Due to the current lapse of federal funding, Berkeley Lab websites are accessible, but may not be updated until Congress approves funds for Fiscal Year 2014.

The price at which electricity from large-scale solar power projects in the western United States is being sold has fallen by more than two-thirds in the last five years, according to a new report released today by Lawrence Berkeley National Laboratory. The report—"Utility-Scale Solar 2012: An Empirical Analysis of Project Cost, Performance, and Pricing Trends in the United States"—focuses exclusively on ground-mounted solar projects with capacity ratings greater than 2 MWAC. Such large-scale solar projects are commonly referred to as "utility-scale" solar.

The progress revealed by the report comes despite the utility-scale segment of the market being relatively young. "Although a number of concentrating solar power ("CSP") projects using parabolic trough reflectors have been operating in California since the 1980’s, very few CSP projects have been built since then, and utility-scale photovoltaic ("PV") projects have really only entered the market in the last five to seven years" notes Berkeley Lab report author Mark Bolinger. In that short time, utility-scale PV has rapidly grown to become, for the first time in 2012, the largest segment of the overall PV market in the United States—a position it is expected to retain through at least 2016. Utility-scale PV has also far surpassed CSP as the dominant utility-scale solar technology, though a number of major new CSP projects—including two large power tower facilities and three large parabolic trough projects—are under construction and should reach commercial operations in the coming months.

As a result of this rapid growth, there is now a critical mass of utility-scale project-level data ripe for analysis. This report—the first edition in a new Berkeley Lab series to be revisited annually—analyzes project-level empirical data in four key areas: installed project costs or prices, operating costs, capacity factors, and power purchase agreement ("PPA") prices. "With the growth in the market in recent years, we are now able to systematically review actual market data to directly observe what large-scale solar projects cost to build and operate, how they are performing, and at what price they are selling electricity," notes report co-author Samantha Weaver. Data for some of these areas—operating costs in particular—are still sparse, but availability should improve greatly in the coming years as the strong pipeline of utility-scale solar projects—for both PV and CSP—continues to be built out.

Key findings in each of the four areas covered by the report include the following:

Installed Project Prices: Installed PV project prices have fallen by nearly one-third since the 2007-2009 period, from around $5.6/WAC to $3.9/WAC (in real 2012 dollars) on average for projects completed in 2012 (with some projects higher and others lower, and with further reductions evident in 2013). Most of the decline has been concentrated among projects using crystalline silicon ("c-Si") modules or panels, as the cost gap between c-Si and thin-film modules steadily eroded over this period. In response to falling c-Si module prices, there has been a marked increase in the proportion of projects using c-Si (rather than thin-film) modules.

O&M Costs: Although publicly available O&M cost data are extremely limited at present, the data suggest that actual costs to date have largely been in line with pro forma operating cost projections. For PV, O&M costs appear to be in the neighborhood of $10-$20/MWh. For CSP (parabolic trough, no storage), O&M costs are higher due to the thermal components, and come in around $25-$30/MWh.

Capacity Factors: PV capacity factors vary by region, by module type (c-Si versus thin film), and by whether a project is installed at a fixed tilt or uses a tracking device. In some of the best locations in the Southwest, PV projects using single-axis trackers are able to achieve capacity factors in excess of 30%. In lieu of trackers, and enabled by the sharp decline in module prices, some projects have instead opted to oversize the PV array relative to the capacity rating of the inverters as a way to boost production during morning and evening hours (and thus increase overall capacity factor in AC terms). As expected due to their ability to perform well at high temperatures, thin-film projects appear to have higher capacity factors than c-Si projects in the Southwest, where ambient temperatures are highest, but not elsewhere. On the CSP side of the market, parabolic trough systems that have been operating in California for more than 20 years were still (in 2012) achieving capacity factors in excess of 20% (solar output only, with no storage and not counting fossil fuel augmentation), which is comparable to newer trough projects without storage.

Power Purchase Agreement ("PPA") Prices: Driven primarily by lower installed PV project prices (which, in turn, have been driven primarily by declining module prices), as well as expectations for further cost reductions in future years, levelized PPA prices have fallen by more than two-thirds since 2008, or by roughly $25/MWh per year on average. Some of the most-recent PPAs for PV projects in the West have levelized PPA prices as low as $50-60/MWh (in real 2012 dollars), which, in some cases, is competitive with wind power projects in that same region. PV appears to be particularly competitive when considering its time-of-delivery pricing advantage (i.e., the fact that solar generation correlates better with peak demand periods), which amounts to roughly $20/MWh (in California and at current levels of penetration) relative to a flat block of base-load power, and roughly $25/MWh relative to wind power.

These and other findings from the report will be presented via a free webinar on Wednesday, October 2, at 3 PM eastern time. Registration for the webinar can be found here:


Meanwhile, the full report, along with a presentation slide deck that summarizes the report, can be downloaded for free from the link below.

This research was supported by funding from the U.S. Department of Energy’s SunShot Initiative.