A key reason for promoting market-based instruments (MBIs) for environmental policy has been their potential efficiency gains relative to command-and-control and regulatory instruments, whether in terms of resource allocation or implementation and enforcement costs. They epitomise in policy the ‘cult of efficiency’. However, efficiency is but one of several criteria used to assess policy performance. If others are used, such as distributional equity, how do MBIs compare with other instruments, and how do MBIs compare amongst themselves? Does their performance according to each criterion define their limits? This analysis examines these questions; and in particular, how MBIs perform when equity criteria are taken into account.

Market-based instruments (MBIs) are policy instruments that use market forces to achieve policy objectives. In the realm of natural resource and environmental policy, they have recently left the esoteric realm of specialised economics and have come out in the open, out in the public arena of policy debate. This development has been concomitant with the increased importance of environmental policies, notably to manage global warming, pollution, water scarcity and biodiversity issues. Accordingly, the costs of implementing ever more demanding environmental policies are rising, to the point where individuals are beginning to feel the pinch. What MBIs purport to do is to achieve at least cost whatever policy objectives have been chosen. Whence their recent popularity. As such, MBIs are tools to achieve economic efficiency. Indeed, MBIs epitomise the ‘cult of efficiency’.

At the same time, concern is rising regarding the possible ethical implications of relying on MBIs to improve resource allocation. MBIs specifically involve rights over public goods rather than private goods – rights to pollute, rights to develop, rights to conservation, and so forth. Although MBIs operate in much the same way as markets do in general, the fact that they operate on rights over public goods poses issues of equity over access to resources and the flows of benefits they generate. Unfortunately, like any market, MBIs are not designed to achieve any particular form of equitable distribution. They therefore have no reason to do so, and it is unlikely they will by themselves achieve equity by any standards. Their raison d’être is economic efficiency, not distributional equity.

If however, for one reason or another, equity is a concern, then the question arises as to whether such concerns will, or should, impose limits on the use of MBIs. And if so, what is the nature of those limitations. The purpose of this paper is to examine this question. The next section asks “Why equity?”, assesses the importance of equity concerns, and delves into the different concepts of equity: “What equity?”. Much of the difficulty in analysing equity considerations in economics stems from the fact that there is a multiplicity of competing equity principles, just as there are of justice principles. Economists have traditionally shied away from this, arguing that a lack of consensus on fundamental principles prevented economic analysis from achieving scientific status. As we shall see, this view, marshalled by Robbins in 1935, may be based on an outdated notion of scientificity. A third section presents the results of a quick survey carried out for the purpose of this investigation, aimed at nailing down the key problem. Section four then proceeds to present three paradigmatic examples of MBIs which pose inescapable equity issues. These examples are used to introduce key issues of equity when using MBIs. Section five draws some lessons for the future from the examples examined, and highlights some unresolved problems. Section six concludes by providing a tentative answer to the title question.

Why equity? – What equity?

The context in which MBIs are used

Equity over and above efficiency?

Before asking whether equity concerns should impose limits on the use of MBIs, it will be useful to recall the context in which MBIs are used. As mentioned above, the purpose of MBIs is economic efficiency, not equity. However, their efficiency will only be as good as the markets they rely on or that they have created. Markets can fail and lead to inefficient outcomes. Therefore, efficiency may be a sufficient reason to impose limits on the use of MBIs. If an MBI is predicted, in specific circumstances, to reflect poorly functioning markets, then it may be dismissed as a policy tool, at least as a stand-alone tool. It is very easy to design markets that will not work. For example, water trading markets have been shown, in several cases, to be too thin to work properly and achieve any further efficiency in water allocation than more traditional, regulatory tools. Before 2003, private water trading in the Murray Darling Basin of eastern Australia was such an example, described by Challen (2000). In Europe, and particularly in France, as analysed by Godard (2001), the institutional framework within which MBIs would be made to work is inappropriate and does not allow them to yield their potential efficiency gains. Alternative tools, such as direct negotiations for water scarcity allocations, are given an edge (Thoyer, 2001, 2003).

In light of this, the title question really means: Should equity impose limits over and above efficiency? This does not presume of the type of relationship that might exist between the two criteria. They might be in competition, leading to a trade-off, or they might be complementary. At this stage, there seems to be no study clarifying when each might be true and under what conditions.

Equity, efficiency, and other criteria for assessing policy instruments

Although one may question the merits of MBIs relative to other policy instruments in an exclusive “either/or” framework, it must be remembered that, in practice, a package of instruments are usually implemented together, and that each instrument is used for different purposes. For example, MBIs are usually put forward for efficiency gains and for private information revelation. Command-and-control regulations are often defended for their effectiveness and the degree of control and reliability they might provide. While taxation might be used for purposes as different as revenue generation, incentive for new technology adoption, and even equity if done in a progressive way. As a result, performance criteria will vary. The family of criteria that come under the heading of ‘equity’ (that is, distributional equity) form only one amongst others.

An overview is provided in Table 1, where policy instruments have been grouped into three broad categories: CAC (command and control) regulations, incentives and MBIs, and ‘suasion’ instruments, such as education, information and persuasion. Even in terms of general tendencies, different policy instruments will perform differently on different criteria. Regarding MBIs, they tend to be efficient, as that is what they are designed for, though of course they are only as efficient as the markets they are associated with. If well designed, they also tend to be reliable, since, to achieve their ends, they rely on agents’ own motivations, such as profit maximization. In terms of effectiveness and cost-effectiveness, the outcome is uncertain: it will depend on the cost for agents to achieve the policy goals. If these costs are high, then MBIs will be either ineffective, if the regulator decides not to pay the agents, or cost-ineffective, if it decides to pay them, but at a high price. In such cases, the regulator might prefer to tackle the source of the high costs, perhaps by first encouraging new technologies that will bring the costs down.

Table 1: Performance criteria for policy instruments

Legend:  +     Tends to positively contribute to the criterion
             -      Tends to negatively contribute to the criterion
             +/-  Will contribute either positively or negatively, depending on the case
             0     Tends to have no effect on the criterion
             ?      Effect unknown
 

With regard to equity, suasion will usually not change the existing distribution of income or access to resources, although some individuals or groups might have easier access to information than others. However, suasion policies typically try to reach everyone. CAC regulations will tend to yield equitable or inequitable outcomes depending on whether equity is written into the policy or not. As for MBIs, it is unclear at this stage how they will perform in terms of some equity criterion. As mentioned earlier, MBIs are not designed for equitable outcomes. The outcome will mainly depend on whether the relationship between efficiency and some form of equity is substitutable or complementary; that is, whether there is a trade-off or a positive interaction.

Why equity?

If among the several criteria of Table 1 we focus on equity, we may ask: why equity, if with MBIs we are in the business of efficiency? The answer is a pragmatic one. If a policy is seen as unfair, it may simply not be implemented: the constituency will not participate and, unless coercion is used, the policy will fail. So we must ask: what use is a potentially more efficient policy that will not be implemented? Equity is part of the picture, simply because part of, and perhaps a majority of the constituency are concerned with equity outcomes. In this case, there will be a trade-off between efficiency and participation, or between efficiency and probability of acceptance of the policy.
At this stage, then, it might seem we have run into a conundrum. Because MBIs are not designed to achieve equitable outcomes, however defined, they have no reason to achieve them, and so they are unlikely to do so. But more importantly, MBIs will lead to no more equity than markets generally do. As a result, the issue of MBIs and equity is really no different from that of markets and equity in general. Now the literature in this field is impressive: tons of ink have been spilled over this relationship. What is equally impressive is the lack of any clear statement. Why is that so? Why is the literature so inconclusive? This is a serious shortcoming if equity in any form is an inescapable social and political issue.

As most readers know, the answer lies with the multiplicity of equity principles. Equity is an under-defined concept which really refers to a family of concepts or principles. Or to be more precise, the notion of ‘equity’ refers to the distribution of some quantity among a set of individuals or groups, without specifying the precise distribution rule. Different definitions of equity correspond to different distribution rules, which for some reason are all deemed equitable in some precise way. Cazorla and Toman (2000) review, using previous studies, the different distribution rules referred to in the literature, and come up with twelve (12) different notions of equity. These are shown in Table 2.

Table 2: Alternative equity criteria for climate change policy

Source: Cazorla and Toman, 2000.

The reader is referred to the paper by Cazorla and Toman (2000) as well as to Rose and Kverndokk (1999), for a discussion of these twelve equity principles and how they translate into specific distribution (or allocation) rules. The authors also examine the empirical consequences of applying one or the other of these equity principles. Their results are shown in Table 3. What we can learn from their exercise is that some criteria, namely egalitarianism and consensus, tend to produce rather extreme distributions, while other criteria, such as sovereignty, vertical equity and horizontal equity (as well as ability to pay, maximin and Pareto compensation), tend to produce less extreme outcomes. This raises an important point, which is illustrated by the heated debates on how greenhouse gas emission restrictions should be distributed amongst countries: should decision makers focus on the philosophical justification of the principles themselves, or on the pragmatic outcomes they produce, or somehow on both? For example, the two equity principles of sovereignty and vertical equity seem to be worlds apart from the point of view of their philosophical foundations, as per Table 2, but they can produce, as per Table 3, relatively similar outcomes. How does one choose between what may be called “value fundamentalism” and what may be called “outcome pragmatism”?

Table 3: Costs of different allocation rules in 2020, PV in 10⁹ x 1990 dollars

Source: Cazorla & Toman, 2000. (Only part of the original table is shown.)

What equity?

Given the reality of several (apparently) competing equity principles, how does one choose amongst them? This question may be interpreted in two ways, normatively and positively. Normatively, one looks for a meta-principle that would allow us to choose amongst the twelve equity principles. Positively, one observes how different individuals, groups or communities choose between them, and in what circumstances. Let us consider each of the two approaches in turn.

From a normative standpoint, there has been an attempt to find basic meta-principles in the field of justice, a concept that encompasses the notion of equity. Elster (1992, 1993) proposes such a meta-principle, which he calls “ethical individualism”. Ethical individualism purports that justice (and therefore equity) should be attached to individual human beings, reflecting humans as ends in themselves rather than as means to other ends, where for instance the individual is sacrificed to society as a whole or to the Hobbesian State – a principle which should guarantee against totalitarianism. Two direct consequences follow. One, theories of justice (and therefore of equity) should allocate goods among individuals. Two, this allocation should be made on the basis of information about individuals, rather than on aggregates.

On the face of it, Elster’s principle of “ethical individualism” would seem to exclude efficiency – the Pareto criterion in the list of twelve – as a valid equity principle. Efficiency focuses on social welfare as a whole, on the sum total of individual welfare, rather than on every individual’s welfare taken separately. This depends on whether one accepts this meta-principle of Elster or not. Whatever one’s choice, Elster raises the question of the nature of the relationship between the concepts of efficiency and equity. If one accepts Elster’s meta-principle, then efficiency and equity are antagonistic in principle, and their relationship is in terms of a trade-off. If one does not accept Elster’s meta-principle, then one is led to Julian Le Grand’s (1991) position, whereby Pareto efficiency is just a special form of equity, a form that may be called ‘collective equity”, as opposed to “individual equity”. From Le Grand’s standpoint, talking of a trade-off between efficiency and equity is meaningless. It is no different than considering any two equity principles from the list of twelve in Table 2.

This brings us to the crux of the problem. As far as we know, there is no way to rationally prefer one equity principle over another. From a normative point of view, all twelve of them are equally justifiable. Or to be more precise, they are all equally justifiable now or then. By this is meant that depending on circumstances, one may appeal to different equity principles. This may seem as rationalistic anathema, a sign of pure arbitrariness. But the question is, how does the choice of equity principle by specific decision makers vary with specific circumstances? There seems to be only one study that has examined this in some detail. It is the study by Konow in 2001 – ‘Fair and square: the four sides of distributive justice’ – in which he examines the role of context, an issue we shall examine below. This leads us away from the normative approach and into the positive approach; but before doing so, some consequences from the ‘undecidability’ of equity criteria follow that we must point out.

Amartya Sen has defined equity as equality of something. Equality of income, of rights, of opportunity, of reward per unit effort, and so on. His question, “equality of what?”, will suffer from the undecidability problem, as there are many candidates for the equality relationship. This in turn has direct consequences for the choice of a social welfare function, and much of Sen’s work deals in one way or another with this issue. However, he ends up doing no better than anyone else: as long as one considers the problem from a normative standpoint, one is faced with the “now or then equal justifiability” of each of the twelve equity criteria. Multi-criteria analysis tries to escape this trap by pretending to tilt towards the positive approach, since the weightings of the different criteria emerge from a social consensus process: from discussions around the table. However, even this process represents one of the twelve equity criteria, an example of ‘process equity’ (see category C in Table 2).

To conclude, the normative approach has so far led to a logical impasse. It may be that we do not yet know how to think the normative problem in a way that will allow us to escape the merry-go-round of competing equity principles. Or it may be that the problem has not yet been sufficiently studied from an empirical point of view, whether from a social or a psychological perspective. This leads us to the positive approach, where the problem is to understand how different decision makers choose equity criteria to make distributive decisions. In order to illustrate the problem in reference to MBIs, I have carried out a quick survey of several colleague economists on the issue of allocating greenhouse gas emission restrictions across countries. Although the sample was very small, it was sufficient to demonstrate the wide variations in choice of criteria – a rather remarkable result.

A survey on allocating greenhouse gas emission restrictions across countries

After a short introduction to the Kyoto protocol, four questions were put to respondents. Each question was preceded by an example using specific countries, for the sake of concreteness. The first question asked whether emission restrictions should be based on total national emissions, or on per capita emissions. The second question focused on increases relative to some past point of reference (say 1990 emissions), and opposed increases in absolute terms, measured in tonnes of CO2 equivalents, to increases in relative terms, measured in percentage points. The third question was similar to the first, but opposed total national emissions to emissions per dollar of national GDP. The fourth question carried on from the second and asked whether restrictions should be based on total cumulated emissions over the long term, reflective of the nation’s development history, or on current or recent emissions, reflecting the nation’s current development effort. Respondents were asked to choose one alternative in each question and to explain their choice. All the colleagues sampled were, on purpose, natural resource or environmental economists.

The motivations behind this survey design were the following. The choice of the four questions in that particular order was meant to expose the tensions existing between the different questions. At the same time, the formulation of each question was meant to give expression to the tensions existing within each question. These choices were made in order to identify people’s multiple standards of value (if such were the case) and to thereby elucidate the multi-dimensional nature of the problem.

Remarkably, given the very small size of the sample, on all questions the result was never far from a 50-50 split. Even more remarkable were the motivations, or justifications, given for each response. The criteria invoked were not solely, or even mainly, related to a notion of equity. Out of the eight (8) identifiable principles, only three (3) had clear equity connections. In what follows, it is worth referring to the criteria shown in Table 1.

The principles invoked may be grouped into three categories. The first consists of the three principles of effectiveness, pragmatism, and ‘ethical presentism’ (as Elster called it – see below), which have nothing to do with equity. Effectiveness referred to the actual impact of GHG emissions on the planet, leading the respondent to choose total emissions over emissions per capita or per dollar GDP, absolute increases over relative increases, and long term accumulated emissions over recent or current ones. Pragmatism referred to issues of measurability and whether an international consensus could be reached or not. Pragmatism, for example, argues against using long term accumulated emissions on the grounds, firstly, that they will be hard to measure and, secondly, that it will be even more difficult to agree on a common starting date from which to count emissions. The ethical presentism principle considers in this context past emissions to represent a sunk cost to the global community, and therefore, as all sunk costs, that they should not enter the accounting equation.

The second category included the three principles of deterrence, avoidance of moral arbitrariness, and disutility of change. The deterrence principle was invoked to avoid creating perverse incentives. For instance, using the emissions per capita principle was seen to possibly influence a country’s population strategy, while the emission per dollar GDP principle might tempt powerful countries to export their more polluting industries into weaker or poorer nations. Reference to national aggregates was seen as reflecting moral arbitrariness, since national boundaries were the result of the vagaries of history. The fact that some countries were large, populated or richly endowed with natural resources was seen to be a random variable. The implication is that a distribution principle cannot be based on a random variable, something which reflects the deeper principle of accountability, to be examined below. The third principle, put forth by a renowned agricultural economist, was that of disutility of change. The idea here is that the disutility of changing habits, namely reducing, or changing the pattern of, consumption is greater than that of not acquiring new habits, namely those of richer more developed countries. The relationship of these three principles with equity, if it exists, remains unclear.

The third category included the principles of equal opportunity and environmental debt. Equal opportunity referred to developing countries being allowed today the same possibilities of economic development as the more developed countries have had in the past, largely thanks to their unconstrained emissions generated by less efficient technology and accelerated growth. Obviously, the choice influences one’s decision to account for a country’s past emissions or not. The other principle, environmental debt, is also related to past emissions, and considers them to have been generated at a cost to other countries, in the form of direct externalities, and that now is the time to pay up. These two principles are clearly related to notions of equity. Equal opportunity reflects the principle of horizontal equity: treat alike entities in a like manner. Environmental debt reflects both a polluter pays and a compensation principle.

Efficiency was not once invoked as an allocation principle – an interesting fact, given the identity of the respondents. More importantly, it seems that references to different distribution principles reflect different points of view, different angles from which the question is considered. If this is the case, then further empirical investigations using the techniques of experimental psychology should be carried out. Then some correspondence may be discovered between specific points of view, or perceptions, and choice of distribution principles. The equity choice might turn out to correspond to a specific point of view, or aspect of the question. It is like trying to comprehend a three-dimensional object with a two-dimensional viewing technique. These investigations could lead to more efficient negotiations and bargaining strategies on Kyoto type issues.

We shall now focus our attention to how MBIs can generate equity issues, and, given the weakness of current theoretical progress, we shall do so using three typical examples. Each example will generate specific equity issues.

MBIs and equity: three examples

Example 1: Auctioning conservation contracts

In reference to Latacz-Lohmann and van den Hamsvoort’s work (1997, 1998), and to the Bush Tender experience in the state of Victoria (Stoneham et al., 2000, 2002), Hailu and Schilizzi (2003 a,b) have investigated the auctioning of conservation contracts to landholders. In particular, they have examined what happens when auctions are repeatedly held and bidders can learn over time from previous outcomes. The traditional setting is a one-off, static auction, where bidders make decisions based only upon a priori expectations. In this case, the authors examined what happens when auction dynamics are introduced. We refer the reader to the original papers for the auction setup and details.
Results relevant to our concern are given in Figure 1 (a) and (b). The following two diagrams help explain the outcomes of the learning process. The opportunity cost of bidder participation is indicated on the x-axis. Contracting bidders start by bidding (and getting paid) different prices, reflecting their individual opportunity costs. The program payment rates are marked by asterisks in Figures 1(a) and 1(b). In the first period the bid to environmental quality ratios for the 41 winners fall in the range of 0.50 to 0.93 (see Figure 1(a)). But these differences in bid ratios disappear over time. The spread in these prices is reduced as the infra-marginal bids catch up with the marginal winning bid, eventually forming a narrow band of bid ratios ranging from 0.90 to 0.99, as shown in Figure 1(b) (for the 15th period).

Figure 1: Distributions of bid and government rates for periods 1 and 15.

Just above this narrow band of 28 winning bids for that round is another band of ‘active’, but currently not selected, competitors. These two narrow bands represent the two components of the ‘active’ bidders, some of whom replace each other from one period to the next –a kind of “basket of crabs” effect, in part due to the learning algorithm used.

This study shows two related trends with equity and efficiency implications. One is that bidder learning and marking-up (or down) of their bids end up crowding out legitimate competitors, while the other is that government payments increase over and above true opportunity costs. Informational rents increase within the first two rounds and continue to increase slightly for some time. 44% of the bidders never obtain contracts. Of course, most of these, though not all, are those with higher costs. The key point is that the total cost of the 41 lowest cost bidders is within reach of the government’s fixed budget; however, the auction mechanism only allows the hiring of the services of 28 bidders, and these are not necessarily the lowest cost among the 41. This difference of 13 potential winners may be seen as an inequity effect induced by the auction mechanism. 32% of competitors are crowded out by the rent-capturing effect.

Here, inequity is intimately linked with inefficiency. It is because the repeated auction, as modelled in this example, is inefficient – due to information rents being extracted by winning bidders – that inequity arises. The form of equity involved here, in reference to the list of twelve in Table 2, is horizontal equity, that is, like people should be treated in a like manner; in this case, given equal opportunities to compete.
This study thus illustrates two issues regarding equity and efficiency. Firstly, an MBI can easily under-achieve its potential efficiency gains. Secondly, inefficiency can generate inequity. In this case, there are three aspects: rent extraction from the government, crowding out of legitimate competitors, and formation of an exclusive closed-shop effect, whereby only a subgroup of bidders are ever allowed to compete for contracts. The corresponding equity principles are undue rewards to rent seekers, a violation of the principle of accountability to be seen hereafter; horizontal equity, linked to a right to compete for those whom the government’s budget could have hired; and an egalitarian principle, linked to equal opportunity to compete. At the bottom of this process lie the initial differences, the heterogeneity between bidders. The repeated auction mechanism appears to be a process that amplifies such initial differences, however small initially. This in itself raises an equity issue. Should an MBI ‘conserve’ rather than amplify initial inequalities? Indeed, should it reduce them?[1]

Example 2: Farm level impacts of a greenhouse emissions tax policy

Flügge and Schilizzi (2003) examine a greenhouse gas restriction policy and how it could affect certain agricultural regions in Western Australia. A tax on the amount of CO2 equivalents emitted is considered. The policy is assumed to be nationwide, and two different agricultural regions are examined: the Great Southern Region (GSR), which is livestock dominant, and the Eastern Wheatbelt Region (EWR), which is crop dominant. The initial motivation for this distinction was the fact that ruminant livestock (sheep and cattle) contribute significantly to global warming through methane emissions, the global warming potential of which is 21 to 23 times higher than that of CO2. By contrast, dryland crop production contributes very small amounts. The differences in farming systems are due to regional soil and climate discrepancies. An EWR farm is more flexible to CO2 restriction policies than a GSR farm. This difference in flexibility is crucial to the way in which farms can respond to a CO2 restriction policy. Farms are modelled as minimising the cost of the restriction policy by readjusting their farm plan. The reader is referred to the original paper for further details regarding the setup of the study.

Figure 2 shows one of the outcomes of the study. The GSR farm is much more vulnerable, economically, than a EWR farm to a CO2 taxation policy. The former will hit zero profits at a tax rate much lower than will an EWR farm. This means that, should the tax be of the order of $50 a tonne, the GSR farms would go bankrupt while the EWR farms would survive. Lower tax rates would in any case create more difficulties in one region than in the other. This is because one region (the GSR) has fewer options than the other for switching to less CO2-intensive systems. This lack of flexibility is mainly due to agro-climatic differences, something beyond the control of farmers in the region.

Figure 2: Total farm abatement at each tax rate for GSM and EWM farms.
 

Because of this, the government will be faced with the question of helping the disadvantaged region to survive, or to let it go under, subject to a complete rehaul of land use in that area, for example, a conversion to (agro)forestry. Such a conversion is not unthinkable, but it might entail replacing current farmers with another category of landholders, leaving open the question of what would become of the current farmers. Clearly, this raises equity issues.

This example illustrates the equity foundations put forth by Konow in his 2001 paper. Rather than the different notions of equity shown in Table 2, Konow focuses on the foundations of equity as a justice principle. He identifies three: accountability, efficiency, and needs. The accountability principle is related to individual equity; the efficiency principle is related to collective equity; and the needs principle is related to the right to survival. In the example above, we may ask on what basis the government would bail out regions who couldn’t adapt to a greenhouse restriction policy? If, as previously mentioned, one agrees that soil and climate differences are beyond the control of farmers – accounting for the investments they could have made in the past to limit the consequences – then the principle of accountability would demand that they cannot equitably be left to shoulder all the burden of their situation: the government must help, at least to allow for a transition of some sort.

Konow (2001) derives these three foundational principles from experimental studies and empirical observations. He concludes two things. One, that people will use all three of them jointly, but with different weightings. Two, that the accountability principle, at least in societies of western culture, will trump the other two principles, so that the best model to describe people’s use of these principles in the West is a weighted average between the accountability principle and a composite of all three – a distinction he calls ‘specific justice’ and ‘generic justice’. The ‘reported justice’ he observes in surveys is a combination of these two.

The next point is that the specific weighting varies not only across people, but for the same person across circumstances. Here he proposes the fourth ‘corner’ of his justice quadrangle: context, and suggests that the notion of justice, and therefore equity, is context-dependent, but not context-specific. To quote, “The chief reason that justice has remained an elusive concept is because the greatest challenge to formulating and verifying a positive theory of justice is related to issues of context. One aspect of context is the relative importance attached to each of the three justice principles in a particular situation.” This distinction appears to be an important one. Context-specificity implies that there are no general principles. Considering the list of twelve in Table 2, a decision maker would take his pick given his strategic positioning in the decision problem – a situation diametrically opposed to John Rawls’ (1971) ‘original position’, described as deciding behind a ‘veil of ignorance’. This somewhat cynical view is one taken by the economist Peyton Young, when he writes “Equity is a complex idea that resists simple formulations. It is strongly shaped by cultural values, by precedent, and by the specific types of goods and burdens being distributed. To understand what equity means in a given situation we must therefore look at the contextual details” (H. Peyton Young, 1994, p. xii).

Konow opposes context-dependence to context-specificity by suggesting, on the base of empirical analyses, that justice and equity principles do exist, but different contexts lead to different interpretations of these principles. However, Konow does not provide a full account of how different aspects of context determine such interpretations. Arguably, this would require an extensive and laborious program of empirical investigations. At any rate, Konow’s work appears to bring us to the frontier of research in this area. No doubt there is much yet to be done.

Example 3: The initial allocation of tradable permits

In the field of tradable permits, the question of the initial allocation of permits, in particular the choice between grandfathering and auctioning, raises important equity and efficiency issues. These are summarized in Table 4, taken from an unpublished OECD document. The table considers the possible objections to grandfathering and the possible solutions to these objections. A similar table could be made for auctioning. As can be seen, both options have positive and negative implications, and this explains why the issue is as yet unresolved and the subject of ongoing debate. The question we ask here is, what are the criteria and principles involved?

Table 4: Grandfathering of permits: objections and possible solutions

Source: Unpublished OECD document.

One can identify four such principles in Table 4. These are: fairness to newcomers into the industry; the polluter pays principle; competitiveness of domestic industries with other countries if no global agreement exists on the initial allocation; and efficiency arguments. Fairness to newcomers and competitiveness with other countries both instantiate the principle of horizontal equity: treat like entities in a like manner. Both it and the polluter pays principle are founded on the accountability principle. Newcomers cannot control for the fact they are newcomers: their position must be considered as a random variable. By contrast, polluters are accountable for the pollution they generate. The efficiency arguments against grandfathering include monopolistic power, dynamic inefficiency and the loss of innovative drive for new technologies, and static efficiency associated with lower transaction costs.

Over and above criteria for initial allocations, this example highlights the deeper problem of equity targets. There are three possible targets: initial states, intermediate processes, and final states. Initial states include endowments and entitlements. Processes include negotiations, markets, and administrative decision making. Final states include distributional outcomes and efficiency. The obvious question then is, which stage should equity target, and should it target all stages? If so, how?

Robert Nozick (1974) has an indirect answer in terms of a justice chain principle. In the more general terms of justice, he identifies justice in initial appropriation, justice in subsequent transfers, and justice in rectification of any of the above. Nozick then suggests that an allocation will be just if the initial allocation is itself just, all subsequent transfers are just, and any rectification to the initial allocation and to subsequent transfers is also just. Translated in terms of equity, this would require that all initial endowments, entitlements etc. be equitable, that all processes used to decide these endowments and entitlements be themselves equitable, and that the final distributions be also equitable. Whether the same equity principle, or principles, should preside over all these stages is as yet an question open. But certainly everything that has been said until now, in terms of equity criteria and foundational principles, could apply to each of the three targets. As may be painfully obvious, much work is still needed to clarify our thinking on these matters.

Lessons and unresolved problems

Lessons

From the above three examples, what can we learn regarding the use of equity as a criterion for evaluating the performance of MBIs, over and above economic efficiency? Firstly, we saw (example 1) that efficiency and equity criteria can interact and be complementary. In other words, the use of an MBI will be inequitable, in a certain very precise sense, because it is inefficient. The lesson here is, equity and efficiency do not necessarily relate in terms of a trade-off.

Secondly, we saw (our survey) that different equity criteria generate inner tensions between them. Invoking one in one circumstance may conflict with the need to invoke another in another circumstance. This problem has not yet, to our knowledge at least, been studied in the economics literature, though work by psychologists exists. However, this is not solely a psychological problem, but also a cultural and a logical one. The precise conditions underlying the nature of the need for consistency across decisions has not yet been properly researched in this context. There is something there we genuinely do not understand and cannot conceptualise.

Thirdly, it appears, mostly on empirical grounds (see Konow’s work), that different evaluation criteria, including equity and efficiency criteria, need to be combined in some way. This raises the question of which combination principle ought to be used. This leads us to the forefront of research in this field, and I shall therefore only be able to hint at several avenues of research.

The easiest combination principle, and one favoured by economists until now, is a linear combination of criteria, a weighted average. This linear and static approach underlies much of the literature on the choice of social welfare functions. Of course, the question then becomes how the weighting should itself be chosen. From a normative point of view, this regression, potentially ad infinitum, poses the problem of the inescapability of the ultimate criterion, not a very satisfactory prospect. From a logical point of view, we have the choice between two topologies. If we assume a standard topology, each criterion must itself be justified on the basis of a more fundamental criterion, a process which indeed leads to the ultimate criterion problem. But if, following Godard’s (1999) analysis, we assume a Möbius strip topology, as he does in order to relate the anthropocentric and ecocentric views in environmental valuation, then the problem of the ultimate criterion disappears. The Möbius strip topology refers to the twisted slip of paper that has only one side, rather than two, and where any point on the paper can be made to continuously communicate with any other: the slip of paper has no ‘edge’. In the example analysed by Godard, ecocentrism underlies all human values in that without the supporting functions of Mother Nature, no human activity, including valuation, would be possible. At the same time, anthropocentrism underlies our appreciation of nature, no matter how highly or how little we value it.

Multi-criteria analysis purports to circumvent the problem by allowing the weighting to emerge from a social consensus, from discussions around the table. However, this is relying on a specific equity principle, that of process equity, or consensus equity (number 9 in Table 2). Although acceptable, it is not the only one acceptable, as we are now well aware, and so multi-criteria analysis does not really help in this regard.
A second approach to the combination problem is what one may call dynamic transitioning. Rather than considering the different equity criteria all at once, as if in competition with each other, one considers them in succession, in correspondence with the sequence of circumstances. An example of this is Young and Shi’s (2003) dynamic duty of care approach. In a nutshell, they suggest that the use of MBIs, as well as other policy instruments, should be made to work within clear definitions of rights and duties, in particular of environmental duty of care. However, the level of care, they say, should not be fixed, but increase over time. In this way, the compensation problem linked to the payment to landholders for conservation work would be a temporary scheme, rather than a permanent one, and would pay for the costs of transition (see Challen, 2000, on the economics of transition costs following changes in property right structure). The logic of dynamic transitioning with changing performance criteria is yet to be studied.

On the pragmatic side, when several equity criteria, possibly upheld by different competing groups of people, are in the balance, there seems to be an over-arching principle governing the choice process: that to avoid conflict, war, and to seek consensus. This is usually done in a very primitive way, on a trial and error and feel-your-way-through basis, provided that whoever is leading the process has the collective good as a goal. This is far from being guaranteed. The policy maker may represent private interests within the polity. The balancing of private and public interests reflects the problem of balancing individual and collective equity criteria – the major difficulty in balancing out efficiency, a collective equity criterion, and distributive justice, an individual equity criterion.

Unresolved problems

In conclusion, we must humbly acknowledge that we probably still do not know how to think, how to conceptualize the collective versus individual equity problem that underlies much of the tensions between equity criteria. The philosopher Hegel, in his Phenomenology of Mind (1807), and later in his Philosophy of Right (1821), suggested we think dialectically between the individual and the collective. Should we go back and read Hegel for a fresh start? His thinking was very innovative for his time, as it was inherently dynamic, since time was an active factor.
Another unresolved problem, which runs through most of the economic literature, is the scandal of the optional principle of diminishing marginal utility. In my view, its resolution is tied up with the questions we have reviewed in this paper, though a rigorous analysis of this linkage needs to be done. Economists are well aware of this thorn in the discipline’s side, where, for example, DMU is used to justify a positive discount rate for increasing future consumption, while it is ignored in the cost-benefit analysis that incorporates that discount rate – a rather stark example of internal contradiction.

Perhaps to make any progress on the analysis of equity in allocation problems, some serious inter-disciplinary work is necessary. Economists may need to work more closely with behavioural scientists, anthropologists, analytical philosophers, and perhaps mathematicians, to name a few.

Conclusions

A reminder is in order. What use is a potentially more efficient policy instrument that will not be adopted by the collective because of equity issues? A market-based instrument may offer potential efficiency gains compared to a more traditional command-and-control instrument, but this potential may remain unrealized if its implementation raises equity issues. Godard (2001) is one author who has analysed why MBIs are so much more popular in the USA than they are in continental Europe. The differences in social security systems could well follow the same pattern. It may be that Europeans have a different sense of equity than Americans have – they may be referring, because of historical reasons or otherwise, to different equity principles. The existence of institutional structures which allow the introduction of MBIs in the USA more easily than they do in Europe also raises questions as to the nature and pace of institutional change, and how perceptions of equity influence this change. Whatever the case may be, equity considerations directly affect the acceptance of MBIs by the constituency.

So, should equity concerns impose limits on the use of MBIs? The answer is yes, but no differently than efficiency concerns may also impose limits. Also, equity concerns will also impose limits on other policy instruments, just as they will on MBIs. Finally, equity concerns over the use of MBIs are no different than those over markets in general.

What then is the nature of those limits? If we adopt a normative approach, the limits will stem from the decision-maker’s own sense of equity and, more generally, of ethics. From a positivistic point of view, these limits can come in two forms, depending on whether MBIs are implemented in a liberal or in a coercive manner. In a liberal context, such limits will be in the form of the probability of acceptance, or adoption, of the policy instrument. In this case there will be a trade-off between the efficiency gains expected from the MBI and the probability of its acceptance by the community. The policy maker can then compare the ‘expected efficiency gains’ of different policy instruments in terms of the potential efficiency gains multiplied by an estimate of the probability of acceptance. As an example, Challen (2000) examined the thinness of water trading markets in the Murray Darling Basin of eastern Australia as a cause of their inefficiency. Market thinness reflects a lack of adoption by private agents of the tradable permit opportunities. Of course, other criteria than equity will be at work.

In a coercive context, potential efficiency gains from MBIs can be offset by political risks, in terms of lost votes and/or social unrest, which itself generates economic costs. A World Bank document (Huber et al., 1998) reviews the problems that Latin America has been facing when trying to implement MBIs. Equity issues were present in most of the cases. An important difference here is whether one considers political risks to the specific government in place, or to society as a whole. In the first case, we are dealing largely with private interests, in which case, from an ethical point of view, it may be argued that the social benefits, in terms of efficiency gains, are worth more to society than the political risks to a military junta or other form of undemocratic government. But this leads us to the joint problem of market failure and government failure, a different topic altogether.

References

Bouchaud, J-P. and Mézard, M. 2000, ‘Wealth Condensation in a simple model of economy’, arXiv:cond-mat/0002374, UMR 8548, CNRS and Laboratoire de Physique Theorique de l’ENS, Email: boucaud@spec.saclay.cea.fr

Cazorla, M. and Toman, M. 2000, ‘International Equity and Climate Change Policy’, Climate Issues Brief No. 27, Washington DC: Resources For the Future.

Challen, R. 2000, Institutions, Transaction Costs and Environmental Policy: Institutional Reform for Water Resources, Cheltenham, UK: Edward Elgar.

Elster, J. 1992, Local Justice, New York: Russell Sage Foundation.

Elster, J. 1993, ‘Ethical individualism and presentism’, The Monist, vol. 76, pp. 333-48.

Flügge, F. and Schilizzi, S. 2003, ‘Greenhouse Gas Abatement Policies and the Value of Carbon Sinks: Do Grazing and Cropping Systems have Different Destinies?’, Paper presented at the 47th Annual Conference of the Australian Agricultural and Resource Society, Fremantle, Western Australia, 12-14 Feb. 2003.

Godard, O. 2001, Les instruments économiques au service de la lutte contre la pollution atmosphérique - le cas des permis négociables. Laboratoire d’Econométrie, Ecole Polytechnique, Cahier no. 2001-011

Godard, O. 1999, ‘Sur l’éthique, l’environnement et l’économie. La justification en question’, Unpublished : Laboratoire d’Econométrie de l’Ecole Polytechnique : http://ceco.polytechnique.fr/chercheurs/godard/

Hailu, A. and Schilizzi, S. 2003a, ‘Investigating the performance of market-based instruments for resource conservation: the contribution of agent-based modelling’, Paper presented at the 47th Annual Conference of the Australian Agricultural and Resource Society, Fremantle, Western Australia, 12-14 Feb. 2003.

Hailu, A. and Schilizzi, S. 2003b, ‘Learning in a “Basket of Crabs”: An Agent-Based Computational Model of Repeated Conservation Auctions’, Paper presented to the international Workshop on the Economics of Heterogeneous Interactive Agents (WEHIA), Kiel, Germany, 31 May 2003.

Hegel, G. 1807, The Phenomenology of Mind. Translation and introduction by J.B. Baillie, Allen & Unwin: London, 1949.

Hegel, G. 1821, The Philosophy of Right. Translation by T.M. Knox, ‘Hegel’s Philosophy of Right’, Oxford University Press : London, New York, 1967.

Huber, R.M., Ruitenbeek, J. and Serôa Da Motta, R. 1998, ‘Market Based Instruments for Environmental Policymaking in Latin America and The Caribbean: Lessons from Eleven Countries’, World Bank Discussion Paper No. 381, World Bank, Washington D.C.

Kingwell, R.S. and Pannell, D.J. (Eds), 1987, MIDAS, a bioeconomic model of a dryland farm system, Pudoc: Wageningen, NL.

Konow, J. 2001, Fair and square: the four sides of distributive justice, Journal of Economic Behavior & Organization, vol. 46, pp. 137–164.

Latacz-Lohmann, U. and van der Hamsvoort, C. 1997, ‘Auctioning conservation contracts: a theoretical analysis and an application’, Amer. J. Agric. Econ., vol. 79, pp. 407-418.