Chapter 2 of 5 · 11 min

The J-curve

Money goes out for years before any comes back. How deep the hole gets, and when the LP is back to even.

By the end of this chapter you can

Run an LP's cash flows year by year: calls, fees and distributions
Find the trough and the year it happens
Find the year the LP is back to even
Work out the distributions needed to break even by a target year

The intuition

Planting an orchard. For the first few years you buy saplings, pay for water and fencing, and harvest nothing. Only when the trees mature does money start to come back, and it takes several good harvests before you have recovered what you spent. Plot your bank balance over time and it dips, bottoms out, and climbs: a J.

An LP in a buyout fund sees the same shape. The GP calls capital to buy companies over the first few years and charges fees from day one, but companies bought in year one are not sold until year five or later. So the LP's cumulative cash position falls first and rises only when distributions arrive.

The key idea

Each year: net cash = distributions − capital calls − fees. Keep a running total. The lowest point is the trough; the first year the total is back to zero is breakeven.

Why it works

The conventions here: a ten-year fund. Capital is called in four equal instalments in years 1–4. The fee is a flat share of the commitment every year for all ten years, called on top. Distributions are a flat amount each year from a start year. No discounting.
Why it dips. Calls and fees go out; nothing comes back until the first exits. Fees are the part with nothing to show for it yet.
Why it climbs. After year 4 only the fee goes out, so once distributions start, each year adds to the total.
Two numbers describe the curve: how deep the trough is, and how long until breakeven. Neither is the fund's final return. A fund that holds longer and sells for more can break even later and still return more.
Why a young fund shows a negative IRR. Fees have been paid and deals are held near cost, so early IRRs are negative almost by design. It is a question of timing, not a verdict.
What flattens it: lower early fees, earlier exits or dividend recaps, and buying into a fund part-way through on the secondary market, so someone else sat through the trough.

$1,000M committed; 25% called each year in years 1–4; a 2% fee every year; $300M a year distributed from year 5

Years 1–4: −250 − 20 each year	−1,080 by year 4, the trough (108%)
Year 5: −20 + 300	−800
Year 6	−520
Year 7	−240
Year 8	+40, back to even
Needed a year to be even by year 7: (1,000 + 20 × 7) ÷ 3	380

At half the fee the trough is −1,040: 40 shallower.

The formulas

Capital call = commitment ÷ 4, in years 1–4

Four equal instalments.

Fee = fee rate × commitment, every year

Paid whether or not money is invested.

Net cash in a year = distributions − call − fee

What the LP gains or loses that year.

Cumulative cash = running sum of net cash

The J-curve itself.

Distribution needed by year T = (commitment + fee × T) ÷ (T − start year + 1)

Everything paid out by then, spread over the years of distributions.

Worked example

Run the years one at a time, keeping a running total. The trough is the lowest the total goes.

Drawing the numbers…

See it move

Same commitment. Change the fee, when distributions start and how large they are.

Drawing the numbers…

Try this

Raise the fee. The whole curve shifts down: the trough deepens and breakeven comes no sooner.
Raise the distributions. The climb steepens, breakeven comes sooner or stays put, and the trough is never deeper.
Start distributions a year later. Another year of fees with nothing coming back: the trough deepens and breakeven comes no sooner.
Compare the two trough bars. Halving the fee always makes the hole shallower, which is why LPs fight hardest over fees in a fund's early years.

Run it backwards

Same fund, reversed: how much must it distribute each year for the LP to be back to even by a target year?

Drawing the numbers…

By the target year all the capital has been called, and a fee has been paid every year. That total is what the distributions must add up to. Divide by the number of distribution years up to and including the target.

It turns a vague question, 'is this fund on track?', into one number to hold the GP to.

Traps

Leaving the fee out of the trough.

The fee is called every year on the whole commitment, on top of the capital. The trough is deeper than the capital called.

Stopping the fee when the capital calls stop.

Here it runs for all ten years. After year 4 it is the only cash going out.

Reading a late breakeven as a bad fund.

Breakeven is about timing. A fund that holds longer and sells for more can break even later and return more.

Counting distribution years from year 1.

Count from the start year to the target year, including both.

Treating an early negative IRR as poor performance.

Fees paid and deals held at cost make it negative by design. Worry when it persists past year five to seven, or comes from write-downs.

Say it in the interview

The interviewer asks

“What is the J-curve, and why does it matter to an LP?”

Say yours out loud first, then compare.

Check yourself

4 fresh questions, with new numbers. Answer each one correctly to finish the chapter. Get one wrong and you will see the full working, then you can try it again with new numbers.

Answers within 1% are marked right. Type the number; $, %, x and M are fine. First tries count toward Learned: the topic is Learned once every chapter is done and 75% of first tries were right.

0 of 4

Drawing your questions…

Remember

Net cash = distributions − calls − fees; keep a running total.
Trough = the lowest total; breakeven = the first year back at zero.
Fees deepen the trough; earlier, bigger distributions shorten it.
A young fund's negative IRR is the J-curve, not a verdict.