Valuing The Gap using a discounted cash flow model[1]
Overview:
The DCF approach we will follow is designed to estimate the equity value of the firm by first valuing all of the assets of the firm regardless of the source of capital. Next, the value of debt claims on the firm is subtracted to arrive at the equity value of the firm. This perspective is analogous to a focus on ROA rather than ROE when studying a company's financial ratios. The following steps are employed in the approach:
1. Forecast free cash flows (FCF) which are equal to earnings before income less adjusted taxes (EBILAT) minus net investment for each year over an appropriate forecast horizon (5-15 years).
2. Calculate the value of the firm beyond the forecast horizon (continuing value) using an appropriate perpetuity model.
3. Determine the weighted average cost of capital (WACC) used to discount forecasted FCF.
4. Calculate the equity value of the firm. It is equal to the sum of the present value of the FCF over the forecast period, plus continuing value, plus excess cash and marketable securities on hand currently, less the market value of debt.
5. Check for reasonableness and consistency
- Forecast balance sheet items, ensuring they are consistent with your income statements and cash flow calculations.
- Review forecasted financial statement values and the ratios implied by these values for consistency and to verify they reflect the conclusions drawn from your business strategy analysis.
- Perform the analysis for different scenarios to determine the sensitivity of your valuation to assumptions about key profit drivers, discount rates and investment levels.
Step 1. Forecasting free cash flow. The DCF approach requires the forecasting of free cash flows defined as earnings before interest less adjusted taxes minus net investment or:
FCF = EBILAT – net investment
EBILAT: The FCF calculation requires the calculation of EBILAT. This begins with an estimate of EBIT which, in turn, typically begins with the sales forecast. A forecast of The Gap's sales based on forecasts of growth is presented below:
Year / 1991 / 1992 / 1993 / 1994 / 1995 / 1996 / 1997 / 1998 / 1999 / 2000 / 2001Sales
/ 2519
/ 3023
/ 3597
/ 4244
/ 4966
/ 5760
/ 6625
/ 7552
/ 8609
/ 9815 / 11189
While an intricate analysis of sales growth can be performed, often some basic assumptions of uniform yearly expansion in the number of stores and growth in sales per square foot per store will yield forecasts similar to those based on more detailed analyses. For example, the forecasts shown above are determined by assuming a starting growth rate of 20% in fiscal 1992 with a 1% drop-off in growth over the next 5 years. The decline to a steady 15% growth rate after 1997 is consistent with the geometric rate of growth in comparable store sales per square foot over the past 5 years. Emphasis on comparable store sales is justified because as time goes on, comparable store sales will represent a greater and greater proportion of total sales. Sales are then assumed to grow at a 3% in perpetuity, approximately the long-run rate of inflation.
Operating Expenses: Estimates of cost of good sold and SG&A are required to arrive at a forecast of EBIT. Unless significant changes in the firm's business strategy or cost structure are anticipated there will be little variation in the values of these variables as a percentage of sales over the forecast period. However, even small changes in estimated percentages can have a significant impact on projections. In this example, CGS as a percentage of sales is assumed to begin at 60% in fiscal 1992 –slightly higher than the previous year- with an increase of .5% in each year, reaching at 65.5% in the terminal year. The rationales behind this assumed growth in the CGS% include 1) prices are subject to competition and, 2) costs will rise more than proportionally as operations continue to expand.
SG&A: SG&A has been approximately 23% recently. This constant rate is assumed throughout the forecast period (see attached spreadsheet for summary calculations). Note that the net effect of these assumptions is that profit margin is projected to fall to around the 5% by the end of the forecast period, consistent with the historical performance of the specialty retail industry.
Cash Taxes on EBIT:Assuming that depreciation has been included in CGS and SG&A (but goodwill amortization has not) the revenues less expenses forecasted above result in forecasted EBIT, i.e., pretax income without a charge for debt. The next step is to calculate cash taxes on EBIT.
Note if you begin your estimate of cash taxes with the forecasted after debt total tax provision (current and deferred) you must adjust this amount for any tax shields that will arise from interest costs of debt, interest income on marketable securities, and tax effects of non-operating income.
Once taxes on EBIT are estimated, they should be adjusted for any forecasted change in deferred taxes to arrive at forecasted cash taxes on EBIT. For example, if the credit balance in deferred taxes is expected to increase in a given year, the amount by which it rises should be subtracted from cash taxes on EBIT. It has been assumed that the predictable component of changes in the deferred taxes balance will be immaterial for the forecast period (see spreadsheet). EBILAT is equal to EBIT less cash taxes on EBIT.
Depreciation Addback: Because depreciation is not a cash flow it should be added back to EBILAT to arrive at gross cash flow from operations. Note that the tax impact (shield) from depreciation expense has been accounted for in the calculation of cash taxes. Thus, the net effect of depreciation on EBILAT after the addback is (1 - tax rate) X depreciation expense.
Net Investment: The forecast of net investment must account for changes in the level of working capital (current operating assets and liabilities) required to support operations, capital expenditures, property and plant acquisitions/disposals, investments in goodwill, and changes in other operating assets and liabilities.
Changes in working capital: Increases in current operating assets (liabilities) represent decreases (increases) in working capital. For example, the extension of customer credit through accounts receivable or the payment of supplier accounts payable tie up or use company funds and thus, represent investments. Historical financial statement data for current assets (other than cash) and liabilities suggest estimates of investment in these items of 15% and 13% of sales, respectively. In addition, an increase in the level of cash deemed necessary to support operations is also considered a change in working capital. In the case of The Gap we forecast an increase in operating cash equal to 3.5% of sales growth. It is difficult to justify operating cash balances in excess 2 or 3% of sales for most firms even when cash needs are highly seasonal.
Capital Expenditures: It is often the case that this component of net investment is significant for companies projecting sizable growth. The Gap is no exception. Some information is available in the financial statements to help estimate the level of expenditure required. The company spent $227 mil in fiscal 1991 carrying out store expansion and remodeling. The ratio of net PP&E (including lease rights and other assets) to sales following this expenditure was 23%. Management expects to spend around $230 million on CAPX in the coming year for an expansion program similar in magnitude and nature to this year. Based on management's stated goal of 20% sales growth and the estimate of depreciation for fiscal 1992, it appears that the ratio of PP&E to sales will once again be 23%. Therefore, in the absence of more detailed information, CAPX is set such that net PP&E remains at 23% of sales over the forecast period. What are the implications of this procedure for The Gap’s projected asset turnover?
Note this valuation does not address the issue of capitalizing operating leases. The reason is that altering the accounting treatment of this item should not, in principle, affect discounted free cash flow. Can you explain why this is the case?
Step 2 Estimate the value of cash flows beyond the forecast horizon. As a practical matter, most analysts do not forecast FCF beyond 10 years. Thus, some calculation must be made for that part of firm value that will be realized beyond the forecast horizon. There are many methods available to the analyst for estimating value beyond the terminal year (see chapter 6 of your text for descriptions). All of these methods are flawed in some way. We will discuss some of the more exotic methods of estimating continuing value later in the semester. For now, a simple perpetuity method is employed. This method requires an estimate of a stable EBILAT and net level of investment estimate (including operating cash) at the end of the forecast period. The stable estimate of FCF is divided by the target WACC (see step 3 below) at the end of the forecast period, resulting in a perpetuity calculation of the continuing value of the firm. Keep in mind that this perpetuity is estimated at the end of the forecast horizon so we must discount it back as a lump sum for the number of years in the forecast horizon to get its actual NPV. In The Gap example, the FCF in year 10 is divided by the estimated WACC of 14%.
An important implicit assumption underlying the simple perpetuity method is that the marginal rate of return on investment is equal to the WACC at the end of the forecast period. That is, operating returns on new investment are exactly matched to the cost of generating those returns. This implies that you may assume sales growth of zero and still get the same answer that you would with detailed forecasts of non-zero sales growth. Thus even if the firm’s sales and assets continue to grow, it is assumed that this growth fails to add value to the firm (see table 6-A in your text for an example).
Finally, the assumption of zero sales growth beyond the terminal year will have the unwanted impact of causing the firm to contract if inflation is expected after the terminal year (a good bet for the long run). One way to deal with this shortcoming is to allow for 2-4% growth (a reliable estimate of long-run inflation) in the perpetuity. This requires adjustment of the terminal year FCF in the numerator by the inflation rated and adjustment of the WACC in the denominator by the same rate, i.e.,
Where g includes the long run inflation rate. More generally, if you believe some new investment (sales) will generate returns in excess of the WACC beyond the forecast horizon, this can also be captured by increasing the value of g. However, you should keep in mind that the larger the abnormal return you build into your estimate of terminal FCF, the more exaggerated the impact of assuming positive growth in perpetuity. Positive growth with high returns implicit in the terminal FCF estimate may be appropriate for growth firms or risky R&D firms but less appropriate for firms in mature and competitive industries.
Step 3Determine the WACC. To be consistent with our approach that values the entity as a whole without regard to capital structure, it is necessary to estimate the "charge" for capital used to generate revenues. Because The Gap employs both debt and equity funds we require a weighted average of the opportunity cost to these two types of investors. The general formula is:
where rd = pre-tax cost of debt
re = cost of equity
t = the marginal tax rate
Ve = market value of equity
and, Vd = market value of debt
Note that the market value rather than the book value of investors’ claims must be estimated. A number of other types of financing can also be included in the calculation of the WACC. Can you think of another source of financing other than debt and common stock that could be considered in calculating The Gap's cost of capital?
To calculate the cost of capital in the case of The Gap it would be conceptually sound to use the yield to maturity on the firm's outstanding market debt in the estimate the cost of debt capital. However, because some of the debt is in the form of commercial loans and no information is provide in the case about current yields, the calculation used in this example is based on the coupon rate of 8.9%. The book value of debt, $142 million, is used as lieu of its market value to determine the weights applied to derive the WACC.
An estimate of the cost of equity capital can be obtained in a variety of ways. For example, it can be estimated from the CAPM as follows:
where is an estimate of the covariance of The Gap's security return with that of the market portfolio estimated from a time-series regression, is the expected return on the market (time subscript omitted), and is the risk-free rate (e.g., the T-bill rate, or the rate on intermediate-term Treasury bills). The last term in brackets on the right hand side of the equation above is referred to as the market risk premium. An empirical estimate of the long-run geometric market risk premium is approximately 7.6%, see Ibbotson [1992]. The Gap’s is estimated to be 1.3. At the time of the case, the ten-year Treasury bond rate stood at approximately 6.3%. Thus, the CAPM estimate of the cost of equity capital would be 16%. However, The Gap is a member of the largest size decile of firms in the economy and an adjustment for size may be appropriate. Table 6-4 in your text summarized the relation between firm size decile membership and returns since 1926. The value weighted average return over all firms is about 12.3%. The average return to the largest decile of which the GAP is a member is 11.2%, suggesting a reasonable estimate of the cost of equity for the Gap would be approximately 1% lower than calculated using the CAPM. This downward adjustment leads to an estimated cost of equity capital of 15%.
Note that the $57/share and 142.5 million shares outstanding imply a total market value of The Gap's equity of $8094 million. But plugging the actual market value of the firm into the calculation of WACC involves circular reasoning (since we are trying to determine what that market value should be!). Thus, it is necessary to guess at the firm's market value, use the guess to determine the weights to apply in the WACC and determine if the estimated WACC leads to a projected equity value of the firm equal to your original guess. It almost certainly will not the first time around. Continue to iterate the process of guessing the value of equity to determine weights for a WACC that, once applied to the FCF and continuing value estimates, yields a value of equity equal to the guess you started with. The iteration process should converge systematically. For example, if you guess a value of equity that leads to a WACC that once applied to your FCF estimates yields an equity value of the firm lower than the original guess; the WACC estimate is too high. This means you will have to lower your guess of the value of equity (to an amount between the previous guess and the value of the firm calculated using the WACC implied by the previous guess) to decrease the WACC. This procedure has the effect of moving the equity value of the firm toward your guess without exceeding it. After a few iterations you should arrive at a value of equity which is internally consistent with the WACC. The WACC calculated for the GAP in this case is approximately 14%.
How would you estimate the WACC if the current capital structure were not expected to remain the same in the future? Can you think of business situations where this complication must be considered?
Step 4 Calculate the equity value of the firm. After the present value of FCF over the forecast horizon and the value of the perpetuity beyond the forecast horizon have been determined, the next step is to add excess cash and marketable securities on had currently. The logic behind this treatment of financial assets is that they are assumed to be zero NPV potential investments. That is, the cost of funds to hold these financial assets is equal to the returns they generate.
The equity value of the firm is determined by subtracting the market value of debt from the total equity value forecasted using steps 1-4.
Because it is not usually reasonable to assume cash flows arrive at the end of the period it is sensible to make adjustment for their average timing. The adjustment in this case is for one half year, essentially an assumption that cash flows are received evenly throughout the year. Some care should be taken with assumption as it could have a substantial impact on the valuation.
The forgoing calculations lead to an equity value of $2213 million. With 142 million shares outstanding, this implies a price per share of $15.59, considerably lower than the market price of $57
Step 5: Following through with the Valuation
Review forecasted values and ratios for consistency and reasonableness
Often the first attempt at a valuation results in an answer that surprises you (were you surprised at you valuation of The GAP? Why?). Errors in calculation are frequent. Perhaps more important are errors of internal consistency and failure to forecast numbers that reflect the conclusions of your business strategy analysis. One way to check for internal consistency is to produce a set of pro forma financial statements to ensure they articulate. An abridged version of this check is provided for in the spreadsheet. A second method of ensuring internal consistency and reasonableness is to check the profitability, liquidity and activity ratios implied by the projections. For example, one can ask, did forecasted ROEs and profit margins converge to levels anticipated after an analysis of the firm's business strategy and competitive environment? Does the firm have the necessary liquidity to carry through with its planned expansion? Once again the spreadsheet includes a running summary from year to year of important ratios.