Determination No. R-2017-198
DETERMINATION by the Canadian Transportation Agency (Agency) of the methodology to be used by federally-regulated railway companies to determine the working capital amounts and capital structure for regulatory purposes.
 In accordance with its findings on the issues involved in determining the capital structure for regulated railway companies as described in this Determination, the Agency defines the Balance Sheet Method for determining the capital structure of regulated railway companies.
 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 (UCA) unless otherwise stated. Appendix A to this Determination lists the UCA accounts included in the regulatory balance sheet.
Net rail investment
 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:
- The net book value of property assets, defined as Property (UCA Account 29) less Accumulated Amortization – Property (UCA Account 33); the accounts included in UCA Account 29 and UCA Account 33 are listed in Appendix B to this Determination;
- 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 1-17) less the average of the beginning and closing balances of Current Liabilities (UCA Accounts 41-59);
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).
Sources of capital
 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:
- The following non-current liabilities included in the capital structure at the cost rate of debt:
i. Long-term debt (UCA Account 65);
ii. Lease Obligations (UCA Account 67);
b. The following non-current liabilities included in the capital structure at a cost rate of zero percent:
i. Deferred Liabilities (UCA Account 61);
ii. Future Income Taxes (Non-Current) [UCA Account 63];
iii. Other Deferred Credits – Long-term (UCA Account 69);
iv. Investment Tax Credits (UCA Account 74);
v. Stock-Based Employee Compensation Liabilities – Non-Current (UCA Account 75);
vi. Deferred downsizing, restructuring or environmental accrual costs not included elsewhere;
c. The following included in the capital structure at the cost rate of common equity:
i. Share Capital (UCA Account 81);
ii. Contributed Surplus (UCA Account 83);
iii. Retained Earnings (UCA Account 85);
iv. Net Investment in Rail Assets (UCA Account 87); and,
v. Minority Shareholders’ Interest in Subsidiary Companies (UCA Account 71);
d. Net cash balances or intercompany transaction balances not deducted from long-term debt.
Capital structure weights
 Capital structure weights will be based on equating the book value of the net rail investment as defined under the title Net rail investment to the book value of the sources of capital as defined under the title Sources of capital.
 The capital structure weights are the percentage amounts of each source of capital (under the title Sources of capital, in a., b., and c.) used to finance the net rail investment.
 The Agency annually determines the cost of capital for the Canadian National Railway Company (CN) and the Canadian Pacific Railway Company (CP) which is used to determine the volume-related composite price index (VRCPI) for the transportation of western grain, and interswitching rates, as well as for other regulatory purposes requiring costing determination. The cost of capital is the return that each company must earn for each dollar of capital invested, in order to pay income taxes and interest on its long-term debt, and provide a reasonable return to investors. As part of the cost of capital determination for a company, the Agency must determine its capital structure, which describes the proportions of various sources of financing used by the company to acquire its capital assets. The capital assets include working capital, which comprises cash and the materials and supplies required by the company to maintain day-to-day operations.
 In financial analysis, the standard methodology for determining the capital structure of a company is to first determine the total value of the company’s capital assets and set it equal to the total financing costs (long-term debt, shareholders’ equity, etc.) identified on the company’s balance sheet, then determine the proportion of each type of financing. A widely-used methodology in financial analysis, and generally considered the most accurate for determining the amount of the working capital requirements for a company, is a lead-lag studyFootnote 1 . This methodology reviews company transactions over a period of a year, to identify the time lag between when the company provides services and when it receives payment for those services, as well as when it buys and pays for materials and supplies, so as to estimate the amount of capital required on hand for day-to-day operations.
 In a formal review of cost of capital methodologies in 1985 (1985 review), the Agency permitted CP to deviate from the standard methodology because CP did not have its own balance sheet at the time, as it was part of a larger industrial conglomerate. In Letter Decision No. LET-R-98-2002 (2002 Letter Decision), the Agency directed CP to use the “balance sheet method” in future capital structure determinations as it was no longer part of a conglomerate and had its own balance sheet. However, the Agency did not define what it meant by the balance sheet methodology, and did not correct continuing CP deviations because it considered those deviations relatively minor. The Agency, though, continued to monitor the CP cost of capital submissions for materiality of the deviations.
 In Decision No. 131-R-2016, the Agency noted that the CP deviations were no longer minor, and launched a public consultation in August 2016 to solicit industry views on a proposed specification of the methodology to be used in determining the capital structure. In the same consultation, the Agency also noted that, while lead-lag studies were used to determine the working capital requirements for CN and CP in 1992, those studies have not been updated since, due to the complexity and expense of such studies. The Agency is concerned that, as a result, the 1992 working capital allowances, which are still being used in the annual Agency cost of capital determinations, may no longer be appropriate for modern railway operations.
 In 2011, the Agency conducted a detailed review of all aspects of the methodologies involved in determining the cost of capital methodologies (2011 reviewFootnote 2 ), similar in scope to the 1985 review. The issues under review in this Determination address only a small part of the issues that were considered in the 1985 and 2011 reviews. Paragraphs 7 to 25 of Decision No. 425-R-2011 (2011 Decision) places the current review in its proper context and defines the terms used in this Determination.
 With a view to prescribing a uniform and unambiguous methodology to be used to determine the capital structure for regulatory purposes of all regulated railway companies, the Agency undertook industry consultations from August to November 2016, and from April to May 2017, in order to seek industry views and comments on issues relevant to capital structure determination. Details of the consultation, including the participants consulted, the issues under consultation, and the participants who responded to the consultation, as well as summaries of the views and comments expressed by the respondents, are presented in Appendix C.
 Section 7 of the Railway Costing Regulations (RCR) provides as follows:
For the purposes of sections 264, 272, and 276 to 278 or for other purposes pertaining to rates for the carriage of goods,
(a) costs shall be variable costs […], and shall include the increases or decreases in rail operations expenses resulting from changes in the volume of traffic, after allowing a reasonable period of time for adjustment in view of the traffic to be handled [emphasis added]
(b) there shall be included in the variable costs an allowance for cost of capital based on a rate of return, including an allowance for income tax, that, in the opinion of the Committee, is appropriate for CP Rail (a Division of Canadian Pacific Limited) applied to the variable portion of the net book value of the assets related to the movement of the traffic; [emphasis added]
 Capital structure determination is a component of the cost of capital determination.
ANALYSIS AND FINDINGS
Inclusion of land in the asset base (net rail investment)
 The Agency currently includes land in the determination of net rail investment, and a review of historical cost of capital determinations since 1985 indicates that land has been included in the annual net rail investment determinations.
 Both CN and CP pointed out that the Consultation Document omitted land (UCA Account 101) from the capital assets that comprise the net rail investment. They submitted that land, as one of the capital assets used in the production of rail transportation services, should be included in the determination of the net rail investment. McMillan disagreed, submitting that the RCR exclude assets such as land from the asset base.
 McMillan did not state on what grounds it based its claim that the RCR excludes land from the asset base, but subsection 7(b) of the RCR offers possible context for this position.
 The question, as it appears from McMillan’s assertion, is whether land may not be considered a variable portion of the assets. This question is addressed by reference to the definition of variable costs in the RCR, which is provided in subsection 7(a).
 It is clear that, while a given parcel of land may support rail operations covering a wide range of traffic volumes, as traffic operations expand, there may arise a need for additional land to accommodate the increased traffic, as in, for example, when offering new services, expanding existing yards or terminals, etc., or there may arise a need to sell off existing land, as in, for example, the discontinuance of service in a given location. Land may be considered as fixed for regulatory purposes if there is evidence to show that the railway company has made no new land transactions over a long enough periodFootnote 3 in the past and, more importantly, has put measures in place to ensure that it makes no new land transactions in the future. This is not the case for CN and CP.
 This means that, while for most regulatory cost determinations involving existing services land may be considered to be fixed, on a system level, land cannot be considered fixed and is variable over a reasonable period of time to allow for adjustment in the form of purchase, lease, or sale of land to reflect the traffic to be handled. As the cost of capital is determined at the system level, land meets the criteria for a variable cost as prescribed in the RCR.
 In the Canadian Transport Commission Railway Transport Committee’s Decision on the Cost of Capital Methodology dated July 31, 1985 (1985 Decision), the Agency addressed the question of whether land should be considered variable with traffic and included in costs. All parties in the review (the Provinces, Wheat Pools, CN, CP and VIA) agreed that land has a value, and rejected the treatment of land at zero value.
 The Agency accepts CN and CP’s submission that land must be included in the capital assets to be used in the determination of net rail investment.
Values of assets included in net rail investment
 The Agency currently includes the book values of assets in its determination of the net rail investment. CP submitted that market values have the advantage of reflecting the true value of the assets at any time, and that a return on capital based on the book value of land does not provide a fair return according to the Supreme Court definition of that term. CN submitted that “book value” is an accounting device that has no bearing on required financing or investor returns, and that equipment and investment are not bought and sold at book values, but at market values. CN submitted that the entire capital structure must be based on market values.
 In the 2011 review, the Agency invited participants to submit their views on whether it should continue to use an asset base that is based on book value, or instead adopt a market value asset base. The views expressed by CN and CP in this review mirror their views expressed in the 2011 review. The Agency concluded in the 2011 review, with reasons detailed in paragraphs 78 to 84 of Decision No. 425-R-2011, that use of market values to determine net rail investment would not be reasonable, reliable, or pragmatic, and decided to continue to use book values to determine the capital assets included in net rail investment.
 The use of market values in place of book values in determining the net rail investment was not part of this consultation exercise but was extensively consulted and decided upon by the Agency in its 2011 review. The Agency rejects the use of market values for the reasons detailed in its 2011 Decision, and will continue to use an asset base that is based on book values, as prescribed in the RCR.
Use of the market values in place of book values in the capital structure
 The Agency currently uses the book values of long-term debt and shareholders’ equity in determining the capital structure.
 As part of its over-arching submission that the capital structure should be based on market values, not book values, CN argued that financing and returns are all driven by market values, and that investors who buy the company’s shares or bonds expect returns on the market value of their investment, not on their book value.
 The method of valuation of capital structure items was not part of this consultation. However, in the 2011 review, the Agency invited the industry to submit its views on whether the Agency should use market values in place of book values when determining the relative weights of long-term debt and common equity in the railway companies’ capital structure. At that time, CN had submitted similar arguments advocating a capital structure based on market values.
 The Agency presented a detailed analysis of the concept of a market value capital structure in paragraphs 85 to 112 of its 2011 Decision, including a listing of a number of substantive shortcomings associated with that approach, and concluded that the adoption of a market-based capital structure would not be reasonable, reliable or pragmatic.
 The Agency rejects the use of market values for the capital structure items and the capital structure weights for the reasons detailed in its 2011 Decision, and will continue to rely on a capital structure based on book values.
Deferred taxes, investment tax credits and deferred costs
 The Agency currently includes deferred taxes, investment tax credits, and deferred costs in the capital structure, and assigns them a zero cost rate.
 CN submitted that deferred taxes, investment tax credits and deferred costs do not belong in the capital structure, and that these are non-cash items that are required solely for accounting purposes and have no bearing on the company’s financing. Both CN and CP submitted that only long-term debt and shareholders’ equity, those sources for which cost rates are calculated, should be included in the capital structure. McMillan’s finance expert, Dr. Gould, disagreed with that position, submitting that deferred taxes comprise tax charged to customers but not paid to government until a very distant date in the future (if it is paid at all), and that accumulated deferred taxes should be considered to be a loan to the company from its customers.
 The issue of whether or not these items should be included in the capital structure was not part of this consultation. However, in the 2011 review, the Agency considered the issue of whether or not to include deferred taxes in the capital structure, and what cost rates to use. The Agency presented a detailed analysis of the issue in paragraphs 125 to 144 of the 2011 Decision, leading to its conclusion that the accumulated deferred taxes comprise a pool of capital derived from profit, that the capital functions as an interest-free loan from customers that could be used for any business purposes including investment, and that the capital must therefore be considered as a zero cost source of funds.
 The Agency had reviewed in detail the same issue of whether or not to include deferred taxes and related items in the capital structure in the 1985 review, and had come to an identical conclusion, that deferred taxes should be considered as interest-free loans from the customers on which no returns should be paid to the company, and that the same principle should apply to investment tax credits and related items.
 The Agency finds, as it did in the 1985 and 2011 Decisions, that including deferred taxes, investment tax credit, and related items in the capital structure as a source of funds, and assigning to them a zero cost rate, continues to be the most appropriate approach in the capital structure determination methodology.
Netting cash balances against long-term debt
 CP presented two principal arguments to support its proposal that the Agency should continue to allow CP to deduct its net cash balance from its long-term debt. CP suggested that, because the Agency’s 2002 order for CP to use the balance sheet approach to determine its capital structure did not define the specifics of the balance sheet approach, and because the Agency continued to accept CP’s current approach for determining CP’s capital structure, then, by implication, the Agency was accepting CP’s current approach as the definition of the balance sheet approach. CP then detailed why its current approach is justified in considering inter-company transactions as a form of debt owed by the regulated company, i.e. CP, to the parent company, i.e. CP Ltd
 With respect to the argument that the Agency has allowed the cash flow approach in the past and therefore it must have been the right approach, the Agency points out that it is not bound to continue to use an approach simply because that approach was accepted in the past. The Agency retains the flexibility to consider the merits of an approach and how it meets the Agency’s regulatory objectives as circumstances change. The Agency, in the past, considered the net cash balances netted against long-term debt by CP as not significant enough to materially affect the cost of capital determinations, but in light of recent increases in the amounts of the cash balances, the Agency is reviewing the merits of the approach based on accepted accounting and economic principles.
 CP’s proposed justification for the approach is that intercompany transactions, including cash transfers, may be considered to be debt instruments between the regulated company and its parent. However, the Agency observes that CP, in its cost of capital submissions to the Agency starting in 2010, included intercompany transactions as part of the equity and not as part of debt. That is, CP’s actual practice does not support its argument. Despite CP’s representations in this consultation about cash being part of intercompany transfers, CP files intercompany transfers under equity and cash under debt in its cost of capital submissions. The question then becomes whether the net cash balance, from any source, can be considered to be debt either from an accounting or economic perspective for regulatory purposes under the Canada Transportation Act, S.C., 1996, c. 10, as amended.
 In accounting practice, long-term debt is any debt that has a maturity greater than one year. It may be an amount borrowed for capital investment and other business purposes, and it is recorded as such on the company books. Cash is considered to be part of working capital as businesses would normally carry an amount of cash required for day-to-day operations. Therefore, CP’s inclusion of cash in long-term debt, whether or not it is as part of intercompany transactions, is contrary to standard accounting practice.
 From an economic perspective, the cost of capital is intended to recognize the cost incurred by the railway companies to make capital investments. Thus, the cost of capital covers, among others, the portion of capital that was acquired through borrowing, for which a cost rate is calculated based on the actual amount of interest incurred by the railway company on its long-term debt. For any amount to be considered as long-term debt, there must be a bona fide loan arrangement between non-arm’s length companies, with proper documents to support the loan, and with defined interest and repayment schedules. Cash satisfies none of these requirements. The net cash balance does not represent a loan on which the railway company pays interest and on which interest must be recognized for regulatory purposes in the cost of capital, nor does it represent the actual payment of a loan so as to reduce the amount of the interest.
 This means that, from both the accounting and economic perspectives, the net cash balance cannot be considered in the cost of capital determination as either long-term debt on which a cost rate is required or as a past payment to reduce long-term debt.
 CP also argued that discontinuing its current approach, a net adjustment of intercompany advances including cash against debentures, would be severely disruptive because it has built up its regulatory balance sheet over a long period using that approach.
 Starting in the 2011-2012 crop year, CP included inter-company advances in its equity section while classifying net cash against debt. CP has been following this practice for six years. Therefore, there is no discontinuity if intercompany transactions continue to be classified in equity. The Agency observes that the net cash balances that CP has used to reduce debt since 2010-2011 has resulted in an average of less than 0.02% impact on CP’s cost of capital. The Agency does not consider this impact to be material, and is not convinced by the position advanced by CP.
 The Agency finds that the net cash balance or net intercompany transactions should not be deducted from long-term debt.
Revision of the current working capital allowances
 Respondents were unanimously of the view that the 1992 amounts of the working capital allowances for CN and CP used in current cost of capital determinations may no longer reflect current railway operations and need to be updated.
 The Agency finds that, as the amounts of the working capital amounts determined in 1992 may no longer reflect the current working capital requirements by CN and CP, there is a need to remove those amounts from future capital structure determinations for CN and CP, and to replace them with updated working capital allowances determined separately for CN and CP.
Methodology for determining the working capital allowance
 The 1985 Decision discussed the definition of working capital. The Decision stated that a consensus was reached among the parties that working capital for rail regulatory purposes should be defined differently from the classical accounting definition of the term. The parties agreed to use instead the term current working assets, comprising cash and the materials and supplies required to support the day-to-day operations of the railway company. The parties also agreed that only the portion of materials and supplies financed by the railway companies’ investors should be included in the definition of current working assets.
 In this consultation, only CP referenced the 1985 definition of current working assets in place of working capital. That is, the parties in this consultation did not make the same distinction in the definition of working capital as did the parties in 1985. Respondents proposed three methodologies for determining the appropriate working capital allowance in future capital structure determinations. Agency staff proposed a fourth methodology. The Agency assessed each of the proposals against the following five criteria:
- Conformance with accepted accounting practice;
- Conformance with regulatory economic principles, that is, that railway companies’ investors should receive a return on only the amount of working capital financed by investors;
- Accuracy of the estimate for regulatory purposes;
- Ease of determination; and,
- Frequency of updating.
 MacMillan’s finance expert, Dr. Gould, advocated a lead-lag study, submitting that it would not only provide an accurate estimate of the current working capital allowance, but would also provide an assessment of the change that has taken place in the 25 to 30 years since the last study, and, most importantly, provide a benchmark against which future working capital estimates and the results of other methodology proposals can be judged. CP stated its willingness to undertake a lead-lag study, but pointed out that, due to the expense and cumbersome nature of such a study, this would imply that the working capital allowance would continue to be updated very infrequently going forward.
 A lead-lag study conforms to standard accounting practice, and is accepted and easily understood by accountants and financial analysts. It also conforms to regulatory economic principles when the sampling design for the study is structured to result in only the portion of investor-supplied working capital. A lead-lag study is generally accepted as theoretically sound, and a properly conceived and executed lead-lag study would be considered to result in the most accurate estimate of the amount of the working capital requirements.
 Nevertheless, a number of problems associated with using a lead-lag study to determine annual working capital requirements for CN and CP can be identified.
 First, as noted by both CN and CP, a lead-lag study is complicated, expensive, and time-consuming to implement, as it requires reviews of thousands of invoices, freight movements, and other sources of indirect financing. To perform this study, extensive consultations would be required between Agency and CN and CP staffs to develop an appropriate sampling design for the review of each railway company’s financial transactions in a selected year. Given existing Agency resources, the Agency estimates that the implementation of a lead-lag study in future cost of capital determinations could not reasonably be expected within two years.
 Second, the amount of time and financial resources required by both the Agency and the railway companies to undertake a lead-lag study makes it impractical to employ for any determination that requires frequent updating. It is the primary reason why a lead-lag study has not been undertaken since 1992.
 Therefore, the resources needed to implement a lead-lag study and to undertake future updates continue to make the methodology extremely impractical for use in capital structure determinations for regulatory purposes.
CP’s balance sheet formula
 CP proposed a formula for calculating the working capital allowance which it claims is based on the definition of working capital agreed with the Technical Committee in the 1985 Decision.
 CP’s proposed formula is not one currently recognized by the accounting profession, and may not be easily understood by the general public, accountants, and financial analysts. As well, the formula may require further extensive studies and reviews of railway company accounting to determine whether the formula does indeed result in the investor-financed portion of the amount of capital needed on hand to facilitate the day-to-day operations of the railway company. Further, CP’s proposal that the formula be expanded to include unspecified influences from several other expense items, including capital depreciation, taxes, labour, and intercompany transactions, renders the proposed formula too complex and opaque for a determination to be made as to whether or not it meets the intended regulatory economic principles.
 Due to the complexity and opacity of the proposed formula, the expected accuracy of the formula cannot readily be determined. In particular, the formula does not account for cash, which is a principal component of the 1985 Decision’s definition of working capital.
 Given the lack of clarity concerning the validity of the proposed formula and the need to conduct extensive reviews of railway companies’ accounting in order to establish the validity of the methodology, and given that the potential revision of the formula could bring forth numerous other expense items to consider, as proposed by CP, it is not clear how easily such a methodology can be implemented or updated.
CN’s modification to the classical accounting formula
 CN proposed that the classical accounting definition of working capital, which is current assets less current liabilities, represents an acceptable measure of the investor-financed cash and materials and supplies required for the day-to-day operations of the railway companies, and should be used to determine the amount of the working capital allowance. CN also proposed that, to dampen the volatility that may be present in a single estimate of working capital taken at the end of the fiscal year, a multi-year rolling average of two or three years could be used.
 The Agency accepts that the classical definition is recognized by the accounting profession as valid, is easily understood by the general public, accountants and financial analysts, and could be easily implemented and updated on an annual basis as it uses the standard financial accounts of the railway companies. However, it is unclear to the Agency the extent to which a multi-year rolling average of working capital requirements would be accepted by the profession to be a valid reflection of the requirements in any given year.
 None of the other respondents disputed CN’s assertion that the classical definition provides only the amount of the investor-supplied working capital. However, the classical definition results in an estimate of the working capital requirements at one point in time, which may not necessarily reflect the amount of capital required for day-to-day operations during the year. Further, CN’s proposed modification of using a multi-year average of end-of-year point estimates makes the result even less reflective of the working capital requirements during a given year. Thus, the result may not conform to regulatory economic principles.
Agency staff’s modification to the classical accounting formula
 Agency staff has proposed an alternative modification to the classical accounting definition. In this modification, the annual working capital allowance would be estimated as an average of 12 monthly estimates of the requirements, where each monthly estimate is determined as the average of the opening and closing balances of current assets for the month, less the average of the opening and closing balances of current liabilities for the month.
 This proposed modification to the classical definition would be easily understood by the accounting profession, financial analysts and the general public, and it could be easily implemented and updated on an annual basis as it uses the standard financial accounts of the railway companies. The proposed modification reflects estimates of working capital requirements taken during the year, and therefore conforms more closely to the 1985 definition of the working capital requirements.
 This modification to the classical formula might not produce a result as accurate as an estimate of the working capital requirements resulting from a lead-lag study. However, given the extensive and intensive resource requirements for a lead-lag study, which can be performed only very infrequently, as compared with the relative ease of determination and updating of the proposed modification to the classical formula, the parties may not consider the additional accuracy of the lead-lag study to be worth the additional expense and uncertainty. Considering the costs and benefits of using the staff-modified classical formula as opposed to a lead-lag study, the former appears to be more cost effective and may be deemed sufficient for the purpose of the cost of capital rate determination.
 The Agency finds that its staff’s proposed modification to the classical formula would allow the working capital allowance to be estimated with reasonable accuracy for regulatory purposes, would be easily implemented, would allow for annual updating of the estimates, and would not impose an undue burden on the railway companies.
Specification of long-term debt in the capital structure
 Considering the sources and uses of funds in the capital structure from an economic perspective, the face value of the long-term debt and any premiums received on the bonds function as sources of funds, while any discounts on the face value of the debt function as assets. This suggests to the Agency that, for the economic purpose of determining the cost of capital, long-term debt is more properly included in the capital structure at its face value, the discounts on long-term debt are more properly included as assets in the net rail investment, and premiums on long-term debt are more properly included with the other sources of funds deemed to have zero cost rate.
 This view is contradicted by CP and McMillan, who both suggested that in standard accounting treatment the unamortized premiums and discounts are used to adjust the face value of long-term debt in the capital structure, meaning that the reported long-term debt is the net value after adjustment with the premiums and discount. On the other hand, the economic view appears to be supported by CN, whose proposed balance sheet and accompanying capital structure includes the long-term debt at face value and premiums on the long-term debt as a non-current liability.
 A further complication arises from the determination of the cost rate of debt, which is weighted by the long-term debt in the cost of capital determination. The cost rate of debt is the average cost to the company of financing each dollar of debt capital. It is calculated as a weighted average of the interest rates on all the existing non-matured debt instruments issued by the company. If the net values of the debt instruments are used as the weighting factors, then the appropriate cost rates are the effective interest rates, which adjust the coupon rates for any premiums or discounts. Calculation of the effective interest rates is not simple, considering the many different bonds over time with different maturities and coupon rates issued by railway companies, and places an additional burden on the railway companies.
 However, if the face values of the debt instruments are used as the weighting factors, then the appropriate cost rates are the unadjusted coupon rates. Using the net values of debt as the weighting factors for coupon rates would result in an understatement or overstatement of the cost rate of debt, depending on whether there is a net premium or a net discount.
 In paragraphs 155 to180 of the 2011 Decision, the Agency considered in some detail the methodology for determining the cost rate of debt, and concluded:
 To the extent that the method for measuring the yield on long-term debt is attempting to reflect the actual financing cost of existing debt, the Agency finds that the coupon rate method is the most reasonable, reliable and pragmatic of the three models examined. [Emphasis added]
 Given the current Agency methodology, which specifies that the cost rate of debt must be a weighted average of the coupon rates on existing debt instruments, it is reasonable to conclude that the decision also implies that the face values of the debt instruments would be the appropriate weighting factors. That is, that the long-term debt reported in the capital structure must reflect the face values of the debt instruments, which also corresponds with the implied specification of the long-term debt in the UCA as the face value.
 The Agency finds that the appropriate specification of long-term debt in the capital structure, to conform to the long-standing definitions of assets and liabilities, and to conform to the existing Agency order regarding determination of the cost of debt rate, is the face value of the debt.
Treatment of non-current liabilities in the capital structure
 The defining feature of non-current liabilities is that the company identifies a liability that is payable in a future period. Under standard accounting conventions, once the amount of the future contingent liability can reasonably be estimated, and there is reasonable likelihood that the liability will be incurred, the company recognizes the expense in the current period and records the amount on its balance sheet as a deferred liability. The fact that the expense is recorded in the current period but not paid out immediately creates a pool of funds that can be used by the company for any business purpose, including financing capital investment, due to the timing difference. Similar to deferred taxes, this pool may be considered to be a loan to the railway company from its customers, and on which the company pays no interest.
 CN submitted that deferred liabilities and other deferred credits must be considered to be a cost-free source of funds in order to be consistent with Agency decisions on deferred taxes and similar items.
OTHER DEFERRED CREDITS – LONG-TERM
 The unamortized premiums from issuance of bonds/long-term debts represent funds received from bond purchasers in excess of the face value of the long-term debt, and for which the underlying bonds have not yet reached maturity. The interest payment reflects the bond’s stated coupon rate, which is based on the face value of the long-term debt. Other long-term deferred credits include payments received by the company in advance of future services to be provided by the company, which may be considered to be an advance loan from the customer on which the railway company pays no interest.
 Therefore, the premiums on the long-term debt and advance payments represent funds that can be used to finance net rail investment, and on which no interest or dividends are paid by the company. The same principles applied to deferred taxes apply to other deferred credits reported by CN and CP.
 The Agency finds that both deferred liabilities and other deferred credits represent sources of funds that may be used to finance capital investment, and therefore should be included in the determination of the capital structure. The Agency also finds that on both these sources of funds, the railway company does not pay any interest or dividends, and that, therefore, no interest or dividends are to be allocated to those sources of funds in the capital structure.
Regulatory Balance Sheet Accounts Current Assets
|UCA Account Number||UCA Account Description|
|5||Accounts Receivable – Trade|
|7||Other Accounts Receivable|
|9||Allowance for Doubtful Accounts|
|11||Material and Supplies|
|15||Other Current Assets|
|17||Future Income Taxes (Current)|
|UCA Account Number||UCA Account Description|
|27||Long-Term Intercorporate Investments|
|31||Construction in Progress|
|33||Accumulated Amortization − Property|
|UCA Account Number||UCA Account Description|
|49||Notes and Other Loans Payable|
|51||Income and Other Taxes Payable|
|52||Future Income Taxes (Current)|
|54||Stock-Based Employee Compensation Liabilities – Current|
|57||Long-Term Debt Maturing Within One Year|
|58||Lease Obligations Due Within One Year|
|59||Other Current Liabilities|
|UCA Account Number||UCA Account Description|
|63||Future Income Taxes (Non-Current)|
|69||Other Deferred Credits − Long-term|
|71||Minority Shareholders’ Interest in Subsidiary Companies|
|73||Donations and Grants|
|74||Investment Tax Credits|
|75||Stock-Based Employee Compensation Liabilities – Non-Current|
|UCA Account Number||UCA Account Description|
|87||Net Investment in Rail Assets|
Property Accounts in Gross Investment and Accumulated Depreciation Accounts
|Gross Investment Account||Accumulated Depreciation Account||Account Title|
|106||206||Paved Concrete Trackbed (PACT System)|
|107||207||Other Track Material|
|111||211||Track Laying and Surfacing|
|113||213||Used Track Material in Store|
|121||221||Fences, Snow Sheds and Rock Sheds|
|125||225||Other Right-of-Way Property|
|131||231||Office and Common Buildings|
|133||233||Office and Common Buildings Moveable Equipment and Machinery|
|137||237||Passenger Station Moveable Equipment|
|141||241||Roadway Building Machines and Moveable Equipment|
|143||243||Equipment Repair Shops|
|145||245||Shop Machinery and Moveable Equipment|
|151||251||Rail Communication Systems|
|159||259||Rail Freight Terminals|
|177||277||Intermodal Terminal Handling Equipment|
|179||279||Trailers, Semi-Trailers, Containers, Chassis and Bogies|
|189||289||Other Non-Revenue Rolling Stock|
INDUSTRY CONSULTATIONS SUMMARY
The Agency undertook three rounds of consultations with industry stakeholders from August to November 2016, and from April to May 2017, to solicit their views and comments on the methodology to be employed in determining the capital structure of railway companies for regulatory purposes.
In the first round of consultations in August 2016, the Agency solicited industry proposals for a definition of the Balance Sheet Approach to determine the capital structure, and proposed a list of capital items, identified on a railway company’s balance sheet, that may be included in net rail investment and in the capital structure. The Agency also asked stakeholders to propose a methodology to update the current working capital allowances that would facilitate more frequent updating. A Consultation Document, which asked stakeholders to address specific questions, was provided to consultation participants.
Stakeholders were asked to provide their initial responses by October 21, 2016. The following parties responded and made submissions with respect to the capital structure and the working capital allowance methodologies:
- Canadian National Railway Company (CN);
- Canadian Pacific Railway Company (CP);
- McMillan LLP, representing Teck Resources Limited and its affiliates Teck Coal Limited and Teck Metals Limited;
- Forest Products Association of Canada (FPAC) and Western Canadian Shippers’ Coalition (WCSC), in a joint submission;
- Western Grain Elevator Association (WGEA) in support of the McMillan LLP submission;
- Canadian Canola Growers Association (CCGA) in support of the McMillan LLP submission;
- Freight Management Association of Canada (FMAC); and,
- Saskatchewan Association of Rural municipalities (SARM).
In the second round of consultations, the Agency posted the initial responses from all parties on the it’s website, and invited all participants to submit comments on the responses by a prescribed deadline, November 18, 2016. Comments on the initial responses were received from the following participants and also posted on the Agency’s website:
- Canadian Pacific Railway Company (CP);
- McMillan LLP, representing Teck Resources Limited and its affiliates Teck Coal Limited and Teck Metals Limited;
- Western Grain Elevator Association (WGEA) and Canadian Canola Growers Association (CCGA) in support of McMillan’s comments; and,
- Aikins, MacAuley & Thorvaldson LLP (Aikins), on behalf of the Forest Products Association of Canada (FPAC) and Western Canadian Shippers’ Coalition (WCSC).
Finally, on April 11, 2017, the Agency undertook follow-up consultations with three targeted participants who had provided their views and comments in the two previous rounds, i.e. CN, CP, and McMillan, for the purpose of clarifying additional issues that arose during the review of the consultation responses. Letters were also sent to all stakeholders who had provided their opinions to the Agency in the early rounds of the consultation, to inform them of the follow-up questions that were sent to the targeted stakeholders. All three targeted stakeholders provided their responses to the consultation questions by the final deadline of May 19, 2017.
ISSUES AND STAKEHOLDER VIEWS
Issue 1: Capital structure methodology
Stakeholders were invited to provide feedback, for consideration by the Agency, on the proposed Balance Sheet Approach to determine the capital structure for CP and CN, which was defined in the Consultation Document provided to stakeholders as comprising the following items:
- That the net rail investment comprises the following components:
i. The net book values of property assets identified in the Uniform Classification of Accounts and Related Railway Records (UCA) accounts 102 to 195 (full list of accounts in Appendix B);
ii. The pension asset (UCA account 197) as defined in Agency Decision No. 97-R-2012;
iii. The working capital allowance; and,
iv. Deferred charges and other assets.
2. That the Capital Structure comprises the following components:
i. The long-term debt;
ii. The deferred taxes, investment tax credits and deferred downsizing costs; and,
iii. The shareholders’ equity.
3. That the net cash balance cannot be applied to reduce long-term debt.
The Agency has directed in the UCA that accounting practices for regulatory purposes should follow the Generally Accepted Accounting Principles GAAP unless specifically instructed to deviate from GAAP by an Agency decision. Items 1 and 2 list the asset types to be considered when determining the capital structure and the net rail investment. The components and elements of both the capital structure and the net rail investment correspond to GAAP, and are defined in previous Agency cost of capital decisions.
The respondents to the consultation were invited to provide their thoughts on the following question:
Does the proposed Balance Sheet Approach list of items adequately include all appropriate elements that the Agency should take into consideration in determining the prescribed railway companies’ capital structure, and if not, what components should be included or excluded, based on what elements of GAAP? Please provide clear justification for any elements or components that you think should be added or excluded.
ROUND 1 OF THE CONSULTATION – AUGUST 26 TO OCTOBER 14, 2016
CN indicated that it routinely buys land properties for developing its network (siding extensions, spurs to serve new customers, even entire shortlines, etc.) or sells land properties of lines no longer perceived as economically viable. According to CN, these additions and retirements of land do impact the amounts of investments that have to be financed and therefore should be taken into account in determining the net rail investment for the purpose of calculating the cost of capital. CN submitted that the net rail investment should also include the net book values of property assets identified in UCA account 101 –Land.
With respect to the capital structure components, CN submitted that deferred taxes, investment tax credits and deferred downsizing costs are adjustments required for accounting purposes and have no real bearing on the company’s sources of financing. CN submitted that these deferred costs are only accounting devices and not real funds obtained from a financing source outside the corporation, and that any amounts they represent on the balance sheet are ultimately financed by investors and therefore should attract the returns expected by these investors in the same proportion as debt and equity. As well, CN submitted that investment tax credits, similar to donations and grants, should be applied against the property investments that generated those credits to begin with, and should be used to reduce the value of properties in the determination of the net rail investment.
Finally, CN agreed that the net cash balance should not be applied to reduce long-term debt. CN explained that long-term debt is well identified on the company’s books and requires the returns expected by bondholders, and that any cash balance, net of short-term liabilities, is part of the working capital that should attract the cost of capital returns required by investors.
CP pointed out that the proposed definition of net rail investment in the Consultation Document does not include the value of land, and noted that the Canada Transportation Act and related regulations do not make any explicit statement that land used in railway operations should not be provided a return for cost of capital. CP submitted that two plausible reasons why one might consider excluding land from the net rail investment may be:
- that the land was originally provided to the railway as a grant, and thus does not represent invested capital; and,
- that the Railway Costing Regulations specify that the rate of return for cost of capital be applied to the variable portion of the net book value of assets related to the movement of traffic.
CP then presented a summary of the history and the uses of the CP land grant in 1881 to explain its position that the land grant represents earned compensation for risks assumed and obligations met by the company in the construction of the original main line across undeveloped territory, and thus represents property owned by shareholders on which they should expect to receive normal returns. CP added that, in addition, much of CP’s land has been acquired on a commercial basis since the completion of the original main line, and that, therefore, exclusion of land from the net rail investment implies excluding not only any of the remaining original land grant, but also the exclusion of all land acquired on a commercial basis. CP submitted that, in that case, the express exclusion of land may inadvertently create a disincentive to acquire land that may be required, for example, to lengthen sidings or to build double mainline track.
CP also disagreed with the proposal that the net cash balance cannot be applied to reduce long-term debt. CP pointed out that, as the Consultation Document notes, the Agency’s 2002 Letter Decision did not define what is meant by the “balance sheet method”, and suggested that it is not even clear whether this balance sheet method is a reference to an adjustment for intercompany financing or whether something else in the capital structure was being altered.
CP submitted that the cash flow method was designed to measure “off-balance sheet” financing and transform it into “on-balance sheet” financing for the purpose of determining the regulatory cost of capital rate, and that, in the absence of a clear definition of the balance sheet method, it can reasonably be presumed that the actual method employed to calculate CP’s capital structure, which has been reviewed and approved annually by the Agency, should be taken to describe what the Agency meant by the balance sheet method.
CP submitted that its cash flow method is an accurate adjustment to the railway’s capital structure to reflect changes in intercompany financing, and reflects the method intended by the 2002 Letter Decision. CP submitted that, therefore, the proposal to eliminate the method must be based either on a fundamental flaw in the reasoning behind the adjustment, or on some fundamental change in the environment since 2002. But, according to CP, its corporate structure has remained effectively unchanged since 1996 – it is still a legal entity owned by a parent company – and effectively the same today as it was at the time of the 1985 Decision.
CP noted that previous Agency decisions make it clear that the reasoning behind the cash flow adjustment is to provide an empirical method to separate the capital structure of the regulated entity from that of the parent company, and that this goal is still valid today, within the spirit of the existing regulations, as the capital structure of the regulated entity, the parent company and other subsidiaries are distinct, but not independent, being legally and operationally intertwined.
McMillan LLP responded to the consultation as solicitors for Teck Resources Limited and its affiliates Teck Coal Limited and Teck Metals Limited. McMillan also submitted letters from the Western Grain Elevator Association and the Canadian Canola Growers Association stating those organizations’ support for the McMillan submission.
McMillan engaged Dr. Lawrence I. Gould to provide his opinion on the issues raised in the consultation. Dr. Gould is a Senior Scholar at the Asper Business School at the University of Manitoba, and previously was Head, Department of Accounting and Finance at the University of Manitoba, as well as Chairman, Finance and Business Economics at McMaster University. In his 40-year career, Dr. Gould has been employed as a consultant on financial theory as applied to cost of capital calculation, and has testified on financial matters before numerous regulatory boards and commissions including the Agency and the Canadian Radio-television and Telecommunications Commission (CRTC).
In his submission, Dr. Gould agreed that a regulated company must acquire land, plant, and equipment in order to provide services to its customers, and that, in addition, the operation of the business also requires current assets such as cash, accounts receivable, and materials and supplies. He explained that the rate base is essentially the property that is used and useful in providing the service, that any capital in the rate base must be provided by someone, and that customers should be charged the appropriate cost of each type of capital.
Dr. Gould then categorized capital into three different sources: capital which arises from the ordinary business operations of the firm; capital which arises due to tax policies; and capital which is provided by investors. He explained that:
- capital arising through the operations of the firm, such as trade credit, has zero cost to the company and should be used as fully as possible;
- to the extent that capital is obtained from deferred income tax, the capital represents a forced loan from customers to the regulated company, equal to the increases in accumulated deferred taxes, on which the company pays no interest, and therefore obtained at zero cost to the company;
- investment tax credit, similarly to deferred taxes, is in essence a forced no-interest loan from the company’s customers, and represents capital obtained at zero cost to the company; and,
- the balance of the capital required for the firm must be provided by investors, and may take the form of common equity, preferred stock, debt, or some combination such as convertible preferred stock; the varying costs and risks of these different forms create the need to decide on the investor-supplied capital.
Dr. Gould submitted that accumulated deferred income taxes and the investment tax credit should be included in the weighted average cost of capital (WACC) calculation as a zero-cost source of funds. He also submitted that it is not correct to reduce long-term debt by the amount of the cash balances, and that this practice must be discontinued. He noted that netting cash balances to reduce long-term debt increases the measured proportion of equity in the capital structure and, as the cost of equity capital is higher than the cost of debt capital, this has the effect of raising the WACC.
FPAC and WCSC’s joint response
FPAC and WCSC indicated their support for the principle that the methodology should be reasonable, reliable and pragmatic, and also be consistent with the objective of providing federally-regulated railway companies with a fair and reasonable return. They noted that, however, the methodology should not produce results that overcompensate or provide windfalls for federally-regulated railway companies, and that it should support and be consistent with the remedial objectives of the statutory provisions concerned.
FMAC noted that it would not be possible for them to provide comments with regard to the inclusion or exclusion of specific items, but in general would object to any modifications that would have the impact of increasing the cost of capital.
SARM did not provide any comments on the issue of capital structure methodology.
ROUND 2 OF THE CONSULTATION – OCTOBER 21 TO NOVEMBER 18, 2016
CP’s comments on the responses
On the issue of netting the cash balance, CP submitted in response to CN’s submission that, as CN, for historical reasons, has not been subject to an intercompany financing adjustment, CN is likely unaware of the reasoning and the function of the adjustment. CP indicated that, as adjustments for intercompany financing have been central to the development of CP’s capital structure over time, the discontinuance of the practice for CP will introduce a material inconsistency in the development of CP’s capital structure going forward.
CP also submitted, in response to McMillan’s submission, that a plain reading of Dr. Gould’s conclusion on intercompany financing suggests that he was not aware of significant factors in the operation of this adjustment, and that, therefore, Dr. Gould’s conclusion should not be relied on. According to CP, the fact that the method was carefully considered in a number of regulatory decisions over several decades indicates that the justification for the methodology is sound, or at the very least, that it was sound at the time. CP submitted that, in its opinion, the aggregate impact of the intercompany financing adjustments have been to reduce CP’s revenues since 1973, and that it has not benefitted financially from the historical intercompany adjustments.
McMillan’s comments on the responses
McMillan submitted that, in asking for land to be included in net rail investment, CP seeks a cost of capital rate that is determined from components of its business that are not caused by the traffic it handles; that is, from the total asset base. McMillan submitted that the logical extension of CP’s assertion is that its Canadian customers should pay for the costs of its capital incurred for businesses that are unrelated to rail traffic, and that to do so would be contrary to (i) the central aim of cost development, (ii) the reason why CN and CP are partially regulated under the Act, and (iii) its need for capital to provide rail, as opposed to other, services.
McMillan presented comments prepared by Dr. Lawrence Gould on the submissions by other consultation participants:
- On the issue of whether to include land in net rail investment, Dr. Gould noted that CP acknowledges that a plausible reason to exclude land from the net rail investment is that the land was originally provided to the railway as a grant, and thus does not represent investor capital. Dr. Gould then submitted that, however, CP’s claim that lands obtained from grants were earned compensation and thus represent property owned by the shareholders on which they should expect to receive returns is based on an incorrect assumption, and that CP is making the mistake of trying to use financial theory that applies to unregulated companies to regulated companies. Dr. Gould submitted that as investors did not provide the funds for land acquired by grants, any returns on this land would represent an excess return on the common equity over and above the return required.
- On the issue of whether the net cash balance can be applied to reduce long-term debt, Dr. Gould agreed with CN’s submission that it is not appropriate.
Aikins’ comments on the responses
Aikins submitted that CN and CP’s proposal to include land in net rail investment represents a radical departure from the Agency’s established methodology, and that such a proposal can reasonably be expected to increase significantly the equity component of the railways’ capital structure and inflate the cost of capital rates determined by the Agency. Aikins submitted that, in the absence of concrete data supporting a conclusion that a fundamental change in the treatment of land or the valuation of assets included in net rail investment is required to allow CN and CP to raise sufficient capital for investment, FPAC and WCSC do not consider such a radical change to be warranted.
Issue 2: The working capital allowance
No new studies to update the working capital amounts for CN and CP have been undertaken since 1992. As a result, the working capital allowances that were determined for CN and CP in 1992 are still used annually in determining the capital structure of CN and CP. These pre-set amounts may not reflect the current working capital allowance requirements for CN and CP for two reasons. First is the significant increases in operations for both CN and CP (for example, the annual operations expenses for CN and CP combined have increased close to 40 percent from about $6.7 billion in 1992 to about $9.3 billion in 2014), and second is the significant increase in the use of electronic banking and payment methods that have taken place in the Canadian economy over the last 30 years.
The Agency wanted stakeholders to comment on whether to update the working capital allowances for CN and CP to reflect their current requirements. However, the Agency, being mindful of the potential burden on the railway companies of conducting extensive and repeated lead-lag studies, asked stakeholders to propose a methodology that would allow the determination of the annual working capital allowances to reflect current operations, and yet would not impose an undue burden on the railway companies. The consultation invited the participants’ views on the following questions, along with any additional comments and suggestions, for consideration by the Agency:
- The working capital allowances in CN and CP’s capital structure were determined in 1992 and maintained to date. Is there is a need for a working capital allowance adjustment in today’s railway operating environment?
- If yes, what methodology would be appropriate for determining the amount of the required working capital allowance?
- How would the methodology proposed in 3) distinguish investor-supplied cash versus supplier-financed cash, with specific reference to the interrelationship between cash, inventory, and accounts payable?
- How often should the working capital allowances be updated?
ROUND 1 OF THE CONSULTATION
CN noted that while the last determination of the working capital was in 1992, it may have been based on data from 1985 or 1986, and that significant changes to both its qualitative and quantitative operations have since occurred. CN also pointed out that while it is generally accepted that a lead-lag study is the most accurate methodology for estimating the working capital requirements, it also requires the most resources and effort, and suggests that this may have been the reason why a lead-lag study has not been performed since 1992.
CN proposed that, instead of a lead-lag study, the simpler and more classical definition of working capital be used: net short term current assets, or current assets minus current liabilities. CN submitted that this is an acceptable measure of the short term working capital required, and covers the time period between when services are rendered and when revenues are collected. CN added that to dampen the volatility that may be present in a single point estimate taken at the end of the fiscal year, as opposed to an average over an entire year, a multi-year rolling average of two or three years could be used.
CN submitted that its proposed methodology avoids the issue of investor-supplied versus supplier-financed working capital as, subtracting the current liabilities removes the supplier-financed assets or inventory, and that any remaining working capital, conceptually, belongs to the investors. CN submitted that “any vendor-supplied finance is ultimately baked into the price of goods and services that CN buys, and therefore all financing costs are ultimately borne by investors”.
Finally, CN noted that this definition would also allow the working capital allowance to be updated every year, and would thus follow the business cycles as CN and its customers take longer or shorter times to pay their invoices.
CP pointed out that, while a lead-lag study for determining the working capital allowance is theoretically sound, it is complicated, and time-consuming to implement because, considering that a wide variety of materials are used in railway operations – each with unique procurement cycles and payment terms – the lead-lag study requires examining thousands of invoices, service contracts, freight movements, and other sources of indirect financing. CP suggested that this may be the primary reason that the current working capital allowance has not been updated since 1992. CP was of the opinion that the current requirement has changed.
CP submitted that there are several options for determining a new working capital allowance, specifically:
- Update the 1992 figure by applying indexation;
- Carry out a new lead-lag study;
- Apply an alternative method, based on accounting figures.
CP submitted that, while the 1992 figure could be updated through indexation using, presumably, the VRCPI, given that the current allowance is based on a study that is nearly 25 years old, should be possible to do better.
CP indicated that it is willing to undertake an updated lead-lag study if, in the opinion of the Agency, that is the best way to update the working capital allowance. CP noted that, however, there is no reason to presume that a new lead-lag study is any less cumbersome than it was in 1992, and that a decision to rely on lead-lag studies may, by extension, imply that the working capital allowance would continue to be updated very infrequently going forward.
CP proposed instead that the company accounting system, which includes accounts to deal specifically with working capital items, namely accounts payable, accounts receivable, and materials and supplies, be used to determine the working capital allowance. Specifically, CP proposed that the working capital allowance can be determined using the following figures from railway accounts:
|Working Capital Allowance||=||Materials and Supplies||-||Accounts Payable||+||(Freight Expense / Freight Revenue)||x||Accounts Receivable|
CP submitted that this equation captures the value of materials which have been paid for but not used, as well as the value of materials that have been consumed, but for which the railway company has not yet been reimbursed. CP added that, however, there are other factors that need to be considered in the calculation of the working capital amount, including treatment of depreciation expenses, treatment of tax expenses, treatment of labour expenses, pre-paid taxes paid by customers, accrued labor expenses, cash, pre-paid expenses, and intercompany amounts.
CP submitted that such a method, which relies on railway accounts, is practical and reasonable, and it would allow for annual updating of the working capital allowance.
Dr. Gould, in his response on behalf of McMillan, noted that significant changes in the operations of the railways, changes in the capital markets, and technological changes in payment methods make it very likely that working capital allowances have changed over the last 25 years. Dr. Gould recommended that a lead-lag study be conducted for both CN and CP in order to determine the current working capital allowance. He submitted that a new lead-lag study would provide an accurate current amount as well as a measure of the change that has taken place since the last lead-lag study, and would provide a benchmark to evaluate the accuracy of any different methodologies proposed by the railways or other interested parties.
Dr. Gould did not consider that it would be necessary to conduct lead-lag studies on an annual basis. He submitted that changes in the working capital allowance could be indexed for short-term changes in operations until the next lead-lag study, which may initially be set for three years later, at which time the resulting change in the working capital allowances could be evaluated to determine the optimal time periods for repetitions of the lead-lag studies.
FMAC noted that if there is a need for adjustment, it was not possible for it to provide meaningful comments until the railways or the Agency have offered proposals and the justification for such proposals. FMAC reiterated, however, that in general it would object to any modifications that would have the impact of increasing the cost of capital.
SARM was of the opinion that, as no new studies have been made to update the working capital amounts since 1992, a new study should be undertaken. SARM submitted that updated data will allow for a more accurate reflection of the current working capital allowance requirements.
ROUND 2 OF THE CONSULTATION
McMillan’s comments on responses
Dr. Gould noted that CN and CP both acknowledge that a lead-lag study is theoretically sound and is the most accurate way of determining the working capital allowance, but they recommend simplified determinations based on balance sheet accounts because of the higher costs of a lead-lag study. Dr. Gould pointed out, however, that while CN and CP raise valid concerns about the costs of conducting lead-lag studies, after 25 to 30 years it is reasonable to require that a lead-lag study should be done for both CN and CP in order to determine the current working capital allowance, and he restated the benefits of a new lead-lag study as he had previously listed.
Issue 3: The specification of long-term debt
Companies such as CN and CP often raise money in the market by selling bonds. A bond promises to pay the holder a certain fixed amount (the face value) at some point in the future, as well as provide interest payments (coupon rate) until the bond’s maturity. In practice, some market realities affect the bond sale. When the coupon rate is less than the yield that the market expects, investors demand a discount to the face value of the bond, and when the coupon rate is higher than the yield that the market expects, investors are willing to pay a premium for the bond. Therefore, the company issuing the bond may receive an amount that is lower (discount) or higher (premium) than the face value of the bond.
The definition of Long-Term Debt (LTD) in the UCA does not specify whether the account should reflect the face value or the net value (that is, the face value net of any premiums or discounts) of the debt. Careful reading of the UCA suggests that the definition implies the face value, as the UCA also provides specific accounts in which railway companies are required to record any unamortized discounts or premiums on the LTD. However, for capital structure determination purposes, the lack of specific direction as to whether the LTD included in the capital structure should reflect the face value or the net value has led to one railway using the net value and the other the face value in determining the capital structure.
The follow-up consultation invited the targeted stakeholders to provide their opinions on the following consultation question, for consideration by the Agency:
For consistency going forward, staff proposes that the Agency makes explicit the definition of LTD, in both the UCA and for capital structure determination purposes, as the face value of the debt as reported in UCA Account 65. Staff also proposes that, for capital structure determination, the discounts on LTD (UCA Account 35) be included as an asset in Net Rail Investment as part of Deferred Charges, and that premiums on LTD (UCA Account 69) be included as part of non-current liabilities. Staff is asking for your opinion and any concerns you may have with respect to this proposal.
FOLLOW-UP CONSULTATION – APRIL 11 TO MAY 19, 2016
CN re-iterated its long-standing objections concerning the inclusion of deferred income taxes in the capital structure and the use of book values in place of market values of equity and assets in determining the capital structure. CN also submitted that, notwithstanding its objections as noted, if the Agency insists on using the balance sheet method then it must use a balance sheet that corresponds to the Agency’s regulatory costing applications.
CN submitted that, while the Agency’s methodology includes all the rail-related assets of the company in determining the capital structure, the resulting cost of capital is applied in regulatory cost applications solely in determining the cost of rail properties. CN further submitted that, in order to determine the capital structure needed to support only property assets, the balance sheet must be re-arranged so that only the property assets are on the asset side, and all the other assets are shifted to the liabilities side. Using this proposed balance sheet, CN then appeared to indicate from the calculations presented that the long-term debt to be considered is the face value.
CP submitted that Agency staff’s proposed approach is incorrect in the context of the regulatory cost of capital determination. CP presented a number of examples to illustrate its position that, by stating debt at its face value, the size of the obligation that is incurred is represented, but not the amount of capital raised or the real cost, and that this does not meet the objective of the cost of capital exercise.
CP submitted that the correct method to record long-term debt for the purpose of the regulatory capital structure is to include the face value of the debt net of any premiums or discounts, and then apply the nominal interest rate. CP noted that this approach is equivalent to its current practice, the main difference being that CP reports the face value net of the unamortized discounts or premiums.
McMillan once again engaged its finance expert, Dr. Lawrence I. Gould, to provide his opinion on the issues raised in the consultation. Dr. Gould provided examples of the standard accounting treatment of premiums and discounts in relation to the annual interest rate, and concluded that if the LTD is recorded at face value, as the Agency instructs railway companies to do in the UCA, then unamortized discounts and expenses on bonds payable, and unamortized premiums on bonds payable, should be considered as valuation accounts used in combination with the LTD for the purpose of capital structure determination. He also concluded that the unamortized bond discounts and expenses should reduce the LTD and the unamortized bond premiums should increase the LTD.
Issue 4: The allocation of non-current liabilities
Past cost of capital methodology reviews have made no specific prescriptions for the allocation of two types of non-current liabilities to the capital structure. The two unspecified items are Deferred Liabilities (UCA Account 61) and Other Deferred Credits – Long-term (UCA Account 69), both of which are considered long-term liabilities which affect the overall capital available to a company.
Deferred liabilities may arise from many different railway activities. For CN and CP, the causes include, but may not be limited to, the following:
- Downsizing and restructuring costs: significant labour downsizing and/or network restructuring costs that may be payable over a future period;
- Environmental remediation costs: significant costs expected over a future period to remediate environmental contamination, clean up toxic spills, and provide other environmental remedies;
- Work-related employee disability or death: significant costs that may be payable over a future period, incurred to settle employee claims for injury, disability or death;
- Legal claims: significant costs that may be payable over a future period, incurred to settle claims from third parties for damages, injury, disability or death arising from accidents, fires, or other causes; or
- Other significant costs payable in a future period.
Other Deferred Credits reported by CN and CP include the following activities:
- Unamortized premiums received on long-term debt; and,
- Payments received in the current period for future services.
The Agency noted in the Consultation Document that the lack of specific direction in the treatment of these accounts in previous reviews has led to inconsistencies in the allocation to the capital structure of these accounts, with one railway company including the accounts with long-term debt and the other railway company including them with equity. As part of the Agency’s intent to prescribe a consistent unambiguous methodology for determining the capital structure for all regulated railway companies, it invited participants to provide their views and comments on the following questions:
- Should the following accounts be included in the capital structure? Give reasons as to why they should or should not:
a. Deferred Liabilities;
b. Other Deferred Credits – Long-term.
2. If the non-current liabilities identified in 1(a) and 1(b) are to be included in the capital structure, what cost rate should each be assigned – the cost rate of debt, the cost rate of equity, a zero cost rate, or some other cost rate? Please provide justification for your recommendation.
3. If some other cost rate is appropriate for either of the non-current liabilities identified in 1(a) or 1(b), what methodology should be used to determine the appropriate rate?
4. Within each of the accounts corresponding to 1(a) and 1(b), should the treatment for all individual accounting items be the same in the capital structure, or is it necessary to make a case by case determination in order to assign a cost rate to each individual item? That is, for example, within the individual items in the accounts classified as Deferred Liabilities, such as environmental remediation or restructuring costs, should all such items be assigned the same cost rate, or are there significant differences in items within the account such that an item by item assignment of cost rates would be required?
5. Is there a set of consistent principles that would guide the treatment of any new non-current liabilities that arise, and if so, what are those principles?
6. Are there any other related issues or comments that you would like the Agency to take into consideration?
CN noted that the non-current liabilities are part of CN’s balance sheet, and as such are part of its capital structure. CN submitted that as the Agency costing model does not support the cost of financing all non-current assets but only the properties, all the other non-current assets (other than properties) should be netted out of the non-current liabilities before considering them for the capital structure. As indicated earlier, CN presented a proposed balance sheet and associated capital structure to reflect this view.
CN then indicated that, to be consistent with previous Agency rulings on the cost rate of deferred income taxes and other tax credits, all these other net liabilities should be included in the capital structure at a 0 percent cost rate, and demonstrated it accordingly in its proposed capital structure.
CP noted that, in the Agency Decision No. LET-R-17-2017 accompanying its determination of CP’s 2017-2018 cost of capital for grain transportation purposes, the Agency referred to paragraph 7(b) of the Railway Costing Regulations, which directs that the cost of capital be based on “the net book value of the assets related to the movement of the traffic”. CP submitted that the Decision, on the basis that goodwill is an intangible asset that is not utilized in the movement of traffic, excluded goodwill from CP’s regulatory asset base.
CP submitted that the same reasoning should be applied generally to deferred liabilities and other deferred credits. CP submitted that these accounts contain items that are often intangible, and/or often are not related to the movement of traffic, such as Workers Compensation Board payment liabilities, and estimates of future environmental cleanup costs. CP submitted that these accounts should not be included in the capital structure.
CP then repeated a long-standing objection it has made (together with CN) on every cost of capital methodology review; namely, that, in general, an item should not be included in the capital structure if it does not have an associated rate, as the purpose of developing a regulatory cost structure is to provide an appropriate rate of return to debt holders and investors. CP submitted that, for that reason, the capital structure should include only debentures, leases and equity, and that, therefore, CP does not support determining deemed cost rates for liabilities that have no intrinsic cost rates.
Dr. Gould explained that generally, accounting standards require that a liability be recognized if the occurrence of a cost in the future is deemed likely and that the amount of the cost is known or can be reasonably estimated. He submitted that these so-called deferred liabilities represent future obligations on the company whose presence affect other sources of funds employed by the company, such as debt sources, and that they should be included in the capital structure. Dr. Gould also explained that other deferred credits arise when payment is received in advance for the provision of a service or sale of an asset, and that they should be included in the capital structure.
Dr. Gould submitted that there are similarities between deferred taxes and other deferred liabilities, in that in all cases the expense is recognized in the current year and recovered from customers in the rates, but is not paid to the third party, only recorded as a deferred liability until it is paid. He submitted that the amount of the deferred liabilities is recovered from customers in their rates, and held as capital on which the company pays no capital cost. Dr. Gould likened it to an enforced loan to the company from its customers on which no interest is paid by the company. Dr. Gould submitted that, therefore, deferred liabilities, similar to deferred taxes, should be included in the capital structure at a zero cost rate.
Dr. Gould submitted that, with respect to other deferred credits, the portion that comprises unamortized premiums on the LTD be considered as valuation amounts used in combination with the face value of the LTD for the purposes of capital structure determination, that is, to increase the LTD. He submitted that the remaining portion of other deferred credits represents capital that has been supplied to the company by non-investors at zero cost to the company, and that these other deferred credits should be included in the capital structure at zero cost.
Other issues arising in the consultation
DEFERRED TAXES, INVESTMENT TAX CREDITS AND DEFERRED COSTS
CN submitted that the capital structure should include only two sources of financing: long-term debt and shareholders’ equity. CN submitted that deferred taxes, investment tax credits and deferred downsizing costs are adjustments required by accounting rules and not a source of financing, and should therefore not be included in the capital structure.
CN also questioned the Agency’s inclusion of investment tax credits in liabilities. CN is of the opinion that investment tax credits should be subtracted from the property investments that generated those credits to begin with, similar to the treatment afforded to donations and grants. CN submitted that unlike liabilities such as bonds, equity or deferred costs, investment tax credits do not represent amounts that are owed or have to be paid to anyone in the future, and, like donations and grants, should be used to reduce the value of properties in the determination of net rail investment.
MARKET VALUE OF EQUITY
CN submitted that shareholders’ equity should be evaluated using the market value of equity and not the book value.
McMillan, however, submitted that the use of market value of assets in the cost of capital determination would not only defy the scheme of the Act and the purpose of regulation, but would impose on railway customers a burden that is disconnected from the provision of rail transportation services, as it would result on a return on capital that is greater than what the railway needs to provide those services.
MARKET VALUE OF LAND
CP submitted that using the book value of land in determining the capital structure, as prescribed by the Railway Costing Regulations, has the advantage that book values are readily determined, relatively stable from year to year, and are based on the original cost to the enterprise. CP also submitted that market values, on the other hand, have the advantage of reflecting the true value of the assets today, and that, consequently, the market value is the cost to the investor of having their capital invested in the enterprise, as opposed to liquidating that investment to employ in alternative investments or for other uses.
CP finally concluded that, to the extent that the book value of land is never restated from its original acquisition cost, even if it was acquired over 100 years ago, any actual changes in the value of those lands, or any lands acquired since, have no prospect of ever being reflected in their book values, and that, therefore, a return on capital based on the book value of land does not provide a fair return according to the Supreme Court definition of that term. CP proposed that a fair market value of land be included in the capital structure determination, and proposed some considerations for establishing that fair market value.