What is the return on investment for solar panels in the UK?

On our modelling of a median-priced 4 kWp system over 25 years, the internal rate of return ranges from slightly below zero to about 9.4% a year. The spread comes from self-consumption and the export tariff, not from anything about the equipment.

What is a realistic IRR for domestic solar?

For a household using 23% of its generation on the best open untied export rate, our model gives about 6% before any inverter replacement. On the median export rate the same household is nearer 1%. Neither figure is a promise — they are outputs of assumptions we state in full.

Why use IRR rather than payback?

Payback ignores everything after the break-even point and treats a pound in year twenty as identical to a pound today. IRR uses the whole cash flow and expresses it as an annual rate, which is what makes it comparable with other uses of the same money.

Does the model include the inverter being replaced?

As a sensitivity rather than a baseline, because we have no sourceable UK price for replacement. A £1,200 replacement in year twelve takes our central case from about 6% to about 4.9%, which is the honest size of that uncertainty.

What degradation rate do you assume?

0.5% a year, which sits within the range manufacturers warrant. It is an assumption rather than an observation — we have found no independent UK dataset on real degradation, and we would rather label it than bury it.

Is solar a good investment?

We are not financial advisers and will not answer that. What we can do is show the rate of return under stated assumptions so that you can compare it with the alternatives available to you, at your own tax position and risk tolerance.

The return on domestic solar: ROI and IRR, modelled properly

Payback treats twenty-five years of income as if the last pound were worth the same as the first. A rate of return does not. On our modelling the same 4 kWp system returns anywhere from below zero to about 9.4% a year — and the difference is not the panels.

Same panels, same roof, same price — and a return from below zero to 9.4%.

Over 25 years, on a median-priced 4 kWp system generating 3,424 kWh a year:

Case	25-year income	Simple payback	IRR
14% self-consumption, 4.1p export	£5,766	25.9 years	−0.8%
23% self-consumption, 4.1p export	£7,360	20.3 years	1.2%
23% self-consumption, 12p export	£12,235	12.2 years	6.0%
60% self-consumption, 4.1p export	£13,899	10.8 years	7.4%
60% self-consumption, 12p export	£16,430	9.1 years	9.4%

Assumes 0.5% annual degradation, prices held flat, and no inverter replacement. The full model is at the foot of this page.

Why payback is the wrong tool on its own

Simple payback — the figure everyone quotes, including on our own page — has two defects that matter for a twenty-five year asset.

It stops counting at break-even. A system that pays back in twelve years and then runs for another thirteen is not equivalent to one that pays back in twelve and dies. Payback cannot see the difference.

It treats all pounds as equal. A pound in year twenty-four is not worth a pound today, whatever your view of interest rates.

IRR fixes both. It takes the whole cash flow — the negative one at the start, the positive ones each year — and expresses the result as the annual rate that makes them balance. That is a number you can hold against the alternatives.

Read the table again and notice what is constant: the panels, the roof, the location, the price paid.

What changes is how much of the generation is used at home and which export tariff the household is on. Between them they move the return by more than ten percentage points.

This is why we are so insistent about self-consumption and so rude about single-figure payback claims. The equipment is the least variable part of the calculation.

The inverter sensitivity

The baseline above excludes inverter replacement, because we have no source for what one costs. That is an omission with a direction, so here is its size — with the caveat that these three figures are illustrative, not sourced:

Inverter replacement in year 12	IRR, 23% self-consumption at 12p export
None (baseline)	6.0%
£800	5.3%
£1,200	4.9%
£1,600	4.5%

So a replacement costs roughly one to one and a half percentage points of return. It does not change the conclusion in any row; it does move a marginal case.

Reading the negative row honestly

The top row of the answer box is a negative return. A household using 14% of its generation, on the median open untied export rate, does not get its money back over twenty-five years on these assumptions.

We publish that because it is what the arithmetic gives, and because that household exists: a house empty on weekdays, no storage, whichever export tariff came with the installation.

Three things change it, in order of size:

Raising self-consumption — storage, an EV, a heat pump, or somebody at home
Switching export tariff, which is free
Paying less than the median per kW, which is a matter of getting three quotes

None of them requires different technology.

What this model does not price

Rising electricity prices. Holding the import price flat is deliberate. If prices rise, every row improves; the model shows what is arithmetic rather than what is a forecast.

Any effect on house value. We have consistently declined to put a figure on this because we do not have a source we trust. Until we have joined EPC data to sold prices ourselves, we are not going to add a number to a financial model.

Non-financial reasons. Some people want to generate their own electricity. A model has nothing to say about that, and pretending otherwise would be false precision of a different kind.

If you want to compare with keeping the money

That is solar against a savings account, which is a different comparison from the one most people make — and which has a wrinkle in solar’s favour that is rarely mentioned.

Sources

Solar photovoltaic (PV) cost data Department for Energy Security and Net Zero · Accessed 20 August 2026 · OGL v3.0
MGD 003: Solar PV Self-Consumption, issue 2.0 (1 April 2022) MCS · Accessed 17 August 2026 Read locally. MCS restricts reproduction, so we cite clauses and quote sparingly.
Energy price cap unit rates and standing charges Ofgem · Accessed 17 August 2026 · OGL v3.0

Contains public sector information licensed under the Open Government Licence v3.0.

Last verified: 20 August 2026 How we research this Suggest a correction

Was this page helpful?