Decision No. 125-R-1997

On July 31, 1985, the Railway Transport Committee of the Canadian Transport Commission (hereinafter the RTC) issued The Cost of Capital Methodology Decision in the matter of issues pertaining to the Canadian Transport Commission's Cost of Capital Methodology for Regulated Railways; and in the matter of proposed amendments to the Railway Costing Regulations related to the Cost of Capital (hereinafter the 1985 Decision).

On July 1, 1996, the Canada Transportation Act, S.C., 1996, c. 10 (hereinafter the CTA) came into effect. Pursuant to section 196 of the CTA, the 1985 Decision continues to the extent that it is not inconsistent with the CTA as provided for under the transitional provisions of the CTA and National Transportation Act, 1987, R.S.C., 1985, c. 28 (3rd Supp.). While the application of the costing methodologies has been narrowed under the new legislation, with for example, the elimination of rail subsidies, the Agency still has a statutory ongoing costing responsibility in other areas, such as the setting of the maximum rate scale for the transportation of western grain and the setting of interswitching rates.

Section 32 of the CTA states, in part, that the Agency may vary any decision, if, in the opinion of the Agency, there has been a change in the facts or circumstances pertaining to the decision since it issued.

Since the issuance of the 1985 Decision, there have been substantial changes in the railway industry. As a result, it has become evident that certain components of the 1985 Decision may no longer be relevant or may now require revision.

The Agency initiated the review process of the 1985 Decision by letter dated September 27, 1996 to interested participants that were involved in the process that resulted in the 1985 Decision as well as other participants that had indicated an interest in cost of capital matters since 1985. These participants were requested to provide their comments and positions on specific issues to be reviewed regarding inputs into cost of capital rate determinations. By November 1, 1996, responses were received from the Canadian Pacific Railway Company (hereinafter CPR), the Canadian National Railway Company (hereinafter CN) and from the Provinces of Alberta, Manitoba and Saskatchewan and the prairie Wheat Pools (hereinafter the Provinces/Wheat Pools).

Major issues which were commented on included:

• Whether CPR's cost of common equity rate should continue to be derived from Canadian Pacific Limited's (hereinafter CPL) cost of common equity rate for cost of capital rate determinations;
• Whether CN's cost of common equity rate should continue to be based on CPR's cost of common equity rate or whether separate rates should be developed for each railway;
• Whether the Agency should use results from market-based models on an individual basis or use a blend of the Capital Asset Pricing Model (hereinafter the CAPM), the Discounted Cash Flow Model (hereinafter the DCF) and the Equity Risk Premium model;
• Whether to continue with the business risk adjustment, in respect of CPR, of one half of one percentage point reduction to CPL's cost of common equity rate;
• Whether to continue with the grain risk adjustment of one percentage point reduction to CPR's cost of common equity rate; and,
• Whether the write-down of assets incurred by CPR and CN in 1995 should be reflected in the net rail investment and in the capital structure of the railway companies.

On December 23, 1996, the Agency provided a Staff Report to the participants (attached as Appendix A) which summarized the participants' positions and provided initial recommendations on the issues to be re-examined. In addition, the Agency advised participants that three issues were to be examined through a consultative hearing. The three issues were: the business risk adjustment of CPR; the grain risk adjustment; and the property issue (which dealt with the treatment of the write-down of assets incurred by both railway companies).

Prior to the commencement of the consultative hearing, CPR, CN and the Provinces/Wheat Pools filed written evidence.

The consultative hearing took place on January 27 and 28, 1997. While the Provinces/Wheat Pools originally intended to participate in the consultative hearing, they chose to withdraw from the oral portion of the process on January 27, 1997 and, by agreement with the Agency and participants, continued to participate through the filing of written material only, including the filing of rebuttal evidence. The railway companies expanded on their written evidence through witness presentations during the consultative hearing. Final arguments were filed in writing with the Agency on and before February 7, 1997.

This Decision will deal with both the matters raised in the Staff Report, on which all parties have had the opportunity to comment, and the three issues addressed during the consultative hearing.

I ISSUES DEALT WITH THROUGH STAFF REPORT

On December 23, 1996, the Agency provided participants with a Staff Report (Appendix A) for further comments. The Staff Report summarized positions of participants to date, and provided staff recommendations on all matters relevant to the review with the exception of the three issues subsequently dealt with at the consultative hearing. Only relatively minor comments on the Staff Report were received by the Agency.

SUMMARY OF ISSUES AND AGENCY FINDINGS

1. Cost of Common Equity Rate

In the 1985 Decision, it was decided that since CPR did not have publicly traded shares, the cost of common equity rate for CPR would be based on the cost of common equity rate for CPL adjusted for any differences in business and financial risks between the two organizations.

As a result of the repeal of the Western Grain Transportation Act, R.S.C., 1985, c. W-8 (hereinafter the WGTA), the coming into force of the CTA and the privatization of CN in 1995, the following issues were identified for review:

• Should CPL still stand as the benchmark for rate determination?
• Should CN and CPR have their own cost of equity rates?
• If so, when would sufficient data be available for calculation?
• Should there be an average rate?

AGENCY FINDINGS

After consideration of the positions of CPR, CN and the Provinces/Wheat Pools, the Agency finds that the use of CPL's cost of common equity rate as a benchmark is still appropriate since CPL is the only publicly traded company that has the historical market information available to determine the cost rate.

CPR's cost of common equity rate will therefore continue to be derived from the cost of common equity rate of CPL. This process will continue unless other market information becomes available to the Agency which would allow the Agency to determine CPR's own cost of common equity rate. CPR's cost of common equity rate will subsequently be used in the determination of CN's cost of common equity rate. This approach will continue until the Agency determines that there is sufficient market information available to allow for a separate determination of CN's cost of common equity rate. In respect of CN, the Agency expects that sufficient data will be available for calculation by the year 2002.

2. Market Driven Models

In the 1985 Decision, three market-based methods were put forward to measure CPL's cost of common equity: the DCF method, the CAPM and the equity risk premium approach. According to the DCF approach, the cost of common equity rate is equal to the sum of the current dividend yield plus the expected future growth rate in dividends. The CAPM measures the cost of common equity rate as the sum of a risk-free rate of return plus the risk premium on a portfolio (representing the market as a whole) multiplied by the riskiness of CPL relative to the riskiness of the market as a whole (known as the "beta" of CPL). The risk premium approach measures the yield on current Government of Canada bonds plus a premium for the additional risk of common equity over Government of Canada bonds.

The 1985 Decision provided that annual cost of capital rate determinations would be made on the basis of informed judgement following the assessment of the results of each of the three methods.

Because the economic environment has changed, interested persons have expressed concerns in regard to which method is the most appropriate for use as well as the inputs to the methods.

The following issues were identified:

• Should the Agency use a blend of various models such as the DCF, the CAPM and Equity Risk Premium?
• Should the risk-free rate used in the CAPM be taken from a specific source (i.e. Econoscope, Consensus Forecasts) at a certain point in time?
• Should the risk premium element be revised? If so, which bond rates (e.g. T-Bills, Government of Canada Bonds) should be used, and which maturity of such bonds (e.g 10 years, 30 years) should be used?

AGENCY FINDINGS

After consideration of the positions of CPR, CN and the Provinces/Wheat Pools, the Agency finds that an assessment of the three market-based methods (i.e. the DCF, the CAPM and Equity Risk Premium) will continue to be performed to estimate CPL's cost of common equity. As in the 1985 Decision, the outcome of all three methods will continue to be assessed annually and weight will be given to the most appropriate model or a combination of models. The Agency notes that there is judgement involved in estimating the results of each model. In respect of the various inputs required for these models, primary weight should be given to Canadian data and studies.

The Agency has examined each of the following elements specific to the CAPM model separately.

a) Risk-Free Rates

In applying the CAPM, the Agency finds that, as in the 1985 Decision, the use of both short-term (1-3 years) and long-term (10+ years) Government of Canada bonds as proxies for a risk-free rate of return is appropriate.

For the purposes of applications based on forecasted data, various sources for the risk-free rates were considered. The availability of the data is a factor to be considered in order to identify the most accurate and reasonable rates to be used in the determination of a cost of equity rate for maximum rate scale purposes. For the maximum rate scale, the Agency finds that the short and long-term rates should be based on the Government of Canada bond rates quoted in the Globe and Mail during the month of January. In addition, the Agency will monitor these bond rates during the period in which the maximum rate scale is being developed in order to assess the reasonableness of the January bond rates. For the purpose of developing interswitching rates, the Agency will rely on the same sources of information and the same process for the month of May.

b) Market Risk Premiums

For the market risk premium, a consistent term, based on the term used for the risk-free rates, must be used in the calculation of the risk premium on a market portfolio. In order to respond to changes in market conditions, the Agency will assess the market risk premium on an ongoing basis in order to determine the appropriate input to the CAPM Model.

c) Beta

In regard to the beta component of the CAPM, which indicates the systematic risk, the Agency will continue to assess CPL's historical beta adjusted for the long-term tendency of the beta to converge to unity.

3. Financial Risk of CPR

In the 1985 Decision, it was decided that to obtain the cost of common equity rate for CPR from that of CPL, an adjustment, amongst others, had to be made to reflect any differences in financial risk between the two companies. The RTC decided that a downward adjustment of one percentage point to CPL's cost of common equity rate to arrive at CPR's cost of common equity rate would have been appropriate for 1982 for the financial risk adjustment. In the 1985 Decision, it was recognized that the capital structure of CPR contained less debt than the capital structure of the publicly traded CPL and therefore found that there was less financial risk associated with CPR than CPL.

Since the 1985 Decision, the economic environment has changed as has the status of CPR versus CPL.

The following issue was identified:

• Is the adjustment still relevant?

AGENCY FINDINGS

After consideration of the positions of CPR, CN and the Provinces/Wheat Pools, the Agency finds that since both companies essentially have the same capital structures at this time, CPR has the same level of financial risk as CPL. However, the Agency will continue to monitor the financial position of CPR on an annual basis. If the capital structures of CPL and CPR remain similar, no adjustment will be considered. If the capital structures differ significantly, the Agency will consider an adjustment to CPL's cost of common equity rate to reflect that difference.

4. Business Risk of CN

In the 1985 Decision, the RTC noted that it was desirable to establish CN's own capital structure and its own rates for the various components of the capital structure. Measurement of CN's cost of common equity rate was complicated by the fact that it was a crown corporation and therefore the methods used for measuring CPR's cost of common equity rate were not applicable to CN. With respect to the business risk, it was noted that CN's operations were roughly similar to those of CPR and that both CN and CPR were broadly diversified in that they served almost all sectors of the Canadian economy. It was decided that the business risk for CN was the same as that for CPR.

Since the 1985 Decision, the economic environment has changed as has the status of CN.

The following issue was identified:

• Is any adjustment required?

AGENCY FINDINGS

Until there is sufficient data available to calculate CN's cost of common equity, the Agency finds that the circumstances of business risk for CN continue to be similar to those of CPR, and therefore, has determined that the same risk adjustment factor used for CPR shall apply to CN.

5. Financial Risk of CN

The RTC stated in the 1985 Decision that the financial risk of CN differed from CPR in two major ways:

• CN's capital structure had a significantly higher proportion of debt than that of CPR which resulted in theory in a higher financial risk.
• CN was a crown corporation and it was much less probable that a crown corporation would suffer bankruptcy than it was for a private firm; in theory, this resulted in a lower risk of incurring all the expenses associated with a bankruptcy.

It was the RTC's view that the reduction in financial risk to CN due to avoidance of bankruptcy and guaranteed access to financial markets, had equal weight to the increase in financial risk due to CN's higher debt ratio. Therefore, it was judged that CN had the same level of financial risk as CPR.

Since the 1985 Decision, the economic environment has changed as has the status of CN as a crown corporation.

The following issue was identified:

• Is the adjustment still relevant?

AGENCY FINDINGS

The Agency notes that with the privatization of CN, the debt burden has been reduced, and in fact, the debt ratio for CN is now similar to that of CPR.

The Agency finds that, at this time, based on the most recent information available, CN has the same level of financial risk as CPR. However, the Agency will continue to monitor the financial position of CN on an annual basis. If the capital structures of CN and CPR remain similar, no adjustment will be considered. If the capital structures differ significantly, the Agency will consider an adjustment to CN's cost of common equity rate to reflect that difference.

6. Working Capital Allowance

In the 1985 Decision reference was made to two methods of quantifying the amount of the investor-supplied cash component of current working assets for CN and CPR's rail operations. The first method was the use of a lead/lag study. The second alternative was the balance sheet approach. The RTC decided that it would consider the results of either alternative.

In 1991, the Agency decided to revise the requirements for working capital submissions. One of the measures adopted was that the working capital determinations would be subject to a quadrennial review consistent with the WGTA Costing Review cycle. The lead/lag approach was used in determining the working capital allowance for CN and CPR for base year 1992. The next review period would have been scheduled for 1996.

Because of the repeal of the WGTA and because of the extensive lead/lag studies that are required to set a new base for the working capital allowance, the following issues were identified:

• Should the Agency maintain the current process: using 1992 as the base year study and continue indexing on an annual basis?
• Should the Agency proceed with a requirement for the railway companies to conduct in-depth lead/lag studies for 1996 as part of the four-year process (1996 base year and subsequently indexed over the next 3 years)?
• Should the Agency consider other alternatives?

AGENCY FINDINGS

By letter dated March 12, 1996, CN and CPR were requested to provide the Agency with their comments and/or suggestions on various aspects surrounding the working capital determination. By letter dated April 10, 1996, CPR indicated that as a result of the elimination of the Quadrennial Costing Review, the need for a 1996 working capital determination was highly questionable.

The Agency has determined that the process which involves using the 1992 base year study and indexing on an annual basis shall be maintained. The Agency may, however, reevaluate this approach on an "as required" basis.

7. Capital Structure

The 1985 Decision dealt with the development of the appropriate rail capital structure, which includes the proportion of debt and equity used in determining the cost of capital rate.

Because of restructuring by CN and CPR, there may be an impact on elements of their capital structures. The following issues were identified:

• Should the Agency continue to use the parent company's (i.e. CPL) financial structure with adjustments to reflect the rail portion using the cash flow approach?
• Should the Agency consider other options (i.e. in the case of CN, the rail operations represent approximately 90% of CN's activities. Is its capital structure as presented in the Annual Report to Shareholders an alternative to the cash flow approach)?

AGENCY FINDINGS

After consideration of the positions of CPR, CN and the Provinces/Wheat Pools, the Agency has determined that for CPR, the same methodology used in prior years (i.e. cash flow approach) shall continue to be used for calculating CPR's capital structure.

With respect to CN, in years prior to 1992, CN's portion of the long-term debt was determined by allocating 100% of the corporate long-term debt to rail excluding the portion that was clearly identified as non-rail. In 1995, CN's rail activities represented over 90% of the corporate activities. The Agency finds that for the cost of capital purposes, using the corporate long-term debt reduced by the identifiable non-rail long-term debt is appropriate in assessing CN's long-term debt for cost of capital purposes.

8. Interswitching Rates

According to the 1985 Decision, a cost of capital rate was to be established for the movement of rapeseed oil and meal (canola) pursuant to Order in Council P.C. 1976-894. This Order in Council was repealed in 1995. Because this rate was also used for the development of interswitching rates, it is now necessary to establish a method for determining the cost of capital for this purpose. The 1997 interswitching cost of capital rate was developed using the 1996/97 maximum rate scale cost of capital rate with some adjustments to the various elements of the cost of common equity rate.

The following issues were identified:

• For the determination of the 1998 interswitching cost of capital rate as well as for subsequent years, should the railway companies submit a separate determination? or,
• Should the Agency continue to rely on an existing cost of capital rate such as the maximum rate scale rate and apply the appropriate adjustments to the elements of the cost of common equity rate as was done in the development of the 1997 interswitching rate?

AGENCY FINDINGS

Through this review process, participants agreed to use, for the cost of capital rate for regulated interswitching purposes, the cost of capital rate determined for the maximum rate scale, with appropriate adjustments to the element of the cost of common equity rate. These adjustments are: i) using appropriate risk-free rates, when the CAPM Model or the Equity Risk Premium Model are used in the determination of CPL's cost of common equity rate; and, ii) the exclusion of any grain risk adjustment factor. Therefore, the Agency has determined that this method will be applied in future interswitching rates determinations.

9. Income Tax Allowance

In its 1979 Income Tax Decision dated August 17, 1979, the RTC stated that in calculating a cost of capital for subsidy purposes, the issue of whether or not a particular railway company paid or deferred income taxes would have to be considered. In the past, for the determination of cost of capital rates for the purposes of transporting western grain and the development of interswitching rates, an allowance for income tax has been included.

The following issue was identified:

• Should there be an allowance for income tax in the development of the maximum rate scale and the interswitching rates?

AGENCY FINDINGS

The Agency finds that there has not been a change in the facts or circumstances with respect to the issue of an income tax allowance, and therefore an income tax allowance shall continue to be included in the determination of cost of capital rates for the purposes of transporting western grain, the development of interswitching rates and other applications for rate-making purposes.

II ISSUES DEALT WITH THROUGH THE CONSULTATIVE HEARING

The consultative hearing addressed three specific issues: i) the business risk of CPR, ii) the grain risk adjustment and, iii) the property issue. All participants to the consultative hearing provided written and/or oral evidence and final arguments on the three issues, with the following exceptions: CN did not address the issue of business risk of CPR, while the Provinces/Wheat Pools did not address the property issue.

ISSUES AND AGENCY ANALYSIS AND DECISION

1. Business Risk of CPR

In the 1985 Decision, it was decided that the cost of common equity rate for CPR would be based on that of CPL with adjustments to reflect any differences in business risk between the two entities. The RTC decided that a downward adjustment to account for such differences of one half of one percentage point to CPL's cost of common equity rate to account for such differences was appropriate. The RTC noted that CPR served almost all sectors of the Canadian economy and as a result, its business activities should be even more highly correlated with the performance of the economy as a whole than are those of CPL.

Since the 1985 Decision, the economic environment has changed as has the status of CPR versus CPL. In light of these changes, participants were asked whether the traditional adjustment was still relevant.

Position of CPR

It is CPR's opinion that the one half of one percentage point adjustment for the business risk of CPR relative to that of CPL is no longer warranted.

CPR presents its position using both qualitative and quantitative statements, a statistical analysis of systematic risk using the method of beta analysis, and a statistical analysis of total risk. In addition, CPR responded to the Provinces/Wheat Pools' position. CPR provided an assessment of the reasonableness of its results and made reference to the 1985 Decision.

According to CPR, its traffic is heavily dependent on export markets and the traffic mix is correlated more with the export performance of the primary sector than with the economy as a whole. CPR is faced with more competition because of a trend towards trade liberalization and greater emphasis on competitive forces in the regulatory regimes in the United States and Canada. CPR also argues that CPL's portfolio of business investments has significantly changed since 1985. The effect of CPL's divestitures of certain elements of its investment portfolio increases the proportion of CPL's business activities represented by CPR. As a result, the financial results for CPL are more closely tied to CPR than was the case in 1985. CPR's proportion of CPL's identifiable assets has significantly increased from 1985 to 1995. Similarly, CPR's proportion of CPL's revenues has also significantly increased over the same period.

Systematic Risk

Dr. Evans, on behalf of CPR, argues that the Agency should not consider simply systematic risk in measuring the business risk but should consider the totality of risk (both systematic and non-systematic). Dr. Evans provides evidence to show that CPR was more risky in 1995 than it had been in 1985 using analysis of beta factors, bond ratings and the variability of the operating income of CPR.

Dr. Evans focused on the estimation of the systematic risk as measured by the beta factor: beta measures the extent to which the rates of return on an individual security or asset follow those of the stock market, in this case, the Toronto Stock Exchange (TSE 300). Using market information for CPL and CPL's publicly-traded entities (i.e., PanCanadian Petroleum Limited and Laidlaw Inc.) and certain assumptions concerning the remaining entities not publicly traded, Dr. Evans "derived" a beta factor for CPR. Dr. Evans illustrated, through his analysis, that CPR is riskier than CPL since its derived beta is equal to 1.5 while the beta for CPL is 1.3. This use of a beta factor was, according to Dr. Evans, based on the fact that the beta used in the CAPM is frequently used as a measure of relative risk and not strictly for systematic risk.

Dr. Evans used the bond rating approach to measure the relative risks of CPL and CPR. CPR now has its own bond rating which was not the case in 1985. The ratings for CPL are more favourable than for CPR. Therefore, Dr. Evans concluded that the bond rating analysis supports his contention that CPR is riskier than CPL. Dr. Evans also illustrated that the bond ratings for PanCanadian Petroleum are more favourable than those for CPL and CPR. This leads to the conclusion that the bond ratings for the remainder of CPL's investment portfolio would have a stand alone bond rating lower than the rating for CPL. Since CPR comprises the dominant portion of the non-PanCanadian Petroleum holdings of CPL, CPR is considered to have a lower bond rating than CPL and therefore is relatively more risky than CPL.

Dr. Evans undertook an analysis of the coefficient of variation of operating income rates of return on total assets to illustrate that CPR is riskier than CPL. According to Dr. Evans, the results indicated that the more disperse and less stable the operating income rates of return, the higher the coefficient of variation and accordingly, the greater the business risk. In this case, Dr. Evans illustrated that the coefficient of variation of the operating income of CPR was higher than for the coefficient of variation of the operating income of other divisions of CPL.

In response to the Provinces/Wheat Pools' position, based on the Patterson model¹ , that a "fundamental" beta for CPR should be derived by the Agency, Dr. Evans provided comments on the reasonableness of this model. As Dr. Patterson's model was estimated from data for the period of 1969-87, reliable use of this model would require the Agency to compile information on CPR for the 1969-87 period and then, having applied the model to historical data, make judgemental adjustments for changes in circumstances and risks. In short, CPR asserts that the use of the Patterson model in the present context raises as many questions as it answers. In addition, Dr. Evans refers to Dr. Patterson's own allusion to the limitations of the fundamental beta approach which are that this model requires extensive accounting data and that the estimates are subject to a significant amount of statistical prediction error.

Dr. Evans also addressed the Provinces/Wheat Pools' conclusions concerning risk based on statistical correlation. Dr. Evans indicated that it was inappropriate to reach conclusions, such that CPR was more closely correlated with the economy than CPL. For the same reason, Dr. Evans was critical of the RTC finding, in the 1985 Decision, that the business risk of CPR was less than the business risk of CPL. In support of his contention that the rationale in the 1985 Decision is inappropriate, Dr. Evans states that beta factors measure the sensitivity of the rate of return associated with the business activity to the rate of return for the stock market as a whole (in this case, the rate of return on TSE 300 share index). On the other hand, R-squared measures correlation rather than sensitivity. There is no necessary relationship between the rate of return sensitivity to the stock market and the extent to which underlying business activities may or may not be correlated with general economic activity.

As a result of its analysis and that of Dr. Evans, CPR concluded that the downward adjustment for the business risk of CPR was not warranted.

Position of the Provinces/Wheat Pools

The Provinces/Wheat Pools provided evidence that support their claim that the one half of one percentage point reduction for business risk differences between CPL and CPR should continue.

The Provinces/Wheat Pools claim that although the differences between CPL and CPR have been reduced in recent years, the differences are still substantial. The Canadian rail activities of CPR are still a minority of overall CPL activities as CPR's Canadian revenues in 1995 represented only about one-third of CPL's revenues and therefore it cannot be assumed that the risks of the two entities are the same.

The Provinces/Wheat Pools submit that if the Agency finds that CPR's activities are still well correlated with the economy as a whole, then the Agency should continue to make a business risk adjustment. On the other hand, if it is found that this does not continue to be the case, the Provinces/Wheat Pools submit that the Agency should revise the level of business risk allowance for CPR relative to CPL using a "fundamental" beta approach, to estimate a beta for CPR as described by Dr. Patterson.

The Provinces/Wheat Pools submit that the use of the CAPM means that the Agency accepts that the main form of business risk faced by CPR is systematic risk. If the Age.cy beapes that the non-systematic risks are significant as determinants of return expectations for investors in CPL, then, conceptually, it cannot use the CAPM.

The Provinces/Wheat Pools also question the validity of calculating a "derived" beta for CPR as presented by Dr. Evans. According to the Provinces/Wheat Pε_ls, the beh_ta of the parent can be viewed as the weighted average of the beta factors of each of the parent's divisions. The Provinces/Wheat Pools indicated that Dr. Evans used this model when only two of the seven divisional beta factors were known and made a number of highly questionable assumptions. They took the position that the methodology is valid only when the beta for just one division is unknown.

In respect of the bond rating analysis, the Provinces/Wheat Pools submitted that it is inconsistent to use CPL and PanCanadian Petroleum ratings to claim that CPR is exposed to greater risk. The bond rating approach is intended to rate the relative riskiness of bonds of a company and not its equity investment.

With respect to the use of coefficient of variation as presented by Dr. Evans, the Provinces/Wheat Pools submitted that since the coefficient of variation is a ratio with average return on the denominator, it is not, by definition, a pure risk measure. It confounds risk with return and, hence, cannot be used to measure risk for determining the cost of common equity.

The Provinces/Wheat Pools further submitted that the standard deviation of operating income and, therefore, the coefficient of variation of operating income, attempts to measure total operating risk, not systematic risk. Consequently, the Provinces/Wheat Pools concluded that relying on this analysis is incompatible when using the CAPM to obtain the cost of common equity rate for CPR.

AGENCY ANALYSIS AND DECISION

Agency Analysis

The Agency notes that with the passage of time since 1985, the business portfolio of CPL has changed, and that many of the components that would have provided a diversification of risk in 1985 are now absent from the portfolio in 1996. By the same token, the importance of CPR in the CPL portfolio has grown to the point that if, in 1995, one were to consider CPR and PanCanadian Petroleum together, they would constitute about three quarters of the value and earnings of that portfolio, and CPR alone now amounts to some fifty percent of the value of earnings and some forty percent of the value of the assets of CPL. This is a significant change in the underlying circumstances relating to the relationship between CPL and CPR since 1985.

The Agency has examined Dr. Evans's arguments in his analysis of the relative riskiness of CPL and CPR. The Agency is aware that his calculations of the residual beta for CPR at some 1.5 is based on some broad assumptions, and that his efforts have gone beyond the normal use for this sort of model. Nonetheless, the Agency is persuaded in logic that the beta for CPR must be greater than for CPL, even though the precise amount of difference cannot be calculated as he has suggested. The Agency also accepts the notion that the relative riskiness of corporate entities are reflected to some extent in their bond ratings, even though there are many other circumstances that will dictate the exact rating for each at a given point in time. Finally, the analysis that indicated that the earnings of CPR were more volatile than those of CPL, as measured by the coefficient of variation of the earnings of each company, is also indicative of a more risky CPR. In summary, each of the measures proposed by Dr. Evans supports the position that since 1985, CPR has become at least as risky as CPL, and this is a change in circumstance since that former time.

The Agency has considered the position of the Provinces/Wheat Pools that the Agency should undertake the calculation of a fundamental beta for CPR, rather than relying on the CAPM model to determine the cost of common equity rate. This fundamental beta approach as described by Dr. Patterson was criticized by Dr. Evans. The Provinces/Wheat Pools referred to the use of this model in the Teleglobe Canada Inc.² case which was detailed in a Canadian Radio-Television and Telecommunications Commission (hereinafter the CRTC) decision. The Agency has reviewed this Decision and notes that Dr. Patterson first estimated the relationship between observed levels of systematic market risk and the accounting variables that appear to determine that risk, using data from 61 firms. Dr. Patterson applied this relationship to the corresponding accounting variables for Teleglobe Canada Inc. in order to determine what its market beta would be if it were publicly traded. The CRTC concluded that Dr. Patterson's failure to account for the tendency of his model to overestimate the market beta factors of firms at the low-risk end of his sample resulted in an overestimate of Teleglobe Canada Inc.'s market beta. For the purposes of this methodology, the Agency finds that there are difficulties in the calculation of the fundamental beta arising from the assumptions that must be made. Further, the Agency accepts Dr. Evans' opinion that this model has limitations in that it requires extensive accounting data and the estimates are subject to a significant amount of statistical prediction error. Therefore, the Agency finds that the application of this model is impractical and of limited utility. As a result, the Agency has decided not to adopt this approach.

The Agency considered the Provinces/Wheat Pools statement that if the Agency continues to place reliance on the CAPM exclusively, it implies that only the systematic risk is to be assessed. The distinction between systematic and non-systematic risk is a unique feature of the CAPM which is not covered under the DCF or the Equity Risk Premium approaches. These approaches take into account the total risk faced by investors in making their investment decisions. The Agency also recognizes the need to examine total risk. As indicated in the Staff Report and expressed elsewhere in this Decision, in the past, the Agency assessed results from the three market-based models, i.e. the DCF, the CAPM and Equity Risk Premium, and used the most appropriate results from a model or combination thereof to provide the Agency with CPL's cost of common equity rate. This rate was subsequently adjusted for the business risk of CPR, if necessary. The use of different cost of common equity rate estimation techniques meant that the cost of common equity rate was not measured only by reference to whether the risk attributes are systematic or non-systematic.

As stated previously in this Decision on the issue of market driven models, the outcome of all three models will continue to be assessed annually by the Agency and weight will be given to the most appropriate model or a combination of models. In this way, the Agency applies its informed judgement on deriving a cost of common equity rate from the results of the various estimation techniques.

Agency Decision

For the above reasons, the Agency finds that while statistical analyses presented by each of the participants are informative they do not lead to a conclusive result on their own. The Agency has decided that for cost of capital rate determinations for the costing of 1997 railway operations after the issuance of this Decision and for the next crop year (1997-98), the business risk adjustment to the cost of common equity rate for CPL to reach the cost of common equity rate for CPR will be zero. The Agency will continue to monitor the situation and based on its informed judgement will make a determination on whether an adjustment is required, on an annual basis.

2. Grain Risk Adjustment

According to paragraph 38(2)(b) of the WGTA, the cost of capital to be used for the provisions of the WGTA had to be adjusted for any differential risk of grain traffic. During the hearing process that led to the 1985 Decision, three approaches were examined to determine whether or not grain traffic had different risk characteristics from other freight traffic and, if so, whether an adjustment to the applicable cost of common equity rate was justifiable. The three approaches were:

• To measure empirically whether grain traffic has historically been more or less risky than freight traffic as a whole.
• To compare the risks of grain traffic to the risks of regulated public utilities.
• To directly compare grain traffic with other specific types of freight traffic.

In the 1985 Decision, the RTC found that grain traffic for the purposes of the WGTA was less risky than freight traffic in general. It was determined that this differential risk led to a cost of common equity rate under the WGTA that was one percentage point below the cost of common equity rate determined for CPR's operations as a whole.

Because of the repeal of the WGTA, the change in the economic environment and concerns expressed about the relative sensitivity of grain traffic to the business cycle compared to the sensitivity of all freight traffic, participants were asked to address the following questions to assist the Agency in its review:

• Should the risk of transporting grain be different than the one associated with transporting other commodities?
• How are the grain transportation characteristics different from other commodities?
• What is the impact of the repeal of the WGTA on the transportation of grain?

Position of CPR

CPR states that each commodity carried by CPR is characterized by a unique set of transportation characteristics and that the transportation markets for each commodity are very different from one another. CPR submits that it is not possible to categorize the non-grain commodities as a relatively homogeneous unit and to infer differences in transportation characteristics between grain and non-grain traffic.

According to CPR, the repeal of the WGTA has three major consequences for the railway companies:

• The termination of the federal subsidy means that the shippers pay the full amount of the freight rate; it removed the artificial distortion that encouraged production of bulk grains for offshore export, and provided a greater incentive to increase the domestic consumption of grain through value-added activities and the production of higher-valued crops; it has increased the choices available to producers as to which crops to grow and which markets to serve. Elimination of the WGTA will result in a change in cropping patterns, an increase in value-added processing on the Prairies and a change in the direction of movement such as increased exports to the United States, and increased exports by modes other than rail. The effect of these changes is to increase the risk associated with expected traffic volumes because of the greater variability of volumes from the total volume available.
• The payment of the full rate by the shipper has increased the modal competition (both intramodal and intermodal) faced by CPR associated with the movement of grain. The competitive choices available to shippers has expanded, thus increasing the risk to CPR from investments in grain-related assets.
• The only difference between the regulatory regime for the movement of grain in western Canada and the movement of all other commodities is the existence of a maximum rate scale for the movement of grain.

CPR submits that since the transportation demand characteristics for each commodity are different, the variations in the movement patterns are not correlated with one another nor with the economy in general. Consequently, the incremental addition of grain to the portfolio of CPR traffic changes the variability of total traffic to about the same degree as other commodities. There is no evidence to suggest that the correlation between the changes in the volume of grain and the changes in the volume of non-grain traffic differs from the correlation of the volume changes among the non-grain commodities.

CPR provided empirical evidence to support its claim that the transportation of grain had essentially the same characteristics as the transportation of other commodities. The evidence showed that:

• the reduction in the variability of total traffic as a result of the incremental carriage of grain was relatively small;
• the incremental carriage of the other commodities (with the exception of automotive and import/export containers) made a greater contribution towards the reduction in the total variability of railway traffic than did grain;
• the average relationship between changes in the volume of grain and the changes in the volume of non-grain commodities is not materially different from the average relationship between the changes in the volume of traffic among the non- grain commodities; and,
• grain is not distinguished by characteristics that suggest that it is uniquely less "risky" than other commodities.

As a result of the evidence, it is CPR's position that there is no justification for a reduction to the cost of common equity rate for CPR on the basis that the carriage of grain is of a lower risk than the carriage of other traffic.

CPR refers to the legislative/regulatory risk as the one area of risk associated with the movement of grain that makes it materially more risky than other commodities. An important aspect of this risk is the uncertainty about the outcome of the required 1999 Review of Division VI of the CTA, and the associated uncertainty as to whether rates for the transportation of grain will continue to be regulated. This uncertainty represents a source of risk that is unique to grain. This uncertainty means that there is substantial and unique risk associated with investments in grain-related assets.

Position of CN

According to CN, the repeal of the WGTA has the following major consequences for the railway companies:

• It has exposed western export grain shippers to the full cost of rail transportation. This legislation and other government support programs had previously insulated farmers from the economic realities of grain production, marketing and transportation decisions.
• It has introduced new competitive market dynamics into the western grain economy and introduced new risks in the volume and rate levels of western grain traffic for each railway.
• It has provided shippers with all the competitive access and dispute resolution provisions of the CTA, while imposing a cap on the rail rates.

CN states that the demand for grain transportation differs in one major respect from that of most other commodities. For most other commodities, the uncertainty with respect to the total demand for transportation (of all modes) arises primarily from the volatility of the market demand for the product being transported. In the case of grain, however, weather is a source of volatility that applies to both ends of the production-consumption chain. Weather conditions in Canada affect the production of Canadian grain, while weather conditions in competing exporting countries and importing countries affect the global supply and demand situation.

CN further states that in deciding whether to make an adjustment for grain risk, the Agency must consider both the volatility of an asset on its own, and its correlation with other assets of a well- diversified (but not necessarily the market) portfolio. CN asserts that the 1985 Decision implies that the latter outweighs the former by a sufficient margin to justify a downward adjustment and acknowledges that the 1985 Decision may have been correct under the WGTA regime. However, in CN's view, conditions have since changed to such an extent that the volatility of grain traffic has overtaken whatever benefits are derived from its supposed independence from the business cycle.

CN concludes that in its view a careful evaluation of the current and future grain transportation market will show that the risks are at least as high as those of the railway's other commodities.

Position of the Provinces/Wheat Pools

With respect to the first issue, the Provinces/Wheat Pools used simple regression analysis to examine the relationship of annual changes in rail tonnage under the WGTA to annual changes in national economic activity as measured by real GDP. The results demonstrated that those annual changes are not significantly correlated. The Provinces/Wheat Pools conclude that in the context of the fundamental beta formula, the demand beta of railway grain tonnage is zero. The fluctuations in grain tonnage that exist do not correlate with overall economic activity and therefore do not contribute any systematic risk to CPR or CPL. The fundamental beta for grain is zero.

The Provinces/Wheat Pools suggest that if the systematic risk of grain traffic of the railway companies is zero, and if the CAPM is a valid methodology for estimating the cost of common equity rate of the railway companies, then the conclusion is that the appropriate cost of common equity rate for grain, as measured by the CAPM, is the risk-free rate.

If, however, it is concluded that the appropriate rate of common equity for grain exceeds the risk-free rate because of non-systematic risks, then, according to the Provinces/Wheat Pools, the conclusion the Agency should reach is that the CAPM is not a valid methodology. Therefore, the CAPM cannot be consistently used in the process for which the cost of common equity rate for the movement of grain is established. Thus the cost of common equity rate must be estimated with other methods.

If the Agency concludes that the business risk of CPR as a whole is substantially the same as for the economy as a whole, then the difference between grain and other commodities is the difference between a beta of one (for CPR which is equivalent to the economy as a whole) and a beta of zero, for grain. If it is decided that the risks for CPR as a whole are less than for the economy as a whole, then the risk differential for grain would be reduced, but it would be exactly offset by an increasing risk differential for the business risk of CPR relative to CPL; the result for grain would be unchanged.

Regarding the difference between grain transportation characteristics and the transportation characteristics of other commodities, the Provinces/Wheat Pools stated that grain volumes are not particularly sensitive to freight rate levels. In a regulated environment, the railway companies can charge the full grain rates allowed without substantial attrition of traffic. For other commodities, especially resource commodities, there is the potential of substantial traffic attrition in times of declining commodity prices. When that occurs, the railway companies are required to reduce freight rates in order to protect these movements.

The repeal of the WGTA ended the federal subsidy for the transportation of western grain so that producers now pay the full rate, approximately double the amount they paid previously. The Provinces/Wheat Pools claim that this results in competition from other modes of transport, diversification of the western agricultural economy, and in lower export of grains. However, according to the Provinces/Wheat Pools, there would appear to be little potential for truck competition because of the great lengths of haul associated with the western grain movement. Also, the economics of the movement of Canadian grain to American ports for export does not make it a competitive option. The repeal of the WGTA has not resulted in a reduction in acreage in western Canada planted with the six principal export grains. Also, it is not certain that additional value-added processing on the Prairies has the impact of reducing railway traffic levels, since, as in the case for canola, the outputs themselves are frequently export products that travel by rail to destination.

The Provinces/Wheat Pools state that there is no reason to expect that any of the areas of risk associated with the repeal of the WGTA would be systematic since they are not correlated with the economy. Therefore, they are not relevant to the estimation of the cost of common equity rate for grain transportation on the basis of the CAPM.

AGENCY ANALYSIS AND DECISION

Agency Analysis

CPR and the Provinces/Wheat Pools provided interpretations of the relative variability of grain traffic to traffic created by other commodities. The assumption underlying this evidence is that more stable and less volatile traffic flows produce less risk. The Agency notes the evidence introduced by CPR that every other bulk commodity, and every other commodity except automotive traffic and import/export containers, has a greater effect in reducing the variability of total traffic than does grain. Thus, the addition of grain traffic to the total CPR traffic mix does relatively little compared with other traffic to reduce the variability, and so the riskiness of the traffic mix.

The Agency also notes that the quantity of grain shipped can vary from year to year. Quality may also vary from year to year depending, for example, on weather conditions. This, in turn, creates a volatility in the revenues and the returns generated from the movement of grain by rail. The Agency is of the opinion that grain volumes have varied and will continue to vary to a considerable extent depending on the circumstances of the weather in Canada during the growing and the shipping season, and to an unrelated extent on the climatic conditions, and economic, international trade and political situations in other parts of the world that produce grain and consume grain.

The Agency is of the view that the variability of revenues and returns, and of the intensity of use of assets provided by the railway companies for the movement of grain, creates a certain riskiness not dissimilar to the riskiness of other commodities. The Agency accepts the argument that grain is subject to additional risk; however, the rates at which the grain may move cannot be raised above the statutory maximum established pursuant to the CTA. Consequently, there is a "downside" risk to the movement of grain that does not attach to other commodities. The railway companies may reduce their rates, such as in the negotiation of incentive rates, to encourage efficient operations by grain shippers, but there is no provision for the railway companies to raise their rates above the present maximum to compensate for, for example, relatively more costly operations over light steel railway lines and remote delivery point locations with small amounts of grain to be moved.

This situation is fundamentally different than that applying to other commodities where the commodity moves on a regular cycle subject to contract with firm commitments to take the product. Coal is such a commodity.

The Agency concludes that with the repeal of the WGTA in 1995, there have been a number of attendant changes to the way the business of growing, handling and transporting grain is conducted. Where previously there had been a significant subsidy provided by the federal government, now grain producers are required to pay the entire cost of moving grain from their farms to a terminal point. Therefore, the Agency concludes that producers have been given a greater incentive to select the lowest cost combination of transportation by truck and rail or truck to a local processor. The Agency agrees that there is the risk that significant quantities of grain that customarily moved by rail will not do so in the future. Grain may be trucked to a competitive railway, or directly to market, or to a local facility for value-added processing, such as crushing of oilseeds or livestock production using feed grains. The Agency also concludes that there is a risk that the total quantity of grain produced may decline as producers withdraw marginal land from grain production, since its products can no longer support the unsubsidized costs of transportation to market. Further, grain producers may themselves diversify into other lines of business, such as livestock production, thereby reducing the quantity of grain to be moved by rail. The Agency notes the current evidence that some of the value-added production is moving by truck and that other products will be subject to truck competition in the future.

The Agency also concludes that the announced, but not as yet completed, sale of the government hopper-car fleet adds a new element of risk to the business of transporting grain. The railway companies now face the uncertainty of who will own and control the operation of these cars, which constitute approximately fifty percent of the available grain car fleet. Potential car allocation problems and conditions of use for these cars are also elements that contribute to uncertainty.

The Agency notes that the Provinces/Wheat Pools have argued that the reduction in the cost of capital rate made by the RTC in the 1985 Decision was unrelated to the passage at that time of the WGTA and its coming into force at that time. As the Provinces/Wheat Pools state in their final argument:

The Provinces/Wheat Pools concluded that it would be to the detriment of the grain producers to eliminate the reduction in the cost of capital on the basis that the WGTA has been repealed in 1995.

However, the Agency has reviewed the 1985 Decision in this regard and notes the following passage:

The Agency has concluded on the basis of this element of the 1985 Decision, taken in the context of the entire matter heard at that time, that the RTC was cognizant of the effects of the implementation of the WGTA, even if it was unable to reach empirical determinations of the precise impact. It was for that reason, among others, that the RTC exercised its informed judgement on the issue and reduced the cost of common equity rate by one percent. In the same vein, therefore, the Agency in the current case must consider the effects of the repeal of the WGTA. The Agency concludes that such an important change in the legislation governing the transportation of western grain adds to the risk faced by the railway companies.

The Agency finds that uncertainty, in and of itself, constitutes risk. The Agency acknowledges that this is a time of great change for the Canadian grain industry. The Agency accepts the evidence of Dr. Evans in which he states that uncertainty itself or the inability to predict the future translates into risk.

For the above reasons, the Agency concludes that the carriage of grain today is no less risky than the carriage of other commodities.

Agency Decision

Accordingly, the Agency has determined that for the next crop year (1997-98), the grain risk adjustment to the cost of common equity rate of CPR to determine the cost of common equity rate for the movement of grain by CPR will be zero. The Agency will continue to monitor the situation and based on its informed judgement will make a determination on whether an adjustment is required, on an annual basis.

3. Property

Because CN and CPR were both faced with a significant reevaluation of their assets in 1995, participants were asked to address whether the write-down of assets should be reflected in the net rail investment and consequently in the capital structure elements which are used in the determination of the cost of capital rates.

Position of CPR

CPR discussed the issue of generally accepted accounting principles that governed the write-down of assets incurred in 1995. CPR recognized that the write-down should be based on the conservatism principle which rules that the least favourable accounting alternative be shown in the shareholders' equity. Consequently, CPR proceeded with an evaluation of its eastern assets based on the future net recoverable amount expected, and allocated the reduction in value to a special charge in the balance sheet, deferred charges, under the properties' net accumulated depreciation amount. The offset was presented as an expense in the income statement. The reduction in value of the assets will be captured by amortization of the deferred credit account to the Income Statement over an approximate 20 year period. CPR states that the use of a deferred credit in the balance sheet using a lump-sum amount was necessary as the company's records were not sufficiently detailed to allow a branch line by branch line or subdivision by subdivision computation.

The asset write-down was an accounting entry that simply changed the timing in the charging of the undepreciated cost of assets to income. The asset write-down had no impact on the costs incurred by CPR to acquire assets and must be recovered out of the revenues received for the movement of traffic using those assets. Rate regulation focuses on asset costs and not on asset values. Write-downs and depreciation are non-cash charges and neither have any impact on income taxes. Similarly, they have no impact on the rates that must be charged by the railway company to recover its total costs, including the acquisition costs of these assets.

On the issue of whether the write-down of assets should be considered in a similar fashion as inert assets, it is CPR's opinion that the assets are still used and useful and therefore are not to be considered inert. CPR indicated that the assets in question satisfied the three criteria as set out in section 3060 of the Chartered Institute of Chartered Accountants (CICA) handbook in order to determine what constitutes a capital asset: i) it must be an asset that is held for use by the company, ii) it must be the company's intention to continue using that asset over its foreseeable useful life, and iii) the asset or group of assets must not be held for sale in the ordinary course of business. CPR claims that the eastern Canadian rail assets meet those three criteria. The type of write-down contemplated by section 3060 of the CICA handbook is not permitted for assets that would be inert or otherwise withdrawn from service.

It is therefore CPR's position that the write-down of assets is not comparable to a declaration of inert assets and that for the purposes of indexing the maximum rate scale for the movement of grain in western Canada and for other regulatory purposes, the net rail investment and the capital structure should not reflect the asset write-downs.

Position of CN

CN provided similar evidence to CPR. CN holds that the write-down of assets is not relevant to any costing applications, and that the purposes of costing are quite different from those of accounting.

CN therefore concluded that the write-down of assets was not to be treated as a declaration of inert assets and should therefore be excluded from the calculation of the net rail investment and consequently, the capital structure.

Position of the Provinces/Wheat Pools

The Provinces/Wheat Pools were of the opinion that the motivation for the write-down of assets is their diminished economic utility. In that context, they are similar to assets considered inert or not used and useful, and as a result, the position of the Provinces/Wheat Pools is that it would be appropriate to write them down in the capital structure.

AGENCY ANALYSIS AND DECISION

Agency Analysis

The Agency notes that the assets that were involved in the asset write-down continue to be used by the railway company to carry revenue traffic, and so can be considered used and useful. The Agency therefore concludes that they are not inert assets. The Agency accepts the position of the railway companies that the costs incurred in the acquisition of these assets should remain on the books of the railway companies. This means that the assets continue to be in both the cost base and the physical asset base of the carriers.

The Agency notes the decision regarding the 1996-97 crop year cost of capital rate, wherein it decided that in order to be consistent with the price indices development for the maximum rate scale, the write- down of assets was not included in the determination of the net rail investment and the capital structure for the 1996-97 crop year. Based on the evidence and arguments presented on this issue, the Agency is not convinced that there has been any change in fact or circumstances that merits a departure from this position.

Agency Decision

Accordingly, the Agency excludes the write-down of assets incurred by both railway companies in 1995 from the net rail investment and consequently from the capital structure for the determination of various cost of capital rates.

SUMMARY OF ISSUES RAISED DURING THE CONSULTATIVE HEARING

CONCLUSIONS

1. Business Risk of CPR

The Agency finds that while statistical analyses presented by each of the participants are informative they do not lead to a conclusive result on their own. The Agency has decided that for cost of capital rate determinations for the costing of 1997 railway operations in existence after the issuance of this Decision and for the next crop year (1997-98), the business risk adjustment to the cost of common equity rate for CPL to reach the cost of common equity rate for CPR will be zero. The Agency will continue to monitor the situation and based on its informed judgement will make a determination on whether an adjustment is required, on an annual basis.

2. Grain Risk Adjustment

The Agency has determined that for the next crop year (1997-98), the grain risk adjustment to the cost of common equity rate of CPR to determine the cost of common equity rate for the movement of grain by CPR will be zero. The Agency will continue to monitor the situation and based on its informed judgement will make a determination on whether an adjustment is required, on an annual basis.

3. Property

The Agency excludes the write-down of assets incurred by both railway companies in 1995 from the net rail investment and consequently from the capital structure for the determination of various cost of capital rates.

---

1. Patterson, C.S., The Cost of Capital: Theory and Estimation, Quorum Books, 1995, p. 221 ↑
2. Telecom Decision, CRTC 91-21, Ottawa 19 December 1991 ↑