Hydrogen vs electrification: Why ENA’s Gas Vision is a house of cards

The Gas Vision 2050 update study recently released by Energy Networks Australia and several gas industry organisations states (page 9) “Net-zero emissions can be reached with hydrogen at half the cost of electrification.”

This remarkable claim is based on a study by Frontier Economics for gas industry associations, available from Energy Networks Australia’s website, along with the Gas Vision report. As with many energy papers, this headline statement is poorly worded, based on many assumptions, and over-simplifies the outcomes of the Frontier study.

Some key detail from the Frontier report provides important insights and illustrates the fragility of this claim.

Over half of Frontier’s net ‘cost of electrification’ is incurred in Western Australia, and a quarter in Victoria (Figure 22, shown below). Frontier point out that much of this ‘cost of electrification’ is due to high estimated costs of electrifying industrial heat processes.

The other major factor is the estimated cost of increased electricity generation and supply network capacity to replace the winter gas peak [created mainly by inefficient appliances heating inefficient buildings].

The Frontier study has some important limitations.

Many debatable assumptions are made about details of industrial heat demand, efficiencies of gas and electricity use, demand profiles and trends, and other issues. No sensitivity studies are provided, so we don’t know the size of the ‘error bands’.

For example, innovation in industrial heat, including lower process temperatures, dematerialisation and technology change is accelerating from a low base. Recently, David Leitch referred to analysis that showed use of renewable electric Mechanical Vapour Recompression in alumina refining could halve operating costs compared with gas, though capital costs are still high.

Boston Metal’s metal oxide electrolysis technology shows promise as an electric alternative to hydrogen for production of steel and other metals – it works like aluminium smelting. Frontier rely on a 2015 study that shows temperatures of industrial heat supplied, not the, often lower, temperatures actually required for processes.

Peak demand for building heating is assumed to remain high – despite building thermal efficiency improving through building regulation and other policies. Electric heat pump efficiency is assumed to achieve a Coefficient of Performance (COP) of 2.5 – 2.5 units of heat per unit of electricity consumed: present Minimum Energy Performance Standards require a COP of 3 and best products achieve a COP of over 5, already twice as efficient as Frontier’s assumed value for 2050.

The study considers the costs of supplying energy provided by gas and electricity at zero emissions in 2050. On page 7, the report notes “Our analysis does not address transition from 2020 to 2050, but deals only with outcomes in 2050”.

So the relative outcomes regarding Australia’s cumulative carbon budget are not evaluated. This is what really drives global heating, which depends concentrations of greenhouse gases in the atmosphere, and hence on how quickly emissions are reduced.

In a recent workshop presentation on building electrification I pointed out that factors affecting the cost and rate of electrification for home heating would include many factors, such as:

• Relative size of summer compared with winter electricity peaks (Victorian 2019 summer peak was around 2000MW higher than winter);
• Mix of solar, wind, hydro and storage – seasonal and daily electricity availability and cost;
• Building thermal efficiency;
• Improvement in efficiency of both electric and gas appliances and equipment in all sectors which could potentially free up significant electricity supply capacity;
• Role of EV charging and export to grid;
• Demand response, management, storage, targeting of electricity efficiency measures;
• Demand-side gas options – eg fuel cell + heat pump or conversion of traditional gas heaters.

The Frontier study compares the incremental cost of electrification and two other scenarios relative to the unstated cost of a ‘baseline’ scenario, which is not costless – indeed it is likely to involve investment of hundreds of billions of dollars and emission of billions of tonnes of carbon emissions over the 30 years to 2050. So the estimated additional $27.3 billion cost of the electrification scenario relative to the baseline scenario, and $14.3 billion relative to the zero carbon fuels scenario, may be very modest in the overall energy investment picture.

Without knowing the cost of the baseline scenario, statements cannot be made about the total cost of any of the scenarios. The study can only tell us the relative incremental costs of the scenarios on top of the baseline scenario cost. So Energy Networks Australia’s claim regarding the high cost of electrification is not supported by Frontier’s work.

Frontier comment (p.6) that:

“There is significant uncertainty about technological developments and costs over the period to 2050. This means that the actual costs of the scenarios that we have examined will change over time, and new alternative scenarios will emerge over time. Policies to achieve net zero emissions that are broad-based, rather than focused solely on promoting the electrification of all stationary energy end-use, will enable energy sector participants and their customers to respond flexibly to these technology and cost changes to lower costs.”

I agree that there is much uncertainty about future technology and costs, and that broad-based policies are needed. But these should apply across the board, not just to electrification. They should be based on a range of transparent, independent research studies. And Australia’s ‘forgotten fuel’, energy efficiency must have much higher priority to reflect its multiple benefits.