How a JV Equity Waterfall Actually Works

A joint venture equity waterfall decides who gets paid, in what order, when a deal distributes cash, and it's the single most common source of confusion in REPE interviews, mostly because most explanations skip straight to the promote split without building up the tiers that come before it. Here's the full structure, tier by tier, using the European (whole-fund) convention: proceeds distribute in strict tier order, one tier fully clearing before the next one opens.

Tier 1: Return of capital

Before anyone earns a return, the LP and GP get their own contributed capital back, pro rata by ownership percentage:

LP Capital = Total Equity × LP %
GP Capital = Total Equity × GP %

This tier isn't a "return" in any meaningful sense: it's just handing investors back the principal they put in. Nothing has been earned yet.

Tier 2: Preferred return

Next, both partners receive a preferred return, a minimum compounded return on their contributed capital, split the same pro rata way, before any promote (the GP's outsized share of profit) kicks in:

Total Preferred Return = Total Equity × ((1 + Pref Rate) ^ Hold Years − 1)

An 8% preferred return over a 5-year hold isn't 8% × 5 = 40% of capital; it compounds, the same way any other multi-year return does. This is the tier that protects the LP: no matter how the deal ultimately performs, the LP is contractually first in line (after return of capital) for this minimum return before the GP earns anything beyond its own pro rata share.

Tier 3: GP catch-up

This is the tier that trips up almost everyone the first time. The GP catch-up gives the GP 100% of distributions (cutting the LP out entirely, temporarily) until the GP's cumulative take (its Tier 2 pref plus this catch-up) reaches a target percentage of total profit distributed so far:

GP Catch-Up = (Catch-Up % × Total Preferred Return − GP Preferred Return) ÷ (1 − Catch-Up %)

The intuition: the preferred return tier paid the GP only its small pro rata share of the pref (say, a 5% GP stake gets 5% of the total pref), even though the GP is supposed to ultimately earn a much larger promoted share of total profit. The catch-up tier exists purely to true the GP's share back up to that target ratio before the deal moves into the tiers where profit actually gets split at the promote rate. Once the catch-up completes, the GP has received exactly its target percentage of total profit distributed through this point: not a dollar more, not a dollar less.

Tiers 4+: Escalating promote

Everything above a second (and often third) IRR hurdle splits at a promote ratio that gets richer for the GP as returns get better, the whole incentive-alignment mechanism of the structure:

Hurdle 2 Distribution = Total Equity × (1 + Hurdle 2 Rate) ^ Hold Years
Tier Pool = MAX(0, MIN(Total Proceeds, Hurdle 2 Distribution) − Prior Tier Watermark)
LP Share = Tier Pool × (1 − GP Split)
GP Share = Tier Pool × GP Split

A typical structure might split 80/20 between the first and second hurdle, then 70/30 above the second hurdle, then 60/40 (or steeper) above a third, each tier's pool clamped between the previous tier's watermark and the current hurdle, using exactly the same MAX(0, MIN(...)) clamping pattern that shows up in any tiered payout structure. If the deal never generates enough proceeds to reach a given hurdle, that tier's pool is simply zero: nothing above it ever triggers.

Why "watermark" is the right way to think about hurdle tiers

Each hurdle represents a cumulative dollar total that would have to be distributed for the pooled equity to reach that hurdle's IRR: not a return rate applied tier-by-tier. That's why the math above uses MIN(Total Proceeds, Hurdle Distribution): a tier's pool is capped at whatever's actually available, and floored at zero if a lower hurdle wasn't even reached. Thinking in terms of these cumulative watermarks, rather than trying to reason about "tier 3's return" in isolation, is what makes a multi-tier waterfall tractable to build without getting the boundary conditions wrong.

Deal-by-deal vs. whole-fund: always ask

Everything above describes a single-deal (or whole-fund, treated as one pool) waterfall. Some structures instead calculate the waterfall separately for each individual deal in a portfolio, which changes the math meaningfully: a GP can earn promote on an early winning deal under a deal-by-deal structure even if a later deal in the same fund loses money, something a whole-fund structure would net out first. Before building or discussing any waterfall, confirm which convention is in play: it's one of the first clarifying questions a real GP negotiation would ask, and it should be one of the first things you ask in an interview too.