Solar Electric Power Generation (U.S.) — An Investor's Primer

North American Industry Classification System (NAICS) 2022 code 221114. Covers companies whose main business is running solar power plants that feed the grid. Federal statistics below are the authoritative figures; industry and company data are labeled by source.

---

1. Overview

This industry is the business of owning and operating large solar power plants — mostly ground-mounted photovoltaic (PV) fields of 1 megawatt (MW) and up — that sell electricity into the grid. It has gone from a rounding error to a major piece of the U.S. power system: utility-scale solar produced about 5% of the electricity from large U.S. power plants in 2024 and grew another 34% in 2025 [6][8]. Two things make it distinctive for investors. First, it is capital-heavy and labor-light: you spend most of the money up front to build the plant, there is no fuel to buy, and a big site runs with a handful of technicians. Second, it is not mostly owned by regulated utilities — roughly 83% of capacity is held by independent power producers (IPPs) and infrastructure funds that sell power under long-term contracts [10]. That shapes how you get in. A public-market investor buys shares of the utilities and renewable platforms that own solar (there is no pure-play public solar-generator stock). A private investor — an infrastructure fund, a direct developer, a municipal utility, or a public-private partnership — buys projects, portfolios, development companies, tax credits, or project debt directly.

---

2. What it is and how it's structured

Scope. NAICS 221114 is establishments "primarily engaged in operating solar electric power generation facilities" that deliver power to transmission or distribution systems [1]. In practice the investable core is utility-scale solar (plants ≥1 MW), which is how the U.S. Energy Information Administration (EIA) measures the fleet [5][6].

What it excludes (each is a different NAICS code):

So this industry is not panel manufacturers, not rooftop installers, and not stand-alone batteries — even though batteries are increasingly built alongside solar ("solar-plus-storage") [1].

Who owns it. Ownership is unusually tilted away from regulated utilities. Using 2024 EIA plant data compiled by the American Public Power Association (APPA), the ~130 gigawatts (GW) of solar capacity breaks down as [10]:

Generation splits the same way — about 86% of 2024 solar output came from IPPs, 14% from utilities [11]. But that understates the role of public power and cooperatives, which are major buyers: they contract for far more solar than they own outright (electric cooperatives alone owned or contracted for more than 6.3 GW at the end of 2025) [39]. Government bodies mostly show up as customers, landowners, and permitting authorities, not plant owners.

---

3. How big it is

Two federal programs measure the business, and they measure different things. Both are our ground-truth figures:

Two things stand out. The workforce is tiny relative to the assets — under 9,000 people run the whole employer base — because solar plants need almost no staff once built [2][3]. And the industry is not very concentrated: the top 4 firms earn 43.3% of receipts and the top 50 earn 87%, but the Herfindahl-Hirschman Index (HHI, a standard concentration gauge) is just 615.9, well below the 1,500 line regulators treat as "unconcentrated" [2]. In plain terms: a few large platforms sit atop a long tail of independent developers.

The Census revenue figure understates the industry's real economic weight, and it is worth being clear why. First, government-owned generation is largely excluded from these business counts — though, as Section 2 shows, direct government ownership of solar is under 1% of capacity, so that exclusion is small here. The bigger reasons are structural: the $5.51 billion counts only the receipts of establishments primarily classified as solar generators, so it misses solar owned inside big diversified utilities (classified elsewhere) and the many single-project entities; and it excludes the value of federal tax credits, which are worth billions a year (Section 5). One deep-research estimate put total industry economic activity nearer $12–18 billion a year once energy sales, certificates, capacity payments, and credits are included (author estimate, not a federal statistic — treat as directional) [Opus report].

Physically, the fleet is large and growing fast. At year-end 2024 the U.S. had about 123 GW of utility-scale PV nameplate capacity plus 1.4 GW of solar-thermal capacity [6]. That fleet generated 219.8 terawatt-hours (TWh) in 2024 — about 5.1% of all large-plant U.S. electricity — and output jumped ~34% to roughly 296 TWh in 2025 [7][8]. Counting rooftop systems (outside this industry), all solar is closer to 7% of total U.S. electricity. A structural quirk to remember: solar plants run at a ~23% capacity factor [9] — a 100 MW plant produces like a 23 MW around-the-clock unit — which is why nameplate GW and actual output diverge so much.

---

4. The investable universe

There is no pure public-market play. Every listed company that owns U.S. utility-scale solar also owns something else — a regulated utility, wind, storage, gas plants, or foreign assets — so no share price and no market cap is "solar generation." Public investors buy solar as one ingredient inside a larger business. Below, solar scale comes from company filings; yields are date-stamped (15 July 2026); there is no meaningful solar-only revenue or market-cap line to report.

Not generators (linked suppliers, excluded from this industry): First Solar (FSLR, panels), Nextracker (NXT) and Array Technologies (ARRY, trackers), Shoals (SHLS, wiring), Canadian Solar (CSIQ) [—supply chain, cited in reports].

ETFs: Invesco Solar (TAN, 0.70% expense ratio) and iShares Global Clean Energy (ICLN, 0.39%) — but both are global and manufacturer-heavy, not baskets of U.S. utility-scale owners [40][41].

Major private and other owners (where most of the industry actually lives): Invenergy (largest privately held developer/owner), Brookfield / BlackRock-GIP / KKR / Blackstone / EQT (infrastructure funds), Arevon (>5.3 GW operating), Origis, EDF Renewables, Lightsource bp, Silicon Ranch (Shell-backed), Intersect Power, Pine Gate, RWE (~4.3 GW U.S. solar), plus Enel, EDP and others [21][22][23][28]. On the public-power side: Tennessee Valley Authority (TVA), Los Angeles Department of Water and Power, Salt River Project, municipal utilities, and rural cooperatives — mostly as buyers.

---

5. How the money works

Two business models share this one NAICS code.

A. The competitive / IPP model (the ~83% majority). An independent owner builds a plant and earns money from some mix of:

• a power purchase agreement (PPA) — a 15-to-25-year contract to sell the plant's output at a set price. This is the anchor cash flow that makes the project financeable.
• merchant sales into wholesale markets when output is uncontracted;
• capacity payments (modest, because solar produces little at the evening peak);
• renewable energy certificates (RECs) sold to buyers with clean-energy mandates; and
• federal tax credits — often the single largest value driver.

Because there is no fuel, the economics are almost entirely up-front cost + cost of capital + tax credits. Lawrence Berkeley National Laboratory (LBNL) put the 2023 build cost at $1.43 per AC watt, down ~75% since 2010; unsubsidized levelized cost of energy (LCOE) around $46/MWh, and about $31/MWh after credits [11][12]. Newly signed long-term PPAs averaged about $35/MWh (LBNL, 2023 vintage) [11]; a broader current-market index (LevelTen) ran higher at $56.76/MWh in late 2024, from ~$45 in Texas to ~$80 in the mid-Atlantic [27]. The catch: as more solar is built, it floods the midday market and pushes down the price it can capture — LBNL found solar's average market value fell about 35% to ~$32/MWh in 2024 [12][13]. This "cannibalization" is the core reason batteries get added.

The tax credits — the decisive lever. A project chooses one of two federal credits:

• Investment Tax Credit (ITC, §48E): 30% of project cost when wage/apprenticeship rules are met, plus up to +10 points each for domestic content and for building in an "energy community" — potentially ~50% [15].
• Production Tax Credit (PTC, §45Y): a per-kilowatt-hour credit for 10 years, favored by low-cost, high-output plants [15].

Since 2023 these credits can be sold for cash ("transferability") — a market that reached roughly $30 billion in 2024, priced near 92–95 cents on the dollar [19]. Tax-exempt owners (municipal utilities, cooperatives, tribes) can instead take the credit as a cash refund ("elective/direct pay"), which is what makes public-private solar deals work [18].

B. The regulated-utility model (~16%). When a regulated utility builds and keeps solar, the plant enters its rate base (invested capital), and the utility earns its state-approved return on equity (ROE) on it. Florida Power & Light (FPL) is the model: ~$5.5 billion of 2025 rate-base growth (including solar) at an 11.70% earned ROE, with a 2026–29 settlement setting a 10.95% ROE midpoint [16]. Here, tax-credit benefits generally flow to customers (lower rates), not shareholders — the opposite of the IPP model. This is why regulated names (DUK, SO, D) are lower-risk, steadier-return ways to own the buildout, while IPPs and funds keep the credit upside and the volatility.

---

6. What drives demand

• Electricity demand is rising again after ~15 flat years — led by data centers (about 4.4% of U.S. power in 2023, potentially 6.7–12% by 2028), plus manufacturing and electrification [32]. Solar can be built faster than nuclear, gas, or new transmission, so it captures much of the near-term buildout — but data centers need round-the-clock power, so solar increasingly competes paired with storage.
• Cost leadership: utility solar's ~75% cost decline since 2010 makes it the cheapest bulk energy in most regions before credits [11].
• Federal tax credits (through the coming cliff — Section 9).
• State clean-energy mandates (renewable portfolio standards) that create REC demand.
• Corporate procurement — hyperscalers and Fortune 500 buyers signing PPAs.
• Coal and old-gas retirements needing replacement energy.

Planned 2026 additions tell the story: developers reported 43.4 GW of new utility-scale solar for 2026 — 51% of all planned U.S. generating capacity (versus 30.8 GW actually added in 2024 and 27.2 GW in 2025) [14][35].

---

7. Regulation

• Federal Energy Regulatory Commission (FERC): oversees wholesale power sales and, critically, grid interconnection. Its Order No. 2023 replaced slow one-by-one studies with "cluster" studies and stricter readiness rules to attack the queue backlog [20]. Order No. 1920 pushes long-term regional transmission planning, which over time reduces the congestion and curtailment that hurt solar revenue.
• State public utility commissions (PUCs): the gatekeepers for the regulated model — they approve which solar goes into rate base and set the allowed ROE.
• U.S. Treasury / Internal Revenue Service: administer the ITC/PTC, transferability, direct pay, and the domestic-content and "prohibited foreign entity" sourcing rules — arguably the most economically consequential regulator for competitive solar [15][18].
• U.S. Environmental Protection Agency (EPA): operating solar has no smokestack, so EPA is not its main regulator; construction still triggers stormwater and wetlands permits, and EPA rules on fossil plants indirectly help solar's relative position [41].
• Trade agencies (Commerce / USTR): tariffs raise panel costs — Section 301 duties on Chinese wafers, polysilicon, and cells rose to 50% effective January 1, 2025, on top of antidumping duties on Southeast Asian imports [32][33].
• Nuclear Regulatory Commission (NRC): governs nuclear plants only — not applicable to solar, noted for completeness.

---

8. Competitive dynamics and consolidation

Development is fragmented (hundreds of developers originate projects), but ownership consolidates into a smaller set of well-capitalized IPPs, yield vehicles, and infrastructure funds that can handle tax-credit complexity and hold 30-year assets. Capital is the moat: because returns hinge on cost of capital and credit monetization, large balance sheets win. The pattern shows up in deal flow — Duke Energy sold its commercial-renewables arm to Brookfield for ~$2.8 billion, and AES is being taken private by a Global Infrastructure Partners / EQT consortium at ~$33.4 billion enterprise value [37][38]. The binding competitive constraint is increasingly grid access, not demand: LBNL's queue study counted ~1,400 GW of generation (about 956 GW of it solar) waiting to connect at year-end 2024, with only ~13% of past requests ever reaching operation and a median wait over four years [13][37]. A mature, studied, permitted interconnection position is now worth more than the panels themselves.

---

9. Risks

1. The federal tax-credit cliff (dominant near-term risk). Under the 2025 budget law (Public Law 119-21, the "One Big Beautiful Bill Act"), wind and solar projects generally must begin construction by July 4, 2026 or be in service by December 31, 2027 to claim the ITC/PTC [17][26]. This pulls a rush of investment into 2026–27 and risks an "air pocket" after.
2. Interconnection and transmission delays — multi-year queues, high upgrade costs, low completion rates [37].
3. Price cannibalization and curtailment — high midday solar depresses the price solar earns; California curtailed 3.4 TWh of wind and solar in 2024 (93% of it solar) [13][14].
4. Trade and supply chain — tariffs and sourcing rules raise panel costs and can disqualify credits [32][33].
5. Interest rates — these are long-life, capital-heavy assets with no fuel cost to offset higher financing expense.
6. Merchant "tail" risk — much of a project's value can sit in uncertain power prices after the PPA expires.
7. Counterparty risk — a fixed PPA is only as good as the buyer's credit.
8. Land, permitting, and local opposition to large ground-mount sites.

---

10. How to invest and the outlook

Public-market route. There is no pure play, so you pick a lens: regulated-utility solar (NEE's FPL, plus DUK, SO, D) for lower-risk, rate-base-driven returns and steadier dividends; contracted "yield" vehicles (Clearway at ~5.4%, Brookfield Renewable at ~4.8%) for high, contracted-cash-flow yields with modest growth; or IPP/platform growth (NextEra's overall story) [16][17][25][34][35][36]. Judge yield vehicles on cash available for distribution and remaining contract length, not just the dividend; judge regulated names on allowed ROE and rate-base growth. Suppliers (First Solar, Nextracker) and ETFs (TAN, ICLN) give indirect, non-generator exposure [40][41].

Private-market route. Ways in, from safest to most speculative: buy operating project equity (contracted, infrastructure-like cash flows); provide tax equity or buy transferable credits for a fixed, lower-risk return without owning assets [19]; lend project or holdco debt; take platform/development equity in a developer (higher risk, higher return); or structure a public-private partnership where a municipal utility, cooperative, or tribe uses direct pay to capture the 30%+ credit as cash [18]. The key underwriting question is the same everywhere: how far is the project along the ladder from a speculative land position to a contracted, interconnected, tax-qualified operating plant?

Outlook. Near term (2025–27), expect a construction rush to beat the tax-credit deadline, keeping installs at record levels — solar is over half of all planned 2026 capacity additions [35]. After 2027, growth likely moderates and turns selective: with credits curtailed, economics lean more on falling equipment costs, rising power prices, data-center demand, and — increasingly — storage to firm output and capture evening value. The demand backdrop is extraordinarily strong, but the constraints have shifted from demand to grid access, supply chains, and tax policy. For investors, the split is the whole point: regulated-utility solar is the lower-volatility way to ride the buildout; IPP and fund exposure offers more upside with materially more policy, interconnection, and price risk. The single most important variable to track is the tax-credit cliff.

---

Sources

1. U.S. Census Bureau, "2022 NAICS 221114 — Solar Electric Power Generation." https://www.census.gov/naics/?details=221114&year=2022
2. U.S. Census Bureau, 2022 Economic Census (EC2200BASIC) — firms, receipts, concentration ratios (CR4 43.3%, CR8 59.3%, CR20 77.3%, CR50 87%), HHI 615.9. https://data.census.gov/table/ECNBASIC2022.EC2200BASIC
3. U.S. Census Bureau, County Business Patterns 2023 — 779 establishments, 8,867 employees, $1.22B annual payroll (NAICS 221114). https://data.census.gov/profile/221114
4. U.S. Energy Information Administration, "Utility-scale U.S. solar electricity generation continues to grow," Today in Energy. https://www.eia.gov/todayinenergy/detail.php?id=63324
5. U.S. EIA, Electric Power Annual, Table 4.3 — existing capacity by source (year-end 2024: 122,907 MW PV; 1,405 MW solar thermal). https://www.eia.gov/electricity/annual/html/epa_04_03.html
6. U.S. EIA, Electric Power Annual, Table 3.1.B — net renewable generation (2024 utility-scale solar 219.8 TWh; ~5.1% share). https://www.eia.gov/electricity/annual/table.php?t=epa_03_01_b.html
7. U.S. EIA, "Utility-scale solar generation grew 34% in 2025" (≈296 TWh). https://www.eia.gov/todayinenergy/detail.php?id=67367
8. U.S. EIA, Electric Power Annual, Table 4.8.B — capacity factors (2024 PV 23.2%). https://www.eia.gov/electricity/annual/html/epa_04_08_b.html
9. American Public Power Association, "2026 Public Power Statistical Report" (2024 EIA-860 ownership: non-utility 83.3%, IOU 16.0%, coop 0.4%, public power 0.3%, federal <0.1%). https://www.publicpower.org/system/files/documents/2026_Public-Power-Statistical-Report.pdf
10. Lawrence Berkeley National Laboratory, "Utility-Scale Solar, 2024 Edition" ($1.43/Wac; ~24% capacity factor; ~$35/MWh PPA; LCOE ~$46/$31; ~75% cost decline since 2010; market value ~$45/MWh 2023). https://emp.lbl.gov/publications/utility-scale-solar-2024-edition
11. LBNL, "Utility-Scale Solar, 2025 Edition" (solar market value fell ~35% to ~$32/MWh in 2024). https://emp.lbl.gov/publications/utility-scale-solar
12. U.S. EIA, Electric Power Annual, Table 1.1 — net generation by producer type (2024 solar: IPP 188.4 TWh / ~86%, utility 30.3 TWh / ~14%). https://www.eia.gov/electricity/annual/table.php?t=epa_01_01.html
13. U.S. EIA, "Solar curtailments increase in California" (3.4 TWh curtailed 2024; solar 93%). https://www.eia.gov/todayinenergy/detail.php?id=65364
14. Internal Revenue Service, "Clean Electricity Investment Credit (§48E)" and "Clean Electricity Production Credit (§45Y)." https://www.irs.gov/credits-deductions/clean-electricity-investment-credit; https://www.irs.gov/credits-deductions/clean-electricity-production-credit
15. NextEra Energy, 2025 Form 10-K (NEER 12,794 MW gross / 10,504 MW net solar; FPL earned ROE 11.70%, 2026–29 settlement 10.95% midpoint). https://www.sec.gov/Archives/edgar/data/37634/000075330826000015/nee-20251231.htm
16. Clearway Energy, 2025 Form 10-K (12.9 GW portfolio, ~12-yr contract life). https://www.sec.gov/Archives/edgar/data/1567683/000162828026010952/cwen-20251231.htm
17. IRS, "Elective pay and transferability" (§6417 direct pay; §6418 transfer). https://www.irs.gov/credits-deductions/elective-pay-and-transferability
18. Crux, "2024 Transferable Tax Credit Market Intelligence Report" (~$30B of transfers; ITC ~92.5¢, PTC ~95¢). https://www.cruxclimate.com/insights/2024-transferable-tax-credit-market-key-takeaways
19. Federal Energy Regulatory Commission, "Explainer on the Interconnection Final Rule" (Order No. 2023). https://www.ferc.gov/explainer-interconnection-final-rule
20. Arevon Energy, "Arevon Surpasses $10 Billion in Operating Assets" (>5.3 GW operating). https://arevonenergy.com/news/releases/arevon-surpasses-10-billion-in-operating-assets-expanding-its-american-energy-portfolio/
21. Origis Energy, "Projects and Impact." https://origisenergy.com/our-projects-impact/
22. EDF Renewables North America, project disclosure. https://www.edf-re.com/
23. U.S. Congress, Public Law 119-21 ("One Big Beautiful Bill Act"; construction by July 4, 2026 / in service by Dec 31, 2027). https://www.congress.gov/119/plaws/publ21/PLAW-119publ21.pdf
24. AES Corporation, 2025 Form 10-K (10,961 MW operating U.S. clean energy). https://www.sec.gov/Archives/edgar/data/874761/000087476126000063/aes-20251231.htm
25. Brookfield Renewable Partners, 2025 Form 20-F (~14 GW U.S. wind/solar/storage). https://www.sec.gov/Archives/edgar/data/1533232/000153323226000011/bep-20251231.htm
26. LevelTen Energy, "Q4 2024 PPA Price Index" (North American solar ~$56.76/MWh; ERCOT ~$45 to PJM ~$80). https://www.leveltenenergy.com/post/levelten-energy-q4ppi
27. RWE, "U.S. operating renewable and storage capacity reaches 10 GW" (~4.3 GW solar). https://www.rwe.com/presse/rwe-clean-energy/
28. U.S. Department of Energy / LBNL, data-center electricity report (4.4% of U.S. power in 2023; 6.7–12% by 2028). https://www.energy.gov/articles/doe-releases-new-report-evaluating-increase-electricity-demand-data-centers
29. U.S. EIA, "Developers plan a record 86 GW of U.S. generating additions in 2026" (solar 43.4 GW / 51%; 2024 adds 30.8 GW; 2025 adds 27.2 GW). https://www.eia.gov/todayinenergy/detail.php?id=67205
30. LBNL, "Queued Up: 2025 Edition" (~1,400 GW generation + 890 GW storage in queues; solar ~956 GW; ~13% built; median >4 yr). https://emp.lbl.gov/publications/queued-2025-edition-characteristics
31. U.S. EPA, "Stormwater discharges from construction activities." https://www.epa.gov/npdes/stormwater-discharges-construction-activities
32. Office of the U.S. Trade Representative, Section 301 tariff increases on solar wafers, polysilicon, and cells (50%, effective Jan 1, 2025). https://ustr.gov/about-us/policy-offices/press-office/press-releases/2024/december/
33. U.S. Department of Commerce, solar circumvention / antidumping determinations (Southeast Asia). https://www.commerce.gov/news/press-releases/2023/08/
34. NextEra Energy, dividend information (annualized $2.4928); 46. share price $89.09 on July 15, 2026 → 2.80% yield. https://www.investor.nexteraenergy.com/stock-information/dividend-history
35. Clearway Energy, Q1 2026 dividend ($0.4676/qtr); 48. price $34.46 on July 15, 2026 → 5.43% yield. https://investor.clearwayenergy.com/
36. Brookfield Renewable, distributions ($0.392/qtr); 50. price $32.52 on July 15, 2026 → 4.82% yield. https://bep.brookfield.com/bep/stock-information/distributions
37. AES Corporation, GIP/EQT-led take-private (~$15/share, ~$33.4B enterprise value). https://www.aes.com/
38. Duke Energy, "Duke Energy to sell utility-scale commercial renewables business to Brookfield for $2.8 billion." https://news.duke-energy.com/releases/
39. National Rural Electric Cooperative Association, "Co-op Renewable Growth Update" (>6.3 GW owned or contracted, year-end 2025). https://www.cooperative.com/
40. Invesco Solar ETF (TAN, 0.70% expense ratio). https://www.invesco.com/us/en/financial-products/etfs/invesco-solar-etf.html
41. iShares Global Clean Energy ETF (ICLN, 0.39% expense ratio). https://www.ishares.com/us/products/239738/ishares-global-clean-energy-etf

Note on figures: NAICS 221114 business statistics (firms, revenue, establishments, employment, payroll, concentration) are our authoritative ingested federal figures (2022 Economic Census and 2023 County Business Patterns). Physical capacity, generation, ownership, and market data are drawn from EIA, LBNL, APPA, and company filings as cited. The ~$12–18B "total economic activity" figure is a deep-research author estimate, not a federal statistic.