IP Tax and Holding Structure Toolkit: Entities, Transfers, Royalties, and Documentation

By Casey Scott McKay · 2024-05-16

Where a company's intellectual property sits determines who is taxed on the income it produces, and the evidence that decides the question is generated almost entirely by the intellectual property function. This toolkit collects what a practitioner needs. It works through the structures that recur, the arm's length standard that governs intercompany pricing, and the analysis that allocates returns to the entities performing development, enhancement, maintenance, protection, and exploitation rather than to the entity holding legal title. It then addresses the mechanics an intellectual property lawyer controls: executed assignments, recordation, portfolio schedules, territorial allocation for royalty sourcing, and the quality control that keeps a licensed mark valid while simultaneously evidencing a function. It closes on the annual maintenance cycle that most structures never receive.

IP and Technology > IP and IT in Corporate Transactions | Toolkit | Published 16 May 2024 - Updated 20 January 2026 | Casey Scott McKay - marksy.us

Summary. Where a company's intellectual property sits determines who is taxed on the income it produces, and the evidence that decides the question is generated almost entirely by the intellectual property function. This toolkit works through the structures that recur, the arm's length standard governing intercompany pricing, and the analysis allocating returns to the entities performing the substantive functions rather than to the entity holding legal title. It addresses the mechanics an intellectual property lawyer controls — executed assignments, recordation, portfolio schedules, territorial allocation, and quality control — and closes on the annual maintenance cycle most structures never receive.

Keywords: IP holding company toolkit · transfer pricing · section 482 · section 367(d) · cost sharing arrangements · DEMPE functions · GILTI and FDII · royalty sourcing · withholding tax · intercompany licence · recordation · chain of title · substance evidence · state addback statutes · contemporaneous documentation

Start Here

The instinct is to treat this as somebody else's subject. Tax counsel design the structure, the intellectual property team files the applications, and the two conversations happen in different rooms.

That separation is where the failures come from. When a tax authority examines an intangible structure, the first documents it requests are the ones the intellectual property team produced: assignments, inventor declarations, licence agreements, recordation records, correspondence with foreign associates, and invention disclosure files showing who did the work.

Those documents either support the structure or destroy it. A licence granting rights the licensor never owned; an assignment recorded three years after the transfer it purports to effect; a patent naming inventors employed by a different entity from the one claiming the return — each is an ordinary housekeeping failure and each is a transfer pricing adjustment waiting to happen.

And the reverse is true. A structure designed without the intellectual property lawyer produces licences that do not match the registrations, entities holding marks they cannot support with use, and migrations that lose priority dates or break chains of title.

Two further framing points. There is no property right in the tax analysis: the question is always who performed which functions and bore which risks. And the regulatory system is a second, parallel exclusivity regime, so a portfolio strategy built without the regulatory and tax calendars describes half the position.

This toolkit works through structures, pricing, substance, mechanics, and maintenance.

The Structures That Recur

The domestic holding company. A separate entity owns the group's patents and marks and licences them to the operating companies. The historical driver was state tax; the durable drivers are centralised ownership, cleaner chains of title, ring-fenced enforcement, easier financing, and a single licensing counterparty.

The state tax rationale has largely closed. Geoffrey, Inc. v. South Carolina Tax Commission, 437 S.E.2d 13 (S.C. 1993), established economic nexus through the use of marks in a state; addback statutes now disallow related-party royalty deductions in most separate-reporting states; combined reporting nets the transaction to nothing in most large states; and South Dakota v. Wayfair, Inc., 585 U.S. 162 (2018), confirmed the direction of travel.

The offshore principal. A foreign affiliate owns or co-develops the intangibles, bears development risk, and earns the residual return, licensing to distribution affiliates that earn a routine margin. This is the structure that produced two decades of litigation and the entirety of the international base erosion project.

The cost sharing arrangement. Affiliates share development costs in proportion to reasonably anticipated benefits, each owning the results in its territory. Governed by 26 C.F.R. § 1.482-7, it avoids a royalty entirely and requires a platform contribution payment for pre-existing intangibles brought in.

And the domestic repatriation, which the 2017 changes made comparatively attractive: 26 U.S.C. § 951A subjects most intangible-rich foreign subsidiary income to current tax, 26 U.S.C. § 250 offers a deduction for foreign-derived intangible income earned domestically, and 26 U.S.C. § 59A reaches outbound royalties directly. A number of large groups responded by moving portfolios back, which is an intellectual property project run on a tax timetable that nobody costed.

Pricing and the Arm's Length Standard

26 U.S.C. § 482 is short and consequential. It permits allocation of income, deductions, credits, or allowances among commonly controlled organisations to prevent evasion or clearly to reflect income, and for intangibles it adds that income must be commensurate with the income attributable to the intangible.

The commensurate-with-income standard permits periodic adjustment. A licence priced correctly in year one can be re-priced in year five if the intangible turns out to be far more valuable than anyone projected — which converts a pricing decision into a permanently open question and is why the projections underlying a transfer are examined as closely as the transfer itself.

The methods appear in 26 C.F.R. § 1.482-4 — comparable uncontrolled transaction, comparable profits, profit split, and unspecified methods — with the best method rule selecting among them on the reliability of the available data.

The comparable uncontrolled transaction method is the taxpayer's preference because it produces a defensible number from a real licence, and it is the hardest to sustain because genuinely comparable licences of unique intangibles rarely exist. Medtronic, Inc. v. Commissioner generated years of litigation over exactly that comparability question.

Profit split is where large intangible cases end up, and it is where the functional analysis does its work.

Cost sharing has its own settled point. Altera Corp. v. Commissioner, 926 F.3d 1061 (9th Cir. 2019), upheld the regulation requiring stock-based compensation in the shared cost pool, and certiorari was denied.

And the outbound transfer is a deemed sale. 26 U.S.C. § 367(d) treats a transfer of intangible property to a foreign corporation as a sale for contingent payments over the useful life, with the definition of intangible property now including goodwill, going concern value, and workforce in place — reversing prospectively the result taxpayers obtained in Amgen.com litigation typified by Amazon.com, Inc. v. Commissioner, 934 F.3d 976 (9th Cir. 2019), and Veritas Software Corp. v. Commissioner, 133 T.C. 297 (2009).

Substance and the Functional Analysis

The OECD base erosion project changed the analytical frame more than it changed any statute. Returns from an intangible accrue to the entities that perform and control development, enhancement, maintenance, protection, and exploitation, and that bear the associated risks — rather than to the entity holding legal title.

Which is severe for the classic offshore principal. An entity in a low-tax jurisdiction owning patents on paper, employing a handful of people, and outsourcing research, prosecution, enforcement, and commercialisation is entitled to a return on the funding it provided and little more.

Protection is the function intellectual property lawyers own, and it is routinely mislocated. If enforcement decisions are made in the operating company, litigation is instructed by operating company counsel, and prosecution strategy is set by an in-house team employed elsewhere, then the protection function does not sit with the title holder — and the correspondence file proves it.

Country-by-country reporting gives authorities a group-level view of where revenue, profit, employees, and assets sit, and a structure allocating substantial profit to a jurisdiction with two employees is visible on the first page.

The modified nexus approach conditions preferential regime benefits on research actually conducted in the jurisdiction, ending the practice of acquiring a portfolio and relocating its income.

So substance has to be built and evidenced. People, with names and roles. Decisions, recorded in board minutes as actual choices rather than ratifications. Money, with prosecution and renewal invoices addressed to and paid by the owner and proceeds received by it. Instructions, with foreign associates instructed in the owner's name by someone employed by it. Contracts, describing what affiliates actually provide. And records: invention disclosures, quality control inspections, and a current portfolio schedule.

Build a substance file and update it annually. It takes a day, and it is the document that answers an information request in a week rather than three months.

The Mechanics an IP Lawyer Controls

A memorandum is not a transfer. Every migration needs executed instruments, and their absence is the most common structural defect found on examination and on diligence.

Build the schedule first: every registration and application by jurisdiction, with number, class or family, status, and renewal or expiry date, plus unregistered rights, domain names, and social accounts.

Execute assignments naming the specific rights, dated, signed by a person with authority, with consideration recited. For marks, ensure the assignment carries the goodwill of the business symbolised by the mark as 15 U.S.C. § 1060 contemplates, because a mark cannot be assigned in gross.

Record everywhere recordation is available. 35 U.S.C. § 261 makes an unrecorded patent assignment void against a subsequent bona fide purchaser without notice; section 1060 does the same for marks; and copyright recordation is available under 17 U.S.C. § 205. Foreign offices vary in requiring notarisation, legalisation, translation, or evidence of the underlying transaction.

Notify the associates and change the correspondence address, so instructions and renewal notices flow to the new owner from the transfer date.

Assign the ancillaries: pending proceedings, existing licences, security interests requiring consent, and any coexistence or settlement agreements binding the rights.

Then draft the intercompany licence properly, because most are two pages and cannot support the structure they exist to create.

Schedule the portfolio with numbers and jurisdictions and a mechanism for additions. A grant of "all intellectual property" cannot be priced, sourced, or enforced.

Allocate territory expressly, because royalty sourcing under 26 U.S.C. § 861 and 26 U.S.C. § 862 follows the place of use, and a single worldwide rate creates an allocation problem solved later on worse evidence.

Define the royalty base, support the rate with a benchmarking study, specify exclusivity and field of use, allocate withholding under 26 U.S.C. § 1441 with the documentation supporting any treaty rate under 26 U.S.C. § 894, address improvements, and provide for termination and reversion.

Include real quality control terms for marks — standards, approval, inspection, and records — because they protect the mark against abandonment for naked licensing under 15 U.S.C. § 1127 and evidence the maintenance function at the same time. One programme solves two problems.

And confirm the royalty is invoiced and paid rather than accrued, which is a recurring and trivially avoidable examination finding.

Asset-Specific Notes

Patents. The cleanest asset to move. Watch three things: inventor assignments must be effective where the inventors are employed, because several jurisdictions impose mandatory remuneration rules; pending applications must transfer as well as granted patents with the prosecution record updated; and government funding creates obligations that travel with the invention.

Trademarks. The hardest, because of the goodwill requirement and because the licence back must carry real quality control. Use evidence is a second problem: the entity making specimen filings must show use by it or by a controlled licensee.

Copyright. Transfer requires a signed writing under 17 U.S.C. § 204; registration should be maintained because it conditions statutory damages and fees under 17 U.S.C. § 412; and the termination rights in 17 U.S.C. § 203 survive any corporate structure.

Trade secrets. Nothing to record, so the transfer is evidenced by the confidentiality architecture itself, and a transfer to an entity whose people have never had access transfers nothing — because the reasonable measures analysis under 18 U.S.C. § 1839 locates the asset where the controls are.

Software. Copyright, patents, secrets, open source obligations that travel with the code, and third-party components whose licences may restrict transfer. Run the open source scan before the migration.

And the domestic characterisation points. Research expenditures capitalised under 26 U.S.C. § 174; acquired intangibles amortised over fifteen years under 26 U.S.C. § 197; the research credit under 26 U.S.C. § 41 substantiated from invention disclosure records; and patent sales capable of producing capital gain under 26 U.S.C. § 1235 where all substantial rights transfer, notwithstanding 26 U.S.C. § 1221 — with "all substantial rights" being an intellectual property question that field-of-use limits and retained practice rights can defeat.

Documentation and the Four Numbers

Contemporaneous means before the return is filed, not after the information request arrives.

The master file and local file describing the group's business, its intangibles, its intercompany financial activities, and the local entity's transactions.

The country-by-country report, and a candid look at how the structure appears on it.

The functional analysis, which the intellectual property team should read and correct, because it is written by people who do not know who instructs the foreign associates and it is the document that decides the outcome.

The benchmarking study with comparables, rejections, and adjustments.

The executed intercompany agreements, matching what actually happens.

And the substance file.

Then reconcile the four numbers that describe the same portfolio and are produced by four teams that do not speak. The transfer pricing royalty rate; the valuation produced for a migration, an impairment test, or a financing round; the damages number advanced in litigation; and the number offered or demanded in a licensing negotiation.

All four are discoverable, all four are comparable, and the inconsistency is found quickly — by an examiner, by an opposing damages expert, or by a buyer's diligence team.

Reconcile them annually, not necessarily to make them identical but to be able to explain the differences in a paragraph, which is easy contemporaneously and nearly impossible three years later.

And review the inputs the intellectual property team owns: useful life against actual expiry dates adjusted for term adjustment, extension, and terminal disclaimers; scope against the claim charts; validity exposure against pending challenges; and the comparability of any benchmark licence against the rights, fields, and exclusivity it actually covered.

Annual Maintenance

The most common failure in this area is not a bad structure. It is a good structure nobody has looked at in six years, during which the business acquired three companies, moved its research, changed who runs enforcement, and launched a product line the licence does not cover.

The annual cycle. Update the portfolio schedule and re-attach it to the licence. Confirm every new registration and application stands in the correct name, and record any assignments outstanding. Recalculate, invoice, and settle the royalties. Recalculate cost sharing benefit shares and bring in any acquired intangibles with a documented contribution. Refresh the benchmarking study on the agreed cycle. Perform and document the quality control inspections. Update the substance file and re-run the functional analysis if anything material changed. Reconcile the four numbers.

Trigger a full review on an acquisition, a relocation of research, a change in who instructs prosecution or enforcement, a new product line or business model, a material change in tax law, a financing or impairment test, or a dispute in which the portfolio's value is asserted.

And name the owner. In most groups the structure belongs to tax, the portfolio belongs to legal, and the maintenance belongs to nobody — which is why good structures decay into indefensible ones without anyone deciding that they should.

A Worked Example

A mid-sized software business with a domestic headquarters, a development centre abroad, a foreign subsidiary that holds the intellectual property and contracts with customers across a region, and marks registered in twelve countries.

The structure as designed. The foreign entity owns the code and the marks, licences the parent to sell domestically, and earns the regional margin.

The facts as they are. The code is written by employees of a different subsidiary. The product roadmap is set at headquarters. Enforcement against an infringer was instructed and paid for by the parent. The marks were "transferred" by a reorganisation memorandum with no assignments executed and nothing recorded. The holding entity employs eleven people, all in sales.

What follows. Development sits with the development subsidiary; enhancement and exploitation direction sits at headquarters; protection sits at headquarters; and the holding entity performs distribution. On a functional analysis it is entitled to a distribution return rather than the residual. It does not own the marks at all. And the royalty from the parent is exposed under 26 U.S.C. § 59A.

The remediation is unglamorous and effective. Execute and record the assignments, or accept that title sits where it sits and price accordingly. Relocate the protection function to match the title, or relocate the title to match the function. Put the development arrangement on a contractual footing describing what that entity actually does and bears. Schedule the portfolio into the licence with territorial allocation. And reconcile the transfer pricing number with the valuation used in the last financing round.

A fortnight of intellectual property housekeeping, and it is the difference between a structure that survives an examination and one that does not.

Where This Is Heading

The direction of travel is toward substance and transparency, and it has been for a decade. Country-by-country reporting made group structures visible; the modified nexus approach tied preferential regimes to local research; the functional analysis relocated returns from title holders to function performers; and the global minimum tax now being implemented further compresses the benefit of locating income in low-tax entities.

Structures built purely on rate arbitrage are ending, and the ones that survive reflect where the business actually operates.

Which is good news for the intellectual property lawyer, because a structure that follows the business is one the file can support: title where the people are, licences matching the registrations, prosecution instructed by the owner, enforcement decided by the owner, and quality control genuinely exercised — all of which good portfolio management requires anyway.

The remaining tension is with speed. Tax structures are designed on transaction timetables, and portfolio migration takes months of assignments, recordations, and associate notifications across dozens of jurisdictions. That mismatch is the source of most documentation failures, and it is fixed by involving the intellectual property team at design rather than at implementation.

And the enduring point is simple. The entity taxed on the income from an intangible is decided by facts, and most of those facts are recorded in documents the intellectual property team creates as a matter of routine. Creating them well costs almost nothing; reconstructing them under examination costs a great deal, and frequently fails.

Collaboration, Financing, and Encumbrances

Joint development agreements should allocate background, foreground, and improvements separately and say what happens to jointly conceived subject matter, because under 35 U.S.C. § 262 each joint owner may exploit and licence without accounting to the others — almost never what the parties intend and always the default if the agreement is silent.

Inventorship is not negotiable. It is determined by contribution to conception under 35 U.S.C. § 116, and an agreement purporting to allocate inventorship rather than ownership is void as to the former.

University collaboration brings publication rights, institutional policies, and government funding conditions, and the publication calendar is what most often destroys foreign novelty under 35 U.S.C. § 102.

Security interests split their perfection. Article 9 filings in the debtor's jurisdiction, plus federal recordation for patents and marks and copyright recordation under section 205 — and prudent practice is to do both, because the case law on the interaction is not uniform.

Existing encumbrances survive a migration, and moving encumbered assets without the secured party's consent is usually an event of default.

Anti-assignment terms in inbound licences frequently prohibit intra-group transfer or trigger consent on change of control, and the discovery typically happens after the restructuring is announced.

Insolvency changes the licensor. 11 U.S.C. § 365(n) protects licensees on rejection, and Mission Product Holdings, Inc. v. Tempnology, LLC, 587 U.S. 370 (2019), held that rejection of a trademark licence does not terminate the licensee's rights — which licensees will negotiate around once a holding structure changes who the licensor is.

And record the reversion path, so that a wind-up, sale, or relocation of the holding entity does not strand the rights.

The State Layer

Most companies will never build an offshore principal, and the state-level version of the same question arrives far more often.

The classic move was a holding company in a state that did not tax intangible income, owning the group's marks and licensing them to operating companies in states that did — with the royalty deductible where paid and untaxed where received.

Three developments closed it: economic nexus through the use of marks in the taxing state; addback statutes disallowing the deduction for related-party royalties in a majority of separate-reporting states; and combined reporting in most large states, which consolidates the group so the intercompany royalty nets to nothing.

What survives is the exception structure. Most addback statutes carve out payments where the recipient is subject to tax elsewhere at a comparable rate, where the transaction had a valid business purpose, where the payment was at arm's length, or where the recipient on-paid to an unrelated party. Each of those exceptions is proved with documents the intellectual property team holds — the licence, the royalty calculation, the benchmarking study, and evidence that the licensor actually performs functions.

Which produces the practical rule. A state holding company built for tax alone is a losing structure; one built for centralised ownership, cleaner title, ring-fenced enforcement, and a single licensing counterparty — that happens also to be tax-efficient in some states — is defensible, provided it genuinely does what it says.

And "genuinely does what it says" is measured concretely: its own employees or a real services agreement, its own bank account, board minutes recording actual decisions, instructions to outside counsel issued in its name, enforcement proceeds received by it, and quality control exercised and documented.

Migration Timetable

Weeks one and two. Build the schedule: every registration and application by jurisdiction, plus unregistered rights, domains, and accounts. Identify encumbrances, in-licences with anti-assignment terms, pending proceedings, and required consents.

Weeks three and four. Confirm chain of title for every asset and remediate the gaps — missing inventor assignments, unrecorded prior transfers, entities that no longer exist — before transferring anything, because a migration inherits every defect it does not fix.

Week five. Obtain valuations, and have the intellectual property team check useful life against real expiry dates adjusted for term adjustment, extension, and terminal disclaimers, and scope against the claim charts.

Weeks five to seven. Execute the assignments and the intercompany licence together, effective from the same date.

Weeks seven to twenty. Recordation, tracked to confirmation in every jurisdiction. Notify associates, change correspondence addresses, update docketing, substitute parties in pending proceedings, obtain secured party consents, and amend third-party licences referencing the former owner.

Weeks twenty to twenty-four. Build the substance file, the functional analysis, and the contemporaneous documentation. Reconcile the four numbers. Set the maintenance calendar and name its owner.

Throughout. Keep the working file — board papers, business case, valuations, and correspondence — because "why does this structure exist" is the first question in every examination and the contemporaneous answer is worth far more than a later reconstruction.

And do not let the effective date be fixed before the recordation lead times are known, because that single sequencing error produces most of the documentation failures in this area.

What Clients Actually Ask

"Can we just move the IP to a lower-tax jurisdiction?" You can move it anywhere, provided assignments are executed and recorded, the outbound transfer is priced under 26 U.S.C. § 367(d), and the receiving entity performs functions justifying the return it will claim. Moving title alone earns a funding return, not the residual.

"Do we need a royalty at all?" Not if the affiliates jointly develop under a cost sharing arrangement, in which case each owns its territory. Otherwise yes, and it must be priced, invoiced, and paid — an intercompany royalty accrued and never settled is a recurring finding.

"What rate?" Whatever a benchmarking study or profit split supports, documented at the time. There is no default, and a rate chosen because it is round is a rate that will be adjusted.

"Can we do this before year end?" The assignments, yes. The recordations, generally not.

"What if we get audited?" You produce the master file, the local file, the functional analysis, the benchmarking study, the executed agreements, and the substance file. A group that has them answers in a week; a group that does not spends two years assembling them under pressure.

"Is this aggressive?" A structure that follows where the business actually operates is not. One allocating the residual return to an entity with no functions is, and the current framework is designed specifically to find it.

"Who should own this internally?" Somebody named, with the calendar and the schedule and the authority to ask why a new registration was filed in the wrong name.

Common Failure Patterns

The transfer that never happened. A reorganisation memorandum, no assignments, no recordations, and a holding company that owns nothing — found in diligence, in litigation standing challenges, and in examinations, and remediable only prospectively.

The two-page licence. No schedule, no territorial allocation, no defined royalty base, and no quality control. It cannot be priced, sourced, or enforced, and it is the most common document in this area.

The absentee owner. Title in one entity, instructions from another, invoices paid by a third, and proceeds received by a fourth. The functional analysis follows the conduct, and the conduct is documented in the correspondence file.

Naked licensing. Marks licensed intra-group with no quality control exercised, creating simultaneous exposure to abandonment under 15 U.S.C. § 1127 and to a finding that the licensor performs no maintenance function.

The stranded rights. A restructuring leaving registrations in a dissolved entity, discovered at renewal when the office refuses the filing.

The unmaintained cost sharing arrangement. Executed, never recalculated, with acquisitions never brought in.

The inconsistent numbers, as set out above.

And the stale structure, correct when built and invalidated by an acquisition, a research relocation, and a change in who runs enforcement — with nobody having revisited it because nobody owned it.

Every one of these is an intellectual property file problem with a tax consequence, and every one is cheap to prevent and expensive to remediate.

Working With the Tax Team

The conversation usually goes wrong in the same way. Tax presents a structure as settled; legal is asked to implement it; the timetable is fixed; and the intellectual property questions surface as obstacles rather than as inputs.

Ask to see the functional analysis in draft. It is the document that decides the outcome, it is written by people who do not know who instructs the foreign associates, and correcting it takes an hour.

Give tax the four facts it needs and never asks for: where the inventors are employed; who instructs prosecution and enforcement; what the portfolio actually covers, claim by claim, against the products; and the real remaining term.

Explain the trademark constraints early. Goodwill must accompany the mark, quality control must be genuinely exercised, and use evidence must be attributable to the filing entity. These are not negotiable and they surprise tax teams every time.

Quantify the migration timetable honestly, with recordation lead times by jurisdiction, before the effective date is chosen.

Flag encumbrances and anti-assignment terms in week one, not after the structure is announced.

And offer the reciprocal benefit. A well-documented structure requires a clean portfolio schedule, current chain of title, and a functioning quality control programme — all of which the intellectual property team wants anyway and rarely gets funded. The tax project is frequently the only occasion on which a group will pay for portfolio hygiene, and a practitioner who recognises that gets both jobs done at once.

If the structure is already broken, fix it prospectively and say so. Execute now, record now, attach the schedule now, and relocate the functions now, with a memorandum explaining what was done and when. Retroactive documents are worse than late ones, and an examiner presented with an honest remediation dated correctly is in a very different frame of mind from one presented with an assignment backdated to make a memorandum true.

A Closing Note

Intellectual property lawyers tend to think of their documents as instruments — an assignment transfers, a licence permits, a recordation perfects. That is correct and incomplete.

Those same documents are also evidence, and their second audience is not a court but an examiner asking who is entitled to the income an asset produces. The assignment never executed, the licence granting rights the licensor does not hold, the schedule never attached, the instruction letter sent from the wrong entity — none of these will ever be litigated as property questions, and all of them can cost more in adjustment than a portfolio is worth.

The habits that prevent it are the habits that make a portfolio well run. Execute what you approve. Record what you execute. Schedule what you licence. Instruct in the name of the owner. Keep the invention records. Exercise the quality control you promised.

Six habits, and they cost less than one hour of the specialist advice that will otherwise be required to explain their absence.

Do them and the tax structure has a foundation. Skip them, and the structure rests on a memorandum — which is exactly what an examiner will say when the file arrives.

And the six habits are worth stating to the business in exactly those terms, because they ask for administrative discipline rather than money — which is the version of this advice most likely to be accepted, and the only version that survives a change of personnel in either the tax function or the legal one.

One final practical suggestion. Attach the six habits to events the business already has. Execution and recordation to the closing checklist; the portfolio schedule to the annual renewal review; the instruction convention to the outside counsel engagement; the invention records to the disclosure process; and the quality control inspections to the existing supplier or licensee audit calendar. A control bolted onto an existing routine survives; a standalone compliance task does not.

That is the difference between a structure that stays defensible for a decade and one that quietly stops being true in year three.

A Suggested Reading Path

Start with the doctrine in Where the Royalty Lands.

Then the practice in Structuring IP Ownership for Tax and Business Reasons.

Then the audit in the IP holding structure checklist.

For chain of title, Trademarks in the Deal, Conducting a Cross-Regime IP Audit, and the assignment recordal checklist.

For valuation, What Is a Brand Worth? and Valuing and Monetizing a Trademark Portfolio.

For quality control, the trademark license quality control checklist and the Brand Licensing Program Toolkit.

For portfolio governance, the IP Audit and Portfolio Governance Toolkit and the IP audit checklist.

For transactions, the Trademark Transactions Toolkit and Trademark Due Diligence in Mergers and Acquisitions.

And for the secrecy layer that a migration frequently disturbs, the Trade Secret Protection Toolkit.

Primary Authorities

| Authority | Proposition | |---|---| | 26 U.S.C. § 482 | Arm's length; commensurate with income | | 26 C.F.R. § 1.482-4 | Intangible pricing methods | | 26 C.F.R. § 1.482-7 | Cost sharing arrangements | | 26 U.S.C. § 367(d) | Outbound intangible transfers | | 26 U.S.C. § 951A | GILTI | | 26 U.S.C. § 250 | FDII deduction | | 26 U.S.C. § 59A | BEAT on outbound royalties | | 26 U.S.C. § 861 | Royalty sourcing by place of use | | 26 U.S.C. § 862 | Foreign source royalties | | 26 U.S.C. § 1441 | Withholding on royalties | | 26 U.S.C. § 894 | Treaty eligibility | | 26 U.S.C. § 174 | Capitalised research expenditures | | 26 U.S.C. § 41 | Research credit | | 26 U.S.C. § 197 | Fifteen-year amortisation | | 26 U.S.C. § 1235 | Patent transfers as capital gain | | 26 U.S.C. § 1221 | Self-created intangibles | | 35 U.S.C. § 261 | Patent assignment recordation | | 35 U.S.C. § 262 | Joint owners exploit independently | | 15 U.S.C. § 1060 | Trademark assignment with goodwill | | 15 U.S.C. § 1127 | Abandonment; naked licensing | | 17 U.S.C. § 204 | Copyright transfers in writing | | 17 U.S.C. § 205 | Copyright recordation | | 18 U.S.C. § 1839 | Trade secret; reasonable measures | | 11 U.S.C. § 365(n) | Licensee protections in bankruptcy | | Altera Corp. v. Commissioner | Stock compensation in the cost pool | | Amazon.com, Inc. v. Commissioner | Pre-2017 intangible definition | | Veritas Software Corp. v. Commissioner | Buy-in valuation and useful life | | Geoffrey, Inc. v. South Carolina Tax Comm'n | Economic nexus through marks | | South Dakota v. Wayfair, Inc. | No physical presence requirement | | Mission Prod. Holdings v. Tempnology | Rejection does not terminate a licence | | OECD DEMPE analysis | Returns follow functions, not title | | OECD documentation and CbCR | Master file, local file, transparency |

Forms and Templates

The Assignment Agreement Template is the instrument this entire toolkit turns on, and it should be executed for every transfer, name the specific rights, carry the goodwill where marks are involved, and be recorded wherever recordation is available. The License Agreement Template supplies the intercompany licence, whose operative provisions are the portfolio schedule, the territorial allocation that makes sourcing possible, the defined royalty base with supported rate, the withholding allocation, the improvements terms, the quality control obligations, and the termination and reversion. The Portfolio Inventory Template adapts into the combined register this area requires: registrations by jurisdiction and owner of record, unregistered rights, domains and accounts, encumbrances, and inbound licences with their consent requirements. Beyond those, keep four records: a substance file covering people, decisions, money, instructions, contracts, and records; a functional analysis corrected by the intellectual property team; a quality control inspection log; and an annual four-number reconciliation with the differences explained.

Related Toolkits and Checklists

The IP Audit and Portfolio Governance Toolkit covers the inventory and chain of title work that every structure depends on. The Trademark Transactions Toolkit covers assignment and licensing mechanics. The Brand Licensing Program Toolkit covers the quality control that keeps a licensed mark valid and evidences the maintenance function. The Brand Valuation and Monetization Toolkit covers the valuation inputs, and the Trade Secret Protection Toolkit covers the asset that a paper transfer cannot move.