No, IRR is not the same as the discount rate, though they share the same underlying math and get mixed up constantly. The discount rate is an input you bring to the analysis: the return your capital needs to earn, whether that’s your cost of funds or a hurdle set by policy. The internal rate of return is an output that falls out of a project’s own cash flows: the effective annual yield those cash flows produce on their own. You compare the two to decide whether to invest. If the IRR exceeds the discount rate, the project is worth pursuing; if it falls short, it isn’t.
The Discount Rate Is Something You Choose
The discount rate is an external benchmark. It reflects what it costs your organization to raise money, or what you’d give up by putting cash into this project instead of the next-best alternative. Three versions show up most often: the weighted average cost of capital, a company-set hurdle rate, and market-based reference rates like the IRS applicable federal rates.
WACC blends the cost of equity with the after-tax cost of debt, weighted by how much of each the company uses. Because interest is tax-deductible at the 21% federal corporate rate, the debt side of that blend comes in lower than the stated interest rate. A firm funded half by 10% equity and half by 5% debt does not land at a 7.5% WACC; the tax shield pulls the effective debt cost below 5% and the blended rate with it.
Many firms set their hurdle rate deliberately above WACC to absorb estimation error and project risk. Research from the Reserve Bank of Australia found companies commonly using hurdle rates of 12% to 15% while their estimated WACC sat around 6%. The cushion is intentional: it screens out projects that would barely clear the cost of capital and could underperform if conditions shift.
For certain debt transactions, the IRS publishes applicable federal rates each month under Section 1274(d) of the Internal Revenue Code, tied to yields on outstanding U.S. Treasury obligations and split into short-term, mid-term, and long-term tiers.1Office of the Law Revision Counsel. 26 USC 1274 – Determination of Issue Price in the Case of Certain Debt Instruments Issued for Property These serve as minimum benchmarks for imputed interest on loans and seller-financed transactions.2eCFR. 26 CFR 1.1274-4 – Test Rate
Whichever version you use, the discount rate answers one question: what is the minimum return this money needs to earn to justify tying it up here?
IRR Is Something the Project Tells You
IRR works in the opposite direction. In a standard present-value calculation you plug in a discount rate and solve for a dollar amount. IRR flips the equation: set net present value to zero and solve for the rate. The number that comes back is the annualized yield the project’s cash flows generate on their own, without reference to any outside benchmark.
If you invest $100,000 today and receive $30,000 a year for four years, there is exactly one discount rate that makes the present value of those four payments equal the $100,000 you put in. That rate is the IRR, and here it works out to roughly 7.7%. The figure doesn’t care what your WACC is, what Treasury yields are doing, or what your board considers acceptable. It’s purely a function of how much goes in, how much comes back, and when.
That self-contained quality is what makes IRR intuitive. A project with a 15% IRR feels more concrete than one with a $47,000 NPV because the percentage is easy to compare against other rates you already know. It’s also the reason IRR can mislead, as the later sections show.
How the Two Numbers Work Together
The standard investment test compares them directly. If the IRR exceeds the discount rate, the project adds value. If it falls below, the project destroys value. If they are exactly equal, the project breaks even on a present-value basis and there is no compelling financial reason to proceed.
Say a company’s WACC is 8% and an equipment upgrade produces an IRR of 12%. The four-point spread means the project earns more than the capital costs to fund it, and NPV comes back positive. If the IRR were 6%, the company would do better returning cash to shareholders or paying down debt, because the project can’t cover its own financing cost.
A useful way to picture the relationship: as you increase the discount rate, NPV falls, and the rate at which NPV hits zero is the IRR. Every discount rate below that point produces a positive NPV; every rate above it produces a negative one. The IRR marks the boundary between the two.
Where IRR Can Mislead You
Because IRR and the discount rate feed into the same decision, it’s tempting to treat them as interchangeable measures of the same thing. They aren’t, and three specific problems show why.
The Reinvestment Assumption
When you calculate NPV using a discount rate, the math implicitly assumes that cash flows received during the project’s life get reinvested at that discount rate. If your WACC is 8%, NPV assumes every intermediate payment earns 8% until the project ends. That’s usually realistic; most companies can find something that earns roughly their cost of capital.
IRR makes a bolder assumption. It assumes every intermediate cash flow gets reinvested at the IRR itself. A project with a 25% IRR assumes you can find 25% reinvestment opportunities for every dollar generated along the way. For a typical corporate project, that isn’t realistic. The higher the IRR, the more the metric overstates the project’s true economic value. This matters most on long-lived projects with large intermediate payments, where the two methods can diverge enough to flip a decision.
Multiple IRRs on the Same Project
IRR solves a polynomial equation, and polynomials can have more than one solution. A standard project with an upfront cost followed by positive returns has only one sign change in its cash flows and produces a single IRR. But projects where cash flows flip between positive and negative more than once can produce multiple IRRs, and none of them may be economically meaningful.
This shows up more often than people expect. Mining operations that require a large upfront investment, generate revenue for years, and then face significant closure and remediation costs have at least two sign changes. Staged developments needing a second major capital injection years after launch create the same pattern. When a spreadsheet returns an IRR for one of these projects, it’s returning whichever solution its algorithm landed on first, which may not be the right one. NPV doesn’t have this problem: plug in a discount rate, get one answer.
Conflicting Rankings Between Projects
IRR and NPV can rank mutually exclusive projects in opposite order, and this is where picking the wrong metric costs real money. Consider two options. Project A costs $50,000 and returns $75,000 in one year, an IRR of 50%. Project B costs $500,000 and returns $650,000 in one year, an IRR of 30%. IRR says pick A. But Project B generates $150,000 of profit against A’s $25,000. At any reasonable discount rate, B’s NPV is far larger. The percentage flatters A; the dollars favor B.
Timing mismatches produce similar conflicts. A project that returns most of its cash early shows a higher IRR than one with the same total cash flows spread over a longer period, even when the longer project builds more wealth at the company’s actual cost of capital. For decisions that come down to maximizing dollar value, NPV is the more reliable tool.
The Modified IRR
The modified internal rate of return was built to fix both the reinvestment assumption and the multiple-solution problem. Instead of letting the math assume reinvestment at the IRR, MIRR asks you to specify two rates: a finance rate for discounting negative cash flows to the present (typically the borrowing cost), and a reinvestment rate for compounding positive cash flows forward to the project’s end (typically the cost of capital).
Because MIRR collapses everything into a single outflow and a single terminal inflow, there is only one sign change and only one solution. And because the reinvestment rate is set at something realistic rather than the project’s own yield, the result is more conservative and usually closer to the truth. When MIRR and NPV both use the cost of capital as the reinvestment rate, they agree on project rankings.
Which Metric to Use When
IRR earns its place as a communication tool. Telling a board a project returns 14% against a 9% hurdle is immediately understandable. It works well for standalone go/no-go decisions on conventional projects, where cash flows follow the normal pattern of money out, then money in.
NPV is the stronger choice for most other situations: ranking mutually exclusive projects, evaluating irregular cash flow patterns, comparing projects of different lifespans, or any decision where precision matters more than a clean percentage. The profitability index, which divides NPV by the initial investment, gives a ratio-based alternative when capital is limited.
MIRR sits between the two. It produces a percentage like IRR but uses realistic reinvestment assumptions like NPV. Running all three often gives the clearest picture: NPV for the decision itself, MIRR for the realistic return, and IRR as a quick check that the ballpark is right.