The Generation Professionals Group is for utility professionals who work in biomass, coal, gas/oil, hydro, natural gas, or nuclear power generation fields. 


You need to be a member of Energy Central to access some features and content. Please or register to continue.


Columbia Generating Station Market Test


A recent Bloomberg news service article[i] reported that nuclear power reactors’ profits have been squeezed by cheap natural gas and renewables. The revelation came with a better understanding of the economic magnitude. More than half of U.S. nuclear reactors are bleeding cash, racking up losses totaling about $2.9 billion a year.

Nationally, nuclear power plants are getting paid $20 to $30 a megawatt-hour for their electricity. Meanwhile, it costs them an average of $35 a megawatt-hour to run. That puts 34 of the nation’s nuclear power plants out of the money, appearing to operate below their economic break-even point. 

Unless subsidized, any facility running below its economic break-even point will eventually have to shut down. The economic shutdown rule states that “to produce in the short run, a firm must earn sufficient revenue to cover its variable costs.”


Could it be true that Columbia’s variable running costs are higher than other available options?  Because the Columbia constitutes about 3.5 percent of the regional power capability[ii], annual monetary gains or losses from its operation are a multibillion-dollar proposition.

What’s a market test?  We market-test every day when comparing the value of the goods and services we purchase.  Consider this market test example. All other things being held equal, which service station would you buy gasoline from?  Would it be the service station selling fuel for $5 per gallon, or would you purchase from the one selling exactly the same product for $2.50 per gallon?  My bet is that you’ll be pulling into the less costly service station along with me.

In 1998, BPA stated it would conduct a market test of Columbia to assess its economic competitiveness.  This resulted from customer complaints that it would never be cost-effective.[iii]

BPA’s market test commitment stated: “This biennial test will determine whether the market value of the WNP-2 (now Columbia) output recovers annual operating costs of the plant. As recommended in the Cost Review, BPA will evaluate plant termination if operating costs are projected to exceed revenues achievable at market prices.” BPA also stated its intention to solicit input, in a public process, on the precise nature of this market test.

The market test might have happened, but the exigencies of the 2000-2001 energy crisis took over.  Many factors, including drought in the Pacific Northwest and fraudulent market manipulation, created a “perfect storm” for capacity shortfalls and volatile power prices.[iv] 

BPA suddenly had to acquire 3,300 megawatts during a period of extremely high prices.

Columbia reportedly saved the region $1.4 billion dollars[v] compared to buying wholesale power from the Mid-Columbia (Mid-C) trading hub.

Years passed, and the market test issue was raised in 2013 in a battle between two consultants. 

One concluded that Columbia’s power was significantly more expensive than wholesale power purchased from the Mid-C hub.  The report found if BPA could substitute Mid-C hub electricity for Columbia’s through 2043, the region would save at least $1.7 billion. The other report concluded that continuing to operate Columbia could save consumers $1.6 billion through 2043, compared to constructing a new natural gas generation system.

Certainly, both are correct.  However, they addressed two entirely different questions. Even its supporters admit Columbia’s current running costs exceed Mid-C hub prices.  However, permanently closing Columbia, and losing its capacity, deserves serious consideration.

“Clearly, the two studies were prepared by experts and assess an important resource in the Northwest power supply. The fact that they arrive at polar opposite conclusions is a puzzle, but also is an important issue for the region. Independent analysis of future resource costs is critical to making the best decisions about future sources of electricity. There are uncertainties on both sides of the issue, such as the future cost and performance of CGS, safety considerations of nuclear power, and costs of alternative power supplies.”[vi]

This multibillion-dollar puzzle certainly warrants a transparent evaluation on a regular basis. 


Are there potentially billions of dollars in gains or losses from running Columbia compared to purchasing electricity from the Mid-C hub?

As an economist, the challenge was to develop a transparent first-order model to evaluate the market test hypothesis. The math is pretty simple: mainly subtraction and multiplication. 

The transparent data used to populate the model came from two sources. The Mid-C hub alternate came from the Northwest Power and Conservation Council.[vii] It included the Mid-C’s historic and forecast annual wholesale electricity price data.  Columbia’s historic and forecast variable running cost data was obtained from Energy Northwest.[viii]

For example, in 2015 the Mid-C hub price was $21.70 per megawatt-hour.  Energy Northwest reported Columbia’s variable running cost was $50.50 per megawatt-hour that year.  This means Columbia cost $28.80 more per megawatt-hour of electricity than the Mid-C hub.

Columbia’s energy production was also obtained from Energy Northwest[ix], and the plant produced 8,142,000 megawatt-hours of electricity in 2015.  By multiplication, Columbia’s cost about $234 million more than the equivalent Mid-C hub electricity.

Columbia became the higher cost resource beginning in 2009, and projections of its costs versus the wholesale market indicate that it will remain the higher cost resource through 2028. Regional ratepayers could cumulatively pay about $1.64 billion more for Columbia’s power than from using Mid-C hub electricity as a result. 

The Council provided two sensitivity studies to bracket its Mid-C hub price forecast.  Using the high Mid-C forecast, Columbia’s losses were lower, but ratepayers could still pay a cumulative $550 million more for Columbia’s power than for the Mid-C hub. However, using the Council’s low Mid-C forecast, Columbia’s potential losses are staggering.  Ratepayers could pay a cumulative $3.1 billion more for Columbia’s power than for using the Mid-C hub by 2028.

Economic impacts for individual BPA customers can be approximated by using their Tier One Cost Allocator (TOCA), one of BPA’s customer charge billing determinants.  For example, a hypothetical customer with a 1 percent TOCA could cumulatively pay about $16.4 million more for Columbia’s power than for the Mid-C hub.[x]  

BPA’s largest customer is Snohomish County PUD #1, which could pay a cumulative $172 million more for Columbia’s power than for the Mid-C hub power under the medium-price forecast. The bracket for the two sensitivity studies ranged between a loss of between $60 million and $340 million.

What alternatives to Columbia might provide better economic choices for our future?


It’s well-proven and reliable: Energy efficiency is the lowest-cost and the highest-priority resource to meet new demand.  Regionally, by saving more than 6,000 average MW since 1978, energy efficiency has proven an effective alternative to constructing new thermal power plants -- at about 20 percent of their cost. Because energy efficiency costs about two-thirds of BPA’s basic power rate, it also helps mitigate Columbia’s negative economic impacts, which contributed to BPA’s increased power costs since 2009. 

Three years ago, the Swedish Academy of Sciences awarded the Nobel Prize to the inventors of the blue LED bulb. In industrial economies, replacing wasteful inefficient lighting with LEDs would reduce electricity consumption by about 15 to 25 percent, the Academy said.

Regionally, 15 percent waste is roughly equal to the electrical energy now consumed in Idaho and Montana. That’s right: In theory, both Idaho and Montana could be powered with the energy now wasted by using inefficient lighting.  Wasted electrical energy from inefficient lighting is the third-largest “state” in the Pacific Northwest.  Put another way, wasted electrical energy from inefficient lighting represents a potential resource equal to more than 2.5 Columbia power plants.[xi]

Perhaps you’re thinking that while LED lighting may be a significant opportunity, how long will it take for it to reach a dominant market penetration? LEDs are projected to have an 84 percent market share of lumen-hour sales in the general lighting market by 2030.  Nationally, LEDs will save consumers around $26 billion at today’s energy prices, and will reduce energy consumption equal to the use of about 24 million U.S. homes.[xii]


A first-order market test suggests Pacific Northwest electric customers could pay about $1.64 billion more for Columbia’s power than for the equivalent amount from the Mid-C hub through 2028.  

While unlikely, another “perfect storm” could wipe out the anticipated benefits gained from Mid-C hub replacement power.  However, long-range national forecasts indicate power costs will continue to be stable through 2050.[xiii]

To best ensure its ratepayers are provided reliable power at the least cost, BPA should conduct its own market test to assess Columbia’s economic competitiveness, as it originally pledged doing in 1998. 

BPA also stated its intention to solicit input, in a public process, on the precise nature of this market test. This market test should evaluate Columbia’s termination if its running costs are projected to exceed market prices.

Energy efficiency measures, such as substituting LEDs for inefficient lighting, can more than offset Columbia’s lost capacity in the regional power market.  We can avert these potentially massive economic losses. Why pay more when you can use less and get the same result for less money?

[i] More Than Half of America's Nuclear Reactors Are Losing Money.  June 14, 2017.

[ii] Capability is the maximum amount of energy the plants are capable of producing over the course of an average year.

[iii] BPA Cost Review Implementation Plan Fact Sheet #8, November 1998.

[iv] Energy Crisis of 2000/2001.  SEE:

[v] Columbia Generating Station Regional Value.  SEE:

[vi] Hydrothermal Power Program.  SEE:

[vii] Northwest Power and Conservation Council 2016 Wholesale Electricity Price Forecast.


[viii] Energy Northwest Annual Reports 2006-2016.  SEE:

[ix] Fiscal Year 2018 Columbia Generating Station Long Range Plan (2017-2028).  SEE:

[x] The remaining “Top 10” BPA customers are: Cowlitz County PUD #1, Seattle City Light, PNGC Power, Tacoma Public Utilities, Clark County PUD #1, Eugene Water & Electric Board, Benton County PUD #1, Flathead Elec Cooperative, and Central Lincoln PUD. Combined, they represent about 52% of BPA’s total power sales with a cumulative equivalent estimated loss of $845 million.

[xi] LED technology uses semiconductors, and will respond to a variant of Moore’s Law; future products will be better and cheaper than today’s. Several manufacturers have ENERGY STAR®-qualified bulbs surpassing 100 lumens per watt. Comparatively, the typical efficiency of CFL bulbs is 55 to 70 lumens per watt. Traditional incandescent bulbs, which do not meet current efficiency standards, provide 13 to 18 lumens per watt.

[xii] LED Lighting Forecast. SEE:

[xiii] Annual Energy Outlook 2017: Total Energy Supply, Disposition, and Price Summary Table. 


Phil Lusk's picture

Thank Phil for the Post!

Energy Central contributors share their experience and insights for the benefit of other Members (like you). Please show them your appreciation by leaving a comment, 'liking' this post, or following this Member.


Michael Paoli's picture
Michael Paoli on Oct 25, 2017 10:31 pm GMT

The above post by Phil Lusk is based on an apples-to-apple pie comparison of spot market prices to the actual costs of producing wholesale power. As chief researcher for the Guacamole Fund, a group focused on “a non-nuclear future,” we believe the author’s examination of nuclear power economics only considers part of the value proposition provided by our region’s sole source of clean nuclear energy.

In contrast, the Northwest’s Public Power Council recognizes Columbia Generating Station as a linchpin for clean energy diversity, grid resiliency and low-cost predictability. Experts at the PPC understand that daily prices for energy at the Mid-Columbia trading hub do not represent the full cost or value of producing wholesale power. In fact, using spot market prices as the sole basis for comparison with firm power generation undervalues all firm power resources, not just nuclear.

The Mid-Columbia spot market is a daily bilateral market for wholesale energy. The amount of electricity actually traded in this incremental market is small when compared to the overall average megawatts required to power the Northwest. Prices in this spot market are driven by short-run variable costs – such as fuel and variable operations and maintenance expenses – of incremental generation in the Pacific Northwest. Examination of actual Mid-Columbia market prices and regional generation patterns demonstrates that daily spot prices do not allow generation owners to recover their fixed costs, such as depreciation, interest expense, labor and other fixed O&M expenses.

To illustrate, at certain times, such as when regional loads are low to moderate and hydro and wind generation are high, wind and hydro are the incremental sources of generation in the Northwest and their low variable costs drive the Mid-Columbia spot market price. At other times, such as during high system demand and low hydro and wind generation, natural gas-fired generation is the incremental source of generation and its somewhat higher variable costs set the spot market price.

Recent changes in the regional generation fleet have made it more difficult to recover fixed costs in the spot market. Large amounts of wind power have been added in the region (more than 8,000 megawatts to date), and surplus solar generation from California is imported into the Northwest. These resources, with low variable operating costs, are the incremental sources of generation driving down spot market prices more often than not.

This reduces the effectiveness of the spot market as a mechanism to recover power plant fixed costs, further negating the validity of the Mid-Columbia price index as a benchmark for valuing generation. Furthermore, the spot market does not value the capacity, resilience and other attributes that power plants provide.

Since the daily spot market for wholesale power is not an effective mechanism for recovering fixed costs of generation, how are such costs currently recovered in the Pacific Northwest? The answer lies in the fact that most wholesale power is either generated by, or sold via bilateral contracts to, utilities who then sell it to their retail customers at cost-based rates. As a result, most fixed costs of generation are recovered directly from consumers in retail utility rates, rather than via the spot market.

Mr. Lusk’s argument lacks this important understanding of the full range of cost factors and how fixed costs are recovered. A market price index for daily spot market energy transactions is not a valid or accurate representation of the actual value of power produced by Columbia Generating Station or any other firm, long-term power supply resource.

The full “all in” economic value of Columbia is further strengthened by the plant’s environmental contribution.

The Northwest Power and Conservation Council concluded in its Seventh Northwest Power Plan that “it’s not possible to entirely eliminate carbon dioxide emissions from the power system without the use of nuclear power or emerging technology breakthroughs.” And studies by the International Energy Agency, the U.N.’s Intergovernmental Panel on Climate Change and the U.S. Energy Information Administration suggest that we absolutely cannot prevent the rapid pace of climate change without preservation of current nuclear resources and aggressive investment in new nuclear.

The current 99 nuclear plants in the United States provide nearly 20 percent of the country’s power, and an impressive 60 percent of our country’s clean energy. And what is, or will be, the price on carbon? Columbia alone prevents 3.6 million metric tons of carbon dioxide emissions annually compared to the best-case natural gas replacement option. For a region facing imminent carbon constraints from nine coal and 29 gas plants, the zero-carbon nature of our existing nuclear facility will result in an even greater premium on its value.

(Energy Northwest staff contributed to this response.)

Get Published - Build a Following

The Energy Central Power Industry Network is based on one core idea - power industry professionals helping each other and advancing the industry by sharing and learning from each other.

If you have an experience or insight to share or have learned something from a conference or seminar, your peers and colleagues on Energy Central want to hear about it. It's also easy to share a link to an article you've liked or an industry resource that you think would be helpful.

                 Learn more about posting on Energy Central »