Yallourn’s “brown fire” future will depend on exports, and ramping ability

The Yallourn brown coal generator in the Latrobe Valley is profitable, but will depend on exports to NSW to remain so as the amount of variable renewable energy sources (wind and solar) constructed south of the NSW border increases.

We expect maintenance capex to remain high and that taking union boss Geoff Dyke at face value, as confirmed by Yallourn’s 2019 Summer performance, reliability will be an issue.

And an increased ramping requirement – as the share of renewables grows – will increase the difficulties.

Unlike NSW, Victoria is taking steps to build new generation in front of coal closing. This is a two-edged sword as the new generation makes the closure of the old generators more certain. However, we are equally sure that Victoria and the Southern region are on the right track. Even though prices in Victoria are higher than in NSW, Victoria’s energy future is becoming more secure.

The trick is in the timing to make sure that enough new generation is built, perhaps including some dispatchable power, to ensure that even an early closure of Yallourn can be managed.

The safety valve is that Yallourn can export even as local supply increases.  We still think another NSW generator will close before Yallourn or before Tarong or Gladstone in Queensland. But perhaps we should run a poll to see what readers think? Please let us know in comments below.

About Yallourn

Yallourn sells  about 10 TWh of electricity per year of power from its 1480 MW of registered capacity.

The power station was acquired by its current owners EnergyAustralia in 2000.  Yallourn W was built in the 1970s but the first small power plant on the site was built about 100 years ago.

The Morwell river was diverted in 2002 to allow access to further coal sources following depletion of coal from the East Field mine. In late 2007 subsidence in the mine wall resulted in the river flooding the mine, damaging the coal conveyor.

Existing coal resource will last until 2032, the currently estimated closure, and there is no doubt more accessible coal in the region.

Yallourn’s profitability seems to have declined

Superficially, taking public statements at face value Yallourn would seem to be highly profitable. See table below for our assumptions.

Figure 1. Source: Company, ITK, AEMO

How reasonable are these numbers?  In our view the fuel and variable opex numbers are on the low side.  Nevertheless, the overall ebit (earnings before interest and tax) for EA on rolling 12 months basis is as follows including ITK’s estimate of the June 2019 six months.

Figure 2. Source: Company

The break up between retail (electricity & gas), wholesale (generation code) and overheads is always heavily dependent on the intersegment transfer pricing policy and what constitutes “corporate” expense.

For what it’s worth, corporate overheads have been higher than retail profits for the past 12 months the way EA presents the data. According to EA there’s about $300 m  of corporate costs a year, that’s a decent size head office if taken at face value and considering that HK head office would have yet another layer of overheads.

Figure 3. Source: Company

The main point is that most of the ebit comes from generation and is split between Yallourn, Mt Piper and the gas and renewable generation PPAs.

Yallourn typically is about 10/19 of total generation by EA and fuel costs are lower in Victoria and generally prices are higher at the pool level than in NSW. So these numbers are consistent with Yallourn having earned something like A$400-$600 m in the 12 months to Dec 18.  Equally though, that won’t be the case this year .

All brown coal stations in Victoria have had to deal with higher coal royalties. These rose from 7.6 cents to 22.8 cents per GJ – still less than the NSW rate of 25.2 cents. We estimate that increase added around $20 m per year to costs. The royalty represents around $3.20/MWh of opex.

Reliability and Safety

It’s fairly well-known that the Hazelwood power station was not seen as an exemplar of the modern safe, clean reliable power station prior to its closure. Leaving aside any issues of carbon, industry gossip was that there were many issues at that power station. The Hazelwood boilers had had safety improvement notices issued, for instance.

It’s not unreasonable to expect ageing power stations and ageing coal mines to have issues. Pressure is high, control systems not necessarily the latest, moving parts wear etc.

It’s always hard to know how seriously to take unverifiable comments made by people with agendas, but here is a quote from “The Age” from June 25 2019.

“Geoff Dyke, of the Construction, Forestry, Mining and Energy Union, said …… he believed the reasons behind Yallourn’s closure would be due to reliability issues rather than financial issues. The plant is in very poor condition and would need a lot of maintenance to be able to run all the way through to 2032,” he said.”

In November 2018 a worker, Graeme Edwards, was killed in a circuit breaker arcing incident at Yallourn.

ITK does factor in a lot of maintenance, across the entire black and brown coal sector in NSW and Victoria we allow broadly about $10 MWh as the annual maintenance capital cost of keeping the power stations going.

What about volume risk

It seems obvious that unlike Mt Piper, Yallourn will have a tough time ramping up and down on a daily basis.  Victoria’s renewable energy target of 40% by 2025 and 50% by 2030 is expressed, we think, on a production basis and not consumption.

As such ITK’s expectation is that as wind and solar output steps up in Victoria, South Australia and Tasmania and including ongoing significant behind the meter increases, then the first plan will be to try to export more energy to NSW.

This is where transmission constraints and later the likely desire for NSW not to be overly dependent on imports may come into play. To the extent that exports are difficult brown coal will lose market share to wind and solar. The exact extent may depend on factors such as Tasmanian dam levels.

What we can say is that given ITK and industry consensus that Yallourn’s variable cost of production is, say, $15 MWh it is nearly as sensitive to volume loss as it is to price decline. $1 MWh in price is $10 m reduction in ebitda for Yallourn but its easier to see it as say a 30% reduction in either volume or price.

Figure 4. Source: Company, AEMO, ITK

Price sensitivity is high but so is volume sensitivity due, as I say, to the high contribution margin (price less variable cost). We inflated coal and opex categories at 3% per year. High but that’s the business in ITK’s view.

Of course, if you really want to think about downside you can put both volume and price decline in. It’s far from impossible.  ITK has allowed $50 m depreciation and we think maintenance capex will be at least that high. But it’s just a guess as there is no good publicly available information.

Volumes will hold up mostly but due to riskier exports to NSW

Figure 5 shows ITK’s current expectations for aggregate brown coal output. We presently estimate a fall on average to 3.9 GW in 2025 compared to 4.1 GW for FY19. That fall is in the mid morning through mid afternoon representing, solar output round the NEM.

We also show our present estimate of net exports in 2025. These are ITK’s Southern Region to NSW exports so from South Australia to NSW as well as Victoria. But they essentially represent coal generation exports as wind and solar in Victoria and even in South Australia can underbid brown coal forcing the coal North.

If there is a transmission or other issue then subject to AEMO reliability issues the reduced export capability will likely come at brown coal’s expense. We see some kind of risk that either the SA-NSW interconnector is delayed, or that works around Snowy 2.0 and the ISP Stage 1 and 2 projects end up causing transmission issues during implementation.  This shouldn’t happen but…

Figure 5. Daily brown coal profile. Source: NEM Review, ITKe

It’s also possible that black coal generation in NSW may compete for market share to limit exports although the higher fuel cost makes that an unwinnable battle as was proved long ago when the NEM was first introduced.

Optimistically assuming our forecast to be correct our guess is that much of the daily ramp requirement will indeed fall on Yallourn since LYA and LYB have 11%-14% better heat rates that translate to about $1 MWh cost advantage.

Coal’s share of Victorian production is down to 71% for the last 30 days in Victoria from 77% in the same period last year (PCP). Wind is up to 15% from 10%. Across the NEM wind and solar including rooftop is running at 15.5% of annualized NEM output. That’s virtually identical with the all time record set in March 2019 and will surely be exceeded this Summer by some distance.

Figure 6. Fuel share. Source: NEM Review

 

Figure 7 Source: NEM Review

 

Figure 8 Source: NEM Review