1. Time value of money may be ignored. For investments whose consequences are largely concentrated within a few years (say 10 years), this problem usually can be managed by discounting future cash flows, generally called "present value" calculations.

Future cash flows are "discounted" to their present value equivalents. In other words, what amount today, if invested at appropriate returns ("cost of capital" or "discount rate"), will yield the future cash flow being discounted. E.g. $621 would be the present value of $1,000 in five years at a discount rate of 10% because $621 invested now with a return of 10% compounded annually would yield $1,000 in five years. PV = FV/(1 + i)ⁿ is the equation for this calculation.

The discount (interest) rate used becomes crucial, increasingly so for cash flows taking place in more distant years. Selecting the appropriate rate can be a difficult decision, can be a source of manipulation, and should be clearly revealed. Long term investments, especially those extending over generations, make discounting very problematic because future benefits and costs become inconsequential through such calculations. This raises fundamental issues of distributive justice across generations.

2. Benefit/Cost ratios are easily manipulated. The most obvious problem is that by using net benefits, any project in which calculated benefits exceed calculated costs can be shown to have a ratio as high as one desires. Rules limiting this manipulation are themselves ultimately arbitrary, favoring some projects over others. A solution to this problem is to not use ratios for setting priorities among competing projects. Instead, comparisons of net present value may be used, although this too has several problems (e.g. problem 1, above).

3. What is measured? This includes innumerable questions and problems. A particularly obvious questions is whether to limit the benefits and costs calculated to those that can be measured in monetary terms. There are no solutions, except careful thought, integrity, and recognition that there is no complete measure and that benefit-cost analysis is more useful as a way of thinking than as a decision rule.

4. How to measure those matters that are considered? This too is an irresolvable set of problems. An example is what value to place upon life. One way to deal with this is through cost-effectiveness analysis in which the objective (say lives saved) is not given a monetary value but is used to compare costs per unit of objective (e.g. life saved) achieved. This too raises its own questions and results in various measures, a common one in health being QALYs (Quality Adjusted Life Years; see AcademyHealth Glossary).