CBO Statement of Peter R. Orszag …

            CBO

        Statement of
       Peter R. Orszag
          Director



   Issues in
Climate Change
    Presentation for the
CBO Director's Conference on
     Climate Change


         November 16, 2007




     CONGRESSIONAL BUDGET OFFICE
       SECOND AND D STREETS, S.W.
         WASHINGTON, D.C. 20515
                                     Note
Materials presented at this conference are available on CBO's Web site in a special
collection on the topic of climate change (www.cbo.gov/publications).
Issues in Climate Change
Global climate change is one of the nation's most significant long-term policy
challenges. Human activities are producing increasingly large quantities of green-
house gases, particularly carbon dioxide (CO2). The accumulation of those gases
in the atmosphere is expected to have potentially serious and costly effects on
regional climates throughout the world. The magnitude of such damage remains
highly uncertain. But there is growing recognition that some degree of risk exists
for the damage to be large and perhaps even catastrophic.

Reducing greenhouse-gas emissions would be beneficial in limiting the degree of
damage associated with climate change. However, decreasing those emissions
would also impose costs on the economy--in the case of CO2, because much eco-
nomic activity is based on fossil fuels, which release carbon in the form of carbon
dioxide when they are burned. Most analyses suggest that a carefully designed
program to begin lowering CO2 emissions would produce greater benefits than
costs.

Employing incentive-based policies to reduce CO2 emissions would be much more
cost-effective than using more-restrictive command-and-control approaches (such
as imposing technology standards on electricity generators). Incentive-based poli-
cies use the power of markets to identify the least costly sources of emission
reductions. Thus, they can better reflect technological advances, differences
between industries or companies in their ability to make low-cost emission reduc-
tions, and changes in market conditions.

Policymakers can choose between two general forms of incentive-based poli-
cies--those that limit the overall level of emissions (so-called quantity instru-
ments) or those that reduce emissions by raising their price (so-called price instru-
ments). The simplest price-based mechanism would be a tax on emissions. Under
a tax, a levy would be imposed on each ton of CO2 emissions or on each ton of car-
bon that is contained in fossil fuels (and which is ultimately released in the form of
CO2). The simplest quantity-based mechanism would be a cap-and-trade program.
Under such a program, policymakers would set a limit (cap) on total emissions
during some period and would require regulated entities to hold rights, or allow-
ances, to the emissions permitted under that cap. After allowances were initially
distributed, entities would be free to buy and sell them (the trade part of the pro-
gram). Reducing emissions to the level required by the cap would be accomplished
mainly by reducing demand for carbon-based energy through increasing its price.1
Those price increases could provide an effective financial incentive for firms and
households throughout the economy to take actions that would decrease emissions.

1. Emissions could also be reduced to some extent through carbon sequestration, which is the cap-
   ture and long-term storage of CO2 emissions underground (geological sequestration) or in veg-
   etation or soil (biological sequestration). For more information, see Congressional Budget
   Office, The Potential for Carbon Sequestration in the United States (September 2007).
The size of the required price increase would depend on the extent to which emis-
sions had to be reduced--larger reductions would require larger price increases to
reduce demand sufficiently.

Policymakers would have several key decisions to make in designing a cap-and-
trade program. One such decision would be whether to sell emission allowances or
give them away. Policymakers' decisions about how to allocate the allowances
could have significant effects on the overall economic cost of achieving a given
cap on CO2 emissions, as well as on the distribution of gains and losses among
U.S. households. Another key decision for policymakers is determining the appro-
priate level at which to set the cap. More stringent caps would lead to lower emis-
sions, which in turn would reduce future risks but raise near-term costs. The choice
of stringency is further complicated by the fact that the benefits (reductions in
future damage) and costs of alternative levels of stringency are both uncertain.


Trends in Emissions and Temperatures
Human activities--industry, transportation, power generation, land use--are pro-
ducing large quantities of greenhouse gases, particularly carbon dioxide. Those
gases are accumulating in the atmosphere more rapidly than natural processes can
remove them. Atmospheric concentrations of carbon dioxide have risen from 280
parts per million (ppm) in the pre-industrial era to about 380 ppm today. Concen-
trations of other greenhouse gases have also risen, but so have concentrations of
gases that have a cooling effect. The overall net increase in warming potential is
currently roughly equivalent to that of CO2 alone.

Rising concentrations of greenhouse gases are gradually warming the global cli-
mate, contributing to an increase of about 1.4°F in the average global temperature
since the middle of the 19th century. Concentrations will continue to rise--and the
climate will probably continue to warm--unless global emissions are reduced well
below their current levels. Stabilizing emissions (that is, the amount of greenhouse
gases released each year into the atmosphere) is merely the first step in a long pro-
cess of stabilizing concentrations (that is, the total amount of greenhouse gases
remaining in the atmosphere).

At present, however, global emissions are rising rapidly, and depending on the
growth of emissions and the climate's response, the global climate could warm by
another 2°F to 12°F or even more over the next century. Roughly 1°F will result
from accumulations that have already occurred. At the higher end of the range of
projections, the amount of warming to come would be at least as great as the
amount that has occurred since the depths of the last ice age and could produce
unexpected, rapid, and very costly changes in the Earth's climate.




                                         2
Uncertainty About Potential Damage
The climate problem is substantially complicated by uncertainty. Those uncertain-
ties make it difficult to determine the full range of possible outcomes, the likeli-
hood of particular outcomes, and the most appropriate policy response to address
such potential outcomes. The important uncertainties include these:

B   Future levels of greenhouse gas emissions and concentrations that will result
    from global population trends, technological developments, and economic
    growth. As a result of those uncertainties, any projection of cumulative emis-
    sions of greenhouse gases (measured in terms of CO2 equivalent) over the next
    century could easily err by plus or minus 50 percent. The accumulation of those
    gases in the atmosphere also depends on how rapidly they will be absorbed by
    the oceans and forests--another source of significant uncertainty.

B   The magnitude and timing of the global climate's response to rising concentra-
    tions of greenhouse gases. Given current scientific uncertainties, any projection
    of the climate's full response to a given increase in greenhouse-gas concentra-
    tions could err by plus or minus 50 percent. Moreover, that full response will
    unfold at an uncertain pace over decades to centuries.

B   The types, magnitude, and timing of damage that will result from changes in cli-
    mate. Some of the changes from warming are likely to be beneficial in some
    regions--for example, milder winters, longer growing seasons, and greater
    rainfall. However, many of the changes are likely to impose economic and
    social costs in other regions--for example, by melting ice caps and ice sheets;
    raising sea levels; altering agricultural seasons, ecological zones, and the range
    of infectious diseases; affecting the acidity of the oceans; increasing the vari-
    ability of weather patterns; and increasing the severity of storms and droughts.
    All of those types of effects are likely to be more severe the greater (and more
    rapid) the degree of warming. In addition to those largely expected effects,
    experts argue that there is a small but uncertain chance that warming could trig-
    ger abrupt and unforeseen changes in climate that could be associated with
    unexpectedly large economic costs. Finally, some effects, such as the extinction
    of species, are not only difficult to project but difficult to quantify in economic
    terms.


Choosing Policy Goals in the Face of Uncertainty
The significant uncertainties about potential future damage and the presence of
possible catastrophic risk complicate the process of setting realistic goals for cli-
mate policy: Targets for emissions cannot guarantee that concentration goals will
be reached, and even choosing a target for atmospheric concentrations of green-
house gases will not guarantee that a temperature target will be achieved, let alone
guarantee that the amount of damage will be limited. For example, if policies were
implemented to keep concentrations from rising above 550 ppm of CO2 equivalent



                                           3
by 2100, warming over the next century would most likely be around 4°F, but
there is a small probability that warming could be nearly twice as large.

An important implication is that effective climate policies should consider not only
the most likely outcomes near the middle of the range, but also the less likely but
more damaging possible consequences of climate change. The possibility of
unlikely outcomes involving extreme costs provides an economic motivation for
greater action to moderate the growth of emissions than might be justified by more
likely outcomes. Just as individuals tend to avoid risky behavior and buy insurance
to reduce the harm from unlikely but extreme events, some of society's resources
may best be devoted to addressing the most severe risks associated with climate
change even if the worst outcomes ultimately do not materialize.

In the context of climate change, reducing the risk of future damage involves
reducing the trajectory of future emissions: The amount of damage that would be
likely to result from a ton of emissions today would be less if the trajectory of
future emissions was expected to stabilize the atmospheric concentration of green-
house gases at a relatively low level. However, reducing emissions would impose
economic costs. The difficulty for policymakers is to determine the appropriate
cost to be incurred today to reduce what may be a small risk of a potentially cata-
strophic event in the future.

Different methods of incorporating information about risks and calculating their
present value can lead researchers to come up with very different estimates of the
benefits of reducing greenhouse-gas emissions even when they assume a similar
trajectory of future emissions. For example, one well-known, recently updated
study, by William Nordhaus, estimates that the benefit of reducing a ton of emis-
sions today under a trajectory of emissions that would lead to a concentration of
550 ppm would be just over $9 per metric ton of carbon dioxide.2 In contrast, a
comprehensive report by the British government, the Stern Review, estimates that,
given the same trajectory, the benefit of reducing a ton of emissions would be over
three times as high. The difference in the two estimates stems largely from differ-
ent ways of incorporating and valuing small probabilities of large future damage.

Policies that stabilize the atmospheric concentration of CO2 at lower levels reduce
the potential for larger damage in the future but also impose greater costs in the
near term. Differences in the measurement and evaluation of potential future dam-
age lead researchers to come to different conclusions about what targets to aim for.
For example, the Nordhaus study concludes that an ideal policy, pursued by all
countries simultaneously, would involve gradually raising the price of emissions
and would ultimately limit CO2 concentrations to about 650 ppm, resulting in a
50 percent chance of limiting the ultimate increase in global temperature to a little

2. Value inferred from estimates of optimal emissions price for carbon in 2015 in William
   Nordhaus, The Challenge of Global Warming: Economic Models and Environmental Policy
   (September 11, 2007), available at http://nordhaus.econ.yale.edu/dice_mss_091107_public.pdf.




                                               4
over 6°F. The Stern Review, in contrast, suggests that it would be most appropriate
to aim to stabilize the atmospheric concentration of greenhouse gases at between
450 ppm and 550 ppm of carbon dioxide equivalent. Such lower concentrations
would be much more likely to keep global temperature increases below 6°F than
the level recommended by Nordhaus but are also likely to be considerably more
expensive.

Climate policy is further complicated by issues of timing, geography, international
political economy, and history. Timing matters because unlike short-term invest-
ments whose costs and benefits are enjoyed by the same individuals, climate pol-
icy could impose significant costs on current generations who will not live to see
the vast bulk of the benefits, which will instead accrue to generations that have not
yet been born--and who are likely to be much richer than the current generations.
Geography matters because much of the cost is likely to fall disproportionately on
fossil-fuel-producing regions and on developed regions with extensive infrastruc-
ture based on fossil fuels, but much of the benefit would tend to accrue to develop-
ing countries in tropical and subtropical regions, which would be expected to bear
much of the damage associated with climate change. International political econ-
omy issues are central because countries can directly control only their domestic
emissions; as a consequence, managing the atmospheric stock of greenhouse gases
will require ongoing, long-term cooperation among all countries with significant
emissions. Finally, history matters because developed countries are responsible for
the bulk of emissions that have already occurred, and developing countries--
which will contribute most of the growth in emissions over the coming century--
may be unwilling to incur significant costs in restricting emissions unless devel-
oped countries bear the bulk of the costs, at least until developing countries have
achieved substantial income gains.


Comprehensive Policy Design
Designing policies to address climate change is complicated by uncertainty about
the damage that might result from unchecked emissions and uncertainty about the
cost of reducing those emissions. The cost of reducing emissions will depend on
various factors, including the magnitude of future emissions (which, in turn,
depends on the pace of population and economic growth), relative prices for fossil
fuels, the development of new technologies for producing energy without creating
emissions, and the willingness of individuals to adopt those technologies.3 A prag-
matic climate policy will probably involve a sequence of decisions based on the

3. For a more detailed description of the sources of uncertainty associated with the cost of reduc-
   ing emissions (as well as the benefits), see Congressional Budget Office, Uncertainty in Ana-
   lyzing Climate Change: Policy Implications (January 2005).




                                                 5
gradual accumulation of information and the resolution of uncertainties.4 For
such an approach, policies that can be easily modified over time would offer
advantages.

A flexible approach to dealing with climate change could include three different
policy strategies:

B   Researching the problem and developing technologies to address it,

B   Adapting to a warmer climate, and

B   Reducing greenhouse-gas emissions.

In addition, a comprehensive climate policy would inevitably involve coordinating
U.S. policies with those of other countries that are major emitters of greenhouse
gases.

Research and Development
Research is an essential part of any comprehensive strategy to address potential
changes in the climate. Research is likely to provide benefits by helping to resolve
uncertainties (including uncertainties about the physical damage that might result
from climate change as well as the substantial uncertainties about how to evaluate
that damage) and by leading to the development of technologies to cut emissions.
Such technologies could include improvements in energy efficiency; advances in
low- or zero-emissions technologies (such as nuclear, wind, or solar power); and
the development of sequestration technologies. Federal support would probably be
most cost-effective if it went toward basic research on technologies that are in the
early stages of development. Such research, which often creates knowledge that is
beneficial to other firms but does not generate profits for the firm conducting the
research, is likely to be underfunded in the absence of government support.

Adaptation
Some degree of warming is inevitable from emissions that have already occurred,
and even very aggressive emission restrictions are unlikely to stabilize concentra-
tions for many years to come. In light of the potential for future temperature
increases, adaptation could play an important role in any effective climate strategy.
Unlike mitigation policy, which could be implemented largely with a single instru-
ment--for instance, a tax on emissions or an aggregate emissions cap--efforts to
promote adaptation are likely to be more diffuse, involving numerous policies that

4. That sequence is sometimes referred to as an "act, then learn, then act" process. A discussion of
   that approach is provided in A.S. Manne and R.G. Richels, Buying Greenhouse Insurance: The
   Economic Costs of CO2 Emission Limits (Cambridge, Mass.: MIT Press, 1992). Another useful
   discussion is found in L. James Valverde A. Jr., Henry D. Jacoby, and Gordon M. Kaufman,
   "Sequential Climate Decisions Under Uncertainty: An Integrated Framework," Journal of
   Environmental Modeling and Assessment, vol. 4, no. 2-3 (1999), pp. 87­101.




                                                 6
address different areas. Those policies, many of which are likely to yield benefits
even if climate change proved to be relatively benign, could include the following:
B   Promoting the efficient use of water resources (which are likely to become
    scarcer in some regions) through prices that reflect scarcity or through the
    establishment of markets for water;

B   Encouraging the development of low-cost technologies for desalinating
    seawater;

B   Encouraging the preservation of green corridors that would allow plant and ani-
    mal species to migrate as their habitat changed;

B   Facilitating the relocation of people living in low-lying areas that are prone to
    increased flooding;

B   Improving health care and pest control; and

B   Encouraging the development and use of drought-resistant crops.

Reducing Emissions
Policy choices about how much to reduce greenhouse-gas emissions involve sig-
nificant trade-offs: If left unchecked, emissions could ultimately lead to costly and
perhaps even catastrophic damage, yet emissions restrictions impose economic
costs. Reducing carbon dioxide emissions could be particularly challenging
because carbon-based fossil fuels are an integral part of the global economy,
accounting for almost 90 percent of global energy use. As a result, policies for
reducing CO2 emissions must be designed in a way that facilitates a smooth, cost-
effective transition to a less fossil-fuel-dependent economy, or to the sequestration
of fossil-fuel-related emissions. (The design of efficient policies for reducing CO2
emissions is discussed below.)

International Coordination
Because the causes and consequences of climate change are global, the most cost-
effective policy approaches require a coordinated international effort. In particular,
reductions in emissions could be achieved most cost-effectively if low-cost oppor-
tunities for mitigation were taken advantage of, regardless of where those opportu-
nities were located around the globe. Despite the need for international coopera-
tion to address climate change, the nature of the climate problem may make
international agreement difficult to reach because the distribution of likely costs
and benefits leaves countries and regions with considerably divergent interests.
Furthermore, developing countries, which contributed a relatively small share of
historical emissions but are expected to contribute a growing share of future ones,
may object to having their development constrained by emissions restrictions.
Finally, the challenges associated with enforcing a global solution may make some



                                           7
nations reluctant to participate, adding a source of uncertainty about how cost-
effective the policies will be. Regardless of the challenges, U.S. policies will be
most effective if they are part of a global effort.


Incentive-Based Policies for Reducing CO2 Emissions
Any effort to limit CO2 emissions would have two principal effects: It would pro-
duce long-term economic benefits by avoiding some future climate-related dam-
age, and it would impose immediate economic costs by reducing the use of fossil
fuels.

Employing incentive-based policies to reduce emissions would help minimize the
cost of reducing emissions by any given amount because they would use the power
of markets to identify the least expensive sources of emission reductions. Thus,
they can better accommodate technological advances, differences between indus-
tries or companies in their ability to make low-cost emission reductions, and
changes in market conditions.

Two alternative incentive-based approaches for reducing CO2 emissions are to tax
them or to establish a cap-and-trade system for them. Either a tax or a cap would
be most efficient (that is, would best balance expected benefits and costs) if it was
designed to gradually become more stringent over time--meaning the tax would
gradually rise or the cap would become tighter. Such an approach would best
reflect the present value of avoided future damage (the benefit of reducing a ton of
emissions), which would take on greater weight as larger potential damage became
closer in time. Further, such an approach would allow a smooth transition to a less
carbon-intensive economy, allowing firms and households time to gradually
replace capital equipment with alternatives that are more efficient, use less carbon
intensive fossil fuels (such as natural gas rather than coal) or use renewable energy
sources (such as wind or solar).

Efficiency Advantages of a Tax on CO2 Emissions
Although both types of incentive-based approaches are significantly more efficient
than command-and-control policies, studies typically find that over the next sev-
eral decades, a well-designed and appropriately set tax would yield higher net ben-
efits than a corresponding cap-and-trade approach. A tax creates relative certainty
about the cost of emission reductions each year, because firms will undertake such
reductions until the cost of decreasing emissions by another ton just equals the tax
on an additional ton of emissions. A cap-and-trade program, by contrast, creates
relative certainty about the total quantity of emission reductions each year,




                                          8
because the cap limits total annual emissions. In terms of the impact on the cli-
mate, however, it does not matter greatly whether a given cut in emissions occurs
in one year or the next.5
From that perspective, a tax has an important advantage: It allows more emission
reductions to take place in years when they are relatively cheap. Various factors
can affect the cost of emission reductions from year to year, including the weather,
the level of economic activity, and the availability of new low-carbon technologies
(such as improvements in wind-power technology). By shifting emission-
reduction efforts into years when they are relatively less expensive, a tax can yield
a given quantity of emission reductions at a lower cost than can a cap-and-trade
program with specified annual emission levels. In addition, by avoiding the poten-
tial volatility of allowance prices that might result from a rigid annual cap, a tax
could be less disruptive for affected companies. Provided that the tax was set at a
level that reflected the expected benefit of reducing an additional ton of CO2
emissions, the tax would provide a motivation for firms and households to reduce
emissions up to the point at which the cost of doing so was equal to the resulting
expected benefits.

The relative advantages of a tax and a cap-and-trade program could change over
time as new information became available. For example, because a cap creates rel-
ative certainty about the level of emissions, it could become more efficient than a
tax if scientists determined that additional emissions were likely to trigger a sharp
increase in damage, or if new technologies offered the opportunity to make
extremely large cuts in emissions at a low and fairly constant cost. Analysts who
have tried to define more precisely the conditions under which a cap would be
more efficient than a tax have found those conditions to be quite narrow and not
likely to be relevant in the near term. Specifically, scientists would need to have
fairly precise knowledge about the level of an emissions threshold--beyond which
additional emissions would trigger a sharp increase in total global damage--and
such a threshold would have to be sufficiently close that policymakers would want
to make very large cuts in emissions each year to avoid crossing it.6 In the absence

5. Although it is difficult to measure, the long-term cumulative nature of climate change implies
   that the benefit of emitting one less ton of CO2 in a given year--referred to as the marginal ben-
   efit--is roughly constant. In other words, the benefit in terms of averted climate damage from
   each additional ton of emissions reduced is roughly the same as the benefit from the previous
   ton of emissions reduced, and shifting the reductions from one year to another does not materi-
   ally affect the ultimate impact on the climate. In contrast, the cost of emitting one less ton of
   CO2 in a given year--the marginal cost--tends to increase with successive emission reduc-
   tions. The reason is that the least expensive reductions are made first and progressively more-
   expensive cuts would then have to be made to meet increasingly ambitious targets for emission
   reductions.
6. See William A. Pizer, Climate Change Catastrophes, Discussion Paper 03-31 (Washington,
   D.C.: Resources for the Future, May 2003).




                                                 9
of those conditions, a tax offers a more efficient approach for reaching a multiyear
emission-reduction target.

Although a tax is generally a more efficient policy, the efficiency of a cap-and-
trade approach can be enhanced by various design features. In addition, some par-
ticipants in the policy discussion believe that analytical comparisons of a tax and a
cap-and-trade system ignore the idea that policymakers may be more inclined to
set a tight cap than a correspondingly high tax.

Enhancing the Efficiency of a Cap-and-Trade System
Policymakers could capture some of the efficiency advantages of a tax, while
maintaining the structure of a cap-and-trade program, by adding features that
would help keep the price of allowances in line with the anticipated benefits of
emission cuts. For example, a price cap--typically referred to as a safety valve--
and a price floor could keep the price of allowances from climbing too far above or
falling too far below the anticipated benefits of emission reductions. The govern-
ment could implement a safety valve by agreeing to sell as many allowances as
firms wanted to buy at a specified price. (If the safety valve was triggered, emis-
sions would exceed the level of the cap.) A price floor could be implemented if
policymakers decided to sell a significant fraction of the allowances in an auction
and set an auction reserve price. Alternatively, rather than setting a price floor, pol-
icymakers could allow firms to "bank" allowances when the cost of reducing
emissions was low and to use those allowances in the future when costs were
higher. Banking would keep the price of allowances from falling too low, provided
that prices were expected to be higher in the future.


Key Decisions in Designing a Cap-and-Trade Program
By establishing a cap-and-trade program, policymakers would create a new com-
modity: the right to emit CO2. The emission allowances--each of which would
represent the right to emit, say, one ton of CO2--would have substantial value.
Based on a review of the existing literature and the range of CO2 policies now
being debated, the Congressional Budget Office (CBO) estimates that the value of
those allowances could total between $50 billion and $300 billion annually (in
2006 dollars) by 2020. The actual value would depend on various factors, includ-
ing the stringency of the cap (which would need to grow tighter over the years to
keep CO2 from continuing to accumulate), the possibility of offsetting CO2 emis-
sions through carbon sequestration or international allowance trading, and other
features of the specific policy selected.

Policymakers would need to decide how to allocate the allowances that would cor-
respond to each year's CO2 cap. One option would be to have the government cap-
ture their value by selling the allowances, as it does with licenses to use the elec-
tromagnetic spectrum. Another possibility would be to give the allowances to
energy producers or some energy users at no charge. The European Union has used



                                          10
that second approach in its two-year-old cap-and-trade program for CO2 emis-
sions, and nearly all of the allowances issued under the 12-year-old U.S. cap-and-
trade program for sulfur dioxide emissions (which contribute to acid rain) are dis-
tributed in that way.

Consequences of Allowance-Allocation Decisions
Policymakers' decisions about whether to sell the allowances or to give them
away would have significant implications for the distribution of gains and losses
among U.S. households and for the overall cost of the policy (for a given level of
stringency).

Distributional Consequences. The ultimate distributional impact of a cap-and-
trade program would be the net effect of two distinct components: the distribution
of the cost of the program (including the cost of paying for the allowances) and the
distribution of the allowances' value (because someone will pay for them, some-
one will benefit from their value). Market forces would determine who bore the
costs of a cap-and-trade program, but policymakers would determine who received
the allowance value. The ultimate effect could be either progressive or regressive.

Obtaining allowances--or taking steps to cut emissions to avoid the need for such
allowances--would become a cost of doing business for firms that were subject to
the CO2 cap. However, those firms would not ultimately bear most of the costs of
the allowances. Instead, they would pass along most such costs to their customers
(and their customers' customers) in the form of higher prices. By attaching a cost
to CO2 emissions, a cap-and-trade program would thus lead to price increases for
energy and energy-intensive goods and services that contribute the most to those
emissions. Such price increases would stem from the restriction on emissions and
would occur regardless of whether the government sold emission allowances or
gave them away. Indeed, the price increases would be essential to the success of a
cap-and-trade program because they would be the most important mechanism
through which businesses and households were encouraged to make investments
and behavioral changes that reduced CO2 emissions. The rise in prices for energy
and energy-intensive goods and services would be regressive--that is, they would
impose a larger burden, relative to income, on low-income households than on
high-income households.

In addition to imposing relatively large burdens on low-income households, the
higher prices that would result from a cap on CO2 emissions would reduce demand
for energy and energy-intensive goods and services. Thus, those price increases
would create losses for some current investors and workers in those sectors. Such
investors could see their stock values decline, and workers could face the risk of
unemployment as jobs in those sectors were cut. Stock losses would tend to be
widely dispersed among investors, because shareholders typically have diversified
portfolios. In contrast, the costs borne by existing workers would probably be con-
centrated among relatively few households and, by extension, their communities.



                                         11
Although the price increases triggered by a cap-and-trade program for CO2 emis-
sions would be regressive, the policy's ultimate distributional effect would depend
on policymakers' decisions about how to allocate the emission allowances. Those
allowances would be worth tens or hundreds of billions of dollars per year. Who
received that value would depend on how the allowances were distributed.

Lawmakers could more than offset the price increases experienced by low-income
households or the costs imposed on workers in particular sectors by providing for
the sale of some or all of the allowances and using the revenue to pay compensa-
tion. Conversely, giving all or most of the allowances to energy producers to offset
the potential losses of investors in those industries--as was done in the cap-and-
trade program for sulfur dioxide emissions--would exacerbate the regressivity of
the price increases. On average, the value of the CO2 allowances that producers
would receive would more than compensate them for any decline in profits caused
by a drop in the demand for energy and energy-intensive goods and services that
cause emissions. As a result, the companies that received allowances could experi-
ence "windfall" profits, with the government regaining only part of that windfall
through corporate income taxes.

Those profits would accrue to shareholders, who are primarily from higher-income
households, and would more than offset those households' increased spending on
energy and energy-intensive goods and services. Low-income households, by con-
trast, would benefit little if allowances were given to energy producers for free,
and they would still bear a disproportionate burden from price increases.

Overall Economic Cost Consequences. The ways in which lawmakers allocate
the revenue from selling emission allowances could affect not only the distribu-
tional consequences but also the total economic cost of a cap-and-trade policy. For
instance, the government could lessen the cost by using the revenue from auction-
ing allowances to reduce existing taxes that tend to dampen economic activity--
primarily, taxes on labor, capital, or personal income. Research indicates that a
CO2 cap would exacerbate the economic effects of such taxes: The higher prices
caused by the cap would lower real (inflation-adjusted) wages and real returns on
capital, which would be equivalent to raising marginal tax rates on those sources
of income. Using the allowance value to reduce such taxes could help mitigate that
adverse effect of the cap. Alternatively, policymakers could choose to use the rev-
enue from auctioning allowances to reduce the federal deficit. If that reduction
lessened the need for future tax increases, the end result could be similar to dedi-
cating the revenue to cutting existing taxes. The decision about whether or not to
sell the allowances and use the proceeds in ways that would benefit the economy
could have significant impacts for the economywide cost of a cap-and-trade
program.



                                        12
Consequences of Stringency Decisions
Researchers have produced a large number of studies of the near-term and long-
term costs of achieving various levels of stringency in reducing emissions. Several
recent analyses focus on the likely costs to the U.S. economy of various legislative
proposals put forward in recent years. Within the federal government, both the
Environmental Protection Agency and the Department of Energy's Energy Infor-
mation Administration have analyzed some versions of those legislative proposals
at the request of Members of Congress. In addition, researchers at the Joint Pro-
gram on the Science and Policy of Global Change at the Massachusetts Institute of
Technology (MIT) have produced a comprehensive analysis of several long-term
emission targets that approximate some of the legislative proposals currently under
consideration. The National Commission for Energy Policy has also proposed
a comprehensive energy policy strategy that would include restrictions on
greenhouse-gas emissions, and Duke University's Nicholas Institute has analyzed
at least one of the bills.

Like the proposals themselves, the scenarios analyzed in those studies specify a
number of different levels of stringency. In general, all of the proposals involve an
increasing level of stringency over time. Some impose hard caps, but others pro-
vide for a price limit or safety valve. Most allow for at least some degree of bank-
ing and borrowing of emission allowances. Many scenarios involve limited sec-
toral coverage, usually exempting most emissions from nonelectric energy use in
buildings (such as home heating). Many proposals and scenarios allow covered
entities to purchase offsets from one or more types of sources--in uncovered sec-
tors, through sequestration of carbon in domestic forests and soils, or from interna-
tional sources--as a substitute for their own actions.

Most of the models used for those studies are of a type referred to as general-
equilibrium models, in which idealized households supply labor and capital to
idealized firms that supply a variety of goods and services back to the households.
Most of the models incorporate forward-looking behavior; that is, the idealized
agents in the model correctly anticipate future developments in the model and
adjust their behavior accordingly. With such assumptions, most of the models
implicitly assume that households and firms respond rather quickly, rationally, and
flexibly to changes in policy--an assumption that may overstate many people's
abilities to accommodate such changes.

Many models include considerable detail about the different types of technology in
the energy sector. One model, the Energy Information Administration's National
Energy Modeling System--NEMS--is not a general-equilibrium model, but is
instead a highly detailed representation of the energy sector integrated with a mac-
roeconometric model of the U.S. economy. Some models include the entire global
economy, linking countries through international trade flows; others include only
the domestic economy and include only rather cursory representations of interna-
tional linkages. Most of the modeling efforts draw on the Environmental Protec-



                                         13
tion Agency's estimates of abatement costs for carbon dioxide from nonenergy
sources and for other greenhouse gases, though most other groups modify those
estimates in some way. Those models estimate only the policies' costs; they do not
provide for feedbacks from the mitigation policies to climate benefits.

Different models provide somewhat different types of cost estimates for the mod-
eled policies. All of them provide estimates of the annual prices of allowances and
offsets (which can differ, depending on whether the use of offsets is restricted or
not), or of the level of the annual tax rate on emissions necessary to achieve a
given level of emissions. All provide estimates of macroeconomic impacts, such
as changes in inflation-adjusted gross domestic product and personal consumption
expenditures. Some models provide more comprehensive estimates of changes in
households' economic well-being--what economists refer to as welfare impacts.

Even for a given policy proposal, the models' cost estimates can vary because of
differences in their assumptions about future economic growth and emission
trends, about how firms and households will respond to changing prices, and about
what types of technologies are likely to be available and accepted at different
times. None of the models explicitly provides for a channel through which policies
might influence the pace of development of new technologies, although in some
models, such influences are modeled implicitly. The models also differ in whether
or not they consider the effects on domestic costs of climate policies pursued in
other countries; such effects can be quite important.

Considering the extraordinary complexity of the domestic and global economies
and the wide range of activities associated with greenhouse-gas emissions, no sin-
gle modeling framework or modeling scenario provides a fully comprehensive
treatment of all the potential effects. A realistic assessment of the results requires
modelers and users to consider a range of models and results to draw broader
conclusions.
With a wide range of modeled policies, a wide range of baseline projections, and a
wide range of reasonable differences in assumptions about how the economy
might respond to policies, the results vary considerably. Nevertheless, a few robust
insights emerge:
B   First, there is significant uncertainty about the price required to induce a spe-
    cific level of emission reductions, and significant uncertainty about the level of
    reductions that would be induced by a specific price. Higher prices induce
    greater reductions in emissions; conversely, greater stringency in constraining
    emissions requires higher prices and imposes greater welfare impacts.

B   Second, even substantial reductions in emissions are generally associated with
    relatively modest impacts on consumption--especially compared with the
    expected growth in consumption over the next few decades.



                                          14
B   Third, in the near term--that is, roughly through 2025--many reductions are
    likely to come from relatively inexpensive sources of greenhouse gases outside
    the energy sector. Within the energy sector, reductions in overall energy use are
    likely to be relatively modest; instead, reductions are likely to come predomi-
    nantly from the electric utility sector, mainly by shifting fuel use away from
    coal and toward natural gas and nuclear power. Sequestration and international
    offsets could yield significant reductions at relatively modest cost, assuming
    moderate administrative costs.

B   Finally, over the longer term, domestic and international offsets are likely to
    become increasingly less important, and most reductions are likely to come
    from the energy sector. Technology will play a critical role in that process, but
    current understanding provides little consensus as to which technologies will
    dominate. It appears likely that a combination of several technologies could
    play a role: carbon capture and sequestration, nuclear power, and biofuels.
    Other renewable energy sources and conservation appear likely to play rela-
    tively limited roles.




                                          15