
Carrying value as at Dec. 31, 2019
	Derivatives used for hedging	Derivatives held for trading (FVTPL)	Amortized cost	Total
Financial assets
Cash and cash equivalents⁽¹⁾	—	—	411	411
Restricted cash	—	—	32	32
Trade and other receivables	—	—	462	462
Long-term portion of finance lease receivable	—	—	176	176

Risk management assets
Current	71	95	—	166
Long-term	607	33	—	640
Other assets (Note 21)	—	—	47	47
Financial liabilities
Accounts payable and accrued liabilities	—	—	413	413
Dividends payable	—	—	37	37
Risk management liabilities
Current	1	80	—	81
Long-term	1	28	—	29
Credit facilities, long-term debt and finance lease obligations⁽²⁾	—	—	3,212	3,212
Exchangeable securities	—	—	326	326


Carrying value as at Dec. 31, 2018
	Derivatives used for hedging	Derivatives held for trading (FVTPL)	Amortized cost	Other financial assets (FVTPL)	Total
Financial assets
Cash and cash equivalents⁽¹⁾	—	—	89	—	89
Restricted cash	—	—	66	—	66
Trade and other receivables	—	—	731	25	756
Long-term portion of finance lease receivables	—	—	191	—	191

Risk management assets
Current	60	86	—	—	146
Long-term	629	33	—	—	662
Other assets	—	—	37	15	52
Financial liabilities
Accounts payable and accrued liabilities	—	—	496	—	496
Dividends payable	—	—	58	—	58
Risk management liabilities
Current	1	89	—	—	90
Long-term	1	40	—	—	41
Credit facilities, long-term debt and finance lease obligations⁽²⁾	—	—	3,267	—	3,267

The fair value of a financial instrument is the price that would be received to sell an asset or paid to transfer a liability in an orderly transaction between market participants at the measurement date. Fair values can be determined by reference to prices for that instrument in active markets to which the Corporation has access. In the absence of an active market, the Corporation determines fair values based on valuation models or by reference to other similar products in active markets.

Fair values determined using valuation models require the use of assumptions. In determining those assumptions, the Corporation looks primarily to external readily observable market inputs. However, if not available, the Corporation uses inputs that are not based on observable market data.

The Level I, II and III classifications in the fair value hierarchy utilized by the Corporation are defined below. The fair value measurement of a financial instrument is included in only one of the three levels, the determination of which is based on the lowest level input that is significant to the derivation of the fair value.

Fair values are determined using inputs that are quoted prices (unadjusted) in active markets for identical assets or liabilities that the Corporation has the ability to access at the measurement date. In determining Level I fair values, the Corporation uses quoted prices for identically traded commodities obtained from active exchanges such as the New York Mercantile Exchange.

The Corporation’s commodity risk management Level II financial instruments include over-the-counter derivatives with values based on observable commodity futures curves and derivatives with inputs validated by broker quotes or other publicly available market data providers. Level II fair values are also determined using valuation techniques, such as option pricing models and interpolation formulas, where the inputs are readily observable.

In determining Level II fair values of other risk management assets and liabilities, the Corporation uses observable inputs other than unadjusted quoted prices that are observable for the asset or liability, such as interest rate yield curves and currency rates. For certain financial instruments where insufficient trading volume or lack of recent trades exists, the Corporation relies on similar interest or currency rate inputs and other third-party information such as credit spreads.

The Corporation may enter into commodity transactions for which market-observable data is not available. In these cases, Level III fair values are determined using valuation techniques such as mark-to-forecast and mark-to-model. For mark-to-model valuations, derivative pricing models, regression-based models and historical bootstrap models may be employed. The model inputs may be based on historical data such as unit availability, transmission congestion, demand profiles for individual non-standard deals and structured products, and/or volatilities and correlations between products derived from historical price relationships.

The Corporation also has various commodity contracts with terms that extend beyond a liquid trading period. As forward market prices are not available for the full period of these contracts, the value of these contracts is derived by reference to a forecast that is based on a combination of external and internal fundamental modelling, including discounting. As a result, these contracts are classified in Level III.

The Corporation has a Commodity Exposure Management Policy that governs both the commodity transactions undertaken in its proprietary trading business and those undertaken to manage commodity price exposures in its generation business. This Policy defines and specifies the controls and management responsibilities associated with commodity trading activities, as well as the nature and frequency of required reporting of such activities.

Methodologies and procedures regarding commodity risk management Level III fair value measurements are determined by the Corporation’s risk management department. Level III fair values are calculated within the Corporation’s energy trading risk management system based on underlying contractual data as well as observable and non-observable inputs. Development of non-observable inputs requires the use of judgment. To ensure reasonability, system-generated Level III fair value measurements are reviewed and validated by the risk management and finance departments. Review occurs formally on a quarterly basis or more frequently if daily review and monitoring procedures identify unexpected changes to fair value or changes to key parameters.

Information on risk management contracts or groups of risk management contracts that are included in Level III measurements and the related unobservable inputs and sensitivities, is as follows, and excludes the effects on fair value of certain unobservable inputs such as liquidity and credit discount (described as “base fair values”), as well as inception gains or losses. Sensitivity ranges for the base fair values are determined using reasonably possible alternative assumptions for the key unobservable inputs, which may include forward commodity prices, commodity volatilities and correlations, delivery volumes and shapes.

As at Dec. 31

2019

2018

Description

Base fair value

Sensitivity

Base fair value

Sensitivity

Long-term power sale – US

737

+46
-139

801

+116
-116

Unit contingent power purchases

(6)

+1
-1

+4
-4

Structured products – Eastern US

+2
-2

+5
-5

Full requirements – Eastern US

+3
-3

—

Long-term wind energy sale – Eastern US

(28)

+20
-20

(39)

+21
-21

Others

—

+7
-7

+3
-3

For periods beyond 2021, market forward power prices are not readily observable. For these periods, fundamental-based forecasts and market indications have been used to determine proxies for base, high and low power price scenarios. The base price forecast has been developed by using a fundamental-based forecast (the provider is an independent and widely accepted industry expert for scenario and planning views). Prior to the second quarter of 2018, the base price forecast was developed using an additional independent industry forecast. Forward power price ranges per MWh used in determining the Level III base fair value at Dec. 31, 2019, are US$20 to US$28 (Dec. 31, 2018 - US$20 to US$35). The sensitivity analysis has been prepared using the Corporation’s assessment that a US$3 to US$9 (Dec. 31, 2018 - US$6) price decrease or increase in the forward power prices is a reasonably possible change.

The contract is denominated in US dollars. With the weakening of the US dollar relative to the Canadian dollar from Dec. 31, 2018 to Dec. 31, 2019, the base fair value and the sensitivity values have decreased by approximately $11 million and $2 million, respectively.

Under the unit contingent PPAs, the Corporation has agreed to purchase power contingent upon the actual generation of specific units owned and operated by third parties. Under these types of agreements, the purchaser pays the supplier an agreed upon fixed price per MWh of output multiplied by the pro-rata share of actual unit production (nil if a plant outage occurs). The contracts are accounted for as FVTPL.

The key unobservable inputs used in the valuations are delivered volume expectations and hourly shapes of production. Hourly shaping of the production will result in realized prices that may be at a discount (or premium) relative to the average settled power price. Reasonably possible alternative inputs were used to determine sensitivity on the fair value measurements.

This analysis is based on historical production data of the generation units for available history. Price and volumetric discount ranges per MWh used in the Level III base fair value measurement at Dec. 31, 2019, are nil (Dec. 31, 2018 – nil) and 2.2 per cent to 2.8 per cent (Dec. 31, 2018 – 2.2 per cent to 16.9 per cent), respectively. The sensitivity analysis has been prepared using the Corporation’s assessment of a reasonably possible change in price discount ranges of approximately 1.0 per cent to 2.0 per cent (Dec. 31, 2018 – 1.1 per cent to 1.9 per cent) and a change in volumetric discount rates of approximately 8.6 per cent to 10.5 per cent (Dec. 31, 2018 – 8.6 per cent and 27.3 per cent), which approximate one standard deviation for each input.

The Corporation has fixed priced power and heat rate contracts in the eastern United States. Under the fixed priced power contracts, the Corporation has agreed to buy or sell power at non-liquid locations or during non-standard hours. The Corporation has also bought and sold heat rate contracts at both liquid and non-liquid locations. Under a heat rate contract, the buyer has the right to purchase power at times when the market heat rate is higher than the contractual heat rate. As at Dec. 31, 2019, the Corporation did not have any open positions on heat rate contracts.

The key unobservable inputs in the valuation of the fixed priced power contracts are market forward spreads and non-standard shape factors. A historical regression analysis has been performed to model the spreads between non-liquid and liquid hubs. The non-standard shape factors have been determined using the historical data. Basis relationship and non-standard shape factors used in the Level III base fair value measurement at Dec. 31, 2019, are 91 per cent to 112 per cent and 63 per cent to 116 per cent (Dec. 31, 2018 – 75 per cent to 109 per cent and 63 per cent to 104 per cent), respectively. The sensitivity analysis has been prepared using the Corporation’s assessment of a reasonably possible change in market forward spreads of approximately 4.0 per cent to 6.0 per cent (Dec. 31, 2018 – 4.2 per cent to 6.9 per cent) and a change in non-standard shape factors of approximately 4.0 per cent to 10.0 per cent (Dec. 31, 2018 – 4.0 per cent to 9.3 per cent), which approximate one standard deviation for each input.

The key unobservable inputs in the valuation of the heat rate contracts are implied volatilities and correlations. As there are no open positions on Level III heat rate option contracts, the implied volatilities and correlations used in the Level III base fair value measurement at Dec. 31, 2019, are nil and nil (Dec. 31, 2018 – 25 per cent to 84 per cent and 70 per cent), respectively. The sensitivity analysis was prepared using the Corporation’s assessment of a reasonably possible change in implied volatilities ranges and correlations of approximately nil and nil, respectively (2018 – 37 per cent to 49 per cent and 30 per cent, respectively).

The Corporation has a portfolio of full requirement service contracts, whereby the Corporation agrees to supply specific utility customer needs for a range of products that may include electrical energy, capacity, transmission, ancillary services, renewable energy credits and Independent System Operator costs.

The key unobservable inputs used in the portfolio valuation include delivered volume and supply cost. Hourly shaping of consumption will result in a realized cost that may be at a premium (or discount) relative to the average settled price. Reasonable possible alternative inputs are used to determine sensitivity on the fair value measurement. The sensitivity analysis has been prepared using the Corporation’s assessment that a reasonably possible change in the expected portfolio delivery volumes and portfolio’s realized cost of supply of (+/-) 5 per cent and (+/-) US$1 per MWh, respectively.

In relation to the acquisition of Big Level (See Note 4(J)), the Corporation has a long-term contract for differences whereby the Corporation receives a fixed price per MWh and pays the prevailing real-time energy market price per MWh as well as the physical delivery of renewable energy credits ("RECs") based on proxy generation. Commercial operation of the facility was achieved in December 2019, with the contract commencing on July 1, 2019, and extending for 15 years after the commercial operation date. The contract is accounted for at fair value through profit or loss.

The key unobservable inputs used in the valuation of the contract are expected proxy generation volumes and forward prices for power and RECs beyond 2024 and 2022, respectively. Forward power and REC prices per MWh used in determining the Level III base fair value at Dec. 31, 2019, are US$38 to US$60 and US$9 (Dec. 31, 2018 – US$42 to US$68 and US$7 to US$8), respectively. The sensitivity analysis has been prepared using the Corporation’s assessment that a change in expected proxy generation volumes of 10 per cent (2018 – 10 per cent), a change in energy prices of US$6 (2018 – US$6) and a change in REC prices of US$1 (2018 – US$1) as reasonably possible changes.

Commodity risk management assets and liabilities include risk management assets and liabilities that are used in the energy marketing and generation businesses in relation to trading activities and certain contracting activities. To the extent applicable, changes in net risk management assets and liabilities for non-hedge positions are reflected within earnings of these businesses.

Commodity risk management assets and liabilities classified by fair value levels as at Dec. 31, 2019, are as follows: Level I - $3 million net liability (Dec. 31, 2018 – $3 million net asset), Level II – $9 million net asset (Dec. 31, 2018 – $19 million net liability) and Level III – $686 million net asset (Dec. 31, 2018 – $695 million net asset).


	Year ended Dec. 31, 2019					Year ended Dec. 31, 2018
	Hedge	Non-hedge	Total	Hedge	Non-hedge	Total
Opening balance	689	6	695	719	52	771
Changes attributable to:
Market price changes on existing contracts	77	8	85	(7)	(9)	(16)
Market price changes on new contracts	—	14	14	—	4	4
Contracts settled	(57)	(19)	(76)	(90)	(42)	(132)
Change in foreign exchange rates	(31)	(1)	(32)	67	5	72
Transfers into (out of) Level III	—	—	—	—	(4)	(4)
Net risk management assets at end of period	678	8	686	689	6	695
Additional Level III information:
Gains recognized in other comprehensive income	46	—	46	60	—	60
Total gains included in earnings before income taxes	57	21	78	90	—	90
Unrealized gains (losses) included in earnings before income taxes relating to net assets held at period end	—	2	2	—	(42)	(42)

Other risk management assets and liabilities primarily include risk management assets and liabilities that are used in managing exposures on non-energy marketing transactions such as interest rates, the net investment in foreign operations and other foreign currency risks. Hedge accounting is not always applied.

Other risk management assets and liabilities with a total net asset fair value of $4 million as at Dec. 31, 2019 (Dec. 31, 2018 – $2 million net liability) are classified as Level II fair value measurements. The significant changes in other net risk management assets during the year ended Dec. 31, 2019, are primarily attributable to favourable market prices on existing contracts.

The fair values of the Corporation’s debentures, senior notes and exchangeable securities are determined using prices observed in secondary markets. Non-recourse and other long-term debt fair values are determined by calculating an implied price based on a current assessment of the yield to maturity.

The carrying amount of other short-term financial assets and liabilities (cash and cash equivalents, trade accounts receivable, collateral paid, accounts payable and accrued liabilities, collateral received and dividends payable) approximates fair value due to the liquid nature of the asset or liability. The fair values of the loan receivable (see Note 21) and the finance lease receivables (see Note 8) approximate the carrying amounts.

The majority of derivatives traded by the Corporation are based on adjusted quoted prices on an active exchange or extend beyond the time period for which exchange-based quotes are available. The fair values of these derivatives are determined using inputs that are not readily observable. Refer to section B of this Note 14 above for fair value Level III valuation techniques used. In some instances, a difference may arise between the fair value of a financial instrument at initial recognition (the “transaction price”) and the amount calculated through a valuation model. This unrealized gain or loss at inception is recognized in net earnings (loss) only if the fair value of the instrument is evidenced by a quoted market price in an active market, observable current market transactions that are substantially the same, or a valuation technique that uses observable market inputs. Where these criteria are not met, the difference is deferred on the Consolidated Statements of Financial Position in risk management assets or liabilities, and is recognized in net earnings (loss) over the term of the related contract. The difference between the transaction price and the fair value determined using a valuation model, yet to be recognized in net earnings, and a reconciliation of changes is as follows:


As at Dec. 31	2019	2018	2017
Unamortized net gain at beginning of year	49	105	148
New inception gains (losses)	3	(14)	12
Change in foreign exchange rates	—	5	(7)
Amortization recorded in net earnings during the year	(43)	(47)	(48)
Unamortized net gain at end of year	9	49	105