Carry Capitalisation: When Unrealised Carry Becomes Participating Capital

Author: Gert-Tom Draisma / www.TristanFinance.com

First published: 24th of September 2026

Latest update: 3rd of October 2026

Status: First draft

A particularly unusual carried-interest structure arises where carried interest is not merely calculated and accrued to the GP, but is periodically converted into an ownership interest in the fund.

This creates a fundamentally different economic dynamic from a conventional carried-interest arrangement.

Under a conventional structure, the GP may have an entitlement to carried interest calculated by reference to the fund's performance. That entitlement may be realised, unrealised, accrued, distributed or retained, but it remains conceptually distinguishable from the investor capital on which subsequent fund economics are allocated.

Under a carry-capitalisation structure, that distinction progressively disappears.

At a specified calculation date, typically annually, the waterfall is applied to determine the GP's carried-interest entitlement, including carried interest attributable to unrealised appreciation. The resulting carried interest is then converted into an ownership interest or capital balance of the GP.

From that point onwards, the GP's capital participates in subsequent fund economics substantially as investor capital would.

The process can therefore be represented as:

Fund Economics

↓

Periodic Waterfall Calculation

↓

Carry Entitlement

↓

Conversion of Carry into GP Capital

↓

GP Capital Becomes Participating Capital

↓

Subsequent Fund Economics Allocated Between LP and GP Capital

↓

Next Waterfall Calculation

↓

Further Carry Adjustment

The structure is recursive:

Carry Changes Ownership → Ownership Changes Future Allocations → Future Allocations Change Carry → Carry Changes Ownership Again

The complication is that the GP can simultaneously occupy two different economic positions.

It is an owner of participating capital for purposes of investment returns, while remaining the recipient of carried interest calculated on the economics attributable to the LPs.

That distinction is essential to understanding the structure.

1. Carry Becomes Capital

Assume that the LPs initially contribute €100 million to a fund and the GP contributes no capital.

During the first period, the fund appreciates to €150 million.

Assume for the moment that the applicable waterfall determines that €10 million of the €50 million gain belongs to the GP as carried interest.

Instead of recording €10 million merely as accrued carry, the fund transfers €10 million of capital ownership from the LPs to the GP.

The capital accounts become:

The €10 million allocated to the GP is no longer merely an amount potentially payable to the GP.

It has become participating capital.

Therefore:

Carry Entitlement → Capital Ownership

This is the defining state transition.

2. Previously Capitalised Carry Participates in Subsequent Performance

Suppose that during the next period the fund generates another €30 million of investment gain.

The €30 million is not allocated entirely to the LPs because the LPs no longer own 100% of the participating capital.

Immediately before the gain:

• LP capital is €140 million;
• GP capital is €10 million;
• total participating capital is €150 million.

The €30 million gain is therefore allocated:

LP:

€30m × 93.333% = €28m

GP:

€30m × 6.667% = €2m

Before calculating any new carried interest, the capital accounts become:

The €2 million earned by the GP is not new carried interest.

It is investment return earned on capital that the GP already owns.

Therefore:

Return on GP Carry Capital ≠ New Carried Interest

This distinction becomes increasingly important as the GP's ownership grows.

3. The GP Can Be an Investor for One Purpose and the Carry Recipient for Another

The fact that GP capital participates in fund performance does not mean that the GP should automatically be included in the economic population on which new carried interest is calculated.

This would create an obvious problem.

If the GP owns €12 million of the fund after receiving its share of investment performance, the GP should not ordinarily pay carried interest to itself on the investment return earned by its own capital.

The structure therefore contains at least two economic populations.

For investment performance:

Investment-Return Population = LP Participating Capital + GP Participating Capital

For carried interest:

Carry-Paying Population = Relevant LP Economics

The GP can therefore be included in one allocation and excluded from another.

This illustrates an important general principle:

The Same Partner Can Belong to Different Economic Populations for Different Economic Events

The applicable population must therefore be determined separately for each event.

4. There Is No Universal Allocation Percentage

The structure becomes even more interesting where different types of income and expense use different allocation bases.

Assume that:

• realised investment gains and losses are allocated according to total participating capital;
• unrealised investment gains and losses are allocated according to total participating capital;
• investment income is allocated according to total participating capital;
• ordinary fund expenses are allocated according to total participating capital;
• management fees continue to be borne only by the LPs.

The calculation populations are then:

Investment Gains and Losses → Total Participating Capital

Investment Income → Total Participating Capital

Ordinary Fund Expenses → Total Participating Capital

Management Fee → LP Fee-Bearing Population

Carried Interest → Relevant LP Economics

There is therefore no single allocation percentage that can be attached to the GP or an LP and applied to every transaction.

Instead:

Economic Event → Applicable Economic Population → Allocation Basis → Allocation

This is precisely why calculation populations need to be treated as part of the economic specification rather than merely as a data attribute.

5. Commitment Percentage and Capital Ownership Percentage Diverge

The GP may have made no original commitment to the fund.

Nevertheless, after the first carry conversion it owns €10 million of the €150 million of participating capital.

Its participation percentage is therefore:

€10m / €150m = 6.667%

even though its commitment percentage remains:

0%

Consequently:

Commitment Percentage ≠ Capital Ownership Percentage

and:

Capital Ownership Percentage ≠ Original Investor Participation Percentage

The distinction becomes increasingly significant as successive carry allocations change the capital ownership of the fund.

6. New Contributions Dilute Existing GP Capital

Assume again that:

• LP capital is €140 million;
• GP capital is €10 million.

The GP therefore owns 6.667% of participating capital.

The LPs subsequently contribute another €50 million.

Capital becomes:

The GP's participation falls to:

€10m / €200m = 5.000%

The GP has not lost any capital.

It still owns €10 million.

However, the additional LP contribution has diluted its percentage participation in future fund economics.

Therefore:

New Participating Capital → New Ownership Split → New Future Allocation Percentages

A model cannot therefore assume that the ownership percentage established at the previous carry calculation remains constant.

7. Distributions Can Also Change Ownership Percentages

The same principle applies to distributions.

If a distribution is made proportionately to all participating capital, the absolute capital balances decline but the relative ownership percentages may remain unchanged.

If a distribution is made disproportionately, the ownership percentages can change.

Similarly, a distribution may be classified differently depending on whether it represents:

• investment proceeds;
• income;
• return of capital;
• a carry distribution;
• a distribution from GP participating capital;
• or another contractual category.

The model must therefore determine both the amount and economic classification of each distribution.

The capital roll-forward becomes:

Opening Capital + Contributions + Allocated Economics − Distributions ± Other Capital Events = Capital Before Carry Adjustment

Only after this state has been established can the next carry calculation be performed.

8. Downside Performance Creates Two Separate Questions

A down year exposes the most important feature of this structure.

Suppose the GP owns €10 million out of €150 million of participating capital and the fund subsequently loses €30 million.

The ordinary investment loss is allocated according to participating capital:

LP loss:

€30m × 93.333% = €28m

GP loss:

€30m × 6.667% = €2m

Capital becomes:

The GP has therefore lost €2 million as an investor.

But this does not answer the carried-interest question.

The fund's decline may also mean that some of the carried interest previously allocated to the GP is no longer supported by the current cumulative waterfall.

These are two completely different economic effects:

Loss on GP Participating Capital ≠ Reversal of Previously Capitalised Carry

The first results from the GP owning capital.

The second results from recalculating the carried-interest entitlement.

Both may occur in the same period.

9. Negative Incremental Carry

Assume that €10 million of carry was previously capitalised.

Following a subsequent decline, the cumulative waterfall is recalculated and now supports only €6 million of cumulative carried interest.

The change in carry entitlement is:

€6m Current Cumulative Carry Entitlement − €10m Previously Capitalised Carry = (€4m)

There is therefore €4 million of negative incremental carry.

This creates a fundamental design question:

What happens to previously capitalised carry when the current cumulative waterfall no longer supports it?

There is no single answer.

At least two materially different structures are possible.

10. Architecture A — Reversible Carry Capitalisation

Under a reversible structure, the capital accounts continuously reflect the carried-interest entitlement produced by the current cumulative waterfall.

If cumulative carry increases, capital moves from the LPs to the GP.

If cumulative carry decreases, capital moves back from the GP to the LPs.

Thus:

Positive Incremental Carry → Transfer from LP Capital to GP Capital

and:

Negative Incremental Carry → Transfer from GP Capital Back to LP Capital

In the previous example, €4 million would be transferred back from GP capital to LP capital.

The GP may therefore experience two separate reductions during the same period:

1. its share of the investment loss; and
2. reversal of carry that is no longer supported by the waterfall.

These should never be combined into a single unexplained movement.

11. Architecture B — Irreversible Capitalisation with Loss Recovery

A different structure may treat the periodic carry allocation as crystallised once transferred into GP capital.

Previously capitalised carry is not transferred back merely because subsequent performance deteriorates.

Instead, the decline creates a deficit that must be recovered before additional carry can be generated.

Conceptually:

Previously Capitalised Carry Remains GP Capital

↓

Fund Performance Declines

↓

Current Carry Entitlement Falls Below Previously Crystallised Carry

↓

Loss Carryforward / High-Water-Mark Deficit Arises

↓

No Further Carry Until Deficit Is Recovered

Under this architecture, the GP still suffers its ordinary share of losses on its participating capital.

What it does not suffer is a separate transfer of previously crystallised capital back to the LPs.

The two architectures can therefore produce materially different ownership states after exactly the same investment performance.

12. Reversal and High-Water-Mark Protection Are Not the Same

Suppose:

• GP capital immediately after previous carry capitalisation: €10 million;
• GP share of subsequent investment loss: €2 million;
• current cumulative waterfall indicates €4 million less carry than previously recognised.

Under a reversible structure, the GP could first suffer the €2 million investment loss and then potentially suffer a further capital transfer resulting from the €4 million negative incremental carry, subject to the precise calculation mechanics.

Under an irreversible structure, the GP suffers the €2 million investment loss but the €4 million carry deficit is instead carried forward as a barrier to future carry.

Therefore:

Loss Participation ≠ Carry Reversal ≠ Loss Carryforward

These are three separate economic concepts.

A calculation model must not substitute one for another.

13. High-Water Marks and Loss Carryforwards

The irreversible approach resembles the economic logic found in structures that use high-water marks or loss carryforwards.

The essential idea is that performance previously rewarded should not ordinarily generate another performance allocation merely because value first falls and subsequently recovers.

Suppose the fund rises from €100 million to €150 million and carry is crystallised.

It subsequently falls to €120 million and later recovers to €150 million.

Without a high-water-mark or equivalent loss-recovery mechanism, the recovery from €120 million to €150 million could potentially generate carry for a second time.

A properly designed structure prevents that result.

Conceptually:

Previous Performance Already Rewarded → Subsequent Loss → Recovery of Loss → No New Carry Until Previous Performance Level Is Recovered

Only performance beyond the relevant threshold can generate additional carry.

The exact mechanism, however, is contractual. A high-water mark, loss carryforward, cumulative waterfall and reversible capital account can produce related but not necessarily identical outcomes.

14. The Effective Date of New GP Capital Is Critical

Carry capitalisation also requires a precise rule determining when newly allocated carry begins participating in fund economics.

Suppose the fund earns €50 million during Year 1 and the year-end waterfall determines that €10 million belongs to the GP.

If that €10 million is treated as though it had been GP capital throughout Year 1, the GP would participate in the very investment gain from which its carry was calculated.

That can create circularity.

Unless the governing arrangement deliberately requires such treatment, the cleaner sequence is:

Year 1 Economics Allocated to Opening Capital

↓

Year-End Carry Calculated

↓

Carry Transferred into GP Capital

↓

GP Capital Participates from Defined Effective Date Thereafter

Therefore:

Calculation Date → Carry Allocation → Ownership Effective Date

must be explicitly defined.

Otherwise:

New Carry → Participation in Gain Generating That Carry → More Carry → More Participation

can create an unintended circular calculation.

15. The Annual Calculation Has Three Separate Stages

A robust implementation should separate three calculations.

Stage 1 — Allocate Current-Period Economics

Begin with opening participating capital.

Allocate:

• realised gains and losses;
• unrealised gains and losses;
• investment income;
• applicable expenses;
• contributions;
• distributions;
• other relevant economic events.

Each event is allocated using its appropriate economic population.

This produces:

Capital Before Current Carry Adjustment

Stage 2 — Calculate Cumulative Carry Entitlement

The waterfall is then calculated using the relevant LP economics.

The GP's own participating capital is not automatically included in the carry-paying population.

The calculation determines:

Current Cumulative Carry Entitlement

Stage 3 — Determine the Carry Adjustment

Compare current cumulative carry entitlement with the relevant carry previously capitalised:

Incremental Carry Adjustment = Current Cumulative Carry Entitlement − Previously Recognised Carry Entitlement

The result may be:

• positive;
• zero;
• or negative.

The contractual architecture then determines how that adjustment affects capital ownership.

This produces:

Closing Capital After Carry Adjustment

which becomes the opening capital for the next period.

16. A Fully Worked Multi-Year Example

Consider a simplified fund with the following terms:

• initial LP contribution: €100 million;
• GP initial contribution: nil;
• annual calculation date: 31 December;
• preferred return: 8%;
• carried interest: 20%;
• for simplicity, assume the preferred-return amount relevant to each annual cumulative calculation is stated directly in the example rather than introducing the full timing mechanics developed in Chapter 4;
• GP capital participates in investment gains and losses;
• management fees are borne only by the LPs;
• carry is calculated only on the relevant LP economics;
• newly capitalised carry begins participating after the annual calculation date.

The objective is not to demonstrate a universal contractual structure. It is to demonstrate the interaction between participating GP capital and recurring carry calculations.

17. Year 1 — Initial Carry Capitalisation

The LP contributes:

€100m

Assume the fund generates €50 million of investment appreciation during Year 1.

Before carry:

Assume the applicable cumulative waterfall determines that the GP is entitled to €10 million of carry.

The €10 million is transferred from LP capital to GP capital.

Closing capital becomes:

The GP now owns:

€10m / €150m = 6.667%

of participating capital.

Year 1 therefore creates the first recursive state:

Carry₁ → GP Capital₁ → Ownership for Year 2

18. Year 2 — GP Capital Earns Its Own Return

Assume Year 2 produces €30 million of investment gain.

Opening ownership is:

• LP: 93.333%;
• GP: 6.667%.

The gain is allocated:

LP:

€30m × 93.333% = €28m

GP:

€30m × 6.667% = €2m

Capital before the Year 2 carry calculation is therefore:

The €2 million allocated to the GP is investment return.

It is not Year 2 carry.

Now assume the cumulative waterfall calculated solely on the relevant LP economics determines that cumulative carry should be €14 million.

Previously recognised carry was €10 million.

Therefore:

Year 2 Incremental Carry = €14m − €10m = €4m

The €4 million is transferred from LP capital to GP capital.

Closing capital becomes:

The GP's €16 million closing capital consists of:

• €10 million previously capitalised carry;
• €2 million investment return on that capital;
• €4 million newly capitalised carry.

These components have different economic origins even though they are now part of the same GP capital account.

19. Year 3 — A Down Year

Assume Year 3 produces a €36 million investment loss.

Opening capital is:

• LP: €164 million;
• GP: €16 million;
• total: €180 million.

The ownership percentages are:

LP:

€164m / €180m = 91.111%

GP:

€16m / €180m = 8.889%

The €36 million loss is therefore allocated:

LP:

€36m × 91.111% = €32.8m

GP:

€36m × 8.889% = €3.2m

Capital before the Year 3 carry recalculation becomes:

The GP has already lost €3.2 million because it owned 8.889% of the participating capital.

Now assume the cumulative waterfall is recalculated and supports only €8 million of cumulative carry.

Previously recognised cumulative carry was €14 million.

Therefore:

Incremental Carry Adjustment = €8m − €14m = (€6m)

The structure now reaches its critical fork.

20. Year 3 Under Reversible Capitalisation

Under Architecture A, the €6 million negative incremental carry is transferred back from GP capital to LP capital.

Capital becomes:

The GP's capital has fallen from €16 million to €6.8 million.

But that €9.2 million reduction consists of two different effects:

€3.2m Investment Loss

plus:

€6.0m Carry Reversal

The distinction must remain visible.

21. Year 3 Under an Irreversible / High-Water-Mark Structure

Under Architecture B, the €6 million negative incremental carry is not transferred back.

The GP retains the €12.8 million remaining after its share of the investment loss.

Closing capital is therefore:

However, the carry system records a €6 million carry deficit or equivalent loss-recovery balance.

No new carry can be generated until the relevant deficit has been recovered according to the contractual methodology.

The same fund therefore ends Year 3 with two possible ownership structures:

The fund value is identical.

The ownership is not.

Therefore:

Same Fund NAV + Different Carry-Capitalisation Architecture = Different Ownership

22. Year 4 — Recovery

Assume the fund subsequently recovers.

The consequences now differ because Year 4 begins with different ownership states.

Under the reversible structure, the GP begins with €6.8 million.

Under the high-water-mark structure, it begins with €12.8 million but also carries the relevant carry deficit.

The same investment gain will therefore be allocated differently between LP and GP capital before any new carry is calculated.

This illustrates the recursive nature of the structure.

The Year 3 carry treatment has changed Year 4 ownership.

Year 4 ownership changes the allocation of Year 4 investment performance.

That allocation changes the LP economics on which Year 4 carry is calculated.

Thus:

Year 3 Carry Rule → Year 4 Ownership → Year 4 Economic Allocation → Year 4 Carry

The historical architecture continues to affect future periods even after fund performance has recovered.

23. The Two Structures Do Not Necessarily Reconverge Automatically

It may be tempting to assume that once the fund returns to its previous NAV, both methodologies produce the same result.

That is not necessarily true.

The reversible structure transferred capital back to the LPs.

The irreversible structure left more capital with the GP but imposed a carry deficit.

During the recovery period, those different capital balances participate differently in investment performance.

Consequently:

Same Current NAV ≠ Same Current Ownership

and:

Same Current NAV ≠ Same Historical Waterfall State

The calculation must preserve the path by which the current state was reached.

24. Carry Must Be Calculated on LP Economics

The worked example also exposes another implementation requirement.

Suppose GP capital earns €2 million of investment return.

That €2 million should not ordinarily be treated as additional LP profit when calculating carry.

Otherwise the GP would effectively earn carried interest on the return generated by its own capital.

The calculation therefore needs to distinguish:

Total Fund Performance

from:

Performance Attributable to Carry-Paying LP Capital

The carry engine must operate on the latter population where that is what the governing economics require.

Thus:

Fund Return → Allocate Between Capital Owners → Identify LP Economics → Apply LP Waterfall

rather than:

Fund Return → Apply Carry Percentage to Entire Fund

This distinction is fundamental to implementing the structure correctly.

25. Management Fees Create Another Economic Population

Suppose management fees are borne only by the LPs.

The GP's capital participates in investment performance but does not bear management fees.

A €2 million management fee therefore reduces LP capital by €2 million without reducing GP capital.

This changes the ownership percentages.

If LP capital before the fee is €140 million and GP capital is €10 million:

Total capital:

€150m

After a €2 million LP-only management fee:

LP capital:

€138m

GP capital:

€10m

Total:

€148m

The GP's percentage ownership increases from:

€10m / €150m = 6.667%

to:

€10m / €148m = 6.757%

without the GP receiving any additional carry.

This demonstrates again why:

Change in Ownership Percentage ≠ Necessarily New Carry

Ownership can change because different economic events are allocated to different populations.

26. Income and Expenses Require Explicit Classification

The same issue applies to other items.

The model must determine whether each item participates according to:

• total participating capital;
• LP capital only;
• commitment;
• a specific investment population;
• a particular class;
• a carry pocket;
• or another contractual allocation basis.

Potential items include:

• interest income;
• dividend income;
• realised gains;
• unrealised gains;
• realised losses;
• unrealised losses;
• management fees;
• fund administration expenses;
• broken-deal costs;
• financing expenses;
• tax expenses;
• transaction expenses.

The correct allocation cannot be inferred merely from the fact that the GP has become an investor.

The governing economics determine the population.

27. Carry Capitalisation Is Not the Same as Carry Distribution

Another important distinction is between capitalising carry and paying carry.

If €10 million of carry is transferred to the GP's capital account, no cash necessarily leaves the fund.

The GP may own €10 million of capital while receiving no cash.

Therefore:

Carry Generated ≠ Carry Capitalised ≠ Carry Distributed

A later distribution of GP capital is a separate event.

This matters for:

• liquidity;
• tax;
• accounting;
• clawback;
• reporting;
• participant allocations.

The model should preserve these states separately.

28. Capitalised Carry May Itself Become Realised or Unrealised

Once the GP owns participating capital, its capital account can contain value attributable to several sources.

For example:

GP Capital = Capitalised Carry + Return on Capitalised Carry − Losses − Distributions ± Other Allocations

The distinction between those components may matter later.

A €20 million GP capital account does not necessarily mean that €20 million of carry has been generated.

Part of the €20 million may represent investment return earned after carry was capitalised.

Conversely, cumulative carry generated may exceed current GP capital if the GP has subsequently suffered investment losses or received distributions.

Therefore:

GP Capital Balance ≠ Cumulative Carry Generated

29. A Capital Account Does Not Eliminate the Need for Carry History

Because GP capital contains both carry and subsequent investment performance, the current capital balance cannot by itself tell the model how much carry has historically been generated.

The system must retain separate histories for at least:

• carry generated;
• carry capitalised;
• carry reversed;
• return on GP participating capital;
• losses allocated to GP participating capital;
• GP contributions, if any;
• GP distributions;
• current GP capital.

Therefore:

Current GP Capital ≠ Sufficient Carry History

This becomes especially important where negative incremental carry or high-water-mark mechanics exist.

30. Why the Structure Is Stateful

The result in any period depends on prior periods.

Year 4 cannot be calculated correctly from Year 4 NAV alone.

The model needs to know:

• how much capital the GP owned at the beginning of Year 4;
• how that capital arose;
• whether prior carry was reversible;
• whether a loss carryforward exists;
• whether a high-water mark exists;
• which economic populations apply;
• what contributions and distributions occurred;
• which expenses were LP-only;
• the relevant hurdle state.

Therefore:

Current-Year Data Alone ≠ Sufficient Calculation Data

The calculation is inherently stateful:

Previous State + New Economic Events + Allocation Rules + Waterfall Rules = New State

31. Carry Is Both an Output and a Future Input

In a conventional waterfall, carry can largely be viewed as an output:

Economic Events → Waterfall → Carry

In a carry-capitalisation structure:

Economic Events → Waterfall → Carry → Capital Ownership

and subsequently:

Capital Ownership → Economic Allocation → Waterfall → New Carry

Carry therefore occupies two positions in the calculation architecture.

It is:

1. an output of the current waterfall; and
2. an input into future economic allocations.

That feedback loop is the defining feature of the edge case.

32. Similar Concepts in Evergreen and Periodically Crystallising Structures

Although the precise structure described here is unusual, the underlying economic issues are not confined to traditional closed-ended private equity funds.

Evergreen, semi-liquid and other periodically valued private-market structures can contain performance allocations or incentive mechanisms determined periodically by reference to NAV.

Hedge-fund structures have long dealt with related questions through periodic incentive allocations, crystallisation, high-water marks and loss-recovery mechanisms.

These structures are not necessarily economically or legally identical to the carry-capitalisation mechanism described here.

The comparison is nevertheless useful because they raise similar questions:

• when performance participation crystallises;
• whether previously recognised participation can reverse;
• how subsequent losses are treated;
• whether losses must be recovered before new incentive allocations arise;
• how ownership or capital accounts change;
• which capital participates in subsequent performance;
• and how new subscriptions are treated relative to existing investors.

The edge case is therefore unusual, but the underlying problem is increasingly relevant as private-market fund structures become more varied.

33. Tax Consequences

Converting an unrealised carried-interest entitlement into an actual participating ownership interest may have tax consequences that differ from merely accruing an amount of carried interest.

Relevant questions may include:

• when the GP is considered to have acquired the ownership interest;
• whether unrealised carry creates taxable income;
• the tax basis of the resulting interest;
• taxation of subsequent investment returns on that interest;
• the distinction between carry and return on capital;
• treatment of subsequent losses;
• treatment of reversals;
• consequences for individual carry participants.

These questions depend heavily on jurisdiction and legal structure.

The waterfall calculation should therefore determine the economics first and allow the tax analysis to follow from the actual legal and economic structure.

The tax treatment is considered separately in Chapter 8 — Tax.

34. Accounting Consequences

The structure also creates accounting questions.

The accounting records must distinguish between:

• the original LP capital;
• GP capital created through carry allocations;
• subsequent returns on GP capital;
• subsequent losses allocated to GP capital;
• additional carry allocations;
• reversals where applicable;
• distributions from GP capital.

The accounting treatment should not obscure the underlying economics.

In particular:

Carry Allocation ≠ Return on Existing GP Capital

even where both ultimately increase the same GP capital account.

The accounting implications are considered further in Chapter 10 — Accounting.

35. Data Consequences

This structure is particularly demanding from a data perspective because the current calculation depends on the complete historical evolution of ownership.

The system must be able to reconstruct:

Opening Ownership

↓

Economic Events

↓

Allocation by Population

↓

Capital Before Carry

↓

Cumulative Carry Calculation

↓

Incremental Carry Adjustment

↓

Closing Ownership

for every relevant calculation period.

A current snapshot is insufficient.

This connects directly to the principles discussed in Chapter 11 — Carry Data.

36. Modelling Consequences

A model implementing this structure cannot simply calculate:

Carry = Carry Percentage × Profit

Nor can it treat GP ownership as a static input.

The model must allow ownership to be generated by the waterfall itself.

The architecture is therefore recursive across periods, while remaining sequential within each period.

A robust implementation should preserve:

• opening capital by economic population;
• event-level allocation rules;
• capital ownership before carry;
• LP-only waterfall state;
• cumulative carry entitlement;
• previous carry entitlement;
• incremental carry adjustment;
• carry reversal or loss-recovery balance;
• closing capital ownership.

These modelling issues are considered further in Chapter 12 — Carry Modelling.

37. Control Requirements

Several controls are particularly important.

The first is a capital reconciliation:

Opening Capital + Contributions + Allocated Profit/Loss − Distributions ± Carry Transfers = Closing Capital

The second is an ownership reconciliation:

LP Capital + GP Capital = Total Participating Capital

The third is a carry reconciliation:

Previous Cumulative Carry + Incremental Carry Adjustment = Current Cumulative Carry

where the architecture uses reversible cumulative carry.

The fourth is a population control:

Was every economic event allocated to the correct economic population?

The fifth is an effective-date control:

Did newly capitalised carry begin participating only from the contractually correct date?

These controls are considered further in Chapter 13 — Controls & Assurance.

38. The Central Economic Distinctions

The structure depends on maintaining several distinctions that can otherwise easily become blurred:

Carry Generated ≠ Carry Capitalised

Carry Capitalised ≠ Carry Distributed

Capitalised Carry ≠ Return Earned on Capitalised Carry

Loss on GP Capital ≠ Reversal of Carry

Carry Reversal ≠ Loss Carryforward

Commitment Percentage ≠ Capital Ownership Percentage

Investment-Return Population ≠ Carry-Paying Population

GP Capital Balance ≠ Cumulative Carry Generated

Current NAV ≠ Complete Waterfall State

These distinctions are not merely terminology.

Each can change the numerical result.

39. Why This Is an Edge Case

This structure is an edge case because the waterfall does something more than allocate existing economic value.

It changes the ownership base on which future economic value will itself be allocated.

In most waterfall calculations, ownership can conceptually be treated as an input into the calculation.

Here, ownership is also an output.

That output becomes an input into the next calculation.

Therefore:

Ownershipₜ → Carryₜ → Ownershipₜ₊₁ → Carryₜ₊₁

The relationship is recursive.

A calculation cannot be performed correctly without preserving the state created by previous calculations.

40. Final Principle

The complete architecture can be represented as:

Opening Ownership

↓

Allocate Current Fund Economics

↓

Determine LP Economics

↓

Calculate Current Cumulative Carry

↓

Compare with Previously Recognised Carry

↓

Determine Positive or Negative Incremental Carry

↓

Apply Carry Capitalisation, Reversal or Loss-Recovery Rules

↓

Determine Closing Ownership

↓

Use Closing Ownership as Opening Ownership for the Next Period

Or, more compactly:

Carry → Capital → Ownership → Return → Carry

The central lesson is:

A carried-interest calculation can do more than determine how existing economic value is divided. In some structures, it changes the ownership structure through which future economic value will itself be divided.

That creates a calculation in which carry is simultaneously a result of past performance and a determinant of future economic participation.

For that reason, the structure should not be modelled merely as a periodic carried-interest accrual.

It is a recursive capital-allocation system.

Need assistance with your Carried Interest Challenges? Reach out to us:

[email protected]

Or ask your questions here:

The Carried Interest Bible © 2026 Table Bay Investments Ltd, All rights reserved.