Question 1
The rationale of an Apache acquisition of MW Petroleum is plausible, yet there are outstanding concerns. By completing a deal, Apache stands to benefit from several aspects. First, MW isa large company which has more than double Apache’s reserves and it includesproperties that are well-suited to Apache’s operating capabilities. Moreover, on behalf of MW, Amoco operated fields accounting for approximately 80% of MW’s production. Such high operating percentage would promise Apache significant cost-saving opportunities. The acquisition would more than double Apache’s reserves, it would shift Apache’s oil-to-gas ratio from 20-80 to 40-60, and would further diversify Apache geographically. Shifting the oil-to-gas ratio is important because gas prices were highly volatile during this time period while oil was less volatile and, therefore, more predictable. Apache’s revenue stream was high levered and the company’s acquisition driven growth strategy would be more difficult when there is an instability of gas prices.
With an acquisition, Apache could also recognize several synergies that would make the deal even more attractive. The properties of MW would diversify Apache geographically and make it more stable in the US market.These include acquired technical data (further exploration possibilities) and several production optimization techniques. Apache would have access to Amoco’s data, which could lead to undiscovered reserves despite that both parties agreed that the probability of a large find was improbable. Furthermore, Apache chairman Raymond Plank devised a strategy to optimize production. This strategy includes tasks such as recompletion, plugback, well-deepening, and repair. These tasks would lower costs, extend well life, and/or increase well production rates.
There are several problems for Apache in trying to reach an agreement. First and foremost, the acquisition requires significant investment that Apache does not have. Apache cannot raise funds easily because of its Ba3 rating and the current tight lending environment. In addition, the price of oil and gas were becoming volatile in 1980’s and 1990’s so that the expected return of this business would be hard to predict. Another problem Apache could face is the uncertainty about the quantities of oil and gas to be produced from MW fields and the expenses of producing. The ability of managing costs would be important to Apache’s profitability.
Fortunately, Amoco is willing to negotiate on financing options. Apache’s debt will nevertheless be expensive with a credit rating of Ba3 in a time where even AAA-rated debt pays approximately 9%. In addition to high debt costs, oil and gas revenues are difficult to predict due to global political and economic events. To put this into perspective, the annualized standard deviation of oil price changes at the beginning of 1991 was just over 50% per year. Lastly, there is a risk that company estimates of reserves are too optimistic and that Apache may get less in production than estimated. All of these factors could have significant impact on the attractiveness of the acquisition.

Question 2
In order to estimate the value of MW Petroleum Corporation, cash flows were discounted at the WACC. Weights were found to be 23.43% for debt and 76.57% for equity given a book value of debt in 1990 of $200 million and a market capitalization of equity calculated by multiplying the 1990 year end stock price by the shares outstanding. To find the cost of equity, the capital asset pricing model was used. The risk-free rate is assumed using the 30-year U.S. Government Bond, which is 8.24%. A market risk premium of 6% is assumed given oil and gas market volatility. The beta was 0.64, which is the mean asset beta for six independent companies, including Apache. The cost of debt was estimated using the January 1991 industrial bond yields for BBB rated corporations. While Apache had a Ba3 rating, which equates approximately to a B- on an S&P scale, Amoco had mentioned that it could possibly assist in financing concerns; Amoco had a AAA debt rating. Therefore, potentially a twenty-plus percent rate was lowered due to the possibility of Amoco assistance. The tax rate of 30% was assumed given that taxes for 1989 and 1990 were roughly 30%. The resulting WACC estimate is 11.16%. Given the WACC, the aggregated cash flows were then discounted to achieve a net present value of $531.91 million.

Question 3
By using the binomial trees to value the proven undeveloped, the probable, and the possible reserves, the value of MW is estimated at $138.19 million. In assuming five years deferral and a convenience yield of 7%, volatilitiesof 30%,40%, and 50% are used for the proven undeveloped, probable, and possible reserves, respectively. The benefit from physically holding the asset in the inventory and using it on the machines during the process would lead to a convenience ratio within the 5-10% interval.First, the up, down, and risk-neutral values were calculated for the proven undeveloped reserves at 1.37, 0.75, and 0.43, respectively.According to the binomial tree (see Table 7), the result is to exercise the option, or drill, today with a project value of $50.55 million. The up, down, and risk-neutral values in probable reserves are 1.51, 0.68, and 0.40, respectively. Based on the lattice (see Table 8), the option should also be exercised today with a project value of $49.32 million. Finally, the up, down, and risk-neutral values in possible reserves are 1.67,0.61, and 0.38, respectively. Judging from the tree (see Table 9), the option should not be exercised until t=2 with a project value of $188.53 million. However, that is under the assumption that prices continuously rise, and that what happens in the following years is unforeseeable. For this reason, it is important to value the option of waiting at t=0 instead, which is $38.32 million.In total, when the DCF of proven developed ($416.33 million) is added to the three lattices, the value isequal to $554.52 million. The value from the binomial tree is higher than the standard DCF calculation because of the flexibility of the waiting option that the binomial tree model brings for possible reserves.

Question 4 In the binomial tree valuation, Apache has the choice of purchasing or waiting to purchase MW. In evaluating when to exercise this option, Apache considers best and worst case scenarios in terms of pricing for each successive year. Given these expectations, Apache would drill as soon as possible (today) for the proven undeveloped and probable oil fields. This is because the net present value of drilling now is higher than the net present value of waiting. However, Apache would choose to wait until t=2 to drill in the possible oil field and only if the best case scenariosare realized. The reason that the possible field would not be exercised until the t=2 is that the oil field is more risky and requires further capital requirements for advanced recovery techniques. The crystal ball analysis shows that with the convenience yield interval of 5-10%, the possibility that the NPV of waiting is larger than the NPV of drilling is 0% in the proven undeveloped oil field. Also for the probable oil fields, the possibility that the NPV of waiting is larger than the NPV of drilling is 8.11% and this probability is found to be immaterial. However,the NPV of waiting for the possible oil field is larger than the NPV of drilling with a 100% certainty. This analysis also suggests that waiting would be more beneficial with the possible oil field.

Question 5 If the Black-Scholes method was used instead of the tree analysis, the option value would be lower. The Black-Scholes model is a special case of the binomial model where the number of binomial steps is infinite and the option is a European option. On the other hand, lattice models are frequently used to value American options;yet, for European options, if step-size gets infinitely small, the value in the tree will approach the Black-Scholes value.The Black-Scholes method does not consider the steps along the way where there could be the possibility of early exercise of an American option. Rather, it only calculates the option price at one point in time - only at expiration. Where an American and a European option are otherwise identical, a European option price would be less than or equal to the price of an American option. The European option price (and the result with the Black-Scholes) is lower because it does not have an early exercise option, hence it is less risky (the exercise can only be at the expiration date, so there is less uncertainty about timing). When Black-Scholes is used,there is a need to distill real life variables into the variables used in Black-Scholes. Also, the Black-Scholes model calculates the price while ignoring dividends (the convenience yield in this case) paid during the life of the option. Finally,it can be mentioned that the Black-Scholes method is much easier to implement than lattice models.

Appendix

Table 1 (Assumptions)

Table 2 (DCF Aggregated)

Table 3 (NPV with lattices)

Table 4 (Assumptions for undeveloped)

Table 5 (Assumptions for probable)

Table 6 (Assumptions for possible)

Table 7 (Lattice Undeveloped)

Table 8 (Lattice Probable)

Table 9 (Lattice Possible)

Table 10 (CrystalBall Analysis for Undeveloped)

Table 11 (CrystalBall Analysis for Probable)

Table 12 (CrystalBall Analysis for Possible)