Negative NPV but Positive IRR: What Does It Mean?
It looks like a contradiction — the project earns a positive return, yet the model says it destroys value. It isn't a contradiction at all. Here is exactly what is happening, why it is completely normal, and what the right decision is.
A project can have a positive IRR and a negative NPV at the same time when its internal rate of return is above 0% but below your discount rate (the hurdle rate or cost of capital). In plain terms: the project makes money, but not enough money to beat what you could earn elsewhere. The correct decision is to reject it — and both metrics actually agree on that.
The Short Answer
The internal rate of return (IRR) and net present value (NPV) are two views of the same cash flows, so people expect them to always tell the same story. When one looks "good" (a positive IRR) and the other looks "bad" (a negative NPV), it feels like the numbers are fighting each other.
They are not. The single fact that resolves the whole puzzle is this:
IRR is simply the discount rate at which NPV equals zero. So whether NPV is positive or negative depends entirely on how your chosen discount rate compares to the IRR — not on whether the IRR is above zero.
If your discount rate is higher than the IRR, the NPV is negative — even when the IRR itself is a healthy-looking positive number like 8%. A positive IRR only means the project returns more than 0%. It says nothing about whether that return beats your cost of capital.
What NPV and IRR Each Measure
To see why this happens, it helps to remember what each metric is actually calculating.
Net Present Value (NPV) discounts every future cash flow back to today using a discount rate you choose — usually your cost of capital or required rate of return — and subtracts the initial investment. It answers: "In today's money, how much value does this project create above my required return?" A positive NPV means value created; a negative NPV means value destroyed.
NPV = −Investment + CF1/(1+r)1 + CF2/(1+r)2 + … + CFn/(1+r)n
Internal Rate of Return (IRR) flips the question around. Instead of choosing a rate, it solves for the rate that would make NPV exactly zero. It answers: "What annual return does this project actually earn on the money invested?"
IRR is the value of r for which NPV = 0
That definition is the key to everything that follows. Because IRR is defined through NPV, the two are not independent opinions — they are the same equation read in two directions.
The Hidden Link Between NPV and IRR
For a normal ("conventional") project — an upfront outflow followed by a series of inflows — NPV falls steadily as the discount rate rises. Discount future cash more aggressively and it is worth less today, so NPV shrinks. Plot NPV against the discount rate and you get a downward-sloping line that crosses zero exactly once. That crossing point is the IRR.
Because NPV slides downward as the rate rises and crosses zero at the IRR, three simple rules always hold for a conventional project:
- Discount rate below IRR → NPV positive
- Discount rate equal to IRR → NPV zero
- Discount rate above IRR → NPV negative
Read that third rule again. A project with an IRR of 8% will show a negative NPV the moment you discount it at anything above 8% — say a 10% or 12% cost of capital. The IRR is still positive. The NPV is still negative. Both are correct. That is your entire scenario in one sentence.
Why a Positive IRR Can Still Give a Negative NPV
The word "positive" is doing a lot of quiet mischief here. "Positive IRR" only means the return is above zero percent. That is a very low bar. It tells you the project returns more cash than you put in on an undiscounted basis — nothing more.
But money has a cost. Your capital could be earning your required rate of return somewhere else — in another project, in the market, or simply paying down debt. That required return is your hurdle rate. A project is only worthwhile if it clears the hurdle.
So there are really three zones, not two:
| Situation | IRR | NPV at your hurdle rate | Verdict |
|---|---|---|---|
| Return beats your hurdle | IRR > hurdle rate | Positive | Accept |
| Return is positive but below your hurdle | 0% < IRR < hurdle | Negative | Reject |
| Project loses money outright | IRR < 0% | Negative (worse) | Reject |
The middle row is exactly the "negative NPV but positive IRR" case. The project genuinely earns a return — it just earns less than you require. Taking it would leave you worse off than deploying the same money at your hurdle rate.
A Full Worked Example
Numbers make this obvious. Suppose you can invest $10,000 today in a project that returns $2,500 per year for five years. Over the five years it pays back $12,500 in total — clearly more than the $10,000 you put in, so its return is positive.
Solving for the rate that sets NPV to zero gives an IRR of about 7.93%. Now suppose your cost of capital — your hurdle rate — is 12%. Let's discount every cash flow at 12%:
| Year | Cash flow | Discount factor at 12% | Present value |
|---|---|---|---|
| 0 | −$10,000 | 1.0000 | −$10,000.00 |
| 1 | $2,500 | 0.8929 | $2,232.14 |
| 2 | $2,500 | 0.7972 | $1,992.98 |
| 3 | $2,500 | 0.7118 | $1,779.45 |
| 4 | $2,500 | 0.6355 | $1,588.80 |
| 5 | $2,500 | 0.5674 | $1,418.57 |
| NPV at 12% | −$988.06 | ||
There it is in black and white. The IRR is a positive 7.93%, yet the NPV at a 12% hurdle is a negative $988. Nothing is broken. The project earns 7.93%, but you demanded 12%, so discounting at 12% shrinks the inflows below the $10,000 you paid. The $988 shortfall is the value you would lose by accepting a 7.93% project when 12% was available elsewhere.
Watch what happens if the hurdle rate were only 6% instead of 12%. Discounting the same cash flows at 6% gives an NPV of roughly +$531 — now positive. Identical project, identical cash flows; only the hurdle changed. NPV's sign is a verdict on the project relative to your required return, not an absolute property of the cash flows.
The NPV Profile Chart
The clearest way to see all of this at once is the NPV profile — a plot of the project's NPV against a range of discount rates.
Everything about your scenario is visible in one picture. The curve is positive on the left (low rates), crosses zero at the IRR, and turns negative on the right (high rates). Your hurdle rate of 12% lands in the red zone, so NPV is negative — while the IRR, sitting comfortably above 0%, is positive.
What a Positive IRR Really Tells You
This is where most explanations go wrong, so let's be precise. A positive IRR tells you the project earns a return greater than zero. That is not the same as "the project is a good investment." The bar that actually matters is not zero — it is your cost of capital.
Think of it like a savings decision. A friend offers to pay you 3% on a loan. That is a positive return. But if your bank pays 5% risk-free, lending to your friend at 3% makes you poorer than the alternative, despite the positive rate. The IRR is the 3%; your hurdle rate is the 5%; the negative NPV is the wealth you forgo by choosing the worse option.
Treating any positive IRR as a green light. "The project returns 8%, so it must be worth doing" ignores the only comparison that counts: 8% versus what you require. Below the hurdle, a positive IRR is still a value-destroying decision.
Do the Two Metrics Actually Conflict?
Here is the reassuring part. In this specific situation — a single, conventional project — NPV and IRR do not disagree at all. People only think they do because they misread "positive IRR" as "accept."
Line the two decision rules up side by side:
| Rule | Says | Result in our example | Decision |
|---|---|---|---|
| NPV rule | Accept if NPV > 0 | NPV = −$988 (< 0) | Reject |
| IRR rule | Accept if IRR > hurdle rate | 7.93% < 12% | Reject |
Both say reject. The mistake was never in the maths — it was comparing the IRR to zero instead of to the hurdle rate. Once you compare IRR to your required return, the "conflict" vanishes and the two metrics march in perfect step, exactly as theory promises for a conventional project.
When IRR Genuinely Misleads
To be a complete analyst, it is worth knowing that IRR can mislead — just not in the case above. There are three well-known situations where you should lean on NPV rather than IRR:
- Non-conventional cash flows. If the cash flows change sign more than once (for example, a big cleanup cost at the end of a mining project), there can be multiple IRRs or none at all. NPV stays well-defined; IRR becomes ambiguous.
- Mutually exclusive projects of different size. A small project can have a higher IRR while a larger project creates more total value. IRR is a percentage and ignores scale; NPV measures dollars of value and should win.
- Different timing of cash flows. IRR implicitly assumes you can reinvest interim cash flows at the IRR itself, which is often unrealistic. NPV assumes reinvestment at the discount rate, which is more defensible.
When NPV and IRR truly disagree — usually with mutually exclusive projects or odd cash-flow patterns — trust NPV. It measures value in currency, not percentages, and it never suffers from the multiple-solution problem. Our guide on NPV vs IRR covers these cases in depth.
But note carefully: the plain "negative NPV, positive IRR" case is not one of these. It is not a flaw in either metric — it is simply a project whose return falls short of your hurdle.
Common Misconceptions
- "A positive IRR means the project is profitable enough." No — it means the return is above 0%. Profitability that matters is measured against your cost of capital.
- "NPV and IRR are contradicting each other, so one must be wrong." Neither is wrong. They are the same equation; the IRR is just the rate where NPV hits zero.
- "If I lower my discount rate, the project becomes genuinely better." Lowering the rate turns the NPV positive on paper, but only lower it if your true cost of capital really is lower. Manufacturing a low rate to force a positive NPV is self-deception.
- "A negative NPV project never earns anything." It can earn a positive return — it just earns less than you require. Negative NPV measures a shortfall against the hurdle, not an outright loss.
Key Takeaways
- IRR is the discount rate at which NPV equals zero — so the two are linked by definition, never truly independent.
- NPV is negative whenever your discount rate is above the IRR, even if the IRR is a positive number.
- "Negative NPV but positive IRR" means 0% < IRR < hurdle rate: the project earns a return, just not enough to clear your cost of capital.
- The right decision is to reject — and both the NPV rule and the correctly applied IRR rule agree on it.
- Always compare IRR to your hurdle rate, never to zero.
- When NPV and IRR genuinely conflict (scale, timing, non-conventional flows), follow NPV.
Frequently Asked Questions
Can a project have a positive IRR and a negative NPV at the same time?
Yes, and it is common. It happens whenever the IRR is above 0% but below your discount rate (hurdle rate). The project earns a positive return, but that return is smaller than the return you require, so discounting the cash flows at your higher rate produces a negative NPV.
Does a negative NPV mean the project loses money?
Not necessarily. A negative NPV means the project fails to clear your required rate of return. It may still generate a positive cash return overall — it simply earns less than your cost of capital, so it destroys value compared with the alternative use of your money.
Should I accept a project with a positive IRR but a negative NPV?
No. Reject it. A negative NPV means the IRR is below your hurdle rate, so the project does not earn enough to justify the capital. Both the NPV rule and the correctly applied IRR rule (accept only if IRR is above the hurdle rate) tell you to pass.
Why do NPV and IRR seem to disagree here?
They only appear to disagree because "positive IRR" is misread as "good." The IRR rule is to accept when IRR exceeds the hurdle rate, not when IRR exceeds zero. Compared correctly against the hurdle rate, IRR and NPV give the same reject decision for a conventional project.
What is the relationship between IRR and the discount rate?
The IRR is the specific discount rate that makes a project's NPV equal to zero. If your chosen discount rate is below the IRR, NPV is positive; if it is above the IRR, NPV is negative; if it equals the IRR, NPV is exactly zero.
Can NPV be positive while IRR is negative?
For a normal project with an upfront cost followed by inflows, no. A negative IRR means the project returns less than 0%, so its NPV is negative at any non-negative discount rate. Positive-NPV-with-negative-IRR only appears with unusual, non-conventional cash-flow patterns.
Which should I trust when NPV and IRR conflict?
Trust NPV. It measures value in actual currency, works with any cash-flow pattern, and never produces multiple answers. IRR can mislead with mutually exclusive projects of different size, different timing, or cash flows that change sign more than once.
What is a hurdle rate?
The hurdle rate is the minimum return you require from a project, usually set to your cost of capital or WACC. A project is worth doing only if its IRR exceeds the hurdle rate — equivalently, only if its NPV at that rate is positive.
Related reading on this site:
- NPV vs IRR — a full comparison and when each metric wins.
- What is Discounted Cash Flow (DCF)? — where the discount rate comes from.
- NPV Calculator — test your own cash flows and discount rate instantly.