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

First published: 24th of September 2026

Latest update: 29th of September 2026

A private equity fund is governed by rules.

Some of those rules come from law and regulation. Some arise from accounting standards, tax law or regulatory requirements. Some are established through separate agreements with individual investors.

But the central document governing the economic relationship between the investors and the manager is normally the Limited Partnership Agreement, or LPA.

For anyone seeking to understand carried interest, the LPA is indispensable.

It tells us what investors have committed to provide, what the GP may do with that capital, when capital may be called, how investments and expenses are allocated, when proceeds may be distributed, which investors may be excluded from particular investments, how profits are divided and, ultimately, when the carried-interest participants become entitled to receive money.

The LPA therefore connects almost everything we have discussed so far:

\[ Commitments \rightarrow Drawdowns \rightarrow Investments \rightarrow Allocations \rightarrow Distributions \rightarrow Waterfall \rightarrow Carried\ Interest \]

But an LPA is not an instruction manual written for the person who must eventually calculate carried interest.

It is a legal agreement.

The relevant economic rules may be spread across hundreds of pages, definitions, schedules, amendments, side letters and subscription documents. One provision may refer to another provision, which uses a defined term whose meaning depends upon yet another definition.

A single word can matter.

Contribution.

Investment.

Realised Investment.

Distributable Proceeds.

Cost.

Preferred Return.

Commitment.

Unfunded Commitment.

Carried Interest.

Each may have a specific contractual meaning that differs from the intuitive meaning of the word.

Consequently, one of the most important principles in carried-interest work is:

Never calculate first and read the LPA afterwards.

The calculation is an implementation of the agreement.

The agreement comes first.

1. What is an LPA?

The limited partnership agreement establishes the contractual relationship between the parties to a limited partnership.

In a conventional private equity structure, the principal parties are:

• the General Partner, or GP; and
• the Limited Partners, or LPs.

Other parties or economic classes may also exist. Depending upon the structure, these can include founder partners, special limited partners, carried-interest partners or vehicles and other classes established for particular economic purposes.

The GP normally has responsibility for managing or controlling the partnership, frequently delegating investment-management functions to an affiliated manager or adviser.

The LPs provide most of the capital but generally do not conduct the day-to-day management of the partnership.

The LPA defines the relationship between them.

It is simultaneously:

• a governance document;
• an economic agreement;
• an operating framework;
• an allocation rulebook;
• and, from our perspective, the foundation of the carried-interest calculation.

2. The LPA is the economic constitution of the fund

It is useful to think of the LPA as the economic constitution of the fund.

It establishes the boundaries within which the GP operates.

It may determine:

• the purpose of the fund;
• its duration;
• investment restrictions;
• the investment period;
• borrowing powers;
• commitment mechanics;
• drawdowns;
• defaults;
• expenses;
• management fees;
• distributions;
• recycling;
• recallability;
• allocations of profit and loss;
• preferred return;
• carried interest;
• clawback;
• transfers;
• successor funds;
• governance;
• conflicts;
• amendments;
• and termination.

Not every LPA is organised in precisely this way.

Indeed, two economically similar funds can have LPAs organised very differently.

That is one reason why carried-interest calculations cannot safely be built by searching for a section headed Carried Interest and implementing whatever appears there.

The economics of carried interest may depend upon provisions appearing throughout the agreement.

3. The LPA records a bargain

An LPA should not be viewed merely as legal machinery.

Behind it is a commercial bargain.

The LP essentially says:

I will commit capital to this fund and permit you considerable discretion over its deployment, subject to agreed restrictions.

The GP essentially says:

I will organise and manage the fund, source and manage investments, operate within the agreed rules and return the resulting proceeds according to the agreed economic arrangements.

The economics of that bargain include several different forms of compensation and risk.

LPs supply capital and bear investment risk.

The manager may receive management fees.

The GP or associated parties may commit their own capital.

Members of the investment team may participate in carried interest.

The LPA translates this bargain into enforceable rules.

Carried interest is therefore not something added to the fund after its investment performance has been determined.

It is embedded in the economic bargain from the beginning.

4. There is no universal LPA

Private equity practitioners frequently use expressions such as:

“Normally the LPA says…”

or:

“European waterfalls work like…”

Such statements can be useful descriptions of market practice.

They are dangerous substitutes for reading the agreement.

There is no universal private equity LPA.

Two funds may both have:

• 20% carry;
• an 8% preferred return;
• a five-year investment period;
• and a ten-year term,

yet produce different carried-interest outcomes because their definitions and mechanics differ.

One fund may include management fees in the capital that must be returned before carry.

Another may not.

One may calculate preferred return from the date capital is contributed.

Another may use another contractual date.

One may permit extensive recycling.

Another may restrict it.

One may use a whole-of-fund waterfall.

Another may calculate carry investment by investment.

Even where the headline economics appear identical:

\[ 20\%\ Carry + 8\%\ Hurdle \]

the actual economics may be different.

The headline terms are not the model.

The LPA is the model specification.

5. Read the definitions first

If there is one part of an LPA that deserves disproportionate attention, it is the definitions section.

Defined terms can determine the outcome of an entire calculation.

Suppose the waterfall says:

“First, 100% to the Limited Partners until they have received an amount equal to their Contributions.”

That looks straightforward.

But what are Contributions?

Do they include:

• amounts used to acquire investments?
• management fees?
• organisational expenses?
• broken-deal expenses?
• taxes?
• amounts subsequently recalled?
• temporary investments?
• bridge financing?
• equalisation amounts?

The answer may be contained entirely in the definition.

Similarly, if preferred return accrues on Unreturned Contributions, we need to know exactly what causes a Contribution to become returned.

A distribution?

Any distribution?

Only a return-of-capital distribution?

A deemed distribution?

A recallable distribution?

A distribution net of withholding?

Those questions cannot be answered from the headline waterfall.

They depend upon definitions.

6. Defined terms form a network

Definitions should not be read independently.

Suppose:

Preferred Return

is calculated on:

Unreturned Contributions

which are defined by reference to:

Contributions

less:

Returned Capital

and Returned Capital depends upon:

Distributable Proceeds.

The calculation chain is:

\[ Distributable\ Proceeds \rightarrow Returned\ Capital \rightarrow Unreturned\ Contributions \rightarrow Preferred\ Return \]

An error in the interpretation of the first definition can therefore flow through the entire waterfall.

This is why sophisticated LPA interpretation resembles building a dependency map.

For each important calculated amount, ask:

What does this number depend upon?

Then follow those dependencies until they terminate in actual transactions or factual information.

That is essentially how the eventual carried-interest model should work as well.

7. Never assume ordinary language has its ordinary meaning

Legal drafting frequently assigns specific meanings to familiar words.

Investment might include only portfolio investments.

Or it might include bridge investments.

Cost might include transaction expenses.

Or it might not.

Proceeds might include dividends.

Or dividends might be classified separately as income.

Realisation might include a refinancing.

Or only a disposal.

Commitment may change following transfers, defaults or additional closings.

Distribution may include amounts deemed distributed even when no cash moves.

Therefore:

\[ Ordinary\ Meaning \neq Contractual\ Meaning \]

unless the LPA confirms otherwise.

This becomes particularly important when translating legal language into software or spreadsheets.

A model cannot understand ambiguity.

Eventually every relevant contractual concept must become a rule.

8. The parties matter

Before interpreting the economics, identify the parties and economic classes.

A simple fund may contain:

\[ GP + LPs \]

A more complicated structure might contain:

\[ GP + LPs + Founder\ Partner + Carry\ Partner \]

There may also be parallel partnerships, feeder vehicles, alternative investment vehicles and co-investment structures.

Why does this matter?

Because a payment to the GP is not necessarily carried interest.

It might be:

• a return on the GP's own investment;
• a management fee;
• reimbursement of expenses;
• a priority allocation;
• or carried interest.

Likewise, carried interest may not be paid directly to the GP at all.

It may be allocated to a separate carried-interest partnership or vehicle.

Before calculating who receives what, therefore, establish:

Who are the economic participants, and in what capacity can each receive money?

9. Fund structure and governing documents

The LPA does not exist in isolation.

A private equity fund may have a collection of governing and offering documents including:

• the LPA;
• private placement memorandum;
• subscription agreement;
• deeds of adherence;
• side letters;
• amendments;
• investment-management agreement;
• advisory agreements;
• carried-interest arrangements;
• and constitutional documents of related entities.

The old fund-accounting text correctly distinguishes, for example, the PPM, subscription/deed-of-adherence documentation and side letters as separate documents surrounding the LPA.

For our purposes, their functions are important.

The PPM describes the investment proposition being offered.

The subscription documentation records an investor's admission and commitment and obtains representations from that investor.

The LPA establishes the general contractual rules of the partnership.

The side letter may modify how some of those rules apply to a particular investor.

Consequently:

\[ Fund\ Rules \neq LPA\ Alone \]

in every case.

10. The fund's purpose

The LPA normally describes the purpose and permitted activities of the partnership.

At first sight this may appear far removed from carried interest.

It is not.

Carried interest rewards performance generated within a particular investment mandate.

The LPA may establish restrictions concerning:

• asset classes;
• industries;
• geography;
• investment size;
• concentration;
• diversification;
• leverage;
• follow-on investments;
• related-party transactions;
• and other investment activities.

These provisions define the investment universe within which the GP is permitted to generate the profits ultimately entering the waterfall.

11. Fund term

Private equity funds are generally established for a finite period.

A common structure might provide:

\[ 10\ Years + Extensions \]

but the precise term varies.

Extensions may require:

• GP discretion;
• LP Advisory Committee approval;
• LP consent;
• or some combination.

The term matters because the fund cannot simply hold investments indefinitely.

Eventually investments must be realised, liabilities settled and the partnership wound up.

That makes the LPA's termination and liquidation provisions relevant to carried interest.

A carry model may look attractive while significant unrealised assets remain.

The ultimate economics are not known until the portfolio is realised.

This is one reason clawback exists.

12. The closing process

A private equity fund does not necessarily raise all of its commitments on one day.

There may be:

First Closing

followed by one or more:

Subsequent Closings

and eventually:

Final Closing.

For example:

\[ First\ Closing = €600m \]\[ Second\ Closing = +€250m \]\[ Final\ Closing = +€150m \]\[ Final\ Fund\ Size = €1bn \]

This creates an obvious fairness issue.

If early investors have already funded investments before later investors join, how should the later investors be placed economically into the same position?

That is the purpose of equalisation or closing adjustments.

13. Subsequent closings and equalisation

Suppose Investor A commits €100 million at first closing and funds €20 million.

Six months later Investor B joins with the same €100 million commitment.

If B simply begins participating from that date, A and B do not have equivalent economic positions.

B has avoided funding the first six months while potentially obtaining exposure to investments acquired during that period.

The LPA will therefore usually contain mechanics designed to rebalance the investors.

Depending upon the structure, B may contribute amounts representing its share of earlier contributions plus an equalisation amount or interest-like adjustment.

Some of those amounts may be distributed to existing investors.

Some may be retained by the fund.

Some may have particular treatment for preferred-return or commitment purposes.

Equalisation is therefore not merely a closing administration exercise.

It modifies the capital history from which carry is ultimately calculated.

14. Commitment

The LPA defines the commitment mechanics.

As discussed in the previous chapter:

\[ Commitment \neq Cash\ Already\ Paid \]

It represents the investor's contractual obligation to provide capital when validly called.

The LPA may determine:

• the commitment amount;
• commitment currency;
• increases or reductions;
• transfers;
• defaults;
• recycling;
• recallability;
• and circumstances in which commitment can no longer be called.

This means that the commitment ledger we built in the previous chapter is ultimately an implementation of LPA provisions.

15. Drawdown powers

The LPA normally gives the GP authority to issue capital calls for permitted purposes.

Those purposes may include:

• investments;
• follow-on investments;
• management fees;
• partnership expenses;
• liabilities;
• reserves;
• repayment of borrowing;
• indemnification;
• taxes;
• and other permitted fund purposes.

The distinction matters because different categories of drawdown may receive different treatment elsewhere in the agreement.

For example, the preferred-return definition might exclude certain temporary or bridge-related contributions.

Or recallability provisions may apply differently to management fees and investments.

Therefore the capital-call provision should never be read by itself.

16. The investment period

The investment period, sometimes called the commitment period, is one of the major phases in a fund's life.

During this period the GP generally has broad authority to call capital for new investments.

A typical period might be five years, although actual terms vary.

The LPA determines when it starts and when it ends.

Termination may occur through:

• passage of time;
• deployment of a specified amount;
• key-person events;
• GP removal;
• launching a successor fund;
• LP action;
• or other contractual events.

Once the investment period ends, the GP's ability to deploy capital generally becomes more restricted.

17. The end of the investment period does not necessarily end drawdowns

This is an important distinction.

An investor may assume:

Investment period over = no more capital calls.

That is generally too simplistic.

After the investment period, capital may still be callable for permitted purposes such as:

• follow-on investments;
• existing contractual commitments;
• management fees;
• expenses;
• liabilities;
• indemnities;
• taxes;
• reserves;
• or other specifically permitted items.

Therefore:

\[ End\ of\ Investment\ Period \neq End\ of\ All\ Capital\ Calls \]

For carry modelling, the distinction matters because additional contributions can continue changing the investor's capital and preferred-return balances.

18. Follow-on investments

The LPA normally addresses the GP's ability to invest additional capital into existing portfolio companies.

These are follow-on investments.

Suppose the fund originally invests €50 million in Company A.

Three years later Company A requires another €15 million.

That additional investment may be needed to:

• finance growth;
• fund an acquisition;
• protect the original investment;
• provide liquidity;
• refinance debt;
• or rescue a company experiencing difficulty.

Follow-ons matter to carried interest because they increase the capital attributable to an investment and can change the economics of a later realisation.

In a deal-by-deal carry structure, correct attribution becomes especially important.

19. Bridge and temporary investments

LPAs may distinguish genuine portfolio investments from temporary uses of cash.

The old book identifies bridge or temporary investments as short-duration investments made while cash is awaiting deployment or distribution.

That distinction can matter considerably for carried interest.

Depending upon the LPA, temporary investments may be excluded from:

• invested-capital calculations;
• preferred-return calculations;
• catch-up calculations;
• or other elements of the waterfall.

Again, classification matters because:

\[ Cash\ Invested \]

does not necessarily equal:

\[ Capital\ Relevant\ to\ Carry \]

The contractual definition controls.

20. Excused and excluded investors

The previous chapter showed what happens economically when an investor does not participate in a particular investment.

The LPA provides the legal machinery.

An investor may be excused or excluded because participation could cause:

• legal problems;
• regulatory problems;
• tax consequences;
• sanctions issues;
• policy conflicts;
• or other circumstances contemplated by the agreement.

Individual investors may also have negotiated additional rights through side letters.

Suppose:

\[ A=B=C=D=25\% \]

but D is excused from Investment X.

The LPA determines how X is allocated among A, B and C and how subsequent proceeds, expenses and losses follow that participation.

This is where legal drafting becomes investment-level accounting.

21. An excusal can persist for years

The immediate effect of an excusal is easy to understand:

D does not fund Investment X.

The long-term consequences are more complicated.

Years later X may:

• pay dividends;
• require follow-on capital;
• incur expenses;
• be partially sold;
• be refinanced;
• produce escrow proceeds;
• incur a loss;
• or be fully realised.

The system must remember that D did not participate.

This is why investor-level investment participation must remain linked to the original LPA event.

The contractual exception creates a data attribute that may survive for the entire life of the investment.

22. Defaults

An LP that fails to satisfy a valid capital call may become a defaulting investor.

LPAs commonly provide substantial remedies because a default can harm the entire partnership.

The GP may have entered into binding investment obligations relying upon committed capital.

Potential contractual consequences can include, depending on the agreement:

• interest;
• penalties;
• suspension of rights;
• forced transfer;
• dilution;
• forfeiture;
• compulsory sale;
• set-off against distributions;
• or other remedies.

For carried-interest calculations, a default can create complex allocation consequences.

Who funds the shortfall?

Do other investors acquire a greater participation?

What happens to the defaulting investor's existing economics?

Again, the answer is not generic.

It is contractual.

23. Management fees

The management-fee provisions are economically important even though management fees are not carried interest.

A common headline arrangement might be expressed as:

\[ Management\ Fee = 2\% \times Fee\ Base \]

But the interesting question is:

2% of what?

During the investment period, the fee base may be commitments.

Afterwards it may change to:

• invested capital;
• acquisition cost;
• net invested capital;
• NAV;
• unrealised investment cost;
• or another contractually defined amount.

The fee rate itself may also step down.

Consequently:

\[ Fee = Rate \times Defined\ Base \]

Both terms must be obtained from the LPA.

24. Why management fees matter to carry

Suppose LPs contribute:

\[ €100m \]

for investments and:

\[ €10m \]

for management fees.

Before carry can be paid, must the LPs receive:

\[ €100m \]

or:

\[ €110m \]

back?

That depends upon the waterfall definitions.

Thus management fees affect carried interest even though they are not themselves carry.

This illustrates an important principle:

A provision can be outside the carried-interest section and still change the carried-interest result.

The same applies to expenses.

25. Fee offsets

The manager or its affiliates may receive transaction fees, monitoring fees, director fees or other income associated with portfolio investments.

The LPA may provide that some or all of these amounts offset the management fee.

Suppose:

\[ Gross\ Management\ Fee = €20m \]

and qualifying portfolio fees are:

\[ €4m \]

with a 100% offset.

Then:

\[ Net\ Management\ Fee = €16m \]

If management fees are funded by investors and form part of capital that must be returned before carry, the offset indirectly affects the waterfall.

Once again:

\[ Fee\ Provision \rightarrow Capital\ Calls \rightarrow Contributed\ Capital \rightarrow Waterfall \]

The LPA is an interconnected system.

26. Fund expenses

The LPA allocates responsibility for expenses between the fund and the manager.

Possible fund expenses include:

• legal costs;
• audit;
• tax;
• administration;
• custody;
• valuation;
• broken-deal expenses;
• financing expenses;
• insurance;
• regulatory expenses;
• litigation;
• due diligence;
• consultants;
• and portfolio-related costs.

The precise allocation can materially affect LP economics.

For carried-interest purposes, two questions are particularly important:

Who bears the expense?

and:

Does the expense enter the capital base relevant to the waterfall?

Those are different questions.

27. Broken-deal expenses

Private equity funds investigate many transactions that they never complete.

The costs of those unsuccessful transactions can be substantial.

An LPA may specify how broken-deal expenses are borne.

This becomes particularly interesting where:

• co-investors participated in the proposed transaction;
• an affiliate would have invested;
• one or more LPs were excused;
• or the proposed investment would have been allocated differently from the fund generally.

Who should bear the abandoned transaction cost?

There is no mathematical answer.

The contractual allocation rule is required.

And that cost may subsequently affect contributed capital and carry.

28. Allocations

One of the most important words in private equity is allocation.

Unfortunately, it can describe several different things.

We might allocate:

• a drawdown;
• an investment;
• an expense;
• income;
• realised gain;
• unrealised gain;
• realised loss;
• a distribution;
• or carried interest.

Those allocations do not necessarily use the same percentages.

The LPA therefore needs to be interpreted carefully whenever it refers to allocations.

29. Cash allocation and profit allocation are not necessarily the same

This distinction is particularly important.

Suppose Investor A receives €10 million cash.

That does not automatically mean €10 million of accounting profit is allocated to A.

Conversely, an investor may receive an allocation of profit without receiving the same amount of cash at that moment.

We therefore distinguish:

\[ Cash\ Allocation \]

from:

\[ Profit/Loss\ Allocation \]

and from:

\[ Capital\ Account\ Allocation \]

For carried-interest purposes, the relationship between these can matter greatly.

The old accounting text makes this same distinction, noting that LPAs can contain separate rules for allocation of cash and allocation of profits and losses.

For us, the lesson is broader:

Never assume that one allocation rule governs every economic layer of the fund.

30. Distributions

The LPA establishes the GP's authority and obligations concerning distributions.

Distributions may arise from:

• disposal proceeds;
• dividends;
• interest;
• refinancing;
• return of unused capital;
• refunds;
• temporary investments;
• and other receipts.

The LPA may permit the GP to retain amounts for:

• reserves;
• liabilities;
• expenses;
• taxes;
• indemnities;
• follow-ons;
• or other anticipated obligations.

It may also permit distributions in specie.

But the crucial question for this book is not merely:

Can the GP distribute the cash?

It is:

How must the distributable amount be divided?

That takes us to the waterfall.

31. The waterfall

The distribution waterfall establishes the priority in which economic value is distributed among the relevant participants.

At its most simplified, a whole-of-fund waterfall might resemble:

\[ Distributable\ Proceeds \]\[ \downarrow \]

Return of Capital

\[ \downarrow \]

Preferred Return

\[ \downarrow \]

GP Catch-up

\[ \downarrow \]

Residual Split

For example:

\[ 80\%\ LP / 20\%\ Carry \]

This diagram is useful.

It is also dangerously incomplete.

Every box contains definitions.

32. “Return of capital” is not self-explanatory

Suppose LPs have contributed €120 million.

Of that:

• €100m funded investments;
• €12m funded management fees;
• €5m funded expenses;
• €3m funded other amounts.

If the waterfall says capital must first be returned, does that mean:

\[ €100m? \]\[ €112m? \]\[ €117m? \]

or:

\[ €120m? \]

The answer depends upon what the LPA defines as the relevant contribution.

This is why shorthand descriptions of waterfalls can be misleading.

Two funds described commercially as:

“8% preferred return, 20% carry”

can produce materially different distributions.

33. Preferred return

The preferred return, often called the hurdle, establishes a priority return to investors before carried interest participates according to the applicable waterfall.

A simplified formulation might be:

\[ Preferred\ Return = 8\%\ per\ annum \]

But that tells us very little.

We need to know:

• 8% on what?
• From what date?
• Until what date?
• Simple or compounded?
• Compounded when?
• On gross or net contributions?
• What reduces the balance?
• How are recallable distributions treated?
• What happens after partial returns?
• How are equalisation contributions treated?
• How are multiple currencies handled?

The percentage is often the easiest part of the calculation.

34. The preferred-return base

Suppose an LP contributes €10 million.

A year later the fund distributes €4 million.

What balance earns preferred return thereafter?

Perhaps:

\[ €6m \]

But not necessarily.

What if the €4 million represented profit rather than returned capital?

What if it was recallable?

What if the LPA applies distributions first against preferred return?

The preferred-return base depends upon the waterfall mechanics.

Thus:

\[ Preferred\ Return = f( Amount, Date, Classification, Prior\ Distributions, LPA\ Rules ) \]

It is a stateful calculation.

The answer today depends upon everything that happened before today.

35. Catch-up

After LPs have received the required priority return, many waterfalls contain a catch-up.

The purpose is to move the carried-interest participant towards its agreed share of overall profits.

Consider a simplified structure with:

• 8% preferred return;
• 20% carry;
• full catch-up.

After LP capital and preferred return have been satisfied, a portion of subsequent distributions may go disproportionately—or even entirely—to the carry participant until the intended sharing relationship has been reached.

Thereafter residual profits might be divided:

\[ 80\%\ LP \]\[ 20\%\ Carry \]

The catch-up is therefore not an additional return unrelated to carry.

It is part of the mechanism through which the agreed carried-interest percentage is achieved.

We will analyse its mathematics separately in the carried-interest chapters.

36. Whole-of-fund versus deal-by-deal waterfalls

One of the most consequential distinctions is whether carry is calculated principally at:

fund level

or:

investment level.

Under a whole-of-fund waterfall, LPs generally recover specified fund-level amounts before carry becomes distributable.

Under a deal-by-deal structure, carry may become payable from successful realised investments while other investments remain unrealised.

Consider:

\[ Investment\ A: +€50m \]\[ Investment\ B: -€30m \]

A deal-by-deal waterfall may permit carry relating to A before the economic loss on B has fully entered the realised calculation.

That creates greater potential for overdistribution of carry.

Hence the importance of clawback.

37. Carried interest

The LPA or related governing documents determine the economic entitlement to carried interest.

The headline percentage might be:

\[ 20\% \]

But, as we have already seen, the percentage is only one variable.

The actual carry depends upon:

\[ Carry = f( Contributions, Distributions, Preferred\ Return, Catch\text{-}up, Realised\ Gains, Losses, Expenses, Allocation\ Rules, Timing, Definitions ) \]

This is why a carried-interest model cannot sensibly begin with:

“Put 20% in cell B12.”

The model begins with contractual interpretation.

38. Carry distributions versus carry entitlement

Another distinction will become important later.

A calculation may indicate that the carried-interest participant has economically earned a particular amount.

That does not necessarily mean the entire amount is immediately distributable.

The LPA may contain:

• escrow;
• holdbacks;
• reserves;
• carry deferral;
• guarantees;
• clawback protections;
• or other mechanisms.

Thus:

\[ Calculated\ Carry \neq Carry\ Paid \]

and:

\[ Carry\ Paid \neq Carry\ Ultimately\ Retained \]

The final amount retained may not be known until the fund's economics have substantially or completely played out.

39. Clawback

Clawback exists because carried interest can sometimes be distributed before the final economics of the fund are known.

Suppose the GP receives carry following several successful early exits.

Later investments perform poorly.

At the end of the fund, the GP may have received more carry than it would have been entitled to had the final fund result been known from the beginning.

Conceptually:

\[ Carry\ Paid > Final\ Carry\ Entitlement \]

The difference may be subject to:

\[ Clawback \]

subject to the precise contractual terms.

Clawback therefore represents an important reconciliation between interim distributions and final economics.

40. Clawback is not merely “give back excess carry”

That shorthand hides substantial complexity.

A clawback provision may require consideration of:

• taxes paid by carry recipients;
• net-of-tax limitations;
• guarantees;
• several versus joint liability;
• caps;
• escrow;
• interim clawback testing;
• timing;
• individual carry recipients;
• departed employees;
• currency movements;
• and subsequent adjustments.

Therefore the final carry calculation can become both a fund-level calculation and a recipient-level recovery problem.

We will return to this in detail later.

41. Recycling and recallability

The LPA determines whether and when proceeds can be reused or recalled.

This can directly affect the amount of capital deployed over the life of the fund.

A €1 billion fund may, through permitted recycling, invest more than €1 billion of gross capital over its life.

For carried-interest purposes, the questions include:

• Was the amount actually distributed?
• Was it designated recallable?
• Did it restore commitment?
• Was it subsequently recalled?
• Does it count again as contributed capital?
• How does it affect preferred return?

These are precisely the questions developed in the previous chapter.

The LPA supplies the answers.

42. Borrowing and subscription facilities

Modern funds frequently have borrowing powers.

The LPA may permit borrowing at fund level and may restrict:

• amount;
• duration;
• purpose;
• security;
• and repayment.

Subscription facilities can create a particularly important issue for performance and carry.

Suppose an investment is made on 1 January using a fund-level facility.

LP capital is not called until 1 July.

Economically the LP has been exposed to the investment since January, but the LP's cash contribution occurs six months later.

If preferred return begins only when LP capital is actually contributed, the facility can change the timing of the preferred-return calculation.

Thus financing provisions can affect waterfall economics even though they appear outside the carried-interest section.

43. Currency

The LPA normally identifies the fund's base or functional currency and may contain provisions governing commitments and transactions in other currencies.

As discussed previously, multi-currency structures can create differences between:

• commitment currency;
• drawdown currency;
• investment currency;
• distribution currency;
• and reporting currency.

For carried interest, we need to know which currency is used for the relevant calculation and what conversion conventions apply.

A USD profit and EUR profit cannot simply be added without a conversion rule.

Therefore currency provisions can become waterfall provisions.

44. Transfers

LP interests may be transferred, subject to the terms of the LPA and usually GP consent.

A transfer creates an important carried-interest question:

What happens to the economic history?

Suppose LP A has:

• €100m commitment;
• €60m contributed;
• €40m remaining commitment;
• €20m distributions;
• accrued preferred return;
• participation in six investments.

LP B acquires A's interest.

The fund cannot treat B as a brand-new investor with no history.

The transferred interest carries an economic history with it, subject to the applicable transfer mechanics.

This makes transfer processing another example of why carry calculations depend upon persistent historical data.

45. Side letters

The LPA may establish the general rule.

A side letter can establish a different rule for a particular investor.

This is one of the most important practical complications in fund administration.

A side letter may provide rights concerning matters such as:

• fees;
• reporting;
• excuse rights;
• regulatory treatment;
• tax;
• confidentiality;
• transfers;
• co-investment;
• notification;
• most-favoured-nation provisions;
• or other investor-specific arrangements.

Therefore:

\[ General\ LPA\ Rule + Investor\ Exception = Applicable\ Investor\ Rule \]

A carried-interest or allocation system that reads only the LPA can therefore still be wrong.

46. Most-favoured-nation provisions

A Most Favoured Nation, or MFN, provision may permit an investor to elect certain more favourable terms subsequently granted to another investor.

This can create a second layer of complexity.

The applicable terms may not simply be:

What does this investor's signed side letter say?

They may also depend upon:

Which rights granted elsewhere has the investor validly elected?

A complete operational rulebook therefore may need to combine:

\[ LPA + Side\ Letter + MFN\ Elections + Amendments \]

This is why document control is part of carried-interest control.

47. Subscription documents and deeds of adherence

The LPA tells us the general rules.

The subscription agreement or deed of adherence tells us important facts about a particular investor.

These can include:

• identity;
• commitment;
• investor classification;
• tax status;
• regulatory status;
• representations;
• contact information;
• and agreement to be bound by the fund documents.

The investor's commitment may therefore not be found merely by reading the generic LPA.

The LPA provides the framework.

The subscription documentation populates that framework with investor-specific facts.

48. Amendments

LPAs can change.

An amendment may alter:

• the fund term;
• investment period;
• management fees;
• investment restrictions;
• borrowing powers;
• distribution mechanics;
• or potentially economic provisions.

This creates an important modelling principle:

Rules can be time-dependent.

Suppose the LPA is amended on 1 January 2030.

Transactions before that date may have arisen under one rule set.

Transactions after it may arise under another.

A model that simply overwrites the old rule can lose the historical basis of prior calculations.

Version control therefore matters.

49. The LP Advisory Committee

Many funds establish an LP Advisory Committee, usually abbreviated LPAC.

The LPAC can have functions relating to:

• conflicts;
• related-party matters;
• valuations;
• extensions;
• waivers;
• investment restrictions;
• key-person events;
• and other matters specified in the LPA.

The LPAC does not normally manage the fund.

Its role is instead part of the governance architecture between broad GP discretion and matters requiring investor oversight.

For carried-interest purposes, LPAC decisions can matter when they alter or approve circumstances that affect investments, valuations or fund duration.

50. Key-person provisions

Private equity investors frequently commit capital partly because of the individuals expected to manage it.

LPAs therefore commonly contain key-person provisions.

If specified individuals cease devoting the required time to the fund, consequences may include:

• suspension of the investment period;
• restriction on new investments;
• LPAC involvement;
• investor approval requirements;
• or termination of the investment period.

A key-person event can therefore change the fund's ability to deploy remaining commitments.

That in turn changes future fees, investments, returns and ultimately carry.

Governance provisions can have economic consequences.

51. Removal of the GP

The LPA may contain mechanisms permitting removal of the GP.

The provisions may distinguish between:

for-cause removal

and:

no-fault removal.

The thresholds and consequences can differ substantially.

For carried interest, the crucial issue is what happens to the GP's and carry participants' economics after removal.

Possible outcomes may include continuation, reduction, forfeiture or crystallisation of some economic rights, depending entirely upon the documents.

The important conceptual point is:

Carry is not always determined solely by investment performance. Governance events can affect entitlement.

52. No-fault divorce and suspension mechanisms

Some LPAs permit investors, subject to specified voting thresholds, to suspend the investment period or terminate aspects of the GP's authority without establishing misconduct.

These provisions provide an important counterweight to the discretion delegated to the GP.

Economically, they can change:

• future deployment;
• management-fee basis;
• fund duration;
• successor-fund timing;
• and potentially carry-related outcomes.

Again, provisions that appear to concern governance can flow into the economics.

53. Successor funds

An LPA may restrict the GP from launching or closing a successor fund until a specified point.

For example, the restriction might continue until:

• a certain percentage of commitments has been invested or reserved; or
• the investment period has ended.

The purpose is intuitive.

LPs committed to Fund I because they expected the investment team to focus on Fund I.

They do not want the GP immediately diverting its attention to Fund II.

The provision can also affect management fees, allocation of opportunities and team incentives.

54. Co-investment and parallel vehicles

Private equity transactions are not always undertaken solely through the main fund.

There may be:

• co-investment vehicles;
• parallel funds;
• alternative investment vehicles;
• employee vehicles;
• feeder structures;
• or other related entities.

Suppose the main fund invests €80 million and a co-investment vehicle invests €20 million.

Transaction expenses total €5 million.

Who bears those expenses?

80/20?

Only the fund?

According to another allocation methodology?

What about broken-deal expenses?

What about proceeds?

What about opportunities that the main fund could have taken itself?

These questions can affect both LP economics and the profit base from which carry is generated.

55. Allocation of investment opportunities

Where a manager operates multiple funds, separately managed accounts or co-investment vehicles, the governing arrangements may address allocation of investment opportunities.

This is fundamentally a conflict-management issue.

But it also affects carry.

If a profitable investment is allocated:

\[ 70\%\ Fund \]

and:

\[ 30\%\ Co\text{-}Investment \]

then only the fund's portion normally contributes to the fund's own economic result.

Allocation decisions therefore determine the pool of assets from which fund-level carried interest can arise.

56. Conflicts of interest

Private equity structures contain inherent potential conflicts.

The GP or its affiliates may:

• manage multiple funds;
• receive fees;
• allocate investment opportunities;
• arrange transactions between affiliated funds;
• value illiquid assets;
• allocate expenses;
• arrange co-investments;
• and participate in carry.

The LPA establishes mechanisms for managing these conflicts.

These may involve:

• disclosure;
• LPAC consent;
• investor consent;
• independent valuation;
• procedural safeguards;
• or explicit contractual permissions.

For carried-interest practitioners, conflicts matter because many of them concern allocation.

And allocation determines economics.

57. Books, records and reporting

The LPA normally establishes obligations concerning:

• books and records;
• financial statements;
• investor reporting;
• tax information;
• valuation;
• inspection rights;
• and reporting deadlines.

For our purposes, this section has a deeper significance.

The LPA creates economic rules.

The fund's records must preserve sufficient information to demonstrate that those rules have been followed.

For carried interest, that means retaining the lineage between:

\[ LPA\ Rule \]

and:

\[ Transaction \]

and:

\[ Calculation \]

and:

\[ Distribution \]

A number without that lineage may be correct.

It is not yet demonstrably correct.

58. Reading an LPA for carried interest

When approaching an unfamiliar fund, I would not start by reading every page with equal intensity.

I would first build an economic map.

Identify:

1. Parties and classes

Who can contribute and who can receive?

2. Commitments

What has each investor promised?

3. Contributions

What counts as contributed capital?

4. Investment period

When and for what purposes can capital be called?

5. Allocations

How are investments, expenses, profits and losses shared?

6. Distributions

What constitutes distributable proceeds?

7. Recallability and recycling

Can returned amounts be used again?

8. Preferred return

What earns it, from when, and how?

9. Catch-up

How does the carry participant catch up?

10. Residual split

What is the final profit-sharing ratio?

11. Clawback

How is overdistributed carry corrected?

12. Investor-specific exceptions

What do side letters change?

Only after mapping these relationships would I build the calculation.

59. Build a rule matrix

For complex funds, one of the most useful practical tools is an LPA rule matrix.

For example:

This transforms hundreds of pages of legal drafting into a structured specification.

But the matrix does not replace the LPA.

It points back to it.

60. Then build the dependency map

The next step is to identify how the rules interact.

For example:

\[ Commitment \rightarrow Management\ Fee \]\[ Commitment \rightarrow Drawdown\ Allocation \]\[ Drawdowns \rightarrow Contributions \]\[ Contributions \rightarrow Unreturned\ Capital \]\[ Unreturned\ Capital \rightarrow Preferred\ Return \]\[ Distributions \rightarrow Returned\ Capital \]\[ Returned\ Capital \rightarrow Preferred\ Return\ Base \]\[ Preferred\ Return \rightarrow Catch\text{-}up \]\[ Catch\text{-}up \rightarrow Residual\ Carry \]

This is the skeleton of the carry model.

The spreadsheet or software comes later.

61. Translate legal language into calculation rules

Consider an illustrative clause:

“Distributable Proceeds shall first be distributed to the Limited Partners until they have received aggregate distributions equal to their Unreturned Contributions.”

A model cannot execute prose.

It needs a rule.

For example:

\[ Tier\ 1 = \min( Available\ Proceeds, Unreturned\ Contributions ) \]

But even this translation is incomplete until we define:

\[ Unreturned\ Contributions \]

and determine whether the calculation is:

• investor-specific;
• fund-wide;
• investment-specific;
• currency-specific;
• or affected by previous distributions.

Thus every mathematical formula should be traceable to contractual language.

62. Do not simplify too early

A common modelling error is to translate the LPA immediately into a simplified spreadsheet structure.

Suppose the LPA contains five categories of contributions.

The modeller decides that they all behave similarly and combines them into:

Capital.

That may work for several years.

Then a transaction occurs where one category receives different treatment.

The historical distinction has been lost.

The better principle is:

Preserve contractual distinctions in the data even when they currently produce the same result.

If two categories genuinely behave identically, the model can aggregate them for presentation.

But the underlying data should retain their identity.

This is particularly important for long-lived funds.

63. Ambiguity is itself a finding

Sometimes the LPA does not provide a clear answer.

This should not be hidden by the model.

Suppose the agreement defines preferred return but is unclear about the treatment of a particular recallable distribution.

The modeller should not quietly choose the interpretation that seems most sensible and embed it in a formula.

Instead:

\[ Ambiguous\ Contract \rightarrow Documented\ Interpretation \rightarrow Appropriate\ Approval \rightarrow Implemented\ Rule \]

The interpretation may require fund counsel, the GP, the administrator, auditors or other appropriate parties.

A model should implement a decision.

It should not silently make the legal decision itself.

64. Small words can move large amounts of money

Consider the difference between:

“all Contributions”

and:

“Contributions attributable to Investments”.

Or:

“realised Investments”

and:

“all Investments”.

Or:

“may”

and:

“shall”.

Or:

“distributed”

and:

“deemed distributed”.

In a €5 billion fund, one phrase can affect tens of millions of euros of carry.

This is why LPA interpretation requires a different form of precision from ordinary financial modelling.

The mathematics may be simple.

The difficult part is determining which mathematics the contract requires.

65. The LPA should be read dynamically

An LPA is not merely a collection of static rules.

Many provisions depend upon the stage of the fund.

During fundraising:

\[ Closing\ Rules \]

dominate.

During the investment period:

\[ Commitment + Drawdown + Investment + Fee\ Rules \]

dominate.

During harvesting:

\[ Realisation + Distribution + Carry \]

become increasingly important.

Towards termination:

\[ Reserves + Final\ Distribution + Clawback \]

become critical.

The same provision can also behave differently before and after a particular event.

Private equity economics are therefore both:

\[ Transaction\ Dependent \]

and:

\[ Time\ Dependent \]

The carry model must be capable of reflecting both.

66. The LPA and the accounting records must eventually meet

The LPA is legal language.

The accounting system records transactions.

The carry model applies economic rules.

Ultimately all three need to reconcile:

\[ LPA \]

defines:

\[ Economic\ Rights \]

which are reflected through:

\[ Transactions\ and\ Allocations \]

which enter:

\[ Accounting\ Records \]

and feed:

\[ Carry\ Calculations \]

which determine:

\[ Distributions \]

A disconnect anywhere in this chain can produce an incorrect result.

This is why carried-interest work sits at the intersection of law, accounting, finance and data.

67. The LPA is necessary, but not always sufficient

Even after carefully reading the LPA, we may still need:

• subscription documents;
• side letters;
• amendments;
• LPAC approvals;
• transfer agreements;
• closing documentation;
• investment-allocation records;
• historic notices;
• and documented interpretations.

Therefore the relevant question is not merely:

“Do we have the LPA?”

It is:

Do we have the complete set of documents and decisions that determine the economic rights of every participant?

For a mature fund with many investors, that can be a substantial information-management exercise.

68. Why the LPA matters so much for carried interest

We can now connect this chapter directly to the purpose of this book.

The previous chapter established that carried interest depends upon an economic history:

\[ Commitments + Contributions + Investments + Distributions + Time \]

This chapter adds the missing component:

\[ Rules \]

Thus:

\[ Economic\ History + LPA\ Rules = Waterfall\ Inputs \]

and:

\[ Waterfall\ Inputs + Waterfall\ Mechanics = Carried\ Interest \]

This is the architecture we will use throughout the remainder of the book.

69. The central lesson

An LPA should not be regarded as a legal document that sits in a folder after the fund closes.

It is an executable economic specification written in legal language.

Every capital call implements part of it.

Every allocation implements part of it.

Every distribution implements part of it.

Every preferred-return calculation implements part of it.

Every carry payment implements part of it.

And every clawback calculation ultimately asks whether those earlier implementations produced the economic result the parties agreed.

The job of the carried-interest practitioner is therefore not simply to understand a waterfall formula.

It is to translate:

\[ Contract \]

into:

\[ Economic\ Rules \]

into:

\[ Data \]

into:

\[ Calculation \]

into:

\[ Cash \]

while preserving the ability to travel all the way back in the opposite direction.

From any carry payment, we should be able to ask:

Why was this amount paid?

and trace the answer backwards through:

\[ Carry\ Payment \]\[ \uparrow \]\[ Waterfall\ Calculation \]\[ \uparrow \]\[ Distributions,\ Contributions\ and\ Allocations \]\[ \uparrow \]\[ Applicable\ LPA\ Provisions \]

That traceability is what turns a calculation into a defensible carried-interest calculation.

And now that we understand both the economic events and the contractual rulebook governing them, we have the two foundations required to begin examining the mechanics at the centre of this book:

how private equity profits are actually allocated between the investors and the carried-interest participants.

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

[email protected]

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