A contribution to the NPV-IRR debate

This paper proposes a contribution to the century-old debate between npv and irr: we argue that the irr(s) of a project are the interest rate(s) where an agent may switch from finding the proje ct profitable to finding it detrimental (or the other way around). We propose a path to deal with the case when the agent’s reference rate is not unique, and we show how our interpretation accommodates most of the so-called flaws of the irr, even when it is not unique. We show that it is possible to measure the sensitivity of every irr of a project w.r.t. the exchanged amounts.