Where’s Waldo? The wobbly assumptions underpinning the NEG

Print Friendly, PDF & Email

ESB price assumptions don’t make sense, and no company would make such a crucial decision based on such flimsy evidence.

Figure 2: Energy by fuel NEG and NO NEG cases. Source: ESB
Print Friendly, PDF & Email

“Sometimes I’ve believed as many as six impossible things before breakfast.” Lewis Carrol

NEG modelling is likely agenda driven, but it doesn’t matter

UPDATED: The fuller set of information released by the ESB is in our view still incomplete. No matter as it’s only modelling anyway and conclusions depend on the input assumptions.

In our view nothing short of a spreadsheet with a full set of outputs would really suffice. How can policy decisions be made on the basis of a few charts?

What company management would commit to a policy without very detailed scenario analysis? For all we know perhaps COAG members do get the spreadsheet. We, the public, have to make do with some pictures.

In any case we don’t think any decisions are going to be made on the basis of the modelling. Our view remains that the modelling was an expensive agenda-driven exercise designed to come up with a headline that prices will be lower with the NEG than without.

In our limited view you need very, very good glasses to see why that’s the case on the basis of the modelling graph results shown.

In the case of ACIL Allen’s modelling for the ESB we understand that the assumptions were provided by the ESB and run through ACIL’s proprietary “powermark” model.


Assumptions of note include:

– Only the first stage of the Queensland and Victorian targets were modelled. Any later stages were ignored. No explanation for this was provided except the general statement that the modelling was designed to outline differences between NEG and no NEG cases.

– Coal prices are assumed to decline from US $120/t down to US$60/t. FWIW we think this could be a reasonable assumption.

– On the technology capital cost assumptions we have two comments:

  • ACIL assumes a  pretax WACC rises from 6.3% to 8.9% over the forecast period as interest rates “normalize”. Higher WACC penalizes renewables compared to thermal generation. More to the point I have never read  of a “normal” interest rate in the finance textbooks I studied. This must be a new concept in finance. Very broadly speaking it also runs contrary to a 30 year trend of falling global interest rates. Nevertheless it could be correct.
  • It’s unclear whether the rising capacity factors of wind and even PV are accounted for.

The ESB still did not explicitly provide a graph, let alone a table, of new investment expected under the NEG case out to 2030

NEG scenario impact well camouflaged in graphs

In any case shown below are four charts. Two of the charts show the energy and generation mix in the NEG case and two in the NO-NEG case.

We haven’t identified which is which because we are sure our intelligent readers can do that for themselves. Can’t you?

Figure 1 NEG and no NEG. You cut, I choose, Source: ESB report

Spot the difference? Neither can I. Seriously. It’s ridiculous.

If you look at energy side by side the same picture emerges. In 2030, as near as you can tell from a graph the share of coal fired generation  (brown and black combined is virtually identical.

One of the two scenarios has a tiny bit more brown coal relative to black coal. We couldn’t get these graphs to exactly line up so we might be missing something, but not much. The text of the report states that the renewable share in 2030 is 36 per cent under the NEG scenario compared to 34 per cent under the no NEG scenario.

Figure 2: Energy by fuel NEG and NO NEG cases. Source: ESB

Can these supply differences explain the price differences?

The ESB notes that one factor expected to drive prices lower is a 5% higher level of contracting if the NEG is passed as compared to the no NEG case. We understand that this is explicitly allowed for in the Powermark model and forces more thermal generation to be offered at SRMC.

Not sure I’d want to be contracting at the prices modelled in this report but there you go.

Clearly this provides an incentive to the generation sector to avoid contracting or to arrange things so that power is not offered at SRMC. It is axiomatic that if all power is offered at SRMC the sector will go broke.

I can recall Paul Simshauser when he was at AGL getting analysts to play a bidding game where the whole point was to demonstrate that SRMC bidding was a poor long run strategy.

Those contracting incentives must be very powerful because it stretches my imagination (never very good) that such small differences in generation by fuel outlined above could produce the much larger price differentials the modelling has come up with.

Of course demand in both scenarios is identical The price difference is modelled to something like a more than a $15/MWh difference in nominal dollars by 2025.


Figure 2 ESB modelled prices


Final thoughts:

As stated above, I doubt the modelling will do anything other than be used as a pawn in negotiations. My personal view is it presents an unlikely scenario of what will actually happen in both the NEG and no-NEG cases.

Secondly, it’s worth restating that high prices are not an indication of market failure. Market failure, normally, is defined as a lack of activity, a lack of trading, a lack of bids and offers, or a lack of clear, transparent rules and reporting.

The REC market and the gas market probably show more evidence of market failure than the electricity market. Perhaps the OTC market also falls into that category.

Thirdly, we don’t think enough thought is being given to market design. This policy is being presented as “the policy.” However, to many of us in the industry we question whether the “energy only” market is going to work properly in a world where most of the energy is zero marginal cost.

The question, is who is getting spilled at lunchtime? And yet you still can’t get a clear enough price signal to invest in a pumped hydro plant in South Australia.

Anyhow, that’s a question for another day.

David Leitch is principal of ITK. He was formerly a Utility Analyst for leading investment banks over the past 30 years. The views expressed are his own. Please note our new section, Energy Markets, which will include analysis from Leitch on the energy markets and broader energy issues. And also note our live generation widget, and the APVI solar contribution.

Print Friendly, PDF & Email

  1. Perhaps they’ve managed to publish the same graph twice by mistake. 😛

    • Joe 1 year ago

      Ha, ha, ha.

  2. MaxG 1 year ago

    “How can policy decisions be made on the basis of a few charts?”
    When you want to push an agenda, you make up some graphs for the sillies and try to push it through — isn’t this what the NEG is all about? Hogwash!

    • Joe 1 year ago

      Lucky or us Ministers Shane, Lily and Anthony are no stooges.

  3. bruce mountain 1 year ago

    Thanks David. I went through much the same exercise as you. I don’t know whether to laugh or cry at this stuff. Australia’s energy polity at one time deserved respect. How must these folk be feeling about themselves now, must be a touch painful? On the contracting, I have heard others lend some credibility to the bizarre idea that the NEG will result in higher contracting and this will depress spot prices. Besides being an assumption lacking inany credible basis, the concept itself is nonsensical: why would the purchase of association rights affect dispatch? These are not financial contracts they are completely independent “association rights”. They absolutely do not affect dispatch and will have no $/MWh price associated with them that producers would then consider in their bidding decisions.

    • Andy Saunders 1 year ago

      Yes, and the firming contracting will be non-transparent (unless someone knows something I don’t) – certainly more so than the energy-only NEM and hence likely to be much more of a market failure… creating a hidden market likely full of poor bid-offer spreads and high friction.

      I like your term “association rights” (although maybe they should be “association obligations”?)

  4. Matt S 1 year ago

    Bravo David. Always grateful for your analysis

  5. Peter F 1 year ago

    David two complicated but important points on the finances.
    1. A coal plant takes seven years from commitment to build. Over that time an average of 40% of the project cost is committed. Project finance is obviously more expensive than long term operating finance because execution risks ( see Dattelen IV) and delivery delays and regulatory or fuel price changes can all impact the project. The project finance (40% x 8% x 7 yrs) adds about 17-24% to the notional contract cost. In the case of a windfarm there are fewer risks and shorter delivery and a solar farm again even less so project financing adds about 7-10% to the windfarm and 4-7% to the solar farm.

    When you add these costs in, coal in Australia will cost about $3.5-4 m /MW, Wind about $1.9 m and solar about $1.3 m. allowing for respective capacity factors of 65% coal, 44% wind (new wind farms are already doing better) and 29% for tracking solar the adjusted capital cost per annual MWh capacity is $615- $700 for coal. For wind it is around $500 and for solar $510.
    So in fact a rise in interest rates will aid renewables because the capital cost per MWh is now lower than coal.

    2.There is a further compounding factor. Once the plant is up and running if prices fall to say $40/MWh none of the operators will be able to repay their loans and all will go broke. The coal plant will be a stranded asset with a huge cleanup liability because $40 won’t even cover the SRMC. The solar farm can be on sold and make money at $12/MWh to cover maintenance and operation. The wind farm will be on sold and make money at $20-30/MWh so the risk to the financier is much lower. Therefore the ongoing finance costs for the renewables will be 1-2% less than the coal plant. If interest rates go up that differential will widen.

    I could do all the DCFs with appropriate interest rates and different lifetimes etc, but now at almost any risk free interest rate, the capital and interest repayments on wind and solar are less than coal. If coal fell to the US average CF of 53% then owning a coal plant would be a bloodbath

    • David leitch 1 year ago

      Coal plants can run at 85% capacity factors and have done so in Australia when there is a market. Overnight capital cost is the conventional benchmark and project finance for the “right” project can be attractive.

      • Peter F 1 year ago

        The best coal plants can run at 85% CF when they are new and demand is relatively stable. Even after the closure of Hazelwood the average in Australia is about 66%, India 58%, USA 53%, Germany 50%, China 46%. Demand is becoming less stable. In the good old days minimum demand was 40% of peak now it is heading to 30% in Victoria and zero in SA. Thus all plants will have to cycle more so CF will fall for all but the very lowest SRMC plants.

        You might be able to claim that your new high efficiency plant will run at 75% for a couple of years because it is cheaper to run than old plants but as you have said if you run on the short run cost for too long in a fairly short time you will go broke

        Project finance is pretty much always more expensive for construction than operation, because timing and execution risks have been to be included in the risk premium, therefore whatever the “attractive” rate is it will cost more money for coal than wind or solar.

    • juxx0r 1 year ago

      Thanks Peter, this is exactly right. Risk is now going against the traditional sources of power.

  6. Les Johnston 1 year ago

    Withholding the details of all assumptions highlights the shallowness of LNP claims of being unfairly criticised. Why should informed people accept a policy merely because a politician proposes that policy? All propositions must be subject to critical analysis. Scientists and engineers are self critical and work hard to undermine the flaw in their own argument. The challenge to the ESB/LNP is to do likewise.

Comments are closed.

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