For nearly fifty years, partners who held a "limited partner" interest under state law treated their share of partnership profits as outside the reach of self-employment tax. They reported a salary-like guaranteed payment on Schedule SE, paid 15.3% on that slice, and walked away from the rest. Then in late 2023 the U.S. Tax Court told an investment firm called Soroban Capital Partners that the words "limited partner, as such" in Section 1402(a)(13) were not a state-law label — they were a fact pattern. In May 2025, after a full trial, the Tax Court applied that test and held that Soroban's three principal partners owed self-employment tax on tens of millions of dollars of distributive share they had been excluding for years.
If you are a fund manager, a service partner in a private equity firm, a law firm equity holder, or a physician in a state-law LP, the floor under your old planning strategy has moved. Here is what changed, how the new test works in practice, and what to do about it before your next K-1 cycle.
A Forty-Year-Old Statute That Was Never Defined
Section 1402(a)(13) was written in 1977, when the limited partnership was almost exclusively an investment vehicle. The provision excludes from net earnings from self-employment "the distributive share of any item of income or loss of a limited partner, as such, other than guaranteed payments described in section 707(c)." The phrase "as such" was tacked on at the last minute. Nobody outside Congress wondered what it meant for forty years because, in 1977, the working assumption was that limited partners were retired dentists in oil-and-gas tax shelters, not the people running the fund.
In 1997, the IRS tried to draw the line. It proposed regulations under Reg. 1.1402(a)-2 that would have used a three-factor test based on management authority, personal liability, and time devoted. The proposal triggered a political firestorm. Congress passed Section 935 of the Taxpayer Relief Act of 1997, imposing a moratorium that barred Treasury from issuing temporary or final regulations on the subject through July 1, 1998. The moratorium was lifted on schedule, but the proposed regulations were never finalized. For the next quarter-century, partnerships and the IRS argued in a doctrinal vacuum.
The vacuum slowly filled with case law. In Renkemeyer, Campbell & Weaver, LLP (136 T.C. 137, 2011), the Tax Court rejected the limited partner claim of three attorneys in a Kansas LLP, holding that Congress meant to exclude only passive investors — people whose earnings were "in the nature of a return on investment" rather than payment for personal services. Castigliola (T.C. Memo. 2017-62) applied the same logic to LLC members in a Mississippi law firm. Both cases involved professionals, both involved pass-through entities that were not even state-law LPs, and both stopped short of resolving the harder case: a real state-law LP whose limited partners actually do work.
That case arrived as Soroban.
Soroban I: The Tax Court Picks the Test (2023)
Soroban Capital Partners LP is a Delaware-domiciled investment manager. Three principals — Eric Mandelblatt, Gaurav Kapadia, and Scott Friedman — held interests directly and indirectly through Soroban Capital Partners GP LLC, the general partner. Each principal received a Section 707(c) guaranteed payment for services and reported self-employment tax on that portion. The principals also received massive allocations of ordinary business income through their limited partner interests, and that distributive share — totaling more than $140 million across 2016 and 2017 — was excluded from SECA under Section 1402(a)(13).
The IRS adjusted Soroban's returns and Soroban filed a petition. In November 2023, ruling on cross-motions for partial summary judgment, the Tax Court (161 T.C. No. 12) held that whether a partner qualifies as a "limited partner, as such" cannot be decided on state-law status alone. The phrase "as such," the court reasoned, signals that Congress was looking at substance, not form. A "functional analysis" — the same test introduced in Renkemeyer — would have to be applied. The court denied summary judgment to both sides and set the case for trial.
The decision quietly rewrote the planning landscape for every fund manager in the country. A state-law LP designation was no longer a safe harbor. The question became: what does the functional test actually require?
Soroban II: The Functional Analysis in Practice (2025)
The trial happened in 2024 and the Tax Court issued its opinion on May 28, 2025, as T.C. Memo 2025-52. After a 100-page record review, the court held that none of the three principals qualified as limited partners for Section 1402(a)(13) purposes. Their full distributive share was reclassified as self-employment income.
The court declined to draw a fixed multi-factor checklist. Instead, it announced that the functional test takes account of all relevant facts and circumstances, organized around three lines of inquiry:
- The source of the partnership's income. Where does the money come from? Is it generated by capital that has been put at risk, or by the labor of the partners?
- The roles of the partners in generating that income. Are the partners' time, skills, and judgment essential to producing the partnership's revenue? Do they negotiate trades, manage portfolios, sign on behalf of the entity, hire and fire employees, sit on investment committees, or represent the firm externally?
- The relationship between distributive share and capital contribution. Are the allocations economically consistent with a return on invested capital, or are they really compensation that has been routed through a "limited partner" interest?
Applied to Soroban, the facts were brutal for the principals. They treated the firm as their full-time job, logging 2,300 to 2,500 hours per year. They were contractually required to devote full attention to the firm and were prohibited from pursuing outside investment opportunities. Two of the three principals contributed zero capital. The firm's marketing emphasized the personal expertise of these three individuals as the reason clients invested. The court concluded that the distributive share was a payment for services rendered, dressed in limited partner clothing.
A second Tax Court opinion delivered around the same time, Denham Capital Management, LP (T.C. Memo. 2024-114), reached the same conclusion on factually similar grounds against another investment manager. Together they marked the IRS Compliance Campaign on the SECA limited partner exception as a multi-front offensive, not a one-off challenge.
Who Is at Risk?
The functional test sweeps far beyond hedge funds. Any organization that books distributive share allocations to "limited partner" interests held by people who actually run the business should reassess. That includes:
- Hedge funds and private equity managers structured as state-law LPs with partner-level investment committees, portfolio managers, or trading principals.
- Real estate sponsors and developers where general partner economics flow through tiered LP interests held by working partners.
- Law firms organized as LPs or LLPs with equity partners managing matters, supervising associates, and originating business.
- Medical, dental, and veterinary practices organized as multi-owner LPs where the doctors are also the partners.
- Consulting and accounting firms where senior partners draw modest guaranteed payments and large profit allocations.
- Family investment LPs when the family members operate the business rather than passively holding it.
In short, if there is daylight between the labor that produces the partnership's income and the "limited partner" who collects it, the IRS now has a credible argument that the daylight does not exist.
The Numbers That Make This Painful
Self-employment tax matters because the rates are not trivial. In 2026, SECA is 15.3% on the first $184,500 of net earnings from self-employment (12.4% Social Security plus 2.9% Medicare), then 2.9% Medicare on every additional dollar. Add the 0.9% Additional Medicare Tax on combined earned income over $200,000 (single) or $250,000 (joint), and the marginal rate on a partner's distributive share above the wage base sits at 3.8%. On a $5 million allocation, that is $190,000 of tax that previously was not on the table — per partner, per year — plus interest and potentially Section 6662 accuracy-related penalties if the position is found to lack substantial authority going forward.
For most investment firms, the Soroban exposure runs into eight or nine figures across the open statute period.
What Still Works
The Tax Court did not abolish Section 1402(a)(13). It tightened the definition of who gets to claim it. Genuinely passive limited partner interests still qualify. The opinion offers four directional planning takeaways:
- Separate the labor from the capital. Where a partnership has both active managers and passive investors, hold the limited partner interest in a separate entity that does no work and receives no compensation. The labor is paid through W-2 wages from a management company or through Section 707(c) guaranteed payments; the limited partner interest reflects only invested capital.
- Make the capital meaningful. Distributive share that looks proportionate to capital contributed, holding period, and risk borne is harder to recharacterize. Allocations that are wildly disproportionate to contributed capital are a red flag.
- Document the absence of services. If a partner is truly passive, the partnership agreement, operating procedures, board minutes, and committee rosters should show it. Hours logs, email patterns, and management role assignments will all become discovery exhibits if the IRS opens an examination.
- Treat Section 707(c) honestly. Guaranteed payments for services are still subject to SE tax under both old and new law. The temptation to compress salary into a small guaranteed payment and shift the rest to "limited partner" distributive share is exactly the structure the Soroban court took apart.
What does not work: relabeling, restructuring solely on paper, or pointing to the state-law definition. The functional test looks at what people actually do.
Bookkeeping and K-1 Implications
Partnerships have to decide, every year, how to report each partner's share of self-employment earnings in box 14 of Schedule K-1 with code A (net earnings/loss from self-employment). The IRS revisited the draft 2022 Form 1065 instructions to add language requiring partnerships to apply the Renkemeyer functional test; the language was removed before the final version was published, leaving the reporting decision in the partnership's hands.
Operationally, that means three concrete bookkeeping changes for any partnership with limited partner activity at risk:
- Time-tracking by partner. Even rough records of hours by partner, role, and activity become valuable evidence of passivity or activity. This is the single most important record the Tax Court looked at in Soroban.
- Capital account hygiene. Maintain accurate tax-basis capital accounts with clear contribution and distribution histories. A reviewing agent will compare cumulative contributions to cumulative distributive share. Wide gaps invite recharacterization.
- Compensation-vs.-allocation reconciliation. Document the methodology that decides what goes through a Section 707(c) guaranteed payment versus a distributive share. The decision should track the work the partner did, not the SE tax result the partner wanted.
The partnership's accounting records have to support these distinctions before the K-1 is filed, not after the IDR arrives. Plain-text, version-controlled financial records make it dramatically easier to reconstruct contribution histories, hours by partner, and the calculation behind a guaranteed payment when an audit lands two or three years later.
The Litigation Is Not Over
The Soroban principals have appealed to the Second Circuit, and at least one parallel taxpayer obtained a favorable Fifth Circuit ruling that resets the test on slightly different grounds. Until the Supreme Court resolves a circuit split or Congress amends the statute, partnerships outside the Second Circuit may have litigation leverage that Soroban itself lacked. But that leverage is exactly that — leverage in audit. Filing positions today on the assumption that Soroban will be reversed is a higher-risk bet than rebuilding the structure on supportable ground.
Keep Your Partnership Books Audit-Ready From Day One
The Soroban era is going to be won and lost on documentation. Partnerships that can produce clean capital account histories, defensible time records, and a coherent narrative tying distributive share to invested capital will negotiate from a position of strength. Partnerships that cannot will be writing checks with interest. Beancount.io provides plain-text accounting that gives you complete transparency over every contribution, allocation, and guaranteed payment — version-controlled, queryable, and exportable to whatever your tax counsel needs to see. Get started for free and build the kind of partnership records that hold up under a functional-analysis review.