It has been well established that estimated beta factor for identical company can vary significantly depending on the data inputs of the regression analysis. Time period, return interval and proxy for market portfolio are key factors that have a substantial impact on the estimated parameter of the regression analysis. Different results of estimated betas consequently lead to various assessments of required rate of return. Capital asset pricing model proposes that theoretically appropriate required rate of return depends on risk-free rate, market risk premium and asset’s sensitivity to non-diversifiable risk, often represented by the quantity beta. The CAPM returns the asset-appropriate required return or discount rate at which future cash flows produced by the asset should be discounted given asset’s relative riskiness. It is intuitive that different estimates of discount rates result in diverse approximations of asset’s value.
Impact of beta on the value of the company X
Let us first consider company X that needs to be evaluated. Initial step is to estimate beta and for purpose of this example several different betas will be estimated. More specifically 16 different betas will be estimated. We consider two different scenarios. First scenario describes company X as a large well known company; consequently, stock of company X is frequently traded. In the second scenario stocks of company X are not frequently traded and thin-trading is present. For both scenarios two different estimation periods, two different return intervals and two different proxies for market portfolio are used to estimate beta. The tables below represent the estimated regression coefficients for both scenarios:
Estimates vary depending on the data used and as expected betas in the second scenario are lower compared to the first scenario. As mentioned, different betas result in diverse required rates of return in accordance with CAPM. To calculate required rate of return we made assumptions regarding risk free rate and market risk premium. Note that risk free rate equals 3.5% and market risk premium equals 6.2%. Further we assume that the nominal value of the company X in the future period is 1 million EUR. To simplify we assume that in one case the period is 1 year and in second case this period is 5 years and only a simple discounted method is used to determine the present value of such company. The figure below represents the present value of the company X with respect to different betas.
Present value of the company X fluctuates from EUR 944 thousand to EUR 902 thousand in the case with 1 year period between the time of valuation and nominal value., in a five year period impact is even bigger.
The example described above demonstrates how different ways of estimating beta can yield significantly diverse approximations of present value of some asset and concludes a series of articles regarding the issues with estimating beta.
There is no universal answer on how to tackle the beta estimation. The most appropriate method should be determined by the characteristics of the analyzed company.