Assessment of Wal-Mart valuation using different methods

To test the assumption of a discount rate of 7% as given in the outline of the case, we calculated the required rate of return for the Wal-Mart stock using CAPM . Using rWalMart = Rf + βWalMart [E(RM) – RF], we find the required rate of return to be 7.01% and in line with the information given in the case outline. Perpetual dividend growth model:

The standard method of calculating a stock price using the perpetual dividend growth model is done by assessing a company’s dividend one year into the future adding the future expected growth rate. The formula is written as: P0 = D1/(Ke − g), where Ke is the investor required return, D1 is next year’s dividend and g is the expected …show more content…

We recognize the calculated theoretical value is considerable higher than the market value of the Wal-Mart stock, which could provide a strong indication for investing. However, we also recognize the weakness of the three-stage model thus we are careful to draw too decisive conclusion. We are especially concerned about the sensitivity and impact on the end theoretical price when making just marginal changes in the input factor. Valuation models in general are sensitive to the input factors, but we believe the extra complexity in the three-stage model amplifies even small forecast errors. We understand that the three-stage model can be very useful for companies approaching the transition phase between growth phase and consolidation, but Wal-Mart does in our opinion not fulfill this characteristic. Thus, leading us to conclude the three-stage model isn’t particularly applicable to valuation studies of Wal-Mart.

Price/Earnings multiple approach:

The debate on whether to use trailing or projected price/earnings multiple is ongoing. We have found evidence in literature that there is no clear preference on which method to use as both have advantages and disadvantages . We have no strong view or preference generally speaking, but in the case of Wal-Mart we believe there is a case for use trialing data rather than projected data. For reference we have calculated both using P0 = EPS * P/E .

We see the range of estimates is very wide,