Executive Summary
Across the 32 metros our solar payback calculator models in Arizona, Texas, Florida, North Carolina and Georgia, a 2026 cash purchase pays back in a median 12.2 years at each state’s average residential rate (EIA, July 2026), and in 10.7 to 13.9 years from the fastest metro to the slowest. The federal residential solar credit (Section 25D) expired for homeowner-purchased systems installed after December 31, 2025, so a 2026 cash purchase no longer subtracts 30%; with that 30% off, the same systems would pay back in a median 8.5 years, which is roughly where a solar lease or PPA that captures the commercial Section 48E credit (for systems placed in service by December 31, 2027) lands. Florida has the fastest state median (11.8 years; Miami is the fastest metro at 10.7) and Texas the slowest (13.3 years; Houston 13.7). California’s payback is withdrawn: under NEM 3.0 exported power is credited far below the retail rate, and the payback calculator values every kilowatt-hour a system offsets at the retail rate and sets no export credit or self-consumption share, so it cannot price NEM 3.0.
Paybacks come from our solar payback calculator’s metro models: each metro’s typical system size and 2026 cash cost from the Dreamy Leads Research data layer, and its calibrated first-year bill offset, priced at EIA’s July 2026 state-average residential rates (Electric Power Monthly Table 5.6.A, released 2026-09-24). Incentive and net-metering rules come from DSIRE and the utilities’ own tariffs (Table 2).
Key Findings
- Across the 32 metros the payback calculator models in Arizona, Texas, Florida, North Carolina and Georgia, a 2026 cash purchase pays back in 10.7 years (Miami) to 13.9 years (Tallahassee), with a median of 12.2 years, at each state’s average residential rate (EIA, July 2026).
- By state median: Florida 11.8, Arizona 12.1, Georgia and North Carolina 12.5 each and Texas 13.3 years.
- California’s payback is withdrawn. It has the highest rate of the six states (33.61¢/kWh), but under NEM 3.0 exports are credited far below that rate, and the payback calculator values every kilowatt-hour a system offsets at the retail rate and sets no export credit or self-consumption share, so it cannot price NEM 3.0. Its installed cost does not depend on NEM: the calculator’s 4 California metros run 8.6–9.0 kW at $24,940–$26,100 cash ($2.90 per watt, the same price per watt as the other five states).
- The federal §25D residential solar tax credit expired for systems installed after December 31, 2025, so these are 2026 cash paybacks with no federal credit. With 30% off the cost, the same model gives 7.5–9.7 years (median 8.5), roughly what a lease or PPA capturing the §48E commercial credit (for systems placed in service by December 31, 2027) can offer.
How Solar Payback Is Calculated
The payback period is the number of years it takes for cumulative electricity bill savings to equal the net out-of-pocket system cost. The net cost reflects applicable state incentives and utility rebates (the federal §25D credit expired for 2026 purchases; a lease or PPA may still capture 30% via §48E, for systems placed in service by December 31, 2027). The formula is:
Payback (years) = Net System Cost / Annual Electricity Savings
In the payback calculator, first-year savings are the metro’s calibrated bill offset per installed kW × system size × the electricity rate. The bill offset is set so the tool reproduces the data layer’s modelled payback for that metro; it is not panel output, and it sets no export credit or self-consumption share: every kilowatt-hour it counts is valued at the retail rate. All three inputs differ significantly across states and utility territories.
A critical assumption: this study holds today’s rates flat, with no escalation. If rates instead rose 3.5% a year (an illustrative figure, not a forecast), the same systems would pay back 1.5–2.4 years sooner (the median metro in about 10.3 years instead of 12.2), and 25-year savings would be about 56% higher. If rates fall, as Florida’s average did over the past year (EIA), payback runs longer.
6-State Payback Comparison
| State | Avg. residential rate, state average (EIA, July 2026) | Metros modelled | Typical system (kW) | Installed cost (2026 cash) | First-year savings | Payback range | Median payback |
|---|---|---|---|---|---|---|---|
| Florida | 15.03¢ | 16 | 8.8–10.4 | $25,520–$30,160 | $1,836–$2,767 | 10.7–13.9 yrs | 11.8 yrs |
| Arizona | 15.38¢ | 4 | 8.6–9.2 | $24,940–$26,680 | $1,904–$2,280 | 11.7–13.1 yrs | 12.1 yrs |
| Georgia | 16.27¢ | 4 | 8.2–8.4 | $23,780–$24,360 | $1,815–$2,047 | 11.9–13.1 yrs | 12.5 yrs |
| North Carolina | 15.16¢ | 4 | 8.2–8.4 | $23,780–$24,360 | $1,829–$1,997 | 12.2–13.0 yrs | 12.5 yrs |
| Texas | 15.88¢ | 4 | 9.2–9.6 | $26,680–$27,840 | $1,947–$2,263 | 12.3–13.7 yrs | 13.3 yrs |
| California | 33.61¢ | 4 | 8.6–9.0 | $24,940–$26,100 | Withdrawn (NEM 3.0) | Withdrawn (NEM 3.0) | — |
| All 32 modelled metros | — | 32 | 8.2–10.4 | $23,780–$30,160 | $1,815–$2,767 | 10.7–13.9 yrs | 12.2 yrs |
Sources: rates, EIA Electric Power Monthly Table 5.6.A (July 2026, released 2026-09-24; state-average residential price); system sizes, installed costs and bill offsets, the solar payback calculator’s metro data. Payback = installed cost ÷ first-year savings (bill offset per kW × kW × rate), the calculator’s own formula; no rate escalation or degradation.
Arizona’s 25% state tax credit (up to $1,000, Table 2) is not in the calculator’s cost. Taken off the same systems, it trims Arizona’s metros to 11.3–12.6 years (from 11.7–13.1). Texas rates vary by retail electric provider in the ERCOT market; the Texas row uses the state average.
Model assumption: the Arizona, Georgia and North Carolina figures value every kilowatt-hour the system produces at the state-average retail rate, as if all of it offset power bought from the grid. Those utilities credit exported power at far less: APS at 5.554¢/kWh (Rate Rider RCP, from September 1, 2026) and TEP at 5.13¢/kWh (Rider R-14 RCP, from October 1, 2025); Georgia Power at 7.2188¢/kWh (the 2026 solar avoided-cost rate of 3.2188¢ plus the 4¢ renewable generation adder; Georgia Power rooftop-solar FAQ; Rate RNR-11, for 2026); Duke Energy Carolinas at 4.53¢/kWh (Rider RSC, effective July 1, 2026), so real savings run lower, and payback longer, for a home that exports part of its output. Duke Energy Progress’s export credit was not verified.
Why This Study Does Not Price a California Payback
California has the highest residential electricity rate of the six states, averaging 33.61¢/kWh (EIA, July 2026, state average) — more than double Texas’s 15.88¢. Intuitively, a high rate should produce the fastest payback. NEM 3.0 (Net Energy Metering 3.0), which took effect for new solar customers in California in April 2023, breaks that link: excess solar power exported to the grid earns only $0.05–$0.08/kWh (export credit) — a 75% reduction from the retail-rate credit previously available under NEM 2.0 — so a California system is worth what its owner uses on site, and much less for what it exports. Pricing that needs a self-consumption share, and the payback calculator values every kilowatt-hour a system offsets at the retail rate and sets no export credit or self-consumption share, so it cannot price NEM 3.0. Earlier versions of this study printed California paybacks from that model; they are withdrawn (2026-09-30), and California is left out of every range, median and ranking on this page. Installed cost does not depend on NEM, so it stays in Table 1: the calculator’s 4 California metros run 8.6–9.0 kW at $24,940–$26,100 cash ($2.90 per watt, the same price per watt as the other five states). For how export credits work, see our NEM 3.0 solar economics tracker.
State Incentive Comparison
| State | State Tax Credit | Sales Tax Exemption | Property Tax Exemption | Net Metering Policy | Utility Rebates Available |
|---|---|---|---|---|---|
| Arizona | 25% (up to $1,000) | Yes — full exemption | Yes — 100% | Net billing — exports credited below retail (APS Resource Comparison Proxy; SRP export price plans) | Limited |
| North Carolina | None | Yes — full exemption | Yes — 80% exemption | Duke Energy: new customers on Residential Solar Choice or Bridge rider, monthly exports credited at avoided cost (retail-credit rider closed 2023-09-30) | Duke Energy: up to $500 |
| Florida | None | Yes — full exemption | Yes — 100% | Investor-owned utilities: retail kWh credit, year-end excess paid at as-available rate (Fla. Admin. Code R. 25-6.065); municipal utilities and co-ops set their own terms | FPL: limited |
| Texas | None | Yes — full exemption | Yes — 100% | No statewide net metering — buyback set by the retail plan or local utility (PUCT) | Austin Energy: up to $2,500 |
| Georgia | None | Yes — full exemption | Yes — 100% | Georgia Power: instantaneous netting, exports at solar avoided cost (monthly netting closed to new customers) | Georgia Power: $450 |
| California | None | Yes — full exemption | Yes — 100% | NEM 3.0 — $0.05–$0.08/kWh export | LADWP, PG&E: varies |
Source: DSIRE database (April 2026); EIA state incentive summaries; utility tariff schedules. Net metering column (checked 2026-09-16): APS solar transition (RCP); SRP solar price plans; Public Utility Commission of Texas; Georgia Power RNR-11 tariff.
System Size Sensitivity Analysis
In the calculator’s model, payback does not change with system size: installed cost scales at the metro’s cost per watt and first-year savings scale with every installed kW, so a 4 kW and a 10 kW system in the same metro pay back in the same number of years (Miami 10.7, Tallahassee 13.9). Real quotes are not that tidy: fixed costs such as permitting and interconnection weigh more on a small system, and a larger system exports more of its output, which Arizona’s net billing, Georgia Power and Duke Energy credit below the retail rate (Table 2), so an oversized system can take longer than shown. An earlier version of this page printed a Florida-vs-Arizona size table from the model Table 1 used to rest on; it could not be reproduced and has been withdrawn. To test a size against your own quote, use the payback calculator.
The Battery Storage Question
Adding a battery storage system (Powerwall 3 at $9,200 installed, or Brightbox at similar cost) extends payback by 3–5 years on its own economics — the battery itself does not produce power and must be justified by backup value or time-of-use rate arbitrage. In Florida, where time-of-use rates are limited, battery ROI is primarily a resilience calculation. In Arizona, where APS has aggressive peak-demand charges from 3–8pm, a battery that shifts solar consumption to peak hours can improve the combined system’s economics, but this study does not model time-of-use rates, so it puts no number on that gain.
Related Reading
Data & tools used in this analysis
Frequently Asked Questions
How long does it take for solar panels to pay off in the South?
For a 2026 cash purchase, the 32 metros our payback calculator models across Arizona, Texas, Florida, North Carolina and Georgia pay back in 10.7 to 13.9 years, with a median of 12.2 years, at each state's average residential rate (EIA, July 2026). State medians run from 11.8 years in Florida to 13.3 in Texas. The model values all output at the retail rate; utilities in Arizona, Georgia and North Carolina credit exports below it (APS 5.554 cents per kWh; TEP 5.13 cents per kWh; Georgia Power 7.2188 cents per kWh; Duke Energy Carolinas 4.53 cents per kWh), so paybacks there run longer for homes that export. California's payback is withdrawn: NEM 3.0 credits exports far below retail, and the model sets no self-consumption share. The federal Section 25D residential credit expired for systems installed after December 31, 2025; with 30% off the cost the same systems would pay back in a median 8.5 years, and a lease or PPA may still capture 30% through the Section 48E commercial credit (for systems placed in service by December 31, 2027).
Which Southern state has the best solar return on investment?
On payback, Florida leads the five states we model with a median of 11.8 years across its 16 modelled metros (Miami is the fastest at 10.7), followed by Arizona (12.1), Georgia and North Carolina (12.5 each) and Texas (13.3), at each state's average residential rate (EIA, July 2026). Florida metros also post the highest first-year savings in the study (up to $2,767). California is not ranked: its payback is withdrawn under NEM 3.0.
What factors most influence solar payback period?
The five dominant factors are: (1) electricity rate — at the same cost and output a higher rate pays back faster, but only for power used on site or credited at retail: California has the highest rate of the six, yet NEM 3.0 credits exports far below it, which is why this study does not price a California payback; (2) system cost per watt — our calculator prices every modelled metro at $2.90 per watt installed (cash, before incentives), and real quotes vary around that by installer and equipment; (3) net metering policy — retail-rate credit vs. avoided-cost buy-back rates can materially lengthen payback; (4) sun hours — varies from 4.5/day (NC mountains) to 6.5/day (AZ desert); (5) available incentives — state/utility rebates typically reduce net system cost (the federal §25D credit expired for 2026 purchases; a lease or PPA may still capture 30% via §48E, for systems placed in service by December 31, 2027).
Methodology
- Model: the solar payback calculator’s 32 metro models in Arizona, Texas, Florida, North Carolina and Georgia. Payback = installed cash cost ÷ first-year savings, where first-year savings = the metro’s calibrated bill offset per installed kW × system size × rate — the calculator’s own formula. The bill offset is a calibration (set so the tool reproduces the data layer’s modelled payback for the metro), not panel output. No rate escalation, degradation or financing.
- Utility rates: EIA Electric Power Monthly Table 5.6.A, July 2026 (released 2026-09-24, eia.gov): each state’s average residential price, the same value for every metro in that state — a state average, not a utility tariff. Texas retail prices vary by retail electric provider. Time-of-use rate impacts not modeled.
- System size and cost: the data layer’s typical system size per metro and its 2026 cash cost before any state credit.
- Federal credit: not applied — the residential §25D credit expired for homeowner-purchased systems installed after Dec 31, 2025, so a 2026 cash purchase carries the full cost basis with no 30% federal credit. Arizona’s state credit (25%, max $1,000) is shown separately under Table 1; leased/PPA systems may still capture 30% via the §48E commercial credit (for systems placed in service by December 31, 2027).
- Net metering: not priced. The model values every kilowatt-hour a system offsets at the state-average retail rate and sets no export credit or self-consumption share. No utility or regulator we checked publishes a self-consumption share for these territories, so the figures are not re-priced; each state’s export treatment is disclosed instead. Model assumption: the Arizona, Georgia and North Carolina figures value every kilowatt-hour the system produces at the state-average retail rate, as if all of it offset power bought from the grid. Those utilities credit exported power at far less: APS at 5.554¢/kWh (Rate Rider RCP, from September 1, 2026) and TEP at 5.13¢/kWh (Rider R-14 RCP, from October 1, 2025); Georgia Power at 7.2188¢/kWh (the 2026 solar avoided-cost rate of 3.2188¢ plus the 4¢ renewable generation adder; Georgia Power rooftop-solar FAQ; Rate RNR-11, for 2026); Duke Energy Carolinas at 4.53¢/kWh (Rider RSC, effective July 1, 2026), so real savings run lower, and payback longer, for a home that exports part of its output. Duke Energy Progress’s export credit was not verified. Florida pays year-end surplus at the as-available rate (Table 2). California is withdrawn because NEM 3.0 makes the gap too large to leave unpriced.
- System degradation: not applied. At 0.5%/yr, output in year 13 — the year the median metro pays back — is about 5.8% below year one, so real paybacks run slightly longer than shown.
- Correction (2026-09-26): earlier versions of this table used rates it described as February 2026 EIA figures (they were not) and an annual-savings column about 1.4 to 1.5 times production times rate, so its paybacks could not be reproduced from its stated method. Table 1, the size analysis, the summary, the findings and the FAQ were rebuilt from the calculator on 2026-09-26.
- Correction (2026-09-30): California paybacks, and the six-state ranges and medians that included them, are withdrawn: the model sets no export credit or self-consumption share, so it cannot price NEM 3.0. California keeps its rate, system size and cash cost in Table 1. Model-assumption disclosures with each utility’s own export rate were added for Arizona, Georgia and North Carolina.
