The Problems with IRR
January 3, 2012 2 Comments
Internal Rate of Return works well in many situations, but in not quite all. In the typical investment, you expect to have a single negative cash flow on day one (your cash outlay to acquire the investment) followed by a series of periodic positive cash flows. The last of these will be the proceeds of sale when you finally dispose of the investment. In such a scenario, IRR usually works pretty well.
The wheels can start to come off, however, when you use IRR to help you choose among alternative investments. Particularly vexing is a situation where your investment timeline expects to encounter some negative cash flows. Perhaps you’re projecting a significant increase in the interest rate on your financing; or you expect to have some major (but unfunded) repairs; or you want to play “what if?” to see what will happen if you lose an important tenant and seek to replace that tenant quickly. Any of these possibilities could throw your projected cash flow for a future year into the negative.
That’s where the arcane math behind IRR throws you a curve. In general, if you have more than one change of sign in the series of cash flows (and you must include the initial investment as one of the cash flows), then you may encounter “non-unique” results. That’s a polite way of saying the same of facts can give you more than one answer, which clearly is not helpful.
Consider this example from the classic text,
Year 0 Initial Investment: (25,000)
Year 1 Cash Flow: 150,000
Year 2 Cash Flow: (275,000)
Year 3 Cash Flow: 150,000
In this series of cash flow, the sign changes three times; therefore, there could be as many as three different internal rates of return, i.e., rates at which you could discount these cash flows so that their NPV would equal zero. Indeed, there are three such rates: 0%, 100% and 200%, and they’re all mathematically correct.
The IRR is of little value if it presents you with multiple solutions for the same set of data and invites you to pick one of those solutions.
If IRR’s relationship with negative cash flows is occasionally dysfunctional, it doesn’t get along as well as it should with positive cash flows either. Conventional wisdom has long held that IRR assumes that positive cash flows can be reinvested, until the end of the holding period, at the same rate as the IRR itself. There are also those who assert that IRR actually makes no assumption at all as to the rate of reinvestment of positive cash flows.
For our purposes the distinction may be literally academic because in either case the IRR does not attend to how positive cash flows are handled in the real world. You will reinvest positive cash flows at the best rate you can reasonable obtain, and that rate is likely to be closely tied to the size of the cash flow. If your cash flow is large, you may be able to reinvest it in another piece of real estate. If it is small, then passbook savings may be your only option.