GE, SIE, MHVYF: A Primer on How the Power Market Shapes the Market for Gas Turbines, Part 2 – Which Technologies Will Utilities Choose?

admin
Print Friendly
Share on LinkedIn0Tweet about this on Twitter0Share on Facebook0

______________________________________________________________________________

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

______________________________________________________________________________

January 9, 2019

GE, SIE, MHVYF:

A Primer on How the Power Market Shapes the Market for Gas Turbines, Part 2 –

Which Technologies Will Utilities Choose?

In our note of Nov. 15, A Primer on How the Power Market Shapes the Market for Gas Turbines[1], we analyzed the two principal drivers of global gas turbine orders, the growth in global power demand and the need to replace retiring generation capacity. Our analysis found that, despite decelerating power demand growth, retirements will drive a marked recovery in global gas turbine orders, which we see rising from 31 GW in 2020 to 49 GW in 2025 and 70 GW in 2030 — implying 5-year growth rates of 9.4% p.a. over 2020-25 and 7.6% p.a. over 2025-30. As this global wave of retirements will be concentrated in the developed economies of North America and Europe, we expect these regions to become the two largest markets for gas turbine orders, accounting for over 50% of orders through 2030.

Which gas turbine technologies and manufacturers will benefit most from this global recovery? This will reflect the factors influencing utilities’ choice of generation technologies, including the evolving roles of combined cycle and open cycle gas turbine generators in the supply of power and the economic incentives utilities face to invest in different power plant configurations and gas turbine models. In this research report, we use the granular data available on the composition and output of the U.S. gas turbine fleet to analyze how these factors are shaping utilities’ choice of gas turbine technologies. We conclude that G/H/J class turbines in combined cycle configurations should remain the dominant technology during the next decade, but that open cycle configurations, particularly with aeroderivative turbines, could capture increasing market share in the 2030s in response to the rising penetration of intermittent renewables resources.

Portfolio Manager’s Summary

Gas turbine orders:

  • Our forecast of global generation capacity additions models gross annual additions of firm generating capacity as the sum of (i) the capacity required each year to replace retiring power plants plus (ii) the net new capacity required to meet the annual growth in global power demand.
  • The historical cyclicality of capacity additions will predictably give rise to echo booms as the capacity installed during these surges reaches the end of its useful life.
    • The historical timing of capacity additions around the world suggests that global retirements of fossil fuel generation capacity will average ~58 GW annually over 2021-2025, 88 GW annually over 2026-2030, 123 GW annually over 2031-2035, and 142 GW p.a. over 2036-2040 (Exhibit 4).
    • Globally, as a share of gross annual additions of fossil generation capacity, the need to offset retirements will far outweigh the need to add net new capacity to meet the growth in load.
    • Because this global wave of retirements will be concentrated in the developed economies of North America and Europe, we expect these regions to become the two largest markets for fossil fuel capacity additions, including gas turbines, over this period (Exhibit 5).
  • With respect to net new capacity additions, our forecast assumes that the ratio of the growth in power demand to the growth in GDP falls to half the level observed over the last 10 years in each region of the globe, or ~1.1% annually over the next decade, an assumption consistent with the observed deceleration in power demand growth over the last five years in every region of the world as ratios of power demand to GDP growth have fallen.
  • Critically, we expect gross annual additions of firm generating capacity to be dominated by gas.
    • Globally, expected additions of nuclear generating capacity are relatively small.
    • While we expect wind and solar energy to expand rapidly, the intermittent nature of these resources severely limits their value as sources of firm capacity.
    • In regions that historically have relied heavily on coal, we have assumed that coal-fired capacity additions will continue to comprise the same share of gross capacity additions that they have over the last decade.
    • However, coal’s share of gross capacity additions is declining in China and India, due to changes in government policy, and is very low in North America and Europe, the two markets facing the largest capacity retirements through 2040.
  • Based on the assumptions above, we see total gas turbine orders rising from 31 GW in 2020 to 49 GW in 2025 and 70 GW in 2030, implying 5-year growth rates of 9.4% p.a. over 2020-25 and 7.6% p.a. over 2025-30. (See A Primer on How the Power Market Shapes the Market for Gas Turbines[2]).

Service revenues:

  • The service revenues of gas turbine manufacturers are a function primarily of the hours of operation of the gas turbines they have supplied.
  • We expect global gas fired generation to expand over the next ten years, although not as fast as gas turbine orders, and therefore foresee continued growth in gas turbine service revenues.
    • We expect gas fired generation to grow at an average annual rate of ~2.1% over the next ten years. This will represent a marked slowing from the 4.1% p.a. compound annual growth in global gas fired generation realized over 2006-2016 (Exhibits 7 & 11).
  • Global gas fired generation will continue to grow through the early 2030s before plateauing and then declining in the second half of the 2030s, reflecting continued growth in wind and solar generation from a very substantial base (Exhibit 9).
  • The rapid growth of renewable generation will suppress the output of higher cost fossil fuel power plants.
    • Retiring coal and nuclear generating capacity will be replaced primarily with gas fired capacity, but the lost output of these retiring plants will be offset, to a significant degree, by an expanding supply of wind and solar energy, rather than gas fired generation.
    • While we expect global power demand growth to average 1.1% annually over the next ten years, we expect renewable generation to expand at 5.0% p.a., increasing renewables’ share of total global generation from ~25% in 2018 to ~37% by 2028 (Exhibits 7-9).
    • By contrast, we expect fossil generation to fall from ~64% of the global power supply today to ~51% by 2028 (Exhibit 9), and the average capacity factor of the global fossil fleet to fall from ~44% today to ~40% by 2028. Coal fired generation will fall first, later gas (Exhibit 11).

Factors influencing the composition of gas turbine orders:

  • We have used the granular data available from the U.S. government on the composition and output of the U.S. gas turbine fleet to analyze shifts in the mix of generation over time, how these have affected the operating profiles of existing gas turbine power plants, and the implications for utilities’ choice of gas turbine technologies.
  • In planning to meet their capacity needs over the next decade, we expect utilities to favor combined cycle gas turbine power plants over open cycle gas turbines. This will benefit sales of the most energy efficient industrial scale gas turbines, such as the G/H/J class turbines, as well as GE and Mitsubishi as the currently dominant manufacturers of these models.
  • Over the last 15 years (2003-2017), 75% of gas fired capacity additions in the United States have comprised combined cycle gas turbine (CCGT) power plants, and only 25% open cycle gas turbines (GT) (Exhibit 13).
  • Underlying trends in the capacity factors of U.S. combined cycle and open cycle gas turbine generators suggest that this historical breakdown in capacity additions is likely to continue.
    • Over the last 15 years, the average capacity factor of U.S. combined cycle gas turbine generators has increased by more than half, from 30% in 2003 to 47% in 2017, while that of open cycle gas turbine generators has remained roughly constant at ~7% (Exhibit 14).
    • The ability to defray the construction and other fixed costs of a combined cycle gas turbine power plant over 50% more operating hours has significantly enhanced the new build economics of CCGTs. GTs, by contrast, have enjoyed no such improvement.
  • Another factor eroding the new build economics of GTs vs. CCGTs has been the rise in reserve margins over the last 15 years in most regions of the United States (Exhibit 19). As reserve margins have improved, GTs have operated fewer hours and earned lower gross margins during these hours, rendering investment in these plants less attractive.
  • CCGTs, by contrast, have become more compelling investments as the drop in the price of natural gas has allowed these plants to undercut coal fired generators, increasing their hours of operation and improving generation gross margins (Exhibits 15 & 16).
  • Over the last 15 years the most energy efficient turbine classes – F, G, H & J – have registered high and rising levels of operating hours per unit, while the least energy efficient turbine classes – B class and similar turbines – have registered persistently low average hours of operation.
    • We see this as a reflection of utilities’ technology choice: in building combined cycle gas turbine plants, with their high and rising capacity factors (Exhibit 17), utilities have favored the most energy efficient gas turbines.
    • We believe this trend also underpins the rapid increase in the market shares of the more energy efficient G, H & J class turbines, which have been displacing the F class turbines (Exhibit 22).
  • Finally, announced additions of gas turbine capacity in the U.S. over 2019-21 are consistent with the continued dominance of F through J class gas turbines, and a rising market share for the most energy efficient G/H/J class (Exhibit 29).
  • Standing to benefit are the leading manufacturers of these models, GE for F class turbines and GE and MHI for G/H/J class (Exhibit 30).

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

Source: SSR analysis

Details

Forecast of gas turbine orders:

As global electricity consumption continues to rise, it will drive a commensurate increase in the need for firm power generation capacity that can be dispatched as needed to meet peak demand (“firm capacity”). In our note of Nov. 15, A Primer on How the Power Market Shapes the Market for Gas Turbines[3], we modeled gross annual additions of firm generating capacity as the sum of (i) the net new capacity required to meet the annual growth in global power demand[4] and (ii) the capacity required each year to replace retiring power plants. We then estimated the proportion that gas turbines would contribute to these gross annual additions of firm generating capacity in each region of the world to arrive at forecast of global gas turbine orders.

In our base case forecast, we see total gas turbine orders rising from 31 GW in 2019 to 51 GW in 2025 and 73 GW in 2030, implying 5-year growth rates of 9.1% p.a. over 2020-25 and 7.6% p.a. over 2025-30 (Exhibit 2). Reflecting the scale of expected capacity retirements in the mature economies of North America and Europe, and the preference in these regions for gas over coal fired capacity additions, we see these regions dominating global gas turbine orders in the coming decade. We expect the United States to be the single largest market for gas turbines over the coming decade, with 30% of forecast orders, followed by Europe with 23% (Exhibit 3). (For a sensitivity analysis of this base forecast, please see our research report of June 19, 2018, GE, SIE,MHVYF: The Impending Recovery in Gas Turbine Orders).[5]

Exhibit 2: Base Case Forecast of Global Gas Turbine Orders (GW)

______________________________________________________________________________________________

Source: Energy Information Administration, International Energy Agency, Organization for Economic Co-operation and Development, SSR analysis and estimates

Exhibit 3: Forecast Gas Turbine Orders by Region (Cumulative GW Over 2019-2030)

______________________________________________________________________________________________

Source: Energy Information Administration, International Energy Agency, Organization for Economic Co-operation and Development, SSR analysis and estimates

In Exhibit 4, we present our estimate of the required annual additions of firm generating capacity, from 2020 through 2040, broken down into (i) the replacement capacity required to offset power plant retirements and (ii) the net new generation capacity needed to meet the growth in power demand. Importantly, we expect that from 2025 on the need to offset capacity retirements will significantly outweigh the need for net new capacity additions.

In Exhibit 5, we present our estimate of how the required gross annual additions of firm generating capacity will be met through a combination of net new nuclear capacity, growth in firm renewable capacity and gross additions of fossil capacity. To forecast future gross additions of fossil

generation capacity, and gas turbine generation capacity in particular, we have subtracted from our estimate of gross annual additions of firm generating capacity, first, expected additions of nuclear capacity and second, the firm capacity contributed by rapidly expanding wind and solar generation,[6] both as forecasted by the IEA (International Energy Agency). This analysis suggests that from 2025 on a majority of gross additions of firm generation capacity will be fossil fired(Exhibit 5).

Exhibit 4: Required Net Annual Additions Exhibit 5: Technology Breakdown of Forecast

Of Firm Generation Capacity (GW) Additions of Firm Generation Capacity (GW)

 

Source: Energy Information Administration, International Energy Agency, Organization for Economic Co-operation and Development, SSR analysis and estimates

We expect the bulk of the fossil generation capacity added globally through 2040 to be gas fired (Exhibit 6). Our analysis assumes that, in regions which historically have relied heavily on coal, coal-fired capacity additions will continue to comprise the same share of gross capacity additions that they have over the last decade. Critically, however, coal’s share of gross capacity additions is negligible in North America and Japan, two markets facing large capacity retirements through 2040, as well as in the Middle East, a region with a very substantial need for net new capacity additions. In the case of China, the market with the largest need for net new capacity additions, the government has stated that it intends to limit total coal generation capacity to 1,100 GW, and we have respected this cap in our forecasts.

Exhibit 6: Gross Additions of Gas and Coal Fired Generation Capacity by Region (GW)

Gas Coal

 

Source: Energy Information Administration, International Energy Agency, Organization for Economic Co-operation and Development, SSR analysis and estimates

For a fuller discussion of the assumptions and methodology underpinning these conclusion, please see the Appendix to this note.

Service Revenues Trends

While we expect global gas turbine orders to rise rapidly in the coming decade, we expect the power output of the global gas turbine fleet to expand much more slowly, and indeed to lag the pace of the last ten years. Consequently, while the growth in manufacturers’ revenues from equipment sales is expected to accelerate markedly, growth in manufacturers’ high margin service revenues, which are largely a function of the operating hours of their gas turbine fleets, will slow.

Our base case forecast of global gas turbine orders sees these increasing at a 7.5% average annual rate over the next ten years (2018-2028); by contrast, we expect the power output of the global gas turbine fleet to expand at only 2.1% p.a. over this period. This is approximately half the pace of the last ten years: over 2006-2016, the latest ten year period for which we have global data, global gas turbine generation grew at a 4.1% average annual rate.

Two factors drive this expected slowing in the growth of gas fired generation. First, we have assumed that global growth in power demand will slow in the years ahead, to an estimated 1.1% p.a. over the next ten years (2018-2028). More specifically, we assume that the ratio of power demand growth to GDP growth falls to half the level observed over the last 10 years in each region of the globe. This assumption reflects the secular tend, particularly observable over the last ten years, of declining electricity use per dollar of GDP. (See the Appendix to this note, and particularly Exhibits 31, 32 and 33.)

Second, we expect a substantial increase in the MWh of renewable energy produced by the globe’s rapidly expanding wind and solar generation capacity. This rising supply of zero variable cost renewable energy will limit the need for generation from higher cost fossil fuel power plants.

Specifically, we have assumed that the global demand for power will be supplied on the basis of economic dispatch, with the lowest variable cost generation resources, renewable energy and nuclear, being dispatched first to meet demand. Based on the forecast of the International Energy Agency, or IEA, we assumed that renewable generation (including hydro) expands at a 5.0% compound annual rate over 2018-2028, or slightly below the 5.4% CAGR evident over the last ten years for which we have global data, 2006-2016 (Exhibit 7). The IEA also foresees nuclear generation growing at 2.4% p.a. over the next ten years, driven primarily by the expansion of nuclear generating capacity in China.

This combination of slowing demand growth and substantial increases in renewable energy imply a declining role for fossil fuel generation: our 2018-2028 forecast of 5.0% annual growth in renewable generation and 2.4% annual growth in nuclear generation significantly exceed the estimated 1.1% annual growth in global electricity consumption. As fossil fuel generation is increasingly crowded out, the output of the global fossil fuel fleet is expected to fall at 1.1% average annual rate over the next ten years (Exhibits 8 & 9).

Exhibit 7: Historical and Forecast Growth Exhibit 8: Forecast Growth Rate of Electricity

Rate in Renewable, Nuclear & Fossil Consumption and Renewable, Nuclear, Fossil,

Generation 2018-2028 vs. 2006-20161 Coal and Gas Fired Generation, 2018-2028

 

1. 2006-2016 is the latest ten year period for which global generation data are available.

Source: International Energy Agency, Energy Information Administration, Organization for Economic Cooperation and Development (OECD), SSR analysis and estimates

Exhibit 9: Historical and Forecast Output of Global Generating Fleet (Millions of MWh)

2000-2016 Forecast, 2017-2030

 

Source: International Energy Agency, U.S. Energy Information Administration, Organization for Economic Cooperation and Development (OECD), SSR analysis and estimates

We expect this global decline in fossil fuel generation to be absorbed primarily by the coal fired fleet. Our analysis of the age and expected useful life of the coal fired power plants in operation today suggest that these will also account for the bulk of fossil fuel capacity retirements; by contrast, coal fired capacity will represent a relatively small portion of fossil fuel capacity additions, which we expect to be dominated by gas (Exhibit 6). Specifically, we expect negligible additions of coal fired generation capacity in North America and Japan, two markets facing large capacity retirements through 2040, as well as in the Middle East, a region with a very substantial need for net new capacity additions. In the case of China, the market with the largest need for net new capacity additions, the government has stated that it intends to limit total coal generation capacity to 1,100 GW, and we have respected this cap in our forecasts.

A further headwind to coal fired generation is our assumption that the capacity factor[7] of the global coal fired fleet continues to decline, although at a slower rate, while that of the global gas fired fleet remains stable. Over the last 12 years, the capacity factor of the global coal fired fleet has fallen fairly steadily, from a high of ~68% in 2005 to 54% in 2016 (Exhibit 10). Various factors have contributed to this decline, including (i) the dramatic fall in the price of natural gas in North America, which triggered a marked shift of generation from coal to gas fired power plants as a result; (ii) the rapid global growth of zero variable cost wind and solar generation, which has eroded the hours of operation of higher cost coal fired generating units; and (iii) environmental limits on coal fired generation in many countries in Europe and Asia. In our forecast, we assume the trend toward lower coal fired capacity factors continues, at a relatively steady rate of 1.0 percentage point per annum through 2040.

Exhibit 10: Historical and Forecast Capacity Factors of Coal and Gas Fired Fleets

2000-2016 2017-2040

 

Source: International Energy Association, Energy Information Administration, Organization for Economic Cooperation and Development (OECD), SSR analysis and estimates

Reflecting our assumptions of slowing global power demand growth, rapidly rising renewable generation, and an ongoing trend of gas fired generation being substituted for coal for both economic and environmental reasons, our forecast is for a coal fired generation to fall at an estimated 4.2% average annual rate over 2018-2028 (Exhibits 8 & 11).

Gas fired generation, by contrast, we expect to continue to grow, albeit a slower rate. Facing the same headwinds of slow power demand growth and rapidly rising renewable generation, we see the growth in global gas fired generation decelerating from a pace of 4.1% p.a. over 2006-2016 to 2.1% p.a. over the next ten years. As noted above, gas turbine manufacturers’ service revenue growth, which is largely a function of the operating hours of their gas turbine fleets, is therefore expected to slow, even as their revenue from equipment sales is expected to accelerate markedly.

Exhibit 11: Historical and Forecast Growth of Coal and Gas Fired Generation (TWh)

2000-2016 2017-2040

Source: International Energy Association, Energy Information Administration, Organization for Economic Cooperation and Development (OECD), SSR analysis and estimates

Looking beyond 2030, our assumptions of ongoing stagnation in power demand, combined with the continued growth of renewable generation, begin to constrain the forecast output of the gas turbine fleet, whose generation is expected to peak in the mid-2030s and then begin to decline (see the blue line in Exhibit 12.

Exhibit 12: Breakdown of Global Power Generation by Energy Source, 2000-2040E

Millions of MWh %

Source: International Energy Association, Energy Information Administration, Organization for Economic Cooperation and Development (OECD), SSR analysis and estimates

The expected scale of renewable generation in the 2030s may begin to affect technology choice by fossil fuel generators. Even given our assumption that the growth of renewable generation decelerates to 3.5% p.a. over the decade of the 2030s, we see renewable generation rising from ~25% of global power output today to ~40% by 2030 and ~50% by 2040. With virtually all of this increase expected to come from intermittent wind and solar resources, gas fired generators will face not only a decline in their average capacity factor but also an increase in the intermittency of dispatch, as gas turbines are forced to offset random variations in wind and solar generation. These expected changes in the operating conditions of the global gas turbine fleet may erode the current popularity of fuel efficient, industrial scale gas turbines in combined cycle configuration and increase demand for highly flexible open cycle gas turbines, including aeroderivative designs capable of quick starts and rapid ramping of their power output.

Factors influencing the composition of gas turbine orders:

In this section, we will capitalize on the compendium of information maintained by the Energy Information Administration on the U.S. power generating fleet to analyze the key factors influencing the volume and composition of gas turbine orders. As noted above, we expect North America to be the largest market for gas turbines through 2030, accounting for some 30% of forecast gas turbine orders over this period, followed by Europe with 23% (see Exhibit 3). Given the scale of the North American market, and the structural similarities between the competitive power markets of North America and Europe, we believe an analysis of the U.S. data can provide useful insights into the factors influencing utilities’ choice of generation technology.

Since the turn of the century, gas turbine orders in the United States have skewed heavily in favor of combined cycle rather than open cycle gas turbines. U.S. additions of open cycle gas turbine capacity last exceeded combined cycle gas turbine capacity additions in 2001 (Exhibit 13, left hand chart). Over the last 15 years, combined cycle gas turbine capacity additions have comprised 75% of total U.S. gas turbine capacity additions (Exhibit 13, right hand chart)

Exhibit 13: Open & Combined Cycle Gas Turbine Capacity Additions in the U.S.

Breakdown of Annual Capacity Additions, 1987-2017 1987-2002 Average vs. 2003-2017 Average

 

_________________________________________________

Source: U.S. Energy Information Administration, SSR analysis

Underlying trends in the capacity factors[8] of U.S. combined cycle and open cycle gas turbine generators suggest that this historical breakdown in capacity additions is likely to continue. Over the last 15 years, the average capacity factor of U.S. combined cycle gas turbine generators has increased by more than half, from 30% in 2003 to 47% in 2017, while that of open cycle gas turbine generators has remained roughly constant at ~7% (Exhibit 14). The ability to defray the construction and other fixed costs of a combined cycle gas turbine power plant over 50% more operating hours has significantly enhanced the new build economics of CCGTs. Open cycle gas turbines, by contrast, have enjoyed no such improvement.

Exhibit 14: Capacity Factors of the U.S. Open and Combined Cycle Gas Turbine Fleets

_________________________________________________

Source: U.S. Energy Information Administration, SSR analysis

The falling price of natural gas over the last ten years has also enhanced the new build economics of CCGTs. As illustrated in Exhibit 15, the increase in the capacity factor of the U.S. combined cycle gas turbine fleets reflects the marked drop in CCGTs’ variable cost of operation as the price of natural gas has fallen. Lower variable costs have allowed CCGTs to undercut coal fired power plants (Exhibit 16), increasing CCGTs’ hours of operation, widening their generation gross margin and materially improving their profitability. Coal fired capacity factors have declined markedly as a result, so that over the last three years the capacity factors of the coal and combined cycle gas turbine fleets have been roughly equal (Exhibit 17).

Exhibit 15: CCGT Capacity Factors Have Increased Markedly as the Gas Price Has Fallen

_________________________________________________

Source: U.S. Energy Information Administration, SSR analysis

Exhibit 16: Estimated Variable Operating Exhibit 17: Capacity Factors of U.S. Gas

Cost of Gas Fired CCGTs and Coal Fired Turbine & Steam Turbine Generating Fleets

Steam Turbines, Ohio River Valley

 

_________________________________________________

Source: U.S. Energy Information Administration, SSR analysis

With lower variable costs of operation, much lower construction costs, and capacity factors comparable to coal, combined cycle gas turbines now offer a levelized cost of energy[9] (LCOE) that is roughly one third to one half below that of new coal fired plants (Exhibit 18). The secular increase in the capacity factors of U.S. combined cycle gas turbine generators since 2003 has materially improved their new build economics not only with respect to open cycle gas turbines but coal fired power plants as well.

Exhibit 18: Levelized Cost of Energy of Conventional Generation Technologies1

_____________________________________________________________________________________________

1. Levelized cost of energy represents the constant price a generator would need to receive for its power output over the life of the project to recover its operating costs, ensure the return of its invested capital, and earn a return equivalent to its after-tax weighted average cost of capital.

Source: Lazard’s Levelized Cost of Energy Analysis – Version 12.0, November, 2018

Another feature of the U.S. power market that has favorably influenced the new build economics of combined versus open cycle gas turbines has been the rise in reserve margins[10] over the last 15 years in most regions of the United States . As noted above, U.S. additions of open cycle gas turbine capacity last exceeded combined cycle gas turbine capacity additions in 2001. The prior year, when developers were breaking ground on these open cycle gas turbine plants, the average reserve margin in the United States was only 10% (Exhibit 19), one third below the ~15% reserve margin deemed prudent by system planners. By contrast, in 2017 the average U.S. reserve margin stood at 28%, or almost twice the target level. For peaking plants, such as open cycle gas turbines, low reserve margins tend to be reflected in higher capacity factors (Exhibit 17) and materially higher generation margins during peak demand hours, increasing the expected revenues and gross margins of these plants and enhancing their new build economics. By contrast, peaking plants in power markets with high reserve margins operate fewer hours and earn lower gross margins during hours of peak demand. Thus a developer assessing the new build economics of open cycle gas turbines today would likely find them to be far less favorable than at the turn of the century.

Exhibit 19: U.S. Reserve Margin (Excess of Exhibit 20: U.S. Peak Demand & Firm Capacity,

Firm Capacity Over Peak Demand in %) 2000-2017 (2000 = 100)

 

_________________________________________________

Source: U.S. Energy Information Administration, SSR analysis

The tight reserve margins prevailing at the turn of the century also influenced utilities’ views as to the relative urgency of adding generation capacity, tending to favor open cycle gas turbines whose combined development and construction periods (2-3 years) were shorter than those of CCGTs (3.5-5.5 years). Influencing utility managements were a series of incidents where low regional reserve margins had forced utilities to shed load; these culminated in the California energy crisis of 2000-2001, when San Francisco suffered repeated blackouts. The rapid pace of demand growth at the time raised concerns that that reserve margins would only erode further. In 2000, U.S. power demand was 25% higher than it had been a decade earlier, and expectations were that it would continue to growth at the 2.3% average annual rate of the last ten years. Many utilities in 2000 thus felt strong pressure to add capacity, and were prepared to capitalize on the shorter development and construction schedules of open cycle gas turbines to do so. In 2017, by contrast, the U.S consumed 3.0% less electricity than it had a decade earlier; worse, peak power demand had dropped by ~9.0% over the preceding decade (Exhibit 20). Today, the rapid deployment of capacity is no longer a priority in most regions, and technology choice is shaped primarily by considerations of cost and long run returns.

In summary, developers in the U.S. today are attracted to combined cycle gas turbines because of their favorable new build economics versus competing gas and coal fired generation technologies. The factors underpinning the economic advantages of CCGTs are unlikely to chage dramatically in the near to medium term. CCGTs have offered lower construction costs than coal fired plants for decades, and this advantage has widened as increasingly stringent air emissions regulations have forced coal fired power plants to add expensive emissions controls. More recently, CCGTs have become not only cheaper to build but also cheaper to operate than coal fired units, as the fracking revolution has driven down the cost of gas supplies in North America – causing CCGT capacity factors to rise and further reducing their levelized cost of energy. Increasing exports of LNG from North America and recent large finds natural gas in the eastern Mediterranean and elsewhere should support improved economics of gas fired generation globally. The conditions that might favor investment in open cycle gas turbines, such as low reserve margins, high capacity factors for peaking plants, and pressure to add capacity quickly, no longer apply.

Rather, with reserve margins robust, the trend has been for development dollars to flow into new, highly energy efficient CCGTs, capable of displacing higher cost coal and gas fired competitors on the supply curve. Looking forward to the wave of coal and nuclear retirements we anticipate over the next two decades, we see the need to replace the capacity of these base load power plants as yet another factor favoring low LCOE CCGTs as the technology of choice for future capacity additions.

The tendency for combined cycle gas turbine plants to dominate additions of gas turbine capacity over the last 15 years has had important implications for utilities’ choice of generation technology. As gas turbine technology has developed over time, gas turbines have been grouped into a series of classes based on turbine inlet combustion temperatures, a proxy for the energy efficiency of the turbine’s design. As can be seen in Exhibit 21, over the last 15 years the most energy efficient turbine classes – F and G/H/J – have registered high and rising levels of operating hours per unit, while the least energy efficient turbine classes – B class and similar turbines – have registered persistently low average hours of operation. We see this as a reflection of utilities’ technology choice: in building combined cycle gas turbine plants, with their high and rising capacity factors (Exhibit 17), utilities have favored the most energy efficient gas turbines. We believe this trend underpins the rapid increase in the market shares of the more energy efficient G/H/J class turbines, which have been displacing the previously dominant F class. (Exhibit 22).

Exhibit 21: Hours of Operation of Different Classes of Gas Turbines in the United States

Average Yearly Hours of Operation per Unit Total Hours of Operation of All Units in Class

 

_________________________________________________

Source: U.S. Energy Information Administration, SSR analysis

Exhibit 22: Breakdown of U.S. Gas Turbine Installations in MW by Class of Turbine

_________________________________________________

Source: U.S. Energy Information Administration, SSR analysis

The growing market share of F and G/H/J class turbines, combined with the tendency for the average annual hours of operation of these units to rise over time, has driven a significant increase in the total operating hours of the F and G/H/J class turbine fleets, materially benefiting the service revenues of the historically dominant manufacturers of these turbine classes, GE and Siemens. As illustrated in Exhibit 23, the total operating hours of GE’s and Siemens’ U.S. gas turbine fleets have more than quadrupled over the last 15 years (2001-2016, the last years for which data are available).

Exhibit 23: Total Hours of Operation of Each Manufacturer’s Gas Turbine Fleet

_________________________________________________

Source: U.S. Energy Information Administration, SSR analysis

The evolving U.S. market shares of GE, Siemens, Mitsubishi and other gas turbine suppliers are presented in Exhibits 24 and 25. To a significant degree, changes in market share amongst these competing manufacturers can be explained by shifting leads in gas turbine technology, as new models are designed and introduced to the market. Thus GE’s market dominance over 2000-2005 – years that accounted for two thirds of all gas turbine capacity additions in the United States over the last 30 years — reflected the timely introduction and widespread acceptance of GE’s F Class turbine (Exhibit 22). GE’s U.S. market share peaked around 2007, and subsequently eroded significantly, due to (i) Siemens introduction of a competing F Class machine (Exhibit 26), (ii) the loss of a substantial portion of the market for large capacity industrial turbines to more the energy efficient G/H/J Class turbines produced by Siemens and Mitsubishi (Exhibits 22 and 27), and (iii) Mitsubishi’s share gain in the market for aeroderivative gas turbines,[11] a market GE had previously dominated (Exhibit 28). After bottoming in 2013, however, GE has enjoyed a continuous recovery in its market share through 2017 as it has (i) regained share from Siemens in the market for F class turbines (Exhibit 26), (ii) introduced a new H class turbine (currently the global leader in energy efficiency), allowing it to regain a share of the market for the most energy efficient industrial turbines (Exhibit 27), and (iii) clawed back share from Mitsubishi in the market for aeroderivatives (Exhibit 28).

Exhibit 24: Gas Turbine Capacity Additions by Manufacturer

Annual in MW Market Share in % (3-Year Rolling Average)

 

_________________________________________________

Source: U.S. Energy Information Administration, SSR analysis

Exhibit 25: Breakdown of U.S. Gas Turbine Sales by Manufacturer, 2000-2017

GW %

 

_________________________________________________

Source: U.S. Energy Information Administration, SSR analysis

Exhibit 26: F Class Gas Turbine Sales by Manufacturer, 2000-2017

GW %

 

_________________________________________________

Source: U.S. Energy Information Administration, SSR analysis

Exhibit 27: G, H & J Class Gas Turbine Sales by Manufacturer, 2000-2017

GW %

 

_________________________________________________

Source: U.S. Energy Information Administration, SSR analysis

Exhibit 28: Aeroderivative Gas Turbine Sales by Manufacturer, 2000-2017

GW %

 _________________________________________________

Source: U.S. Energy Information Administration, SSR analysis

Finally, we have analyzed the data collected by the Energy Information Administration on planned additions of gas turbine capacity over the next three years (2019-2021) to estimate the future U.S. market shares of the principal as turbine manufacturers. Because combined cycle gas turbine plants have longer combined development and construction periods than open cycle gas turbine plants, the EIA data is particularly helpful in estimating planned additions of CCGT capacity and the F through J class turbines that CCGT plants frequently deploy. The EIA data may understate, by contrast, future additions of aeroderivative gas turbines, which can be developed and built more quickly.

While it is difficult, therefore, to make positive assertions based on the EIA data on planned capacity additions, the EIA data appear at least to be consistent with the continued dominance in the U.S. of F through J class gas turbines (right hand chart of Exhibit 29), with the most energy efficient G/H/J class turbines capturing a rising share over time (compare Exhibit 29 with Exhibit 22). The EIA data also suggest that GE will continue to capture a larger share of the U.S. gas turbine market than it did in 2011-2015, at the expense of Siemens (compare the left hand chart of Exhibit 29 with Exhibit 25).

Exhibit 29: Breakdown of Announced Gas Turbine Capacity Additions in the United States Over 2019-2021 by Manufacturer and Turbine Class

Manufacturer Turbine Class

_________________________________________________

Source: U.S. Energy Information Administration, SSR analysis

In Exhibit 30, we analyze which manufacturers are gaining leadership in the markets for F and G/H/J Class turbines. To do so, we have combined historical data on U.S. gas turbine capacity additions with data on announced future capacity additions to create a five year frame of reference (2017-2021), long enough to smooth out year-to-year fluctuations in manufacturers’ market share. Portending well for GE’s future sales, the company appears to maintain its traditional dominance in the market for F Class turbines and, critically, is capturing a much larger share of the market for G/H/J class turbines than it has historically (compare Exhibit 30 with Exhibit 27).

Exhibit 30: Breakdown by Manufacturer of Historical and Announced Gas Turbine Capacity Additions in the United States Over 2017-2021 Turbine Class

F Class G/H/J Class

 

_________________________________________________

Source: U.S. Energy Information Administration, SSR analysis

Over the longer term – 2030 and beyond — we believe that the rapid growth of renewable generation in the United States may weaken utilities’ current preference for combined cycle gas turbines. Over the decade of the 2030s, we anticipate that wind and solar energy will supply ~20% to 30% of total U.S. electricity demand.[12] The gas fired generating fleet will then be subject to much more intermittent dispatch, as gas turbines are forced to offset random variations in wind and solar generation. As renewable generation continues to rise, moreover, it will suppress the output of higher cost gas turbine generators, whose average capacity factor may begin to decline as a result. If so, these changes in the operating conditions of the gas turbine fleet could erode the current popularity of fuel efficient, industrial scale gas turbines in combined cycle configuration and increase demand for highly flexible open cycle gas turbines, including aeroderivative designs capable of quick starts and rapid ramping of their power output.

Appendix: The Methodology and Assumptions Underpinning Our Forecasts

In our note of Nov. 15, A Primer on How the Power Market Shapes the Market for Gas Turbines[13], we analyzed the two principal drivers of global gas turbine orders, the growth in global power demand and the need to replace retiring generation capacity. In that analysis, we modeled gross annual additions of firm generating capacity as the sum of (i) the capacity required each year to replace retiring power plants plus (ii) the net new capacity required to meet the annual growth in global power demand. To forecast future gross additions of fossil generation capacity, and gas turbine generation capacity in particular, we have adjusted our estimate of gross annual additions of firm generating capacity by subtracting, first, expected additions of nuclear capacity and second, the firm capacity contributed by rapidly expanding wind and solar generation.

As global electricity consumption continues to rise, it will drive a commensurate increase in the need for firm power generation capacity that can dispatched as needed to meet peak demand (“firm capacity”).[14] Critically, the growth of renewable generation, supplied as it is by intermittent resources such as wind and solar, contributes only a very limited amount of firm capacity. Even across large fleets of wind and solar generating units, the firm capacity value of these resources — the level of generation likely to be continuously available on a round-the-clock basis – is a fraction of their total potential or nominal capacity. The continuously available output of regional wind fleets, for example, can be estimated at just 10% of their nominal capacity, and even solar farms can be counted on for only 40% of their nominal capacity, and then only during daytime hours. Only a fraction of the growth in renewable generation will thus be available to offset the need for firm capacity. To supply the gross capacity additions needed to meet the growth of power demand and to replace retiring power plants, the growth of renewable generation capacity must be supplemented by nuclear and fossil capacity additions.

To estimate the need for future gross additions of firm generation capacity, we have adopted conservative assumptions for future demand growth as well as for the retirement of existing firm capacity resources, such as fossil and nuclear power plants.

With respect to net new capacity additions, our forecast assumes that the ratio of (i) growth in electricity consumption to (ii) growth in GDP falls to half the level observed over the last 10 years in each region of the globe (see Exhibit 31). While conservative, this assumption is consistent with the historical deceleration in power demand growth observed in every region over the last decade (see Exhibit 32), as ratios of power demand to GDP growth have fallen (Exhibit 33).[15] In the developed economies, the slowing of power demand growth has been particularly marked: power demand growth over the last five years has ranged from zero to 0.3% p.a. across the OECD countries. In certain developing regions, including China, India and the Middle East, power demand growth remains strong, requiring capacity additions well in excess of expected retirements. By the late 2020s, however, net additions of new firm capacity are likely to slow in these regions as well, reflecting (i) increasing energy efficiency in these economies and (ii) the gradual deceleration of economic growth in these rapidly developing economies.

Exhibit 31: Historical and Forecast Growth in Global Power Generation (Annual % Change)

_______________________________________________________________________________________________________

Source: U.S. Energy Information Administration, Organization for Economic Cooperation and Development (OECD), SSR analysis and estimates

Exhibit 32: Power Demand Growth by Exhibit 33: Average Annual Decline in

Region, 2006-2016 and 2011-2016 CAGRs Electricity Use per Dollar of GDP, 2011-20161

 

________________________________________________________________________________________________________

1. Constant 2010 US$; GDP calculated at purchasing power parity

Source: U.S. Energy Information Administration, Organization for Economic Cooperation and Development (OECD), SSR analysis and estimates

We base our estimate of future power plant retirements on the dates that currently operating generating units entered service and the historical pattern of retirements for power plants of similar technology and fuel. The Energy Information Administration of the U.S. Department of Energy maintains data on the capacity, primary fuel, generation technology, and commercial operation date of the approximately 16,000 generating units in operation in the United States. By calculating the age at retirement of the generating units that have been retired from service, we compiled a historical data base that allows us to estimate the likely retirement dates of similar plants in operation today. On the basis of this data, we have assumed, somewhat conservatively, average useful lives of 60 years for nuclear, coal, gas and oil fired steam turbine generators and 35 years for gas and oil fired combustion turbines and combined cycle gas turbine generators.[16] Adding these estimated useful lives to the estimated commissioning dates of the generation capacity currently in service around the world today allows us to forecast when this capacity will be retired.

In Exhibit 34, we present our estimate of the required annual additions of firm generating capacity, from 2020 through 2040, reflecting the sum of (i) the replacement capacity required to offset power plant retirements and (ii) the net new generation capacity needed to meet the growth in power demand. Even given our conservative assumptions for the future growth of firm power demand, our analysis suggests that global power demand growth will drive a need for ~57 GW annually of net new firm capacity through 2040. Importantly, however, we expect that from 2025 on the need to offset capacity retirements will significantly outweigh the need for net new capacity additions.

In Exhibit 35, we present our estimate of how the required gross annual additions of firm generating capacity will be met through a combination of net new nuclear capacity, growth in firm renewable capacity and gross additions of fossil capacity. As can be seen there, we expect that from 2025 on gross additions of fossil generation capacity will comprise the bulk of the firm generation capacity added globally.

Exhibit 34: Required Net Annual Additions Exhibit 35: Technology Breakdown of

Of Firm Generation Capacity (GW) Additions of Firm Generation Capacity (GW)

 

______________________________________________________________________________________________

Source: Energy Information Administration, International Energy Agency, Organization for Economic Co-operation and Development, SSR analysis and estimates

Our analysis suggests that global retirements of fossil fuel generation capacity will average ~58 GW annually over 2021-2025, 88 GW annually over 2026-2030, 123 GW annually over 2031-2035, and 142 GW p.a. over 2036-2040 (Exhibit 36). Because this global wave of retirements will be concentrated in the developed economies of North America and Europe, we expect these regions to become the two largest markets for fossil fuel capacity additions, including gas turbines, over this period (Exhibit 37).

Exhibit 36: Estimated Retirements of Fossil Generation Capacity by Region (GW)

______________________________________________________________________________________________

Source: Energy Information Administration, International Energy Agency, Organization for Economic Co-operation and Development, SSR analysis and estimates

Exhibit 37: Gross Annual Additions of Fossil Generation Capacity by Region (GW)

______________________________________________________________________________________________

Source: Energy Information Administration, International Energy Agency, Organization for Economic Co-operation and Development, SSR analysis and estimates

We expect the bulk of the fossil generation capacity added globally through 2040 to be gas fired (Exhibit 38). Our analysis assumes that, in regions which historically have relied heavily on coal, coal-fired capacity additions will continue to comprise the same share of gross capacity additions that they have over the last decade. Critically, however, coal’s share of gross capacity additions is negligible in North America and Japan, two markets facing large capacity retirements through 2040, as well as in the Middle East, a region with a very substantial need for net new capacity additions. In the case of China, the market with the largest need for net new capacity additions, the government has stated that it intends to limit total coal generation capacity to 1,100 GW, and we have respected this cap in our forecasts.

Exhibit 38: Gross Additions of Gas and Coal Fired Generation Capacity by Region (GW)

Gas Coal

 ______________________________________________________________________________________________

Source: Energy Information Administration, International Energy Agency, Organization for Economic Co-operation and Development, Organization for Economic Co-operation and Development, SSR analysis and estimates

Based on the assumptions above, we see total gas turbine orders rising from 31 GW in 2019 to 51 GW in 2025 and 73 GW in 2030, implying 5-year growth rates of 9.1% p.a. over 2020-25 and 7.6% p.a. over 2025-30 (Exhibit 39). Reflecting the scale of expected capacity retirements in the mature economies of North America and Europe, and the preference in these regions for gas over coal fired capacity additions, we see these regions dominating global gas turbine orders in the coming decade. We expect the United States to be the single largest market for gas turbines over the coming decade, with 30% of forecast orders, followed by Europe with 23% (Exhibit 39). (For a sensitivity analysis of this base forecast, please see our research report of June 19, 2018, GE, SIE,MHVYF: The Impending Recovery in Gas Turbine Orders).[17]

Exhibit 39: Base Case Forecast of Global Gas Turbine Orders (GW)

______________________________________________________________________________________________

Source: Energy Information Administration, International Energy Agency, Organization for Economic Co-operation and Development, Organization for Economic Co-operation and Development, SSR analysis and estimates

Exhibit 40: Forecast Gas Turbine Orders by Region (Cumulative GW Over 2019-2030)

______________________________________________________________________________________________

Source: Energy Information Administration, International Energy Agency, Organization for Economic Co-operation and Development, Organization for Economic Co-operation and Development, SSR analysis and estimates

©2019, 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. Available at http://www.ssrllc.com/publication/ge-sie-mhvyf-a-primer-on-how-the-power-market-shapes-the-market-for-gas-turbines-historical-developments-their-implications-for-the-future/ 
  2. Available at http://www.ssrllc.com/publication/ge-sie-mhvyf-a-primer-on-how-the-power-market-shapes-the-market-for-gas-turbines-historical-developments-their-implications-for-the-future/ 
  3. Available at http://www.ssrllc.com/publication/ge-sie-mhvyf-a-primer-on-how-the-power-market-shapes-the-market-for-gas-turbines-historical-developments-their-implications-for-the-future/ 
  4. In our modelling, we have assumed that peak power demand, and thus the need for firm capacity, will rise in tandem with total electricity consumption. 
  5. Available at http://www.ssrllc.com/publication/ge-sie-mhvyf-the-impending-recovery-in-gas-turbine-orders/
  6. Critically, the growth of renewable generation, supplied as it is by intermittent resources such as wind and solar, contributes only a very limited amount of firm capacity. Even across large fleets of wind and solar generating units, the firm capacity value of these resources — the level of generation likely to be continuously available on a round-the-clock basis – is a fraction of their total potential or nominal capacity. The continuously available output of regional wind fleets, for example, can be estimated at just 10% of their nominal capacity, and even solar farms can be counted on for only 40% of their nominal capacity, and then only during daytime hours. Only a fraction of the growth in renewable generation will thus be available to offset the need for firm capacity. To supply the gross capacity additions needed to meet the growth of power demand and to replace retiring power plants, the growth of renewable generation capacity must be supplemented by nuclear and fossil capacity additions. 
  7. Annual output in MWh over potential annual output (MW of installed capacity x 365 x 24). 
  8. Annual output in MWh over potential annual output (MW of installed capacity x 365 x 24). 
  9. Levelized cost of energy represents the constant price a generator would need to receive for its power output over the life of the project to recover its operating costs, ensure the return of its invested capital, and earn a return equivalent to its after-tax weighted average cost of capital. 
  10. The excess of firm capacity over peak demand, expressed as a percentage of peak demand. 
  11. As the name implies, aeroderivative gas turbines are derived from commercial jet turbine engines. Smaller and less energy efficient than turbine classes F through J, aeroderivatives are nonetheless prized for their operational flexibility, including quick start and rapid ramp capabilities, as well as for their ease of maintenance. Aeroderivatives are deployed primarily as peakers or for use in isolated locations (e.g. off-shore oil platforms). 
  12. See our research report of May 21, 2018, Half of U.S. Generating Capacity Will Retire by 2040:What Is the Impact on Power Supply, Power Markets, Coal Burn and Rail Volumes?, available at http://www.ssrllc.com/publication/half-of-u-s-generating-capacity-will-retire-by-2040-what-is-the-impact-on-power-supply-power-markets-coal-burn-and-rail-volumes/
  13. Available at http://www.ssrllc.com/publication/ge-sie-mhvyf-a-primer-on-how-the-power-market-shapes-the-market-for-gas-turbines-historical-developments-their-implications-for-the-future/ 
  14. In our modelling, we have assumed that peak power demand will rise in tandem with total electricity consumption. 
  15. See our research report of February 14, 2018, Power Demand Growth: What Does International Power Demand Growth Tells Us About the Outlook for the U.S. Power Sector?, available at http://www.ssrllc.com/publication/power-demand-growth-what-does-international-power-demand-growth-tells-us-about-the-outlook-for-the-u-s-power-sector/
  16. Rather than assume that each steam turbine generator is retired on the 60th anniversary of its in-service date, and each gas turbine generator on its 35th anniversary, we have constructed distributions of potential retirements dates around these expected dates. For example, for all the steam turbine generating units whose 60th year of operation falls in 2030, we have assumed that the actual retirement dates of this cohort of units is distributed over 19 years centered on 2030, i.e., from 2021 through 2039. We assigned the highest probability (10%) to retirement in 2030, the 60th year of operation, and declining probabilities for each year above and below 2030 (i.e., 9% probability of retirement in 2029 or 2031, 8% probability in 2028 or 2032, 7% in 2027 and 2033 and so forth). 
  17. Available at http://www.ssrllc.com/publication/ge-sie-mhvyf-the-impending-recovery-in-gas-turbine-orders/