This action might not be possible to undo. Are you sure you want to continue?
Alan Ingham, Jie Ma and Alistair M. Ulph June 2005
Tyndall Centre for Climate Change Research
Working Paper 74
How do the costs of adaptation affect optimal mitigation when there is uncertainty, irreversibility and learning?
Alan Ingham1 , Jie Ma2, Alistair M. Ulph3
Economics Division, School of Social Sciences, University of Southampton, Highfield, Southampton SO17 1BJ, UK. E-mail: email@example.com. Corresponding author 2 Economics Division, School of Social Sciences, University of Southampton, Highfield, Southampton SO17 1BJ, UK. E-mail: firstname.lastname@example.org 3 Faculty of Humanities, University of Manchester, Oxford Road, Manchester M13 9PL, UK. E-mail: email@example.com. Tyndall Centre Working Paper No. 74 June 2005
Please note that Tyndall working papers are "work in progress". Whilst they are commented on by Tyndall researchers, they have not been subject to a full peer review. The accuracy of this work and the conclusions reached are the responsibility of the author(s) alone and not the Tyndall Centre.
* We would like to thank the Tyndall Centre for Climate Change Research for providing financial support.
Summary In this paper we survey the literature on the economic analysis of mitigation and adaptation in the context of climate change. We present an overview of the main elements of what formal economic analysis there has been of adaptation and mitigation. This uses the familiar economic framework of expected utility maximization to capture behaviour. Mitigation and adaptation are two different ways which societies can reduce the damages that might be caused by climate change relationship which shows how mitigation and adaptation reduce the damage costs caused by climate change, and known costs of adaptation and of mitigation. A single social planner can choose the optimal mix of adaptation and mitigation to minimize total social costs in general the optimal combination of mitigation and adaptation will require the use of both strategies. This just reflects the rather mild assumption that the first bit of mitigation or adaptation is cheap relative to the marginal reduction in damage costs they bring about. This confirms the view of Kane and Yohe (2000) and Parry et al. (2000) that we need to have an integrated approach to adaptation and mitigation, and we cannot rely on either mitigation alone or adaptation alone to deal with climate change. It is sometime claimed that this joint determination of mitigation and adaptation means that adaptation and mitigation are complementary Thus, an increase in damage costs would be expected to lead to an increase in both adaptation and mitigation1. But in the context we have sketched so far, in technical economic terms, adaptation and mitigation are substitutes (Kane and Yohe (2000)), in the sense that if, say, the cost of adaptation fell, the optimal response would be to do more adaptation but less mitigation2. This just reflects the fact that these are two different ways of reducing damage costs, so if one becomes more expensive we should make more use of the other. This has obviously important policy implications. A central focus is the extent to which adaptation and mitigation are substitutes, in the sense that if, say, the cost of adaptation fell, the optimal response would be to do more adaptation but less mitigation. We introduce uncertainty and learning, and argue that a crucial determinant of the costs of adaptation is the rate at which those who respond to climate change, such as individual households, farmers, firms and so on, can learn about a changing environment. Now the above argument is based on a simple partial economic analysis, and it is conceivable in principle that with general equilibrium effects adaptation and mitigation might become complements. For example, if increased mitigation caused energy prices to fall this could make it more attractive to use some forms of
This is like an income effect with normal goods. In this context we are essentially treating mitigation as a single activity and adaptation as a single activity. Of course there are many different methods of mitigation and adaptation, and we do not claim that every form of mitigation is a substitute for every form of adaptation (or even that every form of mitigation is a substitute for every other form of mitigation, and similarly for adaptation). This claim only applies in an aggregate sense.
adaptation. Nevertheless it is sometimes believed that there may again be a kind of complementarity between mitigation and adaptation in that mitigation `buys time' for adaptation. It is not clear what underlies this belief. In part it might reflect issues to do with the ability to learn IPCC (2001) notes there are still significant uncertainties attached to climate change. This raises the important question of how such uncertainty, and the prospect of getting better information in the future, affecting optimal policy towards mitigation and adaptation. Introducing uncertainty would lead to an increase in current mitigation relative to a model in which all parameters took their certainty equivalent values, and empirical models showed this effect to be significant. However introducing the possibility of learning, together with the irreversibility of emissions, has ambiguous effects on the optimal current policy towards mitigation, although in empirical applications this effect is small. In this section we begin by asking how these results are affected when we allow for both mitigation and adaptation. It is straightforward to show that if one introduces an exogenous risk of climate change damages into the simple models of the previous sections, then the main results go through in a straightforward way: an increase in the risk of climate change will cause the optimal levels of mitigation and adaptation to rise, but adaptation and mitigation remain substitutes. Kane and Shogren (2000) obtain slightly different results by using a static model in which adaptation and mitigation play asymmetric roles: the level of damage costs can be reduced by adaptation, but mitigation reduces the risk of climate change, so risk is now endogenous. They show that an exogenous increase in risk has an ambiguous effect - it always leads to an increase in adaptation but the effect on mitigation depends on whether an exogenous increase in risk causes an increase or decrease in the marginal effectiveness of mitigation in reducing risk, and they give examples of how this effect could go either way. However, we show in IMU that it remains the case that adaptation and mitigation are substitutes, in the formal economic sense in which we have been using this term. This is confirmed by some empirical examples. The above analysis is static and so does not allow for important dynamic aspects of the problem: the possibility that over time one can acquire better scientific information which may reduce current uncertainties and the fact that the atmospheric concentration of greenhouse gases is effectively irreversible. This raises an important timing question: should we moderate the current level of action to deal with climate change while we wait to get better scientific information, or should we increase the level of current action in case we learn that climate change is a much more serious problem that we currently expect but then find that because of irreversibility the costs of taking effective action are prohibitive. When include both adaptation and mitigation, although only for the simple case of quadratic cost functions3. We show that the same ambiguity that arose when we only have mitigation applies also when we have adaptation as well: current actions (mitigation and adaptation) with learning may be higher or lower than current actions with no learning depending on the precise parameters of the model.
In Ulph and Ulph (1997) it was shown that even for the simple ``textbook''\ case of linear marginal damage and abatement costs the ``irreversibility effect''\ remains ambiguous.
A contribution of the paper is to provide a single coherent analytical framework in which to demonstrate the formal economic analysis of adaptation and mitigation, in which the information available, preferences and objectives, and economic constraints are explicitly set out so that optimizing behaviour can be modelled. We also extend the existing literature in the context of climate change, which has focussed just on issues of mitigation, to include both mitigation and adaptation. 1 Introduction. In this paper we present a survey of the literature on the economic analysis of mitigation and adaptation in the context of climate change. Since the focus is on economic analysis we shall concentrate on papers written by economists, but we shall also use the framework to comment on more general discussion of issues to do with mitigation and adaptation. In the first two sections we present an overview of the main elements of what formal economic analysis there has been of adaptation and mitigation, beginning in section 2 with the analysis ignoring uncertainty and learning. A central focus of this section is the extent to which adaptation and mitigation are complements or substitutes. In section 3 we introduce uncertainty and learning, and argue that a crucial determinant of the costs of adaptation is the rate at which agents can learn about a changing environment. Since this paper is designed to be read by non-economists, we provide in a companion paper (Ingham, Ma and Ulph (2004) - henceforth referred to as IMU) some formal modelling to support the claims made in section 3. We make no claim for much originality in the analysis, but we hope that one contribution of the paper is to provide a single coherent analytical framework in which to demonstrate the formal economic analysis of adaptation and mitigation, albeit using the simplest possible models to allow us to derive the results as cleanly as possible. The one area where the analysis goes beyond what we have seen in the literature is our analysis of uncertainty, learning and irreversibility where we extend the existing literature in the context of climate change, which has focussed just on issues of mitigation, to include both mitigation and adaptation. In the rest of the paper we comment on some of the wider discussion of adaptation which does not draw directly on formal economic analysis. The analysis of uncertainty and learning in section 2 uses the familiar economic framework of expected utility maximization to capture behaviour under uncertainty and in section 4 we discuss criticisms of that approach and survey alternative approaches, such as tolerable windows. In section 5 we consider the issue of what constitutes adaptive capacity and how might we measure it.
2 Optimal Mix of Mitigation and Adaptation 2.1 Introduction to Mitigation and Adaptation In this Section we begin by exploring some basic issues concerning the relationship between mitigation and adaptation. In the research on climate change and climate variability, mitigation refers to the limitation of greenhouse gas emissions in order to prevent the future climate impacts on society; adaptation refers to adjustments in individual, group, and institutional behavior in order to reduce society's vulnerabilities to climate (Pielke (1998) and Smit et al. (2000)). In principle, mitigation and adaptation are two different ways which societies can reduce the damages that might be caused by climate change. However, Klein et al. (2003) identify that there are three differences between mitigation and adaptation4. The first is related to the temporal and spatial scales on which they are effective. The benefits of mitigation activities carried out today will be evidenced in several decades because of the long residence time of greenhouse gases in the atmosphere, whereas many effects of adaptation measures should be apparent immediately or in the near future. In addition, mitigation has global benefits, whilst adaptation typically takes place on the scale of an impacted system, which is regional at best, but mostly local. The second is the extent to which their costs and, in particular, their benefits can be determined, compared and aggregated. Irrespective of the diversity of mitigation options, they all serve to reduce greenhouse gas emissions and in view of its global benefits it is irrelevant where in the world the mitigation takes place. Expressed as CO2-equivalents, the emission reduction achieved can be compared with that of other mitigation options and if the implementation costs are known, the cost-effectiveness of these options can be determined and compared. The benefits of adaptation are more difficult to express in a single metric, impeding comparisons between adaptation options. In principle, the benefits of adaptation are the climate-related damage costs one avoids by taking adaptation measures (assuming that climate change would have adverse consequences). Thus, if one quantifies the potential impacts of climate change on a system assuming no adaptation, as well as its residual impacts assuming adaptation, the benefits of adaptation are given by the difference between the two. From the value thus obtained one can subtract the costs of implementing the adaptation options to arrive at the net benefits of adaptation. However, the practice of assessing and comparing adaptation benefits is fraught with difficulties related to the uncertainty about and differences between the impacts avoided. The third difference between mitigation and adaptation concerns the actors and types of policies involved in their implementation. Mitigation primarily involves the energy and transportation sectors in industrialized countries, and to an increasing extent the energy and forestry sectors in developing countries. In addition, the agricultural sector plays a role in mitigation. Compared to
Also see the references cited by Klein et al. (2003)
adaptation, the number of sectoral actors involved in mitigation is limited. Moreover, they are generally well organized, linked closely to national planning and policy making, and used to taking medium to long-term investment decisions. Over the past decade, incentives and opportunities created by national and international climate policy have increasingly stimulated mitigation activities by the energy and forestry sectors. In contrast, the actors involved in adaptation represent a large variety of sectoral interests, including agriculture, tourism and recreation, human health, water supply, coastal management, urban planning and nature conservation. Whilst these sectors have in common that they are potentially impacted by climate change, decisions as to whether or not to adapt are taken at different levels, ranging from individual farmers to national planning agencies. For these actors, climate change is typically not of immediate concern. Moreover, in spite of the potential magnitude of climate change they often have little incentive to incorporate adaptation into decisionmaking, either because policy and market failures do not encourage medium to long-term planning, because responsibilities for action are unclear or because adaptation is concerned with collective goods such as safety, human health and ecosystem integrity. In the next sub-sections we explore the implications of these differences for the optimal mix of adaptation and mitigation strategies. 2.2 Some simple cases To analyze the links between adaptation and mitigation we begin with the simplest case where we ignore dynamic aspects and uncertainties, so there is a known relationship which shows how mitigation and adaptation reduce the damage costs caused by climate change, and known costs of adaptation and of mitigation. A single social planner can choose the optimal mix of adaptation and mitigation to minimize total social costs5. It is straightforward to show (see section 1 of IMU) that in general the optimal combination of mitigation and adaptation will require the use of both strategies. This just reflects the rather mild assumption that the first bit of mitigation or adaptation is cheap relative to the marginal reduction in damage costs they bring about. This confirms the view of Kane and Yohe (2000) and Parry et al. (2000) that we need to have an integrated approach to adaptation and mitigation, and we cannot rely on either mitigation alone or adaptation alone to deal with climate change.
Klein et al. (2003) cast doubt on whether such an exercise in optimization is worthwhile. They point out that striking the appropriate balance between mitigation and adaptation will be a tedious process and the optimal mix of response options will vary by country and over time, as local conditions and costs change. Striking the balance will be particularly challenging because of some unique characteristics of the problem; long time horizons; non-linear and irreversible effects; the global nature of the problem; social, economic, and geographic differences amongst affected parties; and the fact that institutions needed to address the issue have only partially been formed. Given these characteristics, as well as the widely differing interests, values and preferences within and between societies, there is no single optimal mix of mitigation and adaptation. In addition, uncertainty about climate and socioeconomic change strongly affects the outcome of any optimization exercise. As soon as new information becomes available, the optimal mix will be different. While we agree that the optimal mix of adaptation and mitigation depends on all the factors identified above, it is precisely because of the complexity of the issues that we believe careful modelling of these choices is important
It is sometime claimed that this joint determination of mitigation and adaptation means that adaptation and mitigation are complementary (IPCC (1996), Pielke (1998)). Thus, an increase in damage costs would be expected to lead to an increase in both adaptation and mitigation6. But in the context we have sketched so far, in technical economic terms, adaptation and mitigation are substitutes (Kane and Yohe (2000)), in the sense that if, say, the cost of adaptation fell, the optimal response would be to do more adaptation but less mitigation7. This just reflects the fact that these are two different ways of reducing damage costs, so if one becomes more expensive we should make more use of the other. This has obviously important policy implications. For example, early generations of models of the economic response to climate change tended to focus only on mitigation strategies, so can be considered as treating the costs of adaptation as infinite. Later models have given more attention to adaptation options and found them to be relatively cheap and so have concluded that there is rather less need for mitigation (see for example recent surveys by Mendelsohn (2003) and Tol (2003)). Now the above argument is based on a simple partial economic analysis, and it is conceivable in principle that with general equilibrium effects adaptation and mitigation might become complements. For example, if increased mitigation caused energy prices to fall this could make it more attractive to use some forms of adaptation, but such general equilibrium effects would need to be quite powerful to outweigh the direct partial effect, and we are not aware of evidence which shows such results. There is another sense in which it is sometimes thought there may be a complementary relationship between adaptation and mitigation. Suppose an increase in the optimal use of air conditioning to adapt to climate change caused such an increase in power generation and associated emissions of greenhouse gases that it was necessary to increase mitigation, then adaptation and mitigation might be considered complements. However, as we argue in IMU, this could not happen if adaptation and mitigation are being optimally regulated. This argument is closely related to the third point made by Klein et al. (2003) quoted above - that adaptation and mitigation often involve different actors. Mitigation decisions inevitably require some form of government intervention because of the public good nature of climate change damages, whereas it is sometimes thought that adaptation decisions can be made more locally, even individually8. However the question of whether decisions can be devolved to individuals is just as much an issue for adaptation as for mitigation. Thus getting energy prices right, i.e., to reflect the social costs of energy use, will be just as important for adaptation choices that use energy (e.g. the use of air conditioning) as for mitigation decisions, such as switching to low petrol consuming cars. The example above is sometimes referred to as a case of “maladaptation” (Scheraga and Grambsch (1998)) whereby private decisions to adapt may not be optimal and require offsetting
This is like an income effect with normal goods. In this context we are essentially treating mitigation as a single activity and adaptation as a single activity. Of course there are many different methods of mitigation and adaptation, and we do not claim that every form of mitigation is a substitute for every form of adaptation (or even that every form of mitigation is a substitute for every other form of mitigation, and similarly for adaptation). This claim only applies in an aggregate sense. 8 It is sometimes argued that mitigation is more of a public good while adaptation is more of a private good (see for example Mendelson (2000), Hanemann (2000), Kane and Shogren (2000))
mitigation. But it is rather an example of mal-regulation; provided energy use is optimally regulated such an outcome cannot arise and mitigation and adaptation are substitutes9. There is a related issue arising from the first point by Klein et al. (2003) concerning the different spatial aspects of mitigation and adaptation, with mitigation dealing with a global public good, while adaptation often provides purely local benefits. As we show in IMU this has the obvious implication that if we move away from the assumption of a single social planner (who would have to be a global government of some type) and recognize that individual nation states may set their adaptation and mitigation policies independently then the noncooperative outcome will involve each state setting too little mitigation and too much adaptation, relative to the case of a single global social planner. This again reflects the fact that adaptation and mitigation are formally substitutes for each other. As we note in section 2.1, these operate at very different levels, with mitigation operating at national/international level because of the public good/externality failure involved. Adaptation is much more local, and for the most part requires little government intervention. Klein et al. draw the conclusion that the two policies should be kept separate. But we think this actually means something more subtle. The key point is that it is difficult to construct sensible empirical policy models that operate correctly at the right level of scale for the two types of response. Therefore in building models of mitigation, which need to operate at appropriate national/international level, one will have to make some kind of assumption about adaptation (to get the optimal level of mitigation), but one should not then assume that this tells us much about actual policy on adaptation. The final extension we make in this section is to introduce dynamic considerations. As Klein et al. (2003) note mitigation and adaptation differ in their temporal aspects in that the effects of mitigation in one period will produce benefits for all future periods while adaptation is often thought to produce benefits only for the period in which adaptation takes place. However while this may be true, a forward looking social planner will want to choose time paths for both mitigation and adaptation which are optimal, and in a broad sense all the results for the static model carry over to a dynamic model, including the notion that mitigation and adaptation are substitutes10. Nevertheless it is sometimes believed (see for example Parry et al. (2000)) that there may again be a kind of complementarity between mitigation and adaptation in that mitigation `buys time' for adaptation. It is not clear what underlies this belief. In part it might reflect issues to do with the ability to learn, but that cannot be captured in the analysis so far which assumes certainty. Partly this could just reflect the fact that it takes time to accumulate a stock of adaptive capital -- if one has many decades available then the costs of moving a coastal city inland will be relatively small
Scheraga and Grambsch (1998) note that many existing policies may be sub-optimal, but do not link this to the issue of “maladaptation”. 10 Of course with many time periods we can formally think of mitigation and adaptation at different dates as different types of mitigation and adaptation, and as noted in footnote 4, with many types of mitigation and adaptation we cannot claim that every form of mitigation is a substitute for every form of adaptation.
compared to the costs of doing this within a few decades, essentially because the former can rely on the natural process of replacing obsolescent capital to achieve adaptation, whereas in the latter one has to retire capital early. The increased difficulty of adaptation the faster is the rate of change in climate can also reflect problems for non-human agents whereby over time species might be able to migrate or mutate to adapt to climate change, but if it is too rapid they could be wiped out, or conversely, slowing the rate of climate change through mitigation may allow crops to adapt to become drought-resistant. However, what we are concerned with here is how the rate of change of climate might affect human adaptation. Whatever the underlying explanation, a crude way of capturing this in our analysis might be to assume that in later periods the costs of adaptation also depend on the level of mitigation in earlier periods, with more mitigation in earlier periods (and hence slower rate of change of climate in later periods) reducing the cost of adaptation in later periods. It can be shown that now mitigation and adaptation could indeed be complements, since lowering costs of mitigation increases mitigation, which slows climate change and makes adaptation more effective. 3 Introducing Uncertainty and Learning The previous discussions have ignored the important issue of uncertainty about future damage costs from climate change. But as IPCC (2001) notes there are still significant uncertainties attached to climate change (see also Roughgarden and Schneider (1999), Goodess, Hulme and Osborne (2001), Brooks and Adger (2003), Dessai et al. (2003) for attempts to quantify the different dimensions of climate change risks). This raises the important question of how such uncertainty, and the prospect of getting better information in the future, affecting optimal policy towards mitigation and adaptation. In an earlier paper (Ingham and Ulph (2004)) we have reviewed the literature on uncertainty and learning in the context of climate change, but we focussed on the implications only for mitigation strategies. We showed that introducing uncertainty would lead to an increase in current mitigation relative to a model in which all parameters took their certainty equivalent values, and empirical models showed this effect to be significant11. However introducing the possibility of learning, together with the irreversibility of emissions, has ambiguous effects on the optimal current policy towards mitigation, although in empirical applications this effect is small. In this section we begin by asking how these results are affected when we allow for both mitigation and adaptation, on which there has not been much formal analysis. It is straightforward to show that if one introduces an exogenous risk of climate change damages into the simple models of the previous sections, then the main results go through in a straightforward way: an increase in the risk of climate change will cause the optimal levels of mitigation and adaptation to rise, but adaptation and mitigation remain substitutes. Kane and Shogren (2000) obtain slightly different results by using a static model in which adaptation and mitigation play asymmetric roles: the level of damage costs can be reduced by adaptation, but mitigation reduces the risk of climate change, so risk is now endogenous. They show that an exogenous increase in risk has an ambiguous effect - it always leads to an increase in adaptation but the effect on mitigation depends on whether an exogenous increase in risk causes
In terms of the social cost of carbon, the marginal value of a reduction in current emissions of carbon, the introduction of uncertainty causes the social cost of carbon to rise by a factor of between 2 and 3.
an increase or decrease in the marginal effectiveness of mitigation in reducing risk, and they give examples of how this effect could go either way. However, we show in IMU that it remains the case that adaptation and mitigation are substitutes, in the formal economic sense in which we have been using this term. This is confirmed by some empirical examples Kane and Shogren provide from the agriculture sector of US. They find that mitigation and adaptation are clear substitutes. However, there is the possibility of maladaptation. For example, when people turn to energy intensive practises to overcome a significant loss in agricultural productivity associated with climate change, more agrochemical use and irrigation would increase the demand for fossil-based energy, which will then add carbon dioxide into the atmosphere, and thus would reduce the effectiveness of adaptation. The strength of this negative effect will depend on the severity of the loss of agricultural productivity expected from such a change in climate, inducing adaptation investments. The above analysis is static and so does not allow for important dynamic aspects of the problem: the possibility that over time one can acquire better scientific information which may reduce current uncertainties and the fact that the atmospheric concentration of greenhouse gases is effectively irreversible. This raises an important timing question: should we moderate the current level of action to deal with climate change while we wait to get better scientific information, or should we increase the level of current action in case we learn that climate change is a much more serious problem that we currently expect but then find that because of irreversibility the costs of taking effective action are prohibitive. This question, which is closely linked to the `precautionary principle' has been much studied both theoretically and empirically, and, as already noted, in our earlier survey of the literature (Ingham and Ulph (2004)) we found that the answer to this timing question was theoretically ambiguous (in the sense that in a general model the answer can go either way) but empirically small. However almost all the analysis focuses on the case where policy-makers have a single action to deal with climate change - mitigation. An important question is how far the results carry over to the case where the policy maker has two sets of responses - mitigation and adaptation. Might it be the case that while the result for mitigation is ambiguous, the result for adaptation is clear-cut? Or might it be the case that one uses the presence of two instruments to deal with climate change to resolve the ambiguity by, say, doing more mitigation and less adaptation when faced with irreversibility and learning, but with the net effect being ambiguous? To the best of our knowledge there is no theoretical work which addresses the timing question when the policy maker can use both mitigation and adaptation. In IMU we extend the model of Ulph and Ulph (1997) to include both adaptation and mitigation, although only for the simple case of quadratic cost functions12. We show that the same ambiguity that arose when we only have mitigation applies also when we have adaptation as well: current actions (mitigation and adaptation) with learning may be higher or lower than current actions with no learning depending on the precise parameters of the model. The above analysis of learning makes the rather stark contrast between a scenario in which no learning takes place, and one in which learning occurs, which is modelled as the extreme case of exogenous perfect learning taking place at a given moment of time. In reality the process by which agents collect information on which to update
In Ulph and Ulph (1997) it was shown that even for the simple ``textbook''\ case of linear marginal damage and abatement costs the ``irreversibility effect''\ remains ambiguous.
their expectations about the future risk of climate change, and hence to adapt their behavior, is a much more complex process than this simple model suggests, and has been at the heart of a lot of debate about the impact of adaptation on mitigation. As noted in section 2.2, if, as our analysis suggests, adaptation and mitigation are substitutes, then the inclusion of adaptation in models of climate change which previously ignored adaptation is likely to lead to a significant reduction in the optimal level of mitigation (see for example Mendelson (2000, 2003)). However this conclusion has come in for a lot of criticism. There are some technical criticisms of the methodology (mainly the use of cross-section analysis) used to assess how agents respond to different climatic environments (see e.g. Hanemann (2000)). But the main criticism is that the modelling ignores the process of learning. The comparison between models which exclude or include adaptation is sometimes expressed as the difference between the ‘dumb farmer’ and the ‘clairvoyant farmer’ hypotheses (see e.g. Hanemann (2000)). Assuming that agents do not respond to climate change - the ‘dumb farmer’ - means that the burden of dealing with climate change falls on mitigation. However much of the early modelling of adaptation assumed that agents could immediately adjust to changes in climate - the `clairvoyant farmer'. This implicitly assumed that costs of adaptation are rather low, and so leads to much less reliance on mitigation. The reality is likely to be somewhere between these two extremes, and a crucial aspect of this is how quickly agents learn from information, such as say weather patterns, and adjust their behavior. A number of studies have noted this point (e.g. Yohe (1996), Kane and Yohe (2000), Lempert et al. (2000), and Reilly and Schimmelpfennig (2000)) have sought to characterize different features of society which might affect society's ability to adapt, but rather fewer studies have explicitly modelled the learning process and its impact on adaptation. Schneider et al. (2000) consider 3 stylized models of adaptation - no adaptation (dumb farmer), perfect adaptation (clairvoyant farmer) and an intermediate case in which agents adapt with a 20 year lag relative to perfect adaptation. They show that for small changes in climate (one-third increase in greenhouse gas concentrations), adapting with a lag is close to no adaptation, while with a large change in climate (doubling concentrations) adapting with a lag is close to perfect adaptation. Perhaps the best analysis to date of learning and adaptive behavior is by Kelly, Kolstad and Mitchell (2002).They set up a model of an individual firm (say a farm) deciding what inputs to use, but where the firm's production process is subject to random shocks, and so profits are in turn uncertain. The firm has to choose its inputs to maximize expected profits. However the process generating these shocks is subject to a once and for all change. If the firm knew what this change was, it would simply adapt and choose a new mix of inputs to maximize expected profits given the new process generating the shocks. But the firm cannot directly observe the change in the process (i.e., it cannot directly observe climate change) - all it can observe is how the pattern of shocks it has experienced has changed. Kelly, Kolstad and Mitchell (KKM) assume the firm uses the information it gets from observing the shocks to gradually update its understanding of the process generating the shocks (using Bayesian learning), and eventually learn what the new process is. They distinguish between the costs of adaptation, which are essentially the difference between the long-run steady state costs the firm faced with the old and the new process generating the shock (i.e.,
it is the difference in costs the firm would face if it could adapt instantaneously to the new process generating uncertainty) and the costs of adjustment, which reflect the extra costs the firm faces in the process of learning because it has not fully learned the change in the distribution of shocks. KKM apply this analytical framework to an empirically estimated model of farm behavior in five states of the USA, to see how farmers respond to changes in climate and weather (where climate is the process which generates the (random) pattern of weather). They then simulate the effects of a change in climate which takes place gradually over a 100 year period, and compare profits when farmers have to learn about the change in climate to profits if farmers were fully informed about the change in climate. They show that while climate change has a long-run beneficial effect, raising profits per acre by $1.86 per acre in present value terms (so the ‘costs’ of adaptation are negative) the costs of adjustment are $5.47 per acre, so the net costs of climate change are $3.61 per acre. While these are not large percentage changes in overall profits of farming, the analysis shows that when one allows for the process by which agents have to learn about climate change the costs of climate change may be much higher than the costs if one assumes instantaneous perfect adaptation. In terms of our earlier discussion this implies that properly modelling the process by which agents learn about climate change in order to adapt their behaviour may significantly raise the costs of adaptation and hence require more mitigation than would be the case if one just assumed perfect learning and hence perfect adaptation. While we believe that the analysis of learning by Kelly, Kolstad and Mitchell is the most sophisticated we are aware of in the context of climate change, there are a number of limitations to their model from a more general perspective of the economics of information and learning. (This is not a criticism of their work, rather it shows how much more there is to do to study the link between learning and adaptation properly). There are five aspects we consider. (i) KKM use a model of passive learning - agents just take actions to maximize expected profits given their current beliefs about climate, and update their beliefs about climate as they get observations of the weather. But this may overstate the time it takes for agents to learn, because they could become active learners - i.e., if they suspected climate change was taking place they might try out a range of different crops or input patterns to see which worked best, i.e., they could use experiments to learn more about climate change. (ii) KKM assume that the only information available to agents is observations about the weather in their own location. But agents may be able to observe changes in patterns of weather across the globe, and draw inferences more generally about climate change. (iii) The model employed by KKM is a representative agent model. But in an economy agents also learn from others, partly through social networks through which agents share information (families, neighbourhoods, social groups, etc.), partly through agents observing directly what other agents do, and partly by observing the consequences of what other agents do, for example through prices. This has a number of implications. At one level the ability of agents to learn from others implies that information may disseminate very quickly in an economy. One area where this has
been widely studied by economists is in the field of financial markets where agents can observe what happens to prices of assets. Even if there are only a relatively small number of `informed' agents in a market, their behaviour, buying or selling assets in response to information, can change prices which in turn ‘reveals’ their information to uninformed agents. Economists have long argued for the importance of prices as conveying information in an economy. But there are problems. This kind of learning behaviour can generate pathologies - such as herd behaviour and speculative bubbles. Furthermore, if there are costs to agents acquiring information, but agents realize that the information they gain will be quickly learned by others, then this substantially reduces their incentives to sink the costs to acquire information. The classic situation where this arises is research and development, where there can be substantial costs to developing new products or technologies and if these can be easily imitated then the incentive to engage in R&D investment is sharply reduced. The standard policy response is either to provide patent protection or to carry out the research in public laboratories and make the results freely available. (iv) KKM assume that the only changing source of uncertainty is climate and the random weather patterns it generates. But in an economy there is a multitude of sources of uncertainty facing agents. One, just referred to, is change in technology and products, which can wipe out firms and industries. In addition there can be changes in consumer preferences, or changes in government policies. Climate change is just one other source of uncertainty and one other factor agents need to take on board. (v) The final point is that markets are not only a means of conveying information between agents, they are also a mechanism for selecting which firms survive. In the idealized model of a market economy, competition selects the efficient firms (both in the static sense of cost-minimizing and in the dynamic sense of innovating new technologies and products that consumers want) and weeds out the inefficient. This would suggest that the competitive process will be one of the mechanisms for ensuring that those who successfully and quickly adapt to climate change will flourish, while those who do not will go out of business. Of course this idealized model of market behavior is far from a description of actual economies, and economists are aware that there are deep reasons why actual markets will not perform this way (e.g. Dutta and Radner (1999)). The point is simply that this kind of consideration is absent from most analyses of learning and adaptation. 4 Catastrophes In the previous discussions, both adaptation and mitigation can reduce the risk due to climate change and climate variability. Generally speaking, they are substitutes, i.e., for the purpose to reduce the (potential) damage cost, you can either use more adaptation and less mitigation if adaptation is relatively cheap or you can use less adaptation and more mitigation if mitigation is relatively cheap. But it should be noted that such arguments are based on a notion that the climate change is continuous and slow and climate variability is small, i.e., the change of climate and climate variability is marginal. So through adaptation, society can compensate for the effects of climate change by undertaking actions. This approach can be thought of as being consistent with the notion of weak sustainability, i.e., that if natural capital is eroded by climate change then there is the possibility of its replacement by physical capital of some form.
However, it is thought that various aspects of climate change may not allow for the possibility of substitution. One important reason for this is that climate models can generate all sorts of sudden and large events considered to be catastrophic in nature which come with large damage costs , for example substantial loss of biodiversity because of the inability of species to adapt rapidly, and hence no possibility of substitution through physical capital. This situation is closely related to the notion of strong sustainability. In such situations the scope for adaptive options may be very limited. What does this imply for adaptation and mitigation policies? Three types of catastrophic effects modelled in Integrated Assessment Models (IAMs) have been characterized by Wright and Erickson (2003). These are: 1. Low Probability -- High Impact Events 2. Threshold Phenomena 3. Lack of Knowledge with readily resolved uncertainty For catastrophes of type 1, the issue is one of surprise as opposed to climate effects being smooth and gradual, as in is the ‘translated climate’ approach used by Peck and Teisberg (1994) or Mendelsohn and Neuman (1999), where the distribution function of climate parameters is shifted by climate change. With catastrophes of type 1 there is a switch to a completely different distribution function of climate parameters and the magnitude of effects is much larger than might be expected from what are termed `regular' parts of the climate domain. For these concerns to be of importance, events must be of low probability -- otherwise they would not be a surprise (there is some notion of incomplete information about what the climate response are) -- and high impact because otherwise the impacts would not be especially noticeable. For catastrophes of type 2, there does not need to be a low probability attached to the event. In fact if the threshold is crossed then effects will occur with certainty. These thresholds are those for which “significant” effects are to be noticed although no definition is made of what “significant” is. Nordhaus (1994) and Peck and Teisberg (1992) are cited as examples of these. For catastrophes of type 3, it is noted that whether uncertainty can be resolved has an important impact on the results obtained from IAM policy optimizations. Catastrophe here is seen as having a Hazard Function -- i.e., the event classed as a catastrophe cannot be determinably foreknown. These catastrophe types are then related to the scientific background of climate events causing geophysical catastrophe, such as runaway greenhouse, rapid sea level rise, or ocean circulation change. The linkage of these scenarios with their catastrophe characterizations gives particular implications for climate economics. For example, catastrophes can give rise to the possibility of negative discount rates arising out of negative economic growth, and affect how damage assessments should be incorporated, and consequences for adaptation. Incorporating such catastrophic effects into conventional IAMs is not straightforward. In a study of ocean/atmospheric circulation changes as a possible consequence of climate change Keller et al. (2000) suggest an inverse approach. They subject DICE
to a constraint of maintaining the Thermo-Haline Circulation (THC) system, and then estimate the cost that THC shutdown would need to exceed to make this an efficient policy. However Wright and Erickson (2003) see the role of damage assessment in models such as those of Keller et al. (2000) and Mastandrea and Schneider (2001) as problematic in that they use perfect foresight which conflicts with catastrophe as necessarily involving uncertainty of crossing thresholds. The consequences of catastrophes of type 3, where there is decreased predictability, can lead to the possibility that mitigation and adaptation could be complements. Wright (2000) uses an option value model and shows that reductions in mitigation which lead to increased variability of climate can lead to reductions in adaptation. This arises from those situations in which the adaptive strategy requires predictability for effectiveness. It also can lead to increased pre-adaptation damages. Further the predictability decreases arising from increased variability do not decay with time. This is closely related to the argument we made in section 2, that if the costs of adaptation depend on the stock of greenhouse gases because more mitigation slows the rate of change of climate and hence makes adaptation easier, then adaptation and mitigation may be complements. This effect does not arise in Bayesian models of learning such as Kelly and Kolstad (1999) in which expectations are based on past observations. Whilst learning is slow in this model, after several decades uncertainty is resolved. The increased difficulty in making predictions is also analyzed by Fisher and Rubio (1997) where the increase in hydrological uncertainty leads to an increase in the size of optimal water infrastructure. However, predictability decreases arising from increased variability do not decay with time, and represent an enduring degradation of information. Consequently more observations are needed to establish an understanding of climate change. Omitting increased climate variability into adaptation models leads to an overestimate of the gains to be made from adaptation, and a consequent under estimate of climate change damages. Wright and Erickson (2002) consider timing in relation to adaptation. The key feature of their model is that they consider climate change as being characterized by the extreme levels and the degree of variability rather than the mean value (such as mean temperature). They determine adaptation timing within an optimal stopping model, which extends the Conrad (1997) model. Increasing temperature variability here delays the optimal adaptation time and increases cumulative damages. They conclude that failing to take into account the dynamic effects of variability on adaptation timing leads to overstating the damage reductions to be obtained from adaptation and so makes mitigation a more attractive strategy. Climate variability delays action by individual actors, the delay depending on characteristics of the particular sector that they are in. 4.1 Catastrophic Effects and the Tolerable Windows Approach The models of catastrophic effects in the previous section attempted to incorporate such effects into conventional IAM models based on choosing optimal mitigation and adaptation policies. However this involves solving a complex decision problem which has dynamic as well as uncertainty aspects. The mathematical complexity of the problem will be enhanced by the presence of non linearities such as sudden increases
in damage costs, and when these include considerable uncertainty as to when they occur, and how large they are, correspond to “surprises” or “catastrophes”. An alternative approach, the Tolerable Windows Approach (TWA), effectively denies the possibility of simple trade-offs between mitigation and adaptation policies in the presence of catastrophic effects. Rather what this approach does is to establish a critical range of values (tolerable window) for a variable such as the stock of greenhouse gases and requires that mitigation policies are set to ensure that this variable always remains within this critical range. Adaptation policies are therefore only relevant as long as mitigation policies keep climate within the tolerable window. A key question then is how the tolerable windows are set. This requires either knowledge or a judgment about what are the most important elements of uncertainty, and this is why the TWA is seen as being inappropriate by some. For example Tol (1999) sees the constraints, known as guardrails, imposed as being dependent on geological records and so possibly an extreme position. He suggests that the requirement that future climate scenarios not fall outside past experience reflects a naturalistic fallacy, or status quo effect, that deviations from past experience are inherently undesirable. He has similar criticisms of the ecological considerations for the quantities corresponding to a Safe Window. This difference between the views of natural scientists and economists as to the seriousness of large emission increases is supported by surveys conducted by Nordhaus (1994) and Morgan and Keith (1994) of subjective estimates of the extent to which a changing climate would create economic damage. For social scientists a 6o C increase in temperature would lead to a 6% of GWP damage, whilst for natural scientists this would be 37%. One reason for this large range is that social scientists may regard there as being more possibility of substitution and adaptation possibilities. The tolerable windows approach starts from what is thought to be acceptable climates and then works back to the adaptation and mitigation strategies that are consistent with these. Further reasons that are given for its use are: 1. IAMs are complex models for which, in practice, dynamic optimization may not be feasible except for drastic simplification of the climate model. 2. The TWA is the inverse problem to that of dynamic optimization models. TWA asks the question `If we have a view as to where we wish the climate and economy to be at some time in the future, what paths would be consistent with that?' 3. It is argued that TWA better addresses the questions posed by policy makers without having to calculate the full solution to the optimization problem, because policy makers are primarily concerned with unacceptable climate change impacts or unacceptable mitigation costs. TWA is thus seen reflecting the stated preferences of policy makers in the context of a detailed IAM. 4. Another reason why TWA is thought to appeal to policy makers is that it provides a simpler way of dealing with timing issues. If climate change and emissions negotiations take time, then the time at which policy must start, and Business as Usual (BAU) ends, becomes an important issue. The set of paths arising out of the TWA
analysis may show (and in the examples to be discussed later, do indeed show) that there is a date after which no tolerable path can be the same as that of BAU. 5. The optimal control problem may display an aspect of separability so that the process of optimization can be divided into one of determining a set of emission paths that satisfy the scientific climate change model and constraints and then selecting one of those paths according to other (presumably economic) criteria. 6. The TWA approach is seen by some as avoiding perceived problems with the use of the cost-benefit analysis approach to climate change decision analysis (for example, Bruckner et al. (1999) and Toth (2000), (2002)). One perceived difficulty is that it is thought that there is a major problem in obtaining reliable monetary valuation of climate change impacts which result in temperatures and related impacts outside of those recently experienced by individuals whose valuations are sought. A second difficulty is that CBA is interpreted as being a tool of aggregate `utilitarian' analysis. We expand on this later in the section on distributional consequences. Finally, CBA is seen as being an tool designed to produce a single optimal path in models such as DICE . However the outcome of the CBA calculation may not turn out to be acceptable to policy makers where the resulting emissions/climate outcomes fall outside (maybe well outside) of current experience. Other economic criteria can then be considered such as cost effectiveness or ones having a Rawlsian approach towards distributional considerations. So that if a set of possible emission paths is determined, then the one with the most desirable distributional consequences in terms of the worst off could be selected. However it should be noted that in the examples produced the economic constraints are an important component of the TWA approach and whilst authors sometimes claim that they are avoiding the problems of using cost-benefit analysis, in setting limits they believe are set at an arbitrary level, CBA is often used as a justification for what that level is. A more detailed discussion of models that use the TWA is given in the appendix. 5 Other Considerations 5.1 Adaptive Capacity 5.1.1 What is meant by Adaptive Capacity? In section 2 we presented some very simple analysis of adaptation and mitigation designed to address some issues about the relationship between them. In those simple models a key determinant of the amount of mitigation that will be done relative to adaptation is the cost of adaptation relative to the cost of mitigation, and we recognized that this might vary across different countries or groups within a country. However in thinking a bit more carefully about why and how costs of adaptation might vary across countries leads to a rather richer way of construing costs of adaptation. We have already noted in section 3 that a proper analysis of costs of adaptation requires careful consideration of the process by which agents may learn about climate change, and that this opens up a very rich set of issues in the economics of information which have not yet been fully addressed in the context of climate change. Yohe (2001) also points out that there may be limitations in existing evaluations of adaptation options, and while economic costs and benefits are
important aspects of such evaluations, they are not sufficient to capture the appropriateness of all adaptation options. A fuller analysis would have to concentrate on the roles and responsibilities in adaptation of individuals, communities, corporations, private and public institutions, governments and international organizations. A popular way of trying to encapsulate these considerations is in the concept of adaptive capacity of a particular group. Adaptive capacity has some similarity to the Tolerable Windows Approach in that it determines the vulnerabilities of the climate to change, as opposed to considering emissions scenarios and working out the consequences of those. In this subsection, we briefly look at some definitions of adaptive capacity and then review some attempts to measure it. Tol (2002) points out that adaptation is carried out by individual agents, and there is a long history of adaptation to weather and climate change. The role for government in adaptation is to create the appropriate conditions for it to take place. While, in general, adaptation operates at different scale from mitigation, Tol introduces the notion of facilitative adaptation, meaning roughly enhancing adaptive capacity. But as he notes while there is some broad discussion of what drives adaptive capacity (see, for example, our summary of Yohe and Tol (2002)), there is not a lot of hard evidence to support this. In terms of enhancing adaptive capacity Tol also makes the point that getting markets to work properly may be an important part of adaptive capacity because they provide the incentives and flexibility which government policy may lack. Markets are also important as a device for learning, a point we made in previous sections. Adaptive capacity is seen as being an attractive concept because it reduces the many varied forms of adaptation (seen as being almost infinite in number by Smit and Pilifosova (2003)) to the assessment and promotion of a single variable. However this aggregation means that will be no simple way of doing this. There have been three somewhat similar definitions of adaptive capacity. Smit et al. (2000), define adaptive capacity as the potential or capability of a system to adapt to (to alter to better suit) climatic stimuli.13 For Folke et al. (2002) adaptive capacity is the ability of a socialecological system to cope with novel situations without losing options for the future, and resilience is key to enhancing adaptive capacity. In social systems, the existence of institutions and networks that learn and store knowledge and experience, create flexibility in problem solving and balance power among interest groups play an important role in adaptive capacity14. According to Klein (2003), adaptive capacity can be defined as the ability to plan, prepare for, facilitate and implement adaptation measures. 5.2.2 How is adaptive capacity determined and measured? If one is to make these definitions of adaptive capacity operational, so that one can either compare countries in terms of their adaptive capacities, or assess how a country's adaptive capacity varies over time, either for exogenous reasons or in response to deliberate policy measures to improve adaptive capacity, it would be desirable to have some way of measuring adaptive capacity.
Its synonym adaptability is defined as the ability , competency or capacity of a system to adapt to (to alter to better suit) climatic stimuli. 14 Adaptive capacity in ecological systems is related to genetic diversity, biological diversity, and the heterogeneity of landscape mosaics.
Kolstad and Toman (2001) present a simple model of greenhouse gas emissions and climate change, but then argue that the model should be extended to allow for adaptation. This would involve accumulating stocks of human and physical capital for adaptation such that for any stock of greenhouse gases and temperature the damages caused by climate change to economic productivity and household welfare is a decreasing function of these stocks. However they do not say how they would measure these stocks of human and physical capital for adaptation. A rather cruder approach might be to use the level of economic output (per capita) as a proxy for adaptive capacity. But this just amounts to saying that richer societies have higher adaptive capacities than poorer ones. A rather more sophisticated approach is provided by Yohe and Tol (2002) who offer a practically motivated method for evaluating adaptive capacity by assessing the potential contributions of various adaptation options to improving systems' coping capacities. First, the determinants of adaptive capacity include a variety of system, sector, and location specific characteristics. These are: 1) The range of available technological options for adaptation, 2) The availability of resources and their distribution across the population, 3) The structure of critical institutions, the derivative allocation of decision-making authority, and the decision criteria that would be employed, 4) The stock of human capital including education and personal security, 5) The stock of social capital including the definition of property rights, 6) The system's access to risk spreading processes, 7) The ability of decision-makers to manage information, the processes by which these decision-makers determine which information is credible, and the credibility of decision-makers, themselves and 8) The public's perceived attribution of the source of stress and the significance of exposure to its local manifestations. Some of these determinants will operate on a macro-scale in which national or regional factors play the most significant role, but other determinants will function on a more micro-scale that is precisely location specific. But the local manifestations of the macro-scale determinants of adaptive capacity are their most critical characteristics. So, overall, adaptive capacity depends on local characteristics. These determinants are used to construct a measure of an economic system's `coping capacity' and its robustness. This is constructed in a max-min way using subjective scores for each options determining characteristics, and the overall effectiveness of each adaptation option. For each adaptation option, overall feasibility is the minimum
of the subjective scores attributed to each determinant. The product of this overall feasibility and the subjective score for efficacy gives a `potential contribution to coping capacity', and the maximum of these over all adaptation options gives the coping capacity index. The robustness of this index is measured by the ratio of the average of the 2nd and 3rd highest adaptation options to the highest. As far as we are aware this is one of the few attempts to provide a formal way of measuring adaptive capacity. This is not surprising as Klein (2003) points out15 that adaptive capacity is not a concept that can be measured in a straightforward way. According to Klein (2003), factors that determine adaptive capacity to climate change include economic wealth, technology and infrastructure, information, knowledge and skills, institutions, equity and social capital. It therefore follows that most industrialized countries have higher adaptive capacities than developing countries. For example, whilst Bangladesh and The Netherlands have a similar physical susceptibility to sea-level rise, Bangladesh lacks the economic resources, the technology and the infrastructure that The Netherlands has at its disposal to respond to potential impacts. He suggests that adaptive capacity could be measured either by using Integrated Assessment Models or an Agent Based Modelling approach. Examples of the latter are Berkhout, Hertin, Gann (2003), and Tompkins and Hurlston (2004). Integrated assessment provides a methodologically rigid but consistent approach to top-down analysis of vulnerability, in which adaptive capacity is represented by a series of indicators, ranging from income per capita and literacy to corruption and religion16. Agent-based modelling combines empirical research with modelling techniques to conduct a more bottom-up analysis of adaptive capacity. Whilst indicators may be a useful tool for assessing and predicting adaptive capacity, their use in integrated assessment does not contribute to the understanding of how adaptive capacity develops. Such understanding is a prerequisite to the task of enhancing adaptive capacity where societies are vulnerable. Systems in which the behaviour of each agent depends on that of the other agents---so-called complex systems---produce emergent properties. Society can be considered a complex system, where agents are decision-makers from the local to the national or even international level, and adaptive capacity can be thought of as an emergent property. Agent-Based Modelling (ABM) is a method by which one investigates and describes complex systems and their emergent properties. ABM keeps track of the complicated relationships between agents in a system and is a way of modelling interactions of non-rational, or non-market oriented, decision rules. Suitable problems for ABM are those in which individual decision-makers have a mix of incentives: social and political pressures create an incentive to act in one way, whereas economic selfinterest creates an incentive to act in another. An agent-based model can provide both scientists and decision-makers with insights into how extreme negative outcomes might emerge from the positive feedbacks of the social and political pressures. In addition, it can offer a method for testing whether small changes made to the system at the outset might produce more desirable results.
Also see the references cited by Klein (2003). Assumptions on how these indicators affect adaptive capacity, based on a literature survey and empirical research, together with scenarios of how these indicators change over time, will allow adaptive capacity to be combined with impact potential to arrive at a measure of vulnerability.
ABM therefore seems particularly suitable for analyzing the process of adaptation, including the conditions that prevent effective adaptation or even promote maladaptation. It has only appeared relatively recently, both because it requires a lot of computing power and because the theoretical foundations on which it is built theories of evolving networks and positive feedback loops - are relatively recent. However, whilst adaptive capacity may not be something that can be precisely defined and measured, it may be possible to measure its determinants, such as the flexibility of the economy to price and technology shocks. 5.3 Distributional Aspects Of Adaptation and Mitigation As well as there being an important temporal distinction between mitigation and adaptation as discussed in section 2, there are also spatial considerations. These are seen by many as being very important as the spatial distribution of impacts and costs will have implications for principles of equity and for economic development, and for many commentators these equity issues may well dominate issues arising from concern for efficiency. Mitigation costs will tend to be borne by those countries that are substantial emitters of greenhouse gases, whilst those bearing climate change damage costs and where there is the greatest potential scope for adaptation may well be countries which have relatively low emissions, and may have low levels of resources and adaptive capacity. Damages may be asymmetrically felt across the developed/developing country divide as climate change becomes more of a problem in the future, and this will raise questions of inter-country and intergenerational equity. Climate change is likely to increase world- and country-scale inequity, both within the present generation and between present and future generations, particularly in the developing countries. This can be handled within cost benefit analysis providing that appropriate distributional weights are used. However opponents of the use of traditional cost-benefit methods, see this as being a major problem, especially in international climate policy negotiations. Whilst there has been work on empirical estimates of climate change impact costs, Tol et al. (2004). spatial distributional considerations have not received a great deal of formal attention. There are a number of conceptual issues involved in trying to capture distributional considerations, such as the nature of the social welfare function to be used. Particular welfare functions can have very strong implications. For example, Tol (2003) in his criticism of the use of cost benefit analysis in climate change models, uses a welfare function which is a weighted sum of the logarithm of income in different countries. This implies that if income falls to zero in any country as a result of climate change impacts then global welfare tends towards minus infinity independently of what happens in other countries. Tol uses this to illustrate the possibility of infinite variance of welfare in climate change policy models. However, it could be argued that in the eventuality of crop failure in certain parts of the world, there is the possibility of migration or income transfer. The extent to which this problem arises will also depend on how fine is the resolution for the disaggregation of countries. Where countries are aggregated into large blocks then the analysis implies there will automatically be transfers of income between different countries in the block. So the objective function may need to be very carefully specified so that results obtained reflect what are possibly arbitrary choice of functional form or of
aggregation. However the general implication is that if we take distributional issues seriously, then poor countries should be protected by better off countries from exposure to significant costs, whether damage costs or mitigation or adaptation costs. For the purpose of this survey, the most relevant issue is that the range of adaptation and mitigation strategies may depend on a country's stage of development. For some developing countries it is thought that mitigation would be an expensive strategy, either because they contribute so little to global emissions that mitigation would have a trivial effect, or because to mitigate would mean foregoing significant development benefits. However, adaptation could also be expensive if the adaptability of individuals depends on the stock of human capital. Developing countries may also face significant distortions in capital markets, making capital-intensive forms of mitigation or adaptation expensive. Where such constraints occur the use of international transfers of aid either in cash or know-how are a potential solution, and models used need to include such policy options. The general point is that, consistent with the analysis so far, countries can face quite different relative costs of damages, mitigation and adaptation, so the optimal strategy for each country will reflect these differences. Furthermore, it is poorer countries that are currently subject to weather extremes such as drought and floods. That is perhaps why they are poor. An expected consequence of climate change is an increase in the number of these extreme weather events. Smit and Pilifosova (2001) suggest that it is these extreme weather events rather than changes in long term average temperatures that most require adaptation. The low adaptive capacity of poor countries poses special problems. The various impact and adaptation options for differing parts of the world are set out in chapter 19 of the report of Working Group II IPCC (2001). The IPCC report provides a comprehensive list of potential climate change impacts for most of the world's regions and economic sectors. It suggests that high adaptive capacity is seen as requiring : a stable and prosperous economy, a high degree of access to technology at all levels, well delineated roles and responsibilities for adaptation strategies, systems in place for the national, regional and local dissemination of climate change and adaptation information, and an equitable distribution of access to resources. LDCs are seen as being unlikely to fulfil these requirements. This is supported by comments on the ability to cope with risk. Attitudes to risk are frequently mentioned as a source of variation in adaptability. Economic affluence and stability, institutions and infrastructure, access to capital and information are all seen as reasons why adaptive capacity is greater in developed countries. However measures of adaptive capacity are seen as being difficult to construct. Hence empirical evidence for these statements is hard to find. Indeed, the report cites a study by Ramakrishnan (1998) who suggests that low to middle intensity farming as found in LDCs will have greater adaptive capacity than the highly intensive agricultural systems found in wealthier economies. The requirements for enhancing adaptive capacity place a strong emphasis on national action. So a lack of funds and resources at a national level is problematic. This centralized emphasis is criticized by Tol (2003) who sees adaptation as essentially a single agent response; adaptation takes place at the finest possible or local scale. There is little role here for governments as incentives for adaptation already exist at an individual level. Where governments do have a role is in providing a framework for facilitative adaptation. This, however, could be a cause of maladaptation, since mitigation and facilitative adaptation will compete for resources.
Tol. et al. (1998) suggest that distributional considerations at a governmental level can be taken account of through financing measures. However where adaptation takes place at an individual level, distributional considerations will influence what adaptation takes place and hence what the policy response is. At an international level, distributional considerations focus on the impact on poor and less developed countries. An aspect that has received some attention (Babiker et al. (2000), is on the effect that comes though international trade where poorer countries may suffer disproportionately, and that this can have an effect on health and disease, so that maladaptation can occur. Tol and Dowlatabadi (2001) suggest that for climate change policies in developed countries that substantially reduce their rate of growth there will be a more than off-setting effect on growth in LDCs and consequent health impacts. The variation of adaptation across regions brings up complex issues of justice and fairness. The climate change problem enhances equity discussions because the socioeconomic conditions driving emissions may determine what are the adaptive and mitigative capacities of various countries. Climate change promises to bring an uneven distribution of consequences. Poorer nations that have less adaptive capacity are more vulnerable to climate change damages, and so more in need of successful adaptation. Some evidence for this has been provided by Yohe and Tol (2002) who show that poorer people are more likely to fall victim to natural catastrophe than are richer people, and more densely populated areas are more vulnerable. These of course will tend to be areas of deprivation. They find a positive relationship between income and vulnerability. Reasons given for this are the stocks of human and social capital in developed compared to developing countries. People in more egalitarian societies are less likely to fall victim of natural violence than are people in societies with a highly skewed income distribution. Climate policy in practice has become embroiled in two separate questions. One is the need to estimate the impacts of a wide range of plausible climatic futures, and the other is the need to estimate the relative adaptive capabilities of future societies so as to assess the equity implications of the consequences of global warming. This complicates the question of how much mitigation there should be and how much adaptation if the typically proposed mitigative activities slow the economic growth rates of those very countries that need to build adaptive capabilities by growing economically (e.g., IPCC, (2001)). However the argument could work in reverse as in Tol and Dowlatabadi referred to above. So the direction of the argument is by no means clear. And this applies more generally. For example developed countries which have substantial coastal capital could face greater adaptation costs in the short to medium run, than developing countries which have a much smaller capital stock. But, if developing countries are to be exempted from mitigation because of the development consequences of that and they are allowed to emit unabated amounts of greenhouse gases, then the risks of severe climate change impacts, including irreversibilities, will increase. Schneider (2004) suggests that the best way to approach this dilemma is to assess the range of possible climate change outcomes, their costs, and the distribution of those costs, and then to weigh those impacts against the costs and benefits of a host of mitigation options carried out in various countries. Side payments or other schemes to redress the inequity issue will have to be part of future climate policy negotiations if they are to be acceptable to a majority of nations.
In this way the development and equity problems that might be enhanced by climate change can be de-coupled from the climate change problem itself. Adaptation might seem cheaper in a cost-benefit analysis that aggregates all costs and benefits since the rich country, with a much larger share of world GDP, will be able to adapt more easily. But, that policy may not be fair in its distribution across rich and poor countries, which is the concept behind distributive justice/equity. Efficiency versus equity dilemmas can lead to alternative political views of what should be done, and are also connected to the question of who should pay to abate risks. Most discussion of the distributional effects comes through a discussion of the various degrees of vulnerability and adaptive capacity, for example Huq et al. (2003). Reduced adaptive capacity arises from the absence of potential government support in poor countries. So that for two case studies of Bangladesh and Mali, whilst various adaptive actions are available these are dependent on governmental action such as information provision and little had been done to incorporate these into national policy. Kane and Yohe (2000) set out some of the reasons why adaptation might be more difficult in developing countries. They conclude that institutions matter in determining what adaptation can take place. This is related to the efficient flow of timely and credible information. It cannot be presumed that existing institutions will cope, and that the institutional gaps will be filled. So that where there is a lack of social capital, adaptation will be more difficult. As noted earlier, Kolstad and Toman (2001) present a simple model of greenhouse gas emissions and climate change which allows for the incorporation of adaptation as well as mitigation activities as an extension. It has relevance to the discussion of distributional and developmental issues as adaptation here involves accumulating knowledge and physical capital stocks such that for any given GHG stock and temperature level, the level of economic productivity, and the direct damages of climate change on household well-being, are smaller than without the investment. In contrast to what is developed above, adaptation also could be assumed to occur naturally as a by-product of economic growth, so that the partial derivative of the damage function with respect to the level of economic activity, is negative. The argument here is that wealthier societies are less dependent on the natural systems than developed countries. Natural systems figure prominently in sustaining welfare in poorer societies, so that developed countries have higher adaptive capacities than developing countries. Developed countries may be able to pursue a strategy of Weak Sustainability with respect to Climate Change whereas developing countries need to pursue one of Strong Sustainability. 6. Conclusions In this paper we have surveyed the literature on economic approaches to climate change. We began by applying some simple analytical models from which a key conclusion emerged. Under a wide range of circumstances, including allowing for uncertainty, irreversibility and learning, adaptation and mitigation can be thought of as substitutes. This carries the important implication that the lower are the costs of adaptation the less reliance should be placed on mitigation. However determining
exactly what this insight tells us about actual policies is complicated for three reasons. First, as we have noted, adaptation and mitigation policies operate at very different geographical scales, in different time periods and with very different agents, especially in terms of the need for government policy. Trying to build a comprehensive model which incorporates all mitigation and adaptation options is therefore infeasible. Second, the costs of adaptation depend crucially on how individual agents learn about how climate change is occurring and there have been very few models which give us a real handle on how this occurs in practice. Such modelling that has been done takes a very simple approach to learning and ignores a wide range of aspects of the economics of information and learning. So we are a long way from fully understanding how quickly society will be able to adapt to climate change. Third, there are other factors, other than the ability to learn, which determine the capacity of societies to adapt to climate change. However, we do not yet have a full understanding of these factors, or a means of translating these factors in an agreed way into predictive indicators of capacity to adapt, let alone an understanding of how this relates to the costs of adaptation. Such work as is available suggests that adaptive capacity is likely to be correlated with income, so costs of adaptation are likely to be highest in the poorest societies, which are also the societies which might be most exposed to unmitigated climate change impacts. This makes even more acute the need to think carefully about both the welfare basis of climate change policy and the distinction between climate change policies and development policies. While the conclusion that adaptation and mitigation policies are broadly substitutes for each other is a reasonably robust result, the one area where this may not be the case is when consideration is given to catastrophic effects and extreme variability of climate, when mitigation, as a means of reducing the risk of catastrophes or reducing extreme variations in climate, may help adaptation to occur and so build a complementarity between mitigation and adaptation. But again building such factors into a comprehensive model for determining the optimal mix of adaptation and mitigation policies is extremely difficult, and some of the attempts to get round these complications, as in the Tolerable Windows Approach, might be thought as avoiding the complications rather than solving them. In summary, while there has been progress in developing some of the conceptual frameworks for thinking about issues of adaptation and mitigation, there is much yet to be done if we are to translate these concepts into models that would provide practical tools for policy making. References Barker, T. (2001). “Representing the integrated assessment of climate change, adaptation and mitigation”, Tyndall Centre Working Paper. Berkhout, F., Hertin, J., and Gann, D. M. (2004). “Learning to adapt: Organisational adaptation to climate change impacts”, Tyndall Centre Working Paper. Bruckner, T., Petscel-Held, G., Toth, F., Fussel, H-M., Helm, C., Leimbach, M., and Schellnhuber, H-J. (1999). “Climate change decision-support and the Tolerable Windows Approach”, Environmental Model Assessment, vol. 4, pp. 217-234.
Conrad, J.M. (1997). “Global warming: when to bite the bullet”, Land Economics, vol. 73, pp. 164-173. Dessai, S., Adger, W., Hulme, M., Kohler, J., Turnpenny, J., and Warren R. (2003). “Defining and experiencing dangerous climate change”, Tyndall Centre Working Paper. Dutta, P. and Radner, R. (1999), “Profit maximization and the market selection hypothesis”, The Review of Economic Studies, vol. 66, pp. 769-798. Fankhauser, S., Smith, J. B., and Tol., R. S. J. (1999). “Weathering climate change: some simple rules to guide adaptation decisions”, Ecological Economics, vol. 30, pp. 67-78. Folke, C., Carpenter, S., Elmqvist, T., Gunderson, L., Holling, C. S., Walker, B., Bengtsson, J., Berkes, F., Colding, J., Danell, K., Falkenmark, M., Gordon, L., Kasperson, R., Kautsky, N., Kinzig, A., Levin, S., Maler, K.-G., Moberg, F., Ohlsson, L., Olsson, P., Ostrom, E., Reid, W., Rockstrom, J., Savenije, H. and Svedin, U. (2002). “Resilience and sustainable development: building adaptive capacity in a world of transformation”, Environmental Advisory Council, Sweden. Greenough, G., McGeehin, M., Bernard, S., Trtanj, J., Riad, J., and Engelberg D. (2001). “The potential impacts of climate variability and change on health impacts of extreme weather events in the United States”, Environmental health Perspectives, vol. 109, pp. 191-198. Haurie A. (2002). “Integrated assessment modeling for global climate change: An infinite horizon optimization viewpoint”, NCCR-WP4 Working Paper, Swiss National Centre of Competence, University of Geneva. Ingham A. and Ulph A. (2004)), “Uncertainty, irreversibility, precaution, and the social cost of carbon”, in Helm D. (ed.), “Climate Change and Policy Response”, Oxford University Press, Oxford. Ingham A., Ma J. and Ulph A.(2004), “Some Simple Analytics of Adaptation and Mitigation”, (mimeo.) Kane, S. and Shogren, J. F. (2000). “Linking adaptation and mitigation in climate change policy”, Climatic Change, vol. 45, pp. 75-102. Kane, S. and Yohe, G. (2000). “Societal adaptation to climate variability and change: an introduction”, Climatic Change, vol. 45, pp. 1-4. Keller, K., Tan, K., Morel, F.M.M., and Bradford, D.F. (2000). “Preserving the ocean circulation: Implications for climate policy”, Climatic Change, vol. 47, pp. 17-43. Klein, R. J. T. (2003). Part 3 of “Resilience and Adaptation to Climate Change: An Interdisciplinary Perspective”, PhD Thesis, Christian-Albrechts-Universitat zu Kiel.
Klein, R. J. T., Schipper, E. L. and Dessai, S. (2003). “Integrating mitigation and adaptation into climate and development policy: three research questions”, Tyndall Centre Working Paper. Kolstad, C. D. and Toman, M. (2001). “`The economics of climate policy”, Discussion Paper, Resources for the Future. Kriegler, E. and Bruckner, T. (2002). “Global and Annex I emissions corridors for the 21st century”, Integrated Assessment and Decision Support, Proceedings of the 1st Biennial Meeting of IEMSS, vol. 3, pp. 151-156. Kriegler, E. and Bruckner T. (2004). “Sensitivity analysis of emissions corridors for the 21st century”, Climatic Change, vol. 66, pp. 345-387. Leimbach, M. and Bruckner, T. (2001). “Influence of economic constraints on the shape of emission corridors”, Computational Economics, vol. 18, pp. 173-191. Parry, M., Arnell, N., McMichael, T., Nicholls, R., Martens, P., Kovats, S., Livermore, M., Rosenzweig, C., Iglesias, A. and Fischer, G. (2000). “Millions at risk: defining critical climate change threats and targets”, Global Environmental Change, vol. 11, pp. 181-183. Pielke, R. A., Jr. (1998). “Rethinking the role of adaptation in climate policy”, Global Environmental Change, vol. 8, pp. 159-170. Pizer, W. (2003). “Climate change catastrophes”, RFF Discussion Paper. Schneider S. (2004), “Climate change”, website: http://stephenschneider.stanford.edu/Climate/ClimateFrameset.html. Smit, B., Burton, I., Klein, R. J. T. and Wandel, J. (2000). “An anatomy of adaptation to climate change and variability”, Climatic Change, vol. 45, pp. 223-251. Smit, B. and Pilifosova, O. (2001). “Adaptation to climate change in the context of sustainable development and equity”, in Climate Change 2001: Impacts, Adaptation, and Vulnerability - Contribution of Working Group II to the Third Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge: Cambridge University Press. Smit, B., and Pilifosova, O. (2003). “From adaptation to adaptive capacity and vulnerability reduction”, in (J. B. Smith, R. J. T. Klein, and Saleemul Huq, eds.) Climate Change, Adaptive Capacity And Development, London: Imperial College Press. Tol, R. (1999). “Safe policies in an uncertain climate: An application of FUND”, Global Environmental Change, vol. 9, pp. 221-232. Tol, R. (2001). “Equitable cost-benefit analysis of climate change”, Ecological Economics, vol. 36, pp. 71-85.
Tol, R. (2002) “Adaptation and mitigation: Trade-offs in substance and method”, mimeo. Tol, R.S.J. (2003), “Is the uncertainty about climate change too large for expected cost-benefit analysis?”, Climatic Change, vol. 56, pp. 265-289. Tol, R.S.J., Downing T.E., Kuik O.J., and Smith J.B. (2004), “Distributional aspects of climate change impacts”, Global Environmental Change, vol. 14, pp. 259-272. Tompkins, E. and Hurlston, L-A. (2003). “Report to the Cayman Islands' Government: Adaptation lessons learned from responding to tropical cyclones by the Cayman Islands' Government, 1988 – 2002”, Tyndall Centre Working Paper. Toth, F. (2002). “State of the art and future challenges for integrated environmental assessment”, Paper presented at EFIEA Conference on Sustainability Assessment. Ulph, A. (1998). “Learning about global warming”, in (N. Hanley and H. Folmer, eds.) Game Theory and the Environment, Cheltenham: Edward Elgar, pp. 255-286. Ulph, A. and Ulph D. (1997). “Global warming, irreversibility and learning”, Economic Journal, vol. 107, pp. 636-650. Wright, E. and Erickson, J. (2002), “Climate variability, economic adaptation, and investment timing”, mimeo. Wright, E. and Erickson, J. (2003). “Incorporating catastrophes into integrated assessment: Science, impacts, and adaptation”, Climatic Change, vol. 57, pp. 265-286. Yohe, G. W. (2000). “Assessing the role of adaptation in evaluating vulnerability to climate change”, Climatic Change, vol. 46, pp. 371-390. Yohe, G. W. (2001). “Mitigative capacity -- the mirror image of adaptive capacity on the emissions side”, Climatic Change, vol. 49, pp. 247-262. Yohe, G. W. (2003). “Estimating benefits: Other issues concerning market impacts”, OECD Workshop on the Benefits of Climate Policy: Improving Information for Policy Makers. Yohe, G. W. and Tol, R. S. J. (2002). “Indicators for social and economic coping capacity -- moving toward a working definition of adaptive capacity”, Global Environmental Change, vol. 12, pp. 25-40.
Appendix: How Does TWA Work? The TWA model is set out in several papers, e.g. Bruckner et al. (1999), Yohe and Toth (2000), Toth (2002). Paths are set out for variables that are of concern to policy makers. Two different types of TWA models have been used in practical applications. One set of models use a complex IAM but a rather simple set of constraints and with a single set of fixed levels. A second set of models consider a variety of parameter values both for the climate and economic policy aspects, and a more complex set of constraints to define the guardrails. However, in order to accommodate this a relatively simple climate model is used. Typically these are for the levels and rate of change of variables as functions of time. These paths are known as `guardrails'. They could be derived from those paths that are needed for catastrophe avoidance. Because of this the `guardrails are seen by some as being analogues of critical loads (Bruckner et al. (2003), especially figure 1). However, they are usually stricter than those as policy may be not to just avoid catastrophe but to keep Climate Change impacts at a lower level within the domain of ‘regular behavior’. How this works can be seen from some examples. Bruckner et al. (1999) provide an example based on a tolerable climate window formulated by the German Advisory Council (WBGU).Guardrails are specified on temperature, rate of change of temperature, sea level and rate of change of sea level. The levels of the guardrails are defined by recent climate history (over the last 120,000 years) (see Yohe and Toth (2000) p. 106.) and by expert opinion. A separate guardrail is defined to accommodate mitigation costs. It is not specified where this comes from, and is stated to be just an example. This is specified by a maximum reduction in carbon emissions of 10% p.a. Whilst the study recognizes that other constraints such as ones on GDP or welfare ought to be included, the fact that they are not and there is not a clear way in which they can be must be seen as a considerable weakness in the application. In general, TWA studies reject CBA as an overall methodology, but do not make any suggestion for a replacement. Instead, CBA is used as a justification for the mitigation cost constraints that are imposed as guardrails. These constraints are then combined with an IAM. The results of this exercise give the permissible paths for CO2 concentration and emissions, global temperature and sea level against time. Leimbach and Bruckner (2001) calculate an `Emissions Corridor' which is the set containing paths which satisfy the guardrail constraints. To do this requires a complete model inversion. The algorithm they use starts from an initial position and for a given time increment maximizes and minimizes the level of emissions that satisfy the constraints. The set that is obtained will contain all paths satisfying all the guardrail constraints over time, subject to the way in which it is converted into a discrete problem, and the size of the increment being used. However it will not be the case that any path in the emission corridor obtained will be feasible, as at each stage the maximization and minimization only depends on the initial value of the state variable for that incremental step and not on their past history. So, for example, a certain high level of emissions may be feasible only if emissions before that incremental stage were very low. They focus on the influence of economic constraints on the emissions corridor so they consider a single constraint arising out of climate itself. The economic constraints relate to 1) Net welfare loss. This is for a maximum loss of per capita GDP on a regional basis, for example 6%. 2) A maximum rate of change of consumption increase. Again there is no particular justification for these
rates of change which are set out as illustrations. The minimum rates of change are different for different regions and are 1\% for the OECD, 1.5% for former USSR to 2% for China and Rest of the World. One of the main implications is that effective policy towards emission reduction needs to start much earlier in North America than in Europe or the developing world. The calculated emission corridors show bulges in the middle compared with a narrowing at the end, due to the fact that there are a large range of intermediate emission levels which are consistent with the constraints when integrated through the climate model. Bruckner et al. (1999) point to several possible extensions. The first is to disaggregate the socioeconomic aspects to a regional level, so that regional as well as global restrictions can be imposed (this is the ICLIPS model). These restrictions are intended to be for per capita GDP changes as well as temperature. The second is to incorporate uncertainty. This is done by using 5 sets of constraint levels, with a certain TWA analysis for each set. So uncertainty is not included in the differential equations which therefore correspond to a `learn-then-act' approach. Just as they suggest that CBA should only be seen as one of a range of tools (including TWA), so they also suggest that uncertainty models that extend CBA should be used alongside other tools. Bruckner et al. propose three ways in which the TWA approach could be extended to incorporate aspects of uncertainty. 1. Rather than use a `certainty equivalent' approach to CBA of replacing uncertain parameters by their most likely values, use regions of uncertainty or ``borders of ignorance''. Where parameter values in these regions may lead to catastrophic consequences, then safe or prudent strategies would avoid those regions. 2. Incorporate uncertainty into the solution method, with the hope that ``worst-case''\ situations could be determined within the model rather than at a prior stage. It is not made clear how this is to be done though. 3. Where reliable probability distributions exist then use these within the constraints defining the guardrails, so that strict constraints are replaced by a requirement that they are exceeded only with a set probability. However, it is not indicated how the level should be determined other than as a `normative', and possibly arbitrary, setting by policy makers under scientific advice. Kriegler and Bruckner (2002, 2004) undertake sensitivity analyses of the TWA approach. A simple 3 differential equation climate model is used. This relates temperature to carbon emissions. They state that the simplicity of the climate model is necessary in order to accommodate the complexities arising out of the sensitivity analysis. This sensitivity is then used to compare the Emissions Corridors, and consequently the admissible responses for two different groups of countries, those contained in Annex I and those not. The climate impact response for the Annex I countries compared to the global response, shows that there is a greater burden on emissions rights in those countries. It also shows that for high climate resilience the socioeconomic parameters make little difference but as resilience decreases so the starts to evolve a trade off between the two, keeping maximum admissable cumulative emissions approximately constant. Overall, Kriegler and Bruckner find that the climate (or resilience ) parameters determine the width of the corridor whilst the
economic parameters determine the slope of the corridor boundaries. The implication that they obtain from this is that it will be the economic parameters that determine the flexibility of emissions reduction and how gradual the emissions path has to be and where climate resilience is low, early policy will be needed.
The trans-disciplinary Tyndall Centre for Climate Change Research undertakes integrated research into the long-term consequences of climate change for society and into the development of sustainable responses that governments, business-leaders and decision-makers can evaluate and implement. Achieving these objectives brings together UK climate scientists, social scientists, engineers and economists in a unique collaborative research effort. Research at the Tyndall Centre is organised into four research themes that collectively contribute to all aspects of the climate change issue: Integrating Frameworks; Decarbonising Modern Societies; Adapting to Climate Change; and Sustaining the Coastal Zone. All thematic fields address a clear problem posed to society by climate change, and will generate results to guide the strategic development of climate change mitigation and adaptation policies at local, national and global scales. The Tyndall Centre is named after the 19th century UK scientist John Tyndall, who was the first to prove the Earth’s natural greenhouse effect and suggested that slight changes in atmospheric composition could bring about climate variations. In addition, he was committed to improving the quality of science education and knowledge. The Tyndall Centre is a partnership of the following institutions: University of East Anglia UMIST Southampton Oceanography Centre University of Southampton University of Cambridge Centre for Ecology and Hydrology SPRU – Science and Technology Policy Research (University of Sussex) Institute for Transport Studies (University of Leeds) Complex Systems Management Centre (Cranfield University) Energy Research Unit (CLRC Rutherford Appleton Laboratory) The Centre is core funded by the following organisations: Natural Environmental Research Council (NERC) Economic and Social Research Council (ESRC) Engineering and Physical Sciences Research Council (EPSRC) UK Government Department of Trade and Industry (DTI) For more information, visit the Tyndall Centre Web site (www.tyndall.ac.uk) or contact: External Communications Manager Tyndall Centre for Climate Change Research University of East Anglia, Norwich NR4 7TJ, UK Phone: +44 (0) 1603 59 3906; Fax: +44 (0) 1603 59 3901 Email: firstname.lastname@example.org
Tyndall Working Papers are available online at http://www.tyndall.ac.uk/publications/working_papers/working_papers.shtml Mitchell, T. and Hulme, M. (2000). A Country-byCountry Analysis of Past and Future Warming Rates, Tyndall Centre Working Paper 1. Hulme, M. (2001). Integrated Assessment Models, Tyndall Centre Working Paper 2. Berkhout, F, Hertin, J. and Jordan, A. J. (2001). Socio-economic futures in climate change impact assessment: using scenarios as 'learning machines', Tyndall Centre Working Paper 3. Barker, T. and Ekins, P. (2001). How High are the Costs of Kyoto for the US Economy?, Tyndall Centre Working Paper 4. Barnett, J. (2001). The issue of 'Adverse Effects and the Impacts of Response Measures' in the UNFCCC, Tyndall Centre Working Paper 5. Goodess, C.M., Hulme, M. and Osborn, T. (2001). The identification and evaluation of suitable scenario development methods for the estimation of future probabilities of extreme weather events, Tyndall Centre Working Paper 6. Barnett, J. (2001). Security and Climate Change, Tyndall Centre Working Paper 7. Adger, W. N. (2001). Social Capital and Climate Change, Tyndall Centre Working Paper 8. Barnett, J. and Adger, W. N. (2001). Climate Dangers and Atoll Countries, Tyndall Centre Working Paper 9. Gough, C., Taylor, I. and Shackley, S. (2001). Burying Carbon under the Sea: An Initial Exploration of Public Opinions, Tyndall Centre Working Paper 10. Barker, T. (2001). Representing the Integrated Assessment of Climate Change, Adaptation and Mitigation, Tyndall Centre Working Paper 11. Dessai, S., (2001). The climate regime from The Hague to Marrakech: Saving or sinking the Kyoto Protocol?, Tyndall Centre Working Paper 12. Dewick, P., Green K., Miozzo, M., (2002). Technological Change, Industry Structure and the Environment, Tyndall Centre Working Paper 13. Shackley, S. and Gough, C., (2002). The Use of Integrated Assessment: An Institutional Analysis Perspective, Tyndall Centre Working Paper 14. Köhler, J.H., (2002). Long run technical change in an energy-environment-economy (E3) model for an IA system: A model of Kondratiev waves, Tyndall Centre Working Paper 15. Adger, W.N., Huq, S., Brown, K., Conway, D. and Hulme, M. (2002). Adaptation to climate change: Setting the Agenda for Development Policy and Research, Tyndall Centre Working Paper 16. Dutton, G., (2002). Hydrogen Energy Technology, Tyndall Centre Working Paper 17. Watson, J. (2002). The development of large technical systems: implications for hydrogen, Tyndall Centre Working Paper 18. Pridmore, A. and Bristow, A., (2002). The role of hydrogen in powering road transport, Tyndall Centre Working Paper 19. Turnpenny, J. (2002). Reviewing organisational use of scenarios: Case study - evaluating UK energy policy options, Tyndall Centre Working Paper 20. Watson, W. J. (2002). Renewables and CHP Deployment in the UK to 2020, Tyndall Centre Working Paper 21. Watson, W.J., Hertin, J., Randall, T., Gough, C. (2002). Renewable Energy and Combined Heat and Power Resources in the UK, Tyndall Centre Working Paper 22. Paavola, J. and Adger, W.N. (2002). Justice and adaptation to climate change, Tyndall Centre Working Paper 23. Xueguang Wu, Jenkins, N. and Strbac, G. (2002). Impact of Integrating Renewables and CHP into the UK Transmission Network, Tyndall Centre Working Paper 24 Xueguang Wu, Mutale, J., Jenkins, N. and Strbac, G. (2003). An investigation of Network Splitting for Fault Level Reduction, Tyndall Centre Working Paper 25 Brooks, N. and Adger W.N. (2003). Country level risk measures of climate-related natural disasters and implications for adaptation to climate change, Tyndall Centre Working Paper 26 Tompkins, E.L. and Adger, W.N. (2003). Building resilience to climate change through adaptive management of natural resources, Tyndall Centre Working Paper 27
Dessai, S., Adger, W.N., Hulme, M., Köhler, J.H., Turnpenny, J. and Warren, R. (2003). Defining and experiencing dangerous climate change, Tyndall Centre Working Paper 28 Brown, K. and Corbera, E. (2003). A MultiCriteria Assessment Framework for CarbonMitigation Projects: Putting “development” in the centre of decision-making, Tyndall Centre Working Paper 29 Hulme, M. (2003). Abrupt climate change: can society cope?, Tyndall Centre Working Paper 30 Turnpenny, J., Haxeltine A. and O’Riordan, T. (2003). A scoping study of UK user needs for managing climate futures. Part 1 of the pilotphase interactive integrated assessment process (Aurion Project), Tyndall Centre Working Paper 31 Xueguang Wu, Jenkins, N. and Strbac, G. (2003). Integrating Renewables and CHP into the UK Electricity System: Investigation of the impact of network faults on the stability of large offshore wind farms, Tyndall Centre Working Paper 32 Pridmore, A., Bristow, A.L., May, A. D. and Tight, M.R. (2003). Climate Change, Impacts, Future Scenarios and the Role of Transport, Tyndall Centre Working Paper 33 Dessai, S., Hulme, M (2003). Does climate policy need probabilities?, Tyndall Centre Working Paper 34 Tompkins, E. L. and Hurlston, L. (2003). Report to the Cayman Islands’ Government. Adaptation lessons learned from responding to tropical cyclones by the Cayman Islands’ Government, 1988 – 2002, Tyndall Centre Working Paper 35 Kröger, K. Fergusson, M. and Skinner, I. (2003). Critical Issues in Decarbonising Transport: The Role of Technologies, Tyndall Centre Working Paper 36 Ingham, A. and Ulph, A. (2003) Uncertainty, Irreversibility, Precaution and the Social Cost of Carbon, Tyndall Centre Working Paper 37 Brooks, N. (2003). Vulnerability, risk and adaptation: a conceptual framework, Tyndall Centre Working Paper 38 Tompkins, E.L. and Adger, W.N. (2003). Defining response capacity to enhance climate change policy, Tyndall Centre Working Paper 39
Klein, R.J.T., Lisa Schipper, E. and Dessai, S. (2003), Integrating mitigation and adaptation into climate and development policy: three research questions, Tyndall Centre Working Paper 40 Watson, J. (2003), UK Electricity Scenarios for 2050, Tyndall Centre Working Paper 41 Kim, J. A. (2003), Sustainable Development and the CDM: A South African Case Study, Tyndall Centre Working Paper 42 Anderson, D. and Winne, S. (2003), Innovation and Threshold Effects in Technology Responses to Climate Change, Tyndall Centre Working Paper 43 Shackley, S., McLachlan, C. and Gough, C. (2004) The Public Perceptions of Carbon Capture and Storage, Tyndall Centre Working Paper 44 Purdy, R. and Macrory, R. (2004) Geological carbon sequestration: critical legal issues, Tyndall Centre Working Paper 45 Watson, J., Tetteh, A., Dutton, G., Bristow, A., Kelly, C., Page, M. and Pridmore, A., (2004) UK Hydrogen Futures to 2050, Tyndall Centre Working Paper 46 Berkhout, F., Hertin, J. and Gann, D. M., (2004) Learning to adapt: Organisational adaptation to climate change impacts, Tyndall Centre Working Paper 47 Pan, H. (2004) The evolution of economic structure under technological development, Tyndall Centre Working Paper 48 Awerbuch, S. (2004) Restructuring our electricity networks to promote decarbonisation, Tyndall Centre Working Paper 49 Powell, J.C., Peters, M.D., Ruddell, A. & Halliday, J. (2004) Fuel Cells for a Sustainable Future? Tyndall Centre Working Paper 50 Agnolucci, P., Barker, T. & Ekins, P. (2004) Hysteresis and energy demand: the Announcement Effects and the effects of the UK climate change levy, Tyndall Centre Working Paper 51 Agnolucci, P. (2004) Ex post evaluations of CO2 –Based Taxes: A Survey, Tyndall Centre Working Paper 52
Agnolucci, P. & Ekins, P. (2004) The Announcement Effect and environmental taxation, Tyndall Centre Working Paper 53 Turnpenny, J., Carney, S., Haxeltine, A., & O’Riordan, T. (2004) Developing regional and local scenarios for climate change mitigation and adaptation, Part 1: A framing of the East of England, Tyndall Centre Working Paper 54 Mitchell, T.D. Carter, T.R., Jones, .P.D, Hulme, M. and New, M. (2004) A comprehensive set of high-resolution grids of monthly climate for Europe and the globe: the observed record (1901-2000) and 16 scenarios (2001-2100), Tyndall Centre Working Paper 55 Vincent, K. (2004) Creating an index of social vulnerability to climate change for Africa, Tyndall Centre Working Paper 56 Shackley, S., Reiche, A. and Mander, S (2004) The Public Perceptions of Underground Coal Gasification (UCG): A Pilot Study, Tyndall Centre Working Paper 57 Bray, D and Shackley, S. (2004) The Social Simulation of The Public Perceptions of Weather Events and their Effect upon the Development of Belief in Anthropogenic Climate Change, Tyndall Centre Working Paper 58 Anderson, D and Winne, S. (2004) Modelling Innovation and Threshold Effects In Climate Change Mitigation, Tyndall Centre Working Paper 59 Few, R., Brown, K. and Tompkins, E.L. (2004) Scaling adaptation: climate change response and coastal management in the UK, Tyndall Centre Working Paper 60 Brooks, N. (2004) Drought in the African Sahel: Long term perspectives and future prospects, Tyndall Centre Working Paper 61 Barker, T. (2004) The transition to sustainability: a comparison of economics approaches, Tyndall Centre Working Paper 62 Few, R., Ahern, M., Matthies, F. and Kovats, S. (2004) Floods, health and climate change: a strategic review, Tyndall Centre Working Paper 63 Peters, M.D. and Powell, J.C. (2004) Fuel Cells for a Sustainable Future II, Tyndall Centre Working Paper 64
Adger, W. N., Brown, K. and Tompkins, E. L. (2004) The political economy of cross-scale networks in resource co-management, Tyndall Centre Working Paper 65 Turnpenny, J., Haxeltine, A., Lorenzoni, I., O’Riordan, T., and Jones, M., (2005) Mapping actors involved in climate change policy networks in the UK, Tyndall Centre Working Paper 66 Turnpenny, J., Haxeltine, A. and O’Riordan, T., (2005) Developing regional and local scenarios for climate change mitigation and adaptation: Part 2: Scenario creation, Tyndall Centre Working Paper 67 Bleda, M. and Shackley, S. (2005) The formation of belief in climate change in business organisations: a dynamic simulation model, Tyndall Centre Working Paper 68 Tompkins, E. L. and Hurlston, L. A. (2005) Natural hazards and climate change: what knowledge is transferable?, Tyndall Centre Working Paper 69 Abu-Sharkh, S., Li, R., Markvart, T., Ross, N., Wilson, P., Yao, R., Steemers, K., Kohler, J. and Arnold, R. (2005) Can Migrogrids Make a Major Contribution to UK Energy Supply?, Tyndall Centre Working Paper 70 Boyd, E. Gutierrez, M. and Chang, M. (2005) Adapting small-scale CDM sinks projects to low-income communities, Tyndall Centre Working Paper 71 Lowe, T., Brown, K., Suraje Dessai, S., Doria, M., Haynes, K. and Vincent., K (2005) Does tomorrow ever come? Disaster narrative and public perceptions of climate change, Tyndall Centre Working Paper 72 Walkden, M. (2005) Coastal process simulator scoping study, Tyndall Centre Working Paper 73 Ingham, I., Ma, J., and Ulph, A. M. (2005) How do the costs of adaptation affect optimal mitigation when there is uncertainty, irreversibility and learning?, Tyndall Centre Working Paper 74
This action might not be possible to undo. Are you sure you want to continue?
We've moved you to where you read on your other device.
Get the full title to continue reading from where you left off, or restart the preview.