The Northwest Power and Conservation Council’s Fifth Northwest Power Plan
responds to an energy crisis that developed over a period of about five years,
largely unnoticed, and then burst upon the Northwest in the fall and winter of
2000 and the spring of 2001. The Council dissects the crisis — its causes and
impacts — in the power plan and then makes a number of policy recommendations
for how to avoid a repeat of the crisis in the future.
The Council analyzed the energy crisis and responded to it in the
Northwest Power Plan. The following discussion is condensed
from the power plan and from other Council documents and analyses that addressed
the causes and impacts of the crisis.
The Western electricity crisis has been referred to as the “perfect storm” — the result of the confluence of a number of adverse trends and events. It had
its roots in several years of under-investment in generating and conservation
resources. It was triggered by the onset of poor hydropower conditions in the
late spring of 2000 leading to the second-lowest
Columbia River runoff since
1929. The crisis was made much worse by a deeply flawed electricity market
design in California and opportunism by some of the participants in that market.
And many believe it was prolonged by the reluctance of the Federal Energy
Regulatory Commission to impose West-wide price caps.
The near-record low runoff in 2001 resulted in almost 4,000 average megawatts
less hydroelectric energy available than in an average year, and even less
compared to the relatively wet years of 1995-1999. The reduced hydropower
generation affected not only the Northwest, but also California and the
Southwest. Net exports from the Northwest Power Pool area for the May through
September period were about 2,700 average megawatts less in 2000 and 2001 than
in the preceding three years. However, these conditions and the California
market problems likely would not have triggered the crisis had it not been for
the extremely tight power supply in the Northwest and the West leading into
2000. The gap between power supply and power demand widened steadily through the
1990s and reached 4,000 average megawatts — nearly enough power for four cities
the size of Seattle — by 2000.
During most of the late 1990s, construction of new power plants in the
Northwest and, for that matter, the rest of the West, effectively was at a
standstill. Wholesale electricity prices were low and were expected to stay low,
thus sending an economic signal to power plant developers that it was not the
right time to build — it seemed unlikely that a new power plant could compete
with market-priced electricity. Similarly, utility investment in energy
conservation during that period was less than half the cost-effective levels
identified by the Council.
Concerned by the growing deficits, the Council undertook a study of regional
power supply adequacy. That study, released in early 2000, estimated that the
probability of being unable to fully serve Northwest load (the “loss of load”
probability) would climb to 24 percent by 2003 even when accounting for the
ability to import power from the Southwest in the winter and to draft reservoirs
behind hydropower dams beyond normal limits in emergencies. The analysis also
indicated that 3,000 megawatts of new power supply would be necessary to bring
the loss-of-load probability down to the industry-accepted criterion of 5
percent. However, the report failed to emphasize that the probable leading
indicator of the scarcity was volatility in power prices. Surging prices
in the winter and spring of 2000 and 2001 would make that clear.
Neither the Council’s study nor any of the other indicators of the
increasingly inadequate power supply prompted a rush to build new power plants
or invest in energy efficiency. Some new plants were under development.
However, they were not enough, soon enough, to avert the crisis. Why did the
Northwest and the rest of the West allow loads and resources to get so far out
Causes of the crisis
One explanation is the infatuation with the competitive wholesale power
market that was prevalent in the late 1990s. Why should a load-serving entity
build new resources or enter into long-term contracts when the invisible hand of
the competitive market would take care of long-term supply? A long period of low
spot market prices seemed to validate this view. However, it should have been
clear that the market was not taking care of supply
electricity continued to grow, but very few new power plants were being built.
Wholesale prices in the years immediately preceding the summer of 2000 were
generally below what it would take for a new generator to fully recover its
costs, in part because of greater-than-average hydropower production during that
period. Few independent power producers were willing to undertake the risk of
building a plant without having a significant portion of a plant’s capability
committed to long-term contracts. This was particularly so in the Northwest,
where good hydropower conditions can depress market prices for extended periods.
Fear of retail competition also kept utilities from making commitments to new
resources. During the mid-to-late 1990s, there was a great deal of discussion
about retail competition. Some states, such as Montana and, on a more limited
basis, Oregon, opened their retail markets to competition. Others were
considering it, and there was speculation that Congress might impose retail
competition. In the face of these developments, utilities were concerned that if
they were forced to open their service territories to competition, they might
lose customers to competitors and their investments in new resources would be
“stranded” — that is, the utility would not be able to fully recover costs of
new resources or long-term contracts.
The growing deficits should have been seen as a sign that a reasonable level
of investment in new resources would not become stranded. Nonetheless, concerns
about retail competition and stranded costs undoubtedly played some part in
slowing resource development.
Another contributing factor was uncertainty with regard to the role the
Bonneville Power Administration would play in serving future Northwest
loads. Most utility and direct-service contracts with Bonneville were to expire
in October 2001. Decisions about the signing of new contracts for subsequent
service did not begin until 2000. This meant that both Bonneville and its
customers were uncertain about whom would have the responsibility for acquiring
new resources until the electricity crisis was practically upon us. In the end,
Bonneville found itself in the position of having to acquire 3,300 megawatts in
a relatively short time during a period of extremely high prices. Had there not
been the uncertainty, Bonneville or the utilities may have taken steps to
acquire resources earlier that would have lessened the impacts of 2000-2001.
Finally, it seems clear that electricity planning in the 1990s, including
that of the Council, failed to fully appreciate and factor into decisions the
risks facing the industry. In particular, these included the risks associated
with reliance on a potentially volatile wholesale market and risks associated
with gas-fired generation that depends on the also volatile natural gas market.
If planning had done a better job of reflecting the risks and potential impacts,
might load-serving entities have taken action to mitigate those risks? In
February 2000 the Council released a report that put a spotlight on the region’s
worsening resource condition. However, by then it was too late to elicit much of
a response from the region.
Response to the crisis
Ultimately, Northwest utilities, independent developers, businesses,
governments and citizens responded to the electricity crisis with ingenuity and
effectiveness. There were three primary responses: new generation, both
small-scale and larger conventional generation; load reduction
By December 2001, almost 1,300 megawatts of new permanent generation had
entered service, approximately 1,100 megawatts of which was gas-fired combustion
turbines. Another almost 3,800 megawatts was under construction, almost 2,900
megawatts were permitted, and over 10,000 megawatts were in the permitting
process. The great majority were gas-fired plants, and most of those were
combined-cycle units. However, there were several hundred megawatts of wind
power developed as well. The developers were primarily Independent Power
Producers (IPPs). This pattern was seen throughout the West.
By 2003, approximately 4,000 megawatts of new capacity had come on line in
the Northwest since January of 2000. An additional 1,400 megawatts was partially
complete, although construction had been suspended. With the exception of
approximately 970 megawatts of wind (by the summer of 2006), the great majority
of the generation was gas-fired. While the amount of new generation is
impressive, most of it effectively “missed the party.” By the time the
generation became operational, prices had fallen and along with them, the
profits anticipated by the developers.
As a result, there are hundreds of megawatts of under-utilized new generating
capacity in the region, most developed and owned by independent power producers.
The good news is that the capital risk associated with this capacity is borne by
the investors rather than by electricity consumers.
In response to increasing wholesale power prices, demand for electricity in
the region began falling in late 2000. By 2002, loads were 2,800 average
megawatts below loads in 2000 on an average annual basis, a drop of 13 percent.
This load reduction was accomplished through two means: efficiency (energy
conservation) and, primarily, demand response
While the efficiency response was impressive, demand response made up the
great majority of the load reduction. Demand response means a reduction in
electricity use unrelated to the efficiency of the facility, equipment or
process. It can be accomplished through a reduction or cessation in the
electricity-using activity (for example, making sure unnecessary lights are
turned off, only running one shift in a factory or shutting down entirely) or by
switching to a different source of electricity (installing self-generation) or a
different energy source altogether (e.g., switching to direct use of natural
gas). All three methods were employed in 2000-2001.
Demand response was accomplished through a number of different inducements.
These included appeals to the public-spiritedness of consumers by public
figures, price signals, and utility “buyback” offers — offers by utilities to
pay for reduced consumption. The governors of the Northwest states raised the
visibility of the severity of the electricity situation and made public appeals
for cutbacks. Some industrial customers exposed to market prices responded in a
variety of ways to the sharp increases in wholesale prices, including fuel
switching, self-generation, cutbacks and shutdowns, albeit at some significant
economic expense. Sixty-three percent of the load reductions came about through
various forms of buybacks, over 90 percent of which came from the aluminum
industry. In the residential sector, programs like “20-20” and its variants
offered ratepayers a percentage reduction in their bill for reducing their
consumption by the same percentage relative to the same period in the previous
year. None of these load reductions came cheap, but they were cheaper than the
alternative of paying the market price for the electricity.
As impressive as the load reductions were, they came too late to avoid
several months of extremely high wholesale prices. Load reduction did not really
begin taking effect in a significant way until more than seven months after the
onset of wholesale prices that were several hundred percent higher than normal.
Had there been a more rapid response of loads to wholesale prices, it might have
partially mitigated the high wholesale prices that the region was experiencing.
Similarly, had investment in conservation continued at cost-effective levels
throughout the 1990s there would have been at least a couple hundred megawatts
fewer loads exposed to the high prices.
The third leg of the response to the electricity crisis was changes to the
operation of the hydroelectric system that increased generation. The most
significant change was reduction in bypass spill at the John Day, The Dalles,
and Bonneville projects. Bypass spill (running water over a dam’s spillways
instead of through the turbines) is intended to reduce injury and mortality of
out-migrating juvenile salmon and steelhead. However, from a power supply
standpoint, spill is energy lost. Most of the spill reduction took place in
2001. In total, reducing spill called for in NOAA Fisheries’ 2000 Biological
Opinion (BiOp) added an additional 4,500 megawatt-months to the region’s energy
supply, much of that coming in late spring and early summer when power prices
were still at extremely high levels. It also allowed storing additional water in
Canadian reservoirs in case poor water conditions continued into the winter of
The use of spill reduction also highlighted the conflict between fish and
power. Some viewed it as an example of the power system being willing to violate
fish operations instead of making the needed investments in an adequate power
supply. Others viewed it as a reasonable and prudent step given the high cost
and poorly demonstrated biological effectiveness of spill. The debate continues
It is tempting to believe that the factors that led to and prolonged the
Western electricity crisis are no longer of concern. Have we learned our lesson?
Certainly the possibility of additional jurisdictions moving to retail
competition is much diminished if not eliminated. There is also a renewed
enthusiasm on the part of many utilities and their regulators for the vertically
integrated utility where the utility owns generation and is less reliant on “the
market.” Similarly, many utilities now have experience with demand management
programs that could, if maintained, serve them in good stead should another
crisis begin to emerge.
In many respects these are positive developments that represent a retreat
from excesses of the late 1990s. However, we believe it would be a mistake to
think it could not happen again. Market prices will fall again, and the “wrong”
economic signal again will be sent to power plant developers as demand for power
increases. The lessons learned during the energy crisis must be built into the
structure of our electricity system if we hope to avoid another crisis — or at
least soften its impact.
It is likely we will continue to see a mix of vertically integrated
utilities, a federal power-marketing agency, local distribution utilities and
competitive wholesale suppliers in the regional power system for the foreseeable
future. This mix will have elements of federal, state and local regulation and
competition. This mix results in uncertainty regarding roles and
responsibilities and lacks some of the elements necessary for it to function
The challenge for the Council and the region is to determine what will make
such a system function effectively in the future.
In the Fifth Power Plan, completed in December 2004, the Council forecasts
demand for electricity 20 years into the future, as required by the Northwest
Power Act. The most-likely forecast, according to the plan, is that demand will
grow from 20,080 average megawatts in 2000 to 25,423 average megawatts by 2025,
an increase of 1 percent per year.
Here is a synopsis of the plan’s recommendations for meeting this demand:
Conservation — improved energy efficiency — costs less than new generating
plants and provides a hedge against market, fuel, and environmental risks. The Council recommends that the Northwest increase and sustain its efforts to
secure cost-effective conservation immediately. The targets in the power
plan are ambitious but achievable: 700 average megawatts between 2005 and 2009;
and 2,500 average megawatts over the 20-year planning period.
Demand response — agreements between utilities and customers to reduce
demand for electricity during periods of high prices and limited supply — helps
stabilize prices and prevent outages. The Council recommends developing
500 megawatts of demand response between 2005 and 2009 and larger amounts
The power plan anticipated more than 1,100 megawatts of new wind power
between 2005 and 2014, based on what was known at the time about state
incentives for wind power and utilities’ integrated-resource plans. However, wind power development has proceed at a much faster pace. Between
January 2005 and late 2006, more than 970 megawatts of wind power were completed
or were under construction in the Northwest, and construction of another 660
megawatts or more was expected by 2008. Wind project developers have requested
integration services and facilities to add more than 3,000 additional megawatts
of wind power in the region over the next several years.
This rapid development seems to be spurred by many factors and assumptions,
including: 1) that federal tax credits for wind power plants will continue; 2)
that controls on greenhouse gas emissions from power plants will be enacted in
the future, encouraging wind power; 3) that construction costs for wind plants
will continue to decrease; 4) that wind power, an intermittent resource, can be
integrated into the existing power system at reasonable costs; and 5) that large
areas of land with access to high-voltage transmission will be available at
Prepare for new power plants:
risk-management strategy, the power plan defines a schedule for siting and
permitting new power plants in anticipation of construction. With siting
and permitting completed, actual construction can begin quickly when conditions
are best. If the projects are not needed, the expended costs are
relatively small. The schedule includes up to 5,000 megawatts of wind
power to be developed through the end of the 20-year planning period, 425
megawatts of high-efficiency coal-fired generation (begin construction by
January 2012) and, late in the planning period, additional natural gas-fired
plan identifies four policy issues that are critical to the future of the
region’s power supply:
- The future role of the Bonneville Power Administration
The Council recommends that Bonneville sell electricity from the existing
Federal Columbia River Power System to eligible customers at cost. Customers that request more power should be required to pay the additional cost. The Council recommended that Bonneville implement this change through new
long-term contracts to be offered by 2007. The Council also believes that
Bonneville must continue its commitment to support conservation, renewable
energy and fish and wildlife mitigation.
- High-voltage transmission
Adequate transmission is critical to any of the new generating resources
identified in the plan. The move toward deregulation and expansion of
wholesale electricity markets, along with changes in technology, altered the
character of the traditional transmission system. Questions of how to plan
for, build, pay for and effectively manage the region’s transmission system are
becoming critically important. The Council supports and is an active
participant in regional efforts to resolve these problems.
- Regional power system adequacy standards
An adequate power system has a high probability of being able to maintain
service when the region experiences a poor water year, unexpected growth in
demand for power or the failure of some new resources to perform as planned. The power plan evaluates alternative regional adequacy standards and their
interaction with the western United States power system. The Council is
committed to working with regional utilities and regulators to develop a
standard that will assure an adequate power supply while being fair and
equitable to all parties.
- Integrating fish and wildlife recovery with power planning
The Council’s Northwest Power Plan and Columbia River Basin Fish and Wildlife
Program coordinate planning for future power supplies and for fish and wildlife
affected by hydroelectric dams. The Council committed in the power plan to
improve coordination of power and fish and wildlife issues with other entities
in the region.