At present, electricity cannot be stored economically. So production of electricity must match consumption on a real-time basis. As electricity demand increases in short-time periods (i.e. heat wave when everyone switches on their air conditioner), additional generation capacity must be brought online quickly but for relatively short periods of time. This results in some proportion of the capital stock (i.e. the existing generation fleet) being utilised for much lower periods of time than in other industries.
Based upon this unique feature of electricity markets, investors in new plant capacity consider the economics of the technology being developed against a backdrop of different growth trajectories across the various demand categories.
- Generation plant with relatively high capital costs but low operating costs is used to meet base load demand (demand that occurs for most of the time). Historically, black and brown coal generation which is slow to start or shut down has been used to meet base load demand and such plant typically operates at a 75–90 per cent annual capacity factor.
- Intermediate demand (nominally the higher ‘‘daytime demand’’) is generally met by plant with medium capital and operating costs and flexible operating capacity (i.e. can be ramped up quickly). Combined cycle gas turbine (CCGT) plant is generally used to meet intermediate demand and typically runs at an annual capacity factor of between 40 and 60 per cent.
- Generation plant with relatively low capital costs but high operating costs is used to meet peak demand (demand that only occurs on the hottest and coldest days of the year, or during power system contingency conditions such as unexpected plant outages). Open cycle gas turbine (OCGT) plant or hydro generation which can be ramped up very quickly is generally used to meet peak demand and typically operates at annual capacity factors of between 5 and 30 per cent.
In the wholesale electricity market, generators bid the quantity of electricity they are willing to supply and the price they want to receive for each dispatch interval (each interval is five minutes in length). Generators are dispatched in price order from lowest to highest up to the level required to match demand.
All dispatched generators receive the market-clearing price in their region, which is the price of the most expensive dispatched generator. The specific bids made by each generator are a combination of several factors including their short-run operating costs and plant technology. Gas-fired generators are usually quick to start and flexible but fuel costs (gas) are more expensive than coal.
Much of the debate in relation to renewable energy at the moment is based upon partial analysis. One of the critical points that people misunderstand relates to the ‘merit-order effect’. The proposition is simple: because renewables have very low short-run marginal costs (i.e. the wind or sun is free) then increasing the proportion of renewables in the mix reduces wholesale electricity prices.
However, what does the scientific literature say about this? Edenhofer et al (2013, p. 519) make the salient point: “lower average prices caused by higher renewable energy penetration lead to a reduction of overall capacity, which in turn increases the frequency of scarcity events and respective scarcity prices. According to theory this will bring the market back to the long-term equilibrium in which long-run average costs and average revenues are balanced for all capacities and where, as a direct result, the capacity level is efficient”. One of my own papers sums it up nicely by quoting Gelabert, Labandeira, and Linares 2011 p. S65: “the merit order effect is transient because the reduction in prices will in turn result in lower investments and therefore higher prices in future periods.”
My next blog post will talk about the ageing existing thermal fleet and the need for an orderly transition as our electricity system decarbonises. Ill also try and cover off how the contract market interacts with the wholesale market.
This article is based upon a number of my papers. You can find them at my Google Scholar profile.