User:Camiloprats/sandbox

Environmental Cost-Benefit Analysis
Despite criticism, Cost-benefit Analysis is the most widespread public policy evaluation method. For that reason; Environmental Cost-Benefit Analysis has been recognized as one of the major tools in order to introduce environmental means like climate change or sustainability in policy making. Pearce et al. (2006) point to three core issues when considering Environment issues in Cost-benefit analysis: environmental non-market values; equity and discounting considerations.

Non-market values
“Central to cost-benefit analysis as it applies to environmental issues is the idea of an externality – a third party detrimental (or beneficial) effect for which no price is exacted” (Pearce et al. 2006)). Environmental goods and services have been valued by different methods in order to extract its economic value, where pollution is probably one of the most cited and studied examples of this non-market goods. This new methods have allowed the rise of new global market arrangements like the Kyoto Protocol and the carbon markets.

New methods and tools have been developed in order to capture the full economic value of an environmental asset to account it as a benefit or a cost. Revealed Preference Methods also called market-based methods and stated-preference methods or survey methods are the two main theoretical frameworks to discover this shadow prices. Revealed-preference methods, seeks these non-market values as implicit values that could be extracted from traditional markets. Hedonic Price or Travel Cost methods are probably the most common revealed-preference methods.

Stated-preference methods allow searching the willingness to pay for a determined environmental change by survey techniques. During last decades stated preference methods have strongly influenced environmental policy around the world where contingent valuation and choice modelling are two of the core methodological lines.

While Revealed-preference methods present limitations in order to capture the total economic value when compared with stated preference methods, these methods are supported by traditional markets which make them easy to get legitimized. In the other side, stated preference methods have been more flexible to capture the complexity of the environmental phenomena, but reliability and survey design are some of many relevant technical problems that are needed to overcome.

Equity
Traditional Cost-Benefit Analysis seeks efficiency in the allocation of scarce resources letting aside equity considerations. This allocation is based in a Pareto-optimum approach which sets that these efficiency levels must improve the situation at least of one individual without making any other individual worse off. Cost-Benefit Analysis assumes that in this allocation there will be winners and losers, where winners potentially could compensate losers (Kaldor-Hicks criteria). It assumes pre-eminence of the greatest benefit to the greatest number of people and regards any social preferences as to be addressed by other means (Atkinson and Mourato, 2008: 328). Since 1970s international organizations have supported the inclusion of distributional considerations within Cost-Benefit Analysis (UNIDO, World Bank, OECD).

The widespread of the efficiency rational criteria of Cost-Benefit Analysis has been criticized by different groups within the civil society. Ethical, practical and technical considerations have been exposed regard the need of the inclusion of distributional considerations in Cost-Benefit Analysis. As Pearce et al. (2006) argue environmental considerations have reinforced such claims. As stated above, distribution has been “addressed by other means” like tax systems, but there are two considerations against this separation between efficiency and equity in the application of Cost-Benefit Analysis: first, as Boardman et al. (2006) explain there is a political need of understanding how different projects and policies affect in different degrees different individuals or groups (as for example low and high income groups) and Cost-Benefit Analysis provide useful information about it; and secondly this information allow in some sense the legitimization process of public intervention as pointed by Pearce et al. (2006).

Distributional Cost-Benefit Analysis aim to incorporate equity concerns in the project and policy appraisals based on the assumption of diminishing marginal utility of income. This assumption implies that one extra dollar of benefit or cost, not only decrease with one extra unit, but also has a different weight for low and high income groups (aᵢ≠1). Nevertheless, as Pearce et al. (2006) explain there are multiple levels related with equity and distribution that need to be considered beyond income.

One of the most relevant proposals to deal with different weights is the Kristom’s hierarchies which consider different options related with implicit or explicit considerations in the Cost-Benefit Analysis. A first level of the hierarchy seeks to identify distributional impacts within the project making them explicit and provide detailed information for the decision maker. A second level further, go beyond signalling in a more normative way identifying the “tipping point” where a project fulfils with predetermined distributional conditions. The weight (aᵢ) is the quotient between the average income and individual (household) income (ȳ/yᵢ), for low income groups aᵢ>1 and for high income groups aᵢ<1. Lastly in the hierarchy, explicit distributional weights are identified as a measure of societal preferences (or elasticity “e”: aᵢ = (ȳ/yᵢ)ᵉ ) towards increments in income of individuals (aversion to inequality). This inequality aversion measure is obtained by revealed weights as for example from the tax system or public expenditure or by empirical evidence from past studies.

It had being criticized the distortion of high weighting in major project and policy outcomes and the obviously redistributive inefficiency of any scheme of this type. But this procedure allows generating information respect different alternatives of distributive schemes and its inefficiency level, which may help identify permissible levels. In the case of explicit distributional weights, there is no consensus at all about the magnitude of “e”, and probably as a tacit consensus this magnitude must be calculated in each specific case. Some proposals as for example Pearce (2003) analyse specific magnitudes for the particular case of climate change policy Cost-Benefit Analysis which are focused in a long term horizon rather than most of projects and policies that have short-term expected impacts.

Some other broader critics for distributional proposals (that are not going to be fully covered here) are related with: the complexity of identifying winner and losers; double counting; interest conflicts and groups influences; the fact that politics face multiple objectives at the same time (political gap); differences between policies, projects and programs evaluation framework.

Despite the problems recognized here and also considering some structural complications of rational social (or individual) choice as the Impossibility Theorem of Arrow (1951), this mix between normative and positive approaches is probably our “best game in town” (Pearce, 2000: 74). This claim could be supported considering the fact that most of these critics could be applied to other non-market and market methods (even non-distributional Cost-Benefit Analysis) that further more have not support in a “second best”.

It is relevant to note that the results of these approaches go beyond the pure methodological issue. Distributional Cost-Benefit Analysis findings are helping to understand the societal attitude towards inequality and how to deal in a better way with government failures (reduction of administrative costs). But probably the most valuable consideration is that these improvements are nested efforts that are reinforcing each other rather than isolated views. In that way for example, the use of stated preference methods as proposed by Atkinson et al. (2000) might be understood as an interesting approach to complement this search towards individual and societal attitudes regard distribution. Atkinson’s proposal may be also a useful line to solve the question raised by Pearce et al. (2006) about: what we want to distribute? (pp. 224-225).

These requirements reflect increasing demand of civil society for a more equal distribution among different groups but also new demands about intergenerational equity considerations. As stated by (Pearce et al., 2006: 222-234)”renewed impetus to the debate stems from the complexity of contemporary environmental problems, in particular the distribution of responsibility and suffering across group and generations”.

Discounting
The concept of sustainability has a transversal support among different sectors, but the methods and actions necessaries to achieve this goal are under debate. Uncertainty about the future and its relation with current needs is probably the hardest element to address. If uncertainty is going to be considered as a claim for a strong sustainable approach, probably a precautionary principle would remain at the core of the decision making. On the other hand, we could claim in this debate that “more is better than less”, as without the information provided by environmental market signalling a fair balance between current necessities and future generations could potentially reinforce inequality patterns.

Traditionally the treatment of future generations in Cost-Benefit Analysis had being discussed in the discounting rate and uncertainty arenas. Discounting refers to the process of assigning a lower weight to a unit of benefit or cost in the future than to a unit now (Pearce et al., 2006:184). This definition is based in the assumption of “impatient individuals” where people value money more in the present rather than money in the future so money have a diminishing rate. This fact is based in the perception of risk and uncertainty about the future. Future benefits and costs are computed with a discount rate and then aggregated to obtain the Net Present Value (NPV).

Discounting appears to be inconsistent with the rhetoric and spirit of “sustainable development” – economic and social development paths that treat future generations with far greater sensitivity than has hitherto been the case (Pearce et. al, 2006, pp. 184). But, there is a strong rationale behind this “Tyranny of Discounting” that must be contrasted with the “zero discounting” alternative. A zero discounting implies 0% rate of discount, which mean equal treatment for current and future generations. Nevertheless this seems fair, not discounting implies sacrifices (savings) or in stronger sense “impoverishment” of current generations, as stated by Olson and Baily (1981), for someone further in the future (decades, centuries, millenniums, etc.). This is also supported in the assumption that those future generations will be richer than us, so they could bear in an easiest way costs that in another way must assume the poorest (current) generations.

Behavioural approaches have complemented the positive discussion about discounting. Empirical evidence in recent studies suggests that people discounting decrease along time, evidence that propose declining rates (a hyperbolic) rather than constant rates (exponential) as in the standard Cost-Benefit Analysis. If we accept empirical observations as a source to set normative rules, the fact that this kind of discounting reflects in some way preferences, as Pearce et al. (2006) claim that the evidence is not “overwhelming”, but it could be said that have some “legitimacy”.

Some Time Declining Rates approaches based on the last findings explained above had treated uncertainty in different ways. Weitzman (1998) treat uncertainty as related with future interest rates. This proposal establishes probability-weighted averages for discount factors for different time periods, set by different interest rate scenarios. The average of the different interest rates scenarios, for a fixed given time period, gives a “certainty equivalent discount factor” that diminishes with time.

Gollier (2002) treat uncertainty at the economic level. In simple, traditionally the social discount rate is determined by (Ramsey formula) adding the rate of time preference (reflecting people’s impatience) plus the product between the growth rate of future consumption and elasticity of marginal utility consumption as a measurement of aversion in future income changes (consumption possibilities). Gollier (2002) add a new consideration to the Ramsey formula, that could be explained as a reaction to the economic cycles, where higher is the uncertainty (and so aversion) of the wealth effect in future, people tend to save more in the present (prudence effect).

A third approach about the discounting rate assumes that the “Social Choice” is not a “tyranny” thus there is no difference between current and future generations when we consider who is the decision maker. The relevance of this approaches, goes beyond the particular assumptions of each one but in their possible linkages. For example, Pearce et al. (2006) state about Weitzman and Gollier models that they were “[…] produced separate but related rationales for time-declining discount rates”(pp. 186); also recognize some “parallels” between the results of Weitzman and the social choice models in their tendencies in the long run (lower discount rates).

Since these improvements in Cost-Benefit Analysis intergenerational equity considerations, a severe clash with “time inconsistencies” have been pointed at theoretical and practical levels. The complexity of changing discount rates is discussed by authors like Henderson and Bateman (1995) where the best theoretical model becomes in some way undefined. In a more practical area, as policy decision making, this inconsistency become clearer in the sense that new political regimes, may not act in concordance with previous decisions.

The Stern Review (2006), as one of the most relevant sources of discussion about the discount rate respect to climate change, emphasise the complexity of setting discount rates arguing that “there is no presumption that is constant over time” (pp. 32) and must be equalized regard to the different “paths” and variations of present outcomes as: inequality, consumption patterns and uncertainty. The Stern Review (2006) is probably one of the empirical studies related with discounting and climate change with major impact in the debate about future and current generations. Stern presents a low discount rate of 1.4% composed by a 0.1% time preference, elasticity for marginal utility of 1, and a per capita growth rate of 1.3%. Sterner and Persson (2008) summarizes the implications of these numbers: “The major difference between the discount rate in the Stern Review and most other benefit-cost analysis of climate change is that Stern uses a very low pure rate of time preference. This implies that Stern takes a very egalitarian view of intergenerational distribution”(pp.65).

Critics about the inconsistency between observed interest rates and the discounting rate have been through by Nordhaus (2007) and incompatibilities with the saving patterns have been argued by Dasgupta (2006). However it is possible to argue that the discussion about discounting is not clearly aligned behind a normative-ethical approach or a positive-empirical one, taking the position that better fit or make more sense in each case. In that way, Sterner and Persson (2008) reject the criticism of Nordhaus and Dasgupta respect to Stern Review stating that “We are not simply observing the market as we do in positive or empirical studies; we are providing arguments for public action that involve the provision of very complex public goods” (pp.65).

Lastly, is possible to say that the results of the Stern Review had raised some scepticism (due to its low discount rate) respect the claim of a supposed advantage of future generations. There is some debate respect this point in a Malthusian sense, where limits of growth could impose higher costs on future generations. But as Sterner and Persson (2008) expose these considerations are “discredited” (pp.67) by technological change but its logic has to be kept in mind because the logic of perfect substitutability may not be certain at all. This doubt is supported by the Stern Review (2006) where was found for example that there is not a direct economic substitutability between the global agricultural output (GDP) and the estimated total economic loss of this activity. As proposed by Sterner and Persson (2008) seems logic to have some sensitivity regards to the changes in relative prices that could even generates an abatement ”on the same order of magnitude as changing the discount rate”(pp.70).

It is necessary to evaluate in which extent these efforts are concordant with the current economic projections and the capability of implementation according to these trends. In the case of climate change we depend on a global deal where the big problem is the free rider figure in a world which current growth is led by developing countries where is difficult to impose heavy restrictions in their current trends. Despite the major impact of a discount rate policy, this consideration might be a crucial one in the threshold of this second industrial (and urban) revolution of China and India.

Legitimization and participation
The trade-off between efficiency and equity is not conclusive despite recent Cost-Benefit Analysis improvements. Nevertheless, Cost-Benefit Analysis is contributing developing new ways to understand projects distributive impacts and human behaviour about inequality. Despite these findings, distributional and further more intergenerational considerations make the Cost-Benefit Analysis results more complex to understand and legitimize. Likewise, as more in depth social costs are considered, more fragile become some traditional economic assumptions. For example, as pointed by Sterner and Persson (2008) if we are going to consider the “social disruptions” of climate change as migrations, wars, refugees, civil conflicts “it seems reasonable to assume very low “substitutability” with material consumption” (pp.72).

Cost-Benefit Analysis applications are a mix of normative and positive deliberations, and probably most of the critics rely in the absence of participation in this conceptual definition. As Pearce (2006) states, what needs to be distributed is also pending. This paternalist attitude rest legitimacy to the process. Moreover, this positive-normative mix may be less defensible if a specialist view does not clarify its criteria when makes the differentiation between individual and societal choice, but further more between consumer and citizen.

In the past century, most of the discussion about the traditional development models faced the problem of dealing with the trade-off between equity and efficiency. This debate was based on the belief that some basic goals related with human life conservation (as for example child mortality reduction) were largely achieved or with well-known solutions. The claim for sustainable development, popularized by the Brundtland Report (1987), revealed a step back in the development goals. The threat of climate change raised again the shadow over some basic goals related with human life conservation and social disruptions. Facing the second decade of the new century, inequalities among current and future generations seem unsolved by the Kuznetsian's predictions.

These considerations make important to review how Cost-Benefit Analysis is treating these structural development deficits, as the outcomes of development models are in theory supported by their methodologies. In that sense Cost-Benefit Analysis is widely spread as one of the most relevant tools in the decision making, so how this method perform may have an extended influence in the development achievements.