Coordinating Lead Authors:

Kirsten Halsnæs (Denmark) [SRC:9], Priyadarshi Shukla (India) [SRC:4],

Lead Authors:

Dilip Ahuja (India), Grace Akumu (Kenya), Roger Beale (Australia), Jae Edmonds (USA) [SRC:5], Christian Gollier (Belgium) [SRC:5], Arnulf Grübler (Austria) [SRC:5], Minh Ha Duong (France), Anil Markandya (UK) [SRC:5], Mack McFarland (USA), Elena Nikitina (Russia), Taishi Sugiyama (Japan), Arturo Villavicencio (Equador), Ji Zou (PR China) [SRC:1],

Contributing Authors:

Terry Barker (UK) [SRC:2], Leon Clarke (USA) [SRC:1], Amit Garg (India) [SRC:1],

Review Editors:

Ismail Elgizouli (Sudan), Elizabeth Malone (USA),

This chapter should be cited as:

Halsnæs, K., P. Shukla, D. Ahuja, G. Akumu, R. Beale, J. Edmonds, C. Gollier, A. Grübler, M. Ha Duong, A. Markandya, M. McFarland, E. Nikitina, T. Sugiyama, A. Villavicencio, J. Zou, 2007: Framing issues. In Climate Change 2007: Mitigation. Contribution of Working Group III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change [B. Metz, O. R. Davidson, P. R. Bosch, R. Dave, L. A. Meyer (eds)], Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA

2.1.1 Introduction

This section introduces the relationship between sustainable development (SD) and climate change and presents a number of key concepts that can be used to frame studies of these relationships. Climate change and sustainable development are considered in several places throughout this report. Chapter 12 provides a general overview of the issues, while more specific issues relating to short- and long-term mitigation issues are addressed in Chapters 3 ( Section 3.1 ) and 11 ( Section 11.6 ). Sectoral issues are covered in Chapters 4 - 10 (Sections 4.5.4 , 5.5.5 , 6.9.2 , 7.7 , 8.4.5 , 9.7 , and in 10.6). Furthermore, the IPCC 2007b [NPR, 2007] ) addresses SD and climate change in Chapters 18 and 20.

2.1.2 Background

The IPCC’s Third Assessment Report (TAR; IPCC, 2001 [NPR] ) included considerations concerning SD and climate change. These issues were addressed particularly by Working Group II and III, as well as the Synthesis report. The TAR included a rather broad treatment of SD ( Metz et al., 2002 [JoC, ARC] ). The report noted three broad classes of analyses or perspectives: efficiency and cost-effectiveness, equity and sustainable development, and global sustainability and societal learning.

Since the TAR, literature on sustainable development and climate change has attempted to further develop approaches that can be used to assess specific development and climate policy options and choices in this context ( Beg et al., 2002 [JoC, ARC] ; Cohen et al., 1998 [JoC, ARC] ; Munasinghe and Swart, 2000 [NPR, ARC] ; Schneider, 2001 [PoC, JoC] ; Banuri et al., 2001 [NPR, ARC] ; Halsnæs and Verhagen, 2007 [SRC, 2007] ; Halsnæs, 2002 [SRC] ; Halsnæs and Shukla, 2007 [SRC, 2007] , Markandya and Halsnæs, 2002a [NPR, SRC] ; Metz et al. 2002 [JoC, ARC] ; Munasinghe and Swart, 2005 [NPR, ARC] ; Najam and Rahman, 2003 [JoC, ARC] ; Smit et al., 2001 [NPR] ; Swart et al.,. 2003 [JoC, ARC] ; Wilbanks, 2003 [JoC] ). These have included discussions about how distinctions can be made between natural processes and feedbacks, and human and social interactions that influence the natural systems and that can be influenced by policy choices ( Barker, 2003 [JoC, SRC] ). These choices include immediate and very specific climate policy responses as well as more general policies on development pathways and the capacity for climate change adaptation and mitigation. See also Chapter 12 of this report and Chapter 18 of IPCC 2007b [NPR, 2007] ) for a more extensive discussion of these issues.

Policies and institutions that focus on development also affect greenhouse gas (GHG) emissions and vulnerability. Moreover, these same policies and institutions constrain or facilitate mitigation and adaptation. These indirect effects can be positive or negative, and several studies have therefore suggested the integration of climate change adaptation and mitigation perspectives into development policies, since sustainable development requires coping with climate change and thereby will make development more sustainable ( Davidson et al., 2003 [JoC, SRC] ; Munasinghe and Swart, 2005 [NPR, ARC] ; Halsnæs and Shukla, 2007 [SRC, 2007] ).

Climate change adaptation and mitigation can also be the focus of policy interventions and SD can be considered as an issue that is indirectly influenced. Such climate policies can tend to focus on sectoral policies, projects and policy instruments, which meet the adaptation and mitigation goals, but are not necessarily strongly linked to all the economic, social, and environmental dimensions of sustainable development. In this case climate change policy implementation in practice can encounter some conflicts between general development goals and the goal of protecting the global environment. Furthermore, climate policies that do not take economic and social considerations into account might not be sustainable in the long run.

In conclusion, one might then distinguish between climate change policies that emerge as an integrated element of general sustainable development policies, and more specific adaptation and mitigation policies that are selected and assessed primarily in their capacity to address climate change. Examples of the first category of policies can be energy efficiency measures, energy access and affordability, water management systems, and food security options, while examples of more specific adaptation and mitigation policies can be flood control, climate information systems, and the introduction of carbon taxes. It is worth noticing that the impacts on sustainable development and climate change adaptation and mitigation of all these policy examples are very context specific, so it cannot in general be concluded whether a policy supports sustainable development and climate change jointly or if there are serious tradeoffs between economic and social perspectives and climate change (see also Chapter 12 of this report and Chapter 18 of IPCC 2007b [NPR, 2007] ) for a more extensive discussion).

2.1.3 The dual relationship between climate change and Sustainable Development

There is a dual relationship between sustainable development and climate change. On the one hand, climate change influences key natural and human living conditions and thereby also the basis for social and economic development, while on the other hand, society’s priorities on sustainable development influence both the GHG emissions that are causing climate change and the vulnerability.

Climate policies can be more effective when consistently embedded within broader strategies designed to make national and regional development paths more sustainable. This occurs because the impact of climate variability and change, climate policy responses, and associated socio-economic development will affect the ability of countries to achieve sustainable development goals. Conversely, the pursuit of those goals will in turn affect the opportunities for, and success of, climate policies.

Climate change impacts on development prospects have also been described in an interagency project on poverty and climate change as ‘Climate Change will compound existing poverty. Its adverse impacts will be most striking in the developing nations because of their dependence on natural resources, and their limited capacity to adapt to a changing climate. Within these countries, the poorest, who have the least resources and the least capacity to adapt, are the most vulnerable’ (African Development Bank et al., 2003 [NPR] ).

Recognizing the dual relationship between SD and climate change points to a need for the exploration of policies that jointly address SD and climate change. A number of international study programmes, including the Development and Climate project ( Halsnæs and Verhagen, 2007 [SRC, 2007] ), and an OECD development and environment directorate programme ( Beg et al., 2002 [JoC, ARC] ) explore the potential of SD-based climate change policies. Other activities include projects by the World Resources Institute ( Baumert et al., 2002 [NPR] ), and the PEW Centre ( Heller and Shukla, 2003 [NPR, SRC] ). Furthermore, the international literature also includes work by Cohen et al., 1998 [JoC, ARC] ; Banuri and Weyant, 2001 [NPR, ARC] ; Munasinghe and Swart 2000 [NPR, ARC] ; Metz et al., 2002 [JoC, ARC] ; Munasinghe and Swart, 2005 [NPR, ARC] ; Schneider et al., 2000 [PoC, JoC] ; Najam and Rahman, 2003 [JoC, ARC] ; Smit et al., 2001 [NPR] ; Swart et al., 2003 [JoC, ARC] ; and Wilbanks, 2003 [JoC] ).

2.1.4 The Sustainable Development concept

Sustainable development (SD) has been discussed extensively in the theoretical literature since the concept was adopted as an overarching goal of economic and social development by UN agencies, by the Agenda 21 nations, and by many local governments and private-sector actors. The SD literature largely emerged as a reaction to a growing interest in considering the interactions and potential conflicts between economic development and the environment. SD was defined by the World Commission on Environment and Development in the report Our Common Future as ‘development that meets the needs of the present without compromising the ability of future generations to meet their own needs’ ( WCED, 1987 [NPR, MoS] ).

The literature includes many alternative theoretical and applied definitions of sustainable development. The theoretical work spans hundreds of studies that are based on economic theory, complex systems approaches, ecological science and other approaches that derive conditions for how development paths can meet SD criteria. Furthermore, the SD literature emphasizes a number of key social justice issues including inter- and intra-generational equity. These issues are dealt with in Section 2.6 .

Since a comprehensive discussion of the theoretical literature on sustainable development is beyond the scope of this report, a pragmatic approach limits us to consider how development can be made more sustainable.

The debate on sustainability has generated a great deal of research and policy discussion on the meaning, measurability and feasibility of sustainable development. Despite the intrinsic ambiguity in the concept of sustainability, it is now perceived as an irreducible holistic concept where economic, social, and environmental issues are interdependent dimensions that must be approached within a unified framework ( Hardi and Barg, 1997 [NPR] ; Dresner, 2002 [NPR] ; Meadows, 1998 [NPR] ). However, the interpretation and valuation of these dimensions have given rise to a diversity of approaches.

A growing body of concepts and models, which explores reality from different angles and in a variety of contexts, has emerged in recent years in response to the inability of normal disciplinary science to deal with complexity and systems – the challenges of sustainability. The outlines of this new framework, known under the loose term of ‘Systems Thinking’, are, by their very nature, transdisciplinary and synthetic ( Kay and Foster, 1999 [NPR] ). An international group of ecologists, economists, social scientists and mathematicians has laid the principles and basis of an integrative theory of systems change ( (Holling 2001 ) ). This new theory is based on the idea that systems of nature and human systems, as well as combined human and nature systems and social-ecological systems, are interlinked in never-ending adaptive cycles of growth, accumulation, restructuring, and renewal within hierarchical structures ( Holling et al., 2002 [NPR] ).

A core element in the economic literature on SD is the focus on growth and the use of man-made, natural, and social capital. The fact that there are three different types of capital that can contribute to economic growth has led to a distinction between weak and strong sustainability, as discussed by Pearce and Turner 1990 [NPR] ( and Rennings and Wiggering 1997 ) ). Weak sustainability describes a situation where it is assumed that the total capital is maintained and that the three different elements of the capital stock can, to some extent, be used to substitute each other in a sustainable solution. On the other hand, strong sustainability requires each of the three types of capital to be maintained in its own right, at least at some minimum level. An example of an application of the strong sustainability concept is Herman Daly’s criteria, which state that renewable resources must be harvested at (or below) some predetermined stock level, and renewable substitutes must be developed to offset the use of exhaustible resources ( (Daly, 1990 ) ). Furthermore, pollution emissions should be limited to the assimilative capacity of the environment.

Arrow et al., 2004 [PoC, JoC] , in a joint authorship between leading economists and ecologists, present an approach for evaluating alternative criteria for consumption ^[1] , seen over time in a sustainable development perspective. Inter-temporal consumption and utility are introduced here as measurement points for sustainable development. One of the determinants of consumption and utility is the productive base of society, which consists of capital assets such as manufactured capital, human capital, and natural capital. The productive base also includes the knowledge base of society and institutions.

Although institutions are often understood as part of the capital assets, Arrow et al. 2004 [PoC, JoC] ) only consider institutions in their capacity as guiding the allocation of resources, including capital assets. Institutions in this context include the legal structure, formal and informal markets, various government agencies, inter-personal networks, and the rules and norms that guide their behaviour. Seen from an SD perspective, the issue is then: how, and to what extent, can policies and institutional frameworks for these influence the productive basis of society and thereby make development patterns more sustainable.

The literature includes other views of capital assets that will consider institutions and sustainable development policies as being part of the social capital element in society’s productive base. ( Lehtonen 2004 ) ) provides an overview of the discussion on social capital and other assets. He concludes that despite capabilities and social capital concepts not yet being at the practical application stage, the concepts can be used as useful metaphors, which can help to structure thoughts across different disciplines. Lehtonen refers to analysis of social-environmental dimensions by the OECD 1998 [NPR] ) that addresses aspects such as demography, health, employment, equity, information, training, and a number of governance issues, as an example of a pragmatic approach to including social elements in sustainability studies.

Arrow et al., 2004 [PoC, JoC] ) summarize the controversy between economists and ecologists by saying that ecologists have deemed current consumption patterns to be excessive or deficient in relation to sustainable development, while economists have focused more on the ability of the economy to maintain living standards. It is concluded here that the sustainability criterion implies that inter-temporal welfare should be optimized in order to ensure that current consumption is not excessive. ^[2] However, the optimal level of current consumption cannot be determined (i.e. due to various uncertainties). Theoretical considerations therefore focus instead on factors that make current consumption more or less sustainable. These factors include the relationship between market rates of return on investments and social discount rates, and the relationship between market prices of consumption goods (including capital goods) and the social costs of these commodities.

Some basic principles are therefore emerging from the international sustainability literature, which helps to establish commonly held principles of sustainable development. These include, for instance, the welfare of future generations, the maintenance of essential biophysical life support systems, more universal participation in development processes and decision-making, and the achievement of an acceptable standard of human well-being ( Swart et al., 2003 [JoC, ARC] ;( Meadowcroft, 1997; ) WCED, 1987 [NPR, MoS] ).

In the more specific context of climate change policies, the controversy between different sustainability approaches has shown up in relation to discussions on key vulnerabilities; see Section 2.5.2 for more details.

2.1.5 Development paradigms

Assessment of SD and climate change in the context of this report considers how current development can be made more sustainable. The focus is on how development goals, such as health, education, and energy, food, and water access can be achieved without compromising the global climate.

When applying such a pragmatic approach to the concept of SD it is important to recognize that major conceptual understandings and assumptions rely on the underlying development paradigms and analytical approaches that are used in studies. The understanding of development goals and the tradeoffs between different policy objectives depends on the development paradigm applied, and the following section will provide a number of examples on how policy recommendations about SD and climate change depend on alternative understandings of development as such.

A large number of the models that have been used for mitigation studies are applications of economic paradigms. Studies that are based on economic theory typically include a specification of a number of goals that are considered as important elements in welfare or human wellbeing. Some economic paradigms focus on the welfare function of the economy, assuming efficient resource allocation (such as in neoclassical economics), and do not consider deviations from this state and ways to overcome these. In terms of analyzing development and climate linkages, this approach will see climate change mitigation as an effort that adds a cost to the optimal economic state. ^[3] However, there is a very rich climate mitigation cost literature that concludes that market imperfections in practice often create a potential for mitigation policies that can help to increase the efficiency of energy markets and thereby generate indirect cost savings that can make mitigation policies economically attractive ( IPCC, 1996 [NPR] , Chapters 8 and 9; IPCC, 2001 [NPR] , Chapters 7 and 8). The character of such market imperfections is discussed further in Section 2.4 .

Other development paradigms based on institutional economics focus more on how markets and other information-sharing mechanisms establish a framework for economic interactions. Recent development research has included studies on the role of institutions as a critical component in an economy’s capacity to use resources optimally. Institutions are understood here in a broad sense, as being a core allocation mechanism and as the structure of society that organizes markets and other information sharing ( Peet and Hartwick, 1999 [NPR] ).

In this context, climate policy issues can include considerations about how climate change mitigation can be integrated into the institutional structure of an economy. More specifically, such studies can examine various market and non-market incentives for different actors to undertake mitigation policies and how institutional capacities for these policies can be strengthened. Furthermore, institutional policies in support of climate change mitigation can also be related to governance and political systems – see a more elaborate discussion in Chapter 12 , Section 12.2.3 .

Weak institutions have a lot of implications for the capacity to adapt or mitigate to climate change, as well as in relation to the implementation of development policies. A review of the social capital literature related to economic aspects and the implications for climate change mitigation policies concludes that, in most cases, successful implementation of GHG emission-reduction options will depend on additional measures to increase the potential market and the number of exchanges. This can involve strengthening the incentives for exchange (prices, capital markets, information efforts etc.), introducing new actors (institutional and human capacity efforts), and reducing the risks of participation (legal framework, information, general policy context of market regulation). All these measures depend on the nature of the formal institutions, the social groups of society, and the interactions between them ( Olhoff, 2002 [NPR] ). See also Chapter 12 of this report for a more extensive discussion of the political science and sociological literature in this area.

Key theoretical contributions to the economic growth and development debate also include work by A. Sen 1999 [NPR] ) and P. Dasgupta 1993 [NPR] ) concerning capabilities and human well-being. Dasgupta, in his inquiry into well-being and destitution, concludes that ‘our citizens’ achievements are the wrong things to look at. We should be looking at the extent to which they enjoy the freedom to achieve their ends, no matter what their ends turn out to be. The problem is that the extent of such freedoms depends upon the degree to which citizens make use of income and basic needs’. ( Dasgupta, 1993 [NPR] , pp. 54). Following this, Dasgupta recommends studying the distribution of resources, as opposed to outcomes (which, for example, can be measured in terms of welfare). The access to income and basic needs are seen as a fundamental basis for human well-being and these needs include education, food, energy, medical care etc. that individuals can use as inputs to meeting their individual desires. See also Section 2.6 , where the equity dimensions of basic needs and well-being approaches are discussed in more detail.

In the context of capabilities and human well-being, climate change policies can then include considerations regarding the extent to which these policies can support the access of individuals to specific resources as well as freedoms.

The capability approaches taken by Sen and Dasgupta have been extended by some authors from focusing on individuals to also covering societies ( Ballet et al., 2003 [NPR] ;( Lehtonen, 2004 ) ). It is argued here that, when designing policies, one needs to look at the effects of economic and environmental policies on the social dimension, including individualistic as well as social capabilities, and that these two elements are not always in harmony.

2.1.6 International frameworks for evaluating Sustainable Development and climate change links

Studies that assess the sustainable development impacts of climate change (and vice versa) when they are considering short to medium-term perspectives will be dealing with a number of key current development challenges. This section provides a short introduction to international policy initiatives and decisions that currently offer a framework for addressing development goals.

A key framework that can be used to organize the evaluation of SD and climate change linkages is the WEHAB ^[4] framework that was introduced by the World Summit on Sustainable Development in 2002 ( WSSD, 2002 [NPR] ). The WEHAB sectors reflect the areas selected by the parties at the WSSD meeting to emphasize that particular actions were needed in order to implement Agenda 21. Seen from a climate change policy evaluation perspective it would be relevant to add a few more sectors to the WEHAB group in order to facilitate a comprehensive coverage of major SD and climate change linkages. These sectors include human settlements tourism, industry, and transportation. It would also be relevant to consider demography, institutions and various cultural issues and values as cross-cutting sectoral issues.

Climate change policy aspects can also be linked to the Millennium Development Goals (MDG) that were adopted as major policy targets by the WSSD. The MDGs include nine general goals to eradicate poverty and hunger, health, education, natural resource utilization and preservation, and global partnerships that are formulated for the timeframe up to 2015 ( UNDP, 2003a [NPR] ).

A recent report by the CSD (Commission on Sustainable Development) includes a practical plan for how to achieve the Millennium Development Goals ( CSD, 2005 [NPR] ). Climate change is explicitly mentioned in the CSD report as a factor that could worsen the situation of the poor and make it more difficult to meet the MDGs. Furthermore, CSD 2005 [NPR] ) suggests adding a number of energy goals to the MDGs (i.e. to reflect energy security and the role that energy access can play in poverty alleviation). Adding energy as a separate component in the MDG framework will establish a stronger link between MDGs and climate change mitigation.

Several international studies and agency initiatives have assessed how the MDGs can be linked to goals for energy-, food-, and water access and to climate change impacts, vulnerability, and adaptation (African Development Bank et al., 2003 [NPR] ), and an example of how the link between climate change and MDGs can be further developed to include both adaptation and mitigation is shown in Table 2.1 . A linkage between MDGs and development goals is also described very specifically by Shukla 2003 [NPR, SRC] and Shukla et al. 2003 [NPR, SRC] ) in relation to the official Indian 10^th plan for 2002 – 2007 . In the same way, the Millenium Ecosystem Assessment (MEA) presents a global picture of the relationship between the net gains in human well-being and economic development based on a growing cost through degradation of ecosystem services, and demonstrates how this can pose a barrier to achieving the MDGs ( MEA, 2005 [NPR] ).

Measuring progress towards SD requires the development and systematic use of a robust set of indicators and measures. Agenda 21 ( 1992 ) explicitly recognizes in Chapter 40 that a pre-requisite for action is the collection of data at various levels (local, provincial, national and international), indicating the status and trends of the planet’s ecosystems, natural resources, pollution and socio-economy.

The OECD Ministerial Council decided in 2001 that the regular Economic Surveys of OECD countries should include an evaluation of SD dimensions, and a process for agreeing on SD indicators. These will be used in regular OECD peer reviews of government policies and performance. From the OECD menu of SD issues, the approach is to select a few areas that will be examined in depth, based on specific country relevance ( OECD, 2003 [NPR] ).

The first OECD evaluation of this kind was structured around three topics that member countries could select from the following list of seven policy areas ( OECD, 2004 [NPR] ):

• Improving environmental areas:

• - Reducing GHG emissions
• - Reducing air pollutants
• - Reducing water pollution
• - Moving towards sustainable use of renewable and non-renewable natural resources
• - Reducing and improving waste management

• Improving living standards in developing countries.
• Ensuring sustainable retirement income policies.

Most of the attention in the country choice was given to the environmental areas, while evaluation of improving living standards in developing countries was given relatively little attention in this first attempt.

The use of SD indicators for policy evaluations has been applied in technical studies of SD and climate change ( Munasinghe, 2002 [ARC] ; Atkinson et al., 1997 [NPR, ARC] ; Markandya et al., 2002 [NPR, SRC] ). These studies address SD dimensions based on a number of economic, environmental, human and social indicators, including both quantitative and qualitative measurement standards. A practical tool applied in several countries, called the Action Impact Matrix (AIM), has been used to identify, prioritize, and address climate and development synergies and tradeoffs ( Munasinghe and Swart, 2005 [NPR, ARC] ).

All together, it can be concluded that many international institutions and methodological frameworks offer approaches for measuring various SD dimensions, and that these have been related to broader development and economic policies by CSD, the WSSD, and the OECD. Many indexes and measurement approaches exist but, until now, relatively few studies have measured climate change in the context of these indexes. In this way, there is still a relatively weak link between actual measurements of and climate change links.

2.1.7 Implementation of Sustainable Development and climate change policies

SD and climate change are influenced by a number of key policy decisions related to economic, social and environmental issues, as well as by business-sector initiatives, private households and many other stakeholders, and these decisions are again framed by government policies, markets, information sharing, culture, and a number of other factors. Some of the decisions that are critically important in this context are investments, use of natural resources, energy consumption, land use, technology choice, and consumption and lifestyle, all of which can lead to both increasing and decreasing GHG emission intensities, which again will have implications for the scope of the mitigation challenge. Seen in a longer-term perspective these decisions are critical determinants for development pathways.

There has been an evolution in our understanding of how SD and climate change mitigation decisions are taken by societies. In particular, this includes a shift from governments that are defined by the nation/state to a more inclusive concept of governance, which recognizes various levels of government (global, transnational/regional, and local), as well as the roles of the private sector, non-governmental actors and civil society. Chapter 12 , Section 12.2.3 , includes a comprehensive assessment of how state, market, civil society and partnerships play a role in sustainable development and climate change policies.

2.2.1 The ‘public good’ character of climate change

Mitigation costs are exclusive to the extent that they may be borne by some individuals (nations) while others might evade them (free-riding) or might actually gain a trade/investment benefit from not acting (carbon leakage). The incentive to evade taking mitigation action increases with the substitutability of individual mitigation efforts and with the inequality of the distribution of net benefits. However, individual mitigation efforts (costs) decrease with efficient mitigation actions undertaken by others.

The unequal distribution of climate benefits from mitigation action, of the marginal costs of mitigation action and of the ability to pay emission reduction costs raises equity issues and increases the difficulty of securing agreement. In a strategic environment, leadership from a significant GHG emitter may provide an incentive for others to follow suit by lowering their costs ( Grasso, 2004 [NPR] ; ODS, 2002 [NPR] ).

Additional understandings come from political science, which emphasizes the importance of analyzing the full range of factors that have a bearing on decisions by nation states, including domestic pressures from the public and affected interest groups, the role of norms and the contribution of NGOs to the negotiation processes. Case studies of many MEAs (Multilateral Environmental Agreements) have provided insights, particularly on the institutional, cultural, political and historical dimensions that influence outcomes ( Cairncross, 2004 [NPR] ). A weakness of this approach is that the conclusions can differ depending on the choice of cases and the way in which the analysis is implemented. However, such ex-post analysis of the relevant policies often provides deep insights that are more accessible to policymakers, rather than theoretical thinking or numeric models.

2.2.2 Long time horizons

Climate policy raises questions of inter-generational equity and changing preferences, which inevitably affect the social weighting of environmental and economic outcomes, due to the long-term character of the impacts (for a survey see Bromley and Paavola, 2002 [NPR] ).

However, studies traditionally assume that preferences will be stable over the long time frames involved in the assessment of climate policy options. To the extent that no value is attached to the retention of future options, the preferences of the present generation are implicitly given priority in much of this analysis. As time passes, preferences will be influenced by information, education, social and organizational affiliation, income distribution and a number of cultural values ( Palacios-Huerta and Santos, 2002 [NPR] ). Institutional frameworks are likely to develop to assist groups, companies and individuals to form preferences in relation to climate change policy options. The institutions can include provision of information and general education programmes, research and assessments, and various frameworks that can facilitate collective decision-making that recognizes the common ‘global good’ character of climate change.

At an analytic level, the choice of discount rates can have a profound affect on valuation outcomes – this is an important issue in its own right and is discussed in Section 2.4.1 .

2.2.3 Irreversibility and the implications for decision-making

Human impacts on the climate system through greenhouse gas emissions may change the climate so much that it is impossible (or extremely difficult and costly) to return it to its original state – in this sense the changes are irreversible ( Scheffer et al., 2001 [JoC] ; Schneider, 2004 [PoC, JoC] ). Some irreversibility will almost certainly occur. For example, there is a quasi-certain irreversibility of a millennia time scale in the presence, in the atmosphere, of 22% of the emitted CO₂ ( Solomon et al., 2007 [NPR, PoC, 2007] ). However, the speed and nature of these changes, the tipping point at which change may accelerate and when environmentally, socially and economically significant effects become irreversible, and the cost and effectiveness of mitigation and adaptation responses are all uncertain, to a greater or lesser extent.

The combination of environmental irreversibility, together with these uncertainties ( Baker, 2005 [JoC] ; Narain et al., 2004 [NPR] ;( Webster, 2002; ) ( Epstein, 1980 ) ) means that decision-makers have to think carefully about:

a) The timing and sequencing of decisions to preserve options.

b) The opportunity to sequence decisions to allow for learning about climate science, technology development and social factors ( Baker, 2005 [JoC] ; Kansuntisukmongko, 2004 [NPR, MoS] ).

c) Whether the damage caused by increases in greenhouse concentrations in the atmosphere will increase proportionally and gradually or whether there is a risk of sudden, non-linear changes, and similarly whether the costs of reducing emissions change uniformly with time and the depth of reduction required, or are they possibly subject to thresholds or other non-linear effects.

d) Whether the irreversible damages are clustered in particular parts of the world or have a general effect, and

e) whether there is a potential that these irreversible damages will be catastrophically severe for some, many or even all communities ( Cline, 2005 [NPR] ).

Just as there are risks of irreversible climate changes, decisions to reduce GHG emissions can require actions that are essentially irreversible. For example, once made, these long-lived, large-scale investments in low-emission technologies are irreversible. If the assumptions about future policies and the directions of climate science on which these investments are made prove to be wrong, they would become ‘stranded’ assets. The risks (perceived by investors) associated with irreversibility of this nature further complicate decision-making on abatement action ( (Keller et al., 2004; ) ( Pindyck, 2002; ) Kolstad, 1996 [ARC] ; Sullivan et al., 2006 [NPR] ; Hamilton and Kenber, 2006 [NPR] ).

Without special actions by governments to overcome their natural inertia, economic and social systems might delay too long in reacting to climate risks, thus leading to irreversible climate changes. Ambitious climate-protection goals would require new investments (physical and intellectual) in climate-friendly technologies (efficiency improvements, renewables, nuclear power, carbon capture and storage), which are higher in cost than current technologies or otherwise divert scarce resources. From an economic point of view these investments are essentially irreversible. As the scale of the investment and the proportion of research and development costs increase, so the private economic risks associated with irreversibility also increase. Therefore, in the presence of uncertainty concerning future policy towards GHG emission reduction, future carbon prices or stabilization targets, investors are reluctant to undertake large-scale irreversible investments (sunk costs) without some form of upfront government support.

2.2.4 Risk of catastrophic or abrupt change

The possibility of abrupt climate change and/or abrupt changes in the earth system triggered by climate change, with potentially catastrophic consequences, cannot be ruled out ( Meehl et al., 2007 [NPR, MoS, 2007] ). Disintegration of the West Antarctic Ice Sheet (See Meehl et al., 2007 [NPR, MoS, 2007] ), if it occurred, could raise sea level by 4-6 metres over several centuries. A shutdown of the North Atlantic Thermohaline Circulation (See Meehl et al., 2007 [NPR, MoS, 2007] ) could have far-reaching, adverse ecological and agricultural consequences (See IPCC, 2007b [NPR, 2007] , Chapter 17), although some studies raise the possibility that the isolated, economic costs of this event might not be as high as assumed (See Meehl et al., 2007 [NPR, MoS, 2007] ). Increases in the frequency of droughts ( Salinger, 2005 [NPR, MoS] ) or a higher intensity of tropical cyclones (See Meehl et al., 2007 [NPR, MoS, 2007] ) could occur. Positive feedback from warming may cause the release of carbon or methane from the terrestrial biosphere and oceans (See Meehl et al., 2007 [NPR, MoS, 2007] ), which would add to the mitigation required.

Much conventional decision-making analysis is based on the assumption that it is possible to model and compare all the outcomes from the full range of alternative climate policies. It also assumes there is a smooth trade-off between the different dimensions of each policy outcome; that a probability distribution provides an expected value for each outcome, and that there is a unique best solution – the one with the highest expected value. Consequently, it could suggest that a policy which risked a catastrophically bad outcome with a very low probability might be valued higher than one which completely avoided the possibility of catastrophe and produced merely a bad outcome, but with a very high probability of occurrence.

Assumptions that it is always possible to ‘trade off’ more of one dimension (e.g. economic growth) for less of another (e.g. species protection) – that there is always a price at which we are comfortable to ‘dispense with’ a species in the wild (e.g. polar bears), an ecological community or indigenous cultures are problematic for many people. This also applies to assumptions that decision-makers value economic (and other) gains and losses symmetrically – that a dollar gained should always assumed to be valued equally to one that is lost, and that it is possible and appropriate to assume that the current generation’s preferences will remain stable over time.

Recent literature drawing on experimental economics and behavioural sciences suggests that these assumptions are an incomplete description of the way in which humans really make decisions. This literature suggests that preferences may be lexicographical (i.e. it is not possible to ‘trade off’ between different dimensions of alternative possible outcomes – there may be an aversion at any ‘price’ to losing particular species, ecosystems or communities), that attitudes to gains and losses might not be symmetrical (losses valued more highly than gains of an equivalent magnitude), and that low-probability extreme outcomes are overweighted when making choices (Tversky ( and Kahneman, 1992; ) ( Quiggin, 1982 ) ). This literature suggests that under these circumstances the conventional decision axiom of choosing the policy set that maximizes the expected (monetary) value of the outcomes might not be appropriate. Non-conventional decision criteria (e.g. avoiding policy sets which imply the possibility, even if at a very low probability, of specific unacceptable outcomes) might be required to make robust decisions ( Chichilnisky, 2000 [JoC] ; Lempert and Schlesinger, 2000 [JoC] ; Kriegler et al., 2006 [NPR] ).

No one analytic approach is optimal. Decision-making inevitably involves applying normative rules. Some normative rules are described in Section 2.2.7 and in Section 2.6 .

2.2.5 Sequential decision-making

Uncertainty is a steadfast companion when analyzing the climate system, assessing future GHG emissions or the severity of climate change impacts, evaluating these impacts over many generations or estimating mitigation costs. The typology of uncertainties is explored fully in Section 2.3 below. Uncertainties of differing types exist in key socio-economic factors and scientific phenomena.

The climate issue is a long-term problem requiring long-term solutions. Policymakers need to find ways to explore appropriate long-term objectives and to make judgments about how compatible short-term abatement options are with long-term objectives. There is an increased focus on non-conventional (robust) decision rules (see Section 2.2.7 below), which preserve future options by avoiding unacceptable risks.

Climate change decision-making is not a once-and-for-all event. Rather it is a process that will take place over decades and in many different geographic, institutional and political settings. Furthermore, it does not occur at discrete intervals but is driven by the pace of the scientific and political process. Some uncertainties will decrease with time – for example in relation to the effectiveness of mitigation actions and the availability of low-emission technologies, as well as with respect to the science itself. The likelihood that better information might improve the quality of decisions (the value of information) can support increased investment in knowledge accumulation and its application, as well as a more refined ordering of decisions through time. Learning is an integral part of the decision-making process. This is also referred to as ‘act then learn, then act again’ ( Manne and Richels, 1992 [NPR, ARC] ;( Valverde et al., 1999 ) ).

Uncertainties about climate policies at a decadal scale are a source of concern for many climate-relevant investments in the private sector (for example power generation), which have long expected economic lives.

It is important to recognize, however, that some level of uncertainty is unavoidable and that at times the acquisition of knowledge can increase, not decrease, uncertainty. Decisions will nevertheless have to be made.

2.2.6 Dealing with risks and uncertainty in decision-making

Given the multi-dimensionality of risk and uncertainty discussed in Section 2.3 , the governance of these deep uncertainties as suggested by Godard et al. 2002 [NPR] , p. 21) rests on three pillars: precaution, risk hedging, and crisis prevention and management.

The 1992 UNFCCC Article 3 (Principles) states that the Parties should take precautionary measures to anticipate, prevent or minimize the causes of climate change and mitigate its adverse effects. Where there are threats of serious or irreversible damage, lack of full scientific certainty should not be used as a reason for postponing such measures, taking into account that policies and measures to deal with climate change should be cost-effective in order to ensure global benefits at the lowest possible cost. ^[5]

While the precautionary principle appears in many other international treaties, from a scientific perspective the concept of precaution is subject to a plurality of interpretations. To frame the discussions on precaution, three key points should be considered first.

First, ‘precaution’ relates to decision-making in situations of deep uncertainty. It applies in the absence of sufficient data or conclusive or precise probabilistic descriptions of the risks ( Cheve and Congar, 2000 [NPR] ; Henry and Henry, 2002 [NPR] ) or in circumstances where the possibility of unforeseen contingencies or the possibility of irreversibility ( Gollier et al., 2000 [SRC] ) is suspected.

Second, in addition to that uncertainty/risk dimension, there is also a time dimension of precaution: the precautionary principle recognizes that policy action should not always wait for scientific certainty (see also the costs and decision-making sections of this chapter).

Third, the precautionary principle cuts both ways because in many cases, as Graham and Wiener 1995 [NPR] ) noted, environmental choices are trade-offs between one risk and another risk. For example, mitigating climate change may involve more extensive use of nuclear power. Goklany 2002 [NPR] ) has suggested a framework for decision-making under the precautionary principle that considers trade-offs between competing risks.

There is no single agreed definition of precautionary decision-making in the scientific literature.

The risk of catastrophes is commercially important, particularly for reinsurers that are large companies whose business is to sell insurance to other insurance companies (see IPCC, 2007b [NPR, 2007] , Chapter 7, Box 7.2). In the context of globalization and consolidation, many reinsurers are actively developing new instruments to trade some of their risk on the deeper financial markets. These instruments include options, swaps and catastrophe bonds.

At the same time, governments are also developing new kinds of public-private partnership to cope with market failures, uncertainties and really big cataclysms. On a global scale, it can be argued that the best form of insurance is to increase the systemic resilience of the human society through scientific research, technical, economic and social development. This requires the broad participation of society in order to succeed.

Mills 2005 [JoC] ) concludes that the future role of insurance in helping society to cope with climate change is uncertain. Insurers may rise to the occasion and become more proactive players in improving the science and crafting responses, or they may retreat from oncoming risks, thereby shifting a greater burden to governments and individuals.

2.2.7 Decision support tools

Decisions concerning the appropriate responses to climate risks require insights into a variety of possible futures over short to very long time frames and into linkages between biophysical and human systems, as well as ethical alternatives. Structured analysis – both numerical and case-based – can ‘aid understanding by managing and analyzing information and alternatives’ (Arrow et al., 1996a [NotFound] , referenced in ( Bell et al., 2001 ) ). Integrated Assessment Models (IAMs) in particular have improved greatly in terms of the richness with which they represent the biophysical, social and economic systems and the feedbacks between them. They have increasingly explored a variety of decision rules or other means of testing alternative policies. Without structured analysis it is extremely difficult, if not impossible, to understand the possible effects of alternative policy choices that face decision-makers. Structured analysis can assist choices of preferred policies within interests (for example at the national level) as well as negotiating outcomes between interests (by making regional costs and benefits clearer).

The use of projections and scenarios is one way to develop understanding about choices in the context of unpredictability. These are discussed in detail in Chapter 3 .

A large number of analytical approaches can be used as a support to decision-making. IPCC 2001 [NPR] ) Chapter 10, provides an extensive overview of decision-making approaches and reviews their applicability at geopolitical levels and in climate policy domains. The review includes decision analysis, cost-benefit analysis, cost-effectiveness analysis, tolerable windows/safe-landing/guard-rail approaches, game theory, portfolio theory, public finance theory, ethical and cultural prescriptive rules, and various policy dialogue exercises. Integrated assessment, multi-attribute analysis and green accounting approaches are also commonly used decision support tools in climate change debates.

A major distinction between cost benefit-analysis, cost-effectiveness analysis, and multi-attribute analysis and different applications of these relates to the extent in which monetary values are used to represent the impacts considered. Cost-benefit analysis aims to assign monetary values to the full range of costs and benefits. This involves at least two important assumptions – that it is possible to ‘trade off’ or compensate between impacts on different values in a way that can be expressed in monetary values, and that it is possible to ascertain estimates of these ‘compensation’ values for non-market impacts, such as air pollution, health and biodiversity. By definition, the benefits and costs of climate change policies involve many of such issues, so climate change economic analysis embodies a lot of complicated valuation issues. Section 2.4 goes more into depth about approaches that can be used to value non-markets impacts and the question of discounting.

In multi-attribute analysis, instead of using values derived from markets or from non-market valuation techniques, different dimensions (impacts) are assigned weights – through a stakeholder consultation process, by engaging a panel of experts or by the analyst making explicit decisions. This approach can use quantitative data, qualitative information or a mixture of both. Developing an overall score or ranking for each option allows alternative policies to be assessed, even under conditions of weak comparability. Different functional forms can be used for the aggregation process.

Policy optimization models aim to support the selection of policy/decision strategies and can be divided into a number of types:

• Cost-benefit approaches, which try to balance the costs and benefits of climate policies (including making allowances for uncertainties).
• Target-based approaches, which optimize policy responses, given targets for emission or climate change impacts (again in some instances explicitly acknowledging uncertainties).
• Approaches, which incorporate decision strategies (such as sequential act-learn-act decision-making, hedging strategies etc.) for dealing with uncertainty (often embedded in cost-benefit frameworks).

Another approach is to start with a policy or policies and evaluate the implications of their application. Policy evaluation approaches include:

• Deterministic projection approaches, in which each input and output takes on a single value.
• A stochastic projection approach, in which at least some inputs and outputs take on a range of value.
• Exploratory modelling.
• Public participation processes, such as citizens juries, consultation, and polling.

IAMs aim to combine key elements of biophysical and economic systems into a decision-making framework with various levels of detail on the different sub-components and systems. These models include all different variations on the extent to use monetary values, the integration of uncertainty, and on the formulation of the policy problem with regard to optimization, policy evaluation and stochastic projections. Current integrated assessment research uses one or more of the following methods ( Rotmans and Dowlatabadi, 1998 [NPR, MoS] ):

• Computer-aided IAMs to analyze the behavior of complex systems
• Simulation gaming in which complex systems are represented by simpler ones with relevant behavioral similarity.
• Scenarios as tools to explore a variety of possible images of the future.
• Qualitative integrated assessments based on a limited, heterogeneous data set, without using any model.

A difficulty with large, global models or frameworks is that it is not easy to reflect regional impacts, or equity considerations between regions or stakeholder groups. This is particularly true of ‘global’ cost-benefit approaches, where it is particularly difficult to estimate a marginal benefit curve, as regional differences are likely to be considerable. Such approaches have difficulty in assisting decision-making where there are many decision-makers and multiple interests and values to be taken into account.

Variants of the safe landing/tolerable windows/guard rails approach emphasize the role of regional/national decision-makers by providing them the opportunity to nominate perceived unacceptable impacts of climate change (for their region or globally), and the limit to tolerable socio-economic costs of mitigation measures they would be prepared to accept to avoid that damage (e.g. Toth 2004 [NPR, ARC] ). Modelling efforts (in an integrated assessment model linking climate and economic variables, and with explicit assumptions about burden sharing through emissions allocations and trading) are then directed at identifying the sets of feasible mitigation paths – known as ‘emissions corridors’ – consistent with these constraints. To the extent that there is some overlap between the acceptable ‘emissions corridors’, the conditions for agreement on mitigation action do exist.

Green accounting attempts to integrate a broader set of social welfare measures into macro-economic studies. These measures can be related to a broad set of social, environmental, and development-oriented policy aspects. The approach has most commonly been used in order to integrate environmental impacts, such as local air pollution, GHG emissions, waste generation, and other polluting substances, into macro-economic studies. Green accounting approaches include both monetary valuation approaches that attempt to calculate a ‘green national product’ (where the economic values of pollutants are subtracted from the national product), and accounting systems that include quantitative non-monetary pollution data.

Halsnæs and Markandya 2002 [NPR, SRC] ) recognize that decision analysis methods exhibit a number of commonalities in assumptions. The standard approach goes through the selection of GHG emission-reduction options, selection of impact areas that are influenced by policies as for example costs, local air pollution, employment, GHG emissions, and health, definition of baseline case, assessment of the impacts of implementing the GHG emission-reduction policies under consideration, and application of a valuation framework that can be used to compare different policy impacts.

Sociological analysis includes the understanding of how society operates in terms of beliefs, values, attitudes, behaviour, social norms, social structure, regarding climate change. This analysis includes both quantitative and qualitative approaches, such as general surveys, statistics analysis, focus groups, public participation processes, media content analysis, Delphi etc.

All analytical approaches (explicitly or implicitly) have to consider the described elements, whether this is done in order to collect quantitative information that is used in formalized approaches or to provide qualitative information and focus for policy dialogues. Different decision-making approaches will often involve very similar technical analysis in relation to several elements. For example, multi-criteria-analysis, as well as cost-benefit analysis (as, for example, applied in integrated assessment optimization modelling frameworks) and green accounting may use similar inputs and analysis for many model components, but critically diverge when it comes to determining the valuation approach applied to the assessment of multiple policy impacts.

2.3.1 How are risk and uncertainty communicated in this report?

Communicating about risk and uncertainty is difficult because uncertainty is multi-dimensional and there are different practical and philosophical approaches to it. In this report, ‘risk’ is understood to mean the ‘combination of the probability of an event and its consequences’, as defined in the risk management standard ISO/IEC Guide 73 ( 2002 ). This definition allows a variety of ways of combining probabilities and consequences, one of which is expected loss, defined as the ‘product of probability and loss’. The fundamental distinction between ‘risk’ and ‘uncertainty’ is as introduced by economist Frank Knight 1921 [NPR] ), that risk refers to cases for which the probability of outcomes can be ascertained through well-established theories with reliable complete data, while uncertainty refers to situations in which the appropriate data might be fragmentary or unavailable.

Dealing effectively with the communication of risk and uncertainty is an important goal for the scientific assessment of long-term environmental policies. In IPCC assessment reports, an explicit effort is made to enhance consistency in the treatment of uncertainties through a report-wide coordination effort to harmonize the concepts and vocabulary used. The Third Assessment Report common guidelines to describe levels of confidence were elaborated by Moss and Schneider 2000 [NPR, PoC] ). The actual application of this framework differed across the three IPCC working groups and across chapters within the groups. It led to consistent treatment of uncertainties within Working Group I (focusing on uncertainties and probabilities, see Sommerville et al., 2007, Section 1.6) and Working Group II (focusing on risks and confidence levels, see IPCC, 2007b [NPR, 2007] , Section 1.1), although consistency across these groups was not achieved. The authors of Working Group III did not systematically apply the guidelines.

The most important insight arising from an interdisciplinary assessment of uncertainty is its conceptual diversity. There is no linear scale going from ‘perfect knowledge’ to ‘total uncertainty’. The literature suggests a ‘pedigree’ approach for characterizing the quality of information (for example the NUSAP approach by van der Sluijs et al., 2003 [NPR, MoS] ). This involves examining at least the amount and reliability of evidence ^[6] supporting the information and the level of agreement of the information sources.

The degree of consensus among the available studies is a critical parameter for the quality of information. The level of agreement regarding the benefits and drawbacks of a certain technology describes the extent to which the sources of information point in the same direction. Table 2.2 ’s vocabulary is used to qualify IPCC findings along these two dimensions. Because mitigation mostly involves the future of technical and social systems, Table 2.2 is used here to qualify the robustness of findings, and more precise expressions regarding quantified likelihood or levels of confidence are used only when there is high agreement and much evidence, such as converging results from a number of controlled field experiments.

Where findings depend on the future of a dynamic system, it is important to consider the possibility of extreme or/and irreversible outcomes, the potential for resolution (or persistence) of uncertainties in time, and the human dimensions. Rare events with extreme and/or irreversible outcomes are difficult or impossible to assess with ordinary statistics, but receive special attention in the literature.

Box 2.1 Risk and uncertainty vocabulary used in this report

Uncertainty cannot always be quantified, and thus the vocabulary displayed in Table 2.2 is used to qualitatively describe the degree of scientific understanding behind a finding or about an issue. See text for discussion of Table 2.2 ’s dimensions, the amount of evidence and the level of agreement.

Table 2.2: Qualitative definition of uncertainty

Source: IPCC Guidance Notes on risk and uncertainty ( 2005 ).

2.3.2 Typologies of risk and uncertainty

The literature on risk and uncertainty offers many typologies, often comprising the following classes:

Randomness: risk often refers to situations where there is a well-founded probability distribution in typologies of uncertainty. For example, assuming an unchanged climate, the potential annual supply of wind, sun or hydropower in a given area is only known statistically. In situations of randomness, expected utility maximization is a standard decision-making framework.

Possibility: the degree of ‘not-implausibility’ of a future can be defined rigorously using the notion of acceptable odds, see de Finetti 1937 [NPR] ( and Shackle 1949 ) ). While it is scientifically controversial to assign a precise probability distribution to a variable in the far distant future determined by social choices such as the global temperature in 2100, some outcomes are not as plausible as others (see the controversy on scenarios in Box 2.2). There are few possibility models related to environmental or energy economics.

Knightian or Deep Uncertainty: the seminal work by Knight 1921 [NPR] ) describes a class of situations where the list of outcomes is known, but the probabilities are imprecise. Under deep uncertainty, reporting a range of plausible values allows decision-makers to apply their own views on precaution. Two families of criteria have been proposed for decision-making in this situation. One family associates a real-valued generalized expected utility to each choice (see Ellesberg, 2001 [NPR] ), while the other discards the completeness axiom on the grounds that under deep uncertainty alternative choices may sometimes be incomparable (see ( Bewley, 2002; ) Walley, 1991 [NPR] ). Results of climate policy analysis under deep uncertainty with imprecise probabilities ( Kriegler, 2005 [NPR] ; Kriegler et al. 2006 [NPR] ) are consistent with the previous findings using classical models.

Structural uncertainty: is characterized by « unknown unknowns ». No model (or discourse) can include all variables and relationships. In energy-economics models, for example, there can easily be structural uncertainty regarding the treatment of the informal sector, market efficiency, or the choice between a Keynesian or a neoclassical view of macro-economic dynamics. Structural uncertainty is attenuated when convergent results are obtained from a variety of different models using different methods, and also when results rely more on direct observations (data) rather than on calculations.

Fuzzyness or vagueness: describes the nature of things that do not fall sharply into one category or another, such as the meaning of ‘sustainable development’ or ‘mitigation costs’. One way to communicate the fuzzyness of the variables determining the ‘Reasons for concern’ about climate change is to use smooth gradients of colours, varying continuously from green to red (see IPCC, 2001a [NPR] , Figure SPM 2, also known as the ‘burning embers’ diagram). Fuzzy modelling has rarely been used in the climate change mitigation literature so far.

Uncertainty is not only caused by missing information about the state of the world, but also by human volition: global environmental protection is the outcome of social interactions. Not mentioning taboos, psychological and social aspects, these include:

Surprise: which means a discrepancy between a stimulus and pre-established knowledge ( Kagan, 2002 [NPR] ). Complex systems, both natural and human, exhibit behaviour that was not imagined by observers until it actually happened. By allowing decision-makers to become familiar (in advance) with a number of diverse but plausible futures, scenarios are one way of reducing surprises.

Metaphysical: describes things that are not assigned a truth level because it is generally agreed that they cannot be verified, such as the mysteries of faith, personal tastes or belief systems. Such issues are represented in models by critical parameters, such as discount rates or risk-aversion coefficients. While these parameters cannot be judged to be true or false they can have a bearing on both behaviour and environmental policy-making. Thompson and Raynor ( 1998 ) argue that, rather than being obstacles to be overcome, the uneasy coexistence of different conceptions of natural vulnerability and societal fairness is a source of resilience and the key to the institutional plurality that actually enables us to apprehend and adapt to our ever-changing circumstances.

Strategic uncertainty: involves the fact that information is a strategic tool for rational agents. The response to climate change requires coordination at international and national level. Strategic uncertainty is usually formalized with game theory, assuming that one party in a transaction has more (or better) information than the other. The informed party may thus be able to extract a rent from this advantage. Information asymmetry is an important issue for the regulation of firms by governments

and for international agreements. Both adverse selection and moral hazards are key factors in designing efficient mechanisms to mitigate climate change.

2.3.3 Costs, benefits and uncertainties

In spite of scientific progress, there is still much uncertainty about future climate change and its mitigation costs. Given observed risk attitudes, the desirability of preventive efforts should be measured not only by the reduction in the expected (average) damages, but also by the value of the reduced risks and uncertainties that such efforts yield. The difficulty is how to value the societal benefits included in these risk reductions. Uncertainty concerning mitigation costs adds an additional level of difficulty in determining the optimal risk-prevention strategies, since the difference between two independent uncertain quantities is relatively more uncertain than related to the individual.

How can we decide whether a risk is acceptable to society? Cost-benefit analysis alone cannot represent all aspects of climate change policy evaluation, and Section 2.2 on Decision-making discusses a variety of tools. In the private sector, another practical way to deal with these risks has been to pay attention to the Value-At-Risk (VAR): in addition to using the mean and the variance of the outcome, a norm is set on the most unfavourable percentile (usually 0.05) of the distribution of outcomes at a given future date.

However, in the language of cost-benefit analysis, an acceptable risk means that its benefits to society exceed its costs. The standard rule used by public and private decision-makers in a wide variety of fields (from road safety to long-term investments in the energy sector) is that a risk will be acceptable if the expected net present value is positive. Arrow and Lind 1970 [NotFound] ) justify this criterion when the policy’s benefits and costs have known probabilities, and when agents can diversify their own risk through insurance and other markets. For most of the economic analysis of climate change, these assumptions are disputable, and have been discussed in the economic literature.

First, risks associated with climate change cannot easily be diversified using insurance and financial instruments. Atmospheric events are faced by everyone at the same time in the same region. This reduces the potential benefit of any mutual risk-sharing agreement. A solution would be to share risks internationally, but this is difficult to implement, and its efficiency depends upon the correlation of the regional damages. Inability to diversify risks, combined with the risk aversion observed in most public and private decision-makers, implies that there is an additional benefit to preventive efforts coming from the reduced variability of future damages. If these monetized damages are expressed as a percentage of GDP, the marginal benefit of prevention can be estimated as the marginal expected increase in GDP, with some adjustments for the marginal reduction in the variance of damages.

Second, in most instances, objective probabilities are difficult to estimate. Furthermore, a number of climate change impacts involve health, biodiversity, and future generations, and the value of changes in these assets is difficult to capture fully in estimates of economic costs and benefits (see Section 2.4 on costs). Where we cannot measure risks and consequences precisely, we cannot simply maximize net benefits mechanically. This does not mean that we should abandon the usefulness of cost-benefit analysis, but it should be used as an input, among others in climate change policy decisions. The literature on how to account for ambiguity in the total economic value is growing, even if there is no agreed standard.

Finally, Gollier 2001 [SRC] ) suggests that a sophisticated interpretation of the Precautionary Principle is compatible with economic principles in general, and with cost-benefit analyses in particular. The timing of the decision process and the resolution of the uncertainty should be taken into account, in particular when waiting before implementing a preventive action as an option. Waiting, and thereby late reactions, yield a cost when risks happen to be worse than initially expected, but yield an option value and cost savings in cases where risks happen to be smaller than expected. Standard dynamic programming methods can be used to estimate these option values.

2.4.1 Definitions

Mitigation costs can be measured at project, technology, sector, and macro-economic levels, and various geographical boundaries can be applied to the costing studies (see a definition of geographical boundaries in Section 2.8 ).

The project, technology, sector, and macro-economic levels can be defined as follows:

• Project: A project-level analysis considers a ‘stand-alone’ activity that is assumed not to have significant indirect economic impacts on markets and prices (both demand and supply) beyond the activity itself. The activity can be the implementation of specific technical facilities, infrastructure, demand-side regulations, information efforts, technical standards, etc. Methodological frameworks to assess the project-level impacts include cost-benefit analysis, cost-effectiveness analysis, and lifecycle analysis.
• Technology: A technology-level analysis considers a specific GHG mitigation technology, usually with several applications in different projects and sectors. The literature on technologies covers their technical characteristics, especially evidence on learning curves as the technology diffuses and matures. The technology analysis can use analytical approaches that are similar to project-level analysis.
• Sector: Sector-level analysis considers sectoral policies in a ‘partial-equilibrium’ context, for which other sectors and macro-economic variables are assumed to be given. The policies can include economic instruments related to prices, taxes, trade, and financing, specific large-scale investment projects, and demand-side regulation efforts. Methodological frameworks for sectoral assessments include various partial equilibrium models and technical simulation models for the energy sector, agriculture, forestry, and the transportation sector.
• Macro-economic: A macro-economic analysis considers the impacts of policies across all sectors and markets. The policies include all sorts of economic policies, such as taxes, subsidies, monetary policies, specific investment programmes, and technology and innovation policies. Methodological frameworks include various macro-economic models, such as general equilibrium models, Keynesian econometric models, and Integrated Assessment Models (IAMs), among others.

In comparing project, technology, sector, and macro-economic cost estimates it is important to bear in mind that cost estimates based on applying taxes in a macro-economic model are not comparable with abatement costs calculated at other assessment levels. This, for example, is because a carbon tax will apply to all GHG emissions, while abatement costs at project, technology or sector level will only reflect the costs of emission reductions.

Private and social costs: Costs can be measured from a private as well as from a social perspective. Individual decision-makers (including both private companies and households) are influenced by various cost elements, such as the costs of input to a production process, labour and land costs, financial interest rates, equipment costs, fuel costs, consumer prices etc., which are key private cost components. However, the activities of individuals may also cause externalities, for example emissions that influence the utility of other individuals, but which are not taken into consideration by the individuals causing them. A social cost perspective includes the value of these externalities.

External costs: These typically arise when markets fail to provide a link between the person who creates the ‘externality’ and the person who is affected by it, or more generally when property rights for the relevant resources are not well defined. ^[7] In the case of GHG emissions, those who will eventually suffer from the impacts of climate change do not have a well-defined ‘property right’ in terms of a given climate or an atmosphere with given GHG concentrations, so market forces and/or bargaining arrangements cannot work directly as a means to balance the costs and benefits of GHG emissions and climate change. However, the failure to take into account external costs, in cases like climate change, may be due not only to the lack of property rights, but also the lack of full information and non-zero transaction costs related to policy implementation.

Private, financial, and social costs are estimated on the basis of different prices. The private cost component is generally based on market prices that face individuals. Thus, if a project involves an investment of US$ 5 million, as estimated by the inputs of land, materials, labour and equipment, that figure is used as the private cost. That may not be the full cost, however, as far as the estimation of social cost is concerned, because markets can be distorted by regulations and other policies as well as by limited competition that prevent prices from reflecting real resource scarcities. If, for example, the labour input is being paid more than its value in alternative employment, the private cost is higher than the social cost. Conversely, if market prices of polluting fuels do not include values that reflect the environmental costs, these prices will be lower than the social cost. Social costs should be based on market prices, but with eventual adjustments of these with shadow prices, to bring them into line with opportunity costs.

In conclusion, the key cost concepts are defined as follows:

• Private costs are the costs facing individual decision-makers based on actual market prices.
• Social costs are the private costs plus the costs of externalities. The prices are derived from market prices, where opportunity costs are taken into account.

Other cost concepts that are commonly used in the literature are ‘financial costs’ and ‘economic costs’. Financial costs, in line with private costs, are derived on the basis of market prices that face individuals. Financial costs are typically used to assess the costs of financing specific investment projects. Economic costs, like social costs, assess the costs based on market prices adjusted with opportunity costs. Different from social costs, by definition they do not take all externalities into account.

2.4.2 Major cost determinants

A number of factors are critically important when determining costs, and it is important to understand their character and role when comparing mitigation costs across different studies, as occurs in Chapters 3 - 11 of this report, which compares costs across different models and which are based on different approaches.

The critical cost factors are based on different theoretical and methodological paradigms, as well as on specific applications of approaches. This section considers a number of factors including discounting, market efficiency assumptions, the treatment of externalities, valuation issues and techniques related to climate change damages ^[8] and other policy impacts, as well as implementation and transactions costs, and gives guidance on how to understand and assess these aspects within the context of climate change mitigation costing studies. For a more in-depth review of these issues see IPCC, 2001 [NPR] , Chapters 7 and 8.

2.4.3 Mitigation potentials and related costs

Chapters 3 - 11 report the costs of climate change mitigation at global, regional, sectoral, and technology level and, in order to ensure consistency and transparency across the cost estimates reported in these chapters, it has been agreed to use a number of key concepts and definitions that are outlined in this section. Furthermore, the following paragraphs also outline how the concepts relate to mitigation cost concepts that have been used in previous IPCC reports, in order to allow different cost estimates to be compared and eventual differences to be understood.

A commonly used output format for climate change mitigation cost studies means reporting the GHG emission reduction in quantitative terms that can be achieved at a given cost. The potential terminology is often used in a very ‘loose’ way, which makes it difficult to compare numbers across studies. The aim of the following is to overcome such lack of transparency in cost results based on a definition of major cost and GHG emission reduction variables to be used when estimating potentials.

The term ‘potential’ is used to report the quantity of GHG mitigation compared with a baseline or reference case that can be achieved by a mitigation option with a given cost (per tonne) of carbon avoided over a given period. The measure is usually expressed as million tonnes carbon- or CO₂-equivalent of avoided emissions, compared with baseline emissions. The given cost per tonne (or ‘unit cost’) is usually within a range of monetary values at a particular location (e.g. for wind-generated electricity), such as costs less than x US$ per tonne of CO₂- or carbon-equivalent reduction (US$/tC-eq). The monetary values can be defined as private or social unit costs: private unit costs are based on market prices, while social unit costs reflect market prices, but also take externalities associated with the mitigation into consideration. The prices are real prices adjusted for inflation rates.

2.5.1 Integrating mitigation and adaptation in a development context – adaptive and mitigative capacities

The TAR ( IPCC, 2001 [NPR] ) introduced a new set of discussions about the institutional and developmental context of climate change mitigation and adaptation policies. One of the conclusions from that discussion was that the capacity for implementing specific mitigation and adaptation policies depends on man-made and natural capital and on institutions. Broadly speaking, institutions should be understood here as including markets and other information-sharing mechanisms, legal frameworks, as well as formal and informal networks.

Subsequent work by Adger 2001a [NPR] ) further emphasizes the role of social capital in adaptation. Adger refers to a definition by ( Woolcock and Narayan 2000, ) p. 226), which states that social capital is made up of ‘the norms and networks that enable people to act collectively’. According to Adger there are two different views within the main areas of the international literature that are important to climate change issues namely: 1) whether social capital only exists outside the state, and 2) whether social capital is a cause, or simply a symptom, of a progressive and perhaps flexible and adaptive society. The first issue relates to how important planned adaptation and government initiatives can be, and the second considers the macro-level functioning of society and the implications for adaptive capacity.

Adger observes that the role that social capital, networks and state-civil society linkages play in adaptive capacity can be observed in historical and present-day contexts by examining the institutions of resource management and collective action in climate-sensitive sectors and social groups, highlighting a number of such experiences in adaptation to climate change. The examples include an assessment of the importance of social contacts and socio-economic status in relation to excess mortality due to extreme heating, coastal defence in the UK, and coastal protection in Vietnam, where the adaptive capacity in different areas is assessed within the context of resource availability and the entitlements of individuals and groups ( Kelly and Adger, 1999 [NPR] ). A literature assessment ( IPCC, 2007b [NPR, 2007] , Chapter 20) includes a wider range of examples of historical studies of development patterns, thus confirming that social capital has played a key role in economic growth and stability.

IPCC 2001 [NPR] ), Chapter 1 initiated a very preliminary discussion about the concept of mitigative capacity. Mitigative capacity (in this context) is seen as a critical component of a country’s ability to respond to the mitigation challenge, and the capacity, as in the case of adaptation, largely reflects man-made and natural capital and institutions. It is concluded that development, equity and sustainability objectives, as well as past and future development trajectories, play critical roles in determining the capacity for specific mitigation options. Following that, it can be expected that policies designed to pursue development, equity and/or sustainability objectives might be very benign framework conditions for implementing cost-effective climate change mitigation policies. The final conclusion is that, due to the inherent uncertainties involved in climate change policies, enhancing mitigative capacity can be a policy objective in itself.

It is important to recognize here that the institutional aspects of the adaptive and mitigative capacities refer to a number of elements that have a ‘public-good character’ as well as general social resources. These elements will be common framework conditions for implementing a broad range of policies, including climate change and more general development issues. This means that the basis for a nation’s policy-implementing capacity exhibits many similarities across different sectors, and that capacity-enhancing efforts in this area will have many joint benefits.

There may be major differences in the character of the adaptive and mitigative capacity in relation to sectoral focus and to the range of technical options and policy instruments that apply to adaptation and mitigation respectively. Furthermore, assessing the efficiency and implementability of specific policy options depends on local institutions, including markets and human and social capital, where it can be expected that some main strengths and weaknesses will be similar for different sectors of an economy.

As previously mentioned, the responses to climate change depend on the adaptive and mitigative capacities and on the specific mitigation and adaptation policies adopted. Policies that enhance adaptive and mitigative capacities can include a wide range of general development policies, such as market reforms, education and training, improving governance, health services, infrastructure investments etc.

The actual outcome of implementing specific mitigation and adaptation policies is influenced by the adaptive and mitigative capacity, and the outcome of adaptation and mitigation policies also depends on a number of key characteristics of the socio-economic system, such as economic growth patterns, technology, population, governance, and environmental policies.

It is expected that there may be numerous synergies and tradeoffs between the adaptive and mitigative capacity elements of the socio-economic and natural systems, as well as between specific adaptation and mitigation policies. Building more motorways, for example, can generate more traffic and more GHG emissions. However, the motorways can also improve market access, make agriculture less vulnerable to climate change, help in evacuation prior to big storms, and can support general economic growth (and thereby investments in new efficient production technologies). Similarly, increased fertiliser use in agriculture can increase productivity and reduce climate change vulnerability, but it can also influence the potential for carbon sequestration and can increase GHG emissions.

2.5.2 Mitigation, adaptation and climate change impacts

The discussion on mitigation and adaptation policy portfolios has a global as well as a national/regional dimension. It should be recognized that mitigation and adaptation are very different regarding time frame and distribution of benefits. Dang et al. 2003 [JoC, ARC] , Table 1) highlights a number of important commonalities and differences between mitigation and adaptation policies. Both policy areas can be related to sustainable development goals, but differ according to the direct benefits that are global and long term for mitigation, while being local and shorter term for adaptation. Furthermore adaptation can be both reactive (to experienced climate change) and proactive, while mitigation can only be proactive in relation to benefits from avoided climate change occurring over centuries. Dang et al. 2003 [JoC, ARC] , Table 4) also points out that there can be conflicts between adaptation and mitigation in relation to the implementation of specific national policy options. For example, installing air-conditioning systems in buildings is an adaptation option, but energy requirements can increase GHG emissions, and thus climate change.

In relation to the trade-off between mitigation and adaptation, Schneider 2004 [PoC, JoC] ) points out that when long-term integrated assessment studies are used to assess the net benefits of avoided climate change (including adaptation options) versus the costs of GHG emission reduction measures, the full range of possible climate outcomes, including impacts that remain highly uncertain such as surprises and other climate irreversibility, should be included. Without taking these uncertain events into consideration, decision-makers will tend to be more willing to accept prospective future risks rather than attempt to avoid them through abatement. It is worth noting here that,when faced with the risk of a major damage, human beings may make their judgment based on the consequences of the damage rather than on probabilities of events. Schneider concludes that it is not clear that climate surprises have a low probability, they are just very uncertain at present, and he suggests taking these uncertainties into consideration in integrated assessment models, by adjusting the climate change damage estimates. The adjustments suggested include using historical data for estimating the losses of extreme events, valuing ecosystem services, subjective probability assessments of monetary damage estimates, and the use of a discount rate that decreases over time in order to give high values to future generations.

In this way the issues of jointly targeting mitigation and adaptation has an element of decision-making under uncertainty, due to the complexity of the environmental and human systems and their interactions. ( Kuntz-Duriseti 2004 ) ) suggests dealing with this uncertainty by combining economic analysis and precautionary principles, including an insurance premium system, hedging strategies, and inclusion of low-probability events in risk assessments.

A common approach of many regional and national developing country studies on mitigation and adaptation policies has been to focus on the assessment of context-specific vulnerabilities to climate change. Given this, a number of studies and national capacity-building efforts have considered how adaptation and mitigation policies can be integrated into national development and environmental policies, and how they can be supported by financial transfers, domestic funds, and linked to foreign direct investments ( IINC, 2004 [NPR] ; CINC, 2004 [NPR] ). The Danish Climate and Development Action Program aims at a two-leg strategy, where climate impacts, vulnerabilities, and adaptation are assessed as an integral part of development plans and actions in Danish partner countries, and where GHG emission impacts and mitigation options are considered as part of policy implementation (Danida, 2005 ).

Burton et al. 2002 [JoC] ) suggest that research on adaptation should focus on assessing the social and economic determinants of vulnerability in a development context. The focus of the vulnerability assessment according to this framework should be on short-term impacts, i.e. should try to assess recent and future climate variability and extremes, economic and non-economic damages and the distribution of these. Based on this, adaptation policies should be addressed as a coping strategy against vulnerability and potential barriers, obstacles, and the role of various stakeholders and the public sector should be considered. Kelly and Adger 2000 [JoC] ) developed an approach for assessing vulnerabilities and concluded that the vulnerability and security of any group is determined by resource availability and entitlements. The approach is applied to impacts from tropical storms in coastal areas inVietnam.

On a global scale, there is a growing recognition of the significant role that developing countries play in determining the success of global climate change policies, including mitigation and adaptation policy options ( Müller, 2002 [NPR] ). Many governments of developing countries have started to realize that they should no longer discuss whether to implement any measures against climate change, but how drastic these measures should be, and how climate policies can be an integral part of national sustainable development paths ( SAINC, 2003 [NPR] ; IINC, 2004 [NPR] ; BINC, 2004 [NPR] ; CINC, 2004 [NPR] ; MOST, 2004 [NPR] ).

2.6.1 Development opportunities and equity

Traditionally, success in development has been measured in economic terms – increase in Gross National Income (GNI) per capita remains the most common measure ^[18] . Likewise, income distribution has been one of the key components in equity, both within and between countries, and has been measured in terms of inequalities of income, through measures such as the ‘GINI’ coefficient. ^{[19] [20]} Although a great deal has been written in recent years on the components of well-being, the development literature has been slow to adopt a broader set of indicators of this concept, especially as far as equity in well-being is concerned, despite the fact that some authors have argued that absolute changes in income and other indicators of human well-being (e.g. education, mortality rates, water, sanitation etc.) are just as important as the distribution within these indicators ( Maddison, 2003 [NPR] ; Goklany, 2001 [NPR] ).

Probably the most important and forceful critic of the traditional indicators has been Sen 1992 [NPR] , 1999 ). Sen’s vision of development encompasses not only economic goods and services but also individuals’ health and life expectancy, their education and access to public goods, the economic and social security that they enjoy, and their freedom to participate freely in economic interchange and social decision-making. While his criticism is widely acknowledged as addressing important shortcomings in the traditional literature, the ideas still have not been made fully operational. Sen speaks of ‘substantive freedoms’ and ‘capabilities’ rather than goods and services as the key goals of development and provides compelling examples of how his concepts can paint a different picture of progress in development compared to that of changes in GNI. It remains the case, however, that actual indicators of equity still do not cover the breadth of components identified by Sen.

The UNDP Human Development Index (HDI) is an important attempt to widen the indicators of development, and initially included per capita national income, life expectancy at birth and the literacy rate. However, it is important to recognize that no single all-encompassing indicator can be constructed, will be understandable or useful to either policymakers or the public, so different indexes have to be used that reflect different issues and purposes.

Rather than synthesizing these three components into a single index, as the HDI has done, we can also look at changes in the inter-country equity of the individual components. Table 2.3 ^[21] provides data for the period 1980 – 2001 for per capita national income (GNI) and life expectancy at birth (LE) and from 1990 to 2001 for the literacy rate (ILL). The increase in average GNI has been much faster over this period than those of life expectancy and literacy rates. The increase in coefficient of variations for GNI per capita (by 6%) and life expectancy (by 14%) therefore show an increase in dispersion over this period, indicating a wider disparity of these parameters across countries. However, literacy rates have become more equal, with a decline in the coefficient of variation by 22% (see Table 2.3 ). However, a study by Goklany 2002 [NPR] ) concluded that inequality between countries does not necessarily translate into inequality between individuals.

Table 2.3: Measures of Inter-country Equity

As Sen notes, the problem of inequality becomes magnified when attention is shifted from income inequality to inequality of ‘substantive freedoms and capabilities’, as a result of a ‘coupling’ of the different dimensions – individuals who are likely to suffer from higher mortality and who are illiterate are also likely to have lower incomes and a lower ability to convert incomes into capabilities and good standards of living. While this is certainly true at the individual level, at the country level the correlation appears to be declining.

This wider analysis of equity has important implications for sharing the costs of mitigation and for assessing the impacts of climate change (see Chapter 1 for a more detailed discussion of climate change impacts and the reference to the UNFCCC Article 2). As generally known, the impacts of climate change are distributed very unequally across the planet, hurting the vulnerable and poor countries of the tropics much more than the richer countries in the temperate regions. Moreover, these impacts do not work exclusively, or even mainly, through changes in real incomes. The well-being of future generations will be affected through the effects of climate change on health, economic insecurity and other factors. As far as the costs of actions to reduce GHGs are concerned, measures that may be the least costly in overall terms are often not the ones that are the most equitable – see Sections 2.6.4 and 2.6.5 for a further discussion of the links between mitigation policy and equity.

2.6.2 Uncertainty as a frame for distributional and equity aspects

Gollier, 2001 [SRC] outlines a framework for assessing the equity implications of climate change uncertainty, where he considers risk aversion for different income groups. The proposition (generally supported by empirical evidence) is that the relative risk aversion of individuals decreases with increasing wealth ( Gollier, 2001 [SRC] ), implying that the compensation that an individual asks for in order to accept a risk decreases relative to his income with increasing income. However, the absolute risk aversion – or the total compensation required in order to accept a risk – increases with wealth. It means that a given absolute risk level is considered to be more important to poorer people than to richer, and the comparatively higher risk aversion of poorer people suggests that larger investments in climate change mitigation and adaptation policies are preferred if these risks are borne by the poor rather than the rich.

A similar argument can be applied in relation to the equity consequences of increased climate variability and extreme events. Climate change may increase the possibility of large, abrupt and unwelcome regional or global climatic events. A coping strategy against variability and extreme events can be income-smoothing measures, where individuals even out their income over time through savings and investments. Poorer people with a lower propensity to save, and with less access to credit makers, have smaller possibilities to cope with climate variability and extreme events through such income-smoothing measures, and they will therefore be more vulnerable.

2.6.3 Alternative approaches to social justice

Widening our understanding of equity does not provide us with a rule for ranking different outcomes, except to say that, other things being equal, a less inequitable outcome is preferable to a more inequitable one. But how should one measure outcomes in terms of equity and what do we do when other things are not equal?

The traditional economic approach to resource allocation has been based on utilitarianism, in which a policy is considered to be desirable if no other policy or action is feasible that yields a higher aggregate utility for society. This requires three underlying assumptions:

(a) All choices are judged in terms of their consequences, and not in terms of the actions they entail.

(b) These choices are valued in terms of the utility they generate to individuals and no attention is paid to the implications of the choices for aspects such as rights, duties etc.

(c) The individual utilities are added up to give the sum of utility for society as a whole.

In this way the social welfare evaluation relies on the assumption that there is a net social surplus if the winners can compensate the losers and still be better off themselves. It should be recognized here that philosophers dispute that efficiency is a form of equity.

This approach has been the backbone of welfare economics, including the use of cost-benefit analysis (CBA) as a tool for selecting between options. Under CBA all benefits are added up, as are the costs, and the net benefit – the difference between the benefits and costs – is calculated. The option with the highest net benefit is considered the most desirable. ^[22] If utilities were proportional to money benefits and ‘disutilities’ were proportional to money costs, this method would amount to choosing to maximize utilities. Since most economists accept that this proportionality does not hold, they extend the CBA by either (a) asking the decision-maker to take account of the distributional implications of the option as a separate factor, in addition to the calculated net benefit; or (b) weighting costs or benefits by a factor that reflects the relationship between utility and the income of the person receiving that cost or benefit. For details of these methods in the context of climate change, see Markandya and Halsnaes 2002b [NPR, SRC] ). ^[23]

An alternative approach to allocating resources, which is derived from an ethical perspective and has existed for at least as long as the utilitarian approach described above (which has its modern origins in the late 18^th century by Jeremy Bentham), is based on the view that social actions are to be judged by whether or not they conform to a ‘social contract’ that defines the rights and duties of individuals in society. The view was inspired by the work of Kant and Hegel and finds its greater articulation in the writing of Rousseau and the French 19^th century philosophers. ^[24] In this position, for example, a society may predetermine that an individual has the right to be protected from serious negative health damage as a result of social actions. Hence no action, even if it increased utility, could be tolerated if it violated the rights and duties of individuals.

Modern philosophers who have developed the ‘rights’ view include Rawls, who argued that it is not utilities that matter but the distribution of ‘primary goods, which include, in addition to income, “rights, liberties and opportunities and… the social basis of self respect”’ ( Rawls, 1971 [NPR] ). Rawls argued further that social justice demanded that society be judged in terms of the level of well-being of its worst-off member. At the other end of the political spectrum, Nozick and the modern libertarians contend that personal liberties and property rights have (with very few exceptions) absolute precedence over objectives such as the reduction of poverty and deprivation ( Nozick, 1974 [NPR] ).

More recently, however, some ethical philosophers have found fault with both the ‘modified’ utilitarian view and the rights-based approach, on a number of grounds. Sen, for example, has argued that options should be judged not only in terms of their consequences, but also in terms of procedures. He advocates a focus on the capabilities of individuals to choose a life that one has reason to value. A person’s capability refers to the alternative combinations of ‘functionings’, where functionings can be more popularly described as ‘lifestyles’ ( Sen, 1999 [NPR] , pp. 74-75). What matters are not only the realized functionings, but also the capability set of alternatives, differently from a utilitarian-based approach that focuses only on the outcomes. In particular, the freedom to make the choices and engage in social and market transactions is worth something in its own right.

Sen criticizes the ‘rights-based’ equity approaches for not taking into consideration the fact that individuals are different and the actual consequences of giving them specific rights will vary between individuals, so rights should be seen in the context of capabilities. Both apply, because individuals have different preferences and thereby value primary inputs, for example, differently, and because their capability to use different rights also differ. Along these lines, Sen further argues that his capability-based approach can facilitate easier inter-personal comparisons than utilitarianism, since it does not suggest aggregating all individuals, but rather presenting information both on the capability sets available to individuals and their actual achievements.

What implications does this debate have in the context of climate change? One is that rights and capabilities need to be viewed in an international context. An example of an approach based on global equity would be to entitle every individual alive at a given date an equal per capita share in the intrinsic capacity of the earth to absorb GHGs. Countries whose total emissions exceeded this aggregate value would then compensate those below the value. In accordance with a utilitarian approach this compensation would be based on an estimate of the aggregate economic welfare lost by countries due to climate change, seen in relation to their own emissions. In contrast, the capability-based approach would argue for reduced capabilities associated with climate change.

As suggested above, societies do not (in practice) follow a strict utilitarian view of social justice and they do indeed recognize that citizens have certain basic rights in terms of housing, medical care etc. Equally, they do not subscribe to a clear ‘rights’ view of social justice either. Social choices are then a compromise between a utilitarian solution that focuses on consequences and one that recognizes basic rights in a more fundamental way. Much of the political and philosophical debate is about which rights are valid in this context – a debate that shows little sign of resolution. For climate change there are many options that need to be evaluated, in terms of their consequences for the lives of individuals who will be impacted by them. It is perfectly reasonable for the policymakers to exclude those that would result in major social disruptions, or large number of deaths, without recourse to a CBA. Equally, choices that avoid such negative consequences can be regarded as essential, even if the case for them cannot be made on CBA grounds. Details of where such rules should apply and where choices can be left to the more conventional CBA have yet to be worked out, and this remains an urgent part of the agenda for climate change studies.

As an alternative to social-justice-based equity methods, eco-centric approaches assign intrinsic value to nature as such ( Botzler and Armstrong, 1998 [NPR] ). This value can be specified in terms of diversity, avoided damages, harmony, stability, and beauty, and these values should be respected by human beings in their interaction with nature. In relation to climate change policies the issue here becomes one of specifying the value of nature such that it can be addressed as specific constraints that are to be respected beyond what is reflected in estimates of costs and benefits and other social impacts.

2.6.4 Equity consequences of different policy instruments

All sorts of climate change policies related to vulnerabilities, adaptation, and mitigation will have impacts on intra- and inter-generational equity. These equity impacts apply at the global, international, regional, national and sub-national levels.

Article 3 of the UNFCCC 1992 [NPR] , sometimes referred to as ‘the equity article’) states that Parties should protect the climate system on the basis of equity and in accordance with their common but differentiated responsibilities and respective capabilities. Accordingly, the developed country Parties should take the lead in combating climate change and the adverse effects thereof. Numerous approaches exist in the climate change discourse on how these principles can be implemented. Some of these have been presented to policymakers (both formally and informally) and have been subject to rigorous analysis by academics, civil society and policymakers over long periods of time.

The equity debate has major implications for how different stakeholders judge different instruments for reducing greenhouse gases (GHG) and for adapting to the inevitable impacts of climate change.

With respect to the measures for reducing GHGs, the central equity question has focused on how the burden should be shared across countries ( Markandya and Halsnaes, 2002b [NPR, SRC] ; Agarwal and Narain, 1991 [NPR] ; Baer and Templet, 2001 [NPR] ; Shukla, 2005 [SRC] ). On a utilitarian basis, assuming declining marginal utility, the case for the richer countries undertaking more of the burden is strong – they are the ones to whom the opportunity cost of such actions would have less welfare implications. However, assuming constant marginal utility, one could come to the conclusion that the costs of climate change mitigation that richer countries will face are very large compared with the benefits of the avoided climate change damages in poorer countries. In this way, utilitarian-based approaches can lead to different conclusions, depending on how welfare losses experienced by poorer people are represented in the social welfare function.

Using a ‘rights’ basis it would be difficult to make the case for the poorer countries to bear a significant share of the burden of climate change mitigation costs. Formal property rights for GHG emissions allowances are not defined, but based on justice arguments equal allocation to all human beings has been proposed. This would give more emissions rights to developing countries – more than the level of GHGs they currently emit. Hence such a rights-based allocation would impose more significant costs on the industrialized countries, although now, as emissions in the developing world increased, they too, at some point in time, would have to undertake some emissions reductions.

The literature includes a number of comparative studies on equity outcomes of different international climate change agreements. Some of these studies consider equity in terms of the consequences of different climate change policies, while others address equity in relation to rights that nations or individuals should enjoy in relation to GHG emission and the global atmosphere.

Equity concerns have also been addressed in a more pragmatic way as a necessary element in international agreements in order to facilitate consensus. Müller 2001 [JoC] ) discusses fairness of emission allocations and that of the burden distribution that takes all climate impacts and reduction costs into consideration and concludes that there is no solution that can be considered as the right and fair one far out in the future. The issue is rather to agree on an acceptable ‘fairness harmonization procedure’, where an emission allocation is initially chosen and compensation payments are negotiated once the costs and benefits actually occur.

Rose et al. 1998 [SRC] ) provide reasons why equity considerations are particularly important in relation to climate change agreements. First, country contributions will depend on voluntary compliance and it must therefore be expected that countries will react according to what they consider to be fair, ^[25] which will be influenced by their understanding of equity. Second, appealing to global economic efficiency is not enough to get countries together, due to the large disparities in current welfare and in welfare changes implied by efficient climate policies.

Studies that focus on the net costs of climate change mitigation versus the benefits of avoided climate change give a major emphasis to the economic consequences of the policies, while libertarian-oriented equity studies focus on emission rights, rights of the global atmosphere, basic human living conditions etc. ( (Wesley and Peterson, 1999 ) ). Studies that focus on the net policy costs will tend to address equity in terms of a total outcome of policies, while the libertarian studies focus more on initial equity conditions that should be applied to ex ante emission allocation rules, without explicitly taken equity consequences into consideration.

Given the uncertainties inherent in climate change impacts and their economic and social implications, it is difficult to conduct comprehensive and reliable consequence studies that can be used for an ex ante determination of equity principles for climate change agreements. Furthermore, social welfare functions and other value functions, when applied to the assessment of the costs and benefits of global climate change policies, run into a number of crucial equity questions. These include issues that are related to the asymmetry between the concentration of major GHG emission sources in industrialized countries and the relatively large expected damages in developing countries, the treatment of individuals with different income levels in the social welfare function, and a number of inter-generational issues.

Rights-based approaches have been extensively used as a basis for suggestions on structuring international climate change agreements around emission allocation rules or compensation mechanisms. Various allocation rules have been examined, including emissions per capita principles, emissions per GDP, grandfathering, liability-based compensation for climate change damages etc. These different allocation rules have been supported with different arguments and with reference to equity principles. An overview and assessment of the various rights-based equity principles and their consequences on emission allocations and costs are included in Rose et al. 1998 [SRC] ), Valliancourt ( and Waaub 2004 ) Leimbach 2003 [MoS] ( Tol and Verheyen 2004 ) ( and Panayotou et al. 2002 ) ).

While there is consensus in the literature about how rules should be assessed in relation to specific moral criteria, there is much less agreement on what criteria should apply (e.g. should they be based on libertarian or egalitarian rights-based approaches, or on utilitarian approaches).

A particular difficulty in establishing international agreements on emission allocation rules is that the application of equity in this ex ante way can imply the very large transfer of wealth across nations or other legal entities that are assigned emission quotas, at a time where abatement costs, as well as climate change impacts, are relatively uncertain ( Halsnæs and Olhoff, 2005 [SRC] ). These uncertainties make it difficult for different parties to assess the consequences of accepting given emission allocation rules and to balance emission allocations against climate damages suffered in different parts of the world ( (Panayotou et al., 2002 ) ).

Practical discussions about equity questions in international climate change negotiations have reflected, to a large extent, specific interests of various stakeholders, more than principal moral questions or considerations about the vulnerability of poorer countries. Arguments concerning property rights, for example, have been used by energy-intensive industries to advocate emission allocations based on grandfathering principles that will give high permits to their own stakeholders (that are large past emitters), and population-rich countries have, in some cases, advocated that fair emission allocation rules imply equal per capita emissions, which will give them high emission quotas.

( Vaillancourt and Waaub 2004 ) ) suggest designing emission allocation criteria on the basis of the involvement of different decision-makers in selecting and weighing equity principles for emission allocations, and using these as inputs to a multi-criteria approach. The criteria include population basis, basic needs, polluter pays, GDP intensity, efficiency and geographical issues, without a specified structure on inter-relationships between the different areas. In this way, the approach primarily facilitates the involvement of stakeholders in discussions about equity.

2.6.5 Economic efficiency and eventual trade-offs with equity

For more than a decade the literature has covered studies that review the economic efficiency of climate change mitigation policies and, to some extent, also discuss different emission allocation rules and the derived equity consequences ( IPCC, 1996 [NPR] , Chapter 11; IPCC, 2001 [NPR] , Chapters 6 and 8). Given that markets for GHG emission permits work well in terms of competition, transparency and low transaction costs, trade-offs between economic efficiency and equity (resulting from the distribution of emission rights) do not need to occur. In this ideal case, equity and economic efficiency can be addressed separately, where equity is taken care of in the design of emission allocation rules, and economic efficiency is promoted by the market system.

In practice, however, emission markets do not live up to these ideal conditions and the allocation of emission permits, both in international and domestic settings, will have an influence on the structure and functioning of emission markets, so trade-offs between what seems to be equitable emission allocations and economic efficiency can often occur ( Shukla, 2005 [SRC] ). Some of the issues that have been raised in relation to the facilitation of equity concerns through initial emission permit allocations include the large differences in emission permits and related market power that different countries would have ( Halsnæs and Olhoff, 2005 [SRC] ).

2.7.1 Technology and climate change

Recognizing the importance of technology over the long-term introduces an important element of uncertainty into the climate change debate, as direction and pace of future technological change cannot be predicted. Technological innovation and deployment are responsive to climate policy signals, for example in form of carbon taxes, although the extent and rate of this response can be as uncertain as the timing and magnitude of the policy signal. Reducing such uncertainties, for instance through long-term, predictable policy frameworks and signals, are therefore important. The usual approach consists of formulating alternative scenarios of plausible future developments. These, however, are constrained by inherent biases in technology assessment and uncertainties concerning the response of technological change to climate policy. There is also widespread recognition in the literature that it is highly unlikely that a single ‘silver bullet’ technology exists that can solve the climate problem, so the issue is not one of identifying singular technologies, but rather ensembles, or portfolios of technologies. This applies to both mitigation and adaptation technologies. These technologies have inter-dependencies and cross-enhancement (‘spillover’) potentials, which adds another important element of uncertainty into the analysis. Despite these problems of uncertainty and ignorance, insights are available from multiple fields.

Extensive literature surveys on the importance of technological change on the extent of possible climate change and on feasibility and costs of climate policies are provided by Clarke and Weyant 2002 [NPR, SRC] Grubb et al. 2002 [MoS, ARC] Grübler et al. 1999 [SRC] Jaffe et al. 2003 [NPR] and Löschel 2002 [MoS] ) among others. Quantitative illustrations have been published in a number of important scenario studies including the IPCC SAR ( IPCC, 1996 SRES [NPR] SRES (IPCC, 2000 [NPR] ), the scenarios of the World Energy Council WEC Nakicenovic et al., 1998a [NPR, SRC] ) as well as from climate policy model inter-comparison projects such as EMF-19 (Energy Modelling Forum) ( Weyant, 2004b [ARC] ), the EU-based Innovation Modeling Comparison Project (IMCP) ( Edenhofer et al., 2006 [NPR, MoS, ARC] ) and the multi-model calculations of climate ‘stabilization’ scenarios summarized in the TAR ( IPCC, 2001 [NPR] ). In a new development since the TAR, technology has also moved to the forefront of a number of international and national climate policy initiatives, including the Global Energy Technology Strategy ( GTSP, 2001 [NPR] ), the Japanese ‘New Earth 21’ Project ( RITE, 2003 [NPR] ), the US 21 Technology Roadmap ( NETL, 2004 [NPR] ), or the European Union’s World Energy Technology Outlook ( WETO, 2003 [NPR] ).

The subsequent review first discusses the importance of technological change in ‘no-climate policy’ (or so-called ‘reference’ or ‘baseline’) scenarios, and hence the magnitude of possible climate change. The review then considers the role of alternative technology assumptions in climate policy (‘stabilization’) scenarios. The review continues by presenting a discussion of the multitude of mechanisms underlying technological change that need to be considered when discussing policy options to further the availability and economics of mitigation and adaptation technologies.

2.7.1.1 Technological change in no-climate policy (reference) scenarios

The importance of technological change for future GHG emission levels and hence the magnitude of possible climate change has been recognized ever since the earliest literature reviews ( Ausubel and Nordhaus, 1983 [NPR] ). Subsequent important literature assessments (e.g. Alcamo et al., 1995 [NPR, MoS, SRC] ; Nakicenovic et al., 1998b [ARC] ; Edmonds et al., 1997 [NPR, SRC] ; SRES, 2000 [NPR] ) have examined the impact of alternative technology assumptions on future levels of GHG emissions. For instance, the SRES 2000 [NPR] ) report concluded technology to be of similar importance for future GHG emissions as population and economic growth combined. A conceptual simple illustration of the importance of technology is provided by comparing individual GHG emission scenarios that share comparable assumptions on population and economic growth, such as in the Low Emitting Energy Supply Systems (LESS) scenarios developed for the IPCC SAR ( 1996 ) or within the IPCC SRES 2000 [NPR] ) A1 scenario family, where for a comparable level of energy service demand, the (no-climate-policy) scenarios span a range of between 1038 (A1T) and 2128 (A1FI) GtC cumulative ( 1990 - 2100 ) emissions, reflecting different assumptions on availability and development of low- versus high-emission technologies. Yet another way of illustrating the importance of technology assumptions in baseline scenarios is to compare given scenarios with a hypothetical baseline in which no technological change is assumed to occur at all. For instance, GTSP 2001 [NPR] and Edmonds et al. 1997 [NPR, SRC] , see also Figure 3.32 in Chapter 3 ) illustrate the effect of changing reference case technology assumptions on CO₂ emissions and concentrations based on the IPCC IS92a scenario by holding technology at 1990 levels to reveal the degree to which advances in technology are already embedded in the non-climate-policy reference case, a conclusion also confirmed by Gerlagh and Zwaan, 2004 [NotFound] . As in the other scenario studies reviewed, the degree to which technological change assumptions are reflected in the scenario baseline by far dominates future projected emission levels. The importance of technology is further magnified when climate policies are considered. See for example, the stabilization scenarios reviewed in IPCC TAR ( 2001 ) and also Figure 2.1 below.

Figure 2.1: Emission impacts of exploring the full spectrum of technological uncertainty in a given scenario without climate policies. Relative frequency (percent) of 130,000 scenarios of full technological uncertainty regrouped into 520 sets of technology dynamics with their corresponding carbon emissions by 2100 . Also shown is a subset of 13,000 scenarios grouped into 53 sets of technology dynamics that are all ‘optimal’ in the sense of satisfying a cost minimization criterion in the objective function. See text for further discussion. 1 Gt C = 3.7 Gt CO₂

Source: Adapted from Gritsevskyi and Nakicenovic, 2000 [MoS, ARC] .

Perhaps the most exhaustive examination of the influence of technological uncertainty to date is the modelling study reported by Gritsevskyi and Nakicenovic 2000 [MoS, ARC] ). Their model simulations, consisting of 130,000 scenarios that span a carbon emission range of 6 to 33 GtC by 2100 ( Figure 2.1 ), provided a systematic exploration of contingent uncertainties of long-term technological change spanning a comparable range of future emissions as almost the entirety of the no-climate policy emissions scenario literature (see Chapter 3 for an update of the scenario literature). The study also identified some 13,000 scenarios (out of an entire scenario ensemble of 130,000) regrouped into a set of 53 technology dynamics that are all ‘optimal’ in the sense that they satisfy the same cost minimum in the objective function, but with a bimodal distribution in terms of emissions outcomes. In other words, considering full endogenous technological uncertainty produces a pattern of ‘technological lock-in’ into alternatively low or high emissions futures that are equal in terms of their energy systems costs. This finding is consistent with the extensive literature on technological ‘path dependency’ and ‘lock-in phenomena’ (e.g. ( Arthur, 1989 ) ) as also increasingly reflected in the scenario literature (e.g. Nakicenovic et al., 1998b [ARC] and the literature review in Chapter 3 ). This casts doubts on the plausibility of central tendency technology and emissions scenarios. It also shows that the variation in baseline cases could generate a distribution of minimum costs of the global energy system where low-emission baseline scenarios could be as cheap as their high-emission counterparts.

The results also illustrate the value of technology policy as a hedging strategy aiming at lowering future carbon emissions, even in the absence of directed climate policies, as the costs of reducing emissions even further from a given baseline are ceteris paribus proportionally lower with lower baseline emissions.

2.7.1.3 Technological change and the costs of achieving climate targets

Rates of technological change are also critical determinants of the costs of achieving particular environmental targets. It is widely acknowledged that technological change has been a critical factor in both cost reductions and quality improvements of a wide variety of processes and products. ^[27] Assuming that technologies in the future improve similarly to that observed in the past enables experts to quantify the cost impacts of technology improvements in controlled modeling experiments. For instance, Edmonds et al. 1997 [NPR, SRC] , compare Figure 3.36 in Chapter 3 ) analyzed the carbon implications of technological progress consistent with historical rates of energy technology change. Other studies have confirmed Edmonds’ ( 1997 ) conclusion on the paramount importance of future availability and costs of low-emission technologies and the significant economic benefits of improved technology that, when compounded over many decades, can add up to trillions of dollars. (For a discussion of corresponding ‘value of technological innovation’ studies see Edmonds and Smith 2006 [NPR, SRC] ) and Section 3.4 , particularly Figure 3.36 in Chapter 3 ). However, to date, model calculations offer no guidance on the likelihood or uncertainty of realizing ‘advanced technology’ scenarios. However, there is an increasing number of studies (e.g. Gerlagh and van der Zwaan, 2006 [NPR] ) that explore the mechanisms and policy instruments that would need to be set in place in order to induce such drastic technological changes.

The treatment of technological change in an emissions and climate policy modeling framework can have a huge effect on estimates of the cost of meeting any environmental target. Models in which technological change is dominated by experience (learning) curve effects, show that the cost of stabilizing GHG concentrations could be in the range of a few tenths of a percent of GDP, or even lower (in some models even becoming negative) – a finding also confirmed by other modelling studies (e.g. Rao et al., 2005 [NPR, ARC] ) and consistent with the results of the study by Gritsevskyi and Nakicenovic 2000 [MoS, ARC] ) reviewed above, which also showed identical costs of ‘high’ versus ‘low’ long-term emission futures. This contrasts with the traditional view that the long-term costs ^[28] of climate stabilization could be very high, amounting to several percentage points of economic output (see also the review in IPCC, 2001 [NPR] ).

Given the persistent uncertainty of what constitutes ‘dangerous interference with the climate system’ and the resulting uncertainty on ultimate climate stabilization targets, another important finding related to technology economics emerges from the available literature. Differences in the cost of meeting a prescribed CO₂ concentration target across alternative technology development pathways that could unfold in the absence of climate policies are more important than cost differences between alternative stabilization levels within a given technology-reference scenario. In other words, the overall ‘reference’ technology pathway can be equally, if not more, important in determining the costs of a given scenario as the stringency of the ultimate climate stabilization target chosen (confer Figure 2.2 ).

In a series of alternative stabilization runs imposed on the SRES A1 scenarios, chosen for ease of comparability as sharing similar energy demands, Roehrl and Riahi 2000 [ARC] ) confer also IPCC 2001 [NPR] ) have explored the cost differences between four alternative baselines and their corresponding stabilization targets, ranging from 750 ppmv all the way down to 450 ppmv. In their calculations, the cost differences between alternative baselines are also linked to differences in baseline emissions: advanced post-fossil fuel technologies yield both lower overall systems costs as well as lower baseline emissions and hence lower costs of meeting a specified climate target (confer the differences between the A1C and A1T scenarios in Figure 2.2 ). Their findings are consistent with the pattern identified by Edmonds et al. 1997 [NPR, SRC] and Gerlagh and van der Zwaan 2003 [JoC, MoS] ). Cost differences are generally much larger between alternative technology baselines, characterized by differing assumptions concerning availability and costs of technologies, rather than between alternative stabilization levels. The IEA 2004 [NPR] ) World Energy Outlook also confirms this conclusion, and highlights the differential investment patterns entailed by alternative technological pathways. ^[29] The results from the available literature thus confirm the value of advances in technology importance in lowering future ‘baseline’ emissions in order to enhance feasibility, flexibility, and economics of meeting alternative stabilization targets, in lowering overall systems costs, as well as in lowering the costs of meeting alternative stabilization targets.

Figure 2.2: The impacts of different technology assumptions on energy systems costs and emissions (cumulative 1990 – 2100, systems costs (undiscounted) in trillion US$) in no-climate policy baseline (reference) scenarios (based on the SRES A1 scenario family that share identical population and GDP growth assumptions) and in illustrative stabilization scenarios (750, 650, 550 and 450 ppm respectively). For comparison: the total cumulative (undiscounted) GDP of the scenarios is around 30,000 trillion US$ over the 1990 – 2100 time period.

Source: Roehrl and Riahi 2000 [ARC] ).

A robust analytical finding arising from detailed technology-specific studies is that the economic benefits of technology improvements (i.e. from cost reductions) are highly non-linear, arising from the cumulative nature of technological change, from interdependence and spillover effects, and from potential increasing returns to adoption (i.e. costs decline with increasing market deployment of a given technology). ^[30] (A detailed review covering the multitude of sources of technological change, including the aforementioned effects, is provided in Chapter 11 , Section 11.5 , discussing so-called ‘induced technological change’ models).

2.7.2 Technological change

Changes in technology do not arise autonomously – they arise through the actions of human beings, and different social and economic systems have different proclivities to induce technological change. The range of actors participating in the process of technological change spans the full range of those that use technology, design and manufacture technology, and create new knowledge.

The process of technological change has several defining characteristics. First, the process is highly uncertain and unpredictable. Firms planning research toward a well-defined technical goal must plan without full knowledge regarding the potential cost, time frame, and even the ultimate success. Further, the history of technological development is rife with small and large examples of serendipitous discoveries, (e.g. Teflon) whose application is far beyond, or different, than their intended use.

A second defining characteristic of technological change is the transferable, public-good nature of knowledge. Once created, the value of technological knowledge is difficult to fully appropriate; some or all eventually spills over to others, and in doing so the knowledge is not depleted. This characteristic of knowledge has both benefits and drawbacks. On the one hand, an important discovery by a single individual, such as penicillin, can be utilized worldwide. Knowledge of penicillin is a public good and therefore one person’s use of this knowledge does not preclude another person from using this same knowledge – unlike for capital or labour, where use in one task precludes use in an alternative task. On the other hand, the understanding by potential innovators that any new knowledge might eventually spill over to others limits expected profits and therefore dampens private-sector innovative activity. Thus intellectual property rights can serve both as a barrier and an aid in technology change. A final, third feature of technological change is its cumulativeness, which is also frequently related to spillover effects.

There are numerous paradigms used to separate the process of technological change into distinct phases. One approach is to consider technological change as roughly a two-part process, which includes:

(1) The process of conceiving, creating, and developing new technologies or enhancing existing technologies – the process of advancing the ‘technological frontier’.

(2) The process of diffusing or deploying these technologies.

These two processes are inextricably tied. The set of available technology defines what might be deployed, and the use of technology affords learning that can guide R&D programmes or directly improve technology through learning-by-doing. The two processes are also linked temporally. The set of technologies that find their way into use necessarily lags the technological frontier. The useful life of technologies – their natural turnover rate – helps to drive the time relationship. Car lifespans can be in the order of 15 years, but the associated infrastructure – roads, filling stations, vehicle manufacturing facilities – have significantly longer lifespans, and electric power plants may be used for a half-century or more; hence, the average car is substantially younger than the average coal-fired power plant and much of its associated infrastructure. The nature of the capital stock (e.g. flexifuel cars that can use both conventional petrol and ethanol) is also important in determining diffusion speed.

2.7.2.3 The public-sector role in technological change

Given the importance of technology in determining both the magnitude of future GHG emission levels as well as feasibility and costs of emission reduction efforts, technology policy considerations are increasingly considered in climate policy analyses. Ongoing debate centers on the relative importance of two differing policy approaches: technology-push (through efforts to stimulate research and development) and demand-pull (through measures that demand reduced emissions or enhanced efficiency). Technology-push emphasizes the role of policies that stimulate research and development, especially those aimed at lowering the costs of meeting long-term objectives with technology that today is very far from economic in existing markets. This might include such measures as public-funded R&D or R&D tax credits. Demand-pull emphasizes the use of instruments to enhance the demand for lower-emission technologies, thereby increasing private incentives to improve these technologies and inducing any learning-by-doing effects. Demand-pull instruments might include emissions taxes or more direct approaches, such as renewable portfolio standards, adoption subsidies, or direct public-sector investments (see Figure 2.3 ).

Figure 2.3 : Technology development cycle and its main driving forces. Note that important overlaps and feedbacks exist between the stylized technology life-cycle phases illustrated here and therefore the illustration does not suggest a ‘linear’ model of innovation.

Source: Adapted from Foxon 2003 [NPR, MoS] and Grubb 2005 [Ambiguous] ).

Two market failures are at issue when developing policies to stimulate technology development. The first is the failure to internalize the environmental costs of climate change, reducing the demand for climate-friendly technologies and thereby reducing private-sector innovation incentives and learning-by-doing. The second is a broad suite of private-sector innovation market failures that hold back and otherwise distort private-sector investment in technological advance, irrespective of environmental concerns (confer ( Jaffe et al., 2005 ) ). Chief among these is the inability to appropriate the benefits of knowledge creation. From an economic standpoint, two market failures require two policy instruments: addressing two market failures with a single instrument will only lead to second-best solutions (see, for example, Goulder and Schneider, 1999 [PoC, JoC] ). Hence, it is well understood that the optimal policy approach should include both technology-push and demand-pull instruments. While patents and various intellectual property protection (e.g. proprietary know-how) seeks to reward innovators, such protection is inherently imperfect, especially in global markets where such protections are not uniformly enforced by all governments. Similarly, in the early adoption of technology learning-by-doing (by producers) or learning-by-using (by consumers) may lower the cost to all future users, but in a way that may not fully reward the frontrunners. ^[33] Similarly, lack of information by investors and potential consumers of innovative technologies may slow the diffusion of technologies into markets. The ‘huge uncertainties surrounding the future impacts of climate change, the magnitude of the policy response, and thus the likely returns to R&D investment’ exacerbate these technological spillover problems ( (Jaffe et al., 2005 ) ).

The outstanding questions revolve around the relative combinations of instruments and around how effective single-policy approaches might be. Within this context, a number of authors (e.g. Montgomery and Smith, 2005 [NPR] ) have argued that fundamental long-term shifts in technology to mitigate greenhouse gas emissions cannot be achieved through emissions-constraining policies alone, and short-term cap and trade emission-reduction policies provide insufficient incentives for R&D into long-term technology options. Conversely, Popp 2002 [ARC] ) demonstrated how energy R&D is responsive to price signals, suggesting that without emissions constraints R&D into new low-emission technologies may face a serious lack of incentives and credible policy signals. The argument that emissions-based policies will induce long-term technology innovation relies primarily on two lines of reasoning ( Goulder 2004 [NPR] ; Grubb, 2005 [Ambiguous] ). The first is that the anticipation of future targets, based on a so-called announcement effect, will stimulate firms to invest in research and development and ultimately to invest in advanced, currently non-commercial technology (the credibility and effectiveness of this effect, however, being challenged by Montgomery and Smith, 2005 [NPR] ). The second is that early investment, perhaps through incentives, mandates, or government procurement programmes, will initiate a cycle of learning-by-doing that will ultimately promote innovation in the form of continuous improvement, which will drive down the cost of future investments in these technologies. This issue is especially critical in the scaling up of niche-market applications of new technologies (e.g. renewables) where mobilizing finance and lowering investment risks are important (see, for example, IEA, 2003 [NPR] , or Hamilton, 2005 [NPR] ). In their comparative analysis of alternative policy instruments Goulder and Schneider 1999 [PoC, JoC] ) found that when comparing a policy with only R&D subsidies to an emissions tax, the emissions-based policies performed substantially better.

Irrespective of the mix between demand-pull and technology-push instruments, a number of strong conclusions have emerged with respect to the appropriate policies to stimulate technological advance. First, it is widely understood that flexible, incentive-oriented policies are more likely to foster low-cost compliance pathways than those that impose prescriptive regulatory approaches ( (Jaffe et al., 2005 ) ). A second robust conclusion is the need for public policy to promote a broad portfolio of research, because results cannot be guaranteed since it is impossible to ex ante identify

technical winners or losers ( GTSP, 2001 [NPR] ). A third conclusion is that more than explicit climate change or energy research is critical for the development of technologies pertinent to climate change. Spillovers from non-energy sectors have had enormous impacts on energy-sector innovation, implying that a broad and robust technological base may be as important as applied energy sector or similar R&D efforts. This robust base involves the full ‘national systems of innovation’ ^[34] involved in the development and use of technological knowledge. Cost and availability of enabling infrastructure can be especially important factors that limit technology uptake in developing countries. ^[35] Here enabling infrastructure would include management and regulatory capacity, as well as associated hardware and public infrastructure.

2.7.3 The international dimension in technology development and deployment: technology transfer

Article 4.5 of the Convention states that developed country Parties ‘shall take all practicable steps to promote, facilitate, and finance, as appropriate, the transfer of, or access to, environmentally sound technologies and know-how to other Parties, particularly developing country Parties, to enable them to implement the provisions of the Convention’, and to ‘support the development and enhancement of endogenous capacities and technologies of developing country Parties’.

Similarly Article 10(c) of the Kyoto Protocol reiterated that all Parties shall: ‘cooperate in the promotion of effective modalities for the development, application, and diffusion of, and take all practicable steps to promote, facilitate and finance, as appropriate, the transfer of, or access to, environmentally sound technologies, know-how, practices and processes pertinent to climate change, in particular to developing countries, including the formulation of policies and programmes for the effective transfer of environmentally sound technologies that are publicly owned or in the public domain and the creation of an enabling environment for the private sector, to promote and enhance the transfer of, and access to, environmentally sound technologies’.

Technology transfer is particularly relevant because of the great interest by developing countries in this issue. This interest arises from the fact that many developing countries are in a phase of massive infrastructure build up. Delays in technology transfer could therefore lead to a lock-in in high-emissions systems for decades to come (e.g. Zou and Xuyan, 2005 ). Progress on this matter has usually been linked to progress on other matters of specific interest to developed countries. Thus Article 4.7 of the Convention is categorical that ‘the extent to which developing country Parties will effectively implement their commitments under the Convention will depend on the effective implementation by developed country Parties of their commitments under the Convention related to financial resources and the transfer of technology’.

The IPCC Special Report on Methodological and Techno-logical Issues on Technology Transfer (SRTT) ( IPCC, 2000 [NPR] ) defined the term ‘technology transfer’ as a broad set of processes covering the flows of know-how, experience and equipment for mitigating and adapting to climate change amongst different stakeholders. A recent survey of the literature is provided in ( Keller 2004 ) ) and reviews with special reference to developing countries are included in Philibert 2005 [NPR] and Lefevre 2005 [NPR, ARC] ). The definition of technology transfer in the SRTT and the relevant literature is wider than implied by any particular article of the Convention or the Protocol. The term ‘transfer’ was defined to ‘encompass diffusion of technologies and technology cooperation across and within countries’. It also ‘comprises the process of learning to understand, utilize and replicate the technology, including the capacity to choose and adapt to local conditions and integrate it with indigenous technologies’.

This IPCC report acknowledged that the ‘theme of techno-logy transfer is highly interdisciplinary and has been approached from a variety of perspectives, including business, law, finance, micro-economics, international trade, international political economy, environment, geography, anthropology, education, communication, and labour studies’.

Having defined technology transfer so broadly, the report ( IPCC, 2000 [NPR] , p. 17) concluded that ‘although there are numerous frameworks and models put forth to cover different aspects of technology transfer, there are no corresponding overarching theories’ (emphasis added). Consequently there is no framework that encompasses such a broad definition of technology transfer.

The aforementioned report identified different stages of technology transfer and different pathways through which it is accomplished. These stages of technology transfer are: identification of needs, choice of technology, and assessment of conditions of transfer, agreement and implementation. Evaluation and adjustment or adaptation to local conditions, and replication are other important stages. Pathways for technology transfer vary depending on the sector, technology type and maturity and country circumstances. Given this variety and complexity, the report concluded that there is no pre-set answer to enhancing technology transfer.

There is no international database tracking the flow of ESTs (environmentally sound technologies). Little is known about how much climate-relevant equipment is transferred, and even less about the transfer of know-how, practices and processes, and most international analyses rely on proxy variables. It is well known that the nature of financial flows from OECD countries to developing countries has changed over the last 15 years. Overseas development assistance (ODA) has declined and been overtaken by private sources of foreign direct investments ( WDI, 2005 [NPR] ). International financial statistics only reflect the quantity and not the quality of FDI. They also say nothing about what fraction is a transfer of ESTs. Despite its decline, ODA is still critical for the poorest countries, particularly when it is aimed at developing basic capacities to acquire, adapt, and use foreign technologies.

IPCC 2000 [NPR] , p. 22) summarized the historical experience as a ‘failure of top-down, technology-focused development’. Some developing country policymakers believe that payments for technology are beyond their means and that international technology transfer contributes little to technological development in the recipient country ( UNDP, 2000 [NPR] ). Many failures of technology transfer have resulted from an absence of human and institutional capacity ( IPCC, 2000 [NPR] , p. 118).

There are several modes to encourage technology transfer to developing countries, from technical assistance and technology grants, to capacity building and policy development cooperation. The priorities for these modes shift as host countries develop economically. Technology demonstration projects can play an important role early in the industrialization process. As the economy grows, policy development cooperation, such as assistance to develop energy-efficiency standards or to create an enabling environment for technology diffusion, becomes more important. ( Ohshita and Ortolano 2003 ) ) studied past experiences of demonstration projects using cleaner energy technologies in developing countries through assistance by international organizations as well as developed countries. They found that demonstration projects raised awareness of cleaner energy technologies in the technology transfer process, but were not very successful in diffusing the technologies more widely in the target developing countries. For China in particular, demonstration projects played an important role in the past, when the economy began shifting from a centrally planned system to a more open, market-based system. There is increasing recognition that other modes of technology diffusion may now be more suitable for China. Given the continued high growth of the Chinese economy, donors have been shifting their assistance programmes from technology demonstration to policy development assistance ( Ohsita, 2006 [NPR] ).

Figure 2.4: A general framework for factors affecting technology transfer and subsequent innovation.

Figure 2.4 shows one attempt to create a framework for all forms of technology transfer. In all forms technology transfer, especially across countries, at least seven characteristics are important. These are:

1. The characteristics of the technology.

2. The characteristics of the originator of the transfer.

3. The enabling (or disabling) environment in the country of origin.

4. The conditions of the transfer.

5. The characteristics of the recipient.

6. The enabling (or disabling) environment in the host country.

7. The ultimately valuable post-transfer steps, i.e. assimilation, replication and innovation.

Each of these characteristics are discussed below.

Characteristics of the originator of the transfer. Initially, there was a widespread tendency to think of technology transfer in supply-side terms – the initial choice and acquisition of technology ( (Brooks, 1995 ) ) and a lack of corresponding focus on the other factors that influence the successful outcome of technology transfer, such as enabling environment, institutions and finance.

The environment in the country of origin can be conducive or disabling for technology transfer. The public sector continues to be an important driver in the development of ESTs. Of the 22 barriers listed in the technical summary of the IPCC Report ( 2000 ) as barriers to technology transfer, 21 relate to the enabling environment of recipient countries. Many governments transfer or license the patents arising out of publicly funded efforts to the private sector as a part of their industrial policy, and then the transferred patents follow the rules of privately owned technologies ( IPCC, 2000 [NPR] , p. 25).

One should also consider the ‘imperfect’ nature of technology markets:

(1) While some of the components of technology are of a public-good nature, others have an important tacit nature.

(2) Technology markets are normally very concentrated on the supply side, and bargaining power is unevenly distributed.

(3) The strategic nature of technologies normally includes limiting clauses and other restrictions in transfer contracts (for a discussion see Arora et al., 2001 [NPR] ; Kumar, 1998 [NPR] ).

Technology Denial Regimes ^[36] in the country of origin also sometimes constitute a barrier to technology transfer, especially for multiple-use technologies. Thus supercomputers can be used for climate modelling and global circulation models and also to design missiles.

The conditions of the transfer. Most technologies are transferred in such a way that the originators also benefit from the transfer and this helps to establish strong incentives for proper management and maintenance of the technologies. The conditions of the transfer will primarily depend on the transfer pathway used, as mentioned above. Common pathways include government assistance programmes, direct purchases, trade, licensing, foreign direct investment, joint ventures, cooperative research agreements, co-production agreements, education and training and government direct investment. Developing countries have argued for the transfer of ESTs and corresponding know-how, on favourable, concessional and preferential terms (Agenda 21, 1992, Chapter 34). There have been instances in the pharmaceutical industry when certain drugs benefiting developing countries have been licensed either free or on concessionary terms.

The characteristics of the recipient. The recipient must understand local needs and demands; and must possess the ability to assess, select, import, adapt, and adopt or utilize appropriate technologies.

The enabling (or disabling) environment in the host country. Many of the barriers to technology transfer that are listed in the IPCC Report ( IPCC, 2000 [NPR] , p. 19) relate to the lack of an enabling (or a disabling) environment in the recipient country for the transfer of ESTs. A shift in focus, from technology transfer per se to the framework represented in Figure 2.4 , leads to an equal emphasis on the human and institutional capacity in the receiving country. A crucial dimension of the enabling environment is an adequate science and educational infrastructure. It must be recognized that capacity building to develop this infrastructure is a slow and complex process, to which long-term commitments are essential.

A recipient’s ability to absorb and use new technology effectively also improves its ability to develop innovations. Unfortunately, the capacity to innovate and replicate is poorly developed in developing countries ( STAP, 1996 [NPR] ). However, the engineering and management skills required in acquiring the capacity to optimize and innovate are non-trivial. The technology-importing firm needs to display what has been called ‘active technological behaviour’. Firms that do not do this are left in a vicious circle of technological dependence and stagnation ( UNDP, 2000 [NPR] ).