Utility-Scale Solar, 2024 Edition

Empirical Trends in Deployment, Technology, Cost,

Performance, PPA Pricing, and Value in the United States

Joachim Seel1, Julie Mulvaney Kemp1

Anna Cheyette, Dev Millstein, Will Gorman, Seongeun Jeong,
Dana Robson, Rachman Setiawan, and Mark Bolinger (consultant)

Lawrence Berkeley National Laboratory

Energy Markets and Policy Department

1 Corresponding authors

October 2024

This material is based upon work supported by the U.S. Department of

Energy’s Office of Energy Efficiency and Renewable Energy (EERE)
under Solar Energy Technologies Office (SETO) Agreement Number
38444 and Contract No. DE-AC02-05CH11231. Photo credit: Intersect Power
Utility-Scale Solar, 2024 Edition

Disclaimer
This document was prepared as an account of work sponsored by the United States Government. While this document is believed
to contain correct information, neither the United States Government nor any agency thereof, nor The Regents of the University of
California, nor any of their employees, makes any warranty, express or implied, or assumes any legal responsibility for the
accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use
would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by its trade
name, trademark, manufacturer, or otherwise, does not necessarily constitute or imply its endorsement, recommendation, or
favoring by the United States Government or any agency thereof, or The Regents of the University of California. The views and
opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency
thereof, or The Regents of the University of California.

Ernest Orlando Lawrence Berkeley National Laboratory is an equal opportunity employer.

Copyright Notice
This document has been authored by an author at Lawrence Berkeley National Laboratory under Contract No. DE-AC02-
05CH11231 with the U.S. Department of Energy. The U.S. Government retains, and the publisher, by accepting the article for
publication, acknowledges, that the U.S. Government retains a non-exclusive, paid-up, irrevocable, worldwide license to publish or
reproduce the published form of this manuscript, or allow others to do so, for U.S. Government purposes

Acknowledgements
The authors thank Ammar Qusaibaty, Juan Botero, Michele Boyd, and Becca Jones-Albertus of the U.S. Department of Energy
Solar Energy Technologies Office for supporting this work. The authors also thank the many individuals from utilities, the solar
industry, state agencies, and other organizations who contributed data to this report and/or who responded to numerous inquiries
and requests. Without the contributions of these individuals and organizations, this report would not be possible.

Photo credit: Utility-Scale Solar, 2024 Edition

Intersect Power's 415 MWDC/320 MWAC 2
Radian solar facility, completed in 2023.
http://utilityscalesolar.lbl.gov
Utility-Scale Solar, 2024 Edition
Purpose and Scope:
New in this year’s edition:
– Summarize public and private data on key trends in U.S. utility-scale solar sector
– Analyses of PV projects in newly
– Focus on ground-mounted projects >5 MWAC designated Energy Communities
• There are separate DOE-funded data collection efforts on distributed PV
(e.g., trackingthesun.lbl.gov) – Discussion of PV performance
degradation rates
– Focus on historical data, emphasizing the most-recent full calendar year
– Comparison of Berkeley Lab’s
Data and Methods: PPA data with data from
– See summaries at the beginning of each section LevelTen1 and Trio2

Funding: – Comparison of PV market value

with generation costs and
– U.S. Department of Energy’s Solar Energy Technologies Office climate + health benefits
Products in addition to this slide deck: – Discussion of solar withdrawal
– This report deck is complemented by an Excel data file that lists all data behind each rates and study status in
graph (and more), a written executive summary, and interactive visualizations. interconnection queues
All products are available at: utilityscalesolar.lbl.gov

These two companies operate renewable energy marketplaces and publish PPA pricing trends: Utility-Scale Solar, 2024 Edition
1 https://www.leveltenenergy.com/ppa 3
2 https://www.trioadvisory.com/publications http://utilityscalesolar.lbl.gov
Other recent publications from our team related to utility-scale solar

Annual Reports Siting and Community Engagement Value of Renewable Energy

Queued Up: 2024 Edition Developer Practices and Perspectives The Renewables and Wholesale Electricity Prices
Berkeley Lab’s annual report documents the growing on Community Engagement for Utility- (ReWEP) Tool
backlog of new power generation, particularly solar, Scale Renewable Energy in the United The ReWEP tool allows users to explore trends in nodal wholesale
wind, and storage, seeking transmission connections. States energy pricing and their relationship to renewable generation.
Hybrid Power Plants: 2024 Edition A survey of professionals shows that renewable Grid Value and Cost of Utility-Scale Wind and Solar:
developers use community engagement strategies
This annual briefing tracks existing hybrid plants in but favor limited public input, falling short of full
Potential Implications for Consumer Electricity Bills
the U.S. while also synthesizing data from PPAs and citizen empowerment.
interconnection queues to shed light on future This research quantifies the market value of wind and solar over
growth. time, exploring how contractual and market structures influence
Laws in Order: An Inventory of State consumers’ ability to benefit from cost savings.
U.S. State Renewables Portfolio & Clean Renewable Energy Siting Policies
Renewable-Battery Hybrid Power Plants in Congested
Electricity Standards: 2024 Status Update This report outlines state and territorial authorities
responsible for siting and permitting large-scale
Electricity Markets
This report provides a status update on state wind and solar projects, alongside an interactive Berkeley Lab’s analysis of hybrid renewable-battery plants in
renewable portfolio and clean electricity standards. map for exploring state-specific information. congested U.S. regions reveals optimal energy and capacity value
U.S. Large-Scale Solar Photovoltaic for solar and wind hybrid projects.
Perceptions of Large-Scale Solar Project
Database Neighbors Solar and Storage Integration in the Southeastern U.S.
In collaboration with the USGS, the USPVDB creates This study evaluates how varying levels of solar and storage would
A survey of residents living near large-scale solar
an accurate, comprehensive, and publicly accessible affect electricity system costs, reliability, and operations in the
projects provides insights into local perceptions
national large-scale PV database of large-scale PV Southeast U.S. by 2035.
that can inform future large-scale solar
facilities. deployment.
Join our mailing list to receive notice of future publications: https://emp.lbl.gov/mailing-list
Utility-Scale Solar, 2024 Edition
4
http://utilityscalesolar.lbl.gov
Report Contents

Deployment and Technology Trends

Capital Costs (CapEx) and Operation & Maintenance (O&M) Costs

Performance (Capacity Factors)

Levelized Cost of Energy (LCOE) and Power Purchase Agreement (PPA) Prices

Wholesale Market Value, Air and Climate Benefits, and Net Value

PV+Battery Hybrid Plants

Concentrating Solar Thermal Power (CSP) Plants

Capacity in Interconnection Queues

Summary

Utility-Scale Solar, 2024 Edition

5
http://utilityscalesolar.lbl.gov
Regional boundaries applied in this analysis include the seven
independent system operators (ISO) and two non-ISO regions

Source of the Irradiance data:

https://nsrdb.nrel.gov/

Utility-Scale Solar, 2024 Edition

6
http://utilityscalesolar.lbl.gov
Deployment and Technology Trends

Utility-Scale Solar 2024 Edition

http://utilityscalesolar.lbl.gov 7
Deployment and Technology Trends: data and methodology

Deployment Trends Data: National and state-level deployment data are sourced from the
Energy Information Administration (EIA), the American Clean Power Association (ACP), Wood
Mackenzie/SEIA Solar Market Insight Reports, and Berkeley Lab datasets.

Technology Trends Data: Project-level metadata are sourced from a combination of Form EIA-
860, FERC Form 556, state regulatory filings, interviews and websites of project developers and
owners, and news and trade press articles. We independently verify much of the metadata—
such as project location, fixed-tilt vs. tracking, azimuth—via satellite imagery.

Methods: Because we collect data from a variety of unaffiliated and incongruous sources, the
data must be synthesized and cleaned in multiple steps before becoming useful for analytic
purposes. In some cases, we essentially create new data by piecing together various snippets of
information that are of less consequence on their own.

Utility-Scale Solar, 2024 Edition

8
http://utilityscalesolar.lbl.gov
The utility-scale sector has the greatest share of the U.S. solar market

Annual Solar Capacity Additions (GW) Cumulative Solar Capacity (GW) Wood Mackenzie and SEIA report that the
utility-scale sector added 22.5 GWDC of new
70 700 solar capacity in 2023, accounting for 70%
Utility-Scale CSP
Utility-Scale PV of all new solar capacity. Annual growth
60 600 rose by 77% compared to 2022 and set a
Commercial PV
Residential PV new record.
50 500
PV is in GW DC and CSP is in GW AC Utility-scale solar contributed 65% of
40 400 cumulative solar capacity (and 69% of
Columns show annual capacity
solar generation) in 2023; this share is
Areas show cumulative capacity
30 300 projected to rise to nearly 70% by 2027.

20 200 Our data analysis focuses on a subset of

this sample—all projects larger than 5
10 100 MWAC—based on their completion date:
– 2022: 153 new projects totaling 10.4 GWAC
0 0
or 13.4 GWDC
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023

2032e
2024e
2025e
2026e
2027e
2028e
2029e
2030e
2031e

2033e
2034e
– 2023: 221 new projects totaling 18.5 GWAC.
Sources: Wood Mackenzie/SEIA Solar Market Insight Reports, Berkeley Lab The subset of 217 projects with known DC
capacity total 23.9 GWDC.

Utility-Scale Solar, 2024 Edition

Note: Graph above shows utility-scale solar as >1 MWDC while most of this report uses >5 MWAC. 9
http://utilityscalesolar.lbl.gov
More than 50% of new U.S. grid capacity came from solar in 2023

Annual Capacity Additions (GWAC) Solar Capacity Additions (% of Total) Utility-scale (35%) and distributed
60 60% (17%) solar accounted for a combined
Utility-Scale Solar (>1 MW) 52% of all capacity added to U.S.
Distributed Solar (<1 MW) Total Solar
(right axis) grids in 2023.
50 Wind 50%
Storage
Other It is the first year that solar made up
40 Gas 40% more than half of new US grid
Coal capacity.
30 30%
Solar has added more capacity than
any other fuel since 2021, contributing
20 20%
>40% of capacity additions each year,
>30% in 7 of the last 8 years, and
10 10% >20% in each of the last 11 years.

0 0% Storage continues to expand as well:

2007 2009 2011 2013 2015 2017 2019 2021 2023 6.9 GW of storage were added to U.S.
grids in 2023, up from 4.6 GW in 2022
Bars represent annual capacity additions in GWAC (left axis), and 3.9GW in 2021.
Line represents solar’s capacity share of annual additions (right axis)

Note: Graph above shows utility-scale solar as >1 MWAC while most of this report uses >5 MWAC. Utility-Scale Solar, 2024 Edition
10
http://utilityscalesolar.lbl.gov
 Solar generation’s market share was 5.6% across the U.S. in 2023,
but reached >25% in California and Nevada
Preliminary 2023 Solar market share can vary considerably depending on whether it is
Solar generation as a % Solar generation as a %
of in-state generation of in-state load
calculated as a percentage of total generation or load (e.g., Vermont).
Top 20
States
All Solar
Utility-Scale
All Solar
Utility-Scale As a percentage of in-state generation, California’s solar market share
Solar Only Solar Only
reached 28% in 2023, while Nevada, Massachusetts, Vermont, and Hawaii
California 28.2% 16.7% 28.9% 17.1%
Nevada 25.9% 22.1% 29.8% 25.4% all surpassed 15%.
Massachusetts 24.4% 8.6% 11.4% 4.0%
Vermont 18.9% 8.6% 8.2% 3.7% The utility-scale sector’s contribution varies by state: a minority in the
Hawaii 18.9% 5.9% 23.1% 7.2% Northeast and Hawaii, a majority in Southwest states and the overall U.S.
Utah 13.9% 11.3% 14.6% 11.9%
Rhode Island 11.8% 5.1% 14.7% 6.4%
Arizona 10.2% 6.3% 13.8% 8.6%
Maine 9.9% 4.9% 10.6% 5.2%
North Carolina 9.3% 8.8% 9.0% 8.4%
Colorado 9.0% 6.1% 9.4% 6.4%
New Mexico 8.2% 6.4% 11.2% 8.8%
New Jersey 7.4% 2.4% 7.1% 2.3%
Delaware 7.0% 3.1% 3.1% 1.4%
Idaho 6.9% 5.4% 4.3% 3.3%
Florida 6.7% 5.3% 7.1% 5.6%
Virginia 6.6% 5.8% 4.7% 4.2%
Maryland 6.3% 2.6% 4.2% 1.7% You can explore this data over time at
Georgia 6.1% 5.7% 5.7% 5.4% https://emp.lbl.gov/capacity-and-generation-state
Texas 5.8% 5.0% 6.5% 5.6%
TOTAL U.S. 5.6% 3.9% 6.2% 4.3%

Note: Table above shows utility-scale solar as >1 MWAC (most of this report uses >5 MWAC). Percentages represent Utility-Scale Solar, 2024 Edition
annual averages. Data is based on an early EIA data for 2023, findings may be revised with final data. 11
http://utilityscalesolar.lbl.gov
New utility-scale solar projects were built in the eastern Midwest,
the mid-Atlantic, and southern United States in 2023
Fixed-tilt ( ) projects are increasingly only being built
on particularly challenging sites (e.g., due to terrain or
wind loading) or in the least-sunny regions in the
northeast.

Other high-latitude states such as Oregon, Minnesota,

Wisconsin, and Michigan, added predominantly
tracking projects in 2023 ( ).

In 2023, storage ( ) hybrid projects hit the ground in

record numbers. Batteries were added to already
existing (15) and new (37) PV projects. Solar-rich CA
added the most storage capacity (1657 MW), followed
by the non-ISO West (1046 MW).

Note: A 6MWAC project in Alaska is not shown in the map above. Larger icons represent greater capacity. Utility-Scale Solar, 2024 Edition
12
You can explore this data interactively at https://emp.lbl.gov/technology-trends http://utilityscalesolar.lbl.gov
Utility-scale solar has been built throughout the United States

Utility-scale PV is well-represented throughout the

nation, with the exception of the central “wind belt”
states in SPP, Montana, and Wyoming.

Projects larger than 100 MW were built in 2023 in

northern MISO, with solar growth in PJM occurring
in Virginia, Ohio, and Pennsylvania, and solar
expansion in Texas extending beyond the
panhandle.

Montana and South Dakota had their first large

solar projects (3x80 MW) in 2023, as did
Alaska (6 MW). West Virginia’s first project was
completed in 2024 (80 MW).

Only North Dakota and New Hampshire still await

their first utility-scale solar projects in our sample.

Utility-Scale Solar, 2024 Edition

You can explore this data interactively at https://emp.lbl.gov/technology-trends 13
http://utilityscalesolar.lbl.gov
 Texas and the non-ISO Southwest added the most utility-scale
solar capacity in 2023
PV project population: 1,506 projects totaling 80.2 GWAC After a temporary decline in 2022, utility-
scale solar deployment set again new
records in 2023, with multiple GW of
additions in ERCOT (4.2 GWAC), the non-
ISO Southeast (3.1 GWAC), PJM (2.8 GWAC),
CAISO (2.7 GWAC), MISO (2.7 GWAC) and
the non-ISO West (2.5 GWAC) regions.

Taking a state perspective, Texas led the

nation with 4.3 GWAC. California’s USS
growth accelerated in 2023 to 2.6 GWAC—its
greatest deployment since 2016. Florida
(1.8 GWAC), Ohio (0.9 GWAC), and
Wisconsin (0.8 GWAC) scored 3rd to 5th
place.

In cumulative deployment, ERCOT (15

GWAC) is still lagging CAISO (19 GWAC), and
the non-ISO Southeast (17 GWAC) although
the gap is narrowing.
Utility-Scale Solar, 2024 Edition
You can explore this data interactively at https://emp.lbl.gov/capacity-and-generation-state 14
http://utilityscalesolar.lbl.gov
Half of new solar capacity is built in Energy Communities
Annual Capacity Additions (GWAC) >5 MWAC The Inflation Reduction Act offers a tax credit adder for
18 Projects built in new solar projects located in “Energy Communities”,
USS Projects Not in Energy Communities "2023 Energy Projects built
Multiple Criteria
which are areas with:
16 Communities" in Energy
Coal Community before IRA took - Employment or tax revenue from coal, oil, natural
Communities
14 Fossil Fuel Employment + Unemployment Community effect
gas and greater unemployment than national mean
12 - A closed coal mine or coal power plant
10 - Contaminated properties (brownfields)
49%
8
51% 49%
6 52%
4 52% 40%

52% 66% 45% 44%

2 31%
34%
0
2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 H1
2024

New projects built since the passing of the IRA may not yet have been
intentionally sited to capture the bonus (interconnection processes take
several years). Nearly half of the new solar capacity built since 2022 is
located in areas qualifying as Energy Communities.

Note: Some solar projects developed on brownfield sites may not be accurately classified as part of Utility-Scale Solar, 2024 Edition
15
an Energy Community due to data availability issues. *H1 2024 data is preliminary from EIA 860M. http://utilityscalesolar.lbl.gov
Projects with tracking technology dominated 2023 additions

PV project population: 1,504 projects totaling 80.2 GWAC

Annual Capacity (GWAC) Cumulative Capacity (GWAC) Projects using single-axis tracking
20 100
have consistently exceeded fixed-tilt
Columns show annual capacity additions (left scale) installations since 2015, and
18 90
Areas show cumulative capacity (right scale) dominated again in 2023, with 96% of
16 80
Tracking all new capacity using tracking—the
14 70 greatest ever.
12 60
10 50 Upfront cost premiums for trackers
8 40 have generally fallen over the years,
6 30
resulting in favorable economics in
Fixed-Tilt most of the United States thanks to
4 20
increased generation (though 2023
2 10
saw again an uptick in cost
0 0 premiums—discussed later).
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023

Utility-Scale Solar, 2024 Edition

You can explore this data interactively at https://emp.lbl.gov/technology-trends 16
http://utilityscalesolar.lbl.gov
Use of c-Si modules grew again in 2023

PV project population: 1,494 projects totaling 79.3 GWAC c-Si modules have been the dominant
Annual Capacity (GWAC) Cumulative Capacity (GWAC) module technology at large-scale solar
20 100 projects in the US since 2015. After a
Columns show annual capacity additions (left scale) temporary decline in relative growth in
18 90
Areas show cumulative capacity (right scale)
16 80 2022, c-Si modules expanded their
c-Si market share again in 2023 to 72% of
14 70
12 60
newly installed capacity.
10 50
Thin-film modules grew in popularity
8 40
between 2018 and 2021 as they were
6 30
Thin-Film not subject to Section 201 import
4 20
tariffs. In 2023 they reached a new
2 10
record annual deployment of 5 GWAC.
0 0
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023

Note: The 2023 sample includes 6 projects (0.8GWAC) without conclusive module type data
which are excluded from the graph above
Utility-Scale Solar, 2024 Edition
You can explore this data interactively at https://emp.lbl.gov/technology-trends 17
http://utilityscalesolar.lbl.gov
Solar projects built were built in very solar rich areas in the early 2010s. Project
locations and associated resource quality have become much more diverse since then.

PV project population: 1,506 projects totaling 80.2 GWAC The average long-term global horizontal
Long-Term Average Annual GHI at Newly Built Sites (kWh/m2/day) irradiance (GHI) at newly built sites declined
0 5 10 15 20 25 from 2013 through 2017 as the market
6.0 expanded to less-sunny states. This metric
All PV
rebounded slightly in 2023 to
Fixed-Tilt PV 4.59 kWh/m2/day.
5.5 Tracking PV
Fixed-tilt PV is increasingly relegated to
lower-insolation sites, while tracking PV is
5.0
increasingly pushing into those same areas
(note the decline in its 20th percentile).
4.5
Exceptions are fixed-tilt installations in windy
regions (Florida), on brownfields and landfill
4.0 sites, and on particularly challenging terrain.
About 25% of these projects now have a
Markers show median values, with 20th and 80th percentiles
south-western orientation to maximize
3.5 evening production.
n 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023
10 35 44 38 65 89 152 167 96 110 164 157 153 221
GW 0.2 0.5 1 1.3 3.2 2.9 7.4 4.1 4 4.6 9.6 12.5 10.4 18.5 All else equal, the buildout of lower-GHI sites
dampens sample-wide capacity factors
Commercial Operation Year
(reported later).

Note: We use NREL’s NSRDB to estimate long-term solar resource quality for each new USS project. Utility-Scale Solar, 2024 Edition
18
Bars are sample-wide medians, markers show distribution for fixed-tilt and tracking projects. http://utilityscalesolar.lbl.gov
The median inverter loading ratio (ILR) is higher for fixed-tilt
projects than tracking projects
PV project population: 1,500 projects totaling 79.6 GWAC As module prices have fallen (faster
Inverter
0 Loading Ratio5(DC:AC) 10 15 20 25
than inverter prices), developers have
oversized the DC array capacity
1.50 All PV relative to the AC inverter capacity to
Fixed-Tilt PV enhance revenue and reduce output
1.40 Tracking PV variability.

1.30
In 2023, the median inverter loading
ratio (ILR: MWDC to MWAC ratio) was
1.20 1.34, and was higher for fixed-tilt
installations (1.38) than for tracking
1.10 projects (1.32).
Markers show median values, with 20th and 80th percentiles
1.00
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 All else equal, a higher ILR should
n 10 35 44 38 65 89 150 167 96 110 164 157 153 217 boost capacity factors (denominated
GW .2 0.5 1 1.3 3.2 2.9 7.4 4.1 4 4.6 9.6 12.5 10.4 18.
in AC terms and discussed later in the
Commercial Operation Year report).

Utility-Scale Solar, 2024 Edition

19
http://utilityscalesolar.lbl.gov
 Capital Costs (CapEx)
and
Operation & Maintenance (O&M) Costs

Utility-Scale Solar 2024 Edition

http://utilityscalesolar.lbl.gov 20
Capital and Operation & Maintenance Costs: data and methodology

CapEx Data:
– Project-level capital expenditure (CapEx) estimates are sourced from a combination of Form EIA-860, Section 1603 grant data from the U.S. Treasury, FERC
Form 1, data from applicable state rebate and incentive programs, state regulatory filings, company financial filings, interviews with developers and owners, trade
press articles, and data previously gathered by NREL.
– CapEx estimates for projects built from 2013-2022 have been cross-checked against confidential EIA-860 data obtained under a non-disclosure agreement. The
close agreement between the confidential EIA data and our other sources in most cases provides comfort that our data collection process yield reputable CapEx
estimates.
CapEx Methods:
– We present data in $/WAC terms to facilitate cost comparison between generators of multiple fuel types. The accompanying data file on our project website also
provides detailed data in $/WDC terms.
– We define cost scope in close alignment with EIA’s 860 Schedule 5B (p29) to include:
• construction costs (civil and structural costs, equipment and installation, electrical and instrumentation, indirect costs (incl. overhead and profits) and owner costs (incl.
tie-in and potential transmission network upgrades)). For a detailed analysis of interconnection costs of utility-scale solar see https://emp.lbl.gov/interconnection_costs.
• construction finance costs.
O&M Data:
– Plant-level operation and maintenance costs, capacity, net generation, and construction year are sourced from FERC Form 1 Annual Reports, which are filed by
major electric utilities.
O&M Methods:
– We exclude O&M cost observations from the year a plant was constructed to avoid data based on a partial year of operations.
– We also exclude projects ≤5 MW in size, consistent with our definition of utility-scale.
– We present data for combined operations and maintenance costs in $/kWAC (capacity denomination) and $/MWh (generation denomination) terms.

Utility-Scale Solar, 2024 Edition

21
http://utilityscalesolar.lbl.gov
Installed costs of PV have fallen by 8% since 2022, to $1.43/WAC
($1.08/WDC) in 2023
Sample: 1,279 projects totaling 66.5 GWAC Since 2010, costs for installed utility-scale PV have
fallen by 75% (or 10% annually):
▪ Capacity-weighted means (reflecting the average
costs of solar capacity) decreased in real terms from
$1.56/WAC in 2022 to $1.43/WAC in 2023.
▪ Medians (reflecting typical project costs) decreased
from $1.61/WAC in 2022 to $1.33/WAC in 2023.

Despite strong inflationary pressures, we have not

observed cost increases in real dollar terms in recent
years, unlike some other industry observers.

The lowest 20th percentile of project costs fell in real

terms from $1.3/WAC ($1.0/WDC) in 2022 to $1.2/WAC
($0.9/WDC) in 2023.

Historical sample is robust (covering 96% of installed

capacity through 2022). 2023 data covers 34% of new
Note: The 2023 sample does not yet include data contributions from EIA 860, cost projects (75) or 37% of new capacity (6.9 GWAC).
findings may thus be revised as higher quality data will become available.

Detailed statistics in $/WDC are shown in the accompanying data workbook. Utility-Scale Solar, 2024 Edition
22
http://utilityscalesolar.lbl.gov
Tracking projects cost $0.2/W more than fixed-tilt projects on average

Sample: 1,279 projects totaling 66.5 GWAC We focus here on cost differences between projects
using tracking and fixed-tilt mounting. The graph shows
capacity-weigthed average costs by mounting type
across our sample but does not control for other factors
that influence total project costs (equipment, labor, land,
grid interconnection, project size…).

Trackers can sustain some higher upfront costs

because they deliver more energy per installed
capacity.

Over time tracking projects have often been more

expensive, at least on average across our sample, but
the cost premium has fluctuated and at times even
reversed (like in 2016).

Beginning in 2020, tracker installations boomed. By

2023, 96% of all new capacity used trackers, and
tracking projects ($1.4/WAC or $1.1/WDC) were slightly
more expensive than fixed-tilt projects ($1.2/WAC or
$0.9/WDC).

Utility-Scale Solar, 2024 Edition

23
http://utilityscalesolar.lbl.gov
Larger solar projects (>50 MW) cost 13% less than smaller (5-50 MW)
per MW of installed capacity in 2023
Sample in 2023: 76 projects totaling 7.1 GWAC Differences in project size could
potentially explain cost variation—
we focus only on 2023 for this slide.

Cost savings seem to occur

especially in projects larger than 50
MWAC at ~$1.4/WAC vs. $1.6/WAC for
smaller projects.

In $/WDC terms, prices seem to

decline especially among the largest
projects:
❑ $1.28/WDC for 5-20 MW
❑ $1.23/WDC for 20-50 MW
❑ $1.18/WDC for 50-100 MW
❑ $1.05/WDC for 100-700 MW

Utility-Scale Solar, 2024 Edition

24
http://utilityscalesolar.lbl.gov
Operation and maintenance (O&M) costs have decreased by 73%
since 2012, but remained flat the past 3 years
PV project population in 2023: 145 projects totaling 7.4 GWAC
Regulated utilities report solar O&M costs for plants that they own, Cost Scope (per guidelines for FERC Form 1):
representing a mix of technologies and at least one full operational year. • Includes supervision and engineering, maintenance, rents, and training
• Excludes payments for property taxes, insurance, land royalties,
These O&M costs are only one part of total operating expenses. performance bonds, various administrative and other fees, and overhead

Projects built since 2019 report much lower O&M costs in their first three
Median O&M costs for the cumulative sample have declined from years of operation compared to older ones, potentially due to a narrower
about $39/kWAC-year or $22/MWh in 2012 to about $11/kWAC-year scope of service agreements. Starting in year 6 of a project’s life there does
or $7/MWh in 2023. not appear to be a sustained upward or downward trend in O&M costs.

Utility-Scale Solar, 2024 Edition

Note: Detailed project-level data are shown in the accompanying data workbook. 25
http://utilityscalesolar.lbl.gov
Performance (Capacity Factors)

Utility-Scale Solar 2024 Edition

http://utilityscalesolar.lbl.gov 26
PV performance analysis: data and methodology
Net generation data are sourced largely from EIA Form 923. These data reflect net generation and thus exclude
energy used by the plant itself. They also exclude energy that was curtailed (for economic or system stability
reasons). Outliers and low-quality data are dropped from the analysis. We exclude observations from the first
calendar year of the plant’s operation.

Net Capacity Factors (AC) measure a plant’s performance, representing the ratio of its actual annual generation
delivered to the grid to the maximum possible annual output if it operated continuously every hour of the year.
𝐴𝑛𝑛𝑢𝑎𝑙 𝑁𝑒𝑡 𝐺𝑒𝑛𝑒𝑟𝑎𝑡𝑖𝑜𝑛 (𝑀𝑊ℎ)
𝐴𝑛𝑛𝑢𝑎𝑙 𝑁𝑒𝑡 𝐶𝑎𝑝𝑎𝑐𝑖𝑡𝑦 𝐹𝑎𝑐𝑡𝑜𝑟 =
𝐶𝑎𝑝𝑎𝑐𝑖𝑡𝑦 𝑀𝑊𝐴𝐶 ∗ 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 ℎ𝑜𝑢𝑟𝑠 𝑖𝑛 𝑦𝑒𝑎𝑟

We use MWAC capacity terms in our capacity factor calculations to facilitate comparisons with other bulk system
generator types.

Annual generation can vary based on weather and climate variability, system degradation, system uptime, or
curtailment. We thus present primarily cumulative net capacity factors, which represent the average capacity factor
over the lifetime of a project up until the most recent reported period (i.e., no future modeled generation data).

Utility-Scale Solar, 2024 Edition

27
http://utilityscalesolar.lbl.gov
PV performance varies widely among projects, driven by resource
availability and project design choices
Sample: 1,253 plants totaling 61.4 GWAC
The cumulative net capacity factor is
typically around 24% but ranges from 7% to
35% among all projects in our sample.

Project-level variation in PV capacity factor

is driven by:
❑ Solar Resource (GHI): Strongest solar resource
quartile has ~9 percentage point higher capacity
factor than lowest resource quartile

❑ Tracking: Adds ~4 percentage points to capacity

factor on average, with improvement from tracking
more pronounced in higher solar resource areas

❑ Inverter Loading Ratio (ILR): Highest ILR

quartiles have on average ~2 percentage point
higher capacity factors than lowest ILR quartiles

Utility-Scale Solar, 2024 Edition

You can explore this data interactively at https://emp.lbl.gov/pv-capacity-factors 28
http://utilityscalesolar.lbl.gov
Tracking boosts capacity factors by roughly 5 percentage points
in high-insolation regions
Sample: 1,253 plants totaling 61.4 GWAC Not surprisingly, capacity
factors are highest in
California and the non-ISO
West, and lowest in the
Northeast (ISO-NE and
NYISO).

Tracking yields more benefits

compared with fixed-tilt
installations in regions with
strong solar resources,
leading to a greater proportion
of tracking projects in those
regions.
Notes: The NYISO tracking sample comprises just projects, possibly driving unexpected results.
Capacity factors represent weighted means by capacity (MWAC).
Utility-Scale Solar, 2024 Edition
You can explore this data interactively at https://emp.lbl.gov/pv-capacity-factors 29
http://utilityscalesolar.lbl.gov
Since 2013, competing drivers have caused average capacity factors
by plant vintage to stabilize
Sample: 1,253 plants totaling 61.4 GWAC Cumulative capacity-weighted capacity factors
improved for projects from 2010 to 2013 due to
increased DC-oversizing, adoption of single-axis
tracking, and better solar resources.

Since 2013, capacity factors have stagnated due to

mixed factors: while tracking has become
widespread (50% to 90%) and ILR has seen minor
growth, new projects have expanded into less sunny
regions (average GHI decreased from 5.30 to 4.32
kWh/m²/day).

In our latest cohort (145 projects, 2022 vintage),

Columns show capacity-weighted capacity factor (left axis) annual output declined by 1% (absolute) or 4%
Lines show changes in major drivers, indexed by value in 2016 (right axis) (relative) compared to 2021 projects. Key drivers
remained stable, and irradiance-based modeling
even suggested a slight performance increase. We
Flat trend since 2014 is not necessarily negative, but rather a sign of will continue monitoring potential performance issues
a market that is expanding geographically into less-sunny regions as more data emerges.

Utility-Scale Solar, 2024 Edition

You can explore this data interactively at https://emp.lbl.gov/pv-capacity-factors 30
http://utilityscalesolar.lbl.gov
 Plant output declines with age, but the performance of newer
projects has fallen at a slower rate compared to older projects
Sample: 905 plants totaling 37.4 GWAC Annual system-level performance degradation varies by
cohort, with newer plants degrading 0.9% per year
compared to 1.47% for older plants.
Fixed effects regression model defined by:
𝐶𝐹𝑓,𝑡 𝑎𝑐𝑡𝑢𝑎𝑙 = 𝐶𝐹𝑓,𝑡 𝑖𝑑𝑒𝑎𝑙 + 𝑆𝑓 + 𝐴 𝑇 + 𝜖𝑓,𝑡
where:
𝐶𝐹𝑓,𝑡 𝑎𝑐𝑡𝑢𝑎𝑙 = Actual capacity factor of plant f at time t (raw
empirical data, but grossed up for curtailment in CAISO and
ERCOT)
𝐶𝐹𝑓,𝑡 𝑖𝑑𝑒𝑎𝑙 = “Ideal” capacity factor of plant f at time t, estimated
based on physical plant characteristics and solar resource at the
site
𝑆𝑓 = Site-level fixed effects of plant f to control for differences in
capacity factor across plants
Note: Sample includes plants built through 2020 (model requires two years of 𝐴 𝑇 = Age fixed effects at time t to control for differences in
performance with weather data available through 2022). All four slopes are capacity factor within plants
statistically significant, but commercial operation date (COD) ≥ 2015 is not 𝜖𝑓,𝑡 = Residual of plant f at time t
statistically different from COD ≥ 2013 or ≥ 2017.

Note: For greater detail on methods, see Bolinger et al. (2020) System-level Performance and Utility-Scale Solar, 2024 Edition
31
Degradation of 21 GWDC of Utility-Scale PV Plants in the United States. http://utilityscalesolar.lbl.gov
 Levelized Cost of Energy (LCOE) and
Power Purchase Agreement (PPA) Prices

Utility-Scale Solar 2024 Edition

http://utilityscalesolar.lbl.gov 32
LCOE analysis: data sets and methodology
Methods:
- For our project-level LCOE estimates of solar projects we follow the formula published in NREL’s Annual Technology Baseline.
- We use LCOE as proxy for generation costs in later parts of this presentation. It is important to note that additional integration costs
(transmission needs beyond what is captured via interconnection costs and LMP congestion components or ancillary service costs) are not
fully accounted for here.

Data and Assumptions:

– LCOE will be presented first without and then later with inclusion of federal tax credits (assuming labor requirements for ITC and PTC are met,
including Energy Community adders where applicable, but assuming no Domestic Content adders).
– Project-level variation:
• Capex: LCOE is only calculated for projects with empirical cost estimates, only costs of solar components are used for PV-battery projects.
• Net Capacity Factor: We use empirical annual NCF estimates based on EIA 923 data when available. For missing and future years we assume
annual degradation rates ranging between 1.47% (pre-2013) and 0.9% (post-2016). For projects without any reported generation (e.g., most recent
COD cohort) we use the regional average NCF of recent projects. NCF is levelized over the project design life.
– Cohort-level variation:
• OpEx is levelized and declines from $41/kWDC-yr in 2007 to $16/kWDC-yr in 2023 (in 2023$, based on prior LBNL and NREL Benchmarks)
• Project design life increases from 21.5 years in 2007 to 35 years in 2021 and thereafter (prior LBNL research).
• Weighted average cost of capital (WACC):
– based on a constant 70%/30% debt/equity ratio and time-varying market rates.
– Combined income tax rate of 38.25% pre-2018 and 24.95% post-2017.
– 5-yr MACRS; forward-looking inflation expectations range from -0.2% (early Covid pandemic) to 4.2%.
– Real WACC for 2023 COD projects is 2.58%.

Utility-Scale Solar, 2024 Edition

33
http://utilityscalesolar.lbl.gov
Average LCOE (without the ITC/PTC) has been largely stable since
2021
Sample: 1,261 projects totaling 66.3 GWAC Utility-scale PV’s average LCOE has
fallen by 80% since 2010, driven by lower
capital costs and operating expenses, as
well as increased project design life.

Counteracting these beneficial longer-

term trends are falling national average
capacity factors since 2016 and less
favorable financing terms since 2020.

Average LCOE (not including any tax

credits) increased slightly among projects
coming online recently, from $45/MWh in
Note: LCOE estimates depicted here do not include tax credit benefits. 2022 to $46/MWh in 2023.
Only preliminary data is available for new solar projects coming online in 2023.
Findings may shift as more Capex and project-specific performance data become available.
Utility-Scale Solar, 2024 Edition
See interactive visualization at https://emp.lbl.gov/capex-lcoe-and-ppa-prices-region 34
http://utilityscalesolar.lbl.gov
LCOE varies between regions due to differences in solar resource
quality, project costs, and system size
Sample: 214 projects totaling 16.9 GWAC
Lower-insolation regions (ISO-NE,
NYISO, PJM, MISO) will always have
higher LCOEs than higher-insolation
regions (ERCOT, CAISO, the non-ISO
West and Southeast), but the difference
has narrowed over time.

Among projects coming online in 2022

and 2023, large projects in the non-ISO
West, ERCOT, and CAISO had the
lowest cost ($37, 41 and $42/MWh),
while smaller projects in ISO-NE and
NYISO had the highest cost in our
Note: LCOE estimates depicted here do not include tax credit benefits. sample ($76 and $78/MWh).
Only preliminary data is available for new solar projects coming online in 2023.
Findings may shift as more Capex and project-specific performance data become available.

Utility-Scale Solar, 2024 Edition

See interactive visualization at https://emp.lbl.gov/capex-lcoe-and-ppa-prices-region 35
http://utilityscalesolar.lbl.gov
Most projects installed in 2023 seem to benefit more from the
Production Tax Credit than the Investment Tax Credit
Available federal tax credits lower the effective LCOE
shown on the previous slides. The graphs show post-
incentive LCOE variation by project capex and
performance, both for the Investment Tax Credit (ITC,
left) and the new Production Tax Credit (PTC, center).
Introduced by the Inflation Reduction Act, the PTC is
paid for the first 10 years ($28.5/MWh and rising with
inflation) – levelized over a 35-year project lifetime it
reduces LCOE by ~$14.5/MWh.

The right column compares the benefit of each tax

credit, with red squares showing where the ITC
results in a lower LCOE (higher-cost, lower-
performing projects) and the blue squares showing
where the PTC is preferable (lower-cost, higher-
performing projects).

For the 2023 COD cohort, preliminary data indicates

that most projects (54 out of 74) would benefit more
from the Base PTC than the ITC (although the PTC
comes with some challenges such as greater
performance risk that we do not account for in this
comparison).

The lower column repeats the analyses but includes

the bonus adder available for projects sited in Energy
Communities (more details on the next slide).

Utility-Scale Solar, 2024 Edition

36
http://utilityscalesolar.lbl.gov
2023 solar projects in Energy Communities have an average after-tax credit
LCOE of $26/MWh, compared to $34/MWh in the rest of the country
Sample: 75 projects totaling 7.1 GWAC (23 EC projects, 3.1 GWAC) About half of all new utility-scale solar capacity in
2023 qualifies for the Energy Community Tax Credit
Adder. Not considering potential Domestic Content
adders, these project are eligible for a 40% ITC or a
$31.6/MWh PTC over the first 10-years ($2023) –
equivalent to $16.1/MWh levelized over a project’s
lifetime.

Using a simple multivariate regression model, solar

projects in Energy Communities in our sample are
on average $0.22/WAC cheaper than similarly
designed projects outside of Energy Communities,
contributing to a lower LCOE before any tax credits
are applied.

The ITC benefit over the PTC increases if additional

adders (like Energy Community or Domestic
Content) are available for a project, but even among
Note: Only preliminary data is available for new solar projects coming online in 2023. our 23 projects in Energy Communities, 16 have a
Findings may shift as more Capex and project-specific performance data become available. lower LCOE with the PTC than the ITC.
Utility-Scale Solar, 2024 Edition
37
http://utilityscalesolar.lbl.gov
 Power Purchase Agreement (PPA) price analysis:
data sets and methodology
PPA prices are from utility-scale solar plants built since 2007 or planned for future installation, and include:
– 472 PV-only contracts totaling 36.8 GWAC
– 104 PV+battery contracts totaling 13.0 GWAC of PV capacity and 7.8 GWAC / 30.9 GWh of battery capacity (presented in a later section)
– 5 concentrating solar thermal power (CSP) contracts totaling 1.2 GWAC (presented in a later section)
PPA prices reflect the bundled price of electricity and RECs as sold by the project owner under the PPA
– Dataset excludes merchant plants, projects that sell renewable energy certificates (RECs) separately, and most direct retail sales
– PPAs are priced to recover both capital and other ongoing operational costs while accounting for the receipt of state and federal incentives (e.g., the ITC) and, as a
result, do not simply reflect solar generation costs. Ultimately PPA prices reflect marketplace conditions, including the supply of ready-to-build plants, cost of capital,
and demand for energy, capacity, and RECs.
Data collection
– We gather PPA price data from a combination of FERC Electric Quarterly Reports, FERC Form 1, Form EIA-923, state regulatory filings, company financial filings,
and trade press articles. We prioritize data quality over quantity in this process. That is, we only include a PPA within our sample if we have high confidence in all of
the key variables such as execution date, starting date, starting price, escalation rate (if any), time-of-day factor (if any), and term.
– To augment our PPA price sample, and to gain visibility into corporate PPA pricing (which is not well-represented within our sample), we also compile LevelTen
Energy1 and Trio2 data on PPA offers (25th percentile). These often reflect shorter contract durations and target voluntary and corporate offtakers, though fewer
contract specifics are known relative to the PPA data we collect directly.
Levelization methodology
– We deflate the nominal dollar price series to 2023 dollars using a GDP deflator (actual deflators historically, along with projected future deflators), and then levelize
the resulting price series using a 4% real discount rate.
• For PPA prices we collect, prices are levelized over the full term of each contract, after accounting for any escalation rates and/or time-of-delivery factors.
• For LevelTen Energy and Trio, we assume the reported prices are for 12-year, flat-priced (in nominal dollars) PPAs that commence in the following calendar year.

1 https://www.leveltenenergy.com/ppa Utility-Scale Solar, 2024 Edition

38
2 https://www.trioadvisory.com/publications
http://utilityscalesolar.lbl.gov
Since 2016, levelized PPA prices have tracked the LCOE accounting
for tax credits of utility-scale PV
Sample: 1,266 projects totaling 66.3 GWAC
This graph contrasts solar LCOE with and without tax
credits - choosing either PTC or ITC for each project
that results in lowest cost (not including Domestic
Content adder).

With new tax credits becoming available in 2023 we see

for the first time since 2010 a widening gap between
pre- and post-incentive LCOE.

While generation-weighted average LCOE increased

slightly in 2023 before the application of tax credits
($46.5/MWh vs. $44.8 in 2022), post-credit LCOE
continued to fall ($30.5/MWh vs. $33.3/MWh).

Since 2016, levelized PPA prices charted by plant COD

have closely tracked or hovered slightly below the
LCOE with tax credits. This suggests a pass-through of
Note: Only preliminary data is available for new solar projects coming online in 2023. these tax credits and a competitive PPA market.
Findings may shift as more Capex and project-specific performance data become available.
Utility-Scale Solar, 2024 Edition
39
http://utilityscalesolar.lbl.gov
Individual PPAs typically follow the national price trend, though
there are high-price outliers for projects in NY, HI, and New England
Full sample: 472 PPAs, 36.8 GWAC Post-2015 sample: 263 PPAs, 21.1 GWAC
Levelized PPA Price (2023 $/MWh) Levelized PPA Price (2023 $/MWh)
CAISO CAISO
250 West (non-ISO) 180 West (non-ISO)
MISO MISO
160
SPP SPP
200 ERCOT 140
25 MW ERCOT
PJM PJM
120
NYISO NYISO
150
Southeast (non-ISO) 100 Southeast (non-ISO)
550 MW ISO-NE ISO-NE
Hawaii 80 Hawaii
100
60

50 40
20
0 0
2006 2008 2010 2012 2014 2016 2018 2020 2022 2016 2017 2018 2020 2021 2022 2024
PPA Execution Date PPA Execution Date

o Power Purchase Agreement (PPA) prices are levelized over the full term of each contract, after accounting for any escalation
rates and/or time-of-delivery factors, and are shown in real 2023 dollars
o Contract term is between 20 and 25 years (inclusive) for 76% of projects in the full sample
o >95% of the sample is currently operational
o Aided by the 30% ITC, PPAs in our sample executed in 2021 or later are usually priced around $20-$30/MWh for projects in
CAISO and the non-ISO West, and $35-$47/MWh for projects elsewhere in the continental United States

Utility-Scale Solar, 2024 Edition

You can explore this data interactively at https://emp.lbl.gov/pv-ppa-prices 40
http://utilityscalesolar.lbl.gov
Average PPA prices in the Lower 48 fell by ~88% (or ~19%/year) from
2009-2019, but have been stagnant or rising ever since
Average Levelized PPA Price (2023 $/MWh) This graph focuses on national and
regional average PPA prices, rather
180 than project-level (as in the prior slide).
160 CAISO Southeast (non-ISO) The generation-weighted national
140 average was $35/MWh in 2023 (based
on a small sample of 7 PPAs), up
120 considerably from 2019’s low of
100 $23/MWh.
PJM
80 Year-Region combinations with fewer
than 2 PPAs are excluded from the
60
MISO graph (dashed line segments indicate
40 that the line is skipping over such
Lower 48 years).
20 ERCOT
West (non-ISO)
0 The graph reflects PV-only pricing, not
2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 PV+battery (PV+battery PPA prices are
PPA Execution Year
presented separately, in a later
section).

Utility-Scale Solar, 2024 Edition

You can explore this data interactively at https://emp.lbl.gov/pv-ppa-prices 41
http://utilityscalesolar.lbl.gov
LevelTen Energy and Trio’s utility-scale PV PPA price indices match
the increasing trend seen in the LBNL sample since 2021
LevelTen PPA Price Index (Levelized 2023 $/MWh, 25th percentile of first-year offer) Average Levelized PPA Price (2023 $/MWh)
$70 60
PJM Trio
$60 National
50
Continental Index MISO Sample
$50 (P25)
CAISO LevelTen
SPP 40
$40 National
ERCOT Sample (P25)
$30 30

$20 20
Note:
Note: LevelTen does not consistently report PPA prices for ISO-NE or NYISO
$10 LBNL Lower 48 Sample small
The Continental Index represents the value across all North American offers in 10 sample
LevelTen’s Marketplace irrespective of ISO.
$0 in 2024
2Q20

3Q23
1Q19
2Q19
3Q19
4Q19
1Q20

3Q20
4Q20
1Q21
2Q21
3Q21
4Q21
1Q22
2Q22
3Q22
4Q22
1Q23
2Q23

4Q23
1Q24
2Q24
0
2018 2019 2020 2021 2022 2023 2024

Trio PPA Price Index (Levelized 2023 $/MWh, 25th percentile of first-year offer)
PPA Execution Year

$70 To augment our PPA price sample, and to gain visibility into corporate PPA
PJM
$60 pricing (which is not well-represented within our sample), we present
MISO LevelTen Energy and Trio’s PPA price indices.
$50
SPP
$40
Drivers of PPA price increases in recent years include:
CAISO ERCOT
$30
• High interest rates leading to higher financing costs
• Long lead times for high- and medium-voltage equipment
$20
• Supply constraints by these equipment lead times and long
$10 interconnection and permitting timelines
Note: Trio does not report PPA prices for ISO-NE or NYISO
$0 • High demand from corporations and utilities in advance of 2025 and 2030
4Q21

2Q22
1Q21

2Q21

3Q21

1Q22

3Q22

4Q22

1Q23

2Q23

3Q23

4Q23

1Q24

2Q24

emissions targets Utility-Scale Solar, 2024 Edition

42
http://utilityscalesolar.lbl.gov
 Solar PPA prices are competitive with wind. Recent declines in natural gas
prices give existing gas-fired generators a near-term cost advantage.
Levelized PPA and Gas Price (2023 $/MWh) 2023 $/MWh
200 50
PV PPA prices
180
160 40
140
120 30 Gas
Levelized 20-year
100 Solar
EIA gas price projections
80 20
STEO + AEO2023 reference case natural gas fuel cost projection
60 Median PV PPA price
(with full range of projections)
40 (with 10th and 90th percentile range)
10
20
Wind PPA prices
0 0
2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2024 2026 2028 2030 2032 2034 2036 2038 2040
PPA Execution Date and Gas Projection Year

The left graph shows that solar PPA prices have largely closed the gap with wind, and some contracts are competitive with levelized gas price
projections.
The right graph compares recent (2022-24 execution date) solar PPA prices (extending over their contract terms through 2040) to the range of gas
price projections from the EIA’s Annual Energy Outlook 2023 (AEO 2023) and Short-Term Energy Outlook (STEO). Gas price projections through
2025 have dropped considerably compared to expectations one year ago. PV can help hedge against fuel price risk in the short to medium-term,
and by the mid-2030s most PV is projected to be competitive with the cost of burning fuel in an existing combined-cycle natural gas unit (NGCC).
Note that PV PPAs are priced to recover both capital and other ongoing operational costs—for a new NGCC, this would add another ~$20-$80/MWh
(per Lazard data) to the projected fuel costs shown in the graphs.
Utility-Scale Solar, 2024 Edition
43
http://utilityscalesolar.lbl.gov
 Wholesale Market Value,
Air and Climate Benefits, and
Net Value

Utility-Scale Solar 2024 Edition

http://utilityscalesolar.lbl.gov 44
Wholesale market value analysis: data sets and methodology

We estimate the wholesale market value for each utility-scale PV project larger than 1 MW (as reported on Form EIA-860). Each project-
level estimate may be prone to some biases - greater emphasis should thus be placed on the aggregate generation-weighted averages
which we calculate for all seven ISOs and ten additional balancing authorities.

We draw from project-level modeled hourly solar generation (using NREL’s System Advisor Model and site- and year-specific insolation
data from NREL’s National Solar Radiation Database and NOAA’s High Resolution Rapid Refresh Model) and de-bias the generation by
leveraging ISO-reported aggregate solar generation and plant-level reported generation by Form EIA-923. Hourly curtailment data is
either derived from plant-level reports (ERCOT: HSL minus MW) or allocated from ISO-level reports (CAISO).

Energy value is the product of hourly solar generation by plant or county and concurrent wholesale energy prices
– Plant-level debiased hourly solar generation σ 𝑃𝑜𝑠𝑡𝑐𝑢𝑟𝑡𝑎𝑖𝑙𝑚𝑒𝑛𝑡 𝐺𝑒𝑛𝑒𝑟𝑎𝑡𝑖𝑜𝑛ℎ ∗ 𝑊ℎ𝑜𝑙𝑒𝑠𝑎𝑙𝑒 𝑅𝑇 𝐸𝑛𝑒𝑟𝑔𝑦 𝑃𝑟𝑖𝑐𝑒ℎ
– Real-time energy price from 𝐸𝑛𝑒𝑟𝑔𝑦 𝑉𝑎𝑙𝑢𝑒 =
σ 𝑃𝑟𝑒𝑐𝑢𝑟𝑡𝑎𝑖𝑙𝑚𝑒𝑛𝑡 𝐺𝑒𝑛𝑒𝑟𝑎𝑡𝑖𝑜𝑛ℎ
• nearest LMP node (ISOs, CAISO’s + SPP’s EIM BAs)
• gateway node from nearby ISO / FERC Lambda for some BAs

Capacity value is the product of a plant’s or county’s capacity credit and capacity prices
– Capacity credit based on plant-level profile; varies by month, season, or year σ 𝐶𝑎𝑝𝑎𝑐𝑖𝑡𝑦 𝐶𝑟𝑒𝑑𝑖𝑡𝑇 ∗ 𝑁𝑎𝑚𝑒𝑝𝑙𝑎𝑡𝑒 ∗ 𝐶𝑎𝑝𝑎𝑐𝑖𝑡𝑦 𝑃𝑟𝑖𝑐𝑒𝑇
– Capacity prices from respective ISO region; prices vary by month, season, or year 𝐶𝑎𝑝𝑎𝑐𝑖𝑡𝑦 𝑉𝑎𝑙𝑢𝑒 =
σ 𝑃𝑟𝑒𝑐𝑢𝑟𝑡𝑎𝑖𝑙𝑚𝑒𝑛𝑡 𝐺𝑒𝑛𝑒𝑟𝑎𝑡𝑖𝑜𝑛 𝑇
– Estimate bilateral capacity prices for regions without organized capacity markets
– Focus on annual value of solar for projects with a full calendar year of operation
– Calculate capacity value for all solar, even if some solar does not participate in capacity markets

For more information, see Berkeley Lab’s publication: “Solar-to-Grid: Trends in System Impacts, Reliability, and Market Value Utility-Scale Solar, 2024 Edition
in the United States with Data Through 2020.” https://emp.lbl.gov/publications/solar-grid-trends-system-impacts-0 45
http://utilityscalesolar.lbl.gov
Market value vs. Generation costs and Environmental benefits:
Scope of value comparisons and methods
• Total wholesale market value is simply the sum of solar’s energy and capacity value
– It represents the “replacement costs” of what an offtaker would have to pay in the wholesale market had they not procured solar generation. Revenues for a solar
project owner are set by their PPA terms and may differ from our estimate. However, in a market with little friction, we expect long-term convergence.
– It does not include any potential additional revenue streams (ancillary service (AS) revenues, renewable energy credits, infrastructure deferral, or resilience that
are not already internalized in wholesale energy and capacity markets).
– It is based on the real-time LMP market and thus reflects the marginal solar value. It does not fully consider sub-hourly variability and forecast errors.
– It excludes broader sectoral impacts such as merit-order effect on power prices or reduced natural gas demand and associated price declines.

• Generation costs are approximated by LCOE (with and without tax credits), but do not include:
– Full integration costs (AS) or transmission needs (beyond LMP congestion components and interconnection network upgrade costs).
– The full cost to the Treasury of federal investment and production tax credits.
– Other costs and benefits to local communities and ecosystems.
– LCOE sample is matched to market value sample on slide 53, but only reflects a subset of solar projects that contribute to the environmental benefits on slide 54
(LCOE projects are greater than 5MWAC with some 2022 COD projects missing).

• Health and climate benefits are approximated by the marginal avoided emission rate * damage per ton of pollutant emitted
– Avoided emissions rates are regression results leveraging hourly generation data by source type, accounting for imports and exports between regions, and time
shifting of impacts through redispatch of hydropower. Coal and gas emissions data are used to determine the emissions avoided from solar in each region.
– Health benefits are a function of the total mass of pollutants avoided and where those reductions occur based on a suite of reduced-complexity air quality health
impact models.
– Climate benefits are calculated as a function of the social cost of carbon, as described in EPA’s Report on Social Cost of Greenhouse Gases (2023).
– We represent central estimates ($125/MWh) from range of plausible values (5%: $38/MWh , 95%: $303/MWh)

Utility-Scale Solar, 2024 Edition

46
http://utilityscalesolar.lbl.gov
Solar curtailment was 3.0% in CAISO and 7.3% in ERCOT in 2023.
Other ISOs and regions do not yet report curtailment.
The rate of curtailment was much higher in ERCOT (7.3%) than
Quarterly and Annual Solar
10.0% in CAISO (3.0%) in 2023, even though solar’s penetration rate
10%
9.4% is far lower in ERCOT (7%) than CAISO (27%). Most of
9% ERCOT’s curtailment occurs in the western part of Texas, driven
8%
7.3%
by transmission/pipeline congestion and excess local electricity
7% 6.7% production (prices for natural gas, a byproduct of oil production,
Q4
6%
6.1%
were at times negative in 2023 in the Permian Basin).
5.0%
Q3
5%
4% Q2 CAISO had 2,051 GWh of solar curtailed in 2023, equivalent to
3.4% 3.0% 3.0%
3% 2.4% 2.6%
the annual output of a hypothetical 816 MWAC tracking PV
2% 1.6%
project operating at an average CA capacity factor of 28.7%
Q1
0.8% 1.0%
1.0% (which would have been 29.6% if not for curtailment).
1% 0.6%
0.1% 0.2%
0%
ERCOT had 2,500 GWh of solar curtailed in 2023, equivalent to
CAISO ERCOT
the annual output of a hypothetical 1142 MWAC tracking PV
project operating at an average TX capacity factor of 25.0%
(which would have been 27.0% if not for curtailment).

Note: The data shows ISO-wide solar curtailment relative to total solar production Utility-Scale Solar, 2024 Edition
47
(sum of utility-scale and distributed solar) http://utilityscalesolar.lbl.gov
Solar's energy and capacity value varied by location

Solar’s average energy and capacity value in 2023

varies from one region to another: It was lower in
CAISO at $27/MWh, but high in many southeastern
balancing authorities ($48-65/MWh), ERCOT
($67/MWh), SPP ($58/MWh), and the Pacific
Northwest ($48-58/MWh).
But value also varies within regions, driven by
transmission congestion, solar resource quality or
differing use of technology like trackers.
For example, in CAISO the northern zone has
typically higher average values than the southern
zone. Solar in southern SPP and NYISO was nearly
$20/MWh more valuable than solar in the north of the
ISOs.
Other markets like ISO-NE show very little variation
in annual average value between projects (10th vs.
90th percentile had a difference of less than
Note: Marker size shows project capacity while marker color shows market value. $2/MWh).
Utility-Scale Solar, 2024 Edition
You can explore this data interactively at https://emp.lbl.gov/utility-scale-solar-value 48
http://utilityscalesolar.lbl.gov
 Solar’s average energy and capacity value across the U.S. was $45/MWh in
2023, similar to pre-pandemic levels
Average Solar Value, with 10th/90th Percentiles of Combined Value (2023 $/MWh) Average Solar Value, with 10th/90th Percentiles of Combined Value (2023 $/MWh) Average Solar Value, with 10th/90th Percentiles of Combined Value (2023 $/MWh)

140 140 140

Capacity Value
Capacity Value
120 120 120
2012-2023
2012-2023 Energy Value
100 100 100
Capacity Value Energy Value Energy Value
2012-2023
80 80 80

60 60 60

40 40 40

20 20 20

0 0 0

Independent System Operator (ISO) Western Non-ISO Balancing Authorities (BA) Southeastern Non-ISO Balancing Authorities (BA)

Energy value typically makes up the bulk of total market value. After high natural gas prices in 2022, solar’s average energy value across the US
returned to more normal levels in 2023 of $34/MWh.

Capacity value is more significant in the non-ISO regions and can add $30-40/MWh in some BAs, where capacity prices are high (e.g., SERC
region) and where the solar profile is still well aligned with peak netload hours.

Variation across years mostly reflects fluctuations in wholesale power prices, but also shows how increasing solar penetration can dampen solar’s
value (e.g., CAISO).

In 2023, combined energy and capacity value was lowest in CAISO ($27/MWh) and highest in ERCOT ($67/MWh). Even though ERCOT reduced
its price cap from $9000/MWh to $5000/MWh in 2023, summer heat waves and associated record demand levels contributed to high prices that
allowed solar to capture value at 2022 levels.

Note: The data shows generation-weighted average annual market value of all large-scale (1 MW+) Utility-Scale Solar, 2024 Edition
49
solar projects in select Balancing Authorities. http://utilityscalesolar.lbl.gov
 Solar’s value factor tends to decline as solar serves a higher share
of a region’s load
Solar Value Factor Solar Market Share Solar Value Factor Solar Market Share Solar Value Factor Solar Market Share

Solar Market 28%

200% 28% 200% 200% 28%
Solar Market Share Solar Value 2012-2023 Solar Market
Solar Value Solar Value 2012-2023
Share 2012-2023 Factor Share
Factor Factor

150% 21% 150% 21% 150% 21%

100% 14% 100% 14% 100% 14%

50% 7% 50% 7% 50% 7%

0% 0% 0% 0% 0% 0%

Western Non-ISO Balancing Authorities (BA) Southeastern Non-ISO Balancing Authorities (BA)
Independent System Operator (ISO)

The columns represent the solar value factor (left axis), the dots show growth in solar market share (right axis)

The 'Value Factor' is defined as the ratio of solar’s total market value (including both energy and capacity) to the market value of a
'flat block' of power (i.e., a 24x7 block). It indicates whether solar’s total revenue is above or below the average wholesale
revenue, with generators delivering electricity during high-value hours achieving a value factor above 100%.“

It controls for fluctuations in energy and capacity prices across years (and across ISOs) and focuses instead on the impact of
solar’s generation profile (and penetration) on value.

Most regions with the highest solar market shares show Value Factors less than 100%, even just 38% in CAISO. However, in
many southeastern BAs solar still provides above-average value despite approaching 10% penetration.

Note: IPCO and SCEG did not have solar projects larger than 1 MWAC operating in the early years. Utility-Scale Solar, 2024 Edition
50
Solar market share in those years only reflects contribution of distributed PV. http://utilityscalesolar.lbl.gov
Solar’s generation profile was the largest source of value differences
between solar and a flat block in 2023
Wholesale Market Value in 2023 (2023 $/MWh) Solar Value Relative to Flat Block Value (% difference) $58
60 50% $46
curtailment Solar market $67
location
value ($/MWh)
50
25%
profile $36
$41
40
Capacity 0%
Value
30
$34
-25%
20

Energy -50%
10 Location
Value
Curtailment
$27 Profile
0 -75%
Flat block Flat block Flat block Solar profile Value of $71 $39 $40 $31 $51 $32 $35
at all nodes at solar nodes at solar nodes at solar nodes flat block
($/MWh) CAISO ISONE NYISO PJM ERCOT MISO SPP
(with curtailment)

Across the seven ISOs, solar projects were usually sited

at locations with above average energy values. The large Solar’s generation profile has the largest impact and
amount of solar deployed in areas with lower relative either hurts (in CAISO, ISO-NE) or helps (in SPP,
value (particularly CAISO) yields a value factor of 82% MISO, ERCOT, PJM, NYISO) solar’s value relative to
across all solar projects in the ISOs. a flat block. Curtailment is becoming a growing issue
for solar in ERCOT.

Note: Numbers and figures shown here only reflect market value in the year 2023 in the seven ISOs Utility-Scale Solar, 2024 Edition
51
and do not include data from other years or non-ISO regions. http://utilityscalesolar.lbl.gov
Market Value vs. PPAs: Rising prices for new PPAs started to
exceed solar’s wholesale market value in some regions in 2023
PPAs provide the power purchaser a hedge value
Solar Value and PPA Price (2023 $/MWh) for price fluctuations over 10 to 20 years. While we
show price trends by individual years, a true
120
Capacity Value benefit accounting should span the length of the
Average Levelized PPA Price
PPA contract. PPA prices are influenced by solar’s
(by execution year) Energy Value
100 LevelTen generation costs, solar’s wholesale market
PPA Price “replacement costs”, and broader supply and
demand dynamics.
80
Trio
PPA Price Solar’s market value has declined over time within
60 several regions. Falling PPA prices had largely
kept pace until PPAs started rising in 2021.
40
Temporarily high energy prices in 2022 more than
compensated for emerging PPA price increases,
20 but PPAs have begun to exceed wholesale market
value in 2023 in CAISO, MISO, PJM, and NYISO,
0 indicating potential future economic challenges
based on solar’s wholesale market value in these
2012-23 2012-23 2012-23 2012-23 2012-23 2012-23 2012-23 2012-23 2012-23
regions.
CAISO ERCOT SPP MISO PJM NYISO ISO-NE West Southeast
(non-ISO) (non-ISO)
In contrast, solar offered greater value than what it
is paid for in PPAs in ERCOT and SPP in 2023.

Note: We do not have sufficient PPA data to present robust trends for every balancing authority. PPAs are Utility-Scale Solar, 2024 Edition
52
indexed here by execution date and span a longer time than the annual wholesale market value estimates. http://utilityscalesolar.lbl.gov
 Market Value vs. LCOE (with tax credits): Newer solar projects had greater
value than their generation costs in 2023, yielding over $1BN in benefits
Sample: Only includes >5 MWAC projects with LCOE and wholesale market value data: 1,079 projects totaling 55.9 GWAC

National average energy and capacity market value has In 2023, recent projects in our sample offer, in
been greater than levelized generation costs (after tax aggregate, a net market value of about $1.1 billion
credits) for new utility-scale solar projects since 2020. that could be passed on to end-use customers. Older
Plants built in 2022 delivered on average $15/MWh projects had generation costs that were higher than
more wholesale market value in 2023 than their LCOE. their wholesale market revenues alone in 2023.

Note: For greater detail on methods and impacts, see Wiser et al (2024) Grid Value and Cost of Utility-Scale Solar, 2024 Edition
53
Utility-Scale Wind and Solar: Potential Implications for Consumer Electricity Bills. http://utilityscalesolar.lbl.gov
Combined climate, air quality, and market values of historical solar fleet
are greater than generation costs and incentives in most regions in 2023
Solar is offsetting electricity generation by coal and natural
gas plants, thereby reducing emission-associated health and
climate damages.
Employing statistical analyses of empirical generation and
emission records, we can infer impacts for regions with
greater solar generation share (shown in graph).
Using avoided emission values from the scientific literature
and air quality models, we estimate the U.S. health benefits
from solar generation in 2023 equaled $24/MWh, based on
reductions to SO2 and NOx emissions. Solar also reduced
global damages caused by climate change at $101/MWh,
based on reductions of CO2 emissions and standard
estimates of the Social Cost of Carbon.
We can subsequently compare market and environmental
value with the weighted generation costs of the solar projects
The estimated aggregate net value of utility-scale solar that contribute to these benefits (COD 2007-2022). The net
generation across the examined regions was $13.7 BN in value is positive in all regions except ISO-NE (which has
2023 (or $16.5 BN when using after tax credit LCOE). both higher solar costs and a relatively clean grid in 2023).

Note: For more detail on methods and discussion of caveats, see Millstein et al (2024) Climate and Utility-Scale Solar, 2024 Edition
54
air quality benefits of wind and solar generation in the United States from 2019 to 2022. http://utilityscalesolar.lbl.gov
 PV+Battery Hybrid Plants
(for more of Berkeley Lab’s analysis of hybrid power plants, see https://emp.lbl.gov/hybrid)

Utility-Scale Solar 2024 Edition

http://utilityscalesolar.lbl.gov 55
Deployment of PV-battery hybrid plants set a record with 5.3GW
greenfield and retrofit capacity in 2023
Sample: 155 projects totaling 12.4 GWAC of PV, 7.0 GWAC of battery capacity, and 22.7 GWh of battery energy
Annual PV Capacity (GWAC) Cumulative PV Capacity (GWAC) Annual (left axis) and Cumulative (right axis) Storage Capacity coupled with PV Hybrids
90
(GWAC)
3.5 7
Columns show annual capacity additions (left scale) Hybrids Retrofits West (non-ISO)
18 Columns show annual capacity
Areas show cumulative capacity (right scale)
New Hybrids 75 3.0 Southeast (non-ISO) 6
SPP Areas show cumulative capacity
15
60 2.5 PJM 5
12 NYISO
New Solar Standalone 2.0 MISO 4
45
9 ISO-NE
1.5 Hawaii 3
30
6 ERCOT
1.0 CAISO 2
15
3
0.5 1
0 0
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023
0.0 0
-3 Solar Standalone -15 2016 2017 2018 2019 2020 2021 2022 2023
retrofitted to Hybrid

The large-scale PV+battery hybrid build-out started slowly in 2016, Most of the new hybrid storage was built in CAISO (22 plants, 1.7
with just 1-12 plants/year built through 2020. The market started in GW storage capacity with ~3.5h storage energy). Hybrids had
earnest in 2021 with 39 hybrid installations. Following steady their first big year in the solar rich non-ISO West (20 plants, 4.0
growth in 2022, 2023 was another record year for newly built GW capacity with ~3.8h energy). Hybrid additions declined
hybrids (37 plants, 4.4 GWAC-PV) while storage retrofits to existing slightly in ERCOT (3 plants, 0.2 GW storage capacity with ~1.3h
stand-alone solar projects declined a bit (15 plants, 0.9 GWAC-PV). energy) and ISO-NE (only 1 project in the MA Smart program).

Utility-Scale Solar, 2024 Edition

56
http://utilityscalesolar.lbl.gov
Greenfield PV+battery project costs fell 15% in 2023 relative to 2022

Sample: 91 plants totaling 9,605 MWAC of PV and 5,360 MW / 16,854 MWh of batteries with CODs from 2018-2023
Combined PV+Battery Costs (2023 $/WAC-PV)
Mean PV and Battery Costs of newly-built Hybrids (2023 $/WAC-PV)
5 Bubble area = storage duration
10
4.5
Combined System Costs
4 $3.66 $3.72 $3.89 5h duration
8
3.5 Battery
3 $1.07 Solar $0.92 Component
Projects with
Component $2.68
6 Battery COD
2.5 $2.24
$1.71 pre-2023
$1.03
2
4
1.5 $2.80
$2.59 $2.06
1 $2.06 $1.72 $1.73 $1.64
$1.50 $1.46 2
0.5 Projects with
Battery COD
0 2023
PV only PV+Bat PV only PV+Bat PV only PV+Bat PV only PV+Bat PV only PV+Bat 0
0% 20% 40% 60% 80% 100% 120% 140% 160% 180% 200%
2019 2020 2021 2022 2023
Battery:PV Capacity

In our newly-built hybrid sample, average combined costs Unfortunately, we do not have a robust sample of separate
have fallen 11% from $2.68/WAC-PV in 2022 (n=22) to battery and PV costs among recent newly-built projects (2023)
$2.24/WAC-PV in 2023 (n=16). as recent EIA 860 capex data is not yet available. In 2022,
Associated average storage duration in our cost sample median storage costs were $500/kWh. As shown above,
increased from 2.9h in 2022 to 3.2h in 2023, and the combined PV+battery costs generally scale with increased
battery:PV capacity ratio increased from 0.6 in 2022 to 0.7 in battery capacity (relative to the PV capacity) and storage
2023. duration.
Note: “Combined System Costs” and “Component System Costs” in left graph may deviate due to Utility-Scale Solar, 2024 Edition
57
different sample sizes contributing to each cost estimate. http://utilityscalesolar.lbl.gov
PPA prices for PV+battery have approximately doubled since
2019/20 lows; Hawaii historically at a premium
Bubble area = battery:PV capacity ratio Bubble area = PV capacity

• Both graphs show same data from sub-sample of 93 plants (retrofits not included); the only difference is what the
bubble size represents
– Hawaii (orange): 22 plants, 0.8 GWAC PV, 0.8 GWAC battery (third round of Hawaii PPAs expected soon)
– Other States (blue): 71 plants, 10.5 GWAC PV, 5.8 GWAC battery
– Storage duration ranges from 2-8 hours; 80 plants have 4-hr duration (the other 13 are 5x2 hr, 1x2.5, 1x3, 1x3.7, 4x5, and 1x8 hr)
• Upward price trend among PPAs on the mainland, with prices in 2024 approximately twice typical prices in 2020
– Rate of hybrid PPA price growth exceeds that of stand-alone solar, which saw increases of ~50-65% since 2020/2021 (see previous
“LCOE & PPA Prices” section)

Utility-Scale Solar, 2024 Edition

58
http://utilityscalesolar.lbl.gov
PPAs that price the PV and storage separately enable us to calculate a
“levelized storage adder,” which has increased in recent years
Graph Sample: 66 PV hybrid projects with 5.9 GWAC of batteries (all 4h duration) in CA (35), NV (16), NM (11), AZ (3) and OR (1)

“How much does storage add to

my standalone PV PPA price?”

A larger capacity battery adds more to a PPA price than a smaller Increased PPA prices for the battery component explain some, but not
battery, when normalized for the PV plant size. This relationship all, of the recent increase in hybrid PPA prices. Levelized price increase
between “levelized storage adders” and Battery-to-PV capacity since 2020 ($2023):
ratios is roughly linear. • Hybrid PPAs: ~$30/MWh-PV (see prior slide)
Retrofits tend to have higher “levelized storage adders” than • Storage Adder: ~$23/MWh-PV
greenfield projects. • All PV PPAs (not just hybrid): ~$10/MWh-PV

Utility-Scale Solar, 2024 Edition

59
http://utilityscalesolar.lbl.gov
Concentrating Solar Thermal Power (CSP)
Plants

Utility-Scale Solar 2024 Edition

http://utilityscalesolar.lbl.gov 60
Sample description of CSP projects
CSP project population: 16 projects totaling 1,781 MWAC After nearly 400 MWAC built in the late-
1980s and early-1990s, no new CSP was
built in the U.S. until 2007 (68 MWAC), 2010
(75 MWAC), and 2013-2015 (1,237 MWAC).

Prior to the large 2013-15 build-out, all

utility-scale CSP projects in the U.S. used
parabolic trough collectors.

The five 2013-2015 projects include:

– 3 parabolic troughs (one with 6 hours of
storage) totaling 750 MWAC (net) and
– 2 “power tower” projects (one with 10
hours of storage) totaling 487 MWAC (net).

Utility-Scale Solar, 2024 Edition

61
http://utilityscalesolar.lbl.gov
With no recent CSP installations in the U.S., empirical installed cost
data are dated
CSP cost sample: 7 projects totaling 1,381 MWAC Small sample of 7 projects using
Installed Costs (2023 $/WAC)
different technologies makes it hard
110 MWAC with 10 hours of storage to identify trends. Newer projects (5
12
11
250 MWAC with 6 hours of storage
built in 2013-15) did not show cost
10 declines, though some included
9
75 MWAC storage or used new technology
8 68 MWAC 250 MWAC each
7 (power tower).
6 377 MWAC
5 PV costs have continuously declined
4 and are now far below the historical
3
2 CSP Trough CSP costs. While international CSP
CSP Tower projects seem to be more
1
PV Capacity-Weighted Mean (for reference)
0 competitive with PV, no new CSP
2006 2008 2010 2012 2014 2016 2018 2020 2022 projects are currently under active
Commercial Operation Year
development in the U.S.

Utility-Scale Solar, 2024 Edition

62
http://utilityscalesolar.lbl.gov
Most newer CSP projects continue to underperform relative to long-
term expectations
CSP capacity factor sample: 7 projects totaling 1,394 MWAC Power Towers: Ivanpah’s (377 MW) capacity factor
fell in 2023 to just 17.3%, well below long term
Annual Capacity Factor (solar portion only) expectations of 27%. Two of Ivanpah’s generators
40% reported no power generation over a combined three-
Capacity-Weighted Annual Average PV in Solana month period in 2023. Crescent Dunes (110 MW with
States with CSP (CA, NV, AZ) 10 hours of storage) performed at just 8.3% capacity
35%
factor in 2023.
Mojave
30%
Trough with storage: Solana (250 MW trough
project with 6 hours of storage) performed at 35.4%
25% Genesis capacity factor in 2023, an increase from the previous
Ivanpah two years but below long-term expectations of >40%.
20%
SEGS III-IX Troughs without storage: Mojave and Genesis
15% (both 250 MW net) were at 27-28% capacity factor in
Nevada Solar
One 2023. Both have performed better than the old SEGS
10% projects (now decommissioned and repowered with
PV) and the 2007 Nevada Solar One project.
5% Crescent
Dunes
SEGS I & II Only Solana, Genesis, and Mojave have matched or
0% exceeded the average capacity factor among utility-
2008 2010 2012 2014 2016 2018 2020 2022 scale PV projects across CA, NV, and AZ.

Utility-Scale Solar, 2024 Edition

63
http://utilityscalesolar.lbl.gov
Though CSP was once competitive, PV PPA prices have declined
dramatically. Without new CSP PPA data, current comparisons are difficult
CSP PPA price sample: 5 projects totaling 1,237 MWAC
Levelized PPA Price (2023$/MWh) When PPAs for the most recent
$250 PV in CA, NV, AZ (for comparison) batch of CSP projects (with CODs
CSP trough, no storage of 2013-15) were signed back in
$200 CSP trough, 6 hours storage 2009-2011, they were still mostly
250 MW competitive with PV.
CSP tower, no storage
$150 CSP tower, 10 hours storage

$100 But CSP has not been able to keep

pace with PV’s price decline. Partly
$50
as a result, no new PPAs for CSP
The offtaker cancelled this PPA in projects have been signed in the
October 2019, following prolonged
underperformance. U.S. since 2011 – though the
$0
technology continues to advance
2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024
overseas.
PPA Execution Date

Utility-Scale Solar, 2024 Edition

64
http://utilityscalesolar.lbl.gov
Capacity in Interconnection Queues

Utility-Scale Solar 2024 Edition

http://utilityscalesolar.lbl.gov 65
 Looking ahead: Strong growth in the utility-scale solar pipeline

Sample: Active bulk-power interconnection requests from 51 interconnection queues. 1085 GW of solar was in the
queues at the end of 2023—312
Capacity in Queues at Year-End (GW)
GW of this total entered the queues
1,200
in 2023 (the remainder entered in
Striped portion indicates the amount of capacity proposed as part of a hybrid plant
earlier years and remain active).
1,000

800
Entered queues in the year shown
571 GW of the 1085 GW of solar in
Entered queues in an earlier year
the queues (i.e., 53%) includes a
600 battery in a PV hybrid
configuration.
400

200 Solar (both in standalone and

hybrid form) is the largest resource
0 within these queues, followed
2014-2023 2014-2023 2014-2023 2014-2023 2014-2023 2014-2023 2014-2023
closely by storage, with wind and
Solar Storage Wind Gas Other Nuclear Coal
gas a distant 3rd and 4th. (All other
resources are negligible in
comparison.)

Note: Not all projects will ultimately be built as many withdraw during the interconnection process. Utility-Scale Solar, 2024 Edition
66
For more details on methods see Berkeley Lab’s Annual Queued Up Report. http://utilityscalesolar.lbl.gov
 Looking ahead: Continued broadening of the market

Sample: Data from 51 interconnection queues across the U.S. Most regions of the country saw
growth in the amount of queued
Solar Capacity in Queues at Year-End (GW) solar, with CAISO, the non-ISO
300
West and Southeast leading the
Striped portion indicates the amount of capacity proposed as part of a hybrid plant
way in 2023
❑ MISO and PJM did not accept
200 new interconnection requests in
Entered queues in the year shown 2023, so all solar in those queues
Entered queues in an earlier year entered in an earlier year
100 98% of the solar capacity in
CAISO’s queue at the end of 2023
was paired with a battery; in the
0 non-ISO West, that number was
2014-23 2014-23 2014-23 2014-23 2014-23 2014-23 2014-23 2014-23 2014-23
West MISO CAISO PJM ERCOT Southeast SPP NYISO ISO-NE
also high, at 81%
(non-ISO) (non-ISO) ❑ Both regions are grappling with
“duck curve” issues due to solar’s
relatively high market share

Note: Not all projects will ultimately be built as many withdraw during the interconnection process. Utility-Scale Solar, 2024 Edition
67
For more details on methods see Berkeley Lab’s Annual Queued Up Report. http://utilityscalesolar.lbl.gov
Most active solar proposes to be online by 2028, but the historical
completion rate for solar projects requesting interconnection is low
Solar Capacity in Queues at 2023 Year-End (GW)
Process phase:
175 Current Status for Requests Submitted 2000-2018
IA Draft/Executed
By capacity of requests:
150 Facility Study
Solar 10% 15% 2% 73%
SIS/Cluster 1
125
Feasibility Study Solar + Battery 7% 47% 4% 43%
100
In Progress/Unknown Operational Active
75 Not Started By number of requests: Suspended Withdrawn

Solar 13% 10% 1% 75%

50
Solar + Battery 10% 37% 3% 49%
25

0 0% 20% 40% 60% 80% 100%

<2023 2023 2024 2025 2026 2027 2028 2029 2030 >2030
Proposed Online Year

• Few solar projects have requested interconnection with a proposed • If historical patterns persist, only ~10% of solar capacity requesting
online date of 2029 or later interconnection will ultimately get built and become operational

• Proposed online dates are included in the developer’s original • Developers withdraw interconnection requests for myriad reasons:
interconnection request and may differ from actual online date ❑ Some reasons are based in the interconnection process, such as high
cost to interconnect and study delays
❑ 132 GW of active solar requests were already past their proposed online
❑ Some reasons arise outside of the interconnection process, such as
date at the end of 2023
failure to secure financing or an offtaker, permitting issues, or insufficient
• 130 GW of solar capacity have an interconnection agreement (either resources to complete all proposed projects
draft or executed) – these projects are the most likely to be completed

Note: Not all projects will ultimately be built as many withdraw during the interconnection process. Utility-Scale Solar, 2024 Edition
68
For more details on methods see Berkeley Lab’s Annual Queued Up Report. http://utilityscalesolar.lbl.gov
Summary

Utility-scale PV continued to lead solar deployment in 2023, with Texas adding the most new capacity. 89% of new projects and 96% of new capacity feature
single-axis tracking.

The capacity-weighted installed cost of solar projects that came online in 2023 fell to $1.43/WAC ($1.08/WDC), down 8% from 2022 and 75% from 2010.
Median prices (perhaps better reflecting typical project costs) fell from $1.61/WAC to 1.33/WAC .

Average capacity factors range from 17% in the least-sunny regions to 31% in the sunniest. Single-axis tracking adds more than five percentage
points to capacity factors in the regions with the strongest solar resource.

The generation-weighted LCOE from utility-scale PV has declined by 80% since 2010 to $46/MWh (without tax credits) or $31/MWh (with tax credits)
in 2023. Levelized PPA prices have kept pace, but prices on newly executed agreements have trended upward the last few years. Since 2021 prices
have typically ranged from $20-45/MWh in CAISO and the non-ISO West to $35-$85/MWh elsewhere, with the highest prices in PJM and MISO.
The market value of solar fell in 2023 to $44/MWh on average, as energy prices returned to more normal levels. The generation costs of newer
projects is less than their wholesale market replacement costs. Rising prices for new PPAs are now higher in some regions than solar’s energy and
capacity market value in 2023. When accounting for climate and health benefits, solar generation delivered nearly $14 billion net-value in 2023.

Interest in hybridization (pairing PV with batteries) continued to set new records in 2023 (37 new greenfield plants, 4.4 GWAC-PV and 15 project retrofits,
0.9 GWAC-PV). Recent (2022-2024) PV+battery hybrid PPAs in the continental US had a capacity-weighted average of $55/MWh-PV.

Across all 7 ISOs and 44 additional utilities, there were 1085 GW of solar in interconnection queues at the end of 2023. More than half of this proposed solar
capacity is paired with battery storage, with the highest concentration of these PV+battery hybrid plants in CAISO (98%) and the non-ISO West (81%).

Utility-Scale Solar, 2024 Edition

69
http://utilityscalesolar.lbl.gov
For more information
Explore this report deck, a written technical brief, an extensive workbook with
all underlying data, and interactive visualizations: http://utilityscalesolar.lbl.gov

For questions or technical assistance requests, contact:

Joachim Seel ([email protected])
Julie Mulvaney Kemp ([email protected])

Read our other solar and wind work at: https://emp.lbl.gov/utility-scale-

renewable-energy-storage

Join our mailing list to receive notice of future publications:

https://emp.lbl.gov/mailing-list

Follow us on X @BerkeleyLabEMP

This material is based upon work supported by the U.S. Department of

Energy’s Office of Energy Efficiency and Renewable Energy (EERE) under
Solar Energy Technologies Office (SETO) Agreement Number 38444 and
Contract No. DE-AC02-05CH11231. TheUtility-Scale
authors are solely
Solar responsible
2024 Edition for
any omissions or errors contained herein.
http://utilityscalesolar.lbl.gov Photo credit: Intersect Power