Faster renewable adoption did not raise household electricity bills more — if anything, the opposite, 2015-2024

Summary

No. Looking at all 50 states from 2015 to 2024, the states that added the most renewable generation share did not see bigger increases in household electricity prices. If anything, the relationship runs the other way: states with the fastest renewable-share growth saw average residential price increases of about 22% versus 28% for the states with the slowest (or negative) renewable growth, and a state-level regression finds a weak negative slope (-0.34 percentage points of price growth for every +1 percentage point of renewable-share growth, p≈0.05, n=50). This matches the consensus in the published literature: utility-scale renewables are generally associated with flat-to-lower rates, and the recent run-up in U.S. electricity bills is driven mainly by other factors — grid/infrastructure spending, natural gas price swings, extreme-weather/wildfire costs, and load growth from data centers — not by the pace of renewable buildout. California is the one state that fits the popular narrative (large renewable growth AND the largest price increase in the country), but it is a genuine outlier, not the source of the aggregate finding: dropping California from the regression makes the negative relationship stronger, not weaker.

Step 1 — What the published research already says

Before touching the data, a literature scan turned up a reasonably deep and fairly consistent body of work:

• MIT CEEPR working paper (Knittel & Argosino, 2026), using 25 years (1998-2023) of U.S. electricity/utility data: state Renewable Portfolio Standard mandates have "virtually zero impact" on residential prices, and utility-scale wind and solar are "robustly linked to lower residential prices" because of lower transmission/distribution/maintenance costs passed through to consumers. The counterintuitive finding is that rooftop solar is significantly correlated with higher prices for non-solar households, because it forces a one-way distribution grid to handle two-way power flows, raising O&M costs, and because rate structures often let solar owners avoid paying for grid upkeep (a cost shifted onto other ratepayers).
• Hannah Ritchie / Our World in Data (Feb 2025), plotting 2023 state-level solar+wind generation share against residential price: no significant cross-sectional correlation. States like Iowa, South Dakota, Kansas, and New Mexico combine high wind/solar shares with cheap electricity; states like Connecticut and New Hampshire have expensive power with almost no wind/solar. California stands out because of its large solar share specifically, but Nevada has a similar solar share at roughly half California’s price — undercutting a simple "solar = expensive" story. Looking at within-state trends over time, she also notes national average prices actually fell slightly even as solar and wind grew, while Texas, Iowa, North and South Dakota show renewables growth with flat or falling prices.
• World Resources Institute / Energy Innovation pieces on the 2024-2025 price run-up attribute it mainly to utility capital spending (utilities sought a record ~$30.5B in rate increases, much of it grid investment and disaster-related costs), not renewables compliance costs (estimated at roughly 4% of bills on average).
• A contrasting, frequently-cited counter-argument (Mackinac Center/ZeroHedge-style pieces) points to a strong correlation (~0.93 cited) between wind+solar additions and price increases across Michigan and its four neighboring states since 2019 — a real but much narrower, five-state, six-year comparison that does not generalize once the full 50-state panel is examined (see Step 3).

Net: the literature leans toward "no positive relationship, and possibly a negative one for utility-scale renewables," with important caveats about rooftop solar and rate design. This corpus's own 50-state data (below) is used to test that directly over the specific 2015-2024 window the question asks about, rather than taking the literature's word for it.

Step 2 — Data and method

Renewable share: energy.state_energy_mix (EIA electric power operational data rollups), renewable_pct = renewable generation ÷ total generation, by state, for 2015 and 2024 (most recent full year loaded).

Household electricity prices: energy.eia_electricity_prices, residential sector (sector_code='RES'), avg_price_cents_kwh, by state, for 2015 and 2024. Regional/Census-division rollup rows (e.g. ENC, PACC, NEW) and the national 'US' row were excluded, restricting to the 50 states plus DC; DC was then dropped from the regression as a non-state outlier per this corpus's own diagnostic warning.

Measures: for each state, the change in renewable-generation share in percentage points (2024 minus 2015) and the percent change in the average residential price (nominal cents/kWh, 2015→2024). Nominal dollars are used because the comparison is across states in the same two years — a common national inflation shift over 2015-2024 (nominal U.S. average price rose 30.3%, from 12.65 to 16.48 ¢/kWh) affects every state equally and does not change the cross-state ranking; it would only matter for a real-terms trend within one state over time.

This is a bivariate, cross-sectional (not causal) design: one predictor, no controls for population growth, wildfire/storm exposure, generation-fuel mix (e.g. gas dependence), market structure (regulated vs. deregulated), or grid-investment cycles — all factors the literature above identifies as bigger price drivers than renewables share per se. The finding below should be read as descriptive: renewable growth was not accompanied by systematically larger price increases, not that renewables caused prices to fall.

Step 3 — Result

Across the 50 states (DC excluded), the Pearson correlation between 2015-2024 renewable-share change (percentage points) and 2015-2024 residential price change (percent) is r = -0.28 (p = 0.052, n = 50) — a weak, borderline-significant relationship, and its sign is negative: states that added renewable share faster tended to see slightly smaller price increases, not larger ones. An OLS regression of price % change on renewable-share change gives a slope of -0.34 percentage points of price growth per +1 percentage point of renewable-share growth (SE 0.17, p = 0.052).

Splitting the 50 states into the 15 with the fastest renewable-share growth (Iowa +43.8pp, New Mexico +34.8pp, Kansas +26.4pp, Nebraska +26.1pp, Nevada +21.4pp, Colorado, Wyoming, California, Utah, South Dakota, North Dakota, Oklahoma, Texas, Montana, Illinois) versus the 15 slowest/negative (New Jersey, Louisiana, South Carolina, Connecticut, New Hampshire, Arkansas, Vermont, Alabama, Pennsylvania, Tennessee, Idaho, Alaska, Maine, Washington, Oregon — several of which actually saw their renewable share fall, largely due to hydropower variability in the Pacific Northwest): the fast group averaged +22.4% residential price growth versus +28.3% for the slow group. A Welch t-test on this 15-vs-15 split is not itself statistically significant (p = 0.30) — consistent with a real but modest effect that a 50-state, single-predictor design has limited power to pin down precisely.

Robustness / California check: California is the standout outlier in the scatter — a large renewable-share gain (+18.1pp) paired with by far the largest price increase in the country (+88.2%, from 16.99 to 31.97 ¢/kWh), matching the popular narrative that renewables push up prices in the state most associated with the policy. A leave-one-out sensitivity check (refitting the regression with each state dropped in turn) shows this is not what is producing the aggregate negative correlation: dropping California from the regression makes the negative slope stronger, not weaker (-0.46, SE 0.12, p < 0.001), because California's own extreme combination of high renewable growth and high price growth is actually pulling the fitted line toward zero/positive, not driving a spurious negative result. No state drop flips the sign of the coefficient; the coefficient ranges from -0.28 to -0.46 across all 50 leave-one-out refits. Several individual states' omission does move the p-value across the 0.05 threshold (the sensitivity tool's own summary marks this specification 'not robust' by that stricter criterion), which is exactly what a borderline p=0.05 result on n=50 with one predictor implies — the direction is stable, the precision is not.

Step 4 — Why California is not a counterexample to the aggregate finding, and what actually seems to explain the outliers

California illustrates the literature's exact caveat: it is a large-solar, not primarily large-wind, state, and per the MIT CEEPR findings rooftop solar (which California has adopted heavily) is linked to higher costs via grid two-way-flow strain and cost-shifting rate design, unlike utility-scale wind/solar. California's price was already high (over 20¢/kWh) before its renewables buildout accelerated, and its recent price surge is widely attributed in the literature search above to wildfire liability and grid-hardening costs layered on top of the renewables transition, not renewables alone. Maine, Washington, and Oregon sit at the other extreme (renewable share fell, largely due to hydropower output variability, yet prices still rose 28-56%) — a reminder that hydro variability, not a shift away from renewables policy, explains their falling share, and that price growth in the Pacific Northwest and New England has other drivers (e.g. Maine and New England natural-gas-constrained winters) entirely independent of renewables share.

What This Report Does Not Answer

• Causal attribution ('did renewables cause the price rise'): Answered as a descriptive/correlational relationship only; the single-predictor design cannot support a causal claim, and the report says so explicitly rather than overclaiming