Supplier assessment based on corporate social responsibility criteria in Indian automotive and textile industry sectors

Abstract

Mandatory disclosure of corporate social responsibility (CSR) reporting forces supply chain members to consider CSR issues in developed countries. Researchers have viewed CSR content within different themes, such as human (social), ethical and environmental perspectives. So far, in India, very little research has been done to study the impact of human issues in employment (social), ethical issues in business operations and pollution (environmental) issues at both regional and specific sector level. This paper emphasises the perspective view of supplier selection, considering CSR issues as opposed to traditional methods using conventional selection criteria such as cost, quality, delivery and service. It employs six CSR criteria such as discrimination, abuse of human rights, child labour, long working hours, unfair competition and pollution to assess small- and medium-sized suppliers who produce and deliver products to the automotive and textile industry sectors. The results of this study indicate that to prevent discrimination against pay variation, minimise cases of abuse of human rights and incidence of child labour, suppliers must put in place proper policies and procedures regarding employment contract, identification and verification records. Furthermore, to reduce unfair competition, suppliers of both automotive and textile sectors may introduce processes for investigating reports of unethical conducts and take severe disciplinary action. The suppliers of both automotive and textile sectors believe that the competitive advantage can be achieved through environmental oriented business practices. However, they differ in their approaches in attaining this objective.

Keywords:

• corporate social responsibility
• Indian manufacturing sector
• regression analysis
• supplier selection
• small and medium enterprises

1. Introduction

There is global recognition of the importance of corporate social responsibility (CSR) issues in sustainable development (Okafor et al. 2008). CSR has been defined as ‘the voluntary integration, by companies, of social and environmental concerns in their commercial operations and in their relationships with interested parties’ (Graafland and Ven 2006, pp. 111–123). The economic, social and environmental benefits achieved while adopting socially responsible behaviours go beyond the boundaries of a single firm and involve wider communities. To be effective in terms of CSR, companies need all firms in their supply chains to act in a socially responsible manner (Enderle 2004). CSR-focused businesses can proactively promote public interest by encouraging community growth and development and voluntarily eliminating practices that harm the environment (Williams 2003, Emmanuel et al. 2008).

In a recent study, Chapple and Moon (2005) found that adoption and implementation of CSR policies in Asian firms in general and in Indian firms in particular takes place in three stages, with community involvement being the first stage, followed by socially responsible production operations and human relations as the second and third stage. The examples for companies engaging in the first two stages can be seen in some small- and medium-sized Indian firms who contribute their profit to orphanages or use renewable energy sources for production processes. However, it is hard to find companies that have introduced right to work initiatives from a CSR perspective. This has motivated us to consider human (social) issues within a work environment as the first theme of CSR. The human issues cover criteria such as discrimination, abuse of human rights, child labour and long working hours (O'Rourke 2003). As unfair practices are common among small- and medium-scale firms in India, we have taken ethical issue in business operations as the second theme. The criterion of unfair competition represents ethical issue in this study. Recent research on global warming has spread awareness of pollution and environment issues (Visser 2008). This has led us to consider pollution as the third theme (Figure 1).

Figure 1 CSR themes and criteria.

In developing countries, the practices adopted and difficulties experienced by small- and medium-sized enterprises (SMEs) in transferring socially responsible behaviours within their firms to their suppliers have not been extensively analysed (Ciliberti et al. 2008). The assessment of suppliers by manufacturers involves the firms in incorporating diverse management systems and different strategies for organising their supply chains (Ciliberti et al. 2008). Conventionally, suppliers to these manufacturers are expected to supply high-quality products and components at lowest cost and at shortest lead times. However, with the inclusion of CSR issues in supplier selection decision making, firms have started paying attention to CSR issues while managing their supply processes (Beske et al. 2008). This has prompted several firms to adhere to greater environmental responsibility and pursue environmental certification such as ISO 14000. Though some attempts to integrate environmental issues in a conventional supplier selection decision were made (Beske et al. 2008), there have been very limited research conducted to integrate CSR issues in the supplier selection process in the context of India. This study addresses this research gap. It employs six CSR criteria such as discrimination, abuse of human rights, child labour, long working hours, unfair competition and pollution to assess small- and medium-sized suppliers who produce and deliver products to the automotive and textile industry sectors.

The rest of the paper is organised as follows. A review of the literature on CSR-based supplier assessment criteria is given in Section 2. Research hypotheses were developed and discussed in Section 3. This is followed by an overview of the research methodology and results and discussion in Sections 4 and 5, respectively. A brief summary and conclusions are given in Section 6.

2. Literature review

2.1 CSR-related criteria

Theoretical bases for the construction of supplier selection can be derived from transaction-cost economics and the resource-based view of the firm. The underlying premise of transaction-cost economics is that firms are driven by the objective of profit maximisation. So, for supplier assessment, criteria such as the price were given paramount importance (Weber 1990, Weber and Current 1993, Byrne 1998, Gupta and Krishnan 1999, Simpson et al. 2002). Several studies have attempted to identify the underlying dimensions of quality management and to relate them to organisational success. Supply quality and on-time delivery have been consistently identified as two key supplier selection criteria (Weber and Current 1993, Dickson 1996, Karpak and Kasuganti 1999, Kannan and Tan 2002). Furthermore, supplier selection research has also highlighted the importance of service to meet buyer needs. There are two facets to this. First, suppliers should have the ability to meet buyers' demand in a timely manner and respond rapidly to changes in demand. Second, the supplier should be able to do so in a manner consistent with price expectations of the buyer (Kannan and Tan 2002). However, these traditional criteria may not be appropriate in all situations. In the current business environment, CSR-related criteria are becoming critical for the supplier assessment and selection decision-making process. In the following section, we briefly discuss the six CSR-related criteria selected for the purpose this study.

2.1.1 Human issues

Studies suggested that CSR programmes could be used to create strategic benefits for firms such as centrality, specificity, voluntarism and visibility (Burke and Logsdon 1996, Ahmad and Balu 2006, Aguilera et al. 2007, Azadegan et al. 2008). It also provides a structure for firms to obtain other benefits, such as financial performance, fairness to suppliers and customers, corporate reputation, social change, good human relations and inter-organisational learning (Cochran and Wood 1984, Vonderembse and Tracy 1999, Murphy and Poist 2002, Graafland and Ven 2006, Peterson and Anderson 2006, Rodriguez et al. 2006, Dawkins and Ngunjiri 2008).

Human issues in CSR include discrimination, abuse of human rights, long working hours and child labour (O'Rourke 2003). There are laws and policies that protect workers from illegal discrimination and harassment in employment based on race, colour, religion, sex, sexual orientation, marital status, nationality, ancestry, disability, age and so on. Discrimination includes sexual harassment, hate speech, verbal abuse and obscene telephone calls (Jay 2010). Abuse of human rights includes health and safety violations in the workplace, murder, torture and forced displacement at the hands of military and security forces protecting company facilities (Caroline 2008). To prevent child labour, the manufacturing companies and their suppliers should develop and implement policies and procedures to discover the workers who are below minimum age and take corrective actions. There are evidences to believe that by maintaining personal records of the employee, including evidence of date of birth, suppliers can minimise the incidence of child labour (Jay 2010).

2.1.2 Unfair competition

To avoid unfair competition, a company should have in place procedures to prevent corruption, bribery, excessive gift giving, extortion, inappropriate advertising and unethical competitive marketing and non-disclosure of confidential information about customers and suppliers (Emmelhainz and Adams 1999). Transparent procedures for disciplinary action against misconduct help to provide a credible framework. SMEs often reflect the values and principles of the owner/manager. However, only a few enterprises have put their values into a code of conduct, a statement of good business practice and a set of simple rules articulating the company's vision, values, responsibilities and ambitions (Rajagopalan and Zhang 2009).

2.1.3 Pollution

Carter et al. (2002) suggested that issues related to environment, diversity, safety and human rights are some of the criteria that could be captured under the rubric of CSR. Pollution is the introduction of contaminants into an environment that cause instability, disorder, harm or discomfort to the ecosystem, i.e. physical systems or living organisms (Sarkis 1998). Rao and Holt (2005) and Zhu and Sarkis (2004) investigated the relationships between environmentally focused practices and firms' financial performance and found that there is a positive relationship between these two variables.

One of the most widely cited references is the Wal-mart standards of ethical sourcing used to oversee all its contract manufacturers worldwide. Currently, the standards for code of conduct for suppliers include provisions for evaluating health and safety, environment, fair compensation, hours of labour, a seventh day of rest, avoiding forced/prison labour, underage labour and discrimination and ensuring compliance with applicable local laws and regulations, freedom of association and collective bargaining and rights concerning foreign contract workers (Emmelhainz and Adams 1999, Kolk and Tulder 2002, O'Rourke 2003).

Details of the criteria and sub-criteria used in this study are provided in Tables 1 and 2.

Table 1 CSR criteria and sub-criteria (independent variable).

Criteria	Sub-criteria	Definition	Source
Discrimination	A2. Policies against discrimination	Discriminations in hiring age and gender	Charter et al. (2001), Gray et al. (1995), Jay (2010), Kolk and Tulder (2002) and Williams (2003)
	A3. Procedures to prevent discrimination	At the time of recruitment
Abuse of human rights	B1. Signed employment contract	The percentage of workers having a written and signed employment contract	Ahmad and Balu (2006), Caroline (2008), Carter et al. (2002), Dawkins and Ngunjiri (2008), Lindgreen et al. (2009), Kolk and Tulder (2002) and Handler (2005)
	B2. Resignation notice	Procedure allowing workers to resign with no penalty after giving reasonable notice
	B3. Safeguarding	Safeguarding of workers from potentially hazardous machinery
	B4. Disciplinary aspects	Investigation procedures for disciplinary aspects
	B6. Training on handling of hazardous environment	Providing training on handling of hazardous materials
	B7. Written rules	Displaying the rules and regulations
	B8.Warning procedure	Procedure for issuing warning to the worker for breaking the company rules
Child labour	C2. Identification procedure	Procedures for discovering workers below 18 years	Kolk and Tulder (2002) and Williams (2003)
	C3. Verification of employment	Documented procedure to verify that your suppliers do not use child labour
	C4. Record on number of applications below 18	During the last 12 months how many persons below the employment age 18 applied for job
	C5. Evidence for date of birth	Personnel records including evidence of the birth date of each worker
Long working hours	D2. Regular rest days	All the workers have one day rest in a seven-day period	Charter et al. (2001), Jay (2010) and Kolk and Tulder (2002)
D3. Practice of allocating overtime to limited workers	Limited workers will be given overtime (not all the workers)
	D4. Basis on which worker accept overtime	Overtime is given voluntary or only when required
	D5. Legal working hours limit	Legal limit of working hours for that area
Unfair competition	E1. Bribery	Procedures to prevent corruption including bribery	Ahmad and Balu (2006), Choi and Wu (2009), Choi and Kim (2008), Graafland and Ven (2006), Kannan and Tan (2002), Degraeve and Roodhooft (2006) and Sharland et al. (2003)
E2. Advertisement	Procedures to ensure standards of fair business including advertising and competitive marketing
	E3. Non-disclosure of confidential information	Information about customers, channel partners, suppliers, workers and other business partners
	E4. Ethical violation	Procedure for workers to report ethical violations, harsh treatment
	E6. Disciplinary actions	Process for investigating reports of ethical or legal misconduct and for taking disciplinary actions
	E7. How many legal violations	During the last 12 months, how many confidential notifications regarding ethical or legal violations were investigated
	E8. Procedure to avoid bribery	Programme to avoid excessive gift giving, extortion or embezzlement on the part of suppliers
Pollution	F1. Procedure for recycling	Procedure for the management of waste water including monitoring, characterisation and treatment	Lee (2009), Carter et al. (2000), Charter et al. (2001), Handfield et al. (1997), Jay (2010), Rao (2002), Thanaraksakul and Phruksaphanrat (2009), Ueltchy and Tate (2009), Montabon et al. (2007) and Williams (2003)
	F2. Management of solid waste	Solid wastes including collection, separation and disposal
	F3. Environmental permit	Environmental permits and managing competence with the permit requirements
	F4. Liquid waste closed loop recycling	Programme aimed at reducing the amount of waste water and solid waste generated
	F6. Safety training to the workers	Providing safety training to the workers

Table 2 CSR criteria and sub-criteria (dependant variable).

Criteria	Sub-criteria	Definition	Source
Discrimination	A1. Variation in pay	Discriminations in pay	Charter et al. (2001), Gray et al. (1995) and Jay (2010)
Abuse of human rights	B5. Number of abuse cases	During the last 12 months, how many abuse cases have occurred	Ahmad and Balu (2006) and Caroline (2008)
Child labour	C1. Labour policy	Scope of your labour policy	Kolk and Tulder (2002) and Williams (2003)
Long working hours	D1. Limit on overtime	For each worker the limit on overtime	Charter et al. (2001) and Jay (2010)
Unfair competition	E5. Legal misconduct	Programme to protect the identity of workers who report suspected cases of ethical or legal misconduct	Ahmad and Balu (2006), Choi and Wu (2009), Choi and Kim (2008) and Graafland and Ven (2006)
Pollution	F5. Recovery of discarded products	Programme to facilitate recovery of discarded products components or materials at the end of their useful life product	Carter et al. (2000), Charter et al. (2001) and Handfield et al. (1997)

3. Research hypothesis

In this section, we develop hypotheses based on the six CSR criteria selected for this study.

3.1 Discrimination and variation of pay relationship

Labour laws prohibit discrimination and harassment in employment and the workplace. For example, in the USA, laws enforced by the Department of Fair Employment and Housing (DFEH) protect employees from illegal discrimination and harassment based on race, colour, religion, sex (pregnancy or gender), sexual orientation, marital status, national origin (including language use restrictions), ancestry, disability, medical condition (cancer/genetic characteristics), age (40 and above), denial of family, medical care leave and denial of pregnancy disability leave or reasonable accommodation (California Fair Employment and Housing Act Part 2.8).

Previous studies suggested that discrimination at workplace has profound impact on employee pay variation. For example, Berik et al. (2003) found that unequal pay for equal work among employees is prevalent in East Asian countries. In a recent study, concluded that unequal pay between male and female workers in the Chinese industry is a common phenomenon. In this study, we have investigated the relationships between discrimination in terms of policies and procedures against discrimination and variation in pay. These hypotheses are as follows:

Hypothesis H1a	=	Policies against discrimination are positively related to the variation in pay.
Hypothesis H1b	=	Procedures to prevent discrimination are positively related to the variation in pay.

3.2 Relationships between abuse of human rights and abuse cases

The right to life and liberty, freedom of expression, equality, economic, social and cultural rights, including the right to participate in culture, the right to food, the right to work and the right to education are the basic human rights envisioned by the United Nations (http://un.org/en/rights/index.shtml).

The abuse of human rights exists in many forms, from health and safety violations in the workplace to murder, torture and forced displacement at the hands of military and security forces protecting company facilities (Caroline 2008). Allegations of abuse are reviewed for the rights denied, or rights impacted, using the rights laid out in the Universal Declaration of Human Rights and other international bodies such as International Labour Organisation.

For rights protection, workers should have a written and signed transparent employment contract. In addition, they should be subjected to policies and procedures that allow them to resign without penalty, after giving reasonable notice. Similarly, firms should lay out procedures for safeguarding workers from potentially unsafe equipment and conditions. In this study, we investigate the relationships between the forms of abuse of human rights such as the existence of a signed employment contract, resignation notice procedures, safeguarding (safety barriers, interlocks, etc.), disciplinary aspects, training related to the handling and monitoring of hazardous materials and the existence of rules and regulations. We test the following hypotheses:

Hypothesis H2a	=	Existence of signed employment contract is positively related to the number of abuse cases.
Hypothesis H2b	=	Existence of resignation notice is positively related to the number of abuse cases.
Hypothesis H2c	=	Existence of a safeguarding procedure is positively related to the number of abuse cases.
Hypothesis H2d	=	Procedures for disciplinary aspects are positively related to the number of abuse cases.
Hypothesis H2e	=	Training on handling and monitoring of hazardous waste is positively related to the number of abuse cases.
Hypothesis H2f	=	Existence of written rules is positively related to the number of abuse cases.
Hypothesis H2g	=	Existence of warning procedure is directly related to the number of abuse cases.

3.3 Child labour and child labour policies

International Labour Organisation has estimated that 158 million or 19% of children aged between 5 and 14 in the Asia–Pacific region are economically active (ILO 2002, Edmonds 2003). Many are engaged in the worst forms of child labour, in which they are enslaved, forcibly recruited, trafficked, forced into illegal activities or exposed to hazards. Article 32(1) of the UN Rights of the Child (1989) calls for the recognition of the right of children to be protected from economic exploitation and from performing any work that is likely to be hazardous or that would interfere with their education, or be harmful to their health or physical, mental, spiritual, moral or social development.

To prevent child labour, manufacturing industries and their suppliers should follow procedures to identify workers who are below minimum age and take corrective action. They should maintain adequate personal records including evidence of the birth date of each worker. Firms should check whether they have policies to avoid child labour in apprentice programmes and ensure that younger workers are protected from abuse and are not working out of standard daylight hours. To assess the relationships between the existence of child labour and child labour policies, we developed the following hypotheses:

Hypothesis H3a	=	Existence of appropriate labour policy is positively associated with the reduction of child labour.
Hypothesis H3b	=	Existence of proper verification of employment is positively associated with the reduction of child labour.
Hypothesis H3c	=	Keeping organised record on the number of applications below 18 years is positively associated with the reduction record on child labour.
Hypothesis H3d	=	Keeping of date of birth is positively associated with the reduction of child labour.

3.4 Work hour policy and long working hours

Overtime is the amount of time someone works beyond eight hours of normal working hours. Most nations have designed overtime laws to dissuade or prevent employers from forcing their employees to work excessively long hours. In India, there are relevant laws and acts such as the Indian Minimum Wages Act, 1948; Contract Labour (Regulation and Abolition) Act, 1970, and Inter-State Migrant Workmen (Regulation of Employment and Conditions of Services) Act, 1979, to regulate long working hours.

One common approach to regulate overtime is to legislate that workers be paid at a higher hourly rate for overtime work. Overtime laws, attitudes towards overtime and hours of work vary from country to country and between different economic sectors. The exact ways in which long working weeks affect culture, public health and education have been widely debated (Dembe et al. 2005). A review of literature suggested that, generally, overtime depends on policies and procedures relating to rest days, work allocation among limited workers and total working hours. On the basis of these aspects of overtime, we developed the following hypotheses:

Hypothesis H4a	=	Follow-up of regular rest days is positively related to the limit on overtime.
Hypothesis H4b	=	Practice of allocating overtime to limited workers is positively related to the limit on overtime.
Hypothesis H4c	=	Legal working hours limit is positively related to the limit on overtime.
Hypothesis H4d	=	Total working hours is positively related to the limit on overtime.

3.5 Unfair competition

Firms should strictly adhere to procedures to prevent corruption, bribery, excessive gift giving, extortion, unfair business (including some advertising and competitive marketing) and ensure non-disclosure of confidential information about customers and suppliers. Firms should take disciplinary action against legal misconduct in a timely manner. Small and medium businesses often reflect the values and principles of the owner/manager. However, many Indian firms have put their values into a code of conduct in the form of a statement of good business practice, or as a set of simple rules articulating the company's vision, goals and objectives, and responsibilities. We test the following hypotheses:

Hypothesis H5a	=	Bribery in firms is positively related to the procedures to avoid bribery.
Hypothesis H5b	=	Advertisement issued by firms is positively related to the procedures to avoid bribery.
Hypothesis H5c	=	Non-disclosure of confidential information is positively related to the procedures to avoid bribery.
Hypothesis H5d	=	Ethical violation is positively related to the procedures to avoid bribery.
Hypothesis H5e	=	Legal misconduct is positively related to the procedures to avoid bribery.
Hypothesis H5f	=	Disciplinary actions are positively related to the procedures to avoid bribery.
Hypothesis H5g	=	Number of legal violations is positively related to the procedures to avoid bribery.

3.6 Relationships between pollution and recycling practices

Environmental degradation is both a global and a local problem of increasing concern throughout society. Recycling involves processing of the used materials into new products to reduce the waste of potentially useful materials, the consumption of fresh raw materials, energy usage and air and water pollution. The literature suggested that if firms adhere to environmentally oriented practices through recycling can help minimise the level of pollution (Zhu and Sarkis 2004). To test the relationships between pollution and recycling practices, we have developed the following hypotheses:

Hypothesis H6a	=	Management of solid waste is positively associated with the existence of the recycling procedure.
Hypothesis H6b	=	Environmental permit is positively associated with existence of the recycling procedure.
Hypothesis H6c	=	Management of liquid waste is positively related to the procedure of recycling.
Hypothesis H6d	=	Recovery of discarded products is positively related to the procedure of recycling.
Hypothesis H6e	=	Safety training to the workers is positively related to the procedure of recycling.

4. Research methodology

This study was conducted using responses from the small- and medium-sized suppliers to a leading oil seals manufacturer and a textile ancillary parts manufacturer in India who have an annual turnover of INR 1.8 billion and INR 4.0 billion (US$ 1 = INR 45), respectively. The manufacturing and textile sectors are chosen as they have been widely criticised for having sweatshops and, therefore, are potentially risky organisations to violate the three major CSR themes considered in this study (Emmelhainz and Adams 1999).

Suppliers to automotive manufacturers are engaged in production operations such as assembly of rubber and coil spring and the manufacturing operations of oil seals. Textile ancillary suppliers are involved in supplying machinery parts and components to the textile manufacturer. On the basis of the prequalification procedures suggested by Commercial Toolkit Supplier Selection (2009), 36 suppliers of automotive manufacturers and 33 textile ancillary part suppliers have been selected for this study. They were also assessed based on their financial strength and resources, documented skill on previous contracts, overall experience in the industry, as well as experience of the key personnel who execute the work, quality of production and ability to complete the order within a prescribed time and cost.

A questionnaire with 35 questions related to CSR issues was designed and used to collect data. Out of 35 questions, 29 were related to the independent variables and 6 questions were related to the dependent variables. Data were collected through interviewing the top executives of the suppliers, which lasted for about 30 min each. The top managers perception on the CSR-related questions was gathered using a Likert scale of 1–5 (1, strongly disagree and 5, strongly agree). The survey contained other questions relating to general information of the firms including contact information, respondent information, number of employees and annual turnover for the last five financial years between 2003 and 2007.

4.1 Data analysis

Reliability of the six constructs (criteria), in terms of Cronbach's α values, was evaluated by assessing the internal consistency of the items under each construct. The results are shown in Table 3. In the case of the automotive manufacturer suppliers, Cronbach's αs varied between 0.6 and 0.781 and are acceptable (values ≥ 0.6; Hair et al. 2006). In the case of the textile ancillary parts suppliers, values of the α varied between 0.631 and 0.804 and are also within the acceptable level.

Table 3 Factor loading and Cronbach's α for various sub-criteria.

Criteria	Sub-criteria	Automotive manufacturer suppliers	Cronbach's α	Textile ancillary parts suppliers	Cronbach's α
Discrimination	A1	0.778	0.628	0.804	0.742
	A2	0.764		0.915
	A3	0.865		0.815
Abuse of human rights	B1	0.811	0.599	0.651	0.695
	B2	0.672		0.642
	B3	0.776		0.788
	B4	0.818		0.787
	B5	0.832		0.833
	B6	0.432		0.754
	B7	0.653		0.528
	B8	0.611		0.769
Child labour	C1	0.868	0.616	0.908	0.631
	C2	0.798		0.682
	C3	0.655		0.733
	C4	0.856		0.787
Long working hours	D1	0.907	0.664	0.776	0.641
	D2	0.741		0.757
	D3	0.643		0.755
	D4	0.436		0.841
Unfair competition	E1	0.849	0.697	0.647	0.669
	E2	0.62		0.604
	E3	0.847		0.669
	E4	0.776		0.658
	E5	0.889		0.763
	E6	0.763		0.726
	E7	0.692		0.747
Pollution	F1	0.676	0.781	0.762	0.804
	F2	0.829		0.594
	F3	0.467		0.793
	F4	0.651		0.668
	F5	0.882		0.807
	F6	0.484		0.918

5. Results and discussion

Several regression models were developed and analysed. Table 4 reports the results of the analyses for the automotive suppliers and Table 5 reports the results for textile ancillary parts suppliers. These tables include β-coefficient values, t-values and values for the significance level to test the hypotheses. The results are discussed in the following sub-sections.

Table 4 Regression results of automotive manufacturers suppliers.

Criteria	Dependent variable	Independent variable	β	Mean	Standard error	t-Value	Significance level	Hypothesis
Discrimination	A1	A2*	0.669	0.385	0.077	6.601	0.000	H1a: supported
		A3*	0.368	0.592	0.122	2.853	0.007	H1b: supported
Abuse of human rights	B5	B1*	0.561	4.472	0.218	3.596	0.001	H2a: supported
		B2	0.136	1.806	0.138	0.851	0.402	H2b: not supported
		B3	0.034	3.102	0.145	0.189	0.851	H2c: not supported
		B4	0.061	2.822	0.093	0.411	0.684	H2d: not supported
		B6	0.128	0.288	0.104	0.764	0.449	H2e: not supported
		B7	0.194	0.685	0.206	1.317	0.194	H2f: not supported
		B8	0.005	1.801	0.147	0.832	0.410	H2g: not supported
Child labour	C1	C2	0.096	3.722	0.227	0.662	0.513	H3a: not supported
		C3	0.243	3.805	0.876	1.790	0.084	H3b: supported
		C4*	0.724	0.555	0.607	3.579	0.001	H3c: supported
		C5	0.106	0.805	0.575	0.577	0.568	H3d: not supported
Long working hours	D1	D2*	0.523	1.812	0.133	3.408	0.002	H4a: supported
		D3**	0.033	0.750	0.133	0.216	0.830	H4b: supported
		D4	0.309	0.192	0.234	2.229	0.030	H4c: supported
		D5	0.100	0.283	0.269	0.589	0.559	H4d: not supported
Unfair competition	E5	E1	0.090	3.371	0.147	0.720	0.478	H5a: not supported
		E2	0.053	2.914	0.144	0.418	0.679	H5b: not supported
		E3	0.165	2.228	0.162	0.948	0.352	H5c: not supported
		E4	0.165	1.885	0.160	1.339	0.192	H5d: not supported
		E6*	0.975	2.542	0.160	6.204	0.000	H5e: supported
		E7	0.104	4.485	0.165	0.850	0.403	H5f: not supported
		E8	0.282	2.584	0.115	2.503	0.016	H5g: supported
Pollution	F5	F1*	0.686	1.444	0.259	2.872	0.008	H6a: supported
		F2**	0.534	1.083	0.255	1.866	0.070	H6b: supported
		F3	0.054	1.333	0.325	0.166	0.869	H6c: not supported
		F4	0.162	1.666	0.171	0.806	0.427	H6d: not supported
		F6	0.220	4.641	0.165	2.278	0.028	H6e: supported
Notes: *Statistically significant predictor variable in the regression model at α = 5%. **Statistically significant predictor variable in the regression model at α = 10%.

Table 5 Regression results of textile ancillary parts suppliers.

Criteria	Dependent variable	Independent variable	β	Mean	Standard error	t-Value	Significance level	Hypothesis
Discrimination	A1	A2	0.938	0.725	0.061	16.254	0.000	H1a: supported
		A3	0.017	2.000	0.153	1.432	0.162	H1b: supported
Abuse of human rights	B5	B1*	0.612	1.121	0.095	0.507	0.000	H2a: supported
		B2	0.570	2.666	0.045	0.232	0.617	H2b: not supported
		B3	0.036	3.810	0.057	0.187	0.818	H2c: not supported
		B4*	0.078	0.393	0.078	4.985	0.000	H2d: supported
		B6	0.597	3.375	0.218	1.656	0.113	H2e: not supported
		B7	0.340	3.900	0.209	0.482	0.121	H2f: not supported
		B8	0.138	2.475	0.391	0.318	0.752	H2g: not supported
Child labour	C1	C2	0.051	1.242	0.334	1.553	0.853	H3a: not supported
		C3	0.503	1.484	0.357	1.405	0.132	H3b: not supported
		C4	0.367	0.272	0.395	1.239	0.172	H3c: not supported
		C5	0.056	0.303	0.688	0.997	0.813	H3d: not supported
Long working hours	D1	D2	0.848	0.223	0.149	0.816	0.327	H4a: not supported
		D3	0.363	0.191	0.116	0.999	0.421	H4b: not supported
		D4	0.371	0.750	0.413	1.352	0.185	H4c: not supported
		D5	0.659	0.950	0.099	8.019	0.000	H4d: supported
Unfair competition	E5	E1	0.134	0.212	0.139	0.589	0.327	H5a: not supported
		E2	0.079	3.601	0.168	0.299	0.563	H5b: not supported
		E3*	0.173	3.454	0.176	3.110	0.006	H5c: supported
		E4*	0.437	4.181	0.141	1.947	0.005	H5d: supported
		E6**	0.271	3.121	0.162	1.850	0.063	H5e: supported
		E7	0.116	2.121	0.207	0.672	0.403	H5f: not supported
		E8	0.505	4.825	0.129	7.460	0.000	H5g: supported
Pollution	F5	F1	0.154	4.454	0.269	0.358	0.582	H6a: not supported
		F2	0.265	1.484	0.282	0.116	0.276	H6b: not supported
		F3	0.270	1.636	0.303	0.270	0.350	H6c: not supported
		F4*	0.519	1.727	0.174	2.519	0.008	H6d: supported
		F6	0.09	0.875	0.266	1.608	0.118	H6e: not supported
Notes: *Statistically significant predictor variable in the regression model at α = 5%. **Statistically significant predictor variable in the regression model at α = 10%.

5.1 Automotive sector

Out of 29 hypotheses developed and tested for the automotive sector, 12 hypotheses were found to be statistically significant either at level 0.05 or at level 0.10. These hypotheses are H1a and H1b in discriminant sub-criteria; H2a in abuse of human rights sub-criteria; H3b and H3c in child labour sub-criteria; H4a, H4b and H4c in long working hour sub-criteria; H5c, H5e and H5g in unfair competition and H6a and H6e in pollution sub-criteria. The implications of these results are discussed below:

•	The respondents from the automotive suppliers agree that if there are proper policies and procedures in place to prevent verbal abuse, then the employees have the opportunity to showcase their performance. This ultimately yields to a reduction in the variation of pay among employees. This finding will assist manufacturers to choose suppliers who are less likely to discriminate employees on pay.
•	To avoid abuse of human rights, suppliers of the automotive suppliers must have proper employment contracts for all workers. What it means for the manufacturing sector is that while assessing the potential suppliers they must ask for employment contacts for their employees as an indicator of the degree of abuse of human rights at work.
•	The existence of proper verification of employment and record of the number of under-aged applications is critical to the reduction of child labour among the automotive suppliers. Therefore, it is vital that the manufacturers enquire for these documents while making assessment about the suppliers.
•	It is possible for the suppliers to minimise long working hours by two ways. One, by following up of regular rest day and two, by introducing practice of allocating overtime to the limited number of workers, instead of all workers.
•	Unfair competition in the businesses can be reduced by introducing processes to investigate reports of unethical conducts and taking severe disciplinary actions.
•	For competitive advantage, automotive manufacturers are increasingly using the recovered products for their newer vehicles. One way to take advantage of the current business environment is to put in place recycling procedures to manage solid waste.

5.2 Textile sector

The overall analysis of the textile sector reveals that 10 out of 29 hypotheses were supported at either level 0.01 or level 0.10. These include H1a and H1b in discrimination sub-criteria; H2a and H2d in abuse of human rights sub-criteria; H4d in long working hours; H5c, H5d, H5e and H5g in unfair competition and H6d in pollution sub-criteria. None of the hypotheses from the child labour sub-criteria were statistically supported. The results of the analysis are discussed below:

•	The respondents in textile ancillary suppliers believe that the existence of the policies against discrimination and procedure to prevent discrimination will avoid variations in pay. This mandates textile ancillary suppliers to insist on policies and procedures to avoid discrimination in future.
•	In the case of abuse of human rights, textile ancillary suppliers believe that in addition to having proper employment contracts, existence of procedure to monitor disciplinary aspects would reduce abuse cases.
•	With respect to long working hours, textile ancillary suppliers differ from automotive sector suppliers in terms of allocation of overtime to limited workers and regular rest days. This implies that due to unpredicted demand fluctuations, occasionally the suppliers are not in a position to allocate overtime to only limited skilled workers and so vary the rest days.
•	Within unfair criteria, practices indicate that bribery is used as a means to violate certain rules and regulations.
•	Except safety training to workers, other aspects of pollution taken for study such as management of solid waste, environmental permit, management of liquid waste and recovery of discarded product were not supported by the textile sector.

6. Summary and conclusions

Traditionally, criteria such as cost, quality, lead time and service are considered to assess suppliers. However, with the recent change in the business practices, these criteria may not be appropriate. The supplier assessment criteria related to CSR issues are increasingly becoming important. However, very limited research has been conducted to integrate CSR issues in the supplier selection process of the small- and medium-sized suppliers in the context of India. This study addresses this gap. It employs six CSR criteria such as discrimination, abuse of human rights, child labour, long working hours, unfair competition and pollution to assess small- and medium-sized suppliers who produce and deliver products to the automotive and textile industry sectors.

Using a questionnaire, data were collected from the top executives of 36 suppliers in the automotive industry and 33 suppliers of the textile ancillary parts in the textile industry. Overall, 29 hypotheses were developed and tested. Eleven hypotheses in the automotive industry and eight hypotheses in the textile sector were supported by the data. The implications of the results indicate that to prevent discrimination against pay variation, minimise cases of abuse of human rights and incidence of child labour, suppliers must put in place proper policies and procedures regarding employment contract, identification and verification records. Furthermore, to reduce unfair competition, suppliers from both automotive and textile sectors may introduce processes for investigating reports of unethical conducts and take severe disciplinary actions.

The suppliers of both automotive and textile sectors believe that the competitive advantage can be achieved through environmental oriented business practices. However, they differ in their approaches in attaining this. The automotive suppliers would like to put in place recycling procedures to manage solid waste and minimise pollution, whereas the suppliers of textile sector prefer to engage in the closed-loop liquid recycling process. This difference can be explained by their nature of operations processes and types of products.

This study was conducted in the context of automotive and textile sectors in India. In future, this study can be replicated in other industry sectors of similar sized suppliers and in other developing countries. A comparison of the findings from different studies could be benchmarked for best practice.

Acknowledgements

The authors are grateful to the anonymous reviewers and editors for their constructive comments that improved the presentation and content of this paper.

Additional information

Notes on contributors

Venkatesan Baskaran

1

Shams Rahman

2

Notes

1. Email: [email protected]

2. Email: [email protected]