There is already much literature which compares the pros and cons of using the internal-rate-of-return (IRR) as a measure of a project’s merit compared to using the net present value (NPV). However, the effect of on each measure of the delay of a project is rarely mentioned.
The IRR has a number of well-known drawbacks (such as it not existing at all if a project has only positive cash flows, as would be the case for the future cash flow of a project that is past its start-up investment phase); more generally, there are as many possible IRR values as there are changes in the sign of the cash flows, so a unique solution would often not exist (especially in risk or stochastic models). Similarly, the use of an IRR may favour the selling of a project’s earlier than would otherwise be the case, and so may lead to short-term or biased behaviour. These points are covered in more detail and with examples in my book Financial Modelling in Practice. On the other hand, in a sense, the IRR represents (where it uniquely exists) the actual return experienced by project participants, and in that sense cannot be argued with!
Regarding the issue of time delays, if we take a simple example of a project which consists of an initial investment (cash outflow) followed by a cash inflow at some later date, then on brief reflection it becomes clear that the IRR is relatively insensitive to any delay in the positive cash inflow. For example, for an investment of (minus) 10 with a future value returned of (plus) 10, the IRR is zero whereas the NPV (for any positive discount rate) will be negative. As the cash inflow is pushed further out in time, the IRR remains at zero (hence the IRR is insensitive to a delay of the proejct), whereas the NPV would fall further. One can easily experiment in Excel with various more realistic scenarios (e.g. investment of minus 10 leading to a future return of 20); such tests will confirm the basic behaviour of the relative insensitivity of IRR compared to the NPV insofar as time-shifting is concerned.
An important point in this respect is that firms whose business are project-based (and which use IRR as their chief investment criteria) may be sub-optimising their business due to a lack of pressure (resulting from the financial analysis of the projects) to implement the projects in a timely manner; such firms may be lead to believe that project delays are of little material concern; on the other hand, shareholders and analysts using discounted cash flow (NPV) techniques will more significantly mark down the value of such firms, so that the market value may be significantly reduced compared to what it otherwise would have been.