Investors beware: “Cheap" renewables are very expensive

Chris Leithner

Leithner & Company Ltd

For years, it’s not merely been conventional wisdom; it’s become a pillar of the zeitgeist and a litmus test of acceptable opinion: electricity generated from intermittent sources, particularly solar panels and wind farms, is much cheaper than baseload power from coal- and gas-fired plants. Respected sources underpin this orthodoxy. On 11 December 2020, for example, the Commonwealth Scientific and Industrial Research Organisation stated:

A new report from CSIRO … has found that solar and wind continue to be the cheapest source of energy in Australia … This is projected to continue to be the case throughout the projection period to 2050.

On countless occasions, the mainstream media has repeated (or uncritically accepted) claims such as CSIRO’s. “Renewable energy is the cheapest form of energy, and that includes transmission and storage,” Chris Bowen, the new Minister for Climate Change and Energy, told ABC 7.30 on 9 June. On 1 June, however, ABC News reported:

Australia is on the precipice of a UK-style energy crisis that could send many of its power retailers broke and fuel a surge of households unable to pay their bills … The soaring cost of wholesale energy has triggered widespread alarm among observers and there are fears a significant chunk of Australia’s retail electricity market will be wiped out in the crunch.

The consequences of the crisis extend well beyond Australia’s energy markets. “Here we are,” commented Ian Verrender on 6 June, “faced with a critical failure in energy and electricity supplies that threatens to wreak havoc on the economy, shut down major manufacturing industries and cause a potential spike in both unemployment and inflation.” According to The Sydney Morning Herald (1 June), this is the “biggest energy crisis in 50 years.” Moreover, Australia is hardly alone: the world presently faces a “much bigger” energy crisis than it did in the 1970s, the Executive Director of the International Energy Agency (IEA), Fatih Birol, told the German daily, Der Spiegel, on 1 June. “Back then it was just about oil. Now we have an oil crisis, a gas crisis and an electricity crisis simultaneously.” 

Today’s crisis shouldn’t surprise you

How is this possible? If solar and wind produce the cheapest power, and since intermittent sources have generated a significantly greater share (and coal-fired generators a materially lower percentage) of Australia’s electricity over the past decade (for details, see Decarbonisation: A doubter’s guide for conservative investors), why has the wholesale price of power on the National Energy Market recently reached an all-time high?

 The current crisis isn’t a mystery: power prices haven’t surged despite the fact that renewables are cheapest; they’ve skyrocketed because “renewable” energy is intermittent, and unreliable sources of power are expensive. Taking their full cost – including their considerable external costs into consideration, today’s crisis should surprise nobody.

It doesn’t surprise Leithner & Company. Moreover, we welcome situations where “what everybody knows” is actually highly questionable. Members of a herd – including advisers, funds managers, journalists and others – don’t take the time and effort to think for themselves. Indeed, people seek safety in numbers precisely in order to let others do their thinking. This abdication eventually causes their clients and customers to suffer severe losses (see also Are you a customer, client or partner?).

“Experts” regularly succumb to hope (and hype) and thus misprice certain assets – sometimes egregiously. Their mistakes create attractive opportunities for conservative investors who’re prepared to conduct their own analyses, draw their own conclusions and when necessary, defy the conventional wisdom. Today’s energy sector in Australia and elsewhere is a prime example. 

The cost of electricity

Table 1 summarises the cost of electricity along two dimensions (a full elaboration appears in the full report (pdf) appended to this article). Reading along its left-hand column, the total cost of generation (and analogously for the other links of the chain) is the sum of fixed (capital) cost, variable (running) cost and external cost. Fixed (capital) costs are those that don’t vary according to the quantity of output; variable costs are those that do. Fixed and variable costs vary from one type of generation to another, and from one stage of production to another.

 Table 1: The Cost of Electricity – Two Dimensions

Many members of the public – egged by lazy journalists, venal politicians and self-interested promoters – mistakenly think: “sunshine and wind are costless; therefore renewable energy is (or should be) virtually free.” They simply ignore intermittent generation’s considerable capital and external costs.

Power generation’s external costs (“Externalities”)

The fixed and variable costs of generating and supplying electricity are hard enough to estimate. What’s most difficult is what’s seldom mentioned: identifying – never mind quantifying – its externalities. An externality (“external cost”) is an indirect, usually unquantified, sometimes undetected and thus almost always uncompensated cost to a third party that arises from another party’s (or parties’) activities. Pollution is probably the most commonly-cited example: its producers seldom pay its cost or otherwise compensate its “consumers” (which is often society as a whole).

Enthusiasts of intermittent energy vociferously denounce coal- and gas-fired electricity generation’s externalities – but furiously deny the external costs of solar and wind power. It’s seldom recognised but vital to acknowledge: the production of electricity from sunshine and wind generates substantial external costs.

Most notably, the intermittency of their output places considerable strains upon the grid. These strains are very costly to manage – and in extreme cases can cause blackouts – yet solar and wind generators incur none of the costs of their unreliability: electricity transmission and distribution firms, and ultimately consumers, do.

In short, the actual, total (including external cost) of renewable power is significantly greater than its market price.

The “levelised cost of energy”

Different methods of generating power incur different types of cost; their magnitudes and thus their total costs also vary considerably. Yet comparisons can paint a confusing and even misleading picture. Among the many complications: some technologies are more expensive to build but cheaper to run; conversely, others’ fixed (capital) costs are comparatively low but their variable costs are relatively high. As a result, small changes of any one of a large number of assumptions, such as discount rates, can greatly affect estimates of total costs. How, then, to compare them?

The levelised cost of energy (LCOE) compares total costs of different methods of generation on a consistent basis. The Independent Review into the Future Security of the National Electricity Market (2017), headed by the Chief Scientist at the time, Dr Alan Finkel, provides background and elaboration. It states: “LCOE is a measure of the average cost of producing electricity from a specific generating technology. It represents the cost per megawatt hour (MWh) of building and operating a generating plant in order to break even over (the plant’s) assumed financial life. Key inputs to calculating the LCOEs include capital costs, fuel costs, fixed and variable operating and maintenance (O&M) costs, financing costs, and assumed usage rates for each technology type. The LCOEs do not include transmission or distribution costs.”

Likewise, they typically exclude external costs. Conceptually and empirically, the LCOE is therefore hardly fool-proof. Yet it remains the best available metric for comparing the costs of various types of power generation.

Five recent studies 

Table 2 summarises key estimates from five recent studies that attempt to quantify the LCOEs of various types of power generation (for details, see pp. 7-10 of the full report). To varying extents, these studies disclose their assumptions and methods. From these disclosures it’s evident that their conclusions rest upon a variety of assumptions and complex analyses. Yet they also contain a crucial similarity:

All of them strive (for example, by estimating the effects of carbon taxes) to take into account the externalities of coal- and gas-fired power – but none attempts to “internalise” the external costs of solar and wind generation. As a result – and like governments’ policies – they punish fossil-fuels and privilege intermittent generation.

Table 2: Levelised Cost of Energy ($US per MWh), Five Studies and Nine Types of Generation

Similarly, these studies estimate the LCOEs of solar and wind on a “standalone” basis – that is, without the backup from batteries or gas that they often require.

Incorporating backup into the calculations doesn’t just greatly increase renewables’ LCOEs: it renders them uncompetitive. In 2020, Bloomberg New Energy Finance estimated that the LCOE in Australia of wind power backed by battery storage is $US87/MWh, and that the LCOE of solar power backed by battery storage is $US118/MWh. These “bundled” estimates are at least twice as large as the CSIRO’s “standalone” estimates.

Considered as a whole, these studies flatly contradict “what everybody knows”:

Despite the favourable treatment they receive from analysts and policymakers, “green” methods of generation aren’t demonstrably cheaper. The mean ranking of the four varieties of solar and wind is 6.0; the mean of coal-fired, gas-fired and the two varieties of nuclear power is 4.75. If anything, in aggregate the tried-and-true forms of generation are cheaper than the “trendy" ones.

Electricity’s price is innately volatile

Why has the wholesale price of electricity in the National Energy Market recently skyrocketed to an all-time high? Three points, which pp. 10-11 of the full report detail, explain it:

  1. Although the retail price of power in Australia is heavily regulated, its wholesale price is relatively unregulated.
  2. Consumers’ demand for power is inelastic. A sudden shortfall of supply relative to demand (as occurs, for instance, when the wind doesn’t blow) thus causes its price to skyrocket, and a surplus of supply causes it to plunge.
  3. The innate characteristics of electricity exacerbate its wholesale price’s volatility. Among the essential goods and services that underpin a modern economy, in one crucial respect electricity is unique: its consumption occurs almost instantaneously after its production. As a result, a sudden shortfall of supply relative to demand can cause the grid to become unstable; an unbalanced grid, in turn, can cause “brownouts” and outages.

The influx of “Cheap Renewables” can cause the wholesale price to skyrocket 

We’re now in a position to address two critical questions:

  1. If over the past decade the cost of generating renewable power has fallen greatly and renewables’ share of total power output has risen sharply, why has the wholesale price of electricity recently skyrocketed to an all-time high?
  2. More generally and over longer periods, can the increasing proportion of “cheap” renewables in the generating mix actually raise power’s wholesale price?

The analysis on pp. 11-13 of the full report concludes:

The rising incidence of “firming,” which is a consequence of the increasing role of “cheap” renewable in the mix, when combined with the imperative of balancing supply and demand and power’s inelasticity of demand, doesn’t just produce short-term upward spikes of wholesale prices: over longer periods, it also boosts wholesale prices. Hence renewables’ rising share of total generation can raise the wholesale price of power even as renewables’ LCOEs decline. In short, the growing incidence of intermittent generation increases the external costs it imposes upon the system.

Testing 3 hypotheses against Australian reality 

As a rough rule, the higher is intermittent generation as a percentage of total output, (1) the higher will be the level and (2) the greater will be the volatility of wholesale power prices. Over the past decade, intermittent generation has provided a rising share of power into the National Energy Market; that is, during the first decade of the century it provided little or no power, but since 2010 has contributed a rapidly-rising and now significant percentage (for details, see Decarbonisation: A doubter’s guide for conservative investors).

Throughout this period, this percentage has been highest in South Australia. According to data compiled by OpenNEM, “wind and solar met 60.2% of South Australia's electricity demand over the 12 months to 7 February 2021. Wind energy accounted for 42.2%, followed by rooftop solar (13.7%) and utility scale solar (4.3%).” Accordingly, and notwithstanding the sharp decrease of solar and wind generation’s LCOE but given its high and rising external costs, we can expect to observe that:

  1. the real (that is, CPI-adjusted) wholesale price of wholesale power was roughly flat in 2000-2010 but thereafter has risen;
  2. since 2010, the real wholesale price has become more volatile;
  3. since 2000, given its high share of intermittent generation, the price won’t be lower, could be higher and will be more volatile, in SA than elsewhere.

Figure 1, which plots the combined, CPI-adjusted average wholesale price of power per quarter in NSW, Queensland and Victoria, supports the first hypothesis (Q2 of 2022 covers the period 1 April-1 June). It groups these three states together because their grids are relatively well interconnected. As a result, their prices and volatility are similar: the mean and standard deviation in NSW are $43 and $26 respectively; in Queensland, they’re $44 and $31; and in Victoria they’re $41 and $29. Before 2010, the quarterly mean was mostly below its overall (23-year) mean; since then, it’s mostly been above it. The average price in the most recent quarter has spiked to an all-time high; other spikes occurred in Q1of 2001, Q1-Q3 of 2017 and Q1 of 2019.

Figure 1: The Wholesale “Spot” Price ($A per MWh) of Power, Mean for NSW, Queensland and Victoria per Quarter, CPI-Adjusted

Figure 2, which plots the standard deviation of wholesale prices’ average price per quarter over selected intervals, upholds – at least in NSW, Queensland and Victoria – the second hypothesis. Since 2012, the variability of prices in these states has risen dramatically. So has their dispersion in South Australia – yet the variation in 2007-2012 exceeded the one in 2022.

Figure 2: Standard Deviation, Wholesale Spot Price of Electricity ($A per MWh), Quarterly Data, CPI-Adjusted

Figure 3 compares these three states’ quarterly spot price of wholesale power to the corresponding average in South Australia. It shows that volatility in the form of short-term spikes has been much more numerous and severe in SA than in the eastern states. Indeed, the current spike in SA isn’t the severest: the one in Q1 of 2008 was much worse, and the one in Q1 of 2019 was only slightly worse. Because SA’s wholesale price has fluctuated so much for so long, it shows little upward trend.

Figure 3: The Wholesale Spot Price ($A per MWh) of Power, Mean for NSW, Queensland and Victoria per Quarter, CPI-Adjusted