Not Of Any General Interest
I'm helping a friend of mine who's taking a finance course. The textbook for the course is called Fundamentals of Corporate Finance
by Brealey, Myers and Marcus. In general the book is solid, but I did find one gaping error.
In Chapter 7, while discussing some of the problems with IRR, they give an example of an office building with two options. Under scenario one, they construct the office building for $350,000 and sell it one year later (after completion) for $400,000. In the second scenario, they lease it out for three years at 16,000 per annum, and sell it at the end of the third year for $450,000. They point out correctly that the IRR under the first scenario is 14.29% and the IRR under the second scenario is 12.96%. However, they claim that an NPV calculation of the cash flows at a 7% discount rate is higher for the second scenario, and therefore the second scenario is the better option. Here are the cash flows they present (000s):
Scenario 1: -350 +400
Scenario 2: -350 +16 +16 +466.
What did they miss? It's a bit tricky, but they missed what the developer did with the $400,000 at the end of the first year. Assuming he could reinvest it safely at 7%, there would be additional cash flows of $28 (000s) in years two and three, which have to be factored into the analysis. If you add those cash flows back in and do the NPV analysis, you'll find that the first scenario is indeed the better option.
Note: This is not to say that there are not other problems with IRR. But the example given is quite plainly mistaken.