THE ECONOMICS OF SOLAR ENERGY

There is a wide variety of solar energy technologies and they compete in different energy markets, notably centralized power supply, grid-connect­ed distributed power generation and off-grid or stand-alone applications. For instance, large-scale PV and CSP technologies compete with technol­ogies seeking to serve the centralized grid. On the other hand, small-scale solar energy systems, which are part of distributed energy resource (DER) systems, compete with a number of other technologies (e. g., diesel genera­tion sets, off-grid wind power etc.). The traditional approach for compar­ing the cost of generating electricity from different technologies relies on the “levelized cost” method. The levelized cost (LCOE) of a power plant is calculated as follows:

Подпись:image098л-х(1 і r)T (1 + r)7 -1

where OC is the overnight construction cost (or investment without ac­counting for interest payments during construction); OMC is the series of annualized operation and maintenance (O&M) costs; FC is the series of
annualized fuel costs; CRF is the capital recovery factor; CF is the capac­ity factor; r is the discount rate and T is the economic life of the plant.

In this section, we discuss the economics of grid connected PV and CSP under various scenarios. One of the main challenges to the economic anal­ysis of power generation technologies is the variation in cost data across technology type, size of plant, country and time. Since fuel costs are highly volatile and capital costs of solar technologies are changing every year, an economic analysis carried out in one year might be outdated the next year. Nevertheless, the analysis presented here could help illustrate the cost com­petitiveness of solar energy technologies with other technologies at present.

We have taken data from various sources including Lazard (2009), NEA/IEA (2005, 2010), EIA (2007, 2009) and CPUC (2009). The data were available for different years, so we adjusted them using the GDP deflator and expressed them in 2008 prices for our analysis. Moreover, the existing calculations of LCOE for a technology vary across studies as they use different economic lives, capacity factors and discount rates. Some studies account for financial costs (e. g., taxes and subsidies) (Lazard, 2009; CPUC, 2009), while others include only economic costs (NEA/IEA, 2005, 2010). Therefore, we have taken the maximum and minimum values of overnight construction costs for each technology considered here from the existing studies to reflect the variations in overnight construction costs, along with the corresponding O&M and fuel costs, and applied a uniform 10% discount rate and 2.5% fuel price and O&M costs escalation rate to cost data from all the studies. Since our focus is on economic analysis, taxes, subsidies or any types of capacity credits are excluded. Please see Table 2 for key data used in the economic analysis.

Figure 4 presents the results of the levelized cost analysis. Although the costs of solar energy have come down considerably and continue to fall, the levelized costs of solar energy are still much higher compared to conventional technologies for electricity generation, with the exception of gas turbine. For example, the minimum values of levelized cost for solar technologies (US$192/MWh for PV and US$194/MWh for CSP) are more than four times as high as the minimum values of the levelized cost of supercritical coal without carbon capture and storage (US$43/MWh). Among renewable energy technologies, wind and hydropower technolo­gies are far more competitive with fossil fuel and nuclear power plants.

TABLE 2: Key Data Used in Economic Analysis

Technology

Overnight Construction

Cost (US$/kW)

Plant Economic Life (years)

Capacity Factor (%)

Source

Solar PV

Min

2878

25

21

NEA/IEA

Max

7381

25

20

NEA/IEA

Solar CSP

Min

4347

25

34

NEA/IEA

Max

5800

20

26

Lazard

Wind

Min

1223

25

27

NEA/IEA

Max

3716

25

23

NEA/IEA

Gas CC

Min

538

30

85

NEA/IEA

Max

2611

30

85

NEA/IEA

Gas CT

Min

483

25

85

NEA/IEA

(2005)

Max

1575

20

10

Lazard

Hydro

Min

757

80

34

NEA/IEA

Max

3452

20

50

CPUC

IGCC w CSS*

Min

3569

40

85

NEA/IEA

Max

6268

40

85

NEA/IEA

Supercritical

Min

1958

40

85

NEA/IEA

Max

2539

40

85

NEA/IEA

Nuclear

Min

3389

60

20

EIA

Max

8375

20

90

Lazard

Note: * IGCC with carbon capture and storage. лSupercritical coal.

The difference between the minimum and maximum values for the level- ized costs of solar energy technologies (and also other energy technologies) are wide due mainly to large variations in overnight construction costs and to different capacity factors. For example, the overnight construction costs of grid connected solar PV system vary from US$2,878/kW to US$7,381/ kW (NEA/IEA, 2010). Similarly, the overnight construction costs of CSP vary from US$4,347/kW (NEA/IEA, 2010) to US$5,800/kW (Lazard, 2009). The capacity utilization factor of simple cycle gas turbine varies from 10% (Lazard, 2009) to 85% (NEA/IEA, 2010). Furthermore, very different economic lives are assumed for hydro, coal and nuclear plants.

CO

О

 

і Capital Cost eObMCost ■ Fuel Cost

 

$719

7%

 

Подпись: Solar Energy: Application, Economics, and Public Perceptio
image100

FIGURE 4: Levelized Cost of Electricity Generation by Technology (2008US$/MWh) Note: * IGCC with carbon capture and storage. ASupercritical coal.

 

image101

100,000

 

S/kW

 

image102

10.000

 

— г I

 

1,000

 

image103

I J N

 

1000

 

Cii mu latrve Installation (GWp)

 

image104

image105

sources: Earth Policy Institute (2009): DOE (2008b): Stoddard et al. (2007); Charts et al. (2005): Winter (1991)

1Ї76

Produced Silicon PV Module-;

(Global)

Onshore Wind Power Plants (.Denmark.1

Onshore Wind Power Plants

Подпись: [2.6 US D AY) Подпись: (4.3 USO/Wl image108

(UbAJ

[1,4 USD/WI

Подпись: 1.00010.000 100,000 1.000.000

Cumulative Global Capacity (MWJ

Sources: Arvizu et. al (2011)

FIGURE 5: Experience Curves of Renewable Electric Technologies

It is also interesting to observe the contributions of various cost compo­nents (e. g., capital, O&M and fuel costs) to levelized cost. While capital cost accounts for more than 80% of the levelized cost for renewable energy tech­nologies, it accounts for less than 60% in conventional fossil fuel technolo­gies (e. g., coal, gas combined cycle). Fuel costs are the major components in most fossil fuel technologies

Using the concept of experience or learning curves which plot cost as a function of cumulative production on a double-logarithmic scale, imply­ing a constant relationship between percentage changes in cost and cumu­lative output11, existing studies (e. g., Kannan et al., 2006; Hertlein et al., 1991; EWEA, 2008; Ackerman and Erik, 2005; Dorn, 2007, 2008; Neij, 2008), expect significant reductions in the capital costs of solar energy technologies (see Figure 5a). The cost of solar PV has been declining rap­idly in the past, compared not only to conventional technologies such as coal and nuclear, but also to renewable technology such as wind. The 2011 Special Report on Renewable Energy Carried out by Intergovernmental Panel on Climate Change (Arvizu et. al (2011) has also demonstrates re­duction in costs of solar and wind power along with their cumulative in­stalled capacity (see Figure 5b). The “learning rate” of solar PV, CSP and wind are 21%, 7%, and 8%, respectively (Nemet, 2007; Beinhocker et al., 2008).

Considering the declining trend of capital costs as discussed above, we analyzed the levelized costs of solar energy technologies when their capi­tal costs drop by 5% to 25% from the present level. Figure 6 shows how the levelized cost of solar thermal trough, solar thermal tower, photovolta­ic thin-film and photovoltaic crystalline would decline if their capital cost requirements were to fall by up to 25% and how those costs would com­pare to the maximum levelized costs of traditional electricity generation plants. As illustrated in the figure, the minimum values of levelized cost of any solar technologies, including tower type CSP, which is currently the least costly solar technology, would be higher than the maximum values of levelized costs of conventional technologies for power generation (e. g., nuclear, coal IGCC, coal supercritical, hydro, gas CC) even if capital costs of solar energy technologies were reduced by 25%.

Подпись: A Review of Solar Energy: Markets, Economics and Policies 1 83

■ CSPTrough ■ PVThin-Film ■ PVCrystalline ■CSPTower

 

350

 

Reference -5% -10% -15% 20% 25%

 

FIGURE 6: Sensitivity of levelized costs of solar technologies to their capital cost reduction

 

image111

Since fossil fuels such as coal and gas produce negative externalities at the local level (e. g., local air pollution) as well at the global level (e. g., GHG emissions), whereas solar energy technologies do not, it would be unfair to compare solar energy technologies with fossil fuel technologies without accounting for those externalities. Hence, we further analyze the levelized costs of electricity generation technologies, developing a frame­work to capture some of those external costs. The framework accounts for the environmental damage costs of fossil fuels, particularly climate change damage costs. Damage costs of local air pollution are not included due to a lack of data. Since obtaining actual values of damage costs of emissions from different fossil fuel technologies is highly complex, we employed a sensitivity analysis by considering various values of damage costs ranging from US$0/tCO2 to US$100/tCO2. Figure 7 plots the levelized costs of various technologies against the climate change damage costs. The figure demonstrates that the minimum values of levelized costs of solar energy technologies would be higher than the maximum values of the levelized costs of fossil fuel technologies even if the climate change damage costs of 100/tCO2 are imputed to fossil fuel technologies. In other words, even if we assign a climate change damage cost of US$100/tCO2 to fossil fuel technologies, solar energy technologies would still presently be economi­cally unattractive as compared to fossil fuel technologies.

The analysis above shows that climate change mitigation benefits would not be sufficient to make solar energy technologies economically attractive. However, solar energy technologies also provide additional ben­efits, which are not normally excluded from traditional economic analysis of projects. For example, as a distributed energy resource available near­by load centers, solar energy could reduce transmission and distribution (T&D) costs and also line losses. Solar technologies like PV carry very short gestation periods of development and, in this respect, can reduce the risk valuation of their investment (Byrne et al., 2005b). They could enhance the reliability of electricity service when T&D congestion occurs at specific locations and during specific times. By optimizing the location of generating systems and their operation, distributed generation resourc­es such as solar can ease constraints on local transmission and distribu­tion systems (Weinberg et al., 1991; Byrne et al., 2005b). They can also protect consumers from power outages. For example, voltage surges of a

mere millisecond can cause “brownouts,” causing potentially large losses to consumers whose operations require high quality power supply. They carry the potential to significantly reduce market uncertainty accompany­ing bulk power generation. Because of their modular nature and smaller scale (as opposed to bulk power generation), they could reduce the risk of over shooting demand, longer construction periods, and technological obsolescence (Dunn, 2000 quoted in Byrne et al., 2005b: 14). Moreover, the peak generation time of PV systems often closely matches peak loads for a typical day so that investment in power generation, transmission, and distribution may be delayed or eliminated (Byrne et al., 2005b). However, developing a framework to quantify all these benefits is beyond the scope of this study.

image112

7.3 ESTIMATED FUTURE GROWTH OF SOLAR ENERGY AND BARRIERS TO REALIZING GROWTH

Advocates of solar energy claim that it will play a crucial role in meeting future energy demand through clean energy resources. Existing projec­tions of long-term growth (e. g., until 2050) of solar energy vary widely based on a large number of assumptions. For example, Arvizu et al. (2011) argue that expansion of solar energy depends on global climate change mitigation scenarios. In the baseline scenario (i. e., in the absence of cli­mate change mitigation policies), the deployment of solar energy in 2050 would vary from 1 to 12 EJ/yr. In the most ambitious scenario for climate change mitigation, where CO2 concentrations remain below 440 ppm by 2100, the contribution of solar energy to primary energy supply could reach 39 EJ/yr by 2050.

EPIA/Greenpeace (2011) produces the most ambitious projections of future PV installation. The study argues that if existing market supports are continued and additional market support mechanisms are provided, a dramatic growth of solar PV would be possible, which will lead to world­wide PV installed capacity rising from around 40 GW in 2010 to 1,845 GW by 2030. The capacity would reach over 1000 GW in 2030 even with a lower level of political commitment.

A study jointly prepared by Greenpeace International and the Euro­pean Renewable Energy Council (Teske et al., 2007) projects that installed global PV capacity would expand to 1,330 GW by 2040 and 2,033 GW by 2050. A study by the International Energy Agency (IEA, 2008) estimates solar power development potential under two scenarios that are differen­tiated on the basis of global CO2 emission reduction targets. In the first scenario, where global CO2 emissions in 2050 are restricted at 2005 level, global solar PV capacity is estimated to increase from 11 GW in 2009 to 600 GW by 2050. In the second scenario, where global CO2 emissions are reduced by 50% from 2005 levels by 2050, installed capacity of solar PV would exceed 1,100 GW in 2050.

Like solar PV, projections are available for CSP technology. A joint study by Greenpeace, the European Solar Thermal Power Industry (ES – TIA) and the International Energy Agency projects that global CSP capac­ity would expand by one hundred-fold to 37 GW by 2025 and then sky­rocket to 600 GW by 2040 (Greenpeace et al., 2005). Teske et al. (2007) project that global CSP capacity could reach 29 GW, 137 GW and 405 GW in 2020, 2030 and 2050, respectively. IEA (2008) projects that CSP capac­ity could reach 380 GW to 630 GW, depending on global targets for GHG mitigation. In the case of solar thermal energy, the global market could ex­pand by tenfold to approximately 60 million tons of oil equivalent (Mtoe) by 2030 (IEA World Energy Outlook 2006). A more optimistic scenario from the European Renewable Energy Council (2004) projects that solar thermal will grow to over 60 Mtoe by 2020, and that the market will con­tinue to expand to 244 Mtoe by 2030 and to 480 Mtoe, or approximately 4% of total global energy demand, by 2040. It would be also relevant to envisage the contribution of solar energy to the global energy supply mix. According to EREC (2004), renewable energy is expected to supply nearly 50% of total global energy demand by 2040. Solar energy alone is project­ed to meet approximately 11% of total final energy consumption, with PV supplying 6%, solar heating and cooling supplying 4% and CSP supplying 1% of the total. Shell (2008) shows that if actions begin to address the challenges posed by energy security and environmental pollution, sources of energy other than fossil fuels account for over 60% of global electric­ity consumption, of which one third comes from solar energy. In terms of global primary energy mix, solar energy could occupy up to 11% by 2050.

Notwithstanding these optimistic projections, the existing literature identifies a range of barriers that constrains the deployment of solar en­ergy technologies for electricity generation and thermal purposes. These barriers can be classified as technical, economic, and institutional and are presented in Table 3. Technical barriers vary across the type of technol­ogy. For example, in the case of PV, the main technical barriers include low conversion efficiencies of PV modules; performance limitations of system components such as batteries and inverters; and inadequate supply of raw materials such as silicon. In the case of stand-alone PV systems, storage is an important concern, as is the shorter battery life compared to that of the module. Furthermore, safe disposal of batteries becomes dif­ficult in the absence of a structured disposal/recycling process. With re­gard to solar thermal applications, there are two main technical barriers.

Подпись: CO CO TABLE 3: Barriers to the Development and Deployment of Solar Energy Technologies

Подпись: Technical Barriers Подпись: Economic Barriers Подпись: Solar Energy: Application, Economics, and Public Perceptio

PV

• The efficiency constraint: 4% to 12% (for thin film) and under 22% (for crystalline) in the current market (EPIA/Greenpeace, 2011).

• Performance limitations of balance of system (BOS) components such as batteries, inverters and other power conditioning equip­ments (Rickerson et al., 2007, Beck and Martinot, 2004; 0"Rourke et al., 2009).

• Silicon supply: strong demand for PV in 2004 and 2005 outpaced the supply and partly stalled the growth of solar sector (Wenzel, 2008; PI, 2006).

• Cadmium and tellurium supply for certain thin film cells: these two components are by-products from respectively the zinc mining and copper processing and their availability depends on the evolution of these industries (EPIA/Greenpeace, 2011).

• High initial capital cost and the related lack of easy and consistent financing options forms one of the biggest barriers primarily in developing countries (Beck and Martinot, 2004).

• Investment risks seen as unusually high risks by some financial institutions because of lack of experience with such projects (Gold­man et al., 2005; Chaki, 2008

• Cost of BOS is not declining proportional to the decline in module price (Rickerson et al., 2007).

• The fragility of solar development partnerships: many PV projects are based on development partnerships and with the early depar­ture of a partner the revenue to complete, operate and maintain the system may falter (Ahiataku-Togobo, 2003).

Solar Thermal

• Heat carrying capacity of heat transfer fluids.

• Thermal losses and energy storage system issues with CSPs (Herrmann et al., 2004; IEA, 2006a).

• Supply orientation in the design of solar water heaters when product diversity is needed to match diverse consumer de­mand profiles.

• For solar water heating, lack of integration with typical build­ing materials, existing appliances and infrastructure, designs, codes, and standards has hampered widespread application.

• In case of central receiver systems the promising technolo­gies such as the molten salt-in-tube receiver technology and the volumetric air receiver technology, both with energy stor­age system needs more experience to be put for large-scale application (Becker et al., 2000).

• High upfront cost coupled with lengthy payback periods and small revenue streams raises creditworthiness risks.

• The financial viability of domestic water heating system is low.

• Backup heater required in water heating systems to provide reliable heat adds to the cost.

• Increasing cost of essential materials like copper make water heating and distribution costly.

• Limited rooftop area and lack of building integrated systems limit widespread application.

Подпись: A Review of Solar Energy: Markets, Economics and Policies 1 89

TABLE 3: Cont.

 

PV Solar Thermal

Institutional/ • Regulartory Barriers ф

The limited capability to train adequate number of technicians to effectively work in a new solar energy infrastructure (Banerjee, 2005; Dayton, 2002).

Limited understanding among key national and local institutions of basic system and finance.

Procedural problems such as the need to work with several public sector agencies (e. g., in India, MNRE, IREDA, the Planning Com – misson, and the Ministry of Agriculture and Rural Development) (Radulovic, 2005).

Barriers limiting entry of distributed technology platforms into the grid, including potential for access restrictions by conventional utilities (Margolis and Zuboy, 2006); potential burdens include over-complicated procedures for interconnection, metering and billing (Florida Solar Energy Center, 2000).

 

They are limits to the heat carrying capacity of the heat transfer fluids and thermal losses from storage systems (Herrmann et al. 2004; IEA 2006a). In addition, as seen in Table 3, there are constraints with regard to system design and integration as well as operating experience for system optimi­zation. For example, lack of integration with typical building materials, designs, codes and standards make widespread application of solar space and water heating applications difficult. In the case of CSP, technologies such as the molten salt-in-tube receiver technology and the volumetric air receiver technology, both with energy storage systems, need more experi­ence to be put forward for large-scale application (Becker et al., 2000). Moreover, solar energy still has to operate and compete on the terms of an energy infrastructure designed around conventional energy technologies.

The economic barriers mainly pertain to initial system costs. Cost com­parisons for solar energy technologies by suppliers and users are made against established conventional technologies with accumulated industry experience, economies of scale and uncounted externality costs. Solar energy technologies thus face an “uneven playing field,” even as its en­ergy security, social, environmental and health benefits are not internal­ized in cost calculations (Jacobson & Johnson, 2000). Financing is another critical barrier. Financial institutions consider solar energy technologies to have unusually high risks while assessing their creditworthiness. This is because solar energy projects have a shorter history, lengthy payback periods and small revenue stream (Goldman et al., 2005; Chaki, 2008). This implies higher financial charges (e. g., interest rates) to solar energy projects.

Aside from economic and technical constraints, PV and solar thermal technologies face institutional barriers that reflect considerably the nov­elty of the technologies. They range from limited capacities for workforce training, to mechanisms for planning and coordinating financial incentives and policies. Inadequate numbers of sufficiently trained people to prepare, install and maintain solar energy systems is another common barrier. In India, for example, the country invested in the training of nuclear physi­cists and engineers since its independence, while similar requirements for renewable technologies were ignored (Banerjee, 2005).

In some instances, existing laws and regulations could constrain the deployment of solar energy. For example, some applications of small-scale PV systems have had to overcome, cumbersome and inappropriate "interconnection requirements, such as insurance, metering and billing issues, in order to sell excess power generation back into the grid (Florida Solar Energy Center, 2000). However, these potential con­straints can become binding only when other policies in place induce or require use of solar energy in order to overcome its higher cost. Even if interconnection were to be simplified, grid based electricity suppliers would still have to address challenges of integrating significant quantities of episodic, non-dispatchable solar power into the grid (or the high cost of current storage options).