Cost of Capital

I. Cost of Capital

1. Cost of Debt:

Assumptions

a) Current yields on 20-year U.S treasuries remain the same;

b) Nike doesn’t have plans to do large amounts of debt or equity financing, so the D/E ratio won’t change much;

c) Nike’s ratio of short-term liabilities over long-term liabilities remains stable in next 10 years;

d) We need to make projections about the next 10 years’ data on fixed assets, (property, net plant and equipment, net identifiable intangible assets and goodwill, deferred income taxes and other assets) current liabilities (current portion of long-term debt, notes payable) and long-term liabilities (long-term debt, deferred income taxes and other liabilities, redeemable preferred stock). By calculating the ratios each data over sales in 2000 and 2001, we find these ratios relatively stable in these 2 years. So we assume that the ratio will remain as the average value of 2001 and 2002 in the next 10 years.

Conclusion

Yield to Maturity (YTM) is the rate of return anticipated on a bond if held until the end of its lifetime and calculated using the formula bond value = coupon value + par value. Nike’s coupon is 6.75% yielding semi-annually and the par value is $100, so the bond has to pay interest of 3.375 every half a year. Exhibit 4 in the case material provides data of current yield and current price. However, since bonds are usually not par issued, we can’t simply plug in current yield as yield to maturity. In addition, cost of debt consists of short-term paper and long-term bonds. The ratio of notes payable over debt balance is 64% in 2000 and 66% in 2001. By calculation, we find that the ratio of short-term bonds over long-term bonds stable so assume it will remain constant as the average value of 2000 and 2001 (65%) in the next 10 years. Then we calculate the cost of debt by using the formula. Our estimate of cost of debt is 6.73% before tax and 3.97% after tax, which is slightly higher than stated in the case material. [pic 1]

Difference from Ford and Cohen

We didn’t use the method in the case material to calculate cost of debt, which is dividing total interest expense by average debt balance. The reason is that if we use this method, we have to make projections about total interest expense and average debt balance in the next 10 years and the errors in the projection is not negligible.

2. Cost of Equity

We estimated the cost of equity using the capital-asset-pricing model (CAPM).

1) The Risk-Free Rate

The CAPM is a period-by-period model, so a short-term rate, “the one-year Treasury bill rate”, would be a good place to start. But Nike has a long life, so the average one-year rate anticipated over the life of the Nike, rather than today’s one-year rate, is preferred. Then we predict average one-year rate according to the term structure. Over the period from 1959 to 2001, the average return on 10-year treasury bonds was 7.434%, and the average return on one-year Treasury bills was 6.039%. Thus, the term premium was 7.434% - 6.039% = 1.341%. As the current yield on 10-year U.S. Treasury is 5.39%, the average one-year interest rate or risk-free rate expected over the next 10 years is 5.39% - 1.341% = 4.049%. Joanna Cohen just used the current yield on 20-year Treasury bonds as her risk-free rate. The rate will be definitely higher than that we forecasted, because the actual 20-year T-bond is not totally risk-free.

2) Market Risk Premium

Using historical data of “S&P 500 Annual Return” and “10-Year T-bond yield” over 1991-2001, we get market risk premium’s arithmetic mean of 6.87%. We didn’t use the geometric mean of historical data over 1926-1999 as Cohen. As the world economy is changing over time, the data long time ago will be no use.

3) Estimation of Beta

        According to the data from NYSE, we plot Nike monthly returns against monthly returns on the S&P 500 Index. Using a standard regression technique, we fit a straight line through the data points. The slope of the characteristic line is beta. We choose 10-year (1991-2001) monthly data to estimate beta. Even though observations from the distant past will be out of date if firms may switch to another industry, 10-year data set is more convincible than 5-year, as shown in the linear regression output. P-value of 10-year data set is 0.0000105, much less than the 5-year value of 0.00134, because Nike didn’t change its strategy significantly in the past ten years. In addition, Cohen misestimated the beta by simply calculating the last 5 year average. Therefore, using regression of 10-year stock monthly return on S&P 500 monthly return is more reasonable, and we get a beta for Nike of 1.08. Therefore, the cost of equity = 4.049% + 1.08 * 6.87% = 11.47%.