Chapter 6: Electricity

Electricity is not a commodity in the traditional sense of the word. Although ubiquitous in our everyday lives, and uninterrupted access to it is a measure of prosperity and economic development, electricity was not stored, until recently, in commercially large enough quantities, and is traded cross-border only among a relatively small number of countries, when compared to energy commodities discussed earlier.

Yet, in the last two to three decades the deregulation of electricity markets in several economies in the Americas and Europe has led to substantial changes in the way this good is supplied, consumed and traded.

This chapter aims to complete the discussion on energy commodities and also provide the reader with a primer on the key industry characteristics on a global basis, as well as in selected key markets.

Physical characteristics

Electricity is a form of energy caused by the presence of electrical charges in matter. In an atomic nucleus, each proton carries a unit of positive electric charge and each electron circling the nucleus carries a unit of negative electric charge. Electric current is the movement of electrons through an electric conductor driven by differential concentrations of electrons that repel each other. This movement can create heat (e.g. to fire up a kettle or light a conventional lamp), or movement through electromagnetic action (e.g. to move a motor). Electricity moves in closed loops or circuits. When we flip a switch we close the circuit and allow the current to flow through, in order to produce a useful function.

For the total novice, it is worth listing a few key properties of electricity. An electric current is the rate at which electric charges flow through a circuit and is measured in amperes. Potential difference is commonly known as voltage, whereby one volt is the difference in electric potential across a wire when an electric current of one ampere dissipates one watt of power. Electrical power is the amount of work available from an electric current. It is defined as the product of voltage and current and measured in watts, whereby one watt is the work produced by a current of one ampere with a potential difference of one watt.

There are many different ways of producing electricity, but the resulting electrical currents are of two main types: direct (DC) or alternating (AC). A battery, or a photovoltaic cell, produces a steady voltage and a steady direct current. In contrast, a spinning electromagnet generates alternating electricity, whereby the voltage reverses its direction from positive to negative several times per second. In the early stages of the electricity industry, during the late nineteenth and early twentieth centuries, both DC and AC suppliers were competing, as each type had both advantages and disadvantages.[2] In the end AC supply prevailed, primarily because of the efficiency of being able to step up the voltage for long-distance transmission and then step it down again to a voltage appropriate for industrial and residential users of electricity. It is worth noting, however, that DC remains the favourable option for the electrical inter-connection between countries, as it does not have the problems of synchronising the frequency and phase of modern AC three-phase electricity. As a result, countries or regions wishing to exchange electricity convert their supply from AC to DC, transfer it through a high voltage direct current (HVDC) cable and then convert it back again to AC at the other end.

The physical attributes covered above are the bare essentials for understanding the economic fundamentals of electricity and the challenges posed by both demand and supply factors. Let’s start with demand.

Demand characteristics

Non-storability is the key characteristic of electricity, which sets it apart from all other commodities. Theoretically, the fuel used to generate electricity can be thought of as proxy storage. So a tank of natural gas or oil, a load of coal, a lake full of water or rods of uranium can all be thought of as stores of potential electricity. Yet, all of these fuels need to be transformed into electricity first and this can be done in varied amounts of time depending on the fuel itself. Electricity itself cannot be stored as such, except perhaps in batteries, which can be rather expensive and quite cumbersome.[3]

In modern societies where access to electricity is universal [4] we don’t have to think twice before flipping a switch to consume electricity at any time of the day or night. We expect the electricity grid to be able to provide adequate amounts of electricity to satisfy demand from industrial, commercial and residential consumers.

To demonstrate how this works from the point of view of the load on an electricity system, consider Exhibit 1 which shows the diurnal patterns of UK electricity demand during three days in the winter and one day in the summer of the same year, plotted over the course of 24 hours, in 48 half-hourly intervals. Note how demand is at its lowest from midnight to approximately 05:00 – this is known as baseload demand and from the exhibit it can be seen that it was at around 25GW.[5] As people wake up and start preparing to go to work and as industry and businesses start their daily production cycle, demand starts rising quite steeply until it levels off to ca. 40-45GW between 09:30-14:30. From about 15:00, many households begin their late afternoon routine, which typically involves switching high-consumption appliances (e.g. a cooker or an oven for the preparation of supper). In this case, there is also more need for lighting, as dusk starts falling during this time of the year. With these patterns repeating in more and more households, there is usually a peak in demand between 16:30-19:30, when demand reaches its highest point at ca. 45GW – this is known as peakload demand. Thereafter, demand falls steeply, as the various high-consumption household activities come to an end and people eventually go to sleep, by which time demand starts falling towards the baseload point. The pattern is repeated during the weekend, although it is worth noting that the ‘morning rush’ in demand takes place 1-2 hours later than during the week and the level-off plateau is also lower, as there is less requirement for electricity by commercial and industrial consumers.

If we collate the daily data over a period of time, we start getting a fuller picture of how demand varies during the year and from one year to the next, i.e. we can observe seasonal patterns and annual trends. Exhibit 2 demonstrates how the UK daily electricity load fluctuated between since January 2015. Note how the range of loads fluctuates during the year, with the higher levels occurring in December and January and the lower ones in July. 

Similar observations can be made in Exhibit 3, where the data series now refers to demand (measured in GWh) and is extended to the beginning of April 2005. We can also observe a longer-term trend of lower electricity demand across time, with demand peaks being progressively lower year after year, which is partly on account of factors like increased energy efficiency, less demand from industry and slower overall economic growth.

The next step in our analysis is to construct a cumulative distribution function (CDF) of demand loads over a period of time in order to gain better understanding of how much electricity is required for how much of the time. This is what is depicted in Exhibit 4 and is known as the load duration curve. The curve summarises over quarter of a million half-hourly observations of demand loads during the chosen period by clustering the loads into categories (1,000 MW each) and then counting the percentage of times that demand loads fall into each category. How is this graph useful in understanding electricity demand? By simply connecting points on the curve with the two axes we can ascertain how much electricity we need for how much of the time. For example, 60% of the time, the load is at least 30,000 MW. Conversely, 20% of the time the load is more tha 40,000 MW. From the same exhibit we can also observe that the minimum load required at all times is ca. 15,000 MW.

The load duration curve is a useful tool for planning purposes. It gives information about long-term demand requirements and can help an electricity network designer decide what types of generation to build in the system. For example, nuclear or coal generation may be used to cover demand up to a certain threshold, before using additional generation from other fossil fuels or renewables.

Supply characteristics

In 2024, the world produced over 30,000 TWh of electricity, as can be seen in Exhibit 5. The top two countries, China and the US, generated 47% of global electricity and if we add to them India, Russia and Japan the share goes to over 60%. A substantial amount of electricity (ca. 13%) was also produced in Europe. In contrast, the entire continent of Africa generated less than France and Germany put together, which are the first and second largest producers among European countries. The growth of electricity generation has been remarkable since 2000 and is shown in Exhibit 6, where the reader can note the strong growth of Asian economies. This growth in electricity generation is the major new energy trend which will mark the current and future decades, as more people have the desire to consume increased amounts of their energy in the form of electricity.

Where does it all come from, then? Electricity can be generated using a multitude of sources, the largest of which is currently fossil fuels. As demonstrated in Exhibit 7, ca. 60% of global electricity is generated using coal, gas and oil. They are followed by renewables, hydroelectricity (also renewable), nuclear power and finally all other sources. Let’s briefly review the range of available generation sources.

Coal is the most widespread generation fuel. Anything from lignite to high-quality bituminous (steam) coal can be used. Because of its lower calorific value, lignite tends to be used domestically in several countries, e.g. Germany (where it is being phased out) in Europe, China and Indonesia in Asia Pacific. As we have seen in chapter 5, steam coal is used extensively around the world and is in fact one of the largest commodities traded exactly for this reason: to be burnt in order to produce steam, which in turn generates electricity.

Natural gas is the cleanest of the fossil fuels used for generation.[6] In recent years, its use has expanded in North America, Western Europe and Asia Pacific, as numerous combined cycle gas turbine (CCGT) power stations have been built to take advantage of the increased efficiency of using gas and the positive externalities of reducing CO2 emissions.

Oil typically refers to a number of middle or heavy oil products, which can be burnt to produce electricity. In some cases this can be heavy gasoil (i.e. lower quality diesel, with a higher sulphur content), but typically it is heavy fuel oil, one of the least desirable and most plentiful products of the distillation process. Oil-fired capacity was reduced substantially after the first two oil crises, and a lot more electricity has been generated by coal instead. It is, however, still present in most countries around the world.

Nuclear refers to the use of the fission of heavy atoms, typically of uranium, to generate heat, which is then used to produce steam and eventually electricity. Nuclear capacity has declined since the Fukushima accident in March 2011 [7] and this is reflected in the reduced share of nuclear generation in Exhibit 7. Leading nuclear electricity producers include USA, France, Russia, S. Korea, China and Canada.

Hydro refers to hydroelectricity, which is the largest renewable source of generation. Leading hydroelectricity producers include China, Brazil, Canada, USA, Russia and Norway.

Renewables refers to the long list of generation sources which are classed either as inexhaustible or can be replenished regularly, with the exception of hydroelectricity which is listed separately. These include: wind (offshore and onshore), solar (photovoltaic and concentrated thermal), geothermal, biomass, tidal, wave, post-use waste and several more.

The extent to which the different sources of generation are used to provide shares of the total electricity produced in a country or region is referred to as the fuel mix. The choice of fuel mix is a mixture of strategic decisions (e.g. the use of nuclear and renewable generation) and market-based criteria (e.g. the use of fossil fuels, depending on domestic availability and import prices). Exhibit 8 depicts the development of electricity generation by fuel and how this mix has changed over the last two decades. The most notable change has been the surge in renewable generation, followed the strong growth of natural gas and the stagnation of coal and most other sources of generation.

Exhibits 9 and 10 use the US as an example of fuel mix and how electricity has been generated there since the year 2000. In Exhibit 9, note how generation has stagnated since 2005, with a dip in 2009, and also how natural gas has increased its share at the expense of coal, as a result of the advent of US shale gas. China’s case is shown in Exhibits 11 and 12, where we can observe how electricity demand has surged from 2000 onwards and how coal still plays a major role as generation fuel of preference, due to its lower cost. It is also notable, however, that hydroelectricity, wind and solar generation have also made considerable gains since 2011.

It is evident from the previous discussion that electricity networks have several choices as to the range of fuels they may use, as well as how frequently they may use each fuel in the course of a typical day. The latter is due to the fact that different sources of generation have different technical and operational characteristics.

For example, a nuclear plant requires a large initial capital expenditure, but the subsequent operating cost of producing a megawatt-hour of electricity is relatively small. In addition, a nuclear plant normally operates throughout the day, as it is not flexible to shut it down and restart it very quickly. The same can be said for a coal-fired plant and as a result both coal and nuclear generation are used to cover baseload electricity demand throughout the year. Gas and oil-fired plants, on the other hand, are relatively more straightforward to build and incredibly flexible in producing electricity at very short notice (in just a few minutes), but they also have higher operating costs, due to the price of the fuel they use. As a result, they tend to be used to cover high-demand periods or, in the industry parlance, peak shaving.

This relationship may change if the economic fundamentals change. An example of this is mentioned above, whereby low-priced, abundant shale gas has displaced coal in baseload generation in the US. In contrast, the relatively cheaper coal exports from the US to Europe have incentivised the use of the relatively dirtier fuel in the continent, at the expense of gas whose oil-indexed price makes it more expensive.

Finally, renewable electricity is treated as a ‘must-have’ generation, i.e. any amount that can be produced is sent to the electricity network and it is only limited by the operational efficiency of the renewable source (e.g. whether ‘the wind blows’ or ‘the sun shines’).[8]

From the discussion above it is evident that not all sources of generation produce electricity all the time and different sources are used at different points during the day to satisfy varying levels of demand. This variability in the utilisation of a particular plant is known as capacity factor.

The concept is quite simple: the capacity factor is the amount of electricity that a plant produces over a period (say a year), divided by the maximum theoretical or operational capacity that the plant has during the same period. The simple formula used to calculate the capacity factor is

CF = [MWh produced in period t] ÷ [Max capacity MW x hours in period t]

Consider the following example: a coal plant has a nameplate capacity of 30MW and during the course of a year it has produced 185 GWh of electricity. Its annual maximum nameplate capacity is 30MW x 365 days x 24 hours = 262,800 MWh = 262.8 GWh. Hence the capacity factor is simply 185 ÷ 262.8 = 70.4%. If we know that the plant had to shut down for ten 24-hour periods for maintenance, then its annual maximum operating capacity would be 30MW x 355 x 24 = 255.6 GWh, in which case its operating capacity factor is 185 ÷ 255.6 = 72.4%.

[8] Occasionally, the network operator may ask renewable generators to stop producing if there is an oversupply of electricity, which may result in system overload and physical damage to the network infrastructure.

The supply chain of electricity

So far we have only talked about generation of electricity. However, there are several more links in the supply chain that connects the power stations with the final consumers. The electric current produced by most generators is AC [9], but the voltage needs to be stepped up in order to transmit it efficiently over long distances. The network of high-voltage cables and pylons is known as the grid. Electricity is transmitted through the grid at around 400kV [10], but before it can be distributed to industrial, commercial and residential consumers its voltage has to be stepped down. This is done through a series of substations, until eventually the voltage comes down to the familiar 220-240V or 100-120V which is used by households and offices all over the world. In addition to these essential links, some electricity networks also have interconnections within the geographic borders of a country, or across the border linking two or more countries. Cross-border interconnectors are usually large HVDC cables, in order to bypass the technical issues of the difference in voltage and frequency that may exist among neighbouring countries.

It is still quite common in many countries for the entire network to be owned, operated and managed by one state-owned power company, which also meters the electricity at the consumption point and invoices the customers. Since the 1990s, however, electricity markets in several economies have been deregulated and privatised. This development was initiated in the 1980s in South America and subsequently spread to the US, as well as the UK and many more European countries.

This progressive deregulation usually began with the introduction of independent power producers (IPPs, often generators of renewables) who were allowed to enter long-term power purchase agreements (PPAs) with the state utility. The next development was the break up of the state utility into autonomous units, in preparation for their eventual privatisation. This process came to be known as unbundling and was popularised in the UK in the 1990s.

The result of this unbundling was the separation of the supply chain into individual stages and the introduction of competition at each and every stage. Hence:

Generation consists of several generating companies, each with one or more power plants (e.g. coal, gas, nuclear, oil, wind, solar, hydro and so on), which compete with each other to supply various tranches (e.g. baseload or peakload) of electricity.

Transmission consists of one or more independent system operators (ISOs) [11], whose role is to maintain and modernise the high-voltage grid and provide third-party access to both generators and suppliers. See Exhibit 14 for an example.

Distribution refers to the companies which own and operate the sub-stations and the cables (colloquially known as the wires) which bring the low voltage electricity to wholesale and retail consumers.

Balancing demand and supply

So far we have observed how demand fluctuates during the course of a day, week, month of year and also how supply can come from a variety of different sources, which may (or may not) be owned and operated by the same financial entity. Given the non-storability of electricity and the fact that it has to be produced the very second it is demanded, how are supply and demand balanced?

On a daily basis, all participants in the electricity supply chain have a forecast of the diurnal demand pattern for the day ahead, i.e. the one-day forward market. This forecast may be based on past historical data and may also be informed by anticipated events which may create demand peaks, e.g. an anticipated sharp drop in temperatures which is likely to generate additional demand for electrical heating. The forecast usually splits the day into time periods or intervals, with different levels of expected demand for each period. These periods may last for a few hours, a single hour, half-an-hour or even shorter. In most European countries, for example, the interval is hourly, whereas in the UK it is half-hourly.

Every day, each generator (e.g. a nuclear plant, a coal power station or a large wind farm) is prepared to offer a certain amount of electricity for each hourly or half-hourly interval for the next 24-hours (day-ahead market). Likewise, wholesale buyers (e.g. a regional supplier, or a large industrial consumer) put bids for quantities of electricity they will require. Buyers and sellers of electricity can transact on a bilateral basis and notify the ISO, which is ultimately responsible for maintaining system integrity.

Given the multitude of generation sources, how is the decision made whose electricity to use first? This issue is resolved using a very basic premise: the lowest-cost generation source dispatches first until its maximum is reached, before the next, more costly, source is called upon, until the demand at any given moment is satisfied. As a result, there is an ordering of generation sources according to their cost, starting from the cheapest and ending with the most expensive.

This is known as merit order dispatch, and Exhibit 15 demonstrates an example of this. What is shown here is a snapshot of the demand-supply balance of a network at a particular point in time, e.g. during a particular hourly interval during the day. During this interval, demand is determined by the collective activities of consumers and it is inflexible, hence the demand curve is vertical. The supply curve is a staggered line, created by ordering the generators from the least to the most expensive. Right at the bottom of the merit order are generators which are designated as ‘must-run’. This may be a renewable generator, e.g. a wind or solar farm, which are guaranteed to sell their entire production. Generators Nuclear A to Gas CC B are the next ones to be called upon. Each of them produces varying amounts of electricity, depending on their capacity, and these are represented by the width of the bar. The height of the bar indicates their cost or the price at which they are prepared to offer their electricity for. As can be deduced from the exhibit, the last and most expensive generator that needs to be called upon in order to match demand (in this case Gas CC B), is also the one that sets the marginal price for the particular time interval. As we move through the day, demand changes and so the vertical demand curve will shift to the right if demand increases, or to the left if it decreases. In the former case additional, more expensive, generators will be brought in, whereas in the latter case some generators will be asked to rump down production or switch off altogether.

Occasionally, a supply disruption or a system constraint may require the ISO to take off one generator and replace it with a higher-cost one. In Exhibit 16 we can see what happens when generators Coal A and Gas CC A have to come off the grid and are replaced by Pumped hydro and Gas Open, pushing the marginal price up as well. 

As the reader may have noticed from the discussion above, the ISO has a crucial role in ensuring that demand and supply balance every second of the day and the electricity system remains stable at all times. In a deregulated, competitive electricity market, where multiple supply and demand units engage in bilateral transactions, the ISO has continuous oversight of real-time demand requests and full access to generators from whom it can request additional generation when necessary. In the UK, for example, the counterparties can engage in bilateral transactions up to one hour before the beginning of each half-hourly interval. This point in time is known as gate closure, after which the balancing mechanism kicks in, which is overseen by the ISO. Units large enough to be part of the balancing mechanism submit series of bids (if they are consumers) or offers (if they are generators) that the ISO can call against. So if there is an unexpected demand surge, or a supply failure, the ISO can request plants with hot spinning reserve [12] to ramp up production in order to maintain system stability.

Conclusion

In this chapter we examined the economic fundamentals of electricity. Although not a traditional energy commodity, electricity has become the focus of public attention since the 1980s, as the industry has been deregulated and at least partly privatised in several countries around the world. Electricity is traded competitively between generators and wholesale consumers within countries and increasingly across national borders. Two of the three main fossil fuels, coal and natural gas, are predominantly used for electricity generation, hence a discussion of this ‘secondary’ form of energy was necessary.

In the next chapter we will discuss some of the principles of assessing the cost of renewables and other forms of generation, as well as the two key 'modern' renewables: solar and wind. A full discussion of all renewable and alternative forms of generation is beyond the scope of this textbook. The reader is referred instead to the excellent work by Peake (2018), Everett et al. (2012) and Kaltschmitt, Streicher & Wiese (2007), who provide extensive discussion of the technical and economic aspects of renewable sources of energy.

References

Everett, B. et al. (Eds.) (2012). Energy Systems and Sustainability: Power for a Sustainable Future. 2nd ed. Oxford University Press. 

Harris, C. (2006). Electricity Markets: Pricing, Structures and Economics. Wiley. Chichester.

IEA (2023). Electricity Market Report 2023. https://www.iea.org/reports/electricity-market-report-2023

_____ (2022). Electricity Sector. https://www.iea.org/reports/electricity-sector

Kaltschmitt, M., Streicher, W. & Wiese, A. (2007). Renewable Energy: Technology, Economics and Environment. Springer. Berlin, Heidelberg.

Peake, S. (Ed.) (2018). Renewable Energy: Power for a Sustainable Future. 4th ed. Oxford University Press.