A battery rebate that pays for paperwork will end up buying paperwork

Almost a hundred separate people have now rung a hotline to say some version of the same thing: the paperwork says a battery went on the wall, and the paperwork is wrong.

That is an allegation, not a finding, and it should be read that way. But the shape of the allegations is worth more than any single case, because it points straight at where a multi-billion-dollar subsidy is soft. The claims are not that batteries were overpriced, or undersized, or the wrong model. They are that installs were mocked up, that householders’ signatures were forged, and that unlicensed tradesmen did cash jobs. Every one of those is an attack on a document. None of them is an attack on a battery.

The scheme is buying paperwork

A rebate of this kind is not usually handed to a household in cash. It is claimed against an install — a record saying a unit of a given size went in at a given address on a given date, signed off by the customer and certified by someone holding the right licence — and taken off the invoice at the point of sale. The thing the government actually receives, holds and audits is a certified claim. The battery is what the claim is about. It is not what the government sees.

Once that is stated plainly, the alleged behaviour stops being mysterious and becomes arithmetic. If the cheapest route to public money is a convincing form rather than an installed appliance, some operators will take the cheaper route, and the ones who do will be precisely the ones who were never going to be deterred by a warning letter. Designing a subsidy is not moral philosophy. It is deciding what you will pay for, and then paying for that thing rather than a description of it.

The alarming number is the hundred, not the fraud

Someone always tries it on. Every rebate ever written has attracted operators who treat the eligibility rules as the product specification. What should worry anyone responsible for this scheme is not that cheating is alleged but where the allegations came from.

They arrived through a whistleblower channel: from competitors losing work to people undercutting them, from employees, from subcontractors, from customers who noticed something. They did not arrive from the scheme’s routine examination of its own claims. Whistleblower reports are a lagging and self-selecting signal. People only report what they happen to stand next to, and only when the personal cost of reporting is lower than the irritation of staying quiet. If internal controls were doing the work, tip-offs would be confirming things the administrator already suspected. Instead they look like discovery.

Forging a document has never been cheaper

The cost of fabricating a plausible paper trail has collapsed. Elsewhere in this week’s news, shoppers are being warned about an online store selling handmade glassware attributed to an artist who does not exist — a complete commercial identity, invented cheaply, convincing enough to take money. A scanned signature and a job sheet are a much easier forgery than an entire fake artisan.

Verification regimes built in an era when producing convincing documents was slow and expensive have not adjusted. A signature is now close to the least expensive thing in the whole transaction to fake, which makes it the worst possible thing to hang a payment on.

The safety problem underneath the fiscal one

Strip out the money and something worse remains. The licence check is not decoration. It is the mechanism by which a household is assured that stored electricity was connected to their switchboard by someone competent. If unlicensed workers are being papered over with a licensed person’s certification, the state is not merely losing revenue — it is issuing a guarantee it has not verified, to homeowners who will find out whether it was true some years from now.

Money can be clawed back. A bad install cannot be un-done by an audit, and the households concerned have no way of knowing they are in the affected group.

The case for leaving it alone

The honest counter-argument is speed. These schemes exist to get equipment onto roofs and walls quickly, and every pre-payment inspection adds delay and cost to thousands of legitimate jobs to catch a small minority of dishonest ones. There is no denominator in the reporting — we are not told how many installs the tip-offs sit against, and it would be careless to pretend a hundred allegations proves a systemic collapse. It may prove a functioning reporting culture. A scheme that receives complaints is at least a scheme people know how to complain about.

But verification does not have to be universal to work. It has to be random, real, and carry consequences, so that no operator can price fraud as a certainty rather than a gamble. And the cost argument is weaker than it looks: a state government has just committed thirty million dollars to advertising its energy transition. Checking that the hardware the public paid for physically exists is not the expensive part of this policy.

What would settle it

One table would resolve the whole dispute: claims paid, claims inspected, inspections that found nothing there, dollars recovered, installers suspended. Its absence is currently doing a lot of talking.