Valuation of Equity
Valuation of Equity
Valuation of Equity
Thornton, Barry
Jacksonville University
Assistant Professor Economics
bthornt@ju.edu
ABSTRACT
∞
P * t = 1 + ∑ ( ROEt + 1 − r ) * B e , /Bt
t +1−1
i =1
∞
IVt / BVt = 1 + ∑ [( ROEt + 1 − Re) *(1 + gt + i − 1)] /(1 + Re) where
i =1
DATA SET
METHODOLOGY
EMPIRICAL EVIDENCE
Since Ball and Brown (1968), most empirical research on private firm
valuation has been conducted employing a multi-linear regression model where
stock return (or price) appears as the dependent variable and contemporaneous
accounting data appear as the independent variables. Accordingly, accounting
data variables that better explain (higher r2) contemporaneous return (or price)
are considered more “value relevant,” and are generally assumed to provide a
sound theoretical basis for valuation. The critical question to be addressed in
this research is whether an abnormal yield spread is positively correlated to stock
prices. In order to test this hypothesis, a stepwise OLS regression analysis was
performed. The independent variable is abnormal return spread and the
dependent variable is firm value (market value of equity and debt divided by
equity).
RESULTS
If one assumes that price is proportional to the average P/E ratio for a set
of comparable firms, the ex post performance of the P/E model is inferior to other
parsimonious alternative valuation models. Moreover, the results indicate that
predictive errors are negatively correlated with earnings or book values,
suggesting under-pricing (discounts) for the smallest firms. Our results suggest
that multivariate linear regression models incorporating an intercepts, earnings,
book value of equity and market value of invested capital corrects econometric
problems of the proportional models. The results of the study suggest that
earnings, book value, and spread between cost of capital and ROE provide
important incremental information in predicting price. Finally, a surprisingly small
improvement in explanatory power of models estimated on industry sub samples.
If one assumes that all firms have identical AR persistence parameters in
abnormal earnings, that firms have identical discount rates and that non-
accounting information is either value irrelevant or affects all firms in exactly the
same way. In fact, if these assumptions are empirically descriptive, one only
needs to use market multiples (e.g., P/E and P/B ratios) for valuation purposes.
CONCLUSIONS
Table 1
Methods, models and approximations
Profitability Measures
ROA = Return on assets defined as income after taxes divided by the average
total assets, expressed as a percentage.
ROE = Return on equity measured as income available to common equity
divided by average common equity, expressed as a percentage.
SPREAD = Return on common equity (ROE) minus the weighted average cost of
capital.
PRICE/BV = Current market value of common equity divided by the average
book value of equity.
MVIC/PRICE = Invested capital measured as average total long term debt, other
long term liabilities and shareholder equity divided by the current market value of
common equity.
PRICE/BV = Current market value of the common equity divided by average
book value of common equity.
PRICE/EBT = Current market value of the common equity divided by earnings
available to shareholders before tax.
PRICE/NI = Current market value of the common equity divided by earnings
available to shareholder after tax.
MVIC/EBITDA = Invested capital measured as average total long term debt,
other long term liabilities and shareholder equity divided by earnings available to
common equity before tax, depreciation and amortization.
MVIC/EBIT = Invested capital measured as average total long term debt, other
long term liabilities and shareholder equity divided by earnings available to
common equity before interest and tax.
MVIC/DISEARN = Invested capital measured as average total long term debt,
other long term liabilities and shareholder equity divided by discretionary
earnings (FCF).
MVIC/BVIC = Invested capital measured as average total long term debt, other
long term liabilities and shareholder equity divided by the average book value of
invested capital
Regression Models
Regression Model 1: Dependent variable is price/bv; independent variables are
spread, bv/equity, mvic/sales, mivc/ebitda/ mvic/bvic, profit margin, dq/cf, eq/ni.
Regression Model 2: Dependent variable is price/bv; independent variables are
spread, profit margin, ebit/sales, ato, fato, ebit/interest expense, and leverage.
Regression Model 3: Dependent variable is mvic/price; independent variables are
spread, bv/equity, mvic/sales, mivc/ebitda/ mvic/bvic, profit margin, dq/cf, eq/ni.
Regression Model 4: Dependent variable is mvic/price; independent variables are
spread, profit margin, ebit/sales, ato, fato, ebit/interest expense, and leverage.
Table 2
Descriptive Statistics for Sample of Firms
Table 3
Pearson Correlation Coefficients Comparing
Actual Equity Price with Valuation Models
Table 4
Regression analysis for value of firms by different valuation methods
REFERENCES
Abrams, Jay B., (1994), “Discount for lack of marketability: A theoretical model,”
Business Valuation Review, September, 1994
Alderson, Michael J. Betker, Brian L. (1999). "Assessing post-bankruptcy
performance: An analysis of reorganized firms' cash flows," Financial
Management, Summer 1999.
Ang, J. (1991). “Small business uniqueness and the theory of financial
. management,” The Journal of Small Business Finance, Fall 1991