Document Sample

CHAPTER SIXTEEN Creating Accounting Value and Economic Value Stephen H. Penman The web page for this chapter runs under the following headings: What this Chapter is Doing Metrics that Indicate the Hidden Reserves Created by Conservative Accounting and the Release of Hidden Reserves A Spreadsheet Program for Analyzing the Effects of Accounting Methods Accounting Issues in Forecasting: Starbucks Corporation Equivalences: Residual earnings and Discounted Cash Flow Approaches to Valuation Readers’ Corner What this Chapter is Doing Conservative accounting that features so much in this chapter is an accounting policy that keeps the balance sheet low, such that we always expect book value to be below market value. If LIFO inventory methods are consistently applied, then the balance sheet number for inventory will always be low (provided inventory costs are rising). Expensing R&D permanently results in (R&D) assets being omitted from the balance sheet, permanently. Conservative accounting is a feature of GAAP. Chapter 16 shows how conservative accounting affects numbers like ROCE, earnings, earnings growth, residual earnings growth. But most importantly, it shows how conservative accounting is handled in valuation, for conservative accounting affects accounting value added but not economic value added. It also deals with liberal accounting, the opposite of conservative accounting, although this is not common, except in the relative sense of a particular accounting being less conservative than otherwise. Boxes 16.2 and 16.3 provide good summaries. Focus on the valuations under conservative accounting. The chapter shows how the accounting does not affect the valuation, provided that steady state is forecasted. But conservative accounting does affect P/B ratios and P/E ratios, as the chapter demonstrates. Table 16.6 provides a summary. Metrics that Indicate the Hidden Reserves Created by Conservative Accounting and the Release of Hidden Reserves Table 16.7 shows how conservative accounting creates hidden reserves and how these hidden reserves can be liquidated, inflating earnings, if investment slows. This is not a concern for valuation if slowing of investment is forecasted within the forecast horizon (as the example there demonstrates). But one must be aware of earnings that are temporarily inflated by a liquidation of hidden reserves. A paper by Penman and Zhang, “ Accounting Conservatism, the Quality of Earnings, and Stock Returns,” published in The Accounting Review in April, 2002 develops metrics – C scores -- that estimate the amount of hidden reserves developed by the accounting for inventories, advertising and promotion, and R&D. It also develops diagnostics to estimate the amount of hidden reserves released into earnings in any period by the slowing of investment. The paper can be downloaded at: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=201048 A Spreadsheet Program for Analyzing the Effects of Accounting Methods The following programs show how changes in accounting methods for book values change profitability, growth, residual earnings, and P/B and P/E ratios. Make sure you understand Tables 16.1 – 16.5 in the text before proceeding. Here are definitions of the variables: Investment expense rate () is the percentage of investment that is expensed rather than booked to the balance sheet. Investment turnover (t1 and t2) is the amount of sales (as a percentage of investment) that the investment produces. In the program, investment produces sales for two years (as in the tables in the text), so investment turnover for two years has to be specified. As the only expense is the depreciation on the investment, this turnover is the element that determines the “real” profitability. The examples in the text deal only with investments that do not add value. To observe effects for value-added investments, increase these investment turnovers. We denote investment turnover as t1 and t2 for year 1 and year 2, respectively. Depreciation rate (d) is the depreciation rate for year 1. Depreciation expense is equal to depreciation rate times the beginning balance of net operating asset, on top of direct investment expense. Investment growth rate (g) is the annual growth rate for investments. With the assumptions described in the textbook, we can show that (1 g )(t1 1 g ) t2 RNOA t+1 g . (1 g 1 d )(1 ) Given that, you can easily see how return of net operating assets is affected by accounting methods. For instance, RNOA can be induced by higher investment turnover, but reduced by lower investment expense rate and depreciation rate. Most importantly, since RNOAt 1 g Enterprise P/B , rg it is now clear that enterprise P/B ratio is a function of accounting methods too. To see how this program works, click on a panel to get into Excel and change the assumptions in the green area, the spreadsheet will calculate enterprise P/B ratio automatically. The different cases here have the same ingredients, but different targets of the analysis. 1: To show that RNOA and P/B is a investment, we can find the rate Case 2:To show that given any return on funtion of Investment turnover, of investment expense (the amount of conservatism) that rate. investment expense ratio, and depreciation produces target RNOA and P/B ratio. Required rate of return 10% <-change your assumption here 10% expense (Year Investment turnover rate 1) 10% <-change your assumption here 60% 1) Investment turnover (Year 2) 60% <-change your assumption here 55% Investment turnover (Year Depreciation rate (Year 1) 2) 55% <-change your assumption here 50% Depreciation rate (Year Investment growth rate 1) 50% <-change your assumption here 5% Investment growth rate Target RNOA 5% <-change your assumption here 10.6% Target P/B 1.11 1.11 Target RNOA Investment expense rate 10.6% 10.00% 2004 2004 2005 2005 2006 2006 2007 2007 2008 2008 Sales Sales From investments in 2004 From investments in 2004 240.0 240.0 220.0 220.0 From investments in 2005 From investments in 2005 252.0 252.0 231.0 231.0 From investments in 2006 From investments in 2006 264.6 264.6 242.6 242.6 From investments in 2007 From investments in 2007 277.8 277.8 240.0 240.0 472.0 472.0 495.6 495.6 520.4 520.4 Operating expenses (depreciation) Operating expenses (depreciation) From investments in 2004 From investments in 2004 40.0 40.0 180.0 180.0 180.0 180.0 From investments in 2005 From investments in 2005 42.0 42.0 189.0 189.0 189.0 189.0 From investments in 2006 From investments in 2006 44.1 44.1 198.5 198.5 198.5 198.5 From investments in 2007 From investments in 2007 46.3 46.3 208.4 208.4 From investments in 2008 From investments in 2008 48.6 48.6 40.0 40.0 222.0 222.0 413.1 413.1 433.8 433.8 455.4 455.4 Operating income Operating income (40.0) (40.0) 18.0 18.0 58.9 58.9 61.8 61.8 64.9 64.9 Net Operating Asset (NOA) Net Operating Asset (NOA) From investments in 2004 From investments in 2004 360.0 360.0 180.0 180.0 From investments in 2005 From investments in 2005 378.0 378.0 189.0 189.0 From investments in 2006 From investments in 2006 396.9 396.9 198.5 198.5 From investments in 2007 From investments in 2007 416.7 416.7 208.4 208.4 From investments in 2008 From investments in 2008 437.6 437.6 360.0 360.0 558.0 558.0 585.9 585.9 615.2 615.2 646.0 646.0 Investment Investment 400.0 400.0 420.0 420.0 441.0 441.0 463.1 463.1 486.2 486.2 Free cash flow Free cash flow (400.0) (400.0) (180.0) (180.0) 31.0 31.0 32.6 32.6 34.2 34.2 RNOA (%) RNOA (%) 5.0 5.0 10.6 10.6 10.6 10.6 10.6 10.6 Profit margin (%) Profit margin (%) 7.5 7.5 12.5 12.5 12.5 12.5 12.5 12.5 Asset turnover Asset turnover 0.7 0.7 0.8 0.8 0.8 0.8 0.80.8 Growth in NOA (%) Growth in NOA (%) 55.0 55.0 5.0 5.0 5.0 5.0 5.05.0 ReOI ReOI (18.0) (18.0) 3.1 3.1 3.3 3.3 3.43.4 Growth in ReOI (%) Growth in ReOI (%) N/A N/A N/A N/A 5.0 5.0 5.05.0 Growth in cum-dividend OI (%) Growth in cum-dividend OI (%) 127.2 127.2 10.3 10.3 10.3 10.3 AOIG (0.10) AOIG (0.10) 21.1 21.1 0.2 0.2 0.20.2 Growth in AOIG (%) Growth in AOIG (%) N/A N/A -99.3 -99.3 5.05.0 Value of firm Value of firm 400.0 400.0 620.0 620.0 651.0 651.0 683.6 683.6 717.7 717.7 Premium over book value Premium over book value 62.0 62.0 65.1 65.1 68.4 68.3 71.8 71.8 P/B P/B 1.11 1.11 1.11 1.11 1.11 1.11 1.11 1.11 1.11 1.11 Trailing P/E Trailing P/E 24.4 24.4 11.6 11.6 11.6 11.6 11.6 11.6 Forward P/E Forward P/E 22.2 22.2 10.5 10.5 10.5 10.5 10.5 10.5 10.5 10.5 Equity Return Equity Return 10% 10% 10% 10% 10% 10% 10%10% Case 3:To show that given any investment expense rate, we can find the ratio of investment turnover (year 1) that yields the target RNOA and P/B ratio. Required rate of return 10% <-change your assumption here Investment expense rate 10% <-change your assumption here Investment turnover (Year 2) 55% <-change your assumption here Depreciation rate (Year 1) 50% <-change your assumption here Investment growth rate 5% <-change your assumption here Target RNOA 10.6% <-change your assumption here Target P/B 1.11 Investment turnover (Year 1) 60.0% 2004 2005 2006 2007 2008 Sales From investments in 2004 240.0 220.0 From investments in 2005 252.0 231.0 From investments in 2006 264.6 242.6 From investments in 2007 277.8 240.0 472.0 495.6 520.4 Operating expenses (depreciation) From investments in 2004 40.0 180.0 180.0 From investments in 2005 42.0 189.0 189.0 From investments in 2006 44.1 198.5 198.5 From investments in 2007 46.3 208.4 From investments in 2008 48.6 40.0 222.0 413.1 433.8 455.4 Operating income (40.0) 18.0 58.9 61.8 64.9 Net Operating Asset (NOA) From investments in 2004 360.0 180.0 From investments in 2005 378.0 189.0 From investments in 2006 396.9 198.5 From investments in 2007 416.7 208.4 From investments in 2008 437.6 360.0 558.0 585.9 615.2 646.0 Investment 400.0 420.0 441.0 463.1 486.2 Free cash flow (400.0) (180.0) 31.0 32.5 34.2 RNOA (%) 5.0 10.6 10.6 10.6 Profit margin (%) 7.5 12.5 12.5 12.5 Asset turnover 0.7 0.8 0.8 0.8 Growth in NOA (%) 55.0 5.0 5.0 5.0 ReOI (18.0) 3.1 3.3 3.4 Growth in ReOI (%) N/A N/A 5.0 5.0 Growth in cum-dividend OI (%) 127.2 10.3 10.3 AOIG (0.10) 21.1 0.2 0.2 Growth in AOIG (%) N/A -99.3 5.0 Value of firm 399.9 619.9 650.9 683.5 717.7 Premium over book value 61.9 65.0 68.3 71.7 P/B 1.11 1.11 1.11 1.11 1.11 Trailing P/E 24.4 11.6 11.6 11.6 Forward P/E 22.2 10.5 10.5 10.5 10.5 Equity Return 10% 10% 10% 10% Case 4: To show that we can find the depreciation rate that produces target RNOA and P/B ratio. Notice that future RNOA increases as depreciation rate accelerates. Required rate of return 10% <-change your assumption here Investment expense rate 10% <-change your assumption here Investment turnover (Year 1) 60% <-change your assumption here Investment turnover (Year 2) 55% <-change your assumption here Investment growth rate 5% <-change your assumption here Target RNOA 10.6% <-change your assumption here Target P/B 1.11 Depreciation rate (Year 1) 50.0% 2004 2005 2006 2007 2008 Sales From investments in 2004 240.0 220.0 From investments in 2005 252.0 231.0 From investments in 2006 264.6 242.6 From investments in 2007 277.8 240.0 472.0 495.6 520.4 Operating expenses (depreciation) From investments in 2004 40.0 179.9 180.1 From investments in 2005 42.0 188.9 189.1 From investments in 2006 44.1 198.4 198.5 From investments in 2007 46.3 208.3 From investments in 2008 48.6 40.0 221.9 413.1 433.8 455.4 Operating income (40.0) 18.1 58.9 61.8 64.9 Net Operating Asset (NOA) From investments in 2004 360.0 180.1 From investments in 2005 378.0 189.1 From investments in 2006 396.9 198.5 From investments in 2007 416.7 208.4 From investments in 2008 437.6 360.0 558.1 586.0 615.3 646.0 Investment 400.0 420.0 441.0 463.1 486.2 Free cash flow (400.0) (180.0) 31.0 32.6 34.2 RNOA (%) 5.0 10.6 10.6 10.6 Profit margin (%) 7.5 12.5 12.5 12.5 Asset turnover 0.7 0.8 0.8 0.8 Growth in NOA (%) 55.0 5.0 5.0 5.0 ReOI (17.9) 3.1 3.3 3.4 Growth in ReOI (%) N/A N/A 5.0 5.0 Growth in cum-dividend OI (%) 126.5 10.3 10.3 AOIG (0.10) 21.0 0.2 0.2 Growth in AOIG (%) N/A -99.3 5.0 Value of firm 400.0 620.0 651.0 683.6 717.7 Premium over book value 61.9 65.0 68.3 71.7 P/B 1.11 1.11 1.11 1.11 1.11 Trailing P/E 24.4 11.6 11.6 11.6 Forward P/E 22.2 10.5 10.5 10.5 10.5 Equity Return 10% 10% 10% 10% Case 5: To show that we can find the investment growth rate that produces target RNOA and P/B ratio. Required rate of return 10% <-change your assumption here Investment expense rate 10% <-change your assumption here Investment turnover (Year 1) 60% <-change your assumption here Investment turnover (Year 2) 55% <-change your assumption here Depreciation rate (Year 1) 50% <-change your assumption here Target RNOA 10.6% <-change your assumption here Target P/B 1.11 Investment growth rate 5% 2004 2005 2006 2007 2008 Sales From investments in 2004 240.0 220.0 From investments in 2005 252.0 231.0 From investments in 2006 264.6 242.6 From investments in 2007 277.8 240.0 472.0 495.6 520.4 Operating expenses (depreciation) From investments in 2004 40.0 180.0 180.0 From investments in 2005 42.0 189.0 189.0 From investments in 2006 44.1 198.5 198.5 From investments in 2007 46.3 208.4 From investments in 2008 48.6 40.0 222.0 413.1 433.8 455.4 Operating income (40.0) 18.0 58.9 61.8 64.9 Net Operating Asset (NOA) From investments in 2004 360.0 180.0 From investments in 2005 378.0 189.0 From investments in 2006 396.9 198.5 From investments in 2007 416.8 208.4 From investments in 2008 437.6 360.0 558.0 585.9 615.2 646.0 Investment 400.0 420.0 441.0 463.1 486.2 Free cash flow (400.0) (180.0) 31.0 32.5 34.2 RNOA (%) 5.0 10.6 10.6 10.6 Profit margin (%) 7.5 12.5 12.5 12.5 Asset turnover 0.7 0.8 0.8 0.8 Growth in NOA (%) 55.0 5.0 5.0 5.0 ReOI (18.0) 3.1 3.3 3.4 Growth in ReOI (%) N/A N/A 5.0 5.0 Growth in cum-dividend OI (%) 127.2 10.3 10.3 AOIG (0.10) 21.1 0.2 0.2 Growth in AOIG (%) N/A -99.3 5.0 Value of firm 400.0 620.0 651.0 683.6 717.7 Premium over book value 62.0 65.1 68.4 71.8 P/B 1.11 1.11 1.11 1.11 1.11 Trailing P/E 24.4 11.6 11.6 11.6 Forward P/E 22.2 10.5 10.5 10.5 10.5 Equity Return 10% 10% 10% 10% Case 6:To show that given any required rate of return, we can find the combination of investment expense rate and investment growth rate that yields target RNOA and P/B. Required rate of return 8% <-change your assumption here Investment turnover (Year 1) 60% <-change your assumption here Investment turnover (Year 2) 55% <-change your assumption here Depreciation Rate 50% <-change your assumption here Target RNOA 10.6% <-change your assumption here Target P/B 1.11 <-change your assumption here Investment expense rate 2% Investment growth rate -15% 2004 2005 2006 2007 2008 Sales From investments in 2004 240.0 220.0 From investments in 2005 203.4 186.5 From investments in 2006 172.5 158.1 From investments in 2007 146.2 240.0 423.4 358.9 304.3 Operating expenses (depreciation) From investments in 2004 8.6 195.7 195.7 From investments in 2005 7.3 165.9 165.9 From investments in 2006 6.2 140.6 140.6 From investments in 2007 5.2 119.2 From investments in 2008 4.4 8.6 203.0 367.8 311.7 264.3 Operating income (8.6) 37.0 55.7 47.2 40.0 Net Operating Asset (NOA) From investments in 2004 391.4 195.7 From investments in 2005 331.8 165.9 From investments in 2006 281.2 140.6 From investments in 2007 238.4 119.2 From investments in 2008 202.1 391.4 527.5 447.1 379.0 321.3 Investment 400.0 339.1 287.4 243.6 206.5 Free cash flow (400.0) (99.1) 136.0 115.3 97.7 RNOA (%) 9.5 10.6 10.6 10.6 Profit margin (%) 15.4 13.1 13.1 13.1 Asset turnover 0.6 0.8 0.8 0.8 Growth in NOA (%) 34.7682 -15.2 -15.2 -15.2 ReOI 5.7 13.5 11.4 9.7 Growth in ReOI (%) N/A 136.7 -15.2 -15.2 Growth in cum-dividend OI (%) 29.0 4.3 4.3 AOIG (0.10) 7.8 -2.1 -1.7 Growth in AOIG (%) N/A -126.4 N/A Value of firm 450.4 585.5 496.3 420.7 356.6 Premium over book value 58.0 49.2 41.7 35.3 P/B 1.15 1.11 1.11 1.11 1.11 Trailing P/E 13.1 11.4 11.4 11.4 Forward P/E 12.2 10.5 10.5 10.5 10.5 Equity Return 8% 8% 8% 8% Accounting Issues in Forecasting: Starbucks Corporation Chapter 14 makes the point that a valuation has integrity if the relevant information is reflected in forecasts within the forecast horizon, such that periods beyond the horizon can be summarized in a steady-state continuing value calculation. Accounting methods determine how quickly information is reflected in the accounting system and thus how quickly information will show up in forecasted earnings. Accordingly, accounting methods determine the length of the forecast horizon at which a steady-state continuing value can be calculated. For example, if a firm is expected to invest heavily in R&D – which is expensed directly to income under GAAP – losses might be expected for a number of years, so the forecast horizon will have to be long to capture the returns to R&D. The following material appeared in the first addition of the text. The Issue Restated If the future earnings that are forecasted are of low quality, valuations also will be low quality. Forecasted earnings and residual earnings are of good quality if they capture the value generated by a business. As earnings are determined by accounting rules, forecast quality is a matter of accounting principle. If the analyst forecasts earnings but the rules to measure earnings miss some aspect of value, the earnings forecast will be of questionable quality. As analysts typically forecast GAAP earnings, we have a particular interest in how well GAAP captures value. Chapter 16 draws the conclusion that the accounting methods a firm uses should not affect its valuation. Nor should they affect the value calculated from forecasting residual earnings. But there was an important proviso: residual earnings must be forecasted to a point in the future where the firm is expected to be in “steady state.” That is, one must forecast to a forecast horizon where long-run growth in residual earnings can be ascertained so that an appropriate continuing value can be calculated. If one’s forecast horizon is shorter than that for the steady- state forecast, value is not fully captured. So, for example, forecasting residual earnings for just one year ahead often does not yield a high-quality valuation. A continuing value is merely a device for summarizing residual earnings beyond a forecast horizon. So value can be captured by forecasting for longer and longer periods into the future. But the analyst prefers to work with short forecasting horizons for, as she forecasts further out into the future, she becomes less certain about her forecasts and less certain about her valuation. Accounting plays an important role in determining the length of the forecast horizon that captures value, so we can think of the quality of the forecasted accounting in terms of how long a forecast horizon is typically required to capture value. If the accounting delivers a perfect balance sheet (where value equals book value), no forecasting is required at all, so the accounting can be seen as high quality. Thus, as financial assets and liabilities are typically close to market value on the balance sheet (because of use of the effective interest method and mark- to-market accounting), no forecasting is necessary for financing items. So the accounting for financial items is deemed to be good accounting. But not so for operating aspects of the business. Residual earnings must be forecasted and the quality issue is how far in the future must those earnings be forecasted. Or, put another way, for a given forecast horizon of, say, five years, how well does the accounting capture value? GAAP ultimately recognizes value in the income statement (in most cases) if the forecast period is long enough, provided that the earnings are comprehensive. But you can think of situations where the accounting is remiss for shorter forecast periods. The appreciated value of land is not recognized in GAAP earnings until sold, so forecasting earnings over five years before the land is expected to be sold will not capture the earnings from the land. (Marking the land to market on the current balance sheet would solve the problem.) R&D expenditures are expensed under GAAP so one could forecast losses over five years where considerable R&D expenditures are expected, but the revenues from the R&D are not expected until later. The valuation from a five- year forecast would be a poor valuation indeed. Focusing on GAAP, the question is: if we base our valuation on an analyst's forecast of GAAP earnings for the next five years (say), will we get a good-quality valuation? Well, the analyst might be a poor forecaster, but that's another issue. But if we accept the analyst as a good forecaster, would we accept that value is captured by earnings forecasted over five years? The authorities who prescribe GAAP accounting trade off relevance for reliability: they eschew subjective estimates (that may be relevant for valuation) in favor of "hard" (reliable) numbers based on market transactions. In the U.S., GAAP typically does not allow appreciations in the value of land (or other operational assets) to be recognized unless the land is sold, because of uncertainty about whether the value will ultimately be realized. GAAP requires R&D investments to be expensed because of the uncertainty about whether payoffs to the R&D will be realized. Indeed the driving principle for recognizing value under GAAP is the realization principle: value is not added until assets or products are sold. These rules arise from a concern with quality. The restrictions on estimates reduce the opportunities for manipulation, improving quality in one sense, but mean that estimated value may not be recognized within a reasonable time frame, reducing quality. There is a tension in financial reporting that requires a balancing of the need to capture value against the need to maintain credibility. You will see in the next chapter how the estimates that are allowed do indeed introduce quality concerns. We now see how the estimates not allowed also give concerns. To the extent that accountants don't make the estimates, analysts must. As we recognize from Chapter 16, forecasted residual earnings can be constructed to be higher or lower through conservative or liberal accounting, but this in itself is not a concern if steady state is forecasted. What is of concern is whether steady state can be forecasted within a (reasonable) forecast period. Continuing values capture the steady state, so this issue is the same as asking whether residual earnings forecasted at the forecast horizon is, with an anticipation of growth, a good indicator of long-run residual earnings so that a continuing value calculation can be made. Residual operating income (and continuing values) are driven by return on net operating assets (RNOA) and growth in net operating assets (NOA). In the first three subsections here we examine how the accounting affects the quality of forecasted RNOA. In the last subsection we show how growth forecasts also affect the quality of residual income forecasts. The figure below guides you through the analysis. Use this figure as a summary. Forecast Quality Issues The Effect of Growth Forecasts on the Quality Quality of RNOA Forecasts of Forecasted Residual Operating Income Sales Forecast Quality: Revenue Recognition Sales growth forecasts Assets held for sale Equity investments Hidden assets Long-term contracts Steady-state Diagnostic: Sales/NOA Profit Margin Quality: Expense Matching Research and development Advertising and promotion Start-up costs Strategic losses Amortization Compensation expense Steady-state Diagnostic: Expense/ Sales We illustrate the quality problem with Starbucks Corporation. In 1998 Starbucks had the largest chain of coffee shops in the U.S., many stores in Asia, and was beginning an expansion into Europe. It was a favored stock, selling at nearly 6.6 times book value and a P/E of 52. The following is based on its financial reports from 1993 to 1997. Starbucks Corporation. Summary of residual operating income drivers, 1993 – 97. 1993 1994 1995 1996 1997 Sales ($ thousands) 176,541 284,923 465,213 696,481 966,946 Operating income ($ thousands) 7,253 15,051 24,406 31,081 53,252 As a % of sales: Gross margin 54.4 55.3 54.6 51.8 55.3 Store operating expenses 32.1 32.7 32.0 30.3 32.0 Other operating expenses 3.0 3.1 3.0 2.8 2.9 Depreciation and amortization 3.8 4.4 4.8 5.2 5.4 General and administrative 8.4 7.0 6.2 5.3 5.9 expenses Taxes on core income 2.8 2.9 3.4 3.7 3.5 Core profit margin (%) 4.4 5.3 5.2 4.5 5.6 Asset turnover 1.89 2.00 1.74 1.84 1.95 Core RNOA (%) 8.3 10.6 9.0 8.3 10.9 Net operating assets ($ thousands) 93,589 191,416 342,648 412,958 578,237 Growth in NOA (%) --- 104.5 80.5 20.6 40.0 Growth in sales (%) 61.4 63.3 49.7 38.8 ______________________________________________________________________________ Let's roll back the clock to the end of 1992 and suppose that the actual numbers here are as forecasts for 1993 to 1997. The CAPM operating cost of capital for Starbucks at the time (with a beta about 0.9) was about 11%. We could calculate forecasted residual operating income (ReOI) up to 1997 and our problem would be to calculate a continuing value at the end of 1997. Starbucks had few unusual items so core RNOA is close to RNOA. Taking our forecasts of RNOA in 1997 as indicative of long-run RNOA (after 1997), we might set the continuing value equal to zero: the 1997 RNOA of 10.9% is roughly equal to the calculated cost of capital. We would reinforce this calculation by noting that the RNOA from 1993 to 1997 are no more than the cost of capital, core profit margins are no more than the 5.6% in 1997, and asset turnover (ATO) is fairly constant. This seems like a business that earns regularly at or below its cost of capital. We would forecast growth in NOA from opening more and more stores, but this would not change our continuing value calculation: growth in NOA grows residual operating income only if RNOA is greater than the cost of capital. (Of course, we would treat the estimate of the cost of capital with some reservation.) A continuing value measures the premium at the forecast horizon. Our forecasts would set this to zero in 1997. But the market's levered P/B ratio in early 1998 was 6.6 and the unlevered P/B was 6.4. And the levered P/E was 52 with an unlevered P/E of 62. Clearly the market priced Starbucks as if it saw a higher RNOA after 1997 and saw 1997 and prior earnings as low quality indicators of subsequent earnings. Is the accounting missing something? Is the accounting not revealing Starbucks' potential profitability? Or has the market misjudged the potential profitability of the firm? One might be impressed by the sales growth, the earnings growth, and the net operating asset growth. And one might like the coffee. Starbuck’s was a glamour stock at the time. But it seems from the reported margins, turnovers, and RNOA that Starbucks is not able to generate much value over its required return. We are even more concerned when we realize that, because Starbucks expenses advertising, the accounting is conservative: conservative accounting should induce a higher RNOA. But let's check the accounting further. We have couched the Starbucks' quality question as a question of forecast quality and the quality of continuing values. But, rather than rolling back the clock to 1992, place yourself at the beginning of 1998 and think about forecasting future RNOA on the basis of the 1993-97 numbers. Then the GAAP quality question is one of the quality of current and past earnings. Below we discuss a number of aspects of GAAP that raise quality questions and address them to Starbucks' accounting. Recognition of Revenues: the Realization Principle Defers Value Recognition Revenue is value coming into the business, value generated by taking goods and services to the market. With its insistence on reliable evidence, GAAP accounting typically recognizes revenues only when products and services are sold. Estimating revenues that are likely to be earned in the future is not allowed. The accounting authorities feel (justifiably) that, if firms were allowed to book forecasted revenues, the license would be used for abuse. Booking revenues at sale is called the realization principle. Forecasted earnings for the next five years, say, are based on revenues that will be realized within five years. If these revenues (perhaps extrapolated with a growth rate) are a good indication of the ability to earn revenues after the five years, the forecast is of good quality. If not, the value analyst must adjust the forecast or extend the forecast horizon to capture "long-run" sales. Below are the issues to be considered. The quality of forecasted sales growth rates. The continuing value at the end of a forecast period requires a forecasted growth rate for residual operating income. One element is forecasted sales growth. Is the forecasted growth rate for the next five years a good indicator of the permanent growth rate? Starbucks' sales growth from 1993 to 1997 was quite phenomenal, accompanied, with reasonably constant asset turnover, with large growth in net operating assets from new store openings. Could this be continued? In 1998 Starbucks began its foray into Europe with the acquisition of the Seattle Coffee Company, a UK firm. But one might doubt its ability to grow sales at the rates for 1993-97. Considerations are the total size of the market competition that might reduce price and volume the ability to develop other product lines Assets held for sale. Firms sometimes have valuable assets whose earnings are not reported in the income statement until sold. So forecasts of earnings may not reflect their value. Land and property held for speculation are an example. These assets gain value if their market prices increase, but they are held at cost on the balance sheet, subject to impairment, with no gain reported in the income statement. To value them, add the present value of their anticipated gain or loss over the forecast period to the valuation. Or, if their current price is considered an "efficient" one, recognize their current book value at market value rather than cost. Starbucks leases most of its stores so does not have much property for resale. It does own manufacturing facilities. Are these on land that can be sold and has appreciated? Equity investments. If a holding in another firm is greater than 20%, earnings from the investment will appear in future statements under the equity method or through consolidation of accounts. But, if the investment is less than 20%, there could be quality problems. In many countries these investments are held at cost, and in the U.S. they are held at cost if classified as not available-for-sale ("held-to-maturity"). If so, only dividends are recognized in the income statement and dividends, as we have seen, are a poor indicator of value. A solution is to recognize their market value as their value, like the land. Since 1994 U.S. GAAP has required all securities "available-for-sale" or in trading portfolios to be recorded on the balance sheet at market value. This solves the problem if the market price is a sound valuation, an "efficient" price. If investments are carried at cost in a "held-to-maturity" category, and if market prices are available, add the market value to the valuation of other net operating assets. In both cases exclude the book value of the equities and forecasts of dividends from the ReOI valuation of the other assets, then add the market value of the equities. But be careful: using market prices to calculate values is dangerous if one is trying to establish values independent of prices. If you doubt the efficiency of market prices of equities (and why else would you be doing a valuation!), analyze the financial statements of the investee firms to uncover the value in the earnings they generate. This of course has to be done if market prices are unavailable. Starbucks' investments in unconsolidated joint ventures with Dryers Grand Ice Cream, Pepsi-Cola Company and SAZABY (in Japan) are accounted for under the equity method, so this issue does not arise. Hidden assets that can generate additional revenue streams. A firm might have assets or potential assets that can produce revenues in the long run over those forecasted. It might have customer lists that generate revenues for its products but which also can be sold to other firms. It might be positioned to develop products that tie into its existing products. It might have distribution channels which can be used to market other products or which can be sold to other vendors. And it might have contacts -- alliances with other firms or political clout -- that provide opportunities to generate value. The value of these opportunities are difficult to quantify. The opportunities can be seen as options, often called real options to distinguish them from financial options like options on stocks. As part of its investment strategy, a firm might place itself in a position where it could benefit from technological change, shifts in consumer demand, or political developments. Should these occur, it capitalizes on the opportunity. By committing to a strategy, it has effectively bought itself an option to be exercised should events be favorable. Techniques are available to value these real options in a similar way to financial options. These techniques are complicated and are fraught with measurement problems1. Ultimately value from the opportunities must show up in financial statements, they must produce earnings, so the potential revenue streams can, in principle, be forecasted (with considerable uncertainty) and valued using the techniques in this book: what sales are likely to be delivered, at what margins and with what investment in net operating assets? Indeed, as part of their strategy analysis, firms should be identifying these opportunities and estimating their value with the techniques. They should be tracking the value and protecting and enhancing it. But being unarticulated, these strategies and the opportunities they present, are not easily modeled with pro forma analysis. Brand names can produce additional revenue streams. The Starbucks name is well known. Could it be used to market other drinks? A line of coffee makers? Specialty clothes? A chain of restaurants? Is Starbucks the next McDonalds that generates value from franchising under its name? What turnovers and profit margins might these opportunities yield? Long-term contracting. Firms engaging in long-term construction (of ships and planes, for example) sometimes do not recognize revenue until a project is finished. The accounting is called the completed contract method (for recognizing revenue). If the project takes years, revenue recognition is deferred for a considerable time. Firms may, under strict conditions, recognize revenue gradually over the life of a project using the percentage-of-completion accounting method. If so, the deferral of revenue is less of an issue. For long-term contractors, discover whether the firm is using the completed contract method or the percentage-of-completion method. If the former, watch for revenues not recognized. If the latter, watch for manipulation, for the percentage of a completion method requires estimates of total revenues and costs for the project. 1 See the special issue on real options in, The Quarterly Review of Economics and Finance, 1998. The quality of forecasted revenue is a matter of assessing whether there are additional sales not anticipated by the forecast. But it is important to appreciate that additional sales do not necessarily mean higher RNOA. Remember, RNOA = PM x ATO so, if more sales require more net operating assets such that ATO is the same, and if PM remains the same, then RNOA will be unaffected. Thus Starbucks' RNOA for 1993-97 might be a good forecast of subsequent RNOA and thus good quality, even if sales are not a good indicator of subsequent sales. Starbucks may open numerous stores in Europe and Asia, increasing sales and net operating assets, but unless ATO increases, RNOA will not change given it earns the same profit margins on coffee sales. Sales per store is an important driver for Starbucks, not just stores and not just sales. So it is that ATO is a useful diagnostic for assessing the quality of sales: Diagnostic: Sales/NOA Are forecasted ATO's a good indicator of long-run ATO or will the firm increase or decrease sales per dollar invested? Is there idle capacity such that Starbucks can make more sales with the same NOA? The quality of RNOA is determined by the quality of the profit margin as well as the ATO, so we now turn to the profit margin. Mismatch of Expenses with Revenues Can Obscure Core Profitability Revenue is value coming in, expense is value going out. We are concerned about net value added, the value added per dollar of sales. Profit margins measure this net value from sales. Profit margins may not be a good indicator of subsequent margins because of real factors. So, for example, Starbucks can increase its profit margins with an increase in sales if its costs -- rents in store operating expenses and depreciation -- are fixed and there is idle capacity. Sales per store is a driver of margins as well as ATO. But margins can be affected by the accounting also. The profit margin calculation ideally subtracts from revenues the expenses necessary to earn the revenues. If revenues and expenses are appropriately matched, margins are a reliable measure of the profitability of revenues. Applying expenses to revenues to reveal profitability is the matching principle (as described in Chapters 2 and 4). So cost of goods sold is just what it says, the cost of the goods actually sold to get proper matching, not the total expenditure on inventory. If total inventory costs were expensed, there would be a mismatching of revenues and expenses and gross margins would not be a good measure of profits from selling goods. So, any costs for goods not sold are "capitalized" in the balance sheet (in inventories) until they are sold (at which point they are "amortized" to cost of goods sold). But GAAP accounting doesn't always do this matching so well. Ideally all costs should be "capitalized" and amortized to expense only as the revenues for which they are incurred are booked. Typically this is so, at least in principle. Plant costs are capitalized in the balance sheet and expensed against revenue (ideally) as the plant is used up in generating the revenue. Purchased goodwill is capitalized in the balance sheet and (ideally) expensed as its value declines. And, also to accomplish matching, operating liabilities are recognized for expenses necessary to generate revenues of the current period but which will be paid later, like accrued expenses and pensions. But there are exceptions and indeed GAAP, in pursuit of reliable numbers, requires imperfect matching in some cases. Here are the main problem areas. Expensing of R&D. With the exception of certain software development costs, GAAP charges all R&D spending directly to the income statement. The accounting does not distinguish high-quality R&D from low-quality R&D so the investment is not recognized in the balance sheet. Nor are expenses matched to revenues in the income statement. If revenues that the R&D generates do not flow until after a forecast horizon, earnings forecasts will be of low quality. Indeed, some firms incur losses from investing in R&D for several years even though the R&D is very valuable. We are familiar with this problem from the simulation of ratios for pharmaceutical firms with R&D programs in Chapter 16. Once revenues flow from R&D and a steady-state is reached, R&D firms are relatively easy to value. But prior to steady state, RNOA is low and, for startups, is often negative. The diagnostic to track is: Diagnostic: R&D/ Sales If the analyst forecasts a change in this ratio -- because more or less sales are likely to flow from a dollar investment in R&D – the earnings are of low quality. If no change in forecasted, the ratio is in steady state and the earnings are of sound quality. Forecasting long-run profitability when R&D yields a low RNOA is a tricky matter. Will the R&D be successful and, if so, what will be the revenues it will generate? Again, this is an issue of knowing the business. To value the R&D of a start-up biotechnology firm, one has to be a biochemist to understand whether the R&D will pay off. And one has to know the drug market. But one still needs to do financial analysis. Knowing the technology and knowing the market are necessary but the valuation issue is ultimately one of translating this knowledge into forecasts of the long-run residual earnings that will be generated. Advertising and Promotion. Advertising and promotion costs can be incurred to deliver current sales and also future sales. But the accounting typically treats all advertising as an expense when incurred and applies it against current sales. The solution involves a marketing analysis. Will the forecasted advertising produce more sales in subsequent years? The diagnostic to track is: Diagnostic: Advertising Expense/ Sales Start-up costs. Start-up costs for product and facilities development are usually expensed even though they are needed to generate long-term revenue. Strategic losses. Firms may, as a matter of strategy, decide to go into markets where they know they will have to incur losses for a time before they are established and draw customers. Publishing and media ventures, fashion ventures, and new retailers require time to develop name recognition, establish a quality reputation or to drive out competitors. The losses forecasted for the near term may be investments in long-term operations. When Amazon.com began trading, their management insisted that the business model required continuing losses and it would be a number of years before profits were likely. Needless to say, such a firm is difficult to value. Amortization. Expensing an investment immediately is really a very rapid (immediate) amortization. A firm may capitalize an investment but, as a matter of policy, choose high or low amortization rates. This can distort the matching of expenses to revenues. Purchased intangibles are a case in point, and a particular area to watch is amortized goodwill. If a firm has made a recent acquisition using the purchase (rather than pooling) method, it can choose to amortize over one to forty years in the U.S. If the amortization is done over five years, a five-year forecast of earnings will be low quality because subsequent years will not bear the charge. Indeed, there is a question if amortization is warranted at all if the value of the investment does not decline. In the U.K., firms may amortize goodwill only if it is judged to be impaired. This notionally deals with the problem but, as impairment is a matter of judgment, it opens up the accounts to manipulation. This points to a solution: back out any amortized goodwill from a forecast and add it back to current book value. Then ask if current goodwill is a good carrying value or should be written down. That is, continually mark goodwill to market. One must, however, have considerable faith in one's ability to judge impairment. One guiding rule is to write goodwill off such that the forecasted residual income from (the share of) profits forecasted for the subsidiary is always zero. If the forecast of residual income from the investment is zero, the carrying value of the investment must be its value. This may require writing goodwill up. Alternatively one can stick with historical cost accounting, estimate the life of residual earnings from the book values purchased, and amortize over that life. With these points in mind, let's return to Starbucks. Starbucks' expense ratios for 1993- 97 look stable and predictable. So, are its profit margins a good indicator of its long-run ability to get profits from sales? Starbucks has a small amount of R&D but significant advertising costs. Footnotes reveal that store operating expenses, which are in the order of 32% of sales fairly consistently, include advertising, store opening costs and store remodeling costs, as required by GAAP. These depress profit margins but may pay off later with higher sales. The reports do not give the detail on these costs (disclosure quality!) but if the amount each year that produced sales after 1997 were just 2% of sales (and so were capitalized in the balance sheet), profit margins would be higher by 2% and RNOA (with an adjusted turnover for increased net operating assets with the capitalization of the costs) would be 14.4%, 12.3%, 11.8%, and 14.5% for 1994-97. (The profit margins and RNOA for 1994-97 would be lower than these numbers if any of the costs capitalized for future sales produced sales before 1998: the costs would have to be amortized to match the sales.) Starbucks new joint ventures generated $5.9 million in losses from 1994-97 (which were included in other operating expenses). Over the same period it contributed $34.8 million in capital to these ventures in anticipation of profit. The losses might be seen as strategic, start-up losses. One would have to analyze the ventures. Mismatching of revenues and expenses is not always a problem. Proper matching is what we called neutral accounting in Chapter 16. Conservative and liberal accounting disturb the matching. But we also saw in Chapter 16 that this is no problem if there is a steady-state relationship between revenues and expenditures. The accounting induces a change in PM, ATO and RNOA, but the change is permanent, as in the panel examples in Chapter 16. But what if, in the Chapter 16 panel examples, the R&D expenditure forecasted over the four years did not pay off until later? The forecasted RNOA for these four years will be depressed by the expensing of the R&D and will not be at the same level of subsequent RNOA that incorporates the revenues from R&D. A steady-state relationship between revenues and expenses will be deferred to a more distant future and the RNOA for the first four years will be low quality. Profit margins are constant if expense ratios are constant and profit margins are predictably growing if expense ratios are predictably decreasing. So the diagnostic for the quality of forecasted expenses is: Diagnostic: Expense/sales. The R&D/Sales and Advertising/Sales ratios earlier are examples. If expense to sales ratios forecasted for Starbucks for years after 1998 are forecasted to be the same as those for 1997, there is no quality problem. But if one forecasts more (or less) revenues per dollar outlays on advertising, start-up costs and R&D further down the line, the forecasted earnings are low quality. Put another way, one asks: Is advertising, R&D, etc. at the level to sustain sales or is more or less needed in the future per dollar of sales? Missing Accounting: Hidden Dirty Surplus Omits Value Forecasts will be low quality, and some aspect of value will be missed, if earnings forecasted are not comprehensive. We saw in Chapter 8 that GAAP does not recognize the value effects when services are paid with stock issues or shares are issued (usually in exchange for convertible securities) at less than market price. This is a significant issue with stock compensation but potentially any service could be paid for with stock. The issue is not a matter of delayed recognition. It's a matter of non-recognition. We might forecast strong GAAP earnings but will overvalue the firm if we don't reduce the forecast for the value loss anticipated in these share transactions. Remember, we have to be careful that the employees will not run away with the company. Starbucks issues stock options to its employees under an Employee Stock Option Plan. In addition it instituted an Employee Stock Purchase Plan in 1995 permitting eligible employees to apply 10% of their base earnings to purchase up to $25,000 of Starbucks' common stock at a 15% discount from market price. The panel below calculates the implicit additional wages expense from the employee stock plans. The implicit wages paid in the option plan are calculated as laid out in Chapter 8. Proceeds from share issues under the stock purchase plan are in the statement of shareholders' equity and, as these are 85% of market price, the implicit wage cost (market price-issue price) is easily calculated (there is no tax benefit reported). Starbucks Corporation. Adjustments to operating income for stock issues to employees (in thousand of dollars except share and per-share items) 1993 1994 1995 1996 1997 Employee Stock Option Plan Shares issued 1,407 774 946 1,177 1,382 Weighted-average share price 20.38 25.16 16.44 23.87 34.81 Weighted-average exercise price 8.59 9.25 3.34 6.78 9.92 Loss (before tax) 16,591 12,311 12,390 20,128 34,409 Tax benefit 5,243 3,719 4,754 6,808 9,626 Implicit wages expense (after tax) 11,348 8,592 7,636 13,320 24,783 Employee Stock Purchase Plan Proceeds of share issues 263 1,735 2,313 Implicit wages expense 46 306 408 Total implicit wages expense 11,348 8,592 7,782 13,626 25,191 Restated comprehensive operating income (4,095) 6,459 16,624 17,455 28,061 Restated RNOA (%) -4.7 4.5 6.2 4.6 5.7 ______________________________________________________________________________ Note: The tax benefit from issue of shares in the option plan is given in the statement of shareholders' equity. The effect on operating income and RNOA of these quasi-payments to employees is substantial. Reported RNOA is low quality. In 1996 Starbucks' convertible debt with a book value of $79 million was converted into 5,359,769 shares of common stock. At an average 1996 price per share of $23.87 this resulted in a loss of $48.9 million (the market price of share issued minus the book value of the bonds). This is a financing activity, so will not affect RNOA. But the conversion is a loss to stockholders that is not recognized in reported income. Forecasted losses from these share transactions must be recognized to produce a quality forecast. But there is a problem. The loss depends on the forecasted price of the share when options and conversion rights will be exercised. But that price depends on the value of the firm and to get the value of the firm we need the forecasted loss! The circularity is hard to break. But here are some suggestions for accounting for employee stock options: 1. Extrapolate from past losses. Calculate current and past losses from share issues to employees (as we have just done for Starbucks). If they are regular (as a percentage of sales), extrapolate to the future. 2. Forecast future prices from current prices. Forecast stock prices at expected exercise dates by extrapolating from the current stock market price to the future price at the cost of capital and reduce the future price for any expected dividends. This assumes that the market price is efficient and is expected to grow, cum-dividend, at the cost of capital. Deduct expected losses (forecasted price minus exercise price) from forecasted operating income. 3. A liability calculation: calculate the current market price minus exercise price for all options currently outstanding and in the money. This options overhang is the amount of loss that would be incurred if all outstanding options were exercised at today's price. Discounting for the probability the options will not be exercised and for anticipated tax benefits on exercise yields the contingent liability for outstanding options. Reduce the valuation by the amount of the contingent liability. The calculation is correct if the market is efficient and if future option grants are expected to be granted at the money in a fair game. 4. Forecast future prices from an initial calculated value. Value the equity today by using forecasts of operating income before stock compensation. Then forecast future stock prices at forecasted exercise dates by extrapolating from the calculated value at the cost of capital, adjusting for expected dividends. Calculate the future losses and reduce today's value by the present value of the forecasted losses. This method assumes the market will be efficient in the future when options are exercised, but not necessarily so currently. But because the initial calculation of today's value is too high, it also overestimates the loss. Fudge it down? 5. A normal RNOA calculation. Forecast future market prices at expected exercise dates as expected future book value plus a premium based on a normal RNOA for the industry, where the normal RNOA is calculated with normal cash-equivalent compensation expense for employees in the industry. Calculation 1 relies on the past as indicative of normal compensation expense. Calculations 2 and 3 are appropriate if the analyst is comfortable with market efficiency. But this goes against the grain of fundamental analysis. Management may grant options to themselves when their inside information indicates stock prices are low and exercise when stock prices are high: they may play the swings in prices away from fundamentals. Calculations 4 and 5 attempt to measure without reference to market prices, albeit imperfectly. Growth and Accounting Quality RNOA quality is affected by accounting principles. But it is also affected by growth in net operating assets (NOA). Cast back to the panel examples in Chapter 16 and compare Panels B and D. Both panels apply conservative accounting but in Panel D there is growth in NOA. And RNOA in Panel D is lower than in Panel B, 10.6% compared to 11.1%: growth in NOA reduces RNOA if there is conservative accounting. Both panels deal with the same investments that don't add value. Which RNOA is the quality one? Well, if growth in NOA after 2004 is forecasted at the same rate as 2002-2004, both RNOA are quality RNOA. In both cases RNOA forecasted for 2002-2004 is a good indicator of subsequent RNOA and a continuing value can be calculated with this RNOA and the relevant growth rate. But, if growth in RNOA is forecasted to change after 2004, the conservative accounting in Panel D will induce a change in RNOA. Thus the RNOA for 2002 to 2004 will not be a good indicator of subsequent RNOA. Indeed, this is the situation in table 16.7 in Chapter 16. The forecasted leveling off of NOA after 2004 increases forecasted RNOA to 11.1%, the no-growth rate. Hidden reserves are liquidated, as we saw. We would be wrong in calculating a continuing value at 2004 based on the 10.6% RNOA from 2002 to 2004. The continuing value must be based on the zero long-term growth rate rather than 5% rate prior to 2004. But it must also be based on the higher RNOA induced by a change in growth in NOA because conservative accounting is used. In the end, it's a question of the quality of forecasted residual earnings used in continuing values. Residual earnings are driven by RNOA and growth in NOA but growth in NOA will affect RNOA if the accounting is conservative or liberal. Full pro forma analysis (of the type in the panels in Chapter 16) models the evolution of the firm, it RNOA, its growth in NOA and the interaction between the two. In the case of Starbucks, we probably would calculate that the firm could not keep up the high NOA growth rate from 1993 to 1997. Would we expect a decline in the growth rate to generate a higher RNOA? Well, somewhat. Starbucks expenses store opening expenditures and advertising so the accounting is conservative. An extended pro forma after 1997 that anticipated any changes in advertising/sales and store-opening-costs/sales ratios will anticipate changes in RNOA from 1997 levels. Equivalences: Residual earnings and Discounted Cash Flow Approaches to Valuation Table 16.14 in Chapter 16 indicates that the forecasting horizon for Starbucks is likely to be longer if one uses discounted cash flow (DCF) analysis (cash accounting for the future) rather than accrual accounting, returning to the issue raised in Chapter 4: Starbuck’s free cash flows are negative. If one extended the forecast horizon for DCF analysis far enough, one would eventually forecast steady state, and an equivalent valuation to an accrual accounting valuation would result. Here we lay out the situations where DCF valuation and residual earnings valuation are equivalent and where they are not. The material here is somewhat technical but is worth mastering if you wish to be comfortable in choosing a valuation technology. As stated Chapter 4, the discounted cash flow (DCF) valuation model is C I T 1 T V0E t C t I t T 1 T F gCF / F NFOo . Case 3 (A.1a) t 1 F To be compact, we use the summation notation to indicate summed forecasts of discounted cash flows to the forecast horizon, T. This calculation will give the correct valuation if free cash flow is forecasted to grow at the rate g(CF) after T and if net financial obligations (NFO) are at market value. We indicate the anticipated growth rate as g(CF), "growth in free cash flow," to distinguish it from growth in residual operating earnings (ReOI). This calculation is called Case 3 because, like Case 3 of the ReOI valuation, it involves growth in the continuing value. Case 2 applies when there is no growth in free cash flow anticipated after the horizon and accordingly the continuing free cash flow is capitalized at the rate, F 1: C I T 1 T V0E t C t I t T 1 T F 1 / F NFOo . Case 2 (A.1b) t 1 F To compare DCF and ReOI valuation techniques we will work with PPE Inc., the example in Chapters 14 and 15 where we introduced ReOI valuation. The panel below reproduces the pro forma cash flow statement for PPE Inc. and calculates the value of the operations using DCF valuation with a four-year forecast horizon. The forecasted operating income (OI) and change in net operating assets (NOA) from which the free cash flow is calculated are taken directly from earlier panels for PPE in Chapter 15. The value of the operations, 103.68, is the same as that calculated using the ReOI model in Chapter 15 and the value of the equity, after subtracting net financial obligations, is the same value of 103.68 7.7 = 95.98. The free cash flow is forecasted to grow indefinitely after Year 4 at 5% a year and this is incorporated in the continuing value. Indeed, the free cash flow is predicted to grow at 5% from Year 1 onwards, so we can value the firm with a horizon of one year: C1 I1 6.57 V0NOA 103.63 F g(CF) 1.1134 1.05 which is the same valuation as in Chapter 15, allowing for rounding error. PPE Inc.: Pro Forma Cash Flow Statement and Case 3 Discounted Cash Flow Valuation Forecast, year t 0 1 2 3 4 5 Free cash flow OI 9.80 10.29 10.81 11.35 11.92 12.51 NOA 4.52 3.72 3.91 4.10 4.31 4.52 5.28 6.57 6.90 7.25 7.61 7.99 Financing flows Dividends 5.28 3.81 4.10 4.40 4.73 5.08 Debt 0.00 2.76 2.80 2.85 2.88 2.91 5.28 6.57 6.90 7.25 7.61 7.99 Discount rate (0.1134) 1.1134 1.2397 1.3802 1.5368 Growth in free cash flow (%) 24.4 5.0 5.0 5.0 5.0 PV of free cash flow to year 4 5.90 5.57 5.25 4.95 Total PV of free cash flow 21.67 Continuing value (g = 1.05)1 126.03 PV of continuing value 82.01 Value of operations V0NOA 103.68 1 CV = 7.99/(1.1134 1.05) = 126.03 The DCF analysis here gives the same calculation as the ReOI calculation with the same forecast horizon. So, can they be used interchangeably? If so, what was all the fuss in Chapter 4 and in this chapter where we criticized the DCF approach? Does accrual accounting versus cash accounting for the future make any difference? Why doesn't the DCF model work for the Wal- Mart case in Chapter 4 and Starbucks in this chapter? Well, the DCF model works for PPE, Inc. only because it's a special case. We'll show this shortly, but to make a comparison of the two models it is helpful to restate the ReOI model in a form that refers to cash flows. Equivalent Valuation Methods In the appendix to Chapter 5 we pointed out that residual earnings valuation can be done in different ways that yield equivalent valuations. And so can the ReOI valuation. The residual earnings model, we saw, can be stated as the dividend discount model with a continuing value appropriate for Cases 1, 2 and 3. For Case 3 the restatement is: t NI g 1 C S E T T V0T t d t T 1 / E E g E t 1 where g is the anticipated growth in residual earnings after the horizon. The terminal value supplies the expected value of the equity at the horizon, VTE . One thinks of the current value, V0E , as the present value of the terminal value, cum dividend. In the same way the ReOI model can be stated in terms of discounted dividends plus a horizon value. But the dividends are not the cash paid to shareholders. The dividends from operating activities are the cash that the operations pay off to the financing activities. That is, they are the free cash flow from operations that are invested in financial assets or reduce financial obligations. (And they can be negative if more cash is required for operations.) Restated in terms of discounted "dividends," Case 3 of the ReOI model is: O I g 1NO A T T V0E t C t I t T 1 T / F NFOo . Case 3 (A.2a) F g F t 1 where g is now the anticipated growth in ReOI after the horizon. The terminal value is the expected value of the operations at T, VTNOA . One thinks of the current value of the operations, V0NOA , as the present value of their terminal value cum-dividend. Clearly the form of the restated residual earnings and ReOI models is the same but the restated ReOI model forecasts free cash flows rather than dividends, operating income at the horizon rather than net income, and NOA at the horizon rather than CSE to reflect that we are dealing with the value of the operations, not the equity. Then, in both cases, the value of the net debt is subtracted from the value of the operations to yield the equity value. We refer to the restated ReOI model as the cash flow version of the ReOI model.2 The cash flow version of the ReOI model always gives the same valuation as the original ReOI model. We can proof the equivalency for PPE Inc. We valued this firm with a one-year horizon but, as we forecasted ReOI to grow at a constant rate after one year, we get the same value with any horizon. Let's set a four-year horizon and value PPE's operations using the original ReOI model: PPE Inc.: Valuation Using ReOI Model with Four-Year Horizon Forecast, year t 0 1 2 3 4 5 ReOI 1.85 1.95 2.05 2.15 2.25 Discount rate (1.1134t) 1.1134 1.2397 1.3802 1.5368 PV of ReOI 1.67 1.57 1.49 1.40 Total PV of ReOI 6.13 Continuing Value (g = 1.05)1 35.49 PV of continuing value 23.09 NOAo 74.42 Value of operations V0NOA 103.64 2.25 1 CV 35.49 .0634 2 If you are wondering why we call it an ReOI model (given that ReOI is not in the formula), remember it's equivalent to the ReOI model calculation. And the model can be restated as Re OI T 1 T V0E 1 C I t NOA T T / F NFOo F g F t 1 NOA The continuing value here is the forecast of VT , given by the forecasted NOA at T plus a premium based on subsequent Re OI growing at the rate, g. The valuation of 103.64, is the same as in Chapter 15 with a shorter horizon (allowing for rounding error): extending the horizon always gives the same valuation once the long-run profitability has been forecasted. Now let's value the operations using the cash flow version of the ReOI model with the four-year horizon. Using the forecasts of free cash flows and terminal operating income and net operating assets in the pro forma, the valuation is as follows: PPE Inc.: Valuation Using Cash Flow Version of ReOI Model with Four-Year Horizon Forecast, year t 0 1 2 3 4 5 Free cash flow 6.57 6.90 7.25 7.61 Discount rate (1.1134t) 1.1134 1.2397 1.3802 1.5368 PV of free cash flow 5.90 5.57 5.25 4.95 Total PV of free cash flow 21.67 Terminal OIT+1 12.51 Terminal NOAT 90.46 Terminal value (g = 1.05)1 125.98 PV of terminal value 81.97 Value of operations V0NOA 103.64 1 12.51 .05 x 90.46 TV 125.98 .0634 The valuation of the operations here is the same as the ReOI model valuation. And the valuation is also the same as that for the DCF model. Case 2 of the cash flow version of the ReOI model sets g = 1 (in A.2a) to recognize that ReOI will be constant after the horizon: OI V0E t C I t T 1 / T NFOo T F 1 F Case 2 (A.2b) t 1 F This gives the same valuation as Case 2 for the ReOI model and you see it has the same form as the restated Case 2 of the RE model in the appendix to Chapter 5. Here the horizon value is calculated simply by capitalizing the forecasted operating income for T+1. This terminal value is the same as the valuation with an SF2 forecast (see Chapter 14) but with the valuation done at the horizon rather than at time 0. Here ReOI is expected to be constant after the future horizon whereas, with an SF2 forecast, it's expected to be constant for all future years after the current one. Case 1 of the cash flow version of the ReOI model (where ReOI is expected to be zero after the horizon) is: V0E t C I t N O A T / T NFO o T F F Case 1 (A.2c) t 1 This gives the same valuation as Case 1 of the ReOI model; it is the same form as Case 1 of the RE model in the appendix to Chapter 5. Here the NOA at the horizon are expected to earn at the cost of capital so their forecasted book value is indeed the expected value of the operations at that date. So the terminal value is the same calculation as that for the SF1 forecast. The accompanying box summarizes the terminal value and the forecasts involved. Forecasts and Terminal Valuations for the Cash Flow Version of the ReOI Model Case Forecast of ReOI at Horizon Terminal Valuation Case 1 Zero (SF1) VTNOA N O A T Case 2 Constant (SF2) O I T 1 VTNOA F 1 Case 3 Growing at a constant rate OI T 1 g 1NOA T VTNOA F g Are the DCF and Residual Earnings Techniques Equivalent? Having expressed the ReOI model in terms of discounted cash flows, we can compare it with the DCF model directly. Placing their Case 3 versions beside each other, C I T 1 T DCF Model : V0NOA t C t I t T 1 T F gCF / F (A.1a) t 1 F OI g 1NOA T T Re OI Model : V0NOA t C t I t T 1 T / F (A.2a) F F g t 1 Clearly the two approaches give the same valuation for the same horizon if the two terminal values here are the same. Indeed the two approaches gave us the same valuation in the PPE case. But let's modify the forecasts for PPE Inc. a little. In discussing the features of the ReOI model in Chapter 15 we demonstrated that an anticipated investment of 50 million at the end of Year 2 would not affect the current valuation if it was forecasted to earn at the cost of capital. The pro forma with this anticipated investment is laid out below with the forecast horizon extended to seven years. You can see that the 50 million investment has been added to the net operating assets for Year 2. The ATO on this investment is predicted to be the same as that for existing investments, 1.762. But the forecasted profit margin on the sales from the investment is 6.44%, so the expected RNOA is 11.34%, the cost of capital.3 The overall RNOA and profit margins in the pro forma after Year 2 are based on total operating income, net operating assets and sales from the existing investments and the new investment and, are weighted averages of those on the existing and new investments. The drivers yield the same total ReOI and growth in ReOI as before and thus the new investment does not affect the valuation. But look now at the free cash flow forecast. Whereas the firm can be valued using the ReOI model with a forecast horizon of one year, this is not the case for the DCF valuation. Free cash flows are not growing at a constant rate. Indeed free cash flows are predicted to be negative in Year 2 (due to the cash investment) and this is rewarded with higher free cash flow later. But the free cash flow is not forecasted to grow at a constant rate after Year 2. The forecast horizon will have to be quite long for forecasted free cash flow to settle down to a steady state with constant growth. 3 To keep it simple we assume that any depreciation is replaced with further investment at the cost of capital. PPE Inc.: Operating Activities Pro Forma with Anticipated Value-Neutral Investment in Year 2 Forecast year, t 0 1 2 3 4 5 6 7 Income Statement Sales 131.15 137.70 232.67 239.91 247.49 255.48 263.85 Core operating expenses 120.86 126.89 215.65 222.32 229.31 236.67 244.38 Core operating income 10.29 10.81 17.02 17.59 18.18 18.81 19.47 Balance Sheet Net operating assets 78.15 132.05 136.16 140.46 144.99 149.73 154.72 Cash Flow Statement OI 10.29 10.81 17.02 17.59 18.18 18.81 19.47 NOA 3.72 53.90 4.10 4.31 4.52 4.75 4.99 Free cash flow (C-I) 6.57 (43.09) 12.92 13.28 13.66 14.06 14.48 ReOI Drivers RNOA (%) .1383 .1383 .1289 .1292 .1294 .1297 .1300 Profit margin (%) 7.85 7.85 7.32 7.33 7.35 7.36 7.38 Asset turnover 1.762 1.762 1.762 1.762 1.762 1.762 1.762 Growth in NOA (%) 5.00 68.97 3.11 3.16 3.22 3.28 3.33 Residual OI (0.1134) 1.85 1.95 2.05 2.15 2.25 2.37 2.49 Growth in ReOI (%) 5.0 5.0 5.0 5.0 5.0 5.0 5.0 Free Cash Flow Growth Drivers Growth in free cash flow (%) 24.4 (655.9) --- 2.80 2.87 2.92 2.98 Growth in OI (%) 5.05 5.0 57.45 3.33 3.39 3.44 3.49 Growth in NOA (%) 5.00 68.97 3.11 3.16 3.22 3.28 3.33 When Will DCF Analysis Work? We have seen DCF analysis work for the same horizon as the ReOI model for PPE Inc. What are the situations where this occurs? Well, look back at the terminal values for Case 3 of the two models that are compared in expressions (A.1a) and (A.2a). As C T 1 I T 1 OI T 1 N O A T 1 , free cash flow will grow at a constant rate if OI and NOA grow at the same rate. And this occurs when the net operating assets generate operating income at a constant rate, that is, when RNOA is expected to be constant: OI is generated from NOA so, for OI to grow at the same rate as NOA, the rate of return on NOA must be constant. But when this is so, any forecasted growth in ReOI, g, is due only to growth in NOA. The forecast of g 1 N O A T in the terminal value of the ReOI model (in A.2a) is, in this case, a forecast of NOA T 1 , and OI T 1 g 1 NOA T is equal to O I T 1 N O A T which is the same as C T 1 I T 1 . So the anticipated growth in free cash flow, g(CF), equals the anticipated growth in ReOI, g. Under these conditions the two methods are equivalent and the DCF calculation gives the correct valuation. This was the case in the original PPE pro forma. But not so for the pro forma modified for new investment. Comparing the Case 2 version of the ReOI model (in A.2b) with Case 2 of the DCF model (in A.1b), it is clear that the two models give the same valuation when C I T 1 OI T 1 , and perpetually so. That is, the valuations are the same when one forecasts no growth in NOA and constant RNOA after the horizon. The distinguishing feature of the modified pro forma was the negative free cash flow in Year 2. But the determining feature for the horizon point is the changing RNOA that the new investment induced. The changing RNOA results in operating income changing at a different rate from NOA and this yields non-constant growth in free cash flow. Indeed you can see at the bottom of the pro forma with the investment in Year 2 that, not only is RNOA changing each year, but forecasted OI and NOA are also growing at different rates. ReOI on the other hand is forecasted to grow at a constant rate. If forecasted free cash flow is positive, DCF analysis will work if we forecast RNOA to be constant so that any growth in ReOI comes from growth in NOA alone. This of course is the case of an SF3 "simple" forecast (in Chapter 14) but with the forecast applying at the horizon rather than currently. ReOI valuation will work more generally where growth comes from profitability as well as growth in NOA. If one wants to forecast cash flows, one is advised to use the terminal value in (A.2a) rather than that in (A.1a). But this of course effectively transforms the calculation to the ReOI model and, given one must forecast OI and NOA to forecast free cash flows, one might just as well use the ReOI model. Both DCF analysis and ReOI analysis will work well if forecast horizons are very long. But the further one forecasts into the future the less certain one typically is about the forecast. The ReOI model has the feature that in most cases more of the value is captured by the current net operating assets and forecasts over the first few years. Thus it is less susceptible to errors in prediction for later years, particularly in forecasting long-term growth rates. DCF analysis, on the other hand, places much more weight on the terminal value. In the extreme, if free cash flows are forecasted to be negative in the initial years, over 100% if the valuation comes from forecasts for the "long run." In some cases, future cash flows can be in the very distant future. Consider employee pension payments; these often are cash flows well into the future. Accrual accounting accounts for them in the present. DCF analysis has difficulty in capturing value from investments in subsidiaries. The cash flows from the investments are the dividends they pay, but dividends (if any) do not usually capture value. Accrual accounting captures value through the equity method. Under the right conditions DCF analysis can be used interchangeably with ReOI analysis. But one might just as well use the ReOI approach with more general applicability. And the ReOI model has the right mentality. The DCF approach is conceptually flawed because one wants to get at the source of the value generation in the operating activities, not the "dividends" from the operating activities. One wants to "account for the future" in terms of operating income and net operating assets and, having done this, one does not want to "back out the accruals" to get to the cash flows. Cash is king, yes, because investors ultimately want cash. But the valuation problem is one of forecasting what the ultimate cash will be and cash generated in the short term may not be a good indicator of this. Indeed negative free cash flows are often generated to beget the ultimate cash. Readers’ Corner The properties of conservative accounting and its effects of P/B and P/E ratios are modeled and discussed in the following papers: Feltham, J., and J. Ohlson. 1995. Valuation and clean surplus accounting for operating and financial activities.” Contemporary Accounting Research 11 (2): 689-731. Zhang, X. 2000. Conservative accounting and equity valuation. Journal of Accounting and Economics 29: 125-149. Beaver, W., and S. Ryan. 2000. Biases and lags in book value and their effects on the ability of the book-to-market ratio to predict book return on equity. Journal of Accounting Research 38 (Spring): 127-148.

DOCUMENT INFO

Shared By:

Categories:

Tags:
sheet music, printable worksheets, sheet metal, Math Worksheets, fast food, dog food, Printable Pattern, Free excel, diet food delivered, sheet sets

Stats:

views: | 457 |

posted: | 11/15/2010 |

language: | English |

pages: | 33 |

Description:
Free Excel Restaurant Expense Spreadsheet document sample

OTHER DOCS BY unm17024

How are you planning on using Docstoc?
BUSINESS
PERSONAL

By registering with docstoc.com you agree to our
privacy policy and
terms of service, and to receive content and offer notifications.

Docstoc is the premier online destination to start and grow small businesses. It hosts the best quality and widest selection of professional documents (over 20 million) and resources including expert videos, articles and productivity tools to make every small business better.

Search or Browse for any specific document or resource you need for your business. Or explore our curated resources for Starting a Business, Growing a Business or for Professional Development.

Feel free to Contact Us with any questions you might have.