Page images
PDF
EPUB

emissions. To the extent that mitigation costs vary among countries, then the costs of a given level of global mitigation would be lower if international coordination resulted in the most cost-effective mitigation policies being adopted first, than if each country were simply to reduce emissions by an equal amount." Although a growing theoretical literature is currently devoted to this issue, it has been addressed only in a few empirical modelling works to date. Most analyses have assumed either that national polices are adopted unilaterally or that the effects of national mitigation policies are neutral with respect to international competition (i.e., that all countries take the same action or that the effects of mitigation policies on international competitiveness are small). So far no model has been able to account for the gains to be expected from international cooperation in research and development and technology diffusion.

8.3 Patterns of Development and Technological Change

For any country, both the level of greenhouse gas emissions and the costs of mitigation of those emissions depend on a series of factors, including the range of technologies used and the underlying technological and socioeconomic conditions that give rise to final demands for land, energy services, or transportation. Most models explore options and alternatives as well as possible future economic conditions, based upon particular assumptions about the nature of technological change and long-term development paths. This section of the chapter will explore this latter question in more detail, because these assumptions are very often only implicit and because current modelling methodologies and current data do not enable modellers to treat explicitly some of these critical parameters.

8.3.1 Links Between Development Patterns, Technical Change, and Mitigation Costs

8.3.1.1 The Importance of the Socioeconomic Assumptions Underlying Scenarios

Much of the discussion about the costs of greenhouse gas emission reduction - and indeed about energy issues in general - naturally focuses on economic concepts and variables such as income, prices, and growth in GDP. We therefore begin with a rather trite observation: damages to the environment are not caused by a given amount of dollars, yen, pounds, Deutschmarks, or francs, but by the material content of the consumption or production activities that are the counterparts of these monetary values. To put it in another way, greenhouse gas emissions over the long run depend not only on the rate of economic growth but also on the structure and physical content of this growth.

It is well known that countries with rather similar development levels may have very different energy consumption per capita ratios or very different transportation requirements. Some of these disparities are obviously due to natural and geographical characteristics (temperature, population density, etc.), but many of them stem from

differences between the development patterns of these countries. Comparative studies aiming at explaining these differences (Martin, 1992; Darmstadter, et al., 1977) suggest the importance of five considerations that will influence the amount of greenhouse gas emissions, given a certain overall rate of economic growth:

1.

2.

3.

4.

5.

Technological patterns in sectors such as energy, transport, heavy industry, construction, agriculture, and forestry. As discussed below, these patterns encompass individual technological choices and options but also include overall technological systems, with their particular internal consistencies and dynamics.

Consumption patterns. For a given per capita income, parameters such as housing patterns, leisure styles, or the durability and the rate of obsolescence of consumption goods will have a critical influence on long-run emission profiles. Beyond their purely technical aspects, these patterns are also related to the level of education, distribution of income, and degree of dualism in an economy.

The geographical distribution of activities, which encompasses the distribution of human settlements in a given territory, climate impacts on energy demand, and the nature of urban form within a given settlement. The impact of this parameter is threefold: first on the evolution of land uses, second on mobility needs and transportation requirements, and third on the energy used for heating and cooling.

Structural changes in the production system and, in particular, the role of high or
low energy-intensive industries and services. The energy content of industries
such as steel, nonferrous metals, chemicals, or pulp and paper is between four and
six times that of other industries. At the other end of the spectrum, a simple
phone call on a given commodity futures market can generate substantial
economic gains (and e-mail connections can increase the efficiency of researchers)
for a negligible energy content. A shift in the relative size of primary production
and service industries in an economy may or may not affect the overall level of
economic activity but will have significant implications for energy use.

Trade patterns. It is generally argued in the economic literature that removing tariff and nontariff trade barriers enhances overall economic efficiencies. But, because historical experience demonstrates that some form of protectionism was considered necessary to many countries at the early stages of industrialization, and because of the transition problems for removing these barriers (risks of social and economic disruptions), free trade will be implemented only gradually after the Uruguay Round. In the meantime, the world is apparently moving towards the creation of regional trading blocks (European Union, NAFTA, Mercosul). The future of these arrangements is very hard to predict and will alter significantly the access to the best available technologies, the capacity of developing countries to generate high enough internal capital accumulation to finance infrastructures and education, the location of industrial activities, and future land uses (because of the

impact on agricultural markets).

Of course these factors are not ignored by current economic models. They are in some way captured by changes in economic parameters such as the structure of household expenses devoted to heating, transportation, or food, the share of each activity in the total value added, the share of energy costs and transportation costs in the production function of industrial sectors, or import-export elasticities. This type of treatment is convenient for addressing the requirements traditionally posed by policymakers since the beginning of economic modelling just before the Second World War: to provide information on the consequences of economic policies (e.g., a monetary devaluation, a fiscal policy, an incentive to final demand through public investment programmes, etc.) over the short term (1 to 3 years), or to develop consistent economic scenarios to frame sectoral planning and policy (mainly in energy and transportation) over the medium term (4 to 10 years). For these time horizons and objectives it is logical to assume a continuation of historical trends in the main characteristics of development patterns and in the speed and direction of the transformation of these characteristics.

For the longer-term periods under consideration in greenhouse debates, these assumptions cannot be easily maintained, and economic parameters cannot easily be viewed as the sole command variables needed to predict the future of our production and consumption systems. For example, a given amount of added value produced by the steel or chemical industries may correspond to very different levels of material production (and thus energy demand), depending upon the level of sophistication of the final product; in the same way, the differences in household budgets devoted to transportation may not fully express the differences in mobility and transportation patterns prevailing between towns with or without public rail transport systems, or between car uses in towns with very different levels of congestion.

More fundamentally, the dynamics of long-term technological development cannot be fully captured by changes in the capital output ratio (the aggregate amount of economic capital used per unit of output) or by the impact of the rate of investment on overall productivity. These parameters are, of course, important, but the outcome in terms of greenhouse gas emissions will also depend upon dynamic linkages between technology, consumption patterns (mainly with respect to energy requirements), transportation, urban infrastructure, and the rural-urban distribution of population. We will come back later to the attempts of existing models to take these parameters into account, but the lack of knowledge available about their dynamic linkages and about their interactions with economic policies and economic signals over the long run must be underlined at this stage, together with the intrinsic difficulty of predicting innovations and transformations of lifestyles over the long run. While many fields of social science address these issues, such information is not typically available in a form easy to process in a numerical model.

In principle, alternative configurations of the factors determining development patterns could be formally combined to give internally consistent scenarios characterized by various physical and technical characteristics and economic equilibria for a given rate of economic growth. But this does not mean that all of these possible scenarios are viable and that it is possible to achieve a transition towards these long-term pictures without entailing high social and economic costs. What matters here is that these underlying technological and consumption factors are critical not only for the definition of the baseline scenarios but also for the assessment of actual mitigation costs for a given mitigation policy or objective.

As explained earlier, the mitigation costs attached to each possible baseline scenario depend not only on the absolute level of the required mitigation and on the array of available technologies (energy efficiency, fuel switching, biomass planting, other renewable energy development, modal choices in transport) but also on the timing of this mitigation (Grubb, et al., 1994; Hourcade and Chapuis, 1993; Manne and Richels, 1992). In this connection, three groups of partly related issues become very important:

The flexibility (or inertia) of consumption patterns underlying the activity of
greenhouse gas-emitting sectors such as energy, transportation, or cement. This
flexibility parameter determines the speed of adaptation to a given economic or
noneconomic signal and encompasses two aspects. The first is the rate of renewal
of existing end-use equipment, which is likely to be an important factor in the
case of buildings, which have longer lifetimes than most capital stock. A more
critical inflexibility stems from the systemic linkages between consumption
patterns, technology, and the spatial distribution of activities. To take an extreme
illustration, it would be far more costly to move away from oil-based automobile
fuels in big conurbations where urban structure makes the use of cars almost
compulsory than in, say, a European town of about 40,000 inhabitants where it
would be easier to satisfy a significant proportion of intra-urban personal
transportation with electric buses and bicycles. The second aspect is the length of
time required for turnover of the energy supply system. In this connection, the
flexibility of policy response is associated with the size and lead times associated
with new energy supply technologies.

Behavioural characteristics that determine technical change and the evolution of
life styles; this point will be elaborated further with regard to technical adoption
mechanisms. With regard to consumption patterns, the roots of inertia are not
only technical. Anthropology and social psychology demonstrate how far they are
embedded in cultures and habits and, more generally, how individual consumption
behaviours are shaped by social determinants that are hard to change overnight
(Robinson, 1991).

Interactive effects due to feedbacks between the use of certain options and the rest of the economy. Some sectors have a pervasive effect on the rest of the economy,

and drastic adaptations would trigger strong structural shocks on the entire
productive system.

As a whole, the critical role of these factors comes from the fact that the bigger the inertia of the production and consumption systems the more the mitigation costs will be determined by the timing of the required mitigation. This is easy to understand as far as the adoption of new technologies is concerned: to accelerate the replacement of old equipment could be a major source of the costs of mitigation. But this inertia also determines the magnitude of the loss of consumer welfare associated with mitigation. If the range of available alternatives is restricted by the material and spatial features of one's living conditions, the consumer will tend to suffer from welfare losses during the transition period towards consumption and production systems that emit lesser quantities of greenhouse gases.

A special case for inertia in development patterns occurs when such inertia creates irreversible processes of technological change. To a large extent such irreversibilities emerge out of the factors traditionally discussed in the literature on technical innovation: learning curves, economies of scale, increasing informational returns, positive network externalities, barriers to entry, and others. Irreversibility occurs when these factors combine in such a way that a particular trajectory of technological change and development is created that effectively makes impossible alternative choices that were available earlier. This gives rise to a time dependence of technical choices and the occurrence of "lock-in" effects (Arthur, 1988). Beyond such a bifurcation point, market forces will reinforce the first choice in a self-fulfilling process. To give a single example, given the high research and development costs involved in a new automobile engine, it is unlikely that research and development risks will be incurred simultaneously on electric cars, 2L/100 km gasoline engines, and biofuel engines. As with gasoline engines in the early days of the automobile, choice of a particular technology will essentially prevent the development of alternatives.

More generally, beyond technical considerations themselves, the self-reinforcing loops among technical choices, consumer demand, and the geographical distribution of activities and human settlements explain the fact that particular sets of technological and behavioural options can be clustered into consistent packages which, at least for a rather long period of time, foreclose other options in technology and innovation. These clusters are rather systematic in industries relying on network structures, such as energy, transportation, or telecommunications because of the characteristics of their production function (e.g., discontinuities and economies of scale), the need for technical harmonization across the network, and their dynamic interactions with markets. This is particularly clear for transportation systems because of their linkages with urban and spatial dynamics. For example, trends observed in Western Europe over the past several decades could be expected to lead to a doubling of road freight on highways during the following 15 years under the influence of the Single European Market. However, if Austria and Switzerland maintain their policies of limiting international truck freight

« PreviousContinue »