FINC2011 Assignmeent S2 2018 PDF

Preview

Summary

Assignment on Harvey Norman...

Description

480435178 EXECUTIVE SUMMARY Harvey Norman Holdings Ltd. (HVN) is an Australia-based, multi-national company in the Consumer Discretionary sector (ASX 2018). The company is primarily involved in providing franchised retailing services to Harvey Norman franchisees, which sell furniture and consumer electrical products. Harvey Norman is also involved in property investment, the leasing of premises to franchisees and third parties, and the provision of consumer finance.

Harvey Norman’s share price has generally aligned with the performance of the housing sector and the economy at large, peaking at $7.03 on 26 November 2007 at the height of the 2007 housing bubble, before plunging to its lowest at $1.99 on 23 February 2009. In the most recent 5 years, HVN has experienced steady growth in profits and share price. This report conducts a valuation of Harvey Norman Holdings Limited (HVN) stock to inform the investment decisions of existing and prospective stockholders. Extensive research of the company, the lifestyle retail industry, economic forecasts, and academic literature was undertaken. Various applications of the dividend discount model and capital asset pricing model were performed, with their respective advantages and limitations considered.

CALCULATION OF BETA Harvey Norman’s beta represents the security’s volatility to the overall Australian market and the responsiveness of the security’s return to the overall return in the market. To accurately assess Harvey Norman’s beta, the time period and return interval of data, and market index must be carefully chosen.

Beta estimates are most accurate when calculated across a time period in which economic conditions are similar to those predicted in the relevant investment holding period (Coppedge, Lamb, and McCague 2012). As such, a five-year time period was used as the strong performance in the home and lifestyle retail industry in the last five years is expected to continue for at least the next such period (IBISWorld 2016). Thus, historical data from 20 October 2013 to 20 October 2018 was used. Monthly returns are more reliable than higher frequency returns as they incorporate sufficient data to maintain accuracy, while limiting random fluctuations or ‘noise’ arising from

1

480435178 analysing daily or weekly prices that could distort the data (Slaubaugh 1992). The choice to use monthly returns for five years is consistent with both standard convention (Brigham and Gapenski 1997) and experimental evidence that these parameters deliver the most accurate beta forecast (Cenesizoglu et al. 2014). The S&P/ASX 200 market index was used to measure the overall value of the market. Market indexes with more securities provide a better representation of the market portfolio (Damodaran 2009). The ASX200 contains the top 200 ASX listed companies, making it the most suitable representation of the market. The All Ordinaries Index was not used because Harvey Norman is a relatively large company that is less likely to be affected by price movements of smaller firms.

Calculation: The formula for Harvey Norman’s beta is shown below β=

Covariance(r m r p) Variance(r m)

β

= Harvey Norman’s beta

Covariance(r m r p ) = covariance between Harvey Norman’s return and market return Variance(r m ) = variance of market returns Historical monthly data of Harvey Norman’s adjusted closing prices and the ASX 200 market adjusted close were inputted into Microsoft Excel. Their monthly returns were calculated and from this set of data, Microsoft Excel is able to determine the following: Covariance(r m r p ) = 0.001156456 Variance(r m ) = 0.001348833 Applying the beta formula:

β=

Covariance (r m r p) 0.001156456 = =0.857375 ≈ 0.8574 Variance(r m) 0.001348833

HVN’s beta of 0.8574 indicates that HVN stock has lower risk than the market. For every 1%

2

480435178 increase in the market price, HVN’s stock price is expected to increase by 0.8574%.

Comparison: The beta published by Reuters and the Financial Times (2018) are 0.54 and 0.5276 respectively. The published beta is lower than the calculation found in this report due to the difference in methods and parameters. Both Reuters and The Financial Times perform a regression analysis comparing Harvey Norman’s monthly returns over a 3-year interval with the S&P 500 market index (Gozali 2010). Consequently, the published betas are lower because the short time period primarily captures Harvey Norman’s outstanding performance in the most recent two years due to the substantial growth in housing sector (Palmer and Greenblat 2016). This outstanding growth is not likely to be sustained indefinitely, particularly with the rise of the online retailing industry that Harvey Norman is not highly competitive in (Magyer 2016). The longer time period used in this report’s beta calculation is a more comprehensive descriptor of the economy as it encapsulates the full range of Harvey Norman’s performance over a business cycle (Chung 2016). The use of the S&P 500 index may also have contributed to the lower beta estimates as it is an American stock market index, providing justification of Reuters’ and The Financial Times’ beta figures which are similar to each other but contrasting to the beta calculated in this report.

USING CAPM MODEL TO DETERMINE DISCOUNT RATE Harvey Norman’s required rate of return may now be calculated by using the CAPM model. This model states a security’s risk premium is the product of the security’s beta and the market risk premium. Therefore, the security’s rate of return is the sum of its risk premium and the market risk free rate.

The risk-free rate may be determined using the yield of a 10-year Australian government bond, which is approximately 2.71% p.a. as of October 2018 (Bloomberg 2018). Market risk premium of 6.5% p.a. has been stipulated and beta is 0.8574 as calculated above. The CAPM model formula is shown below: Ri=R f +β (r m −r f ) Ri = expected return on Harvey Norman’s security Rf

= risk free rate = 2.71% p.a.

3

480435178 β = Harvey Norman’s beta

≈ 0.8574

r m−r f = market risk premium = 6.5% p.a.

Using these inputs,

Ri=0.0271+0.8574 (0.065) = 0.082831

Therefore, Harvey Norman’s required rate of return

≈ 8.3% p.a.

CONSTANT DIVIDEND GROWTH MODEL A constant dividend growth model assumes dividends grow at a constant rate in perpetuity, and hence valuation of stock with such dividend nature is calculated through an application of the growing perpetuity formula as shown below.

P0 =

E(¿ 1) r −g

P0 = security’s share price at present E(¿ 1) = expected dividend for end of period 1 r = required rate of return = 8.3% p.a. g

= constant dividend growth rate = 4% p.a.

Effective semi-annual rates: However, Harvey Norman pays dividends semi-annually and so effective semi-annual rates must be calculated to match the timing of dividend cash flows in order to appropriately apply the above constant dividend growth formula. Calculations of these effective rates are shown below. For dividend growth rate: Effective semi-annual growth rate = 1.041/2 – 1

≈ 1.9804%

For required rate of return: Effective semi-annual rate of return = 1.0831/2 – 1

≈ 4.0673%

Dividends for the next period may now be estimated by multiplying the most recent dividends with the effective semi-annual growth rate. Note, periods in this calculation are semi-annual E ( ¿1 ) =¿ 0( 1+ g ) =$ 0.18( 1.019804 )=$ 0.18356472

4

480435178 Finally, the constant dividend growth formula may be used.

P0=

E(¿ 1) 0.18356472 = =$ 8.80 r −g 0.040673−0.019804

However, assumptions underlying this value include a constant required rate of return during the perpetual 4% p.a. dividend growth. Furthermore, by applying a semi-annual effective dividend growth rate, it is assumed that interim and final dividends grow at the same rate. The trend shown by Harvey Norman’s pattern of dividend payment suggest a higher final dividend relative to interim dividend, and as calculations above are based on final dividend, this assumption, if incorrect, will result in an overstated valuation. Under the above assumptions, the perpetual 4% p.a. growth of dividends will result in a valuation of Harvey Norman’s stock price as $8.80. The accuracy of this stock price is questionable due to the limitations of the Gordon growth model. Despite being simple, the model is extremely sensitive to the inputs used and can yield misleading results if they are slightly inaccurate. In particular, it is difficult to determine the stable growth rate, g , to be applied. This is caused by the inherent uncertainty associated with estimating expected

inflation and growth in the economy (New York University Stern School of Business n.d.). Further, the model assumes that stock price is solely dependent on the discounted value of future dividends. However, firms may retain a portion of their earnings to re-invest in future projects, which may in fact increase stock price (Michaud and Davis 1982). As such, the model will underestimate the stock price for firms that have a low payout ratio (New York University Stern School of Business n.d.). Consequently, the Gordon growth model has limited application, being only suited to value stocks for firms with high payout ratios, and a stable growth rate lower than the nominal growth of the economy (NewYork University Stern School of Business n.d.).

TWO-STAGE DIVIDEND GROWTH RATE MODEL The stock price will now be calculated assuming the 4% p.a. growth rate only applies for 5 years, before the growth rate reverts to a long-term rate consistent with the forecasted inflation rate. In order to appropriately complete calculations in regard to the above assumption, forecasted inflation for 2023 is required. In the RBA’s most recent Statement of Monetary Policy from August

5

480435178 2018, inflation is only forecasted up until December 2020, which is predicted to be 2.25% p.a. (RBA 2018). However, in 2014 the Australian Energy Regulator (AER) has forecasted beyond this, estimating 2013 inflation as 2.5% p.a. (AER 2014). Although this source is 4 years old and thus factors arising in the last 4 years are not accounted for, it is consistent with the increasing trend and growth rate shown in RBA’s inflation statistics and forecasts (RBA 2018). Furthermore, it is the midpoint of the RBA’s 2-3% inflation target, thus a forecasted inflation rate of 2.5% p.a. will be applied in the following calculations An effective semi-annual forecasted inflation rate must be calculated to match the timing of the semiannual dividends paid by Harvey Norman. 1+ annual rate ¿ ¿ Semi annualrate=¿

= 1.0251/2 – 1 = 0.01242284 ≈ 1.2423%

Therefore, the effective semi-annual forecasted inflation rate, which is the effective semi-annual long term dividend growth rate from the end of year 5 onwards is 1.2423%. Semi-annual time periods will be sued in the following calculations in order to match the timings of the semi-annual dividend payments i.e. end of year 1 is the end of period 2.

Calculation of stock price using this two-stage dividend growth model consist of three steps: 1. Stock at beginning of long run (beginning of Year 6/Period 11) valued using a constant dividend growth model 2. Near term dividends (Years 0-5/Period 0-10) valued 3. Discounting both near-term dividends as well as stock price at beginning of long-run back to present values

Step 1: A constant dividend growth model is applied to value stock at the beginning of the long-run, which is at the end of period 10.

P10=

E (¿11 ) r −i

6

480435178 P10 = undiscounted stock price at end of period 10 E(¿ 1) = expected dividend for end of period 1 r = required rate of return = 4.0673% semi-annually i = long term dividend growth rate = forecasted inflation = 1.2423% semi-annually

To calculate forecasted dividends for Period 11, dividends for Period 10 must first be determined using the short-term growth rate (g) of 4% p.a. or 1.9804% semi-annually. The growing perpetuity formula to determine price at

P10 may then be applied.

E ( ¿10 ) =¿ 0( 1+ g )n =$ 0.18( 1.019804 )10=$ 0.21899 1

1

E ( ¿11 ) =¿10( 1+i ) =$ 0.21899 ( 1.012423 ) =$ 0.221711

P10=

0.221711 =7.8481 =$ 7.85 0.040673−0.012423

Step 2 Near term dividends grow at a constant rate of 4% p.a. and hence a growing annuity formula is applied to value these cash flows over the first 5 years or equivalent 10 periods.

[ ( )] t

PV =

C 1+ g 1− 1+r r−g

PV

= present value of dividends paid during first 10 periods

r = required rate of return = 4.0673% semi annually g

= short term dividend growth rate = 1.9804% semi annually

t = amount of time periods = 10 1

C=E (¿1 ) =¿ 0( 1+ g )

PV =

=

[ ( )]

C 1+ g 1− 1+r r−g

0.18356472

t

=

[ (

0.18356472 1.019804 1− 1.040673 0.040673−0.019804

)

10

]

=¿ 1.61294

Step 3 The present value of dividends paid during the first 10 periods as calculated in step 2, in addition to the PV of P10 , calculated in step 1, will determine

P0 .

7

480435178

10 P0=1.61294+ (7.8481 ÷ 1.040673 )=6.8806 ≈ $ 6.88

Thus the stock price of Harvey Norman under the above growth assumption is $6.88. Note Harvey Norman pays dividends during April and October. These calculations assume dividends are paid at the end of these months, and hence the price above represents the stock price at October 2018.

Impact on stock valuation: Compared to valuation where the 4% p.a. growth of dividends are assumed to apply in perpetuity, the stock price is lower in value when growth is assumed to decrease to forecasted inflation rate of 2.5% p.a. after 5 years. Mathematically, as the denominator of the perpetual growth formula has a lower ‘g’ figure, the denominator (r – g) will increase and as such the whole term will decrease, assuming a constant r and Div0. Intuitively, as dividend streams are expected to grow at a lower rate after the first 5 years, investors should value this security at a lower value as a result of lower amount of future cash flows. Hence, according to this stock valuation method based on the present value of future cash flows, a decrease in forecasted dividend growth rate after the initial 5 years will lead to a lower valuation of Harvey Norman’s security.

“REALISTIC” REPRESENTATION OF CURRENT STOCK PRICE A stock price for Harvey Norman will now be calculated using the company’s payout ratio and return on equity (ROE) to determine a perpetual dividend growth rate. By computing the company’s payout ratio, the ‘sustainable growth rate’, the maximum rate at which a company can grow by relying only on its retained earnings, can be calculated (Grant 2003). This method is more accurate than using historical data as it analyses the firm’s actual internal operations which underpin its growth capacity (Grant2003). Thus, Harvey Norman’s sustainable growth rate was used. g=ROE( 1−p )

ROE = return on equity p = payout ratio A five-year average was taken as this historical period is most representative of the next five-year period, as previously established. Harvey Norman’s five-year average return on equity from 2011 to 2016 was 9.52%, and average payout ratio was 76%. g=0.0952 ( 1− 0.76) =0.022848

8

480435178

1/ 2

Semi annual g=( 1+0.022848 ) −1=0.011360= 0.0114

Using this growth rate, the expected dividend payment for next period was calculated. E ( ¿1 ) =¿ 0( 1+ g )=$ 0.18 ( 1.0114 ) =$ 0.182052

Using the constant dividend growth formula P0=

E(¿ 1) 0.182052 = =6.2191 =$ 6.22 r −g 0.040673−0.0114

Thus, using a more accurate ‘g’ figure, a stock price of $6.22 has been valued

METHOD OF COMPARABLES An alternative method in which Harvey Norman stock may be valued is by multiplying its estimated earnings per share at the end of period 1, with the average price to earnings ratio of comparable companies. The formula is: P V = (E) E V = value of Harvey Norman stock P/E = price to earnings ratio of comparable company E = Harvey Norman estimated earnings per share at end of period 1 E=E 0 ( 1+ g )=32.13 ( 1.0114 ) =33.4952=33.50 cents The comparable company chosen is Myer Australia, due to the similarities between the 2 companies including recent performance, target market, product range and business size. Using the formula, P V = ( E )= 48.64 ( 33.50 ) =1629.44 cents∨$ 16.29 E This method values Harvey Norman stock much higher than done previously in this report, which may speak to the inaccuracy of this stock valuation method, as reliance on analysts’ subjective judgment as to which companies to compare with, as well as the method of estimating growth, this method is inherently biased. While analyst forecasts of earnings per share, the primary method of

9

480435178 growth rate estimates, have indeed been closer to the actual values than those derived from historical data, their accuracy diminishes with longer forecasting periods.

10

480435178 REFERENCES: ASX 2015, Harvey Norman Holdings Limited, ASX, viewed 20 October 2018,

Australian Competition and Consumer Commission 2010, Betas and other CAPM parameters for Australia Post’s reserved letter business, Australian Competition and Consumer Commission, viewed 12 October 2018,...