Environmental and Resource Economics

, Volume 70, Issue 4, pp 835–860 | Cite as

Moving Toward Greener Societies: Moral Motivation and Green Behaviour

  • Lorenzo Cerda PlanasEmail author


I provide an alternative explanation for why societies exhibit varied environmental behaviours. I use a Kantian moral approach at a microeconomic level. I show that two identical societies (in terms of income level and political system) might follow different paths with respect to their “green” behaviour. Additionally, I identify tipping points that could nudge a society from a polluting behaviour to a green one. I find that the perception of environment within the society can be an important factor in this shift.


Green behaviour Kantian morale Moral motivation Tipping points 

JEL Classification

C62 D64 D72 H41 Q50 

1 Introduction

This paper provides an alternative explanation for why countries behave differently with respect to the environment and contributions to global pollution, although they might be quite similar from an economic development point of view. I use a simple micro-founded model in which individuals gain utility from consumption as well as from a moral standpoint. The moral utility comes from the idea that individuals derive satisfaction from doing ‘the right thing’ (at least to some degree)—or according to Immanuel Kant, from behaving according to the imperative principle. Being or acting green could fall into the category of such imperative principle. Using these concepts in addition to a simple political framework, I show that two equivalent societies (i.e., societies with the same income, political system, etc.) can reach two different environmental behaviour equilibria. I also locate how a nudging process could change a society from one equilibrium to another. While I do not claim that this explanation is the only reason for why countries behave differently, this model provides a very straightforward rationale for why this could happen.

Different theories have been developed to explain, to some extent, the dissimilarity of green behaviour among similar countries. One set of theories revolves around a country’s level of development. For example, the Environmental Kuznets Curve (EKC) relates a country’s environmental behaviour to its income level, revealing an inverted U-shaped relationship between the two factors.1 The drawback of such an approach is that while it explains dissimilarities concerning local pollution, it cannot explain the dissimilar behaviour when it comes to global pollutants (for example, \(\hbox {CO}_2\) emissions per capita) among countries with similar income levels. This phenomenon becomes apparent when we compare the United States, Canada, Australia, and their European counterparts Denmark, Finland, Germany, the Netherlands, Sweden and the United Kingdom.2

Another important strand of literature explains green behaviour based on moral motivation. This literature starts from the observation that green actions cannot be explained with a purely homo-oeconomicus theory.3 If individuals only abided by homo-oeconomicus principles, there would be no incentive for them to contribute to maintaining a good environment, since the gain for this action would be negligible. One main concept in the literature is the idea of ‘warm-glow giving’ (Andreoni 1990), in which the sole act of giving provides utility to the agent. This notion has been further developed by Nyborg and Rege (2003) and Nyborg et al. (2006). The former study provides a comprehensive summary of different types of moral motivations (altruism models, social norm models, fairness models, models of commitment, and the cognitive evaluation theory), and focuses on how these different types of moral motivations can crowd out private contributions. Nyborg et al. (2006) present a model in which societies become polarized due to the effect of peer pressure. They find that a society can be completely green (when everyone protects the environment) or completely grey (when everyone chooses to pollute). Although in reality polarization is arguably less extreme, it is clear that this social pressure exists.

I propose the following approach: People are diverse and behave, to a certain extent, according to a Kantian imperative. Individuals try to do ‘the right thing’ even if it goes against their private (or ‘selfish’) homo-oeconomicus tendencies. This Kantian idea has been applied to economics in Brekke et al. (2003), Brekke and Nyborg (2008), Nyborg (2011), Wirl (2011), Roemer (2010), Laffont (1975), and Ulph and Daube (2014). The first three papers use the Kantian approach to define an ideal. In measuring one’s behaviour against this ideal, one gains utility from one’s self-image, which grows as one’s behaviour approaches the ideal. Similarly, Wirl (2011) talks about green players that find the social optimum following a Kantian approach. Roemer (2010) uses a game-theoretic approach that defines a Kantian equilibrium as one that no one wants to deviate from in the same proportion. In Laffont’s paper (1975), agents take into account what their behaviour would or should be with the premise that everyone will do as they do. He then derives a market equilibrium and shows that if agents behave in a Kantian way, they internalize their externalities and achieve an optimal outcome. In a very similar manner, Ulph and Daube (2014) use a Kantian morale where agents assess a hypothetical moral value of adopting a Kantian behaviour, calculating the consequences of their action by asking what would happen if everyone else acted the same way they did.

In my model, I refer to Kantian people in a way similar to Laffont (1975) and Ulph and Daube (2014). Here Kantian means that the agent considers what he or she should do (the right thing), which might differ from what the agent would do if he or she were to act in a purely homo-oeconomicus way. In this case, the agent has to know or find out what the right thing is. This is not exactly what Kant intended in his work (Kant 1785); he described duties (deontology), not responsibilities. He wrote about a maxim (universal law) instead of actions. However, it may be argued that one meets or fails to meet one’s responsibilities as a consequence of one’s actions. Now, if agents thought that their contribution to pollution (or its prevention) was negligible, they would not believe that they were (directly) responsible for climate change. However, it is widely accepted that we are all responsible. In that sense, being Kantian (within the context of this paper) means being a morally responsible person. It means making an assumption about the consequences of our actions (hence the word ‘responsibility’) using a Kantian way of thinking—“assuming everyone behaves as I do”. We have all heard others make statements like: “I don’t pollute (or I do recycle, etc.), because if everyone did so, the effects would be terrible, and I don’t want to contribute to that”. This tells us that the person has assessed the situation, made a decision that the action is morally bad, and is taking the action of refraining from participating in it if at all possible. This practice of responsibility in moral systems may be explained by a variety of primary ethical principles, including the categorical imperative (Kant), utilitarianism, contractualism, cooperation, and compassion, as explained in Baumgärtner et al. (2014). In this paper, I use the Kantian categorical imperative as a primary ethical principle. I can cite Bruvoll et al. (2000) in support of this idea. They find that 88% of surveyed people in Norway, when asked about their motives for sorting waste, agreed or partially agreed with the following statement: “I recycle partly because I think I should do what I want others to do”.

In general, the choice to be green is a costly one; if it were not, of course, it would not make sense not to do so. Therefore, the agent is faced with a trade-off between doing the right thing and being negatively affected by the direct cost of this action. This cost is the one incurred when we compare the decision to make a green action with the one that would be made in a purely homo-oeconomicus way. To be able to effectively assess the situation, the agent must know (or estimate) his moral ‘obligation’ and its cost. He then weighs the options and decides how to behave. As we can see, this is not a general rule (and therefore not strictly Kantian), since a specific person could be swayed toward taking part in green or grey behaviour depending on the cost and his perceived moral implication of his choice. For example, the implications of contaminating the environment when it is already quite polluted are quite different from those in a situation in which the environment is clean. It is clear, though, that people weigh the cost of their green action and their moral responsibility in diverse ways. To account for this, I will assume that society is composed of a continuum of people, ranging from the purely homo-oeconomicus individual to the purely Kantian one. Although the idea of moral obligation has been also addressed in Ulph and Daube (2014), I depart from their work, using damages coming from stock of pollution rather than emissions. In addition, they employ the concept of altruism, while I focus on the Kantian motive.

As we can see, this is a dynamic process. Individuals’ awareness and moral motivations depend on their perception of environmental quality (or pollution). Changes to the environmental state and/or corrections to one’s perceptions can cause the degree of awareness to vary. At the same time, public decisions are made by governments, which generally try to apply the best policies. The problem arises from the fact that global pollution (or transboundary pollution, in some cases) involves usually a prisoner’s dilemma game in the international arena, and in this case it is ‘optimal’ for the country to pollute. Avoiding to bear the cost of green policies is the dominant strategy of the game. In this sense, governments are not likely to implement green policies (concerning global pollutants). However, a government may be willing to incur the cost of implementing a green policy if it has the support of its constituency, as Sweden did in the late 1990s.4 In this case, government is answering to a public demand that will not necessarily lead to the maximization of social welfare. People might be demanding green policies because they feel it is the right thing to do, even if they have to bear an extra cost (for example, the cost of abatement). In this regard I also depart from Ulph and Daube (2014), since I am interested in the results of the dynamics rather than maximizing social welfare.5 Consequently, I introduce a simple political mechanism into the model. The government in power will put a green (or greener) policy in place if the people demand it. Therefore I assume that if individuals behave in a green way, their government is also likely to have a green policy in place. This assumption is in line with what was observed in Germany by Comin and Rode (2013), who found that when people behaved in a greener way, green parties received more votes at elections. It also follows the results of Schumacher (2014), who finds that ‘sufficient concern’ of people triggers green voting, even in the case of global environmental problems. Finally, it is in line with Leiserowitz (2006), who finds that “egalitarian-value based” individuals (similar to those described as embodying the Kantian morale) tend to support green policies.6

On the perception side, a poorer environmental state, which is expressed by Mother Nature through more frequent and severe climatic events, triggers concern. This fact has been comprehensively studied by different surveys (as in Gallup and GlobeScan, among others: Gallup poll,7 Globescan radar8 and Extreme Weather and Climate Change in the American Mind April 20139) and verified using econometric techniques by Krosnick et al. (2006) and Zahran et al. (2006). Lee and Markowitz (2013) performed an overall analysis showing that individuals’ environmental awareness and concern rises after major climatic events, although typically only in the short term.

Combining the previous ideas, we find that pollution can trigger awareness, which can in turn induce action (green behaviour). In this sense, I analyse dynamics that happens in the short term realm. The speed of changes is linked to the socio political process. I do not focus on long-run evolution resulting from the transmission of traits or values, as in Bisin and Verdier (1998), Bisin and Verdier (2001), Bowles (1998), Schumacher (2009), and Buenstorf and Cordes (2008). These works treat the evolution of preferences and/or traits in different ways and, as previously noted, they provide insight into the evolution of societies on a larger time scale. They talk about the transmission of traits from generation to generation. My work differs on this point: I assume a fixed distribution of a trait. The evolution is in behaviour and not in the trait per se.

By combining the elements introduced above—namely, different types of people with regard to their environmental behaviour (their Kantian ‘structure’), the degree of environmentally induced awareness, and a simple political system—I find that two equivalent societies can reach two different environmental behaviour equilibria, even if they share the same structural characteristics (with respect to income, political system, etc.). My main contribution to the literature comes from the mixture of these ingredients: moral motivation from environmental awareness combined with a government that follows the desire of the majority. This combination can yield two distinct equilibria: green and grey. Being a simple model, this shows key drivers of the results. Awareness triggers green behaviour, but maybe not necessarily enough to produce a social change. However, if some political threshold is crossed, the government might apply green policies that in turn reinforce a new (green) equilibrium. This leap could be the result of a shock in perception of pollution or information. The key point is that the combination of social behaviour and policy yields multiple equilibria, and thus similar countries can exhibit dissimilar behaviour.

The logical question that arises is: How can a society be swayed from grey to green? To answer this, I analyse the influence of two factors: individuals’ perception of pollution and the existing political system. A shock to the public’s perception of pollution (such that individuals become more aware of, or concerned with, environmental issues) is an effective mechanism to induce tipping. On the other side, a political framework in which party coalitions are more likely to exist may ease the shift from a grey to green society and vice versa. This could be mainly due to the fact that a more ‘continuous’ political spectrum allows a society to shift toward a relatively greener government and, from there, to greener and greener governments in a cascading process. Applied at a government level, this cascading idea is similar to the one developed by Kuran (1991).10 Continuing along this political angle, the literature has also addressed the determinants of green behaviour by comparing political systems.11 Unfortunately, these results present neither a clear nor consistent view of how political systems might influence a society’s green behaviour. The framework developed in this paper, in particular the idea that tipping can be induced by a more continuous political framework can explain, at least partially, the mixed results found in this literature.

To finalise the model and acknowledge the existence of another psychological ingredient, I add a peer effect, or social approval to the model. It can act as secondary motivator, as in the case of Nyborg et al. (2006). Incorporating this concept into the model reveals that, in fact, an ‘ideological’ peer effect12 makes the transition from a grey to green society a more difficult task to achieve. This comes from the notion that if the society is primarily grey, the agent will have to bear his economic cost plus the new peer pressure cost in order to behave in a green way, thus making the shift harder to accomplish.

The paper is structured as follows: Sect. 2 presents the model and its main features. Section 3 shows possible tipping points and demonstrates how the system can be nudged. Section 4 introduces the concept of social approval as a psychological driver of behaviour. Section 5 concludes and presents a brief discussion of the model.

2 The Model

2.1 Consumption and Pollution

In this simple framework, each person i can either buy green products \((x_{it})\) or grey products \((y_{it})\) at time t. From a consumption point of view, the products are perfect substitutes. However, there are two differences. The first is that the green one does not pollute, whereas the grey one does. The second is that the green product is more expensive than its grey counterpart.13 I assign a normalized price of 1 to the grey good and a price of \((1+\rho )\) to the green one. Therefore, the value \(\rho \) represents the extra amount (with respect to the whole original price) to be paid for a green product. Since the grey product pollutes, I also denote with \(\gamma \) the impact on the environment of the consumption (or more accurately, production) of this type of product. For simplicity, the agent will only choose one or the other, not a mix.14 The agent’s income is also normalized to 1.

Hence the agent can be a ‘grey’ consumer, \((x_{it},y_{it}) = (0,1)\), or a ‘green’ consumer, \((x_{it},y_{it}) = (\frac{1}{1+\rho },0)\).15 On the pollution side, the equation is the standard one:
$$\begin{aligned} p_t = (1-\delta ) p_{t-1} + \gamma \cdot y^s_t \end{aligned}$$
where \(p_t\) is the pollution level at time t, \(\delta \) is the natural decay of pollution level (due to natural absorption), and \(y^s_t\) is the society’s mean grey consumption at time t, which is just the average of grey consumption by all agents.16

2.2 The People

I assume that people care about their utility, which includes consumption and possible damages coming from pollution. As mentioned in the Introduction, some agents will evaluate this utility in a homo-oeconomicus way, and others in a Kantian way, but in general an agent’s utility will be a linear composition of these two. This translates into agents with different attitudes. Therefore we can write each agent’s utility as:
$$\begin{aligned} U_{it} = (1 - \alpha _i) \, u_{it}^h + \alpha _i \, u_{it}^k \end{aligned}$$
where \(u_{it}^h\) and \(u_{it}^k\) are the utilities for agent i at time t from a homo-oeconomicus and Kantian point of view, respectively. Hence the parameter \(\alpha \) (\(0 \le \alpha \le 1\)) represents how homo-oeconomicus \((\alpha \rightarrow 0)\) or Kantian \((\alpha \rightarrow 1)\) the person is, known as his “attitude”. An attitude can be defined as an inherent trait formed from a combination of cultural background and education. It relates to and influences an agent’s moral responsibility. For the purposes of this study, \(\alpha _i = 0\) means that the agent i assesses his utility in the ‘standard’ homo-oeconomicus manner, whereas \(\alpha _i = 1\) means that an agent i is the most morally responsible (Kantian). Moreover, the society is composed of a continuum of people, finite in number, with a given distribution \(f_\alpha \).17
Both \(u_{it}^h\) and \(u_{it}^k\) are constructed in the same way. They have a consumption part \(u(y_{it})\) and a damage part \(d(p_t^j)\):
$$\begin{aligned} u_{it}^j= & {} u(y_{it}) - d(p_t^j) \quad j \in \{ h,k \} \end{aligned}$$
$$\begin{aligned} u(y)= & {} \tilde{u} \left( \frac{1 + \rho y}{1 + \rho } \right) \rightarrow {\left\{ \begin{array}{ll} u(0) = \tilde{u}\left( \frac{1}{1+\rho }\right) \\ u(1) = \tilde{u}(1) \end{array}\right. } \end{aligned}$$
The first term of Eq. (2.3) \(u(y_{it})\), is the classic consumption utility. Since in this set-up the agent has only two possible cases, behaving green or grey, \(y=0\) or \(y=1\),18 I define \(u(\cdot )\) according to Eq. (2.4). Here, \(\tilde{u}\) has the classic properties of \(\tilde{u}'>0\) and \(\tilde{u}''<0\). The damage term \(d(p_t^j)\) also has its classic properties of \(d' > 0\), \(d''>0\) and \(d'(0) = 0\).

2.3 Kantian Behaviour

Since a purely homo-oeconomicus approach cannot explain green behaviour, we must consider the moral motivation mentioned in the Introduction. People behave in a green way because they think it is the right thing to do, not because it is in their best economic interest to do so. But what is the right thing to do in a framework like this one? To tackle this question, I have used the ideas of Immanuel Kant. In his exploration of what was ‘good’ and ‘bad’, he devised the idea that a good action was one that could be tested as a maxim rule, one that everyone would follow, known as the categorical imperative (Kant 1785). If this rule makes things better, then we assume that it is a good rule to follow, and thus that following it makes us good people. In order to use this idea for the present formulation, this dictum can be translated into: Which general rule of action should I follow to maximize (my) welfare, as I perceive it, given that everyone acts according to the same general rule? 19 \(^{,}\) 20

Naturally, this representation could be considered naive in face of reality. Why should each individual expect others to behave as he or she does? Most individuals do not, in fact, believe this. As noted in Brekke and Nyborg (2008), “... the categorical imperative defines one’s moral responsibility vis-a-vis society without referring to others’ actual behaviour, there is no presumption ... that he thinks others will in fact follow his example”. Now, it is reasonable to say that everyone is different with respect to this (Kantian) moral responsibility. Some people are indeed more responsible than others. The weighting parameter \(\alpha _i\) accounts for this fact. Larger values of \(\alpha _i\) mean that the agent is being more responsible (Kantian) than homo-oeconomicus. Hence, those who are fully homo-oeconomicus (or at least, behave as if they are) have \(\alpha _i = 0\). Those whose preferences are fully determined by the Kantian rule have \(\alpha _i = 1\).21

Returning to the utility, the agent will consider (assess) \(u_{it}^k\) as if everyone else were behaving as he or she is. In other words, this part of the utility function is modelled as though the agent assumes that everyone is behaving as he or she does in order to make a decision about how to behave. Strictly speaking, this is not what Kant meant with his categorical imperative, as already discussed in the Introduction. His was not a heteronomous ethic. In this sense, the present formulation is not categorical, autonomous or independent of external influence; on the contrary, it depends on the environment’s quality. Rather, my formulation is in line with the one used by Laffont (1975), which borrows the idea of choosing the good (ethical) rule when assuming that everybody behaves as oneself does, hence the term ‘Kantian morale’.

Regarding pollution, the level considered by the agent is the (estimated) pollution level \(p^j_{it}\) (\(j \in \{h,k \}\)). This in turn depends on two factors: the perceived (past) pollution level, \(p^p_{t-1}\) and the assumed emissions. Until Sect. 3, I assume that the agent has perfect information about the past pollution level, \(p^p_{t-1} = p_{t-1}\).

Coming back to the homo-oeconomicus and Kantian utility functions, we have that for \(u_{it}^h\), the agent understands that he is atomistic with respect to the society and thus he knows that his impact on the environment is negligible with respect to total emissions. This translates to a damage term \(d(p_t)\) that does not vary with his individual decision \(y_{it}\), but depends only on the society’s behaviour \(y^s_t\). On the other hand, \(u_{it}^k\) is the utility taken into account by the agent when using a Kantian view. In other words, the agent considers that everyone behaves as he or she does, implying that society’s emissions \(y^s_t\) will follow his choice \(y_{it}\), as well as \(x^s_t\) with \(x_{it}\). Putting all of the pieces together and rewriting the previous equations, we get:
$$\begin{aligned} p_{it}^k = (1 - \delta ) p_{t-1} + \gamma y_{it} \quad \text {and} \quad p_{it}^h = (1 - \delta ) p_{t-1} + \gamma y_t^s \end{aligned}$$
And replacing these results into the agent’s utility function we arrive at:
$$\begin{aligned} U_{it} = U_i(y_{it}) \equiv u(y_{it}) - \alpha _i \cdot d\big ((1 - \delta ) p_{t-1} + \gamma y_{it}\big ) - (1 - \alpha _i) \cdot d\big ((1 - \delta ) p_{t-1} + \gamma y_t^s\big ) \nonumber \\ \end{aligned}$$

2.4 Agent’s Behaviour

The agent can choose to behave in a way that is green \((y_{it} = 0)\) or grey \((y_{it} = 1)\). Associated levels of utility are given by, respectively
$$\begin{aligned} \text {Green }(y_{it}=0): \quad&u(0) - \alpha _i \cdot d\big ((1 - \delta ) p_{t-1} + 0\big ) - (1 - \alpha _i) \cdot d\big ((1 - \delta ) p_{t-1} + \gamma y_t^s\big )\\ \text {Grey }(y_{it}=1): \quad&u(1) - \alpha _i \cdot d\big ((1 - \delta ) p_{t-1} + \gamma \big ) - (1 - \alpha _i) \cdot d\big ((1 - \delta ) p_{t-1} + \gamma y_t^s\big ) \end{aligned}$$
The agent will behave in a green manner if the first expression is greater than or equal to the second one. Note that the last term in both expressions is just the damage term from the homo-oeconomicus point of view, which for the agent is given, and hence it does not influence his decision. Applying this inequality to the previous expressions and rearranging the terms, we get:
Consumer i is green at time t, i.e. \(y_{it} = 0\) iffwhere
$$\begin{aligned} K(p)\equiv & {} d\big ((1 - \delta ) p + \gamma \big ) - d\big ((1 - \delta ) p\big ); \quad K'(p)> 0. \\ H(\rho )\equiv & {} u(1) - u(0) = \tilde{u}(1) - \tilde{u}(\frac{1}{1+\rho }) ; \quad H'(\rho ) > 0, \end{aligned}$$
As it has been seen here, the utility function in Eq. (2.6) is the basis on which an agent chooses his behaviour: he considers his behaviour most preferable if this function is maximized with respect to \(y_{it}\). The function incorporates the value he attaches to behaving ethically, according to a Kantian-like logic; this value is based on the calculation of what damages and utility from consumption would arise to society if all agents behave like he does. The preference structure used here is compatible with the one used in Ulph and Daube (2014), the only difference being that I allow for perfect substitutes between green and grey products and that I rule out a mix of consumption. The preferences in the present work can be written as in Ulph and Daube (2014) and for a detailed explanation refer to Appendix B.

Concerning the functions K(p) and \(H(\rho )\), two points are worth mentioning: First, K(p) is increasing in p (since \(d'' > 0\)). It is also increasing in \(\gamma \). Second, \(H(\rho )\) is increasing in \(\rho \). The first observation implies that having higher pollution levels, \(p_{t-1}\), will yield more people adopting green behaviour. This is simply because the condition in 2.7 is met for lower values of \(\alpha _i\) when \(K(p_{t-1})\) increases. Therefore, a higher proportion of society will choose to behave in a green way. This is the moral driver of the agent. It increases with the pollution level and with the impact of the ‘morally estimated’ national emissions, \(\gamma \).22 The second point corresponds to the obvious fact that the more expensive the green product is (higher values of \(\rho \)), the higher the cost (in terms of consumption) that will be borne by agents exhibiting green behaviour.23

For a given price of the green product and a perceived pollution level, there is a value of \(\alpha \), say \(\alpha ^*\), that divides the society in two: those behaving in a green way (\(\alpha _i \ge \alpha ^*\)) and those behaving in a grey one (\(\alpha _i < \alpha ^*\)). Hence we can define a function \(\theta (p_{t-1},\rho )\) that tells us the proportion of people exhibiting grey behaviour for the values of \(p_{t-1}\) and \(\rho \), as:24
$$\begin{aligned} {\alpha }^*_t= & {} \theta (p_{t-1},\rho ) \equiv \min \left\{ \frac{u(1) - u(0)}{d\big ((1 - \delta ) p_{t-1} + \gamma \big ) - d\big ((1 - \delta ) p_{t-1}\big )} \; , \; 1 \right\} \nonumber \\= & {} \min \left\{ \frac{H(\rho )}{K(p_{t-1})} \; , \; 1 \right\} \end{aligned}$$
I note two important things about this function \(\theta (\cdot )\). There is a critical level of pollution \(p_{min}\) such that if \(p_{t-1} < p_{min}\) everyone’s behaviour is grey. The intuition for this is straightforward: Since the environment is already so clean, the gains are too small to behave in a green way. To find this value, we solve: \(d\big ((1 - \delta ) p_{min} + \gamma \big ) - d\big ((1 - \delta ) p_{min}\big ) = H(\rho )\). Second, there will be always some people exhibiting grey behaviour: We can always find some positive \(\alpha _i\), such that \(\alpha _i \, K(p_{t-1}) < H(\rho )\), for any given \(p_{t-1} > 0\) and \(\rho > 0\). The inequality is easily verified for \(\alpha _i = 0\): The agent with no Kantian attitude has no incentive at all to behave in a green way. We can graph this function, which relates \(p_{t-1}\) with \(\theta \). For easiness in further discussion, I graph the pollution level on the y-axis and how green the society is on the x-axis (the greenest being on the right side), as in Fig. 1.
Fig. 1

Share of grey behaving people with respect to perceived pollution level

The dashed line shows that a lower value of \(\rho \) (having a cheaper green product) implies more people adopting green behaviour. As we noted before, below a threshold level of pollution \(p_{min}\), everyone behaves in a grey way: \(\theta (p_{t-1}, \rho ) = 1 \; \forall \; p_{t-1} \le p_{min}\). We can also note that as the state of the environment worsens, the society becomes greener. This last observation comes from the fact that \(K(p_{t-1})\) is an increasing function of \(p_{t-1}\). However, it is also intuitive. As the environment worsens, even people with low Kantian attitude (low \(\alpha \)) are confronted with high marginal damages and prefer green behaviour. Green behaviour is not a result of agent thinking that his contribution to the environment will make things better (he assumes his contributions are negligible). It comes from the idea that as his perception of the environment worsens, the agent’s moral motivation increases (K(p)) and therefore more people exhibit green behaviour. In the same fashion, if behaving in a green way gets cheaper (lower values of \(\rho \) and hence lower \(H(\rho )\)), and the moral motivation remains the same, more people will behave in a green way.
Fig. 2

Evolution of pollution

Now we can return to the pollution evolution (Eq. 2.1). Rearranging the terms and defining the change in pollution as \(\Delta p_t = p_t - p_{t-1}\), we have:
$$\begin{aligned} \Delta p_t = \underbrace{\gamma \; \theta (p_{t-1}, \rho )}_{\begin{array}{c} \text {Actual} \\ \text {emissions} \end{array}} \; - \underbrace{\delta p_{t-1}}_{\begin{array}{c} \text {Natural} \\ \text {absorption} \end{array}} \end{aligned}$$
The first term corresponds to present emissions: It is the impact of consumption on the environment (\(\gamma \)) multiplied by the share of people behaving in a grey way (\(\theta (\cdot )\)) multiplied by their grey consumption, which is 1. The second term is the natural absorption of the pollutant.25

I divide the Eq. (2.9) by \(\gamma \) and graph the two terms of the r.h.s. in Fig. 2, where \(p_{t-1}\) is represented on the vertical axis and \(\theta (\cdot )\) on the horizontal one, as in Fig. 1. The curve with the kink stands for the first term, and the straight line denoted by \(-\gamma /\delta \), represents the amount of pollution captured in a natural form. Note that having the curve \(\theta (p_{t-1}, \rho )\) to the left of the line \(-\gamma /\delta \) implies that \(\Delta p_t > 0\), and hence the pollution level increases, and vice versa. Using an historical approach, we begin from the lower left corner and follow the arrows. At this starting point there is no pollution, and therefore everyone exhibits grey behaviour, \(\theta (p_{t-1}, \rho ) = 1\) (recall that the x-axis is inverted). This leads to increasing levels of pollution up to the point where people begin to respond to the issue, \(p_{min}\) (the kink to the right). As society continues to pollute, more people respond accordingly (for bigger values of \(\alpha \)) and it becomes greener. Society then reaches a point where emissions are equal to natural absorption, at (\(\theta ^*\), \(p^*\)). At this point, a proportion \((1 - \theta ^*)\) of the society is exhibiting green behaviour, which translates into an emission level equal to what Mother Nature can absorb at that pollution level \(p^*\). This equilibrium point is stable, since if \(p_t > p^*\) then \(\Delta p_t < 0\) and if \(p_t < p^*\) then \(\Delta p_t > 0\).26

2.5 Endogenizing \(\rho \): Introducing a Political Framework

I now introduce a simple political framework with two parties: the green party and the grey party. They only differ in that each party has a different environmental policy. For simplicity, I assume that the grey party does nothing about the environment, whereas the green party will implement a green-oriented policy. In this set-up, a green policy will simply be some tax/subsidy scheme to reduce the price gap between the green and grey products. This policy could be accomplished by taxing the grey products, subsidizing the green ones, or using both instruments at the same time. To keep the model simple and avoid side effects, I assume that the green government implements a policy that is budget neutral. In other words, it will charge a tax to the grey good, then use that tax revenue to finance a subsidy to the green good, hence lowering \(\rho \). As noted above, the grey government will alter nothing. The implementation or removal of the policy will follow the political cycle, and hence the speed of the update will depend on this cycle. Concerning the level of the tax (and the corresponding subsidy), I will assume that it is determined in order to lower \(\rho \), from say \(\rho _1\), to \(\rho _2\), the latter fixed exogenously.

Following the previous reasoning, it is assumed that green-behaving people will demand green policy, which translates in this framework into smaller values of \(\rho \). As stated in the Introduction, this idea is supported by the results found by Comin and Rode (2013) and Schumacher (2014). But it is also a reasonable assumption. For a given person with an attitude \(\alpha _i\), it will be moral-economically better to behave in a given manner. This behaviour will in turn depend on pollution; therefore, the same person could behave in a green or grey manner depending on the state of the environment. Since it is his optimal choice (from a moral-economical standpoint), it is reasonable to think that the relationship between behaving in a green way and voting in a green way is an increasing one. Again for simplicity, I assume that this relationship is one to one, meaning that a green-behaving person will vote for a green policy. Note that the vote does not only depend on the attitude of the agent. Even the most Kantian person could vote grey if the environment is sufficiently clean. It is also worth noting that this one-to-one assumption does not change the resulting dynamics.

Following the majority rule, we know that when the share of green people is bigger than some threshold \((1 - \overline{\theta })\) (and, hence, \(\theta \le \overline{\theta }\)), a green government will be elected. In a simple case (for graphical illustration) where \(\alpha _i\) is uniformly distributed, we have that \(\overline{\theta } = 1/2\).

In Fig. 3 the two upward sloping curves plot the share of grey consumers, for the cost parameter \(\rho \) that the green and the grey governments impose, respectively. Since the grey government imposes \(\rho = \rho _1\), the upper curve is relevant for \(\theta < \overline{\theta }\), while the lower curve applies when \(\theta > \overline{\theta }\).27 Therefore, the upward-sloping curve that determines the equilibrium is the combination of the solid segments of the two curves. As we can observe in this figure, we have two equilibria: \(\theta ^*_1\) a ‘grey’ equilibrium (on the left side: \(\theta ^*_1 > \overline{\theta }\)) and \(\theta ^*_2\) a ‘green’ one, on the right side (\(\theta ^*_2 < \overline{\theta }\)).
Fig. 3

Two equilibria case

It is interesting to note that the multiple equilibria occur despite the decrease in pollution level on the right side of the discontinuity. Having a better environment gives agents less moral motivation to exhibit green behaviour. But since a green government is in place, behaving in a green way is also cheaper, an effect that overrides the decrease in moral motivation.

Now let us suppose that we have a political framework where alliances between political actors about specific policies (green policies in our case) are more easily formed.28 If this is the case, we can expect to have more discontinuities in the dynamics on the left and right side of \(\overline{\theta }\). If party coalitions are formed (i.e. in a parliamentary regime), an initial green policy could be a mild one. In this case, the tax (and corresponding subsidy) would be only a portion of the tax previously discussed. Thus instead of having one discontinuity, we could have more discontinuities of smaller magnitude.

I illustrate this idea with an example with three discontinuities, as in the following set-up:
  • \(\theta > \overline{\theta }_1 \rightarrow \rho _1\)   (100% grey government)

  • \(\overline{\theta }_1 \ge \theta > \overline{\theta }_2 \rightarrow \rho _2\)   (partially grey government)

  • \(\overline{\theta }_2 \ge \theta > \overline{\theta }_3 \rightarrow \rho _3\)   (partially green government)

  • \(\overline{\theta }_3 \ge \theta \quad \rightarrow \rho _4\)   (100% green government)

with \(\overline{\theta }_1> \overline{\theta }_2 > \overline{\theta }_3\) and \(\rho _1> \rho _2> \rho _3 > \rho _4\). In this case, we can have different outcomes depending if \(\overline{\theta }_1 \lessgtr \theta ^*_1, \; \overline{\theta }_2 \lessgtr \theta ^*_2\) and \(\overline{\theta }_3 \lessgtr \theta ^*_3\).
Different cases are depicted in Fig. 4a–c.
Fig. 4

Different possibilities for multiple equilibria. a “Lock-in” case. b “Cascade” case. c “Overshoot” case

As before (Fig. 3), the system starts from the bottom left corner, where there is no pollution and everyone behaves in a grey way. Again, at some point (\(p_{min}\)), some people start to exhibit green behaviour and the system moves toward \(\theta ^*_1\). In the first case, Fig. 4a, the system will converge to equilibrium \(\theta ^*_1\), getting “locked-in”. However, if different conditions are met, for instance if \(\overline{\theta }_1< \overline{\theta }_2< \overline{\theta }_3 < \theta ^*_1\) as in Fig. 4b, called “cascade” case, the story becomes a different one. When moving toward \(\theta ^*_1\), the system crosses the \(\overline{\theta }_1\) threshold, which activates a lower level of \(\rho \), \(\rho _2\). At this instant, the system moves toward the solid line between \(\overline{\theta }_1\) and \(\overline{\theta }_2\), in a horizontal fashion, since the pollution level \(p_t\) is the same before and after the change of active \(\rho \). At this point, the system continues its path on the upward-sloping curve, now moving toward \(\theta ^*_2\). Again it will cross a threshold, in this case \(\overline{\theta }_2\), repeating the previous process. When the last threshold \(\overline{\theta }_3\) is crossed, the system finally converges at equilibrium point \(\theta ^*_4\). In other words, as the society becomes greener in its behaviour, greener governments get elected. If the conditions are ‘right’, this might produce a cascading process that concludes with the (fully) green government holding power.

A third and similar situation might also happen. If similar conditions are met, but in this case only \(\overline{\theta }_1 < \theta ^*_1\) and \(\theta ^*_1< \overline{\theta }_2 < \theta ^*_2\) (and \(\theta ^*_3< \overline{\theta }_3 < \theta ^*_4\)), a similar process can unfold, as depicted in Fig. 4c. The difference with the previous case is that now, when the threshold is crossed and the systems moves toward the new active \(\rho \) value (moving horizontally to the right), it will cross another threshold and the process is re-started without moving on the upward-sloping curve. In the example depicted here, it crosses all three thresholds, always at the same pollution level, arriving to the right-most active curve. At this point, the system has overshot and it will move toward the equilibrium point \(\theta ^*_4\), but now moving down and left. As one can imagine, different cases can occur, depending on the (relative) values of \(\overline{\theta }_i\) and \(\theta ^*_i\). It might be the case where the greenest equilibrium \(\theta ^*_4\) is not reached and the system converges to another equilibrium, for example \(\theta ^*_3\). If might also be the case where some threshold’s crossings are done in a fashion as in Fig. 4b and others as in Fig. 4c. The point to make here is that the relative position of the equilibria and thresholds determine the final outcome of the society, possibly moving it to a green(er) equilibrium.

On another perspective, the political framework might not be the only source of change in \(\rho \). Another way of endogenizing \(\rho \) is from an evolution in the production process of the green products. As society becomes greener and more green products are bought, it would be logical to expect that: a) economies of scale begin to appear and b) technological innovations occur in green product production methods (in contrast to a more mature grey production technology). Again, this last point could be ‘artificially’ induced by a political agreement in response to the society’s demands. In the end, either method of endogenizing \(\rho \) leads to the same result: The system has a tipping point from which it moves toward a greener equilibrium.

3 A Green Nudge

As we saw in Fig. 4a–c, all we need to nudge the system is to shift the threshold levels \(\overline{\theta }_i\) or move the curve \(\theta (\cdot )\), even temporarily. In this previous example, the two societies were not precisely the same, but I choose this framework in order to exemplify how a ‘jump’ from a grey equilibrium \(\theta ^*_1\) to a greener one \(\theta ^*_4\) might occur. To do so, we should recall that function \(\theta (\cdot )\) depends on the price (differential) of the green product \(\rho \) as well as on the perceived pollution. This last term, as its name indicates, is related to how people perceive the pollution levels, which I assumed in the previous sections to be correctly perceived (\(p^p_{t-1} = p_{t-1}\)). In this Section, we assume that the agent has a perception bias \(\Omega \), such that: \(p^p_{t-1} = \Omega \, p_{t-1}\). Changing the value of \(\Omega \) will shift the curve \(\theta (p_{t-1}, \rho )\) up or down. Hence, if we are able to change this bias, intentionally or by chance, then we could push the system from the grey equilibrium (or path) into the green one. As we expect, this variation does not need to be permanent. When the system has crossed the last threshold (\(\overline{\theta }_3\)), then there is no longer a need for such a nudge to sustain the new equilibrium.29

The change in \(\Omega \) could be due to different causes. The perception of pollution levels depends on information. A higher level of exposure to information about climate change might alter this parameter. Extreme natural events, such as hurricanes hitting more frequently and severely, could have an impact on public perception of the climate status (Lee and Markowitz 2013).

Nevertheless, increasing information or changing perceptions are not the only ways of inducing change. Other avenues exist for nudging the system. One such channel could be political ‘noise’. Specifically, a country could be close to its tipping point when a greener government is elected due to non-environmental reasons (left vs. right, social reforms, etc.). When this government implements green policy measures, this action could also trigger a cascading path toward the green equilibrium. Independent studies by NGOs or the media might also yield the same type of nudge by increasing awareness about the environment and possibly causing constituencies to push for changes to public policy.

In the same vein, it is worth noting that a multi-threshold situation, meaning one in which there is more than one \(\overline{\theta }_i\), may ease the switching process. If we have a similar situation with multiple threshold levels, centred on 1 / 2 for example (meaning half are on the \(\theta (\cdot ) < 1/2\) side and the other half are on the \(\theta (\cdot ) > 1/2\) side), it follows that the cascading process begins at a lower level of greenness of the society (starting at higher levels of \(\theta (\cdot )\)). This implies that the distance, in terms of greenness of the society, between the grey (original) equilibrium and the left-most discontinuity, is smaller in the case with multiple thresholds (\(\overline{\theta }_1\)) than in the case with only one discontinuity (\(\overline{\theta }\)): \({(\theta ^*_1 - \overline{\theta }_1) < (\theta ^*_1 - \overline{\theta })}\). Therefore, the nudge needed to start a switching process (to a greener equilibrium) is smaller in the multi-threshold situation. In this sense, having the latter situation could facilitate this switching process, increasing its likelihood of happening.

4 Social Approval and Social Pressure

So far I have used an absolute moral gain, meaning that the moral or green motivation of the agents comes only from an inner motivation. The agent acts in a given manner because he believes it is the right thing to do (the Kantian idea referred to above). I now introduce what I will call a relative moral gain. In this case, the agent also derives utility from being accepted by his peers and society in general. This idea has already been discussed in Hollander (1990), Nyborg et al. (2006), and Rege (2004), and simply states that the agent derives extra utility coming from the proportion of people behaving as he or she does. This social approval is increasing in the aforementioned proportion, meaning that it will be maximal if everybody behaves as he or she does, null if nobody does, and it is increasing between the two extremes. As one can expect from the cited papers, the element of social approval will make the transition from one equilibrium to the other harder. The aim of this addition is to point out that social mechanisms can affect the transition process and therefore could be another source of explanation of different behaviour among similar countries.

I call \(u^a_{it}\) the satisfaction from social approval that agents get, which will in turn depend on both their own and society’s behaviour as measured by the share of grey consumers \(\alpha ^*_t\). A corresponding weighting parameter \(\beta \) is introduced (\(\beta > 0\)), producing the following version of the agent’s utility function:
$$\begin{aligned} U_{it} = (1 - \alpha _i) u_{it}^h + \alpha _i \, u_{it}^k + \beta \, u^a_{it} \end{aligned}$$
Concerning the form of \(u^a_{it}\), I will assume that the agent gets a social reward from people behaving as he does.30 I take a simple linear specification, assuming that if the agent is behaving in a grey (green) way, his utility will increase with \(\alpha ^*_t\) (with \(1 -\alpha ^*_t\)).31 Hence we have a piecewise function:
$$\begin{aligned} u^a_{it} = \left\{ \begin{array}{ll} 1-\alpha ^*_t &{} \quad \text {if } y_{it} = 0 \qquad \text { (behaving green)} \\ \alpha ^*_t &{} \quad \text {if } y_{it} = 1 \qquad \text { (behaving grey)} \end{array} \right. \end{aligned}$$
We notice that if the society is mainly exhibiting a grey (green) behaviour, agents behaving grey (green) receive more social approval than those behaving green (grey). In other words, the pressure leans toward the leading behaviour present in the society.32 Deriving the counterpart of condition (2.7) for green behaviour for our new set-up, we find:
Consumer i behaves green at time t, i.e. \(y_{it}= 0\) iffIndifferent agent:
$$\begin{aligned} \alpha ^*_t K(\Omega p_{t-1}) + \beta \, (1 - 2 \, \alpha ^*_t)= & {} H(\rho ) \; \iff \nonumber \\ \alpha ^*_t [K(\Omega p_{t-1}) - 2 \beta ]= & {} H(\rho ) - \beta \end{aligned}$$
In other words, when \(S(\alpha ^*_t) > 0\), which occurs when \(\alpha ^*_t < 1/2\), a bigger share of the society will behave in a green manner and vice-versa.33 I define a function \(\tilde{\theta }(p_{t-1},\rho ,\beta )\), which is the generalization of the function \(\theta (\cdot )\): it allows in fact for positive values of \(\beta \), while \(\theta (\cdot )\) is the special case of \(\tilde{\theta }(\cdot )\) with \(\beta =0\).
The introduction of peer pressure might change the way the dynamics goes, depending on the value of \(\beta \). In order to understand the conditions in which the dynamics changes, let us consider the following results:

Result 1: \(\alpha ^*_t = 1\) requires \(K(\Omega p_{t-1}) < \beta + H(\rho )\)    (for all values of \(\beta \)).

Proof: if \(\alpha ^*_t = 1\), then \(S(\alpha ^*_t) = -\beta \) so that according to (4.3), agent i chooses to behave grey if \(\alpha _i \, K(\Omega p_{t-1}) - \beta - H(\rho ) < 0\). This holds for all i iff

\({K(\Omega p_{t-1}) < \beta + H(\rho )}\). \(\square \)

This result shows that for a small pollution stock, \(p_t\) below \(p_{min}\) defined by \(K(\Omega p_{min}) = \beta + H(\rho )\), there is an equilibrium in which everybody behaves grey.

Result 2: \(\alpha ^* = 0\) requires \(\beta > H(\rho )\).

Proof: if \(\alpha ^* = 0\), then \(S(\alpha ^*) = \beta \) so that according to (4.3), agent i chooses to behave green if \(\alpha _i \, K + \beta - H(\rho ) > 0\). This holds for all i iff \({\beta - H(\rho ) > 0}\). \(\square \)

This result shows that if \(\beta \) is critically large, there is an equilibrium with everybody behaving green. Intuitively, the peer pressure is so large that if everybody behaves green, the peer pressure for green behaviour provides so strong incentives that everybody wants to act green.
Result 3:
  1. a)

    An interior solution must satisfy \(\alpha ^*_t \in (0,1)\) and \(\alpha ^*_t = \alpha ^{int}_t \equiv (H(\rho ) - \beta ) / (K(\Omega p_{t-1}) - 2\beta )\);

  2. b)

    If \(\beta < H(\rho )\), an interior solution arises if \(K(\Omega p_{t-1}) > H(\rho ) + \beta \);

  3. c)

    If \(\beta > H(\rho )\), an interior solution arises for \(K(\Omega p_{t-1}) < H(\rho ) + \beta \), but is unstable.

  1. (a)

    follows from Eq. (4.3).

  2. (b)

    If \(\beta < H(\rho )\), \(0< \alpha ^{int}_t < 1\) iff \(K(\Omega p_{t-1}) > H(\rho ) + \beta \).

  3. (c)

    If \(\beta > H(\rho )\), positive \(\alpha ^{int}_t\) requires \(K(\Omega p_{t-1}) < 2\beta \), \(\; \alpha ^{int}_t < 1\) requires \(K(\Omega p_{t-1}) < \beta + H(\rho )\), which together requires \(0< \alpha ^{int}_t< 1 \leftrightarrow K(\Omega p_{t-1})< H(\rho ) + \beta < 2\beta \). To see why this equilibrium is unstable, define \(A(\alpha ^*_t)\) such that if \(\alpha _i = A(\alpha ^*_t)\) then agent i is indifferent between acting green or acting grey given that in society \(\alpha ^*_t\) agents act grey. From (4.3) we see that \(A(\alpha ^*_t)\) is defined by: \(A(\alpha ^*_t) = (H(\rho ) - \beta ) / K(\Omega p_{t-1}) + (2\beta / K(\Omega p_{t-1})) \alpha ^*_t\). In an interior equilibrium we must have \(A(\alpha ^*_t) = \alpha ^*_t\), which is equivalent to \(\alpha ^*_t = \alpha ^{int}_t\). Moreover, the usual stability argument requires \(A' < 1\), i.e. \(K(\Omega p_{t-1}) > 2\beta \). Since \(K(\Omega p_{t-1}) > 2\beta \) makes \(\alpha ^{int}_t\) to fall outside the unit interval for \(\beta > H(\rho )\), there is no stable interior solution.34 \(\square \)

We can now fully characterize the equilibrium for the two cases, \(\beta < H(\rho )\) and \({\beta > H(\rho )}\). If \({\beta < H(\rho )}\), the equilibrium is unique. For small pollution level, such that \(K(\Omega p_{t-1}) < H(\rho ) + \beta \), everybody acts grey, and once \(p_t\) is larger such that \(K(\Omega p_{t-1}) > H(\rho ) + \beta \), we arrive at the interior solution, \(\alpha ^* = \alpha ^{int}\), which implies that \(\alpha ^*\) is then declining monotonically with \(p_t\), but never reaches zero. This can be observed in Fig. 5.
Fig. 5

Share of grey behaving people with social approval, with \(\beta < H(\rho )\)

Different cases, for different values of \(\beta \) have been graphed in Fig. 5. All curves cross at the same point. This is due to the fact that when \(\alpha ^*_t = 1/2\), the peer effect disappears. In other words, it becomes neutral: when half of the society behaves in a green manner and the other half in a grey one, there is no peer effect and they must coincide at this point. We also notice that \(p_{min}\), defined by \(K(\Omega p_{min}) = \beta + H(\rho )\), increases with \(\beta \). The intuition for this is the following: If everyone is exhibiting grey behaviour and is affected by this peer effect, the agent will have less incentive to behave in a green manner. In other words, the pollution will have to be higher in order for someone to care about it and endure this new social pressure.

If \(\beta > H(\rho )\), we have multiple equilibria. In equilibrium everybody can act grey as long as \(K(\Omega p_{t-1}) < H(\rho ) + \beta \). It is always an equilibrium to have everybody behaving green, but this requires coordination among agents. If we rule out the unstable equilibrium and the coordination needed to make everybody switch from acting grey to acting green, the following pattern emerges: everybody acts grey as long as \(K(\Omega p_{t-1}) < H(\rho ) + \beta \), then \(p_t\) rises so that \(K(\Omega p_{t-1})\) rises and once \(K(\Omega p_{t-1}) > H(\rho ) + \beta \) everybody switches to green. After this point, everybody remains behaving green, even when \(p_t = 0\). This trajectory is depicted in Fig. 6.
Fig. 6

Share of grey behaving people with social approval, with \(\beta > H(\rho )\)

The intuition of the main two cases, \(\beta \lessgtr H(\rho )\) is the following: When \(\beta \) is small, the social approval effect is too small to change the mechanisms of the model. However, when \(\beta \) is bigger, \(\beta > H(\rho )\), peer pressure kicks in and multiple equilibria arise.35

Let us now return to the case where \(\beta < H(\rho )\). As we notice in Fig. 7 (the counterpart of Fig. 3), the basic dynamics is the same.
Fig. 7

Two equilibria with social approval

Now, however, the equilibrium points have moved further apart from one another. We can understand that this effect is due to some kind of ‘attractors’ situated in each extreme (\(\alpha ^*_t= 0\) and \(\alpha ^*_t = 1\)). Since in this particular case each equilibrium point is situated in either ‘half’ of the portrait (\(\alpha ^*_t > 1/2\) and \(\alpha ^*_t < 1/2\)), they shift to the left and right side respectively. Therefore, the resulting equilibria are more separated than before: The societies now behave dissimilarly. We can observe this feature observing the horizontal distance between equilibria points \(\{A, A'\}\) (for \(\beta =0\)) and \(\{B, B'\}\) (for \(\beta =0.18\)), in Fig. 7.

Recalling the cascading effect seen in the preceding pages, we notice now that peer effect implies that the threshold needed to switch regimes is bigger,36 and at the grey equilibrium, the pollution is higher. Therefore, even though peer effect itself might not be the main driver to these equilibria, it might be an important factor to consider when analysing the switching from one equilibrium to another. Looking at the green side of the system, we might also note that the peer effect is beneficial for society in terms of pollution: It would drive society toward lower levels of pollution.

5 Conclusions

This model attempts to explain why developed countries that are similar in terms of income level and political system take different actions with respect to the environment. I model people in society to be different, ranging from those who possess stronger ‘Kantian’ attitudes to those who act in a purely homo-oeconomicus manner. Green behaviour is derived from being morally responsible. People with a stronger green attitude follow, to some extent, a Kantian imperative which makes them more prone to exhibit green behaviour. With this set-up, people can either behave in a green fashion (i.e., contribute to the environment) or in a grey way. Adding a simple political framework, in which a green policy is implemented if the majority behaves in a green way, I obtain a dynamic model with multiple equilibria. The model reveals that the same society can arrive at two different equilibria: a green one or a grey one.

Using the result of multiple equilibria, I have also shown that it is possible to switch from a grey trajectory to a green one. Providing information to people, for example, can raise their awareness about the environment and increase the chances of social change. This outcome reinforces the findings of Corbett and Durfee (2004) and Dunwoody (2007), who show that mass media has a large influence over social concerns, which can result in changes to a country’s environmental behaviour.

I have assumed that each person’s green attitude is fixed. In other words, it is as if people were born with this trait and there is no chance of changing it over time. Obviously, this is an extreme assumption. However, the idea here is to create a model that can explain the faster changes in green behaviour observed over the last years.

To tackle this point, we could introduce a second slower evolutionary process of attitude (rather than, or in addition to, taking into account only changes in behaviour). This process could be understood by considering how education, for example, changes attitudes over time. Such a change would be of the type described and modelled in Bisin and Verdier (2000). Adding this extension to the model would not change the previous results (especially the one concerning the nudge), and it would moreover reinforce the idea that specific countries treat the environmental issue differently as a result of cultural considerations.

A second vein of work could be related to the possible interaction with other countries. As mentioned before, this work attempts to provide an explanation of differences among similar countries concerning a global pollutant. For the sake of simplicity no interplay between countries, through the stock of pollutant, has been introduced. Pursuing this line of thought, using a game theoretical framework, could give new insights about this phenomenon.

Another extension might concern the relationship between the change in social awareness and/or behaviour and a more complex political framework model. In particular, the addition of a political mechanism for setting a tax rate could be of interest. Moreover, this process could consider the interaction with foreign nations, linking the political area to the international scenario.


  1. 1.

    The EKC is similar to a traditional Kuznets curve. As a country gets richer and more developed, it begins to pollute more (as measured on a per-capita basis). After reaching a certain developmental level, society begins to grant more importance to the environment; therefore, it starts to pollute less as it becomes richer. Hence we observe an inverted U-shape illustrating the relationship between pollution and income per capita. This literature goes back to the early 1990s, including works by Grossman and Krueger (1991), Shafik (1994), and Panayotou (1993), among others. Although the EKC was used to analyse local pollution, it could also be used to compare countries concerning global pollutants.

  2. 2.

    For the year 2011, the emissions per capita (in metric tons of CO\(_2\)) were: USA, 17; Canada, 14.1; Australia,16.5; Denmark, 7.2; Finland, 10.2; Germany, 8.9; Netherlands, 10.1; Sweden, 5.5; and UK, 7.1. Average of each group, respectively: 15.9 and 8.2. The GDP per capita (in thousands of current US$) were: USA, 49.8; Canada, 52; Australia, 62.1, Denmark, 61.3; Finland, 50.8; Germany, 45.8; Netherlands, 53.5; Sweden, 59.6; and UK, 41. Average of each group, respectively: 54.7 and 52. Source: World Development Indicators, The World Bank.

  3. 3.

    This assertion is considering a public good setting where economics incentives to behave green are few or non-existent.

  4. 4.

    Sweden was the first country to introduce a carbon tax in 1991, and they have been increasing it over time. They are currently aiming to increase it again in order to “point out how to achieve the 2050 vision of zero net GHG emissions” in their Climate Roadmap. (Energy Policies of IEA Countries, Sweden, 2013 review).

  5. 5.

    At least not in the short term. In the long term, green countries could push an international environmental agreement, induced by a moral motive, which would lead to an increase in welfare.

  6. 6.

    From Kant’s work on moral philosophy (Kant 1785), we know that the categorical imperative formulates the equality postulate of universal human worth. Hence, while a Kantian attitude and an egalitarian attitude are not the same, they are closely related and we could proxy one to the other.

  7. 7.
  8. 8.
  9. 9.
  10. 10.

    Kuran talks about the collapse of Eastern Europe’s communist regimes. He divides the society into 10 types of people, ranging from those who are more in favour of a communist government, to those completely opposed to it. He shows that if some sort of threshold is crossed, protests can begin, which can encourage those initially less likely to go against the incumbent regime to join in protesting. This process can lead to a cascading effect, which can in turn trigger the collapse of the whole regime.

  11. 11.

    Persson et al. (2000) used a theoretical model to show that presidential regimes should produce an under-provision of public goods (thus leading to a dirtier environment). On the other hand, Bernauer and Koubi (2004) found the opposite result. They use an econometric study to find evidence that presidential democracies provide more public goods than do parliamentarian democracies. More recently, Saha (2007) tested the previous hypotheses empirically. She finds that the electoral system has no effect on any of the environmental public good supply indicators, and that the nature of the political regime also has no significant impact.

  12. 12.

    This means that green people prefer others to behave in a green way as well, and grey people prefer others to behave in a grey way. Of course, it might be the case that being grey is always considered ‘bad’, even for grey people, as in the case of smoking.

  13. 13.

    If there were a green good that was cheaper than its grey counterpart, agents would automatically choose that good instead of the grey one for purely economic reasons. In the case with multiple products, if some green products were cheaper than their grey counterparts, we could recalculate the pollution produced by a new representative grey good and return to the set-up presented here.

  14. 14.

    The agent’s binary choice is used for its simplicity: It results in closed forms. Also, a model with mixed choice was developed and tested with simulations, arriving to equivalent results. The behaviour functions and the resulting dynamics turn out to be the same.

  15. 15.

    From the budget constraint we have that \((1+\rho )x_{it} + y_{it} = 1\). Since the agent only chooses the green or grey product, we have the case with \(y_{it}=0\) and hence, \(x_{it}=1/(1+\rho )\); or \(x_{it}=0\) with \(y_{it}=1\).

  16. 16.

    From the previous formulation it follows that emissions from the rest of the world are not taken into account (if it were the case that we were considering global pollutants). I proceed in this manner because I want to focus on the country’s own drivers to change, not on those coming from the international community. Taking this caveat into account, it is clear that the moral driver will be focused on national behaviour, the aim of this paper, rather than international behaviour (emissions).

  17. 17.

    As will be made clear in the following pages, the distribution of \(\alpha \) will not change the main results, but assuming a uniform distribution will certainly ease the subsequent calculations and simulations. The analysis will be performed using a uniform distribution. On the other hand, different distributions will simply change the place of the tipping point and the conditions needed to tip, as in Kuran (1991).

  18. 18.

    Since the agent has only two options for \((x_{it},y_{it})\), being these \((\frac{1}{1+\rho },0)\) or (0, 1), I simplify notation by just referring to \(y_{it}\), which can just take the values of 0 and 1. Moreover, from the budget constraint and products’ perfect substitutability, we have that the argument of the function \(\tilde{u}(\cdot )\) is \(x+y=(1+\rho y)/(1+\rho )\).

  19. 19.

    The original categorical imperative (or one of the original versions) was: “So act as if the maxim of your action were to become through your will a universal law of nature” (Kant 1785).

  20. 20.

    For a well-written essay on the relationship between the Kantian imperative and climate change, see Rentmeester (2010).

  21. 21.

    There is also another way of tackling this diversity: We might think that each person has the same Kantian attitude, but that the parameter \(\alpha _i\) instead reflects how ‘optimistic’ each person is. Although this is not the same idea stated here, a development and a proof of its equivalence is given in the Appendix A.

  22. 22.

    As stated in the previous pages, if we are considering global pollutants, the raise of pollution level could be a result of more than that particular country’s emissions. On the other side, the impact of (local) emissions \(\gamma \) is directly linked to the choices of the agent in situations in which he or she has control.

  23. 23.

    An interesting feature to note is that if we consider consumption levels as proportional to some income level w (i.e. comparing \(\tilde{u}(w)\) and \(\tilde{u}(\frac{w}{1+\rho })\)), the private cost of behaving in a green way, \(H(\rho )\), could be increasing, decreasing, or independent of the income level w depending on the functional form of \(\tilde{u}(\cdot )\). Since in this formulation I intentionally leave out the income effect, I use the case where \(\tilde{u}(w) = \ln (w)\) for simulations, which results in a \(H(\rho )\) that is independent of w. For details see Appendix C.

  24. 24.

    For the present and following definitions, I use a uniform distribution of \(\alpha _i\). If this were not the case, we would have a different function \(\theta (\cdot )\), but it would still retain the subsequent properties and results.

  25. 25.

    I use the usual linear natural absorption, as in Calvo et al. (2012) and Breton et al. (2010). A different form for this absorption could been used, but as it can be observed in Fig. 2, we will have in general only one crossing with \(\theta \). Hence, the addition is a more complex form does not add significant new results.

  26. 26.

    We can notice that the behaviour of each agent, and hence of the whole society, is determined by the pollution level, through the function \(\theta (p_{t-1},\rho )\) (which also depends on \(\rho \)). Therefore, finding an equilibrium point for the pollution level gives us the equilibrium situation of the whole society.

  27. 27.

    In other words, a tax and subsidy will be chosen in order to achieve a new differential in green and grey prices equal to \(\rho _2\). Details on the implementation of the policy (tax and subsidy chosen) can be found in Appendix E, where I show that the policy is equivalent to just changing \(\rho \) in the \(\theta (\cdot )\) function; this will shift the curve to the right, as depicted in Fig. 1.

  28. 28.

    This could be due to the fact that there are more than two political actors in place, or only two actors with multiple political dimensions to bargain over, etc.

  29. 29.

    We can see from this reasoning that it is not necessary to know exactly where the thresholds are, but only to be aware of the ability to nudge the system toward a greener equilibrium given these thresholds.

  30. 30.

    I therefore disregard the case of negative social pressure from people not behaving as the agent does. It can be quickly derived that including this second effect does not change the results.

  31. 31.

    This is done for simplicity. Alternatively - and more generally - specifying \(u^a_{it}\) as a function of \(v(\alpha ^*_t)\) (\(1 - v(\alpha ^*_t)\)) when the agent is behaving in a grey (green) way, with \(v(0) = 0\) and \(v(1) = 1\) for normalization and \(v'>0\), does not change the coming results.

  32. 32.

    This was named in the Introduction as an ‘ideological’ peer pressure. Also refer to footnote 12.

  33. 33.

    Since \(v^\prime > 0\) (for the general case with \(v(\cdot )\)), the point when \(S(\alpha ^*_t) = 0\) (neutral social approval: \(v(1 - \theta ) = v(\theta )\)) has to be when \(\theta = 1/2\). Hence if \(\theta < 1/2 \rightarrow S(\alpha ^*_t) > 0\) and vice-versa. Another way of modelling the peer pressure would be to say that the agent reacts to \(\theta _{t-1}\), the share of grey people in the previous period. This would introduce a difference equation into the system where lags would also play a role. In order to keep the model simple, I use the case where the agents instantly respond to society’s behaviour.

  34. 34.

    To see what stability means, consider \(K(\Omega p_{t-1}) < 2\beta \) and \(\alpha ^*_t = \alpha ^{int}_t\). Suppose the “almost indifferent agents”, i.e. the agents i with \(\alpha ^{int}_t< a_i < \alpha ^{int}_t + \epsilon \) for \(\epsilon \) arbitrarily small and positive, make a mistake (they “tremble” in game theoretic parlance) and choose to be grey instead of green. Then \(\alpha ^*_t = \alpha ^{int}_t + \epsilon \) and \(A(\alpha ^*_t) = \alpha ^{int}_t + (2\beta / K(\Omega p_{t-1})) \epsilon > \alpha ^{int}_t + \epsilon \) This implies that all agents i for which \(\alpha _i < \alpha ^{int}_t + (2\beta / K(\Omega p_{t-1})) \epsilon \) want to act grey, thus including the trembling almost-indifferent-agents. They are not willing to revert to green but instead keep behaving grey, while in addition some other agent have now switched to grey. This process will continue until everybody behaves grey. Similarly, if the tremble is in the green direction, we end up having everybody behaving green.

  35. 35.

    This follows the same intuition as the result obtained in Nyborg et al. (2006). In their model there are two extreme equilibria, completely green or completely grey, with a third unstable equilibrium in between.

  36. 36.

    \(\overline{\theta }\) has to be bigger: it has to get to B, which is further left than A.



I want to thank Bertrand Wigniolle, Eugenio Figueroa, and Scott Barrett for their invaluable help and support. I would also like to thank Maria Kuecken and Alessandra Pizzo for their thorough revision, as well as the various conference and seminar participants, two anonymous referees and the editor who have helped me formulate and articulate these ideas.

Supplementary material


  1. Andreoni J (1990) Impure altruism and donations to public goods: a theory of warm-glow giving. Econ J 100(401):464–477Google Scholar
  2. Baumgärtner S, Petersen T, Schiller J (2014) Bringing norms into action—the concept of responsibility. Working paper, University of LünenburgGoogle Scholar
  3. Bernauer T, Koubi V (2004) On the political determinants of environmental quality. Annual meeting of the American Political Science Association, Hilton Chicago and the Palmer House Hilton, Chicago, ILGoogle Scholar
  4. Bisin A, Verdier T (1998) On the cultural transmission of preferences for social status. J Pub Econ 70(1):75–97CrossRefGoogle Scholar
  5. Bisin A, Verdier T (2000) A model of cultural transmission, voting and political ideology. Eur J Polit Econ 16(1):5–29CrossRefGoogle Scholar
  6. Bisin A, Verdier T (2001) The economics of cultural transmission and the dynamics of preferences. J Econ Theory 97(2):298–319CrossRefGoogle Scholar
  7. Bowles S (1998) Endogenous preferences: the cultural consequences of markets and other economic institutions. J Econ Lit 36:75–111Google Scholar
  8. Brekke KA, Nyborg K (2008) Attracting responsible employees: Green production as labor market screening. Resour Energy Econ 30(4):509–526CrossRefGoogle Scholar
  9. Brekke KA, Kverndokk S, Nyborg K (2003) An economic model of moral motivation. J Pub Econ 87(9–10):1967–1983CrossRefGoogle Scholar
  10. Breton M, Sbragia L, Zaccour G (2010) A dynamic model for international environmental agreements. Environ Resour Econ 45(1):25–48CrossRefGoogle Scholar
  11. Bruvoll A, Halvorsen B, Nyborg K (2000) Household sorting of waste at source. Econ Surv 4(2000):26–35Google Scholar
  12. Buenstorf G, Cordes C (2008) Can sustainable consumption be learned? A model of cultural evolution. Ecol Econ 67(4):646–657CrossRefGoogle Scholar
  13. Calvo E, Rubio SJ et al (2012) Dynamic models of international environmental agreements: a differential game approach. Int Rev Environ Resour Econ 6:289–339CrossRefGoogle Scholar
  14. Comin D, Rode J (2013) From green users to green voters. Working Paper 19219, National Bureau of Economic ResearchGoogle Scholar
  15. Corbett JB, Durfee JL (2004) Testing public (un) certainty of science media representations of global warming. Sci Commun 26(2):129–151CrossRefGoogle Scholar
  16. Dunwoody S (2007) The challenge of trying to make a difference using media messages. In: Creating a climate for change, pp 89–104Google Scholar
  17. Extreme Weather and Climate Change in the American Mind April 2013.
  18. Grossman GM, Krueger AB (1991) Environmental impacts of a north American free trade agreement. Technical report, National Bureau of Economic ResearchGoogle Scholar
  19. Hollander H (1990) A social exchange approach to voluntary cooperation. Am Econ Rev 80(5):1157–1167Google Scholar
  20. Kant I (1785) Groundwork of the metaphysics of moralsGoogle Scholar
  21. Krosnick JA, Holbrook AL, Lowe L, Visser PS (2006) The origins and consequences of democratic citizens’ policy agendas: a study of popular concern about global warming. Clim Change 77(1–2):7–43CrossRefGoogle Scholar
  22. Kuran T (1991) Now out of never: the element of surprise in the east european revolution of 1989. World Polit 44(10):7–48 ISSN 1086-3338CrossRefGoogle Scholar
  23. Laffont J-J (1975) Macroeconomic constraints, economic efficiency and ethics: an introduction to kantian economics. Economica 42(168):430–437CrossRefGoogle Scholar
  24. Lee TM, Markowitz E (2013) Disparity in the predictors of public climate change awareness and risk perception worldwideGoogle Scholar
  25. Leiserowitz A (2006) Climate change risk perception and policy preferences: the role of affect, imagery, and values. Clim Change 77(1–2):45–72CrossRefGoogle Scholar
  26. Nyborg K (2011) I don’t want to hear about it: rational ignorance among duty-oriented consumers. J Econ Behav Organ 79(3):263–274CrossRefGoogle Scholar
  27. Nyborg K, Rege M (2003) Does public policy crowd out private contributions to public goods. Pub Choice 115(3–4):397–418CrossRefGoogle Scholar
  28. Nyborg K, Howarth RB, Brekke KA (2006) Green consumers and public policy: on socially contingent moral motivation. Resour Energy Econ 28(4):351–366CrossRefGoogle Scholar
  29. Panayotou T (1993) Empirical tests and policy analysis of environmental degradation at different stages of economic development. Technical report, International Labour OrganizationGoogle Scholar
  30. Persson T, Roland G, Tabellini G (2000) Comparative politics and public finance. J Polit Econ 108(6):1121–1161 ISSN 00223808CrossRefGoogle Scholar
  31. Rege M (2004) Social norms and private provision of public goods. J Pub Econ Theory 6(1):65–77 ISSN 1467-9779CrossRefGoogle Scholar
  32. Rentmeester C (2010) A Kantian look at climate change. Essays Philos 11(1):76–86Google Scholar
  33. Roemer JE (2010) Kantian equilibrium. Scand J Econ 112(1):1–24CrossRefGoogle Scholar
  34. Saha S (2007) Democratic institutions and provision of public good. University of California at Santa Barbara, Economics Working Paper Series qt55f3c17g, Department of Economics, UC Santa BarbaraGoogle Scholar
  35. Schumacher I (2009) The dynamics of environmentalism and the environment. Ecol Econ 68(11):2842–2849CrossRefGoogle Scholar
  36. Schumacher I (2014) An empirical study of the determinants of green party voting. Ecol Econ 105:306–318CrossRefGoogle Scholar
  37. Shafik N (1994) Economic development and environmental quality: an econometric analysis. Oxford Economic Papers, pp 757–773Google Scholar
  38. Ulph DT, Daube MPK (2014) Moral behaviour, altruism and environmental policyGoogle Scholar
  39. Wirl F (2011) Global warming with green and brown consumers. Scand J Econ 113(4):866–884CrossRefGoogle Scholar
  40. Zahran S, Brody SD, Grover H, Vedlitz A (2006) Climate change vulnerability and policy support. Soc Nat Resour 19(9):771–789CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media B.V. 2017

Authors and Affiliations

  1. 1.University of Paris 1 Panthéon - SorbonneParisFrance
  2. 2.Santiago CentroChile

Personalised recommendations