Publicly Traded Competitive Generators: Crawling Towards Extinction

admin
Print Friendly, PDF & Email

______________________________________________________________________________

Eric Selmon Hugh Wynne

Office: +1-646-843-7200 Office: +1-917-999-8556

Email: eselmon@ssrllc.com Email: hwynne@ssrllc.com

SEE LAST PAGE OF THIS REPORT FOR IMPORTANT DISCLOSURES

______________________________________________________________________________

September 7, 2017

Publicly Traded Competitive Generators: Crawling Towards Extinction

Despite the industry having wiped out a significant portion of its borrowings through bankruptcies and financial restructurings, Standard & Poor’s long term issuer ratings of the competitive generators remain deep in speculative territory, ranging from B+ (Calpine, Dynegy, Talen) to BB- (Covanta, NRG, Vistra). In this note, we consider whether the persistent financial weakness of the industry reflects inherent flaws in the merchant generation business model and, if so, whether the industry has a future in the public equity market. We conclude that competitive generation as a business has structural features that impede the recovery of invested capital with an adequate return, and that the industry operates in a regulatory and legislative environment that further erodes its profitability. As a result, the competitive generation industry offers limited opportunity for capital investment at attractive returns, and thus little potential for growth. Holding portfolios of depreciating assets whose cash flows can be expected to decline over time, competitive generators have become increasingly unattractive to public equity investors. Private equity investors, by contrast, can capitalize on this financial profile by eliminating growth capex and allocating all available cash flow to debt service, permitting higher leverage, higher equity returns and concentrated, private ownership focused on operational improvement. Over the last two years, competitive generation assets have steadily migrated into private hands through a series of transactions, including Riverstone’s acquisition of Talen Energy at the end of 2016, the acquisition by Blackstone and Arclight of a portfolio of AEP’s competitive generation assets early in 2017, and the recently announced acquisition of Calpine by Energy Capital Partners. We do not expect this to be the last such transaction; rather we expect Dynegy to be acquired by Vistra or private equity investors, and in the long run also expect to see Vistra and NRG Energy exit the public markets.

  • The competitive generation industry has structural features that impede the recovery of invested capital, and operates in a regulatory and legislative environment that further erodes its profitability. Among the pitfalls facing investors in competitive generation assets are:
  • the inherently cyclical pricing of a high capital cost, low variable cost commodity industry, which can cause power prices to remain below the level required for capital recovery for decades at a time;
  • the declining long run marginal cost of power generation, which continuously pushes older power plants up the supply curve, eroding their profitability from the moment they enter operation;
  • the extreme difficulty of evaluating capital investments in an industry where a project’s generation gross margins are a function of the future relative costs of nuclear fuel, coal, natural gas and oil;
  • the support to the growth of zero cost renewable generation provided by state renewable portfolio standards, and the consequent erosion of the output, prices and revenues of conventional generators;
  • the rising going-forward costs faced by fossil fuel and nuclear generators due to stringent air and water emissions standards that tend to tighten over time.
  • Pricing at variable cost prevents capital recovery. Commodity industries follow a textbook pattern of price formation, in which competition among producers drives the price of the commodity down to the variable cost of production of the marginal supplier. In industries like power, where the cost of capital recovery is high relative to variable cost, prices that reflect the variable cost of the marginal producer can be well below the all-in cost of supply, which includes the cost of capital recovery as well as the variable cost operation. Electricity prices well below the all-in cost of power can persist for years. Under these circumstances, even the most efficient power plants have difficulty generating electricity at a gross margin sufficient to recover their invested capital with an adequate return. (See Exhibit 3.)
    • Despite inadequate returns to capital, capacity rarely leaves the market. Power stations have a useful life of 25 to 50 years, and the bankruptcy code permits failing plants to cram down their financial obligations to creditors to a level commensurate with the market revenues of the plant.
    • Capacity markets, such as those in PJM and New England, exacerbate this problem by ensuring revenues are high enough to keep plants operating, but rarely allowing prices to rise high enough to generate reasonable returns on capital.
    • By contrast, energy-only markets, such as ERCOT, are more likely to see revenues fall to a level that encourages large generation owners to retire portions of their fleet. While we expect to see such retirements announced in the next few years, this will provide only a temporary respite as the surge in prices that follows will encourage the construction of new power plants, likely resulting in oversupply and repeating the cycle.
  • Declining long run marginal cost erodes profitability. Technological innovation has continuously reduced the cost of generation equipment while enhancing its efficiency, steadily reducing the all-in cost of supplying electricity. Older power plants are thus pushed up the supply curve, ceding their position at the bottom of the supply curve to new entrants deploying newer, lower cost technologies. Over time, this process continuously erodes both the output and margin per MWh of existing units.
  • Power project evaluation is opaque. Investors in a power plant must anticipate not only the boom/bust cycle of electricity prices and the potential for the introduction of disruptive new generation technologies, but also changes in the relative costs of the fuels used to generate electricity. Because the power supply curve includes so many different power generation technologies, the cost of supplying electricity is a function of the prices of nuclear fuel, coal, natural gas and petroleum. A power project’s future generation gross margin is thus a function of the relative costs of these fuels over the project’s life.
  • State renewable targets inject zero variable cost competition. State renewable portfolio standards have had the effect of suppressing wholesale power prices and eroding the output of conventional generators. By supporting the construction of new renewable resources, state renewable generation mandates have added zero variable cost generating assets at the bottom of the supply curve, reducing both the hours of operation as well as the prices received by the higher cost conventional generating resources on the system.
  • Markets offer no mechanism for the recovery of environmental costs. Over the last decade, increasingly stringent air and water emissions standards have been imposed on generators, yet power markets offer no mechanism to recover the required environmental capex. These measures have frequently increased the going-forward cost of existing power plants so much as to force their early retirement.
  • Faced with stagnant power demand, the output of existing power plants will fall as planned renewable and combined cycle gas turbine (CCGT) capacity is added. Over the next three years (2017-2019), increased low cost generation from planned additions of wind, solar and new CCGT capacity outpace the expected growth in power demand. As a result, we expect the highest variable cost generators (the marginal, price-setting units) in each of the five RTOs to experience significant declines in power output and capacity factors through 2019 (see Exhibits 7 and 8). (In most of the RTOs, CCGTs are the marginal, price-setting units, but a mix of CCGTs and coal fired power plants are on the margin in PJM.)
  • Wholesale power prices prevailing today do not permit recovery of new CCGT costs with an adequate return. As Exhibit 10 illustrates, the expected gross margins of new CCGTs fall well short of the gross margins required for capital recovery in each of the principal competitive markets. The shortfalls range from 13% of the required gross margin in the New York ISO’s Zone G to 59% in ERCOT, which has no capacity market. If the competitive generation industry cannot recover the capital required for the construction of new generation capacity, its growth prospects are inherently limited.
  • Our conclusion is that the competitive generation industry faces structural impediments to profitability that limit the opportunity for capital investment at attractive returns. The industry thus offers little potential for growth, and is increasingly unattractive to public equity investors.
    • As illustrated in Exhibits 11 and 12, competitive generators’ return on equity and invested capital have consistently fallen below their cost of capital.
    • We find the financial metric favored by the managements of competitive generators, adjusted free cash flow yield, overstates shareholder returns by combining the return on and of invested capital. Recalculated to exclude the recovery of capital invested, the remaining cash available for distribution offers an unacceptably low yield to shareholders. (See Exhibit 15.)
  • Private equity investors, by contrast, can capitalize on the financial profile of competitive generators by eliminating growth capex and allocating all available cash flow to debt service, permitting higher leverage, higher equity returns and concentrated, private ownership focused on operational improvement. Furthermore, private equity is better positioned than public equity investors to assess the option value of generation assets that results from the cyclical volatility of prices for power and fuel.
  • As a result, over the last two years, competitive generation assets have steadily migrated into private hands through a series of transactions, including Riverstone’s acquisition of Talen Energy in 2016, the acquisition by Blackstone and Arclight of a portfolio of AEP’s competitive generation assets early in 2017, and the recently announced acquisition of Calpine by Energy Capital Partners.
  • We are removing Calpine from our list of least attractive IPPs due to its planned sale to Energy Capital Partners. While we still have significant concerns about the valuation of Dynegy, we are not adding it to our list of least attractive IPPs due to the likelihood that it, too, will be acquired, either by Vistra or private equity investors. In the long run, we also expect to see Vistra and NRG Energy exit the public markets.

Exhibit 1: Heat Map: Preferences Among Utilities, IPP and Clean Technology

Source: SSR analysis

Details

The 25 year history of competitive generation in the United States has been a turbulent one. Each of the publicly traded, independent power producers that emerged following the wave of deregulation in the 1990s – Calpine, Dynegy, Mirant and NRG Energy – has undergone bankruptcy. So have the competitive generation subsidiaries of AES (AES Eastern Energy), Edison International (Midwest Generation) and PG&E Corp. (National Energy Group). More recently, the largest leverage buyout in American history, the $32 billion acquisition of TXU by KKR and Texas Pacific Group in 2007, failed spectacularly, ending in bankruptcy only seven years later.

Despite the industry having wiped out a significant portion of its borrowings through bankruptcies and financial restructurings, Standard & Poor’s long term issuer ratings of the competitive generators remain deep in speculative territory, ranging from B+ (Calpine, Dynegy, Talen) to BB- (Covanta, NRG, Vistra). In this note, we consider whether the persistent financial weakness of the industry reflects inherent flaws in the merchant generation investment model and, if so, whether the industry has a future in the public equity market. We conclude that competitive generation as a business has structural features that impede the recovery of invested capital with an adequate return, and operates in a regulatory and legislative environment that is hastening its demise.

In the sections that follow, we will consider each of these risks in turn: the inherently cyclical pricing of a high capital cost, low variable cost commodity industry, which can cause power prices to remain below the level required for capital recovery for decades at a time; the declining long run marginal cost of power generation, which pushes older power plants up the supply curve from the moment they enter operation; the extreme difficulty of evaluating capital investments in an industry where a project’s generation gross margins are a function of the future relative costs of nuclear fuel, coal, natural gas and petroleum; the support to the growth of zero cost renewable generation provided by state renewable portfolio standards, and the consequent erosion of the output, price and revenues of conventional generators; and the rising going-forward costs faced by conventional generators due to the EPA’s increasingly standards for air emissions, cooling water intake and coal ash disposal. We end by summarizing the outlook for competitive generators in each of the nation’s principal competitive markets, and assessing whether these markets can economically support the construction of new power plants.

Our conclusion is that the merchant generation industry combines extremely high economic and financial risk, with little opportunity for capital investment at attractive returns, and therefore little potential for growth. Without the need to access the public markets to fund growth, the management of the existing competitive generation assets is likely best left to private equity investors, who can harvest the declining cash flow of these assets over their remaining useful life.

1. High Capital and Low Variable Costs Lead to Cyclical Pricing, Impeding Cost Recovery

Like many commodity industries, power generation is characterized by an undifferentiated product, high capital costs for new production plant, and low variable costs of production. Such industries tend to be characterized by highly cyclical prices and profit margins. This intense cyclicality frequently limits the ability of participants to recover their investments with an adequate return on capital through the cycle.

Commodity industries tend to follow a textbook pattern of price formation, in which competition among producers drives the price of the commodity down to the variable cost of production of the marginal supplier. In industries like power, where the capital cost of production is high relative to variable cost, prices that reflect the variable cost of the marginal producer can be well below the all-in cost of supply, which includes the cost of capital recovery as well as the variable cost operation. Importantly, electricity prices well below the all-in cost of power can persist for years. Power stations generally have a useful life of 25 to 50 years, and the bankruptcy code permits failing plants to cram down their financial obligations to creditors to a level commensurate with the market revenues of the plant. Once built, therefore, generation capacity tends to have a long life. And as long as prevailing demand can be met from existing production capacity, prices will continue to reflect the variable cost of supply from existing plants. Under these circumstances, even the most efficient producers have difficulty supplying the commodity at a gross margin sufficient to recover their invested capital with an adequate return.

This problem is acute in the power industry, where variable production costs vary between from zero to approximately half of the all-in cost of new generation capacity. Exhibit 3 compares the variable cost of supply from a range of generation technologies with the all-in cost of new power plants. In most U.S. power markets, the marginal, price-setting generating units during the peak demand hours of the day are combined cycle gas turbines. Given that the average heat rate of the U.S. CCGT fleet is ~7.4 MMBtu/MWh, and assuming the price of natural gas is $3.00/MMBtu, the fuel cost of these marginal generating units can be estimated at ~$22.00/MWh; adding variable operation and maintenance expense of ~$4/MWh, the total variable cost of operation of existing CCGTs can be estimated at ~$26/MWh. When power prices settle at this level, full cost recovery is impossible for a new CCGT, whose all-in cost of production can be estimated at ~$49/MWh. It is also impossible for new coal fired steam turbine generators, whose all in cost of production can be estimated at ~$61/MWh; new nuclear plants, whose all-in cost can be estimated at ~$98/MWh; new gas turbine peakers, whose all-in cost can be estimated at $166/MWh; and even new wind and solar power plants, whose all-in cost can be estimated at $32 and $49/MWh, respectively. (See Exhibit 2). Finally, we should emphasize that the variable operating cost of combined cycle gas turbine generators sets the price of power only when they are called upon to run; when they are not, as is frequently the case during off-peak hours, power prices will fall to reflect the variable cost of operation of the existing fleet of coal fired (~$20/MWh), nuclear ($10/MWh) or renewable power plants (zero).

Exhibit 2: The variable cost of operating new power plants is a small fraction of their total cost

__________________________________________________________

Source: Lazard’s Levelized Cost of Energy Analysis – Version 10.0, SSR analysis

Exhibit 3: Prices reflecting the variable cost of generation are therefore inadequate to allow recovery of capital costs of new power plants

__________________________________________________________

Source: Lazard’s Levelized Cost of Energy Analysis – Version 10.0, SSR analysis

Only when demand for power begins to outstrip the capacity of the existing fleet of power plants will prices rise to reflect the all-in cost of supply (i.e., the cost of building as well as operating new generation capacity) rather than the variable cost of production from existing plants. But given time, prices sufficient to recover the all-in cost of new generation capacity will attract new investment to the industry; these capacity additions alleviate the scarcity that led to the price spike; and eventually, as new entrants compete with each other for a share of the market, prices are once again driven down to the variable cost of the marginal supplier.

Thus, wholesale power prices have not been supported at levels reflecting the all-in cost of adding new generation capacity so much as they have been capped by it.

2. Declining Long Run Marginal Costs Further Erode the Industry’s Ability to Earn Adequate Returns

The commodity industries that, faced with this cyclical pricing dynamic, are most likely to earn an adequate return on capital are industries that exploit a limited natural resource whose long run cost of supply is rising. In extractive industries, for example, the lowest cost mineral deposits are developed first; as these are depleted, the industry will move on to develop deposits that are more difficult to extract, geographically more remote or which involve greater political risk. The higher all-in cost of production from these later projects tends to be reflected in a rising real price of the commodity, rendering earlier investments in the extraction of lower cost resources gradually more profitable over time.

Power generation is not one of these industries. The long run cost of supply of power is in large part a function of the cost and efficiency of generation equipment. Whereas the cost of extracting the world’s finite reserves of iron ore or petroleum may rise over time, human ingenuity has continuously reduced the cost of generation equipment while enhancing its efficiency, gradually but steadily reducing the all-in cost of supplying electricity.

This process seems to continue inexorably in the power industry. On occasion, a disruptive technology, such as the combined cycle gas turbine (CCGT) generator, can so reduce the cost of supplying electricity as to render earlier generation technologies economically obsolete. From 1990 through 2010, 75% of all generation capacity added in the United States was gas fired. The previously dominant generation technology, the coal fired steam turbine generator, accounted for only 7% of capacity additions over these two decades. More recently, gas fired capacity additions have fallen behind wind and solar; these two renewable resources have accounted for ~65% of all U.S. capacity additions in 2015 and 2016.

In extractive industries, therefore, the lowest cost mines developed in prior decades tend to enjoy the benefit of rising prices as higher cost operations are added to the high end of the supply curve. In the power industry, by contrast, older power plants tend to be pushed up the supply curve, ceding their previously advantageous position at the bottom of the supply curve to new entrants deploying newer, lower cost and more efficient generation technologies. Inescapably over time, this process erodes both the output and margin per MWh of existing units.

3. The Planning of Investments in Competitive Generation Is Materially More Difficult than in Other Industries

The challenges to investors created by the pricing dynamics described above are compounded by the difficulty of planning investments in competitive generation, which in many ways is materially more difficult than in other industries. Investors in generation must anticipate not only the boom/bust cycle of electricity prices, and the potential for the introduction of disruptive new generation technologies, but also changes in the relative costs of the fuels used to generate electricity, which can materially alter the expected profitability of existing power plants.

In many commodity industries, a single production technology is deployed across the industry: broadly speaking there is one way to extract copper, one way to smelt aluminum, one way to manufacture paper. Electric power, by contrast, is generated from a range of radically different technologies whose input costs bear no relation to each other: photovoltaic cells energized by the sun, turbines powered by wind, hydroelectric dams, nuclear reactors, gas fired combustion turbines, gas fired combustion turbines with heat recovery steam generators, and steam turbine generators powered by boilers burning coal, fuel oil or natural gas.

As a result of the power supply curve including many different power generation technologies, the price of power is a function of the prices of nuclear fuel, coal, natural gas and petroleum. As the relative cost of these fuels shifts, so will the generation gross margin of the power plants that burn them. By way of example, from 2008 to 2009, the average price of natural gas paid by electric power plants in the United States fell from $9.26/MMBtu to $4.93/MMBtu, a drop of 47%. Gas fired power plants were the marginal price setting units in the PJM power market at the time, therefore the load weighted, average price of power in PJM fell commensurately, dropping by 45% from $71/MWh in 2008 to $39/MWh in 2009. A 1000 MW nuclear power plant in the region, operating at a 90% capacity factor at a variable cost of operation of $11/MWh, would have earned a generation gross margin of some $475 million in 2008, but only $220 million in 2009, a drop of 53%.

Because the shape of the power supply curve derives from the relative cost of different generating fuels, themselves set in often cyclical markets, power price forecasts tend to be egregiously wrong, resulting in repeated, dramatic losses for investors in competitive generation assets. Over the five years from 1999 through 2003, 175 GW of new gas fired capacity was added to the U.S. generating fleet, increasing its capacity by almost a quarter. The wave of investment capitalized on abundant and cheap natural gas: over the fifteen years from the deregulation of the well head price of natural gas in 1985 through 2000, the average annual price of natural gas had varied between ~$1.50 and $2.50/MMBtu. Following the huge build-out of the gas fired generating fleet, however, the average price well head price of natural gas in the United States rose from $2.19/MMBtu in 1999 to $7.97/MMBtu in 2008, more than tripling the operating cost of the newly built plants, curtailing their hours of operation and destroying the economic value of the prior investment. The value of coal and nuclear generation surged, as on-peak prices were sustained by the high operating cost of the new CCGTs. By 2007, high gas prices, and the consequent profitability of nuclear and coal fired generation, were self-evidently the new normal, leading a KKR-led consortium to acquire the nuclear and coal fired generator TXU for $45 billion in the largest leveraged buyout in history. Within five years, as gas prices dropped first to $4.93 in 2009 and then to $3.54 in 2012, the debt of the LBO became untenable, bankruptcy was declared in 2014, and lenders lost billions.

4. State Renewable Portfolio Standards Have Suppressed Wholesale Power Prices and the Output of Conventional Power Plants

State renewable portfolio standards have had the effect of suppressing wholesale power prices and eroding the output of conventional generators. By supporting the construction of new renewable resources, state renewable generation mandates have added zero variable cost generating assets at the bottom of the supply curve, reducing both the hours of operation as well as the prices received by the higher cost conventional generating resources on the system.

Currently, 29 states and the District of Columbia – jurisdictions that together account for 65% of U.S. retail electricity sales – have adopted renewable generation mandates, imposing an obligation on electric utilities to supply a stipulated fraction of their retail electricity sales from renewable resources. Another eight states, accounting for 10% of retail electricity sales, have adopted renewable energy goals, which set target levels of renewable generation but impose no mandatory requirement on utilities.

These measures effectively mandate the build-out of renewable generation by requiring regulated utilities to procure renewable generation in stipulated amounts, and permitting these utilities to recover the cost of these purchases in their regulated rates. Once renewable capacity has been added in response to these mandates, the electric output of these facilities is a zero variable cost resource to the system – ensuring that renewable generating units will be dispatched before any conventional resource to meet prevailing demand. The new, zero marginal cost wind and solar plants will push higher cost fossil fuel plants up the supply curve, reducing their hours of operation.

By reducing the hours of operation of higher cost conventional resources, the dispatch of zero marginal cost renewable electricity also has the effect of suppressing the marginal cost of supply, and with it the wholesale price of electricity. To the extent that the highest cost fossil fueled units are no longer required to meet prevailing demand, due to an abundance of renewable energy, power prices will drop to reflect the variable cost of supply of the lower cost units that remain. The generation gross margin of existing conventional generators is thus squeezed further between the loss of output and the erosion of prices.

California’s renewable portfolio standard is one of the most ambitious in the country, requiring the state’s utilities to procure 33% of their electricity from renewable resources by 2020 and 50% by 2030. Reflecting this requirement, wind and solar now account for 28% of all the generation resources in the wholesale power market operated by the California Independent System Operator (CAISO). These zero variable cost renewable resources, while intermittent, can have a material impact of the price of power during the hours when they are available. Exhibit 4 plots two CAISO power supply curves: a blue one, representing all the power generation resources on the system, and an orange one that excludes wind and solar capacity. The two vertical lines on the chart show the average levels of demand corresponding to the four lowest and four highest price hours of each day. Between these two lines, the vertical distance between the two supply curves, representing the difference in the variable cost of supplying electricity with and without the system’s wind and solar resources, ranges from ~$12 to $16/MWh. The implication is that when CAISO’s wind and solar resources are available, the variable cost of generation can be 40% to 60% lower than it would otherwise be – and that power prices during those hours will be commensurately lower.

Exhibit 4: CAISO Power Supply Curve with and without Renewable Resources

______________________

Source: SNL, SSR analysis

5. Regulatory and Legislative Intervention Have Raised Generators’ Cost of Staying in Business, Often Forcing the Early Retirement of Existing Plants

Over the last decade, increasingly stringent EPA air and water emissions standards have imposed on competitive generators rising environmental compliance costs for which there is no recovery. These measures have frequently so increased the going-forward cost of existing power plants as to force their early retirement.

Power generation is a principal source of the air pollutants regulated by the U.S. Environmental Protection Agency under the Clean Air Act. These include sulfur dioxide (SO2), the precursor of acid rain; nitrous oxides (NOx), a precursor of smog; mercury and other toxic metals; and the greenhouse gas carbon dioxide, of which the power industry is the largest single source. Power generation is also the largest single user of water in the country. Generators’ intake of cooling water from ocean bays, rivers and lakes is thus a principal regulatory focus of the EPA under the Clean Water Act. Finally, coal ash, and the water used to sluice it from boilers and transport it to coal ash ponds, is a third area of environmental concern.

Not surprisingly, therefore, the EPA has been active in promulgating regulations to control the air emissions, water use and effluent streams of the power generating fleet. Over the last decade, the EPA has promulgated the Clean Air Interstate Rule (CAIR, which mandates a 40% reduction in SO2 emissions from power plants in 23 eastern states that account for over three quarters of U.S. coal fired generation, the Mercury and Air Toxics Standards (MATS, which require all coal fired power plants nationwide to control for mercury, particulate matter and acid gases, effectively requiring coal fired generators to install flue gas desulfurization technology, activated carbon injection, and bag houses or other particulate matter controls), the Clean Power Plan (CPP, which regulates CO2 emissions from power plants, but which is likely to be significantly cut back under the Trump administration) and cooling water intake and power plant effluent regulations.

No market mechanism exists for competitive generators to recover the incremental investments they must make to upgrade existing power plants to comply with the new rules. Once made, these investments represent a sunk cost. To maximize their output and revenues, competitive generators will continue to offer the output of their plants at their variable cost of generation. In most cases, variable operating costs change little as a result of these investments; among the gas fired generators that are frequently the marginal, price-setting units, they hardly change at all. Because environmental compliance costs place little, if any, upward pressure on power prices, competitive generators enjoy no incremental revenue to offset the capital cost of required upgrades.

Many generators therefore have found it cheaper to retire older, smaller coal fired units than to retrofit them to comply with these rules. Since the Mercury and Air Toxics Standards were promulgated by the EPA in December 2011, the U.S. power industry has either retired or announced the retirement of capacity equivalent to ~15% of the U.S. coal fired fleet. Competitive generators can expect no compensation for their unrecovered investment in these retired units.

6. Given These Challenges, What Is the Outlook for Competitive Generators?

Looking forward, competitive generators will struggle with the consequence of three facts:

  1. Total power output of the United States is stagnant.
  2. New renewable and conventional resources are suppressing the output of existing plants.
  3. The lower variable cost of these new resources will erode power prices.

U.S. power output in 2016 was essentially unchanged from its level 10 years before, even as real GDP has increased by 14%. While total U.S. generation has stagnated, the rapidly rising output of new renewable power plants, particularly wind and solar, has increased non-hydro renewable generation by almost 250 million MWh, or from 2% to 8% of nation’s total power supply. Squeezed between stagnant power demand and surging renewable generation, the combined output of the hydroelectric, nuclear, coal, gas and oil fired power plants in the U.S. has fallen by over 230 million MWh or 6% over the last decade, dropping from 98% to 92% of the nation’s total power output (see Exhibit 5).

Exhibit 5: Total U.S. Generation by Energy Source (Millions of MWh)

___________________________________

Source: Energy Information Administration

We expect this pattern to continue over the next five years: net power demand (gross power demand net of self-generation) will remain flat; the ongoing growth of renewable generation will continue to erode the share of fossil fuel generation in total U.S. power output; as result, coal and gas fired generation will decline, curtailing the capacity factors of existing fossil fuel plants.

Darkening this outlook are large planned additions of combined cycle gas turbine (CCGT) capacity. These new, highly efficient CCGTs, whose fuel consumption is on average 10% lower than that of the existing CCGT fleet, will enter the power supply curve below existing gas fired power plants, pushing these older, more expensive units up the supply curve and reducing their hours of operation. Higher cost coal fired power plants, specifically those burning Appalachian coal, will also be pushed up the curve and see their capacity factors fall. In summary, the existing fleet of fossil fuel plants will be squeezed between stagnant power demand and the growing capacity to supply this demand from lower variable cost units, reflecting the entry into operation of new wind and solar plants and new, more fuel efficient CCGTs.

Below, we assess the impact of these trends on the output and capacity factors of gas and coal fired power plants in the five principal regional transmission organizations (RTOs) covering the principal competitive power markets of the United States: PJM Interconnection (PJM), the Electricity Reliability Council of Texas (ERCOT), the California Independent System Operator (CAISO), the New York ISO (NYISO) and ISO New England (ISO-NE). We have compared the growth in power demand forecast by each RTO (see Exhibit 6) with the increase in generation expected from planned capacity additions[1]. Where new generation exceeds the expected growth in power demand, we have assumed that the output of the highest variable cost

units in the existing fleet falls commensurately. Conversely, if the growth in power demand exceeds the output of the new generating units, we have assumed that the existing plants with the highest variable cost will increase their output to supply the shortfall.

Exhibit 6: Historical and Forecast Electricity Demand Growth by RTO (1)

______________________________________________________________________________

1. Except for CAISO, forecast growth in electricity demand reflects the load forecasts of the RTOs themselves. For CAISO, we have forecast power demand growth based on the experience of the RTO over the last five and ten years.

Source: ERCOT, ISO New England, New York ISO, PJM Interconnection, SNL, SSR estimates and analysis

 

The results of our analysis are summarized in Exhibits 7 and 8. Over the next three years (2017-2019), increases in low cost generation from planned additions of wind, solar and new CCGT capacity outpace the expected growth in power demand. As a result, we expect the highest variable cost generators (the marginal, price-setting units) in each of the five RTOs to experience significant declines in power output and capacity factors through 2019 (see Exhibit 7). (In most of the RTOs, CCGTs are the marginal, price-setting units, but a mix of CCGTs and coal fired power plants are on the margin in PJM.) In CAISO and ISO New England, we expect the capacity factor of the existing CCGT fleet to continue to fall through 2021, reflecting the stagnant power demand combined with continued growth in new CCGT and renewable generation capacity. In the New York ISO and PJM we expect the capacity factor of existing CCGTs to stabilize beginning in 2020, as planned retirements of nuclear and coal fired power plants materially reduce the supply of lower cost generation. We expect a robust recovery in the output and capacity factor of the existing CCGT fleet only in ERCOT, due to continued rapid growth in power demand and a lack of planned capacity additions, although this may be overly optimistic if significant new wind capacity continues to be built in 2020 and beyond.

Exhibit 7: Forecast Capacity Factors of Exhibit 8: Forecast Drop in Power Output of Existing Power Plants (2017-2022) (1) Existing Power Plants (Off of 2016 Base) (1)

______________________________________________________________________________

1. Estimates are for the average expected power output and capacity factors of existing combined cycle gas turbine (CCGT) plants. For PJM, we also present our estimates for the average expected power output and capacity factors of PJM’s existing coal fired fleet

Source: SNL, SSR estimates and analysis

Exhibit 8 presents the forecast percentage decline in the power output of the marginal, price setting units in each of the five regions, as well as the year in which this drop in output reaches its maximum. We expect the capacity factor of the existing CCGT fleets to fall markedly, with the largest and most prolonged output declines likely to occur in CAISO, where we forecast the generation of the existing CCGT fleet will fall by 28% over 2016-2021; the New York ISO, where we expect a 23% drop in CCGT generation over 2016-2019; and ISO New England, where the generation of the existing CCGT fleet could fall by 21% over 2016-2022. We expect smaller, but still very material declines in PJM, with the output of PJM’s existing CCGT fleet

expected to fall by 14%, and that of its coal fleet to fall by 15%, by 2021. Only ERCOT, with its robust growth in power demand, is likely to follow a different trajectory. We expect the output of ERCOT’s existing CCGT fleet to fall by 10% through 2019; thereafter, power demand growth will drive output back almost to 2016 levels by 2021 and higher in 2022.

These expected declines in the power output of existing generating fleets are unlikely, in our view, to be offset by increases in power prices and spark spreads, at least through 2019. In competitive power markets, the prevailing price of electricity reflects the variable cost of generation at the last unit dispatched to meet demand. New, zero marginal cost wind and solar plants, and new, highly fuel efficient CCGTs, push higher cost fossil fuel plants up the supply curve; to the extent that the highest cost fossil fueled units are no longer required to meet prevailing demand, power prices fall to reflect the variable cost of supply of the lower cost units that remain. By way of example, new CCGTs have heat rates of ~6.7 MMBtu/MWh, some 10% below the average heat rate of ~7.4 MMBtu/MWh of the existing CCGT fleet. The implication is that, for each hour that a new CCGT supplants an existing one as the marginal, price-setting unit on the system, the marginal cost of supply and thus the price of power will be 10% lower than would otherwise be the case.

7. Do Market Economics Reward the Construction of New Power Plants?

Do market economics reward the construction of new power plants? To answer this question, we have compared the expected generation gross margin of a new combined cycle gas turbine plant in various markets around the country to the gross margin required by such a plant to recover its fixed costs, including the recovery of the capital invested to build the plant with a competitive return, plus any fixed operation and maintenance expense. The results of our analysis, set out in Exhibits 9 and 10 below, suggest that the construction of a CCGT is not economic in the various competitive power markets around the country.

To estimate the cost of a combined cycle gas turbine generator, we have relied upon data collected by (i) the Energy Information Administration (EIA) and made available in its November 2016 publication, Capital Cost Estimates for Utility Scale Electricity Generating Plants, as well as by (ii) the investment bank Lazard and published in Lazard’s Levelized Cost of Energy Analysis – Version 10.0, published in December 2016.

Based on these sources, our analysis assumes an overnight EPC (engineering, procurement and construction) cost of an H-Class combined cycle gas turbine generator of $1104/kW installed. The EIA publishes regional cost adjustment factors for different regions of the country which we have adopted as well. To capture the cost of capital used during construction, we have assumed that construction disbursements are spread evenly over a two year construction period, and have calculated the required return on this capital by applying our estimate of the pre-tax, weighted average cost of capital of a competitive generator.

Our estimate of the competitive generation industry’s pre-tax WACC is 12.2%, which we calculated assuming a 60/40 debt/equity capital structure, a 7% cost of debt and a 12% after-tax target ROE. To calculate the equivalent ROE on a pre-tax basis, we have used an effective tax rate for 40% for federal and state taxes combined. In Texas, which has no income tax, the effective tax rate is assumed to be the federal rate only, or 35%, causing the pre-tax WACC to fall to 11.6%.

To estimate the annual generation gross margin required for a new CCGT to recover its invested capital with a competitive return, we have amortized the estimated capital cost of the CCGT over an assumed useful life of 30 years at the industry’s pre-tax WACC. To this we added the EIA’s estimate of the annual fixed operation and maintenance expense of a new CCGT. The sum of the plant’s capital cost recovery charge and its fixed O&M expense determines the annual generation gross margin required by the plant to recover its invested capital and fixed costs and earn a competitive return (see Exhibit 9 below).

Exhibit 9: Estimated Cost of a New CCGT by Region, Corresponding Annual Fixed Cost Recovery Charge and Generation Gross Margin per MWh Required for Full Cost Recovery (1)

________________________________

1. To estimate the regional overnight cost of a new H-Class combined cycle gas turbine generator, we have relied upon the Energy Information Administration’s Capital Cost Estimates for Utility Scale Electricity Generating Plants, published November 2016. Capital cost recovery charges and cost of funds used during construction were calculated using a pre-tax WACC of 12.2%, assuming a 60/40 debt/equity capital structure, a 7% cost of debt and a 12% after-tax target ROE. We have used an effective tax rate for 40% for federal and state taxes combined. In Texas, which has no income tax, the effective tax rate falls to 35% and the pre-tax WACC to 11.6%. We have assumed a 30 year useful life for the new CCGTs. Capacity factors reflect the average capacity factors achieved by CCGTs that have entered operation in these regions over the last three years. Required gross margin includes $2/MWh for the recovery of variable non-fuel operation and maintenance expense.

Source: Bloomberg, SNL, Energy Information Administration, Lazard’s Levelized Cost of Energy Analysis – Version 10.0, SSR analysis

We then calculated the equivalent gross margin per MWh by dividing the annual generation gross margin requirement of the plant by its expected power output. For this purpose, we applied a capacity factor equivalent to the average capacity factor achieved by CCGTs that have entered operation in the same region over the last three years. Adding $2/MWh for the recovery of variable non-fuel O&M expense, we arrived at an estimate of the spark spread, or differential between average revenue per MWh and cost of fuel, required for full recovery of a new plant’s capital cost and fixed and variable O&M expense (see Exhibit 9).

Next, we estimated what a new CCGT can expect to earn in different competitive power markets around the country from the combination of generation gross margin and capacity revenues in regions with established capacity markets. We estimated the spark spreads, or gross margin per MWh, that a new CCGT can expect to earn based upon currently prevailing forward prices of electricity and natural gas and assuming an average operating heat rate of 6.5 MMBtu/MWh. For purposes of our analysis, we averaged these implied forward spark spreads over the next ten years and applied the average over each year of the CCGT’s estimated useful life. In markets with established capacity markets, we assumed that the new CCGT would earn annual capacity revenue over its useful life at the average of currently available forward capacity prices.

In Exhibit 10, we present the expected average gross margin of a new CCGT in various competitive power markets around the country, including both its expected generation gross margin and capacity revenues, and compare this to the gross margin required by the plant to recover its capital costs, with a competitive return, plus its fixed and variable non-fuel O&M expense (as derived in Exhibit 9).

As Exhibit 10 illustrates, the expected gross margins of new CCGTs fall well short of required gross margins in each of the markets analyzed. The shortfalls range from 13% of the required gross margin in the New York ISO’s Zone G to 59% in ERCOT, which has no capacity market. While the cost of individual projects can vary from the EIA’s estimates, our analysis suggests that (i) given the EIA’s estimated new build and operating costs for a new CCGT, (ii) given currently prevailing forward prices for electricity and natural gas and known forward prices for capacity, and (iii) applying our estimate of the pre-tax WACC of a competitive generator, the competitive generation industry is not currently in a position to recover the capital required for the construction of new CCGT capacity.

Exhibit 10: Expected Generation Gross Margin of a New CCGT by Region Compared to the Generation Gross Margin Required for Full Cost Recovery (1)

________________________________

1. For purposes of our analysis, we have averaged the next ten years’ forward spark spreads and the next three years’ capacity prices. Forward spark spreads reflect currently prevailing forward price curves for electricity and the estimated cost of the natural gas required to generate it at a CCGT with an assumed heat rate of 6.5 MMBtu/MWh.

Source: Bloomberg, SNL, Energy Information Administration, Lazard’s Levelized Cost of Energy Analysis – Version 10.0, SSR analysis

8. With Low Returns and Little Opportunity to Grow, Competitive Generation is Not Suitable for Public Equity Markets

Unsurprisingly for an industry faced with such a plethora of challenges, returns on capital among competitive generators have been weak. Over the last five years, the returns on invested capital (ROIC) earned by the competitive generators has been well below their cost of capital, and returns on equity have frequently been negative. Specifically, over the last five years, Calpine, Dynegy and NRG Energy have earned average returns on invested capital of 6.0%, (1.8%) an 3.8%, respectively; in contrast, we estimate the after-tax weighted average cost of capital at these three companies at 7.4%, 8.9% and 6.6% respectively. (See Exhibit 11.) Over the same period, the return on equity of Calpine, Dynegy and NRG Energy has averaged 8.7%, (23.8%) and (20.1%), respectively, with Calpine’s stronger performance largely reflecting a gain on an asset sale in 2014. (See Exhibit 12).

Exhibit 11: Return on Invested Capital for Calpine, Dynegy and NRG Energy, 2012-2016

_____________________

Source: SNL, SSR analysis

Exhibit 12: Return on Equity for Calpine, Dynegy and NRG Energy, 2012-2016

_____________________

Source: SNL, SSR analysis

Unable to produce attractive returns on capital the old fashioned way (by earning them), the managements of these three companies have resorted to another time honored approach, the manipulation of financial metrics. Investors have been encouraged to focus on free cash flow yield, the calculation of which divides (i) free cash flow (defined as net income plus depreciation, other non-cash charges, and any items deemed by management to be one-time expenses, less maintenance capex) by (ii) market capitalization. The thought behind this metric is that it represents the cash that could be returned to shareholders and, as can be seen in Exhibit 13, the numbers can look quite attractive. However, we believe this metric is prone to misinterpretation and is thus misleading.

First, cash from operations adds depreciation back to net income, and thus makes no provision, as GAAP requires, to expense over time the capital invested in property, plant and equipment. While free cash flow is calculated by subtracting maintenance capex from cash from operations, maintenance capex is not the cash equivalent of depreciation expense: it is the capex required to keep existing property, plant and equipment in service, not to replace it at the end of its useful life. From a shareholder’s perspective, therefore, the definition of free cash flow used to calculate free cash flow yield (net income plus depreciation expense less maintenance capex) must cover both the return on and the return of investor’s capital; it cannot be compared, therefore, with any commonly used measure of the cost of capital, whether cost of equity or WACC.

If shareholders lose sight of the fact that maintenance capex and depreciation are not equivalent, lenders will not. As the assets of these companies near the end of their useful lives, lenders will require that free cash flow be diverted from shareholders either to replace aging plant or to amortize outstanding debt. The older the plant and the more highly levered the assets, the sooner “free” cash flow must be diverted to buy new equipment or to repay debt. In neither scenario will free cash flow be available for distribution to shareholders.

Second, over the past five years, actual free cash flow (cash from operations less all expenses and capital expenditures, including non-recurring expenses and growth capex), has generally been much lower than adjusted free cash flow as calculated by company managements. (See Exhibit 13.) In part, this is due to the inconvenient tendency of one-time expense to recur, however much managements might wish otherwise. More fundamentally, managements have generally continued to invest in new opportunities, spending the money that, in the calculation of free cash flow yield, is assumed to be available to shareholders. Were these companies deploying this capital at returns in excess of their weighted average cost of capital, shareholders would welcome these investments; however, given that the historic returns on capital at these companies are well below their cost of capital, managements are likely destroying shareholder value by making these investments rather than returning the cash to shareholders.

Exhibit 13: Cumulative Adjusted Free Cash Flow, 2012-2016, as Calculated by

Management, vs. Actual Free Cash Flow (1)

__________________________________

1. While adjusted free cash flow excludes growth capex and expenses deemed by management to be non-recurring, actual free flow represents cash from operations less all expenses and capital expenditures, including non-recurring expenses and growth capex.

Source: SNL, SSR analysis

We believe that any financial metric used to value competitive generators — companies with highly levered balance sheets and aging generation fleets — must take into consideration the pressing need to amortize debt or replace existing assets. Given the failure of these firms historically to earn their cost of capital, and the unattractive economics, based on forward price curves, of investing in new generation assets, we favor a metric that allows for the recovery of the capital invested in the companies over the remaining useful lives of their generation fleets. We therefore propose that free cash flow yield be calculated by, first, setting aside an amount for the recovery of the capital invested and, second, considering any cash flow in excess of this capital recovery as constituting the true yield to investors on their investment.

In our calculations, we have assumed that these companies’ generating fleets have, on average, a remaining useful life of 20 years. We assume, therefore, that the debt and equity capital invested in each competitive generator, (i.e., its enterprise value), must be recovered over this period. In other words, to arrive at a true free cash flow yield, we subtract from management’s calculation of adjusted free cash flow a provision for the return to investors of the debt and equity invested in the company over the remaining useful life of its assets. Based on their managements’ 2017 guidance, we calculate the true free cash flow yield for Calpine, Dynegy and NRG Energy to be 4.3%, 2.5% and 4.5%, respectively. We calculate true free cash flow yield to be materially higher at Vistra and at NRG pro forma for its transformation plan (i.e., with $600 million of cost cuts and $11 billion lower debt). Based on the 2017 guidance provided by Vistra and NRG management, we calculate Vistra’s true free cash flow yield to be 8.0% and NRG’s, pro forma for its transformation plan, to be 7.6% — still below their cost of equity, which we estimate at 9.0% and 7.7%, respectively.[2] (See Exhibit 15.)

Exhibit 14: Management-adjusted Free Exhibit 15: True Free Cash Flow Yield, Based Cash Flow Yield, Based on 2017 Guidance on 2017 Guidance (1)

_____________________1. Our calculation of the true free cash flow yield assumes the recovery from free cash flow of the enterprise value of each company, amortized on a straight-line basis over 20 years, and the distribution to shareholders of the residual. The true free cash flow yield is calculated by dividing this residual by the market capitalization of each firm.

Source: SNL, SSR analysis

In conclusion, the competitive generators have delivered historical returns, whether measured by return on invested capital, return on equity or our own calculation of free cash yield to shareholders, that fall below their cost of capital. Looking forward, we see little opportunity for incremental capital investment at returns in excess of WACC. On the contrary, as discussed in section 7, wholesale power prices in markets around the country are too low for competitive generators to invest capital in new generation assets and be confident of the recovery of their investments, much less of earning an adequate return.

Given their poor historical track records and limited opportunities for growth, we expect competitive generators to continue to exit the public equity market. Investors in the public equity markets generally look to invest in companies whose assets and earnings are growing. U.S. competitive generators, by contrast, face little if any opportunities to grow their portfolios of generation assets with any confidence in reasonable returns and, given the unsatisfactory returns on their invested capital, often seek recovery of their capital through asset sales. We therefore see the competitive generators as owning portfolios of depreciating assets with finite lives that will generate declining cash flows over time. NRG Energy and Vistra have a longer lived source of income from their retail business, which does not require continued capital investment to maintain, but with a majority of the assets and earnings of the companies coming from generation, the fundamental issue remains.[3]

Private equity investors, by contrast, have built portfolios of infrastructure assets with capital provided by investors who expect their investments to be liquidated and their cash returned over time. Because growth is not always a priority, the cash generated by these assets can be allocated to debt service rather than investment, allowing higher levels of leverage and thus highly levered equity returns. High levels of leverage also permit the concentration of ownership in the hands of small partnership of private equity investors who desire to exercise management control to enhance operational performance and further improve financial returns. Furthermore, private equity is better positioned than public equity investors to assess the option value of generation assets that results from the cyclical volatility of prices for power and fuel.

Because the investment objectives of private equity parallel closely the investment opportunity offered by competitive generators, and we have seen several portfolios of competitive generation assets transition to private ownership in recent years. The leveraged buyout of TXU a decade ago was the first and by far the largest example; this was followed more recently by Riverstone’s acquisition of Talen Energy at the end of 2016, the acquisition by Blackstone and Arclight of a portfolio of competitive generation assets from AEP early this year, and the recently announced acquisition of Calpine by Energy Capital Partners. We do not believe that these will be the last such transactions to occur. We expect Dynegy to be acquired by Vistra or a private entity in the near term. Vistra, with its low leverage, and NRG Energy, with its transformation plan, could last for a few more years, but, absent a significant recovery in the wholesale power markets, we expect that both of these will eventually be taken private as well.

©2017, SSR LLC, 225 High Ridge Road, Stamford, CT 06905. All rights reserved. The information contained in this report has been obtained from sources believed to be reliable, and its accuracy and completeness is not guaranteed. No representation or warranty, express or implied, is made as to the fairness, accuracy, completeness or correctness of the information and opinions contained herein.  The views and other information provided are subject to change without notice.  This report is issued without regard to the specific investment objectives, financial situation or particular needs of any specific recipient and is not construed as a solicitation or an offer to buy or sell any securities or related financial instruments. Past performance is not necessarily a guide to future results.

  1. Our estimates of the growth of renewable generation are based upon the expected power output of the renewable power projects under construction or in development today. Our analysis suggests that the output of this backlog of renewable power projects will be required to meet the targets set by existing state renewable mandates. We have assumed that the capacity factors of new wind and solar projects is equivalent to that of wind and solar projects in the same region that have entered service over the last three years.Similarly, we have included in our forecast the generation from new combined cycle gas turbine power plants. In PJM and ISO New England, we have assumed that CCGT projects that are currently under construction or in advanced development and have cleared in capacity auctions, will enter operation as scheduled. In CAISO, we have assumed that CCGT projects that are currently under construction or in advanced development and have entered into a PPA with a utility, will enter service as scheduled. In ERCOT and NYISO, we have assumed that only the CCGT projects that are currently under construction will enter into service as scheduled. We have assumed that the capacity factors of new CCGT projects is equivalent to that of CCGTs in the same region that have entered service over the last five years.
  2. These calculations assume the recovery from free cash flow of the enterprise value of each company, amortized on a straight-line basis over 20 years, and the distribution to shareholders of the residual. The true free cash flow yield is calculated by dividing this residual by the market capitalization of each firm.Investors wishing to apply this analytical framework and prepared to construct a long term forecast of free cash flow could calculate the internal rate of return on an investment in the equity of each company by assuming outstanding debt is amortized over 20 years, and discounting the residual cash flow to shareholders. Our own rough estimate is that the IRR on an equity investment in Calpine, Dynegy and NRG Energy would be negative. We estimate that Vistra and NRG Energy, pro forma for its transformation plan (i.e., with $600 million of cost cuts and $11 billion lower debt) could generate positive returns in the range of 6% to 7%, still well below their cost of equity, which we estimate at 9.0% and 7.7%, respectively.
  3. Assuming a life for the power retailing business beyond 20 years is risky considering the business did not even exist 20 years ago except for large industrial power users. Additionally, increases in storage as costs decline will likely reduce the volatility of power pricing, thus reducing the risk of the retail business and the margins retailers earn.
Print Friendly, PDF & Email