Why Do People Obey Norms?

Abstract

In this chapter presents some reasons why people comply with social norms. First, in large measure, people do what they do because they have learned from those who surround them. The society is sustained by processes favorable to individuals endowed with some docility in following rules. Second, social norms can be sustained if the pecuniary advantage from breaking norms is not sufficient to offset the forgone reputation effect. Third, people comply with norms because the threat of punishment makes it in their interest to do so. Fourth, norms are represented as Nash equilibria of games played by rational agents, and as such they are self-enforcing. Finally, correlated equilibrium allows players’ actions to be statistically dependent on some random signals external to the model.

Appendices

Appendix 1 is based on Teraji (2007).

Appendix 2 is based on Teraji (2013).

Appendix 3 is based on Teraji (2009b).

Appendix 1: Morale and the Evolution of Norms

Appendix 1 is based on Teraji (2007).

2.1.1 Introduction

The literature on socio-economic organizations has recognized the role of team spirit in motivating workers. An organization enjoys good performance as a result of high morale and suffers bad performance as a result of low morale. The economic performance of an organization can vary because of the team atmosphere that prevails. However, in contract theory and principal-agent theory, which are based on material (pecuniary) incentives, values and preferences are taken as given. Thus, the existing literature on incentives cannot offer a way of explaining the diversity of morale observed across organizations. There is something other than material incentives that matters in organizations.

The standard conception of the economic agent as a creature driven by material self-interest has provided powerful theoretical tools in the analysis of diverse problems. However, experimental support for this conception has been fragile. Results from experiments on public goods games, ultimatum games, trust games, and gift exchange games demonstrate that people in fact deviate from self-interest in systematic ways.⁷ Aside from being concerned with their own payoffs, subjects appear to be concerned also with the payoffs of others. Preferences having this property are referred to as being social interdependent or other-regarding. The contents of self-interest are usually not specified by economic theory. Self-interest is not all of human motivation, and it should be recognized as a special case. Departures from the standard self-interest assumption are potentially important for economics.⁸

In conventional economic theory, an individual is assumed to be calculating how to maximize his or her own exogenously given utility independently of any psychological factor. In real life, however, people are affected by psychological and emotional factors. People feel loyalty or jealousy in society, which appears in economic literature only rarely.⁹ Emotions generate behavior (Elster 1998; Frank 1988). An internal norm is a pattern of behavior enforced in part by internal sanctions, including shame, guilt, and loss of self-esteem, as opposed to purely external sanctions, such as material rewards and punishments (Gintis 2003). Neglecting these psychological factors creates the serious risk that economists may not understand the diversity of norms and the changes in economic performances that are induced by changes in norms.

The purpose of this appendix is to go one step further. The analysis deviates from standard incentive theory because it focuses on the evolution of norms. Human beings are embedded in social structures. Individuals can acquire and internalize norms through socialization. Ben-Ner and Putterman (1998, 2000) suggest that the conjunction of (a) a postulated genetic basis for human behavioral predispositions and (b) the demonstrable impact of environment on phenotypic variation in behavior opens up the possibility of a scientific research program for studying the influence of human environments on human preferences. Geneticists developed the distinction between a genotype, the sum of the genetic instructions provided to an organism, and a phenotype, the realized organism dependent on the interaction of those instructions with a particular environment. A similar distinction will be useful for the study of internal norms. I identify socialization as a socio-economic action. Norms acquired in any given environment become generalized reasons for behavior.¹⁰ Values and preferences are transmitted through the internalization of norms. The human mind is an evolved information-processing mechanism, and through this mechanism, adaptation takes place (Tooby and Cosmides 1992). Individuals behave in conformity with a particular social norm in the long run.

Norms are acquired and internalized through an adaptation and imitation process. I introduce the probability of socialization, that is, a way that chance events affect population dynamics. Agents in a certain group are randomly paired to interact. Individuals try to socialize themselves to particular norms, and the actual formation of individual norms is influenced by the distribution of individuals with certain norms in the group. The socio-economic choices of agents with norms are self-enforcing, in the sense that the internalization of norms depends positively on the initial prevalence of agents with certain norms. The process by which norms evolve exhibits generalized increasing returns.¹¹ Norms generally take the form of conventions in the group. In the long run, only one social norm persists. In this context, the social norm is an important element in the dynamics of economic performances because morale is norm-based motivation in organizations.

The model I propose in this appendix has the following features. There are a finite number of workers in an organization. The strategy of an individual is his or her effort level. An individual’s utility function captures self-interest and ‘other-regarding’ preferences. The individual’s underlying motivation is modeled explicitly. Two possible types of other-regarding preferences, which exhibit ‘envy’ and ‘altruism,’ are considered. Each psychological factor, which is denoted by a parameter, enforces a certain internal norm. Individuals are initially endowed with their norms, but they can acquire and internalize norms through socialization. In order to trace the evolution of norms through socialization, I formulate a random-matching model in the group. While individuals act to maximize their utility, it is the prevalence of norms that determines the social norm they obey in the long run. Individuals with different norms will typically take different actions. Chance events have large and persistent effects due to positive feedbacks. It is shown that different social norms generate different equilibrium effort levels within an organization. We have two steady states in the long run: an equilibrium with high morale and an equilibrium with low morale. This multiplicity can explain why economic performance varies across organizations.

Some recent literature has started to analyze the evolution of norms, considering other-regarding preferences or psychological factors. Eaton and Eswaran (2003) examine how preferences evolve by natural selection. They demonstrate that evolutionarily stable preferences can exhibit envy in a given environment. In preference maximization, players are not choosing effort levels to maximize relative fitness. In Kandori (2003), the norm evolves over time, according to the actual effort levels taken by individuals. He shows that, owing to random shocks, the system moves from equilibria with higher effort to those with lower effort. In contrast to this appendix, however, they do not consider that norms or preferences are acquired and internalized through socialization. In the dynamics of the pattern of norms, the model presented here shows that effort levels can vary, depending on the prevailing norm within an organization.

The section ‘The Model’ in Appendix 1 develops the basic model, which characterizes a morale equilibrium and the evolution of norms in a given circumstance. The section ‘Morale and Norms in the Long Run’ in shows the diversity of morale and norms in the long run. The section ‘Path-dependent Inefficiency’ discusses the relationship of this model to some others in the literature. The section ‘Concluding Remarks’ concludes the appendix.

2.1.2 The Model

2.1.2.1 Basic Framework

We consider a group in which there are N individuals.¹² Each individual j chooses his or her effort level, e _j (≥ 0), j ∈ {1, …, N}. Suppose that output is some function of each agent’s effort, given by f(e), where e is an N-dimensional vector of agents’ effort levels. Consider a work situation in which each agent’s compensation is determined as f(e)/N. It is painful to put forth effort, and the pain that an agent feels is given by c(e _j ) that is exogenous, where c′ > 0 and c″ > 0.

Agent i’s utility function is given by:

$$ {U}_i\left(\mathrm{e};{\mu}_i\right)=\frac{f\left(\mathrm{e}\right)}{N}-c\left({e}_i\right)+{\mu}_i\cdot \left(N-1\right)\cdot \frac{f\left(\mathrm{e}\right)}{N}. $$

(2.1)

The first two terms of the right-hand side of (2.1) capture self-interest, and the last term represents ‘other-regarding’ preferences. The parameter μ _i is the weight (or the ‘binding power’ of the internal norm) that agent i’s preferences put on all other agents’ well-being. Also, (N − 1)·f(e)/N is the aggregate compensations of all agents other than i in the N-member group under consideration. The last term is an attempt to formalize the discussion of internal norms, which are social and endogenous.

Suppose there are two possible types of norms that can be internalized by individuals. We say agent i’s internal norm is enforced by ‘envy’ if μ _i ≡ μ ⁻ < 0.¹³ With the binding power of this sort, i’s utility is a decreasing function of the compensations of other agents. On the other hand, we say i’s internal norm is enforced by ‘altruism’ if μ _i ≡ μ ⁺ > 0. Then, i’s utility is an increasing function of the compensations of others. Initially, each agent is endowed with a particular norm.

Each agent chooses his or her effort level, given the parameter μ _i . As the utility function U _i is strictly concave in e _i and the efforts are continuous, there would be a unique utility maximizer for U _i . High (or low) effort level corresponds to high (or low) morale in the group. We next present an equilibrium concept somewhat different from the standard Nash equilibrium of a game.¹⁴

Definition.

Given the binding power of i’s internal norm μ _i , an effort profile e^* is a morale equilibrium if:

$$ {U}_i\left({\mathrm{e}}^{*};\ {\mu}_i\right)\ge {U}_i\left({e}_i,\kern0.1em {\mathrm{e}}_{-i}^{*};\kern0.1em {\mu}_i\right),\forall i,\forall {e}_i. $$

A morale equilibrium holds so that each agent takes others’ effort as given. It is necessary to compare morale equilibria, one with μ ⁺ and one with μ ⁻. We will show that equilibrium effort is higher with μ ⁺ ( > 0) than it would be with μ ⁻ ( < 0).

For analytical simplicity, we assume that f (e) = Σ _j e _j and c(e _i ) = e _i ²/ 2. Thus, (2.1) can be rewritten as:

$$ {U}_i\left(\mathrm{e};\ {\mu}_i\right)=\frac{\varSigma_j{e}_i}{N}-\frac{e_i^2}{2}+{\mu}_i\left(N-1\right)\frac{\varSigma_j{e}_i}{N}. $$

2.1.2.2 The Evolution of Norms

The model is constructed in discrete time, indexed by t = 0, 1, 2, ….

We model the evolution of norms, namely, those that are acquired and internalized through socialization in the group. The model represents each individual’s updating as a process of switching from one norm to another through learning from others. We introduce the socio-economic interaction between individuals in the group. At the beginning of time t, the number of agents with trait μ ⁺ ( > 0), that is, the binding power of the internal norm enforced by altruism, is denoted by n(t). Let τ(t) be the fraction of individuals with trait μ ⁺ at the beginning of time t, that is, τ(t) = n(t) / N. Thus, the fraction of individuals with trait μ ⁻ ( < 0), that is, the binding power of the internal norm which is enforced by envy, is 1 − τ(t) = (N − n(t)) / N at the beginning of t.

We formulate a random-matching model in the group. The model introduces chance events that affect population dynamics through socialization. Individuals try to socialize themselves to particular norms at the beginning of each time. Socialization occurs as follows. With a certain probability of socialization, agent i is matched with an individual in the group and adopts then the norm of that individual. The probability of socialization is allowed to depend on the fraction of individuals in the group. In this model, socialization acts as a type of complementarities.¹⁵ That is, an agent with trait μ ⁻ (or μ ⁺) has more incentives to socialize with an individual with trait μ ⁺ (or μ ⁻) as μ ⁺ (or μ ⁻) is the more widely dominant one in the group. Specifically, socialization to agent i with trait μ ⁻ in t occurs with probability q(τ(t)), which is a continuous, strictly increasing function in τ(t). On the other hand, socialization to agent i with trait μ ⁺ in t occurs with probability q(1 − τ(t)), which is a continuous, strictly increasing function in 1 − τ(t). With probability 1 − q(τ(t)) or 1 − q(1 − τ(t)), agent i is ‘naive’ and does not get randomly matched with somebody else in the group. Moreover, it is assumed that q(0) = 0 and q(1) = 1.

Let P ^{+ −}(t) denote the transmission probability that an agent with trait μ ⁺ is socialized to trait μ ⁻ at the end of time t. The socialization mechanism is then characterized by the following transition probability:

$$ {P}^{+-}(t)=q\left(1-\tau (t)\right)\cdot \left\{\frac{N-n(t)}{N-1}\right\}. $$

(2.2)

This represents the coefficient of norm transmission, where an agent with trait μ ⁺ (at the beginning of t) is matched with an individual with trait μ ⁻ of the remaining individuals, and adopts and internalizes that trait at the end of t. The transition probability, P ^{+ +}(t), that an agent with μ ⁺ sticks to that trait in the socialization mechanism is given by:

$$ {P}^{++}(t)=1-q\left(1-\tau (t)\right)+q\left(1-\tau (t)\right)\cdot \left\{\frac{n(t)-1}{N-1}\right\}. $$

(2.3)

In the term 1 − q(1 − τ(t)) of the right-hand side of (2.3), an agent with μ ⁺ does not get matched with somebody else in the group and sticks to that trait in t. In the last term, an agent with μ ⁺ at the beginning of t has a chance of socialization, but is matched with an individual with μ ⁺ of the remaining individuals in the group (the agent does not change his or her trait).

Similarly, for an agent with trait μ ⁻ at the beginning of t, we get:

$$ {P}^{-+}(t)=q\left(\tau (t)\right)\cdot \left\{\frac{n(t)}{N-1}\right\}, $$

(2.4)

$$ {P}^{--}(t)=1-q\left(\tau (t)\right)+q\left(\tau (t)\right)\cdot \left\{\frac{N-n(t)-1}{N-1}\right\}. $$

(2.5)

Given these transmission probabilities, the fraction τ(t + 1) of individuals with trait μ ⁺ at the beginning of time t + 1 is calculated to be:

$$ \tau \left(t+1\right)=\tau (t)+{P}^{-+}\left(\mathrm{t}\right)\cdot \left(1-\tau (t)\right)-{P}^{+-}(t)\cdot \tau (t), $$

substituting (2.2) and (2.4), we obtain:

$$ \tau \left(t+1\right)-\tau (t)=\left\{\frac{N-n(t)}{N-1}\right\}\cdot \tau (t)\cdot \left\{q\left(\tau (t)\right)-q\left(1-\tau (t)\right)\right\}. $$

(2.6)

This is the equation for the population dynamics through socialization. It is easy to derive conditions that guarantee that the population dynamics of (2.6) converges to a ‘homogeneous’ population in the limit. In Fig. 2.1, the probabilities of socialization, q(τ) and q(1 − τ), are measured vertically: q(τ) slopes upward to τ and q(1 − τ) slopes downward to τ. And q(1 − τ) is a mirror image of q(τ). Assume that τ(0) > 1/2 holds initially. Then, we have ∂{τ(t + 1) − τ(t)}/∂τ(t) > 0 from (2.6). Thus, we have τ(t) → 1 for any τ(0) > 1/2. Everyone adopts the same norm eventually. Similarly, the dynamics implies that τ(t) → 0 for any τ(0) < 1/2.

Fig. 2.1.

Population dynamics through socialization

To summarize:

Lemma 1.1

(Teraji 2007).

For the fraction of individuals having the internal norm which is enforced by altruism, τ(t), in the group at the beginning of time t,

(1) τ(t) converges to 1 if τ(0) > 1/2 holds initially,

and

(2) τ(t) converges to 0 if τ(0) < 1/2 holds initially.

Thus, the process by which norms evolve through socialization exhibits generalized increasing returns. Social norms take the form of conventions, that is, the modes of behavior to which a majority of agents subscribe. People follow a certain norm through socialization. In the presence of generalized increasing returns, chance events have large and persistent effects due to positive feedbacks. In the self-enforcing process that depends on the initial prevalence of norms, only one social norm emerges. The social norm, which is enforced by altruism or envy, persists in the long run if a majority of agents adopt that norm initially. In the socialization mechanism, individuals behave in conformity with a particular social norm eventually.

2.1.3 Morale and Norms in the Long Run

Agent i’s utility function, whose binding power of the internal norm is μ ⁺ or μ ⁻ at the beginning of time t, is given by:

$\begin{array}{l}{U}_i\left(\mathrm{e}(t);{\mu}^{+}\right)=\frac{\varSigma_j{e}_j(t)}{N}-\frac{e_i^2(t)}{2}\\ {}\kern4.25em +\left({P}^{++}(t)\cdot {\mu}^{+}+{P}^{+-}(t)\cdot {\mu}^{-}\right)\cdot \left(N-1\right)\cdot \frac{\varSigma_j{e}_j(t)}{N},\end{array}$

$$ \begin{array}{l}{U}_i\left(\mathrm{e}(t);{\mu}^{-}\right)=\frac{\varSigma_j{e}_j(t)}{N}-\frac{e_i^2(t)}{2}\\ {}\kern4.25em +\left({P}^{-+}(t)\cdot {\mu}^{+}+{P}^{--}(t)\cdot {\mu}^{-}\right)\cdot \left(N-1\right)\cdot \frac{\varSigma_j{e}_j(t)}{N},\end{array} $$

at the end of t. Agent i chooses i’s effort level at the end of each time. Then, there is a unique effort level maximizing the utility function. The utility-maximizing effort levels for agent i in t, e _i ^∗(t; μ ⁻) and e _i ^∗(t; μ ⁺), are given by:

$$ {e}_i^{*}\left(t;{\mu}^{+}\right)= ArgMa{x}_{e_i\ge 0}{U}_i\left(\mathrm{e}(t);{\mu}^{+}\right), $$

$$ {e}_i^{*}\left(t;{\mu}^{-}\right)= ArgMa{x}_{e_i\ge 0}{U}_i\left(\mathrm{e}(t);{\mu}^{-}\right). $$

The first-order conditions are:

$$ {e}_i^{*}\left(t;{\mu}^{+}\right)=\frac{1}{N}+\left({P}^{++}(t)\cdot {\mu}^{+}+{P}^{+-}(t)\cdot {\mu}^{-}\right)\cdot \frac{\left(N-1\right)}{N}, $$

(2.7)

$$ {e}_i^{*}\left(t;{\mu}^{-}\right)=\frac{1}{N}+\left({P}^{-+}(t)\cdot {\mu}^{+}+{P}^{--}(t)\cdot {\mu}^{-}\right)\cdot \frac{\left(N-1\right)}{N}. $$

(2.8)

Suppose that τ(0) > 1/2 holds initially. From lemma, τ(t) → 1 in the limit. Then, we have P ^{+ +}(t) → 1 and P ^{+ −}(t) → 0 in (2.2) and (2.3). Thus, it holds that e _i ^∗(t; μ ⁺) → {1 + μ ⁺ · (N − 1)}/N in (2.7). Furthermore, in (2.4) and (2.5), n(t) / (N − 1) → 1 and (N − n(t) − 1)/(N − 1) → 0 for the remaining individuals other than agent i. Then, we have P ^{− +}(t) → 1 and P ^{− −}(t) → 0. Thus, e _i ^∗(t; μ ⁻) → {1 + μ ⁺ · (N − 1)} / N in (2.8). Similarly, if τ(0) < 1/2 holds initially, we have e _i ^∗(t; μ ⁺) → {1 + μ ⁻ · (N − 1)} / N and e _i ^∗(t; μ ⁻) → {1 + μ ⁻ · (N − 1)} / N in the limit.

Morale is norm-based motivation. The social norm generates the equilibrium effort level in the long run. Everyone chooses the same effort level in equilibrium because of a particular social norm that prevails. When there are multiple equilibria, these equilibria can be ranked. Morale is higher in the high-effort equilibrium than in the low-effort equilibrium. Thus, we have two morale equilibria: in one equilibrium with high morale, every agent’s effort level is {1 + μ ⁺·(N − 1)} / N, and in the other equilibrium with low morale, every agent’s effort level is {1 + μ ⁻·(N − 1)} / N, where μ ⁺ > 0 > μ ⁻.

Thus, we have the following:

Proposition 2.1

(Teraji 2007).

(1) Suppose that τ(0) > 1/2 holds initially. Then, we have a morale equilibrium e = (e, …, e), where e = {1 + μ ⁺·(N − 1)} / N.

(2) Suppose that τ(0) < 1/2 holds initially. Then, we have a morale equilibrium e = (e, …, e), where e = {1 + μ ⁻·(N − 1)} / N.

The evolution of norms allows us to study long-run equilibria with different effort levels. Which norm is persistent determines the equilibrium effort level within an organization. If the norm is enforced by altruism, every agent chooses the high equilibrium effort level in the long run. In this morale equilibrium, every agent is able to maintain high morale. Thus, norms inculcated through socialization can lead members to create incentives within an organization. On the other hand, if the norm, which is enforced by envy, persists, the low equilibrium effort level can be sustained. In this morale equilibrium, everyone shirks. Then, we have the decay of economic performance in the long run. Thus, the economic performance can vary depending on the norm that prevails within an organization.

2.1.4 Path-Dependent Inefficiency

A key finding of path dependence is a property of ‘lock-in’ by historical events (Arthur 1994). In a world of increasing returns to scale (positive feedbacks), initial and trivial circumstances can have important and irreversible influences on the ultimate market allocation of resources. Path-dependent economics is a theory of equilibrium selection with positive feedbacks, and path-dependent outcome is associated with characteristics of persistence and uniformity. The economy has a multiplicity of possible equilibrium solutions, where the dominant solution can be the suboptimal one.

The form of path dependence conflicts with the neoclassical model, where efficient, and therefore predictable, outcomes are attained. Why does an inefficient outcome persist? Where there is a feasible improvement to be gained from moving onto a better path, economic agents will be willing to pay to bring the improvement about. This issue has been addressed by Liebowitz and Margolis (1990, 1994).

In this appendix, effort depends on the prevailing norm within an organization. In a world of effort discretion, the economic performance can vary across organizations. Furthermore, the process by which norms evolve through socialization exhibits generalized increasing returns. One social norm emerges in the process that depends on the initial prevalence of norms. The social norm, enforced by altruism or envy, persists in the long run. Individuals try to socialize themselves to one particular norm. Which norm is persistent determines the equilibrium effort level. It is possible for low-effort organizations to survive in the long run. Then, the norm, enforced by envy, can be a cause of economic inefficiency. If the norm is enforced by envy, there does not exist the private incentives for individuals to choose the high-effort equilibrium. In a world where norms are determinants to effort levels, the path-dependent low-effort equilibrium can exist.

These discussions connect with what Harvey Leibenstein referred to as x-inefficiency. A large number of x-inefficiency studies exist, dating from Leibenstein (1966). He referred to the difference between maximal effectiveness of utilization and actual utilization as the degree of x-inefficiency. According to him, individuals supply different amounts of effort, where effort is a multidimensional variable, under different organizational and environmental circumstances. Individual motivations and interactions between individuals are important. He presented a reasonable vision of human behavior within the organizational context, where x-inefficiency can persist stubbornly. Furthermore, Altman (2000) introduces effort discretion into the path-dependency modeling. Given effort discretion, the quantity and quality of effort can vary. Labor productivity is affected by the quantity and quality of effort inputted into the process of production. The introduction of effort variability allows for the existence of path-dependent high and low productivity firms in the long run.

2.1.5 Concluding Remarks

Individuals may make different decisions, depending on social contexts. Differences in context-dependent behavior can be interpreted as being generated psychologically, ethically, and sociologically.

In this appendix, I have provided theoretical insights into the diversity of economic performances within organizations. The approach taken here can shed light on issues on endogenous norms in society. I have argued that diverse morale and the evolution of norms interact in nontrivial ways. Agents can acquire and internalize norms through socialization. The process by which norms evolve through socialization is self-enforcing, and depends on the initial distribution of individuals with certain norms in the group. In this formulation, different norms typically exhibit different morale equilibria in the long run. The resulting dominance of a certain norm implies that a certain social structure appears more often than otherwise. I have proposed two morale equilibria with different effort levels. If the norm, enforced by altruism, persists in the limit, a ‘good’ morale equilibrium emerges, in which everyone chooses a high effort level and high morale is sustainable within an organization. Norms inculcated through socialization can lead individuals to create incentives. On the other hand, if the norm, enforced by envy, persists in the limit, a ‘bad’ morale equilibrium emerges, in which everyone shirks. Morale can thus be norm-based motivation. Given effort discretion, it is possible to expect a multiplicity of possible equilibrium solutions, where the dominant solution is suboptimal. The economic performance can vary because of the norm that prevails in organizations.

Appendix 2: A Theory of Norm Compliance: Punishment and Reputation

Appendix 2 is based on Teraji (2013).

2.1.1 Introduction

Societies have social norms or, in short, norms; members of the society are required to follow standards of behavior. Social norms are informal rules, as opposed to formal, legal rules promulgated by a court or a legislature. Social norms often direct individuals to undertake actions that are inconsistent with selfish actions. For example, in the dictator game, 50–50 division is generally viewed as norm-compliant (Andreoni and Bernheim 2009).¹⁶ People may deviate from such norms. In the case of legal compliance, individual incentives most often refer to deterrence (Becker 1968). That is, individuals are deterred from criminal activities by a higher fine and by a higher probability of conviction. Unlike legal rules, social norms are not supported by formal sanctions. Why do social norms not simply collapse from the violation? This appendix studies two distinct mechanisms on norm compliance. The incentive to comply with norms derives not only from the enforcement of costly punishment by others but also from reputation building for oneself.

The importance of decentralized punishments (i.e., punishments carried out by individuals without the intervention of a central authority) is documented in experimental studies. Ostrom et al. (1992) show the existence of such punishment opportunities in a common-pool resource use game. The fear of punishment has a positive effect on cooperation. In public goods experiments, subjects begin by contributing on average about half of their endowments to the public account. However, the level of contribution decays over the course of multiple rounds (Andreoni 1995). When costly punishment is permitted, cooperation does not deteriorate. Fehr and Gächter (2000, 2002) indicate that many individuals are willing to punish unfair behavior at a personal cost in public goods games. Costly punishment is administered by ‘third parties’ (Fehr and Fischbacher 2004). Potential punishers are not themselves the victims but have merely witnessed unfair behavior. This is called ‘altruistic’ punishment as individuals sacrifice for no direct benefits (Gintis et al. 2003). It suggests that cooperation has evolved through the sacrifice of altruistic punishers who are ready to incur some costs to prevent unfair behavior. People are fair because they have a psychological motivation to restore fairness.¹⁷ Then, punishment can be seen as a consequence of a sense of fairness.

Certain groups of individuals can maintain a strong reputation over time. Akerlof (1980) develops an economic model to show that social norms that involve pecuniary disadvantage to individuals may persist without erosion.¹⁸ Disobedience to the norm may involve a loss of reputation. People want to achieve the reputation of being fair. People are fair because they care about their reputation. They may not be genuinely fair. They have to be rewarded for good reputation, and they have to be willing to comply with the norm. Individuals are influenced in their convictions by what they think others will do.¹⁹ Conformity to the norm is conditional on expectations about other peoples’ behavior. Norms are constituted by expectations shared by members in a population and are jointly recognized among them.²⁰

Thus, we have a set of solutions to the problem of norm compliance. The first solution is the punishment-based account. Following this account, people comply with norms because the threat of punishment makes it in their interest to do so. Altruistic punishment seems to have a solid foundation in human interaction. However, such costly punishment leads to a large increase in losses for altruistic punishers. The second solution is, on the other hand, the reputation-based account. Social norms can be sustained if the pecuniary advantage from breaking norms is not sufficient to offset the forgone reputation effect. This is related to indirect reciprocity. According to Alexander (1987), indirect reciprocity is arranged in the form of a chain; a person is eventually helped by someone else who may not have been directly helped by him.²¹ Altruistic actions can be sustained if people who support others receive support in return. To achieve such indirect reciprocity, building up a positive reputation is needed.²²

This appendix analyzes the interaction between the potential for costly punishment and building personal reputation. The model considers two groups of agents in a society with one norm. Agents in one group (group i) choose whether to comply with the norm by incurring some cost. They acquire utility from the reputation derived from complying with the norm. This utility depends positively on the proportion of motivated compliers. Individuals may differ in their motivation to comply with the norm. Punishment will be imposed on individuals who deviate from the norm. In the other group (group j), there are agents who value compliance and potentially punish noncompliance (i.e., the sanctioning individuals). The model investigates individual punishment decisions. Agents choose to punish violators at some cost (decentralized punishment).

This appendix asks how individual values evolve endogenously over time and analyzes the long-run dynamics of norm formation. The present framework systematically investigates the different forces to account for the long-run stability of the norm. There are two scenarios as follows. In one scenario, there is some possibility that the erosion of the reputation effect induces individuals to break the norm. Then, the norm is enforced due to a higher level of punishment of noncompliance. Punishment would be used to enforce the norm if a substantial fraction of people has little reputation-derived utility by obeying the norm. This appendix, however, suggests that altruistic punishment may play a limited role in sustaining the norm. In another scenario, everyone is motivated due to reputation formation despite a lower level of punishment by others. For a lower level of punishment, effective reputation building provides a way to sustain the norm.

The section ‘The Model’ in Appendix 2 presents a simple model to consider the problem of norm compliance. The section ‘Analysis’ investigates two scenarios to account for the long-run stability of the norm. The section ‘Concluding Remarks’ concludes the appendix.

2.1.2 The Model

Consider a society populated by a continuum of agents at each period of time t. The population size is constant over time and normalized to 1. We assume that the population is composed of two groups, i and j, which differ in characteristics. The first group i amounts to a share 1/2 of the whole population. Accordingly, the half part of the population belongs to group j. Matching between two groups take place randomly. A member of group i randomly matches with the opponent of group j at each period. This may suggest a large society in which one-shot encounters with unrelated strangers are common and information is rarely transparent.

There is one norm in the society. For simplicity, we assume that it is only possible to either follow the norm or not. An agent of group i (‘he’) chooses an action x ∈ {0, 1} at each period. His action is represented by a discrete variable, one or zero. That is, x = 1 if the agent of group i complies with the norm, and x = 0 if he violates it. Inertia is introduced with the assumption that every agent of group i cannot switch actions at each point in time. He must make a commitment to a particular action in the short run. Opportunities to switch actions arrive randomly; some fraction α, 0 < α < 1, of individuals is drawn randomly from group i and makes a new choice of either x = 1 or x = 0. Thus, we may interpret a norm as a prescription indicating how a person ought to behave at any situation at which he may be called to move.

An agent in group i may deviate from the norm, but this deviation is costly. Punishment (such as ostracism) will be imposed on group i members who deviate from the norm. If deviation from the norm is observed (x = 0), the opponent of group j (‘she’) decides whether to punish the deviator, that is, chooses p on the closed interval [0, 1]. Here, with probability p the agent of group j punishes noncompliance, and with probability 1 – p, she does not. The agent of group j is only an outside party who happens to know that norm violation has occurred. Altruistic punishment is motivated to restore norm compliance even though they are not expected to interact again in the future. It is costly for agents of group j to punish norm violators. Thus, in the model, agents in group i choose whether to comply with the norm, while agents in group j value compliance and potentially punish violators. Punishment is then confined to interactions with others who share the same norm in the population.

Individuals in group i are assumed to be heterogeneous with respect to their social concerns. We denote an agent’s type of group i as g. An agent of type g = 0 is not concerned with the social meaning of a certain action. A higher g implies higher social concerns. The distribution of g is assumed to be exogenous and uniform in the model. Let the uniform distribution of g be F(g), with g ∈ [0, G]. Furthermore, let the density of the uniform distribution be f = 1/G.

In group i, the short-run utility function of an agent of type g is given by:

$$ U=x\left\{R\left(g,\mu \right)-D\right\}-\left(1-x\right)pC. $$

(2.9)

Each individual in group i is assumed to maximize the utility function (2.9), which constitutes the short-run equilibrium. The first term of the utility function (2.9) reflects the reputation value of a norm if the agent of group i chooses to comply with the norm (x = 1). It depends positively on the agent’s type g and the fraction μ of individuals who comply with the norm in group i. The agent internalizes the norm with the reputation value R. The strength of the norm is determined by the proportion of agents who follow it. Group i consists of μ compliers and (1/2 − μ) violators, where 0 < μ < 1/2. The fraction μ is given in the short run and is known by all members of group i. A person who deviates from the norm forgoes reputation-derived utility. The second term reflects a fixed cost, D > 0, of compliance if the agent of group i chooses x = 1. The final term reflects a fixed cost, C > 0, as the penalty for violation if the agent of group i chooses not to comply with the norm (x = 0) and the opponent of group j chooses to punish him with probability p. The punisher may disapprove of the violator, which reduces his level of satisfaction and his well-being.²³ To sum up, the agent of group i has to pay a compliance cost, D, if x = 1 and the penalty for violation, C, if (1 − x) p > 0. It is assumed that D > C.

The reputation value R is assumed to take the following form:

$$ R\left(g,\mu \right)=g\mu . $$

(2.10)

Absent social concerns (g = 0), the agent of group i would not follow the norm. An agent in group i will consider following the norm when the value of his action is greater than the cost of violation. The higher g is and the higher μ (the proportion of norm followers) is, the greater is the utility from following the norm. The value of following it may take the form of abstract reward, such as social approval from others. The reputation value can also be interpreted as the value of social image. People care about how others perceive them, and these concerns influence a wide range of decisions (Andreoni and Bernheim 2009).

The model also encompasses individual punishers who are not direct victims for norm violation. They punish violators to restore norm compliance. For an agent of group j, whether her opponent in group i follows the norm or not is a random variable. Let the probability that her opponent in group i chooses to comply with the norm be Pr(x = 1). The short-run utility function of an individual in group j is given by:

$$ V= \Pr \left(x=1\right)B-\left\{1- \Pr \left(x=1\right)\right\}pT. $$

(2.11)

Each individual in group j is assumed to maximize the utility function (2.11), which constitutes the short-run equilibrium. The first term of the utility function (2.11) reflects a psychological benefit, B > 0, received by the agent of group j when her opponent from group i chooses to follow the norm with probability Pr(x = 1). The second term is a cost T (> B) associated with punishment (p > 0) when her opponent from group i violates the norm with probability 1 − Pr(x = 1). Punishment can be psychologically costly. Punishers suffer a psychological cost if their opponents violate the norm. This psychological cost includes the punisher’s anger (negative emotion) for violation of the norm.

The benefit-cost ratio b for the agent of group j is given by:

$$ b=\frac{B}{T}. $$

(2.12)

As the punisher’s anger (T) increases, the benefit-cost ratio (b) decreases.

The reputation value of the norm affects individual short-run utility as defined in (2.9). Individuals of group i choose to comply with the norm if the utility from so doing is as great as the utility derived from not complying. The reputation value of the norm depends on the proportion of compliers in group i. At each period, the fraction (1/2)α of individuals in group i has a chance of making a new choice. In the aggregate, the proportion of compliers is (1/2)αPr(x = 1), and the proportion of violators is (1/2)α{1 − Pr(x = 1)}. Here, α parameterizes the inertia of the adjustment process. In the adjustment process, it takes some time until individuals find out how many others follow the norm. The proportion μ will increase (or decrease) if (1/2)αPr(x = 1) is larger (or smaller) than (1/2)α{1 − Pr(x = 1)} in group i. Therefore, the long-run adjustment process is described by the following form:

$$ \frac{d\mu }{dt}=\alpha \left[\frac{1}{2} \Pr \left(x=1\right)-\frac{1}{2}\left\{1- \Pr \left(x=1\right)\right\}\right] $$

From the above equation. we have:

$$ \frac{d\mu }{dt}=\alpha \left[ \Pr \left(x=1\right)-\frac{1}{2}\right] $$

(2.13)

Thus, the speed of norm evolution is influenced by the rate α at which the group members acquire the knowledge necessary to appreciate the norm. If Pr(x = 1) differs from 1/2, the proportion μ will increase or decrease at rate α until μ is equal to 0 or 1/2. The relationship between short run and long run is summarized as follows. At each period t, the proportion of norm followers, μ, is given, and agents of group i decide whether to follow the norm or not, given this proportion (short run). The proportion μ is formed over a period of time, and it converges to a steady state (long run).

2.1.3 Analysis

2.1.3.1 Choice of x

An agent, who is drawn randomly from group i, chooses x ∈ {0, 1} to maximize his utility function:

$$ U=x\left(g\mu -D\right)-\left(1-x\right)pC, $$

from (2.9) and (2.10). Maximization of this utility function constitutes the short-run equilibrium. If g μ − D > − p C, he chooses to comply with the norm (x = 1); otherwise, he does not (x = 0). Thus, his best-response function is:

$$ x=\Big\{\begin{array}{cc}1& \mathrm{if}g\mu -D+pC>0,\ \mathrm{or}\ g>\frac{D-pC}{\mu}\\ {}0& \mathrm{if}g\mu -D+pC\le 0,\ \mathrm{or}\ g\le \frac{D-pC}{\mu}\end{array} $$

It is useful to consider the probability of heterogeneous individuals choosing to comply with the norm in group i, Pr(x = 1). Then,

$$ \begin{array}{l} \Pr \left(x=1\right)= \Pr \left(\frac{D-pC}{\mu }<g\right)\hfill \\ {}\kern0.5em =1-F\left(\frac{D-pC}{\mu}\right).\hfill \end{array} $$

Using the density of the uniform distribution, f = 1/G, we have

$$ \Pr \left(x=1\right)=1-\frac{D-pC}{\mu }f. $$

(2.14)

Thus, in the short run, the probability to comply with the norm, Pr(x = 1), is a continuous increasing function in the probability of punishment, p. Only in the long run, the incentive to comply with the norm adapts to the fraction of norm followers, μ. The strength of the norm is endogenously determined by the collective behavior of the group members.

2.1.3.2 Choice of p

An agent in group j optimally chooses to punish the norm deviator in group i with the probability of punishment, p. That is, at each period, the agent of group j chooses p ∈ [0, 1] to maximize her utility function:

$$ V=\left(1-\frac{D-pC}{\mu }f\right)B-\left(\frac{D-pC}{\mu }f\right)pT $$

from (2.11) and (2.14). This maximization also constitutes the short-run equilibrium. The first-order condition is

$$ \frac{\partial V}{\partial p}=\frac{C}{\mu }fB-\frac{D-pC}{\mu }fT+\frac{pC}{\mu }f\kern0.1em T=0. $$

which, for an interior solution, equates the marginal cost of punishment to the marginal benefit of punishment. Using the benefit-cost ratio b in (2.12), we have:

$$ p= \min \left\{\frac{D-bC}{2C},1\right\} $$

(2.15)

For (D − bC) / 2C < 1 in condition (2.15), it follows that:

$$ \frac{\partial p}{\partial b}=-\frac{1}{2}<0. $$

A lower benefit-cost ratio (b) implies a higher probability of punishment (p). In Fig. 2.2, if b ₁ > b ₂, then p ₁ < p ₂. As the punisher’s anger for the norm deviation increases (or T increases), the ratio b decreases (see (2.12)) and the probability p increases. Once a norm deviation is observed, an agent of group j feels anger toward the norm deviator and attempts to reduce his payoff. Thus, anger plays an important role in punishment decisions.

Fig. 2.2

The relationship between b and p

2.1.3.3 Evolution of μ

The proportion of individuals who comply with the norm changes over time. From (2.13) and (2.14), the long-run adjustment process is described by:

$$ \frac{d\mu }{dt}=\alpha \left[\frac{1}{2}-\frac{D-pC}{\mu }f\right]. $$

(2.16)

Using (2.16), a critical value of μ, μ ^*, is defined as follows:

$$ {\mu}^{*}= \min \left\{2\left(D-pC\right)f,\frac{1}{2}\right\}. $$

(2.17)

Condition (2.17) describes the critical value μ ^* as a decreasing function of the probability of punishment, p, for 2(D − pC) f < 1/ 2. Then, the value μ ^* in condition (2.17) does not include the benefit-cost ratio b.

The adjustment process is not necessarily continuous. The proportion μ ( > 0) is decreasing for μ < μ ^*. Then, the adjustment process converges to μ = 0 (everyone in group i deviates from the norm). On the other hand, if μ ^* is less than 1/2, μ is increasing for μ > μ ^*. Then, the adjustment process converges to μ = 1/2 (everyone in group i follows the norm). Thus, there are two stable equilibrium points.

In an (μ ^*, p)-diagram shown in Fig. 2.3, there is one intersection between conditions (2.15) and (2.17). Condition (2.15) describes the probability of punishment (p) for a given benefit-cost ratio (b), which is a straight line; while condition (2.15) describes the critical value μ ^* as a function of p, which is a declining line. Figure 2.3 also describes the effect of a higher probability of punishment (p) due to a lower benefit-cost ratio (b). A lower b (or a higher T) implies a higher p in Fig. 2.2. In Fig. 2.3, a higher p implies that the intersection point between the two lines shifts to the left, which implies a smaller critical level of μ ^* (<1/2). When μ is above the critical value μ ^*, the adjustment process converges to μ = 1/2. Anticipating a higher probability of punishment, more agents of group i consider following the norm.

Fig. 2.3.

The increase in p

On the other hand, Fig. 2.4 describes the effect of a lower probability of punishment (p) due to a higher benefit-cost ratio (b). A higher b (or a lower T) implies a lower p in Fig. 2.2. In Fig. 2.4, a lower p implies that the intersection point between the two lines shifts to the right, which implies a larger critical level of μ ^*. If the fraction of norm followers, which is above a larger μ ^*, is larger, the adjustment process converges to μ = 1/2. For a higher reputation value, more agents of group i consider following the norm. People conform to maintain their reputation. The strength of the norm is determined by the proportion of the members who follow it. Even though p is small, individuals choose to comply with the norm. If the norm is initially widespread, the equilibrium converges to one steady state in which everyone complies with the norm. If instead the initial fraction of norm followers is small, the society ends up in another steady state with opposite features. Depending on initial conditions, the society might converge to one or the other steady state.

Fig. 2.4.

The decrease in p

We conclude the following:

Proposition 2.2

(Teraji 2013).

• (a) A higher p induces individuals to sustain the norm in the society with a smaller μ.

• (b) A larger μ induces individuals to sustain the norm in the society with a lower p.

An agent does not automatically comply with a norm but deviates whenever the gains from deviation are sufficiently large. Proposition 2.2(a) shows that the norm is enforced due to a higher level of punishment for noncompliance. The norm is enforced due to the expectation that the violation will be punished. There is some possibility that the erosion of the reputation effect induces individuals to break the norm. Punishment would be used to enforce the norm if a substantial fraction of people has little reputation-derived utility by obeying it. Punitive behavior is a way to restore norm compliance by penalizing violators. Proposition 2.2(b), on the other hand, shows that everyone is motivated due to reputation formation, despite a lower level of punishment by others. For a lower level of punishment, effective reputation building provides a way to sustain the norm. Then, an agent must believe that sufficiently many other agents will comply with the norm as well. Punishment and reputation are interacting mechanisms. Reputation mechanisms generate an environment where the execution of costly punishment is less frequent without taking away its deterring force. The interaction of two mechanisms, punishment and reputation, not only comes closer to real life but also provides a convenient way to norm compliance.

2.1.4 Concluding Remarks

Even in societies that have strong governments, norms are both a source of law and often a cheap and effective substitute for law (Posner 1997). The incentives for obeying laws are clear enough, but why do people obey norms? We develop a theoretical framework to consider the problem of norm compliance. Some norms often require people to sacrifice for the group. The existence of such norms presents an evolutionary puzzle. This appendix analyzes the interaction between the potential for costly punishment and building personal reputation. The model considers two groups, group i and group j, in a society with one norm. Agents in group i choose whether to comply with the norm by incurring some cost. They acquire utility from the reputation derived from complying with the norm. This utility depends positively on the proportion of motivated compliers in group i. Behaving in a manner that violates the norm is costly. On the other hand, in group j, there are agents who value compliance and potentially punish noncompliance. They are altruistic punishers who are not direct victims for norm violation.

We have a set of solutions to the problem of norm compliance. The first solution is the punishment-based account. Following this account, people comply with the norm because the threat of punishment makes it in their interest to do so. Altruistic punishers are ready to incur some cost to prevent unfair behavior. They have a psychological motivation to restore norm compliance. Although altruistic punishment seems to have a solid foundation in human interaction, such a punishment leads to a large increase in losses for punishers. The second solution is, on the other hand, the reputation-based account. People want to achieve the reputation of being fair. People comply with the norm because they care about their reputation.

The present framework systematically investigates two scenarios to account for the long-run stability of the norm. In one scenario, there is some possibility that the erosion of the reputation effect induces individuals to break the norm. Individuals may have a selfish interest in violating it. Then, the norm is enforced due to a higher level of punishment of noncompliance. Punishment would be used to enforce the norm if a substantial fraction of people has little reputation-derived utility by obeying it. In another scenario, everyone is motivated due to reputation formation, despite a lower level of punishment by others. For a lower level of punishment, effective reputation building provides a way to sustain the norm. The two mechanisms, punishment and reputation, are interacting. The interaction not only comes closer to real life but also provides a convenient way to norm compliance.

Appendix 3: A Model of Corporate Social Performance: Social Satisfaction and Moral Conduct

Appendix 3 is based on Teraji (2009b).

2.1.1 Introduction

Recent corporate scandals involving large companies have highlighted corporate social responsibility (CSR) as one of the principal issues confronting the free-market system. The term ‘CSR’ has been defined in various ways.²⁴ According to one of the most frequently cited definitions, CSR is “a concept whereby companies integrate social and environmental concerns in their business operations and in their interaction with their stakeholders on a voluntary basis” (Commission of the European Communities 2001, p. 6).²⁵ Companies have responsibilities to take ethical and moral issues into account while they make a profit. There is a growing interest in the relationship between corporate governance and the social performance of a company. Contemporary economic theory, however, has argued that so long as economic agents are rational and self-interested, they will be able to affect a better set of consequences than if their intentions are benevolent. In fact, according to Friedman (1970), the social responsibility of a business is to increase its profits. In Friedman’s framework, firms are owned by shareholders who are principals, and business managers are agents with the duty to serve the interests of their principals. Managers are obliged by contract to shareholder value; it is their primary task to maximize profits. Following this view, CSR is a misuse of corporate resources that would be better spent on value-added internal projects or returned to shareholders.

The assumptions of neoclassical economic theory are subject to market forces. In the neoclassical analytical framework, the market system creates a decision-making situation that favors practices to ensure profit maximization at the expense of more morally preferable alternatives. However, the lens of neoclassical economic theory is distorting practical realities. A positive view of managers’ support of CSR is articulated by Freeman (1984) who encourages consideration of external stakeholders, beyond direct profit maximization. Freeman’s broad definition of stakeholders includes any group or individual who might affect the business objective or might be affected by its realization. According to the stakeholder theory approach, it can be beneficial for the company to engage in certain CSR activities that stakeholders perceive to be important; unless stakeholders and their interests are dealt with, they might withdraw their support for the company. Freeman’s (1984) work helps to re-conceptualize the nature of the company to encourage consideration of new external stakeholders, legitimizing new forms of managerial understanding and action. CSR should be understood as a broad concept since it takes in the whole set of philosophical and normative issues relating to the role of business in society. Stakeholder theory suggests a moral relationship between companies and stakeholders.

Corporate social performance (CSP) is a voluntary activity. Moral management yields CSP. Supplying CSP is not just like managing traditional market-based activities. In fact, from empirical studies, the relationship between CSP and profits has not been adequately determined (Kolstad 2007). CSP entails a company’s recognition of broad responsibilities, and it should be concerned with more than just profit. Neoclassical economic theory, however, treats the firm as a black box, completely self-interested entrepreneur that arranges inputs so as to maximize the material well-being. Friedman (1962) questioned the ability of business managers to pursue the social interest as follows:

If businessmen do have a social responsibility other than making maximum profits for stockholders, how are they to know what it is? Can self-selected private individuals decide what the social interest is? Can they decide how great a burden they are justified in placing on themselves or their stockholders to serve that social interest? (Friedman 1962, pp. 133–134)

Instead, Friedman (1962) argues that a socially responsible firm is one that conducts the business in accord with shareholders’ desires, which generally makes as much money as possible while conforming to the basic rules of the society. His position, based on free-market ideology, has come under increasing attack since the time of writing. According to the shareholder perspective, business is about economic and not social goals. In real life, however, economic decisions also have social consequences, and hence, the boundaries between the social and economic world become blurred. Firms have responsibilities toward their environment that go beyond their legal and economic obligations. The stakeholder model reflects the modern understanding of companies as integrated in, rather than separated from, the rest of society. Within our economic system, there exists a growing cadre of companies where social, environmental, and ethical goals are on equal footing with the profit motive.

On the contrary, the behavioral theory of the firm (Cyert and March 1963; Leibenstein 1987; Altman 2001 b) can explain significant elements that are ignored in the neoclassical economic framework. A core idea in the behavioral theory of the firm is the importance of organizational slack or x-inefficiency in organizations. The firm may be operating inside of the production possibility frontier. Behavioral theory focuses on the role of discretionary resources in the corporate social strategy context. Ethical or moral issues may arise when companies engage in CSP and legitimize their activities to stakeholders. As a consequence, the ethical firm can be expected to be relatively more x-efficient, satisfying stakeholders’ interests to some extent, rather than maximizing the material well-being. Thus, the behavioral theory of the firm has the potential to incorporate ethics or morality as a critical component of the underlying motivating structure. The behavioral theory of the firm fits closely with widespread conceptions of how social concerns are dealt with in business.

According to stakeholder theory, CSR includes various kinds of responsibilities or dimensions: each dimension of CSR can be examined in relation to the various stakeholders of the organization. To consider what it means for a company to be socially responsible, this appendix focuses on the ‘socially conscious’ consumer. Managers do business in a way that maintains or improves the consumer’s and society’s well-being. Consumers may expect companies to protect the environment and conduct business ethically. When consumers are given information about a company’s level of social responsibility, they evaluate the company. Lack of awareness of the level of social responsibility is likely to be a major inhibitor of consumer responsiveness to CSR. The socially conscious consumer is a person who takes into account the public consequences of his or her private consumption or who attempts to use his or her purchasing power to bring about social change (Webster 1975). This concept is based on the psychological construct of social involvement. The socially conscious consumer must be aware of social problems, be active in the community, and believe that he or she has the power to make a difference.

Consumers are a major stakeholder group particularly sensitive to CSP. Consumers’ perceptions of a company as socially oriented are associated with a higher level of trust in that company and its products.²⁶ It is important for socially responsible companies to develop consumer trust and engage in CSR programs that are meaningful to their consumers. Trust is a fundamental asset in the company–consumer relationship. Commitment to a brand, defined as trust, is the consumer’s desire to maintain the relationship. Socially oriented companies can use trust to improve their competitive performance (brand loyalty). Brand loyalty includes tolerance to pay a higher price for the product.²⁷ Ethical firms producing a particular product are not producing the same one as unethical firms. Even if the product is identical in all other characteristics, this same product can be differentiated by the extent to which it is produced in accordance with a specified set of ethical standards (Altman 2005). Bhattacharya and Sen (2003) argue how nonproduct aspects of a company, such as CSR, can lead to consumer loyalty and other positive post-purchase outcomes. There are a number of studies in which consumers claim to be ready to pay higher prices for products from socially responsible companies, or to take the social responsibility profile of the producer into consideration when comparing different brands (Creyert and Ross 1997; Ellen et al. 2000; Mohr et al. 2001). From a demand-side perspective, a good reputation increases the value of the brand, which, in turn, increases the company’s goodwill. Companies are faced with the challenge of becoming more attractive to socially conscious consumers. It is important to investigate how corporate decisions are received by consumers. Consumers are engaged in constructing the ethical identity of companies. Managers are encouraged to behave in an ethical manner because information about a company’s ethical behavior is thought to influence consumers’ image of the company.

Corporate identity (‘what the company is’) emerges from the interaction between the company and its consumers. Identity theory at the individual level has been used to explain role-related behavior. Social structures affect the extent to which an individual commits to a particular role. Corporate identity denotes the characteristic way in which an organization goes about its business, or how it behaves and interfaces with the social environment. Corporate behaviors regarding ethical conformity, in light of normative expectations, represent a response mechanism to the social environment. Socialization, through which individuals learn the perspective necessary to perform roles, is expected to prime ethical identification. Corporate identity interplays with social concerns to influence the response to values, norms, and beliefs. CSP can be then regarded as voluntary corporate commitment to exceed the explicit and implicit obligations imposed on a company by society’s prevailing ideas and opinions. On the one hand, we have the interests of the economy arising from a largely unregulated market of companies pursuing their self-interest and, on the other hand, we have the interests of society arising from ethical or moral conformity that may suppress the effects of self-interest. Managers have to legitimate their activities within the field of tension between ethics and economic success. These aspects are strongly reflected in the dimensions of ethical or moral perceptions.

In this appendix, I identify CSP with the private provision of public goods (socially responsible or environmentally friendly activities) or the private redistribution of profits to social causes. Social expenditures represent redistribution through community projects and support, philanthropy, training and educational programs, workers’ rights initiatives, environmental abatement and protection, and alternatives to animal testing (Baron 2008). What motivates a business to make a contribution to these public goods? The model provides a theoretical framework for analyzing corporate incentives to engage voluntarily in socially responsible activities. According to social exchange theory (Blau 1964), individuals are motivated to take voluntary actions when they expect that they could get something in exchange. This is based upon the assumption that people base all their decisions on the calculation of costs and benefits in the pursuit of their own material well-being.²⁸ Some studies assume that private provision of public goods is motivated in part by warm glow (Andreoni 1989, 1990). If the warm-glow motive is strong enough, individuals may continue to make direct donations. CSP is modeled as a response to consumers’ preferences over public goods (Besley and Ghatak 2007). Bagnoli and Watts (2003) study the strategic use of CSP to appeal to ‘green’ consumers with warm-glow preferences for public goods.²⁹ In this appendix, the manager does not engage in strategic CSP in an attempt to maximize only ‘pecuniary’ profits. The manager may act to meet the expectations of socially conscious consumers, and thereby to receive the resulting ‘social satisfaction’ from CSP. The ‘nonpecuniary’ motivation of agents for being involved in activities that benefit others is an equilibrium outcome. The present model is related to Baron (2008) in the organizational forms for private provision of public goods. However, the model developed here provides an alternative explanation for understanding how different social preferences can be created in manager–consumer interactions. The model investigates how the consumer’s taste for CSP is endogenously determined in the social environment. In response, the manager’s taste for CSP is also endogenously determined in the environment.

2.1.2 Overview

Mainstream neoclassical theory assumes that, under conditions of perfect competition, a market economy will allocate scarce resources in an economically efficient way. Economic efficiency is what is called Pareto efficiency, a situation in which no improvement can be made to the material welfare of one person without making someone else worse off. Neoclassical economic theory has paid little attention to economic behavior that is inconsistent with maximizing the material welfare of the individual from a self-interested perspective. Commitment to general social interests is in breach of the conventional neoclassical postulate. In the pursuit of self-interest, perfect market structures and ethical behavior can be inimical to each other. If ethical behavior requires additional inputs into the process of production, the optimal market structures in Pareto-efficient resource allocation terms no longer exist. In order to overcome this problem, it must be understood that the design of current institutions is the result of a reaction to social needs. In this appendix, instead of focusing on narrow self-interested behavior, I examine other configurations that provide the glue that links together individuals, helping realize potential improvements in society. In an alternative context, the firm intends to further public goods beyond its direct interests. The model provides an important foundation for CSP analysis by showing how one might evaluate the CSP of the firm from the ethical or moral perspective.

Contemporary economic theory has treated altruism in different ways.³⁰ For Becker (1976), self-interested behavior and altruism are not inconsistent for utility-maximizing economic agents. What counts in Becker’s analytical framework is that individuals are rational given their objectives and the constraints. For the theory of rational economic behavior, altruistic or ethical behavior can be incorporated into the utility function of the economic agent. The altruist derives pleasure not because the other is assisted but rather because the other’s pleasure is already part of the altruist’s utility.³¹ Thus, the donor would give if this donation increases his or her pleasure more by watching the pleasure of others. Altruistic behavior is caused by the fact that, as a factor in his or her utility function, the person also has the utility of the individuals whom he or she wants to benefit. However, a basic premise of the conventional neoclassical wisdom is that ethical or altruistic behavior cannot persist if it is inconsistent with competitive market forces. The firm has as its primary objective the maximization of its profits. Ethical behavior does not positively affect the firm’s efficiency if the firm is operating at maximum from the start. In the perfectly competitive market, an ethical firm will lose market share to more efficient rivals if it engages in inefficient activities. If managers want to work toward the betterment of society, they should do so as private individuals at their own expense. Yet, social and environmental problems have a direct impact to the company’s welfare. If the company is not concerned with these problems, it may lose social acceptance and support. The company may be coerced into increasing its social output in order to survive. The economic man known to students of Walrasian economics acts on the basis of preferences that are self-regarding (excluding such intrinsic values as altruism, fairness, and vengeance) and are defined over a restricted range of outcomes (excluding honesty as well as concerns about the process rather than simply the outcome of exchange per se) (Bowles and Gintis 2000). The model of the economic man does not consider the process through which the premises from which the man makes decisions are formed. The criteria for neoclassical rationality do not solve the question of what is good for us, or which action we should consider. Departures from the standard economic assumptions are potentially important in the social domain. A focus on processes is a reminder that the very idea of CSR suggests a need to consider the values, motives, and choices of those who are involved in formulating corporate policy. If government does not react to a change in social preferences, the company might be able to act decisively and quickly. Consequently, the company must be motivated to improve its social performance. The CSR programs can provide additional social benefits when such programs are central to the mission. Looking at social aspects is not a new phenomenon for companies. What is new, though, is the intensity and breadth of the efforts made by companies as well as the increasingly strong societal demand for behaving more ethically and responsibly. Companies must learn how far they need to extend their responsibilities, what issues to take up, how to give meaning to those issues, and how to successfully combine economic, social, and environmental strategies.

Nowadays, society at large increasingly expects companies to behave as good corporate citizens.³² Good citizenship metaphor emphasizes voluntary self-restraint and altruism concerning the realization of broad duties. Social responsibility derives from the moral legitimacy the corporation achieves in society. An exclusive focus on profit maximization is too limited as it neglects the company’s responsibilities toward the socially conscious consumers. As consumer awareness of the need for CSP increases, managers increasingly recognize the responsibilities for implementing ethical programs to enhance social welfare. Through evaluation of the manager–consumer relationship, the manager chooses whether to integrate CSP within corporate strategy. CSP commitment increases the company’s trust, making its operations more legitimate to consumers. Whether a company is being ethical or not is dependent on the manager awareness of creating trust in the manager–consumer relationship. Justifying the role of CSP as an extended scheme of corporate governance is an essential part of the requirements of a just society understood as a joint venture for mutual benefit.

In this appendix, CSP is discussed as an outgrowth of ‘collaboration’ between managers and consumers in the social environment. The term ‘collaboration,’ or ‘dialogue,’ describes the involvement of consumers in the manager’s decision-making processes that concern social and environmental issues. The realization of the company’s social responsibilities is shaped by the way the manager interacts with consumers. The model focuses on the views of both managers and consumers about CSP. The manager has a key role in the supply of CSP, using the company’s resources to achieve its social goals. Active CSP is conceptualized as follows. The manager can anticipate social preferences of consumers and transform them into social benefits. Social benefits then reflect the ‘price’ a socially conscious consumer puts on the supply of CSP. If social benefits can be created in manager–consumer interactions, the consumers’ ethical attitudes facilitate the corporate social responsiveness. The collaborative relations can reinforce positive responses with each other. The socially conscious consumers’ attitudes to encourage CSP are influenced by the manager’s view to practice CSP, and vice versa. Thus, both mangers and consumers are influenced by comprehensive perception toward CSP. This establishes the basis for understanding the central role of both ethical self-regulation and consumers’ activism. Ethical self-regulation establishes a new context for manager–consumer interactions, and consumers’ activism becomes effective in this context. This framework thus focuses on how CSP is understood and perceived by both managers and consumers. The model considers two key aspects of social preferences: the consumer’s taste for CSP and the manager’s taste for CSP.

A public good is voluntary contribution to society based on other-regarding attitudes. In the model, public goods can be provided by the firm through its CSP. Consumers may have social preferences for that public good. Ethical managers engage in impartial moral reflection on practicing self-restraint and altruism. CSP is also about managing change at the firm level in a socially conscious manner. Then consumers become aware of a firm’s level of social responsibility. The firm supplies both private and public goods by incorporating social concerns of consumers. The motivation underlying CSP is modeled in terms of the resulting social satisfaction in manager–consumer interactions. Social satisfaction that the manager receives affects whether CSP is provided at the firm level. CSP will change our conception of what a for-profit organization is and how its value is measured. Consumers experience the results of corporate behavior, so they influence the ethical standards by which corporate behavior is judged and evaluate how well managers perform according to those standards. If the company succeeds in managing change in a socially responsible manner, this will have a positive impact for the entire environment.

This appendix offers an analytical framework to evaluate how the manager involves consumers in the decision-making processes. The collaboration is ultimately about exchanging opinions between them; in other words, it is about dialogue. A participatory dialogue allows consumers to translate their own judgments and voices into practice. In the model presented below, consumer engagement is important for the supply of CSP. The dialogue that needs to take place between the parties involved is guided and shaped by a confrontation of views and interests.³³ The dialogue has its added value in coming to some kind of social consciousness. When consumers perceive that they can be increasingly effective in the supply of CSP, they will show more concern for CSP. If consumers are not free to participate in the collaboration, the level of perceived consumer effectiveness is sufficiently low. If consumers are insufficiently included in the dialogue, anticipated social benefits from the collaboration will be limited.

Furthermore, managerial perceptions of CSP are also important. CSP is part of the company’s vision. A company’s stance on CSP may strongly influence how ethically the company is perceived. The manager’s ‘mental model’ or ‘mind-set’ has an important impact on the implementation of consumer dialogue. Mental models function as selective mechanisms or ‘filters’ for dealing with experience. The manager thus has a filter that limits the resulting social satisfaction from the consumer dialogue. The filter makes it difficult for the manager to live up to the participatory ideals of the consumer dialogue. The filter depends on the manager’s social and environmental awareness. The manager’s decisions are thus based on the interplay between individual cognition and social institutions. Managerial perceptions are adapted to certain circumstances. Individuals follow an ‘ecological rationality,’ and rationality thus becomes context-based (Smith 2003). Ecological rationality uses reason to discover the possible intelligence embodied in the rules, norms, and institutions of our heritage that are created from human interactions but not by deliberate human design. People follow rules without being able to articulate them, but they can be discovered. Furthermore, norms are often considered to be the historical legacy of a traditional culture supported and adhered to in a population. The term ethics refers to a set of moral norms that guide people’s behavior. Norms define what actions are considered acceptable or unacceptable according to shared understandings. In this way, this article focuses on moral ‘like-mindedness’ among interacting individuals for CSP to succeed. Morality can be created and utilized in social interactions.

2.1.3 The Model

The firm in the model economy normally produces an identical (numeraire) private good which can be consumed. There are potential shareholders in the firm. The shareholders of the firm are assumed to be in contact with a manager. The manager operates the firm after accepting the contract. Let π be the (pecuniary) operating profit when the project is implemented. After the operating profit is specified, compensation is paid to the manager and the remaining amount of the profit is distributed to shareholders. Let pπ, where 0 < p < 1, denote the manager’s compensation when the profit is specified. Similarly, (1 − p)π is distributed to shareholders when the profit is specified.

Moral conduct refers to a pattern of behavior that goes beyond normal business management. The manager chooses units of social expenditures, g, which affect social welfare. By choosing g, the manager contributes to the private provision of public goods as corporate social performance (CSP). The provision is then bounded. It is assumed that the manager picks up his or her own contribution g in some interval [0, G], where G > 1 is given.

In this model, CSP is discussed as an outgrowth of ‘collaboration’ between the manager of the firm and its consumers in the social environment. The model concentrates on two different dimensions of social preferences: the consumer’s taste for CSP and the manager’s taste for CSP. As a consequence, social preference formation interacts with economic decisions taken by others. The model analyzes the decision problem of the manager who receives the compensation alone and the manager who relies on social values.

Consumers choose whether they consume the private good or become concerned about the CSP of the firm. Consumers have an initial endowment they can allocate between purchasing the private good and personal giving to social causes. They may care about moral management or CSP. The model considers the relative weight between the consumer’s concern for the private good and that for the amount of CSP supplied. Let the consumer’s taste for CSP be described by a fraction α ( ≥ 0). The fraction α represents the value that a consumer receives from the amount of CSP supplied by the firm, g. It can be also considered as the ‘price’ the consumer puts on the perceived benefits from the CSP of the firm. On the other hand, the consumption for the private good becomes 1 − α. Then the consumer benefits from both the private good and the amount of CSP supplied, g. More specifically, the consumer’s utility function is given by:

$$ U= \ln \left(1-\alpha \right)+\alpha g. $$

(2.18)

The greater α is, the more utility a consumer gets from the CSP of the firm.

The manager allocates his or her time between producing the private good and providing CSP (the private provision of public goods). The manager provides CSP by allocating a portion of β ( ≥ 0) of his or her time to social causes. The fraction β is considered as the manager’s taste for CSP. If the fraction β is positive, the manager chooses the amount of CSP supplied, g. On the other hand, the manager earns the payoff ln(1 − β) by allocating 1 − β of his or her time in producing the private good. Thus, the operating profit is specified as:

$$ \pi = \ln \left(1-\beta \right)-\beta g. $$

(2.19)

In (2.19), g reflects the firm’s social expenditures, the cost of CSP. It may be also considered as the private redistribution of profits to social causes. The manager may view CSP as an expense rather than an investment.

Managerial perceptions of CSP are important. The manager may receive a non-economic or psychological benefit reflecting ‘social satisfaction’ associated with the CSP of the firm. In the model, the manager perceives social satisfaction, h, from one unit of CSP provided by the firm. Thus, if the manager’s taste for CSP is β, the manager gets βhg from providing the amount of CSP, g. Social satisfaction h reflects the manager’s social and environmental awareness. From moral management or CSP, the firm has an ethical responsibility to society. The socially responsible firm seeks to meet the expectations of consumers and to get a positive image from consumers in the economy. Thus, social satisfaction h depends on the consumer’s taste for CSP. More specifically, let h = h(α) be a strictly increasing, strictly concave function of α. This implies that social satisfaction h increases when the value the consumer puts on the CSP of the firm increases. It is assumed that h(0) = 0. That is, if consumers are excluded from the collaboration, the value h is zero. When the profit is specified, the manager receives his or her compensation pπ.

The manager’s utility function is then given by:

$$ V=p\pi +\beta h\left(\alpha \right)g. $$

(2.20)

The manager struggles in deciding how to reconcile the benefit from CSP with the cost of CSP. The benefit from socially responsible programs is not guaranteed. To the extent that increased CSP results in improved benefit, the manager will be encouraged to become more socially responsible.

The consumer chooses α to maximize (2.18) subject to α ≥ 0. The first-order condition for the consumer’s choice of α is:

$$ \frac{1}{1-\alpha }+g\le 0. $$

(2.21)

For α > 0, (2.21) holds with strict equality. Then,

$$ \alpha =\Big\{\begin{array}{cc}0& \mathrm{if}g\le 1\\ {}\frac{g-1}{g}& \mathrm{if}g>1.\end{array} $$

(2.22)

The optimal α depends on the amount of CSP supplied by the firm, g. From (2.22), we can say that the consumer is less concerned about the CSP of the firm if the social expenditures are not sufficient (g ≤ 1). The consumer’s benefits from CSP can increase with the amount of the public goods provided. Thus, the fraction α measures consumer’s awareness for CSP. Lack of awareness is likely to be a major inhibitor of consumer responsiveness to CSP.

For a positive α, we have

$$ \frac{d\left(g-1\right)/g}{dg}=\frac{1}{g^2}>0. $$

Thus, α is a strictly increasing function of g if g > 1. The amount of CSP supplied by the firm has a significant impact on consumer responses.

The manager’s choice problem is to choose g and β to maximize his or her utility in (2.20). Using (2.19), (2.20) can be rewritten as:

$$ V=p\left\{ \ln \left(1-\beta \right)-\beta g\right\}+\beta h\left(\alpha \right)g. $$

(2.20ʹ)

In (2.20ʹ), for a positive β, the first-order condition for the manager’s choice of g over the interval [0, G] is:

$$ g=\Big\{\begin{array}{cc}0& \mathrm{if}h\left(\alpha \right)\le p\\ {}G& \mathrm{if}h\left(\alpha \right)>p.\end{array} $$

Thus, we can say that the manager is encouraged to provide CSP if his or her social satisfaction is larger than the ‘marginal’ cost of CSP, or h(α) > p. In the model, h increases in α. Then, the manager’s choice of g is associated with the consumer’s taste for CSP. The manager assesses the social demand for CSP and then determines the optimal level of CSP to provide.

The manager chooses β to maximize (2.20ʹ) subject to β ≥ 0. The first-order condition for the manager’s choice of β is

$$ -\frac{p}{1-\beta }+\left(h\left(\alpha \right)-p\right)g\le 0. $$

For β > 0, the above inequality holds with strict equality. Then, we have:

$$ \beta =\Big\{\begin{array}{cc}0& \mathrm{if}\left(h\left(\alpha \right)-p\right)g\le p\\ {}\frac{\left(h\left(\alpha \right)-p\right)g-p}{\left(h\left(\alpha \right)-p\right)g}& \mathrm{if}\left(h\left(\alpha \right)-p\right)g>p.\end{array} $$

The optimal β depends on the manager’s social satisfaction, h(α), the marginal cost of CSP, p, and the amount of CSP supplied, g.

2.1.4 Discussion

Vogel (2005), examining the links between ethics and profits, argues that:

The emergence of ‘companies with a conscience’ represents a particular vivid expression of the contemporary reconciliation of social values and the business system. These are companies whose vision of social responsibility was integral to their business strategies from the outset. They were formed by individuals with strong personal social commitments who regarded their businesses both as vehicles to make money and as a means to improve society. (Vogel 2005, p. 28)

Vogel (2005, p. 2) defines CSR, or business virtue, as “practices that improve the workplace and benefit society in ways that go above and beyond what companies are legally required to do.” CSR belongs to strategy, and, in this respect, it is a matter of corporate policy. Companies choose to behave in a socially responsible way, in the same way they choose to spend more on marketing or production.

Standard economic analysis, built on the behavioral assumptions of neoclassical theory, pays little attention to ethics and morality beyond direct profit maximization. The self-interested economic agent has no ethical doubts. For instance, in the principal-agent literature (e.g., Grossman and Hart 1983), the principal is the person whose material welfare should be maximized, and the agent should execute the orders of the principal by receiving an adequate compensation. Efficiency, in the Pareto sense, is the main goal of the theory. The agency conception is based on the pecuniary conflict of interests of the principal (shareholder) and the agent (manager).

A model of business as an independent system has been represented by neoclassical economic theory. In a world of independent systems, the firm exists to maximize its profits. In the behavioral theory of the firm, on the other hand, organizational slack is a pre-requisite for being able to afford corporate social strategy. Such strategies rely on individual managers making decisions based on their own values. Nowadays, society at large increasingly expects companies to behave as corporate citizens. Profits are not an end per se; they must be compatible with other social needs. To respond positively to moral responsibility, managers must find an alternative way of doing business. In reality, we recognize that the interactions among various factors of society are complex. From a behavioral theory perspective, this type of complexity is understandable as multiple and conflicting managerial goals. The individual is not separate from society and thus not separate from the corporation. Therefore, business is represented to be part of society, affecting it and being affected by other aspects of it. A relational view thus emerges with corporations embedded in society. CSR is about the basic idea that businesses have to meet society’s expectations in their practices. CSR involvement is fueled by various social demands and enhances the company’s access to various resources.

Consumer attention focuses on a company’s decisions and actions. In particular, consumers may expect companies to behave ethically and to be actively involved in helping society. Consumers then need to be aware of a company’s level of social responsibility. Thus, companies are under pressure to behave in socially responsible ways. Consumer expectations regarding the CSP of the firm are a motivating target and must be considered carefully. Managers must grasp the nature of the values, attitudes, and behaviors of their consumers and respond accordingly. The attitude would serve as a reference point in consumers’ evaluations of a firm’s involvement in CSP. CSP communicates that the company can be trusted to act as a partner that will respond to social needs. CSP is thus internalized in the consumer dialogue process. Managers have to develop communications that provide information about their social performance. Social interactions are driven toward dialogically motivating and sustaining mutual understanding.

People construct the content in a subjective, meaning-creating process. Sense is a continuous process oriented toward placing current experiences in a frame of reference. Social environments are characterized by a shared understanding among individuals about the nature of a particular issue and the available behavioral alternatives. The mental process in the case of CSP is directed at the creation of a common, context-bound view. Corporations are increasingly engaging as a form of social involvement in response to increased expectations for companies to contribute to CSP efforts. Companies bring their resources to rebuild trust in the social environment. By pursuing social and environmental objectives, socially responsible companies increase their trustworthiness, which in turn supports the process of value creation.

The manager needs to carefully examine its mission, vision, and value. Individuals voluntarily work together to create relationships in the pursuit of value creation. Value is a social phenomenon. It is created with the help of others who value what they create. Working with others and for others can be a strong motivation to create new sources of value. Value is an important factor in a process of arriving at a decision about a course of action. Individuals commit to commonly shared values. They may incorporate their moral values into their economic decisions. Organizational ethics is a company’s adoption of desired ethical standards. Ethical standards are formulated from a point of view that goes beyond the interests of a particular individual or group. Ethics sets out what kind of values, norms, and beliefs the good person should cherish. But ethics is only one of a number of dimensions of the decision processes. A dilemma arises in a situation where two (or more) conflicting standards appeal to the consciousness of an individual. An organization encouraging the view that profits are the only consideration will tend to be populated by individuals thinking and functioning on a lower ethical level.

Corporate activities are influenced by the level of moral conduct. The ethical component is silent when there is no moral issue associated with the decision. It comes into play when a moral issue is present. An advantage of behavioral modeling is that individuals can adapt appropriate behaviors to various situations. Wagner-Tsukamoto (2007) reconstructs an economic interpretation of moral agency by providing three levels of moral conduct. (1) Unintentional, passive moral agency. According to classical and neoclassical economic thought, the market economy itself comes with certain ethical ideals. The ‘invisible hand’ of the market best serves consumers in getting a desirable product. Self-interested engagement in the market process itself thus reflects an ethical ideal. (2) Passive, intentional moral agency. Legal laws can be viewed as the enactment of moral minimum standards for all agents.³⁴ This reflects the enforced nature of moral agency. Findings regarding the enforcement of moral agency through legal laws are in line with findings that an agent’s own enforcement of moral conduct is ineffective unless properly sanctioned in economic terms. (3) Active, intentional moral agency. Companies not only sell products on the basis of profitability but also on certain ethical grounds. And such ethical grounds are pursued voluntarily by these companies. Companies engaged in moral agency exceed the minimum standards laid down by public ordering.

Personal belief systems and standards are often related to the social background of an individual. Institutional economics intervenes with institutions understood as incentive structures which order social interactions and resolve problems in organizational behavior (‘governance structure’ as Williamson 1985). However, individuals always choose from a certain ‘bounded’ situation that provides them with the premises upon which they will make their decisions. The present framework focuses on ‘institutions’ for voluntarily creating morality in social structures. The structures relate the company to the social values. Managers stand in a particular relationship to the norms they must enforce. Ethical arguments can be derived from prevailing social norms. Social norms encompass an individual’s perception of whether or not to engage in the behavior as seen in others. If the individual perceives that others would encourage him or her to behave ethically, it is more likely that the individual would intend to behave ethically. The process of social identification and cohesion of social attitudes promote moralization through an internalized focus on value and conformity. Simon (1993) argues that human beings do not behave optimally for their fitness. Decision-making that deviates from neoclassical rationality does not imply irrationality or even errors. Simon coins the term ‘bounded rationality’ to refer to rational choices given the physiological cognitive limitations on the individual decision-makers in terms of acquiring and processing of information. Because of bounded rationality, individuals have a tendency to act on advice and respect norms. Then, ‘docility’ contributes to the fitness of human beings in evolutionary competition. Here, docility means the tendency to depend on suggestions, recommendations, persuasion, and information obtained through social channels as a major basis for choice. Values, norms, and beliefs play a part in determining whether any individual does the right thing. Companies operating at the high level of moral conduct (active, intentional moral agency) take legitimate social interests into account. Morality is thus conceptualized by focusing on the behavioral manifestations of values, norms, and beliefs in social interactions.

Business and society are interwoven rather than distinct entities. A company’s involvement in CSR programs involves answering the requirements of consumers with particular focus on societal issues. This program not only enhances a company’s ethical culture but also its attention to CSR. A company develops its own meaning of CSR as a result of the interaction between the current confrontation with situations linked to CSR and company-specific capabilities. The decision problem in ethics is not only to choose among alternatives but also to choose the alternatives. A proper balance is then created between economic and social aims. Value-driven companies develop their own interpretation of CSR on the basis of firmly embedded values, norms, and beliefs of society. Corporations are then social actors. If their economic actions are guided by social relations, their socially oriented actions are more likely to be shaped by their social relations. The corporate perspective on CSR and the resulting behavioral change are, therefore, products of social interactions.

Generally speaking, the idea of institutions is understood as systems of established and prevalent social rules that structure social interactions (North 1990). Institutional variations, as codified in social structures, create differing environments for the resolution of public issues in which the role of corporations in society is under debate. Institutional variations can result from different perceptions or attitudes toward CSP. The acceptance of consumers into the corporate decision-making processes affects the way specific issues are evaluated, addressed, and resolved within the institutional context. Institutions can be conceptualized as internalized cognitive structures that reflect an individual’s values, norms, and beliefs regarding social conduct. Such internalized structures are meant to dispose the individual toward socially desirable behavior. Cognitive differences among individuals thus result in institutional differences regarding the role of corporations in society. Morality depends on like-mindedness of interacting individuals in order for CSP to prosper. Commitment encourages discretionary behaviors that result in positive goal outcomes. A commitment to good social conduct can be internalized by constructing and promoting moral like-mindedness among individuals, even though there is no apparent economic profit. Behavioral modeling can provide positive sets of ethical behaviors when the situation calls for an ethical response.

2.1.5 Concluding Remarks

The framework presented here identifies CSP with the private provision of public goods or the private redistribution of profits to social causes. In order to consider what motivates a business to make a contribution to public goods, the model focuses on manager–consumer interactions. The model analyzes two dimensions of social preferences: the consumer’s taste for CSP and the manager’s taste for CSP. As a consequence, social preference formation interacts with economic decisions taken by others.

The motivation underlying CSP is modeled in terms of social satisfaction that the manager perceives and receives in the interactions. Consumers experience the results of corporate social behavior, so they influence the ethical standards by which corporate behavior is judged and evaluate how well the manager performs according to those standards. CSP communicates that the company can be trusted to act as a partner that will respond to social needs. The collaborative relations can reinforce positive responses with each other. Consumers’ attitudes to encourage CSP are influenced by the manager’s view to practice CSP, and vice versa.

CSP is considered as levels of economic, ethical, and discretionary activities of a business entity as adapted to the values, norms, and beliefs of society. Managers operating at the low level of moral conduct do not take socially conscious consumers’ interests into account, and their decisions are perceived to be immoral by consumers. No public good is then provided by firms. A behavioral view identifies an alternative source of inertia as the low level of moral conduct insensitive to the values, norms, and beliefs of society. On the contrary, managers operating at the high level of moral conduct focus on meeting the expectations of socially conscious consumers according to ethical requirements. Then the provision of public goods can be improved substantially. The corporate perspective on CSP and the resulting behavioral change are therefore products of social interactions. Managers who rely on social values facilitate the ethical development to be motivated beyond their self-interest. The primary force in determining the level of corporate moral conduct is the resulting social satisfaction in manager–consumer interactions. The social satisfaction that the manager receives is influenced by the consumers’ expectations regarding CSP. More of public goods can be provided when consumers are more likely to value CSP, and managers are attuned to the consumers’ expectations about CSP.