INTRODUCTION

This paper provides a cost-benefit assessment of wind generation in Ontario, Canada for the 2020–2023 period and on a forward-looking basis for the 2027–2030 period. Our work is based on well-established economics literature examining the interaction of wind in various grids and its corresponding cost-benefit.^[2] This literature suggests that the social and climate cost-benefit of wind generation will be grid-specific, with the lower the price of wind on the grid and the more that wind displaces higher-emitting generation, the higher wind’s social and climate benefit. And vice versa. We find a large negative net cost of wind for the 2020–2023 period, reflecting Ontario’s relatively high wind prices and low wind emissions offset.

The second chapter provides a summary of Ontario’s wind roll out policy and how it resulted in its relatively high average price of $151/MWh in the 2020–2023 period. We then review Ontario’s seasonal wind profile and how this is likely to interact with the existing generation mix and emissions intensity. The third chapter presents the results of the regression analysis of the interaction of wind generation with other generation technologies. We apply the regression results to a cost-benefit analysis of wind generation and find that the costs far exceeded the benefits for the 2020–2023 period. We also undertake a forward-looking cost-benefit analysis for the 2027–2030 period and calculate a cost-benefit “break-even” wind price of $46/MWh. The fourth chapter includes the policy discussion, and the fifth chapter is the conclusion.

WIND IN ONTARIO’S ELECTRICITY SECTOR

Ontario’s installed wind capacity of 5.5 GW^[3] has largely evolved within an electricity sector that is unique in North America: a restructured, single-buyer with a system-wide contracts-for-difference (“CfD”) mechanism, majority out-of-market revenues, and high subsidization.^[4]

In preparation for market opening in May 2002, Ontario Hydro was split into several entities. In 2005, the new government established the single-buyer model for generation in Ontario by creating the Ontario Power Authority (“OPA”) responsible for contracting existing and new generation that was not otherwise economically regulated by the Ontario Energy Board (“OEB”). Indeed, virtually all wind resources in Ontario have been centrally procured by the government. To tie the administrative OPA element to the competitive Independent Electricity System Operator (“IESO”) element, the government introduced a sector-wide CfD mechanism in 2005. Generating entities would receive market revenues based on the hourly Ontario energy price (“HOEP”), on top of which they would receive out-of-market CfD payments equal to the difference between their individual “strike price” (set by regulation or contracts) and the HOEP. Those CfD-type payments are funded via the Global Adjustment (“GA”) mechanism, which has generally been fully recovered via rates.

Ontario’s first commercial wind farm went into service in 2002, but it was not until the government implemented the Renewable Energy Supply (“RES”) in 2004 that wind took off in Ontario. Additional rounds occurred in 2005 (“RES II”) and 2007 (“RES III”) and the related Renewable Energy Standard Offer Program (“RESOP”) in 2006. To speed up the rollout of wind and solar, government enacted the Green Energy and Green Economy Act^[5] (“GEA”) in 2009, which would later be renamed the Green Energy Act^[6] before being repealed in 2018. Modelled on German legislation, its key provisions included the rollout of the standard offer feed-in-tariff (FIT) approach to procurement so that wind projects were developed, owned and operated in the form of independent power producers (“IPPs”). Wind contracts generally had “escalation clauses” that increased the rate by one-fifth the rate of inflation. The average contracted wind price for the 2014–2019 period was $143/MWh and increased to $151/MWh for the 2020–2023 period.

The climate benefits of wind will generally depend on its particular profile, including capacity factors over the year, and how that interacts with the existing generation mix and its emissions intensity. On the one extreme, in a relatively high emission grid dominated by coal or oil, for instance, wind will tend to have a relatively higher climate benefit if it can displace coal or oil on a MWh-to-MWh basis. At the other extreme, in a relatively low emission grid dominated by nuclear or hydro with no coal, as is the case of Ontario, we would expect wind to have a relatively lower climate benefit.

To compare Ontario’s average wind profile for the 2020–2023 period, Figure 1 presents average monthly capacity factors from New York state^[7] (NYISO 2024, and previous), Alberta^[8] and the “Lower Plains” states as defined by the U.S. Energy Information Administration (“EIA”) to include Texas, Oklahoma, Kansas, and New Mexico^[9] for selected periods^[10]. Except for Alberta, the other profiles in Figure 1 show some form of an “M” shape, with peaks around March and November and a pronounced trough in July-August. Ontario’s monthly capacity factors are always higher than that of New York, indicating that Ontario has a superior wind profile. However, Ontario’s average capacity factor of 31 per cent is lower than that of Alberta (34 per cent) and of the Lower Plains (38 per cent). Ontario generally compares favourably to these other regions during the peaks; it is Ontario’s more pronounced and prolonged summer trough that brings down its average annual capacity factor.

Figure 1: Average wind capacity factors for 2020–2023^[11], by month

One of the innovations of this report is that the regression and cost-benefit analysis considers this seasonal variation. Indeed, for the rest of the report we use weekly data, week 1 to 52 of the year, to more accurately capture this seasonality. We construct a custom database for the four most recently available years, 2020–2023, based on publicly available data.^[12] We use this database in this chapter to graphically present the results and in Chapter 3 as the basis for the regression and cost-benefit analyses. This hourly^[13] data is only available for transmission-connected generation, which covers 92 per cent of all generation, with distribution-connected capacity making up the remaining 8 per cent, with the ratio for wind being 89 per cent/11 per cent respectively.^[14]

Figure 2 shows the average hourly demand and generation for the years 2020–2023, by week of the year. Table 1 presents averages, standard deviations and correlations. Ontario demand has two troughs (weeks 15 to 20 and weeks 39 to 43) and two peaks (weeks 27 to 34 in the summer and a winter peak from week 49 at the beginning of the winter in one year to week 7 near the end of winter the following year). The summer peak is associated with higher space cooling and the winter peak with higher space heating and industrial use.

Figure 2: Average Ontario demand and generation for 2020–2023, by week

Table 1 shows that demand averaged 15,422 MW and had a normalized standard deviation of 0.08. Wind averaged 1,425 MW with a normalized standard deviation of 0.32, with gas being even higher at 0.42. Nuclear^[15] is positively correlated with Ontario demand (correlation co-efficient r = 0.649).^[16] Gas generation is very strongly correlated with Ontario demand, with r = 0.862, reflecting its “peaking” function. In contrast, wind generation is uncorrelated with Ontario demand, with r = 0.047. Table 1 shows that wind is negatively correlated with gas generation, with r = -0.266, indicating that wind did not efficiently displace gas in Ontario.

Table 1: Descriptive Statistics for Ontario generation and demand for 2020-2023, by week

REGRESSION AND COST-BENEFIT ANALYSIS

This chapter presents the regression analysis to assess how wind generation interacted in Ontario’s nuclear and hydro-dominant grid for the four years from 2020–2023. We apply these regression results to a historical cost-benefit analysis of wind generation for the 2020–2023 period and a forward-looking cost-benefit analysis for the 2027–2030 period.

Our objective is calculating regression coefficients that quantify whether and by how much wind generation is statistically associated with decreases or increases of other types of generation. In our case, we focus on the three largest generation technologies in Ontario, nuclear, hydro and gas. We also model whether and by how much wind generation increases/decreases net exports (NX) from/to other provinces and the US. Our work differs from previous research by specifically considering the seasonal variation of wind by calculating separate week of year regressions over the 2020–2023 period. As described in the previous chapter, we pool hourly data by week of the year and carry out 208 regressions, one for each week of the year (52 weeks) for four variables (gas, hydro, nuclear, and NX).

Figure 3 presents the results of the wind interaction coefficients for the 208 regressions. Statistically significant coefficients are presented by their coefficient results; insignificant results are presented as zero. Overall, the regression results were strong, with relatively high adjusted R2 and other significance parameters. These coefficient results indicate that on average 1.00 MWh of wind generation was statistically associated with the following: a decrease (displacement) of -0.56 MWh of gas, a decrease (displacement) of -0.23 MWh of hydro, an increase (contribution) of 0.17 MWh to NX and had a minimal impact (-0.01 MWh) on nuclear. These results indicate that in Ontario’s low-emissions nuclear and hydro-dominant grid, only about 56 per cent of wind output goes to displacing gas generation.

Figure 3: Wind regression coefficients for 2020-2023, by week

Figure 3 highlights the importance of seasonal variation around these annual averages. During the winter peak of Ontario demand in week 5, for example, it shows that each 1.00 MWh of wind displaced -0.80 MWh of gas. On the other hand, in week 15 1.00 MWh of wind on average displaced just -0.24 MWh of gas. The climate benefits associated with wind displacing gas, therefore, depend on the week of the year.

Figure 4 shows how much gas wind is displacing. To be clear, the displaced gas did not occur — it is an estimate of the gas that would have occurred had wind not existed. It is the gas avoided. During week 5, for instance, wind displaced about 1,302 MW of gas generation per hour. In contrast, during week 15, wind displaced an average of only 375 MW of gas per hour.

Figure 4: Average gas generation and displacement for 2020–2023, by week

Figure 5 shows these climate benefits directly, by showing how much CO₂ is avoided by wind. It shows that on average 1.00 MWh (generation) of wind displaces 0.227 tCO₂ (the wind emissions offset), and that 1.00 MW (capacity) of wind displaces 0.072 tCO₂ per hour (the wind capacity emissions offset). This confirms that the capacity and output avoided CO₂ ratio (0.072/0.227) is the same as average wind capacity factor (31 per cent). From a capacity perspective, Figure 5 shows that the capacity value of wind with respect to climate are lowest in weeks 14 to 34, during which 1.00 MW displaces only 0.043 tCO₂ per hour.

Figure 5: Average tCO₂ reductions due to wind for 2020–2023, by week

This section expands this analysis to assess the cost- benefit of a more comprehensive perspective, including estimating the financial impacts of how wind interacts with the other modelled generation resources and NX, as well as placing a monetary value on the avoided CO₂ emissions in the form of the Social Cost of Carbon (“SCC”). From an Ontario perspective, there are two elements on the cost side, and four elements to the benefit side of the cost-benefit analysis, which we discuss below.

On the cost side the two elements are the expenses associated with wind output and with wind curtailment. Average annual wind output expenses are equal to average output over the 2020–2023 period (12.5 TWh) times the average wind price over the same period ($151/MWh).

Ontario has been a net exporter of electricity since the late-2000s, mostly driven by a condition that IESO refers to as “surplus baseload generation” (“SBG”), which occurs when electricity production from nuclear, hydro, wind, and solar is greater than Ontario demand.^[17] For grid stability purposes IESO must balance surplus and deficit power situations. IESO’s first “escape valve” in surplus situations is to increase exports; the second is to reduce Ontario generation, including wind generation. Such wind reductions are referred to as “curtailment.” As in other jurisdictions, wind IPPs are compensated for curtailment. IESO calculates the estimated capability for every wind turbine in Ontario based on a series of parameters, including available installed capacity, and actual wind speed at the location, based on sensors. The difference between actual and IESO forecast wind generation is referred to as “curtailed wind.” Average annual expenses associated with wind curtailment is equal to average wind curtailment over the 2020-2023 period (1.3 TWh) times the average wind price over the same period ($151/MWh).

Figure 6 shows average hourly wind generation and curtailment for the 2020-2023 period. Curtailed wind is highest during the hydro peak freshet in weeks 16 to 21 and lowest during the Ontario summer demand peak in weeks 27 to 34.

Figure 6: Average Ontario wind output and curtailment for 2020–2023, by week

There are four elements on the benefits side: the financial savings from decreased hydro and gas generation, the increased revenues from increased NX, and the financial benefits from avoided CO₂. We do not include any financial impact of nuclear given wind’s minimal impact on this form of generation.

Our regression-based estimates indicate that wind decreases hydro generation by an average of 2.7 TWh/year over the 2020–2023 period. We calculate the effective price of that reduction by associating wind-related decreased hydro generation with forgone hydro production due to SBG conditions. OPG, which has 84 per cent of Ontario’s hydro resources, reported forgone production of 2.2 TWh/year over the 2020–2023 period^[18] so that for the sector as whole that would be 2.6 TWh/year, very close to the regression-based estimates. OPG was compensated for its forgone hydro generation at $30/MWh based on series of OEB-approved deferral accounts. During this period OPG’s regulated hydro rate was $43/MWh, so the difference between that and the compensated price ($30/MWh) equals the per MWh savings from wind-decreased hydro ($13/MWh).

As discussed above, gas generation in Ontario is used as peaking and to back up wind and solar and not as “baseload,” and is not generally subject to SBG-related reductions. The way gas has been contracted reflects its profile in Ontario. Indeed, about 70 per cent of gas generation is contracted under deemed revenue monthly payments designed to promote the availability of gas capacity when it is needed^[19]. Under this specific contractual arrangement, the financial savings from displaced gas generation is equal to the value of the natural gas and other approved variable costs. The gas generation savings therefore are based on the average 2020–2023 Dawn Hub natural gas price ($4.50/MMBtu) multiplied by the gas saved (54.1 million MMBtu/year). This is equivalent to $34.5/MWh for 7.0 TWh, to which we add $5/MWh as a proxy for the other variable costs.

We calculate revenues from NX by multiplying the average regression — based additional NX for the 2020–2023 period (2.2 TWh) by the average NX price of $37/MWh. For the financial valuation of avoided CO₂ we use a SCC of $50/tCO₂^[20] and multiply it by the avoided emissions (2.9 MtCO₂) associated with the displaced gas.

The summary results of the 2020–2023 cost-benefit analysis are presented in Figure 7, which includes the two cost and four benefit elements as well as the overall cost-benefit. To facilitate comparisons with other scenarios, we calculate the cost-benefit result on a MWh basis, at -$124/MWh. This means that the costs of wind generation in Ontario during the 2020-2023 period far exceeded the corresponding climate and other benefits. This result is driven by the relatively high contracted wind price over the period ($151/MWh) and by our finding that while wind displaced some gas generation, it also displaced lower priced zero-emission hydro and contributed to lower priced NX.

Figure 7: Average cost-benefit of wind generation for 2020–2023, by week

We then move on to estimating the forward-looking cost-benefit analysis for the 2027–2030 period. We chose this period because it is relatively soon from an energy system perspective, and hence the regression parameters that we calculated for the 2020–2023 period are likely to remain reasonably valid. Our analysis serves for two scenarios. One is for the legacy wind projects whose 20-year contracts would expire in and around this period. As it has for other resources whose contracts have expired, there could be a mutual interest between IESO and wind IPPs to re-contract, depending on operational state of the resources. Our study provides an assessment of the price at which such a re-contracting could be cost-beneficial. Our work also serves to provide insight into the cost-benefit of new wind projects.^[21]

For the 2027–2030 scenario we maintain most of the same parameters that we used for the 2020–2023 analysis other than update the natural gas price based on the average 2027–2030 forecast of $6.35/MMBtu.^[22] As a base, we use a (rounded) reference wind price of $80/MWh, based on applying Ontario’s wind capacity factor of 31 per cent to a recent levelized cost of energy (LCOE) study for wind.^[23] Given the recent trajectory of wind LCOEs and uncertainty over its future evolution, we use the same nominal amount of $80/MWh for the 2027–2030 period.

Figure 8 presents the results for the 2027–2030 period, with a cost- benefit result of -$38/MWh. This result is based on a 10 per cent increase in wind generation relative to the baseline amount, but the size-normalized result of -$38/MWh equally applies to both re-contracted legacy and new wind projects. These results suggest that even at the lower reference price of $80/MWh relative to the $151/MWh that held during the 2020–2023 period, the costs associated with wind generation still exceed the corresponding climate and other benefits.

Figure 8: Average cost-benefit of wind generation for 2027–2030, by week

There are an infinite number of possible variations of the baseline and reference amounts to test the sensitivity of the reference 2027–2030 results. For example, we can calculate the “break even” wind price at $46/MWh that would be required to set the 2027–2030 cost-benefit = $0/MWh. The break-even price of $46/MWh is well below both the actual average 2020–2023 price of $151/MWh and the LCOE-based reference price for 2027–2030 of $80/MWh. We can calculate that with an SCC of $0 would result in a break-even wind price of $36 while an SCC=$150/tCO₂^[24] results in a wind price of $67/MWh.

Our regression results are comparable to those of an earlier Ontario study,^[25] suggesting that the results are robust relative to level of data aggregation and to time period. Ontario’s wind emission offset is relatively low, at only 43 per cent of Texas (0.227 vs. 0.53 tCO₂/MWh)^[26], for instance. This reflects that in Ontario 1.00 MWh of wind displaces only 0.56 MWh of gas, a relatively lower-emitting technology, compared to other regions where wind tends to displace relatively more coal and/or gas. Likewise, because of Ontario’s relatively modest wind capacity factor, its wind capacity emissions offset is relatively lower than Texas, at just 37 per cent (0.072 vs. 0.196 tCO₂/MW per hour)^[27].

POLICY DISCUSSION

Our analysis can inform policy options with respect to legacy and new wind projects.

For legacy wind projects whose contracts expire before 2030 the choice faced by owners will be either to decommission or to continue operations either “as is” or under partial/full repowering. Financially, the wind IPPs would recognize that re-contracting at or near $151/MWh is unlikely to be politically or economically feasible and that continuing operations could be done under a new contract with IESO or uncontracted, either a pure HOEP-only market merchant or with a third-party Power Purchase Agreement (“PPA”). From an IESO perspective, our analysis is clear that the societal break-even contract price is about $46/MWh. Assuming that the initial wind project financing in Ontario was for 20 years or less, at contract termination the incremental costs of long-term operation with no or modest partial repowering could well be at or below $46/MWh. In comparison, the relative attractiveness of the HOEP-only alternative would depend on long-term forecasts of the HOEP. The HOEP averaged $30 during 2020–2023 period, with an annual peak of $47 in 2022 during the energy crisis.

One approach would be for IESO to design and offer a wind re-contracting standard offer of $46/MWh for a maximum of a ten-year CfD-type mechanism. Wind IPPs would then be able to determine their decommissioning/continuation business decision based on this standard offer and their specific situation. Some wind operations would shut down, some will recontract with IESO, and some may continue operations either under a third party PPA or be pure merchant. By way of reference, for the Eastern US the average PPA in 2021–2022 was about $65/MWh^[28].

On a stand-alone basis, not considering incremental transmission and other system costs, a similar cost-benefit perspective applies for new wind projects. From an IESO perspective, the same societal break-even contract price of about $46/MWh applies. However, the new build-based reference price results in a large gap between the social price ($46/MWh) and the private cost ($80/MWh). There are a number of options in this regard.

One option is to continue to move forward under the current private wind IPP contractual approach and for the IESO to design a competitive auction process with a maximum “reserve price” of $46/MWh. The reserve price is a critical because if it is set too high it could lead to a low value for money result for the public, but if set too low, wind IPPs may decide not to participate because it does not meet their target weighted average cost of capital (“WACC”). Another possibility is to discard the contractual approach in favour of financing and compensating wind projects based on cost-of-service economic regulation. There is no particular reason that wind should be treated any differently than the majority of generation resources in Ontario or Canada as a whole. The argument that the contractual approach is always superior to economic regulation simply does not hold for wind in Ontario over the last 20 years. Indeed, economic regulation could do a better job of aligning public costs with public benefits.

A third option would be to leverage the larger economies of scale and lower cost of public financing and have new wind projects publicly-owned and operated. This is already the case of about half of the wind capacity in PEI^[29] and is the thrust of the just-announced strategy in Quebec that aims to roll out 10 GW of new publicly-owned wind by 2035 that Hydro-Quebec claims could result in savings of as much as 20 per cent from centralized purchasing and other economies of scale.^[30] The wind assets would enter OPG’s regulated “rate base” and be subject to the lower cost of financing associated with provincially backed Crown corporations, compared to private financing.

CONCLUSION

Our research shows that costs of wind far exceeded its societal and climate benefits for the 2020–2023 period, with average net cost of -$124/MWh. Such a negative result is a combination of Ontario’s relatively low wind emissions offset (0.227 tCO₂/MWh) and high wind prices ($151/MWh). We also undertook a forward-looking cost-benefit analysis for the 2027–2030 period and calculate an average net cost of wind of -$38/MWh based on a reference price of $80/MWh. The cost-benefit “break-even” wind price for the 2027–2030 period is $46/MWh.

Structurally, wind’s value is relatively low in Ontario’s current low-emission nuclear and hydro-dominant grid. Ontario’s average wind capacity factor is relatively low. While wind technology could improve this performance in an absolute sense, it will not change the comparative disadvantage. Further, wind in Ontario is negatively correlated with gas generation, making it relatively inefficient at displacing it. Regardless of the price of wind, these structural shortcomings would remain in the short- and medium-term.

The challenge from a policy perspective is to implement programs that are sustainable over time and that align public costs with public benefits. The overall experience of wind generation in Ontario over the last twenty years has been that costs have far exceeded the benefits. Our hope is that this and other research contributions will provide the type of forward-looking guidance to ensure that any future wind development in Ontario is in the public interest.

• ¹ This is a condensed version of a longer report of the same title prepared by the author for the Macdonald-Laurier Institute (MLI) and released in September 2024: Edgardo Sepulveda, Chasing the Wind: The value of wind generation in a low-emission nuclear and hydro-dominant grid – the case of Ontario (A Macdonald-Laurier Institute Publication, 2024), online (pdf): <macdonaldlaurier.ca/wp-content/uploads/2024/09/20240724_Wind-power-Sepulveda_PAPER-v13-FINAL.pdf>.
• * The author is a regulatory economist with more than thirty years of experience in the telecommunications and electricity sectors. He has advised governments, regulatory agencies, companies, unions, and consumer advocates in more than forty countries. Born in Chile, he is fluent in English and Spanish and has a good working knowledge of French. He received his B.A. (Honours) from the University of British Columbia and his M.A. from Queen’s University, both in Economics. He established Sepulveda Consulting Inc. in 2006.
• ² This includes work on the Texas electricity grid by Joseph Cullen, “Measuring the Environmental Benefits of Wind- Generated Electricity” (2013) 5:4 Am Econ J: Econ Pol’y, 107-33, online: <doi.org/10.1257/pol.5.4.107>; and Kevin Novan, “Valuing the Wind: Renewable Energy Policies and Air Pollution Avoided” (2015) 7:3 Am Econ J: Econ Pol’y, 291-326, online: <doi.org/10.1257/pol.20130268>; and more recent work analyzing the Ontario grid by Pejman Bahramian, Glenn P. Jenkins & Frank Milne, “The displacement impacts of wind power electricity generation: Costly lessons from Ontario” (2021) Energy Pol’y 151, online: <doi.org/10.1016/j.enpol.2021.112211> [Bahramian]; and several regions of the United States by Harrison Fell & Jeremiah X. Johnson, “Regional disparities in emissions reduction and net trade from renewables” (2021) 4 Nature Sustainability, 358–65, online: <doi.org/10.1038/s41893-020-00652-9> [Harrison Fell].
• ³ Independent Electricity System Operator, “Ontario’s Electricity Grid: Supply Mix and Generation” (last visited 22 January 2025), online: <www.ieso.ca/Learn/Ontario-Electricity-Grid/Supply-Mix-and-Generation>.
• ⁴ For an overview of early reforms see Michael Trebilcock & Roy Hrab, “Electricity Restructuring in Ontario” (2005) 6:1 Energy J, online: <journals.sagepub.com/doi/abs/10.5547/ISSN0195-6574-EJ-Vol26-No1-6> for an update, including on the GA and increasing prices, see Edgardo Sepulveda, “Power to the people: Privatization and electioneering have made electricity prices unbearable in Ontario” (1 May 2018), online: <www.policyalternatives.ca/news-research/power-to-the-people>; and for subsidization see Edgardo Sepulveda, “Ontario election: The $6.9 billion budget item that (almost) no one is talking about” (19 May 2022), online: <www.tvo.org/article/ontario-election-the-69-billion-budget-item-that-almost-no-one-is-talking-about>.
• ⁵ Green Energy Act, SO 2009, c 12, s A.
• ⁶ Ibid.
• ⁷ New York Independent System Operator, “NYCA Renewables 2023” (last visited 22 January 2025), online: <www.nyiso.com/documents/20142/29609937/2023-NYCA-Renewables-Presentation.pdf/b4b189e8-e213-baf1-9f81-ac425342a2ea>.
• ⁸ Alberta Electricity System Operator, Annual Market Statistics Report, (Market Analytics, 2024), online: <www.public.tableau.com/app/profile/market.analytics/viz/AnnualStatistics_16161854228350/Introduction>.
• ⁹ See United States Energy Information Administration, “U.S. wind generation falls into regional patterns by season” (30 November 2022), online: <www.eia.gov/todayinenergy/detail.php?id=54819>.
• ¹⁰ Data for Ontario, NYISO and AESO are monthly averages for the 2020–2023 period; for the Lower Plains the data is monthly from 2016 to mid-2022.
• ¹¹ Data for the Lower Plains (TX,OK,KS,NM) is from 2016 to mid-2022.
• ¹² See Independent Electricity System Operator, “Public Reports Data Portal” (last visited 22 January 2025), online: <www.reports.ieso.ca/public>.
• ¹³ For our database we use the hour as the basic unit of analysis and group all hours in seven-day periods from January 1 of every year, from week 1 to week 52. Fifty-two 7-day weeks total 364 days, so we need to add an eighth day to one of the weeks to have the necessary 365 days. Each of the weeks from week 1 to week 51 have seven days thus a total of 672 hours (24 hours x 7 days x 4 years). Week 52 will get an extra day thus having 768 hours (24 hours x 8 days x 4 years). For analytical purposes we exclude the 24 hourly data points for February 29 of 2020, a leap year.
• ¹⁴ Supra note 3.
• ¹⁵ This type of nuclear seasonal “load following” is made possible by planning maintenance outages for Ontario’s fleet of 18 nuclear reactors in a coordinated manner consistent with Ontario demand.
• ¹⁶ The correlation coefficient r measures the strength of the relationship between two variables, going from -1.00 (perfect negative correlation means two variables move in opposite direction), to 1.00 (perfect positive correlation means that two variables move in the same direction all the time), with 0.00 meaning uncorrelated.
• ¹⁷ Ontario Power Generation, OPG Reports 2023 Financial Results, (Toronto: Ontario Power Generation Inc., 2024), online (pdf): <www.opg.com/documents/2023-financial-results-pdf>.
• ¹⁸ Ibid.
• ¹⁹ In summary, for each different gas plant IESO establishes a fixed dollar amount to pay for fixed capital and operational costs, as if there was no gas generation. From that amount IESO subtracts the net revenues that specific generator should have earned (“deemed revenues”) in the market, after paying for the natural gas and other approved variable costs. Deemed hours of generation are those during which the HOEP exceeded the specific operator’s approved net variable costs. To ensure stand-by capacity, this system “tops up” net energy revenues with a form of capacity payment to “make whole” the generators.
• ²⁰ See Bahramian, supra note 2. Based on Government of Canada, “Carbon pricing: regulatory framework for the output-based pricing system” (last modified 31 January 2018), online: <www.canada.ca/en/services/environment/weather/climatechange/climate-action/pricing-carbon-pollution/output-based-pricing-system.html>.
• ²¹ Conceptually, the biggest difference between the cost-benefit analysis of legacy or new projects would be the inclusion in the latter of the system and other costs of adding new wind. This would include new transmission resources to enable the expansion of wind, possibly new back-up or storage facilities and related ancillary services. While this type of detailed modelling is outside the scope of this study, it is important to keep in mind that these incremental costs are likely to be significant. For example, IESO estimates that the average cost of new transmission to 2050 for wind projects is in the range of $25/MWh. See Independent Electricity System Operator, Pathways to Decarbonization (Independent Electricity System Operator, 2022), online (pdf): <www.ieso.ca/-/media/Files/IESO/Document-Library/gas-phase-out/Pathways-to-Decarbonization.pdf>.
• ²² Independent Electricity System Operator,. 2024 Annual Planning Outlook: Resource Costs and Trends (Independent Electricity System Operator, 2024), online (pdf): <www.ieso.ca/-/media/Files/IESO/Document-Library/planning-forecasts/apo/Mar2024/Resource-Costs-and-Trends.pdf>.
• ²³ National Renewable Energy Laboratory, “2022 Cost of Wind Energy Review” (last visited 22 January 2025), online (pdf): <www.nrel.gov/docs/fy24osti/88335.pdf>.
• ²⁴ Government of Canada, “Update to the Pan-Canadian Approach to Carbon Pollution Pricing 2023–2030” (last modified 5 August 2021), online: <www.canada.ca/en/environment-climate-change/services/climate-change/pricing-pollution-how-it-will-work/carbon-pollution-pricing-federal-benchmark-information/federal-benchmark-2023-2030.html>.
• ²⁵ See Bahramian, supra note 2.
• ²⁶ Harrison Fell, supra note 2.
• ²⁷ Ibid.
• ²⁸ U.S. Department of Energy, Land-Based Wind Market Report: 2023 Edition, (Lawrence Berkeley National Laboratory: Wind Energy Technologies Office of the U.S. Department of Energy’s Office of Energy Efficiency and Renewable Energy, 2023) online (pdf): <www.energy.gov/sites/default/files/2023-08/land-based-wind-market-report-2023-edition.pdf>.
• ²⁹ Prince Edward Island Energy Corporation, “What We Do” (last visited 22 January 2025), online (pdf): <www.peiec.ca>.
• ³⁰ Hydro-Québec Charting the Course toward Collective Success: Wind Power Development Strategy, (Québec: Hydro-Québec, 2024), online (pdf): <www.hydroquebec.com/data/a-propos/pdf/wind-power-development-strategy.pdf>.