The NEG will reduce competition, and that’s not good for prices

Print Friendly, PDF & Email


In this note we look at how NEM-wide electricity emissions intensity will move, assuming a linear pathway between 2020 and 2030. We choose two scenarios: the already agreed 26-28 per cent reduction by 2030, and because it is universally recognized that 28 per cent reduction isn’t enough to do our share of keeping global warming to 2°C, we also look at a 50 per cent reduction by 2030.

There is no really strong message from the data. We forecast average NEM-wide emission intensity to fall to 0.65 tCO2/MWh under a 28 per cent reduction scenario and 0.45 tCO2/MWh under a 50 per cent reduction scenario.

This compares with 0.86 in 2016. New open cycle gas won’t meet the threshold from about 2025 and combined cycle gas will be marginal from 2030 under the 50 per cent reduction.

However the gas, like coal, can still be used in a portfolio that meets the guarantee for that year. In fact it’s perfectly possible to see how a little bit of gas can go a long way in a largely renewable portfolio.

State run reverse auctions are still the lowest cost, most effective way to procure new supply

Increasingly, what we don’t like about the federal government’s plan is the crucial and more powerful role of the gen-tailers, and we think that without the carrot of an LRET price or CET price there will be much less incentive for merchant plant.

It seems like the government’s plan may reduce competition, and that is never good for price in the long run. One of the many advantages of state-run reverse auctions is that they essentially introduce new suppliers and new players into the system.

When the states force supply in via a reverse auction, it tends to lower prices. It may crowd out some gen-tailer development, but on balance we see it as a positive. The states do not, of course, own or dispatch the generation. They just provide a CFD that finances the new supply. It’s an effective model that we think should be encouraged.

Demand is tough to forecast, more demand = more renewables

The numbers are straightforward, except in forecasting demand for grid delivered electricity. Forecasts of electricity demand in 10 years are unlikely to be correct. Forecasting is quite simply a mug’s game. That said, every investment will be based on some forecast or another. Investing is essentially a bet on the future.

The factors we consider in forecasting demand are: It seems like the government’s plan may reduce competition and that is never good for price in the long run. One of the many advantages of state-run reverse auctions is that they essentially introduce new suppliers and new players into the system.


  1. Looking around the world
  2. Energy efficiency and price
  3. Competition from distributed electricity
  4. Electric vehicles.

Historic demand for NEM grid delivered electricity over the 2005-2016 period is as follows:

Figure 1 NEM Metered generation. Source: NEM Review

Figure 1 NEM Metered generation. Source: NEM Review

If we look around the world we see many instances of GDP growing, but electricity consumption remaining static. Europe, in general, broadly reflects these characteristics. In Australia, though, population growth provides a natural source of demand growth.

We anticipate grid delivered electricity prices to stay constant, or rise. As such, there won’t be much of a kicker to demand from price elasticity.

Should grid delivered prices fall, that would be another matter. Demand response is likely to grow in significance and it will tend to both reduce volatility and, to an extent, result in some foregone consumption.

Distributed electricity, that is primarily behind the meter PV but increasingly micro grids and household storage, is not yet inherently cheaper than grid delivered electricity, but the consumer clearly sees it as cheaper because the consumer can stay connected to the grid and use it as the supplier of last resort. Ongoing growth in distributed electricity is a certainty under current rules and arguably is socially desirable.

It’s socially desirable because it provides countervailing consumer power to big utilities and because it makes the grid more resilient. Anyhow, for our purposes, it’s enough to say that growth in distributed electricity will continue to act as a negative for grid delivered demand growth.

Electric vehicles are a wild card. There are about 1.1 million new cars sold every year in Australia. For this exercise we assumed that EV sales grow linearly from today’s negligible level to 200K  per year in 2030. Cumulative EVs on the road by 2030 will then be about 1.2 million. We further estimate 15,000 km per year per car, and 19 KWh/100 km.

If, and it’s always a doubt, I’ve punched the formulae into Excel correctly that gives me about 3.5TWh of electricity consumption per year. So even a relatively large forecasting error (100%) would make only modest difference to overall estimates.

Finally it’s worth noting that since the CO2 reduction is expressed in absolute terms, any increase in demand requires a bigger share of low emissions generation.

Emissions intensity estimates detailed

The figure below grows grid delivered demand at 0.5 per cent per year. In the next couple of years we’ve made the falls in carbon bigger than a linear trajectory to reflect the closure of Hazelwood and known new renewable supply. From FY20 we have a linear reduction, reflecting the assumed “emissions guarantee target” [EGT]

Figure 2: NEM wide Emission intensity under two COP21 scenarios. Source: ITKe

Figure 2: NEM wide Emission intensity under two COP21 scenarios. Source: ITKe



Figure 3 Emissions intenstity. Source: ITKe

New-build emission intensity

For thermal generation the numbers we use as rough rules of thumb for new-build carbon intensity are:

Figure 4 Approximate emission intensity new thermal. Source: ITKe

Figure 4 Approximate emission intensity new thermal. Source: ITKe

Emissions intensity are reported on various bases. The main issue is that thermal generator tend to self consume their own power (auxiliary demand). This can be 6-10 per cent of output. We are interested in the carbon intensity of the generation actually metered to the grid, which is higher than if total output is used as the denominator.

Gas will struggle on its own from 2025 to meet emissions guarantee requirements

Using our numbers, open cycle gas will not meet emissions intensity guarantee under either a 28 per cent or 50 per cent reduction from about 2025, and combined cycle gas will struggle from 2030 under a 50 per cent reduction scenario. Some open cycle gas plants, Mortlake, can be more efficient than our benchmark numbers.

This is not to say there is no role for gas. Only a little bit of gas can go a long way if it is used efficiently. Gas can still be used in the same way as coal, if it is part of a portfolio which includes lower emission technologies

We already showed how a 1GW dispatchable  wind, PV, gas and battery generator can be built to replace, say, Liddell with an average emissions intensity of just 0.11 tCO2/MWh

Building a 1GW dispatchable, largely renewable generator in NSW

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.

Hear the discussion on the latest Energy Insiders podcast with China-based Greenpeace analyst Lauri Myllyvirta.


  • howardpatr

    NEG and grid reliability is straight out of Secretary Perry, one of Trump’s new swamp dwellers who like Abbott and Co, who is doing the bidding of the coal industry in the USA.

    The ALP and Greens must demand more renewables in the sleasy document/propsal put forward by the Coalition and the Pierce influence SEB.