While net present value (NPV) calculations are useful when you are valuing investment opportunities, the process is by no means perfect.
The biggest disadvantage to the calculation of NPV is its sensitivity to discount rates. After all, NPV computations are really just a summation of multiple discounted cash flows - both positive and negative - converted into present value terms for the same point in time (usually when the cash flows begin). As such, the discount rate used in the denominators of each present value (PV) computation is critical in determining what the final NPV number will turn out to be. A small increase or decrease in the discount rate will have a considerable effect on the final output.
Let's say you were trying to value an investment that would cost you $4,000 up front today, but was expected to pay you $1,000 in annual profits for five years (for a total nominal amount of $5,000), beginning at the end of this year. If you use a 5% discount rate in your NPV calculation, your five $1,000 payments are equal to $4,329.48 of today's dollars. Subtracting the $4,000 initial payment, you are left with an NPV of $329.28. (To learn more about calculating NPV, see Understanding The Time Value Of Money and Anything But Ordinary: Calculating The Present And Future Value Of Annuities.)
However, if you raise the discount rate from 5% to 10%, you get a very different NPV result. At a 10% discount rate, your investment's cash flows add up to a present value of $3,790.79. Subtract the $4,000 initial cost from this amount, and you're left with a negative NPV of $209.21. Simply by adjusting the rate, you have gone from having an investment that creates $329.28 of value to having one that destroys $209.21 instead.
Of course, you'll want to undertake the investment if 5% is the correct rate to use, and reject it if 10% is the correct rate. But how do you know which discount rate to use? Accurately pegging a percentage number to an investment to represent its risk premium is hardly an exact science. If the investment is very safe, with low risk of loss, 5% may be a reasonable discount rate to use, but what if the investment harbors enough risk to warrant a 10% discount rate? Bottom line, since NPV calculations require a discount rate, there is no way to get around this issue; therefore, it is a big disadvantage to the NPV methodology.
Making matters even more complex is the possibility that your investment won't have the same level of risk throughout its entire time horizon. In our example of a five-year investment, how would you handle a situation in which the investment had high risk of loss for the first year, but relatively low risk for the last four? You can try to use different discount rates for each time period, but this will make your model even more complex and require a lot on your part to peg not only one discount rate accurately, but five. This is another disadvantage to using the NPV model.
Finally, another major disadvantage to using NPV as an investment criterion is that it wholly excludes the value of any real options that may exist within the investment. Consider again our five-year investment example - suppose this is a startup technology company, which is currently losing money but is expected to have the opportunity to expand greatly in three years' time. If you know the company has this valuable real option of expansion in the future, shouldn't you incorporate the value of that option into the total NPV of the investment? Clearly, the answer is "yes", but the standard NPV formula provides no way to include the value of real options. (For further reading, see An Introduction To Real Options.)
Thus, NPV is a useful starting point to value investments, but certainly not a definitive answer that an investor can rely on for all investment decisions. To learn more, check out Discounted Cash Flow Analysis and What's the difference between net present value and internal rate of return?