Alan S Manne and Richard G Richels

Although the Kyoto Protocol (Conference of the Parties 1997) encompasses a number of radiatively active gases, assessments of compliance costs have focused almost exclusively on the costs of reducing carbon dioxide (CO2) emissions.1 There are a number of reasons why this is the case. Carbon dioxide is by far the most important anthropogenic greenhouse gas (Houghton et al. 1996); until recently, few economic models have had the capability to conduct comprehensive multi-gas analyses;2 and the quality of data pertaining to other greenhouse gases (GHGs) is poor (both spatially and temporally). Nevertheless, focusing exclusively on CO2 may bias mitigation cost estimates and lead to policies that are unnecessarily costly. In this chapter, we examine the implications of a multi-gas approach for both short- and long-term climate policy.

A number of anthropogenic gases have a positive effect on radiative forcing (Houghton et al. 1996). We consider the three thought to be the most important: CO2, methane (CH4), and nitrous oxide (N2O). We also consider the cooling effect of sulphate aerosols. We, however, exclude the so-called second basket of greenhouse gases included in the Kyoto Protocol. These are the hydrofluorocarbons (HFCs), the perflu-orocarbons (PFCs), and sulphur hexafluoride (SF6). This omission is not believed to alter the major insights of the analysis.

When dealing with multiple gases, it is necessary to find some way to establish equivalence among gases. The problem arises because the gases are not comparable. Each gas has its own lifetime and specific radiative forcing. Houghton et al. (1996) suggested the use of global warming potentials (GWPs) to represent the relative contribution of different greenhouse gases to the radiative forcing of the atmosphere. A number of studies have, however, pointed out the limitations of this approach (Prinn, Chapter 9, this volume), noting that in order to derive optimal control policies, it is important to consider both the impacts of climate change and the costs of emissions abatement.3 GWPs do neither. Also problematic is the arbitrary choice of time horizon for calculating

GWPs.4 In this chapter, we examine alternatives to GWPs that may provide a more logical basis for action.

The Model

The analysis is based on the MERGE model (a model for evaluating the regional and global effects of greenhouse gas reduction policies). MERGE is an intertemporal general equilibrium model. Like its predecessors, the version used for the present analysis (MERGE 4.0) is designed to be sufficiently transparent so that one can explore the implications of alternative viewpoints in the greenhouse debate. It integrates submodels that provide a reduced-form description of the energy sector, the economy, emissions, concentrations, temperature change, and damage assessment.

MERGE combines a bottom-up representation of the energy supply sector with a top-down perspective on the remainder of the economy. For a particular scenario, a choice is made among specific activities for the generation of electricity and for the production of non-electric energy. Oil, gas, and coal are viewed as exhaustible resources. There are introduction constraints on new technologies and decline constraints on existing technologies. MERGE also provides for endogenous technology diffusion. That is, the near-term adoption of high-cost carbon-free technologies in the electricity sector leads to accelerated future introduction of lower-cost versions of these technologies.

Outside the energy sector, the economy is modeled through nested constant elasticity production functions. The production functions determine how aggregate economic output depends upon the inputs of capital, labor and electric and non-electric energy. In this way, the model allows for both price-induced and autonomous (non-price) energy conservation and for interfuel substitution. It also allows for macroeconomic feedbacks. Higher energy and/or environmental costs will lead to fewer resources available for current consumption and for investment in the accumulation of capital stocks. Economic values are reported in U.S. dollars of constant 1990 purchasing power.

The world is divided into nine regions: (1) the USA; (2) OECDE (Western Europe); (3) Japan; (4) CANZ (Canada, Australia, and New Zealand); (5) EEFSU (Eastern Europe and the Former Soviet Union); (6) China; (7) India; (8) MOPEC (Mexico and OPEC); and (9) ROW (the rest of world). Note that the countries belonging to the Organisation for Economic Co-operation and Development (OECD) (Regions 1 through 4) together with the economies in transition (Region 5) constitute Annex B of the Kyoto Protocol.

Each of the model's regions maximizes the discounted utility of its consumption subject to an intertemporal budget constraint. Each region's wealth includes not only capital, labor, and exhaustible resources, but also its negotiated international share of emission rights. Particularly relevant for the present calculations, MERGE provides a general equilibrium formulation of the global economy. We model international trade in emission rights, allowing regions with high marginal abatement costs to purchase emission rights from regions with low marginal abatement costs. There is also trade in oil, gas, and energy-intensive goods. International capital flows are endogenous.

MERGE can be used for either cost-effectiveness or cost-benefit analysis. For the latter purpose, the model translates global warming into its market and nonmarket impacts. Market effects are intended to measure direct impacts on gross domestic product (GDP), such as agriculture, timber, and fisheries. Nonmarket effects refer to those not traditionally included in the national income accounts, such as impacts on biodiversity, environmental quality, and human health. These effects are even more difficult to measure than market effects.

For Pareto-optimal outcomes—that is, those scenarios in which the costs of abatement are balanced against the impacts of global climate change—each region evaluates its future welfare by adjusting the value of its consumption for both the market and non-market impacts of climate change. The market impacts represent a direct claim on gross economic output—along with energy costs, aggregate consumption, and investment. Nonmarket impacts enter into each region's intertemporal utility function and are viewed as an adjustment to the conventional value of macroeconomic consumption. For more on the model, see our web site: http://www.stanford.edu/group/MERGE/.

The Treatment of Greenhouse Gases and Carbon Sinks

MERGE requires information on the sources of the gases under consideration, their geographical distribution, how they are likely to change over time, and the marginal costs of emissions abatement. Unfortunately, the quality of the data is uneven, particularly for the non-CO2 greenhouse gases. In many instances, we have had to rely on a great deal of judgment to arrive at globally disaggregated time series. Similarly, there is a paucity of data related to the potential for carbon sinks. In this section, we identify the main sources for our estimates. In many instances, however, the cited data required some interpretation to meet the demands of the present analysis. Again, for details, see the computer program shown on our web site.

For purposes of the present analysis, greenhouse gas emissions are divided into two categories: energy related and non-energy related. MERGE tracks energy-related releases of both CO2 and CH4. For the reference case, the model is calibrated so that global CO2 emissions approximate the Houghton et al. (1995) central case no-policy scenario (IS92a). This has been done through the adjustment of several key supply- and demand-side parameters in the energy-economy submodel. Table 25.1 presents estimates of energy- and non-energy-related CH4 emissions for 1990. When a constraint is placed on GHG emissions, the choice of technologies for the energy sector is influenced by their emission characteristics.

We next turn to non-energy-related emissions. In the case of CO2, we must account for other industrial releases (primarily cement production) and the net changes associated with land use. According to Houghton et al. (1995), other industrial emissions are

Table 25.1. Methane emissions, 1990 (millions of tons)

Emissions source

USA

OECDE

Japan

CANZ

EEFSU

China

India

MOPEC

ROW

World

Non-energy-related emissions

Enteric fermentation

7.8

5.2

12.0

60.0

85.0

Rice paddies

0.0

18.0

18.0

24.0

60.0

Biomass burning

0.0

10.0

10.0

20.0

40.0

Landfills

8.0

8.0

2.0

4.0

4.0

2.0

12.0

40.0

Animal waste

5.0

2.0

2.0

2.0

5.0

3.0

6.0

25.0

Domestic sewage

5.0

5.0

1.0

1.0

1.0

1.0

1.0

10.0

25.0

Subtotal

25.8

15.0

1.0

5.0

7.0

43.2

46.0

0.0

132.0

275.0

Energy-related emissions

Gas

3.5

3.3

0.4

1.2

29.2

0.6

0.2

12.0

4.7

55.0

Coal

7.1

2.6

0.0

1.0

10.0

20.8

0.5

0.0

2.9

45.0

Subtotal

10.7

5.9

0.4

2.2

39.2

21.4

0.7

12.0

7.6

100.0

Anthropogenic

36.5

20.9

1.4

7.2

46.2

64.6

46.7

12.0

139.6

375.0

Natural

160.0

Total

535.0

Sources: Houghton et al. (1995) and IEA (1998).

Sources: Houghton et al. (1995) and IEA (1998).

Table 25.2. Potential sink enhancement in 2010 at a marginal cost of US$100 per ton of carbon (million tons of carbon)

Region

Potential sink enhancement

USA

50

OECDE

17

Japan

0

CANZ

50

EEFSU

34

China

25

India

13

MOPEC

25

ROW

250

World

464

Source: Houghton et al. (1996).

Source: Houghton et al. (1996).

relatively small. These are exogenous inputs into MERGE. With regard to land use, we assume that, in the absence of policy, the mass of carbon in the terrestrial biosphere remains constant.

This raises the issue of carbon sink enhancement. The protocol states that Annex B commitments can be met by the net changes in greenhouse gas emissions from sources and removal by sinks resulting from direct human-induced land use change and forestry activities limited to aforestation, reforestation, and deforestation since 1990, measured as verifiable changes in stocks in each commitment period (Conference of the Parties 1997). There is some confusion, however, regarding the treatment of soil carbon. This issue has been flagged for further study in the protocol. For the present analysis, we have adopted the values shown in Table 25.2 for 2010. We suppose that marginal sink enhancement costs are proportional to the quantity of enhancement. We also assume that the potential for sink enhancement increases over time.

Table 25.1 also includes non-energy-related CH4 emissions. Reductions from the reference path are determined by a set of time-dependent marginal abatement cost curves. In 2010 the curve is calibrated based on Reilly et al. (1999).5 For later years, the marginal cost of emissions abatement declines as a result of technical progress.

Nitrous oxide emissions are treated in a manner similar to non-energy-sector CH4 emissions. Table 25.3 reports estimates for 1990. A marginal abatement cost curve for each region is constructed for each commitment period. For 2010 we again rely on the work of Reilly et al. (1999). Similarly, for later years, we assume that the marginal cost of emission abatement declines with technical progress.

Table 25.3. Anthropogenic nitrous oxide emissions, 1990 (millions of tons)

Anthropogenic nitrous

Region

oxide emissions

USA

1.1

OECDE

0.8

Japan

0.1

CANZ

0.3

EEFSU

0.3

China

0.7

India

0.5

MOPEC

0.2

ROW

1.7

World

5.7

Source: Houghton et al. (1995).

Alternative Approaches to GWPs

A multi-gas approach to climate policy raises the issue of trade-offs among gases. In this chapter we explore two alternatives to GWPs—one based on cost-effectiveness and the other based on the balancing of costs and benefits. In each case the relative contribution of each gas to achieving the goal is an endogenous output rather than an exogenous input. That is, we make an endogenous calculation of the incremental value of emission rights for CH4 and N2O relative to CO2 and examine how the relationships might change over time.

In a cost-effectiveness analysis, the goal is to minimize the cost of achieving a particular objective. In the area of climate policy, objectives have included limits on emissions; cumulative emissions; atmospheric concentrations; the rate of temperature change; absolute temperature change; and damage.

For purposes of illustration, we begin by assuming that the goal of climate policy is to limit the increase in mean global temperature over the next two centuries. Figure 25.1 shows the price of emission rights for ceilings of 2° and 3°C. Not surprisingly, the rate of increase of these values is greater with the more stringent target. The calculations are made under the assumption of full where and when flexibility. With the first, reductions take place where it is cheapest, regardless of geographical location. With the second, they take place when it is cheapest.6

Figure 25.2 shows the prices of CH4 and N2O relative to that of carbon. It also shows the 100-year GWPs for each gas.7 Notice that the relative prices vary over time. This is particularly so for CH4. With a relatively short lifetime, a ton emitted in the

2020 2040 2060 2080 2100
0 0

Post a comment