Multiples analysis is simple to understand and apply. The inputs for the multiple arepublicly available, though are vulnerable to accounting manipulation. Also, it isdifficult to obtain a truly comparable large sample of firms. Multiples analysis isbackward-looking, reliant on historical/current data to obtain multiples. It reflectsrelative value rather than the intrinsic value which DCF valuation produces. DCF analysis generates an intrinsic value as it relies on data specific to the firm. DCFanalysis factors in time value of money, and thus is a forward-looking measure.
However, there is uncertainty in forecasting future revenues, especially for privatefirms and those firms that produce little or no cash flows. Assumptions of multiples analysis General assumptions of multiples analysis are that the other firms in the industry arecomparable to the firm being valued. The market, on average, prices these firmscorrectly, but makes errors on the pricing of individual stocks. Exhibit 2 shows aselection of comparable firms, assuming that these firms have the same growth, riskand return as Cox Communications.
There is also the assumption that financialfundamentals such as EBITDA are defined identically in all firms, with the sameaccounting methods and reporting periods. Exhibit 5 assumes a positive linearrelationship between ROIC and the multiple Adjusted Enterprise Value/AverageInvested Capital. Regression analysis and traditional multiples analysis – similarities anddifferences The two analyses both predict the underlying value of the firm. Also, both regression and multiples analyses reflect the past. The future value of the firm is obtained usinghistorical inputs.
Both analyses assume that firms in the same industry arecomparable. Traditional multiples analysis is more arithmetic in its approach. It is based on findingthe average multiple among comparable firms, and then applying it to the firm’sfundamentals. How accurate the valuation is depends on the degree of comparabilityof the firms in the industry. Regression analysis produces a statistical regression line of each comparable firm’smultiple against the fundamentals that affect the value of the multiple. The R-squared of the regression indicates how well that multiple works in the sector.
Afterrunning the regression and establishing the multiple, then it is applied tofundamentals in order to arrive at a firm valuation price. Interpretation of regression results Martin’s regression results produced a higher share price of $50 (see A1), indicatingthat shares were currently undervalued and so Cox Communications does havegrowth potential. Martin’s heavy reliance on the projection of the ROIC value is troubling. The 0. 8%seems to be an arbitrary numerical projection. Any inaccuracy in this projectionwould result in a misleading outcome.
R-squared is 70%. The percentage variation in ROIC cannot be totally explained bythe variation in Adj. EV/Ave. Invested Capital. However, it does tell us that ROICs area substantive prediction of value. The linear relationship between ROIC and themultiple in the regression shows there is a strong relationship. Martin’s DCFanalysis Martin’s weighted cost of equity is 10. 5%. We have calculated the cost of equity tobe 13. 61% (see A2) using a levered beta and a risk premium calculated over a longerhistorical period.
Using the synthetic rating method, after-tax cost of debt is 4. 51% (see A3), which is close to Martin’s estimate. Thus, with a WACC of 12. 53% the stockprice becomes a more conservative $41. 70 (see A4). Martin’s projected EBITDA growth of 16% seems high since her forecasted revenuegrowth is only 14. 2%. Hence, once other expenses are included, it is unlikely thatEBITDA will retain growth at a higher rate than revenue. Conducting sensitivityanalysis with more conservative EBITA growth at 14% and 12% resulted in shareprices of $37. 02 and $33. 6 respectively (see A6). Martin assumes an increase in capital spending in the first 2 years on new digitaltechnology services, and then a fall to a constant rate, resulting in a fall in assetintensity from 2003.
But there is still 0. 8% increase in ROIC as well as a forecastedgrowth in EBITDA of 16% and EBIT of 33%. These forecasts seem contradictory. Also,Martin has not taken into account the value of the empty cable channels. Substituting the terminal value into the horizon value formula using a WACC of 9. 3%and 12. 53% results in a growth rate of 4. 4% and 7. 9% respectively (see A7). Bothstable growth rates seem reasonable, as FCF has been growing at a high rate (up to474% in 2001). So we can conclude that the terminal value of 13x is reasonable, butit also depends on the reliability of the EBITDA forecast. Real options valuation as an alternative to DCF analysis Of the 750 MHz capacity in an upgraded cable plant, there are 17 unused channels(102 MHz), referred to as the “stealth tier”. DCF analysis does not account for theseinvisible but valuable revenue streams. The real option within this project is the “stealth tier”.
It gives the company the right,but not the obligation, to obtain revenues from the unused cable channels,depending on market conditions. It allows the firm to calculate the intrinsic value of the underlying real project, unaccounted for in DCF analysis. Real options analysiswill also allow managers to adjust business strategies according to market situations. Ways the “stealth tier” can be incorporated into DCF and multiples analysis A qualitative assessment of the real option, which can be added to DCF analysis,shows that the value of the stealth tier would increase the value per share that isproduced by DCF analysis (see A8).
The “stealth tier” could also be valued using decision-tree analysis (see A9). Aninvestment timing option could also be incorporated, which would allow CoxCommunication to determine whether it would be profitable to “light up” and utilizethe empty channels today or at a later date. Including the stealth tier option would impact the multiple analyses based oninvested capital (see A10) and also EBITDA.
For an effective valuation however, CoxCommunications would have to be compared to companies that have a similarstealth tier, which is unrealistic. How is the stealth tier like a call option? Applicability of Martin’s optionanalysis The stealth tier is similar to a call option as the company has the right but not theobligation to use the unused capacity to implement further operating capacity. Therefore the decision of whether to use this capacity is a real option as the stealthtier has a value and a cost (see A11), and it is up to