The ETF Tax Loophole That Wall Street Is Exploiting

Imagine you’ve held a single company’s stock for decades. It’s worth millions, but your cost basis is very low. Selling means handing a large check to the IRS. For most investors, that’s just the price of success. But a growing number of wealthy clients—guided by sophisticated wealth managers—are finding a different path rooted in a law that Congress passed more than a century ago to help small business owners incorporate. That law is Internal Revenue Code Section 351.

The paper “Managing Concentrated Public Stock Positions by Seeding an Exchange-Traded Fund,” by Brent Sullivan, editor of Tax Alpha Insider, and Elliot Rozner, a New York-based research analyst, provides the most thorough public analysis to date of how this provision is being applied to exchange-traded funds—and where its limits lie. They explain how Section 351 works in this context; why Congress is hostile to tax-free diversification; and how patterns like “stuffing” and “sequential seeding” can turn an otherwise clean transaction into a target for substance‑over‑form attacks. Their practical message for investors and advisors is simple: Get the economics and documentation right, assume communications and timing will be scrutinized, and don’t confuse a wrapper change with a tax-free exit from concentrated risk.

Using Section 351 to Seed New ETFs

The paper analyzes how wealth managers are using Section 351 of the Internal Revenue Code to seed newly launched ETFs with clients’ appreciated securities, especially concentrated, low‑basis stock. In a typical separately-managed-account–to‑ETF transaction, the client contributes securities to the ETF at launch and receives ETF shares without current gain recognition if Section 351 is satisfied. The cost basis of the seeded positions carries over to the received shares of the ETF.

The authors assembled a dataset of 39 US ETFs launched between 2021 and 2025, with about $8.7 billion of individual investor seed assets, to show that this is no longer a niche technique.

They framed a central question: When is this just a modern application of a century‑old nonrecognition rule, and when does it become tax-free diversification that Congress has repeatedly tried to shut down? Most seeds are routine, but two patterns—“stuffing” and “sequential seeding”—raise sharper risks that the IRS could recharacterize the transaction and impose immediate tax.

How ETF Seeding Works

Section 351 allows a taxpayer to transfer property to a corporation in exchange for stock without recognizing gain if the transferors, as a group, control at least 80% of the corporation immediately after the exchange. Modern ETF seeds typically involve a new fund that intends to qualify as a regulated investment company, so the ETF itself is generally not taxed on distributed income and gains.

For investors, the appeal of seeding an ETF instead of selling outright is threefold:

1. They reposition appreciated holdings into a diversified strategy without an immediate capital gain event.
2. They access the structural benefits of the ETF wrapper, including in‑kind redemptions that help minimize fund‑level realizations, daily transparency, and intraday liquidity.
3. Management fees are paid at the fund level, which can be more tax-efficient than nondeductible, separately billed advisory fees for many individuals under current law.

The catch is Section 351(e), which denies nonrecognition if the transfer to an “investment company” results, directly or indirectly, in diversification of the transferor’s interests. Treasury regulations implement this via the 25/50 diversification tests, applied investor by investor at the time of contribution: no more than 25% of the investor’s total assets in one issuer, and no more than 50% in five or fewer issuers. “Total assets” exclude cash but include most securities, with special rules for government securities and a look‑through for regulated investment company and ETF holdings.

Core Issues Flagged

The authors highlight three key issues that determine whether a seed looks routine or aggressive from a tax standpoint:

1. Is this a business formation or tax-free diversification?

Section 351 was created to facilitate corporate formation, not to let investors swap concentrated stock for diversified portfolios tax-free. Since the 1960s, Congress has repeatedly tightened the rules to curb “swap funds” and other designs that achieve diversification without tax, and ETF seeds now live in that same policy crosshairs. Using a 351 transaction to achieve economies of scale in a new ETF launch with a unique and desirable investment thesis could be a legitimate use of the tax code. Engineering creative ways to satisfy the diversification rules for the sole purpose of diversifying concentrated stock via an ETF wrapper may not satisfy the intent of the code.

2. Does the transaction truly pass the diversification rules in substance, not just form?

While the 25/50 tests are numeric, the regulations also include a “plan” rule that looks at whether other steps—however delayed—are part of a single plan to diversify. Portfolios engineered to barely clear thresholds, especially with last‑minute purchases of government or credit securities, can attract scrutiny if they lack independent portfolio logic.

3. Will the IRS collapse multiple steps into one taxable exchange?

Using the step‑transaction doctrine, courts can combine formally separate steps—borrowing, buying a dilutive sleeve, seeding an ETF, rapid in‑kind removals, and quick share sales—into a single transaction whose result is tax-free diversification. The authors emphasized the end‑result and interdependence tests, and how timing, predictability, and sponsor communications can support or weaken an IRS argument.

Stuffing and Sequential Seeding Risks

Two examples illustrate where investors move from “clever” to “fragile”:

Stuffing: An investor with a 40% weight in Stock A borrows money three months before seeding, buys a large ultrashort credit ETF sleeve with negative carry, and then contributes both the concentrated stock and the new sleeve so Stock A is just under 25% of the contributed portfolio. After launch, in‑kind activity removes Stock A from the ETF, and within a year, the investor sells high‑basis ETF shares to repay the borrowing.

The authors argue these facts strongly support viewing the borrowing, padding, seed, and postlaunch trades as an integrated plan to achieve tax-free diversification, exposing the transaction to challenge under both the plan rule and the step‑transaction doctrine.

Sequential Seeding: An investor with more than 50% in Stock X gradually moves out of the position by seeding three different ETFs over less than two years, always contributing the maximum amount of Stock X that passes the 25/50 tests, plus the previous seeded ETF. Each ETF is held only three to six months before moving to the next.

Here, the risk is that the IRS treats all three seeds as one plan to go from concentrated Stock X to diversified ETF exposure without tax. The more formulaic the pattern and the more contemporaneous evidence of a preset end‑state, the stronger the government’s case; the more each ETF has independent economic significance and genuine market risk, the stronger the investor’s defense.

Across both examples, the same themes recur: short holding periods, negative carry or otherwise uneconomic sleeves, precise threshold engineering, and documented sponsor coordination all increase the risk that the transaction will be recast as a taxable swap.

Key Takeaways for Investors

Sullivan and Rozner close with a practical diligence framework that translates their legal analysis into actionable guidance. The following themes emerge as the most important for practitioners:

1. Communication leaves a trail.

Assume every email, marketing deck, and advisor note is discoverable. Sponsor communications that lead with tax benefits or describe a “plan” to remove the concentrated position postlaunch are exactly the kind of evidence that supports an IRS challenge. Keep communications focused on investment rationale.

2. Borrowed assets are a red flag.

Using leverage specifically to dilute concentration—particularly with negative carry and a clear repayment plan tied to postlaunch ETF share sales—looks engineered. Diversifier sleeves should reflect a durable allocation, not last-minute threshold management.

3. Timing matters enormously.

There is no bright-line safe harbor, but the authors note that two years “arguably should suffice” before postcontribution activity is treated as independent. One year typically does not. The shorter the interval between contribution and exit from the concentrated position, the easier it is for the IRS to argue prearrangement.

4. Document independent rationale at every step.

For sequential seedings in particular, contemporaneous notes explaining why a subsequent seed was triggered by post-first-step developments—not a preset schedule—can make the difference between a respected series of independent transactions and a collapsed taxable exchange.

5. Manager independence is real, but fragile.

The fact that an ETF’s portfolio manager—not the contributing investor—makes postlaunch decisions is a genuine legal protection. But it evaporates if the investor discusses postlaunch intentions with the sponsor. The more coordination, the less independence.

6. Basis uncertainty has real money consequences.

Whether an investor retains lot-level basis tracking or receives an averaged basis across all ETF shares is unresolved law—and the authors reach opposite conclusions from the two leading treatises. Advisors should model both scenarios during diligence, since the answer affects which shares can be sold to minimize future gain.

The Bigger Picture

Sullivan and Rozner are careful to frame the strategy in its proper historical context. Congress has tried to close tax-free diversification loopholes at least three times over the past century—in 1966, 1976, and 1997—and each time, the industry has found a new path through the rules. The authors quote a 1999 congressional press release that compared the phenomenon to a phoenix rising from the ashes. The ETF seed is the latest iteration.

That history is relevant for a practical reason the authors raise in their diligence guide: pending legislation. Congress has noticed the trend. Advisors should monitor proposed statutory changes and treat any effective-date clause as a stand-alone diligence item, particularly for transactions staged across multiple tax years. If Congress moves, it often moves retroactively.

Redemptions, especially where they were discussed in advance, make it easier for the IRS to argue that the seed was part of a plan to sell without paying tax. Longer, risk‑bearing holding periods and genuine manager independence push in the opposite direction.

The paper’s conclusion is balanced. When a client’s portfolio is already genuinely diversified, the contribution reflects a real investment rationale, there is no prearranged understanding with the ETF sponsor about postlaunch disposition, and the investor plans to hold the ETF shares as a long-term allocation, Section 351 seeding is a legitimate and well-established tax-planning tool. The legal authority supporting it goes back a century.

The problems arise at the margins: when assets are acquired at the last minute to manufacture compliance with a numerical test; when the ETF sponsor and the investor have informally agreed on what happens to the concentrated stock after launch; and when the sequence of steps looks less like an evolving investment strategy and more like a scripted plan that was fixed from the beginning.

For advisors navigating this space, Sullivan and Rozner’s paper is the clearest guide currently available—and a useful reminder that in tax law, as in most things, the facts you create are the facts you’ll have to defend.