Determination No. R-2019-229

SUMMARY

[1] This is the first comprehensive review of the cost rate of common equity since 2011. The cost rate of common equity is used in the cost of capital methodology, which, in turn, informs an array of railway-related determinations, including regulated interswitching rates and the Maximum Revenue Entitlement (MRE).

[2] In Determination No. R-2018-254 (2019 interswitching rates Determination), the Agency informed stakeholders that the Agency would examine the methodology that it uses to set interswitching rates, including its cost of capital methodology.

[3] Following up on that Determination, the Agency will, in this Determination, address the following issues with respect to the cost rate of common equity, an important input to the cost of capital methodology:

• Which cost of equity model or combination of cost of equity models should be considered by the Agency?
• Is the current Capital Asset Pricing Model (CAPM) formulation appropriate for determining an adequate rate of return on investment?

[4] For the reasons set out below, the Agency finds that the CAPM remains appropriate for determining an adequate rate of return on investment. Accordingly, the Agency will continue to use the CAPM for determining the cost rate of common equity for federally-regulated railway companies.

[5] However, the Agency has adjusted certain components of the CAPM as follows:

1. For the Canadian cost rate of common equity, the risk-free rate (other than the risk-free rate in the beta calculation) will be determined as the yield on 5‑to‑10‑year Government of Canada marketable bonds.
2. For the U.S. cost rate of common equity, two distinct risk-free rates (other than the risk-free rate in the beta calculation) will be determined using the yields on 5‑year and 10-year U.S. Treasury Bonds.
3. The beta will not be adjusted for mean reversion.

[6] The full methodology that will be used by the Agency to determine cost of equity rates, beginning with the determination of the 2020 regulated interswitching rates, is set out in the Appendix. The methodology presented in the Appendix will also be applied by the Agency in all cost of capital determinations.

BACKGROUND

The Agency's cost of capital methodology

[7] The cost of capital is an estimate of the total return on net investment required by debt holders (cost of debt) and shareholders (cost of common equity) such that debt costs can be paid and shareholders can be provided with a return on investment consistent with the risks assumed for the period under consideration.

[8] The Agency's methodology for calculating the cost of capital of federally-regulated railway companies consists of four distinct assessments:

1. determination of net rail investment;
2. determination of capital structure;
3. determination of capital structure cost rates, which includes the cost rate of debt, deferred taxes and common equity; and
4. calculation of the cost of capital rate.

[9] In Decision No. 425-R-2011 (2011 cost of capital Decision), the Agency defined each of these assessments for the purposes of railway-related regulatory determinations under its jurisdiction. The relevant definitions are outlined in paragraphs 10 to 14 below.

[10] The net rail investment is defined as the gross book value of all the railway company's assets less accumulated depreciation. This first component defines the portion of the railway company's net assets that are providing rail transportation services and are under the Agency's jurisdiction.

[11] The capital structure is defined as the combination of the various sources of capital used to finance the net rail investment. In broad terms, funding can be achieved through borrowing, issuance of debt instruments, deferred taxes and shareholders' equity.

[12] The cost rate of debt is defined as the actual cost of the debt, that is, the interest paid to financial institutions or bond holders for loans made to the railway companies, as recorded in the most recent financial statements of the railway companies and submitted to the Agency.

[13] The cost rate of common equity is defined as the estimated return to equity shareholders that is required for railway companies to maintain access to equity for financing purposes. In 2011, the Agency determined that this return should be estimated using the CAPM.

[14] The annual cost of capital rate for each railway company is defined as the weighted average of the cost rates of debt, equity and deferred taxes, based on the proportions of each type of funding in the railway company's capital structure. When this rate is applied to the net book value of the assets involved, it results in the cost of capital in dollar terms.

Significance of the cost rate of common equity and the cost of capital methodology

[15] The Agency makes cost of capital determinations annually in accordance with subsection 157(1) of the Canada Transportation Act, S.C., 1996, c.10 as amended (CTA), based on its underlying methodology. As it is clear from the above, the cost rate of common equity is a critical component of the cost of capital methodology.

[16] The cost of capital is used in the Agency's railway-related regulatory determinations, including in the determination of costs for setting regulated interswitching rates and the Volume Related Composite Price Index (VRCPI), a component of the MRE. It is also used for other purposes such as resolving disputes about public passenger service providers' use of railway lines and other assets, resolving railway crossings disputes, and providing costing assistance to arbitrators in final offer arbitrations. In each case, the cost of capital rate determined by the Agency is applied to the net book value of railway company assets to determine the cost of capital for that determination.

Transportation Modernization Act and the 2019 interswitching rates Determination

[17] On May 23, 2018, the Transportation Modernization Act, S.C., 2018, c.10, received Royal Assent. It amended the CTA to require the Agency to consider a new element under paragraph 127.1(2)(b):

(2) In determining an interswitching rate, the Agency shall take into consideration

…

(b) any long-term investment needed in the railways.

[18] The Transportation Modernization Act also introduced the requirement, in subsection 127.1(1) of the CTA, that the Agency set interswitching rates annually, by determination rather than by regulation.

[19] On November 30, 2018, the Agency issued the 2019 interswitching rates Determination, the first of its kind in accordance with these provisions. In that Determination, the Agency indicated that it took account of paragraph 127.1(2)(b) of the CTA by providing an allowance for the cost of capital and for depreciation of the railway company's assets in the interswitching rates.

[20] In paragraph 19 of the 2019 interswitching rates Determination, the Agency explained the following:

…. Compensating railway companies with the full economic costs of their operations supports their long-term economic viability in the market. The Agency captures the implicit costs of the Canadian National Railway Company (CN) and the Canadian Pacific Railway Company (CP) through the cost of capital allowance and a depreciation allowance in its costing methodology.

[21] At the same time, the Agency stated the following in paragraph 13 of the 2019 interswitching rates Determination:

The Agency intends to launch consultations in early 2019 to examine the methodology used to set interswitching rates. These consultations will examine a broad range of methodological questions and factors, including car block categories, regional rates, rates developed for federally-regulated short-line railway companies, and the Agency's cost of capital methodology.

2019 interswitching rates consultation

[22] On June 20, 2019, the Agency launched consultations on its approach to setting regulated interswitching rates (consultation). As it is indicated in the consultation's discussion paper, the methodology that the Agency uses to estimate the cost rate of common equity would be reviewed as a result of concerns raised by railway companies that the cost of capital rate produced by the Agency is too low, and given the time that had elapsed since the 2011 cost of capital Decision.

[23] The Agency solicited input on two key issues related to the calculation of the cost rate of common equity (set out in issue 4, question 8 of the discussion paper):

1. Which cost of equity model, or combination of cost of equity models, should be considered by the Agency; and
2. Whether the current CAPM formulation is appropriate for determining an adequate rate of return on investment.

[24] The consultation process included eight bilateral meetings with representatives from Canadian Class 1 railway companies, a short-line railway companies association, industry associations, grain shipper associations, associations representing logistics and freight management, users of freight rail services, and other industry experts.

[25] In addition, the Agency received eleven written submissions from stakeholders, which can be found on the Agency's consultation website. Ten of those submissions addressed the cost of equity issues identified in the questions above, with specific comments from CP; CN; and McMillan LLP, representing Teck Resources Limited, the Western Grain Elevator Association, the Canadian Canola Growers Association, the Mining Association of Canada and the Western Canadian Shippers' Coalition (McMillan).

[26] The Agency has considered the issues related to the cost of equity model based on a review of the submissions received, previous decisions, existing data, an economic literature review, and other regulators' practices.

Which cost of equity model or combination of cost of equity models should be considered by the Agency?

Description

[27] Currently, the Agency uses the CAPM, which is represented by the equation:

$R_e=R_f+\beta MRP$

where:

• $R_e$ is the cost rate of common equity or the expected return of the investment;
• $R_f$ is the risk-free rate of return;
• $MRP$ is the historical market risk premium or excess stock market return over the risk-free rate; and
• $\beta$ is the measure of undiversifiable risk associated with the railway company's stock returns.

[28] The inputs for each component were determined, along with the reasons for their selection, as part of the 2011 cost of capital Decision. The specific inputs for each component can be found in Appendix A to the 2011 cost of capital Decision.

Stakeholder views

[29] In response to this issue, there was strong support from stakeholders to continue estimating the cost rate of common equity of the railway companies using the CAPM.

[30] CP submitted that while there are shortcomings of the CAPM, "in the absence of any new model, the traditional CAPM provides the most pragmatic measure of the cost of equity at this time."

[31] CN submitted that it does not take issue with the CAPM, but rather with the determination of its key variables, namely the beta, which, in CN's opinion, should be a "pure market‑observable" rate with no alteration.

[32] McMillan submitted that the Agency's current CAPM is appropriate for determining an adequate rate of return on investment, and that the constant growth Discounted Cash Flow (DCF) model and the multi-stage DCF model (two potential alternatives) have severe deficiencies for use in regulatory proceedings as:

• DCF models are unlikely to provide accurate results when used in a rigid formula; and
• the procedure used for estimating inputs for the model, in particular growth rates, is unreliable.

[33] Fertilizer Canada indicated that there has not been enough change, since the last review, to justify moving to a new model.

[34] The Forest Products Association of Canada submitted that the current cost of equity model used by the Agency is appropriate.

Analysis and determinations

[35] In 2011, the Agency conducted a comprehensive review of different cost of common equity models including the DCF Model, the Equity Risk Premium Model, and combinations of different models including the averaging of the CAPM and the multi‑stage DCF Model. After considering the merits of all models, as well as considering stakeholder views, the Agency found that the traditional CAPM best met its requirements.

[36] As noted above, stakeholders in the recent consultation also expressed broad support for, and acceptance of, the CAPM.

[37] In these circumstances, where no other theoretical models were suggested as being superior during the consultation, the CAPM will continue to be used to estimate the cost rate of common equity for federally-regulated railway companies.

Is the current CAPM formulation appropriate for determining an adequate rate of return on investment?

[38] With respect to this issue, following a review of stakeholder submissions and the literature, the Agency is of the opinion that there are four CAPM inputs that need to be assessed to ensure that the long-term investment needs of the railway companies are met: (1) the term to maturity for the risk-free rate; (2) the time frame for data used in determining the market risk premium; (3) the use of Canadian and U.S. market data; and (4) the process of adjusting CN's and CP's annual estimated betas. Each of these issues is addressed in detail below.

1) The term to maturity for the risk-free rate

Description

[39] The Agency's current methodology uses 3‑to‑5‑year Government of Canada marketable bond yields as its proxy for the risk-free rate applied in the CAPM calculation for determining the Canadian cost of common equity rate. In the case of its determination of the U.S. cost of common equity, the Agency uses 3-year and 5-year U.S. Government Treasury marketable bonds to develop separate U.S. cost of common equity estimates based on the simple average of the two.

Stakeholder Views

[40] McMillan submitted that the risk-free rate's maturity should match the period of regulation regardless of whether the railway industry has a long-term investment horizon. That is, if the cost of capital is being determined for one year, the one-year Government of Canada bond yield should be used.

[41] However, McMillan also indicated that as short-term bond yields are more volatile than the long-term yields, it is understandable that a regulatory agency might consider stability of rates to be desirable in its determinations. Based on these considerations, McMillan agrees with a yield-to-maturity in the range of three to five years.

Analysis and Determinations

[42] The difference between the rates on short and long-term bonds can be sizeable and can therefore have a considerable impact on the CAPM's calculated outcome. As a general rule, long-term securities have higher yields compared to short-term securities as investors require higher returns for their money to be held longer. This pattern can, however, reverse in times of economic uncertainty.

[43] Given the importance of this component in the calculation of the cost rate of common equity, the Agency assessed the appropriateness of the three-to-five-year range for its risk-free rate through a literature review and an assessment of other regulators' practices.

[44] Academic literature on the selection of a term to maturity for risk‑free rates identifies that both short-term and long-term bonds offer advantages and disadvantages when it comes to the CAPM calculation.

[45] The returns on short-term bonds are more responsive to market changes. This feature increases their ability to communicate current market conditions to the calculation of the cost of common equity. However, this also means that short-term bonds will produce more volatile outcomes year-over-year, which is not conducive to planning by railway companies, shippers or producers. Short-term securities are also less exposed to interest rate and inflationary risks as the period over which interest rates can vary is relatively short.

[46] Long-term bonds are viewed as more stable, as they typically follow short‑term trends but in a less extreme way. The literature has also indicated that an important assumption of the CAPM is that it reflects a single period equilibrium with investors having the same investment horizon. By this reasoning, the risk-free rate's maturity should match the investment horizon of the company for which the cost of equity is being estimated, which would support the use of longer terms to maturity for CN and CP as the majority of their investments are of a long-term nature.

[47] With respect to the practices of other regulators, the Agency's past approach of using the three-to-five-year range represents a shorter time frame than those used by the Surface Transportation Board (STB) in the United States of America, the Economic Regulation Authority of Western Australia (ERA), and the Australian Economic Regulator (AER), all of which use 10-year or greater maturities.

[48] When making the choice on the length of the term to maturity, the Agency seeks to strike a balance between responsiveness and stability. Additionally, the Agency considers that it is important that the risk-free rate's term to maturity takes into consideration the regulated companies' investment horizons, as it is the returns on these investments that is required by the investors for them to continue to provide the railway companies with funds for their long-term investment needs.

[49] Based on these criteria, the Agency determines the input of the risk-free rate in the CAPM for the Canadian cost of common equity as the 5‑to‑10‑year Government of Canada marketable bond. For the U.S. cost of common equity, the Agency will use the 5‑year and 10-year U.S. Government Treasury marketable bonds to develop separate cost of common equity estimates and take the simple average of the two. This approach produces an acceptable balance between responsiveness and stability, while ensuring a better match between the risk-free rate's term to maturity and the life of the regulated assets.

[50] The Agency notes that the risk-free rate within the MRP (discussed below) will also have a yield-to-maturity in the range of five to ten years to align with the change in the risk-free rate within the CAPM equation.

2) The time frame for data used in determining the market risk premium

Description

[51] Since the 2011 cost of capital Decision, the Agency has used as much historical data as possible, subject to the availability of reliable data, to calculate the MRP. Accordingly, the Agency uses an averaging period that comprises return data from 1951 to the current year in the calculation of the MRP in the Canadian CAPM, and from 1954 to the current year in the calculation of the MRP with respect to the U.S. CAPM.

[52] Currently, the Agency calculates the MRP for the CAPM in the following way:

$MRP=\genfrac{}{}{0.1ex}{}{1}{T}{\displaystyle \sum _{t=1}^T}(R_{m,t}-R_{f,t})$

where:

• $MRP$ is the market risk premium;
• $R_{m,t}$ is the market return in year t.
• $R_{f,t}$ is the risk-free return in year t; and
• $T$ represents the number of years over which historical market risk premiums are averaged arithmetically.

Stakeholder Views

[53] McMillan submitted that the Canadian MRP should not take into consideration data prior to 1957, as it relies on data from TSE Corporates, which it feels is of lesser quality than data from the S&P/TSX Total Return Index. McMillan submitted that the Canadian MRP should comprise data from 1957 to the current year, instead of from 1951 onwards.

Analysis and Determinations

[54] In considering McMillan's claim that Canadian market data prior to 1957 are not as reliable as data from the S&P/TSX Total Return Index, which began in 1957, the Agency's analysis of the results over that period for the Canadian and the U.S. markets shows that the market data behave similarly over the period in both jurisdictions. This outcome suggests that Canadian data are equally reliable throughout the period in question.

[55] Accordingly, the Agency determines that it will continue to use an averaging period that comprises return data from 1951 to the current year in the calculation of the MRP in the Canadian CAPM, and from 1954 to the current year in the calculation of the MRP with respect to the U.S. CAPM.

3) The use of Canadian and U.S. market data

Description

[56] Currently, the Agency calculates the cost rate of common equity for CN and CP, for the Canadian market and the U.S. market, and averages them together using a weighted average based on the volume of shares for each railway company that are traded on the Toronto Stock Exchange (TSE) and the New York Stock Exchange (NYSE), respectively.

[57] Prior to 2011, the Agency used only Canadian market data for the cost of common equity estimations. However, the Agency stated the following in the 2011 cost of capital Decision:

The Agency acknowledges that North American financial markets are increasingly integrated and that the railway companies raise capital on an enterprise-wide basis in both the Canadian and U.S. markets, with no distinction being made in the raising of that capital that it will be used exclusively in one market or the other. Further, the Agency recognizes that in this environment CN and CP have to respond to the return expectations of investors in both markets. Therefore, the Agency finds that the concept of incorporating the use of both Canadian and U.S. market data into the CAPM calculation is justified and reasonable, and allows for a better assessment of comparable risk.

Stakeholder Views

[58] McMillan submitted that the averaging methodology of the Canadian and U.S. cost of common equity rates used by the Agency is not an accurate representation of the two markets due to the fact that it is inconsistent with the CAPM theory, that there is no theoretical relationship between share volumes and market weights in the CAPM theory, and that there are major differences in taxation between the two countries for which the approach does not properly account.

[59] McMillan further submitted that as the STB does not use Canadian market data in its estimates of the cost rate of common equity of Class 1 railway companies under its jurisdiction, the Agency should not use U.S. data in its Canadian regulatory determinations.

Analysis and Determinations

[60] While this issue is rather unique based on the Canadian regulatory context, the Agency notes that other Canadian regulators have grappled with the impact of the U.S. market on their regulatory outcomes.

[61] For example, the Alberta Utilities Commission (AUC), in its a, found that despite differences in regulatory regimes in Canada and the United States of America, the integration of the North American market meant that U.S. data should be considered alongside Canadian market data in their determinations.

[62] Similarly, in its 1998 Decision CRTC 98-2, the Canadian Radio-television and Telecommunications Commission (CRTC) adopted a benchmark rate of return for all telephone companies subject to a rate cap. The CRTC stated that in view of the trend of national markets' convergence towards integration, weighted consideration should be given to U.S. data.

[63] In considering McMillan's criticism of the averaging of results for Canada and the United States of America, the Agency notes that CN and CP are listed on both the TSE and the NYSE, and compete for capital in both the Canadian and U.S. markets. As CN and CP rely on foreign investors along with Canadian investors, the use of solely Canadian market data is unlikely to be reflective of the true cost of common equity as it does not take into consideration the relevant opportunity costs of all potential investors in each jurisdiction.

[64] The Agency also notes that while the CAPM is often developed in the United States of America based on U.S. data only, the size and diversity of the U.S. market means that for many American investors, the U.S. stock market may provide a reasonable proxy for the classical market portfolio. Nonetheless, even in the U.S. market, alternatives have evolved to allow for consideration of international capital flows, and the availability of global funds.

[65] Given that the Canadian market is smaller than its U.S. counterpart, the Agency, along with the CRTC and the AUC, have recognized the need to pursue a model that takes into consideration the impact of U.S. markets on the funding requirements of Canadian regulated firms.

[66] In 2011, the Agency reviewed and rejected options including an International CAPM (ICAPM) and a North American CAPM (NACAPM) before settling on the approach of determining a separate cost of common equity for the Canadian and the U.S. submarkets, and merging those country‑specific results.

[67] As in 2011, the Agency finds the modelling of the ICAPM and the NACAPM to be impractical and lacking transparency. It would require a high-degree of subjective input, and would not be easily replicable. Therefore, the Agency rejects the use of the ICAPM and the NACAPM.

[68] At the same time, the proposed use of CAPM results for Canada only would not provide an accurate portrayal of the investment landscape as both CN and CP raise money in both markets, from investors whose expectations are influenced by investment returns in the U.S. market. As a result, the calculation of the cost rate of common equity must take into account the risk that the investors face in each of those markets.

[69] Given the challenges associated with the application of any one model to the relatively unique Canadian situation as described above, the Agency finds that continuing the use of relative volumes of shares traded is the most reasonable approach to averaging the Canadian and U.S. cost of common equity rates. This approach represents a pragmatic way to assess the impact of a broader North American capital market on railway companies' long-term investment needs. It is replicable, reasonably simple, and all the required information is well understood by stakeholders.

4) The process of adjusting CN's and CP's annual estimated betas

Description

[70] In the 2011 cost of capital Decision, the Agency found that it was necessary to adjust beta estimates for mean reversion by using the Blume adjustment, as the Agency uses the beta to forecast a cost of capital rate for a future period in the majority of its regulatory determinations. The review performed in 2011 showed evidence of the beta reverting to a mean of 1.0 between 2001 and 2010 using U.S. data.

[71] The Blume adjustment also adjusts for interest rate sensitivity of regulated companies, conformance with methodologies used by analysts and commercial data services, and beta estimation errors.

Stakeholder views

[72] McMillan submitted that the Agency's current methodology of estimating the beta using five years of weekly return observations and then adjusting it for regression towards the mean using the Blume adjustment is a reasonable approach to measuring the beta.

[73] CN submitted that the beta should be a "pure market-observable" rate with no alteration.

Analysis and Determinations

[74] As stakeholders did not raise any issue with the Agency's use of weekly returns over a five-year period to estimate the beta, the Agency will continue to calculate the beta using the methodology set out in the 2011 cost of capital Decision. However, as set out below, the Agency has assessed the issue of whether the beta should continue to be adjusted, or should instead be a market-observable rate with no alteration.

[75] The Agency calculates two cost of capital rates that are forward-looking rates:

• cost of capital rate for the development of interswitching rates; and
• cost of capital rate for the transportation of western grain, to be used in the VRCPI calculation in accordance with subsection 151(1) of the CTA.

[76] Prior to the coming into force of the Transportation Modernization Act, the Agency was required to review the interswitching regulations (Railway Interswitching Regulations, SOR/88-41, as amended), and the interswitching rates therein, every five years (or when circumstances warranted such a review). Due to the length of time typically required to effect regulatory change, this sometimes resulted in a time lag between the time when the interswitching rates were calculated and the time when regulatory amendments changing the rates came into effect. In this situation, beta estimation errors or the mean reversion tendency of the beta presented potentially significant forecasting errors in the cost of equity portion of the interswitching rates, thus requiring the adjustment of betas in past determinations.

[77] As noted above, one of the amendments to the CTA in the Transportation Modernization Act now requires the Agency to update the interswitching rates annually, by determination. This will reduce the possibility of forecasting errors and the potential problems associated with the mean reversion tendency of the beta.

[78] As such, the Agency determines that an adjustment of the beta is no longer required. The Agency will rely on the raw beta based on the methodology set out in the Appendix in its calculation of the cost of common equity rate for CN and CP. Effectively, this means that the beta will be a "pure market-observable" rate, with no alteration, consistent with CN's proposal.

CONCLUSION

[79] The Agency finds that the CAPM remains appropriate for determining an adequate rate of return on investment. Accordingly, the Agency will continue to use the CAPM for determining the cost rate of common equity for federally-regulated railway companies.

[80] Certain components of the CAPM will be adjusted as follows:

1. For the Canadian cost rate of common equity, the risk-free rate (other than the risk-free rate in the beta calculation) will be determined as the yield on 5‑to‑10‑year Government of Canada marketable bonds.
2. For the U.S. cost rate of common equity, two distinct risk-free rates (other than the risk-free rate in the beta calculation) will be determined using the yields on 5‑year and 10-year U.S. Treasury bonds.
3. The beta will not be adjusted for mean reversion using the Blume adjustment or any other adjustment.

[81] The methodology that will be used by the Agency to determine cost of equity rates, effective immediately for the determination of the 2020 regulated interswitching rates, is set out in the Appendix.

[82] The methodology presented in Appendix A will also be applied by the Agency in all cost of capital determinations, including in the MRE.

[83] In the 2011 cost of capital Decision, the Agency stated that the methodology will be applied until at least 2018. The Agency does not intend to review the methodology affirmed in the present Determination until at least 2024. However, the Agency may make adjustments to that methodology in the event that circumstances warrant.

ADDITIONAL MATTER

[84] CN submitted that it is concerned with the calculation of the weighted average cost of capital, which it believes disproportionately inflates debt levels relative to consolidated actual results, leading to a lower cost of capital rate that is not commercially fair and reasonable.

[85] In Decision No. LET-R-41-2019, the Agency determined the inappropriateness of using CN's consolidated company's balance sheet without providing a detailed reconciliation between CN's parent company's audited balance sheet and the regulatory balance sheet.

[86] This matter is currently with the Agency for determination.

Appendix to Determination no. R-2019-229

This Appendix describes in detail the methodology and the data sources that the Agency will use for making cost of capital determinations for federally regulated railway companies. Cost of capital determinations are made annually for CN and CP, and as required for other railway cost companies in regulatory cost proceedings before the Agency.

For CN and CP, the Agency will make three separate cost of capital determinations annually:

1. Cost of capital rate for the transportation of western grain, to be used in the composite price index calculation under subsection 151(1) of the CTA;
2. Cost of capital rate for the development of interswitching costs and rates; and
3. Cost of capital rate to be used for regulatory purposes other than for calculating the western grain composite price index and establishing regulated interswitching rates.

With the exception of the risk-free rate of return defined below, all of the elements necessary for the development of the three cost of capital rates will be those determined annually in the cost of capital rate for the transportation of western grain.

All balance sheet items, including current assets, current liabilities, long-term debt, shareholder's equity, deferred taxes, investment tax credits, deferred charges, and other accounting items, are as defined in the Uniform Classification of Accounts and Related Railway Records (2014) [UCA] unless otherwise stated.

Asset base

The asset base will be the net rail investment. The net rail investment is defined as the gross book value of all the railway company's assets, less accumulated depreciation, and represents the portion of the railway company's net assets that are providing rail transportation services and are under the Agency's jurisdiction. The net rail investment includes an amount for working capital. The net rail investment will be determined using book values. Based on their most recent year-end financial statements, each railway company will make a submission with respect to their net rail investment, which will be verified by the Agency.

The net rail investment or the asset base comprises the following non-current assets, as determined by the Agency, from the regulatory balance sheet of the railway company:

1. The net book value of property assets, defined as Property (UCA Account 29) less Accumulated Amortization – Property (UCA Account 33);
2. Adjustments to the net book value of property assets to:
   • exclude Used Track Materials in Store;
   • exclude Donations and Grants (as defined in the 1985 Cost of Capital Review); and
   • include Other Accumulated Depreciation (as defined in Agency Decision No. 125-R-1997);
3. The working capital allowance, which will be defined as the average of 12 monthly estimates, each monthly estimate being an average of the beginning and closing balances of Current Assets (UCA Accounts 1 to 17) less the average of the beginning and closing balances of Current Liabilities (UCA Accounts 41 to 59);
4. The Net Pension Asset as defined in Agency Decision No. 97-R-2012;
5. Deferred Charges (UCA Account 35, excluding the Net Pension Asset); and
6. Other non-current assets:
   • Long-Term Intercorporate Investments (UCA Account 27); and
   • Construction in Progress (UCA Account 31).

Capital structure

The capital structure describes the proportions of the various sources of capital used to finance the net rail investment, and will comprise long-term debt, shareholders equity and deferred taxes. The capital structure will be determined using book values. Based on their most recent year-end financial statements, each railway company will make a submission with respect to their capital structure that will be verified by the Agency.

The sources of financing of the net rail investment comprises the following non-current liabilities, as determined by the Agency, from the regulatory balance sheet of the railway company:

1. The following non-current liabilities included in the capital structure at the cost rate of debt:
   1. Long-term
   2. Other Deferred Credits – Long-term (UCA Account 69);
   3. Investment Tax Credits (UCA Account 74);
   4. Stock-Based Employee Compensation Liabilities – Non-current (UCA Account 75);
   5. Deferred downsizing, restructuring or environmental accrual costs not included elsewhere.
2. The following included in the capital structure at the cost rate of common equity:
   1. Share Capital (UCA Account 81);
   2. Contributed Surplus [UCA Account 83];
   3. Retained Earnings (UCA Account 85);
   4. Stock-Based Employee Compensation Liabilities – Non-current (UCA Account 75);
   5. Net Investment in Rail Assets (UCA Account 87); and
   6. Minority Shareholders' Interest in Subsidiary Companies (UCA Account 71).
3. Net cash balances or intercompany transaction balances not deducted from long term debt.

Cost rate of debt

The cost rate of long-term debt will be the weighted rates of interest or premiums actually paid by the company on its debt instruments, as recorded in each railway company's most recent year end financial statements and submitted to the Agency. These submissions will be verified by the Agency prior to its determination of each railway company's cost rate of long-term debt.

Cost rate of deferred taxes

Deferred taxes will be allocated a zero cost rate, as they represent a source of capital for which there are no interest premiums or shareholder obligations payable.

Cost rate of common equity

The cost rate of common equity for each railway company will be determined separately for the Canadian and the U.S. markets using the Capital Asset Pricing Model (CAPM).

The CAPM is defined as:

$R_e=R_f+\beta MRP$

where:

• $R_e$ is the cost rate of common equity or the expected return of the investment;
• $R_f$ is the risk-free rate of return;
• $MRP$ is the historical market risk premium or excess stock market return over the risk-free rate; and
• $\beta$ is the measure of undiversifiable risk associated with the railway company's stock returns.

The Canadian cost rate of common equity will be determined by calculating a single cost rate of common equity using 5 to 10 year Government of Canada bond yields for the risk-free rate of return. The U.S. cost rate of common equity will be determined by calculating two distinct cost rates of common equity, one using 5-year U.S. Treasury bond yields for the risk-free rate of return, and the other using 10-year U.S. Treasury bond yields for the risk-free rate of return, and taking an average^{Footnote 1} of the two rates.

The methodology for determining each of the components of the CAPM is specified below.

Risk-free rate

For the Canadian cost rate of common equity, the risk-free rate will be determined as the yield on 5 to 10 year Government of Canada marketable bonds ^{Footnote 2}. For determining the cost of capital rate for the transportation of western grain, the value used will be the average of daily yield observations of 5 to 10 year marketable bonds for the month of January, as published by the Bank of Canada. For the cost of capital rate for the development of interswitching costs and rates, the value used will be the average of daily yield observations for the month of September, as published by the Bank of Canada. For the cost of capital rate for other regulatory purposes, the value used will be the average yield for the calendar year for which the determination is being made, as published by the Bank of Canada.

For the U.S. cost rate of common equity, two distinct risk-free rates will be determined using the yields on 5-year and 10-year U.S. Treasury bonds ^{Footnote 3}. For determining the cost of capital rate for the transportation of western grain, the values used will be the average of daily yield observations for the month of January on each of the 5-year and 10 year U.S. Treasury bonds, as published by the Federal Reserve. For the cost of capital for the development of interswitching costs and rates, the values used will be the average of daily yield observations for the month of September, on each of the 5-year and 10-year U.S. Treasury bonds, as published by the Federal Reserve. For the cost of capital rate for other regulatory purposes, the values used will be the average yields for the calendar year for which the determination is being made, on each of the 5-year and 10-year U.S. Treasury bonds, as published by the Federal Reserve.

Market risk premium

The market risk premium is the arithmetic average of annual excess returns of the stock market over the risk-free rate, as measured from the year 1951 to the present for the Canadian cost rate of common equity, and from the year 1954 to the present for the U.S. cost rate of common equity. It will be estimated using the equation:

$MRP=\genfrac{}{}{0.1ex}{}{1}{T}{\displaystyle \sum _{t=1}^T}(R_{m,t}-R_{f,t})$

where:

• $R_{m,t}$ is the market return in year t. For the Canadian cost rate of common equity, this value is derived as the percentage change in the closing value of the S&P/TSX Composite Total Return Index ^{Footnote 4} in year t over the closing value of the same market index in year t-1. For the U.S. cost rate of common equity, this value is derived as the percentage change in the closing value of the S&P 500 Composite Total Return Index^{Footnote 5} in year t over the closing value of the same market index in year t-1. Each market return in year t will be estimated using the equation:

$R_{m,t}=\left(\genfrac{}{}{0.1ex}{}{I_{m,t}}{I_{m,t-1}}\right)-1$

where:

• $I_{m,t}$is the closing stock market index value in year t; and
• $I_{m,t-1}$is the closing stock market index value in year t-1.
• $R_{f,t}$ is the risk-free return in year t. For the Canadian cost rate of common equity, this will be estimated as the month of January income return on 5 to 10 year Government of Canada marketable bonds^{Footnote 6} in year t. For the U.S. cost rate of common equity, two sets of risk-free returns comprising the month of January income returns on each of 5-year and 10-year U.S. Treasury bonds ^{Footnote 7} ,will be estimated for each year t.
• $T$ represents the number of years for which reliable market data on market returns and bond yields are available, and will comprise the years 1951 to the present for the Canadian cost rate of common equity, and 1954 to the present for the U.S. cost rate of common equity.

Beta

The Beta for each railway company will be estimated using the most recent five years of weekly data ^{Footnote 8} in the regression equation:

$R_{c,w}-R_{f,w}={\alpha }_c+{\beta }_c(R_{m,w}-R_{f,w})+\epsilon$

where:

• $R_{c,w}$ for the Canadian cost rate of common equity is the weekly return on the stock of company c on the Toronto Stock Exchange in week w ^{Footnote 9} , derived as the percentage change in the company's stock price in week w; and for the U.S. cost rate of common equity is the weekly return on the stock of company c on the New York Stock Exchange in week w ^{Footnote 10} , derived as the percentage change in the company's stock price in week w. Each weekly observation is estimated as:

$R_{c,w}=\left(\genfrac{}{}{0.1ex}{}{P_{c,w}}{P_{c,w-1}}\right)-1$

where:

• $P_{c,w}$ is the closing price of company c in week w; and
• $P_{c,w-1}$ is the closing price of company c in week w-1.
• $R_{f,w}$ for the Canadian cost rate of common equity is the weekly income return on 3 month Government of Canada marketable bonds ^{Footnote 11} , and for the U.S. cost rate of common equity is the weekly income return on 3 month U.S. Treasury bonds ^{Footnote 12} .
• $R_{m,w}$ for the Canadian cost rate of common equity is the weekly return on the S&P/TSX Composite Index, derived as the percentage change in the closing value of the S&P/TSX Composite Index^{Footnote 13} , and for the U.S. cost rate of common equity is the weekly income return on 3 month U.S. Treasury bonds in week w over the closing value of the same market index in week w-1, and for the U.S. cost rate of common equity is the weekly return on the S&P 500 Index, derived as the percentage change in the closing value of the S&P 500 Index ^{Footnote 14} in week w over the closing value of the same market index in week w-1.

Each market return observation will be estimated using the equation:

$R_{m,w}=\left(\genfrac{}{}{0.1ex}{}{I_{m,w}}{I_{m,w-1}}\right)-1$

• $I_{m,w}$ is the closing stock market index value in week w; and
• $I_{m,w-1}$ is the closing stock market index value in week w-1.
• ${\alpha }_c$ is a regression parameter using Ordinary Least Squares (OLS) regression;
• ${\beta }_c$ is the company beta and is a regression parameter in the (OLS) regression; and
• $\epsilon$ is a stochastic error term.

Canada/U.S. cost rate of common equity

The Canada/U.S. cost rate of common equity will be the weighted average of the Canadian and the U.S. cost rates of common equity, with the weights for each railway company, based on the volume of shares traded on the Toronto and New York stock exchanges, respectively ^{Footnote 15} . Weights will be determined as the relative proportions of the daily trading volumes of each company on the Toronto and New York stock exchanges during the most recently ended calendar year. Specifically, the Canada/U.S. cost rate of common equity will be estimated as:

Canada/US Cost Rate of Common Equity = $w_1{C}_1+w_2{C}_2$

where:

• $C_1$ is the Canadian cost rate of common equity;
• $C_2$ is the U.S. cost rate of common equity;
  $w_1=\genfrac{}{}{0.1ex}{}{{\displaystyle \sum _{d=1}^n}{V}_{T,d}}{{\displaystyle \sum _{d=1}^n}(V_{T,d}+V_{N,d})}and{w}_2=\genfrac{}{}{0.1ex}{}{{\displaystyle \sum _{d=1}^n}{V}_{N,d}}{{\displaystyle \sum _{d=1}^n}(V_{T,d}+V_{N,d})}$
• $V_{T,d}$ is the volume traded on the Toronto Stock Exchange on day d, for each day of the most recently ended calendar year;
• $V_{N,d}$ is the volume traded on the New York Stock Exchange on day d, for each day of the most recently ended calendar year; and
• $n$ is the number of trading days in the most recently ended calendar year.

Tax adjusted cost rate of common equity

An income tax allowance, based on the railway companies' submitted Canadian statutory federal and provincial income tax rate, will be added to the Canada/U.S. cost rate of common equity to adjust the value of the Canada/U.S. cost rate of common equity to a before-tax basis. No income tax allowance will be applied to the cost rate of debt as it is income tax deductible.

The tax adjusted cost rate of common equity for each railway company will be determined using the formula ^{Footnote 16} :

Tax adjusted Cost Rate of Common Equity = $\genfrac{}{}{0.1ex}{}{Canada/USCostRateofEquity}{1-CanadianCorporateIncomeTaxRate}$

Weighted average cost of capital

The proportion of each type of funding in the capital structure will be used to weight each cost rate, the sum of which will be the weighted average cost of capital (WACC) expressed in percentage terms. It will be determined using the formula:

$WACC=W_{E}CoE+W_{D}CoD+W_{DT}CoDT$

where:

• $W_E$ is the proportion of equity in the book value of net rail investment;
• $CoE$ is the tax adjusted cost rate of common equity determined as described above;
• $W_D$ is the proportion of long-term debt in the book value of net rail investment;
• $CoD$ is the cost rate of long-term debt determined as described above;
• $W_{DT}$ is the proportion of deferred taxes in the book value of net rail investment; and
• $CoDT$ is the cost rate of deferred taxes, which will be set to a value of zero (0).

Capital structure weights will be based on equating the book value of the net rail investment as defined under the title Asset base to the book value of the sources of capital as defined under the title Capital structure.

The capital structure weights are the percentage amounts of each source of capital (under the title Capital structure, in a., b., and c.) used to finance the asset base.