21, rue d’Artois, F-75008 PARIS C5-204 CIGRE 2018

http : //www.cigre.org

Analysis of the current Carbon Tax implementation in the Chilean Electric

Market and future regulatory developments to allow effective CO2 reduction

F. LEIVA* C. MANCILLA S. JERARDINO R.SAAVEDRA

KAS Engineering
Chile

SUMMARY

The Government of Chile announced GHG emission reduction targets for the short term (2020) and
the medium term (2030), and has implemented a tax package to help achieve those targets. The fiscal
reform of 2014 contains the provisions for taxes on carbon dioxide equivalent (CO 2e, where
‘equivalent’ means that GHGs other than CO2 are expressed in CO2 terms depending on their global
warming potential relative to that of CO2) as well as on local pollutants: particulate matter (PM),
nitrogen oxides (NOX), and sulfur dioxide (SO2). The carbon tax is set at an amount of US$5 per ton
of carbon dioxide (tCO2) emitted by stationary plants of at least 50 megawatts (MW), starting in 2017.
This paper models a limited number of carbon tax scenarios, defined by different values of carbon tax
and different ways of applying the tax. None of these scenarios represents a specific policy of the
Government of Chile. Rather, scenarios are selected to help understand the main underlying
relationships, transition points in the generation mix, and broad types of policy effects.
The scenarios are intended to help start a conversation with the community of policymakers and
stakeholders in Chile’s power sector, which may guide future development of specific policy options

KEYWORDS

Coal, CO2 taxation, Marginal Cost, Levelized Cost of Energy, Regulatory Developments, Greenhouse
Emissions, Paris Agreement.

francisco.leiva.g@gmail.com
CONTEXT OF ANALYSIS OF THIS STUDY
The carbon tax is set at an amount of US$5 per ton of carbon dioxide (tCO2) emitted by stationary
plants of at least 50 megawatts (MW), starting in 2017. The tax package is designed to change the
incentives of private investors to move the generation mix towards lower levels of emissions.
Chile has two international commitments for carbon emissions.
 Short-term target (2020). Chile has voluntarily committed to reducing emissions by 20%
by 2020 with respect to its 2007 trajectory. The commitment was
established at the Conference of the Parties in Copenhagen
(COP15) in 2009, and was reaffirmed in the New York Climate
Summit of 2014.
 Medium-term target (2030) Chile committed to reduce its carbon emissions per unit of gross
domestic product (GDP) by year 2030 by 30% relative to 2007.
For purposes of this paper, the estimated targets for emissions from power generation are: For 2020:
24.2 million tCO2e emitted; for 2030: 21.9 million tCO2e, emitted, or 17.3 million tCO2e if
international concessional finance is provided.
Consistent with Chile’s two commitments, this paper estimates a short term target for 2020 and a
medium term target for 2030. The country’s total targeted emissions for 2020 and 2030 are
estimated based on the 2007 emissions, the carbon intensity target estimated for 2020, and the
published carbon intensity target for 2030. That is, this paper’s estimated targets assume that the entire
energy industry is affected by the tax or equivalent policies. Table 1 summarizes the results of the
estimates based on the calculation explained above.
Table 1: Estimate of Emissions Targets
2030 target
Units Year 2007 2020 target
Non conditional Conditional
Emissions intensity per
tCO2eq/MM of CLP$ 2011 1.02 0.83 0.71 0.56
GDP
Energy sector emissions tCO2eq 68,944,000 60,895,040 55,168,935 43,951,229
Generation sector emissions tCO2eq 27,370,768 24,175,331 21,902,067 17,305,718

How the carbon tax works

The carbon tax (and the tax on local emissions) does not apply in determining the marginal cost of
energy for the purpose of bid price, and therefore has no effect in determining the merit order for
economic dispatch. The carbon tax (and the tax on local emissions) is applied after dispatch has been
defined. If there are generation costs that, as a result of the emissions tax, are above marginal cost,
such additional costs are to be distributed among all power generators that withdraw energy from the
system, prorated on their withdrawals.

INPUTS AND ASSUMPTIONS

For this paper, we developed a unimodal optimization model for the expansion of the energy mix,
assuming that investments in generation take place in a centralized way: the model works as if there
were one agent deciding the investments in power generating units for the entire system. This is done
based on a set of available technologies and with the objective of minimizing total cost, that is, the
sum of operating and investment costs. The model is subject to various constraints: meeting demand,
technical, and regulatory.
The Ose2000 is a multi-nodal, multi-dam market model that allows determining the dispatch of all
power units to be at each stage of the planning horizon, fulfilling the objective of minimizing the
average operation cost over the period of analysis while meeting demand, technical, and regulatory
constraints. This paper uses Ose2000 model to determine the operation of the generation fleet as it
exists and as it evolves according to results from the expansion planning results.

2
Emissions of global GHGs (CO2e) and of local pollutants (NOX, SOX, and PM) are a result of running
the optimization of generation dispatch in Ose2000 software. To do so, each power generation unit has
an emissions factor associated to it, validated by the Ministry of Environment during the course of this
paper.
The generation expansion plan of the SEN for the short term (2020) corresponds to all power
generation units that to date are declared under construction. The set of candidate technologies for the
expansion plan, each with its levelized cost of generation, is shown in Table 2 and Table 3. In the case
of wind energy, two cases are considered given the wide range of load factors that results from data
sources. The first case (Wind I) assumes a load factor of 35%; the second case (Wind II), 30%.
The transmission expansion plan for the SEN for the short term (2020) corresponds to all works that
are under construction or have been tendered to date. For the medium term (2030), network extensions
and new investments in transmission lines are factored in as required by the expansion of the energy
mix.
Table 2: Levelized Cost of Energy by Technology by 2020 (US$/MWh)
Levelized
Investment Fixed O&M Load Non-fuel
Economic Heat rate Cost of fuel cost of
Technology cost cost Factor variable cost
life (Btu/kWh) US$/MMBtu energy
US$/kW (US$/kW/yr) (%) (US$/MWh)
(US$/MWh)
Hydro 40 3,472 73 58.3 - - 0 83.9
Solar PV 25 1,200 32 30 - - 6 68.5
Coal 30 2,800 63 85 8,976 4.2 3 89
Natural gas 30 1,100 24 70 6,309 8.9 2.5 84.7
Geothermal 30 5,800 104 95 - - 2 88.4
Wind I 20 1,550 53 35 - - 7.7 84.4
Wind II 20 1,550 53 30 - - 7.7 97.1
CSP 30 5,800 48 62.3 - - 0 127.5
Biomass 30 2,700 63 85 0 46.9
Table 3: Levelized Cost of Energy by Technology by 2030 (US$/MWh)
Hydro 40 3,472 73 58.3 - - 0 83.9
Solar PV 25 900 32 30 - - 6 55.9
Coal 30 2,765 63 85 8,976 4.25 3 89
Natural gas 30 1,090 24 70 6,309 11.2 2.5 95.9
Geothermal 30 5,800 104 95 - - 2 88.4
Wind I 20 1,491 53 35 - - 7.7 82.1
Wind II 20 1,491 53 30 - - 7.7 94.5
CSP 30 4,300 48 62.3 - - 6 98.4
Biomass 30 2,621 63 85 0 45.8

Analysis of tipping points

To get the best information on how various carbon taxes could affect the national energy mix and the
emissions volumes we identified key tipping points that induce changes.
Tipping point of levelized cost
A cost of US$14 per ton of CO2e emitted in 2030 represents a value of interest insofar as it causes the
levelized cost of energy for natural gas be lower than that of coal. The estimated value of US$14
results from comparing the levelized cost of coal and of gas for the year 2030, and adding the cost of
the tax for both technologies. This is a ‘theoretical level’ that assumes an optimal utilization of power
generation units (that is, utilizing a coal-fired power generation unit as much as possible so to get the
best economic value from the unit while allowing for required maintenance), and that only considers
the two main thermal generation options: coal and gas.
Tipping point of operation
A tipping point of operation may be calculated by incorporating the cost of the tax in the variable cost
of power generation units: with a value of US$69.9 per ton of CO2e for the year 2030, natural gas and
coal generation are inverted in the merit order. This value is defined as ‘tipping point of operation’.
The tax value that makes the variable cost of a coal-fired plant equal that of a gas-fired one is
calculated by solving the following equation:

3
 𝑃𝐶 ⋅ 𝐻𝑅𝐶 + 𝑁𝐹𝑉𝐶𝐶 + 𝑇𝐶𝑂2 ⋅ 𝐸𝐹𝑋𝐶𝑂2 + 𝑇𝐿𝑃 ∙ 𝐸𝐹𝐶𝐿𝑃 = 𝑃𝐺 ⋅ 𝐻𝑅𝐺 + 𝑁𝐹𝑉𝐶𝐺 + 𝑇𝐶𝑂2 ⋅ 𝐸𝐹𝐺𝐶𝑂2 + 𝑇𝐿𝑃 ∙ 𝐸𝐹𝐺𝐿𝑃
Where:
PX: corresponds to the fuel price.
HRX: is the heat rate of the power generation unit.
NFVCX: is the non-fuel variable cost.
TCO2: is the tax level of CO2
EFXCO2: is the emission factor for CO2.
TLP: is the level of the tax on local pollutants.
EFXLP: is the emissions factor for local pollutants.

Scenarios of analysis
No-tax scenario (scenario 1)
This scenario provides the counterfactual of future investment and electricity plant dispatch given the
expected relative prices of fuels and the expected future costs of new and existing technologies but in
the absence of any emissions taxes. The analysis compares other carbon tax scenarios to this
counterfactual no-tax scenario.
‘Tax not included in variable cost’ scenarios (scenarios 2, 3, and 4)
As currently determined by the Law, the emissions tax does not affect marginal cost or dispatch.
Scenarios 2, 3, and 4 reflect the current normative at the three levels of tax described above, not
applied to variable cost of power generation units:
Scenario ‘US$5 not included in VC’: this scenario considers a tax level on CO2 of
US$5/tCO2e that remains unchanged between 2017 and
2030 (as well as after 2030).
Scenario ‘US$14 not included in VC’: this scenario considers a tax level on CO2 that increases
linearly from US$5/tCO2e in year 2017 to
US$14/tCO2e in year 2030 (and stays at US$14/tCO2e
after year 2030.
Scenario ‘US$30 not included in VC’: this scenario considers a tax level on CO2 that increases
linearly from US$5/tCO2e in year 2017 to US$30/tCO2e
in year 2030 and stays at US$14/tCO2e after year 2030.
Results Analysis
Figure 1 and Table 4 show emissions under scenarios 1, 2, 3, and 4. Emissions of CO2e between the
years 2017 and 2019 are the same than the no-tax scenario and the ‘tax not included in VC’. This is
because the expansion plan is the same, and the impact of the tax is not reflected on the system
operating cost. The curve for scenario 2 covers that of the no-tax scenario because these scenarios
have the same emissions.
In 2030, emissions are 6%, and 9%, lower than the no-tax scenario in scenarios 3, and 4, respectively.
Such as currently defined; the carbon tax (scenario 2, flat US$5/tCO2e) would not reduce emissions.

4
Figure 1: Annual Emissions, ‘tax not included in VC’ scenarios (thousands of tCO2e)

Thousands of tCO2eq 60,000

50,000
40,000
30,000
20,000
10,000
0
2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
year

No Tax 5 No CV 14 No CV 30 No CV Meta 2020 Meta 2030

Table 4: Annual Emissions, ‘tax not included in VC’ scenarios (thousands of tCO2e)
1 2 3 4
Year No Tax US$5 No VC US$14 No VC US$30 No VC
2017 30,562 30,562 30,562 30,562
2018 31,663 31,663 31,663 31,663
2019 33,407 33,407 33,407 33,407
2020 35,506 35,506 35,325 35,325
2030 48,693 48,693 45,830 44,068
Red. 2030 - -0% -6% -9%
Red. 2017-30 - 0% -3% -4%

So far, results are shown as an expected average across the 56 hydro resources considered in the
model. The variability that can take place between hydrology scenarios can matter when quantifying
CO2 emissions. As an example (considering that the effect is the same across the 4 simulated
scenarios), we present an analysis for the no-tax scenario.
Figure 2: Maximum, Minimum, and Average Annual CO2 Emissions ‘No Tax’ (tCO2e)

In extreme hydrology cases (driest and most humid), GHG emissions could vary approximately
between 60 million tCO2e and 40 million tCO2e. When water availability is lowest, more energy needs
to be generated using thermal technologies, which increases CO2 emission levels. The opposite
happens when water is abundant.
Figure 3 presents an histogram of the total level of CO2e emissions by hydro resource scenario, as well
as the cumulative probabilities curve associated with the histogram for the years 2020 and 2030.
Over 50% of cases have emissions between 30 and 38 million tCO2e in 2020, and between 45 and 49
million tCO2e in 2030. Maximum and minimum extremes occur only in exceptional cases (less than
2% of cases).

5
Figure 3: Histogram of the Total Level of CO2e Emissions by Hydro Resource Scenario and
Cumulative Probabilities Curve ‘No Tax’ (tCO2e)

FUTURE REGULATORY DEVELOPMENTS

Carbon tax is one of many policy tools within a wider context of GHG mitigation options, and
therefore, only one of the factors that contribute to a reduction in emissions. This paper present three
different changes to current regulations that can help obtain
Tax included in variable cost’ scenarios (scenarios 5, 6, and 7)
Assuming that the tax is included in the variable cost of generation, it can directly affect dispatching.
This is an alternative way of applying the carbon tax that would achieve greater effects on the energy
mix in the short term. Assessing scenarios that include the tax in variable cost informs the discussion
on policies to mitigate emissions, because it quantifies impacts on prices and investment incentives
that the current design of the tax does not envision. Scenarios 5, 6, and 7 reflect this alternative way of
applying the tax to variable cost at the three levels of tax described above:
Scenario ‘US$5 included in VC’: this scenario considers a tax level on CO2 of
US$5/tCO2e that remains unchanged between 2017 and
2030 (as well as after 2030).
Scenario ‘US$14 included in VC’: this scenario considers a tax level on CO2 that increases
linearly from US$5/tCO2e in year 2017 to US$14/tCO2e
in year 2030, and stays at US$14/tCO2e after year 2030.
Scenario ‘US$30 included in VC’: this scenario considers a tax level on CO2 that increases
linearly from US$5/tCO2e in year 2017 to US$30/tCO2e
in year 2030, and stays at US$30/tCO2e after year 2030.
Scenario 5 keeps the tax flat at US$5/tCO2e during the period of analysis. Scenarios 6 and 7 are
modeled by linearly increasing the tax from US$5/tCO2e in 2017 to US$14/tCO2e and US$30/tCO2e in
2030, respectively, and assume that there are no further increments to the tax after 2030. That is, these
scenarios assume that investment decisions towards the end of the analysis period are based on the
expectation that after 2030 the tax will remain at the level reached by 2030.
Emissions from ‘tax included in VC’ scenarios vs no-tax scenario
In general, moderate increments of the carbon tax achieve relatively considerable reductions compared
to the no-tax scenario. Including or not the carbon tax in the variable cost (VC) makes a difference: for
same levels of the tax, emissions reductions are about twice as much when it is included in VC, both
for year 2030 and over the period. Figure 4 and Table 5 show emissions from scenarios 1, 5, 6, and 7.
That is, the no-tax scenario and scenarios where the tax is included in the VC. All three ‘tax included
in VC’ scenarios have the same emissions in the year 2017, given that they all begin with US$5/tCO 2e
as the tax. From then onwards, they diverge following their different tax trajectories to 2030
(unchanged at US$5, US$14, and US$30). In 2030, emissions under scenarios 6 and 7 are 10% and
18% lower than the no-tax scenario, respectively (and 5% and 8% lower over the period, respectively).
Emissions of scenario 5 would only be 0.2% lower than the no-tax scenario by 2030 (and 1% over the
period).

6
Figure 4: Annual Emissions, ‘tax included in VC’ scenarios (thousands of tCO2e)

Thousands of tCO2eq 60,000

50,000
40,000
30,000
20,000
10,000
0
2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
year

5 CV 14 CV 30 CV Meta 2020 Meta 2030

Table 5: Annual Emissions, ‘tax included in VC’ scenarios (thousands of tCO2e)

1 5 6 7
Year No Tax US$5 VC US$14 VC US$30 VC
2017 30,562 29,993 29,993 29,993
2018 31,663 30,566 30,465 30,238
2019 33,407 32,942 32,834 32,656
2020 35,506 35,087 34,747 34,544
2030 48,693 48,592 43,777 39,721
Red. 2030 - -0% -10% -18%
Red. 2017-30 - -1% -5% -8%

Changes in Capacity Payment

The remuneration of the power is subject to regulation in the Chilean electricity market. It is the
Operator of the system who calculates the power recognized to each generating unit, and it is the
National Energy Commission that determines the price at which the power is valued.
The income by power of any generation unit depends on its installed power and the maximum demand
of the system.
Unlike the remuneration of energy, power or capacity corresponds to the payment for a service that
eventually the system will require at some point, and not for the delivery of a "product". Its objective
is to cover the investment costs of a plant that cannot be remunerated with energy revenues.
In the case of a coal unit, since it is a base load plant its capacity payment is low in relation to energy
revenues (15% and 85%, respectively, of total income). Energy revenues of coal reach 115 [MUSD] in
the long term. In contrast, the revenues from capacity payment does not surpass 18 [MUSD] in the
same period. On the other hand, operating costs represent 70% of the total costs by the year 2033. The
tax payment for CO2 emissions and local pollutants sum to 13%.
According to classic marginalistic theory, a generation unit must recover its investment and operating
costs through capacity payment and infra-marginal energy. Coal units that have already reached their
theoretical life span have already recovered their fixed costs. Hence, it is questionable whether these
units should continue to receive capacity payments. In this paper we propose to remove this payment
to those units that have already reached their operational useful life.
According to CNE, the useful life of a coal unit is 24 years. However, international experience shows
that they can operate for up to 40 years.
Retirement of Thermal Power Units
Electricity generation power units have a specified economic life according to their technology. In the
case of thermal power units, a standard measure of economic life is 40 years. Taking this period as a
reference, and considering the commissioning date of the thermal power units of the SEN, Table 6
shows units that, according to this criterion, would complete their economic life between now and
2030. Some of the units in Table 6 have already completed their economic life, and in theory they all

7
should be decommissioned by 2030 (or have been decommissioned already). The retirement of these
units would have an impact both on the level of emissions (GHGs and local pollutants) and on the
development of the national energy mix. However, no thermal power unit has been decommissioned to
date (at least not any large ones) following the criterion of economic life. Table 6 shows that a number
of power units of over 40 years still operate today.
Policies exist that could encourage the early retirement of thermal power units. These would be non-
market interventions: direct regulation to decommission certain types of plants (or levels of emissions
taxes so high that it would be make it financially impossible to continue operating, achieving the same
result).
Table 6: Date of Commissioning of Thermal Power Units (SEN)
Power Unit Year of Entry Year of Exit Technology
Ventanas 01 1964 2004 Coal
Ventanas 02 1977 2017 Coal
Bocamina 1970 2010 Coal
U12 1973 2013 Coal
U13 1975 2015 Coal
U14 1977 2017 Coal
U15 1978 2018 Coal
NTO1 1995 2035 Coal
NTO2 1995 2035 Coal
CTM1 1995 2035 Coal
CTM2 1995 2035 Coal
Guacolda 01 1995 2035 Coal
Guacolda 02 1996 2036 Coal

Results
The following chart and figure present the total emissions considering all changes presented in this
paper.
Table 7: Annual Emissions after regulatory changes (thousands of tCO2e)
Año CO2 SO2 NOx MP
2017 23,289 23.47 29.63 3.35
2018 31,242 30.21 39.41 4.4
2019 27,474 23.67 33.77 3.6
2020 30,331 27.03 37.86 4.03
2030 18,445 14.75 22.98 2.55

Figure 5: Annual Emissions, ‘tax included in VC’ scenarios (thousands of tCO2e)

50,000 5,000
45,000 4,500
40,000 4,000
Thousands of tCO2e

35,000 3,500
MW installed

30,000 3,000
25,000 2,500
20,000 2,000
15,000 1,500
10,000 1,000
5,000 500
0 0

Instaled Coal Capacity No-Tax After Regulatory Changes

8
CO2 emissions experience a remarkable drop after forcefully decommissioning coal plants. From year
2023 onwards, emissions levels remain practically constant. In contrast to the no-tax scenario, CO2
levels experience a 56% reduction, produce namely for two reasons:
i. Lower dispatch of thermal generation: The decrease in the dispatch of coal is
supplemented by an increase in the generation of natural gas from the existing park which
increases by an average of 40% in the long term, and solar generation, whose dispatch
increases by 260% due to the entry into operation of about 4,000 [MW] in units of this type.
Hence, 45% less energy is generated based on fossil fuels than on the no-tax scenario.
ii. Lower emission factor of natural gas vs coal: The CO2 emission factor of an LNG
unit is, on average, 55% less than that of a coal unit. That is, for every GWh of energy
generated based on natural gas instead of coal, 55% less CO2 is emitted. Therefore, even
though the dispatch of LNG units increases, the volume of CO2 emitted is lower than if coal
were dispatched.

CONCLUSIONS
Commit to credible increases of the carbon tax for the medium term (2030). Because electricity
generation plants are long-lived, investors need to have certainty about the future levels of the carbon
tax. Since the analysis makes it clear that US$5/tCO2 carbon tax not applied to the VC will not
achieve the implied emission reduction targets for the electricity sector, it is likely that investors will
be expecting future tax increases.
Study increasing carbon tax paths for the medium to long term (2030-2050) that are consistent with
GHG mitigation targets. As part of the development of the medium term carbon tax path, the
Government should also undertake further studies to identify the tax path that would be consistent
with its emission objectives over the medium to long term. In particular, it will be important to
consider an investment plan that does not lock Chile into a high emission plateau due to investment in
new gas capacity.
Include the application of the carbon tax into the variable cost. The current approach is not consistent
with the sustainability of the electricity sector. In particular, it may have the unintended consequence
of making renewable investments less viable. While the Government may respond to that unintended
effect by raising capacity charges regulated by CNE, it is likely to be more efficient to achieve the
same effect by allowing the marginal cost of electricity to rise. This is because including tax in the
variable cost will lead to more efficient utilization of the existing fleet. It will also avoid sending
perverse signals to new gas investments—and also allow reducing emissions significantly more than
not including it at the same level of tax. In other words, the current approach misses out on the
opportunity of further emissions reductions that could be achieved over the longer term by using the
existing fleet more efficiently and reducing risk in locking in excessive dependence on new gas plants.
Finally, the Government and CNE should review the capacity payment for regulated customers, to
ensure that a combination of the revenue from the average contract prices at the nodes and capacity
payment is sufficient to cover the all-in levelized cost of renewable generation. It is possible that in the
future technologies other than diesel be considered the firming options—namely energy storage.

9
BIBLIOGRAPHY

[1] Ley General de Servicios Eléctricos

[2] Fiscalía Nacional Económica “La Competencia en el Mercado Mayorista en Chile”. Available
at http://www.fne.gob.cl/wp-content/uploads/2014/01/informe_final_FNE_Enero13_2014.pdf
[3] ‘INDC para el Acuerdo Climático de Paris’ Available at
http://www4.unfccc.int/submissions/INDC/Published%20Documents/Chile/1/Chile%20INDC
%20FINAL.pdf.
[4] Data from the ‘Primer Informe Bienal’ and from the website of the Sistema Nacional de
Inventarios de Gases de Efecto Invernadero (SEN Chile). Data on emissions in 2007 available
at http://www.snichile.cl/consulta.
[5] KAS Ingeniería, “Efectos en los precios de la electricidad e impactos ambientales asociados a
políticas energéticas,” 2011.
[6] Interstate Natural Gas Association of America, “Coal-Fired Electric Generation Unit
Retirement Analysis,” 2010.
[7] Centro de Excelencia en Geotermia de los Andes, “CEGA,” Available at:
http://www.cega.ing.uchile.cl/cega/index.php/es/informacion-de-interes-/geotermia-en-chile
(Last accessed: 22 January 2016).
[8] “Development of Solar Thermal Energy (STE) which will reach 4GW by 2040”. International
Energy Agency, “Technology Roadmap,” 2014. Available at:
http://www.iea.org/publications/freepublications/publication/technologyroadmapsolarthermale
lectricity_2014edition.pdf.
[9] Centro Para la Innovación y Fomento de las Energías Sustentables, “Plan 100 Minihidros”.
2015.
Available at: http://cifes.gob.cl/wp-content/uploads/2015/12/Plan-100-Minihidros.pdf.
[10] Comisión Nacional de Energía, “Informe Técnico de Precio de Nudo (octubre 2015),” 2015.
[11] PV Magazine (8 June 2015). South Africa: SunEdison wins solar contracts worth 371 MW,
juwi working on 86 MW project
http://www.pv-magazine.com/news/details/beitrag/south-africa--sunedison-wins-solar-
contracts-worth-371-mw--juwi-working-on-86-mw-project_100019747/#axzz3rP6MdG6w
[12] Reneweconomy (5 November 2015). SunEdison makes record-setting solar bid in India, to
delight of energy minister.
http://reneweconomy.com.au/2015/sunedison-makes-record-setting-solar-bid-in-india-to-
delight-of-energy-minister-84328
[13] CleanTechnica (29 November 2014). Dubai shatters solar price record worldwide.
http://cleantechnica.com/2014/11/29/dubai-shatters-solar-tariff-records-worldwide-lowest-
ever/
[14] Peter Fairley (27 October 2015). Abengoa, Calama thermal solar plant (2014 tender) approved
at 0.115 USc/kWh. IEEE Spectrum, by. Chile´s Hybrid PV-Solar Thermal Power Stations.
http://spectrum.ieee.org/energywise/green-tech/solar/.VjvRmXtKJuE.email

10