Hard and easy cases of the dilemma

Modern theories of institutions recognize that ‘the traditional behavioral assumptions have prevented economists from coming to grips with some very fundamental issues and that a modification of these assumptions is essential to progress further in the social sciences’. Douglass North argues in favour of important modifications of the behavioural assumptions. His starting point is the complex causal relationship between the subjective perception of individuals operating in an institutional environment. He is concerned with the existence of institutions over time, i.e., of the workings of formal and informal constraints on behaviour. These constraints define the opportunity set of individuals. They also partly shape the way people choose in the various social contexts that give meaning to the constraints.

The actor's perspective provides one important modification of the economic approach by accepting that different people can have different preferences and motives. In line with institutional approaches, it is then of interest to ask to what extent our three cases of the dilemma may have contextual qualities which facilitate or hinder environmentally friendly attitudes and behaviour. In previous chapters, it was repeatedly shown that the aggregate level of cooperation differs among the environmental cases. In this chapter we take a closer look at the characteristics of the individual response patterns across the three cases.

The question is whether the three Potential Contributor's Dilemmas somehow demand different intensities of cooperative attitude.

The context of environmental dilemmas

The strategy of internalizing environmental responsibility commits the government to promote the voluntary cooperation of citizens in environmental dilemmas. Do our data show that this strategy has worked? To the extent that it has, is it a wise and viable policy for the future? And what are the limitations on what the strategy can accomplish in various cases of the dilemma? These are questions addressed in the final three sections of this chapter. But first we must follow up on the groundwork laid in the four previous chapters of this part. Chapters 9 and 10 presented the data concerning the acceptance of self-regulation policies, as well as evidence on the extent to which the respondents agree or disagree with the environmental ethos, at the level of motives and preferences.

To get to the point where the effectiveness of self-regulation policy can be assessed, we shall first track the choice intentions of cooperatively disposed respondents. Thus, our question is not only whether reported motives and preferences are congruent with the environmental ethos, but also whether respondents who display the right kind of motives actually intend to cooperate rationally and consistently with those motives. It is by focusing on consistent ethical cooperation, from unconditionally cooperative motives, that we finally capture the degree to which respondents are in full agreement with the content of environmental self-regulation in each case of the dilemma.

A model of reasoned choice

The preference orderings of the respondents measured in chapter 4 reflect considerable diversity in the mix of motivations, beliefs, and social norms to which Arrow's shorthand term of ‘individual values’ refers. This diversity needs to be captured in manageable form, in order to explain the variety of preferences in environmental dilemmas.

In this chapter, we represent the individual values of the respondents in the survey by their responses on two general dimensions of motivation. The first dimension measures the evaluation of environmental collective action in the cases of Chemical Waste, Energy Saving, and Holiday Destination. We ask whether the respondent thinks that collective action for reducing environmental pollution is desirable. The responses on this dimension are regarded as motives of ‘Valuation’. The second dimension measures the respondent's own willingness to take part in collective action. Here we ask whether the respondent is prepared to cooperate in the case at hand, at some cost to himself. Responses on this dimension are regarded as motives of ‘Willingness’. Thus the individual values of respondents are described as a point in a two-dimensional space of motives. Each of the twenty-four possible preference orderings can be characterized as the consistent counterpart of a point in the motive space.

The next two sections explain the details of this procedure. But first we clarify the model of practical reasoning which links together the information on motives, preferences, and choice intentions.

The puzzle of unconditional cooperation

Rational cooperation from unconditionally cooperative motives and preferences is a fairly common response of the people whom we have interviewed. While Sen might perhaps not be overly surprised about these results, given his belief in the efficacy of unconditional norms of cooperation, the results are likely to be viewed with some suspicion by many others in the field of rational choice theory. In this chapter, we shall not be restating our position in the debate with those who support the thick-theory of self-interested rationality. Rather, we want to comment on a more sophisticated thesis concerning the role of morality in collective action. This thesis is supported by a lot of empirical work. It says that in so far as morality has real force in overcoming social dilemmas, it will tend to be a morality of conditional reciprocity, rather than a morality of unconditional cooperation. The preference orderings that correspond to norms of reciprocity are ones like the ‘Assurance Game’-ordering QPRS. Such orderings, which are not associated with a dominant strategy, were discussed in the two previous chapters.

As we shall first argue, the empirical plausibility of the reciprocity thesis depends on circumstances in which rational actors are able to monitor each other's behaviour closely, unlike in the large-scale environmental dilemmas we have been studying. Secondly, we defend our view that many of the motives and preferences that we have observed in these dilemmas actually reflect the social force of an unconditional morality.

Introduction

This chapter gives a detailed overview of environmental policy in the Netherlands. The reason for presenting the Dutch case is that it has gone far in attempting to obtain the commitment of individual citizens to an explicit notion of responsibility, in order to maximize compliance to ambitious environmental planning goals. First, from 1989 onwards, successive governments in the Netherlands have been concerned to develop a framework of indicative planning that integrates environmental considerations into the full range of public policies. We shall describe this framework in the next section. A key feature of the National Environmental Policy Plans, as they are called, is the notion of an environmentally self-regulating community. In such a community, the behaviour of corporate and individual actors is subject to state policies of moral persuasion. What is of particular interest here is that Dutch environmental planning utilizes a specific type of policy instruments, the social instruments. Their purpose, to be further discussed in section 2.3, is to induce the widespread voluntary cooperation of citizens and firms with a set of detailed environmental targets, which are specified in the national plans.

In the final two sections of this chapter, we show how these policies of self-regulation, as we shall label them here, are driven by a project of moral reform, as noted by Albert Weale.

This book has grown out of a research project commissioned by the Dutch Ministry of Environmental Affairs in 1994. The brief of the project was to form a survey database of citizens' responses to environmental collective action problems, in order to report on the effectiveness of ‘self-regulation policy’ in the environmental policy plans of the Netherlands. This type of policy, which exists to the present day, seeks to regulate behaviour in areas of everyday life such as waste collection, private car travel, or economizing on household energy, by obtaining voluntary compliance to environmental goals of reducing pollution, rather than securing compliance through legal compulsion or financial incentives.

Self-regulation policies depend on the challenging notion that government can persuade citizens to overcome environmental dilemmas of the kind that the standard logic of collective action for large groups of rational individuals predicts will arise inevitably. Voluntary cooperation to reduce pollution would thus be ruled out on that logic, unless citizens could be made to act irrationally.

To respond to this challenge, our research design aims to combine theoretical insights into motives and preferences underlying rational choice in environmental dilemmas with empirical methods of survey research, in a way appropriate to assessing the chances and limits of self-regulation policies. In the course of writing the book, and in discussion with numerous colleagues, we have been aware that the results generated within our theoretical perspective on rational choice need to be interpreted with some care.

Can something be done?

The concern about environmental pollution in public policy and public opinion in the USA originates, according to former Vice President Al Gore, with the publication of Rachel Carson's Silent Spring (1962). Its publication made everyone aware of the negative effect of pesticides (DDT) on agricultural production. The environmental movement in Europe got off the ground with The Limits to Growth (1972), the report of the Club of Rome. Concern with the natural environment is nothing new. It dates back to seventeenth-century air pollution in London and to Thomas Malthus's warnings in the eighteenth century about the negative effects of population growth. However, there is an important difference between early and modern concerns. In the early days the public had no influence on the decisions of the political elite in handling environmental affairs. Nowadays, what politicians and policymakers propose or decide is closely followed by public opinion.

The publication of Silent Spring created a shock effect in the USA. As a result DDT was banned and laws protecting clean air, land and water were introduced. The notion of limits to growth of the Club of Rome created a political climate that made environmental politics and policy both possible and necessary. Since 1972, many other studies have been published on the ozone layer, global warming and the greenhouse effect, and the irreversible decline of biodiversity. But no report has yet been able to match the impact of Silent Spring or The Limits to Growth.

The subjectivity of the actor in rational choice theory

The Dutch policies of self-regulation discussed in the last chapter have been generally well received by the public, and we shall report our own findings on their acceptance in chapter 9. However, the question we wish to consider now is whether such policies can be expected to work at all, given what social science can tell us about rational behaviour in large groups. For this, we turn to theories of collective action.

The prevention of environmental degradation is a classical example of a public or collective good. If everyone makes a contribution, pollution will decrease considerably. A cleaner environment is a collective good from which everyone benefits. The problem arises from the fact that these benefits are free, i.e., anyone can enjoy the benefits of a cleaner environment whether or not he or she has made a contribution. According to Mancur Olson, environmental behaviour is vulnerable to the logic of free-ridership. Olson's thesis on collective action has achieved the status of a scientific law in the community of environmental researchers. This ‘law’ states that rational individuals seek to maximize their personal welfare, and will not voluntarily contribute to advance their common good, when they are members of a large group. Olson's logic implies that the answer to the question ‘can the policy of self-regulation really work?’ must be: the policy of self-regulation will never work, because no one will make a contribution.

Introduction

The issue of reasoned agreement with a social ethos is an intricate one. In contrast to the acceptance issue, we do not present new survey material here. Instead we shall spend time in reviewing observed attitudes in environmental dilemmas. The results were presented in part II, from the actor's perspective. It was shown that respondents rationally cooperate or defect in the stylized game form of the dilemma, in their role of ‘Individual’ versus ‘The Others’, depending on their preferences over the game's four outcomes. The observed diversity in preferences is linked to the respondents’ diverse stances on the motivational dimensions of Valuation and Willingness, according to criteria of internal consistency. Moreover, diversity of response occurs not only between individual respondents, but between the same respondent's stances in different dilemmas as well. In short, different people respond differently to the general challenge of the dilemma, and the same people may also respond differently, depending on the area of social behaviour in which that general challenge is posed. Chapters 7 and 8 have focused on this last source of diversity.

In what follows below, we try to bring order into the diversity of responses, by interpreting the data in light of the environmental ethos. Taking up the suggestions of chapter 2, we argue in some detail about how the environmental ethos can be decomposed into two cumulative stages, the first stage concerning the ‘internalization of environmental value’, and the second stage concerning the ‘internalization of personal responsibility’.

Ranking preference orderings

In this chapter we want to discuss how our perspective on rational cooperation fits in with the ‘meta-ranking’ framework, which was developed by Amartya Sen in the seventies. The analysis of motives and consistent preferences extends this framework, and applies it to environmental dilemmas. Sen has done much to defend the idea that cooperation in social dilemmas can be rational, not only in the philosopher's sense of being motivated by coherent reasons, but also in the economist's narrower sense adopted here, where choice follows the dominance rule. In pursuing this theme, Sen was concerned to show that morally committed rational behaviour is not easily accommodated by the standard account of utility maximization or preference satisfaction. But he convincingly argued that one can make good analytic sense of moral behaviour within a somewhat more complex structure of decision-making.

The last chapter interpreted the motives and preferences of respondents as reflecting gradations of agreement or disagreement with the environmental ethos, using the definition of consistency between motives and preferences that we developed in chapter 6. This exercise is closely related to Sen's account of moral commitment. The key concept in that account is a ‘meta-ranking’, a rank order of preference orderings which is based on an underlying structure of moral considerations. We introduce the concept in the next two sections, and then go on to compare it to our definition of consistent preferences in the context of the two-person Prisoner's Dilemma, as originally used by Sen.

Introduction to part III

This part of the book deals with the policy lessons suggested by our study of environmental dilemmas. In the last part, we have been led by the question to what extent and in what ways the attitudes and behaviour of our respondents deviate from the self-interested stance that characterizes an actual contributor's dilemma, taking due account of the differences presented by the three cases of environmental collective action. Now we want to focus on what may follow from this in respect of the design of policies that attempt to deal with these and similar problems. While the general direction of these two questions may seem straightforward enough, the last question is set against the particular background of Dutch environmental policy. The results of part II thus need to be reconsidered in the light of self-regulation policies. As we described in chapter 2, such policies seek to convince citizens that their environmental behaviour presents problems of voluntary collective action, and that they have a moral responsibility to cooperate towards the solution of these problems, at least to the extent sufficient for meeting the quantitative targets for reducing various emissions of pollutants. In the national environmental plans, several of these specific targets have been assigned to the group of consumers. A hallmark of the self-regulation approach is to address the consumers in their roles as citizens in order to obtain active compliance with the objectives of the plans.

Conditions of rational choice

Our empirical application of rational choice theory in the Potential Contributor's Dilemma includes the concept of a strong preference ordering and the dominance rule. Following Arrow and Riker, we assume that a rational individual is able to make a complete and transitive ranking of the outcomes. The content of the preference ordering is rooted in individual values, i.e., the actor orders the outcomes by whatever standards he deems relevant. This assumption of the subjectivity of the actor in rational choice theory implies that the content of the preference ordering can only be specified by the actor himself.

Though the content of an ordering has a free format, it must satisfy the conditions of transitivity and completeness. If the preferences satisfy these conditions, we regard any ordering as the materialization of this actor's own interest, as he or she sees it. From this perspective one cannot disqualify an ordering like QSPR as irrational and claim that only the Prisoner's Dilemma ordering PQRS is the hallmark of a rational actor. The content of a preference ordering should not be considered part of the definition of rationality. This is the lesson of the thin-theory of rationality, which refuses to specify any particular goal. ‘Everybody is presumed to be self-interested, choosing what provides the most satisfaction, but the content of the self-interest is not specified’ (Riker, 1990: 173).