IRR & Limitations — Internal rate of return (IRR) and...

SECTION 1

Historical Context & Motivation

Capital budgeting decisions have shaped the trajectory of corporations for centuries, yet for much of business history, managers lacked rigorous quantitative tools to evaluate potential investments. Before the mid-twentieth century, firms relied heavily on intuitive judgment, simple payback rules, or accounting-rate-of-return measures that ignored the time value of money. As modern finance theory matured, practitioners sought a single metric that could express a project's return in percentage terms—a number that could be compared directly to a firm's cost of capital. The internal rate of return (IRR) emerged to fill precisely that need, becoming one of the most widely cited decision criteria in corporate finance.

1907

Fisher's Rate of Return over Cost

Irving Fisher introduced the concept of comparing the rate of return on investment to the cost of capital, laying the intellectual groundwork for what would later become the IRR framework in his seminal work The Rate of Interest.

1930s

Boulding & Keynes Formalize IRR

Kenneth Boulding (1935) and John Maynard Keynes (1936) independently developed the concept of the 'marginal efficiency of capital,' essentially the discount rate that equates the present value of future cash flows to the initial investment—the modern definition of IRR.

1951

Lorie–Savage Problem

James Lorie and Leonard Savage published a landmark paper identifying capital-rationing problems and highlighting cases where IRR rankings diverge from NPV rankings, sparking decades of academic debate about the reliability of IRR.

1955

Samuelson's Critique & Multiple IRRs

Paul Samuelson and others demonstrated that projects with non-conventional cash flows can produce multiple IRRs, formally establishing one of the method's most significant theoretical limitations.

1990s–Present

MIRR & Modern Practice

The Modified Internal Rate of Return (MIRR) was popularized as a remedy for several IRR shortcomings. Surveys consistently show that more than 75% of CFOs still use IRR alongside NPV, underscoring its practical appeal despite well-known limitations.

The central question that motivates this lesson is straightforward: if a project's IRR exceeds the firm's required rate of return, should the firm always accept it? As we will see, the answer is a qualified no. Understanding both the power and the pitfalls of IRR is essential for making sound capital budgeting decisions.

SECTION 2

Core Principles & Definitions

At its core, the internal rate of return is the discount rate at which a project's net present value equals zero. In other words, it is the break-even cost of capital: if you could borrow at exactly this rate, the project would neither create nor destroy value. When the IRR exceeds the firm's weighted average cost of capital (WACC) or the project-specific hurdle rate, the project appears to add value. When it falls below, the project is value-destroying. This percentage-based framing is intuitive for managers who think in terms of returns, which partly explains IRR's enduring popularity.

IRR Definition

The discount rate r that sets the NPV of all cash flows equal to zero. Solving NPV(r) = 0 yields the IRR.

Decision Rule

Accept a project if IRR > hurdle rate (cost of capital). Reject if IRR < hurdle rate. This rule holds for independent, conventional projects.

Conventional Cash Flows

A project has conventional cash flows when an initial outflow (negative) is followed by a series of inflows (positive). Only one sign change occurs, guaranteeing a unique IRR.

Scale Problem

IRR ignores the absolute dollar magnitude of a project. A small project with a high IRR may create less wealth than a large project with a lower IRR—a critical issue for mutually exclusive choices.

Multiple IRRs

When cash flows change sign more than once (non-conventional), the NPV equation may have multiple roots, producing two or more IRRs—none of which is economically meaningful by itself.

✦ KEY TAKEAWAY

Think of IRR as a speedometer reading: it tells you how fast a project generates returns, but it says nothing about how far you travel—that is, how much total wealth is created. A cyclist sprinting at 30 mph for one mile covers far less distance than a car cruising at 20 mph for one hundred miles. Similarly, a high IRR on a tiny investment may produce far less NPV than a moderate IRR on a large one. Always pair IRR with NPV for the complete picture.

SECTION 3

Visual Explanation — NPV Profile

The most powerful way to understand IRR is through an NPV profile—a graph that plots a project's NPV on the vertical axis against a range of discount rates on the horizontal axis. For a conventional project, the curve starts above the horizontal axis (positive NPV at low discount rates), slopes downward, and crosses zero at exactly the IRR. This crossing point is the geometric interpretation of the IRR: it is the x-intercept of the NPV profile.

The NPV profile plots each project's NPV across a range of discount rates. Project B (pink) has a higher IRR (22%) than Project A (blue, 18%), yet at discount rates below the crossover rate of approximately 7.2%, Project A generates a larger NPV. This illustrates the scale problem: IRR ignores the absolute magnitude of value creation.

Notice several features in the diagram above. First, both curves slope downward because higher discount rates reduce the present value of future cash inflows. Second, Project A's curve is steeper and starts higher because it involves larger cash flows. Third, the crossover rate marks the discount rate at which both projects have identical NPVs—to the left of this point, the NPV ranking favors Project A despite its lower IRR. This is the clearest visual evidence that IRR can lead to incorrect rankings when comparing mutually exclusive projects of different scale.

SECTION 4

Mathematical Framework

The IRR is derived from the fundamental net present value (NPV) equation. Recall that the NPV of a project with cash flows CF₀, CF₁, CF₂, …, CFₙ at discount rate r is computed by discounting each future cash flow back to the present and summing. The IRR is the specific value of r that makes this sum exactly zero. Because the equation is an nth-degree polynomial in (1 + r), there is generally no closed-form algebraic solution for n > 4, and iterative methods (trial-and-error, Newton-Raphson, or financial calculator functions) must be used.

NPV EQUATION

NPV = Σ (CFₜ / (1 + r)ᵗ) for t = 0 to n

Where CFₜ = cash flow at time t, r = discount rate, and n = number of periods.

IRR DEFINITION

0 = CF₀ + CF₁/(1 + IRR)¹ + CF₂/(1 + IRR)² + … + CFₙ/(1 + IRR)ⁿ

IRR is the value of r that satisfies NPV = 0. For conventional cash flows (one sign change), a unique positive IRR is guaranteed by Descartes' Rule of Signs.

DECISION RULE

Accept if IRR > WACC (hurdle rate); Reject if IRR < WACC

This rule is reliable only for independent projects with conventional cash flows. For mutually exclusive projects, always verify with NPV.

An important mathematical nuance arises from Descartes' Rule of Signs, which states that the maximum number of positive real roots of a polynomial equals the number of sign changes in the sequence of coefficients. Applied to the NPV polynomial, if cash flows change sign k times, there can be up to k positive real IRRs. Conventional projects (one sign change) have at most one positive IRR, but non-conventional projects—such as those requiring remediation costs at the end of their life—can produce multiple IRRs, rendering the standard decision rule ambiguous.

MODIFIED IRR (MIRR)

MIRR = (FV of inflows / PV of outflows)^(1/n) − 1

MIRR reinvests cash inflows at the firm's cost of capital (rather than at the IRR itself) and finances outlays at the financing cost, producing a unique rate that avoids the multiple-IRR problem.

SECTION 5

Detailed Breakdown of IRR Limitations

While IRR remains an indispensable tool in the practitioner's toolkit, its limitations are well-documented and consequential. The three most critical shortcomings are the scale (size) problem, the multiple IRR problem, and the reinvestment rate assumption. Each of these can lead to suboptimal investment decisions if IRR is used as the sole criterion.

This NPV profile for a non-conventional project (cash flows: −$100, +$300, −$200) illustrates the multiple IRR problem. The curve crosses the zero line twice, yielding IRR₁ ≈ 11.8% and IRR₂ ≈ 88.2%. The project creates value only when the discount rate falls between these two values—a fact that the standard IRR decision rule cannot capture.

The Three Critical Limitations

Summary of key IRR limitations

Limitation	Description	When It Matters
Scale (Size) Problem	IRR expresses return as a percentage, ignoring the dollar amount invested. A project investing $1,000 at 50% IRR creates $500, while one investing $1 million at 15% creates $150,000.	Mutually exclusive projects of different sizes. Firms that can only choose one must maximize NPV, not IRR.
Multiple IRRs	Non-conventional cash flows (more than one sign change) can produce multiple positive IRRs. Descartes' Rule allows up to k real roots for k sign changes.	Projects requiring significant outlays mid-life or at termination (e.g., environmental cleanup, nuclear decommissioning).
Reinvestment Rate Assumption	IRR implicitly assumes that intermediate cash flows are reinvested at the IRR itself, which may be unrealistic for very high or very low IRRs.	Long-duration projects with large intermediate inflows. MIRR corrects this by assuming reinvestment at the cost of capital.
Timing Problem	When comparing projects with different cash flow timing patterns, IRR may rank them differently from NPV because earlier cash flows are disproportionately impactful in the IRR calculation.	Mutually exclusive projects with different durations or different timing of peak cash flows.

💡 Practical Rule of Thumb

When comparing mutually exclusive projects, compute the incremental IRR—the IRR of the difference in cash flows between the larger and smaller project. If the incremental IRR exceeds the cost of capital, the larger project is preferred despite potentially having a lower individual IRR. Alternatively, simply rely on NPV, which always gives the correct ranking.

SECTION 6

Worked Example — IRR vs. NPV Conflict

Consider a firm with a WACC of 10% that must choose between two mutually exclusive projects. Project Alpha requires an initial investment of $10,000 and generates cash inflows of $8,000 in Year 1 and $6,000 in Year 2. Project Beta requires $50,000 upfront and generates $30,000 in Year 1 and $30,000 in Year 2. Which project should the firm choose?

IRR vs. NPV for Mutually Exclusive Projects

Step 1 — Identify Cash FlowsProject Alpha: CF₀ = −$10,000, CF₁ = +$8,000, CF₂ = +$6,000. Project Beta: CF₀ = −$50,000, CF₁ = +$30,000, CF₂ = +$30,000. Both have conventional cash flows (one sign change), so each will have a unique IRR.

Step 2 — Compute IRR for Project AlphaSet NPV = 0: −10,000 + 8,000/(1+r)¹ + 6,000/(1+r)² = 0. Using trial and error or a financial calculator, we find the discount rate that zeroes out the NPV. Testing r = 30%: NPV = −10,000 + 6,154 + 3,550 = −$296 (slightly negative). Testing r = 28%: NPV = −10,000 + 6,250 + 3,662 = −$88. Testing r = 27%: NPV = −10,000 + 6,299 + 3,720 = +$19. Interpolating, IRR_Alpha ≈ 27.1%.

IRR_Alpha ≈ 27.1%

Step 3 — Compute IRR for Project BetaSet NPV = 0: −50,000 + 30,000/(1+r)¹ + 30,000/(1+r)² = 0. Testing r = 13%: NPV = −50,000 + 26,549 + 23,495 = +$44. Testing r = 13.1%: NPV ≈ −$14. Interpolating, IRR_Beta ≈ 13.1%.

IRR_Beta ≈ 13.1%

Step 4 — Compute NPV at WACC = 10%NPV_Alpha = −10,000 + 8,000/1.10 + 6,000/1.21 = −10,000 + 7,273 + 4,959 = +$2,232. NPV_Beta = −50,000 + 30,000/1.10 + 30,000/1.21 = −50,000 + 27,273 + 24,793 = +$2,066.

NPV_Alpha = $2,232 > NPV_Beta = $2,066

Step 5 — Make the DecisionIRR ranking: Alpha (27.1%) > Beta (13.1%) → Alpha wins. NPV ranking: Alpha ($2,232) > Beta ($2,066) → Alpha also wins. In this case, both criteria agree. However, if Beta's Year 2 cash flow were $32,000 instead of $30,000, its NPV would rise to $3,719, exceeding Alpha's NPV—yet its IRR would only increase to about 14.7%, still far below Alpha's. That scenario illustrates the scale problem: the correct choice is the project that maximizes NPV, not the one with the highest percentage return.

Choose Project Alpha (both IRR and NPV agree at 10% WACC)

SECTION 7

IRR vs. NPV — Strengths & Limitations Compared

The debate between IRR and NPV is one of the most enduring in corporate finance. Each method has clear advantages, and in practice most CFOs use both. The table below summarizes how the two criteria compare across the dimensions that matter most for capital budgeting decisions.

IRR vs. NPV comparison across key dimensions

Criterion	IRR	NPV
Output	A percentage rate of return (%)	A dollar amount ($)
Intuitiveness	Very intuitive; easy to compare to hurdle rates and benchmark yields	Requires understanding of present value; less intuitive for non-finance managers
Scale sensitivity	Ignores project scale—a major flaw for mutually exclusive comparisons	Captures the total dollar value created; fully scale-sensitive
Non-conventional cash flows	May produce multiple IRRs or no real IRR	Always yields a unique, interpretable value
Reinvestment assumption	Assumes reinvestment at the IRR (often unrealistic)	Assumes reinvestment at the cost of capital (more realistic)
Mutually exclusive projects	Can lead to incorrect ranking (use incremental IRR as a workaround)	Always gives the correct value-maximizing ranking
Practical usage	~76% of CFOs use IRR (Graham & Harvey, 2001)	~75% of CFOs use NPV (Graham & Harvey, 2001)

✦ KEY TAKEAWAY

Think of NPV as the total profit from a business venture and IRR as the profit margin. A small boutique may have a 40% profit margin, but a large retailer with a 5% margin can generate far more total profit. In capital budgeting, maximizing shareholder wealth means maximizing total value (NPV), not the percentage return (IRR). Use IRR as a screening tool and sanity check, but let NPV make the final call on mutually exclusive investments.

SECTION 8

Connection to Advanced Theory — MIRR & Real Options

The limitations of IRR have spurred the development of more sophisticated capital budgeting techniques. The Modified Internal Rate of Return (MIRR) directly addresses two of IRR's flaws: the reinvestment rate assumption and the possibility of multiple solutions. MIRR compounds all positive cash flows forward to the terminal date at the firm's cost of capital and discounts all negative cash flows back to the present, then solves for the single rate that equates the two. Because there is only one terminal value and one present value, MIRR always yields a unique answer.

IRR vs. MIRR comparison

Feature	IRR	MIRR
Number of solutions	Potentially multiple (up to k for k sign changes)	Always unique
Reinvestment assumption	At the IRR itself	At the cost of capital (more conservative)
Scale problem	Not addressed	Not directly addressed (still a percentage)
Consistency with NPV	May conflict for mutually exclusive projects	More consistent but not perfectly equivalent

Beyond MIRR, advanced capital budgeting increasingly incorporates real options analysis, which recognizes that managers can adapt investment decisions over time—expanding, deferring, or abandoning projects as new information arrives. Neither IRR nor standard NPV fully captures this managerial flexibility. Real options valuation borrows from financial options pricing theory (Black-Scholes, binomial models) to assign value to the flexibility embedded in capital projects. While a full treatment of real options lies beyond the scope of this lesson, it is important to recognize that IRR and even NPV represent static, now-or-never decision frameworks; real-world capital budgeting often involves dynamic, staged decisions where these metrics serve as useful starting points rather than final answers.

🔭 Looking Ahead

In advanced corporate finance courses, you will encounter the profitability index (PI), which divides NPV by the initial investment to create a ratio that partially addresses the scale problem. You will also study equivalent annual annuity (EAA) for comparing projects with unequal lives, and real options for valuing managerial flexibility. Each builds upon the NPV-IRR foundation established here.

SECTION 9

Practice Problems

PROBLEM 1 — CONCEPTUAL

A firm evaluates two independent projects. Project X has an IRR of 18% and Project Y has an IRR of 14%. The firm's WACC is 12%. According to the IRR decision rule, which project(s) should the firm accept, and why does the 'accept all with IRR > WACC' rule work correctly here?

PROBLEM 2 — BASIC CALCULATION

A project requires an initial investment of $5,000 and produces a single cash inflow of $6,500 at the end of Year 1. Calculate the IRR of this project.

PROBLEM 3 — INTERMEDIATE

Two mutually exclusive projects have the following cash flows. Project A: CF₀ = −$20,000, CF₁ = +$15,000, CF₂ = +$12,000. Project B: CF₀ = −$80,000, CF₁ = +$50,000, CF₂ = +$48,000. The WACC is 10%. Project A has an IRR of approximately 22.5% and Project B has an IRR of approximately 14.5%. Which project should the firm choose, and why might IRR give a misleading signal?

PROBLEM 4 — APPLIED

A mining company considers a project with the following non-conventional cash flows: Year 0: −$500,000 (initial investment), Year 1: +$900,000 (mineral extraction revenue), Year 2: −$350,000 (environmental remediation cost). How many IRRs could this project have? Explain why the standard IRR decision rule is unreliable here and suggest an alternative approach.

PROBLEM 5 — CRITICAL THINKING

Surveys show that roughly 75% of CFOs use IRR in capital budgeting despite its well-documented limitations. Offer a reasoned explanation for why IRR remains so popular in practice, and propose a decision framework that combines IRR and NPV to mitigate IRR's shortcomings while preserving its communicative advantages.

SUMMARY

Lesson Summary

The internal rate of return (IRR) is the discount rate that sets a project's NPV equal to zero. The standard decision rule—accept if IRR exceeds the hurdle rate (WACC)—works reliably for independent projects with conventional cash flows. However, IRR suffers from three critical limitations: the scale problem (ignoring dollar magnitude), the multiple IRR problem (non-conventional cash flows producing more than one root), and an implicit reinvestment rate assumption that may be unrealistic.

For mutually exclusive projects, always rank by NPV rather than IRR because NPV directly measures the dollar value created for shareholders. The crossover rate on the NPV profile identifies the discount rate at which project rankings switch. The Modified IRR (MIRR) resolves the multiple-IRR and reinvestment-rate issues by compounding inflows at the cost of capital and always yielding a unique solution. In practice, the most effective approach is to use IRR as an intuitive communication tool while relying on NPV as the ultimate decision criterion for capital allocation.