DAVID GUIVER – VP SHELL ENERGY AUSTRALIA (TRADING AND SUPPLY)

Welcome to this Shell Energy Australia Q2 wholesale energy market update brought to you by the trading team. It’s a real interesting analysis to share this quarter. We’ve seen a great correlation between the power market and the gas market, really driven from some large generation outages throughout the quarter. Quite different to what we saw in Q1, which was a very low volatile market. So over to the team to talk about Q2.

ANDREW HINES – SENIOR ENERGY TRADER

Thanks, Dave. Hi, I’m Andrew Hines, one of the senior energy traders here at Shell Energy Australia. In last quarter’s review, we spoke about the atypical low spot price outcomes we saw through the normally volatile Q1 period, and the supply and demand mix drivers that contributed to these lower than usual prices. As we can see from this chart, which shows the average daily spot price across each node in the NEM, the low prices that we saw through Q1 did not continue into Q2. In fact, Q2 proved to be a more volatile period. In this quarter’s review, we’re going to have a closer look at the supply side dynamics that have not only driven highest power prices, but strong domestic gas prices as well. To illustrate this, we’re going to look at the breakdown of NEM-wide power generation by fuel source, and see how variations in this mix impact the spot price over the last six months.

As well, we look at variations in the available capacity across each of the fuel sources and how the level of generation over each month as a percentage of the capacity of available generation has an impact on spot prices. So this chart here shows the average total output of all the generators across the NEM, grouped together by the main fuel sources. Brown coal, black call, gas fired, liquid fuels, hydro, solar, and wind. For example, this column here shows the total output of all the black coal fire generators across the NEM. The line here shows the average whole of NEM spot price across the first six months of the year. The line here shows the whole of NEM average spot price all across the first six months of the year. The most obvious trend we see is that the average spot prices have increased since February, with strong average prices, particularly across May and June.

Secondly, based on the increasing total generation, demand over the NEM increase, and we’ve seen strong winter demand, on average, greater than what was seen in summer, which we usually expect to be the high demand period. This is due on one hand to a mild summer, and also a particularly cool start to this winter. Focusing now on the mix of generation, we see that the coal fire output was fairly consistent across the period. May to June saw an increase in wind output and solar decreased with the decreasing daylight hours. The most significant change we saw is a much higher gas output in June, and to a lesser extent in May. Now, there were two highly impactful events on generator capacity this quarter. The explosion at the Callide Power Station in Queensland in late May, and the heightened risk of flooding of the Yallourn Power Station mine in Victoria. Both of these events caused a dramatic reduction in the availability of black coal fire generation across the NEM.

As we can see, more expensive gas fire generation was called upon to satisfy the increased demand, which drove spot prices higher. As we will discuss later, this increased demand for gas fire generation drove gas prices higher, which in turn make gas fire generation more expensive. Delving a little further into the generation dynamics across the NEM. These charts now look at plan availability. The child on the left looks at the average total output as a percentage of a total availability, again grouped by fuel source. Really how the actual output compares to the total capacity available. This is a measure of how hard the different generators are running. For example, 100% would mean that the generator’s running flat out 24/7 for the entire period.

On the right, we look at availability as a percentage of the total capacity. Basically how much of the generating fleet was actually available. Staying with the right chart, we can see the impact of the Callide and Yallourn events on the availability of coal fired units through June. However, we do see that availability was already lower, as April to May is usually a time when generators are on planned maintenance outages, so it’s typically a lower availability period. On the other hand, we can see that there was a strong increase in gas fire generation made available. Moving to the output as a percentage of availability, we see that with less units available, the remaining black coal fire generators had to run harder, hitting over 90% of their available capacity, compared with 70 to 80% that we saw through the summer months. This meant that there were more times when higher price generation was being dispatched.

Also, we see that even with the increase in available capacity, the output percentage for gas power generation was up over double what was seen in summer. So not only was there more gas fire generation capacity available, but that generation that was available was running harder. Both these factors combined to put upward pressure on power prices. Looking at some of the individual fuel source generation characteristics, these charts show the average output by fuel source for the month of January, and then show the changes from this month on month. So looking at gas fire generation, we see that the output levels were relatively flat across January to April, before we saw that increases in May and much more dramatically in June, where we see the generation output has more than doubled compared to January.

Moving on to coal fire generation in Queensland and New South Wales, we see the decreasing output from the Queensland generators, following the outages caused by the Callide event, while there was an increase in output from the New South Wales gen to help balance this out, along with increasing demand. The hydro units across the NEM also increase their output over May to June. The Tasmanian generators especially saw relatively strong increases with additional export to the mainland. All these changes to the generation mix, in particular the increased gas fire generation and the strong winter demand, combined to draw a significantly higher spot price outcomes. The price dynamics across the first six months of this year have been quite atypical. To explore this, Lasanthi is going to compare this year’s results to what we saw last year.

LASANTHI WEERASEKARA – TRADING ANALYST

Hi, my name’s Lasanthi and I’m a Trading Analyst with Shell Energy Australia. In the following slide, we can see how the average output for each fuel source and the average spot price in the NEM for the past six months have differed from the same period last year. Generally we expect price and output outcomes to follow similar trends to what happened in 2020. That is, with average demand highest in the summer, gas and coal units operate at higher and more expensive levels to meet supply needs. As such, we expect average prices to be high during this time of year. As the year progresses, the weather cools down and total demand lowers, making it less likely for price spikes due to extreme demand to occur and for more expensive sources of power to be required for support. Accordingly, average prices tend to decrease noticeably after summer, as it did in 2020.

However, this year we’ve experienced the opposite trend in spot price outcomes and generation output. Total generation was comparatively lower this summer, as milder temperatures suppressed demand. Furthermore, gas and coal output played a lesser role in Q1, as renewable output from wind and solar grew. In short, lower demand and cheaper power from renewables led to considerably lower average spot prices this summer compared to 2020. However, looking at Q2, colder average temperatures this year has led to higher demand than normal. As a result, higher generation price bands, largely from gas, have had to be dispatched to meet this demand. In turn more gas generation puts pressure on gas prices, which therefore makes generation more expensive. Moreover, coal availability has been much lower in Q2 this year. This is due to routine planned outages as well as major unplanned outages, namely from Callide and Yallourn power station units.

As a result, the remaining coal generators had to be dispatched at high levels, and accordingly at more expensive price bands to make up the difference. The result of these conditions is a much higher average spot price going into Q2 this year than we would normally see, and indeed did see, last year. The following charts allow us to take a closer look at how generation output and bid availability from January to June compares to the same period last year. In the output as a percentage of availability graph, we can clearly see black coal and gas was dispatched at higher proportions of their availability, with black coal having 90%, and gas having 30% of their bid availability dispatched in June. In the availability as a percentage of total capacity graph, we can see that gas bid significantly less of its capacity available for most of this year compared to last year, but ramped up their bid availability considerably from April onwards due to high demand outcomes and lower availability from the coal units.

While we tend to see planned coal generation outages in the months following summer, the significant coal capacity that was lost with Callide and Yallourn unit outages prevented coal availability from increasing into the winter months as it did last year. In summary, this year has seen a flip in the average spot price trend. Mild summer temperatures coupled with higher proportions of renewable generation suppressed price outcomes in Q1 this year. However, much colder temperatures heading into winter led to higher demand levels than previous years. Finally, the combination of planned and major unplanned outages constraint dispatchable availability, leading to coal and gas generators being dispatched at more expensive levels to meet demand.

ANDREW HINES – SENIOR ENERGY TRADER

Thanks Lasanthi for that excellent comparative analysis. We’re now going to take a closer look at the two high impact incidents that occurred this quarter, starting with the events that the Callide C Power Station in Queensland. On the 25th of May, a catastrophic failure of the Callide C unit four took out all four generating units at the Callide C Power Station. At the time, this was over 1,100 megawatts of generation and a total of over 1,600 megawatts of capacity. Looking at the events on that day, this chart shows a half hourly output across the Queensland generation fleet by fuel source, along with the Queensland spot price. The impact on spot prices was immediate, with the price spiking as other units were affected. However, the price volatility at this point was dampened somewhat by the tripping of transmission lines which reduced load across the state. However, as the evening peak demand approached and solar output decreased, prices reached the market price cap of $15,100 a megawatt hour.

As we can see the gap caused by reduction in coal fired output was overwhelmingly filled by gas fire generation, along hydro peaking units, as we reached the maximum demand of the evening. The increase in gas fire generation continued through the night as other coal fired units slowly returned to service. Looking out further, this chart shows the average daily output of the Callide units, along with the average daily Queensland spot price. This clearly shows the impact on spot prices caused by the loss of the Callide generation, with prices softening somewhat on the return of the Callide B units through mid June. The other less dramatic event for Q2 three key two was the shutdown of the Yallourn Power Station coal mine due to a heightened risk of flooding. Cracks had emerged in the Morwell river diversion, which was rerouted to run through the Yallourn mine about 15 years ago. Recent floods have resulted in major cracking in the walls of the diversion, raising the risks of further flooding into the mine.

The damage to the mine, which was evacuated after recent wild weather, prompted the Victorian government to declare an energy emergency in June so urgent repair works could be carried out. While the Yallourn units themselves weren’t impacted, their coal supply was, which led to units being shut down to conserve fuel. The impact on spot prices was not as sudden as the Callide C event, with a more steady increase in pricing as greater gas fire generation was needed to fill the gap left by the reduction in Yallourn output. The increasing usage in the already high demand winter period put upward pressure on gas prices, which in turn increased the cost of gas fire generation, putting upward pressure on spot prices. To explore the impacts on gas prices, this quarter we’re introducing Grant, who’s going to give us a deeper discussion on domestic gas prices.

GRANT SHANNON– LEAD GAS TRADER

Hi, I’m Grant Shannon, acting lead gas trader, wholesale gas. Today I’ll be talking you through the Q2 highlights for the east coast gas market following on from Andrew’s analysis of the NEM. This includes review of gas pricing domestically, flows along the Southwest Queensland pipeline, and the response in the Iona gas storage position after the Callide event. Prices around the East Coast gas market were mild in the first quarter. Most markets trading around $6 a gigajoule. There was relatively low volatility and prices were also low, somewhat unusual for the summer quarter. The commencement of Q2 into the shoulder month of April started relatively mild, before a brief price shock in the middle of April, which occurred due to two constraints. Firstly withdrawals from Iona into the DWGM, and also injections at Culken. Following the failure of the Callide C4 unit on the 25th of May, domestic gas prices rallied significantly through to the end of the quarter.

This increase in gas prices across the entirety of the East Coast market was driven principally by the increase in gas generation due to the reduction in the availability of Queensland black coal following the Callide event. This is also combined with increased demand in southern markets due to the commencement of the Australian winter. Despite most gas traded bilaterally ahead of time around the market, events like Callide are almost impossible to predict. Therefore it is likely that marginal gas volumes to support fuel consumption of gas fire generation were purchased at short notice via the short-term trading markets providing upward pressure on prices. The Southwest Queensland pipeline, or SWQP, is the single route for gas flows between Queensland and the southern markets. SWQP is a bi-directional pipeline. Gas flow’s normally dictated by the pricing differential between Queensland and the southern states.

In the Australian summer SQP typically flows from south towards the north, which reflects the Queensland market trading at a premium to the southern markets. This corresponds with northern hemisphere winter, in which there is higher demand for LNG for heating and power generation. Conversely, in the Australian winter, the SWQP typically flows from the north to the south, which reflects the southern markets trading at a premium to Queensland. This is typically driven by increased demand for heating due to the commencement of the Australian winter. At the start of the year, strong Queensland market pricing incentivized high Northern flows along the SWQP, up to 100 terajoules per day greater than the year before, as seen on the chart. As the second quarter began, the SWQP transitioned flow direction from north to south. Queensland coal coalescence gas supplementing supply into the southern markets as cooler weather conditions started to occur. Most years, as winter commences, the SWQP continues to flow gas from north to south, building buffer inventories in the Iona gas storage facility, up to the SWQP capacity of 400 terajoules per day.

These high flows along the SWQP, up to the facility name plate, were seen in 2020. However, this year following the Callide event, there was significantly higher demand for gas in Queensland due to the increase in gas fire generation, which resulted in much less gas transported south than usual along the SWQP. On some days in June, this amounted to several hundred terajoules per day less gas flowing south than the year before as seen in the chart.

In summary, this year, due to the failure of the Callide C4 unit on the 25th of May, there was significantly less gas flow south along the SWQP than previous years. A key part of balancing the east coast gas market, particularly over winter, is the Iona underground gas storage facility, which can hold up to six LNG cargoes worth of gas. As mentioned in the analysis of SWQP flows. Iona inventories typically increase over summer when Southern demand is low, decreasing in winter, as southern demand increases. This year, Iona appeared to be relatively well supplied prior to the Callide event, reaching a total storage inventory of just under 25 petajoules.

Due to the significant increase in gas generation following the Callide event and the reduced southern flow of gas along the SWQP due the high demand of Queensland gas generation, inventories at Iona were withdrawn at record rates during June. Nearly 10 petajoules withdrawn in just under two months. At the start of the third quarter, Iona inventories were abnormally low, coinciding with peak winter demand in southern markets. In summary, the loss of the Callide C4 unit in late May lead to reduced availability of Queensland black coal, which increased demand for gas from gas fire generators, driving increases power and gas prices. This then led to reduced southern flows along the SWQP and increased an early withdrawals from the Iona storage facility.

ANDREW HINES – SENIOR ENERGY TRADER

Thanks, Grant, for that excellent analysis on the domestic gas market. So far, we’ve discussed all the impacts on the physical spot market. Now we’re going to look at how in turn this has impacted on the forward contracting market. This chart shows the end of day settlement price for the ASX futures contracts within year 21/22 for Queensland, New South Wales, and Victoria. With the weakest spot prices in Q1, we can see that the futures contracts decreased in value, with a downward trend across all the nodes. As we entered the outage periods of G2 and spot prices began to increase, we saw futures contracts begin to strengthen across the same period. In the end of May and moving through June, we can see the impact on the futures contracts from the Callide C and Yallourn events.

In the back end of June futures contracts continue to rise as a tightening gas supply put upward pressure on spot prices, which in turn put upward pressure on futures contract prices. So we can see how the spot price dynamics have an impact on the futures contract pricing. Through Q1, where we have mild spot prices, we saw a depression of contract prices. As you can see, when we moved into Q2 and we had the highest spot price outcomes, we saw contract prices begin to increase. With the two high impact events at Callide C and Yallourn, we saw the contract prices jump. With higher spot price outcomes continuing, futures contracts are continuing to rise. Back to you, Dave.

DAVID GUIVER – VP SHELL ENERGY AUSTRALIA (TRADING AND SUPPLY)

Thanks, team. Fantastic update. Really interesting analysis there. It was really quite an amazing quarter, and a real insight to the challenges of the energy transition. Of course, stay up to date, lots more information coming to you. Please subscribe. See you next time.