100% renewables for Australia – not so costly after all

An exploratory study into 100% renewable energy scenarios for Australia has concluded that its impact on consumer electricity prices over the next few decades may be no more than the increases in the last few years to support much criticised network upgrades and the introduction of the carbon price.

The report by the Australian Energy Market Operator (AEMO) – you can access the executive summary here and the draft report here  – canvasses the potential costs and practicality of transforming Australia’s coal-dependent electricity system to 100 per cent renewables, by either 2030 or 2050. It creates two scenarios – depending on the pace of falls in the cost of renewable and storage technologies – but both are considered conservative.

It concludes that the cost could range between $219 and $338 billion and would require wholesale electricity prices of $111-$133/MWh (more than double the current price). Unfortunately, and somewhat controversially, AEMO was not asked to compare these forecasts with “business as usual”, but it does provide one interesting set of data that does put it into some perspective.

The first is the impact on retail prices. See the table below (excuse the draft stamp):

Screen Shot 2013-04-29 at 10.21.03 AM

 

It shows that the impact on consumer electricity costs from a 100% renewables scenario could be as little as 6.6c/kWh, assuming a reasonably optimistic view of technology costs. That compares to the forecast national average increases in retail costs made by the Australian Energy Market Commission from 2011/12 to 2014/15 of 5.4c/kWh.Taking in the two earlier years of  increases, the jump in retail prices has been higher.

This should not come as a surprise. The AEMO study is almost unique in the world in not taking into account BAU. The International Energy Agency, for instance, last year estimated that the world needs to spend an extra $36 trillion on its energy systems by 2050 if it is to meet its 2C scenario. But it points that that this will be more than offset by $US100 trillion in savings through reduced use of fossil fuels.

More recently, the IEA noted that the world needed to spend $5 trillion by 2020 (over and above BAU spending of $19 trillion), but this could be paid for by cutting fossil fuel subsidies. A similar conclusion in the Australian context was reached by the UNSW team led by Mark Diesendorf in their 100 per cent study.

The second thing is to consider wholesale prices. A recent “government Policy” scenario from Treasury has a wholesale price of $110/MWh in 2030 (compared with $111 to $133/MWh for 100% renewables).  That includes a carbon price of around $52/tCO2 in 2030. (Hands up who thinks there will be no carbon price in 2030. Yes, you too, Greg and Tony).

Apart from the lack of comparison with BAU, the AEMO report was hamstrung by a number of other factors, most notably its forced reliance on the technology costs produced last year by the Bureau of Resource Economics.

RenewEconomy has on many occasions questioned those forecasts, which even for 2035 are above current market prices for technologies such as concentrated solar thermal, and assumes, quite bizarrely, no fall in solar PV costs for nearly a whole decade through much of the 2020s.

Still, the AEMO report – although “exploratory” and “limited” in its own words – does come to some useful conclusions. The first is that it says “it is valuable to note that this operational review has uncovered no fundamental limits to 100 per cent renewables.” In other words, it is not a question of can, but how much.

This graphic below probably give the best summation of how the various technologies could be deployed in the various scenarios. Note the large role played by rooftop PV. This would require billions of dollars of investment by households.

This is likely to happen, but only because it is a cheaper option for them than relying entirely on the grid for their electricity, even (or possibly especially) in the 100 per cent renewables scenario. But most scenarios suggest a much greater role for CST and other renewables, particularly wind, which is given only a moderate role in the 2030 scenarios despite its current status as the cheapest renewable source.

Screen Shot 2013-04-29 at 9.29.14 AM

 

The AEMO report says the key question is whether renewables can service peak demand. It makes two observations here – one is that the proliferation of solar PV – is likely to remove the evening summer peak – and the biggest peak will be in meeting winter heating requirements.

It says challenges of meeting peak demand “are generally regarded as not insurmountable”, but it notes that energy storage is key to managing supply and demand, to smooth output, and cover periods when there is not much wind and little sun.

It  modelled three storage technologies – CST with storage such as molten salts, existing pumped hydro, and biogas-fuelled peaking plants,which it noted could add enough “low cost flexibility” to cover any anticipated periods of low generation.

“The modelling shows that the combined dispatch of all three technologies is sufficient to match demand in all four cases, even with the rapid decline of PV generation in the late afternoon,” it noted. Its modelling did not include battery storage or air compression, as it decided that these technologies were still in their formative stages.

AEMO noted that its cost estimates did not include any allowance for the costs of any modifications required to the distribution networks, the cost of acquiring the required land for generation, or the costs of stranded assets (coal and gas fired generators). As for land, it estimates that would require between 2,400 sq kms (50kms by 50kms), and 5,000sq kms.

But it also notes that its modelling results are “highly sensitive” to the assumed technology cost reductions, and any changes to these would see corresponding changes to the modelling outputs.  Given the electricity industry’s propensity to grossly overestimate the cost of renewable technologies, that means there is scope for greatly reduced costs.

And it should be kept in mind, most of Australia’s existing coal and gas fired generation needs to be replaced by 2045 – and as Bloomberg New Energy Finance have pointed out – the cheapest new build generation capacity is already wind, and will soon be joined by solar. That needs to be a critical equation is any assessment of the future, particularly when incorporating environmental costs, as UNSW pointed out.

And there is another missing piece to this assessment – and that is energy efficiency. The modelling is based on AEMO own long term demand forecasts, which have been shown to be pretty hopeless even in recent 12-month forecasts. It does not take into account the sort of energy efficiency gains that could, and should, be contemplated in coming decades.

The IEA, and just about every other major study, suggests energy efficiency should account for at least one third of future scenarios. It points out that it is the energy we don’t use that will be the cheapest and most effective. But that also means a greater diversion from business as usual.

(Final thought: One questions why the study was done without a BAU comparison. It is clear that Labor was a reluctant party to this study, part of an agreement with the Greens as part of the Clean Energy Future package. It was interesting to note that having tried to disguise the costs, it then leaked the final report to paper with the most visibly anti-renewable stance (AFR), and to a report who has written little on energy markets)

Comments

5 responses to “100% renewables for Australia – not so costly after all”

  1. Useful Design Avatar

    The ‘final thought’ was the only part of this story that was unexpected 🙂 When is the final release expected? Are they taking comments on the draft?

  2. Professor Ray Wills Avatar

    While the draft report is a great read, the main holes are still very conservative views on the rates of change and technology improvement. But the most important component above is the fact that most of the expenditure is an alternative spend on energy, not an additional spend on energy.

    BAU must be appraised and included and contrasted.

    As capably outlined above (thanks Giles), these few key points mean all of the above is way cheaper:

    – AEMO assume increase in electricity consumption, albeit modest, based on “relatively low growth in demand, reflecting a less energy-intensive future which is primarily driven by energy efficiency, rooftop PV and demand side participation” — but the averaged trend of the last four years, as reported numerous times in RenewEconomy, has been decline, not just low growth. We won’t need as much generation because of continuing improvements in efficiency, and increased distributed generation from roof tops – therefore smaller investment in future generation, and so the AEMO document underestimates potential here – in a 22 page report the key phrase “energy efficiency” is only used in a sentence twice.

    – the report says no one technology can deliver all – but one can deliver most if storage is cheap. Storage is blind to the source, therefore storage is (as has always been acknowledged) key. By my assessment, storage is where PV was in 2008/9. Market forecasts are for slow falls in the pricing of storage, but logically it will be faster because the demand for storage will be great than the demand for PV as it will be needed by other renewables. Following that rationale, storage will be bottoming out in price by 2017/18, where PV is now.

    – assumptions of price on CSP in 2030 are just silly – will be at least 1/4 of AMEO projected 2030 price of $4.6 million per MW – first CSP’s in WA are likely to be cheaper than the 2030 price in 2014.

  3. Professor Ray Wills Avatar

    Also note above “proliferation of solar PV is likely to remove the evening summer peak and the biggest peak will be in meeting winter heating requirements.”

    Except we are talking 17 more years of climate change and warming temperatures – I note this AFR article today: “Clothing stores face winter of discontent – after yet another weekend of perfect weather, retailers are praying for a cold snap to clear autumn and winter apparel and avoid a blowout in inventories.”

    in April of 2013 it hasn’t been cold (yet) and the warm winter clothes are not selling well. They will be selling even less well in 2030. And less energy for heating (and yes more for summer cooling, but met by PV)

  4. Mark Diesendorf Avatar
    Mark Diesendorf

    Giles, may I clarify one misunderstanding in your piece? The UNSW 100% renewable electricity simulation research group comprises Ben Elliston (PhD candidate and lead author of our papers) and his supervisors Iain MacGill and myself. There is no team leader as such.

  5. Gary Avatar
    Gary

    I think they were a bit quick to dismiss PHES. The generating capacity cost figures they used were based on only 500MW per site. The $4000/kW figure they used would be reduced with more generating capacity per site.

Get up to 3 quotes from pre-vetted solar (and battery) installers.