Jamming the Photocopier – Or How to Assess and Mitigate Xerox Risk

Those of us who live and breathe LMEs know that there always is a new loophole to be found somewhere in lengthy, dense debt documents. Enter Xerox, whose unconventional “non-sub” joint venture structure with a TPG Credit-led consortium (“IPCo Holdings”) permitted it to effectuate a deal-away drop-down of some of the company’s intellectual property.

Typically, credit documents group a borrower’s subsidiaries into three buckets: obligors (also called credit parties or loan parties), non-guarantor restricted subsidiaries and unrestricted subsidiaries.

Drop-downs are typically effectuated using an unrestricted subsidiary or a non-guarantor restricted subsidiary. In today’s market, most broadly syndicated loans and high-yield indentures provide generous capacity for these types of transfers. For example, across instruments reviewed by Octus in 2025:

• Transfer capacity to unrestricted subsidiaries averaged 253% of EBITDA across BSLs, 225% of EBITDA across high-yield senior secured notes and 172% across high-yield senior unsecured notes – giving borrowers ample drop-down capacity.
• 58% of BSLs permitted unlimited (or uncapped) investments in non-guarantor restricted subsidiaries – meaning that a majority of BSLs permit unlimited drop-downs to non-guarantor restricted subsidiaries.
• Structurally senior debt capacity averaged 326% of EBITDA across BSLs, 221% of EBITDA across high-yield senior secured notes and 163% across high-yield senior unsecured notes – which provide borrowers with substantial ability to raise debt at the non-guarantor restricted subsidiary level.

Against that backdrop, market-average capacity for traditional drop-downs to unrestricted subsidiaries and non-guarantor restricted subsidiaries remains generous enough that most current documents already give issuers plenty of room to execute transfers through those standard buckets. Where that level of capacity exists, there may be less need to pursue more aggressive or creative drop-down structures. But where transfer capacity is materially tighter than market, as in Xerox’s secured notes, borrowers may have greater incentive to look for structures that circumvent those limits.

The “non-sub” construct is one example, effectively creating a third category of non-guarantor transferee beyond the usual unrestricted subsidiary and non-guarantor restricted subsidiary framework.

Xerox’s credit documents, especially its secured notes, contained tighter-than-usual restrictions on transfers to unrestricted subsidiaries and non-guarantor restricted subsidiaries, including J.Crew and Envision blockers. As Octus has written, we believe that Xerox circumvented these restrictions first by creating IPCo Holdings, a non-Subsidiary (capital “S”) or “non-sub” joint venture outside the restricted group, and second by recycling investment capacity. That recycling allowed it to drop down material intellectual property and issue drop-down financing in an amount that surprised the market.

In this article, we explain how creditors can assess the risk of getting jammed in a Xerox-style LME. We first explain how to assess “non-sub” risk and suggest potential drafting fixes to mitigate that risk. We then engage in a similar exercise for investment capacity, including the recycled investment capacity issue.

The gating point for assessing Xerox risk is a credit document’s definition of “Subsidiary.”

Xerox’s credit agreement defined “Subsidiary” solely by reference to a greater-than-50% voting control test:

The definition of “Subsidiary” in Xerox’s indentures (first lien and second lien) similarly referenced a greater-than-50% voting control test:

Defining a “Subsidiary” by reference to majority voting control is, in our experience, relatively common. Very similar definitions appear, for example, in Beacon Roofing’s July 31, 2023, credit agreement (since acquired by QXO Inc.); US Foods Inc.’s fourth amended credit agreement dated April 30, 2024; Victoria’s Secret & Co.’s second amended credit agreement dated Dec. 16, 2025 (conformed through first amendment) and 4.625% senior notes due 2029; and Hertz Global Holdings’ 4.625% senior unsecured notes due 2026, 12.625% first lien senior secured notes due 2029, 5% senior unsecured notes due 2029; and 5.5% exchangeable senior notes due 2030.

The exact economic and/or voting ownership structure of IPCo Holdings has not yet been disclosed. We believe that Xerox circumvented the definition of “Subsidiary” in its debt documents by structuring IPCo Holdings in a way that gave the third party equal or majority voting rights while leaving Xerox with operational control and the full common equity economics of the entity.

There are a few potential drafting fixes that may prevent a borrower from forming similar “non-subs.”

A reliable way to formulate an LME blocker is to compare pre- and post-LME credit documents. Here, Xerox’s IPCo credit agreement (or JV credit agreement) defines “Subsidiary” as:

There are two notable additions to this “Subsidiary” definition. First, and most importantly, the definition extends to an entity where a person “otherwise control[s]” that entity’s management. This means that a borrower cannot create a joint venture that gives majority voting rights to a third party while leaving the borrower with operational control and the full common equity economics of the entity (as we suspect happened here).

Defining “Subsidiary” by reference to managerial control as well as greater-than-50% voting control also is fairly common, and is included in, for example, Core & Main’s Feb. 9, 2024, third amended credit agreement; Corpay Inc.’s Nov. 5, 2025, 17th amended credit agreement; Sabre Corp.’s Dec. 9, 2025, 11th amended credit agreement; and Columbus McKinnon Corp.’s Feb. 3, 2026, credit agreement.

Second, this definition explicitly includes an undefined “joint venture.” Including “joint venture” may remove any ambiguity about whether “other business entity” encompasses a joint venture. In our view, that inclusion would very likely encompass the type of “non-sub” entity used in Xerox. Defining a “Subsidiary” as encompassing an undefined “joint venture” is fairly common, and is also done in, for example, Corpay Inc. and Sabre Corp.

That said, including “joint venture” by itself is unlikely to mitigate Xerox risk if the “otherwise controls” prong of the Subsidiary definition is absent. This is to be viewed more as a belt-and-braces inclusion. Therefore, in our view, credit agreements such as Energizer Holdings Inc.’s Sept. 22, 2025, first amended credit agreement and Bausch + Lomb Corp.’s Jan. 2, 2026, fourth amended credit agreement – which include “joint venture” but lack the “otherwise controls” element – remain liable to Xerox risk.

Other drafting approaches may help mitigate, or be instructive about how to mitigate, Xerox non-sub risk. For example, this private company credit agreement defines “Subsidiary” as:

There are two notable aspects of this definition. First, “Subsidiary” is defined by reference to either majority voting control or control of a majority of economic interests. That may prevent a structure with a split voting / economic interest (which we believe is how IPCo Holdings was structured) from falling outside the definition of “Subsidiary.”

Second, the definition prevents an entity from ceasing to be a “Subsidiary” if a split ownership structure that involves transferring voting stock but retaining any nonvoting or a majority of economic interests is subsequently devised. This proviso likely will not mitigate Xerox non-sub risk, where IPCo Holdings never was a “Subsidiary” at the outset. However, this proviso may prevent a “Subsidiary” from being transformed into a non-sub, which may be a worthwhile additional protection.

Another approach can be seen in this private company credit agreement, which defines “Subsidiary” as:

This language appears to be intended to address a Xerox-like scenario, but it runs into the same problem addressed above: The non-sub did not cease to be a “Subsidiary” but was never a “Subsidiary” at the outset. Therefore, this language may not prevent an LME utilizing a Xerox non-sub but also may prevent a “Subsidiary” from being transformed into a non-sub.

This provision also raises questions about whether a deal-away liquidity enhancing drop-down constitutes a “Liability Management Transaction.” As we have written before, such definitions must be parsed carefully to ensure that they are sufficiently protective.

Provided that there is a sufficiently protective definition of “Liability Management Transaction,” a variation of this proviso that states something to the effect that “any Person in which a Restricted Group Member owns holds an ownership interest or otherwise exercises control shall be deemed to be a Subsidiary in connection with a Liability Management Transaction” may mitigate Xerox risk. This formulation would likely capture a scenario where an entity was not a defined “Subsidiary” at the outset.

Successfully negotiating for a protective “Subsidiary” definition is critical to mitigating Xerox risk. But when it comes to defending against LMEs, it is unwise to put all of your eggs in one basket. As we explain below, Xerox circumvented J.Crew blockers, Envision blockers and aggregate caps on investments in non-guarantor restricted subsidiaries. To mitigate their Xerox risk, lenders also should bolster their drop-down blockers.

Xerox’s credit agreement and indentures (first lien and second lien) contained J.Crew blockers. In the credit agreement, this provided as follows:

In the indentures, these stated:

These are standard J.Crew blockers. They prohibit an Unrestricted Subsidiary from owning or exclusively licensing Material Intellectual Property, and the restricted group from transferring Material Intellectual Property to an Unrestricted Subsidiary.

None of these blockers prevented Xerox’s transfer of Material Intellectual Property to IPCo Holdings, because IPCo Holdings was not a “Subsidiary” and, thus, not an Unrestricted Subsidiary.

Xerox’s indentures also contained “Envision blockers,” which limited transfers to Unrestricted Subsidiaries to the Unrestricted Subsidiary investment basket. That basket was sized at the greater of $85 million and 10% of EBITDA and was subject to additional restrictions (first lien and second lien):

Xerox still transferred Material Intellectual Property that we believe to be worth more than $450 million by some margin, because the IPCo Holdings was not a “Subsidiary” and, thus, not an Unrestricted Subsidiary.

In addition, Xerox’s indentures (first lien and second lien) contained aggregate caps of $370 million on investments in or designations of Non-Guarantor Subsidiaries:

Xerox circumvented these caps because IPCo Holdings was not a “Subsidiary” and, thus, not a “Non-Guarantor Restricted Subsidiary.”

We can think of at least three drafting improvements that may mitigate Xerox risk.

First, Lenders may impede a Xerox-style drop-down through a carefully drafted “Pluralsight blocker.” A Pluralsight blocker is an enhanced version of J.Crew protection that restricts leakage of material assets (usually material intellectual property) to any non-guarantor subsidiary, including non-guarantor restricted subsidiaries. It is part of what we at Octus call the suite of “next-generation” LME blockers – blockers that typically are only included in a post-LME document. A Pluralsight blocker is commonly drafted so that it prohibits the transfer to or ownership of material intellectual property or material assets by a non-guarantor subsidiary. That formulation would not encompass a Xerox-style non-sub such as IPCo Holdings, because it only applies to a non-guarantor “Subsidiary.”

A Pluralsight blocker that prohibits the transfer to or ownership of material intellectual property or material assets (ideally material assets) by “any person that is not a Loan Party” likely would encompass a Xerox-style non-sub, because the non-sub is “not a Loan Party.” Here is an example from a private company’s post-LME credit agreement:

(For completeness, we note that this clause came from a credit agreement that did not have an unrestricted subsidiary concept and that made creating an unrestricted subsidiary a sacred right. If that is not the case, the clause should prevent the transfer by the borrower and the restricted subsidiaries, rather than the borrower and “any other Loan Party.”)

That said, lenders may face an uphill battle to obtain a Pluralsight blocker, which are not common outside of post-LME contexts. Of the credit documents that Octus reviewed in 2025, Pluralsight blockers were present in 15% of broadly syndicated loans and 15% of high-yield bonds.

Second, a cap on investments in joint ventures (and similar businesses, if that concept is present) also may prevent a Xerox-style drop-down. In function, this is similar to an Envision blocker, although an Envision blocker is classically defined as an express cap on the amount of investment capacity that can be used to transfer assets to unrestricted subsidiaries. For completeness, we also suggest that any such blocker explicitly prevent any capacity used under the applicable joint venture basket from being reclassified under another basket so that a borrower cannot create additional capacity.

We have written before about the benefits and drawbacks of defining terms in credit documents. Leaving “joint venture” (as well as a concept such as “similar businesses”) undefined arguably can encompass a very broad array of business combinations. But it may also raise the question of what constitutes a joint venture and reward creative deal structures (and arguments in litigation). It may be even more protective to include a broadly defined “Joint Venture” that encompasses a non-sub. Consider this definition from a private company’s post-LME credit agreement:

In our view, this probably would encompass the Xerox non-sub IPCo Holdings, since it encompasses any Person in which a Restricted Group Member owns stock (with no minimum of the amount of stock).

Another approach taken by a different private company credit agreement may be instructive. First, the credit agreement broadly defines “Joint Venture” to include a Person in whom the Borrower or any Subsidiary beneficially owns equity “that is not a wholly owned Subsidiary.” That would likely capture a Xerox-style non-sub.

It then goes on to define “Joint Venture Investments”:

Although this definition contains some credit-specific elements (e.g., ambulatory service centers), there are several protective features with broader applicability, two of which are noteworthy in the Xerox context:

• First, the good faith / bona fide business purpose requirement prohibits investments “for liability management purposes.” That is helpful, although it raises the question of what constitutes “liability management purposes” (as well as a “bona fide business purpose”).
• Second, it extends to Joint Ventures the restrictions applicable to Subsidiaries in Article VII, which are the negative covenants. Those negative covenants include, among other protections, a classically drafted Pluralsight blocker that applies to Subsidiaries that are not Loan Parties. By including Joint Ventures in the definition of Subsidiaries for the purpose of the negative covenants, blockers and capacities that apply to Subsidiaries will apply to the non-sub Joint Venture.

Other elements of this definition, such as an ordinary course of business / past practices requirement, mandating that the joint venture’s equity be pledged as collateral, and a prohibition on joint ventures with the sponsor and its affiliate, also could be protective against a Xerox-type joint venture.

As noted above, however, these clauses come from post-LME credit agreements. Even classic Envision blockers remain rare in the market. Of the credit documents that Octus reviewed in 2025, they were present in 10% of broadly syndicated loans and 16% of high-yield bonds. We expect that similar caps and restrictions on investments in joint ventures, and/or other protections along the lines of the above, will be similarly difficult to obtain outside of a post-LME document.

Alternatively, a debt document can impose an aggregate cap on transfers to “any person that is not a Loan Party.” This also would have the same function as and arguably be more protective than a cap on joint venture investments, because it would apply to non-guarantor restricted subsidiaries, unrestricted subsidiaries and non-subs. Such a provision could limit a range of nefarious transfers. That said, and as noted above, unlimited transfers to non-guarantor restricted subsidiaries are common in broadly syndicated loans. Therefore, we expect it will be difficult for lenders to obtain this protection outside of a post-LME credit agreement.

As we previously explained, we believe that Xerox may also have “recycled” basket capacity to be able to contribute more value to IPCo Holdings.

The exact mechanics of Xerox’s intellectual property contribution to IPCo Holdings have not been disclosed publicly. Our hypothesis about capacity recycling, however, is consistent with certain unusual terms in the IPCo credit agreement. That includes references to two separate intellectual property contributions, neither of which is publicly defined. “First IP Contribution” is listed as a condition precedent to the transaction. “Second IP Contribution” is referenced in the “events of default” section of the IPCo credit agreement, which prescribes that the Second IP Contribution must take place immediately after the incurrence of the term loan and the distribution of proceeds. Finally, the term loan agreement stipulates that all of the proceeds from the term loan are to be immediately distributed as an equity distribution to Xerox Corp., despite that IPCo Holdings is nominally a joint venture. We see no other purpose for the two-step IP contribution other than to replenish basket capacity under the existing secured debt documents.

The language that we believe permitted recycling is included in the joint venture basket in Xerox’s indentures (first lien and second lien). That provided as follows:

If our read is correct, it suggests that company-side advisors are interpreting “returns of capital” prongs as effectively permitting a doubling of capacity under joint venture or other investment baskets.

Baskets that increase capacity based on returns of capital are not uncommon. They often form part of a builder basket (frequently called the “Available Amount” or “Cumulative Credit”). The safest way to prevent recycling may be to remove “returns of capital” language altogether. Alternatively, one could limit use of the returns of capital prong with a dollar or percentage cap.

Although Xerox’s non-sub structure is innovative, it is not unprecedented: Similar non-sub structures have been used in LMEs, albeit infrequently, in other contexts and to different ends.

Trinseo used a nonrestricted sister-level affiliate as the primary obligor and intercompany lender in its September 2023 pari-plus structure. The entity was not a subsidiary of the borrower and so, like Xerox, not a “Subsidiary” under the existing credit agreement. Unlike the split ownership structure that we believe was utilized in Xerox, Trinseo’s non-sub was a wholly owned subsidiary of the topco above the borrower. That topco was not a restricted entity, and thus neither were its nonborrower direct subsidiaries. The purpose of the non-sub structure in Trinseo was likely to get around restrictions that would have made a pari-plus transaction with either an unrestricted subsidiary or a non-guarantor restricted subsidiary difficult.

Robertshaw also utilized a non-sub structure as part of its December 2023 transaction (following its initial uptiering LME). Like we believe happened in Xerox, a split ownership structure was used to establish a non-sub. Robertshaw held 100% of the nonvoting interests in the entity (and none of the voting equity), which it contended made the entity a “non-sub” under the applicable credit agreement. Robertshaw formed the non-sub to circumvent a Wesco / Incora blocker that restricted the company and its “Subsidiaries” from issuing debt for the purpose of manipulating voting thresholds.

The capital raised at the non-sub was used to oust Invesco from its “Required Lender” position through a voluntary prepayment – a clear instance of vote rigging that would have been prevented by the blocker, had it applied to the non-sub. Invesco subsequently challenged the transaction before the bankruptcy court after Robertshaw filed for chapter 11. Judge Christopher Lopez held that Robertshaw breached the applicable credit agreement because the non-sub was a “Subsidiary.” That said, Invesco’s remedy left something to be desired. Justice Lopez did not avoid the transaction or restore Invesco’s Required Lender status. Invesco only was entitled to a claim for a pro-rata portion of the non-sub loan that was not used to prepay its loans, which Judge Lopez found to be $39.4 million.

Xerox’s non-sub drop-down likely can be replicated by other issuers. The definition of “Subsidiary” in Xerox’s pre-LME debt documents is fairly common. Classic “next-generation” blockers, which did not prevent the Xerox drop-down, are hard to obtain outside of a post-LME context. Obtaining enhanced next-generation blockers that target Xerox value leakage risk will be similarly, if not more, difficult. Lenders, therefore, must take a multifaceted approach to mitigate their risk of getting jammed.