Resource and Energy Economics 35 (2013) 380–395
Contents lists available at SciVerse ScienceDirect
Resource and Energy Economics
journal homepage: www.elsevier.com/locate/ree
Market equilibrium in the presence of green
consumers and responsible firms: A
comparative statics analysis夽
Nicola Doni ∗, Giorgio Ricchiuti
Dipartimento di Scienze per l’Economia e l’Impresa, Università degli Studi di Firenze, via delle Pandette 9,
50127 Firenze, Italy
a r t i c l e
i n f o
Article history:
Received 18 February 2012
Received in revised form 16 April 2013
Accepted 17 April 2013
Available online 27 April 2013
JEL classification:
D62
L13
L21
Keywords:
Green consumers
Corporate social responsibility
Vertical differentiation
a b s t r a c t
This paper analyzes how the interaction between green consumers
and responsible firms affects the market equilibrium. The main
result is that a higher degree of responsibility of consumers and/or
firms may both increase and decrease the total abatement and the
social welfare. In general an increment in the degree of CSR of a firm
entails an increase of its total clean-up and a reduction of the aggregate abatement of its rival. When the rival firm has a high degree of
CSR this second effect is stronger than the first and total abatement
falls down. At the same time, when the degree of consciousness of
consumers and/or firms is very high, responsible firms overprovide
environmental quality: in such case a further increment in the level
of social responsibility of a market actor may trigger an increase of
firms’ total clean-up but a reduction in social welfare.
© 2013 Elsevier B.V. All rights reserved.
1. Introduction
In the last two decades a growing body of literature on environmental economics has been devoted
to the analysis of the so called third generation instruments for pollution control: the classic command
夽
We are grateful to Nikos Georgantzis, Aurora Garcia Gallego and the audience at the conferences EARIE 2010, EAERE 2011
and SIE 2011 for the helpful comments to an early version of this paper. We are also deeply indebted with two anonymous
referees and the editor Prof. D. van Soest for their insightful suggestions that improved our paper. Finally, we acknowledge
Nicola Piana Agostinetti and Fabrizio Cipollini for their technical assistance to the numerical simulations. The usual disclaimer
applies.
∗ Corresponding author. Tel.: +39 0554374597.
E-mail addresses: nicola.doni@unifi.it (N. Doni), giorgio.ricchiuti@unifi.it (G. Ricchiuti).
0928-7655/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.reseneeco.2013.04.003
N. Doni, G. Ricchiuti / Resource and Energy Economics 35 (2013) 380–395
381
and control approach can be substituted, or integrated, not only by economic instruments (i.e. taxes,
subsidies and tradable permits) but also by the voluntary market choices of environmentally aware
agents.1 However, the current debate is far from a complete understanding of the actual capabilities
of both individual and firm responsibility as a mean to effectively promote environmental protection
(see Bénabou and Tirole, 2010).
In many sectors firms adopt labeling schemes in order to increase their market share. As noted by
Kotchen (2005) and Besley and Ghatak (2007), environment-friendly goods can be viewed as impure
public goods, in which private and public characteristics are bundled together. As emphasized by
Bagnoli and Watts (2003), the form of this bundling can be either explicit or implicit. The former corresponds to situations in which firms improve the production of environmental quality by increasing
their marginal costs (e.g. the production of energy from renewable sources rather than from fossil
fuels). The latter corresponds to situations in which firms sustain environmental programs whose
benefits and costs are not proportional to their sales (e.g. fixed investments directed to build a public
park close to the plant).
There is convincing evidence that many consumers are willing to pay a price premium to purchase environment-friendly goods (see Bansal, 2008; Kitzmueller and Shimshack, 2012 for empirical
references). A strand of economic literature models the impact of green consumers on the market equilibrium adopting the framework of a vertically differentiated oligopoly. A first group of papers focuses
on how the presence of green consumers interacts with the optimal environmental policy (see Arora
and Gangopadhyay, 1995; Cremer and Thisse, 1999; Moraga-Gonzalez and Padron-Fumero, 2002;
Bansal and Gangopadhyay, 2003; Lombardini-Riipinen, 2005; Bansal, 2008). A second group deals
with the impact of a higher consumers’ consciousness on the market equilibrium and the associated
social welfare (see Eriksson, 2004; Conrad, 2005). In these models, usually the results warn against
a naive confidence in consumers’ responsibility as a solution to environmental problems. Indeed,
rarely the market equilibrium in the presence of green consumers approximates the maximization of
social welfare. Moreover, some authors show that a higher level of consumers’ responsibility is not
always associated to less pollution and higher welfare2 (Rodriguez-Ibeas, 2007; Garcia-Gallego and
Georgantzis, 2009).
Following Garcia-Gallego and Georgantzis (2009), we study a vertically differentiated oligopoly
where two firms overcomply the existing environmental regulation by producing two differentiated
goods (high vs. low quality), while a competitive fringe produces a good with standard quality. We
believe that this framework is more realistic because it allows the analysis of the simultaneous presence of both clean and dirty firms. As far as the production technology is concerned, we assume that
costs and benefits of the environmental quality are increasing and convex in the clean-up effort and
proportional to the level of production, what usually happens when firms abate the pollution associated to their production process or use a cleaner input: following the terminology of Bagnoli and
Watts (2003), we analyze an explicit bundling.3
As usual, we assume that consumers have a different willingness to pay (hereafter WTP) for “clean”
products and we study how an increase in their aggregate WTP affects the market equilibrium. The
main novelty of our paper is that we allow firms to choose their market strategy in accordance with an
objective function that may not coincide with profit maximization. Indeed, in some markets, especially
when the good traded is an impure public good, we observe competition between firms with different
aims. For example in the Fair Trade sector, standard for-profit firms may compete with non-profit firms,
whose main objective is the maximization of the positive externality associated to their production
(Becchetti and Huybrechts, 2008). At the same time, as happens in the energy sector for many countries,
1
See Khanna (2001), for a survey on this historical evolution.
Similar conclusions are reached in a different framework by Calveras et al. (2007). They consider a model in which citizens
first vote the minimum environmental standard and then buy a good produced in perfectly competitive markets. According to
their analysis, a higher level of activism in the society may imply a higher level of pollution.
3
Many existing models adopt this assumption. See for instance Cremer and Thisse (1999), Eriksson (2004), LombardiniRiipinen (2005), Conrad (2005), Rodriguez-Ibeas (2007) and Bansal (2008). Following Bansal, (2008, p. 347), we believe that
this assumption is empirically more relevant than the alternative and frequently used hypothesis of quality costs independent
of the volume of output. However, for the sake of completeness, Doni and Ricchiuti (2011) analyze also the case of fixed costs.
2
382
N. Doni, G. Ricchiuti / Resource and Energy Economics 35 (2013) 380–395
private for-profit firms may compete with a state-owned non-profit firm (see the mixed oligopoly
analyzed in Cremer et al. (1991)).
We propose a more general approach in which firms maximize a convex combination of two different aims4 : profit and the positive social impact of their production. The inclusion of the latter term
is related to the recent behavior of many firms which spend a lot of effort in order to persuade consumers that their production is socially responsible. Therefore, in this paper we draw on two so far
unrelated strands of literature: the former analyzing the influence of green consumers on the market
equilibrium and the latter exploring the behavior of socially responsible firms.
There is not a general consensus regarding to the exact concept of corporate social responsibility
(CSR). There are two polar definitions that can appear in sharp contrast.5 According to a first point of
view, a firm is socially responsible when it takes environment-friendly actions not required by law.
In this light, CSR can be defined without any regard neither to the motivation of the firm’s choices
nor to the impact of such choices on the firm’s profit. From a different point of view, other authors
believe that a firm is truly responsible only when it sacrifices its profit, at least in part, in order to
carry out some social objective. Baron (2001) names the first behavior as strategic CSR and the second
one as altruistic CSR. This second concept of CSR is quite disputed: according to some authors an
altruistic CSR is neither sustainable in a competitive market nor desirable from a social point of view
(see Reinhardt et al., 2008, and literature quoted therein). Moreover, as reported by Kitzmueller and
Shimshack (2012) ‘quantitative empirical data are not consistent with the hypothesis suggesting that
not-for-profit motivations systematically drive observed CSR’. This is probably why in all the existing
models regarding the influence of green consumers on the market equilibrium, firms are assumed to
behave as standard profit maximizers. Consequently the current literature explores only the effect of
the interaction between green consumers and firms engaged in strategic CSR. However, as suggested
by Lyon and Maxwell (2008), we believe that a theoretical analysis of the impact of altrustic CSR on
social welfare is needed. In this view, we assume that the firms’ degree of CSR can be interpreted as the
relative weight they assign to the positive social impact of their production. Therefore, our purpose is
to study the market equilibrium in the presence of green consumers and firms engaged in altruistic
CSR. More specifically, this work aims at analyzing whether a higher degree of responsibility of either
consumers or producers implies a reduction of pollution and an increase of social welfare.
The main result of our paper is that a higher degree of responsibility of consumers and/or firms
may both increase and decrease the total abatement and social welfare, depending on the degree of
responsibility of each agent. In general, an increment in the degree of a firm’s CSR entails an increase of
its total clean-up and a reduction of the aggregate abatement of its rival. When the rival firm has a high
degree of CSR, this second effect is stronger than the first and total abatement falls down. This is due
to the dynamic of market competition: for instance, when the low quality firm increases its CSR and
the high quality firm is non-profit, the former increases the environmental quality of its good while
the latter reduces it because this is the only way it can reduce its price and limit the loss of its market
share. At the same time, when the degree of responsibility of consumers and/or firms is very high,
firms overprovide environmental quality: as a consequence, a further increment in the level of social
responsibility of a market actor may trigger an increase of the firms’ total clean-up but a reduction in
social welfare.
The remaining part of the paper is organized as follows: in Section 2 we present the general model,
introducing the concepts of green consumers and responsible firms, describing their market interactions. In Section 3 we characterize the market equilibrium. In Section 4 we define and analyze the
social welfare. Section 5 concludes. All proofs and figures can be found in Appendix A.
4
A similar assumption is adopted in other frameworks. See for instance De Donder and Roemer (2009) and Ishibashi and
Kaneko (2008).
5
An interesting debate over this issue can be found in the first volume of the Review of Environmental Economics and Policy.
In particular, see Lyon and Maxwell (2008) and Reinhardt et al. (2008).
N. Doni, G. Ricchiuti / Resource and Energy Economics 35 (2013) 380–395
383
2. The model
2.1. The clean-up technology
There is a physically homogeneous good, whose production generates pollution. The environmental
regulator sets an exogenous maximum level of per unit pollutant at e > 0. However, there is a clean-up
technology that may allow firms to abate pollution below the standard level. We assume that both
total clean-up and its costs are proportional to firm’s production, x. Moreover, the abatement costs are
increasing and convex in the clean-up effort, e and the marginal cost for the standard level is zero.6
Formally, for a generic firm i the total emissions are:
∀i = 1, . . . , n,
Yi (ei , xi ) = (e − ei )xi ,
n
and we define Y =
ated cost function is:
Ci (ei , xi ) =
ei2
2
Y
i=1 i
xi ,
the aggregate emissions and E =
n
ex
i=1 i i
the total abatement. The associ-
∀i = 1, . . . , n.
If all firms share this common technology, its convexity implies that the cost effective allocation of
the abatement activity requires that all firms adopt the same level of clean-up.
2.2. The demand side
On the demand side, there is a unit mass of consumers interested in buying only one unit of the
good. Consumers are environmentally conscious,7 taking the firms’ abatement efforts into account
when they choose which product to buy. A generic green consumer selects a product to maximize the
following utility function:
U = V − p + e − Y,
(1)
where V is the (homogeneous) gross utility of consuming one unit of the product, p is its price, is the individual WTP for the marginal increase in firms’ abatement and is the marginal disutility
associated to the negative externality stemming from aggregate pollution Y.8
The differences in income levels and in the degree of environmental awareness justify the heterogeneity in consumers’ WTP for environmental
improvements. Coherently, we assume that is
uniformly distributed in the interval 0, .9 Therefore the aggregate WTP of green consumers is
equal to /2 and it is assumed to be lower than the aggregate marginal disutility of emissions: ≤ 2.
Finally, we assume that V is higher than e and consequently the production of this good is always
socially efficient.
2.3. The supply side
On the supply side, there are n firms. Imagine that a generic firm i adopts a level of abatement ei ,
charges a price equal to pi and sells a market share xi . Thus its profit is:
i =
6
pi −
ei2
2
xi ,
i = 1, . . . , n.
(2)
Our main results hold even if we allow for strictly positive costs of production.
In the working paper version (see Doni and Ricchiuti, 2011) we consider the possibility that only a fraction ˇ of consumers
population is green, while all the other act as free riders. However, the influence of such parameter on both the aggregate
pollution and the social welfare is similar to the influence of , i.e. the maximum WTP in the population.
8
We assume that each consumer takes the total emissions Y as exogenous because her individual contribution to pollution
is negligible.
9
We focus only on the case in which the lowest WTP is 0 in order to simplify our analysis of market equilibrium. Indeed, the
assumption that = 0 ensures that a fraction of consumers buys a standard good. See Section 2.4 for further details.
7
384
N. Doni, G. Ricchiuti / Resource and Energy Economics 35 (2013) 380–395
n
We assume that every consumer buys the product of one firm; consequently,
x = 1 and
i=1 i
Y = e − E. Following Garcia-Gallego and Georgantzis (2009), we allow the coexistence of two kind
of firms. On the one hand, a fringe of firms that provides the good without employing any clean-up
activity. Given that they sell an homogenous product and compete à la Bertrand, they charge a price
equal to 0 (the marginal cost of production when e = 0), and they do not achieve extra profit. On the
other hand, two firms can choose to employ the cleaning technology in order to differentiate their
product and to obtain a strictly positive profit. These two firms are labeled as responsible because they
overcomply the existing environmental regulation. We use H and L to denote the variables associated
to the firms choosing the high and the low level of abatement.
The assumption regarding the existence of only two responsible producers can be justified by noting
that frequently innovation processes are driven by a limited number of firms. Indeed, as sustained by
Kotchen (2009), environment-friendly innovations are frequently introduced by eco-entrepreneurs
where eco-entrepreneurship can be defined as ‘the practice of starting new businesses in response to an
identified opportunity to earn a profit and provide a positive environmental externality.’
We allow responsible firms to have a different willingness to sacrifice their profit in order to increase
their clean-up; they have the following composite objective function that weighs their profit and the
positive externality associated to their abatement activity:
Ji = (i )1−˛i (ei xi )˛i
s.t. i ≥ 0, i = L, H,
(3)
where ˛i ∈ [0, 1]. When ˛i = 0 the firm is a standard profit-maximizer; when ˛i = 1 it is a non-profit firm
that simply wants to maximize the positive impact of its clean-up under the constraint of non-negative
profit.10
In general, a responsible firm pursues two different objectives simultaneously and ˛ is a parameter
signaling the relative importance of the two criteria. More specifically, ˛ can be interpreted as a
measure of the degree of (altruistic) CSR of the firm.11 Such objective function can be interpreted
as the weighted Nash bargaining solution of an efficient negotiation between two different factions
inside the firm: one aiming at maximizing profit and the other aiming at maximizing the positive
externality associated to firm’s production.12 This interpretation is correct if (i) the no agreement
pay-offs are (0,0), as happens when a firm is part of the competitive fringe, that does not obtain any
extra-profit and does not produce any positive externality, (ii) the objective function of each faction
is log-concave in the firm’s strategic choices of p and e (for our case this property is valid. A formal
proof is available upon request). In this view, ˛ represents the relative negotiation power of the faction
supporting the abatement activity inside the firm.
2.4. Firms’ competition
We model competition between the two responsible firms according to the usual framework
adopted in duopoly models of vertical differentiation. There are two stages: in the first one, the
two firms simultaneously choose the clean-up level that can be defined as the (environmental) quality of their product. In the second stage, the two firms observe the choice of their competitor and
simultaneously set the price.
Usually in vertical differentiation models (i.e. Ecchia and Lambertini, 1998), when the lowest consumers’ WTP is equal to zero, the market is only partially covered: there are some consumers that
do not buy any product. This case cannot be applied to our framework because of the presence of a
competitive fringe. The market is completely covered and, in equilibrium, consumers with a low WTP
buy the standard good.
ˆ the consumer which is indifferent
The market share of each firm can be calculated by identifying ,
˜ the consumer which is indifferent between the
between the high or the low quality product, and ,
10
We assume that when ˛i = 1 firm i maximizes the positive impact of its abatement activity also if its profit is equal to 0.
In what follows, we use ˛H and ˛L to indicate the degree of CSR of the firms producing the high and the low level of clean-up.
However, it is important to emphasize that we do not restrict the relative size of their degree of CSR.
12
De Donder and Roemer (2009) apply a similar objective function to explain the choices of a firm where some managers are
interested only in profit while other managers seek to maximize revenues.
11
N. Doni, G. Ricchiuti / Resource and Energy Economics 35 (2013) 380–395
385
low or the null quality product. Straightforward algebra, using Eq. (1), it is easy to see that: ˆ = (pH −
pL )/(eH − eL ) and ˜ = pL /eL . As known, in a vertically differentiated duopoly, the high (low) quality
ˆ ] ([,
˜ ]).
ˆ Then each firm’s market share is:
firm sells to green consumers included in [,
xH =
1
−
pH − pL
,
(eH − eL )
xL =
1
p − p
H
L
eH − eL
−
pL
,
eL
x0 = 1 − xH − xL ,
(4)
where x0 is the total quantity sold by firms of the competitive fringe.
We apply the standard backward induction methodology by first analyzing the price equilibrium
and then the environmental quality equilibrium.
3. Market equilibrium
Using Eqs. (2) and (3), we compute the first derivatives of Ji with respect to pi , ∀i = H, L, and then,
solving the system, we obtain the following equations for the equilibrium prices:
p∗H =
p∗L =
2 + (1 − ˛ )e2 + 2(1 − ˛ )(2 − ˛ )(e − e )]
eH [(2 − ˛L )eH
H L
H
L
H
L
2[(2 − ˛H )(2 − ˛L )eH − (1 − ˛H )(1 − ˛L )eL ]
,
2 + (2 − ˛ )e e + 2(1 − ˛ )(1 − ˛ )(e − e )]
eL [(1 − ˛L )eH
H L H
H
L
H
L
2[(2 − ˛H )(2 − ˛L )eH − (1 − ˛H )(1 − ˛L )eL ]
,
yielding profits:
2
,
H = (1 − ˛H )(eH − eL )xH
(5)
eL
L = (1 − ˛L )(eH − eL ) xL2 ,
eH
(6)
where:
xH =
xL =
eH [(2 − ˛L )(2 − eH ) − eL )]
,
(7)
.
(8)
2[(2 − ˛H )(2 − ˛L )eH − (1 − ˛H )(1 − ˛L )eL ]
eH [(1 − ˛H )(2 − eL ) + eH )]
2[(2 − ˛H )(2 − ˛L )eH − (1 − ˛H )(1 − ˛L )eL ]
In order to identify the duopolists’ maximization problem at the first stage, Eqs. (5)–(8) are substituted in the generic equation (3). The equilibrium levels of clean-up correspond to the solutions
that simultaneously solve the following first order conditions derived from the unconstrained13 maximization problems:
∂JH
∂xH eH − ˛H eL
= 0 ⇔ (2 − ˛H )(eH − eL )
+
xH = 0;
eH
∂eH
∂eH
∂JL
eL ∂xL
eH − 2eL + ˛L eL
= 0 ⇔ (2 − ˛L )(eH − eL )
+
xL = 0.
eH ∂eL
eH
∂eL
(9)
(10)
Moorthy (1988) shows that when vertically differentiated firms behave as profit maximizers (˛ = 0
for both firms), their reaction functions are both positively sloped and so their quality choices are
strategic complements (see Bulow et al., 1985). However, if the high quality firm is a non-profit firm,
˛H = 1, its best response function is negatively sloped.14 Indeed, when the firm H is interested in
maximizing profit, an increment of eL triggers an increase of eH because the profit of the firm H is
proportional to the distance between eH and eL . On the other hand, when the firm H aims at maximizing
13
We neglect the constrain that the firms’ profit must be non-negative as we will verify that such condition is always satisfied
in equilibrium.
14
Indeed, if ˛H = 1 then ∂JH /∂eH ∂eL = −1/2(2 − ˛L ).
386
N. Doni, G. Ricchiuti / Resource and Energy Economics 35 (2013) 380–395
its aggregate abatement, an increment of eL entails a reduction of the market share of the firm H. As
a consequence the firm H finds it convenient to reduce its marginal clean-up in order to cut its price
and to limit its loss in terms of market share.
As far as the best response function of the firm L is concerned, it is possible to verify that it is
always positively sloped, whatever its degree of CSR. Therefore when the high quality firm is a nonprofit quality choices are neither strategic complements nor strategic substitutes. The different sign of
the reaction functions of the two firms when they behave as a non-profit is due to the different effect
that an increase in the environmental quality of the competitor has on their market share. Indeed, if the
firm L behaves as a non-profit and the firm H increases its abatement activity, xL grows. Consequently,
the firm L can exploit this opportunity to increase its marginal clean-up to a higher level.
The system given by Eqs. (9) and (10) can have at maximum one acceptable solution15 (i.e.: such
∗ ≥ e∗ ). When one solution exists, it corresponds to the equilibrium levels of marginal clean-up
that eH
L
of both firms, which depends on the parameters ˛H and ˛L . However, they are not analytically feasible.
Some clear results emerges when one of the duopolists is a non-profit firm and it produces either the
high-quality or the low-quality good.
Proposition 1.
exists.
If ˛L = 1, ˛H < 1 and > 0, then at the first stage no Nash equilibrium in pure strategies
When one of the two responsible producers is a non-profit firm, there is no equilibrium in which it
chooses the low level of marginal abatement with certainty. Indeed, in such case, firm L would mimic
the choice of its competitor, while firm H would prefer a level of marginal abatement strictly higher
than the level of its rival.
As a consequence, in the presence of a non-profit firm, the only Nash equilibrium in pure strategies
is characterized by the fact that this firm adopts the higher level of marginal abatement. Substituting
˛H = 1 in Eqs. (9) and (10) we can achieve the following results.
Lemma 1.
Consider ei∗ and xi∗ the equilibrium level of abatement and the market share of firm i = L, H
∗ ≤ e . If ˛ = 1, ˛ ∈ [0, 1] and ∈ (0, 2], then in equilibrium:
and assume that eL∗ ≤ eH
H
L
1. eL∗ is monotonically increasing in both and ˛L ;
2. xL∗ is monotonically increasing in ˛L and independent of ;
∗ is monotonically increasing in and decreasing in˛ ;
3. eH
L
∗ is monotonically decreasing in˛ and independent of ;
4. xH
L
∗ x∗ + e∗ x∗ is monotonically increasing in and decreasing in˛ .
5. E ∗ = eH
L
H
L L
∗ are proportional to . Hence, if
It is worth noting that, as shown in the proof, both eL∗ and eH
consumers do not have any WTP for the environmental quality, i.e. = 0, there is no space for any kind
of clean-up activity by part of responsible firms. At the same time, a higher consumers WTP induces
the responsible firms to increase the environmental quality of their products, without affecting their
market shares.16
An increment in the degree of CSR of the low quality firm triggers an increase in its aggregate
abatement, but this effect is counterbalanced by the reduction of the total clean-up of the non-profit
firm. This result is coherent with the slope of the best response function of firm H. If firm L becomes
more careful with the environmental impact of its production, it will increase its level of abatement,
but at the same time, it will reduce its mark-up, increasing its supply. On the other hand, firm H cannot
reduce its price without reducing its level of clean-up because it always charges a price equal to its
15
As explained by Motta (1993), the solutions of this system are only the candidate equilibrium of the model. We have checked
that second order conditions hold (a technical proof is available upon request). However, not all the candidate solutions are
valid because, in some cases, firm H may have an incentive to leapfrog the “equilibrium” level of marginal abatement of its rival.
In such cases, an equilibrium in pure strategies does not exist.
16
These properties hold also when ˛H =
/ 1.
N. Doni, G. Ricchiuti / Resource and Energy Economics 35 (2013) 380–395
387
Fig. 1. Nash equilibrium existence.
marginal costs. Therefore, firm H finds it convenient to decrease its level of abatement in order to limit
the reduction of its market share. The last statement shows us that the aggregate effect of an increase
in ˛L entails a reduction of the total abatement and, consequently, an increase in the aggregate level
of pollution.
More general results regarding the market equilibrium in the presence of two responsible firms can
be obtained by means of numerical simulations. The shaded region of Fig. 1 identifies the set of couples
(˛H , ˛L ) in which a Nash equilibrium does not exist: firm H would have an incentive to leapfrog17 firm
L by choosing a level of abatement lower than eL∗ (˛H , ˛L ) in order to increase its market share and
its total abatement. At the first stage we can have zero, one or two Nash equilibria, depending on
the weight that both the responsible firms assign to their profit. Indeed, making the symmetry of the
shaded region with respect to the 45◦ line, we can identify three different regions. When the shaded
region and its symmetric region coincide no Nash equilibrium exists. For all the other couples of values
of (˛i , ˛j ) contained in the shaded region only a Nash equilibrium exists, in which ˛H > ˛L . Finally, for
the couples of (˛i , ˛j ) contained neither in the shaded region nor in its symmetric counterpart, two
Nash equilibria exist.18
Hence, if both firms assign a high weight to the positive externality associated to their own production, the outcome of their strategic interaction is unpredictable. However, when a Nash equilibrium
exists, using numerical simulations we can analyze how the market equilibrium is affected by firms’
degree of CSR. On the basis of the relative graphs (see Appendix A) we can note that:
∗ is monotonically increasing (decreasing) in ˛ only when ˛ is sufficiently low (high). At the
(1) eH
H
L
∗ is monotonically increasing (decreasing) in ˛ only when ˛ is sufficiently low
same time, eH
L
H
∗ is first increasing and then decreasing in ˛ , while for
(high). For intermediate values of ˛L , eH
H
∗ is first decreasing and then increasing in ˛ (see Fig. 2a). Both x∗ and
intermediate values of ˛H , eH
L
H
∗ x∗ are monotone increasing in ˛ and monotone decreasing in ˛ (see Fig. 2b and c);
eH
H
L
H
17
The numerical calculations are available upon request.
The existing literature regarding vertically differentiated duopolies has already stressed the existence of two asymmetric
and mirror image Nash equilibria at the quality stage, i.e. two equilibria that are identical except for the ordering of the firms’
abatement levels (see Moorthy, 1988, p. 159). However, in the present model these two equilibria are no more mirror image (if
/ ˛L ) given the heterogeneity in the firms’ objective functions.
˛H =
18
388
N. Doni, G. Ricchiuti / Resource and Energy Economics 35 (2013) 380–395
Fig. 2. Market equilibrium. Figures are drawn assuming = 1.
(2) eL∗ is monotonically increasing in ˛L , while it is increasing (decreasing) in ˛H only when ˛L is
sufficiently low (high) (see Fig. 2d). Both xL∗ and eL∗ xL∗ are monotone increasing in ˛L and monotone
decreasing in ˛H (see Fig. 2e and f);
∗ x∗ + e∗ x∗ is monotone increasing (decreasing) in ˛ if ˛ is sufficiently low (high) and
(3) E ∗ = eH
H
L
H
L L
monotone increasing (decreasing) in ˛L if ˛H is sufficiently low (high) (see Fig. 2g);
∗ /e∗ is always monotone decreasing in ˛ while it is monotone decreasing in ˛ only if ˛
(4) ∗ = eH
L
H
L
L
is sufficiently high. When ˛L is low, * is first increasing and then decreasing in ˛H (see Fig. 2h).
It is worth noting that both the firms’ market share and total abatement are always increasing in
their own degree of CSR, and decreasing in the degree of responsibility of the rival firm. The overall
impact on the aggregate level of clean-up is ambiguous. However, as stated in point 3, an increase
in the CSR of firm L (H) can either increase or decrease the aggregate clean-up depending on the CSR
of firm H (L). To give an intuition, we could analyze how an increase in ˛i (i = L, H) affects the firms’
marginal abatements given the value of ˛j (with i =
/ j).
Given points 1 and 2, an increase in ˛L has always a positive effect on eL , while it has a positive
(negative) effect on eH when ˛H is low (high). Coherently with the reaction function of firm H, when
˛H is low an increase in eL induces the firm H to react increasing its per unit abatement as well. On
the other hand, almost non-profit firm H reacts to an increase in eL , cutting its price and therefore its
abatement (eH ).
At the same time, an increase in ˛H has a positive effect on eH only when firm L is mainly interested
in profit: indeed, in such case an increment in eH entails an increment in xH too. Conversely, when
firm L is almost non-profit, an increase in ˛H has a negative effect on eH , since the firm H needs to cut
N. Doni, G. Ricchiuti / Resource and Energy Economics 35 (2013) 380–395
389
its price in order to take market share away from the firm L. However in both cases the firm L, given
its reaction function, will change eL according to how the firm H moves eH .
To sum up, when a firm becomes more responsible it is not interested in increasing its level of
marginal abatement per se, but its aggregate clean-up. Therefore its choice is strictly related to the
consequence in terms of market share. The dynamic of the market competition is such that a higher
degree of CSR of a firm entails an increment of its total clean-up and a reduction of the aggregate
abatement of its rival. This second effect is stronger than the first when the rival firm has a high
degree of CSR. Indeed, when firm i becomes more responsible and firm j is almost non-profit, the
reaction of the latter is a reduction of its marginal abatement (ej ). Paradoxically, when both firms are
mainly interested in increasing their abatement activity, an increment in the degree of CSR of one of
them triggers a reduction in aggregate clean-up, i.e. an increase in total pollution.
Finally, a straightforward consequence of what we have just explained above is the statement
in point 4: an increase in the degree of responsibility of firm L always makes the allocation of the
abatement activity more cost effective (i.e. it decreases the value of * , a measure of the distance
between the environmental quality of the two responsible firms), while an increase in the degree of
responsibility of firm H does not have a clear-cut effect on * .
From all these observations we cannot draw any clear inference with regard to the influence of a
higher degree of responsibility on the social welfare. Indeed, an increase in the degree of CSR of one
firm does not have a definite effect on the aggregate abatement and on the cost-effectiveness of the
allocation of such activity. On the other hand, a higher green consumers’ WTP always increases the
aggregate abatement, but also the total costs of the clean-up activity. Therefore, also the impact of
consumers’ responsibility on the total welfare must be deepened.
4. Social welfare
The definition of social welfare in the presence of green consumers is controversial: in a purely utilitarian view, the social welfare is equal to the sum of the firms’ profit and the consumers’ surplus, which
should include the consumers’ WTP for the environmental quality of a product. Consequently, the
evaluation of a policy in welfare terms depends on consumers’ preferences. However, such approach
becomes problematic when policy goals can be achieved through public campaign aiming at increasing the green consumers’ degree of consciousness. As emphasized by Brennan (2006, p. 153), ‘Making
green preferences sensible as a policy instrument may require rejecting the economist’s method of basing
policies on preferences. Such a view better fits the ecologist’s view of environmental policy, where the right
level of pollution is defined by criteria drawn from biology and perhaps environmental ethics, rather than
from estimated willingness to pay for environmental public goods.’
Furthermore the interpretation of consumers’ WTP is also disputed: the premium paid by green
consumers for the purchase of environment-friendly products represents a form of voluntary contribution to the provision of a public good. In the economic literature, there are different ways to
reconcile this behavior with the traditional assumption of self-interested agents. A first attempt is
based on the assumption that green consumers obtain a direct utility by the environmental qualities
of the goods they buy. In this view green consumers derive a warm glow from their responsible action
(Andreoni, 1990), due to social approval or to their internal moral motivation (Bénabou and Tirole,
2006). On the other hand, we could think that green consumers behave as conditional cooperator, who
accept to sacrifice their utility conditional on expectations that others will do the same. Indeed, other
authors (e.g. Ostrom, 2000) emphasizes that in the presence of social dilemmas, if all individuals seek
to maximize their egoistic interest, they are unavoidably trapped in a suboptimal equilibrium. For this
reason truly rational agents can choose to switch to more refined choice criteria and consequently,
their market choices are not a good proxy for their true preferences. Consequently, following such
approach social welfare should not take consumers’ WTP into account.
The majority of the economic literature related to the phenomenon of green consumers follows
the first approach, by including consumers’ WTP in their calculations, but some authors exclude such
factor from their welfare analysis (see for instance Eriksson, 2004; Wirl, 2011). In this paper we follow
the view of the majority, in order to make our results directly comparable with the existing literature.
However, in light of this debate, we have chosen to separate the social welfare into two components:
390
N. Doni, G. Ricchiuti / Resource and Energy Economics 35 (2013) 380–395
the aggregate consumers’ WTP and the terms accounting for the efficiency of the abatement activity.
In this way we can enrich our analysis emphasizing when the results depend on the concept of social
welfare we choose.
The social welfare is defined as the sum of consumers’ utility and firms’ profit:
W=
V − p() + e() − Y ∗ f ()d +
0
p∗i −
ei2
2
xi∗ ,
(11)
i=H,L
where:
(p(), e()) =
⎧
˜
⎪
⎨ (0, 0) if ∈ [0, ]
⎪
⎩
˜ ]
ˆ
(p∗L , eL∗ ) if ∈ [,
∗ ) if ∈ [,
ˆ ]
(p∗H , eH
and:
∗ ∗
Y ∗ = e − eL∗ xL∗ − eH
xH .
(12)
Rearranging we get:
W = WA + WB
where:
WA =
eL∗ xL∗
˜ + ˆ
2
∗ ∗
+ eH
xH
and
WB = V − e +
ˆ + 2
ei∗
−
ei∗
2
(13)
2 xi∗ .
(14)
i=H,L
WA corresponds to the consumers’ aggregate WTP, while WB represents the standard difference
between the benefit and the cost of the firms’ abatement activity.
Proposition 2. WA is monotone increasing in while WB is always first increasing and then decreasing
in , ∀˛H , ˛L ∈ [0, 1] and ∈ [0, 2].
An increment in will increase directly the consumers’ WTP for any given level of quality and
indirectly the environmental quality chosen by firms and therefore WA . Moreover, given that does
not have any effect on the firms’ market shares, the impact of an increment of on WB depends on
the difference between benefits and costs of firms’ marginal abatement: if the maximum WTP is low
(high), the marginal abatements are low (high) and the relative benefits increases faster (slower) than
the relative costs.
The variation of W, the sum of WA and WB , with respect to , is not analytically feasible. However, we get some results from numerical simulations: W is always increasing. The positive effect on
the consumers’ aggregate WTP is always greater than the possible negative impact on the difference
between the benefit and the cost of the firms’ abatement activity.
Regarding the pattern of WA and WB with respect to ˛L and ˛H , it is not feasible to derive analytical
results. This is due to the contrasting effects of an increment in a firm CSR on the total clean-up of
∗ x∗
both firms: indeed, as shown above, when a firm increases its own degree of CSR, eL∗ xL∗ and eH
H
go in opposite direction. Therefore, it is ambiguous how WA is affected by the variation of ˛H or
˛L . Additionally, while an increment in a firm total clean-up is always positively related to WA , its
influence on WB could be either positive or negative, depending on the specific levels of the firms’
marginal clean-up. However, if we limit the analysis to the case in which ˛H = 1 it is possible to derive
some interesting results:
N. Doni, G. Ricchiuti / Resource and Energy Economics 35 (2013) 380–395
391
Fig. 3. W(˛H , ˛L ) for different values of .
Fig. 4. WB (˛H , ˛L ) for different values of .
Proposition 3. If ˛H = 1 then ∀˛L ∈ [0, 1] and ∈ (0, 2] : i) WA is monotone decreasing in ˛L ; ii) if <
1/3(6/5), then WB (W) is monotone decreasing in ˛L ; otherwise it is first increasing and then decreasing.
Therefore, when firm H is non-profit, an increase in ˛L may negatively affect the social welfare.
This negative effect is mainly due to the reduction in the aggregate consumers’ WTP driven by the
reduction in total clean-up (see Lemma 1). Moreover, even W and WB are negatively affected by a
higher ˛L when and ˛L are sufficiently high.
However, in general, numerical simulations show that the social welfare can be both increasing
and decreasing in firms’ CSR, depending on the specific values of ˛L , ˛H and .
W is increasing (decreasing) in ˛H and ˛L when ˛L and ˛H are sufficiently low (high), whatever the
value of is (see Fig. 3a–c). The intuition is that in such cases total abatement increases (decreases) and
consequently also WA grows up (falls). Conclusions are quite different if we choose to consider only
WB . Indeed, its pattern is highly influenced by the value of : for instance, when is sufficiently low,
the impact of a variation in the degree of a firm CSR on WB is similar to the impact on W. Conversely
when is sufficiently high, WB is decreasing in ˛H even if ˛L and ˛H are close to 0 (see Fig. 4c). The
influence of the value of on the pattern of WB is due to the fact that the impact that a variation of ei
(i = L, H) has on the difference between its benefits and its cost depends on its absolute value, that in
equilibrium is proportional to (see Eq. (14)).
The inclusion or the exclusion of consumers’ preferences from the theoretical welfare is a relevant
choice: it may lead to opposite conclusions. For instance, if we identify the social welfare with WB ,
it could be preferable to have a standard duopoly, where firms are both profit maximizers, rather
than a mixed duopoly, where a non-profit producer competes with a for-profit firm. In such case
the presence of a non-profit firm competing with a profit-maximizing firm may harm social welfare.
Moreover, there are couple of parameters (˛i , ˛j ) for which two Nash equilibria exist and WB is higher
when the firm with the highest degree of CSR adopts the low level of clean-up (see Fig. 4c). This never
happens if we define the social welfare as W.
Summing up, a higher degree of responsibility of consumers and/or firms may both increase and
decrease the social welfare, whatever the definition we choose to adopt, depending on the degree of
responsibility of each agent. Consequently it is not always reasonable for consumers and share-holders
to sacrifice their private utility in order to voluntarily contribute to environmental protection.
392
N. Doni, G. Ricchiuti / Resource and Energy Economics 35 (2013) 380–395
5. Conclusions
In this paper we draw on two so far unrelated strands of literature: the green consumers and
the corporate social responsibility of firms. We develop a model where consumers care about the
environmental impact of goods they buy and some firms, following a multidimensional objective,
weigh up both profit and abatement activity. Our analysis focuses mainly on the effects associated to
exogenous changes in aggregate consumers’ WTP for cleaner goods or in the degree of firms’ CSR.
Our main results are that both aggregate clean-up and social welfare may be reduced by an increase
of consumers’ WTP and by a higher degree of firms’ CSR. The key mechanism is related to the dynamic
of market competition: higher degree of CSR of a firm entails an increment of its total clean-up and a
reduction of its rival’s aggregate abatement. This second effect is stronger than the first when the rival
firm has a high degree of CSR. At the same time, when consumers’ WTP is very high, responsible firms
may overprovide environmental quality, triggering a reduction in social welfare. Therefore a higher
responsibility does not necessarily entail lower pollution and higher welfare.
Our results allow us to compare the equilibrium of a standard duopoly, where two standard profit
maximizing firms compete, with a mixed duopoly, where a non-profit firm faces a for-profit firm. The
aggregate clean-up and welfare are higher in the latter case. However, this conclusion changes if we
exclude from the social welfare the component influenced by consumers’ preferences. Indeed, when
consumers’ WTP is very high, the efficiency of firms’ clean-up activity is higher under a standard than
under a mixed duopoly.
To summarize, an increase in the responsibility of either consumers or producers is welfare improving only if the aggregate consumers’ WTP and if the degree of firms’ CSR are sufficiently low; conversely,
when the degree of responsibility of the rivals is already high we can have surprising and unexpected
results.
Future research should extend our analysis in order to check the robustness of our results under
different assumptions. For instance, firms could compete in a different market form: we could assume
that one firm is a Stackelberg leader, and/or that the number of responsible firms is endogenous. Moreover, responsible firms could maximize other kind of objective functions. Finally, both consumers’ WTP
and firms’ degree of CSR could be endogenous: in such case the dynamic properties of the interaction
between green consumers and responsible firms should be analyzed.
Appendix A.
Proof of Proposition 1. When firm L is a non-profit, the market shares of the responsible firms can
be rewritten by substituting ˛L = 1 in formula (7) and (8). We obtain:
xH =
xL =
2 − eL − eH
2(2 − ˛H )
,
[(1 − ˛H )(2 − eL ) + eH ]
2(2 − ˛H )
(15)
.
(16)
The derivative of JL with respect to eL is equal to:
∂JL
∂xL
= eL
+ xL = 2(1 − ˛H )( − eL ) + eH
∂eL
∂eL
(17)
From Eq. (15) we can deduce that xH ≥ 0 ⇔ eL ≤ . This implies that Eq. (17) is always strictly
positive. Consequently, at the first stage firm L would choose eL = eH , the maximum level of marginal
abatement under the constraint that eL must be weakly lower than eH . On the other hand, if ˛H < 1,
firm H always would choose eH > eL , because if eH = eL its profit is equal to 0. N. Doni, G. Ricchiuti / Resource and Energy Economics 35 (2013) 380–395
393
Proof of Lemma 1. Note that if > 0, ˛H = 1 and ˛L ∈ [0, 1] then the solution of the system identifies
the following equilibrium levels of marginal abatement:
eL∗ =
∗
=
eH
2(2 − ˛L )
9 − 8˛L + 2˛2L
;
(18)
.
(19)
2(2 − ˛L )2
9 − 8˛L + 2˛2L
Substituting these solutions in Eqs. (7) and (8), the following equilibrium market shares are
achieved:
xL∗ =
∗
=
xH
(2 − ˛L )
(20)
9 − 8˛L + 2˛2L
(2 − ˛L )2
(21)
9 − 8˛L + 2˛2L
∗ , x∗ and E* in the way stated in the lemma.
Straightforward algebra the parameter affects eL∗ , xL∗ , eH
H
With regard to the impact of the degree of CSR of the low-quality firm on firms’ marginal abatement
choices and on their market shares, the results stated in the proposition stem from the follow-
ing derivatives:
∂xL∗
2˛2 −8˛L +7
∂ e∗
∂x∗
∂x∗
∂ e∗
∂x∗
2(2−˛L )
L
=
> 0; L = 2 ˛ L > 0; H = −
< 0; H = 2 ˛ H < 0;
∂˛L
∂˛L
∂˛L
∂˛L
L
L
(9−8˛L +2˛2 )2
(9−8˛L +2˛2 )2
L
L
2˛L −4
∂E ∗ = 2
< 0.
∂˛L
(9−8˛L +2˛2 )4
L
Proof of Proposition 2.
∗ , e∗ , ˆ and ˜ are all increasing in , it
Given Eq. (13) and knowing that eH
L
follows that the WA is monotone increasing in . The derivative of Eq. (14) with respect to a generic
exogenous parameter z is:
∂ e∗
∂x∗
k
∂WB ∗
( − kei )xi∗ i + ( − ei∗ )ei∗ i .
=
2
∂z
∂z
∂z
i=H,L
2
2
Given that ∂xi∗ /∂ = 0 and ∂ ei∗ /∂ = 0 we obtain that:
2
∂ WB
∂
2
=−
∂ e 2
i
∗
i=H,L
xi
∂
<0
Hence, WB is concave in . Furthermore, as (∂ei∗ /∂) > 0 and (∂xi∗ /∂) = 0, we can note that the
sign of (∂WB /∂) depends on the sign of ( − ei∗ ), ∀i = L, H. The monotony would imply that the sign of
the first derivative, valued in both the lower and the upper bound of , should be the same. However,
∗ and e∗ are lower (higher)
it is easy to check that, whatever the values of ˛i are, when is 0 (2) both eH
L
than and WB is increasing (decreasing) in . Therefore, WB is never monotone in .
Proof of Proposition 3. (i) Substituting Eqs. (18)–(21) in Eq. (13) and deriving with respect to ˛L we
obtain:
(2˛5L − 26˛4L + 128˛3L − 301˛2L + 344˛L − 156)
dW A
=
d˛L
(9 − 8˛L + 2˛2L )4
2
(22)
394
N. Doni, G. Ricchiuti / Resource and Energy Economics 35 (2013) 380–395
It is straightforward to verify that (2˛5L − 26˛4L + 128˛3L − 301˛2L + 344˛L − 156) < 0, ∀˛L ∈ [0, 1].
Indeed, it is monotone increasing in ˛L and its value for ˛L = 1 is less than 0. (ii) Substituting Eqs.
(18)–(21) in Eq. (14) and deriving with respect to ˛L we obtain:
2(2 − ˛L )[(6˛4L − 42˛3L + 108˛2L − 123˛L + 54) − 4˛2L + 16˛L − 18]
dW B
=
d˛L
(9 − 8˛L + 2˛2 )4
(23)
L
This equation is monotone decreasing in ˛L . Moreover, if < 1/3, then
dW B
|
d˛L ˛L =0
< 0. Consequently,
if < 1/3 WB is monotone decreasing in ˛L , otherwise it is first increasing and then decreasing.
Finally, from Eqs. (22) and (23) we get:
(2 − ˛L )[(10˛4L − 62˛3L + 132˛2L − 113˛L + 30) − 8˛2L + 32˛L − 36]
dW
=
d˛L
(9 − 8˛L + 2˛2 )4
L
This equation is monotone decreasing in ˛L . Moreover, if < 6/5, then
dW
|
d˛L ˛L =0
< 0. Consequently, if
< 6/5 W is monotone decreasing in ˛L , otherwise it is first increasing and then decreasing.
References
Andreoni, J., 1990. Impure altruism and donations to public goods: a theory of warm-glow giving. The Economic Journal 100,
464–477.
Arora, S., Gangopadhyay, S., 1995. Toward a theoretical model of voluntary overcompliance. Journal of Economic Behavior and
Organization 28, 289–309.
Bagnoli, M., Watts, S.G., 2003. Selling to socially responsible consumers: competition and the private provision of public goods.
Journal of Economics and Management Strategy 12, 419–445.
Bansal, S., Gangopadhyay, S., 2003. Tax/subsidy policies in the presence of environmentally aware consumers. Journal of
Environmental Economics and Management 45, 333–355.
Bansal, S., 2008. Choice and design of regulatory instruments in the presence of green consumers. Resource and Energy
Economics 30, 345–368.
Baron, D.P., 2001. Private politics, corporate social responsibility, and integrated strategy. Journal of Economics and Management
Strategy 10, 7–45.
Becchetti, L., Huybrechts, B., 2008. The dynamics of fair trade as a mixed-form market. Journal of Business Ethics 81, 733–750.
Bénabou, R., Tirole, J., 2006. Incentives and prosocial behavior. American Economic Review 96, 1652–1678.
Bénabou, R., Tirole, J., 2010. Individual and corporate social responsibility. Economica 77, 1–19.
Besley, T., Ghatak, M., 2007. Retailing public goods: the economics of corporate social responsibility. Journal of Public Economics
91, 1645–1663.
Brennan, T.J., 2006. Green preferences as regulatory policy instrument. Ecological Economics 56, 144–154.
Bulow, J.I., Geanakoplos, J.D., Klemperer, P.D., 1985. Multimarket oligopoly: strategic substitutes and complements. Journal of
Political Economy 93, 488–511.
Calveras, A., Ganuza, J.J., Llobet, G., 2007. Regulation, corporate social responsibility and activism. Journal of Economics and
Management Strategy 16, 719–740.
Conrad, K., 2005. Price competition and product differentiation when consumers care for the environment. Environmental and
Resource Economics 31, 1–19.
Cremer, H., Marchand, M., Thisse, J.F., 1991. Mixed oligopoly with differentiated products. International Journal of Industrial
Organization 9, 43–53.
Cremer, H., Thisse, J.F., 1999. On the taxation of polluting products in a differentiated industry. European Economic Review 43,
575–594.
De Donder, P., Roemer, J.E., 2009. Mixed oligopoly equilibria when firms’ objectives are endogenous. International Journal of
Industrial Organization 27, 414–423.
Doni, N., Ricchiuti, G., 2011. Market equilibrium in the presence of green consumers and responsible firms: a comparative statics analysis. Working Paper n. 07/2011, Dipartimento di Scienze Economiche, Firenze. Available at:
http://ideas.repec.org/p/frz/wpaper/wp2011 07.rdf.html
Ecchia, G., Lambertini, L., 1998. Existence of equilibrium in a vertically differentiated duopoly. In: Motta, M., Lambertini, L.
(Eds.), The Economics of Vertically Differentiated Markets. Edward Elgar, Cheltenam, UK, pp. 83–91.
Eriksson, C., 2004. Can green consumerism replace environmental regulation? A differentiated-products examples. Resource
and Energy Economics 26, 281–293.
Garcia-Gallego, A., Georgantzis, N., 2009. Market effects of changes in consumers’ social responsibility. Journal of Economics
and Management Strategy 18, 235–262.
Ishibashi, K., Kaneko, T., 2008. Partial privatization in mixed duopoly with price and quality competition. Journal of Economics
95, 213–231.
Khanna, M., 2001. Non-mandatory approaches to environmental protection. Journal of Economic Surveys 15, 291–324.
Kitzmueller, M., Shimshack, J., 2012. Economic perspectives on corporate social responsibility. Journal of Economic Literature
50, 51–84.
N. Doni, G. Ricchiuti / Resource and Energy Economics 35 (2013) 380–395
395
Kotchen, M.J., 2005. Impure public goods and the comparative statics of environmentally friendly consumption. Journal of
Environmental Economics and Management 49, 281–300.
Kotchen, M.J., 2009. Some microeconomics of eco-enterpreunership. In: Libecap, G. (Ed.), Frontiers in Eco-Entrepreneurship
Research. Emerald Group Publishing, pp. 24–37.
Lombardini-Riipinen, C., 2005. Optimal tax policy under environmental quality competition. Environmental and Resource
Economics 32, 317–336.
Lyon, T.P., Maxwell, J.W., 2008. Corporate social responsibility and the environment: a theoretical perspective. Review of
Environmental Economics and Policy 1, 1–22.
Moorthy, S.K., 1988. Product and price competition in a duopoly. Marketing Science 7, 141–168.
Moraga-Gonzalez, J.L., Padron-Fumero, N., 2002. Environmental policy in a green market. Environmental and Resource Economics 22, 419–447.
Motta, M., 1993. Endogenous quality choice: price vs. quantity competition. Journal of Industrial Economics 41, 113–131.
Ostrom, E., 2000. Collective action and the evolution of social norms. The Journal of Economic Perspectives 14, 137–158.
Reinhardt, F.L., Stavins, R.N., Vietor, R.H.K., 2008. Corporate social responsibility through an economic lens. Review of Environmental Economics and Policy 2, 219–239.
Rodriguez-Ibeas, R., 2007. Environmental product differentiation and environmental awareness. Environmental and Resource
Economics 36, 237–254.
Wirl, F., 2011. Global warming with green and brown consumers. The Scandinavian Journal of Economics 113, 866–884.