# Case analysis Midland Energy

Topics: Capital asset pricing model, Weighted average cost of capital, Risk-free interest rate Pages: 6 (1333 words) Published: November 24, 2013
﻿Cost of Capital Estimate for Midland Energy Resources, Inc.

In the first section of my report, I list out the main models and methods applied to estimate the cost of capital for Midland’s three divisions, general assumptions made and the corresponding justifications. In the second section, Calculations, I not only compute the cost of capital based on the general assumptions previously made, but also discuss specifics of each division and the additional adjustments or assumptions made to justify my estimates.

SECTION 1: Main models and methods applied and corresponding assumptions 1. Constant Debt-ratio Weighted Average Cost of Capital (WACC):

Assumptions:
WACC: as constant debt ratio is the underlying assumption to derive the WACC model, constant debt ratio should be reasonably assumed to be applied by Midland and its three divisions. According to the case, Midland optimizes its debt levels by regularly reevaluations against its energy price and stock price level and each division has its own target debt ratio. Although the actual capital structure sometimes deviates from the target due to factors such as market value of specific collaterals, it is safe to assume that the debt ratio averages out at the target ratio in the long run, given that the target ratios are not adjusted frequently. Therefore, the debt ratio can be viewed as a constant and thus WACC is applicable. Capital structure: as assumed above, target debt ratio is employed in calculation. Tax rate/t: Effective tax rate should be applied to calculate the cost of capital. The average effective tax rate (39.73%, calculated in Section 2) of that in year 2004, 2005 and 2006 is used as the estimated tax rate. Cost of debt: I basically used the same method as Mortensen did in the case, computing the cost of debt for each division by adding a spread over U.S. treasuries with a similar maturity. I assume that the default risk measured by the spread over treasury gives a reasonable estimate on the amount of premium debt holders require to bear the extra risk over risk-free debt.

Cost of equity: CAPM is applied to estimate cost of equity.
2. Capital Asset Pricing Model (CAPM):

Assumptions:
CAPM: it is assumed that only systematic risk is priced and thus CAPM can be properly applied to calculate cost of equity. Risk free rate: to be consistent with the cost of debt estimate, the same risk free rate of each division is used to compute the corresponding divisional cost of equity. Equity beta: since divisional cost of equity is asked to be estimated, the pure play method is applied here, assuming the industry average provides a fair estimate of divisional equity beta in Midland. First of all, the average equity beta of comparable companies provided in the case is unlevered by formula (2), in order to eliminate the influence imposed by different debt ratio employed by those comparable companies. Formula (2) is derived from formula (1) assuming that debt has a beta of zero. This assumption makes sense as the debt beta is usually substantially smaller than the corresponding equity beta and thus makes little difference in the resulted unlevered beta or asset beta. Then the industry average asset beta is re-levered using the target debt ratio to arrive at the estimated cost of equity. However, this methodology does not apply to the equity beta of Petrochemicals division as no equity betas of comparable companies are provided in the case. The way I deal with this problem is described below in the Calculation session.

Equity market risk premium (EMRP): there exist several ways to estimate EMRP and each one has its advantages and limitations. Even estimates derived by the same approach may vary significantly from one to another due to different parameters used such as time horizon choice in historical risk premium approach. Based on the limited information in the case, I would like to use the conventional historical data approach to estimate EMRP. In...