INSIGHT: The Canadian Equity Risk Premium

Valuation practitioners have a tendency to use U.S. Equity Risk Premium (ERP) estimates for Canadian corporate and asset valuations. In this article, we assess the expected valuation bias when using U.S. ERPs for Canadian corporate and asset valuations by calculating our own U.S. and Canadian ERP estimates using various methodologies.

The Equity Risk Premium is the additional return that a reasonable investor expects to receive on an equity investment above a riskless investment. ERP is a major component of the cost of equity, which is used in corporate valuations and damages quantum applying any sort of discounted cash flow (DCF) approach.

The discount rate (which is, or includes, the cost of equity) used in DCF valuations should represent the relevant risks associated with the currency of the cash flows. Cash flows for Canadian corporate and asset valuations would usually be in Canadian dollars (CAD). Therefore, the discount rate should be calculated using CAD-denominated components to reflect the relevant risks. This means the ERP should represent expected Canadian equity returns.

The ERP is a forward-looking concept that is meant to quantify the expected market risk associated with future cash flows of an equity investment. ERPs can be calculated using contemporaneous data and market expectations. However, some commonly-used ERP estimates are based on historical equity returns, often using distant historical datasets.

In practice, valuation practitioners often rely on ERP estimates that are produced by financial data providers and aggregators. ERP estimates can vary widely for many reasons including: methodology, time period, data source, and the chosen risk-free rate of return.

There are two broad approaches for estimating ERP. The first approach is the backward-looking historical approach, in which the ERP is estimated using actual equity returns of a stock index and long-run average risk-free rate of return for a given time period. The second approach is the forward-looking approach, in which ERP is estimated by using a combination of contemporaneous financial market data and market expectations. We briefly outline the merits and drawbacks of these approaches in this article.

THREE ESTIMATION METHODS

Utilizing the two broad approaches, we use three methods to estimate U.S. and Canadian ERPs: the (i) Historical ERP; (ii) Supply-Side ERP; and (iii) Implied ERP. The Historical ERP and Supply-Side ERP methods use the historical approach while the Implied ERP adopts the forward-looking approach. We estimate ERP for the TSX/S&P Composite (TSX Composite) and the S&P 500 (S&P) to compare estimates of a Canadian ERP and a U.S. ERP as of Dec. 31, 2019, respectively.

Historical ERP

The Historical ERP uses actual historical data to compare equity returns with returns on a risk-free investment over a given time period. The Historical ERP is the difference between the equity and risk-free returns.

Specifically, we calculate the Historical ERP as the difference between the average annual percentage change in index value (including dividend yields) and the average yield of the relevant 10-year government security as of Dec. 31, 2019.

Supply-Side ERP

The Supply-Side ERP also uses a historical approach. However, the Supply-Side ERP strips out fluctuations in historical index values due to changes in the indices’ aggregate price/earnings ratio (PE ratio).

Stock returns can be thought of as the result of: (i) actual corporate financial performance, and (ii) investor sentiment about future performance. PE ratios reflect investor expectations of future stock performance: high PE ratios (a high stock price relative to current earnings) reveal that investors expect high growth in future earnings and low PE ratios reflect minimal growth (or decline) in future earnings. Removing fluctuations in the indices’ aggregate PE ratio removes the impact on equity returns of changing investor sentiment. This leaves the impact of actual corporate financial performance on returns. The impact of actual corporate performance on stock returns is described as the “supply-side” impact; whereas, the impact of changes to investor sentiment about future performance, captured in the PE ratio, is the demand-side impact.

We calculate the Supply-Side ERP as the difference between the average annual change in the PE ratio-adjusted index value (including dividend yields) and the average yield of the relevant 10-year government security as of Dec. 31, 2019.

Implied ERP

The Implied ERP adopts a forward-looking approach, which uses current data to estimate the ERP. The present value of a set of future cash flows is calculated by discounting the cash flows using a discount rate. One can rearrange this relationship to calculate the discount rate needed to achieve a particular present value for a given a set of cash flows.

The Implied ERP uses current market data on the index value, aggregate dividends and forecasts of future dividend growth to calculate the discount rate that equates the present value of future expected dividends to the current index value. This discount rate is the implied average equity return in the future.

The Implied ERP is calculated as the difference between the implied average equity return and the average yield of the relevant 10-year government security as of Dec. 31, 2019.

Risk-Free Rate

The ERP is the equity return premium above the risk-free rate of return. The yield on government securities are used as the risk-free rate, e.g. U.S. Treasuries and Government of Canada bonds. For the purposes of this paper, we have used the 10-year U.S. Treasury yield and the 10-year Government of Canada bond yield to calculate the US and Canadian ERPs, respectively.

Evaluation of the Various Estimation Methods

Each of the ERP estimation methods have their strengths and weaknesses. The following commentary is intended to elucidate these strengths and weaknesses and not pass judgment on which is most accurate or reliable.

The ERP is a forward-looking concept because it is applied as a component of the expected risks to estimated future cash flows. Therefore, the efficacy of the backward-looking approach is grounded in the notion that future equity returns will reflect past equity returns, at least on average. A practical problem of the backward-looking approach is that historical equity returns are highly volatile. Annual returns of the S&P 500 have been between -6.3% and 29.6% in the past 10 years. Therefore, in order to generate stable historical ERP estimates, some advocate for using the maximum available historical data, often back to the early 20th century or beyond.

It is obvious that the equity returns of the early 20th century may not resemble expected future equity returns. The modern developed economies are far more globally-integrated than in the early 20th century. The largest public companies have highly-integrated global supply chains, their products are known and sold globally and market intelligence can be known in real-time. Competition in many consumer markets is now more highly concentrated than it has ever been, but with competitors more frequently being other global giants. The largest public companies are taking greater shares of total economic profits generated in the global economy. Global political, social and environmental events regularly affect U.S. and Canadian stock prices because of risks to supply chains, market access and global brands. There are countless ways in which the economic conditions of the early 20th century bear no resemblance to the modern global economy and, therefore, many reasons why equity returns from the early 20th century may bear no resemblance to future equity returns.

The Supply-Side ERP seeks to remove the impact of investor sentiment to isolate the fundamental equity return generated by corporate performance. A rationale behind the Supply-Side ERP is that past investor sentiment about future earnings growth should not be incorporated into expectations of future equity returns. This rationale both relies on the evolution of past (often distant past) performance as the predictor of future returns and seeks to annul the evolution of past expectations of future returns in the prediction of future returns. Additionally, adjustment to calculate the Supply-Side ERP from the Historical ERP can be highly sensitive to the precise historical time period that is used.

The forward-looking approach may align with the forward-looking nature of the ERP, but these methodologies typically rely on sensitive modeling assumptions compared to the backward-looking methods. Forward-looking methods usually rely on an efficient markets hypothesis, whereby the index value on any day accurately reflects expected future returns. Furthermore, key inputs, such as the current index value and earnings growth forecasts, can have a wide range of feasible values and are highly influential on the ERP estimate.

Some advocate for other approaches entirely. Surveys of the ERP values used by CFO’s, equity analysts and academics can be thought of as a sort of ‘Wisdom of the Crowds’ approach.

DATA

We calculate the ERPs using the TSX Composite to represent Canadian equity returns and the S&P 500 to represent U.S. equity returns.

The TSX Composite is considered the benchmark Canadian stock index as it has historically contained the majority of the total market capitalization of the Toronto Stock Exchange (TSX), the largest stock exchange in Canada. Currently, the TSX Composite represents approximately 70% of the total market capitalization of the TSX with more than 240 companies included.

The S&P is commonly considered as the benchmark American stock index as it contains the 500 largest public companies listed on the New York Stock Exchange or the NASDAQ. The S&P covers a wide range of U.S. industry and is widely recognized as a fair representation of the broad U.S. stock market.

ERP ESTIMATES

Table 3-1: ERP estimates

Methodology

The Historical ERP provides the highest U.S. ERP estimate and the largest difference between the US and Canadian ERP estimates at 1.5 percentage points. Both the US and Canadian Supply-Side ERP estimates adjust their respective Historical ERPs lower: the U.S. Supply-Side ERP is 1.0 percentage point lower while the Canadian Supply-Side ERP is half of a percentage point lower. The PE ratios of the market indices in 2019 have been at close to the highest levels seen over the past 25 years, which contributed to the Supply-Side ERP adjustment to the Historical ERPs.

The Implied ERP method produces the most similar ERP estimates comparing across the US and Canadian ERPs with a difference of 0.4 percentage points. However, the U.S. Implied ERP is the lowest US ERP estimate, while the Canadian Implied ERP is the highest Canadian ERP estimate.

The average of the U.S. ERP estimates, at 6.4%, is notably higher than the average of the Canadian ERP estimates at 5.7%. All else equal, a higher ERP leads to lower valuations because it contributes to a larger cost of equity and discount rate. This will more heavily discount expected future cash flows, which will lead to a lower present value of those cash flows.

CONCLUSION

Our estimates of the U.S.and Canadian ERPs suggest that choosing to use a U.S. ERP may lead to lower Canadian corporate and asset valuations than had a Canadian ERP been chosen. Given that in theory, CAD-denominated cash flows should be valued in consideration of CAD-specific risks and that US ERPs currently appear to be broadly lower than Canadian ERPs, we strongly promote the use of ERPs derived from Canadian assets over U.S. assets for Canadian corporate and asset valuations.

This column does not necessarily reflect the opinion of The Bureau of National Affairs, Inc. or its owners.

Author Information

Alain de Bossart is a Senior Director at FTI Consulting, based in Toronto. Mr. de Bossart in applied economics, econometrics and business valuation for litigation support, regulation and business advisory services.

Jason Kim is a Consultant at FTI Consulting, based in Toronto. Mr. Kim specializes in areas of business valuation and damages quantification for both domestic and international disputes.