Where commercial electricity tariffs are heading in 2026

Commercial electricity tariffs in the UK are commonly discussed as a single number — "we pay 22p" — but that number is the sum of four components, each on its own price-discovery cycle. Wholesale, network charging, policy costs, and supplier margin move on different timeframes and respond to different forces. For a 25-year solar finance model the wholesale-component trajectory is what matters most, and the trajectory is more durable than the headline tariff suggests.

The four components

The components of a typical 22p/kWh commercial tariff in 2026 break down approximately as: wholesale electricity ~9.5p, network charges (transmission and distribution) ~4.5p, policy costs (renewables, social, capacity market) ~4.5p, and supplier margin and cost-to-serve ~3.5p. The split varies by tariff structure, supply point, and contract length, but those are representative weights for a mid-sized commercial site on a non-half-hourly-settled passthrough tariff.

Wholesale: post-energy-crisis normalisation

UK wholesale electricity peaked at over £400/MWh in 2022 and is now (Q2 2026) trading around £85–£105/MWh forward 12–24 months. The decline reflects the resolution of the gas-supply shock, increased deployment of generation capacity, and the price-cap-driven softening of demand. Forward curves price wholesale roughly flat in real terms through 2030. Major investment banks have recently moved their long-run price assumption to £75–£85/MWh real, with risk skewed slightly to the downside given the pace of wind and solar build-out.

For commercial solar economics, the wholesale path is the dominant variable. Solar self-consumption avoids the wholesale component plus the variable network and policy components — together about 17–19p of a 22p tariff. Even if wholesale falls to £70/MWh long-run, the avoided-cost-per-kWh from solar self-consumption stays at 14–16p in real terms because the non-wholesale components stick.

Network charging: structurally upward

Network charges are the part of the tariff that solar economics genuinely benefits from over the long run. Transmission and distribution network operators are funded through RIIO settlement processes that explicitly allocate charging to recoverable allowed revenue — and as electricity demand rises with electrification, the per-unit charges spread over more kWh, but the total network bill is rising fast. Ofgem's RIIO-T3 and RIIO-ED2 frameworks both project material network investment. The likeliest 2030 outcome is network charges as a percentage of total tariff stays roughly flat at 20–22%, but absolute pence rises with the wholesale component.

Policy costs: peak likely passed

Policy costs (Renewables Obligation, FiT, Contracts for Difference, Capacity Market levy, social levies) peaked relative to wholesale in 2020–2021. As legacy CfD and RO contracts are increasingly funded by their own revenue rather than levy, the policy-cost component is forecast to decline as a share of total tariff into the 2030s. From a solar self-consumption perspective this is a slight headwind — every penny removed from policy-cost is a penny removed from avoided cost.

Supplier margin: stable

Commercial supplier margins compressed during the energy crisis and have only partially recovered. Margins on multi-site corporate contracts remain tight (1–3p/kWh range) as well-resourced procurement teams force competitive tendering. We see no structural reason for margin to expand — if anything, AI-driven analytics are tightening procurement further.

The implication for solar finance modelling

The unhelpful but accurate answer to "what tariff should I model?" is: not the headline number. For a 25-year solar finance model the variable that matters is the avoided-cost-per-kWh of solar generation, which equals (wholesale + variable network + policy) — typically 14–16p in 2026 against a 22p headline tariff. We model real-terms tariff inflation at 1–2% per year, with sensitivities at 0% and 3%. The downside scenario (0% real tariff inflation through 2050) still produces 11–14% IRRs on capital-purchased commercial solar at 2026 capex.

Where we see clients getting the model wrong: assuming all of the 22p is avoided. It isn't — the fixed standing charge component (£200–£800/month depending on supply) is unaffected by self-consumption. Subtract the standing-charge-per-kWh-at-current-volume from the headline tariff before computing avoided cost. On a typical commercial site that's a 1–3p/kWh adjustment that some installer models miss.

What we'd do differently in 2026 modelling

Three changes from 2024 modelling: (1) lower the long-run wholesale assumption to £75–£85/MWh real, with sensitivity to £65 downside; (2) drop the policy-cost growth assumption to zero — peak likely passed; (3) increase the network-cost growth assumption modestly given RIIO-T3 visibility. The net effect on a typical 250kWp project IRR: marginally negative versus 2024 assumptions (a 1–2 percentage point reduction), but still firmly inside the "do it" range for any well-positioned profitable trading company.