Tuesday, July 13, 2010


Limitations of Profitability Index

The capital budgeting procedure described above does not always work. It fails in two situations:
  • Multi-period capital constraints
  • Project indivisibility
Multi-period constraints

The serious limitation in using the PI rule is caused by the multi-period constraints. In the above post example, there is a budget limit of 50000$ year 1 also and the firm is anticipating an investment opportunity 0 as in low is year 1. Thus, the decision choices today are as follows:

Project M and N have the first and second ranks in terms of PI. They together have highest NPV and also exhaust the budget in year 0; so the firm would choose them. Further, projects M and N together are expected to generate 20000$ cash flow next year. This amount with the next year’s budget (i.e. 20000$ + 50000$ = 70000$) is not sufficient to accept project O. Thus, by accepting M and N, the firm will obtain a total NPV of 15870$. However, a careful examination of the project’s cash flows reveals that if project L is accepted now it is expected to generate a cash flow of 30000$ after a year, which together with the budget of 50000$ is sufficient to undertake project O next year. Projects L and O have lower PI ranks than projects M and N, but they have higher total NPV of 19820$.

Project indivisibility

The PI rule of selecting projects under capital rationing can also fail because of project invisibility. It may be more desirable to accept many lower ranked similar projects than a single large project. The acceptance of a single large project, which may be top-ranked, excludes the possibility of accepting small projects, which may have higher total NPV. Consider the following:

Suppose that the firm has budget ceiling of 10$ million. Following the ranking by PI, the firm would choose A and C. These projects spend 850000$ of the total a budget and have a total NPV of 180000$. The next best project E needs an investment of 200000$, while the firm has only 150000$. If we examine the various combinations of projects satisfying the budget limit, we find the package of C, E and D as the best. They exhaust the entire budget and have a total NPV of 189000$. Thus, the firm can choose two lower ranked, small projects, E and D, in place of the higher ranked, large project, A. This section procedure will become very unwieldy if the firm has chosen the best package of projects from a large number of profitable projects.

  • Our discussion has shown that the profitability index can be used to choose projects under simple, one-period, capital constraint situation. It breaks down in the case of multi-period capital constraints. It will also not work when any other constraint is imposed, or when mutually exclusive projects, or dependent projects are being considered.

No comments: