Australia's Wealthiest Postcodes Have the Least Rooftop Solar

Rooftop solar has a reputation as a purchase for the comfortable: a system costs a few thousand dollars up front, you need a roof you own, and the household that installs one shifts part of its power bill onto everyone else. From there it is a short step to the idea that solar, and the public money behind it, flows to wealthier Australians.

The map of where the panels actually went does not support that idea. Rank every residential postcode by the ABS index of socio-economic disadvantage, split them into ten equal groups, and count the small-scale solar installations the Clean Energy Regulator has recorded since 2001. The most advantaged tenth of the country has the fewest — the lowest rooftop-solar uptake of any decile, on every way of counting it.

The wealthiest tenth is last#

The most advantaged decile records 132.7 installations for every 1,000 residents. The peak, 218.8, sits in decile 2 — the second-most-disadvantaged group of postcodes. That is a difference of about two-fifths. Whichever denominator you choose, the ordering holds: per 100 occupied dwellings, the wealthiest decile manages 36.3 against a 58.8 peak; the wealthiest decile is last on that measure too.

The relationship is not a clean line, and the finding is reported as what it is. Uptake rises from the most-disadvantaged decile, peaks in the second and third, then falls away across the more advantaged half — an inverted-U rather than a gradient. The most-disadvantaged decile is not the highest; it is second-lowest. What is firm is the top end: sort the deciles by any of these rates and the most advantaged postcodes come last every time.

At the level of individual postcodes the association is weak and noisy — a population-weighted correlation between disadvantage score and installations per resident of −0.08, a rank correlation of −0.21, across 2,321 postcodes. Rooftop-solar uptake in any one postcode is driven by roof type, climate, block size and a dozen other things that swamp the disadvantage signal. The pattern in the chart is real but it is a feature of the aggregate — a shortfall concentrated at the wealthy end — not a tidy slope you could read off a single suburb.

It is not only apartments#

The obvious objection is housing. The wealthiest postcodes include the apartment towers of inner Sydney and Melbourne, and a flat on the fourth floor cannot host a rooftop array. Housing type does explain much of the raw gap. Across the deciles, the share of dwellings that are flats or apartments climbs from under 5% in the mortgage-belt middle to 26% in the most advantaged decile.

So the fair test strips housing out of it: count installations only against owner-occupied detached houses — the homes that can plainly take a panel and whose owner can authorise it. The reversal survives.

Among owner-occupied detached houses, the most advantaged decile records 71.3 installations per 100 homes, against a peak of 99.1 in decile 3 — still the lowest of any group, about 28% below the top. Uptake declines steadily across the whole advantaged half of the country. Housing type accounts for part of why the wealthiest areas lag; it does not account for all of it. Among the very homes best placed to install, the more advantaged the postcode, the less rooftop solar it has taken up.

The bottom of the distribution has its own brake, and it is a different one. The most-disadvantaged decile has the highest rate of renting — about 38% of its dwellings — and a renter rarely installs solar on a landlord’s roof. That is why the poorest decile sits second-lowest rather than at the top: affordability and tenure hold it back, even as the mortgage-belt deciles just above it lead the country.

The suburbs that took up rooftop solar fastest were not the wealthy ones. They were the detached-house, owner-occupied, more-disadvantaged mortgage belt.

Not a recent phenomenon#

One explanation would rescue the intuition: perhaps solar started as a rich household’s toy in the feed-in-tariff years and only later spread down the income scale, and the cumulative count is hiding an early wealthy skew. Splitting the installations by when they happened rules that out. The systems installed by the end of 2014 already show the same shape — decile 2 highest, decile 10 lowest — as the systems installed in 2020–2025.

The wealthiest postcodes were installing the least in 2012, and they are installing the least now. There is no early rich-skew in the postcode record that later reversed; the mortgage belt led from the start.

Who pays for it#

None of this would matter much if solar paid for itself. It does not, quite. Every eligible small-scale installation creates small-scale technology certificates under the Commonwealth’s renewable energy target. Electricity retailers are legally required to buy those certificates and surrender them, and — as the Australian Energy Market Commission sets out in its price-trend work — they recover the cost from customers through their bills. The certificate has a fixed ceiling price of $40, and the scheme runs at scale: the regulator set the 2026 small-scale target at about 24 million certificates. That is on the order of a billion dollars a year, collected across the whole customer base. The AEMC names the small-scale scheme as one of the leading upward pressures on residential prices in recent years.

The subsidy, in other words, is spread evenly across everyone who buys grid electricity — and it is claimed unevenly. A renter in a disadvantaged suburb and an apartment-dweller in a wealthy one both pay the levy on their bill; neither is likely to have a panel to show for it. The households that do — concentrated, the data show, in owner-occupied detached homes across the more-disadvantaged and middle deciles — get both the certificates and the bill savings that follow.

That is a genuine distributional edge, and it is worth stating precisely, because it cuts against the loose version of the same complaint. The people missing out are not “the poor subsidising the rich”. They are renters and apartment-dwellers at both ends of the scale subsidising owner-occupier detached houses in the middle. The wealthiest postcodes are not harvesting this subsidy; by the installation record they are the group harvesting the least of it.

What could explain this instead#

Area is not household. SEIFA ranks postcodes, not people. A disadvantaged postcode still contains a spread of households, and the ones installing solar there may be its better-off owner-occupiers. Every claim here is about the geography of where panels landed — which is exactly what determines who the subsidy’s footprint falls on — not about the income of the individual who signed the contract. The postcode is the unit the data support, and the article keeps to it.

Housing stock, not attitudes. Detached, owner-occupied homes dominate the middle deciles and thin out at both ends. That mechanism is real and is shown above; it explains a large part of the gap. It does not explain the residual decline across owner-occupied detached houses themselves, which is why the finding is not reducible to “wealthy areas have more flats”.

Roofs, sun and space. Wealthy inner-urban postcodes have smaller blocks, more shading, heritage overlays and strata approvals; well-off households also move house more and may value the payback less. These would each depress uptake at the top without any income effect, and the data here cannot separate them. The claim is deliberately narrow: it describes where solar is and is not, not why each household chose.

Counting installations, not homes. The Clean Energy Regulator counts certificate-creating install events, which include replacements and upgrades, so these rates measure installation intensity, not the share of homes with a working panel. Replacements are unlikely to fall so consistently on the disadvantaged-to-middle deciles as to manufacture the pattern, but the measure is an intensity index and the cross-decile comparison, not the absolute rate, is the reliable part.

The rooftop-solar rollout is one of the largest transfers of generation onto private property Australia has attempted, and it was paid for out of everyone’s power bill. The installation record shows it did not follow the wealth map. It followed the roofline — detached, owned, and disproportionately in the suburbs that the same statistics call disadvantaged, while the wealthiest postcodes, and the renters at the bottom, largely watched it go by.