Capital Investment and Non- Constant Marginal Cost of Capital Pt 2

V. Discussion and Conclusion This paper establishes a net present value maximizing methodology that accounts for the presence of a non-constant marginal cost of capital. A simulation was run in order to determine, for a specific marginal cost of capital schedule, whether conceptual and decision errors result from traditional simplifying assumptions and if so, to what magnitude. Our results indicate that over- and under- investment, wealth destruction, and inclusion and exclusion errors are common results of using a simplification technique rather than employing our maximum NPV criterion. Our results have potential impact on a wide variety of research thrusts. We provide an (simpler) alternative explanation for over- and under-investment than those forwarded previously. The investment error occurs simply because of the presence of a non-constant MCC, without explicitly modeling information asymmetries, agency theory, contracting, multiple divisions, or explicit capital rationing. The result also transcends typical phenomena blamed for breakdowns in capital budgeting decision criteria: budget constraints, differences in business risk, nonnormal cash flow patterns within projects, unequal lives, mutual exclusivity, etc. The breakpoints in a stepwise MCC emulate to a degree a budget constraint, but without an absolute barrier. The magnitude of error is hard to ignore, and that the probability of error is substantial. We suspect that accepting these errors would be intolerable for any conscientious financial manager. Theorists, researchers, professors, students, financial managers and funds providers have a significant stake in whether or not this method is developed in the literature, included in textbooks, and practiced in the business world. Table 1. Sample Project Cash Flows Figure 1. IRR Method Project IRR wMCC Decision Victor 10.7728% 9.35% accept Tango 10.5753% 9.35% accept Echo 10.4725% 10.5356% reject Total for Accepted Projects................................................ NPV $41,114.00 $23,201.69 -$4,796.97 $64,315.69 PI 1.080616 1.068240 .996622 1.075666 wMCC calculation for project Echo: Investment $650,000.00 $400,000.00 $370,000.00 MCC 9.35% 11.20% 11.9% Proportion Weight 650,000/1,420,000 .4577465 400,000/1,420,000 .2816901 370,000/1,420,000 .2605634 Project Echo Weighted MCC (wMCC)....... wMCC 4.27993% 3.15493% 3.10070% 10.53556% Figure 2. Alternative Rank-Ordering Project IRR wMCC Decision Echo 10.4725% 9.35% accept Victor 10.7728% 10.95098% reject Tango 10.5753% 11.9% reject Total for Accepted Projects................................................ NPV $90,043.30 - $4,795.47 -$22,471.87 $90,043.30 PI 1.063411 .990597 .933906 1.063411 Table 2. Summary for All Permutations Project Rank Order Echo, Tango, Victor Echo, Victor, Tango Tango, Echo, Victor Tango, Victor Echo Victor, Echo, Tango Victor, Tango, Echo Accepted Projects Echo Echo Tango, Echo Victor, Tango Victor, Echo Victor, Tango Total NPV $90,043.30* $90,043.30* $85,119.30 $64,315.69 $83,912.43 $64,315.69 * NPV Maximizing Solution ** PI Maximizing Solution Figure 3. Marginal Cost of Capital Schedule used in the Simulation Total PI 1.063411 1.063411 1.048363 1.075666* 1.043478 1.075666* Figure 4. Examples of Project Cash Flow Patterns Generated Figure 5. Difference in CF0 per Each Comparison Method (Bin identifiers represent CF0 difference midpoints for each bin) Figure 6. Difference in NPV per Each Comparison Method (Bin identifiers represent NPV difference midpoints for each bin) Table 3. Inclusion and Exclusion Errors Comparison Method: IRR Comparison Method: Arbitrary 11% Discount Rate Comparison Method: Arbitrary 13% Discount Rate Comparison Method: Arbitrary 15% Discount Rate Comparison Method: Weighted Average MCC Figure 7. Frequency of Inclusion/Exclusion Errors per Each Comparison Method Appendix 1 – MACRS Depreciation Schedule