Valuing Small Businesses: Discounted Cash Flow, Earnings Capitalization and the Cost of Replacing Capital Assets

When valuing small businesses, practitioners have a number of models from which to choose. The model most commonly chosen does not have strong theoretical support in the literature. The model advocated in the literature as being theoretically correct is seldom chosen in practice. This article will show that the results obtained under these models can be significantly different. The discussion proposes a modification of the most commonly used method, which will result in valuations more in agreement with the model advocated in the literature. Current literature stresses the importance of the discounted cash flow (DCF) model (Carland and White 1980, Lloyd and Hand 1982, and Burns and Walker 1991). In practice, the earnings capitalization (EC) model and other historical earnings based models are most common. Carland and White (1980, 43) suggest that "the discounted cash flow method ... is infrequently used, as it superficially appears to be a difficult procedure to perform," referring to the complexity of the calculations. Pratt also notes the infrequency of the use of the DCF method, but suggests that the problem is not just complexity of calculations, but rather the speculative nature of the projections necessary to employ DCF (1986). The use of historical earnings in the EC model does not really eliminate the uncertainty of the future, it simply uses prior earnings as a predictor of future performance. Proponents of the EC method argue that where historical earnings are good predictors of future earnings, the two methods are approximately the same. Pratt (1989, 43) states that "in general ... approaches using historical data, if properly carried out, should yield a result that is reasonably reconcilable with what a well-implemented discounted future returns approach would derive." This article challenges the validity of the contention that the two methods are approximately the same, even when historical earnings are good predictors of future earnings. In addition, a refinement of the EC model that will improve its theoretical validity without requiring any additional projections is proposed. The refinement introduced in this article should yield better estimates of firm value. There is general agreement that there is substantial room for improvement (LeClair 1990 and Boatsman and Baskin 1981). The next section compares the economic flows that are assumed to give rise to value under the DCF model and the EC model. The third section focuses on the role of depreciation and capital expenditure under the two methods in a non-inflationary setting, while the fourth section explores the effects of inflation. Finally, the fifth section summarizes the findings. COMPARISON OF AMOUNTS CAPITALIZED UNDER DCF AND EC Lloyd and Hand (1982) state that the appropriate amount to be capitalized under the DCF model is: (1) CF = E + D - CAP + dWC + dLTD, where CF is the annual cash flow measure, E is the annual net income,(1) D is the annual depreciation charge, CAP is the gross annual capital expenditures, dWC is the annual change in working capital, and dLTD is the annual change in long-term debt. It should be noted that the amounts required for calculating CF for any prior period are quite easy to obtain from financial records. However, the DCF model deals with future cash flows, which are not available. Consequently they must be forecasted in some way. The amount that is capitalized under the EC model is just E. This assumes that E (the historical earning) is a good predictor of future earnings. Let us examine the differences between these two measures when E is a reasonable predictor of future earnings. If E is not a reasonable predictor, then EC is inappropriate (Pratt 1989). The DCF model allows each period's cash flow to differ from other periods. EC implicitly assumes that earnings are the same in each future period. …