Diagnosing Common IRR Exam Traps

Why IRR Problems Are Easy to Misread

An internal rate of return (IRR) calculation can go wrong even when the arithmetic is straightforward. Common causes include mixing investor and project signs, using a quarterly rate as though it were annual, accepting one solver output without checking for other roots, and treating a higher IRR as automatically better than a higher net present value.

A reliable solution begins by writing a dated cash-flow timeline and stating the viewpoint and rate period. Then write the NPV equation, solve it, and check the result in that equation. These checks are especially important when cash flows change sign more than once or when projects are mutually exclusive.

Definition: For cash flows $C_0,C_1,\ldots,C_n$ at equally spaced times $0,1,\ldots,n$, measured from one consistent viewpoint, the IRR $r$ is a rate per cash-flow period satisfying $$ 0=\operatorname{NPV}(r)=\sum_{t=0}^{n}\frac{C_t}{(1+r)^t}. $$ This equation requires $r>-1$. An IRR may be absent or there may be more than one; a numerical solver’s output alone does not establish that a root is unique.

Start With Signs and Periods

For a project viewed by its investor, an investment made is normally negative and a receipt is positive. The signs must be consistent with that viewpoint throughout the timeline. Reversing every sign does not change the IRR, since it simply multiplies the entire NPV equation by $-1$. Reversing only some signs changes the equation and can change, remove, or create roots.

The rate period must also match the spacing of the cash flows. If the cash flows are quarterly, the rate $r$ in the equation is a rate per quarter. For a rate $i$ effective annually and $m$ equal periods per year, the equivalent rate per period $q$ satisfies $1+i=(1+q)^m$. Thus $q=(1+i)^{1/m}-1$; in general, dividing an effective annual rate by $m$ is not the correct conversion.

Worked Example: Detecting a Sign Error and Checking a Solver Result

An investor pays $1,000 at time 0, receives $600 one year later, and receives $550 two years later. Determine the annual IRR, and explain why entering the initial outflow as a positive amount while leaving the receipts positive is not a harmless sign convention.

Method. Use the investor’s viewpoint: the initial payment is negative and both receipts are positive. The cash flows are $(-1000,600,550)$, with annual periods. Their NPV equation is

-1000+\frac{600}{1+r}+\frac{550}{(1+r)^2}=0.

Check a candidate. At $r=0.10$, the time-1 discount factor is $1/1.10$, and the time-2 discount factor is $1/1.10^2=1/1.21$. Therefore,

-1000+\frac{600}{1.10}+\frac{550}{1.21} =-1000+545.4545\ldots+454.5454\ldots=0.

So 10% per year is a root. To see whether another rate greater than $-1$ is possible, set $x=1+r$, so $x>0$, and multiply the equation by $x^2$:

-1000x^2+600x+550=0 \quad\Longleftrightarrow\quad 1000x^2-600x-550=0.

The discriminant is $600^2+4(1000)(550)=2{,}560{,}000$, whose square root is 1,600. The roots are $x=(600\pm1600)/2000$, or $x=1.1$ and $x=-0.5$. Since $x=1+r$ must be positive, only $x=1.1$ is admissible, giving $r=0.10$. The IRR is 10% per year.

If a solver is given $(1000,600,550)$, all three entries are positive, so the NPV cannot equal zero for any $r>-1$. That is not a full sign reversal: it incorrectly changes only the initial flow. Entering $(1000,-600,-550)$, by contrast, reverses every sign and preserves the same root. Common mistake avoided: checking that the signs match the chosen viewpoint, rather than assuming any sign change is harmless.

Match the Rate Period to the Timeline

An IRR is quoted per the period used in its equation. If a solver uses quarterly intervals, its output is a quarterly rate, even when the investment description gives the term in years. To express that rate as an effective annual rate, compound four quarterly periods. To convert an annual effective rate to a quarterly rate, take a fourth root. Keep the conversion separate from the original IRR equation so the units remain visible.

Worked Example: Converting a Quarterly IRR to an Annual Rate

A hypothetical investment costs $1,000 now and pays $1,210 exactly two years later. There are no other cash flows. Find the quarterly IRR and its equivalent effective annual rate. Assume equal three-month periods and a constant rate over the term.

Method. Two years contain eight quarters. Let $q$ be the IRR per quarter. The cash-flow equation is

-1000+\frac{1210}{(1+q)^8}=0.

Rearranging gives $(1+q)^8=1210/1000=1.21$, so $q=1.21^{1/8}-1\approx0.02411$. The quarterly IRR is approximately 2.411%.

There are four quarters in one year, so the equivalent effective annual rate is

(1+q)^4-1 =\left(1.21^{1/8}\right)^4-1 =1.21^{1/2}-1 =1.10-1 =0.10.

The effective annual rate is 10.00%. As a check, two annual periods at 10% accumulate $1,000 to $1000(1.10)^2=\$1,210$, matching the payment. Round the quoted rates to three decimal places for the quarterly rate and two decimal places for the annual rate. Common mistake avoided: calling 2.411% an annual IRR or dividing 10% by four to obtain the quarterly rate.

Do Not Stop at the First Solver Root

When all outflows occur before all inflows, an NPV profile often has a single economically relevant root. With cash flows that change sign more than once, however, the equation can have multiple roots. A solver is a numerical method that searches for a root near a starting value; its reported solution need not reveal every root. Check each candidate in the original NPV equation, and use an algebraic transformation or a search over the relevant rate range when possible.

Worked Example: Verifying Two Positive IRRs

A hypothetical project has annual cash flows $(-100,230,-132)$ at times 0, 1, and 2. Find all IRRs greater than zero and explain how a solver result can mislead.

Method. The NPV equation, with $x=1+r$, is

-100+\frac{230}{1+r}-\frac{132}{(1+r)^2}=0, \qquad x=1+r>1.

Multiplying by $x^2$ and changing the sign of both sides gives

$$ 100x^2-230x+132=0 =(10x-11)(10x-12). $$

Thus $x=1.1$ or $x=1.2$, giving $r=10\%$ or $r=20\%$. Both are positive and both satisfy the original NPV equation. At 10%, the NPV is $-100+230/1.1-132/1.21=-100+209.0909\ldots-109.0909\ldots=0$. At 20%, it is $-100+230/1.2-132/1.44=-100+191.6666\ldots-91.6666\ldots=0$.

A solver initialized near 10% may return 10%, while a different starting value may return 20%. Neither result is necessarily a numerical failure: each is a valid root, and the cash flows have more than one sign change. The problem does not have a unique IRR. Common mistake avoided: reporting the first root a solver finds as though it were the only one.

IRR Is Not a Guaranteed Reinvestment Rate

The IRR is defined by the NPV equation. It does not, by itself, promise that interim receipts can actually be reinvested at that rate. One way to understand the distinction is to multiply the IRR equation by $(1+r)^n$:

\sum_{t=0}^{n}\frac{C_t}{(1+r)^t}=0 \quad\Longleftrightarrow\quad \sum_{t=0}^{n}C_t(1+r)^{n-t}=0.

The second expression says the signed cash flows sum to zero when accumulated to time $n$ at the rate $r$. It is an algebraic restatement of the same equation. It does not describe the actual reinvestment opportunities available for receipts between dates. A claim that interim receipts will earn $r$ requires an additional assumption about what happens to those receipts. Keep the mathematical definition separate from that economic interpretation.

When IRR and NPV Rankings Disagree

For mutually exclusive projects, only one can be selected. A larger IRR does not necessarily mean a larger dollar value added at the required rate of return. NPV measures discounted value at the stated valuation rate; IRR reports the rate that makes a project’s own NPV zero. When the rankings conflict, use the decision rule specified by the problem. If the objective is to maximize value at a given required rate, compare NPVs at that rate rather than choosing by IRR alone.

Worked Example: Resolving an IRR–NPV Ranking Conflict

Two hypothetical, mutually exclusive projects each last two years. Project A has cash flows $(-1000,700,500)$; Project B has cash flows $(-1000,100,1150)$, with the entries at times 0, 1, and 2 in dollars. The required annual return is 8%. Determine the IRR ranking and the NPV ranking, and state which project maximizes NPV.

Method. For Project A, set $x=1+r$. Its IRR equation becomes $1000x^2-700x-500=0$. The positive solution is

x=\frac{700+\sqrt{2{,}490{,}000}}{2000} \approx1.13899, \qquad r\approx13.899\%.

For Project B, the equation becomes $1000x^2-100x-1150=0$. Its positive solution is

x=\frac{100+\sqrt{4{,}610{,}000}}{2000} \approx1.12355, \qquad r\approx12.355\%.

Project A has the higher IRR. Now calculate both NPVs at the required rate of 8%. Since $1.08^2=1.1664$,

\begin{aligned} \operatorname{NPV}_A(0.08) &=-1000+\frac{700}{1.08}+\frac{500}{1.1664} \approx \$76.82,\\ \operatorname{NPV}_B(0.08) &=-1000+\frac{100}{1.08}+\frac{1150}{1.1664} \approx \$78.53. \end{aligned}

Project B has the higher NPV at 8%, so it is preferred if the stated objective is to maximize NPV at that required rate. The IRR ranking is not an error; it answers a different question. Project B also has an interim receipt, but its IRR does not guarantee that the $100 receipt actually earns 12.355% after it is received. Round the IRRs to three decimal places and the NPVs to the nearest cent. Common mistake avoided: selecting Project A solely because its IRR is higher, without applying the required-rate NPV decision rule.

A Practical Diagnostic Sequence

A disciplined sequence makes the calculations easier to audit and helps distinguish a genuine multiple-root issue from a sign or unit error.

Set the viewpoint.
Mark payments and receipts consistently; do not change the sign of only one flow to fit a solver convention.

Label the dates and rate period.
Write the number of periods explicitly, including any conversion needed to report an annual effective rate.

Write the NPV equation.
Use the actual cash-flow amounts and dates before relying on a numerical solution.

Check root behavior.
Look for multiple sign changes and test whether the equation can have more than one admissible root.

Verify and interpret.
Substitute each reported IRR into NPV, then apply the stated decision rule. Do not infer a guaranteed reinvestment rate.

Key takeaway: An IRR is a root of a correctly signed, correctly timed NPV equation. Verify the root, its units, and whether it is unique. For mutually exclusive projects evaluated against a required rate, compare their NPVs when the objective is value maximization.

Check Your Understanding

For each problem, write the cash-flow equation before calculating, and check that the answer has the requested rate period.

A project requires $500 now and pays $600 one year later. From the investor’s viewpoint, what is its annual IRR? (Answer: 20% per year)
An investment pays $2,420 after eight quarters for an initial outlay of $2,000. Find the quarterly IRR and the equivalent effective annual rate. (Answer: approximately 2.411% per quarter and 10.00% effective annually)
Annual cash flows are $(-200,460,-264)$ at times 0, 1, and 2. Find all positive IRRs. (Answer: 10% and 20%)
At an 8% required annual rate, Project A has cash flows $(-300,390)$ and Project B has cash flows $(-1200,1440)$, at times 0 and 1. They are mutually exclusive. Which has the higher IRR, which has the higher NPV, and which maximizes NPV? (Answer: A has the higher IRR, 30% versus 20%; B has the higher NPV, approximately $133.33 versus $61.11, so B maximizes NPV.)

What You'll Learn