Will my solar savings be higher in year ten than year one in South Africa?

Answer first: Year-ten savings beat year-one savings in nearly every South African home with a maintained system. Panels lose a fraction of a percent a year; tariffs climb by high single digits. The kWh fade is small, the rand per kWh grows, and the product of the two rises. The risks to that curve are maintenance, not physics. Related: panel degradation · tariff increases and payback · payback.

Two curves: kWh down slowly, rand per kWh up quickly

Panel warranties describe the kWh curve. Example curves on SA-sold modules guarantee at least 95% of rated output after five years, then no more than 0.4% loss a year, to at least 87% at year 25; other OEM sheets sold locally cap year-one loss around 2–3% and then about 0.45–0.55% a year (degradation desk). Read the exact curve on the datasheet for your module.

Tariffs describe the rand curve. NERSA approved 8.76% for Eskom direct customers from April 2026 and 9.01% for municipal supply from July 2026 (NERSA media statement (March 2026)); Eskom’s proposed 2027/28 increase is 8.83% direct and 8.84% municipal, now in public consultation (EWN (4 September 2026)). Beyond 2027/28 there is no published figure — so model the published years and treat later years as your own assumption.

Worked example: the same 6 kWp array, year one and year ten

Take a 6 kWp Highveld array that makes about 9 000 kWh in its first good year, and a household that self-uses 70% of it — 6 300 kWh. At R4.00/kWh — a typical top-block rate before the July 2026 increase — that is R25 200 of avoided purchases in year one.

By year ten, at a 0.5% a year fade after year one, output is about 95–96% of the year-one figure — roughly 8 600 kWh, or about 6 020 kWh self-used. Apply only the two published increases (9.01% then 8.84%) and hold tariffs flat after that: the kWh is worth about R4.75 and the saving is about R28 600. That is the floor case — it assumes no increases at all from 2028 onwards. Any further increase lifts year-ten savings above that.

The arithmetic is ours; the percentages are NERSA’s and the degradation curve is from module warranty sheets. Replace R4.00 with the top-block rate on your own account.

What actually erodes savings over ten years

An inverter fault left unfixed

The inverter is the component most likely to need replacement inside a panel’s lifetime. A dead inverter makes zero kWh, so a week without it costs more than a decade of panel fade. Keep the warranty registration, know who honours it locally, and budget for one replacement — how long inverters last · replacement cost · warranty claims.

A battery worked outside its rating

Batteries earn by shifting midday solar to the evening. Their lifespan is set by cycles at a rated depth of discharge. BlueNova’s SBB MKII manual, for example, rates ≥8 000 cycles at 80% DoD and recommends 80% as the maximum depth (BlueNova SBB MKII user manual v1.2); Solar MD, Pylontech, Freedom Won and other locally supported packs publish their own cycle and throughput warranties. Keep the inverter’s SoC floor at the manufacturer’s recommendation and a 48 V DC LiFePO4 pack — the oldest, most proven home hybrid architecture — lasts well past ten years — battery lifespan · cycle life and DoD.

Dust, new shade and changed habits

Highveld dust and bird droppings shave yield between rains; a neighbour’s new storey or a growing tree shades a string; children leave home and daytime load drops. Each of those reduces self-used kWh more than degradation does. Watch the monitoring app every quarter and compare like-for-like months year on year — dust and soiling · tree shade · monitoring numbers.

Fixed charges and export credits

Metro fixed charges rise every July and solar does not reduce them, so your total bill falls less than your energy saving rises. Export credits are set separately from import tariffs; they have not tracked import increases one-for-one, which is another reason to keep self-use high as the years pass — fixed charges · export credit vs self-use.

Grow the system as the household changes

Year ten is a good time to revisit the design. An EV, a heat pump or a pool changes when and how much you use. A hybrid with spare MPPT capacity and a modular battery adds panels and usable kWh without starting over — expand later · add panels to an existing inverter. Buy from a supplier that will still carry spares and support in a decade; established SA wholesalers such as Herholdts, ARB, Voltex, ACDC, DBS Energy and SONAH Group hold stock and RMA paths for the major brands — wholesaler peers.

Year-by-year habits that protect the curve

• Quarterly: compare monthly yield with the same month last year.
• Yearly: rinse panels after the dry season if yield drops; check isolator and inverter alarms.
• Every July: update your tariff in the payback spreadsheet; the saving should rise.
• Every five years: ask a registered installer to inspect connectors, earthing and the DB.

Why the early years matter most for payback

Payback is decided in the first five to seven years, when panels are at their strongest and every rand saved repays the capital. That is exactly when degradation is smallest: a module at 95%+ of rated output is, for practical purposes, the module you bought. Tariff increases in those years add straight to the saving. The later years — when the system is paid off — are profit, and the rising tariff makes that profit larger each year.

So the comparison that matters is not year one against year 25. It is the cumulative rand saved against the installed price. Track it in a simple spreadsheet: one row a year, kWh self-used from the inverter app, tariff from the July municipal schedule. The line should climb every year. If it flattens, look for an inverter alarm, a dirty array or new shade before blaming the panels.

Selling before year ten

A documented system with yield history is easier to sell with the house. Keep the monitoring export, the CoC and the warranty paperwork together — a buyer sees ten years of rand savings, not a roof of unknown panels — selling a house with solar.

How the way you pay changes year ten

By year ten the savings are the same whichever way you paid; the finance cost is not. On a R70 000 system at prime of 10.75% (SolarGuide worked example, 21 October 2026):

• Cash: R70 000 paid once; years two to ten are pure saving.
• Bond re-advance repaid in seven years: about R1 189 a month and ≈R99 900 in total; years eight to ten are instalment-free.
• Bond over 20 years: about R711 a month and ≈R170 600 in total; in year ten you are still paying for a system halfway through its battery warranty.
• Rent-to-own at 6.5% escalation: from R1 449 a month to about R2 114 by year seven, ≈R155 900 in total.

Year ten is where the cash and short-bond households pull ahead: their system is paid off, tariffs have compounded for a decade, and every kWh from the roof is free. Choose the route that clears the debt before the battery warranty runs out. Full comparison: cash vs bond vs rent-to-own.

Bottom line

Degradation is the smallest number in your solar maths. Tariff increases are the biggest. Keep the inverter running, the battery inside its rating and the panels clean, and your year-ten saving lands above your year-one saving.

Quick questions

Do solar panels lose efficiency every year?

Slowly. SA-sold warranty curves commonly guarantee ≥95% after five years and around 0.4–0.55% per year after that.

Will tariff increases outpace panel degradation?

Yes. A fade of under 1% a year is small against tariff increases of 8–9% a year in 2026/27 and the proposed 2027/28 figure.

What reduces savings over ten years?

An inverter failure left unrepaired, a battery worked outside its rated depth of discharge, dirty panels, new shade, and changes in household load.

Should I budget for an inverter replacement?

Yes. Plan for one inverter replacement inside the panel lifetime and keep the warranty paperwork.