Skip to main content
Log in

The pass-through rates of carbon costs on to electricity prices within the Australian National Electricity Market

  • Research Article
  • Published:
Environmental Economics and Policy Studies Aims and scope Submit manuscript

Abstract

This paper statistically investigates the interaction between a carbon price signal and wholesale electricity spot prices within the Australian National Electricity Market (NEM). While prior studies in Australia have been mainly conducted based on the theoretical and simulation methods, this paper examines the issue by employing an empirical analysis using daily data from July 2010 to October 2013. The findings reveal that carbon costs would indeed be fully passed on to wholesale electricity spot prices resulting in higher electricity prices for consumers and potential windfall profits for some generators.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1

Similar content being viewed by others

Notes

  1. The CPM, renewable energy, energy efficiency, and action on the land were the four key elements of the plan. Since the CPM was seen as the core of the climate change plan, this paper concentrates on the particular impact of the CPM, while the other three key elements are left for future research.

  2. The annual caps for the first 5 years of the ETS flexible price period were to be announced by 31 May 2014, with the price determined by the market. Besides domestic reductions through the ETS for energy-intensive sectors, the plan had also considered emission reductions abroad by linking to the European emissions trading scheme (EU ETS) and the Kyoto flexible mechanisms as key tools to increase the cost-effectiveness of fulfilling its obligations. These links would allow Australian firms covered by the ETS to meet up to 12.5 % of their liabilities using Kyoto units, and up to 50 % of their liabilities using European allowances (EUAs) to ensure significant domestic reductions. Moreover, the price ceiling as a safeguard was planned to be set at $20 above the expected EUA for 2015–2016 and set to rise by 5 % in real terms in 2017 and 2018. However, the new Liberal-National coalition government repealed the carbon tax on 17 July 2014 and plans to commence implementation of its Direct Action Plan (DAP). The DAP will build on the Carbon Farming Initiative and includes an Emissions Reduction Fund (Clean Energy Regulator 2013). At the time of writing, detailed information on this plan is not yet available, but will eventually be accessible at www.environment.gov.au.

  3. Approximately, 85 % of Australia’s electricity consumption is fuelled by coal-fired electricity generation with the emission intensity of 1. Gas-fired plants account for around 8 % of electricity generated, while less than 6 % of output is from hydroelectric generation (ABARES 2011b).

  4. In Australia the financial year starts on 1 July and ends on 30 June of the following year. For example, the 2013 financial year started on 1 July 2012 and ended on 30 June 2013.

  5. Various definitions of the pass-through rate have been proposed in several studies [see, for example, Kate et al. (2005) and Stennek and Verboven (2001)].

  6. See, for example, Frontier Economics (2006); Reinaud (2007); Sijm et al. (2008); Chen et al. (2008); Fell (2008); Fezzi and Bunn (2009); Frontier Economics (2010); Nazifi and Milunovich (2010); Lise et al. (2010); Trueck and Cotton (2011); Fabra and Reguant (2013).

  7. See, for example, ROAM Consulting (2008); McLennan Magasanik Associates (2008); Simshauser (2008); Simshauser and Doan (2009); Nelson et al (2010a); Wild et al. (2012).

  8. It is worth mentioning that relative fuel price change or price signal is only one of a number of elements that could enhance the competitiveness of low carbon technologies to generate electricity. Any discussion of a technological shift would need to take into consideration many other factors that may influence switching activity, such as the variability in plant efficiency, technical factors, environmental regulations and policies (for instance, the Renewable Energy Target scheme) and other specificities of the power market, to fully investigate this change and the extent to which each of these factors is likely to contribute to this shift (IEA 2013). The current paper statistically measures the carbon costs pass-through rates to see if the CPM could further incentivise generators to switch to carbon-efficient technologies and thereby assess the CPM’s likely effectiveness in terms of its contribution to lowering greenhouse gas emissions. The present analysis applies daily observations for the volume-weighted average emission intensity for each state to calculate the costs of the carbon required to generate one MWh of electricity to estimate the CPTRs. However, to analyse precisely whether this technological shift has occurred, further analysis based on average intensity data is needed. According to existing statistics, after the introduction of the CPM, the emission of CO2 from electricity within the NEM fell (by approximately 7.7 %); however, several studies have reported that this reduction can be mainly attributed to the factors unrelated to the carbon tax [see, for detailed information, Frontier Economics (2013) and Quarterly Update of Australia’s National Greenhouse Gas Inventory (2013)]. For example, the shutdown of a substantial amount of brown coal generation in Victoria due to a mine flood has led to a significant reduction in CO2 emissions in this sector (Frontier Economics 2013). Furthermore, the Energy Users Association of Australia (EUAA) has reported that no evidence could be found to support the claim that the CPM has lowered emissions in stationary energy to any meaningful extent (EUAA 2013).

  9. Under the previous government, the Clean Energy Regulator was set up to provide financial assistance to the affected coal-fired electricity generators (industry assistance programs) in adjusting to the carbon price as compensation for the introduction of the CPM. There were three types of assistance: (1) the issue of 41.705 million free carbon units each year from 2013 to 2017; (2) $1 billion of cash payments distributed to eligible generators in June 2012; and (3) financial support to close highly emissions-intensive power stations. All coal-fired generators with an emissions intensity greater than 1 that supplied electricity to a main grid during the period 1 July 2008 to 30 June 2010 were eligible for assistance. However, under the new government, the Emissions Reduction Fund is the centrepiece of the Direct Action Plan, intending to purchase low-cost abatement through reverse auctions.

  10. While the conventional meaning of “windfall profits” refers to just the value of economic rent derived from some fixed asset such as free allowances, in this paper, in the context of a CPM and the power sector, the term “windfall profits” broadly refers to the changes in the profits of existing power generators due to carbon-induced changes in power prices, production costs and sales volumes (Chen et al. 2008). For instance, if the power price is set by a coal-fired plant, given that the cost of carbon is fully passed to the power price (while sales volumes do not change), some generators may benefit from the implementation of the CPM (depending on the fuel generation mix of their installations) as the increase in their revenue may exceed the increase in their costs. Furthermore, the availability of financial assistance to electricity generators under the industry assistance programs (such as the provision of the free allocation of carbon units to eligible generators by the Australian government as discussed above) could increase the potential for windfall gains by generators.

  11. However, it is important to note that in this paper, due to the assumption of a fixed efficiency rate, changes in power prices cannot be attributed to changes in technology which could result in over/underestimation of CPTRs (a detailed discussion on this point can be found in Sect. 4). The impact of the CPM on possible technological changes in generating electricity is left for future research.

  12. Over the past three years, the national average residential retail electricity price has risen by approximately 78 % (AEMC 2013). However, it would not be appropriate to attribute this increase in prices merely to the implementation of the CPM. Australian retail prices are affected (in addition to wholesale electricity prices) by a number of factors such as network costs (including transmission and distribution costs), capital costs, electricity demand, regulation of network businesses, and environmental policies. Any analysis on retail prices would need to take into account these factors and the extent to which each of these major drivers will actually translate into retail prices. Currently, network costs make up approximately 52 % of the national average retail electricity prices, and increases in network costs have significantly contributed to higher retail prices (AEMC 2013). However, the assessment of the carbon costs pass-through rate in this analysis is based on modelling the wholesale electricity spot prices due to the availability of daily data, while the impact of the CPM on the retail electricity market is left for future research.

  13. According to Sijm et al. (2012), this market structure refers primarily to the interaction among the number of firms (as an indicator of the level of market competitiveness or market concentration), the shape of the demand curve, and the shape of the supply curve.

  14. The average emission intensity in Europe is around 0.35; in the US is 0.54; and in Canada about 0.18 tCO2/MWh.

  15. In the power sector, with multiple generators having fixed capacities and different carbon intensity factors, generators can be ranked in ascending order of their short-run marginal costs, known as merit order. So generators with the lowest marginal costs are the first ones to meet demand, and plants with the highest marginal costs are among the last to be brought on line to meet demand.

  16. See, for example, Linares et al. (2006); Sijm et al. (2006); Reinaud (2007); Nordic Council of Ministers (2008); Sijm et al. (2008); Kara et al. (2008); Fell (2008); Fezzi and Bunn(2009); Nazifi and Milunovich (2010); Gronwald et al. (2011).

  17. See, for example, McLennan Magasanik Associates (2006); Simshauser and Doan (2009); Nelson et al. (2010a, 2010b); Kim et al. (2010); Simshauser et al. (2011); Wild et al. (2012).

  18. IndexMundi is a data portal that gathers facts and statistics from multiple sources (more information can be found at: http://www.indexmundi.com/about.html).

  19. The AEMO’s definition of peak hours per day is the period between 07:00 and 22:00 EST.

  20. Following the 2012 IEA report, in this study, the standard generic efficiency rates of 35 and 40 % are assumed for coal-fired and natural gas-fired generation, respectively, regardless of the state in which these technologies operate (IEA 2012). Furthermore, Sijm et al. (2008) use the same rates to analyse the interaction between power, fuel and carbon markets.

  21. The results of unit root tests with structural breaks do not change significantly by allowing for a break in either intercept or trend and are in favour of no breaks in series, and hence not reported. The findings are available upon request.

  22. However, it should be emphasised that in finite samples, the unit root test procedures are known to have limited power against alternative hypotheses and this problem is particularly severe for small samples (see Campbell and Perron 1991). Thus, these results should be used with caution.

  23. It should be noted that small samples could lead to questions about the statistical reliability of results produced by Granger causality tests. Dolado and Lütkepohl (1996) found that in high dimensional (integrated–cointegrated) VAR systems with a small true lag length, the significant reduction of power of the Granger causality test may occur, especially for small samples. In this case, the assumption of standard asymptotic distribution may often cause significant over-rejection. Application of the residual-based bootstrap techniques usually improves the power performance of causality tests. They have also shown that standard asymptotic properties of the Wald test can be ensured by overfitting a VAR process whose order exceeds the true order, as pointed out by Toda and Yamamoto (1995). In this paper, the Granger causality test has also been conducted by overfitting a VAR in levels. The results are similar in sign and statistical significance and hence not reported. They are available upon request.

  24. It is important to note that a major characteristic of electricity spot prices is that they are strongly event-driven and, hence, are more volatile than the forward markets. For example, unexpected plant outages or demand hikes because of hot weather or other factors could cause significant fluctuations in prices of spot markets. However, due to the lack of data and analytical tools, it is often not possible to account for the impact of these events and factors on spot prices in a quantitative way, which may lead to overestimation or underestimation of the CPTRs. Thus, as discussed earlier, concerns could arise regarding the robustness or exactness of the estimated CPTRs, as these estimates may be biased and, hence, inconclusive due to the incidence of other factors besides fuel/carbon costs affecting electricity spot prices. However, in this paper, following Sijm et al. (2008), to account for some demand-related events on electricity spot prices, a demand parameter was included in the regression (detailed discussion can be found in Sects. 4.1 and 4.2). It is important to stress that the findings still must be interpreted prudently. Nevertheless, the author still believes that the model would offer some useful insights with regard to the actual impact of the carbon price on wholesale electricity spot prices in comparison with previous studies. The reasoning behind this is that previous studies have simulated the impact of the carbon price using data with similar deficiencies rather than statistically evaluated the actual impact. For example, Wild et al. (2012) ignores outages in thermal generators for the simulation analysis, and Nelson et al. (2010b) focus on the emission intensities.

  25. An OLS regression analysis is also applied where wholesale electricity spot price (P e) remains in the left hand side of the equation and the fuel price (P f) moves to the right hand side of the equation as an independent variable. The results are similar in sign and statistical significance and are reported in Table 4 in the Appendix.

  26. From a statistical point of view, the estimated b 1 might be overestimated/underestimated due to fundamental assumptions of the model which ignore the impacts of other factors on the wholesale electricity spot prices. However, in practice, as discussed earlier in footnote 10, it could be interpreted as a potential windfall for some generators depending on the fuel generation mix of their installations and the extent of the free allocation of carbon units. Furthermore, given that the Australian Financial Markets Association’s (AFMA) addendum allows generators and retailers to add a carbon price to the contract price, this, in turn, could increase the possibility of windfall profits and “double dipping” by generators and retailers. “Double dipping” here refers to the potential ability of generators and retailers to pass through the carbon cost for every MWh sold, which is in addition to the increased wholesale electricity price in which the carbon price is already incorporated (for more detailed discussion, see the AFMA Carbon Benchmark Addendum 2012).

  27. In addition, for the sake of completeness of the analysis, vector autoregressive (VAR) processes are also used to statistically assess the impact of carbon prices on wholesale electricity spot prices. The VAR models are performed with optimal number of lags (2) chosen by the Schwarz criterion (SC). The estimated CPTRs for each state (based on the VAR models) are reported in Table 5 in the Appendix.

  28. The OLS regression analysis is also applied on the first differences of log prices. The results are similar in sign and statistical significance. They are available upon request.

  29. According to the economic theory, the CPTR is expected to be 1 in the case of a perfect competition market and fixed demand with zero price elasticity (Sijm et al. 2008). Moreover, when the supply is perfectly elastic (i.e. flat line), the pass-through is 100 %.

  30. Since the carbon cost per unit of production is not similar for each generation technology, the merit order of power generation technologies may shift over time by the introduction of the carbon price, depending on the dynamics and interaction of the actual carbon and fuel costs of the plants impacting electricity prices.

  31. The OLS regression analysis is also applied on price series after eliminating these observed price spikes. The results are similar in sign and statistical significance. In this case, the estimated CPTR for QLD decreases to 1.18. They are available upon request.

  32. In Victoria, the demand for electricity is primarily met by brown coal-fired generators which are the most carbon-intensive plants (ABARES 2011b).

References

  • ABARES (2011b) Energy in Australia. Department of Resources Energy and Tourism. ABARES publications

  • ABARES (Australian Bureau of Agricultural & Resource Economics & Sciences) (2011a) Australian energy statistics, energy update. ABARES publications

  • AEMC (Australian Energy Market Commission) (2013) Residential electricity price trends. Final Report, EPR0036, December 2013

  • AEMO (2012) AEMO annual report. Australian Energy Market Operator reports

  • AEMO (2013b) Carbon price–market review. Australian Energy Market Operator reports

  • AEMO (2013c) Market event reports. Australian Energy Market Operator reports

  • AEMO (2013d) South Australian electricity reports. Australian Energy Market Operator reports

  • AEMO (Australian Energy Market Operator) (2013a) Carbon dioxide equivalent intensity index. Australian Energy Market Operator data

  • AFMA (the Australian Financial Markets Association) (2012) The AFMA carbon benchmark addendum. The AFMA report, electronic copy available at: http://www.afma.com.au/afmawr/_assets/main/lib90047/ca%20website%20note%20v4.pdf

  • Campbell J, Perron P (1991) Pitfalls and opportunities: what macroeconomists should know about unit roots. In: Blanchard O, Fischer S (eds) NBER Macroeconomics Annual. MIT Press, Cambridge

    Google Scholar 

  • Chemarin S, Heinen A, Strobl E (2008) Electricity, carbon and weather in France: where do we stand. Ecole Polytechnique Cahier 2008–04

  • Chen Y, Sijm J, Hobbs BF, Lise W (2008) Implications of CO2 emissions trading for short-run electricity market outcomes in northwest Europe. J Regul Econ 34:251–281

    Article  Google Scholar 

  • Clean Energy Regulator (2013) Carbon pricing mechanism. Australian Government Clean Energy Regulator publication, Australia

    Google Scholar 

  • Consulting ROAM (2008) Modelling of carbon pricing scenarios. ROAM Consulting Publication, Brisbane

    Google Scholar 

  • Dickey D, Fuller W (1979) Estimation for autoregressive time series with a unit root. J Am Stat Assoc 74:427–431

    Google Scholar 

  • Dolado J, Lütkepohl H (1996) Making Wald tests work for cointegrated VAR systems. Econometrics Rev 15(4):369–386

    Article  Google Scholar 

  • EUAA (Energy Users Association of Australia) (2013) The impact of emission prices on electricity in the NEM. EUAA Report in June 2013

  • Fabra N, Reguant M (2013) Pass-through of emissions costs in electricity markets. NBER Working Paper No.19613, The National Bureau of Economic Research

  • Fell H (2008) EU ETS and Nordic Electricity. Discussion paper, Resources for the Future, RFF DP 08-31

  • Fezzi C, Bunn DW (2009) Structural interactions of European carbon trading and energy prices. J Energy Mark 2(4):53–69

    Google Scholar 

  • Frontier Economics (2006) The role of CO2 in power markets- in line with competition. paper by Riechmann C, Etten M and Elms N, London

  • Frontier Economics (2010) Energy purchase costs, a final report prepared for IPART. Frontier Economics Publication, Australia, electronic copy available at: www.frontier-economics.com

  • Frontier Economics (2013) Overpowering. Frontier Economics Publication, Australia, electronic copy available at: http://www.frontier-economics.com.au/documents/2014/06/overpowering.pdf

  • Garnaut R (2008) Garnaut climate change review. Cambridge University Press, Melbourne

    Google Scholar 

  • Gronwald M, Ketterer J, Trueck S (2011) The relationship between carbon, commodity, and financial markets: a Copula analysis. Econ Rec 87:105–124

    Article  Google Scholar 

  • IEA (2010) World energy outlook. International Energy Agency Publication, Paris, electronic copy can be found at www.worldenergyoutlook.org

  • IEA (2012) Energy Statistics of OECD Countries. International Energy Agency, electronic copy can be found at www.iea.org.statistics

  • IEA (International Energy Agency) (2013) Gas to coal competition in the U.S. power sector. International Energy Agency, Insights Series 2013

  • Kara M, Syri S, Lehtila A, Helynen S, Kekkonen V, Ruska M, Forsstrom J (2008) The Impacts of EU CO2 emission trading on electricity markets and electricity consumers in Finland. Energy Econ 30(2):193–211

    Article  Google Scholar 

  • Kate A, Niles G (2005) To what extent is cost saving passed on to consumers? an oligopoly approach. Eur J Law Econ 30(2):193–211

    Google Scholar 

  • Kim W, Chattopadhyay D, Park J (2010) Impact of carbon cost on wholesale electricity price: a note on price pass-through issues. Energy 35:3341–3448

    Google Scholar 

  • Kwiatkowski D, Philips P, Schmidt P, Shin Y (1992) Testing the null of stationary against the alternative of a unit root: how sure are we that the economic time series have a unit root? J Econometrics 54:159–178

    Article  Google Scholar 

  • Linares P, Santos FJ, Ventosa M, Lapiedra L (2006) Impacts of the European emissions trading scheme directive and permit assignment methods on the Spanish electricity sector. Energy J 27(1):79–98

    Article  Google Scholar 

  • Lise W, Sijm J, Hobbs B (2010) The impact of the EU ETS on prices, profits, and emissions in the power sector: simulation results with the COMPETES EU20 model. Environ Res Econ 47:23–44

    Article  Google Scholar 

  • McLennan Magasanik Associates (2006) Impacts of a national emissions trading scheme on Australia’s electricity markets. MMA Publication, Sydney

    Google Scholar 

  • McLennan Magasanik Associates (2008) Impacts of the carbon pollution reduction scheme on Australia’s electricity markets. MMA Publication, Sydney

    Google Scholar 

  • Nazifi F, Milunovich G (2010) Measuring the impact of carbon allowance trading on energy price. Energy Environ 21(5):367–382

    Article  Google Scholar 

  • Nelson T, Kelley S, Orton F, Simshauser P (2010a) Delayed carbon policy certainty and electricity prices in Australia. Econ Pap 29(4):446–465

    Article  Google Scholar 

  • Nelson T, Orton F, Kelley S (2010b) The impact of carbon pricing on Australian deregulated wholesale electricity and gas markets. AGL Applied Economics and Policy Research Working Paper No. 23

  • Nelson T, Kelley S, Orton F (2012) A literature review of economic studies on carbon pricing and Australian wholesale electricity markets. Energy Policy 49:217–224

    Article  Google Scholar 

  • Nordic Council of Ministers (2008) Relationship between CO2, fuel and electricity prices and the effect on greenhouse gas (GHG) emissions in Nordic Countries. Report to Nordic Council of Ministers

  • Quarterly Update of Australia’s National Greenhouse Gas Inventory (2013) The Department of Industry, Innovation, Climate Change, Science, Research and Tertiary Education, P.2. ISSN 2201-8883

  • Reinaud J (2007) CO2 allowance and electricity price interaction, impact on industry’s electricity purchasing strategies in Europe. IEA Information Paper, International Energy Agency, OECD/IEA

  • Said S, Dickey D (1984) Testing for unit roots in autoregressive-moving average models of unknown order. Bionmetrika 71(3):599–607

    Article  Google Scholar 

  • Saikkonen P, Lütkepohl H (2002) Testing for a unit root in a time series with a level shift at unknown time. Econometric Theory 18:313–384

    Article  Google Scholar 

  • Sijm J, Neuhoff K, Chen Y (2006) CO2 cost pass-through and windfall profits in the power sector. Clim Policy 6:49–72

    Article  Google Scholar 

  • Sijm J, Hers S, Lise W, Wetzelaer B (2008) The impact of the EU ETS on electricity prices. Final report to the European Commission (DG Environment), Energy Research Centre of the Netherlands, ECN-E-08-007, Petten/Amsterdam

  • Sijm J, Chen Y, Hobbs BF (2012) The impact of power market structure on CO2 cost pass-through to electricity prices under quantity competition: a theoretical approach. Energy Econ 34(4):1143–1152

    Article  Google Scholar 

  • Simshauser P (2008) On emission permit auction vs. allocation and the structural adjustment of incumbent power generators in Australia. Electr J 21(10):30–41

    Article  Google Scholar 

  • Simshauser P, Doan T (2009) Emissions trading, wealth transfers and the wounded bull scenario in power generation. Australian Econ Rev 42(1):64–83

    Article  Google Scholar 

  • Simshauser P, Doan T, Lacey B (2007) The outlook for the economic and environmental performance of Australia’s national electricity markets in 2030. Electr J 20(6):58–75

    Article  Google Scholar 

  • Simshauser P, Nelson T, Doan T (2011) The boomerang paradox part I: how a nation’s wealth creates fuel poverty- and how to defuse the cycle. Electr J 24(1):72–91

    Article  Google Scholar 

  • Stennek J, Verboven F (2001) Merger control and enterprise competitiveness: empirical analysis and policy recommendations. European commission Directorate-General for Economics and Financial Affaires 5:129–194

  • Toda H, Yamamoto T (1995) Statistical inference in vector autoregressive with possibly integrated process. J Econometrics 66(1–2):225–250

    Article  Google Scholar 

  • Treasury (2011) Strong growth, low pollution, modelling a carbon price. The 2011 Treasury report, electronic copy available at: http://www.archive.treasury.gov.au/carbonpricemodelling/content/report.asp

  • Trueck S, Cotton D (2011) Interaction between Australian carbon prices and energy prices. Australasian J Environ Manag 18(4):208–222

    Article  Google Scholar 

  • Wild P, Bell PW, Foster J (2012) An assessment of the impact of the introduction of carbon price signals on prices, production trends, carbon emissions and power flows in the NEM for the period 2007-2009. EEMG working paper No.17, Energy Economics and Management Group, School of Economics, University of Queensland, April 2012

  • Zachmann G, von Hirschhausen C (2008) First evidence of asymmetric cost of pass-through of EU emissions allowances: examining wholesale electricity prices in Germany. Economics Lett 99(3):465–469

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Fatemeh Nazifi.

Appendix

Appendix

See Tables 4 and 5.

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Nazifi, F. The pass-through rates of carbon costs on to electricity prices within the Australian National Electricity Market. Environ Econ Policy Stud 18, 41–62 (2016). https://doi.org/10.1007/s10018-015-0111-8

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10018-015-0111-8

Keywords

JEL classifications

Navigation