GP-LP Alignment: What Stewardship Actually Looks Like on a Term Sheet

Every institutional fund manager claims alignment with its limited partners. The claim appears in the pitch deck, in the due diligence questionnaire, in the side-by-side return attribution that accompanies the capital call. What it rarely appears in is the limited partnership agreement itself, written with sufficient specificity to function as a constraint rather than an aspiration. That gap between what a GP asserts in its marketing materials and what a GP encodes in its legal documents is not an oversight. It is the operating definition of the alignment problem.

The problem is structural, not interpersonal. Alignment language has accumulated in institutional private equity over decades precisely because it is useful without being costly. A firm that describes its culture as LP-friendly, its interests as co-terminus with the interests of its investors, and its governance as stewardship-driven has made claims that are difficult to falsify and impossible to enforce because they live in a register that legal documents cannot reach. The term sheet is the only document in the transaction that can reach them, and most term sheets do not try.

This article describes what GP-LP alignment actually looks like when it is taken seriously at the level of fund documentation. It identifies the provisions that separate encoded alignment from aspirational alignment, examines the institutional industry's use of the term "LP-friendly" and why it has become inadequate as an evaluative standard, proposes a framework for reading an LPA as a stewardship document, and argues for the systematic preference of verifiable mechanical tests over reputational heuristics. It concludes with the application of that framework to The SAVI Capital Model, where each of the four tenets is expressed as a specific LPA mechanic visible to the limited partner before commitment.

Alignment is Either Encoded or Aspirational

The conventional GP-LP relationship rests on a structural misalignment that predates the vocabulary used to manage it. The general partner earns a management fee on committed or invested capital regardless of whether the investment performs. The general partner earns carried interest on profits above a hurdle rate, but the hurdle rate, the carry percentage, the catch-up mechanism, and the conditions under which carry vests are all negotiated variables, and they are negotiated from a position in which the GP controls the template. The result is a document that reflects the GP's preferences with modifications the LP has extracted through negotiation, rather than a document that reflects a genuinely bilateral alignment of economic interest.

ILPA Principles 3.0, the most comprehensive institutional benchmark for LP-protective fund terms, identifies the structural deficiencies that survive even well-negotiated partnership agreements: management fee offsets that are incomplete or capped, carry waterfalls that permit the GP to receive distributions before all invested capital is returned, key-person provisions that are drafted too narrowly to trigger meaningfully, and LPAC consent rights that are advisory rather than binding. These are not fringe concerns. They represent the standard operating baseline of the institutional private equity industry, refined over decades to reflect the GP's structural advantage in the negotiation.

The distinction that matters is between alignment that is expressed and alignment that is encoded. Expressed alignment is a posture. It describes how a firm characterizes its relationship with its investors in language that does not bind the firm to any particular outcome when the relationship is tested. Encoded alignment is a constraint. It specifies, in the operative clauses of the fund document, what the GP may do and may not do, under what conditions the LP has recourse, and what consequences follow when the GP departs from the framework the document establishes. Stewardship as Verifiable Governance develops this distinction in the context of portfolio company governance; the same logic applies at the fund level. A firm that does not encode its alignment claims in the operative clauses of the LPA has not made alignment claims. It has made marketing claims.

The test is simple, and it is the only test that produces verifiable results: what happens to the LP's rights and remedies when the GP's interests and the LP's interests diverge? If the fund document contains specific, enforceable answers to that question, alignment is encoded. If the fund document is silent, or if the answers live in side letters rather than the main instrument, alignment is aspirational. There is no third category.

What Real Alignment Looks Like

Seven provisions distinguish funds in which alignment is encoded from funds in which alignment is asserted. None of them is novel in isolation. What distinguishes a genuinely aligned fund is their simultaneous presence, drafted with sufficient specificity to function as constraints rather than statements of principle.

The first is no-fault removal. A no-fault removal right allows a supermajority of limited partners, typically sixty-six to seventy-five percent by interest, to remove the GP without demonstrating cause. The provision is the structural backstop of every other alignment mechanic in the document: if the GP can only be removed for cause, and if cause is defined narrowly, the LP's practical ability to enforce any other provision is limited by the cost and difficulty of establishing the factual predicate for removal. A no-fault removal right with a low threshold transfers meaningful governance authority to the LP body. Its absence signals a document that protects GP tenure over LP recourse.

The second is a GP catch-up waiver above a defined threshold. The conventional GP catch-up clause allows the GP to receive a disproportionate share of distributions after the LP has recovered its contributed capital and preferred return, until the GP has caught up to its carried interest percentage on total profits. In a standard European waterfall structure, this mechanic operates transparently. In an American waterfall structure with a deal-by-deal carry, the catch-up can be structured to benefit the GP significantly before loss-making investments in the same fund have been reconciled. A catch-up waiver above a defined outperformance threshold, the structure in which the GP foregoes further catch-up once the fund has exceeded a specified multiple or IRR, eliminates this asymmetry and aligns the GP's incremental economic interest with the LP's incremental return.

The third is hurdle structure and clawback mechanics. A preferred return rate that is set and maintained without soft-hurdle provisions, combined with a robust clawback obligation supported by escrow, is the mechanical foundation of carry-based alignment. Phalippou et al. (NBER w28083) document the distributional consequences of carry structures that permit GPs to extract economics from early fund vintages before the full portfolio has matured, and the evidence that LPs frequently cannot practically enforce clawback obligations when a fund has been wound down and carry has been distributed. A clawback supported by a funded escrow during the fund's active life converts a theoretically enforceable right into a practically enforceable one.

The fourth is the key-person provision, which specifies the named investment professionals whose continued involvement is material to the fund's investment mandate and what happens, typically a suspension of the investment period, if those individuals are no longer devoting sufficient time to the fund. The provision is frequently drafted in terms too vague to trigger: a key-person event that requires the departure of all named principals simultaneously, or that can be avoided by re-designating which individuals constitute the key-person group, does not constrain the GP. A functional key-person provision names specific individuals, defines a threshold for their involvement, and triggers meaningful consequences that do not require the LP to initiate litigation to enforce.

The fifth is GP commitment minimums as a percentage of aggregate commitments. A GP commitment expressed as a dollar minimum rather than a percentage of aggregate fund commitments is not an alignment provision. It is a floor that ceases to function as alignment at any fund size above the minimum. A GP commitment expressed as a percentage, typically one to three percent of aggregate commitments under ILPA Principles 3.0 guidance, maintains its alignment function across fund sizes because it scales with the GP's deployment of LP capital.

The sixth is side-letter parity. Side letters that grant economic or governance terms to certain LPs on a non-disclosed basis without offering equivalent terms to other LPs holding comparable commitments are not consistent with a genuinely aligned GP-LP relationship. A most-favored-nation clause with teeth, one that is self-executing rather than requiring notification and election, and that covers economic as well as reporting terms, is the structural mechanism by which the main instrument's alignment commitments are extended to the full LP body rather than to a preferred subset of it.

The seventh is LPAC consent rights. An LP Advisory Committee that holds consent rights rather than advisory rights over material fund actions, GP removal, valuation methodologies, conflicts of interest, fund extensions, and recycling of capital, functions as a governance body. An LPAC that holds advisory rights functions as a consultation mechanism that the GP can acknowledge and disregard. The distinction determines whether LP governance is embedded in the instrument or performed as a courtesy.

What 'LP-Friendly' Has Come to Mean — and Why It's Not Enough

The term "LP-friendly" entered the institutional vocabulary as a shorthand for the cluster of provisions that ILPA and large institutional investors began pushing for systematically in the post-2008 period, when the misalignment mechanics of the preceding fund vintage became visible at scale. It was a useful shorthand when it was first deployed. It described something real: a fund document that had been negotiated past the GP's standard template to include the provisions that the LP community's advocacy organizations had identified as best practice.

The term has since been marketing-grade. A fund described as LP-friendly in its pitch materials may include management fee offsets that apply only to monitoring fees and exclude transaction fees, a key-person provision that requires the departure of principals who hold no meaningful role in day-to-day investment decisions, an LPAC with consent rights over a list of conflicts that does not include the conflicts most likely to arise, and a clawback obligation that is theoretically robust but practically unenforceable because it is unsupported by escrow. Each of these provisions passes a superficial LP-friendliness test. None of them encodes the alignment the term implies. According to Preqin's LP terms benchmarking data, the gap between headline LP-friendly characterizations and the actual strength of individual provisions has widened as the term has become a standard marketing claim rather than a differentiating one.

The institutional investor community has responded to this drift by developing more granular evaluation frameworks. The ILPA Principles 3.0 framework addresses this by moving from general principles to provision-specific guidance with illustrative language, recognizing that alignment claims evaluated at the level of characterization are not evaluable. The shift reflects an institutional consensus that has been building across large allocators: that the unit of analysis for GP-LP alignment is the provision, not the fund, and that provision-level analysis requires legal review rather than marketing review. The Bain Global Private Equity Report has consistently noted that LP satisfaction with fund terms is most highly correlated with the specificity and enforceability of individual provisions rather than with the overall characterization of a fund as LP-friendly or otherwise.

The problem with "LP-friendly" as an evaluative standard is not that it describes the wrong things. It is that it describes them at the wrong level of resolution. Institutional allocators who accept the characterization without performing provision-level analysis are accepting a claim that the fund document itself may not support. The appropriate response is not to replace "LP-friendly" with a different characterization. It is to stop evaluating characterizations and start evaluating provisions.

The Stewardship Reading of a Term Sheet

A term sheet is a stewardship document before it is a financial document. The provisions that govern what a GP may do with LP capital, what governance rights LPs retain over the use of that capital, and what remedies LPs hold when the GP departs from the terms of their relationship are the provisions that determine whether the GP's stewardship of LP capital is a contractual obligation or a stated intention. Reading a term sheet as a stewardship document requires a different analytical frame from reading it as a financial document, because the questions it is designed to answer are different.

The financial reading asks: what is the economics of this fund? What are the fees, the carry, the preferred return, the distribution waterfall? These questions are important and have standard analytical answers. The stewardship reading asks: what are the limits on GP discretion? What triggers LP governance rights? What are the consequences of GP departure from the terms of the relationship? These questions are less frequently asked systematically, and their answers are less frequently in the form that allows comparison across funds, because the provisions that answer them are not standardized in the way that fee and carry provisions are.

The framework for a stewardship reading of an LPA proceeds paragraph by paragraph through the governance provisions of the document with four questions held in mind simultaneously. First: does this provision specify a consequence, or does it specify a principle? A provision that specifies a principle, that the GP will act in the best interests of the fund, that the GP will disclose material conflicts, is not a governance provision. It is a statement of intent that the document does not operationalize. A provision that specifies a consequence, that the LP Advisory Committee must consent before the GP may proceed, that the investment period is suspended upon a key-person event, that carried interest distributed above a threshold is subject to clawback supported by escrow, is a governance provision because it binds the GP to a defined outcome.

Second: is the provision self-executing, or does it require LP action to trigger? A provision requiring LPs to notify the GP, call a meeting, and vote to enforce their rights before those rights become operative transfers the cost of governance to the LP. A provision that operates automatically upon the occurrence of a defined event imposes that cost on the GP. The direction in which enforcement costs flow is one of the most revealing features of a term sheet read as a stewardship document, because it maps directly to the question of which party the document is designed to protect when the parties' interests diverge.

Third: is the provision bilateral, or does it include a GP carve-out? The most common form of apparent alignment provision is one that includes a carve-out that swallows the provision in the cases most likely to arise. A key-person provision that excepts situations in which the GP, in its reasonable judgment, has identified an adequate replacement is not a key-person provision. It is a consultation requirement. The carve-out test is a reliable filter for distinguishing provisions that constrain the GP from provisions that appear to constrain the GP while preserving GP discretion.

Fourth: is the provision auditable? As developed at length in Stewardship as Verifiable Governance, a commitment that cannot be verified cannot be enforced, and a commitment that cannot be enforced does no governance work. A provision is auditable if it specifies observable outcomes that can be measured against a defined standard on a defined cadence. A provision specifying that the GP will operate in accordance with the fund's governance principles is not auditable. A provision specifying that the GP will report compensation ratios across the portfolio on a defined schedule, subject to independent audit, is auditable. The difference is the difference between stewardship language and stewardship architecture. The analysis of Governance Compatibility as Investment Prerequisite demonstrates that this auditability test functions as a pre-commitment screening device as much as a post-hoc evaluation tool: a fund that cannot pass the auditability test at the term sheet stage is a fund whose governance architecture has not been designed to support the stewardship it claims.

Why Mechanical Tests Beat Reputational Heuristics

The institutional allocator community has historically evaluated GP-LP alignment through a combination of reputational heuristics and relationship continuity that is difficult to defend on analytical grounds. A GP with a strong track record, an established brand, and long-standing LP relationships receives the presumption of alignment because the historical relationship has not produced visible failures. The presumption is applied at the fund level, the manager is aligned, rather than at the provision level, this specific provision of this specific fund document encodes this specific constraint on this specific GP behavior. The difference between those two levels of analysis is the difference between trust and verification.

Reputational heuristics fail as evaluative tools for alignment in three systematic ways. First, they are backward-looking in a context where alignment failures are forward events. A GP's track record describes how it has behaved under historical conditions of capital availability, exit-market liquidity, and LP negotiating leverage. It does not constrain how it will behave under future conditions that differ from the historical baseline. The provisions of the fund document, by contrast, constrain future behavior directly and do not depend on the similarity of future conditions to historical ones.

Second, reputational heuristics are selectively presented. The GP controls the narrative of its own track record, selecting the vintages, the return metrics, and the comparison benchmarks that most favorably characterize its historical performance. Research from Harvard Business Review's analysis of PE governance structures documents the systematic gap between how GPs characterize their alignment terms and how those terms perform under legal review, finding that characterizations of alignment terms as LP-friendly frequently overstate the strength of individual provisions relative to ILPA benchmarks. The mechanical test, what does the provision actually say, and what does it actually require of whom under what conditions, is immune to selective presentation because its inputs are the operative language of the document rather than the GP's characterization of that language.

Third, reputational heuristics are subject to incumbency bias. An LP that has committed capital to a GP across multiple fund vintages has an economic and relational interest in the continued success of that GP. That interest creates a systematic tendency to read new fund documents charitably, to attribute deterioration in provision quality to market conditions rather than GP opportunism, and to accept the GP's characterization of its terms rather than conducting independent provision-level analysis. McKinsey's institutional PE research has documented the degree to which LP re-up decisions are influenced by relationship continuity rather than provision analysis, even among allocators with sophisticated governance frameworks. The mechanical test bypasses incumbency bias because it evaluates the current document against an external standard, ILPA Principles 3.0, the seven provisions described in the preceding section, the four-question stewardship framework, rather than against the GP's prior documents or prior characterizations.

The argument for mechanical tests is not that they replace judgment. It is that they produce the evidentiary base on which judgment can operate reliably. A legal review of a fund document that applies the stewardship reading framework described in the preceding section produces a provision-level map of where the GP has encoded alignment and where it has asserted it. That map is the input to a judgment about whether the fund, taken as a whole, encodes the alignment the GP claims. The map cannot be produced by reputational analysis. It can only be produced by reading the document.

Implications for The SAVI Capital Model

The SAVI Capital Model is an alignment architecture in the specific sense that the term requires: each of its four tenets is expressed as an operative LPA provision rather than as a governance principle, and each provision is written to satisfy the four-question stewardship test described above. The model's alignment claims are made at the level of the document, not at the level of the pitch.

Tenet 1, the equitable profit-sharing mechanic, is encoded as a distribution provision that bifurcates fifty percent of net corporate profits for equal distribution across all employees of the financed organization, separate from and additional to customary salaries, before the conventional profit waterfall distributes upward. The provision specifies a consequence, employees receive the distribution, rather than a principle. It is self-executing because the distribution is a defined obligation under the fund document rather than a discretionary action by fund management. It is bilateral without GP carve-out. And it is auditable because the distribution amounts are computable from the same financial statements that govern the balance of the waterfall. The structural logic of this mechanic is developed in the Tenet 1 analysis.

Tenet 2, the executive compensation ratio discipline, is encoded as an operating covenant that sets a maximum ratio of fifteen-to-one to twenty-to-one between the highest-paid executive and the lowest-paid worker in the financed organization, codified in the fund instrument rather than left to the discretion of a portfolio company compensation committee. The provision specifies a consequence, the ratio may not exceed the cap, rather than a principle. It is auditable because the ratio is computable from payroll data on a defined cadence. The encoding mechanics are described in Compensation Ratios, Encoded.

Tenet 3, verifiable stewardship, is encoded as a governance measurement and reporting framework that specifies the indicators, workforce stability, community condition, and institutional health on a five-to-ten-year horizon, against which portfolio company leadership decisions are evaluated, on the cadence the fund document requires, subject to independent audit, with defined escalation pathways for deterioration. The provision satisfies all four tests: it specifies consequences, it is self-executing in that reporting obligations arise automatically, it is bilateral without carve-out, and it is auditable because the indicators are observable and the reports are subject to external review. The LP holds the same enforcement authority over stewardship reporting as over financial reporting because both obligations arise under the same instrument. This is the operational definition of Stewardship as Verifiable Governance.

Tenet 4, the social-impact obligation above five times invested capital, is encoded as a distribution mechanic that redirects the GP's economic participation in returns above the defined outperformance threshold to The SAVI Ministries Endowment, converting extreme outperformance from a private GP benefit into a socially obligated one. The mechanic is self-executing, bilateral, and auditable because the trigger is a defined multiple of invested capital computable from fund-level financials. The design and rationale of this mechanic are analyzed in The Tenet 4 Mechanism.

Taken together, the four tenets constitute an LPA in which the GP's alignment claims are co-extensive with the GP's legal obligations. The allocator evaluating The SAVI Capital Model does not need to evaluate the GP's alignment posture, its culture, or its track record on governance matters, though those factors are available for review. It needs to read the four tenets as a lawyer reads a contract, applying the four-question stewardship test to each provision, and verifying that the provision passes. If it passes, the alignment claim is encoded. If it does not pass, the alignment claim is aspirational. The model is designed to pass.

Institutional allocators for whom GP-LP alignment is a gating criterion rather than a preference, allocators who require that alignment claims survive legal review at the provision level rather than marketing review at the fund level, are the allocators for whom the model is built. The model does not ask those allocators to trust the GP's characterization of its terms. It asks them to read the terms. The difference between those two requests is the difference between stewardship as a posture and stewardship as a discipline. Allocators prepared to make that distinction are invited to request the fund documentation and apply the test themselves.