Dr. Alexandra von Meier
Sonoma State University
Q1. Should the United States rely on a market-based approach to meet the country's
energy needs?
Yes. I know of no viable
mechanisms other than competition among firms and free consumer choice that
can consistently produce outcomes in the best interest of society. Markets,
when functioning properly, can maximize social welfare. In order to assure this
proper function, however, we must carefully distinguish the limitations of market
performance in different situations and recognize the appropriate role for government
intervention where market failures exist. Such intervention should not be considered
antithetical to a market-based approach; rather, it should be considered as
a key enabling tool for efficient market performance.
The production and delivery of energy in particular has inherent properties
leading to some of the classic market failures which, according to the most
basic tenets of economics, prevent markets from achieving socially optimal levels
of production and consumption. These failures include negative externalities
due to environmental impacts of energy production (leading to over-consumption
of energy in general and of environmentally hazardous types of energy in particular);
positive externalities of research and development (leading to under-investment
in energy technology R&D on the part of private firms); price ineleasticity
of demand; imperfect information concerning the costs and benefits of all available
energy options; market power on the part of certain suppliers; and, in the case
of electricity, technical constraints on the realization of market transactions.
I believe that the general tendency in public discourse on energy policy in
recent years has been to underestimate the significance of these market failures.
A top priority for Federal energy policy, then, should be to identify and correct
market failures and approximate the "level playing field" so ubiquitous
in cliché yet so elusive in reality. Here it is indispensable to recognize
all the various forms of implicit and explicit subsidies that have supported
and continue to support today's leading energy resources and technologies, most
notably fossil fuels. Not until these subsidies are equaled in magnitude by
subsidies for competing resources, particularly renewable energy, and not until
a sincere attempt is made to internalize the many short-term and long-term costs
of energy production to environment and human health, is it meaningful to speak
of a functional, competitive energy market.
I believe that in such a functional market, the equilibrium price of energy
- both electricity and fuels - would be somewhat higher than accustomed as a
result of internalized costs, and that it is an appropriate mandate for government
to ease the impact of what would essentially appear as a regressive tax on energy
consumers of lower income. I also believe that the structural changes effected
by an energy market with more realistic prices - including increased end-use
efficiency and the substitution of cleaner energy sources for fossil fuels -
would, in their benefits to society, far outweigh their considerable costs.
Q2. What have been the successes and failures of long-term, Federal energy planning
and could you point to examples?
In an attempt to respond
to this broad question, it seems plausible to begin by considering what might
be the earliest instances of deliberate, long-term Federal energy planning.
The expansion of the electric grid during the earlier part of the 20th century,
and particularly its systematic extension into rural areas in the United States,
surely stands out as an endeavor of momentous significance. Interestingly, this
development was not driven by economic rationale, since extending power distribution
to rural areas was by no means the least expensive way to get electricity to
villages and farms: With a nascent wind industry in the 1930s, power could actually
be supplied more cheaply in many rural locations through small-scale wind turbines.
(Indeed, rural electricity distribution continues not to pay for itself, but
is now subsidized through the electrical rates of customers in densely populated
areas that are less expensive to serve.) However, the notion of a completely
connected electricity infrastructure in and of itself represented a vision of
progress, prosperity, and equality, and was therefore seen as a value to society
beyond its pragmatic and calculable benefits. Seeing that this sort of long-term
social benefit would be under-supplied by competitive markets, Federal government
took on its proper intervening role, creating the Rural Electrification Agency
and expending considerable public funds to connect every American town to the
grid. Though an unfortunate side effect was to put the U.S. wind industry more
or less on hold for half a century, rural electrification must be considered
a "success" of Federal energy policy in terms of having achieved what
it set out to do, and having created what we now consider an absolutely essential
artifact of our civilization.
Rivaling the electrification program in magnitude would be the U.S. Government's
effort to develop nuclear fission as an energy resource. Again, judged by the
criterion of whether the policy accomplished its goals - in this case, to produce
a viable energy resource alternative that could be commercially employed on
a large scale, and to position the United States internationally as a respected
leader in peaceful nuclear technology - the nuclear program has to be considered
a success. This success is notwithstanding the political difficulties nuclear
energy has subsequently encountered, and notwithstanding the fact that its economic
costs turned out to be considerably greater than many had hoped. As with rural
electrification, the driving force behind nuclear research and development was
not economic but based on broader social and political concerns, which to address
the Federal government was indeed the appropriate agent.
In responding to the first oil crisis of 1973, Federal energy planning first
took on its present, self-conscious role that includes not only initiative in
specific technological directions, but a more complete sense of accounting for
all the energy used and resources available, along with a sense of responsibility
to somehow assure that society's projected needs can be met. For the first time
in modern history, a vital natural resource was recognized as being scarce not
only regionally but globally, in a way that might dramatically constrain human
activities. In this sense, the oil crisis represented not just an economic threat
but a profound philosophical challenge to a modern culture that had embraced
the notion of unlimited growth. Government appropriately stepped up to the plate
in attempting to mitigate the impact of the new energy scarcity for Americans
and to develop long-term alternatives to imported fuels.
The diverse specific efforts that ensued, particularly with regard to developing
and implementing new technologies, certainly vary in the extent to which they
"succeeded" as planned. Thus we now have improved techniques for the
production and conversion of fossil fuel energy; we have effective and reliable
means for solar heating and electricity generation; and yet we are still at
a loss in making nuclear fusion produce more energy than it consumes. But it
was and remains fundamentally correct for government to hedge the risk for society
by pursuing a portfolio approach and insisting on a diversity of options, even
if this means investing in some approaches that ultimately turn out to be dead-ends.
What stands out to me as the most compelling success of energy planning in the
late 1970s is the remarkable reduction in the energy intensity of the U.S. economy,
as reflected not only in declining per-capita energy use but also the increasing
ratio of economic productivity to energy use. Clearly, this development was
driven by market forces and did not come without hardship on American citizens.
Yet credit is due to government for lending both financial and moral support
toward energy conservation and efficiency, resulting in technological advances
and structural changes that have produced enduring benefits to society beyond
ameliorating the impact of high energy prices.
By contrast, the abandonment of energy demand reduction as a priority once the
acute energy crisis had relented in the 1980s strikes me as a failure of long-term
planning to which we partly owe our present predicament: a society that has
again become more, not less, dependent on fossil fuels, with clearly detrimental
economic and environmental effects. Though the intention of the Reagan Administration's
approach in particular was to benefit consumers through increased supply, lower
prices, and altogether less worry about any deleterious effects of resource
consumption, sufficient knowledge was available even in the '80s to realize
that increased consumption levels of fossil fuels can, at the most, bring temporary
gains to society.
Domestic as well as global reserves are undeniably going to become more scarce,
meaning more expensive technically to extract and more expensive politically
to claim entitlement to them. At the same time, local as well as global environmental
impacts of burning these resources are undeniably increasing in severity, to
such an extent that industrialized nations other than the United States are
already looking to greenhouse gas emissions as a key constraining factor in
their future fossil fuel consumption - quite apart from resource availability
and price. It should come as no surprise, then, if policies aimed at increasing
domestic supply of fossil fuels and keeping their price low should face increasing
economic, environmental, and political challenges, as they are essentially attempting
to sustain a fiction of abundance and carefree living that is bound to collide
with reality sooner or later.
In this sense, I interpret the energy crisis of 2000-01, which manifested largely
as price spikes due to market power on the part of energy-supplying firms, as
but a gentle foreshadowing of problems ahead. Expanding domestic supplies of
fossil fuels, the major aim of the S.A.F.E. Act of 2001, addresses the problem
of supply shortages quantitatively, and thus, by necessity, only temporarily.
What is needed, however, is long-term Federal planning that addresses the energy
problem qualitatively by promoting structural changes in our economy, directing
us toward dramatic efficiency improvements and renewable resources that can
be used at tolerable levels of environmental impact for centuries to come. Like
the visionary steps undertaken by government in the 20th century that helped
transform the United States into the technological leader it is today, we now
need leadership to transform our energy system once again, not on the basis
of short-term economics, but for the good of society in the long run. Finally,
I would like to offer a comment on transportation. From the standpoint of energy
and environment, the dismantling of regional and long-distance public transportation
systems and the development of highways in their place was arguably the most
tragic planning failure of the 20th century. Yet one must realize that until
the 1970s, transportation was not generally thought of in terms of its energy-consuming
and pollution-causing characteristics. Without our present understanding of
these negative externalities, and failing to extrapolate exponential growth
to its logical conclusion (i.e., hours spent in traffic jams), the plan to build
a huge infrastructure based on individual motor vehicles with grossly inefficient
internal combustion engines made perfect sense at the time, embodying the cultural
values of equality and independence that are at the heart of our national identity.
While giving credit to the highway planners of years past for their resounding
success in realizing a grand vision, it is vital for us now to revisit transportation
policy from a contemporary perspective that recognizes its key role in energy
use and its attendant environmental impacts. I believe that the best thing Federal
government could do for society in this regard is not only to insist on dramatic
improvements in automobile efficiency and emissions for the coming years, but
to offer leadership toward a sensible long-run transition to a transportation
and urban planning infrastructure that is not centered around individual cars.
Q3. How far should a democratic government go in planning and subsidizing for
long-term energy needs?
This question is fundamentally
answered by #1 above. Government should go as far in planning for and subsidizing
energy as is necessary in order to correct market failures. In other words,
government should do that part of the job, and only that part, which competitive
markets cannot accomplish if left to their own devices. In general, this will
mean subsidizing goods and services with positive externalities where insufficient
market incentives exist to provide for the level of supply that maximizes social
welfare. Such policies may involve significant expenditure of funds, but it
is important to note that appropriate government intervention may also include
taxation (in the case of negative externalities) or revenue-neutral incentive
mechanisms.
While this approach is straightforward in theory, and while a number of market
failures in the energy sector are quite obvious, the more difficult problem
in practice is to agree upon the proper magnitude of government intervention.
Thus, for example, the appropriation of funds for energy research and development
in a given fiscal year will be guided by constraints other than a theoretical
derivation of the precise level of support required to attain a socially optimal
level of supply. I submit that arguing about the magnitude of funding levels
should not distract us from the essential point, namely, that the direction
of government intervention be correct - that is, in the best long-term interest
of society.
At different times in history, the nature of this long-term, best interest of
society may or may not be obvious. Today, this interest is painfully obvious,
and not to act on it is unconscionable. In the 21st century, industrialized
nations must reduce their reliance on fossil fuel combustion. Competitive markets
will, eventually, lead us to this outcome - but not before incurring tremendous
and possibly irreparable damage to the earth's physical and biological systems
that support human life, not to mention the social, political, and economic
costs of struggling to hold onto our share of a globally diminishing resource.
Rather than committing our society to this struggle, we know enough about energy
resources and their environmental effects today to initiate an historic transition
to energy efficiency and renewable resources with complete confidence that this
will bring tremendous good for the American people and for all humanity - both
in terms of averting damage and in terms of opening new opportunities for prosperity.
Our knowledge of the vital importance and urgency of this transition should
be an objective guide as to the appropriate extent of government action.
Q4. When discussing resource potential you said in your testimony that what
has happened in the past does not predict the future unless the connection is
shown. How does this pertain to nuclear power and efficiency and renewable energy?
I argued in my testimony
that the potential future contribution of renewable resources cannot be judged
solely on the basis of their limited past contribution. This argument applies
not only to renewables but to any resource, including efficiency and nuclear
energy. For example, it would be fallacious to argue that because nuclear fission
is now supplying only 20% of U.S. electric generating capacity, it could never
be expected to supply more than, say, 60%. Similarly, the fact that nuclear
fusion provides zero commercial energy today does not in and of itself preclude
the possibility of fusion becoming an energy resource in the future. Instead,
we want to know why use of a given resource has been limited and examine whether
or under what conditions the same factors might apply in the future.
In the case of renewables, I would argue that the factors limiting their implementation
to date have primarily had to do with an evaluative framework that incompletely
accounts for their particular capabilities and benefits. For example, one might
argue that solar photovoltaic (PV) power generation has not been widely implemented
on a utility-scale because it is not cost-competitive with other resources.
This leaves open the problem of how cost-competitiveness is defined. An analysis
of PV that takes into account benefits of strategically sited, small-scale arrays
to the transmission and distribution (T&D) system - for example, by avoiding
or deferring capital-intensive upgrades of delivery capacity, avoiding line
losses, and providing reactive power compensation - may very well arrive at
a favorable benefit/cost ratio. Recognizing and accounting for these specific
benefits, however, requires overcoming either a cultural or an institutional
barrier: an institutional barrier because there may not exist a single entity
in an electricity market with the correct incentives to trade off generation
against T&D investments, or a cultural barrier because in traditional utility
planning, where the financial incentives may exist, the philosophical concept
of such trade-offs may not. The larger market context therefore plays an important
role in determining the relative competitiveness of resources and technologies.
We must assume this context to be fundamentally changeable, as new rules in
electricity markets have already challenged many old conventions.
But in order to judge future possibilities, it is necessary to look at the intrinsic
properties of a given resource or technology that are not readily changed. For
example, in the case of PV, one such intrinsic property that could have easily
been a show-stopper is the energy requirement for manufacturing PV cells. If
it were true, as some researchers warned decades ago, that it takes more energy
to produce a PV panel than it ever generates during its lifetime in full sun,
then the technology would be essentially doomed: no change in accounting or
valuation could address the intrinsic problem of losing net energy with every
panel you manufacture, and you will never "make it up in volume."
This was a serious, credible threat to the viability of photovoltaic technology
on a large scale. As it turns out, fortunately, some of the earlier estimates
were based on highly inefficient production processes, and PV modules on the
market today typically recover the energy expended in their manufacture within
months or a few years at the most. But until this was known empirically, it
would have been quite imprudent to bet on substantial growth projections for
PV technology.
In summary, with regard to the renewable resources referred to in my testimony
(particularly solar and wind power), I see no intrinsic problems that would
preclude their implementation on a large scale in the future. Given basic technical
feasibility and proven reliability, the two fundamental constraints are capital
cost and land use requirements, and neither seems particularly frightening or
unreasonable. Indeed, capital costs for solar and wind power have been steadily
declining with increasing manufacturing volumes, and they may well continue
to do so. Even if they remained fixed at today's levels, these capital costs
are well-known and bounded - the sticker price is what you pay - and thus offer
sound and plausible risk-hedging investments. Given this condition, I have every
reason to believe that people will increasingly discover opportunities and niches
for a continually expanding role of small-scale and renewable energy.
The pessimism expressed in my testimony concerning the potential for nuclear
fission must be carefully clarified in these terms. Just as in the case of renewables,
it would be fallacious to argue that, simply because commercial nuclear energy
has not been financially successful in the past, it could not be so in the future.
Again, we must examine why nuclear energy was expensive and whether we have
good reason to believe that these reasons will remain in place, or what it would
take to make them go away. I submit that the specific problem areas listed in
my testimony as implying future costliness of nuclear power do, in fact, reflect
its intrinsic properties - technical, social, and political - that appear to
me very difficult to change.
For example, consider the financial uncertainty entailed by the possibility
of siting and construction delays due to political opposition. While this latent
resistance is a social, not a technical problem, it is beyond the reach of policymakers
to alter. We can expedite licensing procedures, but we cannot simply, as I have
heard one nuclear enthusiast advocate, "change our society to become more
like the French."
Similarly, the complexity of nuclear technology in terms of the numerous support
systems required within and in addition to the generating plant itself is an
intrinsic property that is not susceptible to sweeping change. Though it is
true that plant designs can and have been greatly simplified, a commercial nuclear
reactor will still require peripheral items like back-up cooling, seismic safety,
back-up generators, a spent fuel pool, and a security fence to operate safely.
And a nuclear plant will always require a uranium mine somewhere, an enrichment
facility somewhere, a fuel fabrication plant somewhere, a spent fuel repository
somewhere, a truck to come and pick up the spent fuel, and a reliable electric
grid to feed. It also needs qualified operators, training facilities and simulators,
a regulating agency to look over their shoulders, and a host of technical experts
and political players to reach some consensus about the spent fuel. Finally,
the tragic events of September 11 are a forceful reminder of our technological
society's vulnerability to acts of terrorism - and the efforts that protecting
sensitive targets such as nuclear facilities may entail in the future.
It is because of this inherent complexity and vulnerability of nuclear technology
- the absolute dependence on having a large number of components working together
correctly - that I believe we cannot expect to fully break with the past and
have nuclear energy that is cheap and safe, no matter how smart we are about
improving reactor designs. This is by no means to say that it cannot work or
that we would necessarily go broke trying to get it to work. But I simply cannot
imagine that, once all the various costs are accounted for, the best nuclear
energy system of the future will ever be able to produce electricity less expensively
than a decent wind farm.
Q5. You stated in your testimony that large infrastructure investments and aggregation
are caused by central station electric power generation, presumably nuclear
and fossil stations. Yet you advocate biomass generation as baseload in an all-renewable
electric generation scheme. Won't the biomass steam electric plants be central
stations?
I mentioned infrastructure
associated with coal and nuclear central-station generation in the context of
land use requirements and overall cost, arguing that it is easy to underestimate
land use and costs for these resources if auxiliary facilities and mining are
not accounted for. This does not mean that central station power plants, which
would include biomass generation, are necessarily a bad idea. In fact, power
system engineers know that some proportion of large synchronous generators are
indispensable for the stability of an interconnected a.c. grid, as these machines
are capable of absorbing and thus leveling out transient fluctuations in power
(or a.c. frequency) through their rotational inertia within the split-second
time frame until turbine output is adjusted. (A "large" machine here
would mean on the order of tens to hundreds of megawatts.) There is insufficient
empirical evidence to say exactly what the minimum contribution of large synchronous
generators would have to be in order to assure stability for an interconnected
system on the scale of, say, the Western United States. However, I find it difficult
to imagine that we would ever be pushing this limit even in an all-renewable
scenario, as synchronous machines would be used by biomass, hydropower, and
solar thermal generation. (The only exceptions are photovoltaics and fuel cells
that use d.c.-to-a.c. inverters, and wind turbines with induction generators.)
As for biomass steam generation plants, it should be noted that they tend to
be of smaller size than those fired by fossil fuels, favoring the range up to
100 megawatts (which still leaves them capable to perform the stability service
described above). This practical limit results from diseconomies of scale with
respect to collecting and transporting the bulky fuel, as fresh biomass has
a high water content and a considerably lower energy density than fossilized
biomass. Even though constructing and operating it would get cheaper per megawatt,
at some point it simply isn't worth making a biomass power plant bigger because
the fuel would have to be gathered and trucked from so far away.
For any generation unit - and this is certainly as true for biomass and other
renewables as it is for traditional resources - there is also an infrastructure
and thus an environmental "footprint" associated with manufacturing
the generating equipment in the first place. To be fair, this component will
tend to be worse for smaller-scale technologies and those of lesser power density
in generation, simply because more material is needed in order to build a certain
number of megawatts of generating capacity. This issue deserves to be studied
carefully. Nevertheless, I believe the production and delivery of fuel, and
the management of its waste products, to be the more significant infrastructure
and land-use factor over the lifetime of a generating facility.
Land use and its environmental costs for the case of biomass are by no means
trivial. As agriculture in general can be considered the most environmentally
burdensome human activity on the planet, very careful attention ought to be
paid to cultivation practices for biomass fuel if the use of this energy resource
were to expand considerably beyond using waste products from food crops and
logging. The key question is for what time period agricultural practices can
be reasonably sustained. The same concerns that apply to growing food - soil
depletion, overuse of inorganic fertilizers, aquifer contamination, over-irrigation,
erosion, salinization - may also apply to fuel plantations. Another profound
concern for biomass energy is the net carbon balance: that the rate of replanting
biomass fuel provides a rate of carbon sequestration at least commensurate with
the rate of CO2 release in the fuel's combustion. I am confident that all of
these environmental concerns can be feasibly addressed, but it is crucial for
government to ensure that the correct economic incentives exist.Q6. DOE has
a "Power Park" program where they are investigating large parks with
wind generators, solar panels, hydrogen production, and fuel cells. Do you object
to these, as they would seem ideal for investment by power generating companies
that can aggregate investment? Please comment on the large infrastructure investment
in manufacturing, installing, operating, and repairing hydrogen generators,
fuel cells, and wind generators that are taller than the Capitol dome.I do not
object to "Power Parks" at all; on the contrary, I believe they represent
one important avenue for research, development, and demonstration. My understanding
of the program objective is that it intends to explore efficiencies and economies
of scope that may be achieved by the co-location and coordination of successive
energy conversion steps, such as the generation of electricity, production of
hydrogen, and delivery of heat. The emphasis, therefore, is on the effective
matching and combination of technologies rather than on scale.
Infrastructure investment must be considered in terms relative to the amount
of capacity or energy provided, and relative to the alternative means. I see
no a priori reason why infrastructure investment should be prohibitive for the
technologies under consideration. If one were to envision a network of hydrogen
pipelines, storage devices, and conversion devices on a national scale, the
effort might seem daunting indeed. However, it has to be compared with the existing
infrastructure for fossil fuel delivery and the costs of its various components.
Also, this infrastructure was not put into place all at once.
Q7. In your testimony you mentioned pumped storage hydro as part of your scheme.
How many of these facilities would be needed nationally? Are there enough suitable
sites with acceptable environmental impacts?
To determine the storage
capacity that would be needed on a national scale for an all-renewable electricity
scenario would require careful modeling which, to my knowledge, has not been
done. This determination would be an optimization that trades off the cost of
energy storage capacity against the cost of additional dispatchable generation.
Based on qualitative and anecdotal knowledge (see Q8), I would be surprised
to learn if the need for storage thus calculated came to more than five to ten
percent of the total consumed.
Potential sites for hydro development in the United States have already largely
been established and environmental concerns are likely to prevent the development
of new sites in the future. It may be feasible, however, to add pumped storage
capability to some existing conventional hydro generation sites without dramatic
environmental impacts. While the majority of conventional hydro sites in the
United States already have upper reservoirs (as opposed to "run-of-river"
plants), converting them to storage facilities involves creating a lower reservoir
as well. The volume of the lower reservoir may be much smaller, however, as
its role is not to provide seasonal water storage. According to the U.S. Energy
Information Administration, there were 3207 conventional hydro generators and
139 pumped hydro storage facilities operating in the United States as of 1999.
How many of the existing conventional sites could feasibly be expanded to include
storage capability, taking into account specific topography, engineering, and
environmental factors, would be an interesting and worthwhile investigation.
It is also possible to excavate lower reservoirs underground, and to isolate
hydroelectric storage from natural water sources altogether, thus avoiding impact
on sensitive environments. Finally, while pumped hydro is the most common electric
storage technology today, it is by no means the only option. Solar thermal generation
may include thermal storage capacity for several hours' worth of generation
at a plausible cost. If more storage capability were needed than could reasonably
be supplied by pumped hydro and solar thermal storage, especially on a seasonal
rather than a diurnal cycle, hydrogen can be considered as a storage medium.
The drawback of hydrogen is that the round-trip efficiency (with electricity
being converted into hydrogen by electrolysis and then back to electricity in
a fuel cell) is quite low, necessitating a larger capital investment per unit
of energy finally obtained. Other technologies such as compressed air or superconducting
magnetic energy storage might be stronger candidates for electricity-specific
applications.
In any case, while the availability of acceptable hydroelectric sites is certainly
an important constraint for an all-renewable energy scenario that may force
the use of more costly alternatives, this constraint does not, in my view, pose
an insurmountable problem.
Q8. Have you performed, or do you know of economic studies that project the
cost of electric power in your solar, wind, biomass and hydro scheme? It would
seem that there will be a large generating capacity in biomass steam electric
and hydro idling when solar and wind are at their maximum, or when the wind
is strong at night. How much will it cost? Couldn't hydrogen be used as a storage
medium?
While the literature abounds
with cost projections for individual electric generating resources, I am not
aware of any current study that sketches out an all-renewable electric supply
scenario for the United States and attempts to derive an overall cost. Such
a study (which, incidentally, I think would be well worth funding) should assess
very carefully the plausible contributions from the respective resources in
various geographical areas, taking into account regional resource availability,
the magnitude and time distribution of regional demand, and existing and potential
new transmission capacity.
The basic analytic tool that utilities have used traditionally for planning
and scheduling generation is the "load duration curve," which shows
electric demand (MW) for each of the 8760 hours of the year, displaying these
hours not in temporal sequence but ranked by demand in each hour, starting with
the peak demand hour. The area under the load duration curve represents the
total amount of energy demanded. The process of allocating generation units
to meet this demand can then be worked through a straightforward graphical method
by "filling in" the area from the bottom with baseload generation,
adding non-dispatchable intermittent generation for those hours where it is
available, and finally filling in the remaining hours with dispatchable units.
Pacific Gas & Electric performed one study in the late 1980s that illustrated
how the load duration curve for their service territory could be hypothetically
filled with a combination of resources including hydro, biomass, geothermal,
solar thermal, wind power, pumped hydroelectric storage, and some thermal storage
as part of the solar generation. For a summer-peaking utility like PG&E,
this exercise was aided greatly by the coincidence of solar generation with
the demand peak. Essentially the entire summer and mid-day peak (mostly due
to air-conditioning loads) can be "shaved off" with solar generation
which, though non-dispatchable, has an extremely high availability during the
hours when it is hot. Another convenient factor is that wind speed often tends
to pick up later in the day, complementing the solar resource. For the California
case, then, hydro, biomass, and geothermal units could be operated at a reasonably
high capacity factor, i.e., not idling excessively. Though, to my knowledge,
PG&E didn't cost out this scenario, it certainly demonstrated the basic
feasibility of an all-renewable generation strategy. It would be very interesting
to repeat this exercise on the scale of the entire United States.
The overall cost of an all-renewable energy supply scenario would depend on
the relative contributions from each resource as well as the storage and transmission
needs determined (and traded off against each other) by examining resources
and demand regionally. Still, the marginal costs of the individual components
can give a general sense of the composite. Geothermal and hydropower are among
the lowest-cost generating resources available even compared with fossil fuels;
their contribution will be limited by resource and site availability. Wind power,
which is much less constrained in terms of site availability, can be expected
to provide energy in the neighborhood of five cents per kilowatt-hour. Solar
thermal generation, using parabolic trough technology, is the most expensive
bulk generation at about 10 ¢/kWh (though most analysts expect this figure
to drop with mass production). Transmission capacity is expensive and would
certainly constitute an important factor for large-scale, inter-regional energy
transport. In this context, it should be noted that the construction of new
transmission capacity (and Federal initiative for putting it in place) has already
been the subject of public discussion. The strategic placement of this new transmission
capacity will have important implications for the economic competitiveness of
future generation sites.
Q9. Considering that the waste products from nuclear power plants are contained,
don't your four policy criteria apply to nuclear power? (page 8 of your written
testimony)
Each of my criteria applies
to nuclear energy to some extent, but with qualifications.
--Environmental benefits owing to absence of pollution or greenhouse gas emissions.
While the solid form and "contained" character of nuclear waste might
distinguish it from "pollution" in the minds of technical and scientific
experts, and while it is arguable whether the possibility of future leakage
from a waste repository should be considered "pollution" like other
forms such as air pollution that exist with certainty and in the present time,
the general public certainly appears to view radioactive waste as a highly undesirable
product. I would argue that the political rather than the scientific definition
of "pollution" is more relevant here.
--Continuous resource availability and price stability to offer reliable support
for long-term economic planning.
Uranium reserves are large, but not infinite. Depending on the time horizon,
it could be fair to project nuclear fuel as abundant and cheap, but this time
horizon is inherently limited, while that for the availability of renewable
resources is not. Price stability depends not only on the price of fuel, but
on other components as well, which are inherently less predictable for nuclear
power as a complex and politically controversial technology.
--Potential development of a lucrative and socially responsible export commodity.
Lucrative, perhaps. But exporting nuclear technology is not socially responsible
in my opinion unless the recipients have an established industrial culture that
permits them to operate it safely under any conditions. The proliferation of
weapons-usable materials and know-how is another concern. Finally, the international
transport of nuclear materials has met with political opposition in the past
and may continue to do so. I doubt that large segments of the American public
would agree to the "socially responsible" characterization of nuclear
technology as an export commodity, particularly to developing countries or politically
unstable parts of the world. Such concerns might intensify as of September 11,
2001.
--Intrinsic compatibility with competitive markets.
Nuclear technology does not afford the advantages of flexible scale, distributed
siting options, short lead times, and easy market entry for diverse small firms
that are characteristic of renewable energy and that tend to enable efficient
market function.
Q10. You stated in your testimony that electric power consumers lack the ability
to reduce load. Are you aware of the residential programs that provide load
control devices that are automatically controlled by the utility? Could you
advocate the increased use of load management tools? Can you comment now, after
the savings in projected electric power usage in May, June, and July in California?
California's summer of 2001 has indeed shown remarkable savings in electric
consumption as compared to projections. While these savings have resulted in
large part from unusually mild summer weather (with lower temperatures especially
in Southern California, where air conditioning loads dominate demand), conservation
behavior on the part of consumers certainly had an effect.
While demand-side management (DSM) programs of various scales have been sponsored
by most U.S. utilities, these programs have generally focused on putting in
place energy efficient appliances rather than control devices. For instance,
the Energy Information Administration's estimate of 1998 combined energy savings
through the DSM programs of 971 utilities show approximately 53 billion kWh
savings due to energy efficiency and only 940 million kWh (less than two percent
of the total) due to load management. Few electric customers currently have
automatic load control devices installed.
There remains, therefore, a largely untapped potential for load management,
particularly for loads that are less sensitive as to time of use (e.g., washing
machines and dryers). In order to allow for markets to perform efficiently,
it is important that load management capabilities be combined with real-time
pricing (RTP).
Q11. You say that the Price-Anderson insurance law for nuclear power plant insurance
is a subsidy, which it is, even though the government pays nothing unless the
claims become very large. You advocate free market competition for nuclear power.
Shall we then eliminate the production tax credit for wind generation on the
same basis?
Subsidies such as the production tax credit for wind generation are warranted
in cases where positive externalities or public goods are produced. Most people
would agree that this is the case for the development of wind power, which offers
distinct environmental and risk-hedging benefits that will accrue to society
over decades to come. Therefore, we should continue to subsidize wind power.
An analogous argument can be made for the case of nuclear power, though I personally
do not agree with it. If one believes the availability of nuclear energy as
a resource alternative to constitute a significant public good, then it is legitimate
to call for subsidization of its development and implementation. Indeed, the
continuing cultivation of technological capability and know-how in the nuclear
sector can be considered a public good in and of itself, particularly in view
of the need to address nuclear proliferation issues in the international community.
In my opinion, this latter concern is the most important and justifies some
level of Federal expenditure on nuclear R&D, although its purpose should
be explicit and limited. I am not convinced that the social benefits of nuclear
technology overall exceed its social costs, and therefore do not advocate subsidizing
it as an energy resource.
Q12. In your testimony you say, "land used for fossil fuel extraction and
nuclear facilities becomes permanently sacrificed." Haven't there been
strip mines reclaimed and nuclear plant sites (Shippingport) restored to "greenfield"
conditions?
The decommissioning of the Shippingport plant in the 1980s is indeed widely
recognized as a success, and the site has been released for unrestricted use.
It is important to recognize, however, that this particular case presented almost
uniquely ideal conditions for effective decommissioning:
--At 72 MW, the reactor vessel was considerably smaller than those for PWRs
and BWRs in the 1000 MW range and therefore easier to transport.
--The radioactive inventory of the Shippingport reactor was comparatively small,
estimated by OTA as 16,000 Ci at the time of dismantling (whereas typical commercial
reactors would have an inventory on the order of millions of curies after 30
years of operation).
-- DOE facilities were available for the disposal of contaminated plant components,
including 24,000 cubic yards of low-level waste, at low cost.
These favorable circumstances resulted in a cost of less than $100 million for
the Shippingport case. By contrast, the costs of decommissioning of larger commercial
reactors are typically estimated at least about three to four times that amount.
These estimates are inherently vulnerable to upward adjustment, since it is
unlikely to discover something in the course of decommissioning efforts that
will reduce the cost, but there are many ways the cost can increase. For example,
estimates of decommissioning costs for the Yankee Rowe plant were revised from
$368 million to $508 million due to the increased cost of spent fuel storage.
The high costs of decommissioning, which decline somewhat over time with the
radioactive decay of the inventory, present a general incentive to defer decommissioning
for as long as permissible. Uncertainties about the availability of disposal
sites due to political factors may further delay decommissioning efforts. Current
U.S. regulations allow for 60 years to reach Stage 3 or "unrestricted site
use"; in the U.K., this period has been stretched to 130 years. This is
not "permanent," but it is a long time to wait to do something with
your real estate.
Aside from reactor sites, the high-level and low-level waste disposal sites
that accommodate the materials not only from reactors but from other steps in
the nuclear fuel cycle are in fact "permanent" on any reasonable human
time scale. Accounting for their land use must also include the safety perimeters
or exclusion zones dedicated to disposal facilities.
The problem of reclaiming strip mines is analogous in the sense that economic
incentives tend not to favor the speediest and most thorough clean-up possible.
Also, compared to the nuclear industry, the level of environmental and safety
regulations is less stringent and their enforcement probably less reliable in
the fossil-fuel and mining industry. Thus, the extent to which strip mines can
realistically be expected to be restored to greenfield conditions is questionable
in my mind. Finally, having a former strip mine covered in vegetation and restoring
it to its pre-mining environmental status are quite different things.
Q13. In your testimony you say that "Adapting the design and operation
of nuclear reactors to function in the context of less reliable grids, as may
emerge under competitive market pressures, entails potentially high costs in
addition to those of basic reactor design and operation." Since part of
the design of all power reactors is the ability to safely shut down in the event
of a "loss of off-site power," what additional changes did you have
in mind?
All power reactors must be able to shut down safely when off-site power fails.
This does not mean that they can do so with zero risk (loss of off-site power
ranks at the top among initiating events for potential accident sequences analyzed
by the NRC), nor that they can do so repeatedly without strain on mechanical
and operational systems. Often, a loss of off-site power event at a PWR or BWR
results in a reactor "scram" or rapid emergency shutdown. It is well
recognized in the industry that scramming your reactor frequently is a bad idea
and will cost you money, as it entails material fatigue and possible damage
from rapid thermal expansion and contraction (including the pressure vessel)
and thus a need for increased maintenance and safety inspections, not to mention
cumbersome procedures to bring the reactor back into normal operation each time.
A reactor scram after a turbine trip can be avoided if a sufficient heat sink
is available, which means sufficiently large steam dump valves (for venting
steam to the atmosphere) and enough make-up water for the transient release
while reactor power is ramped down in an orderly manner. Increased thermal inertia,
as is featured by some newer designs like the Pebble Bed Modular Reactor, is
certainly helpful in permitting slow and controlled reactor shutdowns.
While avoiding a reactor scram affords greatly improved safety, it does not
enable a nuclear plant to come back on-line at full power immediately after
the transient disturbance. Following a turbine trip, it is necessary to repeat
the process of "paralleling" the generator to the grid starting at
zero power and then gradually ramping up steam output. For conventional steam
generation plants, this process takes at least several hours; for a nuclear
plant, owing to the many procedural steps that need to be taken in the interest
of safety, it may take a day or more.
It should be noted that standard operating practice in the United States frowns
upon frequent variation of power level even for the purpose of following changes
in demand, based on the reasoning that any change in the operating conditions
(even without unit shutdown) will entail mechanical strain, potential for error,
and thus unnecessary risk. Thus, U.S. nuclear plants are now operated exclusively
as baseload plants. To be fair, perspectives on this issue vary: the French,
for example, designate a number of their reactors as load-following plants under
direct digital control from dispatchers, and they don't appear to lose much
sleep over it (with nuclear providing upward of 75% of electrical energy, they
have little choice). Few would disagree, though, that the present design conditions
for the safest and most cost-effective operation of nuclear reactors involve
being on-line consistently and without interruption at 100% power for as long
a time period as possible.
Q14. In your testimony you stated that nuclear technology cannot be done piecemeal;
that it depends on economies of scale. Later you say that renewable resources
never have a declining cost scale. It seems that many renewable energy advocates
are counting on economies of scale. For example, manufacturing of the current
design of 1MW wind turbines would seem to be very expensive, and [it would seem]
that DOE's "Power Park" program is an effort to take advantage of
economies of scale. Shall we advise DOE to abandon their Power Park program?
As mentioned above (Q6), the aim of the Power Park program as I understand it
is to explore economies of scope rather than economies of scale. The emphasis
is not on making the individual components as large as possible, but on combining
them in smart ways that increase overall efficiency - for example, by avoiding
transportation losses from one energy conversion site to another, or by taking
advantage of waste heat from one process for another.
Economies of scale exist for renewable energy technologies up to a point. For
example, a 1 MW wind turbine will tend to cost somewhat less than two 500 kW
turbines because of the fixed costs associated with some of the materials and
installation labor. However, these economies of scale are quite moderate. The
optimal sizing of a wind rotor and generator also depends on the wind speed
distribution at a given specific site. Thus, bigger is not always better. Finally,
diminishing returns on scale occur past a certain practical limit, which is
on the order of 1 MW for wind machines.
For solar thermal generation, economies of scale exist in the steam generation
component analogous to conventional steam generation. These are balanced by
diseconomies of scale in terms of collecting and transporting the solar heat
(analogous to geothermal steam generation units that gather steam from multiple
wells, where thermal losses begin to matter when the collection area becomes
too large). Photovoltaics are the least scale-sensitive of solar and wind technologies;
minor economies of scale exist with balance-of-systems components like d.c.-to-a.c.
inverters and installation labor.
By contrast, the economies of scale that analysts count on when projecting declining
future costs for renewable energy technologies involve large manufacturing volumes.
The possibility of mass-production and thus cost savings on many identical components
is a fundamental advantage of smaller scale technologies: we will never have
the benefit of rolling thousands of reactor pressure vessels off the assembly
line.
I suspect that in many situations smaller-scale, distributed installations of
renewable energy generation will turn out to be preferable both technically
and economically. I also believe that there is a place for larger-scale and
aggregated projects, particularly in view of drawing investment from established
actors in energy markets. It is certainly worthwhile to obtain the construction
and operating experience with large-scale installations, so as to be able to
address the question of economies and diseconomies of scale for various technologies
with empirical data. Again, an advantage of modular renewable energy technologies
is that one can harmlessly experiment with different sizes and combinations.
The attempt to exploit economies of scale by making larger renewable energy
production facilities has not met with unqualified success in the past. For
example, as was learned in DOE's MOD program, increasing the diameter of wind
rotors eventually compromises efficiency, owing to the fact that wind speed
is not uniform across the swept area. Critics have argued that the motivation
for making solar and wind facilities large and centralized is rooted in cultural
tendencies as opposed to technical rationales. While I suspect that they are
probably right, this in itself doesn't mean that the large plants cannot work
well, nor does it preclude the possibility of large centralized and small distributed
installations co-existing and complementing each other.
In summary, the Power Park program is not primarily aimed at making renewable
energy facilities larger in scale, nor would it be fundamentally wrong or dangerous
to do so (it will probably simply turn out to be somewhat less cost-effective).
I see no reason to abandon the program; on the contrary, the prospect of exploiting
synergies among different technologies deserves ample support.