Photovoltaic module prices have fallen roughly 90% since 2010 — one of the steepest sustained cost declines of any manufactured technology in history.
That part is well known. The second-order effect isn't.
What didn't get cheaper
Wright's Law applies to manufactured goods with scaling production. It does not apply to a technician driving to a site.
Hardware fell 90%. Labour, permitting, interconnection, and service calls didn't. Several have risen with wage inflation.
So the composition of a solar project has shifted. Hardware, once dominant, is now frequently the minority cost in a distributed installation.
The inversion
When hardware was expensive, choosing cheaper components saved a large absolute amount. The downside risk existed but was small relative to that saving.
Now the same percentage discount saves much less — while the downside risk hasn't moved at all. Replacing a failed component in year twelve costs whatever labour costs in year twelve.
The saving collapsed. The exposure didn't.
As hardware prices fall, the cost of buying badly rises relative to purchase price.
This is the inverse of how we're trained to evaluate purchases. In most categories, falling prices lower the stakes of the decision. Here they raise them.
Where it shows up
Degradation divergence. Two panels at identical nameplate capacity can produce meaningfully different output by year fifteen. Invisible in year one — which is exactly why it never surfaces during a sale. Performance warranty length is the usable proxy: a manufacturer guaranteeing output at year 25 is making a claim a 10-year guarantee isn't.
Inverter replacement. Panels are glass and silicon with no moving parts. Inverters are power electronics under continuous thermal cycling, every daylight hour for decades. They're the hardest-working component by a wide margin and will typically need replacing at least once. Yet most proposals model 25 years against a single inverter line item.
Interconnection delay. Physical installation is commoditised — days of work. Net metering approval varies enormously by jurisdiction and is routinely the longest pole in the timeline. A completed system awaiting export approval earns a fraction of its value on a fully paid-for roof.
The generalisable point
When a technology's production cost collapses but its service cost doesn't, quality selection becomes more important, not less. The hardware saving shrinks toward zero while service exposure stays constant.
Battery storage is on the same trajectory and will hit the same inversion.
Buyers still optimising for upfront price are optimising a variable that stopped mattering.
I work on solar system design at [SolarBazaar](https://solarbazaar.io/). The point above is provider-agnostic — worth running against any quote, including ours
Top comments (1)
A strong point solar buyers often focus on the falling hardware price while overlooking the long-term costs of installation, maintenance, replacement, and system performance. The inverter and degradation points are especially important because they directly affect the real ROI.