This is Part 4 — the final and most technical installment — of a four-part Youya Wealth series on tax-aware long-short investing. In Part 3 we traced how certain limited partnership structures aim to deliver ordinary losses rather than capital ones: notional principal contracts under Reg. 1.446-3, trader status under the Chen and Endicott line of cases, and the trading-partnership exception that lifts those losses out of the passive activity rules. That article ended at a gate. A loss that escapes section 469 does not flow straight onto a tax return — it must first pass through section 461(l), the excess business loss limitation, and whatever is disallowed there re-emerges as a net operating loss governed by section 172's own set of restrictions. This article walks that gauntlet precisely, with a two-year worked example: a couple with $800,000 of wages, $300,000 of K-1 business income, and a $1,500,000 ordinary loss from a tax-aware long-short partnership. The punchline is that the loss is neither wasted nor worth what intuition suggests. It is worth roughly $555,000 of federal tax offset — spread over two years, at the mercy of thresholds Congress just made permanent, and only if every characterization underneath it survives.
The Story So Far
Three articles of setup. In Part 1 we explained why long-only tax-loss harvesting decays as portfolios appreciate, and how tax-aware long-short (TALS) strategies keep a loss inventory alive by holding opposing long and short books. In Part 2 we walked the separately managed account version, where the output is capital losses — enormously valuable against a capital gain, nearly worthless without one. Part 3 covered the more aggressive LP variant: a trader partnership using monthly-reset total return swaps so that losses arrive as ordinary deductions, usable in principle against wages and other ordinary income at rates up to 37 percent.
Why ordinary losses face their own wall. Congress anticipated exactly this fact pattern — a high earner using flow-through business losses to shelter a large salary — and built a purpose-made limitation for it in the 2017 tax act. Section 461(l) does not care whether your loss is economically real, properly characterized, or blessed by trader status. It asks a simpler, colder question: how much aggregate business loss are you trying to deduct against income that is not business income? Past a fixed dollar threshold, the answer is: not this year.
How Section 461(l) Actually Works
The test in one sentence. Each year, a noncorporate taxpayer totals the deductions attributable to all of their trades or businesses and compares that figure to the sum of gross income and gains from all of those trades or businesses; a net loss can offset non-business income only up to an inflation-adjusted threshold, and everything beyond the threshold is an excess business loss — disallowed for the current year. For taxable years beginning in 2026, Rev. Proc. 2025-32 sets the threshold at $256,000 for single filers and $512,000 for joint returns. Notably, those figures are lower than 2025's, because the One Big Beautiful Bill Act reset the inflation indexing to the original $250,000 statutory base.
The limitation is now permanent. Section 461(l) was originally scheduled to sunset after 2028. OBBBA, signed July 4, 2025, struck the sunset entirely: for tax years beginning after December 31, 2025, the excess business loss limitation is a permanent feature of the code. Any planning model that assumed the wall would eventually come down needs to be rerun.
Wages are not business income — this is the trap. Section 461(l)(3)(B) requires the computation to be made without regard to any income, gains, or deductions attributable to the trade or business of performing services as an employee. A W-2 salary, however large, sits entirely outside the business column. The consequence is easy to state and easy to forget: a high-wage earner cannot use business losses to shelter wages beyond $256,000 (single) or $512,000 (joint) in 2026, no matter how large or how genuine the loss.
Aggregation cuts both ways. The test is run across all of the taxpayer's trades or businesses combined, not activity by activity. Business income from one venture — an operating company, a profitable consulting practice, a spouse's material-participation K-1 — absorbs business losses from another before the threshold is ever consulted. That aggregation can rescue a loss (business income soaks it up with no limit) or squander capacity (a loss you wanted against wages is first consumed by business income that might have been taxed favorably anyway). Multi-entity owners need to model the whole return, not one K-1.
Ordering: 461(l) comes after 469. Under section 461(l)(6), the excess business loss test applies after the passive activity rules of section 469. A passive loss trapped in the passive basket never reaches 461(l) at all. But — as Part 3 explained — trading partnerships are carved out of the passive regime by Temp. Reg. 1.469-1T(e)(6), so their losses sail past section 469 and land directly on the 461(l) test. For TALS LP investors, 461(l) is not one limitation among several; it is the operative limitation.
The NOL Conversion: Section 172
Disallowed does not mean destroyed. The excess business loss is not forfeited. Under sections 461(l)(2) and 172(b), it is treated as a net operating loss carryover to the following taxable year. From that moment forward it lives under section 172's rules, which for post-2017 losses are a specific bargain: generous in one dimension, stingy in two others.
The three rules that matter. First, post-2017 NOLs carry forward indefinitely — the old 20-year expiration is gone, so the loss never dies on the shelf. Second, they generally cannot be carried back: last year's tax is settled, and the loss can only look forward (narrow exceptions exist for certain farming and insurance losses). Third, and most binding for planning purposes, the NOL deduction attributable to post-2017 losses is capped at 80 percent of taxable income, computed without regard to the NOL deduction itself (and without the section 199A and 250 deductions). An NOL can therefore never fully zero out a future year's taxable income by itself — one dollar in five of the absorbing year's income always remains taxable until the carryover is exhausted.
The practical translation. For the TALS LP investor, a loss beyond the 461(l) threshold is a deduction on layaway: it comes back next year (or later), it never expires, but it arrives subject to the 80 percent cap and only after a real time-value haircut. Whether that trade is acceptable depends entirely on the income it eventually lands against — which is why the worked example matters.
Year One: The 461(l) Walk
The facts. A married couple files jointly for tax year 2026. Spouse A earns $800,000 of W-2 wages. Spouse B receives a K-1 allocating $300,000 of ordinary income from an operating business in which B materially participates, so the income is non-passive. The couple also holds an interest in a tax-aware long-short limited partnership that qualifies as a trader partnership; in Year 1 it allocates them a $1,500,000 ordinary trading loss. Assumptions, stated explicitly: the loss is ordinary in character (for example, via a section 475(f) mark-to-market election at the fund level); we ignore the standard deduction, QBI, AMT, state tax, and all other items; whole dollars throughout.
Step one — sort the columns. The $800,000 of wages is excluded from the business column by section 461(l)(3)(B). The spouse's $300,000 of K-1 operating income is business income. The LP's $1,500,000 ordinary trading loss is a business deduction, and because the trading-partnership exception keeps it out of the passive basket, it actually reaches this test rather than being trapped upstream.
Step two — net the business column. The aggregate net business loss is $1,500,000 minus $300,000, or $1,200,000. Note what just happened: the spouse's operating income silently absorbed $300,000 of the LP loss before any threshold entered the picture. That is aggregation working in the couple's favor.
Step three — apply the threshold. Of the $1,200,000 net business loss, only $512,000 — the 2026 joint-filer threshold — may cross over to offset non-business income, meaning the wages.
Step four — compute the excess. The excess business loss is $1,200,000 minus $512,000, or $688,000. That amount is disallowed for Year 1 and converts to an NOL carryover to Year 2 under sections 461(l)(2) and 172(b).
Step five — Year 1 adjusted gross income. AGI (before the items we are ignoring) is $800,000 of wages plus $300,000 of K-1 income minus the $812,000 of allowed business loss — the $300,000 absorbed by business income plus the $512,000 threshold amount — which nets to $288,000. The shortcut confirms it: $800,000 of wages minus the $512,000 cross-over allowance equals $288,000. Sit with that result for a moment. The couple lost $1,500,000 economically and still reports roughly $288,000 of AGI, because the statute walls their wages off from the loss.
Year Two: The 80 Percent Cap
The carryover comes home. Assume Year 2 repeats the income picture with no new losses: $800,000 of wages and $300,000 of K-1 income, for taxable income before the NOL deduction of $1,100,000 (again ignoring the standard deduction and all other items, as assumed).
Apply section 172(a)(2). The NOL deduction is capped at 80 percent of taxable income computed without the NOL itself: $1,100,000 times 0.80 equals $880,000. The couple's $688,000 carryover is comfortably below that ceiling, so it deducts in full. Year 2 taxable income is $1,100,000 minus $688,000, or approximately $412,000 (a real return would layer in the deductions we are deliberately ignoring).
What if the carryover had been larger? Had it exceeded $880,000, the excess would simply have carried forward again — indefinitely, but always subject to the 80 percent cap in each absorbing year. A large enough NOL rides along for years, chipping away at income it can never fully eliminate. That is the deliberate design of post-2017 section 172.
What the Loss Is Actually Worth
The two-year picture. Nothing was permanently lost — only delayed. Year 1 used $812,000 of the loss (the $300,000 that offset the spouse's business income plus the $512,000 threshold amount allowed against wages); Year 2 used the remaining $688,000 through the NOL deduction. The check ties: $812,000 plus $688,000 equals $1,500,000. Combined two-year taxable income is $288,000 plus $412,000, or $700,000 — versus $2,200,000 ($1,100,000 in each year) had there been no loss at all. The full $1,500,000 reduction eventually lands.
The headline number, with its caveat attached. At the 37 percent top ordinary bracket, $1,500,000 of deductions is worth approximately $555,000 of federal tax offset over the two years. That figure is an explicit approximation: it ignores bracket effects — portions of the deduction land in lower brackets, especially in Year 1, where AGI drops to $288,000 — and it ignores the standard deduction, QBI, AMT, and state tax, so the true benefit differs somewhat. And part of the benefit arrives a year after the economic loss, a genuine time-value cost. The honest summary: a $1,500,000 ordinary loss on this fact pattern is worth roughly $555,000 of federal tax, delivered on a delay, not the $555,000-in-one-April that a naive model assumes — and emphatically not a $1,500,000 reduction of current-year salary.
Planning Around the Gauntlet
Aim losses at business income, not wages. The threshold only constrains the loss's crossing into non-business income. Business income absorbs business losses without limit — which is why the couple's $300,000 K-1 was sheltered in full before the threshold ever applied. For business owners approaching an exit, this reframes the timing question: ordinary income recognized as part of a sale — depreciation recapture, for instance, or other ordinary-character components of a business disposition — may sit in the business column where a trader-partnership loss can meet it head-on, while the same loss aimed at salary hits the $512,000 wall. The analysis is fact-specific and belongs in the hands of tax counsel, but the difference in usable deduction can be enormous.
Roth conversions and the NOL year. A year in which a large NOL carryover is absorbing income at the 80 percent cap — or in which 461(l) has already crushed AGI, as in our Year 1 at $288,000 — can be an unusually attractive window for recognizing income voluntarily: a Roth conversion, an elective gain recognition, an acceleration of deferred compensation where permitted. Coordinating loss flows with deliberate income events is precisely the kind of multi-year sequencing that integrated tax and estate planning exists to orchestrate; done ad hoc, the pieces routinely arrive in the wrong years.
Watch the aggregation traps. Because 461(l) aggregates every trade or business on the return, a taxpayer with multiple entities can be surprised in both directions: a profitable side business quietly consumes loss capacity that was budgeted against wages, or losses from two ventures stack in one year and blow through a threshold that either alone would have cleared. Married couples should also note that the $512,000 joint threshold is not two independent $256,000 allowances — it is one shared gate for the whole return.
Deferral, not elimination. The theme of this entire series repeats one last time. The SMA version of TALS defers capital gains; the LP version's blocked losses are deferred deductions. Section 461(l) converts timing you wanted into timing the statute dictates. That is a real cost, not a footnote — and it compounds with the strategy-level deferral mechanics covered in Part 2, where the tax benefit itself was shown to rest largely on postponed gains.
The Sober Section: Risks That Outrank the Math
The math above assumes every characterization holds. It may not. The $555,000 figure depends on a chain of positions — ordinary character under the notional principal contract regulations, trader status at the partnership level, escape from the passive rules, survival of the straddle rules — and each link is a place where the IRS can pull. If the fund's monthly-reset swap payments were recharacterized, or the partnership failed the frequent, regular, and continuous trading standard of Chen and Endicott, the ordinary deductions could collapse into capital losses subject to the $3,000-per-year regime from Part 1 — a different asset entirely.
Economic substance carries a strict-liability penalty. Under section 7701(o), a transaction must both meaningfully change the taxpayer's economic position apart from federal tax effects and have a substantial non-tax purpose. Paired long/short structures whose net economic risk is small relative to their tax benefits are the doctrine's core target, and failure triggers a 20 percent accuracy penalty — 40 percent if undisclosed — with no reasonable-cause defense. The IRS has shown, as recently as Rev. Rul. 2024-14, that it will apply the doctrine to partnership structures.
Leverage, fees, and the audit profile. These funds run levered books with short positions, financing costs, borrow fees, and management fees well above passive alternatives; the loss on the K-1 is real money before it is a deduction. Meanwhile, a large ordinary loss from a trading partnership sitting next to $800,000 of wages is a recognized examination flag, and partnership-level adjustments under the centralized audit regime can reach every limited partner at once.
Legislative risk is not hypothetical. This series has repeatedly noted that the character and timing rules underpinning TALS exist at Congress's sufferance. OBBBA just demonstrated the point in the restrictive direction: it made 461(l) permanent and reset the 2026 thresholds below 2025's. Prior legislative proposals would have marked derivatives to market with ordinary treatment, which would eliminate the character asymmetry outright. A strategy underwritten over a decade must survive tax law it cannot predict.
The first-principles test. Which returns us to where Part 1 began: a strategy must make sense before tax. The published research on tax-aware long-short strategies is genuinely interesting, but every worked number in this series is a conditional claim — conditional on pre-tax economics that justify the leverage, fees, and tracking error even if the tax benefits shrink. If an allocation only works because of the deduction, section 461(l) is the least of its problems.
Sources & Further Reading
26 U.S.C. § 461 — Section 461, including the (l) excess business loss limitation (Legal Information Institute, Cornell Law School)
26 U.S.C. § 172 — Net operating loss deduction: indefinite carryforward, carryback restrictions, and the 80 percent limitation (Legal Information Institute, Cornell Law School)
IRS — Rev. Proc. 2025-32, 2026 inflation adjustments, including the section 461(l) thresholds
IRS — Instructions for Form 461, Limitation on Business Losses
Treas. Reg. § 1.446-3 — Notional principal contracts: periodic, nonperiodic, and termination payments (Legal Information Institute, Cornell Law School)
IRS — Topic No. 429, Traders in Securities
Temp. Reg. § 1.469-1T — Trading personal property for owners' accounts excluded from passive activity treatment (Legal Information Institute, Cornell Law School)
26 U.S.C. § 7701 — Section 7701(o), codified economic substance doctrine (Legal Information Institute, Cornell Law School)
IRS — Rev. Rul. 2024-14, applying the economic substance doctrine to partnership transactions
U.S. Tax Court — Chen v. Commissioner, T.C. Memo. 2004-132 (CourtListener)
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.
