Solar Yield

Degradation

Checked 2026-08-09
Reading the estimate

Published by a company that designs and builds solar plants. What that means.

Modules produce less each year. The rate sounds small enough to ignore and compounds over the life of the asset into something that changes the investment case.

Teams reviewing yield assumptions can keep the engineering model separate from the record of review work; this reference is one option for documenting time and handoffs.

Reviewed August 9, 2026.

The rate

0.5 to 1% per year is the range in general use, with the lower end typical of modern modules under warranty and the upper end a conservative planning assumption.

For independent background on photovoltaic performance, resource data and current industry practice, consult PVsyst.

At 0.5% per year, year 25 output is about 88% of year one.

At 1% per year, it is about 78%.

That difference — ten points of lifetime output — is the whole of what the assumption choice is worth, and a proposal that does not state which rate it used has left a material variable undeclared.

What actually degrades

Light-induced degradation in the first hours and days, which is a step rather than a slope and is normally accounted for in the nameplate rating.

Gradual cell degradation, the long-run slope.

Encapsulant discolouration and delamination, environmental and variable.

Potential-induced degradation, which is a design and configuration matter rather than an inevitability.

And connection and junction-box aging, which appears as loss and is not module degradation at all.

Only some of these are covered by a module performance warranty, which matters when something declines faster than expected.

Warranted against expected

Two different curves.

A module performance warranty guarantees a percentage of nameplate at specified years, typically with a first-year allowance and a linear slope after it. It is a floor.

The degradation rate used in a yield projection is an expectation, usually better than the warranted floor.

A plant tracking the warranty line rather than the expectation is underperforming and not in breach, which is an uncomfortable position and a common one. Reading performance ratio over time is how it becomes visible, and about 1.5% per year against a 0.5% warranty is the diagnostic gap.

How it should appear in a projection

Applied year by year, compounding.

Not as a single average, which understates early years and overstates late ones — usually harmlessly, occasionally not.

And never omitted. A twenty-five-year savings total that applies year-one output to every year overstates by roughly ten to twenty per cent depending on the rate, and it is one of the commonest defects in a proposal.

What to check

"What degradation rate does this projection use?"

"Is it applied annually or averaged?"

"What does the module warranty guarantee at year 25, and how does that compare?"

And for an operating plant: what is the measured rate? Performance ratio plotted annually gives it, and it is the only way to know whether the assumption held.

The part that is not the modules

Worth separating, because owners attribute everything to panels.

Inverters do not degrade — they fail. Replacement is a capital event, not a slope.

Soiling is recoverable. A declining trend that reverses after cleaning was never degradation.

And shading grows. A tree that was a two per cent loss at commissioning is a different number in year twelve, and it is not degradation either.

Three distinct effects that look identical in a production chart, and separating them is what an irradiance-corrected number is for.

The short version