The Northwest Power Act directs the Council to give priority, in the power plan, to resources “which the Council determines to be cost-effective.” The Act also gives first priority to conservation among all resource types. Northwest Power Act, Section 4(e)(1). In the draft 2021 Power Plan, the Council recommends that the region acquire at least 750 average megawatts of cost-effective conservation by 2027 and at least 2400 average megawatts by 2041.

To determine whether a resource is cost-effective under the Act is a comparative exercise. That is, a resource is cost effective, as the term “cost effective” is defined in the Act, if it has an “estimated incremental system cost no greater than that of the least-cost similarly reliable and available alternative measure or resource, or any combination thereof.” Section 3(4)(A).

Information useful for that comparison begins with the fact that conservation measures are of different types and costs. So, the Council has developed conservation supply curves, which identify conservation measures, their costs, and the amounts potentially available over the planning period assuming what has been the standard pace of implementation. The Council has also developed information about the costs and power system characteristics of different generating resource types, as well as the costs and energy benefits of demand-side resources that reduce the need for additional capacity. And, the Council also develops information about the costs of purchasing power on the wholesale market, which can often be the lowest-cost source for needed generation, rather than building or acquiring the generating resource itself.

The cost-effective comparison is then based, at its simplest, on selecting bundles of conservation measures available at different costs up to the cost of the lowest-cost generation alternative(s), and summing the total. The analysis, however, has to consider a number of factors important under the statute and inherent in producing a power plan that assures an adequate, efficient, economical and reliable power supply.

The most important factor, in the context of this power plan, is that to be cost effective, a resource has to be “reliable and available within the time needed” and “meet or reduce electric power[1] demand” in a way similar to the comparison resource, so that the cost-effectiveness comparison is between conservation measures and “the least-cost similarly reliable and available alternative measure or resource, or any combination thereof.” Moreover, the Council has to make this determination as part of a planning projection into the uncertain future – as the Ninth Circuit recognized in its review of the Council’s first power plan (Seattle Master Builders), the Act “allows the Council the flexibility to define cost effectiveness not in terms of current energy needs, but by reference to whether a resource is forecast to be available within the time needed.”

In the last three power plans, the Council determined that gas-fired generating plants were the lowest-cost generating resource for comparison to the costs of conservation resources within the planning period. [2] Gas-fired power plants provide the full range of power system characteristics needed by the system – not just energy and peaking capacity, but also balancing and flexibility services – as well or even better then energy efficiency measures. So, a direct comparison between the costs of conservation and the costs of this least-cost resource was an apt one, becoming the primary driver for the Council’s determination of the amounts of cost-effective conservation for the Fifth, Sixth and Seventh power plans. The Council had less need to consider factors beyond this direct cost comparison, which in the last plan, for example, brought into the resource strategy conservation measures with a levelized cost of up to and even over $100/MWh. By this method, the Sixth Power Plan (2010) set a five-year regional target for acquiring cost-effective conservation of 1200 aMW, and a twenty-year regional conservation target of nearly 6000 aMW. Similarly, the Seventh Plan (2016) called for the region to “achieve a minimum conservation goal of 1400 aMW by 2021, 3000 aMW by 2026, and 4300 aMW of cost-effective conservation by 2035.”

Things are different in this power plan, significantly so. The costs of renewable generation – solar and wind – have dropped dramatically, at the same time that state laws and other clean policies also require their addition to the system. The result is that utility-scale renewable resources are the lowest cost generating resource, and the power plan’s resource strategy assumes significant addition of renewable resources in the Northwest and the West as a whole. The costs of gas generation have also dropped significantly, but not to the same degree as renewables, and there are policy and other obstacles that will limit the addition of new gas plants other than in isolated cases.

However, renewable resources have different and more limited power system characteristics than either gas plants or many energy efficiency measures. Solar and wind plants provide substantial amounts of energy, but less in the way of capacity value, and their use requires instantaneous balancing and daily ramping flexibility services from other resources – and the more renewable resources that are added, and the more existing thermal resources in region retired (the region is projected to see approximately 4000 mw of coal plant retirements in the next decade), the more of these other power system benefits are needed from other resources, including energy efficiency.

In other words, for this plan, the lowest-cost renewable generating resources do not meet or reduce electric power demand to the same extent as conservation measures, and are not a similarly available and reliable alternative to conservation, at least not completely. What the Council had to do in this plan, as the statute itself recognized might need to happen, is compare energy efficiency against a “combination” of resource alternatives that meet the system needs to determine how much conservation is cost-effective for inclusion in the resource strategy.

While all these considerations are important components of a cost-effective comparison in the current context, not all are easy to quantify. And this is especially so as the Council has to decide in the power plan on a cost-effectiveness level, as the Ninth Circuit noted 35 years ago, based not on current energy needs and costs, but through projections and forecasts and considerations as to what energy needs might be in the future and what resources might be similarly available and reliable to meet those needs and at what costs – and do so here all in the context of a dramatically transforming power system. Hedging against this risk and uncertainty in deciding on an appropriate cost-effective resource strategy has been a crucial part of the Council’s efforts in the 2021 plan to be able to assure the region it can continue to enjoy an adequate, reliable and economical power supply.

Given these considerations, the Council’s technical staff relied on a wide range of information, model analyses and considerations in recommending to the Council an appropriate level of conservation to include in the plan’s resource strategy as cost-effective, a recommendation the Council accepted for the draft plan. At the beginning of the analytical work, the bare comparison of the lowest-cost solar generation costs with the conservation costs and amounts in the supply curves as developed in the baseline model analyses under normal load expectations identified a base amount of conservation needed of roughly 500 aMW by 2027 and approximately 1400 aMW over the 20-year planning period.

The Council staff considered these numbers a starting place for the analysis, not an end result. The Council received comments (such as a letter in June 2020 from the Northwest Requirements Utilities) arguing that this was the result the Council had to accept as the cost-effective level of conservation for the plan’s resource strategy. This is not accurate, for all the reasons noted above. The baseline analyses represented an amount of conservation that results from comparing a generating resource that provides mostly energy benefits (but not always available) with energy efficiency measures providing a range of energy benefits not yet fully captured in the analysis. Moreover, these baseline results, while resulting in a low-cost strategy, do not provide for a fully adequate and reliable system. The baseline results were subsequently run back through the Council’s resource adequacy model (GENESYS) which indicated this would not be an adequate system by itself, and would require something further in terms of resource choices or changes in how system reserves are held in order for the system to be adequate. As a result, the Council concluded that that this baseline result did not represent, for the purposes of determining the cost-effect amount of conservation, a complete comparison of similarly reliable and available alternatives for meeting or reducing electric power needs.

The Council’s technical staff used various analytical techniques and considerations to further inform the determination of the appropriate amount of cost-effective conservation for the region. These included:

  • Adding renewables to the power supply, especially solar, results in a dramatic need to ramp up the addition of other resources during the morning and evening peaks and overnight – a change in the way the power system operates already seen and that will accelerate as more renewables are added and coal thermal generation is retired (and eventually as gas resources need to be retired or utilized much less under state policies). The Council’s analysis of reserve and reliability requirements and cost-effective methods of providing reserves indicated that the resource strategy resulting from the baseline analysis would also require different strategies to preserve an adequate system at the lowest probable cost. Reserves are one method of addressing this ramping. Alternatively, conservation that reduces electric power demand during those periods can meet this need, at a higher cost but with less risk than reserves. Another potential resource for meeting this ramping need is battery storage, which is a higher cost alternative resource relative to 750 aMW of conservation.
  • Similarly, if natural gas generation is considered a more appropriate comparative resource to value the capacity, balancing and flexibility attributes of conservation, significantly more conservation is cost-effective than in the baseline. The Council staff ran a sensitivity in which new renewables were excluded. This resulted in the natural gas generation as the primary low cost, available and reliable resource for comparison and resulted in approximately 750 aMW of conservation being acquired by 2027.
  • If renewable resources are not built in the amount and pace indicated in the baseline analyses, which is an obvious risk in a resource strategy calling for an aggressive and fast-paced build of 1000s of MW of renewables in the Northwest and the West as a whole, the system analysis indicates the region would benefit from the inclusion of additional conservation.
  • Also, one factor in the baseline analyses that allows the region’s power system to tolerate the daily adding and dropping of significant renewable generation is the counter-effect of using the region’s hydropower system to ramp on and off in sync with the system needs. The daily river fluctuations and system generation patterns resulting from the increasing use of the flexibility of the federal hydropower system may not be sustainable, for a number of reasons ranging from impacts to fish migration to the firm power needs of Bonneville customers. As with so many other permutations, a level of conservation above the baseline amount is a cost-effective hedge against this risk and uncertainty, in helping to provide the flexibility and capacity needs that the hydropower system might not provide as fully, at a cost less than other generating resource alternatives.
  • The amount of energy efficiency acquired across the many scenarios shows that energy efficiency is very sensitive to assumed market process. The lower the projected market prices, the less energy efficiency acquired. The baseline analyses assume significant reliance on and effects from a west-wide wholesale power market that will itself be in the midst of a dramatic transformation, adding 100s of GW of mostly solar resources, with sustained exceedingly low and negative power prices. This is a very different world than today, resulting in significant uncertainty around the future market prices. Greater levels of energy efficiency than the baseline help protect the region in the event the dramatic resource or load additions and power market effects are different than expected.
  • Certain conservation measures – such as weatherization - provide a resilience benefit that supports homes and buildings being able to ride through extended power outages. This is an energy benefit difficult to quantify as part of resource costs and benefits.
  • Renewable generation costs are at a historically low level for generating resource costs in comparison to alternatives. As with energy needs, resource costs are a projection or forecast of future conditions. The Council’s 40-year power plan experience has consistently been the opposite of the current moment, with generating resource costs distinctly greater in relative terms, while environmental impacts of generating facilities have always been more significant than non-generating alternatives. Developing an amount of conservation greater than the baseline analysis is a useful hedge against the risk and uncertainty that generating costs could rise or environmental impacts and thus costs might be greater than in the current unusual moment.
  • The baseline analysis assumes a certain moderate rate for the addition of conservation. Accelerating the rate of addition of conservation significantly increases the amount of cost-effective conservation that is developed in the first five years of the plan. And an accelerated pace of conservation development seems a logical strategy for resource adequacy in the face of the retirement of 1000s of MW of coal-fired generation in the next decade, largely replaced by renewable resources. The accelerated implementation scenario tested by the Council indicated close to a 1000 aMW to be cost-effective in the next five years.
  • The baseline analyses assumed moderate load growth within a range of recent and historical experience. But, at least two states in the region, and many more across the West, are putting in place policies that would accelerate the electrification of vehicles and buildings. It is just as likely as not that this trend will further accelerate over the next decade, as will its implementation. And if so, loads could grow at a pace not seen in the 40-year history of the power plan. Analyses indicate that effective and efficient use of all clean resource alternative will be needed in that case for an adequate system, including substantially greater amounts of conservation. Also, the amounts of conservation potentially available will climb, as there will be more units on the electrical grid for conservation measures to apply to – another factor leading to a greater amount of conservation being cost-effective.

Most of the factors could not be quantified with precision, although values can be estimated for all. All indicate that the Council should forecast a greater need for conservation as cost effective than in the baseline comparison of conservation with the lowest-cost renewable generating costs. No one factor directly relates to the amount recommended by staff – some add minor value to the baseline; some would accelerate the cost-effective amount substantially above the recommended target; and others have a forecasted effect somewhere between the two. Balancing all the considerations, the staff recommended, and the Council accepted, that the region acquire at least 750 average megawatts of conservation by 2027 and at least 2400 average megawatts of conservation by 2041, as the amounts deemed by the Council to be cost-effective under Section 3(4) of the Act.

[1] Electric power is defined in the Act to mean electric peaking capacity, or electric energy, or both. Section 3(9).

[2] More precisely, the Council identified power purchased on the wholesale market as the comparison resource and costs, at least with regard to energy. But it did so while recognizing that wholesale market prices were largely driven by the cost of gas-fired combined cycle turbines and their gas fuel costs. For considerations of capacity, the Council also factored in the costs of a single cycle gas-fired peaking plant.