This is Part 3 of a four-part Youya Wealth series on tax-aware long-short investing. Part 1 explained why conventional loss harvesting runs out of raw material, and Part 2 walked through how a long-short extension keeps manufacturing capital losses — losses that are enormously valuable against capital gains but nearly inert against a salary. This installment examines the more aggressive limited-partnership variant: structures that aim to change the character of the loss itself, from capital to ordinary, using monthly-reset total return swaps and partnership-level trader status. The building blocks are each settled law; the combination is not. We explain the machinery neutrally, and we give equal time to the reasons the IRS may not view it charitably.
The Ceiling on Capital Losses
Character is destiny in the tax code. Everything in Parts 1 and 2 of this series operated inside a single tax category: capital gain and capital loss. Within that category, the netting rules are generous — under sections 1211(b) and 1212(b) of the Code, capital losses offset capital gains dollar for dollar without limit, and unused losses carry forward indefinitely, retaining their short- or long-term character. That is what made the Part 2 outcome possible: a stockpile of harvested losses met a $6,000,000 business-sale gain and erased roughly $1,428,000 of federal tax — about 23.8 cents per dollar of loss at the top 20 percent long-term rate plus the 3.8 percent net investment income tax.
But the category has a wall. Once capital losses exceed capital gains, only $3,000 per year ($1,500 married filing separately) may be deducted against ordinary income — wages, bonuses, interest, or the operating income flowing through on a K-1 from a business. The figure is statutory, confirmed in IRS Topic No. 409, and it has not been indexed for inflation; it has sat at $3,000 since 1978. At a 37 percent top ordinary rate, the annual value of that crossover allowance is roughly $1,110. For an executive earning seven figures in W-2 income, or a business owner with substantial K-1 operating income and no large capital gains on the horizon, a mountain of harvested capital losses is close to worthless in the near term.
Hence the second product. The separately managed account format described in Part 2 cannot solve this, because an SMA investor directly owns the positions and realizes losses in whatever character the underlying property produces — which, for stocks held as capital assets, is capital. Solving the character problem requires different instruments and a different wrapper. That is the origin of the limited-partnership variant of tax-aware long-short (TALS): a private fund, typically offered under Section 3(c)(7), that pursues losses which arrive on the K-1 as ordinary items — the kind that can, within limits we cover in Part 4, offset wages and operating income directly.
Enter the Swap: Notional Principal Contracts
The instrument of choice is the total return equity swap. Instead of owning shares, the fund enters a contract with a dealer: the fund receives (or pays) the total return of a stock or basket — price change plus dividends — on an agreed notional amount, and pays (or receives) a financing rate in return. Economically this resembles a leveraged position in the underlying. For tax purposes, however, it lives under an entirely different regime. Treasury Regulation 1.446-3 classifies equity swaps and equity index swaps as notional principal contracts (NPCs), a category with its own timing and character rules that do not track the rules for owning stock.
The regulation draws one line that matters more than any other. Payments under an NPC come in flavors. Periodic payments — amounts payable at intervals of one year or less, calculated by reference to a specified index and the notional amount — are recognized ratably on a daily basis and produce ordinary income or ordinary deductions. Nonperiodic payments are amortized over the contract's term and are likewise ordinary. Termination payments — amounts paid to extinguish, assign, or exchange the contract — are different: Regulation 1.446-3(h) supplies the timing (recognized in the year of termination), but the character generally comes from section 1234A, which treats gain or loss from terminating rights with respect to capital-asset property as capital.
Read that line again, because the entire structure hangs on it. The same economic result — the equity leg of a swap moved against you — can be an ordinary deduction or a capital loss depending on whether the payment that settles it is a periodic payment under the contract's own schedule or a payment to terminate the contract. Ordinary on the reset; capital at the exit. Tax engineering is usually about timing. This is about character, which for a high-earner is worth far more: an ordinary deduction can be worth 37 cents or more per dollar against wages, while a capital loss without offsetting gains is worth $3,000 a year, total.
The Monthly Reset and the Character Split
The design writes the periodic payment into the contract. The LP-format funds pursuing ordinary losses typically use swaps that reset monthly: each month, the appreciation or depreciation of the equity leg is settled in cash as a scheduled periodic payment, and the notional re-strikes at the new level. Because the equity performance flows through those monthly settlements rather than accruing silently to a final termination, substantially all of the swap's economics — including the losses — arrive as periodic payments, which the regulation taxes as ordinary deductions. A losing swap position that is simply allowed to keep resetting therefore bleeds ordinary losses month by month.
The mirror image completes the arbitrage. Now consider the winning legs. If a winning swap were allowed to reset, its gains would be periodic payments too — ordinary income, which would neutralize the ordinary deductions. So the structure does the opposite: winning positions are terminated before the reset date, or the equivalent exposure is held in directly owned securities rather than swaps. Termination payments take capital character under section 1234A; directly owned appreciated stock produces capital gain when sold — or, better, is simply held, deferring the gain the way Part 2 described. The result, in a structure that works as designed, is a systematic character split: losses ordinary, gains capital (and deferred). The investor deducts against 37 percent income now and recognizes at 23.8 percent — or, via charitable transfer or basis step-up at death, potentially never — later.
Name it plainly: this is character arbitrage. Plain-vanilla loss harvesting, and even the SMA form of TALS, only accelerates capital losses within the capital category. The LP swap structure attempts something categorically more aggressive — moving the loss side of a hedged book into the ordinary basket while keeping the gain side in the capital basket. Each individual step cites real authority: the NPC regulation genuinely says periodic payments are ordinary; section 1234A genuinely says terminations are capital. But no statute, regulation, or ruling blesses the combined result, and the taxpayer who claims it is relying on a literal reading of rules that were not written with this trade in mind. That is a tax position, not a tax rule — and it should be priced accordingly.
Why This Design Invites Scrutiny
Start with economic substance. Section 7701(o) codifies a conjunctive two-prong test: a transaction is respected only if it changes the taxpayer's economic position in a meaningful way apart from federal tax effects, and the taxpayer has a substantial non-tax purpose for entering it. A book of offsetting long and short exposures whose net economic risk is small relative to the tax benefits generated is close to the doctrine's core target. The stakes are not merely losing the deduction: failure triggers a strict-liability accuracy penalty under section 6662(b)(6) — 20 percent of the underpayment, 40 percent if undisclosed — with no reasonable-cause defense. The IRS's guidance in Notice 2010-62 and its recent willingness to apply the doctrine to partnership structures in Rev. Rul. 2024-14 suggest this is not a theoretical tool.
Second, the periodic/termination line itself can be attacked. The ordinary treatment depends on monthly resets qualifying as periodic payments. An examiner could argue that a series of engineered resets is, in substance, a series of terminations, or that the arrangement amounts to disguised ownership of the underlying securities. The IRS has made exactly this style of argument before: basket option contracts, which similarly interposed a derivative between a taxpayer and a managed securities portfolio to transform character and timing, were declared listed transactions in Notices 2015-73 and 2015-74. The anti-abuse and embedded-loan rules within Regulation 1.446-3(g) hand examiners additional tools to recharacterize NPC payments that do not reflect economic substance.
Third, the straddle rules lurk over any paired book. Section 1092 defers losses on offsetting positions in actively traded personal property to the extent of unrecognized gain in the offsetting leg, and can suspend holding periods and convert long-term gains to short-term. A structure holding long exposure against short exposure — or swap exposure against physical shares — with substantially diminished risk of loss fits the statutory description uncomfortably well. If section 1092 applies, harvested losses are deferred, not accelerated, which does not merely trim the strategy's benefit; it inverts it. A balanced reading: funds with genuinely diversified, actively traded books and real net exposure have credible answers to all three attacks, and the underlying authorities are established law. But the answers are factual, the IRS has publicly signaled awareness of ordinary-loss swap structures, and under the centralized partnership audit regime an adjustment at the fund level can sweep in every limited partner at once.
Trader Status at the Partnership Level
Ordinary character alone is not enough — the loss must also be a business loss. For the K-1 items to function as ordinary business deductions (and, as Part 4 will show, to convert into a net operating loss when limited), the partnership must be engaged in a trade or business under section 162 — specifically, it must be a trader in securities rather than a mere investor. IRS Topic No. 429 frames the test: the taxpayer must seek to profit from daily market movements in prices rather than from dividends, interest, or long-term appreciation; the activity must be substantial; and it must be carried on with continuity and regularity. Courts weigh typical holding periods, the frequency and dollar volume of trades, and the time devoted to the activity.
The case law shows how easily taxpayers fail. In Chen v. Commissioner, T.C. Memo. 2004-132, the taxpayer executed 323 securities transactions in a year — but roughly 94 percent of them fell within three months, and the Tax Court held that trading concentrated in one quarter of the year was sporadic rather than frequent, regular, and continuous, denying trader status. In Endicott v. Commissioner, T.C. Memo. 2013-199, a retired executive running a covered-call strategy executed as many as 1,543 trades in a year, yet was held to be an investor because he held the underlying stock an average of 35 days: holding periods of one to five months showed he was not trying to catch daily swings. Volume alone does not confer trader status; the pattern and intent of the trading do.
For an LP, the test runs at the fund level — and cuts against tax-motivated design. A limited partner need not trade personally; if the partnership itself conducts substantial, frequent, regular, continuous trading, its expenses are section 162 business deductions and its results flow through to partners in that trade-or-business context. Most institutional TALS funds turn over their books heavily, which helps. But there is an underappreciated tension here: Chen and Endicott require an intent to profit from market movements, and a fund whose positions are visibly managed to harvest tax character — losers held to reset, winners terminated on a tax calendar — hands the IRS an argument that the trading pattern serves the tax result rather than a profit motive. If trader status falls, the ordinary-business-loss claim falls with it, and the deductions may be severely limited at the partner level.
Non-Passive by Regulation: The Section 469 Trading Exception
One trap these structures genuinely do escape. An investor's instinct is that a loss from a partnership in which they are a passive limited partner must be a passive loss, usable only against passive income under section 469. Trading partnerships are the exception, and it is written directly into the regulations: Temporary Regulation 1.469-1T(e)(6) provides that an activity of trading personal property — stocks, securities, and derivatives, per the section 1092(d) definition — for the account of the owners of interests in the activity is not a passive activity, regardless of whether it rises to a trade or business. The regulation's own example is a partnership trading securities for its partners' accounts.
This cuts in both directions. For the loss-seeking investor, it means the ordinary trading losses flowing through the K-1 are not trapped in the passive basket; they reach the rest of the return, subject to basis, at-risk, and the excess business loss limitation. But symmetry applies: trading income from such a partnership is non-passive income that cannot be sheltered by passive losses from other activities, such as real estate. And escaping section 469 does not mean escaping limitation altogether — it means the loss proceeds directly to the next gate. Under the ordering rule of section 461(l)(6), the excess business loss limitation applies after section 469, which is precisely why section 461(l), with its $256,000 single and $512,000 joint thresholds for 2026, is the operative constraint for these structures. That is Part 4's subject.
Who Can Invest: The Qualified Purchaser Gate
The wrapper dictates the audience. The SMA format of TALS is generally available to accredited investors. The LP variant is not: funds relying on the Section 3(c)(7) exemption under the Investment Company Act may admit only qualified purchasers — under section 2(a)(51) of the 1940 Act, individuals (or married couples) owning at least $5 million in investments, excluding the primary residence and business-use property, with family-owned companies needing $5 million and most other entities $25 million. These funds also typically impose lock-ups and deliver K-1s that arrive late in filing season. The gating is not incidental — it reflects a regulatory judgment that leveraged, derivative-heavy private funds with contested tax positions belong only with investors who can absorb both the market risk and the tax-controversy risk. In our planning work with business owners weighing these structures, the suitability question usually resolves the product question before the tax analysis does.
Diligence here is tax diligence. Evaluating a conventional fund means evaluating fees, strategy, and manager skill. Evaluating an ordinary-loss TALS LP additionally means evaluating a legal position: what fraction of the book runs through swaps, how the reset and termination decisions are actually made, whether the fund has real net economic exposure and a credible profit motive, what the offering documents say about economic substance and straddle risk, and who bears the cost if the IRS adjusts the partnership. Those questions belong in front of independent tax counsel and an adviser with no economics in the placement — which is the posture we take in our tax and estate planning work. A structure whose sponsor cannot answer them crisply has answered them.
The Bridge to Part 4: What the Loss Is Worth
Assume everything holds. Suppose the swaps are respected as NPCs, the fund sustains trader status, the straddle rules do not bite, and a large ordinary business loss lands on your K-1 — non-passive, courtesy of the section 469 trading exception. The intuition is that a $1,500,000 ordinary loss now wipes out $1,500,000 of salary. It does not. Congress built a final gate for exactly this fact pattern: the excess business loss limitation of section 461(l), which walls W-2 wages off from the business-loss computation entirely and caps the net business loss usable against non-business income at $512,000 for joint filers in 2026 — a provision the One Big Beautiful Bill Act made permanent. The disallowed excess is not lost; it converts into a net operating loss with its own timing rules and its own 80 percent cap. Part 4 walks the full computation, dollar by dollar, and shows what a seven-figure ordinary loss is actually worth — and when.
Sources & Further Reading
IRS — Topic No. 409, Capital Gains and Losses
IRS — Topic No. 429, Traders in Securities
Legal Information Institute (Cornell) — 26 U.S.C. § 1211, Limitation on Capital Losses
Legal Information Institute (Cornell) — 26 U.S.C. § 1212, Capital Loss Carrybacks and Carryovers
Legal Information Institute (Cornell) — Treas. Reg. § 1.446-3, Notional Principal Contracts
Legal Information Institute (Cornell) — 26 U.S.C. § 1234A, Gains or Losses From Certain Terminations
Legal Information Institute (Cornell) — Temp. Treas. Reg. § 1.469-1T, General Rules (Passive Activity Losses)
Legal Information Institute (Cornell) — 26 U.S.C. § 1092, Straddles
Legal Information Institute (Cornell) — 26 U.S.C. § 7701, Definitions (Including § 7701(o), Economic Substance Doctrine)
IRS — Notice 2010-62, Interim Guidance Under the Codification of the Economic Substance Doctrine
IRS — Rev. Rul. 2024-14, Application of the Economic Substance Doctrine
U.S. Tax Court via CourtListener — Chen v. Commissioner, T.C. Memo. 2004-132
Iowa State University Center for Agricultural Law & Taxation — Endicott v. Commissioner, T.C. Memo. 2013-199 (case annotation)
Journal of Accountancy — Tax Matters: Trader or Investor? The Endicott Decision
Legal Information Institute (Cornell) — 15 U.S.C. § 80a-2, Definitions (including § 80a-2(a)(51), Qualified Purchaser)
SEC Investor.gov — Updated Investor Bulletin: Accredited Investors
We help business owners and executives evaluate tax-aware long-short strategies on their merits — modeling the tax math, the leverage risk, and the exit — with no product to sell and no commissions to earn.
