MPF753 Finance : Despite the Concerns in CAPM

a) All interim dividends must be appropriately annualized before adding up with the year-end final dividends in order to determine the true dollar value of yearly dividends received by the shareholders. For companies that have paid interim dividends, assume that those dividends were paid at the end of the first half of the year and therefore earn six months of interest at a risk-free rate for the next half. The current Australian Government 10-year bonds rate is 2.88% p.a.

 

b) After determining the dollar dividends received by shareholders for each of the past ten years, compute and justify a proxy annual constant growth rate of dividends to be used in Gordon’s model. Use your computed proxy annual constant growth rate to predict next year’s dollar dividend value.

 

c) Finance and look up the closing price of each of your four chosen stocks Use Gordon’s model to solve for the expected return on equity for each of the stocks. Do these expected return figures appear justified given the nature of the business, the overall market conditions and the industrial sectors within which each of your chosen companies operate?

 

d) What do you feel are the most serious methodological problems associated with Gordon’s model? Carry out a review of relevant financial academic literature and identify at least two alternative cost of equity estimation methods. Can these identified methods be better than Gordon’s model? Argue your case.

The four companies selected for this activity are as follows.

• BHP Billiton (BHP) (Mining)
• Commonwealth Bank (CWB) (Banking)
• Woolworths Ltd. (WOW) (Retail)
• Telstra Limited  (Telstra)(Telecom)

1. Further the annualised dividend on an yearly basis for the given period and companies is summarised below.

The annualised dividends have been calculated by the following formula = Final Dividend + Interim Dividend + Interest @ 2.88% pa for 6 months on interim dividend

1. The growth rate of dividends based on the given data can be summarised as shown in the table below.

To get a proxy growth in dividend one way could be take the average of dividend growth in the given years, however the years leading to the global financial crisis i.e. 2006-2008 were exceptionally outperforming years where nearly every business were outperforming. Further during 2008-2009, the brunt of the crisis has been seen while post that during 2009-2011 the recovery phase has led to abnormal growth in dividends. Clearly the data from 2006 -2011 hints towards fluctuation in macroeconomic conditions the world over and hence thus the dividend rate growth may not be a fair indicator of the future growth. Hence as a proxy growth for sustainable dividend growths, only the data from 2012-2015 is taken in to consideration to find the average growth rate which is assumed to continue going forward.

Proxy growth rate for CWB (2012-2015) – 7.09%

Proxy growth rate for BHP (2012-2015) – 5.33%

Proxy growth rate for WOW (2012-2015) – 3.33%

Proxy growth rate for Telstra (2012-2015) – 2.18%

Next year dividend for CWB = 4.23*1.0709 = $ 4.53

Next year dividend for BHP = 1.25*1.0533 = $ 1.3155

Next year dividend for WOW = 1.4*1.0333 = $ 1.446

Next year dividend for Telstra = 0.31*1.0218 = $ 0.3139

1. As per the Gordon constant dividend growth model (Deegan, 2014),

P = D₁/(r-g)

Where P = Price of stock

D₁ = Next year dividend

r = expected return on equity

g = constant dividend growth rate

Further the closing stock price of the given stocks as on June 30, 2015 are captured in the table below (Yahoo Finance, 2015).

Hence expected return (r) (CWB) = (4.53/84.67)+ 7.09% = 12.44%

Hence expected return (r) (BHP) = (1.3155/27.05)+ 5.33% = 10.19%

Hence expected return (r) (WOW) = (1.446/26.96)+ 3.33% = 8.69%

Hence expected return (r) (Telstra) = (0.3139/6.14)+ 2.18% = 7.29%

In order to comment on the viability of the above expected returns, it would be prudent to first estimate the return on ASX 200 during the period of consideration i.e. from July 1, 2012 to June 30, 2015, we get that return on ASX 200 is around 8.3% per annum.

CWB – As the global economy is getting out of crisis especially with the US on the recovery path, it may be estimated that the banking industry would see better times in the future and hence is likely to outperform the market. Thus expected return seems realistic.

BHP – Currently due to slowdown in China, the global commodity prices are at its lows but considering the asset base that BHP has and its market share in various global commodities, even in bad times, it would outperform the market ever so slightly and hence the expected rate seems realistic.

WOW – Considering that it is the largest organised supermarket of Australia, and with the increase in consumer spending as economic growth improves, it may be expected that the given returns are very realistic and would be easily met.

Telstra – The company was facing difficult times in the recent past but is now on the growth path and hence the expected return seems to be highly justified.

• This method can be only used for firms those who have a stable dividend history so that the next year dividend can be predicted with some degree of certainty. For a firm in the growth phase, either dividends would not exist or their quantum would be less or they would be uncertain. It is difficult to apply Gordon’s model to determine the expected return on equity in all such cases.
• The Gordon’s model assumes that a constant stable dividend growth rate is possible and should be used for computation. But in the long run, the firms need to innovate and typically there are drastic changes and hence estimating a constant dividend growth rate with reliability is not possible a number of times. Hence the constant dividend growth rate can be highly biased and influenced by the recent performance of the given firm.

Hence in light of the above, it is imperative that alternative models must be explored which an give more reliable estimates of cost of equity considering it is a key variable used in capital budgeting and investment decision making and simultaneously difficult to predict with accuracy (Patterson, 2005). One of the most credible alternatives to the dividend discount approach for estimating cost of equity is the CAPM (Capital Asset Pricing Model). One of the central tenets of this approach is to compensate the investors for investing in equity which is inherently riskier than debt and thus logically should have a higher cost than debt. This is imperative as investors would only tend to invest in equity in case their extra risk is compensated with potential extra returns. Additionally the CAPM model also intends to compensate with regards to the core principle of time value of money. This is made clear from the formula of the CAPM (Bruner et. al, 2008).

Cost of equity = Risk free rate + Beta * (Stock market return – Risk free rate)

From the above, it is apparent that the first component i.e. Risk free rate seeks to compensate investor for the time value of money so as to provide them with the risk free rate as this is the opportunity cost the equity investor has. The second component in the above formula seeks to compensate the investor for the risk taken by investing in equity. For this, it first calculates the market risk premium (compensation for stock market) which is then multiplied by the beta of the stock which is representative of the relative risk of the stock in comparison with the market index. In the modern world, CAPM is the most widely used method by both practitioners and academicians for estimating the cost of equity which further is used to calculate WACC (Weighted Average Cost of Capital) and extensively used in capital budgeting techniques (Deegan, 2014).

Even though the CAPM technique is used extensively but still there are certain issues with its usage which must be kept in mind. Firstly beta determination is difficult in case of unlisted firms and hence beta determination has to be done usually by finding a suitable peer group in accordance with the underlying business and using its beta for estimating the cost of equity. However in this approach adjustment needs to be made for the differences in the capital structure of the peer group and the firm whose beta is to be estimated. It has been found that often this adjustment is not carried out (Patterson, 2005). Additionally CAPM does not account for any unsystematic risk which is especially of concern with regards to investors those who invest in private unlisted firms since most of these investors are not diversified and tend to identify in only few identified sectors or a particular segment. Clearly CAPM needs to be adjusted for this but this is rarely done (Bali, Cakici & Tang, 2009).

However despite the concerns in CAPM, it may be concluded that this method is better than the Gordon dividend model. This is primarily because of the following reasons (Brealey, Myers & Allen, 2008).

• The inputs of CAPM are essentially based in the past or the present and thus do not require any prediction going forward and thus has less methodological issues as reliable inputs are easily available.
• The cost of equity can be determined for growth companies in a reliable manner and additionally those which do not pay any dividend.
• This model can be deployed for even for calculating the cost of equity of even unlisted firms.

Another alternative method for calculation of equity is the bond yield plus risk premium approach. As per this, since equity is a much riskier investment as compared to the bonds (debt), hence the investors need to be compensated for the additional risk taken in form of s risk premium over and above the cost of debt. Further this approach assumes that the factors which impact cost of equity and cost of debt are essentially the same with the magnitude of impact also being same and hence advocates that a given stock the risk premium remains constant over a period of time (Brealey, Myers & Allen, 2008). However this method is applied only for those companies which have bonds which are traded in secondary financial markets. Additionally the assumption that the risk premium would remain constant over time seems dubious. However despite the shortcoming and limitations, it a simple and effective way of estimating the cost of equity of those companies which have publicly traded bonds. Hence for such companies, this may be a preferable method compared to Gordon dividend model as it has comparatively less methodological issues and also is based on comparatively more realistic assumptions (Patterson, 2005).

References

Bali, T., Cakici, N. & Tang, Y. 2009. ‘The conditional beta and the cross section of expected returns’. Vol. 38 No. 3, pp. 103–137

Brealey, R, Myers, S & Allen, F 2008, Principles of Corporate Finance, 9^th edition, McGraw Hill Publications, New York

Bruner, R., Li, W., Kritzman, M., Myrgren, S., Page, S. 2008, ‘Market integration in developed and emerging markets: Evidence from the CAPM’, Emerging Markets Review, Vol. 9 No.2, pp. 89–103.

Deegan, C.M. 2014. Financial Accounting Theory, 4^th Edition, McGraw-Hill Education Australia, Sydney

Graham, J. & Harvey C. 2001, ‘The theory and practice of corporate finance: Evidence from the field’, Journal of Financial Economics, Vol. 60 No. 3, pp. 187-243

Patterson, C.S. 2005. The Cost of Capital: Theory and Practice, 3^rd edition, Greenwood Publishing Group, New York

Yahoo Finance 2015. Historical Prices of Stocks , Yahoo Finance, Available online from https://au.finance.yahoo.com/q/hp?s=%5EAXJO&a=06&b=1&c=2005&d=08&e=4&f=2015&g=m (Accessed on August 29, 2015)