Back in 2012, Macquarie Infrastructure & Real Assets made a surprise move into the high stakes world of combined cycle gas turbine plants. Six years later, facing losses and low energy prices, the plants are up for sale. What are the odds of the manager achieving its desired returns?
Betting on gas and electricity spot prices is generally viewed as risky. Throw into the mix growing competition from other energy sources, high tax on carbon emissions, plus uncertainty about the scale of obtainable subsidies, then the bet turns into a veritable game of poker. Such are the challenges facing owners of combined cycle gas turbine (CCGT) plants, which use superheated steam and gas to turn two electrical generators.
Owning such assets may make sense for utilities, which have large energy trading teams and relatively low return targets, or even a private equity fund with a healthy appetite for risk. It makes less sense for infrastructure funds, which typically target essential service assets with high barriers to entry, long-term contracts and relatively guaranteed returns that match their limited partners’ long-term liabilities.
This did not stop Macquarie Infrastructure & Real Assets (MIRA) from buying three such plants: one from GE at Baglan Bay in Wales in October 2012; a second one three months later at Sutton Bridge in Lincolnshire from EDF; and a third, the Severn Power station near Newport, Wales, in December 2013 from Dong (now called Orsted). In 2016, it acquired from RWE the rights to develop a 2.4GW CCGT plant at Willington in Derbyshire.
Gordon Parsons, the MIRA director who represented Macquarie on the deals, was sufficiently confident of the plants’ success that he invested some of his own personal wealth into the investment, sources said.
Parsons and Macquarie, which declined to comment, also convinced four relatively low-risk investors – ADIA, Malaysian utility Petronas, the Ontario Pensions Board, and the Canadian Medical Protective Association – also to buy shares in a vehicle called Calon Energy set up to own the power stations.
The shareholders have now decided to cash in on their investment, hiring Jefferies as their financial adviser. The deal is structured to enable investors to buy either the whole lot; the Sutton and Severn plants; or the Willington plant, which is not yet built, and Baglan Bay plants, which needs upgrading.
This raises the question as to whether Macquarie will earn a 13-15% IRR from the sale, the level of return a manager would typically expect from such an investment. Answering this raises a mirror up to the future of energy supply, and the increasingly important role of gas within the energy mix.
An easy answer is that Macquarie has failed. The Sutton, Severn and Baglan plants have been lossmaking for years, although these losses have narrowed of late. Severn Power made a GBP 62.6m loss in the year to 31 March 2016 and a GBP 14.9m loss in the same period last year.
The reason mainly attributed to these losses is the plants’ low sparkspread – the difference between the sale price of electricity on the one hand, and the cost of gas and carbon tax on the other. The gross margin – the profit from the sparkspread – is crucial because on average, sources say, it amounts to some 70% of a CCGT plant’s revenue.
“The sparkspread is a material and perhaps underestimated contributor to the plants’ financial weakness,” one person with knowledge of Calon said.
At a cursory glance, the sparkspread numbers look shocking. Deloitte recently reported the countrywide sparkspread average to be around GBP 2.5 per MWh, while Ofgem said that from September 2016 to September 2017 they varied between GBP 5 per MWh and GBP 27 per MWh. This fits with the “captured clean tax sparkspread” (CTSS) reported by Calon’s plants. In 2016, Baglan’s CTSS was 5.27 per MWh while last year it was GBP 11.59 per MWh.
From these wafer thin margins, Calon must subtract salaries, insurance, network connections costs and other payments. According to one valuation expert, these additional costs amount to between some GBP 20 per MWh and GBP 25 per MWh. On top of this, the CCGT owner subtracts tax and depreciation.
The source of the sparkspread problem lies partly in Macquarie’s assumptions around the impact of coal plant closures on energy prices.
Between 2012 and 2013, the amount of electricity sourced from coal-fired power plants was dwindling. Come 2017, just 6% of the UK’s electricity needs was sourced from this fossil fuel. Macquarie expected this would result in a lower reserve margin – the reserve of power available to meet demand – which in turn would push up prices. One energy investor said: “It wasn’t that Macquarie thought Calon could fill the gap, as it was already part of the reserve margin, but rather that power prices would increase. Basically, it believed the standard demand-supply issue in economics would apply, as demand outstrips supply, prices of the underlying goods go up. In turn, this would help to increase profits.”
In reality, despite many coal plants having come offline, power prices have not risen in line as expected. Experts cite many reasons, a key one being the far higher than expected influx of renewables having connected to the UK’s grid since 2012. One energy specialist said: “Renewable costs have got a lot of cheaper and we are looking at a future [in which] renewables takes a bigger role than expected. In the early part of this decade, people were expecting very tight capacity margins for a period of time that would then to lead to an uptick in sparkspread. However, it has happened less frequently and later than people had expected at the time.” This is not to say there have not been increases in power prices – such as during the cold winter months of 2016 – but simply not enough for assumptions to have met reality.
There is some light at the end of the tunnel, however. Timera, an energy consultancy, says there has been some recovery in sparkspreads over recent months. Also, Macquarie is understood to have produced some fairly punchy assumptions for sparkspreads linked to its unbuilt Willington plant. One insider says Macquarie expects up to GBP 45 per MWh at Willington, but added that “the more realistic view for Willington would be GBP 25-30/MWh.” Based on these assumptions, the likely sale price of Willington is around four or five times Willington’s forecast EBITDA, two times below the seven times EBITDA valuation of many utilities.
Gas prices however remain hard to predict. They soared earlier this year in the wake of the winter cold spell, and generally sourcing of gas has become more uncertain. One Calon employee, who did not wish to be named, said: “Gas is a very complex picture these days. Some comes to the UK from Russia, but it has been controlling its supplies for its political ends. North Sea gas is dwindling. At Calon we are therefore more reliant on LNG from the US, but doing so puts us at the mercy of the worldwide market.”
Given these uncertainties, one stabilising factor could be the top-up of cash from receiving a clearing price – basically a subsidy – at the UK’s low-carbon capacity market auctions. In return for a monthly payment, generators must guarantee a certain level of capacity at any point in the year.
In the 2015 capacity auction, Calon’s Sutton, Severn and Baglan plants were allowed to receive GBP 18 per KW for one year, 2019. On the basis that these plants have a combined capacity of 2,189MW, Calon will next year earn a cash payment from the auction of just under GBP 40m.
This is a handsome sum, although much lower than the amount to be paid to companies planning to build new plants. For instance, Centrica will earn an annual GBP 22.50 per KW for 15 years from October 2020 for its planned 370MW CCGT plant at King’s Lynn in Norfolk. On this basis, it will earn a grand total over the whole period of GBP 124.8m. Furthermore, Calon’s Willington plant last year failed to win a price at a capacity auction, raising questions as to its future.
Calon’s other source of income – said to be around 15% of revenues – is selling electricity under National Grid’s so-called balancing mechanism, which sources said provides around 15% of its income. Under the scheme, National Grid asks plants to provide a top-up amount of electricity during certain peaking periods when consumer demand for energy grows to higher than normal levels (usually in the morning, lunchtime and evening).
On occasion, peaking energy providers, including Calon, have made a small fortune from the balancing mechanism. In November 2015, MPs called for an investigation by Ofgem after it emerged Calon’s Severn Power station had been paid up to GBP 2,500 per MWh, more than 50 times the average wholesale price of electricity.
But this was a one-off. Details of the largest suppliers under the balancing mechanism are hard to come by. Anecdotally, smaller gas turbines – such as those owned by Green Frog and UK Power Reserve – are more able to provide electricity quickly and efficiently given they can be switched on and off more readily than a much larger CCGT plant. Furthermore, plants such as Calon’s Baglan Bay are described as mid-merit, which means their heat rate – the ability to maximize electricity production from burning gas – is lower than other newer CCGT plants. They are also less able to quickly switch on and off their turbines and generators than more advanced plants.
Calon hopes to turn the odds in its favour, however. It plans to upgrade the Baglan Bay plant, and two years ago improved the efficiency and flexibility of its Sutton plant.
Macquarie also hopes to bolster returns by continuing to manage the Calon plants post-sale. That said, it is understood that it only intends to do this for a buyer that does not have specialist asset management or energy specific expertise. This might cause complications. One source said that if such a buyer rejects Macquarie managing the asset post-sale, it may end up having to pay Macquarie upfront the net present value of all fees it would have received.
Furthermore, insiders said Macquarie’s style of ownership of Calon has brought it at times in conflict with some of the company’s other shareholders. One source previously close to Calon said Macquarie wanted to exit within three to five years, while the other investors saw themselves as longer-term investors.
There have also been signs of division at board level. Back in 2014 Calon hired RWE nPower’s chief operating officer, Kevin McCullough, to Calon’s board. One source with inside knowledge said this was done to provide alternative expert opinion at the company’s board. “They wanted an independent management team from Macquarie,” said the person, adding there were “constant run-ins over strategy between Kevin and Macquarie.” It is understood Macquarie recommended and voted in favour of McCullough’s appointment.
A separate insider said ultimately it was Calon’s institutional investors who in the end pushed just as keenly for a sale: “They have had enough of the volatility of this business, with no obvious end in sight.”
Management issues aside, Calon has some important positives. While some developers have failed to raise finance to back new CCGT plants, Calon has remained well supported by debt – first from Macquarie Bank and later by Beal Bank and HSBC. The leverage is said to be fairly average, around 55% of total capital, although one insider said at times Calon “had to bring in debt to keep the power stations liquid.”
Macquarie as a group has successfully helped trade electricity generated by Calon. Furthermore, the shareholders acquired some of the plants at a good time. They bought the Severn plant from Dong at a decent discount to the expected price, for instance. Also, while it gambled unsuccessfully on the Willington plant receiving a subsidy under the capacity market auction, Calon might be successful at a second attempt.
There are other glimmers of hope. In January, the UK government forecast that some 7.5GW of extra CCGT and open cycle gas capacity will be needed between 2018 and 2025. Coal will shortly be all but an irrelevance in the UK’s energy mix. Nuclear is shrouded in political uncertainty while interconnector projects are stumbling in the face of Brexit. However, as one adviser put it, for the Calon sale to succeed, a buyer must gamble on all these situations continuing for the foreseeable future. Meanwhile, the so-called triad subsidy available to CCGT’s competitors – the small scale turbines – looks set to be withdrawn.
MIRA’s foray into the world of CCGTs was doubtless an unusual one for an infrastructure fund. Ultimately, it was just one part of a much wider gamble by MIRA on gas. Its current European gas investments also include Cadent Gas, Open Grid Europe, Compañia Logistica de Hidrocarburos and Società Gosdotti Italia.
As the UK faces up to increased insecurity of energy supply, perhaps buying the plants was a wise move after all. Less likely, however, is that MIRA will meet the high returns it originally sought from Calon.