Australia needs to talk about carbon pricing

With the release this week of the Productivity Commission’s report on cheaper, cleaner energy and net zero, Treasurer Jim Chalmers has been handed some extra ideas – and some dilemmas to navigate.

The PC seems to have limited itself to things the federal government can act on without the states, so some of the recommendations (for example, on planning reform and adapting to climate change) look incomplete.

Nevertheless, it’s heartening that the PC starts off with a call for “comprehensive and consistent incentives to reduce emissions”. Consistency is critical for any industry leader working out which of their assets continues to have value in a net-zero economy. Decision timelines are long, capital requirements are large, and flip-flopping policy can’t be tolerated.

The next good thing is a call for a carbon policy in the electricity sector to guide investment in the industry in the 2030s and beyond. This is well overdue. Governments have been deliberately tilting the playing field in favour of renewables for decades, which has built up an industry capable of supplying 40 per cent of electricity. It’s now time to allow renewables to compete with gas and coal on the terms that really matter: carbon. And the way to do this is with a carbon price – although the PC has been silent on the possible settings.

There are five separate ideas to reduce emissions from heavy vehicles: changes to road-user charges, shifting liability for emissions onto fuel wholesalers, increasing fuel excise, incentives for low-carbon fuels, and incentives for low-emissions vehicles. This will be tempting ground for another skirmish in the climate wars: watch out for spurious claims in the coming weeks about the effects on food prices, and demands that things be ruled in or out.

The reality is that this is an area riddled with inconsistencies and inequities in who pays what for fuel and for carbon, who pays for roads, and how road funding is distributed. Carbon is only one piece of the puzzle, and decisions about it need to be made alongside other reforms. So, here’s hoping government ministers have the courage to leave options on the table.

“The cheap way to change 5 million cooktops is to start today and work through the process over a decade.”

Where the PC report disappoints is in its adherence to the marginal cost curve approach to technology deployment to guide policy towards the lowest-cost options. That thinking was fashionable in the 2010s. But in the years since, huge progress has been made in deploying technologies to cut emissions – whether that’s the solar on your roof, the wind turbine on your farm, or the electric vehicle in your warehouse or garage.

What we’ve learnt from this experience is that marginal cost curves are no more than a point-in-time guess about what’s cheapest – and a guess that often turns out to be wrong. Marginal cost curves from 15 years ago didn’t have electric vehicles or batteries on them – yet these two technologies are creating large shifts in how we travel and how we manage the electricity system.

We’ve learnt that we have to start the expensive things when they are expensive so that they are cheap later. This happens for two reasons: logistics and learning rates.

Models for the costs of emission reductions use marginal cost curves to deploy solutions overnight. If the next cheapest solution in line is to switch all gas cooktops to induction ones, the model will move 5 million households from one to the other in an instant. But in the physical world, doing it overnight wouldn’t be cheap at all. The cheap way to change 5 million cooktops is to start today and work through the process over a decade.

It’s similar in heavy industry. Assets don’t get replaced with new, less polluting ones when the model says it’s time. They get replaced at the end of their lives. The replacement decision has a multi-year runway. And if new technology isn’t proven, available, and economic at the decision point, the emissions reduction won’t be achieved. Conversely, replacing assets before end-of-life requires writing them down, which has balance sheet implications.

In both cases, learning rates change the relative attractiveness of different options. Some technologies, such as solar, were high up on the cost curve a decade ago. But solar had an incredible learning rate: Chinese innovation in manufacturing coupled with Australian learning in deployment has led to some of the lowest-cost rooftop solar in the world. Yes, we may have “overpaid” for some solar in the early days. But as a result, we have a lot more solar now, at a much cheaper price.

Setting Australia up for a flourishing and productive net-zero future is going to require the technology, the capital, and the market signals to change most of the energy-consuming assets in the economy in the next 25 years. We absolutely need comprehensive and consistent incentives, as the PC says. However, these incentives have to fit with the realities of physical assets – because that’s where productivity happens.