Information about http://www.fiscalpolicy.org./northeastRGGI.pdf
Regional Initiatives to Reduce Greenhouse Gasses: The…
Tags: auspices, competitiveness, current state, eastern canadian premiers, electric utilities, electricity sector, england governors, further reductions, governor pataki, greenhouse gas emissions, greenhouse gasses, jim barrett, national model, number of states, power producers, redefining progress, regional initiatives, state budget, state initiatives, state jobs,Pages: 10
Language: english
Created: Wed Dec 29 18:00:06 2004
Display cached document
Page 1
Page 2
Page 3
Page 4
Page 5
Page 6
Page 7
Page 8
Page 9
Page 10
Regional Initiatives to Reduce Greenhouse Gasses:
The Crucial Importance of Auctioning Permits for Jobs,
Competitiveness, and Equity
By J. Andrew Hoerner1 Redefining Progress
A number of states in the Northeastern U.S. have now committed to reducing their
greenhouse gas emissions. Two separate initiatives are underway in the region. The first
is a comprehensive greenhouse gas emissions agreement under the auspices of the
Conference of New England Governors and Eastern Canadian Premiers.2 Its purpose is to
return greenhouse gas emissions to 1990 levels by 2010, with further reductions in
subsequent years.3 The second is an electricity-sector only initiative consisting of state
commitments in response to a call from Governor Pataki of New York for a regional cap-
and-trade system for electric utilities, known as the Regional Greenhouse Gas Initiative
(RGGI).4 Most states in the region have individual state initiatives as well.
There are many open questions concerning the implementation of these agreements. This
report discusses a key issues in the design of regional greenhouse gas emissions reduction
policies: whether the permits are auctioned or grandfathered (i.e. given away to power
producers or utilities). We show that, relative to a grandfathered system, auctioning
improves economically efficiency, help to solve the current state budget crises in the
region, is more distributionally fair, creates in-state jobs and preserves the
competitiveness of the region. Though some have argued that a grandfathered system is
politically easier to put in place, we believe that the much greater social costs of
grandfathering will make grandfathered systems politically unviable in the long run.
Conversely, the greater efficiency and improved equity of an auctioned system will help
to stabilize it politically as it becomes a point of pride for the region and a national
model.
1
The author would like to thank Matt Elliot and Jim Barrett for helpful comments that much improved this
paper, and Paul Baer, Energy Resources Group, U.C. Berkeley, for extremely competent research
assistance on policies to address distributional issues.
2
Participating states are Connecticut, Maine, Massachusetts, New Hampshire, Rhode Island, and Vermont.
Participating provinces are Newfoundland, New Brunswick, Nova Scotia, Prince Edward Island, and
Quebec.
3
By a resolution adopted July 16-18, 2000, the Conference of New England Governors and Eastern
Canadian Premiers agreed to create a joint Climate Change Action Plan and a process to review and update
that plan. See http://www.scics.gc.ca/cinfo00/85007913_e.html for the text of the resolution. The 2001
Climate Change Action Plan sets a short-term goal of reducing greenhouse gas emissions to 1990 levels by
2010, to ten percent below 1990 levels by 2020. The plan's long-term goal is to reduce greenhouse gas
emissions to the level required to avoid any harmful impact on the climate, currently estimated to be 75 to
85 percent below current levels. See http://www.massclimateaction.org/pdf/NECanadaClimatePlan.pdf for
the Climate Change Action Plan.
4
The Pataki initiative proposes to create a regional cap-and-trade program for CO2 emissions from power
plants. Participating states as of August 27, 2004 are Connecticut, Vermont, New Hampshire, Delaware,
Maine, New Jersey, Pennsylvania, Massachusetts, and Rhode Island. See http://www.rggi.org/.
Economic Efficiency
A. Flexibility
It is now generally acknowledged that, where they are feasible, market-based approaches
are among the most cost-effective ways to reduce pollution reduction, 5 because they
allow emissions reductions to be made by the polluter who can achieve those reductions
at the lowest cost. Thus, the broader the coverage of, e.g. an emissions trading system,
the more that high-cost reductions can be exchanged for low-cost ones, and the greater
the savings that can be achieved.
For example, an electric sector trading program such as that proposed under the RGGI
would allow reductions to be accomplished by the firms and plants where they can be
achieved the most cheaply. By including the residential and transportation sectors,
reductions can be made in the sectors that can accomplish them at the lowest cost, further
reducing the total cost of the program. Such universal coverage would make sense under
the Governors' and Primers' agreement, and future expansions of the RGGI to cover
more sectors have been contemplated as well.6
Such comprehensive permitting systems recognize that we do not have perfect
information about the cost of reductions, either now or in the future. It harnesses the
creative power of the market to identify least-cost reductions wherever they may be
found.
B. Revenue recycling effects
In addition to the efficiencies that come with flexibility, emissions permitting systems
can raise revenue that can be used to reduce distorting taxes and thereby improve the
economic efficiency of the entire economy, or can be invested in new clean technologies
to help ease the transition to new, lower-emission ways of living and doing business.
Emission permits that raise revenue and are then used to cut taxes are estimated to have
only a quarter of the total economic cost of grandfathered permits. 7 Studies have also
shown that further economic benefits can be gained if a portion of the revenue is returned
in the form of assistance in adopting new clean technologies to ease (and accelerate) the
transition. 8
5
See, e.g., Robert N. Stavins, Lessons from the American Experiment with Market-Based Environmental
Policies Kennedy School of Government Working Paper No. RWP01-032 (April 2002) and papers cited
therein.
6
"After the cap-and-trade program for power plants is implemented, the states may consider expanding the
program to other kinds of sources."See http://www.rggi.org/about.htm.
7
Lawrence H. Goulder , Ian W.H. Parry and Dallas R. Burtraw "Revenue-Raising vs. Other Approaches to
Environmental Protection: The Critical Significance of Pre-Existing Tax Distortions" RAND Journal of
Economics 28,4 (Winter 1997): 708-731; Goulder, Lawrence H., Ian W. H. Parry, Roberton C. Williams
III, and Dallas Burtraw, 1998. "The Cost-Effectiveness of Alternative Instruments for Environmental
Protection in a Second-best Setting." Journal of Public Economics 72(3): 329-360.
8
See Edmonds, J., Roop, J. M., and Scott, M. J. (2000). Technology and the Economics of Climate Change
Policy (Washington DC: Pew Center of Global Climate Change); J. Andrew Hoerner and Benoît Bosquet
Environmental Tax Reform: The European Experience, Center For A Sustainable Economy: Washington,
DC (February 2001), section 6.4.4 and the citations contained therein.
In assessing the efficiency gains of revenue recycling, whether through cutting other
taxes, providing public services, or financing technology improvements, it is important to
understand that consumer prices (including industrial consumers) are the same whether
the permits are auctioned or grandfathered (given away based on emissions in a base
period). This is because in today's competitive power markets, prices are determined, not
by average costs, but by the cost of the last unit of production (marginal cost). When a
utility produces an additional kWh of power, it has to buy an additional emissions permit,
either from the state under an auction or from another utility under grandfathering. 9 The
result is that the consumer price goes up by the cost of the permit either way the only
difference is whether the money goes to power companies or to the state.
C. Grandfathered permits as a transfer from consumers to industry
In competitive electric generation markets, when the marginal cost of generation
increases by the abatement cost, the price increases by the same amount. This price
increase applies to all units of electricity sold, not just the final units. Thus there is a
transfer of wealth from consumers to producers. Figure 1 shows how this transfer works.
We assume a fixed demand for electricity equal to D to simplify the presentation.
Figure 1.
D S'
P2
S
P1
D
The initial, pre-policy price, P1, is set where the supply curve, S, intersects the demand
curve, D. When a permit system is put in place, it increases the cost of fossil-generated
electricity. This results in a new supply curve, S', which is higher than S for values of
demand that exceed the amount of non-fossil baseload generation. 10 The new price, P2, is
set at the intersection of the new supply curve S' and D. Total abatement cost paid by
generators is the area between S and S'. In the graph above, this is the triangle formed by
S, S', and D. Increased revenue is the rectangle formed by the vertical axis, D, and the
9
Alternatively, the utility can reduce its own emissions on other generation by enough to allow an
additional kWh of generation under the allowances that it already holds. Thus the equilibrium price of
permits, under either an auction or a grandfathering system, is the marginal abatement cost.
10
The quantity of electricity equal to the non-fossil baseload is the point where S' diverges from S.
two price lines, P1 and P2. 11 The revenue represented by this rectangle goes to generating
company stockholders if the permits are grandfathered model and to the public if the
permits are auctioned model.
Note that the total abatement cost paid by utilities is much smaller than the total
additional revenue collected by utilities. It is possible to set an upper bound on this ratio
by making the standard assumption that abatement costs increase as an increasing rate.
This is equivalent to assuming that the (marginal) abatement cost curve is bowed down
(convex). Under this assumption, which is certainly reasonable for moderate reduction
levels such as those now being considered, the total abatement cost is no larger than the
triangle in the graph above, though it could be considerably smaller (because a straight
line is the least "bowed-down" curve possible).12
One way of examining the subsidy to utilities provided by a grandfathered system is to
look at the ratio of increased revenue to abatement costs. To know this ratio precisely, we
would need detailed information on the shape of the cost curves. However, under the
simplifying assumption above, we can easily calculate the ratio of the upper bound of the
cost to total revenues as the ration of the rectangle (the revenue) to the triangle (the total
abatement cost). This measure has the added benefit of being invariant to the particular
level of price increase that is caused by the permitting system, i.e. it is the same
regardless of the difference between P1 and P2. We present the case of NY as a typical
example, based on 2000 data.
In 2000, NY utilities generated a total of approximately 149 million kWh of power. 45
percent of this, or about 68 million kWh, was non-fossil.13 Thus, total revenues would
constitute not less than 2.7 times total abatement costs. 14 Again, this is a lower bound.
For more realistic abatement cost functions, the ratio of new revenues to abatement costs
could be quite a bit greater. Estimates in the literature for national trading systems show
revenues exceeding abatement costs by a factor of roughly five to twenty. 15
11
We show this result with an inelastic (vertical) demand curve in order to simplify the presentation. For
realistic values of demand elasticity and moderate emissions reductions in, say, the zero to 20% range, the
results are similar to the simplified inelastic results above, with very slightly higher social costs and a
slightly lower transfer to producers.
12
If marginal abatement costs increase as the level of abatement per kWh increases, the S' curve is "bowed
down," shrinking the area between it and the initial supply curve, S. This is what one would expect if there
are abatement opportunities at different prices, and firms use the cheapest ones first. That is why the
triangle is an upper bound for this area.
13
New York is a substantial net importer of electricity. The non-fossil share presented here is calculated
based on our best estimate of the non-fossil share of imports in 2000: 34 percent of interstate imports and
80 percent of international imports.
14
The area of the rectangle is Base * Height, or 149 kWh*(P2-P1). The area of the triangle is ½ Base *
Height, or [81 kWh*(P2-P1)]/2.
15
Goulder, L.H. & Bovenberg, A.L. "Neutralizing the Adverse Industry Impacts of CO2 Abatement
Policies: What Does It Cost?"), in C. Carraro and G. Metcalf, eds., Behavioral and Distributional Effects of
Environmental Policies, University of Chicago Press, 2001; Smith, A. E. and Ross, M. T. Allowance
Allocation: Who Wins and Loses under a Carbon Dioxide Control Program? Report prepared by Charles
River Associates for Center for Clean Air Policy, Washington, D.C., (February 2002); Burtraw, D., Palmer,
K., Bharvirkar, R., and Paul, A. 2002. "The Effect on Asset Values of the Allocation of Carbon Dioxide
Emission Allowances." The Electricity Journal, June, pp. 51-62.
The table below shows the ratio of increased revenues to the utility to abatement costs
paid by the utility under the linear upper-bound discussed above and under a quadratic
approximation that is probably closer to the real value. This is shown for each of the
RGGI states. Because the increase in electric costs applies to all units of electricity, while
abatement costs apply only to fossil-generated electricity, the subsidy is greater in states
that have more non-fossil electricity in their generating mix.
Estimate of the Ratio of Power-Industry Subsidy to
Cost of Remediation for RGGI States
Linear upper Quadratic
bound approximation
Connecticut 4.6 6.9
Delaware 2.4 3.6
Maine 4.8 7.1
Massachusetts 2.6 3.9
New Hampshire 6.5 9.8
New Jersey 3.8 5.7
New York 3.5 5.2
Pennsylvania 3.2 4.8
Rhode Island 2.2 3.3
Vermont 3.2 4.9
Calculated by the author from 2000 data drawn from the Energy Information
Administration's State Energy Data System.
Revenue from a Carbon Permitting System
Many of the states in the Northeast are facing severe structural revenue shortages. In this
fiscal environment, turning a natural, non-tax source of new public revenue into a
corporate subsidy seems particularly irresponsible.
The table below shows the revenue from a carbon permitting system with a $20/ton
permit price. These numbers are calculated using a modified version of the State Carbon
Tax Model, developed by the author and others at the University of Maryland's Center
for Global Change. Of course, the actual revenues could be higher or lower, depending
on the magnitude of the carbon reduction, the extent of demand-side reductions that are
achieved through energy efficiency and other policies, the
An "X" in columns 2 or 3 indicates membership in the Conference of New England
Governors and Eastern Canadian Premiers accord and the Pataki agreement, respectively.
Column 5 shows the revenue from a comprehensive tradable carbon permit program,
while column 7 shows the revenue from an electricity-only permit system. Because we
believe that carbon-only systems create economically and politically unacceptable
regional disparities and other perverse incentives,16 columns 4 and 6 show revenues from
what we consider to be a more realistic permitting system. This system includes an
equalizing charge on electricity from nuclear power and large hydro-power equal to the
average permitting fee on fossil fuel-generated electricity.
Table 1. Revenues from Carbon Permitting Systems with a $20 Safety-Valve
by State and Coverage ($mill.)
1. 2. 3. 4. 5. 6. 7. 8.
State New RGGI Comprehensive Comprehensive Electricity: Electricity Motor
England Pollution & Carbon Permit Pollution & : Carbon Fuels:
Compact Energy Permit Energy Permit Carbon
Permit Permit
Connecticut X X 247 213 96 61 90
Delaware X 123 114 63 54 25
Maine X X 142 119 50 33 47
Maryland ? 590 515 301 226 159
Massachusetts X X 536 501 213 178 170
New Hampshire X X 113 94 38 20 41
New Jersey X 862 785 242 164 264
New York X 1,421 1,277 475 331 354
Pennsylvania X 1,823 1,511 865 553 350
Rhode Island X X 74 72 22 20 25
Vermont X X 51 42 11 5 22
Total, New X 1,162 1,042 429 318 394
England Compact
states
Total, RGGI X 5392 4728 2075 1419 1388
states
In the table above, the regional revenues from these more comprehensive pollution and
energy permitting systems are highlighted for the relevant initiative. The total emissions
permitting system revenue is highlighted for the New England Compact, and revenues
from the electricity-only permitting system is highlighted for states that have pledged to
join the RGGI. These revenues would increase for the immediately foreseeable future as
the safety valve rate rises.
It is worthwhile to note that in many cases these revenues would be sufficient to close a
substantial portion of the budget shortfalls plaguing these states.
Distributional Concerns
Lower-income households spend a proportionally larger share of their income on
necessities such as food and energy. As a result, the burden of any initiative that raises
the cost of energy, whether through regulation or market mechanisms, is born
disproportionately by low- and moderate-income families. Conversely, measures that
reduce energy bills tend to provide proportionally larger benefits to these income groups.
16
For instance, a carbon-only permitting system creates an incentive to switch to nuclear power, an
alternative favored by relatively few environmentalists because nuclear plants pose their own
environmental risks and costs. These risks and costs are difficult to compare to those posed by fossil plants.
Therefore, any measure to reduce total greenhouse gas emissions should include policies
to offset negative distributional impacts. Offsetting these impacts can be done with a
modest share of the total revenues generated by a carbon permitting system. 17
Measures to offset the regressivity of energy charges generally fall into four types: (1) tax
measures such as increases to the earned income tax credit or other refundable credits; (2)
transfer measures such as the Low-Income Home Energy Assistance Program (LIHEAP);
(3) targeted and general energy-efficiency measures, such as low-income weatherization
programs, buyback of older fuel-inefficient automobiles, efficiency standards for
appliances, etc.; and (4) pricing measures, such as inverted, lifeline, or basic-block rates
for residential customers of electric and gas utilities.
Carbon Permits and Jobs
When money is spent in a state, it creates jobs in two ways: directly, and indirectly.
Direct job creation comes from the instate jobs used to create and sell the purchased good
or service. Indirect job creation comes about because the inputs used to produce the good
or service may also be created in-state. For instance, when a consumer buys a book, they
are creating jobs directly in retail and publishing, and indirectly in the paper industry and
many other industries.
A specified sum of money spent in a state will create different numbers of jobs depending
on the way it is spent. Again, there are two reasons for this. First, some industries are
more labor-intensive than others, both in terms of direct and indirect production (the
intensity effect). For instance, money spent on education creates almost five times the
number of jobs (and almost four times the wages) that spending a similar amount of
money on gasoline would create. (See Table 2 below). Second, some industries have
most of their supply chain in-state, while others do most of their production out of state
(the locality effect).
The table below shows the job-intensity of expenditures in various sectors of the
economy per million dollars of final demand. Column 2 shows the direct and indirect
jobs created, while column 3 shows the direct and indirect wages created, in millions of
dollars.18 Sectors are ordered from those that create the fewest jobs per dollar of
expenditure to those that create the most. As you can see, expenditures in the energy
sector create far fewer jobs than similar expenditures in most other sectors of the
economy.
17
We are currently doing quantitative estimates of a range of alternative policy packages to offset the
distributional impacts. However, preliminary calculations suggest that the regressive impact on lower-
income households can be fully offset by allocating no more than 10 to 20 percent of the revenues from the
permitting system.
18
These are standard type-I multipliers for the U.S. Leontief input-output table. Source: Laitner, S.,
Bernow, S., & DeCicco, J. "Employment and Other Macroeconomic Benefits of an Innovation-Led
Climate Strategy for the U.S." Energy Policy v.26 no. 5 425:429 (1998).
Table 2. Job and Wage Multipliers for the U.S. National Economy
1. 2. 3.
Sector Employment/$mill. Compensation/$mill
1. Oil refining 13.0 0.41
2. Gas utilities 16.3 0.54
3. Insurance/Real estate 17.6 0.43
4. Oil & Gas extraction 18.1 0.51
5. Electric utilities 19.9 0.64
6. Other mining 24.7 0.75
7. Coal mining 25.5 0.89
8. Motor vehicles 26.0 0.85
9. Pulp & paper 28.1 0.88
10. Primary metals 28.8 0.91
11. Other manufacturing 30.0 0.86
12. Food products 30.2 0.72
13. Metal durables 30.9 0.97
14. Other utilities 31.2 0.91
15. Stone, clay & glass 32.1 0.95
16 Construction 34.2 0.90
17. Financial services 35.6 1.09
18. Wholesale trade 36.6 1.11
19. Agriculture 38.2 0.52
20 Other services 44.2 1.12
21. Retail trade 53.7 1.03
22. Government 54.3 1.57
23. Education 61.9 1.61
Because most of the states in the New England Governors' accord and the Pataki
initiative have little in-state fossil fuel production (with the notable exception of
Pennsylvania) we believe that the differences in jobs created due to intensity effects are
likely to be matched or exceeded by differences caused by locality effects. In essence,
dollars spent on fossil fuels immediately leave the state, while dollars spent on most other
goods will circulate within the state, creating additional jobs.
As a result, the reduction in fuel consumption induced by a carbon permitting system will
cost relatively few in-state jobs. On the other hand, spending from the revenues generated
by a carbon permitting system will probably create many more jobs, whether used to cut
other taxes or to provide essential state services. Conversely, carbon permit systems that
are grandfathered (given away to existing polluters for free) tend to reduce in-state
employment. Grandfathered permits, like auctioned permits, drive up the price of fossil
fuels and fossil-based electricity by constricting the supply. However, if the permits are
sold, the revenues from this price increase will be spent in-state, whether through tax cuts
or direct government expenditures. On the other hand, revenues from grandfathered
permit systems go to the stockholders of energy companies that receive the permits. In
some cases these are out-of-state companies, but even in cases such as an electric utility
with entirely in-state operations, most of the stockholders will typically be out of state.
Hence those moneys will leave the state and not generate in-state jobs.
Estimates of the magnitude of these job effects are contained in Hoerner & Freeman, The
Regional Greenhouse Gas Initiative: A Job Creation Strategy (Redefining Progress,
forthcoming 2004).
Competitiveness
As discussed above, a grandfathered permit system will reduce employment and also
transfer capital out of the state, relative to an auctioned system. In addition, there are even
more serious competitiveness problems posed for the region if grandfathered permits
become the model for a national system.
Recall that, whether auctioned or grandfathered, carbon permits restrict the supply of
fossil fuels and so drive up the price. In a grandfathered system, the benefit of these
higher prices goes to those who receive the grandfathered the permits, i.e. those who have
produced or consumed large amounts of fossil fuels in the past. As most northeastern
states are net importers of energy of all types, this would cause an enormous transfer of
wealth from energy-consuming states to energy-producing states. The result would be
reduced jobs, higher prices and lower growth in the northeast.
Even northeastern utilities might correctly fear such a scenario. The average carbon-
intensity of electricity for the RGGI region is 136 tons of carbon per Gigawatt-hour,
which is 27 percent less than for non-RGGI states. Thus, on a per GWh basis the subsidy
provided by a grandfathered national carbon trading system would be 27 percent greater
for non-RGGI power producers than for northeast utilities. In today's competitive electric
markets, this tilt toward subsidies for out-of-region power producers will erode the
competitive position of in-region power producers, and could ultimately put them out of
business entirely. For the New England Compact states, the contrast is even more
striking, with non-Compact states averaging 37 percent more carbon-intensive than
Compact states.
On the other hand, an auctioned permit system returns the revenue from the constriction
of fuel supply to the public. As a national system, though it would add to the costs of
production for all utilities, it would improve the competitive position of relatively clean
power producers such as those in the Northeast. If the revenues were returned to the
states on a non-carbon basis, such as per capita, per dollar of gross state product, or
through cuts in general federal taxes or increases in general federal services, the result
would be a net transfer of resources to relatively clean regions like the Northeast.
Thus, in addition to the competitiveness effects described in the last section, if the
Governors' accord or the RGGI is to become a model for a national initiative (as seems
likely), the competitiveness of both the region as a whole, and local power producers in
particular, will be benefited by an auctioned system and injured by a grandfathered
system.
Table 3. Carbon intensity of Electric Generation, by State and Region
(2000)
Tons Carbon Utility Non-Utility Total Carbon
from Electric Generation Generation Generation intensity
State or Region Generation (GWh) (GWh) (GWh) (Tons/GWh)
Connecticut 3,057,293 16,993 16,485 33,478 91
Delaware 1,624,985 4,137 1,774 5,911 275
Maine 1,163,105 3 13,048 13,051 89
Massachusetts 6,609,237 1,705 37,443 39,148 169
New Hampshire 1,376,094 12,703 2,242 14,945 92
New Jersey 5,559,387 25,251 32,953 58,204 96
New York 16,006,377 73,189 64,850 138,039 116
Pennsylvania 34,588,377 97,063 108,440 205,503 168
Vermont 37,659 5,308 975 6,283 6
RGGI 70,022,514 236,352 278,210 514,562 136
Non-RGGI 611,416,479 2,779,039 506,341 3,285,380 186
Compact States 12,243,388 36,712 70,193 106,905 115
Non-Compact 669,195,606 2,978,679 714,358 3,693,037 181
National 681,438,994 3,015,391 784,551 3,799,942 179
Calculations by the author from 2000 data from the Energy Information Administration,
State Energy Data System
Summary
To summarize:
· A market-based system of carbon permits or taxes is the most efficient and fair
way of achieving CO2 reductions.
· A system that raises revenue costs only about a quarter of a system that is
grandfathered.
· An auctioned permit system raises sufficient revenue to play a significant role in
helping to resolve major structural budget problems in the region.
· The regressivity of a carbon permit system can be offset by devoting a small
portion of the revenue to tax reduction, energy efficiency policies, and similar
measures.
· A grandfathered permit system would reduce employment, investment, and
competitiveness in northeastern states, while an auctioned permit system would
increase employment.
· If taken as a model for a national system, a grandfathered permit system would
transfer wealth and jobs out of the northeast, while an auctioned system would
transfer resources to the Northeast.