Infratil Update Newsletter September 2018

INFRATIL UPDATE SEPTEMBER 2018
The industrial revolution was an energy
revolution. Replacing wood fires and the toil
of humans and animals with energy from
coal, oil and gas freed mankind to build the
world we know today.
Coal succeeded wood as mankind’s principal
source of fuel in about 1900 when annual
human carbon emissions amounted to
approximately 2,000 million tonnes and CO
2

made up 296ppm of the atmosphere. Now
annual emissions are 35,000 million tonnes
and atmospheric CO
2
is 408ppm.
Now the goal of the 195 countries which
have signed the 2016 Paris Agreement is to
reduce emissions in the hope this reduces
the risk of the planet warming and the
climate becoming unstable.

1
SEPTEMBER UPDATE 2018
As the graph below shows, electricity businesses make up a large
part of Infratil; based in New Zealand, Australia and the USA. They
comprise some retail and ancillary services, but most of the value is
in generation plant.
Given its materiality to Infratil, it’s reasonable to ask about electricity
generation as an investment. Globally the electricity industry is
undergoing major changes which are disrupting value and creating
uncertainty. There are also the personal experiences of consumers, which
may well raise questions about what is happening commercially at the
other end of the wire.
If you are a New Zealand resident, it’s likely your business or household is
paying less for electricity today than five years ago. Even over the last ten
years, while government figures show a modest increase in the average
annual household electricity bill, after stripping out higher GST, line
charges, and consumer prices, what is paid to generators and retailers
has fallen.
However, if you are an Australian reading this, it’s likely your personal
experience is different. As highlighted by the recent Australian Competition
and Consumer Commission finding that “retail electricity prices have
increased by 80 to 90 per cent (in real terms) in the past decade.”
A UK resident is probably feeling the same as an Australian, while the US
situation is different again as in many states wholesale prices have been
driven down by falling renewables costs and lower gas prices. In the US it
has been “merchant generators” (companies that generate electricity and
sell into the wholesale market) which have suffered (many have entered
bankruptcy) rather than consumers.
The divergent experiences illustrate the local character of countries’
electricity industries and, as it happens, the costs regulators often impose
on consumers and generators as they seek to reduce sector emissions and
to define “solutions” for problems often created by other regulations.
In a report on the UK’s cost of energy (published by the UK government last
October) Professor Dieter Helm noted that “The cost of energy is too high,
and higher than necessary to meet Climate Change targets. Households
and businesses have not fully benefited from the falling costs of gas and
coal, the rapidly falling costs of renewables, or from the efficiency gains to
network and supply costs which come from smart technologies. [They] have
not benefited [because] the scale of the multiple interventions in the
electricity market is now so great that few, if any, could even list them all,
and their interactions are poorly understood. Complexity is itself a major
cause of rising costs, and tinkering with policies and regulations is unlikely
to reduce costs. Indeed, each successive intervention layers on new costs
and unintended consequences. It should be a central aim of government to
radically simplify the interventions, and to get government back out of
many of its current detailed roles.”
But to return to the key question. Is generation a good investment for
Infratil? The answer is equivocal. It may be (which of course also means it
may not be too). This Update seeks to outline the pros and cons of
investing in electricity generation, with a focus on New Zealand.
INVESTING IN
ELECTRICITY GENERATION
THE MAKE UP OF INFRATIL’S ASSETS
0
10
20
30
40
50
60
70
80
90
100
%
20182008200920102011201220132014201520162017

New Zealand Energy

Australian Energy

US Energy

Transport

Other

211
INFRATIL
THE REAR VISION MIRROR (NEW ZEALAND)
In New Zealand, the last decade has been a relatively poor one for investors
in generation and energy retailing. The one relevant company with
activities entirely in New Zealand which has been listed through the whole
period is Contact Energy. It experienced a 30% decline in its share price.
Trustpower, benefitting from its Australian activities, now has almost the
same share price as ten years ago (adjusted for the demerger of Tilt
Renewables). Adding in the period’s 16% lift in CPI, on the one hand, and
dividends, on the other, gives a modest real return.
By way of comparison, shares in listed lines company Vector have risen
70%. Its return on capital is set by the Commerce Commission as opposed
to reflecting the outcome of behaviour in competitive markets. Similarly,
the unlisted national grid company Transpower conveyed 37,637GWh
of electricity and charged $1,061 million in the year to 30 June 2017.
A decade ago it conveyed 39,128GWh and had income of $591 million.
The generators have done poorly because weak demand resulted in
over-supply which resulted in low wholesale energy prices and few
opportunities to invest in growth.
Less electricity is being used because of better and more efficient
insulation, appliances and equipment, and perhaps because winters are
a little shorter and warmer. Lower demand depresses market prices.
The following graphs show the composition of the average annual
household electricity bill, expressed as both price (cents per kWh) and
dollars which takes account of price and quantity. Both are adjusted to
take out the effect of consumer price changes. Also graphed are the
real wholesale electricity price over the period, and New Zealand’s
electricity consumption since 1973 which shows the lack of growth over
the last decade.
New Zealand electricity generators built capacity in anticipation of
demand growth, which didn’t occur. Between 1973 and 2008 there were
7 five year periods during which demand grew between 7% and 20%,
averaging 13.5%. Over the following 2 five year periods (from 2008 to
2018), demand fell 0.2% before reviving 1.1%.
The average wholesale price of Trustpower’s generation, in
2018 dollars, fell 22% over the decade. A stable price would
have lifted the value of Trustpower’s 2018 generation by about
$37 million. The hedge market price fell 10% .
Ministry of Business, Innovation and Employment
Morrison & CoMinistry of Business, Innovation and Employment
AVERAGE HOUSEHOLD ELECTRICITY PRICE BREAKDOWN
AVERAGE HOUSEHOLD ELECTRICITY COST BREAKDOWN
NEW ZEALAND ELECTRICITY DEMAND 1973 TO 2018
3 YEAR MOVING AVERAGE WHOLESALE ELECTRICITY PRICE
0
$500
$1,000
$1,500
$2,000
Per Household (2018$)
201820082010201220142016
35
30
25
20
15
10
5
cents/kWh (2018$)
2009
201020112012201320142015201620172018
0

Generation & Retailing

Transmission & Distribution

GST

GST

Distribution

Transmission

Energy and Retail


wholesale energy hedge price

average spot wholesale energy price
200320082013201819781983198819931998
GWh
20,000
0
30,000
50,000
40,000
200820102012201420162018
cents/kWh (2018$)
0
2
4
6
8
10
12
No growth

3PB
INFRATILSEPTEMBER UPDATE 2018
THE REAR VISION MIRROR (GLOBAL)
To quote the consultancy firm McKinsey (May 2018), “analysis of 50 major
publicly listed utilities from Asia, Europe, and North America showed
average total cumulative returns to shareholders of about 1% from July
2007 to July 2017, compared with 55% for the MSCI World Index.” The
poor returns of New Zealand generator-retailers were not unique. However,
while low wholesale electricity prices were universally the culprit, different
factors resulted in the low prices.
New Zealand generators built too much capacity because their forecasts
of demand were optimistic. In many overseas markets too much capacity
was built because of over-generous or badly structured renewable
generation incentives.
An example of the scale of subsidy, and the distortion, was given by a
Financial Times’ columnist writing about his personal experience in the UK.
In 2011 he purchased solar panels for his home for £12,800; encouraged
by being able to sell for 43.3p (CPI adjusted to 2036) any electricity
generated but not used in his home. What he generates and uses has
halved his power bill and he receives an annual cheque of about £1,600
for what he generates and does not use.
Adding icing to the cake, he reported being approached by a broker
offering to pay him £13,428 if he on-sold all the future subsidy payments.
If he accepts that offer, he will end up with someone else maintaining his
little roof-top power station, still having free power to the extent his
household uses its generation, and a net £628 in the bank.
In addition to the distortions of such schemes, McKinsey also noted
problems with badly structured regulation penalising generators and
retailers and skewing the industry’s revenue towards distribution
companies.
Looking forward, McKinsey points to the need for investment returns to
improve. They estimate that over the next decade US$7.2 trillion must be
invested globally to ensure electricity supply keeps up with demand; which
will only happen if positive returns are in prospect. (US$7.2 trillion is over
70 times the value of all New Zealand’s listed companies and almost five
times the value of all Australian listed companies.)
McKinsey’s prescriptions for better returns are simple. Less disruptive
regulation on the one hand, and for generators to transfer more electricity
price-risk to consumers via long-term contracts, on the other hand.
LOOKING FORWARD
In New Zealand today there is ample supply - low prices - poor returns - no
new build. This will change once demand growth revives, or old power
stations are retired.
Today almost no new generation is under construction in New Zealand
because the price of electricity is too low to encourage investment. There
are few other barriers. A household with $5,000 can have photovoltaic
generation installed on the roof and sell the output into the grid. At the
other end of the spectrum, the five large generator-retailers each have
access to hundreds of millions of dollars of investment capability and many
fully-consented development sites. All that is holding back investment
(small and big) is the low electricity price/value.

54
INFRATILSEPTEMBER UPDATE 2018
ELECTRICITY PRICES FROM HERE
Accurately forecasting electricity prices has proven to be difficult. Since
1997, Infratil has published nine Update newsletters which have addressed
aspects of the New Zealand electricity market. Rereading them shows the
challenge of making good forecasts (although they have been more
accurate than those periodically released by government agencies).
The problem is the number of important variables. Over the last two
decades electricity prices have been significantly influenced by less
restrictive regulation, more competition, changes to fuel and plant
economics, and fluctuating demand.
Looking forward, both demand and the cost of supply present
uncertainties:
•
Demand seems likely to depend on which path is taken to reduce
New Zealand’s CO
2
emissions. If every car is converted to battery-electric,
national electricity demand would rise about 25%. On the other hand,
closing New Zealand’s aluminium, steel, and wood processing industries
would also reduce emissions and would drop demand for electricity by
about a quarter.
•
The cost of generation, is a function of both the cost of different types
of generation and the generation mix. For instance, wind and solar
generation costs may fall, but if no gas-fired generation can be used to
provide back-up to intermittent renewables (because of policies to stop
CO
2
emissions) then the average price will probably rise.
As will be apparent, both demand and supply will be impacted by political
decisions, especially with regards to the policies adopted to reduce
New Zealand’s CO
2
emissions.
REDUCING CO
2
EFFICIENTLY WITH LOW-COST ELECTRICITY
New Zealand requires investment in electricity generation to power its
electric, low carbon, future. The investment will be more readily available,
and the electricity will cost less, if government lets the market function.
The electricity generation industry is efficient at responding to price signals,
it has the capacity to expand production to meet growing demand, and its
ownership and competitive structure makes it likely that the lowest-cost
projects will be the ones that get built.
To achieve Climate Change goals, New Zealand needs increased use of
electricity and reduction in the use of motor spirits, gas and coal.
The cheaper electricity is, the lower will be the cost of the switch and the
lower the need for direct subsidies, inducements and restrictions.
If the investors funding new electricity generation perceive a market
operating without distortions or disruptive regulatory intervention they
will price their capital accordingly. No single factor will be more influential
in determining the electricity price required by new power stations than
investors’ required rate of return. The more government seeks to get
involved or to change the rules around generation, the more reluctant
will be investors to fund new power stations and the higher will be the
cost of electricity.
Decarbonisation of the economy will be expensive. Any diversion of
government’s resources into electricity will mean less is available to assist
those most adversely affected by the transition to lower emissions.
HOW MUCH HIGHER DO ELECTRICITY PRICES NEED TO BE
(IF THE MARKET IS LEFT TO ITS OWN DEVICES)?
While little investment in new generation is now occurring, how high prices
need to be before construction restarts depends on the economics of
individual power stations, and the economics of the entire portfolio of
power stations required to reliably provide the additional electricity.
To explain. A home with only photovoltaic panels will have an electricity
cost that is based entirely on the economics of the photovoltaic panels,
but it will have no electricity at night or when it’s cloudy. Adding back-up
for when the sun isn’t shining means additional costs.
Each grid-connected consumer relies on many sources of electricity to
ensure 24/7 supply and therefore their average electricity price reflects the
economics of a portfolio of power stations.
Over recent years, there has not been a single day in New Zealand when
hydro, wind, geothermal and gas-fired generation haven’t all been used for
at least some of the time. The relevance of this should not be understated,
especially in the context of the industry’s three paramount goals; efficiency
(low cost), reliability (24/7 100% availability), and minimum emissions.
There have been suggestions that New Zealand’s electricity industry should
seek to use no gas or coal fired generation in a year of average hydrology.
With current technology this goal is not compatible with low cost electricity
that is reliable.

54
INFRATILSEPTEMBER UPDATE 2018
COMPARING TWO YEARS
The years ended 31 March 2017 and 2018 provide a case study of generation economics and dynamics.
20172018
NZ National Electricity Generation
42,531GWh (85.7% renewable)
NZ National Electricity Generation
42,922GWh (80.8% renewable)
Average Wholesale Electricity Price
5.65c/kWh
Average Wholesale Electricity Price
8.59c/kWh
The price-duration curve graphed above ranks every one of the
17,520 half-hour prices that occurred over the year. It shows
that (moving right to left) 20% of the time the price was below
4.3c/kWh and 50% of the time it was less than 5.3c/kWh.
FY2018’s price-duration curve looks like its predecessor, but its
numbers differ. 20% of the time the price was below 5.5c/kWh and
50% of the time they were below 7.7c/kWh.
The year was wet and windy, and renewables contributed
36,459GWh of generation. Oil, gas and coal fired generation was
only required to produce 6,017GWh.
A dry calm year saw renewable generation fall by 1,794GWh
meaning that, with demand also rising, thermal generation was
called on to produce 2,183GWh more than the prior year.
Generators will have earned approximately $2,400 million selling
at the spot price. This is not entirely accurate as some hours see
more generation than others which is not factored into the
calculation, but it is a reasonable estimate.
Approximate value of generation $3,700 million.
2017 and 2018 illustrate how the average cost of electricity reflects the
cost of the whole portfolio of generation assets. The two years also show
the need for back-ups and the different economics (i.e. cost) of different
forms of generation.
Gas-fired peakers are typically used when there is insufficient wind and
hydro available, and while we don’t have the operational and financial
details of such a power station over the two years, they can be illustrated.
Imagine a hypothetical 10MW gas-fired generator which cost 10c/kWh to
operate (i.e. had a high variable cost). In 2017 that generator would have
operated 116 hours producing net income of $162,000 (average price
27.8c/kWh, but for only 1.3% of the year). In 2018 the same generator
would have been switched on 2,413 hours proving net income of
$1,062,000 (average price 14.4c/kWh operating 27.5% of the year).
In the opposite corner to such high-cost, but controllable, thermal
generation is wind which cannot be switched on to suit demand and hydro
which has considerable variability of annual availability.
In 2017 Trustpower’s hydro power stations generated (coincidentally)
2,017GWh and received an average wholesale price of 5.2c/kWh
making the output worth $105 million. In 2018 Trustpower generated
2,238GWh at an average price of 8.8c/kWh, which made the output worth
$197 million. Trustpower benefitted from having ample water in a year
when others didn’t and when prices were high.
Tilt Renewables’ wind generation over the two years produced a different
picture. Tilt sells all its New Zealand generation on fixed price contracts
so in both years it received 6.3c/kWh. In 2017 it generated 744GWh worth
$47 million and in 2018 it generated 571GWh worth $36 million.
New Zealand’s portfolio of generation has a high reliance on renewables,
which to ensure reliability at least-cost, creates a corresponding high
reliance on back-up from gas plant.
100%80%60%40%20%0%
0
10
20
30
40
50
cents/kWh
100%80%60%40%20%0%
0
10
20
30
40
50
cents/kWh
ELECTRICITY SPOT PRICES (AT OTAHUHU)ELECTRICITY SPOT PRICES (AT OTAHUHU)

76
INFRATILSEPTEMBER UPDATE 2018
FALLING COST AND FALLING VALUE
Reliable, lowest-cost, electricity requires a portfolio of generation which
is mutually compatible and individually efficient. The optimal portfolio
depends on the economics and availability of each generation unit and the
pattern of demand. This means that an individual source of even very low
cost electricity may not lower the market’s average cost, and that a plant
with a very low cost can still run at a loss because of the low value of the
electricity it generates.
If New Zealand had a much higher proportion of wind generation it would
lower the value of electricity when it was windy, but would require more
back-up for when it wasn’t. Whether the resulting average market electricity
cost increased or reduced would depend on the relative impact of the
cheap-wind and the expensive back-up. In either case, even if the
wind-electricity was cheap to generate, that cost could still be less than
the average value of electricity at the times it was being generated.
In Australia, government schemes supported wind generation, but when
wind became a substantial share of generation in some states, reliability of
supply became a problem and government stepped in to subsidise back-up
sources of electricity. Electricity supply became less reliable, more
expensive and required more subsidies because of unbalanced policies
and investment.
The “low-cost and low-value” paradigm is explained at length in the recent
book, Taming The Sun by Varun Sivaram. It notes that solar energy is a
boon in developing markets where the alternative is no electricity or petrol
generators, but not always a good proposition in developed markets.
In California the marginal value of solar generation is low, even zero,
while in Germany the scale of subsidies is absurd.
Graphs like those below are well known. The figures come from a
well-regarded annual report produced by Lazard which shows generation
costs in the USA.
The graphs show the electricity price a new plant requires if it is to cover
operating costs and provide a satisfactory return on invested capital.
A New Zealand version of these curves would look different. New Zealand
does not have the cloudless skies available in the best US solar locations.
While part of the improvement to the economics of wind generation
depicted in the graph relates to the development of plant suited to
relatively low wind speeds. New Zealand has higher wind speeds so the
cost reductions do not apply to the same extent.
However, while the breakeven cost of these sources of electricity have
fallen in the USA, that’s only part of the story, its also necessary to know
what has happened to the value of the electricity being generated and to
the average market price of electricity.
Berkeley Laboratories of California produced a fascinating piece of research
on this topic. The graph over the page on the top left shows the increasing
share of all California’s electricity generation which comes from solar
(the bars) and the average value of that generation as a proportion of
the average market price (the line). In 2012 solar made up 2% of total
Californian generation and, had it been sold at spot prices, would have
received an average US$ 3.8c/kWh which was 126% of the average
electricity price that year. In 2017 solar made up 16% of generation worth
an average US$ 2.5c/kWh which was only 79% of the average market
electricity price.
The graph over the page on the top right shows the average California
market price of solar electricity (as derived by Berkeley Lab) and the Lazard
estimated breakeven electricity price required by new plant. There are two
things to note, the value of solar generation in California is falling faster
than the cost, and the value has been consistently below the cost.
Notwithstanding this, solar generation has, and is, being built in California
thanks largely to generous subsidies.
THE COST OF US SOLAR GENERATIONTHE COST OF US WIND GENERATION
20102012201420162018
0
10
20
30
40
US$/cents/kWh
20102012201420162018
0
10
20
30
40
US$/cents/kWh

76
INFRATILSEPTEMBER UPDATE 2018
STORAGE. THE BIGGEST PROBLEM
The average electricity price consumers face depends on the portfolio
of generation required to provide 24/7 availability with 100% reliability.
As shown in the below graph on the right, on an average day, national
New Zealand demand can fluctuate between 3,000MW and 7,000MW.
The left-hand graph shows weekly and seasonal patterns and how one
year varies from another. The graphs show the fluctuations and
irregularities of demand, which is only half the story as wind and hydro
power stations have their own irregularities.
There are two relevant ways to store energy for peaks in demand and
for when some power stations are unavailable: coal/gas/oil and water.
New Zealand uses both and needs both.
The touted alternative, batteries, are currently irrelevant. They store little
and cost lots. The most efficient battery costs over $300 per kWh of storage.
In other words, to store sufficient electricity to power ten 100 watt
lightbulbs for 1 hour costs at least $300. The average household uses about
7,000kWh a year so to store one household’s annual electricity needs
would cost $2,100,000. Another way of looking at it is that the water stored
in Lake Pukaki alone can generate 1,595GWh of electricity (about 4% of
national annual consumption). To replicate that amount of storage in a
battery would cost $480,000,000,000 ($480 billion).
Batteries are fine on Great Barrier Island which has no mains power, and
they may be viable in special situations where transmission assets are very
expensive and there are short gaps in the availability of grid power, but not
usually. Of course it is possible that the research now being directed at
battery storage improves their viability, also once electric cars are prevalent
their batteries may be available to store energy for more uses than just
powering the vehicle.
THE COST AND VALUE OF SOLAR ELECTRICITY (CALIFORNIA)CALIFORNIA SOLAR ELECTRICITY
NEW ZEALAND ELECTRICITY DEMAND (ROLLING 7 DAYS)TOTAL NEW ZEALAND DEMAND (1 JULY 2018)
201720122013201420152016
18
16
14
12
10
8
6
4
2
0
135
120
105
90
75
60
45
30
15
0
Solar as a % of all Electricity
Solar Electricity price as a % of market average
201220132014201520162017
US$ cents/kWh
0
2
4
6
8
10
12
14

Cost

Value

Solar as a % of all electricity (LHA)

Value of Solar (RHA)
JanMarMayJulAugOctFebAprJunSepNovDec
Total Demand (GWh)
600
630
660
690
720
750
780
810
840
870
900


2017

2018
3AM6AM9AM12PM3PM6PM9PM
MW
3500
4000
4500
5000
5500
6000


Today’s demand

Corresponding day last week

TranspowerEM6Live

8
INFRATILSEPTEMBER UPDATE 2018
WindCCGT*Geothermal
10
8
6
12
4
2
cents/kWh
0

Capital 9% p.a.

CO
2
$30/tonne

Gas @ $12/Gj

O&M and Royalties

Back-up
10
12
8
6
4
2
0
WindGeothermal
cents/kWh
CCGT*

Capital 8.5% p.a.

CO
2
$15/tonne

Gas @ $7/Gj

O&M and Royalties

Back-up
THE COST OF NEW PLANT IS (MOSTLY) FALLING
Although the previous sections have focussed on the portfolio of
generation required to provide 24/7 reliability, the average cost of
generating electricity (and therefore the price to consumers) depends
on the cost of the individual sources of generation as well as their
individual contributions.
The following set of graphs show Morrison & Co’s current and historic
calculations of the breakeven economics for the three types of power
stations which are most relevant for New Zealand.
The graphs show the electricity prices required in dollars of the day.
To better reflect the changes in plant economics they would be adjusted
to reflect that the CPI has risen 9% since March 2011 and 4.5% since
December 2015. Consequently, while the graphs show a 20% drop in
the break-even price required by wind over the seven years, in real terms
it’s 28% lower.
Also provided are the key costs, both to show what has changed over the
seven years, and the relevance of the cost of capital. While the largest cost
of a gas-fired power station is from fuel the main cost of a wind farm relates
to the money used to build it.
SEPTEMBER 2018
Over the seven years, wind has been the winner thanks to a lower
cost of capital and lower plant costs.
However, if the NZ$/US$ were to fall to 60 cents the break-even cost
of wind generation would be back to 8c/kWh. While a 1% hike in the
cost of capital would add about 1c/kWh to the break-even price.
DECEMBER 2015
New geothermal electricity has remained anchored at about
8c/kWh. But access to suitable geothermal liquids is probably
finite making this source of electricity also limited.
MARCH 2011
The economics of gas-fired generation has shown the most
volatility thanks to big changes in local gas prices and
fluctuations in emission costs.
WindCCGT*Geothermal
10
12
8
6
4
2
0
cents/kWh

Capital 8.0% p.a.

CO
2
$30/tonne

Gas @ $8/Gj

O&M and Royalties

Back-up
* Combined Cycle Gas Turbine

98
INFRATILSEPTEMBER UPDATE 2018
CARBON
As the graphs in the previous section show, the price of carbon emissions
matters for gas-fired and, to a lesser extent, geothermal generation.
But as shown in the comparison between the generation in 2017 and
2018, gas/coal fired generation is a small part of the total, and likely to
become smaller. A higher carbon price will make some generation more
expensive, but the overall price impact will be muted. The cost of gas fuel
may be more material than the cost of CO
2
emissions.
In 2016 (the year of the most up to date data) generating the electricity
consumed by the average household produced 0.66 tonnes of CO
2
a cost
of $16.60 for the year (plus GST) at an emission price of $25/tonne.
It is likely that the electricity sector’s carbon emissions will continue to fall
absolutely and per unit of total generation and the price of emissions will
have little bearing on the average electricity price. There have been
suggestions that electricity generation in New Zealand should be placed on
a path towards zero emissions by 2030 (drought and low hydro generation
being extenuating circumstances). This is not a practical goal and would
make it impossible to achieve the largest possible national emission
reduction for the least cost.
Gas-fired generation is an important source of back-up for intermittent
renewables and if it can’t be used in that capacity, the cost of electricity will
rise a lot (unless reliability is sacrificed). High-priced, unreliable, electricity
will make it much harder to encourage people to get out of petrol cars into
electric ones as well as to make the other changes sought to lower national
emissions.
The electricity industry has significantly decarbonised over the last
decade. The CO
2
emissions produced in providing an average
household with its annual electricity requirements has fallen 63%.
Sector emissions caused by burning coal, gas and oil have fallen 66%.
Total sector emissions include those from geothermal generation.
THE AVAILABILITY OF CAPITAL
Before commenting on electricity demand (more briefly than the
commentary on the price of supply) it’s worth drawing attention to research
undertaken by analysts at UBS. They estimated that New Zealand’s big-five
generators are now spending about $180 million a year to keep their
generation plant going, but that over the longer term about $490 million a
year will be required to maintain current generation levels. They also note
that if generation output was required to grow by 1% per annum
investment of an additional $230 million a year would be necessary.
If annual investment in electricity generation has to rise from $180 million
to $720 million, the capital will presumably have to come from a mixture of
additional debt, equity and higher sector earnings. An average price rise of
1c/kWh would, for instance, provide generators with about $300 million
per annum of cash after tax. But whether funds are internally generated or
provided by investors and lenders, capital will only be allocated on the
prospect of a satisfactory return.
The table summarises UBS’s figures of the cash flows of the five main
generators for their most recent financial years.
$ MillionsMercuryGenesisContactMeridianTrustpower
Total EBITDAF *$556$358$484$682$243
NZ Generation EBITDAF*$334$197$339$512$183
Interest($96)($73)($84)($81)($32)
Tax($78)($29)($100)($87)($45)
Operating cash flow$382$256$300$514$166
Dividends($272)($168)($200)($492)($110)
Remaining funds$110$88$100$22$56
Capex($118)($87)($75)($58)($17)
* Earnings before interest, tax, depreciation and amortisations; a proxy for net operating cash flow. The five companies also have earnings from activities unrelated to New Zealand generation.
199019952000200520102015
Sector - Million Tonnes of CO
2
Household - Tonnes
0
1
2
3
4
5
6
7
8
9
10
0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0

Electricity Sector CO
2
Emissions

Tonnes of CO
2
per household

10
INFRATIL
As noted, the difference between the capital now being allocated to
generation and what is required to maintain current levels of output was
estimated by UBS to be approximately $310 million a year. The current
level of capital spending will keep existing power stations ticking over, but
at some point, replacement not just refurbishment is required.
UBS did note that hydro power station dams can have exceptionally long
lives; pointing to a dam built by Egyptian pharaoh Sethi in about 1285BC
as still functioning after 3,300 years. Unfortunately, it is in Syria so evidence
as to this assertion is scant. The oldest hydro power station operating in
New Zealand is Trustpower’s Waipori which started generating in 1907.
IN SUMMARY
Over the last decade electricity demand in New Zealand has been flat, and
because the industry built in anticipation of demand rising, excess supply
resulted in weak prices and poor returns on invested capital.
If demand increases it will remove excess supply and require investment
in capacity. To increase national generation output by 1% per annum will
require over $500 million of additional capital each year. This will only be
forthcoming if prospective returns are satisfactory, which almost certainly
requires higher wholesale electricity prices.
It is however difficult to forecast the required price level as it depends on
both generation technology, financial market variables, and the operation
and regulation of the electricity market. If there are no dramatic changes
to these factors, the likely future price range will be 8-9c/kWh. The average
of the last five years was about 7.1c/kWh.
While technology, finance, and regulations will each contribute to future
electricity prices, the suite of policies adopted to reduce CO
2
emission will
be important. The more policy makers rely on direct initiatives such as
regulating the use of gas-fired generation rather than on placing a price on
CO
2
, the higher will be the cost of electricity.
DEMAND
In the five years to 2004 New Zealand electricity consumption grew on
average by 967GWh per annum. This was then followed by nine years
where annual consumption increased only 49GWh, and then by four years
during which average increases were 281GWh. This century the average
increase so far has been 0.2% per annum.
The policy initiatives now being developed to decarbonise the New Zealand
economy will require individuals and industries to shift from coal/gas/oil
to electricity. But estimating the scale of additional generation required is
more guess than forecast.
Most estimates of the growth required is in the 1-2% per annum range
which over the next three decades amounts to an average annual increase
of 500GWh to 1,200GWh.
While the growth estimates are plausible, conviction seems to be lacking as
no construction of significant new generation capacity is actually
happening.
Impediments to reliable estimates of future electricity demand include
uncertainty about government climate change policies and uncertainty
about the cost of switching from CO
2
fuels to electricity.
Disconcertingly, Transpower the national grid, has shown that
decarbonisation and no electricity demand growth could both occur if the
economy shrinks by 3% per annum. Which is hopefully not what happens.
NEW ZEALAND GENERATION
204920452041203720332029202520212017201320092005200119971993
90,000
80,000
70,000
60,000
50,000
40,000
30,000
20,000
10,000
0
GWh

Net Generation

1% Growth

2% Growth

1211
SEPTEMBER UPDATE 2018
GOVERNMENT POLICIES TO DECARBONISE THE ECONOMY
Leaving out agriculture, waste, and the land sectors (emissions and
sequestration), New Zealand’s emission profile over the 26 years since
1990 is as depicted in the following graphs. Over that period, the annual
CO
2
emissions of the energy, transport and industrial sectors rose from
27.4 million tonnes to 36.2 million tonnes, with 71% of the increase
coming from transport.
A raft of policies are now being developed to reduce these emissions, but
it will take years before there is clarity about how they work. De-carbonising
New Zealand industry and transport could involve transitioning motorists
and manufacturers from oil/gas/coal to using electricity, or it could involve
closing New Zealand’s aluminium, steel, wood processing (and other)
industries and banning cars.
THE COST OF SWITCHING
A price on CO
2
emissions will encourage motorists, manufacturers and
households to switch from coal/gas/oil to electricity. But at present there is
little more than guesses as to where switching will occur between $25 and
$250 per tonne of CO
2
.
•
When Nissan first offered its all-electric Leaf in New Zealand it didn’t sell,
probably because the cost was over $50,000 which was $25,000 more
than a comparable petrol vehicle. However now there is good take up of
second-hand Leafs imported from Japan.
The position of the lines reflects a particular set of costs; for plant, coal and
capital. As those variables change so too will the break-even “indifference” line.
If the price of electricity is very low, then even a low emission price will make
electric milk drying cheaper than using coal. But as the cost of electricity rises,
so too must the cost of coal’s emissions if electricity is to remain viable.
The graph shows that 8c/kWh electricity requires a CO
2
price of about
$65/tonne for new investment to be directed into electric rather than coal
powered milk drying.
The “replacement plant” line indicates that at 8c/kWh an extremely high CO
2

price would be required before Synlait or Fonterra would replace existing plant.
2016199220001996200420082012
45,000
40,000
35,000
30,000
25,000
20,000
15,000
10,000
5,000
0
CO
2
Emissions Million Tonnes

Transport

Electricity

Industrial Processes

Other Industries
NZ CO
2
EMISSIONS: TRANSPORT, ELECTRICITY, INDUSTRIAL
THE % OF CO
2
EMISSIONS: TRANSPORT, ELECTRICITY, INDUSTRIAL
2016199220001996200420082012
0
20
40
60
80
% of each source
100

Transport

Electricity

Industrial Processes

Other Industries
$25$50$75$100
Electricity price - cents/kwh
10
9
8
7
6
5
4
Cost Per Tonnes of CO
2
Emissions
New Plant
Replacing Plant
COAL VS ELECTRIC PRICE INDIFFERENCE
ELECTRICITY PRICES - EMISSION COSTS
It seems that people will take a punt if the car costs less than $20,000
(even if it is still more expensive than a petrol equivalent). They will now
be learning how electric cars compare to petrol ones with regards to
range, maintenance, comfort, safety, reliability, resale, fuel cost,
road-user charges, and so on.
With approaching 10,000 electric vehicles on New Zealand roads,
the knowledge base will be growing quickly. But it will still be some
time before it is possible to anticipate what difference CO
2
prices make.
Few people will switch from petrol to electric just because of higher
priced fuel. They need information about all the factors which
determines a vehicle’s suitability.
•
Recently, milk processing/marketing company Synlait announced the
construction of an electric milk dryer rather than one using gas or coal to
provide heat. First NZ Capital analysed Synlait’s decision; looking at the
relative costs of coal and electric plant and fuel. They also estimated
what it would take for Synlait (and other milk processing companies) to
replace existing coal plant with the electric equivalent.
The graph below shows First NZ Capital’s estimated “indifference curve”.
Where they feel someone looking to dry milk will be indifferent
between using coal or electricity. The calculations are loose, but they
illustrate the relationship between the cost of electricity and the cost of
emissions.

12
INFRATIL
Synlait requires high temperature boilers to dry milk, many other situations
will have economics that are more attractive, for instance heat pumps.
Over time, motorists, manufacturers, and households will do versions of
the above analysis as they look to buy vehicles, plant, equipment and
home appliances. Because the price of energy and emissions are only two
of many variables the relationship will change over time, but as Synlait’s
choice shows, the cost of emissions has become a factor influencing
investment decisions.
THE FINISH LINE: WHAT RETURNS CAN BE ANTICIPATED FROM
INVESTING IN ELECTRICITY GENERATION IN NEW ZEALAND?
It is reasonable to be positive, but there are risks. The positive future
looks thus:
•
Electricity demand rises as motorists, manufacturers, and households
switch from coal/oil/gas to electricity and economic growth is
maintained.
•
Investors and lenders are encouraged by low regulatory risks to provide
the investment necessary to deliver increased electricity output.
•
Electricity prices rise modestly to underpin returns.
A benign picture of a growth sector with investment opportunities and
satisfactory returns.
However, there are risks:
•
If government imposes distortionary policies or goals, such as “zero
electricity sector emission by 2030”, it will increase the risks for investors
and hence required returns.
•
Higher required investment returns will raise the cost of electricity,
which means that stronger measures will be required to reduce CO
2

emissions.
•
Leading to an increased potential for emission reductions to come from
industrial closures, restrictions and declining economic activities.
Fortunately, the “benign picture” is simple and obvious. While the Update
started by noting the experience of zealous and misdirected regulators in
other countries, there is no reason to anticipate those errors being made in
New Zealand.
COMPARING POWER BILLS
We asked friends and relatives in a number of cities to send us copies of
their most recent electricity bills. They make interesting reading, if only to
illustrate the range of outcomes possible. Some of the price differences
reflect different fuel costs, but many are due to regulatory differences.
Wellington: Grey Power Electricity
31 days 660kWh $164.06 (pre GST) 24.9cents/kWh
CostCents/kWh
Energy$53.618.1
Distribution$81.0312.3
Retail & metering$28.684.3
Govt levy$0.720.1
GST$24.613.7
Melbourne: Citipower (exchange rate NZ$/A$ 0.90)
74 days 1,641kWh $373.12 (pre GST) 22.7cents/kWh
CostCents/kWh
Everything$373.1222.7
GST$37.312.3
San Francisco (PG&E) (exchange rate NZ$/US$ 0.70)
34 days 759kWh $301.72 39.8cents/kWh
CostCents/kWh
Energy$116.8915.4
Distribution$128.2016.9
Govt levy$56.637.5
New York (Consolidated Edison) (exchange rate NZ$/US$ 0.70)
32 days 259kWh $108.69 42.0cents/kWh
CostCents/kWh
Energy$35.3313.6
Distribution$43.6916.9
Retailer costs$27.2710.5
Clean energy levy$2.511.0
Tax$9.693.7
Berlin: LichtBlick (exchange rate NZ$/Euro 0.58)
365 days* 2,036kWh €955.57 (pre GST) 46.9cents/kWh
CostCents/kWh
Energy$458.0522.5
Distribution$57.522.8
Metering$14.410.7
Renewable & other levies$425.5920.9
VAT$181.558.9
* Consumption and cost is assessed annually and billed monthly
Surrey, England: Co-op Energy (Exchange rate NZ$/GBP 0.52)
129 days 1,619kWh $552.08 (pre VAT) 34.1cents/kWh
CostCents/kWh
Energy$451.7027.9
Distribution$100.386.2
VAT$25.971.6

INFRATIL UPDATE SEPTEMBER 2018