of four communities.Although none of the communities are prominent in the network, their presence indicates how the idea of sustainability is coined among Indian corporates.The word 'sustainability' is not prominent in these communities, except in one where it is combined with 'environment' and 'company'.Environmental sustainability is clearly a buzz phrase that is mentioned in almost all of the sustainability reports, except in the banking and finance sector companies, as their operations are not linked with the direct use of natural resources.Since many businesses either use natural resources directly in their operations or potentially affect them, the commitment to maintain environmental sustainability appears more often.Another set of prominent words that form a community is development, social, and economic.The community denotes how corporates view the hyphenated relationship between the economic and social development of the society.The sustainability discourse among Indian companies does not address issues related to social development, such as human rights, human development, cultural rights, etc.The phrase 'socio-economic' development appears loosely, and it does not offer any committed scheme regarding the socio-economic development of the society.The other two communities in the sustainability discourse are 'ensuring companies performance' and 'integrating business with sustainability', both of which are a natural consequence of the underlying reality that companies are less likely to address those aspects of sustainability that do not feed into their businesses.
Another pattern relates to the group of the most influential words that appear in the network.These are: sustainability, business, development, and stakeholder.This pattern suggests the influence of the GRI framework on the manner in which sustainability is conceptualized among Indian companies.Out of 45 companies, nine have not defined their own idea of sustainability; instead, they have reported their performance against the indicators of the GRI framework.
A content analysis of the sustainability reports indicated an interesting trend: few firms attempted to ensure that their operations achieved sustainability goals by presenting a 'sustainability infrastructure', including a sustainability governance structure as well as criteria for measuring sustainability performance.For example, some companies (three out of 45) discussed the governance structure that they had put into place with the objective of reviewing their sustainability performance, design policies, and programs, and monitoring the progress.These governance structures as well present the criteria for measuring sustainability performance.Four reports list the sustainability challenges and identify measures to overcome those challenges, while five align their sustainability targets with relevant sustainable development goals (SDGs).These trends suggest that there is an initial push toward a transparent approach to ensure the sustainability of environment, social and economic development, along with their businesses.It can be argued that companies' ideas of sustainability vary across sectors; however, our analysis presents contrary findings.We find that there is a common and standard pattern in defining sustainability, irrespective of the sector and nature of business.Instead, we observe an overall lack of coherence in the manner in which most companies articulate sustainability and how they establish the link between their commitments and mechanisms to deliver those commitments.

How Do Companies Define CSR?
The Indian Companies Act of 2013 has mandated every business organization to have a CSR policy with the intent to provide the broad blueprint of design and implementation of their CSR initiatives.This section presents findings from the TNA of companies' CSR policies as to understand how companies define their CSR initiatives, and how effectively these policies are equipped to achieve the intended (promised) social outcomes.The network measures presented in Table 3 highlight a similar picture as that of the TNA of sustainability.While values of density and average degree seem to suggest a connected network, modularity remains below 0.4; hence, an absence of prominent sub-networks (communities of words) was found, which further indicates the lack of a cohesive company approach toward CSR.
The CSR policies network is more fragmented than the sustainability network (see Figure 3); there is a concentration of words in the CSR policies network of 10% around 26 topics compared with 21% around seven topics in the sustainability network.The main topics detected indicate linguistic patterns as to how companies coin their approach toward CSR, or in other words, how It can be argued that companies' ideas of sustainability vary across sectors; however, our analysis presents contrary findings.We find that there is a common and standard pattern in defining sustainability, irrespective of the sector and nature of business.Instead, we observe an overall lack of coherence in the manner in which most companies articulate sustainability and how they establish the link between their commitments and mechanisms to deliver those commitments.

How Do Companies Define CSR?
The Indian Companies Act of 2013 has mandated every business organization to have a CSR policy with the intent to provide the broad blueprint of design and implementation of their CSR initiatives.This section presents findings from the TNA of companies' CSR policies as to understand how companies define their CSR initiatives, and how effectively these policies are equipped to achieve the intended (promised) social outcomes.The network measures presented in Table 3 highlight a similar picture as that of the TNA of sustainability.While values of density and average degree seem to suggest a connected network, modularity remains below 0.4; hence, an absence of prominent sub-networks (communities of words) was found, which further indicates the lack of a cohesive company approach toward CSR.
The CSR policies network is more fragmented than the sustainability network (see Figure 3); there is a concentration of words in the CSR policies network of 10% around 26 topics compared with 21% around seven topics in the sustainability network.The main topics detected indicate linguistic patterns as to how companies coin their approach toward CSR, or in other words, how companies understand what CSR means to them.Almost all of the companies adopted a similar choice of words to express their understanding as well as vision towards CSR.More specifically, words such as 'positive', 'impact', 'socio-economic', 'development', and 'society' suggest standard verbatim.companies understand what CSR means to them.Almost all of the companies adopted a similar choice of words to express their understanding as well as vision towards CSR.More specifically, words such as 'positive', 'impact', 'socio-economic', 'development', and 'society' suggest standard verbatim.We also carried out a content analysis of the 44 CSR policies to find out how equipped these policies are with respect to the M&E and IE strategies of CSR interventions.While the vision and mission of almost all of the CSR policies made commitments to have a sustainable impact on their surrounding communities, our analysis reflects that efforts to track and measure impact are missing in a large number of CSR policies.We also carried out a content analysis of the 44 CSR policies to find out how equipped these policies are with respect to the M&E and IE strategies of CSR interventions.While the vision and mission of almost all of the CSR policies made commitments to have a sustainable impact on their surrounding communities, our analysis reflects that efforts to track and measure impact are missing in a large number of CSR policies.
Figure 4 presents the grading of CSR policies in terms of having clear approaches and frameworks for overall M&E and specifically IE.Along with overall M&E, we also look at IE, as it is an important and distinct evaluation method with its own methodological challenges [51].We reviewed the available evaluability assessment frameworks developed by Davies (2013) [52] and Holvoet et al. (2018) [53].However, these frameworks present an elaborated methodology to assess program planning and monitoring and evaluation (M&E) systems in national and regional level institutions or large-scale development interventions.CSR policies.Therefore, we developed four distinct criteria to grade CSR policies for both M&E and IE (see Table 4).The analysis shows that out of 44 CSR policies, 22 did not even acknowledge adoption of M&E and/or IE systems and processes in their CSR operations (circled in the left corner at the bottom of the graph).Almost a similar number of policies mentioned that the in-house CSR committee would carry out the monitoring.Interestingly, members of the CSR committee often had industry backgrounds, and a lack M&E and social sector-related experience, while the committee's chair was often the head of the human resource department or a senior level officer from the same department.Figure 4 presents the grading of CSR policies in terms of having clear approaches and frameworks for overall M&E and specifically IE.Along with overall M&E, we also look at IE, as it is an important and distinct evaluation method with its own methodological challenges [51].We reviewed the available evaluability assessment frameworks developed by Davies (2013) [52] and Holvoet et al. (2018) [53].However, these frameworks present an elaborated methodology to assess program planning and monitoring and evaluation (M&E) systems in national and regional level institutions or large-scale development interventions.CSR policies.Therefore, we developed four distinct criteria to grade CSR policies for both M&E and IE (see Table 4).The analysis shows that out of 44 CSR policies, 22 did not even acknowledge adoption of M&E and/or IE systems and processes in their CSR operations (circled in the left corner at the bottom of the graph).Almost a similar number of policies mentioned that the in-house CSR committee would carry out the monitoring.Interestingly, members of the CSR committee often had industry backgrounds, and a lack M&E and social sector-related experience, while the committee's chair was often the head of the human resource department or a senior level officer from the same department.Seven CSR policies proposed to establish an M&E system, while only four proposed carrying out an outcome or impact assessment of their initiatives.Some policies noted that an impact assessment would be carried out only for high-value CSR initiatives, which usually implies long-term programs with substantial financial investments.One policy proposed carrying out an IE Seven CSR policies proposed to establish an M&E system, while only four proposed carrying out an outcome or impact assessment of their initiatives.Some policies noted that an impact assessment would be carried out only for high-value CSR initiatives, which usually implies long-term programs with substantial financial investments.One policy proposed carrying out an IE of only those programs that were going to be implemented for three or more years; another earmarked a separate fund (5% of the total CSR budget) for M&E and IE, which adheres to standard practice among international donors.Finally, about 12 CSR policies proposed establishing an M&E system along with either a needs assessment or baseline study; yet, none proposed including both the elements.Similarly, no policy proposed carrying out rigorous IEs with control and treatment groups and/or pre and post-implementation analysis.It is also important to note that only one CSR policy made the commitment to follow the long-term sustainability approach while designing its CSR initiatives.
In sum, an analysis of CSR policies is not supportive of a clear commitment toward the effective M&E of CSR initiatives.This somehow puts the Companies Act of 2013 into perspective, which makes it clear that a company has to "ensure that the activities as [they] are included in [the] Corporate Social Responsibility Policy of the company are undertaken by the company" (Source: The Companies Act 2013, Section 135, Sub-Section 4b).More specifically, if activities related to M&E systems are not included in the CSR policy, it is highly unlikely that these will be integrated into the implementation design of CSR programs, which raises doubts over the commitments of the company to ensure effective CSR implementation, let alone sustainability.

Implementing CSR Initiatives: Conviction or Compulsion?
It is widely acknowledged in the development community that the monitoring and outcome assessment or IE of an intervention starts with the designing of the intervention [54].In order to have an effective M&E system as well as an effective program design for IE, decisions have to be made regarding needs assessments and baseline surveys, program theory designs outlining causal relationships among inputs, outputs, outcomes and impact, identification of key indicators, data collection methods, and identification of a control or comparison group.In this section, we present an analysis of a program design and implementation of CSR initiatives, with a specific focus on M&E and IE, based on interviews with 15 CSR managers.

Identifying People's Needs and Areas of Interventions
The first step toward having a robust program design is carrying out a needs assessment to identify the needs of target communities.The analysis of CSR policies as well as discussions with CSR managers indicate that needs assessment is not a common practice.In fact, in many cases, the needs as identified by the communities themselves are not the main points of consideration for developing CSR initiatives.Based on an analysis of interview data, we have identified broadly four different patterns of identifying the sectors (health, education, water, sanitation, etc.) and specific needs of the local community to design CSR initiatives.The first pattern depicts that CSR interventions resonate with companies' values, guiding thought and identity.Companies design CSR programs as per their vision documents, which have been approved by their board of governors, irrespective of real needs of the local people.Companies do consider what is helping their supply or value chain, and the same considerations are also prominently addressed in CSR operations.
The second pattern indicates that CSR initiatives are designed after taking into consideration the availability of resources and capacities that are available within the companies.Due to such considerations, a substantial amount of CSR resources is invested sporadically, in the form of donations or one-time investments.For example, several companies buy computers for primary and middle schools in rural areas, or they install a water tank or a water cooler; such initiatives do not require any complex program and implementation design or dedicated staff.A third pattern is that CSR departments do not design their own CSR programs; rather, they outsource them by funding the development initiatives of non-government organizations (NGOs).In such cases, companies have to completely rely on NGOs to identify the needs of the people.In the words of a CSR manager: "What we do is we fund the NGO partners and the partners implement the CSR activities for us.When we decide that we want to work for the upliftment of the women, we scout for the NGOs who work in that area and who are reputed ones.If we like their ideas and their working, we fund them.So the needs identification, beneficiary tracking, everything is done by the NGO partners.We only track their progress on a quarterly basis.We are not involved in the individual identification of the beneficiary.Our NGO partners decide and keep us informed.We accept/trust the output numbers that they give us."(Source: Interview with a CSR manager) The fourth pattern that has emerged is related to designing CSR interventions after the priorities of national and state governments, including broad national-level policy priorities such as livelihoods, primary education, skill development, and sanitation.In this manner, companies as well seek to maintain relationships with the governments, as they are able to show that companies' CSR interventions are feeding into the government's mandate of achieving its policy goals.This observation has also been supported by other studies where CSR initiatives were found to be having a positive influence on the government [55].

Program Design
Program design is one of the most critical features in the entire life cycle of a development intervention.We inquired whether companies design CSR initiatives by adopting well-known approaches such as 'theory of change' and/or 'logical framework analysis'.Theory of change, which is also known as program theory, and logical framework analysis are important concepts in project management as well as program evaluation literature.Both are considered important building blocks to build an effective program design.The theory of change is built on critical assumptions and relationships among various programmatic and non-programmatic factors that help in understanding how outcomes will be achieved [56].Logical framework analysis, on the other hand, coins together "the strategic, planning and operational needs of the organization" in order to achieve outputs, outcomes, and impacts [57].Our analysis points out that CSR companies usually do not have a well-articulated theory of change and logical framework analysis, which is indicative of them usually lacking a robust program design in their CSR initiatives.As most of the CSR operations are carried out by implementing partners, such as NGOs, the presence of a program design is entirely left with the NGO's discretion.A CSR manager said that: "We are more worried about outputs than the program design; we don't intend to disturb the internal matters of our partners by asking them to work out a theory of change or a logical framework analysis for our CSR interventions".(Source: Interview with a CSR manager) We found that the absence of an articulated program design approach in CSR interventions is not only a consequence of choice, but also ignorance on the part of CSR companies.There is a dearth of awareness among the CSR functionaries about these tools as well as their effectiveness, which they consider implementation jargon, and as such should be taken care by NGO partners.This is primarily because the success of an intervention is judged by the outputs expressed in numbers (physical progress), rather than by the change that the intervention is expected to bring about in the behavior of the people (the impact).
Another important aspect of an effective program design is the collection of baseline data that enables companies to keep track of the progress and measure the impact generated by the intervention [58].The practice of conducting a baseline survey before an intervention starts is not prevalent in CSR operations, as only 12 out of 45 companies indicated the use of baseline studies for their CSR initiatives.Please note that this observation is based on an analysis of CSR policies.This statement does not make this claim for the overall operations of all 45 companies.While conducting interviews with CSR managers, we found that in a number of these cases, the indicators used in the baseline surveys were not reproduced in the program design framework; this rendered the baseline information rather meaningless, because it could not be tracked throughout the project.In fact, given the small size of CSR teams, baseline studies are implemented mostly by third parties.This adds to the cost of program implementation, and in turn explains companies' reluctance to invest in baseline information.

M&E Practices
M&E is not merely an activity to monitor and assess the effectiveness and impact of programs; it also exhibits the implementers' commitment toward achieving desired goals.The absence of M&E strategies in the program design also raises doubts regarding the prospects of the program's sustainability.Such apprehensions become even more important when it comes to CSR, which has been mandated by the government in the form of an act.
Discussions with CSR managers revealed that the understanding of M&E varies across companies.Some do not distinguish between 'monitoring' and 'evaluation'; others considered the involvement of a third party in the auditing of CSR finances to be equivalent to third-party evaluation.We also observed that many CSR interventions go through mid-term and end-term evaluation, which they label IE, whereas in reality, these evaluative exercises are limited to verifying the physical progress and outputs of the interventions.Companies are clearly more concerned with outputs, as they are generally eager to grasp what has happened to their investment as quickly as possible.This is further triggered by the yearly reporting toward their governing board, and the national government showing the compliance of mandatory CSR spending.One CSR manager acknowledged that: "We only track the progress on a quarterly basis.Our officers also go and visit the sites, and then, we carry out the financial audit of the implementing partner on a yearly basis".(Source: Interview with a CSR official of a company) Since the time horizon available for companies to utilize the money and report to competent authorities is one year, most of the time-only outputs get reported, as they are the most easily available piece of information for CSR reporting.However, a one-year time horizon is not sufficient to observe the impacts of a program that normally has a much longer gestation period.According to a senior CSR official: "Since companies don't have interest or much stake in CSR, they don't want to measure the impact of their interventions.They are just concerned with the outputs, not with outcomes".(Source: Interview with a CSR official of a company) Adding to the above, there is also a capacity gap among CSR functionaries, as they do not clearly understand how to effectively monitor and measure outcomes.Since CSR functionaries do not have relevant knowledge about these aspects, they are often not able to effectively use evaluation findings, even when they have access to the evaluation reports of their programs.For example, an interpretation of IE reports (impact, mid-term, and process evaluation) authored by academic institutions using state-of-the-art evaluation methodologies, is a challenge for CSR departments.The lack of appreciation for evaluation findings affects program effectiveness, because findings do not take the forms of learning and remedial actions.Another potential use of evaluation results is policy advocacy in the social development sector.If the program was, for instance, implemented at a very large scale, the evaluation results could feed into suggestions for policy changes.However, we found that companies generally resist the advocacy of social development policy, as their primary interest lies in advocacy for business policies.

Discussion
Our analysis provides us with three broad findings.First, companies do not conceptualize sustainability specific to their business operations; rather, they present a rhetorical and standard version of sustainability that remains more or less similar across companies.This finding is supported by a study carried out on Chinese and Korean companies [59].As companies are engaged in different business operations with varied implications for socio-economic and environmental dimensions, a case for distinct approaches to corporate sustainability is made.Second, CSR policies also follow the same trend of rhetorical language of generating 'positive impact' of CSR initiatives on people and creating 'shared values', irrespective of the heterogeneous needs of resources to run the business operations and development needs of communities within which they operate.CSR policies generally do not present a vision, approach, or framework that would ensure that promises are fulfilled.Third, despite the promises made in sustainability reports and CSR policies, efforts to integrate these promises into an implementation of CSR operations are missing.
It is a positive sign that companies have started thinking about integrating socio-economic and environmental sustainability with their business operations.However, the dilemma lies with the word 'social' in the realm of sustainability as well as social responsibility [27].We do not find any sustained efforts by the companies to achieve social sustainability goals in the extant conceptualization of sustainability.While promises have been made in the sustainability reports, there is a dearth of strategies to achieve the same, as frameworks or implementation plans are largely missing.One of the reasons could be that among the four dimensions of sustainability, environmental sustainability has received much attention, while social sustainability is generally the weakest component.While several frameworks and measures have been developed to measure environmental sustainability, other elements of sustainability have not been addressed sufficiently [60].In addition to that, environmental sustainability initiatives are monitored by several regulatory mechanisms and compliances that companies have to fulfill in order to continue to operate.Therefore, the companies are observed as being more responsive to environmental sustainability.
Other elements of sustainability, especially the social dimensions, are essentially voluntary in nature; hence, no sustainability infrastructure is available to help achieve it.Companies, as our analysis points out, are not trying to develop one either.A careful analysis of sustainability reports indicates that more attention has been paid to deal with external factors in the conceptualization of sustainability.The external factors are related to changes that take place in the economic and regulatory environments and affect the business outputs of the companies.Therefore, companies are more prompt in indicating a need to efficiently deal with these external factors.
In order to compensate for the 'social' deficit in sustainability, the term 'social' is hyphenated with 'economic', suggesting that both are given similar attention.However, both are different domains, and both have not been addressed sufficiently in the current measures that are intended to ensure sustainability.The language that is adopted to conceptualize sustainability among the business community fails to clearly distinguish between these two domains of development.This trend is visible in the sustainability reports when they report using GRI standard disclosers.One of the GRI indicators under the social category (referred as G4-SO1 in GRI manual) requires companies to report on their social impact assessment, including gender impact assessments, environment impact assessments, ongoing monitoring processes, and a public disclosure of the results of these impact assessments (Source: GRI Implementation Manual, Available at: https://www.globalreporting.org/resourcelibrary/GRIG4-Part2-Implementation-Manual.pdf).However, most of the companies failed to provide specific details on these aspects.Companies do mention the environmental assessment and social impact assessment in the sustainability reports, but are almost silent on the monitoring processes and other indicators.The social impact assessments mentioned in sustainability reports are usually not the assessments of their CSR programs; these assessments are carried out to fulfill regulatory requirements to initiate new business operations or continue with the existing ones.
The inability to address the 'social' element of sustainability is also reflected in CSR policies.Although companies promise to create 'shared values' and have a 'positive impact' on communities, CSR policies fail to provide a blueprint as to how these promises will be achieved.A study carried out by the Harvard Business School showed that although it is promised, creating 'shared values' is not the reality.The study notes that "most companies practice a multifaceted version of CSR that runs the gamut from pure philanthropy to environmental sustainability to the active pursuit of shared value" [61].CSR policies do not present a clear view of social responsibility of the companies, and indicate a lack of 'willingness' for the same.Since having a CSR policy was made mandatory by the Companies Act 2013, this compulsion is quite clearly visible in the contents of the policies.The length of very few CSR policies exceeds six pages, with the majority of them limited to just two or three pages.Such reports meet the bare minimum criteria to comply with the act.Despite CSR policies making promises about systematic change and community impact, most companies only sporadically invest in CSR initiatives.CSR investments are generally small in size, and are one-time investments, which does not require any intense involvement from the company; as such, outcomes are neither monitored nor evaluated.Although such investments create a positive image of the company among the communities, they do not have the potential to build a sustainable society.
Our analysis also shows a lack of 'readiness' in CSR operations in regard to monitoring and evaluating the outcomes.Having a well-articulated program design of CSR initiatives not only helps in monitoring and measuring outcomes, it also presents a framework for change that the CSR initiatives intend to bring about in the society, as well as a framework of accountability of CSR operations.Lacking such a framework not only restricts the possibility of measuring the impact of CSR interventions, it also weakens the possibility of signaling the sustainability performance of the company to the outside world.Social change is not an overnight process, as it generally takes a considerable amount of time to materialize.CSR interventions do not focus on impact and social change, as these often require a considerable amount of financial investment and time to fructify, whereas companies are interested in immediate outputs.

Conclusions
The rationale behind legislated CSR is that companies must spend a minimum portion of their economic profits on societal well-being, whose resources it utilizes to generate surplus.A firm's ability to adhere to this aspect of the social contract ensures its own sustainability, leads to the creation of shared value (or trust) with catchment communities, and minimizes conflicted situations.Despite such anticipated benefits, the majority of Indian firms choose to invest their profits to enhance their market presence and expand their businesses instead of ensuring higher levels of social welfare.These observations have been supported by other studies as well [62].These types of firms often do not have a long-term vision for their CSR programs.Consequently, they do not find value in monitoring and measuring the CSR outcomes.This also explains the lack of capacity among the CSR professionals to design program-specific M&E frameworks and evaluation strategies that can assist them to improve the social returns from CSR spending.However, if companies are committed to ensuring positive social change from their CSR interventions, they will have to make long-term investments and simultaneously become accountable to the communities.Such a commitment will get reflected in the design of firm's CSR interventions and in the manner in which they measure and monitor these initiatives.
The findings of this study call upon the government to introduce additional provisions in the Companies Act 2013 with respect to adherence to standard M&E practices, and the adoption of robust evaluation designs in order to establish the ground-level impacts of CSR interventions.This is likely to ensure corporate accountability toward the communities in which the firms operate.The findings also provide a strong signal to global CSR reporting and disclosure initiatives such as GRI to further strengthen sustainability reporting frameworks in such a manner that firms, specifically in developing countries, begin to realize the importance of articulating sustainability frameworks that are responsive to their businesses and linking it with their CSR programs.
This study attempts to explore the elements of the continuum of the social dimension of corporate sustainability and CSR policies in the Indian context.However, we do realize the limitation of the study on account of the absence of evidence to claim an external validity of study findings.Specifically, it is not possible to establish whether study findings apply to conditions that are outside the legislated environment.Future research work should seek to establish causal-type linkages between a firm's M&E and evaluation routines and its sustainability performance.This will provide for necessary and sufficient conditions under which both policymakers as well as corporations will see the utility of not only designing and implementing the CSR programs by using firm-specific sustainability frameworks, but also devoting effort and resources to rendering such initiatives amenable to M&E and evaluation processes.

Figure 1 .
Figure 1.Conceptual Approach toward Linking Sustainability with Corporate Social Responsibility (CSR).

Figure 1 .
Figure 1.Conceptual Approach toward Linking Sustainability with Corporate Social Responsibility (CSR).

Figure 2 .
Figure 2. Text Network of the Conceptualization of Sustainability among Indian Companies.Legends: The size of the nodes depicts the degree centrality of the nodes, and the color shows different sub-networks or communities of text.

Figure 2 .
Figure 2. Text Network of the Conceptualization of Sustainability among Indian Companies.Legends: The size of the nodes depicts the degree centrality of the nodes, and the color shows different sub-networks or communities of text.

Figure 3 .
Figure 3.Text Network of CSR Policies.Legends: The size of the nodes depicts the degree centrality of the nodes, and color shows the different sub-networks or communities of text.

Figure 3 .
Figure 3.Text Network of CSR Policies.Legends: The size of the nodes depicts the degree centrality of the nodes, and color shows the different sub-networks or communities of text.

Figure 4 .
Figure 4. Analysis of CSR Policies from M&E and IE Perspectives.

Figure 4 .
Figure 4. Analysis of CSR Policies from M&E and IE Perspectives.

Table 1 .
Network Measures for Text Analysis.

Table 2 .
Text Network Measures and Observations from Sustainability Reports.

Table 3 .
Network Measures for CSR Policies.

Table 3 .
Network Measures for CSR Policies.
Social, enhance, environment Most influential words in the network Community, society, development, sustainable

Table 4 .
Criteria Adopted for CSR Policy Analysis.

Table 4 .
Criteria Adopted for CSR Policy Analysis Ranges from no mention of any M&E system/process/framework to proposing monitoring by the internal CSR committee Ranges from no mention of any IE system/process/framework to proposing internal assessments 1-2 Proposing an M&E system Proposing to assess the impact/outcomes 2-3 M&E system with either needs assessment or baseline study IE with third party 3-4 M&E with both needs assessment and baseline study IE with control group and/or pre and post-intervention assessment Source: Authors' own criteria.