CapTableClash

IRR, MOIC, and Payback Period: When Each One Lies to You

Three ways to score the same deal, and three different answers. Here's the arithmetic that connects them, why a dividend recap moves one number and barely touches another, and what each metric quietly hides.

Every private equity return conversation runs on three numbers. They measure the same pile of cash and they disagree constantly.

That disagreement is the interesting part. An interviewer asking about IRR and MOIC in the same breath is almost never testing the definitions — they're testing whether you know which one is being used to flatter the deal.

The three numbers, plainly

MOIC is multiple on invested capital: total cash returned divided by total cash invested. A $100mm equity check that comes back as $250mm is 2.5x. It has no concept of time in it at all.

IRR is the discount rate at which the deal's cash flows have a net present value of zero. Practically, it's the annualized compound return, and every dollar of timing shows up in it. The mechanics sit in the NPV vs. IRR guide.

Payback period is how long until the original investment has come back in cash. Nothing about profit, nothing about what happens afterward. Just the date the money is whole again.

Three definitions, thirty seconds. Now the part that actually gets scored.

The bridge between MOIC and IRR

For a deal with one cash flow out at the start and one back at exit — the standard simplified LBO case — the two are the same fact stated two ways:

IRR from a multiple and a hold periodIRR = MOIC^(1/n) − 1
2.0x over 5 years2^(1/5) − 1 = 14.9%
2.0x over 3 years2^(1/3) − 1 = 26.0%

Same money back. Eleven points of annualized return separating them, purely because one took two years longer.

Memorize a handful of these. They come up in interviews as sanity checks, and a candidate who can say “that's roughly a 25% IRR” without a calculator sounds like someone who has seen a return model before.

MOICHoldIRR
2.0x3 years~26%
2.0x5 years~15%
2.0x7 years~10%
2.5x5 years~20%
3.0x3 years~44%
3.0x5 years~25%
3.0x7 years~17%

If you'd rather not memorize a grid, the rule of 72 gets you into the neighborhood for anything near a double: 72 divided by the years it takes to double is roughly the IRR. Five years gives 14% against a true 14.9%. It understates as the hold gets shorter — three years gives 24% against a true 26% — so treat it as a floor, not an answer.

Where IRR lies

IRR is exquisitely sensitive to timing, which is exactly the feature and exactly the problem. Anything that moves cash earlier improves it, whether or not the deal got any better.

The dividend recap

A sponsor puts in $100mm and gets $200mm back at the end of year five. That's 2.0x and a 14.9% IRR.

Now run the same deal, except the company borrows a little in year one and pays the sponsor a $50mm dividend, and the exit at year five returns $150mm instead. Total cash back is still $200mm. Still 2.0x.

Year 0− $100mm
Year 1 dividend+ $50mm
Year 5 exit+ $150mm
MOIC2.0x — unchanged
IRR~21%

Six points of IRR appeared out of nothing but a calendar. The limited partners received identical total dollars.

The reinvestment assumption

IRR implicitly assumes every interim dollar returned gets reinvested at the same IRR until the end of the deal. For a 15% deal that's plausible. For a 60% deal it is fantasy, and it's why unusually high IRRs on early distributions should be read with suspicion rather than admiration.

Small dollars, big percentage

A $5mm equity check returned at 3x in eighteen months is a spectacular IRR and $10mm of profit. A $400mm check at 2x over five years is a mediocre IRR and $400mm of profit. A fund gets paid carry on the second one.

Weird cash flow patterns

IRR is the root of a polynomial. When the cash flows change sign more than once — an investment, distributions, then a follow-on capital call — there can be more than one mathematically valid IRR, or none at all. Rare in a clean LBO, common enough in real portfolios that the caveat is worth having.

Where MOIC lies

MOIC's blind spot is the one it announces openly: it has no clock. 3.0x is 3.0x whether it took four years or twelve, and those are a 32% return and a 10% return respectively.

The subtler issue is what “invested” and “returned” include. Gross MOIC is measured at the deal level, before management fees, expenses, and carried interest. Net MOIC is what a limited partner actually keeps. The gap between them is wide, and a number quoted without that label is doing some work.

MOIC also flatters a deal that has been partially written up but not sold. Marks are estimates. Cash is not.

Where payback period lies

Payback is the crudest of the three and the easiest to misread, because it stops counting at exactly the moment the returns start.

Consider a $100mm investment paying back $25mm a year. Payback lands in year four, which sounds fine. If the cash flows stop there, the deal returned 1.0x and earned nothing.

Two structural problems, both worth naming out loud:

  • It ignores everything after the payback date — which is where the entire profit of a good deal lives.
  • In its simple form it ignores the time value of money inside the payback window, treating a dollar in year four as equal to a dollar in year one. The discounted payback period fixes that one, and still ignores the first problem.

What payback is genuinely good for is risk, not return. It answers how long capital sits exposed before the downside is covered, which is why it survives as a screening test in capital budgeting and in credit-adjacent conversations rather than as a headline return metric.

What a fund actually reports

Deal-level MOIC has fund-level cousins, and mixing them up in a growth equity or PE interview is an easy tell:

  • DPI — distributions to paid-in capital. Realized cash returned to LPs divided by capital they've contributed. This is the number that can't be argued with.
  • TVPI — total value to paid-in. Distributions plus the current estimated value of what's still held, over paid-in capital. The unrealized portion is a mark.
  • RVPI — the residual, unrealized half of TVPI on its own.

A fund with a strong TVPI and a thin DPI hasn't returned much yet. Late in a fund's life, that gap is the whole conversation.

One more wrinkle worth knowing: many funds use a subscription credit facility to bridge deals before calling capital from LPs. That delays the start of the IRR clock and lifts reported net IRR without changing a single dollar of profit. It doesn't move MOIC at all — another reason the two get quoted side by side.

How it gets asked

What's the difference between IRR and MOIC?
MOIC is total cash out divided by total cash in, with no regard for time. IRR is the annualized rate that sets the deal's NPV to zero, so it's entirely driven by time. For a single investment and a single exit, IRR is just MOIC raised to the power of one over the hold period, minus one.
A deal returns 3.0x over five years. Roughly what IRR is that?
About 25%. The quick check is that 3x over three years is around 44% and over seven years is around 17%, so five years lands between them. Saying the approximate number confidently is better than reaching for a calculator.
Can a deal have a great IRR and a bad MOIC?
Yes, and it's common. A small check returned quickly — a fast flip, or an early dividend recap on a modest position — can produce a huge annualized return on very few dollars of profit. Funds care about both because carry is paid on realized dollars, not on percentages.
A sponsor does a dividend recap. What happens to IRR and MOIC?
IRR rises, sometimes sharply, because cash comes back earlier. MOIC is unchanged if total proceeds are the same — the recap just moves money forward in time. In practice the added leverage also raises risk and can reduce the eventual exit equity, which is the part worth saying out loud.
Why do LPs look at DPI instead of just IRR?
Because IRR and TVPI both depend on unrealized marks, and marks are estimates made by the manager. DPI is realized cash actually distributed. A fund can carry an impressive IRR for years without returning meaningful capital.
Which metric would you use to compare two deals?
Neither on its own. Pair them: MOIC tells you the size of the profit, IRR tells you how efficiently the capital was used over time, and the hold period reconciles the two. If the deals have different risk profiles or different check sizes, say that too — a 40% IRR on $5mm and a 20% IRR on $400mm are not the same achievement.

Mistakes that show up under a clock

  • Quoting IRR without the hold period. The number is meaningless without it, and the interviewer will ask.
  • Treating MOIC as though it accounts for time, or IRR as though it accounts for size. Each one is missing exactly what the other has.
  • Adding IRRs together, or averaging them across deals. IRRs don't aggregate that way; a pooled cash flow calculation does.
  • Forgetting that gross and net are different numbers. Fees and carry sit between them.
  • Calling a fast, small win a better outcome than a slow, large one purely because the percentage is higher.
  • Using payback period as a return metric. It's a risk-exposure screen, not a measure of profit.

The reason these three get asked together is that no single one of them describes a deal. A candidate who reaches for the pair — and names the hold period unprompted — is already answering the follow-up question.

Get the MOIC-to-IRR conversions automatic first. They're the fastest arithmetic on this page and the most likely to be asked cold.

Practice on CapTableClashBrowse the Concept Guides
Prepared for interview preparation purposes only. Not investment or career advice.