The second round effects of carbon taxes on power project finance

Description
The most problematic area of any carbon policy debate is the treatment of incumbent CO2
intensive coal-fired electricity generators. Policy applied to the electricity sector is rarely well guided
by macroeconomic theory and modeling alone, especially in the case of carbon where the impacts are
concentrated, involve a small number of firms and an essential service. The purpose of this paper is to
examine the consequences of poor climate change policy development on the efficiency of capital
markets within the Australian electricity sector.

Journal of Financial Economic Policy
The second-round effects of carbon taxes on power project finance
Paul Simshauser Tim Nelson
Article information:
To cite this document:
Paul Simshauser Tim Nelson, (2012),"The second-round effects of carbon taxes on power project finance",
J ournal of Financial Economic Policy, Vol. 4 Iss 2 pp. 104 - 127
Permanent link to this document:http://dx.doi.org/10.1108/17576381211228970
Downloaded on: 24 January 2016, At: 21:44 (PT)
References: this document contains references to 28 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 368 times since 2012*
Users who downloaded this article also downloaded:
Linyue Li, Nan Zhang, Thomas D. Willett, (2012),"Measuring macroeconomic and financial market
interdependence: a critical survey", J ournal of Financial Economic Policy, Vol. 4 Iss 2 pp. 128-145 http://
dx.doi.org/10.1108/17576381211228989
Roseline Nyakerario Misati, Esman Morekwa Nyamongo, Lucas Kamau Njoroge, Sheila Kaminchia,
(2012),"Feasibility of inflation targeting in an emerging market: evidence from Kenya", J ournal of Financial
Economic Policy, Vol. 4 Iss 2 pp. 146-159http://dx.doi.org/10.1108/17576381211228998
Paul Bennett, (2011),"The (revised) future of financial markets", J ournal of Financial Economic Policy, Vol. 3
Iss 2 pp. 109-122http://dx.doi.org/10.1108/17576381111133598
Access to this document was granted through an Emerald subscription provided by emerald-srm:115632 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
Authors service information about how to choose which publication to write for and submission guidelines
are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as
providing an extensive range of online products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee
on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
preservation.
*Related content and download information correct at time of download.
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
4
4

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
The second-round effects
of carbon taxes on power project
?nance
Paul Simshauser
Grif?th University, Brisbane, Australia and
AGL Energy, North Sydney,
Australia, and
Tim Nelson
AGL Energy, North Sydney, Australia and
University of New England, Madgwick, Australia
Abstract
Purpose – The most problematic area of any carbon policy debate is the treatment of incumbent CO2
intensive coal-?red electricity generators. Policy applied to the electricity sector is rarely well guided
by macroeconomic theory and modeling alone, especially in the case of carbon where the impacts are
concentrated, involve a small number of ?rms and an essential service. The purpose of this paper is to
examine the consequences of poor climate change policy development on the ef?ciency of capital
markets within the Australian electricity sector.
Design/methodology/approach – The authors conducted a survey of Australian project ?nance
professionals to determine the risk pro?les to be applied to the electricity sector, in the event a
poorly-designed climate change policy is adopted.
Findings – The Australian case study ?nds that if zero compensation results in the ?nancial distress
of project ?nanced coal generators, ?nance costs for all plant rises, including new gas and renewables,
leading to unnecessary increases in electricity prices. Accordingly, an unambiguous case for providing
structural adjustment assistance to coal generators exists on the grounds of economic ef?ciency.
Originality/value – Accordingly, the paper shows that an unambiguous case for providing
structural adjustment assistance to coal generators exists, on the grounds of economic ef?ciency.
Keywords Australia, Electric power generation, Electricity industry, Coal technology, Carbon tax,
Climate change, Government policy, Project ?nance, Electricity prices
Paper type Research paper
1. Introduction
One of the most intriguing issues associated with the global ?nancial crisis was the
complete failure of the roughly 13,000 economists in the US to predict it (Samuelson,
2009). At the time, most economists were focused on in?ationary pressures and an
overheating global economy. The crisis originated in the capital markets, but ?nancial
economics occupies, at best, a peripheral position in mainstream economics. Financial
economics also occupies, at best, a peripheral position inanycarbonpolicy debate. But as
this article subsequently reveals, it must be elevated to centre-stage for similar reasons.
Pricingcarbonis designedto hastenthe exit of coal plant frompower systems. Despite
the pointed and business-disruptive nature of the policy intent, debate on carbon pricing
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1757-6385.htm
JEL classi?cation – D61, L94, L11, Q40, Q, Q4, L1, L9, D6
JFEP
4,2
104
Journal of Financial Economic Policy
Vol. 4 No. 2, 2012
pp. 104-127
qEmerald Group Publishing Limited
1757-6385
DOI 10.1108/17576381211228970
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
4
4

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
and coal-?red generation in Australia does not centre on whether such a framework
should be implemented. All sides of the debate, including coal-?red asset owners, agree it
should. The issue comes down to howplant might exit; withor without compensation. At
one extreme is the asbestos argument; that incumbent generators should not receive
assistance. At the other is the expropriation argument; that full compensation for asset
loss is warranted. Policy is rarely well guided by emotive arguments. Nor will carbon
policy applied to the electricity sector be well guided by macroeconomic theories and
modeling alone given that wealth impacts are concentrated, non-trivial, involve a small
number of ?rms and an essential service. Financial economics, in this instance, has a
central role to play in macroeconomic policy formulation.
To illustrate why, we focus on Australia’s 7,000 MW brown coal power station ?eet.
Brown coal generators have above grid-average CO
2
intensity coef?cients and are
therefore more likely to experience ?nancial distress in the short run. Additionally,
they are privately owned and mostly ?nanced by non-recourse project debt[1].
Modeling results later in this article reveal adverse second-round effects for power
project ?nance (PF) under a “zero compensation scenario” with brown coal plant
?nancial distress. We therefore ?nd an unambiguous case for providing structural
adjustment assistance on the grounds of economic ef?ciency.
This article is structured as follows: Section 2 reviews the theory of structural
adjustment. Section 3 analyses the impact of carbon prices on generator cost structures.
Section 4 analyses capital ?ows and reviews our survey results on PF in Australia’s
National Electricity Market. Section 5 presents our PF modeling results. In Section 6, the
entry cost estimates from Section 5 are translated to our dynamic partial equilibrium
model to produce power system economic ef?ciency losses. Policy recommendations
follow.
2. On the theory of structural adjustment assistance
In current carbon policy debates, advisors and policymakers with a macroeconomic bias
seemto favour a cut-and-run approach, underscored by limited or even zero compensation
to incumbent coal-?red generators. This re?ects a Washington Consensus approach to
reform[2]. This approach observes that carbonpricinghas been well telegraphedandasset
owners have had years to prepare, and government should not provide taxpayer funded
protection to sunset industries or those who produce negative externalities (e.g. asbestos,
tobacco, coal-?red power). Where assistance has historically been provided to generators
as a transitional measure such as in the EUETS, supranormal pro?ts were extracted from
an overgenerous free permit allocation. Moreover, from a transitional perspective,
neo-classical economic theoryandmodelinghas longbeencomfortable withthe notionthat
short run capital losses re?ect the workings of an ef?cient market and new owners will
acquire distressed assets at more appropriate (post-policy) values without any disruption
to supply. History abounds with examples; producers forecast dire consequences when
Australia’s 10 percent goods and services tax (GST) was introduced in 2000. Yet no
assistance was offered, and the economy adjusted without incident.
Simshauser (2008) noted that economists must commence the analysis of any major
reformwiththe notionthat there is nobasis for compensationmechanisms tooffset direct
or indirect losses associated with a policy that is designed to drive economic ef?ciency. If
it were not for this default approach, governments would be unable to function properly
as Pasour (1973), Neary (1982), Johnson (1994), Argy (1999), and many others have noted.
Effects of
carbon taxes
105
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
4
4

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
It is simply impractical to assess economy-wide losses for all policy changes. Besides
which, ef?ciency gains and losses fromreform programs tend to even-out over the long
run, with society considerably better off in the end.
Moreover, in many cases the delivery of assistance would impair the economic
ef?ciency that a policy measure is trying to drive in the ?rst place. For example,
providing production subsidies (rather than structural adjustment assistance) to coal
generators whilst introducing a carbon tax would clearly be a destructive log-rolling
policy[3]; the tax is designed to drive coal generators out of business while production
subsidies are designed to protect ?rms and keep them in business. Furthermore, if
every change included adjustment programs, the outcome would more than likely lead
to moral hazard, whereby investors believe their future actions are protected against
policy change through government intervention. Accordingly, the notion of zero
compensation may have solid foundations in theory and practice.
But for markets to adequately solve for large shocks, all the conditions and
assumptions of economic theory and models must be present. Stiglitz (2002) observed
that one of the great achievements of modern economics has been to demonstrate how
rarely this occurs in practice. To that end, there are clear conditions in economic theory
and in practice where structural adjustment assistance is desirable on the grounds of
economic ef?ciency (Argy, 1999). If a given reform is likely to lead to a material
misallocation of resources, then there is a case for further analysis and intervention. In
most western economies, industries tend to qualify for structural adjustment
assistance where reform shocks are:
.
large;
.
policy driven events;
.
breach long standing expectations; and
.
are likely to produce highly uneven or magni?ed losses in discrete industrial
segments (Argy, 1999).
Given the theory on structural adjustment, policymakers with an energy or ?nancial
economics bias tend to baulk at a Washington Consensus approach to the application of
carbon policy. The 2008 global ?nancial crisis aptly demonstrated that great care must
be taken when guiding policy exclusively via macroeconomic theory and modeling.
Stiglitz (2002) provides a longlist of reformpolicy failures which can be traced back to an
overreliance on macroeconomic constructs. Requisite caution is necessarily heightened
when wealth impacts of a policy reformare concentrated, large, involve a small number
of ?rms with large productive capacity, and the reformtarget is an essential service like
electricity supply. This latter point is critical and distinguishes coal generation from
other products with negative externalities such as asbestos, where substitutes are
immediately available at equivalent cost.
In computable general equilibrium(CGE) modeling, a staple input to macroeconomic
decision making, ?rms and production processes within an industry segment are
essentially passive variables. The equivalent of a 2,000 MW base load power station
could theoretically produce 1 MW in a year in a CGE model, despite being technically
and economically intractable in the real world[4]. On the other hand, the primary tools
used in ?nancial and microeconomic analysis, dynamic multi-period partial equilibrium
models and ?nancial models at the project level, by necessity deal with a level of detail
JFEP
4,2
106
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
4
4

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
entirely unfamiliar to CGE modeling, albeit with an intensely narrow focus. Electricity
sector models typically involve half-hourly resolution of resource allocation across
multiple years and crucially, capture plant-speci?c constraints, transmission
congestion, and regional demand, thus attempting to mimic the rich dynamics of high
impact events on demand, production, price, and in this case, NEM emissions, energy
security, and systemic security.
Coal power plants tend to be very large relative to other plant types and supply a
dominant component of aggregate demand. As Table I later illustrates, about 81 percent
of Australia’s power comes from just 31 coal-?red generation plants. And so electricity
sector modeling tends to reinforce the view that systemic or physical disruption events
arising from policy-induced ?nancial distress of coal plants is more than a theoretical
possibility. Conversely, a macroeconomist would argue that the withdrawal of supply
will be divisible, will raise price, and that the market will quicklyequilibrate at a newand
higher level. A Washington Consensus approach to such matters would be that new
owners would acquire distressed brown coal assets, thus averting collapse and in the
process reset the cost-base of the plant in the post-policy environment, perhaps.
But in this instance it is obvious that unique conditions exist; carbon policy is
designed to drive coal generators out of business. And so ?nancial economic analysis
will reveal that the ?eld of buyers for a terminal coal power station with negative
operating margins and looming, non-trivial closure costs associated with asbestos
removal and mine rehabilitation must surely be zero. Additionally, coal plant cannot be
operated economically on an intermittent basis when it has been purposefully designed,
engineered, manned and more importantly, ?nanced for base load duties.
A systemic shock in the NEM is plausible if a large CO
2
intensive coal facility not
provided with structural adjustment assistance collapsed unexpectedly under the
weight of a carbon price. The reason for this is straightforward enough; forward
electricity hedge contracts formpart of the unsecured market for derivative instruments,
and administrators of moribund plant have broad powers to cancel committed
hedge contracts. This is not contentious. Administrators would only cancel forward
hedge contracts at the very point in time that they are most needed by demand-side
participants; that is, when they are deeply out-of-the-money. Further contagion in
the NEMcould result, causing the ?nancial distress of other energy businesses that were
otherwise stable. A more sobering thought is that given the nature of deregulated
wholesale energy markets and the presence of retail price regulation, no energy retailer
in the NEMis too big to fail on ?nancial grounds under shock conditions, especially with
a market price cap that is 200 times average price[5].
This is not without precedent. Energy and ?nancial economists therefore harbor
reservations about the prospect of market stability under sustained structural shocks to
power systems, not because they fear the wholesale market will not respond correctly,
but because it will respond correctly. Given the regulation of retail electricity tariffs,
there is an imperfect transmission of price movements from wholesale to retail
markets[6]. This issue is material. Price-cap regulation dictates electricity tariff caps to
more than 72 percent of the 8.9 million households in the NEM. And the deregulated
tariffs of the remaining 28 percent of households are constrained to six-monthly
movements. The last time a deregulated wholesale energy market snapped where retail
prices were regulated, two of the largest investor-owned utilities in the US were
bankrupt within six months of the initial shock event ( Joskow, 2001; Bushnell, 2004).
Effects of
carbon taxes
107
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
4
4

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
T
e
c
h
n
o
l
o
g
y
S
i
t
e
s
(
N
o
.
)
C
a
p
a
c
i
t
y
(
M
W
)
S
h
a
r
e
(
%
)
O
u
t
p
u
t
(
G
W
h
)
S
h
a
r
e
(
%
)
C
O
2
i
n
t
e
n
s
i
t
y
(
t
/
M
W
h
)
R
e
p
l
a
c
e
m
e
n
t
(
$
M
)
F
l
e
e
t
v
a
l
u
e
(
$
M
)
F
l
e
e
t
a
g
e
(
Y
e
a
r
s
)
B
r
o
w
n
c
o
a
l
8
7
,
3
3
5
1
3
.
8
5
7
,
0
6
3
2
4
.
8
1
.
1
0
-
1
.
5
5
2
2
,
0
0
5
7
,
8
2
9
3
2
.
2
B
l
a
c
k
c
o
a
l
2
3
2
2
,
2
8
0
4
1
.
9
1
2
9
,
4
0
1
5
6
.
3
0
.
8
0
-
0
.
9
7
4
9
,
0
1
6
2
4
,
1
2
8
2
5
.
4
G
a
s
7
4
1
5
,
2
8
5
2
8
.
7
2
8
,
3
2
2
1
2
.
3
0
.
3
8
-
0
.
7
0
1
8
,
3
4
2
9
,
8
7
2
1
3
.
9
H
y
d
r
o
a
n
d
w
i
n
d
5
0
8
,
3
1
6
1
5
.
6
1
4
,
9
7
0
6
.
5
n
/
a
2
4
,
5
9
4
1
3
,
0
1
6
3
6
.
8
T
o
t
a
l
1
5
5
5
3
,
2
1
6
1
0
0
.
0
2
2
9
,
7
5
6
1
0
0
.
0
0
.
9
2
1
1
3
,
9
5
7
5
4
,
8
4
5
2
4
.
8
N
o
t
e
:
G
r
i
d
-
c
o
n
n
e
c
t
e
d
,
e
x
c
l
u
d
i
n
g
e
m
b
e
d
d
e
d
p
l
a
n
t
S
o
u
r
c
e
:
D
a
t
a
a
d
a
p
t
e
d
f
r
o
m
E
n
e
r
g
y
S
u
p
p
l
y
A
s
s
o
c
i
a
t
i
o
n
o
f
A
u
s
t
r
a
l
i
a
(
2
0
1
0
)
Table I.
Australian power
station ?eet in 2010
JFEP
4,2
108
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
4
4

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
Competitive energy markets were designed to drive productive, allocative, and
dynamic ef?ciency. The NEM has been enormously successful at driving all three
(Simshauser, 2005). But the dynamic ef?ciency objective function was largely
one-directional; providing appropriate signals for new entry and consequences for
excess entry. Energy markets are not typically designed, or well equipped, to deal with
policy-induced lumpy plant exit. And nor should they be since such events must be rare
in practice. But if plant exit is induced by tangential carbon policy, and exit is not
carefully orchestrated, systemic or physical disruption is plausible.
But in our view, these short-run impacts pale into insigni?cance by comparison to the
long run consequences arising fromthe capital markets, which is the prime focus of this
article. Power generation is the world’s most capital-intensive industrial activity.
In Australia, this activity occurs in an economy with a severe structural reliance on
foreign capital (Simshauser, 2010). Consequently, policy-induced disruption to power
generation investments could have adverse impacts on capital market participation
rates, costs of capital, and capital in?ows to the industry. In the balance of this article, we
quantify ef?ciency losses in capital markets relating to remaining and future generating
equipment under conditions of a WashingtonConsensus, free market approachinvolving
zero compensation to brown coal generators.
3. The impact of carbon prices on power station cost structures
Australia’s aggregate generating capacity is 53,216 MW as Table I notes. The ?eet
produces 229,756 GWh and emits about 200 million tonnes per annumof CO
2
equivalent.
Table I distinguishes between brown coal, black coal, gas, and renewables (i.e. hydro
and wind). There are 155 sites of which eight will be intensely affected in the short-run
from carbon pricing. There is 7,335 MW (14.2 percent) of brown coal plant which
produces a quarter of Australia’s aggregate electricity output with emissions intensities
up to 1.55 t/MWh. The average age of the power station ?eet is 24.8 years, with brown
coal averaging 32.2 years. Our rule-of-thumb valuation estimate[7] of the brown coal
?eet is $7.8 billion. Importantly, existing coal-?red plants in the NEM are thought to
have signi?cant remaining technical useful lives. Outhred (2011) estimated this to be in
excess of 20 years, a number few would disagree with.
This data highlights that the number of intensely affected sites in the short run is
minimal, and the value of those affected is small in the context of the aggregate
generation portfolio. The value is also small by comparison to the roughly $6 billion
“annual take” in carbon taxes that will accrue from the power sector at $30/t.
Applying a price to carbon is designed to shift the industry cost structure, and the
most adversely affected will be the brown coal ?eet due to their especially high CO
2
intensity. To determine the degree of asset value loss for a discrete generator, the
emission intensity of the plant must be compared to the rate of carbon pass-through in
the wholesale market. Where a generator’s emission intensity is greater than the rate of
the whole-of-market pass-through, it will incur carbon costs that are greater than
recoverable through the market. It is in these circumstances that a generator is likely to
experience asset value loss through reduced operating margins, lower volumes and a
truncated economic life.
Intheshort-runbeforesigni?cant substitutionof capital canoccur (i.e. givendevelopment
lags of ?ve years), the rate of carbon pass-through in the wholesale market is likely to re?ect
the ?eet average CO
2
intensity of about 0.9 t/MWh (Nelson et al., 2011). In the long-run
Effects of
carbon taxes
109
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
4
4

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
(i.e. beyond ?ve years), the pass-through rate is likely to decay fromthe grid average as new
lower-intensity generation becomes the marginal generator for non-trivial price-setting
periods each year.
The shortening of a generator’s economic life is therefore a function of two variables:
the spread between its individual emissions intensity and the market intensity (reduced
operating margins); and the rate at which the average intensity declines. Policymakers
must establish whether they should provide structural adjustment assistance to debt
and equity capital investors to deal with reduced operating margins, lower volumes and
truncated economic life, and subject to the geographic location of substitution outcomes,
displaced workers. The short run objective of any structural adjustment assistance
therefore needs to enhance the predictability of capacity exit to avoid disruption events;
and as modeling results later in this article reveal, the long run objective should be to
short-circuit second-round effects in the capital markets.
From an entry cost perspective, levelised cost modeling in Simshauser (2011)
highlighted the comparative effects that carbon pricing would have on new coal plant
relative to rival technologies. Note in Figure 1 that the relative carbon tax accruing to
brown and black coal plant is materially higher than combined cycle gas turbine
(CCGT) plant. Consequently, on generalized long run marginal cost (LRMC) modeling
at least, new plant using existing coal technologies is uneconomic.
The analysis in Figure 1 deals with new investment. More important to the carbon
debate is the impact on incumbent plant. Figures 2 and 3 show generalized marginal
running costs of the NEM’s roughly 38,000 MW thermal ?eet before, and after,
a $30/t carbon price. In Figure 2, where carbon prices are excluded, the brown coal ?eet
sits at the bottomof the aggregate supply function with marginal running costs of about
$5/MWh. In contrast, the marginal running cost of base load gas plant (at $4.50/GJ) is
over $30/MWh.
Figure 3 shows the change in the plant pecking order or merit order once carbon
taxes are introduced. The arrows identify the location of brown coal generators in the
newly formed aggregate supply function. Note the CCGT base load plant is creeping up
Figure 1.
Generalised LRMC of
utilities-scale energy
technologies
Source: Simshauser (2011)
0.00
0.20
0.40
0.60
0.80
1.00
1.20
$0
$20
$40
$60
$80
$100
$120
$140
$160
$180
C
O
2

I
n
t
e
n
s
i
t
y

(
t
/
M
W
h
)
U
n
i
t

c
o
s
t

(
$
/
M
W
h
)
Carbon at $30/t
LRMC
CO2 Intensity (RHS Axis)
B
l
a
c
k

c
o
a
l
B
r
o
w
n

c
o
a
l
C
C
G
T
O
C
G
T
M
i
n
i

H
y
d
r
o
N
u
c
l
e
a
r
W
i
n
d
B
i
o
m
a
s
s
H
F
R

G
e
o
t
h
e
r
m
a
l
S
o
l
a
r

P
V
S
o
l
a
r

T
h
e
r
m
a
l
W
a
v
e
$200
JFEP
4,2
110
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
4
4

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
the merit order, dislodging brown coal plant and any black coal plant exposed to
export coal prices. CCGT’s will also become more competitive as the price of carbon
rises due to their low carbon coef?cient.
A characteristic evident from Figures 2 and 3 is the relative size of the incidence of a
carbon tax. On a production-weighted basis, a $30/t carbon tax increases the marginal
running cost of brown coal plant by 10.2 times. For black coal plant, marginal costs
Figure 2.
Generalised marginal
running cost of the NEM’s
thermal ?eet ($/MWh)
$0.00
$10.00
$20.00
$30.00
$40.00
$50.00
$60.00
$70.00
M
a
r
g
i
n

R
u
n
n
i
n
g

C
o
s
t

(
$
/
M
W
h
)
Power Station Capacity (MW)
Vic Brown Coal
Qld Low Black Coal
SA Low Brown Coal
Qld Mid Black Coal
NSW Low Black Coal
SA High Brown Coal
NSW Export Black Coal
CCGT Base Gas
Peaking Gas
Vic Brown Coal
CCGT
Base Gas
0
2
,0
0
0
4
,0
0
0
6
,0
0
0
8
,0
0
0
1
2
,0
0
0
1
0
,0
0
0
1
4
,0
0
0
1
6
,0
0
0
1
8
,0
0
0
2
0
,0
0
0
2
2
,0
0
0
2
4
,0
0
0
2
6
,0
0
0
2
8
,0
0
0
3
0
,0
0
0
3
2
,0
0
0
3
4
,0
0
0
3
6
,0
0
0
3
8
,0
0
0
Figure 3.
Generalised marginal
running cost of the NEM’s
thermal ?eet at $30/t CO
2
$0.00
$10.00
$20.00
$30.00
$40.00
$50.00
$60.00
$70.00
Power Station Capacity (MW)
Qld Low Black Coal
Qld Mid Black Coal
NSW Low Black Coal
Loy Yang A&B
CCGT Base Gas
SA Low Brown Coal
Yallourn
Hazelwood
NSW Export Black Coal
Other Vic Brown Coal
SA High Brown Coal
Peaking Gas
CCGT
Base Gas
Brown coal
generators
M
a
r
g
i
n

R
u
n
n
i
n
g

C
o
s
t

(
$
/
M
W
h
)
0
2
,0
0
0
4
,0
0
0
6
,0
0
0
8
,0
0
0
1
2
,0
0
0
1
0
,0
0
0
1
4
,0
0
0
1
6
,0
0
0
1
8
,0
0
0
2
0
,0
0
0
2
2
,0
0
0
2
4
,0
0
0
2
6
,0
0
0
2
8
,0
0
0
3
0
,0
0
0
3
2
,0
0
0
3
4
,0
0
0
3
6
,0
0
0
3
8
,0
0
0
Effects of
carbon taxes
111
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
4
4

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
increase by 3.0 times, and 1.4 times for CCGT plant. These marginal running cost
multipliers provide some insight into the impact on brown coal plant, and why gas
plant has such a material advantage as carbon prices rise. The marginal running cost
multipliers also highlight why other tax reforms such as the introduction of the
GSTshould not provide policymakers with comfort under zero compensation. The GST
resulted in input costs of goods and services increasing by a multiplier of 1.1 times, a
trivial amount. The carbon impact on brown coal plants is 9.1 times greater than the
GST. Comparisons of carbon policy with previous broader economic reforms in the
context of deciding sunset industry assistance would therefore be misguided at best
and dangerous at worst due to the impacts being an order of magnitude different.
Over the past ten years, there have been a number of studies completedby businesses,
governments and industry associations on the impacts of carbon pricing on coal-?red
power stationasset values. Losses for NEMgenerators have beenmodeledat $11.0billion
by ACILTasman (2011); $16.7 billion by ACILTasman (2008)[8]; $17.5 billion by ROAM
(2008); and $0.1 billion by McLennan Magasanik Associates (2008). The privately owned
brown coal generators have been forecast to experience losses of $7.1 billion, $7.9 billion
and $2.3 billion in the respective 2008 studies[9]. Differences between studies will
invariably be driven by assumptions around market behavior, structure, and the carbon
pass-through rate. We are unaware of any modeling literature in Australia that indicates
brown coal generators will not experience disruptive losses. Based upon these results, it
is not surprising that the structural adjustment assistance proposed for the 2008
Australian ETS legislation (which was defeated in the Senate) involved $7.3 billion in
nominal terms (DCC, 2009). Asset loss will be material due to declining operating
margins, loss of market share and a truncated economic life. None of this is contentious.
It underscores the whole point of carbon pricing.
4. Capital ?ows and project ?nance survey results
An important characteristic of the Australian economy is a severe structural reliance
on foreign capital for investment purposes. This is due to Australia’s low household
saving rates, an above-average proportion of capital-intensive industries such as
energy, metals manufacturing, utilities and mining, along with a small population of
22.5 million people spread over a large area, which raises requirements for transport
and distribution infrastructure (Simshauser, 2010). The out-working of this is best
illustrated by net capital ?ows into the Australian economy, in Figure 4. Note that from
2001 to 2009, about $1 trillion of net foreign capital ?owed in-bound. Since 2006,
foreign capital in?ows have ?nanced about 58 percent of Australia’s new ?xed assets.
Since the utilities sector is the world’s third largest borrower of debt behind
governments and banks, respectively, capital ?ows should be of considerable
importance to carbon policymakers (Simshauser, 2010).
Foreign capital ?ows provide important background context to the debate on
structural adjustment assistance to brown coal generators. If a concentrated portfolio
of assets in the capital-intensive power sector becomes systematically distressed as a
result of policy change, no matter how novel that policy might be, it is unlikely that
?nancing conditions will remain constant thereafter. Table II outlines debt facilities on
Australia’s four largest brown coal-?red power stations.
Syndicateddebt at the four power stations totals $5.212 billion, whichis spread across
36 institutions. About 14 banks are currently active inoriginating debt facilities, withthe
JFEP
4,2
112
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
4
4

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
Banks Loy Yang A Loy Yang B Hazelwood Yallourn Total
ANZ 113 123 64 75 375
NAB 183 123 90 396
BTMU 126 123 47 29 325
Westpac 82 45 120 247
RBS 162 103 120 385
BNP Paribas 81 123 41 71 316
SMBC 126 98 36 260
CBA 221 36 257
Societe Generale 125 64 189
DBJ 150 150
Credit Agricole Indosuez 90 52 142
West LB 71 62 45 178
Mizuho 126 30 156
BOS International 70 52 122
China Construction Bank 108 108
Dexia 70 31 101
STB 100 100
CIBC 98 98
Deutsche Bank 98 98
MUFJ 98 98
United Overseas Bank 71 22 93
KBC Bank 81 81
UniCredit Group 70 70
Aozora 50 50
Other foreign banks (12) 108 0 161 548 817
Total 2,163 1,105 744 1,200 5,212
Source: Data adapted from Reuters, Macquarie Capital, TRUenergy
Table II.
Latrobe valley syndicated
debt (A$ million)
Figure 4.
Net ?ows of foreign
capital to Australia
24%
34%
36%
35%
15%
67%
60%
55% 55%
0%
20%
40%
60%
80%
100%
$–
$25
$50
$75
$100
$125
$150
$175
$200
2001 2002 2003 2004 2005 2006 2007 2008 2009
F
o
r
e
i
g
n

C
a
p
i
t
a
l

(
%
)
F
o
r
e
i
g
n

C
a
p
i
t
a
l

(
$

B
i
l
l
i
o
n
)
Foreign debt
Foreign equity investment
Foreign capital (% of total investment)
From 2001 to 2009, $978 billion in
foreign capital, 74% debt capital.
Source: Data adapted from ABS
Effects of
carbon taxes
113
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
4
4

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
involvement of others by way of syndication. Importantly, if one or more of the brown
coal generators defaults on debt facilities under conditions of zero compensation, capital
market perceptions of the regulatory environment within the Australian electricity
industry will deteriorate further[10]. We therefore opted to test the following proposition:
P1. That the combination of a default on a project debt facility and zero
compensation at a brown coal power station would result in a step-change in
debt ?nance costs for all plant, including future gas and renewable plant.
To test our P1, we issued a survey to Australia’s top 30 PF bankers who represent the
14 active foreign and domestic banks in originating power project ?nancings. The bank
response rate to the 22 question survey was 65 percent. The survey had two primary
sections, and purposes. The ?rst sought bankers’ views on historic and current power
project ?nancings in Australia. Questions focused on spreads for term facilities,
maximumtenors and gearing levels achievable, the number of mandated lead arranging
(MLA) banks required to close a $500 million facility for a gas-?red power station, the
number of banks likely to be included in any syndication, and the number of active
banks in the broader market for a power PF. We selected three points in time; 2006 to
represent the period before the global ?nancial crisis, 2008 as representative of “during
the crisis”, and 2011 given “carbon policy uncertainty”. Summary results from the ?rst
segment of the survey are contained in Table III. Observed spreads for BBB credit-rated
three-year Australian corporate bond issues are also included to provide a benchmark to
compare the movement in PF spreads.
Conditions to minimize the cost of power projects were ideal in 2006, with spreads at
100-120 bps over swap, 12-year tenors (which reduces re?nancing risk, thereby
facilitating) gearing levels of 65 percent þ, and low transaction costs with only three
MLA banks required to close facilities. If 2006 was the low water mark, 2008 must
surely represent the high water mark. Spreads increased nearly four-fold as global
liquidity evaporated in line with the global ?nancial crisis. Figure 5 shows the rapid
deceleration of global liquidity in the market for syndicated debt; it notes debt issuance
across all industries during 2007 was fully US$3.2 trillion, but shrunk three-fold to just
US$1.05 trillion after the Lehman collapse in 2008 (Simshauser, 2010).
In the case of Australian power PF, note from Table III that tenors reduced
from 12 years to just three years in 2008, while the number of MLA’s required to
2006 2008 2011
PF spreads 100-120 bps 400-450 bps 350-400 bps
Spread movement Stable Up 3.8 £ Down 11%
Max tenor 12 years 3 years 7 years
Max gearing 65%þ Approx 55% Approx 60%
MLA banks 3 or less 7-8 banks 7-8 banks
Syndication banks 3 or less Club deal 4-8 banks
Active banks 29 11 14
Spread on BBB bonds 85 bps 360 bps 240 bps
Spread movement Stable Up 4.2 £ Down 33%
Source: AGL Energy Ltd, Bond data adapted from CBA
Table III.
Survey results on
perceptions of PF
facilities in 2006,
2008, and 2011
JFEP
4,2
114
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
4
4

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
close a $500 m power PF deal increased substantially from three to seven-eight banks.
None would take syndication risk; all transactions were club deals (i.e. debt provided
on a “take and hold” basis). Finally, the number of active banks has reduced
considerably although a small number have returned since 2008.
The most striking result in Table III is the PF spread movement for power projects
by comparison to corporate bonds. There was a parallel run-up in bond and PF spreads
between 2006 and 2008 with a multiplier of approximately four times, but a sharp
differential in the retreat to 2011. Table III notes that bond spreads have fallen from
360 bps to 240 bps (33 percent) since 2008 whereas power PF spreads have reduced by
just 11 percent.
We tested whether Australian power PF data is out of step with global trends by
analysing a comprehensive listing of global power project ?nancings from their
recorded inception in 1981 through to the time of writing in Q1 2011. This data
represents 3,140 individual transaction facilities across 101 countries, with a total
facility value of A$2.76 trillion (in 2011$). Summary results are provided in Table IV.
Intriguingly, whereas Australian power PF have shortened in tenor from 12 years in
2006 to seven years in 2011, and margins remain elevated at 350-400 bps as noted
Number of
power
station PF’s
(#)
Average
PF facility
tenor
(Years)
Average PF
facility
spread
(%)
Average PF facility
size
(2011$)
(AUD Million)
Global syndicated debt
(2011$)
(AUD Million)
1981-2007 2,028 11.5 143 938.0 1,902,198
2008-2011 1,112 13.2 236 772.3 858,800
Total 3,140 12.1 176 879.3 2,760,997
Source: Data adapted from Reuters
Table IV.
Aggregate global power
PF deals from 1981
to Q1 2011
Figure 5.
Global market for
syndicated debt
–
500,000
1,000,000
1,500,000
2,000,000
2,500,000
3,000,000
3,500,000
2004 2005 2006 2007 2008 2009 2010
S
y
n
d
i
c
a
t
e
d

D
e
b
t

(
U
S
$
M
i
l
l
i
o
n
)
Fourth Quarter
Third Quarter
Second Quarter
First Quarter
CDO
Market
Collapse
Lehman
Collapse
Source: Data adapted from Bloomberg
Effects of
carbon taxes
115
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
4
4

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
in Table III, global PF data in Table IV presents a very different picture. Of the more
than 1,100 transactions completed globally during 2008-2011, average tenors actually
lengthened from 11.5 years to 13.2 years and margins, while elevated, have averaged
236 bps. The explanation for this is straightforward enough; the rest of the world does
not have the same uncertainty over carbon policy that Australia does[11]. One of
Australia’s most respected power project bankers recently noted that “the investment
community correctly attaches a risk premium for the added uncertainty, which
effectively increases the cost of capital to the industry” (Satkunasingam, 2011, p. 1).
Had Australia’s proposed ETS been legislated back in 2009 when ?rst brought before
the Senate, our view is that PF tenors and margins (in Australia) would have been more
likely to gravitate towards global average trends and the trend in corporate bonds,
that is, spreads below 300 bps and tenors of ten þ years.
This leads us to the second segment of our survey, given uncertainty over carbon
policy. Here, bankers’ views were sought on the potential impact of the ?nancial
distress of an existing coal asset as a direct result of carbon pricing, but crucially,
under conditions of zero compensation. Bankers were asked for their views on any
potential “penalty spread” which might apply to all three types of power project
technologies in Australia. Our reasoning here was twofold. Con?icting signals from
political parties results in risk being priced, and bad debts represent business costs
which must be recovered by ?rms to remain pro?table; this includes banks. The results
from the second segment of our survey are illustrated in Table V.
This survey data from Australia’s top utilities project bankers provides us with
unique insights around the extent of potential economic ef?ciency losses that might
accumulate in the electricity sector if a brown coal power station defaulted on debt
facilities under conditions of zero compensation. Clearly, expanding credit spreads will
increase the underlying cost of power generation through higher interest costs,
shortened tenors and consequent lower gearing levels.
Policy makers may well be indifferent about wealth transfers between incumbent
brown coal equity holders and debt holders. But any policymaker would be rattled by
the 2nd line result in Table V relating to new gas plant (þ150 bps), and policy advisors
to the Greens will no doubt be stunned by the 3rd line result relating to newrenewables
(þ100 bps); and given their opposition to structural adjustment assistance to coal
generators, they should be. Even though penalty spreads in Table V clearly decline
with the emissions intensity of the technologies, renewables are the most
capital-intensive power generating technology with low capacity factors and are
therefore hyper-sensitive to changes in the cost of capital. In the event, new renewables
experience the most profound “hit” to generalized LRMC estimates, and as modeling
results in the next section reveal, they are non-trivial in every respect. In short, policy
Technology bp premium
Existing coal 150-200
New gas 100-150
New renewable 50-100
Source: AGL Energy Ltd
Table V.
Survey results on PF
“penalty spreads”
JFEP
4,2
116
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
4
4

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
uncertainty and zero compensation actually does much more damage to the entry
prospects of new renewable plant than it does to existing coal plant or new gas plant.
To make matters worse, if the new entrant cost of gas and renewable plant is
adversely affected, it does not just affect the economics of singular new entrant plant.
Even though only a handful of plants may be directly affected by this change, the
nature of energy markets means that clearing prices must rise to the cost of entry, and
so if an outcome of zero compensation is a “penalty spread” being added to the
underlying cost of capital and gearing levels reduce, then this will increase wholesale
electricity prices, and in turn, retail electricity prices across the entire 8.9 million
households in the NEM; a sobering thought for policymakers of all persuasions.
It stands to reason that margins on all new plant would be elevated if a brown coal
plant collapses due to zero compensation; we considered this to be entirely predictable
and survey results merely con?rm this. Gas and renewable plant are being developed
in response to (appropriate) government policy settings. But if government policy
disrupts historic investments, policy risk will be priced into the generation sector until
such risk is perceived to have diminished to trivial levels. In Australia, the Government
and the Opposition currently support a 5 percent carbon target, but the Greens
(who hold the voting balance of power in the Australian Senate) do not. And while the
Government and Greens support a price on carbon, the Opposition does not. And the
fact that the Opposition has stated it will repeal government carbon pricing plans will
not help to diminish policy risk perceptions in Australia.
5. Scenarios and project ?nance modeling results
To see how changes in parameters affect the underlying cost structures of benchmark
base load CCGT plant, peaking OCGT plant and renewable plant in the NEM, we have
made use of the PF model from Simshauser (2009) and applied current market
conditions from observed capital market data and the survey results from Tables III
and V to quantify economic ef?ciency losses.
The PF model is a dynamic, multi-period cash ?ow simulation model of a power
project in which all the parameters of a given plant and required PF are incorporated to
determine unit costs, debt sizing, and structuring. The model calculates annual energy
production, revenues, ?xed and variable operations and maintenance costs, fuel costs,
capital works, taxation schedules and establishes PF bullet, and semi-permanent
amortising termfacilities. Re?nancing are undertaken while all structured debt facilities
are extinguished within a 25-year aggregate tenor. Under ideal conditions, two debt
facilities are assumed; a seven-year interest-only “bullet” and a 12-year semi-permanent
amortising facility, with relevant interest rate swaps matching the tenors. In the PF
model, we set the forward curve for base load power to the LRMCof the plant in question
which allows us to put the optimal capital structure in place thus minimizing the overall
unit cost of the plant, albeit within the constraints of debt sizing, namely via debt service
cover ratio (DSCR) and loan life cover ratio (LLCR)[12]. The model speci?cations have
been documented extensively in Simshauser (2009) and therefore we do not propose to
reproduce them here.
For the purposes of our analysis, we have modeled three distinct scenarios:
(1) Certainty scenario. Key assumptions in this scenario are that carbon pricing is
implemented, structural adjustment assistance is well designed, bipartisan
support is achieved and so perceptions over policy uncertainty are diminished
Effects of
carbon taxes
117
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
4
4

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
to trivial levels (for example, as in the EU). Spreads on PF facilities therefore
reduce to levels in-line with reductions in the debt capital markets and global
power PF markets[13]. In our certainty scenario, we assume power PF spreads
reduce by 33 percent and narrow in range, from 350-400 bps to 250-275 bps.
We assume 12-year tenors in line with global data, and gearing levels of
67.5 percent in-line with historic conditions.
(2) Uncertainty scenario. Primary assumptions in the uncertainty scenario are that
well-designed structural adjustment assistance forms a central part of carbon
policy but unresolved policy con?icts remain between the Government, the
Opposition, and Greens on the mechanism and targets, respectively. Margins
for new plant re?ect those from our Table III results at 350-400 bps, seven-year
tenors and gearing of 60 percent.
(3) Meltdown scenario. Key assumptions in this scenario are that zero
compensation applies, that coal generators experience ?nancial distress, and
policy con?icts remain between political parties, representing the worst of all
worlds with spread premiums applying at the rates outlined in Table V and
gearing reverting to 55 percent.
Table VI sets out the assumptions that apply for a 1 £ 400 MW CCGT project,
a 3 £ 175 MW OCGT project and a 200 MW wind project in 2011, with changes to
spreads, tenors and gearing as outlined above.
PF model results for CCGT plant under the three scenarios are illustrated in
Figure 6. Primary debt covenants are within tolerance with DSCR at 2.0 times and the
LLCR at 1.8 times in the certainty (i.e. benchmark) scenario. The bar chart illustrates
headline LRMC including carbon at $30/t. The range in headline LRMC spans
$3.70/MWh, from $74.13 to $77.83/MWh. A shift of roughly $4/MWh may not appear
signi?cant, but given NEM demand of about 200 million MWh per annum, it will result
in cost inef?ciencies of about $800 million pa.
In?ation Taxation
CPI (%) 2.50 Tax rate (%) 30.00
Elec price in?ation (%) 2.13 Useful life (Yrs) 30
Plant costs and prices CCGT OCGT Wind
Debt sizing
parameters
Plant size (MW) 400 525 200 DSCR (times) 1.8-2.2
Capital cost ($/kW) 1,500 980 2,500 LLCR (times) 1.8-2.2
Acquisition price ($M) 600 515 500 Gearing (%) 55-67.5
LRMC statistics Lockup (times) 1.35
LRM C in 2011$ ($/MW h) 76.14 14.22 120.39 Default (times) 1.10
Heat rate (kJ/kWh) 7,000 11,400 –
Unit fuel ($/GJ) 4.50 6.00 – Facilities Swap Spread bps
Variable O&M ($/MWh) 3.00 8.00 1.00 5 year tenor 5.76% 250-525
O&M costs ($M pa) 12.4 6.8 8.6 7 year tenor 5.94% 275-550
Capex ($M pa) 3.0 0.2 0.5 12 year tenor 6.11% 275-550
CO
2
footprint (t/MWh) 0.39 0.59 – Re?nancings 6.11% 250-375
Remnant life (Yrs) 40 30 30 Post tax
equity
15%
Table VI.
PF assumptions for
power plant banked
in 2011
JFEP
4,2
118
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
4
4

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
Next, we analysed the economic inef?ciency that would apply to the ?eet of peaking
OCGT plants, which is shown in Figure 7. We measure peaking plant by their
“carrying cost” or “?xed costs” including pro?t recovery. This can be thought of as the
fair value for call options (in $/MW/h) written by new entrant peaking plant with a
$300/MWh strike price. The marginal running cost of such plant, regardless of ?xed
costs, is about $94/MWh including carbon at $30/t. The PF model produces carrying
costs ranging from $13.09/MW/h to $15.29 MW/h under the three scenarios.
Figure 6.
PF model results for
CCGT under certainty,
uncertainty and meltdown
conditions
0.00
10.00
20.00
30.00
40.00
50.00
60.00
70.00
80.00
90.00
U
n
i
t

c
o
s
t

(
$
/
M
W
h
)
Returns to Equity
Taxation
Interest Payments
Debt Redemption
Capex
O&M
Carbon
Fuel
DSCR is 2.0x
LLCR is 1.8x
Debt sizing :
- Gearing: 55%
- 5 & 7 yr tenor
- 525-550bps over
$74.13/MWh $75.95/MWh $77.83/MWh
Debt sizing :
- Gearing: 67.5%
- 5 & 12 yr tenor
- 250-275bps over
2
0
1
1

C
e
r
t
a
i
n
t
y
2
0
1
1

U
n
c
e
r
t
a
i
n
t
y
2
0
1
1

M
e
l
t
d
o
w
n
Debt sizing :
- Gearing: 60%
- 5 & 7 yr tenor
- 375-400bps over
Figure 7.
PF model results for
OCGT under certainty,
uncertainty, and
meltdown conditions
0.00
2.00
4.00
6.00
8.00
10.00
12.00
14.00
16.00
18.00
U
n
i
t

c
o
s
t

(
$
/
M
W
h
)
Returns to Equity
Taxation
Interest Payments
Debt Redemption
Capex
Fixed O&M
DSCR is 2.2x
LLCR is 1.8x
Debt sizing :
- Gearing: 55%
- 5 & 7 yr tenor
- 525-550bps over
Debt sizing :
- Gearing: 65%
- 7 & 12 yr tenor
- 250-275bps over
Debt sizing :
- Gearing: 60%
- 5 & 7 yr tenor
- 375-400bps over
$14.22/MWh $13.09/MWh $15.29/MWh
2
0
1
1

C
e
r
t
a
i
n
t
y
2
0
1
1

U
n
c
e
r
t
a
i
n
t
y
2
0
1
1

M
e
l
t
d
o
w
n
Effects of
carbon taxes
119
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
4
4

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
This $2.20/MW/h differential is also signi?cant because the NEM will carry about
11,000 MW of peaking plant by 2015. Accordingly, potential economic inef?ciencies
arising from our meltdown scenario amount to $212 million pa by comparison to the
certainty scenario.
Our PF model results for renewable plant (i.e. wind) are shown in Figure 8.
In modeling wind projects, we applied substantially higher gearing levels than
Table VI indicates, and this is appropriate. The certainty scenario is geared at 76 percent
with differential gearing equivalent to our thermal plant modeling. Debt sizing
parameters were also different. Whereas thermal plant typically faces DSCR’s of
1.8-2.2 for debt sizing, the capital intensive nature of wind means that the ratios are
relaxed, inthis instance to 1.35-1.45 times. Our sizingendedupwitha DSCRof 1.42 times,
well within the acceptable envelope.
Note that the LRMC of wind in the certainty scenario is $110.94/MWh. The LRMC of
wind in the meltdown scenario is $127.83/MWh and so the difference in the cost of
wind plant between the certainty and meltdown scenario is a surprisingly large
$16.89/MWh, three times higher than the cost impact on CCGT plant due to the capital
intensive nature of rewnewables. The 2015 Renewable Policy target is about
18,000 GWh, and so economic ef?ciency losses arising from the meltdown scenario
would total $232 million pa.
6. Partial equilibrium modeling results
Higher LRMC for new gas and renewable plant have obvious and non-trivial impacts
on forward electricity prices. As the NEM operates under a uniform, ?rst-price,
energy-only gross pool auction design, the value of spot and forward prices must
ultimately rise to the cost of entry prior to new plant being built. To assess the
economic impact of our different scenarios on electricity prices, we assume ?nancing
Figure 8.
PF model results
for wind under
certainty, uncertainty,
and meltdown conditions
0.00
20.00
40.00
60.00
80.00
100.00
120.00
140.00
U
n
i
t

c
o
s
t

(
$
/
M
W
h
)
Returns to Equity
Taxation
Interest Payments
Debt Redemption
Capex
Fixed O&M
VOM
DSCR is 1.42x
LLCR is 1.55x
Debt sizing :
- Gearing: ~68%
- 5 & 7 yr tenor
- 525-550bps over
$127.83/MWh
Debt sizing :
- Gearing: ~76%
- 7 & 12 yr tenor
- 250-275bps over
Debt sizing :
- Gearing: ~71%
- 5 & 7 yr tenor
- 375-400bps over
$110.94/MWh $120.39/MWh
2
0
1
1

C
e
r
t
a
i
n
t
y
2
0
1
1

U
n
c
e
r
t
a
i
n
t
y
2
0
1
1

M
e
l
t
d
o
w
n
JFEP
4,2
120
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
4
4

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
costs calculated in previous sections begin to bind immediately, but are revealed in the
market from 2015 onwards as new plant is commissioned. CCGT forms the dominant
base and intermediate load technology while OCGT undertake peaking duties.
Individual plant costs under the certainty, uncertainty, and meltdown scenarios are
then applied between 2015 and 2020, which by implication represents investment
origination commitments over the period between 2013 and 2017. We consider any
elevation in margins must ultimately be exhausted as regulatory risk perceptions
return to normal; hence our limited ?ve-year period of analysis.
We utilise the optimal plant mix model (OPM model) from Simshauser and Wild
(2009) to undertake the analysis. This partial equilibrium electricity system model
simulates half-hour resolution and assumes perfect competition and essentially free
entry to install any combination of capacity that satis?es differentiable conditions. The
lumpiness of capacity is a constraint; ?rms may chose either 400 MW CCGT base load
plant or clusters of OCGT plant with unit sizes of 175 MW, the latter based on
conventional “E Frame” gas turbine technology. As this model has been thoroughly
documented in Simshauser and Wild (2009), we do not intend to reproduce it here.
A graphical representation of the half-hourly modeling results is shown in Figure 9.
The top graph in Figure 9 is a transformation of PF model results for CCGT and
OCGT plant into marginal running cost curves. The y-axis intercepts represent annual
?xed costs of the plant, and the slope of the curves represent marginal running costs
(i.e. fuel and variable O&M). The cross-over point, at about 25 percent, identi?es the
annual capacity factor at which all gains from investing in lowcapital cost OCGTplant
are exhausted by the higher capital but more operationally ef?cient CCGT plant.
The bottom graph in Figure 9 shows the 17,520 half-hourly electricity load points
for 2015.This equilibrium-demand data has been plotted in descending order to form a
load duration curve. In establishing our load curve for the NEM, we have aggregated
historical state-based loads. Utilising the methodology in Nelson et al. (2010), load
duration curves were developed for the years 2015-2020, with the OPM model used to
calculate supply-side investment optimality in each of our three scenarios. We assume
average annual growth for each decile of the load duration curve at the historic
ten-year moving average, and apply this to predict demand for each half hour of the
curve in the years between 2015 and 2020.This translates into energy growth rates of
about 1.5 percent per annum, with increases occurring primarily during peak and high
demand periods. Our approach to quantifying the costs of uncertainty is identical to
that in Nelson et al. (2010) in that our analysis is largely quarantined to the period
between 2015 and 2020.Systemaverage cost for each scenario is presented in Table VII.
There is a material difference in NEM-wide system average cost between scenarios.
Headline average costs in the uncertainty and meltdown scenarios are $2.46/MWh and
$4.54/MWh higher than our benchmark certainty scenario, representing increases of
between 2.46 percent and 4.54 percent, respectively. Importantly, Table VII results
exclude renewable plant impacts; Section 5 noted these were surprisingly large in the
meltdown scenario. When combined, it is clear that there are material consequences for
electricity prices associated with implementing a Washington Consensus approach
to carbon policy[14]. System aggregate economic ef?ciency losses across the 2015-2020
period are shown in Figure 10.
The ?rst two bar results inFigure 10 showthe aggregate power systemcost impact of
ongoing policy uncertainty between the Government, Opposition, and the Greens and
Effects of
carbon taxes
121
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
4
4

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
amounts to $4.5 billion without CO
2
pricing, and $4.7 billion with CO
2
pricing. The
reason for the trivial difference between these two scenarios should be obvious enough;
unless the Government provides a framework for a formal carbon price, a shadow
carbon price ?lls the void in any event. In short, both the energy and ?nance industries
viewa carbon price as inevitable whether one exists or not, andthus all of the uncertainty
remains. Our partial equilibriummodeling results for the meltdown scenario (including
CO
2
) puts aggregate economic ef?ciency losses of $8.5 billion, the details of which are
summarized as follows:
Figure 9.
2015 load duration
curve and implied OPM
$0
$100
$200
$300
$400
$500
$600
$700
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
M
a
r
g
i
n
a
l

R
u
n
n
i
n
g

C
o
s
t

(
$
/
k
W
/
a
)
Capacity Factor
Open Cycle Gas
Combined Cycle Gas
0
5,000
10,000
15,000
20,000
25,000
30,000
35,000
40,000
45,000
50,000
0% 10% 20% 30% 40% 50% 60% 70% 80% 90%
E
l
e
c
t
r
i
c
i
t
y

D
e
m
a
n
d

(
M
W
)
Time Exceeded
2,015 Load Duration Curve
CCGT Plant
OCGT Plant
Certainty $/MWh Uncertainty $/MWh Meltdown $/MWh
Headline energy cost (including CO
2
) 82.54 85.00 87.07
Table VII.
NEM system average
cost between 2015
and 2020 (2011$)
JFEP
4,2
122
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
4
4

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
.
Power system economic ef?ciency losses in 2015 total $1.25 billion
(i.e. $1,021 m in headline energy costs and $231 m from renewable energy).
.
Power system economic ef?ciency losses in 2020 total $1.63 billion
(i.e. $1,102 m in headline energy costs and $527 m from renewable energy).
.
Aggregate power system ef?ciency losses between 2015 and 2020 amounts to
$8.6 billion (i.e. $6.4 b in headline energy costs and $2.2 b in additional renewable
energy costs).
.
The present value of economic ef?ciency losses over the 2015-2020 period at a
9 percent private sector real discount rate amounts to $4.9 billion (with economic
ef?ciency losses assumed to be zero until 2015).
7. Policy recommendations and concluding remarks
It is clear that climate change reforms without adequate structural adjustment
assistance run a high risk of a material misallocation of resources in a sector that
provides a universally acknowledged essential service. We noted that this is one trigger
to justify the use of structural adjustment assistance. We also noted that where a reform
results ineconomic shocks that are large; driven bypolicy changes; involving a breach of
long standing expectations and result in highly uneven or magni?ed losses in discrete
industrial segments are another trigger. All of these conditions are satis?ed. Modeling
completed for the Australian Government in 2008 demonstrated that the impact on
brown coal-?red generators is likely to be in the order of $7.9 billion. Our contribution to
understanding the impacts has been to highlight economic ef?ciency losses that might
arise under a Washington Consensus counterfactual scenario of zero compensation.
Our modelingresults provide animportant message – there is a sound public policy case
for providing structural adjustment assistance to intensely affected coal generators.
Our modeling results (in 2011$) concluded economic ef?ciency losses of $1.63 billion pa
Figure 10.
Ef?ciency losses in the
NEM between 2015
and 2020 (2011$)
0.0
1,000.0
2,000.0
3,000.0
4,000.0
5,000.0
6,000.0
7,000.0
8,000.0
9,000.0
Uncertainty Scenario ex-CO2
Uncertainty Scenario incl. CO2
Meltdown Scenario incl. CO2
A
g
g
r
e
g
a
t
e

C
o
s
t

(
$

m
i
l
l
i
o
n
s
)
Impact on Renewable Energy Costs
Impact on Energy Costs
NPV at 9%
Scenario vs. the benchmark Certainty Scenario
Effects of
carbon taxes
123
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
4
4

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
in 2020 and $8.5 billion in aggregate over the period 2015-2020. Such an outcome is
clearly undesirable from a welfare perspective and from an electricity consumer’s
viewpoint. The implications fromthis analysis are clear. Policy certainty and structural
adjustment assistance are critical and requisite elements in relation to carbon policy.
While we have quanti?ed costs of zero compensation, we have not quanti?ed what
level of structural adjustment assistance is necessary to avoid such outcomes. That is
an entirely separate exercise for energy and ?nancial economists. And as we noted at
the outset, economists can provide advice, but the ultimate decision is a matter of
judgment for policymakers.
The costs of our counterfactual zero compensation scenario are substantial, and so
policymakers must turn their attention to how structural adjustment assistance should
be provided. We believe that anadministrative allocationof permits as was prescribed in
the EU has a distinct advantage over other options because it “self-corrects” where
anticipated abatement costs do not materialize. If the value of the permits falls, so too
does the need to provide structural adjustment assistance. This concept is analogous to
the “automatic stabiliser” notion of taxation and transfer payments in a ?scal policy
context. The EU policy may have led to windfall pro?ts, but only because of an
overzealous allocation, not because the policy was somehow fundamentally ?awed.
Importantly, any future application needs to ensure that generators receiving assistance
can close plant at any time subject to maintaining capacity (but not necessarily output)
for system security purposes, and provisions to claw-back any so-called windfall gains
would be entirely appropriate.
In a public policy design, it is necessary to be clear about the objectives being
pursued. The provision of structural adjustment assistance to greenhouse intensive
coal-?red generator should have a short-run objective function of ensuring energy
security and avoiding systemic shocks to energy markets, and a long-run objective of
avoiding economic ef?ciency losses from emerging in capital markets, and in turn,
minimizing electricity price impacts on consumers. In any environment where
the probability of an event occurring is signi?cant, and the consequence extreme, it is
prudent to take action to prevent the event from taking place. This precautionary
principle guides policymakers in all aspects of economic management, and given
electricity is an essential service, carbon policy design should be no different.
Notes
1. The case for structural adjustment assistance to the 22,000 MW black coal ?eet is weaker;
more than 70 percent of black coal plants are owned by State Governments and ?nancial
distress is therefore unlikely to produce material ef?ciency losses in capital markets. Also,
black coal generators have grid-average carbon intensity coef?cients and are therefore less
likely to experience acute ?nancial distress in the short run. To be sure however, over the
long run, the same fate ultimately awaits black coal plant, albeit on a slower-burn trajectory.
2. The term “Washington Consensus” relates to a consensus reached in the 1980s between the
US Treasury, World Bank, and the International Monetary Fund. At its core is a deep belief
in economic reform and the ef?ciency of markets, in particular, ?scal policy discipline, trade
liberalisation, deregulation of capital markets, privatisation of state enterprises, industry
deregulation, establishment of property rights, and so on. The Washington Consensus, and
the virtual blind faith in the ef?ciency of markets, was largely abandoned in late-2010
following the effects of the global ?nancial and economic crisis (Macquarie Securities, 2011).
In contrast, a Keynesian approach for example favours markets and competition, but accepts
JFEP
4,2
124
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
4
4

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
the potential for market failure and the necessary role of government in subsequent
economic stabilisation.
3. Log-rolling policies are implemented in conjunction with reform policies to dampen the
sharpest effects of a reform, in the event making it more politically feasible. An example of
carbon log-rolling policy would be compensation to low income households.
4. Of course, a macroeconomic model would not incorporate any speci?c plant level details. But
the point is that CGE models can produce outcomes which, given the existing plant stock, are
technically intractable due to technical limitations of power systems.
5. Wholesale prices average about $50/MWh. The administered price cap in the NEM spot
market is currently $12,500/MWh.
6. Deregulated wholesale energy markets like the NEM will perform to all economists’
expectations in that prices can and will rise sharply when structural shocks occur. But the
NEMhas demonstrated on several occasions that regulated retail prices are simply incapable
of keeping pace with seismic shifts by comparison to the half hourly spot market clearing
mechanism and the instantly responding forward contract market. Two niche retailers and
one government owned retailer became technically insolvent over such mismatches in the
NEM since 1998.
7. We use a simple new entrant cost multiplier of about $3,000/kW for brown coal plant, over a
50-year term.
8. The 2008 and 2011 ACIL Tasman results re?ect both different input assumptions (one study
was commissioned by the DCC whereas the other utilises ACIL ?gures) and different
reduction scenarios.
9. All ?gures are in $2007 except the ACIL Tasman (2011) study which is in 2011$.
10. Substantive risk premia are already being applied to existing coal plant. Hazelwood attempted
to re?nance a roughly $400 mtermfacility that was nearing the end of its ten-year tenor in 2010.
It is well understood that in the process, the syndicate opened up not only the existing facility,
but another established termfacility despite the fact that it was not maturing. Fully400 bps was
applied to both structures with a tenor of just two-and-a-half years and a cash-sweeping
mechanism. On an incremental basis, the $400 m facility had an effective margin of 700 bps.
11. Estimates on the costs associated with this policy uncertainty can be found in Nelson et al.
(2010).
12. In a project ?nance, three parameters typically limit the size of the facilities; (1) absolute
project gearing; (2) DSCR at 1.8-2.2 times; and (3) LLCR at 1.8-2.2 times the present value of
future cash ?ows.
13. Recall from Table III that spreads for corporate bonds have reduced by 33 percent to 240 bps
since 2008, and global PF facilities now average about 240 bps over swap.
14. These impacts also need to be considered against a background of electricity prices doubling
between FY08 and FY15, being driven by network capital investments (Simshauser et al.,
2011).
References
ACIL Tasman (2008), Impacts of the Carbon Pollution Reduction Scheme and RET – Modeling of
Impacts on Generator Pro?tability, ACIL Tasman, Melbourne.
ACIL Tasman (2011), Energy Security and the Introduction of GHG Emissions Pricing, ACIL
Tasman, Melbourne.
Effects of
carbon taxes
125
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
4
4

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
Argy, F. (1999), “Distribution effects of structural change: some policy implications”, Structural
Adjustment – Exploring the Policy Issues, Productivity Commission, Canberra, AusInfo,
pp. 39-92.
Bushnell, J. (2004), “California’s electricity crisis: a market apart?”, Energy Policy, Vol. 32 No. 9,
pp. 1045-52.
DCC (2009), Proposed Package of Measures to Safeguard the Security of Australia’s Electricity
Supplies, Department of Climate Change, Commonwealth Government, Canberra.
Energy Supply Association of Australia (2010), Electricity Gas Australia, ESAA, Melbourne.
Johnson, O. (1994), “Managing adjustment costs, political authority and implementation of
adjustment programs with special reference to African countries”, World Development,
Vol. 22 No. 3, pp. 399-411.
Joskow, P. (2001), “California’s electricity crisis”, Oxford Review of Economic Policy, Vol. 17 No. 3,
pp. 365-89.
McLennan Magasanik Associates (2008), Impacts of the Carbon Pollution Reduction Scheme on
Generator Pro?tability, MMA, Melbourne.
Macquarie Securities (2011), The Visible Hand: The IMF Leopard Changes is Spots, Macquarie
Economic Research, Sydney, 8 April.
Neary, P. (1982), “International capital mobility, wage stickiness and the case for adjustment
assistance”, in Bhagwati, J. (Ed.), Import Competition and Response, University of Chicago
Press, Chicago, IL, pp. 39-67.
Nelson, T., Kelley, S. and Orton, F. (2011), “The impact of carbon pricing on Australian
deregulated wholesale electricity and gas markets”, AGL Working Paper No. 23, AGL,
Sydney, available at: www.aglblog.com.au
Nelson, T., Kelley, S., Orton, F. and Simshauser, P. (2010), “Delayed carbon policy certainty and
electricity prices in Australia”, Economic Papers, Vol. 29 No. 4, pp. 1-20.
Outhred, H. (2011), Stateline, ABC TV, Sydney, available at: www.abc.net.au/news/video/2011/
02/11/3136902.htm (accessed 30 March).
Pasour, E. (1973), “Economic growth and agriculture: an evaluation of the compensation
principle”, Amercian Journal of Agricultural Economics, Vol. 55 No. 4, pp. 611.
ROAM (2008), Modeling of Carbon Pricing Scenarios, ROAM Consulting, Brisbane.
Samuelson, R. (2009), “Economists ignored history”, Australian Financial Review, 6 July, p. 55.
Satkunasingam, V. (2011), “Australia’s power landscape”, Project Finance International, No. 453.
Simshauser, P. (2005), “The gains from the microeconomic reform of the power generation
industry in east coast Australia”, Economic Analysis and Policy, Vol. 35 Nos 1/2, pp. 23-43.
Simshauser, P. (2008), “On emission permit auction vs allocation and the structural adjustment of
incumbent power generators in Australia”, The Electricity Journal, Vol. 21 No. 10, pp. 30-41.
Simshauser, P. (2009), “On emissions trading, toxic debt and the Australian power market”,
The Electricity Journal, Vol. 22 No. 2, pp. 9-29.
Simshauser, P. (2010), “Resource adequacy, capital adequacy and investment uncertainty in the
Australian power market”, The Electricity Journal, Vol. 23 No. 1, pp. 67-84.
Simshauser, P. (2011), “The hidden cost of wind generation in a thermal power system: what
cost?”, Australian Economic Review, Vol. 44 No. 2.
Simshauser, P. and Wild, P. (2009), “The Western Australian power dilemma”, Australian
Economic Papers, Vol. 48 No. 4, pp. 342-69.
JFEP
4,2
126
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
4
4

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
Simshauser, P., Nelson, T. and Doan, T. (2011), “The boomerang paradox part 1: how a nation’s
wealth is creating fuel poverty”, The Electricity Journal, Vol. 24 No. 1, pp. 72-91.
Stiglitz, J. (2002), Globalisation and its Discontents, Norton & Co, New York, NY.
Further reading
Garnaut (2011), Transforming the Electricity Sector – Update Paper 8, Garnaut Climate Change
Review Update, Commonwealth Government, Canberra.
Simshauser, P. and Doan, T. (2009), “Emissions trading, wealth transfers and the wounded bull
scenario in power generation”, Australian Economic Review, Vol. 42 No. 1, pp. 64-83.
About the authors
Paul Simshauser is Chief Economist at AGL Energy Ltd and Professor of Finance
at Grif?th University.
Tim Nelson is Head of Economic Policy at AGL Energy and an Adjunct Research Fellow at
the University of New England. Tim Nelson is the corresponding author and can be contacted at:
[email protected]
Effects of
carbon taxes
127
To purchase reprints of this article please e-mail: [email protected]
Or visit our web site for further details: www.emeraldinsight.com/reprints
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
4
4

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
This article has been cited by:
1. Paul Simshauser, Tim Nelson. 2015. Australia's coal seam gas boom and the LNG entry result. Australian
Journal of Agricultural and Resource Economics 59:10.1111/ajar.2015.59.issue-4, 602-623. [CrossRef]
2. Tim Nelson, Cameron Reid, Judith McNeill. 2015. Energy-only markets and renewable energy targets:
Complementary policy or policy collision?. Economic Analysis and Policy 46, 25-42. [CrossRef]
3. Paul Simshauser, Tim Nelson. 2015. The Australian east coast gas supply cliff. Economic Analysis and
Policy 45, 69-88. [CrossRef]
4. Paul Simshauser. 2014. The cost of capital for power generation in atypical capital market conditions.
Economic Analysis and Policy 44, 184-201. [CrossRef]
5. Paul Simshauser, Jude Ariyaratnam. 2014. What is normal profit for power generation?. Journal of
Financial Economic Policy 6:2, 152-178. [Abstract] [Full Text] [PDF]
6. Tim Nelson, James Nelson, Jude Ariyaratnam, Simon Camroux. 2013. An analysis of Australia's large
scale renewable energy target: Restoring market confidence. Energy Policy 62, 386-400. [CrossRef]
7. Lauren Solomon, Tim Nelson. 2013. Energy for Life – an evidence?based approach to corporate
citizenship. Sustainability Accounting, Management and Policy Journal 4:2, 236-258. [Abstract] [Full Text]
[PDF]
8. Paul Simshauser, Tim Nelson. 2013. The Outlook for Residential Electricity Prices in Australia's National
Electricity Market in 2020. The Electricity Journal 26, 66-83. [CrossRef]
9. James Nelson, Paul Simshauser. 2013. Is the Merchant Power Producer a broken model?. Energy Policy
53, 298-310. [CrossRef]
10. Tim Nelson, Paul Simshauser, James Nelson. 2012. Queensland Solar Feed-In Tariffs and the Merit-
Order Effect: Economic Benefit, or Regressive Taxation and Wealth Transfers?. Economic Analysis and
Policy 42, 277-301. [CrossRef]
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
4
4

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)

doc_931680524.pdf