Individual firm and market dynamics of CSR activities

Journal of Economic Behavior & Organization 86 (2013) 169–182
Contents lists available at SciVerse ScienceDirect
Journal of Economic Behavior & Organization
j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / j e b o
Individual firm and market dynamics of CSR activities
Franz Wirl a , Gustav Feichtinger b , Peter M. Kort c,d,∗
a University of Vienna, Industry, Energy and Environment, Vienna, Austria
b Vienna, University of Technology, Institute for Mathematical Methods in Economics, Department for Operations Research and Control Systems, Vienna,
Austria
c Tilburg University, Department of Econometrics & Operations Research and CentER, PO Box 90153, 5000 LE Tilburg, The Netherlands
d University of Antwerp, Department of Economics, Antwerp, Belgium
a r t i c l e i n f o
Article history:
Received 26 January 2012
Received in revised form
26 November 2012
Accepted 6 December 2012
Available online 27 December 2012
JEL classification:
C61
D21
D92
L21
Keywords:
CSR
Dynamics
firms’ Interactions
Stability
History and expectation dependence
a b s t r a c t
This paper investigates how firms should plan corporate social responsibility (short CSR). This
dynamic analysis starts with a firm’s intertemporal optimization problem, and proceeds to
analyze interactions with other firms, which are crucial: if CSR is profitable for firm A then
it is most likely also profitable for competitors B and C, and these simultaneous decisions
affect the gain each would achieve from trying to advance its own position. We find that
multiple equilibria exist, irrespective of whether interactions with other firms are taken
into account. Interactions can eliminate or create additional steady states and can lead to a
situation in which history is insufficient to determine the long run outcome among multiple
steady states, so that coordination is beneficial.
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
The term Corporate Social Responsibility (CSR) came into common use in the early 1970s and it refers to a form of corporate
self-regulation. This approach accounts for stakeholders’ interests (advanced in a book by Freeman, 1984) in addition to the
shareholders’ focus on profits and includes thus a number of ethical, social, and environmental targets, where the green
aspect is the most visible one currently. Michael Porter has for some time praised the profitability (win-win) of socially
and, in particular, environmentally responsible behavior, e.g., in Porter and van der Linde (1995a,b), in the annual global
competitiveness reports des World Economic Forum, and most recently in Porter and Kramer (2011). In the last paper, it is
argued that taking a broad view on societal needs is nothing less than a means to ‘reinvent capitalism and unleash a wave of
innovation and growth’. Another broader view is given in Aoki (2011) that mentions CSR as a part in the role of organizations,
which is to cognitively mediate agents’ strategic interactions. Lyon and Maxwell (2008) and Reinhardt et al. (2008) are part of
a recent debate about the pros and cons (in that order). The importance of CSR is also documented by its appearance in many
articles in the public press. We only mention here the surveys in The Economist, first in 2005 (“The good company”, January
∗ Corresponding author at: Tilburg University, Department of Econometrics & Operations Research, PO Box 90153, 5000 Tilburg, The Netherlands.
Tel.: [phone]; fax [phone].
E-mail addresses: [e-mail] (F. Wirl), [e-mail] (G. Feichtinger), [e-mail] (P.M. Kort).
0167-2681/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jebo.2012.12.004

170 F. Wirl et al. / Journal of Economic Behavior & Organization 86 (2013) 169–182
Fig. 1. Evolution of EMAS certifications over time. Permission granted. Copyright with: EMAS.
20) followed in 2008 (in “Just good business”, January 19) and a very recent and more optimistic article (“Schumpeter: Good
business; nice beaches”, May 17, 2012).
The evolution, one can even speak of a wave, of CSR activities can be seen, e.g., by considering how many firms subjected
themselves voluntarily over time to (environmental and social) standards like EMAS (“eco-management and audit scheme”
within the European Union) and ISO14001 (the same objective of good environmental management), as shown in Fig. 1.
Fig. 2 shows the composition by countries. A survey of attitudes to business turns up some intriguing differences about CSR:
first across nations, where Friedman’s view (’the business of business is business’) dominates in developing and emerging
economies according to The Economist (Jan 27th 2011), whereas Germany and other northern and western European com-
panies are much more concerned about CSR. Second, CSR activities differ not only across nations but also widely across
industries. For example, all major oil companies are very CSR conscious (if not always successful, as BP’s failures document)
while much less is heard of other mining industries. Understanding an evolution as the one shown in Fig. 1 requires an
explicit dynamic analysis of CSR activities complementary to the many static investigations. In particular, a dynamic anal-
ysis is needed that allows for a dynamic pattern compatible with Figs. 1 and 2, implying different evolutions and long run
outcomes across countries, regions and industries.
The aim of our model is to address these aspects. In general and more formal terms, two things will be shown: first,
positive interactions between a firm’s current investment and its stock, which seems to be of particular importance for CSR.
For the reasons given, it can make a firm’s long run choice dependent on initial conditions even if the environment remains
constant and the firm operates under diminishing returns. Second, the interactions among firms, which all invest in CSR,
can lead to history dependent, and to even more complex, market outcomes.
The starting point is by and large the optimistic management literature that encourages firms to pick up this win-win
situation associated with CSR. Unruh and Ettenson (2010) even warn managers: “Don’t let your competitors control what
“sustainable” means in your industry.” What this quote and many other management proposals often ignore, however, is
that one must take the reactions of other firms into account, because they face similar incentives. Therefore, this paper
investigates also how such interactions among firms affect outcomes. This integration allows us to address the following
questions: can we explain a CSR-wave? Is it a transient phenomenon or is it sustainable in the long run? Can multiple
equilibria arise such that an individual firm’s activities depend on its history? Does this history dependence extend to entire
industries? Does a history dependent outcome for the firm’s isolated profit maximization problem vanish if we include the
interaction with the other firms in our analysis? Or is the converse possible: where a firm’s decision to fix the other firms’
CSR activities leads to a unique and stable outcome? Does including interactions with other firms yield a history dependent,
maybe even expectation driven, outcome?
The paper is organized as follows: the model is introduced in Sections 3 and 4, after a brief literature review (Section
2). Section 5 derives the necessary optimality conditions for the optimal expansion of a firm’s CSR activities. Section 6
characterizes a single firm’s optimal expansion strategy holding the other firms’ activities constant. Section 7 integrates the
decisions of the other firms within a competitive setting in order to address the above raised questions. Section 8 concludes.

F. Wirl et al. / Journal of Economic Behavior & Organization 86 (2013) 169–182 171
Fig. 2. Regional distribution of ISO 14001 and EMAS certifications.
2. Literature review
A large volume of empirical papers investigates the benefits from CSR. Most of them focus on particular initiatives
in isolation, which may overestimate the effect of particular initiatives if the agents care about the whole CSR portfolio.
The following sample is somewhat selective and focuses on those that bear some connection to our paper. Orlitzky et al.
(2003) find in their meta-study some support for the optimistic claims about CSR. Similarly, Nishitani (2011) obtains that
environmental management adds value for a sample of Japanese firms. This is confirmed in Becchetti et al. (2007) by looking
at how additions and deletions to the Domini Social 400 index affect share prices. Of course, correlation is not causation,
i.e. the found patterns may be due to reverse causality since only profitable firms can afford CSR, or they are spurious. And
indeed, Surroca et al. (2010) merely find an indirect relationship that relies on the mediating effect of a firm’s intangible
resources. Fernández-Kranz and Santaló (2010) study the link between competition and social performance, also empirically,
and obtain that competition fosters CSR. Krishna and Rajan (2009) link CSR with cause marketing by analyzing a game of
competing firms, and theyrun corresponding experiments. Eichholtz et al. (2010) establish that green office buildings fetch
higher rents and selling prices. Chatterji et al. (2009) find that published rankings provide ‘fairly good summaries of past
environmental performance’. Siegel and Vitaliano (2007) verify empirically that firms selling credence goods are more likely
to be socially responsible than firms selling search goods. This is economically intuitive and in line with Heyes (2005) that
suggests signalling as a motive for CSR.
A subclass of the theory papers uses signaling games. Heyes (2005) and Goyal (2006) investigate signaling in connection
with FDI. Recently, Heijnen and van der Made (2012) consider signaling from the other side, more precisely they show
consumers boycotting a firm if it is non-CSR. Baron (2007) presents an agency theoretic framework of corporate social
responsibility and shows, among other things: (i) that CSR is costly when it is an imperfect substitute and (ii) that the
entrepreneurs, but not the shareholders, bear its cost. In a related paper, Baron (2008) suggests that corporate social respon-
sibility is a form of private provision of public goods. Bagnoli and Watts (2003) link CSR to the sales of their private goods.
And more recently, Baron (2010), by using a matching model, advances how self regulation and the existence of social labels
and certification organizations mitigate free riding.
Another branch treats CSR as a kind of product differentiation. Arora and Gangopadhyay (1995) explain overcompliance as
a result of willingness to pay for green goods depending on income and resulting market segmentation. Becchetti et al. (2003)
show in a Hotelling framework that a socially concerned entrant generates a Pareto improvement. Manasakis et al. (2007)
consider CSR in a two-stage game of product differentiation and delegation. In Alves and Santos-Pinto (2008) CSR is linked
to goods being complements if consumers prefer CSR firms. Kopel (2009) employs a product differentiation framework to

172 F. Wirl et al. / Journal of Economic Behavior & Organization 86 (2013) 169–182
study the conditions under which socially responsible firms can develop a first-mover advantage. Toolsema (2009) designs
a duopoly framework to investigate the role of switching costs in a setting with green and nongreen products.
In contrast, the analysis of CSR within a dynamic framework is rare. There exist papers that investigate dynamic versions
of the above sketched product differentiation games, but they do not consider CSR (e.g., Piga, 1998; Cellini and Lambertini,
2002; Hsu and Wang, 2004, and, more recently, Lee et al., 2012). The evolutionary but descriptive approach of CSR in Chen
et al. (2009) applied to emerging economies (as well as the multiperiod supply chain networks combined with multi-criteria
decision optimization in Cruz, 2008 and in Cruz and Wakolbinger, 2008) are complementary to our approach. The following
papers choose a similar focus and method: Becchetti et al. (2010) assume that consumer tastes for CSR increase with a
monopoly’s CSR activities. This is extended to an investigation of the open loop Nash equilibrium among oligopolists in
Becchetti and Solferino (2005). Lundgren (2011) focuses on intertemporal consumer goodwill, which is similar to our model
but without competitors. However, all these dynamic models are structured in a way such that unique and monotonic
saddlepoint paths result, which cannot explain the variance in CSR activities internationally and across industries. As said
before, our aim is to develop a framework that can explain this variance of CSR activities, where at the same time we can
address the question of whether these activities are sustainable.
3. Model
The starting point is similar to Lundgren (2011): a firm considers whether to expand (or to reduce) its CSR activities, with
the aim to maximize its net present value of profits. The following notation is used:
x = afirm′snumberofCSRprojects,
u = changeinafirm′sCSRactivities,
X = averageindustrywideCSRactivity,
 (u, x, X) = instantaneouspayoffforthefirm.
Examples of CSR projects (x) are: environmental reports, philanthropic supports and sponsoring, energy and environmental
management, mentoring and educational programs for workers, family friendly workplaces and alike. Further examples
are higher wages, improvements of working conditions and a switch to local sourcing. The control variable u can be either
negative or positive in order to allow for additions and terminations of such activities and projects. Assuming a competitive
environment, it follows that the firm cannot influence the level of the average industry-wide activity, X.
The firm’s expenditures for CSR consist of, firstly, running existing projects (K(x)) that may include increasing or decreasing
returns, K′′0. Secondly, there are costs for adding new CSR projects, C(u, x), including scrapping if u < 0. With respect to
expansion it makes sense to assume diminishing returns, Cuu > 0. The dependence of these costs, and in particular of the
marginal costs (Cu), on the state x, is presumably nonlinear due to two opposite effects. On the one hand we may have
scale economies and experience, thus Cux ≤ 0 for x small. On the other hand, diseconomies of scale occur as new projects are
getting more and more costly after many projects have already been launched (because the cheaper ones will be selected
first), thus Cux ≥ 0 for x ‘large’. Although one might also include industry-wide cost spillovers, C(u, x, X), where either CX < 0,
i.e., a firm may benefit from the experience in other companies, or CX > 0.1 Their presumably small effect is neglected in the
analysis.
The benefit for an individual firm from CSR is
B(u, x, X) = b(u, x) + f (x, X).
B has positive derivatives with respect to both private variables (u, x), but depends presumably negatively on X. At this
abstract level, benefits B can include many aspects of CSR. Here we think of, e.g., higher revenues by increasing consumers’
willingness to pay for a CSR-promoted good, or for all goods of a firm investing in CSR, less risks and liabilities in the case of
active environmental management and therefore less insurance premia, better access to capital, less sick leaves, and more
effort due to better treatment of workers and providing a family-friendly environment.2 In addition, CSR allows the manager
to obtain private and personal gains from running a green and socially concerned company by accruing public admiration,
invitations, and honors for the firm’s philanthropic engagement and/or support of local industry.
The benefits from a firm’s position in relation to its competitors (f) are separated from the direct benefits (b) only for
reasons of simplicity, which is by no means crucial for the results. Moreover, it is also not entirely out of line with classifying
CSR benefits. For example, let b account for all the firm’s internal benefits, like more effort, less sick leave, less disruption due
to local supplies, less risks and liabilities, etc. and f the gain in the market. Such a gain can be accomplished, e.g., in a Hotelling
1 This could be caused by diminishing returns, e.g., due to resource constraints. A concrete example is the cacao industry and protection of the biosphere.
The more the industry pursues this CSR-strategy – to grow cacao beans in the shadow of the jungle instead of on cleared agricultural plots – the more costly
become additional locations, because the best places have been already taken.
2 There is a large amount of literature, theoretical, empricial and experimental, that supports this: on efficiency wages e.g., Shapiro and Stiglitz (1984),
Salop (1979), Malcomson (1981); on gift exchanges Akerlof (1982) is the seminal paper; on intrinsic motivation we have Ryan et al. (1991), Frey and
Oberholzer-Gee (1997), and Kreps (1997); and for positive reputation effects, see Myers (2009).