Ultra-low solar bid of $0.01997/kWh in the US – not quite so sunny
Aug 5, 2019 08:48 PM ET
PV technology has made significant progress in a short amount of time, leaving solar enthusiasts chuffed. Opponents, on the other hand, brand it as unworkable boondoggle, surviving on the crutches of subsidies and only salving the conscience of the green-minded companies cosseted by political forces. The naysayers do have partial truth in their argument - governments the world over have been supporting the growth of this child with the moniker “solar” with public finances. So, this is the partial truth and the enthusiasts have just enough firepower to say that all energy is, after all, subsidised one way or another – in terms of uncompensated costs of air pollution, congestion and global warming and the consequential irreversible damage these fossil fuels have done to the planet. Perhaps more important and statistically relevant is a report published by the IMF in 2015, which had pegged subsidies to fossil fuels at US$5.3 trillion.
Straight to the point now; the heroic forward march of solar entered into a new trajectory (whether devoid of sense or laden with it, only time will tell) when 8minuteenergy, America’s largest IPP, proposed a PPA rate of US$0.01997/kWh for 400MWac / 530MWdc of solar power plants to be developed in sun-drenched city of Los Angeles in the United States in a recently-concluded auction. With this, 8 Minutes has beat its own pricing leader of US$0.02375/kWh from the 300MW Eagle Shadow Mountain solar project. There is also a US$0.013/kWh adder for the excess electricity later delivered from a co-located 200MW / 400MWh energy storage system.
At first glance, this looks like a mirage in the hot deserts of Arabia, and, obviously, raised a lot of hubbub in the industry around its economic viability. Will it make sense? Let us see.
It is true that almost any external surface can generate solar electricity, and costs are plummeting (not just for the silicon wafers, but also for installation, electronics and storage needed to make the system work), the financial models the developers must have used are testing their limits, we are inclined to believe. It appears that they are calculating development costs well below comparable global benchmarks. It might not come as a surprise if they are forced to stretch this budget to make some returns out of this project.
Yet again, we can use LCOE as the yardstick to unshroud this mystery. We punched some assumptions in our LCOE model and it sprang returns to the tune of 5%, which might just look good in terms of US market. The assumptions go like this:
|LCoE plant life||25|
|Project Configuration||Single-axis tracker|
|Type of PV module||Bifacial p-type monocrytalline|
|Rated power at STC||370 Wp|
|Energy yield DC||2,350 kWh/kWp|
|Degradation||0.70% / yr|
|Gearing ratio (debt/(debt+equity))||50%|
|Cost of debt||3.50%|
|Tenure of debt||15|
|Cost of equity||6%|
|Weighted average cost of capital (WACC)||4.5%|
These numbers are audacious and over-optimistic per se but a deep dive insight can untie the knots in the understanding so far.
It goes without saying that a single-axis tracker mechanism is the obvious choice, given that only trackers can help get the last electron out from the cell by gathering more radiation from the sun. However, it is imperative that tracker manufacturers keep innovating design and tracking algorithms to pump up the yield in the next two years, since this project is slated to go online by April 2023.
Then again, one of the most important assumptions is the energy yield, which has been input as 2,350 kWh/kWp in our model. While this yield, which translates into a utilisation of 26.8%, might seem to be a herculean task to achieve, it is achievable only with bifacial modules, keeping in mind that California is a sun-drenched area and receives good quality sunlight. Over an inverter loading ratio (ILR) – called DC-AC overloading in India) – above 30%, bifacial modules can increase yield by some 7-9%. Interestingly, the fact that bifacial modules will get a pass on the Trump administration's solar import tariffs will help reach these yield levels, without hurting cost economics, as they have recently been excluded from Section 201 tariffs.
The next important factor is the project cost, which, in the US, currently stands between US$1 to US$1.1 on per Wp basis. According to our model, US$0.67/Wp is the capex required to build the project from a clean slate. The typical break up of this pricing is as follows:
|Modules||US$0.24 / Wp|
|BoS||US$0.17 / Wp|
|I&C||US$0.17 / Wp|
|Pre-op cost||US$0.09 / Wp|
As is common knowledge, the most critical component of a solar project is the solar PV modules and, as per our market insights, monocrystalline bifacial modules can, presently, be delivered at site in the US from Southeast Asia at US$0.30/Wp. Considering a historic 10% decline, this will, most probably, hit US$0.24/Wp by 2022, when actual modules would be required. Then again, electrical and structural BoP (balance of plant, which is all the hardware sans modules) constitutes around 25-30% of the total project cost in the US and translates into US$0.17/Wp, assuming lower end of 25% on the basis of market consolidation and pricing maturity. What then remains is the cost of designing, procuring and actually getting hands dirty on the ground with construction, which also forms another 25%-30% of the project cost in this country. At 25%, we have assumed this to be US$0.17/Wp. Finally, we have assumed the pre-operative and other miscellaneous expenses at US$0.09/Wp (around 13.75%, very typical in the US).
All said and done, this factor (project cost) still leaves the model gasping for some fresh air like a bird enveloped in an acrid fug. Tickling an achingly red Excel model, the income tax credits (ITC), like the pristine pinnacles of the Himalayas, whispered a promise of fresh air. ITC stands, currently, at 30% for projects commencing construction before 2020, and will step down to 26% if the project starts ground activity in 2020, further to 22% in 2021. It is only wise to assume that the project should kick off in 2019 to avail this benefit. This takes the project cost down to US$0.47/Wp. Workable? Don’t know yet.
Moving further, Opex is another factor that has a large say on the project’s viability. In the US, it is typically 1.5% of the project cost, which translates into US$7/kWp, a good match with our assumption. There is an annual escalation of 2% over this cost.
To tie all the loose ends together, we need finances. The story unfolds like this: according to National Renewable Energy Laboratory (NREL), the typical PV project financings of large-scale centralized projects owned by independent power producers shows that low-cost debts can be picked up at interest rates around 4% for tenure ranging up to 13 years. The cost of equity, meanwhile, is around 6%. Because debt is a lower-cost source of capital than equity, we have apportioned 50% of capital to debt, which returns a 4.5% weighted average cost of capital (WACC), considering tax for renewable energy projects.
Plugging these numbers in our model threw an equity return of 5%, which leaves very little room for applying sensitivities for these important factors. We were filled with enthusiasm at the beginning of this effort to understand the mind of the bidders but burning the midnight oil does not yield much light, it appears.
"This is like music to my ears," Commissioner Christina Noonan seems to have told the Los Angeles Department of Water and Power (LADWP), who is looking to sign this record-setting PPA. Madam, a lot is at stake, we humbly submit.
For us, it seems to be confusing, devastating, tickling and exhilarating all at the same time, at best. Time to look ahead.
- Poland's Unimot raises PV panels output 3-fold to 45 MW
- Italy includes 2.48 GW of solar in 2022 to appear 25 GW
- Linton to manufacture silicon equipment for photovoltaic panel makers within the United States
- Vertically placed solar system Sunzaun hits market
- Recom to introduce Panther series of solar modules to united state market at Intersolar