That surplus has contributed to sharp falls in the average
selling prices of solar panels, or modules, with knock-on
impacts for manufacturing profit margins.

Strategies to maintain margins, as prices continue to fall,
include mothballing or closing existing factories.

But there is a delicate balancing act where no company wants
to give up its share of a global market which grew at a compound
annual rate of 52 percent from 2002 to 2012 (Chart 1).

Companies are choosing between market share and profit.

Evidence from the larger listed manufacturers suggests that
leading companies at present are split on strategy, with some
continuing to ramp up loss-making capacity, while others have
shelved expansion plans, but only a small minority have
mothballed or closed factories.

That may not bode well for the wider, global industry, where
leaner capacity will be the main route to a return to
profitability, and suggests more value destruction to come as
companies continue to take provisions on inventory, close
factories or file for bankruptcy.

***************************************

Chart 1: (page 14)

Chart 2:

***************************************

CAPACITY CHANGES

The stratospheric growth of the solar industry is
illustrated by the recent expansion of seven of the top 10
producers by shipments which publish relevant data. (See Chart
2)

Those companies were: First Solar, Hanwha Solar
, JA Solar, Jinko Solar, Trina,
Yingli and Canadian Solar.

The manufacturing capacity of those seven module makers
alone at the end of last year corresponded to 44 percent of
actual global demand for solar panels in 2012.

That suggests the scale of industry overcapacity, given
there are 100 or more smaller module manufacturers worldwide.

The seven had combined module production capacity of 13,650
megawatts (MW) as of December, their financial reports show,
compared with actual global demand (as recorded in installed
capacity) last year of 31,095 MW, according to the European
Photovoltaic Industry Association (EPIA).

Regarding their manufacturing strategy, two of the seven cut
capacity in 2012, one left capacity unchanged, and the remaining
four expanded.

Their total manufacturing capacity in aggregate grew by 12
percent or by 1,470 MW.

Three companies provided forecasts for manufacturing
capacity expectations in 2013, two expecting this to remain
unchanged (Jinko and JA Solar) and one possibly to expand
slightly (by 4.2 percent, Trina Solar).

STRATEGY

The leading companies are not a representative sample: they
are the firms with the most resources and clout to ride out the
solar shakeout, and so may be expected to contract less.

There have been plenty of bankruptcies among weaker players
and no doubt capacity destruction is proceeding.

Two opposing strategies now for surviving the shakeout and
becoming a leader in a subsequent, consolidated industry might
be: first, to build a leaner, more sustainable business, or,
second, to continue ramp up in the hope of a revival in prices
soon after more companies have gone to the wall.

An example of the latter, “keep expanding” approach might be
Yingli, as suggested by the trajectory of its continued rapid
growth in factory capacity and supported by statements in its
2012 annual report.

“The size of manufacturing capacity has a significant
bearing on the profitability and competitive position of PV
product manufacturers. Achieving economies of scale from
expanded manufacturing capacity is critical to maintaining our
competitive position,” it said.

Danish wind turbine maker Vestas might be an
exponent of the more cautious, “focus on profit” approach in its
latest annual report.

The wind industry has dealt with similar problems to solar,
including coping with falling power demand and subsidy cuts in
key markets plus global over-capacity.

“We should not aim for higher revenue at any cost. Vestas
will only embark on projects that are profitable to our
customers and our business,” it said in a report which announced
cost cuts following a drop in new orders.

One alternative to expensive capacity expansions is to
outsource more of the supply chain, an approach adopted by
Vestas.

Leading module maker Canadian Solar appears to be an example
in the solar industry, recently announcing that it would achieve
a ten-fold increase in its production capacity of wafers (an
intermediary product in module manufacture) over the next two
years partly through external relationships.

Of the near-2,000 MW expected expansion, 300 MW would be
made internally, 600 MW through a joint venture with GCL Poly
Energy Holdings, and a further 1,000 MW through
long-term supply contracts, it said in its March 2013
presentation.

That could be a safe, half-way house, retaining the
flexibility to ramp up when prices settle and in the meantime
focusing on profitability.

They could achieve even greater benefits by going a step
further, following the lead of their counterparts in Europe, by
linking wholesale markets.

In the United States, operators called balancing area
authorities (BAAs) control electricity transmission, with each
responsible for keeping supply and demand in sync at all times
within its defined area.

The Western Interconnection region, which includes 14
western U.S. states, two Canadian provinces and portions of one
Mexican state, contains 38 such BAAs, linked by interconnectors.

The BAAs make decisions to dispatch power, both within and
between their areas, typically once every hour. Any mismatch in
the meantime is covered by firing up power plants, such as
so-called peaking gas plants, within the balancing area.

The new approach, called an Energy Imbalance Market (EIM),
is based on closer links between BAAs and between grid
operators. It is part of a trend towards integrating wind and
solar power into grids that were mainly designed to handle
steadier sources of supply.

Under a U.S. EIM, a computer algorithm would establish
whether or not there is an energy imbalance between grids and
issue orders in five-minute intervals to power generators across
multiple BAAs to dispatch power as needed at least cost.

The system is voluntary. Power generators can dip in and
out, offering electricity for sale on the EIM or not.

Grid operators gain access to a wider geographical spread of
generation resources and so avoid having to build more reserve
capacity. The more frequent dispatch decisions also increase
predictability and can help utilities avoid having to call up
more expensive, fast-starting power plants.

A recent report by the National Renewable Energy Laboratory
(NREL) calculated that the annual cost savings from more
frequent, faster dispatch decisions would amount to $1.3 billion
in the Western Interconnection region. (“Examination of
Potential Benefits of an Energy Imbalance Market in the Western
Interconnection,” NREL, March 2013)

In addition, the region would save another $146 million a
year from the sharing of resources across a wider area, the NREL
estimated.

Balancing across a wider area is useful to smooth out the
variability of renewable power.

For example, a local effect such as a cloud can cause a big
change in the amount of power generated by one solar farm, but
that variability can be reduced by tapping a number of solar
farms over a wider area.

Two Californian balancing authorities, PacifiCorp and CAISO
(the biggest BAAs in the state), announced a memorandum of
understanding in February 2013 committing to work jointly toward
creating an EIM by October 2014, in a sign of early commitment
to proceed with the wider market.

IN EUROPE

The European Union approach, called market coupling, creates
links between wholesale power markets which often correspond to
EU members’ national borders.

It also uses a computer algorithm to match interconnector
capacity with energy demand data, as supplied by grid operators.

The difference with the U.S. EIM approach is that the
resulting electricity flows are marketed to traders on wholesale
power market exchanges, for them to buy and sell in various
futures contracts.

Under an EIM, the electricity flows are not made available
on wholesale power markets to traders; they remain confined in
bilateral deals between generators and energy consumers.

The purpose of an EIM is to ensure that unfulfilled demand
is matched better with available supply at the lowest price.

The European market coupling system in addition ensures that
power flows from cheaper to more expensive markets to achieve
the aim of price convergence.

This approach may not yet be possible in the United States,
however, because each major region includes a mix of regulated
markets and fully competitive, wholesale markets.

The EU example raises the question whether the U.S. drive to
integrate more renewable power should also aim to speed up the
rollout of competitive wholesale markets and achieve even
greater social benefits by driving down the prices of relatively
expensive areas.

(editing by Jane Baird)

CCS captures carbon dioxide (CO2) emissions from a fossil
fuel power plant and then pipes it to an underground storage
site such as a depleted gas or oil reservoir.

In theory, CCS would allow energy producers to continue to
burn fossil fuels and still meet carbon emission targets. In
practice, the technology is expensive and unproven.

Clustering a number power plants at one end of a central
pipeline and multiple storage sites at the other end would save
money over the long term. More savings could from using CO2 to
help produce oil from aging wells, which would generate revenue
to help cover the costs.

Not only are starting costs higher but also these options
entail a risk that the bigger and longer pipelines required
would never be fully used if the technology fails to prove
itself.

Britain’s Department of Energy and Climate Change (DECC)
published a report this week on the potential for reducing the
costs of CCS, drafted by an advisory group it established a
year. (“The Potential For Reducing The Costs of CCS in The UK”,
May 2013)

The report found two main options for near-term cost cuts.
(See Chart 1) but concluded that even after these savings, CCS
would require some kind of government subsidy through the 2020s.

**************************************

Chart 1: (page 15) goo.gl/Ui6r8

**************************************

CLUSTERS

One cost-cutting option would be to cluster CCS projects
into hubs, which would allow the sharing of pipeline and storage
infrastructure.

“Virtually all of the CCS projects proposed in the UK to
date are based on isolated full-chain schemes in which a single
power station is connected via a single dedicated CO2 pipeline
to a storage site in the UK Continental Shelf,” it said.

“Leveraging early CO2 infrastructure, if it is designed
correctly, can reduce the incremental cost of transport and
storage substantially for later projects.”

The report estimated that clustering would reduce storage
costs from around 25 pounds ($38.27) per megawatt-hour in early
projects to 5 to 10 pounds/MWh. That compares with total CCS
costs estimated at 161 pounds/MWh before any savings.

Transport costs could drop from 18 to 23 pounds/MWh for
early projects carrying 1-2 million tonnes of CO2 per year to 5
to 10 pounds/MWh for large, full pipelines carrying 5-10 million
tonnes, the report estimated.

A report published last year by consultants Mott MacDonald
summed up the benefits and risks from investing more, earlier in
outsized infrastructure. (“Potential cost reductions in CCS in
the power sector”, May 2012)

“In principle, additional sources can be added in the
future, provided CO2 pipeline capacity is sized and designed
accordingly. A coordinated network approach can then lower the
barriers of entry for all participating CCS projects, including
for emitters who subsequently do not have to develop their own
separate transportation and storage solutions,” it said.

“These benefits arising from economies of scale need to be
weighed against the risk of being left with an underutilised or
stranded asset. This means that the development of transport and
storage clusters needs to be carefully coordinated with
capture.”

ENHANCED OIL RECOVERY

Another way to cut costs is to use the CO2 to force out the
dregs of crude oil from nearly depleted fields, through the
enhanced oil recovery process (EOR).

It would require greater upfront investment, by comparison
with non-EOR CCS, to build longer pipelines to transport the CO2
to suitable oilfields. Such a project may need a cluster of at
least 10 oilfields to be economic, according to an industry
source.

But Britain last October dumped the only EOR proposal when
it put together a short list of projects for CCS funding,
presumably on the basis that it was too expensive.

The rejected Don Valley EOR project proposed to transport
the CO2 400 km to two North Sea oil fields.

As for the two that made it to the short list, the Peterhead
project in north-east Scotland would involve a 100 km pipeline
to a depleted gas reservoir, and the White Rose project in
eastern England a roughly 165 km line to a saline aquifer.

The advantage of EOR clearly is that it would generate oil
revenue, which could pay for the entire CO2 storage and possibly
some transport costs, Thursday’s report estimated.

“(The) potential additional EOR benefit is in the range of
5-12 pounds/MWh for gas CCS and 10-26 pounds/MWh for coal CCS,”
it estimated.

It appears that Britain is still far from grasping the
nettle of large, upfront costs, even though they would cut costs
over the longer term.

The implication is that CCS development will remain
piecemeal and more expensive, with a slower larger rollout, if
at all.
($1 = 0.6533 British pounds)

(editing by Jane Baird)

Until now, wind power has been the leading low-carbon
alternative to oil, coal and gas, outside large niche markets
such as Germany, which has seen a huge ramp-up in installed
solar.

But that could change, with deep implications for the health
of both industries if one substitutes the other.

As soon as this year, solar could for the first time surpass
wind in annual global installed capacity, given an expected
contraction in the wind market.

The full costs of wind power generation remain less than
solar because of higher productivity and lower installation
costs, but those advantages are eroding rapidly given current
trends in equipment prices, with a glut of Chinese-made solar
panels sending prices tumbling.

OVER-CAPACITY

Both sectors have struggled to shake off over-capacity, but
in different ways, and with different outcomes for prices.

In solar power, over-capacity is largely a result of global
commoditisation of the finished modules and the influx of cheap
Chinese producers, which now dominate the market. (See Chart 1)

The result has been bankruptcies and continuing, sharp falls
in solar panel, or module, prices.

In wind, over-capacity is a result of subsidy cuts and lower
energy demand in western markets following the financial crisis,
coupled with competition from low gas prices in the United
States and a stagnating Chinese market.

But wind turbines are heavy, complicated pieces of moving
machinery in which sales depend on local servicing.

They have not become commoditised, and three western
producers lead the world market: Vestas, GE and
Siemens. (See Chart 2)

The result of over-capacity has instead been a variety of
cost-cutting strategies, including idled production, while
prices have remained stable.

***************************************

Chart 1:

Chart 2:

Chart 3:

***************************************

AVERAGE PRICE

Recent manufacturing data show that average selling prices
for solar panels, or modules, are now lower than wind turbines
per watt in some markets.

For example, one of the world’s leading turbine makers,
Denmark’s Vestas, reports average selling prices for the first
quarter of this year at 1.04 euros ($1.34) per watt.

That compares with solar module prices among leading
manufacturers at $0.7-$0.8 per watt last year.

The pace of module price falls means that these will also
rival wind turbines in lower-priced Chinese markets this year.
(See Chart 3)

The short-term price trajectory for each suggests that solar
modules will continue to fall, although much more slowly than
previously, while wind turbines remain flat.

Because of their far bigger size, wind turbines are still
cheaper to install per watt.

The installed cost of wind is around 1.25 euros ($1.61) per
watt, according to a recent report by the trade body the Global
Wind Energy Council.

Data for the installed cost for utility scale solar is hard
to glean; Britain’s Department of Energy and Climate Change
reported a total installed cost for solar PV projects over 250
kilowatts as low as 1.2 pounds ($1.83) per watt as of March last
year.

Full energy costs, per generated kilowatt hour, depend on
performance and capacity factor as much as equipment and
installation costs.

Again wind is cheaper but solar PV is catching up, with wide
ranges depending on location.

Wind turbine manufacturers are targeting both incremental
technology development in existing platforms as well as more
fundamental advances, in particular towards bigger turbines for
the offshore market.

Linear growth is expected in the market share for turbines
larger than 3 MW, reaching half the global market by 2020,
according to Madrid-based turbine maker Gamesa.

In the case of solar module makers, the technology focus has
been steady improvements in the efficiency of conversion of
sunlight into electricity.

ANNUAL MARKET

It is plausible that the annual solar market (meaning new
installed capacity) this year will leapfrog wind for the first
time.

Wind turbine makers report a global market of around 48
gigawatts installed last year, but expect that to drop following
a slump in orders in 2012.

Vestas new orders halved last year compared with 2011.

Vestas quoted a forecast by independent research company
Emerging Energy Research for a steep decline in the global
turbine market to 39 GW in 2013. Gamesa has forecast a shallower
decline, to 42 GW.

Meanwhile, the European solar trade body, the European
Photovoltaic Industry Association, estimated last year’s solar
market at 31.1 GW.

It forecasts the market this year at 28-47 GW, a wide range
which especially reflects some uncertainty over the level of
European solar subsidies.

Either way, the solar market will probably pass wind this
decade.

That has implications for the speed of growth for each and
the fortunes of equipment manufacturers, as well as for the cost
of low carbon power.

The case involved domestic content requirements, which block
foreign companies from participating in national support
programmes, and which are intended to suckle local
manufacturers.

The World Trade Organisation’s rejection of Canada’s appeal,
in a complaint brought by Japan and the European Union, should
clarify similar cases, where China claims discrimination in
certain EU countries and the United States has opened a dispute
with India.

It does not, however, address a growing, separate set of
more complicated trade complaints regarding perceived dumping
(which means selling at a loss to gain market share) where
China, India, the European Union and the United States are all
considering or have imposed import duties.

The global industry needs similar clarity on dumping, to
avoid a solar trade war where Germany’s preference for an
amicable resolution with China is encouraging.

Both types of dispute expose a tension in the goals of
renewable energy programmes between jobs and low-cost clean
power.

One aim is to boost “green jobs”, where countries are keen
to grab a foothold in a growing clean technology sector, which
clashes with another to cut costs.

Rules preferring domestic content make little sense right
now, if the aim is to ramp up domestic solar manufacturing when
the global industry is embroiled in a savage shakeout of
over-capacity where some distressed companies are selling
photovoltaic (PV) modules at fire sale prices.

Action against dumping, meanwhile, may directly raise the
price of solar modules through the use of import duties.

ONTARIO

The Ontario case was based on the Canadian province’s “Green
Energy Act” of 2009.

The Act allowed developers to earn a price premium, or
feed-in tariff, for renewable power if they met local content
rules.

Those minimum domestic content requirements were 50 percent
for wind projects and 60 percent for solar, as published in an
Ontario Power Authority “programme overview”.

Japan’s complaint was partly based on Article 3.4 of the
1994 General Agreement on Tariffs and Trade (GATT).

The paragraph states: “The products of the territory of any
contracting party imported into the territory of any other
contracting party shall be accorded treatment no less favourable
than that accorded to like products of national origin in
respect of all laws, regulations and requirements affecting
their internal sale, offering for sale, purchase,
transportation, distribution or use.”

Japan’s complaint – subsequently joined by the European
Union – was supported by a WTO resolution panel in December, and
upheld last week against Canada’s appeal.

NEW CASES

The finding is likely to clarify similar complaints.

For example, India has announced a goal to install 22
gigawatts of grid-connected and off-grid solar power by 2022,
under its Jawaharlal Nehru National Solar Mission (NSM).

Under phase 1, which ends this year, developers could only
use solar modules made in India.

Phase 2 may relax the requirement that all projects meet
domestic content requirement (DCR) rules.

“Some capacity will be kept for bidding with domestic
content requirement,” says a phase 2 consultancy document
published on the ministry’s website, suggesting other capacity
will be available for foreign firms.

India has proposed to widen DCR to cover more advanced thin
film technologies where U.S.-based First Solar is a
world leader.

The Ontario case may not bode well for the Indian programme.

The United States formally initiated a dispute at the WTO on
Feb. 11, by requesting consultations with India in the same
manner as Japan had with Canada in September 2010.

“India’s measures appear to be inconsistent with Article 3.4
of the GATT 1994 because the measures appear to provide less
favourable treatment to imported solar cells and solar modules
than that accorded to like products originating in India,” it
said, referencing identical articles and trade measures to
Japan’s complaint.

Similarly, last November China requested consultations with
the European Union, Greece and Italy, “relating to the feed-in
tariff programs of EU member states including but not limited to
Italy and Greece”.

Italian energy laws have provided a premium on the feed-in
tariff of up to 10 percent for solar PV systems containing a
minimum 60 percent European content by value, various solar
companies report.

TENSION

The latest WTO ruling will probably benefit the industry if
it motivates countries to source solar modules on the basis of
lowest price rather than origin.

But public pressure remains to do the opposite, as
illustrated by a letter in March from an influential group of
U.S. environmental and development non-government organisations
to the U.S. chief trade negotiator.

The letter, signed by 350.org, ActionAid USA, Friends of the
Earth U.S., Greenpeace USA, Sierra Club, among others, expressed
“deep concern” about the country’s solar dispute with India.

They argued that India should be encouraged to develop a
domestic solar industry for three reasons: to increase the share
of solar in the Indian energy market; increase political support
for clean energy programmes by creating local jobs; and to avoid
catastrophic climate change.

It did not mention an impact on costs.

India presently has manufacturing capacity of 848 MW for the
production of solar cells and 1,932 MW for solar modules, the
Ministry of New and Renewable Energy reported last December in
its “Phase II – Policy Document”.

It is targeting by 2020 manufacturing capacity for
4,000-5,000 MW of solar technologies (this appears to refer to
finished modules) and 2,000 MW of solar cells, according to an
overview published on the ministry website.

Just now may not be the optimum time for ramping up
manufacturing capacity, however, either for established
companies already bleeding cash, or for countries planning a new
industry where they can buy more cheaply from companies with
mothballed capacity selling at distressed prices.

The impact of proliferating proposed import duties is a
major unknown, however.

Solar panel, or module, prices have fallen sharply in the
past five years, in response to global over-supply coupled with
falling subsidies and the economic crisis in the world’s biggest
market, Europe.

Falling prices for the raw material polysilicon, which
reached record lows last November have sapped prices for the
finished modules, say manufacturers.

These factors have narrowed module manufacturing margins and
cut solar power generation costs for end users, in turn leading
to a pull-back in subsidies which has further limited module
demand and prices, combining to pose the question: when will
prices stabilise?

China’s Yingli, one of the top module manufacturers
by volume shipped, pointed to a stabilisation of prices in an
upbeat assessment which was not specific on timing.

“We expect that the prices of PV products, including PV
modules, may gradually stabilize or increase slightly due to the
increased demand caused by the quick development of new and
emerging markets and the rationalized supply of PV products on
the market,” the company said in its latest annual report
published in March.

It seems likely that price falls will slow, but rises soon
are unlikely given data which show a continuing fall in
manufacturing costs and selling prices.

HISTORICAL PRICES

Average selling prices for silicon-based modules, as
reported by top-10 manufacturers, have fallen from around
$3.5-$4.5 per watt in 2008 to $0.7-$0.8 per watt last year (See
Chart 1)

Price falls can be attributed both to competitiveness
pressures and structural falls, the latter following advances in
technology and benefits from growing economies of scale.

“We price our standard PV modules based on the prevailing
market prices at the time we enter into sales contracts … and
our silicon-based raw materials costs,” said China’s Trina Solar
, in its latest annual report.

Solar panel prices have fallen quicker than manufacturing
costs, showing the impact of competitiveness pressures.

Yingli’s quarterly and annual reports show average annual
manufacturing costs per watt fell 35 percent in 2012 versus
2011, compared with a 46 percent drop in average selling prices.

***************************************

Chart 1:

Chart 2:

Chart 3:

***************************************

STILL FALLING

Future module prices will be set by a combination of factors
pushing in different directions.

Arguing for a levelling off in module prices, gross margins
have narrowed sharply in the past two years, and these are in
some cases now negative. (See Chart 2)

And average selling prices across manufacturers have
converged, suggesting companies are now operating at the limit
of available technology. (See Chart 1)

Some manufacturers point to stabilisation in prices over the
course of last year.

“Since the first quarter of 2012, the price of PV modules
has remained at a relatively stable level primarily due to the
shake-out of certain uncompetitive production capacity and
increased demand worldwide,” said Yingli.

Arguing for continuing price falls, manufacturing costs are
still falling quarter on quarter, show data from Yingli,
Canadian Solar and JinkoSolar. (See Chart 3)

And module prices fell last year on an annual basis.

Perhaps most importantly, over-capacity remains.

IMPORT DUTIES

The big unknown is a raft of mooted trade protection.

Threatened duties could separate the global market into
silos, where each jurisdiction would favour local manufacturers,
leading to higher selling prices as a result of less competition
and the effect of the duties on import prices.

However, prices could fall, if trade restrictions deepened
local over-capacity by constricting demand.

It is too soon to predict the outcome, as Europe awaits
confirmation of rates, litigation continues in the United States
regarding duties confirmed last year, while China mulls duties
on solar raw materials and India weighs in with its own
threatened trade protection.

The reaction of Chinese manufacturers underlines the threat
which features highly in the “risk factors” section of their
annual reports.

“If as the result of such investigations the European
Parliament imposes anti dumping and countervailing tariffs or
other trade protection measures, our exports to the European
market may be materially and adversely affected,” said Yingli,
40 percent of whose revenues are from Europe.

“The imposition of the AD (U.S. Anti dumping Duty Order) and
CVD (Countervailing Duty Order) and any increase in the rate of
AD or CVD or their respective deposit ratios will significantly
increase the price of our products, negatively affect our sales
in the U.S. and adversely affect our results of operations,”
said China’s Hanwha SolarOne

That was before the U.S. and European imposed and planned
import duties against Chinese products, which may hinder
exporters seeking to tap one of the world’s fastest growing
markets.

The European solar market shrank last year for the first
time this century, and this year will be less than half the
global market for the first time since 2003, according to data
and forecasts from the trade body, the European Photovoltaic
Industry Association (EPIA). (See Chart 1)

Meanwhile meteoric U.S. market growth of some 76 percent in
2012 will slow, exposing First Solar which derived 56 percent of
total net sales last year from utility-scale projects in
California.

Arizona-based First Solar is one of the world’s largest PV
module manufacturers and also builds utility-scale projects for
financial and energy company buyers after its acquisition of two
leading U.S. developers.

It has signalled a shift away from Europe, in particular
after Germany scrapped support for large-scale projects, closing
its manufacturing operations in Frankfurt at the end of 2012.

“We continue to shift our selling efforts from the European
markets, in which we have historically generated a significant
portion of our net sales, to new markets, in which we have not
historically generated any meaningful portion of our net sales,”
it said in February.

It is instead targeting a combination of sun-belt and high
growth markets, such as China, India, the Middle East, South
Africa and Chile, which it terms “new sustainable markets”.

It defines these as markets where: “demand for power that
exceeds current supply; high existing power prices; and abundant
sunshine”, in its 2012 annual report.

The strategy reflects the desperate need of the global
industry to find new demand to mop up a global glut in modules,
where subsidised markets in Europe and the United States are
waning or slowing and anyway fail to match product supply.

**************************************

Chart 1: (page 34)

Chart 2: (page 51)

**************************************

CALIFORNIA

First Solar presently has a U.S. pipeline of massive
projects which it expects to complete by 2015.

They are: a 550 megawatt (MW) project in San Luis Obispo
County, California (Topaz Solar Farm); a 550 MW project west of
Blythe, California (Desert Sunlight Solar Farm); a 290 MW
project in Yuma County, Arizona (Agua Caliente); and a 230 MW
project located just north of Los Angeles, California (Antelope
Valley Solar Ranch One, or “AVSR1″).

The grid-connected portion of the uncompleted Agua Caliente
project is already the largest operating PV power plant in the
world, it says.

Beyond 2015, the United States will continue to represent a
significant portion of the company’s sales.

It has added over the past year new permitted projects in
California and New Mexico which are all smaller than the above,
including: Campo Verde (139 MW); PNM II (22 MW); Macho Springs
(50 MW); Lost Hills (32 MW); and Cuayama (40 MW).

Its less advanced project pipeline, where it has executed
power purchase agreements but not yet secured permits, includes
bigger projects including the 300 MW Stateline project in
California and the 250 MW Silver State South project in Nevada.
(See Chart 2)

NEW MARKETS

First Solar announced at the end of 2011 its long term
strategic plan (LTSP) “to transition primarily to sustainable
opportunities” by the end of 2016.

“Execution of the LTSP will entail a reallocation of
resources around the globe, in particular dedicating resources
to regions such as Latin America, Asia, the Middle East, and
Africa where we have not traditionally conducted significant
business to date,” the company said in an update three months
ago in its 2012 annual report.

A year ago, First Solar had a presence in the United States,
Canada, India, Europe and Australia.

As of its first quarter results, earlier this week, it said
it had added South America, China, the Middle East and North
Africa to the list.

Some of these new initiatives are necessarily limited, as
early stage ambition.

China was the world’s second biggest solar market after
Germany last year, installing some 5,000 MW, according to the
EPIA.

In December, First Solar announced its first commercial
demonstration project, in a collaboration to supply 2 MW of
modules to local developer Zhenfa New Energy Science &
Technology Co. Ltd.

In the United Arab Emirates, First Solar was selected by the
Dubai Electricity & Water Authority to construct a 13 MW power
plant near Dubai, the first stage in the Emirate’s plans for a
solar park with 1,000 MW capacity by 2030.

In its most significant expansion, it announced in January
its acquisition of Solar Chile, a Santiago-based solar
development company with a portfolio of projects at various
stages totalling approximately 1,500 MW in northern Chile,
including the sun-drenched Atacama Desert region.

LOCAL CONTENT

First Solar itself acknowledges the wider risks to its
strategic plan of focusing on new regions, including: grid
stability in emerging economies; local content policies limiting
module imports; the availability of local skilled staff; and
local competition.

Regarding local content, South Africa has ambitious targets
for renewable energy by 2030 in its Integrated Resource Plan.

Published data show rising local content at 29 percent and
48 percent in bidding rounds one and two of the plan which
together allocated 1,049 MW of solar PV.

Regarding inertia, the Middle East presents uncertainty over
the timing of implementing ambitious targets.

A Saudi government advisory body last year recommended a
target for about 41,000 MW of solar energy by 2032, but the
country has installed less than 15 MW.

“While the expected potential of the MENA (Middle East and
North Africa) markets is significant, policy promulgation and
market development are especially vulnerable to governmental
inertia, political instability, geopolitical risk, fossil fuel
subsidization, potentially stringent localization requirements
and limited available infrastructure,” First Solar said in
February.

The higher cost can probably be attributed to its choice of
concentrated solar power (CSP), the competitiveness of which is
being questioned as prices of rival photovoltaic (PV) technology
tumble.

Morocco plans to install at least 2,000 megawatts (MW) of
solar power capacity by 2020 at five sites, which it hopes will
account for 14 percent of total power generating capacity by the
end of the decade.

It will target both CSP and solar PV in its Moroccan Plan
for Solar Energy, or “Solar Plan”, according to the Moroccan
Investment Development Agency, but has so far veered towards
more expensive CSP at one initial project near the southern city
of Ouarzazate.

That contrasts with how developers in California have
increasingly ditched CSP for PV over the past three years as a
global manufacturing glut sent PV costs plummeting.

CSP uses parabolic or other types of mirrors to concentrate
sunlight and create heat and steam to drive a turbine. It is a
technology championed by power equipment producers in Spain,
Morocco’s neighbour and one of its closest diplomatic allies.

Solar PV converts sunlight directly into electricity using a
light-sensitive semiconductor such as silicon.

Morocco would do well to switch to PV, given a far more
developed supply chain, commoditised end product and competitive
power generation, especially given that the country’s economic
troubles make it riskier to experiment with less widely used
technologies.

POWER PURCHASE

Moroccan authorities anticipated the solar plan would cost
$9 billion at its launch in 2009, according to data on the
website of the north African country’s solar energy agency,
Masen.

The plan will be part-financed by a $1 billion Energy
Development Fund, including donations from the Kingdom of Saudi
Arabia and the United Arab Emirates.

Morocco has raised additional funds for the first Ouarzazate
project from institutions including the European Investment
Bank, the French Development Agency (AFD) and the German
development bank KfW.

The plan will be delivered through 25-year power purchase
agreements (PPA) with independent power producers which earn a
fixed rate per unit of solar power they generate.

The aim is to achieve greater energy independence: Morocco
is one of the world’s most energy-poor countries, importing
around 95 percent of its needs, according to the World Bank.

Energy accounts for more than a quarter of the country’s
imports and contributed to a record trade deficit of $23.6
billion last year.

SUNNY

Sun-drenched Morocco wants eventually to export its solar
energy to Europe.

The U.S. National Renewable Energy Laboratory (NREL) has
developed an open-access database measuring solar irradiance
calculated according to the sunlight captured by a panel tilted
southwards, defined in units of kilowatt hours per square metre
per day.

The NREL’s map shows a maximum “Direct Normal Irradiance”
(DNI) in southwest Germany of 3.39, compared with 7.47 at
Ouarzazate in Morocco. (See Chart 1)

It shows that the Moroccan project should achieve far more
competitive power generation.

But Morocco announced last September that it had awarded a
group led by Saudi International Company for Water and Power
(ACWA) a $1 billion contract to build a 160-megawatt (MW) CSP
plant, at 1.62 dirham ($0.194) per kilowatt hour.

ACWA last week confirmed further details including the award
of construction contracts.

That is more support than for smaller PV installations in
Germany, at 0.134 euros ($0.18) per kWh for projects up to 10
MW, and in Britain, at 0.115 pounds ($0.18) for projects above
250 kilowatts, both of which are over 20 rather than 25 years.

Germany has scrapped support for projects over 10 MW.

*************************************

Chart 1: goo.gl/Kks9E

Chart 2: (page 4) goo.gl/EA54k

*************************************

CSP VS PV

One way to compare costs of PV relative to CSP is using a
measure called levelised cost of energy (LCOE), based on total
lifetime costs and energy generation.

Germany’s Fraunhofer Institute last year published a
detailed LCOE study including a comparison of CSP and PV costs
in North Africa, where its central estimate for CSP at 0.163
euros ($0.21) closely matched the contract subsequently agreed
in Morocco.

The Fraunhofer analysis found that CSP was almost twice as
expensive as solar PV and three times more so than wind, and
forecast it would remain far more costly than PV through 2030.
(Chart 2)

“A considerable reduction in costs in recent years has given
PV installations a cost advantage over CSP plants at the same
location,” it concluded.

Masen is aiming to install some 500 MW capacity at its
Ouarzazate site, and to start commissioning this by 2015.

The project will be delivered in three parts: an initial 160
MW of CSP using conventional parabolic mirrors (the contract now
awarded), and two subsequent phases for 300 MW of CSP including
a project using innovative solar tower technology which is even
more expensive.

That appears to leave less than 50 MW for solar PV.

The main advantage of CSP is its ability to control when the
energy is released for consumption.

The first contracted CSP plant at Ouarzazate in Morocco will
provide three hours of storage, according to a press release
published by MASEN on Jan. 23, meaning it can tweak output to
match consumer demand like a conventional power plant.

PV electricity must be consumed at the instant of generation
unless the grid has some kind of storage such as pumped
hydropower.

But the ability to store solar power, known as
dispatchability, is hardly a priority at this early stage when
Morocco’s grid is dominated by dispatchable coal and oil-fired
power.

Masen’s plan to export some of its solar power to Europe via
Spain to help plug the trade deficit appears far-fetched given
how expensive it will be to produce.

At the agreed price, the solar power may also be a drain on
the Moroccan budget if the government has to cover any gap
between the cost of producing solar electricity and the price
the state power utility will pay.
($1 = 0.7642 euros)
($1 = 0.6464 British pounds)

(Reporting by Gerard Wynn; Editing by Tom Pfeiffer)

Countries globally are seeking to expand transmission
capacity, to connect more remote renewable power; link networks
across wider areas to boost stability; and in emerging economies
draw more customers to the grid.

The International Energy Agency estimates some $7.2 trillion
of grid investment is needed globally through 2035, in its
latest World Energy Outlook, which is comparable with that
required in power generation.

Grid investment requires government intervention, given its
massive scale and non-market public benefits for example to
boost security of supply or cut carbon emissions.

Policymakers are turning to private sector finance, funded
by levies on consumer electricity bills, against a backdrop of
tight fiscal policy, and to pension funds in particular given a
pullback in bank lending and a perceived fit with their
long-term liabilities.

The U.S. Mid-Atlantic and Midwest electric grid operator PJM
Interconnection this week opened a new process to allow utility
and non-utility competitors to propose how to meet power system
needs in specific areas.

That competitive approach was prompted by the U.S. Federal
Energy Regulatory Commission’s (FERC) Order 1000, which allows
operators to use bidding to solicit investment in big projects.

Britain has lessons on how to avoid over-payment by
consumers after criticisms of its regime for public contracts
for offshore wind transmission.

UK OFFSHORE WIND

Britain is hoping offshore wind will supply a large portion
of its low-carbon electricity commitments under European Union
renewable energy targets in 2020.

As of the end of 2011, the country had installed 1.8
gigawatts (GW), up 37 percent on the previous year, according to
statistics from the Department for Energy and Climate Change
(DECC).

DECC’s “UK Renewable Energy Roadmap” in July 2011 forecasts
a “central range” for 11-18 GW by 2020. (See Chart 1)

Britain’s National Audit Office estimates the associated
undersea cables would require investment of 8 billion pounds
($12.4 billion) through 2020.

To raise the required funds, Britain has implemented a
competitive licensing regime where investors bid for ownership
of the cable connection.

Bids are based on annual revenue requests which reflect
their estimated cost of financing the cable acquisition or
construction plus operation and maintenance.

Costs are ultimately passed to consumers through the grid
operator National Grid via electric utilities and wind
farms which pay for transmission. (See Chart 2)

That contrasts with most EU countries which have left
responsibility for connection to the grid operator, as in
Germany where disputes over liabilities arising from late cable
construction have held up development.

EU rules require separate ownership of generation and
transmission assets, meaning offshore wind farms cannot own the
subsea cable.

**************************************

Chart 1: (page 45)

Chart 2: (page 17)

Chart 3: (page 22)

Chart 4: (page 12)

**************************************

INVESTMENT RISK

Advisory firm KPMG last December published a detailed review
of the British regime for investors, “Offshore Transmission: An
Investor Perspective”, commissioned by the regulator Ofgem.

KPMG pointed out the low risk and particular benefits, for
example compared with onshore transmission investments and
public-private partnership projects (PPP).

“The evidence to date suggests that OFTOs (Offshore
Transmission Owners) offer strong returns relative to comparable
asset classes with similar risk profiles,” it said, in comments
which perhaps should have alerted the regulator that the scheme
was too generous to investors.

KPMG noted attractions including: a 20-year,
inflation-linked revenue stream; rewards for over-performance;
penalties for under-performance capped at 10 percent of
revenues; and zero exposure to the wind farm’s performance (no
construction risk).

The scheme has transferred risk to consumers who will be
left nursing the full cost if a wind farm fails, is barely used,
or if inflation spikes.

The aim appeared to be to attract competitive bids by
capping risk.

The tenders have attracted a range of bids but the KPMG
study shows that just two firms (Blue Transmission and
Transmission Capital Partners) have won nine out of 11 tenders
as of last December, for assets worth 1.36 billion pounds. (See
Chart 3)

GOOD VALUE?

The idea of full inflation index-linking was to entice
pension funds which seek revenue streams to match their rising
liabilities.

As of December, the UK regime had attracted some pension
fund investment directly or indirectly: AMP Capital is an
Australian pension provider while some of the other equity
sponsors have pension fund investors.

But full index linking was over-generous, according to the
National Audit Office (NAO), which advises on value for money in
public spending, given that the main ownership cost is financing
which is fixed from the outset. (Chart 4)

Returns achieved in early projects were up to 11 percent,
exceeding PPP projects, according to the NAO report published
last June, “Offshore electricity transmission: a new model for
delivering infrastructure”.

Parliament’s public finance advisors, the Public Accounts
Committee (PAC), was more critical.

“The terms of the transmission licences awarded so far
appear heavily skewed towards attracting investors rather than
securing a good deal for consumers,” it concluded in a report in
January.

It suggested possible changes which will provide a check
list for similar regimes to deliver value for investors while
not demanding consumers bear all the risk.

These included: higher penalties for under-performance;
shorter licensing periods than 20 years; alternatives to full
index linking such as flat or partially indexed revenues;
transparency about actual returns; and a claw-back process to
share gains where these were excessive.

The vast majority of electricity is presently consumed at
the instant of generation, putting a premium on electricity
storage.

As supply of intermittent renewable power grows, from
sources such as wind and solar, storage technologies have
additional value to balance variable supply, as well as varying
demand.

That grid balancing value includes benefits that are hard to
monetise, such as reducing the wear and tear on gas-fired power
plants, as these are used increasingly to balance wind and solar
power by ramping up and down – a cost that policymakers may have
to value explicitly to make pumped storage economic.

Governments are funding the development of chemical battery
storage technologies for automated balancing of electricity
distribution at the low voltage, household level.

But hydropower remains by far the most mature, economic,
grid-scale energy storage option.

Pumped storage has an exceptionally sharp ramp-up time,
adding to its grid balancing potential. The operator of
Britain’s Dinorwig plant for example reports a 0 to 1,320
megawatt “pick-up rate” of 12 seconds.

Pumped storage works like a dam in reverse, where the
operator purchases power at cheaper off-peak prices to pump
water from a lower to higher altitude reservoir, releasing it to
generate electricity at times of peak demand.

Paying for some benefits of grid balancing – such as
reducing wear on thermal plants or for an exceptional response
time – is difficult through market mechanisms. It may require
policymakers to intervene in energy markets, as they are doing
increasingly after the roll-out of subsidised renewable power.

RESOURCE

Globally, there are approximately 270 pumped storage plants
operating or under construction, with a combined generating
capacity of over 127 gigawatts (GW), according to the U.S.
National Hydropower Association in its overview, “Challenges and
Opportunities For New Pumped Storage Development”.

Europe is ahead of North America and China. (See Chart 1)

The United States has 40 existing pumped storage projects
with more than 22 GW of storage capacity, according to the NHA.

Most of these were authorised more than 30 years ago.

The U.S. Federal Energy Regulatory Commission (FERC) shows
an upsurge in permit applications for projects since 2007,
presumably with an eye on the new potential for grid balancing
as U.S. wind capacity expands. (See Chart 2)

FERC had issued preliminary permits for total pumped storage
capacity of 49.7 GW, as of March. (Chart 3)

A preliminary permit does not authorise construction but
maintains a place in a queue should the developer apply for a
licence, implying a big gap to realised projects.

Limits on more hydropower include high capital cost and long
planning and construction lead times, while the expected impact
on water availability of climate change, including more extreme
heatwaves and droughts, is a concern.

“Very few financial institutions are willing to finance
these types of long-lead projects through the licensing time
frame,” the NHA said.

**********************************

Chart 1: (page 26) goo.gl/0qLHv

Chart 2: goo.gl/jk68I

Chart 3: goo.gl/P70Xw

**********************************

VALUE?

At present wind, solar and gas-fired power represent the
bulk of added capacity in industrialised economies.

That reflects subsidies including tax credits and a power
price premium for intermittent renewables, as well as a short
process to obtain permits and relatively low fuel cost for
natural gas, which has under-cut pumped storage.

“Pumped storage is the most likely form of large new hydro
asset expansions in the U.S., however justifying investments in
new pumped storage plants remains very challenging with current
electricity market economics,” according to the research
organisation, Electric Power Research Institute (EPRI).

“Even over a wide range of possible energy futures, up to
2020, no energy future was found to bring quantifiable revenues
sufficient to cover estimated costs of plant construction,” it
said in its report, “Quantifying the Value of Hydropower in the
Electric Grid”, published in February.

The EPRI study found numerous ways in which existing pumped
storage operators could increase revenues, including installing
adjustable speed storage pumps to balance better surplus wind
power generation, potentially increasing revenues by 85 percent.

It suggested that explicitly valuing grid balancing services
could pose an “economic tipping point” for new projects.

DIFFERENT ASSET

Arguments for reforming pumped storage markets include two
approaches: first, defining it as a new asset class, where at
present it can be treated as electricity generation,
transmission or consumption; and second, introducing payments
for grid balancing services.

Regarding asset definition, the European electricity
industry body Eurelectric, in its report “Europe Needs Hydro
Pumped Storage: Five Recommendations”, noted last year that
operators can be taxed twice, for both power consumption and
generation.

The U.S. NHA argues that pumped storage could be treated as
a transmission asset, given its grid balancing role.

That would allow it to benefit from predictable, long-term
revenues based on transmission charges instead of wholesale
power market arbitrage, whose profits may be too unclear for
financial backers.

U.S. regulation (Section 219 of the Federal Power Act)
allows investors in transmission facilities an attractive return
through charges on grid users.

Regarding payments for grid balancing services, wholesale
power markets are already evolving to award
government-guaranteed revenues for reserve capacity, demand
response and energy storage, where pumped storage could be
included.

To reform power markets in this way, a philosophical debate
must first be resolved about the role of energy storage compared
with other ways to deal with the intermittency of renewables,
such as building out transmission capacity.

(Editing by Anthony Barker)