ECI Nailed; Sourcing Failed: The Curious Case Of The Recent Repo CCAs

I. Introduction

In November 2025, the Internal Revenue Service (“IRS”) Office of Chief Counsel released two related advice memoranda—CCAs 202548005 and 202548004 (the “Original CCA” and the “Amended CCA,” respectively). The CCAs consider whether certain U.S.-managed offshore hedge funds (the “Funds”) that regularly entered into securities repo transactions with their U.S. prime brokers thereby were engaged in a U.S. trade or business under Code Sec. 864 or derived U.S.-source income subject to gross-basis tax under Code Secs. 871 and 881.¹ The Funds entered into repos and “reverse” repos to borrow and lend cash, respectively, as well as “special” repos to lend hard-to-borrow securities and earn securities borrow fees.

The CCAs’ analysis is striking in its asymmetry, pairing a clear-eyed and economically grounded application of Code Sec. 864 with a sourcing analysis that imposes novel (and decidedly uneconomic) limitations on sourcing income by analogy. The CCAs begin promisingly by determining that the Funds’ repo activity falls within the trading safe harbor of Code Sec. 864(b)(2)(A)(ii) and therefore does not give rise to a U.S. trade or business, demonstrating a sophis- ticated command of the economic distinctions between dealer, banking, and hedge-fund business models. In particular, the CCAs recognize that regularly entering into transactions of a kind entered into by dealers (for example, securities loans) or financing businesses (for example, funding debt at origination) is not in itself sufficient to be a dealer or financing business for purposes of Code Sec. 864. Rather, the CCAs focus on the commercial reality that the Funds did not occupy a market position in which they could earn dealer or financing profits, instead acting as price takers that loaned securities and deployed capital in a manner ancillary to their trading strategy. In this regard, the CCAs may ease concerns that YA Global,² or the recently proposed regulations under Code Sec. 892, have altered the IRS’s understanding of Code Sec. 864, particularly in the context of debt acquisitions.

In contrast, the CCAs’ sourcing analysis is unconventional, as a matter of both substance and procedure. The Original CCA determines that securities borrow fees must be sourced by analogy to the statutory sourcing rule for interest, which generally sources interest to the location of the payor. In reaching this conclusion, the Original CCA relies on the premise that where there is no sourc- ing rule for an item of income, taxpayers may source such income by analogy only to statutory sourcing rules, apparently even in cases where the payment in question more closely resembles a payment whose source has been explicitly addressed in regulations. This premise, which came as a surprise to most practitioners, relies on a literal interpretation of existing authorities that fails to consider their historical context; appears to ignore the reality that sourcing regulations are explicitly authorized by statute; and could in many cases result in arbitrary or elective sourcing outcomes.

Shortly before the Original CCA became public, Treasury published Notice 2025-63 (the “Notice”), announcing its intention to propose regulations pro- viding that securities borrow fees generally are sourced to the residence of the recipient. The Notice thereby effectively reversed the sourcing determination of the Original CCA—albeit without referencing the CCA or providing any substantive explanation for the new recipient-based sourcing rule. The Amended CCA, which became public simultaneously with the Original CCA, redetermined the source of the securities borrow fees earned by the Funds pursuant to the Notice while leaving in place the Original CCA’s determination that taxpayers may not source income by analogy to regula- tory sourcing rules.

This article identifies practical areas in which the CCAs may provide welcome guidance for offshore investors and lenders, focusing on implications for inbound investments in U.S. corporate loans (Part III.B) and securities-backed lending (Part III.C). It then turns to the CCAs’ sourcing analysis, examining why it is unpersuasive and should not alter existing approaches to sourcing income by analogy (Part IV). ³

II. Background

The Funds in the CCAs were managed by a U.S.-based hedge fund manager. The Funds engaged in typical hedge fund trading strategies, including credit and macro trading. They relied on a small group of large U.S. broker dealers to provide prime brokerage services, including trading and execution, financing the funds’ trading activities, and cash management. During the audit year, the Funds entered into thousands of securi- ties repos, reverse repos, and special repos with their U.S. prime brokers.

A. Background on Repos

Before examining the CCAs, it is useful to have a basic understanding of the economic terms and tax char- acterization of securities repos. In general, a securities repo is an agreement by which a repo “seller” agrees to sell underlying securities to the repo “buyer” for cash and simultaneously agrees to later repurchase identical securities from the repo “buyer” for the initial purchase price plus a financing charge, generally on an overnight or short-term basis. Depending on the objectives of the parties, a repo may function principally as a secured cash loan, a cash-collateralized securities loan, or both.

The CCAs’ analysis is striking in its asymmetry, pairing a clear-eyed and economically grounded application of Code Sec. 864 with a sourcing analysis that imposes novel (and decidedly uneconomic) limitations on sourcing income by analogy.

When a repo is intended primarily as a secured cash loan, the repo “seller” (that is, the cash borrower) typi- cally is permitted to elect how to secure the transaction from a basket of highly liquid securities like U.S. Treasuries (a so-called “general collateral” repo).⁴ In contrast, when a repo is entered into primarily because the repo “buyer” wants to acquire specific securities to use them for some purpose (such as to short sell the securities or cover a failed trade), the repo may designate the specific securities acquired by the repo buyer “on spe- cial” (a so-called “special” repo). In such cases, the repo buyer may be willing to economically bear a securities borrow fee that exceeds the financing charge payable by the repo seller in respect of the cash leg, depending on the “specialness” of (demand for) the security, thereby resulting in a positive net securities borrow fee payable to the repo seller.

For U.S. federal income tax purposes, the cash leg of a repo is characterized as a financing (loan), whereas the transfer of legal title to the underlying security may be characterized either as (x) a collateral arrangement (if the repo transfers legal ownership, but not tax ownership, of the transferred security) or (y) a securities loan (if the repo transfers both legal and tax ownership of the trans- ferred security). For these purposes, it is not entirely clear whether the mere right of the repo buyer to rehypothecate the transferred security, or the exercise of that right, is required to cause the repo seller to relinquish tax owner- ship.⁵ A detailed discussion of this question is beyond the scope of this article.

In general, the CCAs use the term “repo” to refer to a general collateral repo in which the Funds are the financed party, the term “reverse” repo to refer to a general collat- eral repo in which the Funds are the financing party, and the term “special” repo to refer to a repo that the Funds entered into to lend specified hard-to-borrow securities “on special” to its U.S. prime brokers in order to earn securities borrow fees. The remainder of this article follows this nomenclature.

B. The Funds’ Repo Transactions

In the taxable year under audit, the Funds executed thou- sands of repos and reverse repos, generally with respect to government securities and corporate debt. In general, the Funds entered into repos to obtain cash financing to support their trading activities and entered into reverse repos to “deposit[] inactive (excess) funds in highly liquid financial assets.” In at least some cases, the Funds entered into repos to synthetically close out existing repos or reverse repos (for example, executing a repo to exit an existing, identical reverse repo).

The Funds represented that “the returns earned from depositing [the Funds’] inactive funds [via reverse repos] do not include any embedded markup for services, unusual risk, etc.” and the Funds were “price taker[s] on this activity and ... did not advertise [themselves] as ... cash lender[s].” The CCAs also determine that the Funds “depended on large banks to finance [their] trading strategies,” highlight- ing that the Funds’ private placement memorandum dis- closed that the Funds “had no special access to financing.” The CCAs conclude that this “implies that [the Funds have] limited bargaining power over [their] financing counter- parties.” The CCAs also observe that the Funds “did not market any strategy of loaning money or matched-book dealing.”

In the case of the special repos, the CCAs note that the Funds “made a request for a quote (‘RFQ’) to the several large banks with which [the Funds] maintained working relationships,” either via phone or electroni- cally, but that the Funds “did not make quotes or respond to RFQs.”

Footnotes

1 All section references herein are to sections of the Internal Revenue Code of 1986, as amended (the “Code”) or the Treasury regulations promul- gated thereunder.

2 161 TC 73 (2023).

3 Under Code Sec. 6110, “written determinations,” including any ruling, determination letter, tech- nical advice memorandum, or CCA, generally may not be used or cited as precedent. See, e.g., Yigal Elbaz, et ux., 109 TCM 1229, Dec. 60,259(M), TC Memo. 2015-49 (“[W]e may not use or cite as precedent IRS Chief Counsel Advice 200842002 in deciding this case.”). Nevertheless, written determinations may provide useful insight into a position the IRS is likely to take on an undecided or novel issue for which no other guidance is available, particularly to the extent that the reasoning that supports the written determination is persuasive. See, e.g., Jasper L. Cummings, Jr., Deep State Revenue Rulings, Tax Notes Today (Feb. 11, 2020) (“[T]his advice may be called chief counsel advice or generic legal advice .... Nevertheless, the IRS knows that this advice will be published and will be relied on by taxpayers, regardless of whether they can cite it for specific purposes.”). In addition, under Reg. §1.6662-4(d)(3)(iii), taxpayers may rely on letter rulings, technical advice, actions on decision, and general counsel memoranda for purposes of establishing “substantial authority” as a defense to penalties.

4 In such cases, the repo seller’s ability to freely substitute the securities collateral suggests that the repo buyer (that is, the financing party) has no specific use for the transferred securities separate and apart from their role as collateral.

5 See, e.g., Matthew A. Stevens, Taxation of Non-Equity Derivatives, 187-1st Tax Mgmt. Port. (BNA), at III.B.1.c.(1) (2025) (“Some commentators have suggested that a transfer of ownership of securities occurs only where the transferee both acquires and exercises the right to sell, exchange, pledge, hypothecate, or otherwise transfer securities. However, the better view appears to be that ownership of securities is transferred where a transferee acquires the mere right to sell, exchange, pledge, hypoth- ecate, or otherwise transfer such securities.”); see also Edward D. Kleinbard, Risky and Riskless Positions in Securities, 71 Taxes 783, 798 (Dec. 1993) (“The differing tax analyses applied to securities loans and repos have been justified as being based on an economic distinction as to how the two transactions are employed in the marketplace. This assertion almost certainly is not true, at least not today. A more accurate explanation of the current state of tax confusion is that the Service, in ruling that repos were money loans, in fact limited its holdings to trans- actions that prohibited the rehypothecation of the repoed securities—a fact that is consistent with traditional secured loans (which, in fact, were what the “repos” considered by the Service amounted to). However, that is inconsistent with securities industry practice with respect to U.S. government securities, which form the vast bulk of repos.”).

To view the full article click here

Originally published by Journal of Taxation of Financial Products

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.