Public Policy and Private CSR Activities: Complements or Substitutes?

Transcription

1 Public Policy and Private CSR Activities: Complements or Substitutes? Mitrokostas Evangelos University of Crete Petrakis Emmanuel University of Crete June 2007 Abstract The present paper examines the conditions under which the regulator can complement the provision of Corporate Social Responsibility (CSR) activities by private firms in an oligopolistic market, in a vertical product differentiation context. Firms alternative strategies are either to engage in CSR or not. The regulator s set of decisions includes no intervention, or certification of firms that engage in CSR. Our main finding is that if there is no credible information disclosure about SR characteristics of the firms products to consumers, no firm will have incentives to undertake CSR effort in equilibrium. However, if the regulator intervenes by providing a certification to the firms that comply with a certain standard of CSR effort, which endows consumers with credible information about the CSR aspects of each firm s product, otherwise unobservable, then both firms will have incentives to engage in CSR activities. Hence in equilibrium, consumers surplus, firms profits and total welfare increase comparing to the benchmark case without CSR activities. JEL Classification: M14, L13, L5. Keywords: Corporate Social Responsibility, Oligopoly, Certification. Department of Economics, University of Crete {mitrokostase@ermis.soc.uoc.gr}. Corresponding author. Department of Economics, University of Crete, Univ. Campus at Gallos, Rethymnon 74100, Greece, Tel: , Fax: {petrakis@ermis.soc.uoc.gr}. 1

2 1 Introduction The large publicity on Corporate Social Responsibility (CSR hereafter) over the last few years has lead many companies to account for the social consequences of their activities. As a result CSR has emerged as a prime issue among firms, exploiting ways to benefit society,and at the same time, benefit from this new challenge. 1 Following the terminology of Porter & Kramer (2006), potential firms benefits engaging in CSR actions may be moral obligation, sustainability, license to operate and reputation. 2 For these benefits to be effective, firms have to convince potential consumers about their social orientation. However, the prevailing approaches to CSR by firms are in most cases fragmented and inconsistent. More specifically, firms try to promote their CSR activities via public relations, media campaigns and glossy CSR reports, which are characterized by restrictions to anecdotal evidence, the lack of the existence of measurable targets and their impact to society (Porter & Kramer, 2006). Moreover CSR effort by firms may include actions within their value chain, which are difficult to be observed by a large scope of consumers. The aforementioned stylized facts reveal that there exists a large confusion among socially conscious consumers over CSR, which may prevent firms from engaging in activities that promote social values. 3 Hence, if firms fail to persuade consumers about their true commitment to social values, they will have no incentives to undertake any CSR activities and a market of lemons problem underlies. Given this evidence, the question that arises is the following: Which are the policy instruments that a regulator can employ in order to increase social awareness among consumers regarding CSR, and what are their effects on market outcomes and social welfare? The present paper addresses and formalizes this question in an oligopolistic market for a final good, where consumers differ with respect to their valuation towards CSR activities. The basic idea behind our model is that firms strategically engage in CSR activities in order to create a "socially friendly image" for their product. We consider that consumers are homogeneous regarding the physical characteristics of the goods, but heterogeneous towards 1 More than half of the top 100 corporations in the 16 more industrialized countries published a CSR report in the year 2005 (Becchetti et al., 2006). 2 For instance, Baron (2001, 2003), Bagnoli & Watts (2003) and Manasakis et. al (2006) under the scope of strategic CSR, formalize situations where firms create a socially friendly image in order to obtain competitive advantage in the market in which they operate. 3 Empirical evidence show that most British consumers do not trust firms reports regarding their CSR practices (see Fliess et al., 2007). 2

3 the valuation of the CSR aspects of each product. More socially conscious consumers have higher valuation for the product of the firm that engages in CSR activities, hence, they are willing to pay a higher price for the "socially friendly" good. 4 On the other hand, CSR may include costly actions by the firm, such as to operate in the interests of employees (investing in workplace safety), suppliers (by supporting local suppliers rather than cheaper alternative sources), and the environment (by reducing emissions of pollutants). 5 Given the above, firms will have incentives to engage in CSR activities only if they know that there is a sufficient mass of consumers that will respond positively to CSR related products. Since CSR is defined as: firms commitment to social and ecological considerations, beyond the law requirements there cannot be any command and control measures, such as compulsory CSR standards in order to impose socially conscious behavior by firms. We thus consider certification as a policy instrument, i.e. a regulator sets certain social and environmental criteria that should be respected during the firm s operational activities and then provides certification to any firm that fulfills these criteria. 6 Following Bottega & De Freitas (2006) we consider the case where the certification is an effective system of information disclosure that permits consumers to distinguish the social characteristics of the products they purchase. 7 Unlike the present paper, the vast majority of the literature on quality certification is based on the seminal paper by Gabszewich and Thise (1979) and concentrates on oligopolistic models in which firms products differ only in their vertical quality characteristics, which are observable by consumers. 8 Our envisaged duopolistic market follows Häckner (2000) along with Garella and Petrakis (2005), using a utility function that combines horizontal and verti- 4 Becchetti et al. (2005) quote the "2003 Corporate Social Responsibility Survey". The main finding of this survey is that the amount of consumers that are socially concerned on their purchasing choices was 62% in 2001 in Europe. 5 See for example Mayer (1999) and Bris & Brisley (2006). 6 Such certifications already exist, however they are still on a primitive stage and are also restricted on fragmented aspects of CSR. For example, the certification SA8000 is specialized in the workers human rights in developing countries ( Page.viewPage&pageId=473. Date last visited: March 2, 2007). Additionally, ISO will certificate SR activities by firms starting from 2008 ( /2000/2122/830949/ / /07 _gen_info/about.html. Date last visited: May 2, 2007). 7 This assumption is in line with recent empirical evidence, according which, EU citisents better trust a certification labeled on the product, comparing to other forms of information about the social characteristics of the products they purchase (see Fliess et al., 2007). 8 Other papers (Crampes and Hollander, 1995; Ecchia and Lambertini, 1997; Maxwell, 1998; Scarpa, 1998; Valletti, 2000; Jinji and Toshimitsu, 2004), explore the issues of consumers and firms gains and losses, and of quality changes, always based on models of pure vertical differentiation. 3

4 cal differentiation aspects of firms products. The vertical differentiation represents the CSR aspects of the production process that are perceived as quality improvement of the final product by socially conscious consumers. Additionaly, since CSR is considered as an experience good it is assumed thar there is no ex ante mechanism that can credibly inform consumers about the CSR characteristics of each product. However in the aforementioned literature the cost to increase quality is assumed to be zero, or fixed. The present paper contributes on this branch of the literature assuming that engaging in CSR increases variable costs. In this context, firms alternative strategies are either to engage in CSR or not. The regulator s set of decisions includes no intervention, or certification of firms that engage in CSR. We investigate two possible scenarios. The first refers to the unregulated equilibrium, assuming that there is an appropriate system of information disclosure that permits consumers to distinguish the social characteristics of the products they purchase, without the need for a policy intervention. We find that in this case both firms endogenous choice will be to engage in CSR, seeking for a competitive advantage in the output competition stage, hence higher equilibrium profits comparing to the benchmark case without CSR activities. The above interaction among competing firms, increases total welfare comparing to the benchmark case, also. However, if one loosens the assumption of credible information disclosure about SR characteristics of the firms products to consumers, firms will fail to persuade consumers about their true commitment to social values, thus, a market of lemons problem arises. More specifically, once consumers have been convinced that one firm has undertaken positive CSR effort, they increase their willingness to pay for the firm s good. The firm has no incentives to spend on CSR activities as these are costly for the firm. Consumers realize the firm s incentives and thus rationally believe that there will be zero CSR activity. The firm, in turn, spends zero on CSR activities in equilibrium. The second scenario refers to the case in which the regulator intervenes, in order to solve the ensuing market of lemons problem, by proposing a certain standard of CSR effort to the firms, and provides a certification to the firms that comply with the standard. This certification endows consumers with credible information about the CSR aspects of each firm s product, otherwise unobservable. Our main finding is that the regulator will set a standard of positive CSR effort up to a level in which both firmswillhaveincentivestocomply. Hence in equilibrium, consumers surplus, firms profits and total welfare increase comparing to the benchmark case without CSR activities. 4

5 Our findings contribute to the existing literature on "strategic CSR", a term that was introduced by Baron (2001) and refers to the case where firms are assumed to be socially responsible because they anticipate a benefit from such a behavior. Baron (2001, 2003) examines CSR under the prism of the strategic choice between public and private politics. His main finding is that private politics and CSR affect the strategic position of a firm in an industry under the existence of activist consumers, who can boycott firms with non-socially friendly behavior. In the same vein, Calveras et al. (2006), assuming a perfectly competitive supply of inputs, compare the effects of formal regulation to firms incentives to provide socially friendly goods as a response to increased activism on behalf of consumers. They argue that substituting formal regulation with firms CSR actions may cause inefficiency, in which non activist consumers free-ride the willingness to pay of activist consumers, lowering formal regulation. Unlike the present paper, the above literature focuses on the difference between the provision of CSR by private firms and the regulator. We concentrate in ways that the regulator can complement the provision of CSR by private firms. The rest of this paper is organized as follows: Section 2 presents the model. In Section 3, the different scenarios are solved and a detailed equilibrium analysis is conducted. Section 4 offers some concluding remarks. 2 The Model We consider a market that consists of two firms, denoted by i, j =1, 2,i 6= j, each producing one brand of a differentiated good. On the demand side, there is a unit mass of consumers composed by individuals who have identical preferences regarding the physical characteristics of the goods. They are, however, heterogeneous regarding their valuation of the CSR activities that are undertaken by the firm that produces the good. In particular, following Häckner (2000), the utility function of the θ-type consumer is given by: U =(a + θs i )x i (θ)+(a + θs j )x j (θ) [x 2 i (θ)+x 2 j(θ)+2γx i (θ)x j (θ)]/2+m (1) where x i (θ), i=1, 2, represents the quantity of good i bought by the consumer of type θ and m is the respective quantity of the composite good. The parameter γ [0, 1] is a measure of the degree of substitutability among goods, with γ =0corresponding to the case of independent goods and γ =1to that of homogeneous goods. Further, s i 0 represents the CSR effort 5

6 that firm i undertakes which, in turn, increases θ-type consumer s valuation for its good by θs i. In other words, θ represents the increase of θ-type consumer s willingness to pay for the firm i s good per unit of CSR effort undertaken by firm i. Thus, the more socially conscious a consumer is, the higher is its θ. While a consumer who does not value the firms CSR activities at all is of type θ =0. We assume that θ is distributed according to a cumulative distribution function F (θ), with density function f(θ), where θ [0, 1]. Thus, θ = R 1 0 θf(θ)dθ represents the average type of consumer in the population. Maximization of utility (1) with respect to x i (θ) and x j (θ) gives the (inverse) demand functions for the θ-type consumer: p i = a + θs i x i (θ) γx j (θ), i =1, 2 (2) where p i and p j are the firms unit prices, while the price of the composite good has been normalized to unity. By inverting (2) we obtain the θ-type consumer s demand for good i: x i (θ) = a(1 γ)+θ(s i γs j ) p i + γp j 1 γ 2 (3) By integrating (3) with respect to θ, we get firm i s aggregate demand function: q i (p i,p j )= Z 1 0 x i (θ)f(θ)dθ = a(1 γ)+ θ(s i γs j ) p i + γp j 1 γ 2 (4) Finally, by inverting (4), we obtain the firm i s (inverse) aggregate demand function: p i (q i,q j )=a + θs i q i γq j,i=1, 2,i6= j (5) Observe that the aggregate demand function corresponds to the demand function of an average type consumer, θ. We assume that both firms are endowed with identical constant returns to scale production technologies. Firm i s total cost function is given by C i (q i,s i )=c(1 + s 2 i )q i. This implies that, for a given CSR effort s i, the firm i s marginal (and average) production cost is constant and equal to c(1 + s 2 i ). Yet, a higher CSR effort increases, at an increasing rate, firm i s unit production costs. This can be justified on the grounds that an individual firm s level of CSR activities, such as improving working conditions for employees, buying more expensive 6

7 inputs from local suppliers, financing recycling and other SR campaigns or introducing green technologies, has an increasingly negative impact on the firm s unit production costs. Firm i s profits can then be expressed as: Π i =(a + θs i q i γq j )q i c(1 + s 2 i )q i (6) Therefore, CSR activities by firm i lead to higher consumers valuation for its product and thus to higher aggregate demand for the firm, but, at the same time, they increase firm i s unit and total production costs. Note however that firms CSR efforts may not be observable, and even in case they are observable, they are not verifiable in the court. Hence, there is a lemons problem in our setup. Once consumers have been convinced that firm i has undertaken a CSR effort s i, and have thus increased their willingness to pay for the firm s good, the firm has no incentives to spend on CSR activities as these are costly for the firm. Consumers realize the firm s incentives and thus rationally believe that there will be zero CSR activity. The firm, in turn, spends zero on CSR activities in equilibrium. To solve for the ensuing lemons problem, we evoke the literature on certification. More specifically, we consider the case in which the regulator intervenes, in order to solve the ensuing market of lemons problem, by proposing a certain standard of CSR effort to the firms, and provides a certification to the firms that comply with the standard. This certification endows consumers with credible information about the CSR aspects of each firm s product, otherwise unobservable. 2.1 The Benchmark case without CSR activities. Before proceeding to the equilibrium analysis, we briefly discuss the benchmark case where no owner engages in CSR and thus s 1 = s 2 = 0. This is a standard Cournot game with differentiated goods, where each owner chooses its output to maximize profits, Π i =(a q i γq j )q i cq i. From the first order condition, the reaction function of owner i is, q i = R C i (q j )= a γq j c 2 (7) 7

8 By symmetry, the equilibrium output, price and profits are, respectively, q C = a c 2+γ ; pc = a +(1+γ)c ; π C = 2+γ (a c)2 (2 + γ) 2 (8) Finally, since all consumers have identical preferences over the physical characteristics of the two goods and there is a unit mass of them in the population, it turns out that each consumer buys a quantity x C = q C from each good. Using (1) and (8), it can be checked that total welfare is given by TW C =(q C ) 2 (3 + γ). 3 Equilibrium Analysis. 3.1 Unregulated equilibrium. The aim of this section is to show that if firms could credibly inform consumers about their CSR effort, then both firms endogenous choice is to engage in CSR. However, in the absence of a reliable information link between firms and consumers, no firm will engage in CSR in equilibrium. Therefore, we begin our analysis by assuming that there is an appropriate system of information disclosure that endows consumers with the necessary information about the CSR characteristics of the products they purchase. We consider a two stage game. In the first stage both firms owners decide whether or not they will engage in CSR activities, while in the second stage they compete in the market a lá Cournot. We solve the game backwards. Hence, in the last stage of the game, owner i sets q i to maximize his firm s profits (6), taking as given the output q j of his rival, along with the CSR efforts, (s i,s j ), chosen in the previous stage. The first order condition (foc) of (6) leads to firm i s reaction function: q i = Ri SR (q j )= a c γq j 2 + θs i cs 2 i 2 (9) Comparing Ri SR (q j ) to the benchmark case with no CSR activities Ri C(q j), in which only the first term of the RHS of (9) appears, we observe that CSR effort has two opposing effects on owner i s output decision. On the one hand, CSR effort s i augments the demand for the firm i s good and thus tends to increase equilibrium output. On the other hand, it increases firm i s unit costs, tending to decrease equilibrium output. Now if s i < θ/c the first effect is 8

9 dominant and the CSR effort undertaken by firm i in the previous stage, since it shifts firm i s reaction function outwards. 9 Solving the system of focs (9), we obtain the equilibrium output: q SR i (s i,s j )= a(2 γ)+ θ(2s i γs j ) c[2(1 + s 2 i ) γ(1 + s2 j )] (4 γ 2 ) (10) Observe that firm i s equilibrium output increases with s i, while it decreases with s j. 10 The higher the firm i s CSR effort s i is, the higher will be the firm i s output. This is so because the owner then earns higher profits per unit of output produced by firm i. On the other hand, when the rival owner sets a higher output for firm j, firm i s owner optimally reacts by reducing its output (due to the strategic substitutability of decision variables). A similar reasoning applies when firm j s CSR effort becomes higher, in which case its owner has incentives to increase firm j s output because he earns higher profits per unit of output produced. In the first stage, each owner i chooses CSR effort s i to maximize his firm s profits, which from the focs of (6) is given by, PRi SR (s i,s j )=[qi SR (s i,s j )] 2. The foc of the latter is equivalent to qi SR (.)/ s i =0. Due to symmetry, the equilibrium CSR effort is: s SR = s SR i = s SR j = θ 2c > 0 (11) Clearly, the CSR effort increases with the social consciousness of the average consumer type θ, while it decreases with the degree of inefficiency of the CSR production technology (as captured by a higher c). Finally, note from (11) both firms owners endogenous choice is to engage in CSR activities. The intuition behind this is that each owner optimally sets a positive level of CSR effort (0 <s SR i < θ/c) that shifts the reaction function of his firm outwards, seeking for a competitive advantage in the output competition stage, hence higher equilibrium profits comparing to the benchmark case without CSR activities. Since positive CSR effort is the optimal response to the competitors choice of CSR effort, any unilateral deviation to zero CSR effort would result the deviant firm to earn lower profits comparing to its previous status. Moreover since both firms selecting zero CSR effort is not an equilibrium, the optimal response to a competitor who sets zero CSR effort is positive CSR effort. The following corollary summarizes: 9 This is in fact the case in equilibrium - see below. 10 Provided that s i,s j < θ - see our discussion above. c 9

10 Corollary 1: In the unregulated market scenario, assuming that consumers can distinguish the CSR characteristics of the products they purchase, both firms owners endogenous choice is to engage in CSR activities. Plugging s SR into(10), (4) and (6) we obtain the equilibrium values for output, price and profits, respectively, q SR = a c 2+γ + θ2 4c(2 + γ) (12) p SR = a + c(1 + γ) 2+γ + θ 2 (3 + γ) 4c(2 + γ) (13) Π SR = " # 2 a c 2+γ + θ2 (14) 4c(2 + γ) Let us now consider the societal effects of owners decisions to engage in CSR. Total welfare is defined as: TW = CS SR net +2Π SR (15) with Π SR and CSnet SR being the overall market profits and net consumers surplus respectively. More specifically, the net consumer surplus of a θ-type consumer is given by the following expression: CS(θ) =(a+θs i )x i (θ)+(a+θs j )x j (θ) (x 2 i (θ)+x 2 j(θ)+2γx i (θ)x j (θ))/2 p i x i (θ) p j x j (θ) (16) In equilibrium, due to symmetry we have s i = s j = ssr and p i = p j = psr. Hence, after some manipulations, eq.(16) and (3) become: CS(θ) =(1+γ)[x (θ)] 2 (17) 10

11 x (θ) = a + θssr p SR 1+γ (18) Hence, from eq.(17), the total net consumers surplus is given by: CS SR net =(1+γ) Z 1 Substituting eq.(18) into (19) and solving gives: 0 [x (θ)] 2 dθ (19) net = 3(a psr ) 2 +3s SR (a p SR )+s SR 2 3(1 + γ) CS SR (20) Plugging eq.(20), (11), (13) and Π SR into (15) one obtains total welfare for the the Unregulated scenario, TW SR. By comparing eq.(12) to eq.(8) we find that for θ 0 then q SR >q C always holds. Hence, in equilibrium, both owners will set output at a level higher than that chosen in the benchmark case. Moreover, from eq.(14) and (8) we obtain that for θ 0 then Π SR >π C always holds. Intuitively, firms profits are affected by two opposing effects. First, since CSR activities are evaluated by consumers positively, these activities increase demand and revenues for the CSR related products. Second, higher CSR effort, increases the firms unit and overall costs. The first effect is dominant in equilibrium and thus profits are higher under Universal CSR. Finally, TW SR >TW C always. The rational behind this result is that according to Corollary 1, it has already been clear that in equilibrium, each firm s owner has a dominant strategy to make positive CSR effort. This interaction among competing firms has a positive effect on total welfare since it increases output, profits and consumers surplus as well. On the other hand, engaging in CSR increases unit cost of production, which decreases total welfare. It is found that the positive effect of increased profits and consumers surplus on total welfare dominates the negative effect of increased costs and thus, TW SR >TW C always. Hence, the following Corollary holds: Corrolary 2: In the unregulated market scenario, assuming that consumers can distinguish the CSR characteristics of the products they purchase, equilibrium output, profits and total welfare are always higher comparing to the benchmark case without CSR activities. However, the above results hold only under the assumption that consumers are fully aware 11

12 about the CSR characteristics of the products they purchase. If one loosens the assumption of credible information disclosure about SR characteristics of the firms products to consumers, firms will fail to persuade consumers about their true commitment to social values, thus, a market of lemons problem arises. In this case no firm will have incentives to undertake CSR effort in equilibrium, and the equilibrium outcomes coincide to the ones observed in the benchmark case without CSR activities. The following Proposition summarizes: Proposition 1: In the unregulated market scenario, in the absence of a reliable information link between firms and consumers regarding the CSR characteristics of firms products, no firm will engage in CSR in equilibrium. 3.2 Certification by the regulator. In this subsection, assuming that there is no appropriate system of information disclosure that endows consumers with the necessary information about the CSR characteristics of the products they purchase, we consider that the regulator proposes a certain standard of CSR effort to the firms, denoted by s R, and provides a certification to the firms that comply with the standard. This certification endows consumers with credible information about the CSR aspects of each firm s product, otherwise unobservable. Each firm s owner can set the CSR effort proposed by the regulator or not engage in CSR activities at all. Each owner may make lower CSR effort than the proposed standard and pretend not to so. Therefore the regulator has to monitor and certify CSR effort made by firms, assuming that, the probability the regulator tracing an owner that reveals untruthful information is almost unity. The fixed cost of monitoring is denoted by M and it is paid by each firm that wishes to be certified. Thus, each firm s owner objective function is now given by the following expression: Π R i =(a + θs i q i γq j )q i c(1 + s 2 i )q i M (21) Where, M =0 inthecasewhichafirm does not engage in CSR, hence, certification is unrequisite. Here we consider a three stage game. In the first stage the regulator fixes a standard of CSR effort s R. In the second stage, given s R,bothfirms owners decide whether or not they will 12

13 engage in CSR activities, while in the last stage owners compete in the market a lá Cournot. We solve the game using backwards induction. Hence, in the last stage of the game, owner i sets q i to maximize his firm s profits now given by eq.(21), taking as given the output q j of his rival, along with the CSR effort s R,chosenby the regulator in the previous stage. Solving the system of focs, and rearranging we obtain the equilibrium output of the second stage: q i (s R )= a c 2+γ + sr ( θ cs R ) 2+γ (22) Plugging eq.(22) into (4) and (21) one obtains firms i s price and profits respectively during stage 2, given by: p i (s R )= a + c(1 + γ) 2+γ + sr [ θ + cs R (1 + γ)] 2+γ (23) Π i (s R )= µ a c 2 + sr ( θ cs R ) 2(a c)+s R 2+γ (2 + γ) 2 ( θ cs R ) M (24) In the second stage both firms decide whether they engage in CSR activities or not. Firms will undertake CSR effort only if their profitability after engaging in CSR, is higher comparing to the benchmark case without CSR activities. Hence, with respect to eq.(24) firms will engage in CSR only if: sr ( θ cs R ) [2(a c)+s R ( θ cs R )] >M,ors R (2+γ) 2 pc θ+ 4c(a c)+ θ 2 4c (a c) 2 +M(2+γ) 2 2c, where s R pc represents firms participation constraint CSR effort. Note that s R pc θ/c for every θ, M 0. Thus, from the analysis of eq.(9), if the above condition holds, the increase in firms profits due to higher demand and revenues from producing CSR related products overcomes the increase in firms costs due to higher CSR effort, and monitoring expenditures, comparing to the benchmark case without CSR activities and therefore, both firms will have incentives to engage in CSR activities. Otherwise owners will have no incentives to comply with the CSR standard. 11 By comparing s R pc to the one obtained in the unregulated equilibrium, we observe that s R pc s SR = θ 2c, for every M 0. Hence firms will now undertake higher CSR effort comparing to their optimal choice under no regulation. In the first stage, the regulator sets CSR effort so as to maximize total welfare now given 11 In this case the prevailing equilibrium coincides with the Benchmark case without CSR. 13

14 by: TW R = CS R net(s R )+2Π i (s R ) (25) Where, CSnet R = 3(a pr ) 2 +3s R (a p R )+s R 2 3(1+γ) represents the net consumers surplus in the certification scenario. By solving the foc and rearranging we obtain the unconstraint minimum CSR effort: s R. 12 Note that if s R s R pc then the regulator will set s R and both firms will comply with the standard. However, if s R >s R pc, thens R does not give incentives to firms to be involved in CSR and the standard is useless. Since the regulators objective is that both firms engage in CSR effortthatwillimprovewelfarehesetss R such that: Thus, the following Corollary holds: s R =min[s R,s R pc] (26) Corollary 3: There exists a minimum CSR effort standard, s R =min[s R,s R pc] > 0, such that both firms owners have incentives to comply with the standard and engage in CSR activities. Observe from eq.(22) and (24) that since s R q R >q C θ+ 4c(a c)+ θ 2 4c (a c) 2 +M(2+γ) 2 2c, then and Π R >π C always holds. Hence, with respect to firms participation constraint output and profits are higher if they comply with the CSR standard, than if they undertake no CSR. By substituting s R, p R i into eq. (20) one obtains the equilibrium value for net consumers surplus in the certification scenario, CSnet. R Comparing CSnetto R the net consumers surplus produced in the benchmark case without CSR activities, CS C we find that CSnet R >CS C always. 13 Hence, in equilibrium, the positive effect of increased profitability and consumers surplus on total welfare dominates the negative effect of increased costs and thus, total welfare in the certification scenario is always higher comparing to the benchmark case without CSR activities (TW R >TW C ). The following Proposition summarizes: Proposition 2: In the certification scenario, given the minimum CSR effort standard, s R = 12 Due to space limits some algebraic formulas are not presented. These are available from the authors upon request. 13 For proof see Appendix. 14

15 min[s R,s R pc] > 0, equilibrium output, profits and total welfare are always higher comparing to the benchmark case without CSR activities. 4 Conclusions The present paper examines the conditions under which the regulator can complement the provision of CSR by private firms. Firms alternative strategies are either to engage in CSR or not. The regulator s set of decisions includes no intervention, or certification of firms that engage in CSR. Our main finding is that if there is no credible information disclosure about SR characteristics of the firms products to consumers, no firm will have incentives to undertake CSR effort in equilibrium. However, if the regulator intervenes by providing a certification to the firms that comply with a certain standard of CSR effort, which endows consumers with credible information about the CSR aspects of each firm s product, otherwise unobservable, then both firms will have incentives to engage in CSR activities. In equilibrium, consumers surplus, firms profits and total welfare increase comparing to the benchmark case without CSR activities. Therefore the introduction of the certificate by the regulator complements firms private incentives for the provision of CSR. The analysis was carried out for a duopolistic market structure. We believe that the duopolistic market provides all essential insights about the firms owners incentives to undertake CSR activities. We are also aware of the limitations of our analysis in assuming specific functional forms. However, it is the nature of the equilibrium conditions that drive our results that allows us to argue that these results will also hold under general demand and cost functions. The use of more general forms would jeopardize the clarity of our findings, without significantly changing their qualitative character. Given the current debate about the market and welfare implications of Corporate Social Responsibility the present paper sheds light on the policy instruments that a regulator may impose, in order to enhance firms incentives to engage in CSR activities in oligopolistic markets. Appendix AppendixA2:Proofofproposition2 Since s R then from eq.(22) and (24) q R >q C θ+ 4c(a c)+ θ 2 4c (a c) 2 +M(2+γ) 2 2c θ 2c, and Π R >π C always holds. 15

16 The net consumers surplus for the benchmark case is given by: CS net = a(q i + q j ) (q 2 i + q 2 j +2γq i q j )/2 p i q i p j q j (27) By imposing symmetry, in equilibrium: q i = q j = q. Hence the demand for the benchmark case P C = a q i γq j, becomes:p C = a (1+γ)q. Inverting gives q = a P C 1+γ.Since in this case all consumers have identical preferences over the physical characteristics of the two goods and there is a unit mass of them in the population, it turns out that each consumer buys a quantity x C = q C.Hencex C = a P C 1+γ. The θ-type consumer s demand for good i in the certification scenario is given by: x(θ) = R [θ+cs R (1+γ)] 2+γ a+θs R P R 1+γ or x(θ) = a+θsr P C s 1+γ. Hence x(θ) x C = sr (θ cs R )(1+γ) 2+γ 0=> x(θ) x C and (1 + γ) R 1 0 [x (θ)] 2 dθ (1 + γ)x C2 for every s R θ/2c, θ [0, 1]. As a result CSnet R CS C will always hold in equilibrium. Hence, since q R >q C, Π R >π C and CSnet R CS C then TW R >TW C. References Bagnoli, M., Watts, S.G., Selling to socially responsible consumers: competition and the private provision of public goods. Journal of Economics and Management Strategy 12, Baron, D.P., Private politics, corporate social responsibility, and integrated strategy. Journal of Economics and Management Strategy 10, Baron, D.P., Private politics. Journal of Economics and Management Strategy 12, Becchetti, L., Giallonardo, L., Tessitore, M.E., Consumer driven market mechanisms to fight inequality: the case of CSR/product differentiation models with asymmetric information. WP 2006/50, ECINEQ. Calveras, A., Ganuza, J.J., Llobet, G., Regulation, corporate social responsibility and activism. Journal of Economics and Management Strategy, forthcoming. European Commission, Promoting a European framework for corporate social responsibility. Green Paper, Directorate-General for Employment and Social Affairs, Unit EMPL/D.1. Garella, P.G., Petrakis, E., Minimum quality standards and consumers information. WP 05-10, University of Crete, Department of Economics. 16

17 Häckner, J., A note on price and quantity competition in differentiated oligopolies. Journal of Economic Theory 93, Fliess, B., Hyung-Jong Lee, Dubreuil, O. L., Agatiello, O., CSR and trade: informing consumers about social and environmental: part I. OECD Trade Policy Working Papers, No. 47, OECD Publishing. doi: / Manasakis, C., E., Mitrokostas, E., Petrakis, Corporate social responsibility in oligopoly. W.P. 0707, University of Crete, Department of Economics. Petrakis, E., E.S., Sartzetakis, and A., Xepapadeas, Environmental information provision as a public policy instrument. Contributions to Economic Analysis & Policy 4 : Iss. 1, Article 14. Porter, M.E., and M.R., Kramer, Strategy and Society: The link between competitive advantage and corporate social responsibility. Harvard Business Review, December, Tietenberg, T., D., Wheeler, Empowering the community: information strategies for pollution control. Paper presented at the Frontiers of Environmental Economics Conference Airlie House, Virginia. 17