Not safe to look up yet
The sharp global slowdown has suddenly obscured the outlook for the solar
technology sector. With demand in Europe (77% of FY08F solar demand) expected to
contract severely, we expect to see reductions in solar module ASPs of 19% q-q in
4Q08 and 13% q-q in 1Q09. Inventory will pile up and drag utilisation to a bottom in
1Q09. Pricing will remain under pressure, in our view, as the poly-silicon market looks
set to tip into oversupply in 1H09. But this will begin the transition that will allow the
industry’s long-term growth prospects to shine through again. We look for PV module
ASP to halve from 3Q08 to 4Q10, at which point solar power will hit grid parity in select
locations, stimulate price-elastic demand and attract a new class of marginal buyer —
the utilities. Meanwhile, falling industry margins in FY09 will push solar companies to
increase cell efficiency to lower cost, allowing silicon usage per watt to fall by 20.5%
between 3Q08 and 4Q10. However, for the next 12 months, it is almost a total eclipse
of earnings growth for the sector. Suntech is our only BUY as superior cell efficiency
gains should support earnings growth of 14.2% y-y in FY09 and 68.7% y-y in FY10.
A year of transition
A faster path to grid parity: end 2010
Cell efficiencies as a route to differentiation
Valuations amid changing returns
Our view
The slowdown in the semi industry will result in about 12% excess polysilicon
finding its way into the solar space. Given an oversupplied poly market next year,
we expect steep price declines and falling industry margins. But this will hasten the
move to grid parity and help companies take the lead in raising cell efficiencies.
Anchor themes
We expect cost reductions to replace government subsidies as the main industry
demand driver, setting the stage for price-elastic demand. We expect ASPs to fall
by 50% over 3Q08-4Q10F, with grid parity being reached by end-2010F.
Woes in the semi industry are prompting suppliers to view solar as a safe haven.
Semi equipment makers are channelling R&D into solar, while polysilicon makers
are switching capacity from semi to solar. Oversupply in poly aside, demand
dynamics look poor, with banks unwilling to extend credit for new demand.
Not safe to look up yet
Year of transition
Next year will bring major change. More than half the demand in Europe
is from households / communities, and this is likely to slow down sharply
as funding baulks. Alongside changes in Spain’s solar policy and a
depreciating euro, price declines look set to accelerate. Facing an
abrupt demand slowdown in 4Q08F, unsold inventory of solar cells will
rise to a record 612MW at end-1Q09F, on our forecasts. That quarter
should mark the low point in industry utilisation, as the industry works to
clear inventory. While new policies in Italy, France and the US hold
promise, these are unlikely to move the needle in 2009F.
Faster path to grid parity: end-2010F
The poly market, so far an area of constraint, looks set to tip into
oversupply in 1H09F. We expect solar industry poly supply to rise
43% y-y in 2009F, versus demand growth of 28% y-y. This will lead to
ASP declines across the value chain. We forecast photovoltaic (PV)
module ASP to halve from 3Q08 to 4Q10F — at this price solar power
will hit grid parity in some places and pave the way for price-elastic
demand acceleration.
Cell efficiencies as a route to differentiation
While margins will fall next year, companies still need to stand apart.
Raising cell efficiency will become a key route to lower costs, allowing
silicon usage per watt to fall by 22.5% from 3Q08 to 4Q10F.
Valuations amid changing returns
Suntech Power (BUY) is the only stock in our coverage that looks set for
earnings growth. E-Ton and Trina are REDUCEs on short-term funding
concerns and stretched balance sheets. We are NEUTRAL on Motech
and Canadian Solar. Canadian is at 0.3x FY09F P/BV, and Motech has
the strongest balance sheet but we initiate with NEUTRAL on concerns
of timing of expansion and higher valuation than peers.
Contents
ASP pressure 3
Solar module: ASP declines to accelerate 4
The changing nature of demand 4
Inventory build looks inevitable 5
Semi pain to benefit solar 6
Solar: a safe haven for semi-equipment and poly suppliers 6
Polysilicon oversupply in 1H09F as demand slows 7
Poly oversupply cannot be avoided 9
Risk of poly ASPs falling towards marginal cost of production 10
Europe: a wider base of countries join 11
Demand slows to 28% y-y in 2009F, after 46% y-y in 2008F 11
Spain changes its solar policy: victim of its own success? 12
New demand centres emerge in Europe 13
New markets unlikely to move the needle 15
US: ITC needs to be accompanied by other measures 15
Japan: controlled pace of installations 17
A faster path to grid parity 18
Grid parity to be achieved by end-2010F 18
Module ASPs to halve over 3Q08 to 4Q10F 19
Valuations amid changing returns 21
Return structure is changing 21
Differentiating through the downturn 22
Balance sheets to be focus as industry fundamentals worsen 23
Sector catalysts 24
Price targets set on peer average P/Es 25
Appendix 27
Value distribution across the solar chain 27
Solar energy versus other alternatives 31
Generating electricity from solar energy 33
Types of government subsidy 36
Routes to improve cell efficiency 38
Sensitivity to global recession; trough valuation 40
Latest company views
Canadian Solar Inc 47
E-Ton Solar Tech 52
Motech Industries 57
Suntech Power 62
Trina Solar 69
Executive summary
Demand slowdown. As consumer spending woes take a toll and government
attention gets directed to bigger economic issues, demand from European countries
(77% of FY08F solar demand) is set to slow. More than half of European demand
comes from households/ communities, where decision making is faster — demand
could see abrupt deceleration (Exhibit 1).
Utilisation to bottom. Unsold inventory of solar cells looks set to rise, with this to
peak at a record 602MW at end-4Q08F followed by 612MW at end-1Q09F. This
should lead to the lowest point in industry utilisations, and record quarterly reductions
in solar module ASPs in 4Q08F and 1Q09F of 19% q-q and 13% q-q respectively,
driven primarily by sharp declines in polysilicon (poly) prices (Exhibit 2).
Cost reductions at PV cell makers. Slowing semi demand will result in excess poly
supply in the traditional semiconductor space which will be diverted to the solar
segment. We estimate this switch alone will add 12% to solar industry poly supply in
2009F. We expect poly pricing to halve by end-4Q10F, thus leading to an overall
decline in solar modules and accelerating progress towards grid parity (ie, price per
unit of solar energy falls to levels of conventional energy from the grid; Exhibit 5).
Grid parity by end 2010F. The market expects solar energy to reach grid parity in
2012F or later; we estimate it can be achieved by end 2010F, on the assumption that
PV system costs will fall by 50% over 2008-10F. This will lead to PV module ASPs of
about US$2.1/watt and with an average of about 1,800 annual solar hours available,
PV generation costs will fall to about US¢19/kWh, similar to grid pricing of select
regions (Exhibit 17).
Potential policy change in big markets. Spain became the only country in the world
to attract a capacity of about 2GW, in only three years since the start of its feed-in tariff
(FIT) programme. While Italy and some other European countries emulate Spain’s FIT
programme to spur critical mass in solar capacity, we have assessed the possible
reasons for Spain to cap the annual installation and reduction of subsidy. Without the
changes in Spanish policies, the monthly burden on the average Spanish household
would have risen to US$57 instead of the planned US$3.5 by 2020F (Exhibit 12).
Hence we see a risk of a repeat of this in some of the largest markets, like Germany.
Bearish. We initiate coverage on five stocks in this report, with a Bearish view on the
sector. Suntech Power (BUY), the world’s second-largest solar module maker, is the
only stock in our coverage that has a clear path to cost reductions, as it defends its
margins by raising cell efficiency. We assess a US$7 price target for Suntech (29.9%
upside), by pegging it to the average P/E of its global peers in China, Europe and the
US. E-Ton and Trina are rated REDUCE with downsides of 21.4% and 17.7%,
respectively, and catalysts expected from pressure on their already stretched balance
sheets given their funding requirements. Motech and Canadian Solar are NEUTRALs.
While Motech has the strongest balance sheet in the sector, we are concerned about
its expansion plans in FY09F and its relatively high valuation at 6.8x FY09F P/E.
Canadian Solar trading at 0.3x FY09F P/BV offers limited downside over the next 12
months; albeit short-term price volatility cannot be avoided owing to a shortfall in
4Q08F unit shipments at 20-25MW, significantly lower than the 60MW shipped in
3Q08.
Demand to see abrupt slowdown
as consumer spending woes take
their toll
Polysilicon pricing to lead cost
reductions at PV cell makers
Spanish policy changed to
contain monthly burden on
Spanish households
Initiate sector coverage with a
Bearish view
1Q09F to mark the lowest point in
utilisations as inventories peak
PV module ASPs to reach gridparity
by end-2010F versus
consensus of 2012F
Solar module: ASP declines to accelerate
The changing nature of demand
The solar industry is entering a new era. The main demand driver so far has been
government subsidies. We expect significant cost reductions will set the stage for
price-elastic demand creation that will replace subsidies as the main driver.
Unfortunately, these cost reductions will be led by an abrupt demand slowdown from
the EU (which comprises 77% of 2008F solar demand), as consumer spending woes
take their toll and government attention switches to bigger economic issues. We think
the demand shock in the pipeline will bring an unprecedented fall in ASPs for solar
modules. However, price-elastic demand is bound to pick up, leading to a new
marginal buyer emerging — namely utilities — in addition to the current user base of
corporate/residential owners.
European woes take centre stage
We expect solar module ASPs to fall 19% q-q in 4Q08F and 34% y-y in 2009F, owing
to a confluence of several negative factors.
We estimate that more than half the demand in the EU is generated by households/
communities wanting to take advantage of lucrative FIT programmes subsidised by
governments. But Europe is going into recession and focusing on conserving cash,
and this will naturally deter fresh investment.
Demand shifts from Spain to countries such as Italy, where ASPs are lower. Spain
has been the second-largest demand driver this year, comprising 32% of industry
demand. In Spain, we estimate ASPs have been as high as US$6-7/watt in 2008.
With the Spanish government capping demand to 500MW in 2009 (in 2008,
installations were around 1.3GW, as there was no cap), demand creation needs to
come from other countries. The problem is that none of the upcoming demand
centres — Italy, the US and France — have ASPs comparable to those in Spain.
The euro has fallen by more than 20% against the US dollar since its peak in July
2008. Cell makers now need to renegotiate prices with their European users. We
estimate euro depreciation will account for 10pp of the estimated 19% q-q decline
in module ASP in 4Q08F.
The above factors only compound the planned 8-10% pa ASP FIT reductions in the
largest demand centre, Germany.
|
