100GW of demand, and the coming inflection point in the US solar market

Share on facebook
Share on twitter
Share on linkedin
Share on reddit
Share on email

Non-linear systems are often difficult to understand. The famous “penny game” is a good example. In this game, a hypothetical person is given one penny (or one euro cent) on the first day, two pennies on the second day, four pennies on the third day, etc., and then asked to guess the total value of the pennies at the end of one month. Very few people guess correctly – US$21 million – or appreciate that 75% of that value is created on the last two days.

Like the penny game, the solar industry is characterized by a number of non-linear dynamics. First, as is well-known, prices are declining non-linearly: the module price (to a developer) has dropped from approximately US$2 per watt last year to US$1 per watt this year. And module price itself is a reflection of multiple non-linear dynamics: polysilicon prices are changing non-linearly; polysilicon usage is changing non-linearly; non-polysilicon processing costs are changing non-linearly; and, importantly, manufacturer margins are changing non-linearly. Recently, these dynamics have been very negative for solar PV producers.

Fortunately, solar PV demand also behaves non-linearly, and this creates incredible potential for the solar industry, particularly in the United States.  In a thorough analysis, we have used local-level electricity prices and local-level irradiation to calculate that market potential, and the results are astoundingly large. At an installed cost of US$4 per watt – for businesses and consumers – distributed solar PV could economically serve as much as 46 billion kWh of US electricity consumption, equivalent to 33GW of solar PV capacity. At US$3 per watt – and some commercial projects are at this level now – addressable electricity consumption rises to 440 billion kWh, equivalent to over 300GW of capacity.

To many readers, this result may seem incredibly large, much like the result of the penny game. Before rejecting it, I would like to present some important characteristics of the United States electricity market that may change your mind.

Overview of the US electricity market

The United States is the world’s second largest consumer of electricity, with annual consumption of approximately 4 trillion kilowatt hours (4 × 1012 kWh) – more than all of Europe combined. The distribution of US electricity consumption resembles a bell-curve, with its apex at US$0.10/kWh (Exhibit 1). Moving from the right along that curve, i.e., from higher prices, toward the center, electricity consumption increases exponentially beginning at US$0.18/kWh. Because the distribution is relatively concentrated in the middle, below US$0.18/kWh the amount of electricity consumed at each price point becomes incredibly large. For example, approximately 150 billion kWh of electricity is consumed at a price of US$0.18/kWh or higher, but an additional 350 billion kilowatts – equivalent to the entire electricity consumption of the United Kingdom – is consumed between US$0.15 – US$0.17/kWh. For reference, at 1,600kWh/m2, 100 billion kWh represents 75GW of solar PV capacity. Moving closer to the apex, consumption increases even faster: approximately 800 billion kWh are consumed between US$0.13 – US$0.15/kWh – significantly greater than total electricity consumption in Germany. As the cost of solar PV electricity approaches these prices, particularly the US$0.18 – US$0.15/kWh range, the demand for solar PV will increase exponentially. It is precisely this dynamic that is being dramatically underestimated. And we are approaching that price range now.

Exhibit 1

As is clear from the data in the Exhibit 1 graph above, it is more accurate to compare the US solar potential to the entire continent of Europe than to any single country. Continuing briefly with this comparison: as is well known, average irradiation in the US is approximately 1,800kWh/m2, again, greater than the average irradiation of Europe. Nearly every solar PV installation qualifies for the federal Investment Tax Credit, an uncapped incentive, which reduces the initial cost of the solar installation by 30%, materially lowering the resulting electricity cost. And, nearly every state has separate, additional incentives in place. Finally, to establish a baseline of solar demand, last year Europe installed approximately 13GW of solar PV, and Europe has installed over 30GW over the last five years (including 2011). With this in mind, 100GW of installations in the United States begins to appear more plausible…

Analysis approach

Because of the large volume of data, calculating the potential demand for solar PV in the United States requires a relatively sophisticated model. I will summarize the approach and results here. Additional details, including sensitivity analyses, are presented in a free research report, available for download here

In summary, the approach is to calculate the point at which the present value of all costs and savings associated with installing a solar array become less than the present value of all payments to the utility. Because the United States has a wide distribution of electricity prices, and a wide distribution of irradiation (ranging from approximately 1,200kWh/m2 in Alaska to nearly 2,400kWh/m2 in Arizona), this analysis must be completed on a market-by-market basis.

At the heart of the analysis lies an understanding of solar PV economics. Operating costs of a solar array are remarkably low, which means that the primary drivers of solar electricity cost are the initial installed cost of the array, irradiation and available incentives. As is well known, this relationship is best expressed with a Levelized Cost Of Energy (LCOE) model. Many detailed LCOE models are available from research organizations and national laboratories. I prefer to use a simple model which ignores the operating costs of the array (a small and decreasing expense), and tax-related depreciation gains (a relatively large benefit to corporations). This model also ignores all state- and local-level incentives and therefore over-estimates the LCOE of solar PV (thereby providing a conservative estimate of solar PV electricity cost and potential demand).

With that model, and assuming an irradiation of 1,800kWh/m2, the relationship between installed cost and LCOE is determined by a factor of 7 without incentives, i.e., US$4.00/W installed cost equates to approximately US$0.28/kWh. The Investment Tax Credit reduces the electricity cost by approximately 30%; in the example above, the cost is reduced to US$0.19/kWh. This relationship is summarized in Exhibit 2 (again, for a complete presentation of economic modeling, see the full report). As is clear from the graph, as the installed cost of solar PV reaches approximately US$3.50 per watt for businesses and residences, the electricity cost begins to intersect with the bulk of US electricity consumption. Again, at this intersection, potential demand for solar PV will increase exponentially, and some installations are reaching that point now.

Exhibit 2

Armed with an understanding of electricity prices, irradiation, and solar economics, the analysis approach is relatively straightforward. The first step is to calculate the present value of all future payments to the utility – this represents the baseline scenario. The next step is to calculate the present value of all future costs and savings associated with installing the array. These components include ongoing (lower) payments to the utility; savings from electricity generated by the PV array and consumed on-site; after-tax revenues from excess electricity sold back to the utility (if any); all costs associated with installing the array; and any applicable incentives (we have only modeled the ITC). All of the input variables vary by market segment, with the exception of the installed cost of the array, which is the control variable. As the installed cost of the array decreases, the amount of electricity that can be economically served increases, depending on the irradiation and the retail price of electricity. We have completed that analysis for the entirety of US electricity consumption (see Exhibit 2 graph above).

These results are summarized in Exhibit 3. At approximately US$5.50 per watt installed cost, a reasonable average for distributed installations in 2011, the amount of electricity that can be economically served by solar PV is very low, only 4 billion kWh, equivalent to 2 – 3GW of solar PV capacity. However, as installed costs decrease, the amount of electricity that can be economically served by solar PV increases exponentially, to 46 billion kWh at US$4.00 per watt, equivalent to 33GW capacity; and to over 400 billion kWh at US$3.00 per watt, equivalent to over 300GW of capacity.

Exhibit 3

Of course, potential is different from actual, and for practical reasons some important constraints have been omitted. Nevertheless, the data indicate that US capacity could comfortably exceed 100GW within five years (Exhibit 3 graph above).

This article presents some of the findings of a detailed analysis entitled 300GW At $3.00 Per Watt? Calculating The U.S. Solar Potential Using Electricity Prices, which is available for download here. I will host a conference call to discuss results in approximately 10 days.  RGK

Read Next

April 12, 2021
Spanish energy company Acciona has secured shareholder approval for an initial public offering (IPO) of its renewables division, providing a boost for the firm’s plans to reach 6GW of installed PV by 2025.
April 12, 2021
Canadian Solar has started mass production of its large-wafer size solar modules, boasting power outputs of up to 665W.
April 12, 2021
British energy company ScottishPower is targeting up to 40MW of solar as part of its Green Hydrogen for Scotland project.
April 12, 2021
Scientists in Australia have used human hair clippings from a Brisbane barbershop to create an ‘armour’ that increases the power conversion efficiency of perovskite solar cells.
April 12, 2021
A consortium between Abu Dhabi-based Masdar, EDF Renewables and Nesma Holding Company has broken ground on a 300MW PV project in Saudi Arabia.
April 12, 2021
The April 2021 episode of the Solar Media Podcast, sponsored by Honeywell, is now available to stream, featuring lessons from Asia on how to push floating solar forward.

Subscribe to Newsletter

Upcoming Events

Solar Media Events
April 13, 2021
Solar Media Events
April 20, 2021
Upcoming Webinars
April 28, 2021
4:00 - 4:30 PM CET