All Insights

Fund Terms

Distribution Waterfalls, Carry, and Clawbacks: What LPs Actually Get Paid

Headline terms tell you the shape of the deal. The waterfall tells you what you actually get paid. A tier-by-tier guide with worked examples.

Matthew S. · Allocator Desk · October 2026 · 15 min read

Every limited partnership agreement contains a section that decides who gets paid, in what order, and when. Most LPs read it once during legal review, sign off on the headline terms (8% preferred return, 20% carried interest, full catch-up), and then never look at it again until a distribution notice arrives with a number nobody on the desk can recreate.

That is a mistake, and it gets expensive. The distribution waterfall is the place where a fund's economics actually live. Two funds with identical "2 and 20" headlines can deliver meaningfully different net outcomes depending on how the waterfall is built, when carry is paid, what counts as a contribution, and whether the clawback has teeth. This guide walks through the mechanics tier by tier, works two numerical examples from start to finish, and then lays out how an allocator desk can monitor waterfall risk across a portfolio rather than discovering it in year eleven.

If you want the broader context on headline economics first, start with our guide to private equity fund fees and terms. This piece goes one level deeper, into the plumbing.

What a distribution waterfall actually does

When a fund sells a company, receives a dividend recap, or collects interest on a credit position, the cash does not go straight to the LPs. It runs through a sequence of tiers defined in the LPA. Each tier must be satisfied before the next one receives anything, which is why the structure is called a waterfall: cash fills the top basin, overflows into the next, and so on.

The standard private equity waterfall has four tiers:

1. Return of capital. LPs receive distributions until they have recovered their contributed capital. Depending on the LPA, "contributed capital" can mean capital for realized investments only, or all capital called to date including fees and expenses. That definitional choice is one of the most important in the document. 2. Preferred return. LPs then receive distributions until they have earned a specified return on their contributed capital, most commonly 8% per year, compounded. This is often called the hurdle. 3. GP catch-up. The GP receives a large share (frequently 100%, sometimes 80% or 50%) of the next distributions until it has "caught up" to its full carried interest percentage of the profits distributed so far. 4. Carried interest split. All remaining distributions are split, typically 80% to LPs and 20% to the GP.

On paper this looks mechanical. In practice, every one of those tiers has definitional choices that shift money between the LP and the GP, and every one of them interacts with timing.

Whole-fund versus deal-by-deal

The single biggest structural question is whether the waterfall runs at the level of the whole fund or deal by deal.

Whole-fund waterfalls (often called European waterfalls) require the fund to return all contributed capital, plus the preferred return on that capital, before the GP receives any carried interest. The GP waits. LPs are protected because carry is only paid once the fund as a whole has demonstrably made money.

Deal-by-deal waterfalls (often called American waterfalls) calculate carry on each realized investment separately. When a single deal exits at a gain, the GP can take carry on that deal even if other investments in the fund are underwater. Most modern deal-by-deal structures include protective features, such as requiring return of capital for all realized deals plus write-downs on unrealized ones, but the core feature remains: the GP gets paid earlier.

Earlier payment is not just a timing nuisance. It creates the possibility that the GP is overpaid relative to the fund's final outcome, which is the entire reason clawback provisions exist. The Institutional Limited Partners Association has long treated the whole-of-fund approach as the LP-preferred standard in its Principles, and its Model LPA was published in both whole-of-fund and deal-by-deal versions so that LPs negotiating either structure have a reference point.

Neither structure is automatically wrong. Deal-by-deal is common in US buyout and real estate funds and can be reasonable when paired with strong clawback mechanics, escrow, and conservative interim tests. But the LP should know which one it signed, and the monitoring burden is very different.

Worked example 1: a whole-fund waterfall from start to finish

Assume a simplified fund where LPs have contributed $100 million in total (including fees and expenses) and the GP commitment is ignored for clarity. The terms are an 8% compounded preferred return, a 100% GP catch-up, and an 80/20 split thereafter. Over the life of the fund, total distributable proceeds come to $180 million, and the accrued preferred return on LP capital, given the timing of calls and distributions, works out to $30 million.

Tier 1: return of capital. The first $100 million goes to LPs. Remaining: $80 million.

Tier 2: preferred return. The next $30 million goes to LPs. Remaining: $50 million. LPs have now received $130 million.

Tier 3: catch-up. The GP is entitled to 20% of all profits distributed so far. Profits distributed to date are the $30 million of preferred return. The GP needs an amount X such that X divided by (30 + X) equals 20%. Solving gives X = $7.5 million. With a 100% catch-up, the GP receives the next $7.5 million. Remaining: $42.5 million.

Tier 4: 80/20 split. LPs receive 80% of $42.5 million, or $34 million. The GP receives $8.5 million.

Totals. LPs receive $100M + $30M + $34M = $164 million. The GP receives $7.5M + $8.5M = $16 million. Total profit is $80 million, and the GP's $16 million is exactly 20% of it.

That last check is the whole point of a full catch-up: once the fund clears the hurdle by enough, the preferred return stops being an economic benefit to the LP and becomes purely a timing feature. The LP still gets paid first, but the GP ends up with a full 20% of profits. A preferred return with a full catch-up protects LPs in mediocre outcomes, not in good ones.

Now change one variable. If total proceeds were only $135 million, the waterfall stops partway through the catch-up: $100 million returns capital, $30 million pays the pref, and the remaining $5 million goes to the GP in catch-up. The GP receives $5 million on $35 million of profit, about 14%. This is where the hurdle earns its keep.

Worked example 2: deal-by-deal carry and the clawback

Now take a deal-by-deal fund with the same headline terms and two investments of $50 million each, again ignoring fees for simplicity.

Year 3. Deal A exits for $90 million, a $40 million gain. Assume the pref on Deal A's capital is cleared and the catch-up is full, so the GP takes 20% of the $40 million profit as carry: $8 million. LPs receive $82 million.

Year 6. Deal B is sold in distress for $10 million, a $40 million loss. There is no carry on Deal B, so LPs receive all $10 million.

Fund-level outcome. Total invested: $100 million. Total proceeds: $100 million. Total profit: zero. Under a whole-fund waterfall, the GP would have earned nothing. Under the deal-by-deal waterfall, the GP has already received $8 million, and LPs have received only $92 million back on $100 million invested.

The clawback is the provision that is supposed to fix this. At the end of the fund (or at specified interim points), the GP must return any carry received in excess of what it would have earned on an aggregate basis. Here, that is $8 million.

The problem is that clawbacks are rarely as clean as that. Most LPAs limit the clawback to the after-tax amount the GP received, on the logic that the GP should not have to repay taxes it already paid. If the agreement assumes a 40% tax rate, the GP repays only $4.8 million. LPs end the fund at $96.8 million on $100 million invested, and the gap is permanent.

There are other common frictions:

• **Timing.** Many clawbacks are only tested at the end of the fund's term, which can be twelve or more years after the carry was paid. Money that should have been returned in year six comes back, if at all, in year thirteen, without interest.
• **Individual versus joint liability.** Carry is usually distributed out to individual partners of the GP. If the clawback guarantee is several (each partner liable only for their share) rather than joint and several, a departed partner who cannot or will not pay leaves a hole that nobody else fills.
• **Creditworthiness.** A guarantee from a GP entity with no assets is a promise, not a protection. LPs should know who actually stands behind the obligation.

The tools that make clawbacks real

Because clawbacks are hard to enforce after the fact, LPs typically negotiate protections that work before the money leaves the fund:

Carry escrow or holdback. A portion of each carry distribution (often 20% to 30%) is held back in escrow and only released once the fund has passed a coverage test or reached final liquidation. This shrinks the amount that could ever need to be clawed back.

Interim clawback tests. Rather than waiting until the end of the fund, the LPA tests periodically (for example, annually after the investment period) whether the GP has been overpaid on a cumulative basis, and requires repayment or suspends further carry if so.

NAV coverage tests. Before carry can be distributed on a realized deal, the remaining unrealized portfolio must be valued at some multiple of remaining invested capital (125% is a frequently cited level). This guards against the GP harvesting carry on early winners while the rest of the portfolio is quietly impaired. The test is only as good as the marks, which is why it connects directly to how LPs interrogate NAV marks.

Joint and several guarantees. The individual carry recipients guarantee the clawback jointly, so the obligation follows the money.

Gross clawback. Less common and harder to negotiate, a gross clawback removes the tax offset, requiring repayment of the full pre-tax amount.

None of these are exotic. They are routine negotiating points, and the right combination depends on the structure. A whole-fund waterfall with a strong hurdle needs less of this machinery. A deal-by-deal waterfall without escrow or interim testing should prompt a hard look.

The definitions that quietly move money

Beyond structure, a handful of definitions in the waterfall section determine more of the outcome than the headline numbers. These are worth flagging during legal review and logging in a format the desk can find later, ideally alongside any related side letter provisions.

What capital gets returned first. Is it capital for realized deals only, or all contributed capital including management fees and fund expenses? Returning fees and expenses before carry is meaningfully more LP-favorable, because otherwise the GP can earn carry on profits that are only profits before fees.

How the preferred return compounds. Annual compounding versus simple interest makes a material difference over long holding periods. So does whether the pref accrues from the date capital is called or from the date it is actually invested.

The effect of subscription lines. When a fund borrows on a credit facility and calls LP capital months later, the preferred return clock usually starts only when LP capital arrives. That shortens the measured holding period, raises IRR, and makes the hurdle easier to clear, sometimes by a lot. We cover this in more detail in our piece on subscription lines and NAV facilities. Some LPs negotiate for the pref to accrue from the date the fund draws on the facility.

Recycling and reinvestment. If the fund can recycle early distributions back into new deals, the waterfall needs to handle those amounts consistently. Recallable distributions can make it look as though capital has been returned when it may be called again.

Treatment of fee offsets. Transaction, monitoring, and director fees charged to portfolio companies are often offset against management fees. Whether the offset is 100% or something less, and whether it flows through the waterfall correctly, affects net returns.

Catch-up percentage. A 100% catch-up gets the GP to its full 20% fastest. An 80% or 50% catch-up slows that down and leaves more value with LPs in the band just above the hurdle.

Distributions in kind. When a fund distributes public shares rather than cash, the value assigned to those shares for waterfall purposes matters. A valuation based on a single day's closing price can diverge sharply from what an LP actually realizes on sale.

GP commitment and the waterfall

The GP's own commitment usually participates in the fund alongside LPs, but typically without paying management fees or carry on its own capital. That is normal and appropriate. What deserves attention is how the GP commitment is funded. If it is satisfied through management fee waivers (the GP forgoes fees in exchange for a deemed contribution that participates in profits), the alignment is weaker than a cash commitment, because the GP's "capital" is only at risk if the fund produces profits to begin with. This is worth noting in the re-up decision, where alignment carries real weight.

Reconciling a distribution notice

The most useful habit an LP desk can build is checking each distribution notice against the waterfall rather than simply booking the cash. In practice that means:

1. Rebuild the cumulative position. Total contributions to date, total distributions to date, and the accrued preferred return. The capital account statement should support these numbers, and if it does not, that is the first question. 2. Identify the current tier. Is the fund still returning capital, paying the pref, in catch-up, or in the split? The distribution notice should say, and the answer should be consistent with the cumulative position. 3. Check the carry figure. If carry was paid, confirm that it is consistent with the tier and the catch-up mechanics. In a deal-by-deal fund, check whether escrow was withheld and in what amount. 4. Track cumulative GP carry against a whole-fund shadow calculation. Even in a deal-by-deal fund, a simple running estimate of what carry would be on an aggregate basis tells you whether a clawback exposure is building. This is the single best early warning indicator. 5. Log the result. The outcome of each check, any discrepancy, and the GP's explanation belong in the fund record, not in someone's inbox.

This sounds like a lot, but most distributions are straightforward and the check takes minutes once the cumulative figures are tracked. The value comes from the handful of notices where something does not add up.

Waterfalls and performance metrics

Waterfall mechanics feed directly into the metrics LPs report. Net IRR, net TVPI, and DPI are all calculated after carry, so a fund that has paid early carry under a deal-by-deal structure can show weaker DPI than its gross performance suggests, and stronger later if a clawback is paid. Conversely, a fund whose net TVPI includes unrealized gains may have significant accrued carry embedded in its NAV. Many GPs report NAV net of accrued carry, but not all present it the same way.

When comparing funds, especially across vintages, it helps to know which structure each fund uses and whether reported NAV is net of hypothetical carry. Our guides to DPI, TVPI, and MOIC and to benchmarking with IRR, TWR, and PME cover how those comparisons can mislead.

Waterfalls in GP-led secondaries

Continuation vehicles introduce a second layer. When a GP moves an asset from an older fund into a new vehicle, the old fund's waterfall typically crystallizes carry on the transfer price, and the new vehicle starts a fresh waterfall, often with a tiered carry structure that steps up at higher return multiples. An LP that rolls into the new vehicle is effectively paying carry twice on the same asset, once at the transfer and again on future gains. That can still be the right decision, but it should be a decision rather than a default. Our GP-led secondaries framework goes into the evaluation in detail.

Portfolio-level monitoring

For an allocator with dozens of fund commitments, the waterfall is not a one-time legal question but an ongoing data problem. A practical monitoring setup tracks, for each fund:

• Waterfall structure (whole-fund or deal-by-deal) and catch-up percentage.
• Preferred return rate and compounding basis.
• Clawback terms: tax offset, guarantee type, escrow percentage, interim test frequency.
• Cumulative carry paid to date and a shadow whole-fund carry estimate.
• Current tier and the date the fund is expected to enter carry.
• Any carry escrow balance.

With that information in one place, the desk can answer questions that usually take days: which funds have paid carry while still below 1.0x DPI, where clawback exposure is building, and which GPs are approaching the end of their term with escrow still outstanding. It also turns the next fundraise into a better conversation, because the desk can point to concrete history rather than impressions.

Questions to ask a GP

During diligence or at the annual meeting, a short set of questions surfaces most of what matters:

• Does the waterfall return all contributed capital, including fees and expenses, before carry?
• When does the preferred return start accruing, and how is it affected by the credit facility?
• If carry has been paid, what is the current cumulative clawback exposure on an aggregate basis?
• What percentage of carry is held in escrow, and what triggers its release?
• Is the clawback guarantee joint and several, and who are the guarantors?
• How are distributions in kind valued for waterfall purposes?

A GP who answers these quickly and consistently usually runs a clean back office. Hesitation is informative, and so is a GP who needs to "check with fund admin" on the cumulative carry figure.

The bottom line

The headline terms tell you the shape of the deal. The waterfall tells you what you actually get paid. For LPs, the work is not complicated, but it is unglamorous: read the definitions, model the structure, check each distribution, and track carry and clawback exposure across the portfolio. Funds that run clean waterfalls rarely mind the scrutiny. The ones that do mind are the ones worth scrutinizing.

Frequently asked questions

What is a distribution waterfall in private equity?

A distribution waterfall is the section of a fund's limited partnership agreement that sets the order in which cash is distributed between LPs and the GP. A typical structure returns contributed capital to LPs first, then pays a preferred return, then allows the GP to catch up, and finally splits the remaining proceeds, usually 80% to LPs and 20% to the GP.

What is the difference between a European and an American waterfall?

A European, or whole-fund, waterfall requires the fund to return all contributed capital plus the preferred return before the GP earns any carried interest. An American, or deal-by-deal, waterfall calculates carry on each realized investment, so the GP can be paid earlier, even while other investments are losing money. Deal-by-deal structures rely on clawback provisions to correct any overpayment.

What is a GP catch-up?

After LPs receive their preferred return, the catch-up tier gives the GP a large share of subsequent distributions until the GP has received its full carried interest percentage of total profits. With a 100% catch-up and 20% carry, once the fund clears the hurdle by enough, the GP ends up with exactly 20% of all profits.

Why do clawbacks often fail to make LPs whole?

Most clawbacks are capped at the after-tax amount of carry the GP received, are tested only at the end of the fund, and may rely on several rather than joint guarantees from individual partners. Each of those features can leave LPs with less than the full overpayment returned, which is why escrow, interim tests, and joint and several guarantees matter.

How does a subscription line affect the preferred return?

The preferred return usually accrues from the date LP capital is contributed. When a fund uses a credit facility and calls LP capital later, the measured holding period is shorter, the IRR is higher, and the hurdle is easier to clear. Some LPs negotiate for the preferred return to accrue from the date the fund draws on the facility instead.

Sources and scope

The worked examples in this article are original, simplified illustrations and do not represent any specific fund. Real waterfalls vary widely, and the governing terms are always those of the specific LPA. For reference frameworks, see the ILPA Principles and the ILPA Model LPA.

Topics

• Waterfalls
• Carried Interest
• Clawback
• LPA
• Fund Terms