De Minimis Exception to the Swap Dealer Definition

1Public Law 111–203, 124 Stat. 1376 (2010), available at https://www.gpo.gov/fdsys/pkg/PLAW￾111publ203/pdf/PLAW-111publ203.pdf. 2The CEA is found at 7 U.S.C. 1, et seq. 3See generally 7 U.S.C. 6s. 4 Dodd-Frank Act section 712(d)(1). See the definitions of ‘‘swap dealer’’ in CEA section 1a(49) and § 1.3 of Commission regulations. 7 U.S.C. 1a(49); 17 CFR 1.3. 5See Dodd-Frank Act section 721. 6 7 U.S.C. 1a(49)(A). In general, a person that satisfies any one of these prongs is deemed to be engaged in swap dealing activity. See also the definitions of ‘‘swap’’ in CEA section 1a(47) and § 1.3 of Commission regulations. 7 U.S.C. 1a(47); 17 CFR 1.3. 7 7 U.S.C. 1a(49)(D). 8 7 U.S.C. 1a(49)(A). 9Further Definition of ‘‘Swap Dealer,’’ ‘‘Security￾Based Swap Dealer,’’ ‘‘Major Swap Participant,’’ ‘‘Major Security-Based Swap Participant’’ and ‘‘Eligible Contract Participant,’’ 75 FR 80174 (proposed Dec. 21, 2010). 10Further Definition of ‘‘Swap Dealer,’’ ‘‘Security-Based Swap Dealer,’’ ‘‘Major Swap Participant,’’ ‘‘Major Security-Based Swap Participant’’ and ‘‘Eligible Contract Participant,’’ 77 FR 30596 (May 23, 2012). 11See 17 CFR 1.3, Swap dealer. As discussed in more detail in section II, the Commission notes that a joint rulemaking with the SEC is not required to amend the De Minimis Exception, pursuant to paragraph (4)(v) of the De Minimis Exception. See 17 CFR 1.3, Swap dealer, paragraph (4)(v); 77 FR at 30634 n.464. COMMODITY FUTURES TRADING COMMISSION 17 CFR Part 1 RIN 3038–AE68 De Minimis Exception to the Swap Dealer Definition AGENCY: Commodity Futures Trading Commission. ACTION: Final rule. SUMMARY: The Commodity Futures Trading Commission (‘‘Commission’’ or ‘‘CFTC’’) is amending the de minimis exception within the ‘‘swap dealer’’ definition in the Commission’s regulations by setting the aggregate gross notional amount threshold for the de minimis exception at $8 billion in swap dealing activity entered into by a person over the preceding 12 months. DATES: This rule is effective November 13, 2018. FOR FURTHER INFORMATION CONTACT:
Matthew Kulkin, Director, 202–418– 5213, mkulkin@cftc.gov, Rajal Patel, Associate Director, 202–418–5261, rpatel@cftc.gov, or Jeffrey Hasterok, Data and Risk Analyst, 646–746–9736, jhasterok@cftc.gov, Division of Swap Dealer and Intermediary Oversight; Bruce Tuckman, Chief Economist, 202– 418–5624, btuckman@cftc.gov or Scott Mixon, Associate Director, 202–418– 5771, smixon@cftc.gov, Office of the Chief Economist; or Mark Fajfar, Assistant General Counsel, 202–418– 6636, mfajfar@cftc.gov, Office of General Counsel, Commodity Futures Trading Commission, Three Lafayette Centre, 1155 21st Street, NW, Washington, DC 20581. SUPPLEMENTARY INFORMATION:
Table of Contents
I. Background
A. Statutory and Regulatory Background
1. Statutory Authority
2. Regulatory History
3. Policy Considerations
4. De Minimis Calculation
B. The Proposal
II. Final Rule—$8 Billion Threshold
A. Proposal
B. Summary of Comments
1. Set Threshold at $8 Billion
2. Increase Threshold
3. Allow Threshold to Decrease
4. Other Comments
C. Final Rule and Commission Response
1. Rationale for Not Reducing AGNA
Threshold to $3 Billion
2. Rationale for Not Increasing AGNA
Threshold
3. Response to Other Comments
III. Proposed Rule Amendments Not Adopted
A. Swaps Entered into by Insured Depository Institutions in Connection With Loans to Customers
1. Proposal
2. Summary of Comments
3. Commission Response
B. Swaps Entered Into to Hedge Financial or Physical Positions
1. Proposal
2. Summary of Comments
3. Commission Response
C. Swaps Resulting From Multilateral
Portfolio Compression Exercises
1. Proposal
2. Summary of Comments
3. Commission Response
D. Methodology for Calculating Notional Amounts
1. Proposal
2. Summary of Comments
3. Commission Response
IV. Other Matters Discussed in NPRM
A. Dealing Counterparty Count and Dealing Transaction Count Thresholds B. Exception for Exchange-Traded and/or Cleared Swaps
C. Exception for Non-Deliverable Forwards
V. Related Matters
A. Regulatory Flexibility Act B. Paperwork Reduction Act
C. Cost-Benefit Considerations
1. General Costs and Benefits
2. Direct Cost and Benefits
3. Section 15(a)
D. Antitrust Considerations
I. Background
A. Statutory and Regulatory Background
1. Statutory Authority
Title VII of the Dodd-Frank Wall Street Reform and Consumer Protection Act (‘‘Dodd-Frank Act’’) 1 established a statutory framework to reduce risk, increase transparency, and promote market integrity within the financial system by regulating the swap market. Among other things, the Dodd-Frank Act amended the Commodity Exchange Act (‘‘CEA’’) 2 to provide for the registration and regulation of swap dealers (‘‘SDs’’).3 The Dodd-Frank Act directed the CFTC and the U.S. Securities and Exchange Commission (‘‘SEC’’ and together with the CFTC, ‘‘Commissions’’) to jointly further define, among other things, the term ‘‘swap dealer,’’ 4 and to exempt from designation as an SD a person that engages in a de minimis quantity of swap dealing.5 CEA section 1a(49) defines the term ‘‘swap dealer’’ to include any person who: (1) Holds itself out as a dealer in swaps; (2) makes a market in swaps; (3) regularly enters into swaps with counterparties as an ordinary course of business for its own account; or (4) engages in any activity causing the person to be commonly known in the trade as a dealer or market maker in swaps (collectively referred to as ‘‘swap dealing,’’ ‘‘swap dealing activity,’’ or ‘‘dealing activity’’).6 The statute also requires the Commission to promulgate regulations to establish factors with respect to the making of a determination to exempt from designation as an SD an entity engaged in a de minimis quantity of swap dealing.7 CEA section 1a(49) further provides that in no event shall an insured depository institution (‘‘IDI’’) be considered to be an SD to the extent it offers to enter into a swap with a customer in connection with originating a loan with that customer.8
2. Regulatory History
Pursuant to the statutory requirements, in December 2010, the Commissions issued a proposing release (‘‘SD Definition Proposing Release’’) 9 further defining, among other things, the term ‘‘swap dealer.’’ Subsequently, in May 2012, the Commissions issued an adopting release (‘‘SD Definition Adopting Release’’) 10 further defining, among other things, the term ‘‘swap dealer’’ in § 1.3 of the CFTC’s regulations (‘‘SD Definition’’) and providing for a de minimis exception in paragraph (4) therein (‘‘De Minimis Exception’’).11 The De Minimis Exception states that a person shall not be deemed to be an SD unless its swaps connected with swap dealing activities exceed an aggregate gross notional amount (‘‘AGNA’’) threshold of $3 billion (measured over the prior 12- month period), subject to a phase-in period during which the AGNA

12 17 CFR 1.3, Swap dealer, paragraph (4)(i)(A). Paragraph (4)(i)(A) also provides for a de minimis threshold of $25 million with regard to swaps in which the counterparty is a ‘‘special entity’’ (excluding ‘‘utility special entities’’ as provided in paragraph (4)(i)(B) of the De Minimis Exception) as defined in CEA section 4s(h)(2)(C), 7 U.S.C. 6s(h)(2)(C). This final rule would not change the AGNA threshold for swaps with special entities. 13See Order Establishing De Minimis Threshold Phase-In Termination Date, 81 FR 71605 (Oct. 18, 2016); Order Establishing a New De Minimis Threshold Phase-In Termination Date, 82 FR 50309 (Oct. 31, 2017). 14See SD Definition Adopting Release, 77 FR 30632–34. In making their determination, the Commissions considered the limited and incomplete swap market data that was available at that time and concluded that the $3 billion level appropriately considers the relevant regulatory goals. Id. at 30632. The Commissions found merit in determining the threshold by multiplying the estimated size of the domestic swap market by a
0.001 percent ratio suggested by several
commenters. Id. at 30633.
15 17 CFR 1.3, Swap dealer, paragraph (4)(ii)(B). 16 17 CFR 1.3, Swap dealer, paragraph (4)(ii)(C). 17See Swap Dealer De Minimis Exception Preliminary Report (Nov. 18, 2015), available at http://www.cftc.gov/idc/groups/public/@swaps/ documents/file/dfreport_sddeminis_1115.pdf. For the Preliminary Staff Report, staff analyzed data from April 1, 2014 through March 31, 2015. 18The comment letters are available on the Commission website at http://comments.cftc.gov/ PublicComments/CommentList.aspx?id=1634. 19See Swap Dealer De Minimis Exception Final Staff Report (Aug. 15, 2016), available at http:// www.cftc.gov/idc/groups/public/@swaps/ documents/file/dfreport_sddeminis081516.pdf. For the Final Staff Report, staff analyzed data from April 1, 2015 through March 31, 2016. 20 Given that all of the CEA section 4s requirements have not yet been implemented by regulation, the term ‘‘registered SD’’ refers to an entity that is a provisionally registered SD. See 17 CFR 3.2(c)(3)(iii). 21 81 FR 71605. 22 82 FR 50309. 23 Dodd-Frank Act, Preamble (indicating that the purpose of the Dodd-Frank Act was to promote the financial stability of the United States by improving accountability and transparency in the financial system, to end ‘‘too big to fail,’’ to protect the American taxpayer by ending bailouts, to protect consumers from abusive financial services practices, and for other purposes). See also De Minimis Exception to the Swap Dealer Definition, 83 FR 27444, 27446 (proposed June 12, 2018). 24For example, registered SDs have specific requirements for risk management programs and margin. See, e.g., 17 CFR 23.600; 17 CFR 23.150– 23.161. 25For example, registered SDs are subject to external business conduct standard regulations designed to provide counterparty protections. See, e.g., 17 CFR 23.400–23.451. 26SD Definition Adopting Release, 77 FR 30628 (‘‘On the one hand, a de minimis exception, by its nature, will eliminate key counterparty protections provided by Title VII for particular users of swaps and security-based swaps.’’). See also 83 FR 27446. 27 77 FR 30629 (‘‘The statutory requirements that apply to swap dealers . . . include requirements . . . aimed at helping to promote effective operation and transparency of the swap . . . markets.’’). See id. at 30703 (‘‘Those who engage in swaps with entities that elude swap dealer or major swap participant status and the attendant regulations could be exposed to increased Continued threshold is set at $8 billion.12 The phase-in period was originally scheduled to terminate on December 31, 2017, and the AGNA threshold was scheduled to decrease to $3 billion at that time. However, as discussed below, pursuant to paragraph (4)(ii)(C)(1) of the De Minimis Exception, the Commission issued two successive orders to set new termination dates, and the phase-in period is currently scheduled to terminate on December 31, 2019.13 When the $3 billion AGNA threshold was established, the Commissions explained that the
information then available regarding certain portions of the swap market was limited, and that they expected more information to be available in the future (following the implementation of swap data reporting), which would enable the Commissions to make a more informed assessment of the De Minimis Exception and to revise it as appropriate.14 In recognition of these limitations and in anticipation of additional swap market data becoming available to the CFTC through the reporting of transactions to swap data repositories (‘‘SDRs’’), paragraph (4)(ii)(B) of the De Minimis Exception was adopted, which directed CFTC staff to complete and publish for public comment a report on topics relating to the definition of the term ‘‘swap dealer’’ and the de minimis threshold as appropriate, based on the availability of data and information.15 Paragraph (4)(ii)(C) of the De Minimis Exception provided that after giving due consideration to the staff report and any associated public comment, the CFTC may either set a termination date for the phase-in period or issue a notice of proposed rulemaking to modify the De Minimis Exception.16 In November 2015, staff issued a preliminary report concerning the De Minimis Exception (‘‘Preliminary Staff Report’’).17 After consideration of the public comments received in response to the Preliminary Staff Report,18 and further data analysis, in August 2016 staff issued a final staff report 19 concerning the De Minimis Exception (‘‘Final Staff Report,’’ and together with the Preliminary Staff Report, ‘‘Staff Reports’’). The data analysis in the Staff Reports provided some insights into the effectiveness of the De Minimis Exception as currently implemented. For example, staff analyzed the number of swap transactions involving at least one registered SD,20 which is indicative of the extent to which swaps are subject to SD regulation at the current $8 billion AGNA threshold. Data reviewed for the Final Staff Report indicated that approximately 96 percent of swap transactions analyzed involved at least one registered SD. To provide additional time for more information to become available to study the De Minimis Exception, in October 2016 the Commission issued an order, pursuant to paragraph (4)(ii)(C)(1) of the De Minimis Exception, establishing December 31, 2018, as the new termination date for the $8 billion phase-in period.21 To enable staff to conduct additional analysis, in October 2017 the Commission further extended the phase-in period to December 31,
2019.22 Generally, the extensions
provided additional time for Commission staff to conduct further data analysis regarding the De Minimis Exception, and gave market participants additional time to begin preparing for a change, if any, to the AGNA threshold.
3. Policy Considerations
(i) Swap Dealer Registration Policy Considerations The policy goals underlying SD registration and regulation generally include reducing systemic risk, increasing counterparty protections, and increasing market efficiency, orderliness, and transparency. Reducing systemic risk: The Dodd￾Frank Act was enacted in the wake of the financial crisis of 2008, in significant part, to reduce systemic risk, including the risk to the broader U.S. financial system created by interconnections in the swap market.23 Pursuant to the Dodd-Frank Act, the Commission has adopted regulations designed to mitigate the potential systemic risk inherent in the previously unregulated swap market.24 Increasing counterparty protections:
Providing regulatory protections for swap counterparties who may be less experienced or knowledgeable about the swap products offered by SDs (particularly end-users who use swaps for hedging or investment purposes) is a fundamental policy goal advanced by the regulation of SDs.25 The Commissions recognized that a narrower or smaller de minimis exception would increase the number of counterparties that could potentially benefit from those regulatory protections.26 Increasing market efficiency, orderliness, and transparency:
Increasing swap market efficiency, orderliness, and transparency is another goal of SD regulation.27 Regulations

counterparty risk; customer protection and market orderliness benefits that the regulations are intended to provide could be muted or sacrificed, resulting in increased costs through reduced market integrity and efficiency. . . .’’). See also 83 FR 27446. 28See, e.g., 17 CFR 23.200–23.205; 17 CFR parts 43 and 45; 17 CFR 23.502–23.503. 29See 77 FR 30628. See also 83 FR 27446. 30See 77 FR 30628–30, 30707–08. See also 83 FR 27446–47. 31 In considering the appropriate de minimis threshold, ‘‘exclud[ing] entities whose dealing activity is sufficiently modest in light of the total size, concentration and other attributes of the applicable markets can be useful in avoiding the imposition of regulatory burdens on those entities for which dealer regulation would not be expected to contribute significantly to advancing the customer protection, market efficiency and transparency objectives of dealer regulation.’’ 77 FR 30629–30. See also 83 FR 27446–47. 32 77 FR 30628–29 (‘‘[T]he de minimis exception may further the interest of regulatory efficiency when the amount of a person’s dealing activity is, in the context of the relevant market, limited to an amount that does not warrant registration. . . . In addition, the exception can provide an objective test . . . .’’). See also 83 FR 27446–47. 33 77 FR 30707–08 (‘‘On the other hand, requiring market participants to consider more variables in evaluating application of the de minimis exception would likely increase their costs to make this determination.’’). See also 83 FR 27446–47. 34 77 FR 30629, 30707–08. See also 83 FR 27447. 35 77 FR 30629. See also 83 FR 27447. 36 77 FR 30628–29. See also 83 FR 27447. 37 77 FR 30628. See SD Definition Proposing Release, 75 FR 80179 (The de minimis exception ‘‘should apply only when an entity’s dealing activity is so minimal that applying dealer regulations to the entity would not be warranted.’’). See also 83 FR 27447. 38 17 CFR 1.3, Swap dealer, paragraph (4)(i)(A); Interpretive Guidance and Policy Statement Regarding Compliance With Certain Swap Regulations, 78 FR 45292, 45323 (July 26, 2013). See also 83 FR 27447. 39See 17 CFR 1.3, Swap dealer, paragraph (5); 77 FR at 30620–24. See also 83 FR 27447. 40See 17 CFR 1.3, Swap dealer, paragraph (6)(i); 77 FR at 30624–25. See also 83 FR 27447. 41See 17 CFR 1.3, Swap dealer, paragraph (6)(ii); 77 FR at 30625–26. See also 83 FR 27447. 42See 17 CFR 1.3, Swap dealer, paragraph (6)(iii); 77 FR at 30611–14. See also 83 FR 27447. 43See 17 CFR 1.3, Swap dealer, paragraph (6)(iv); 77 FR at 30614. See also 83 FR at 27447. The floor trader exclusion was also addressed in no-action relief. See CFTC Staff Letter No. 13–80, No-Action Relief from Certain Conditions of the Swap Dealer Exclusion for Registered Floor Traders (Dec. 23, 2013), available at https://www.cftc.gov/idc/groups/ public/@lrlettergeneral/documents/letter/13-80.pdf. 44See Determination of Foreign Exchange Swaps and Foreign Exchange Forwards Under the Commodity Exchange
Act, 77 FR 69694, 69704–05 (Nov. 20, 2012); Further Definition of ‘‘Swap,’’ ‘‘Security-Based Swap,’’ and ‘‘Security-Based Swap Agreement’’; Mixed Swaps; Security-Based Swap Agreement Recordkeeping, 77 FR 48208, 48253 (Aug. 13, 2012). 45See 17 CFR 32.3; Commodity Options, 77 FR 25320, 25326 n.39 (Apr. 27, 2012). 46See 78 FR 45292; CFTC Letter No. 18–13, No￾Action Position: Relief for Certain Non-U.S. Persons from Including Swaps with International Financial Institutions in Determining Swap Dealer and Major requiring SDs, for example, to keep detailed daily trading records, report trade information, and engage in portfolio reconciliation and compression exercises help achieve these market benefits.28 (ii) De Minimis Exception Policy Considerations Consistent with Congressional intent, ‘‘an appropriately calibrated de minimis exception has the potential to advance other interests.’’ 29 These interests include increasing efficiency, allowing limited swap dealing in connection with other client services, encouraging new participants to enter the market, and focusing regulatory resources.30 The policy objectives underlying the de minimis exception are designed to encourage participation and competition by allowing persons to engage in a de minimis amount of dealing without incurring the costs of registration and regulation.31 Increasing efficiency: A de minimis exception based on an objective test with a limited degree of complexity enables entities to engage in a lower level of swap dealing with limited concerns about whether their activities would require registration.32 The de minimis exception thereby fosters efficient application of the SD Definition. Additionally, the Commission is of the view that the potential for regular or periodic changes to the de minimis threshold may reduce its efficacy by making it challenging for persons to calibrate their swap dealing activity as appropriate for their business models. Further, the Commission is mindful that objective, predictable standards in the de minimis exception increase efficiency by establishing a simple test for whether a person’s swaps connected with swap dealing activity must be included in the de minimis calculation. On the other hand, more complexity in the de minimis calculation potentially results in less efficiency.33 Allowing limited ancillary dealing: A de minimis exception allows persons to accommodate existing clients that have a need for swaps (on a limited basis) along with other services.34 This enables end-users to continue transacting within existing business relationships, for example to hedge interest rate or currency risk. Encouraging new participants: A de minimis exception also promotes competition by allowing a person to engage in some swap dealing activities without immediately incurring the regulatory costs associated with SD registration and regulation.35 Without a de minimis exception, SD regulation could become a barrier to entry that may stifle competition. An
appropriately calibrated de minimis exception could lower the barrier to entry of becoming an SD by allowing smaller participants to gradually expand their business until the scope and scale of their activity warrants regulation (and the costs involved with compliance). Focusing regulatory resources:
Finally, the de minimis exception also increases regulatory efficiency by enabling the Commission to focus its limited resources on entities whose swap dealing activity is sufficient in size and scope to warrant oversight.36 As noted in the SD Definition Adopting Release, ‘‘implementing the de minimis exception requires a careful balancing that considers the regulatory interests that could be undermined by an unduly broad exception as well as those regulatory interests that may be promoted by an appropriately limited exception.’’ 37 A narrower de minimis exception would likely mean that a greater number of entities would be required to register as SDs and become subject to the regulatory framework applicable to registered SDs. However, a de minimis exception that is too narrow could, for example, discourage persons from engaging in limited swap dealing activity in order to avoid the burdens associated with SD regulation.
4. De Minimis Calculation
Generally, a person must count towards its AGNA threshold all swaps it enters into for dealing purposes over the preceding 12-month period. In addition, each person whose own swaps do not exceed the AGNA threshold must also include in its de minimis calculation the AGNA of swaps of any other unregistered affiliate controlling, controlled by, or under common control with that person (referred to as ‘‘aggregation’’).38 Pursuant to various CFTC regulations, certain swaps, subject to specific conditions, need not be considered in determining whether a person is an SD, including: (1) Swaps entered into by an IDI with a customer in connection with originating a loan to that customer; 39 (2) swaps between affiliates; 40 (3) swaps entered into by a cooperative with its members; 41 (4) swaps hedging physical positions; 42 (5) swaps entered into by floor traders; 43 (6) certain foreign exchange (‘‘FX’’) swaps and FX forwards; 44 and (7) commodity trade options.45 In addition, the Commission understands that persons have applied CFTC interpretive guidance and staff letters so as not to count towards the AGNA threshold, subject to certain conditions, certain cross-border swaps 46 and swaps resulting from

Swap Participant Status (May 16, 2018), available at https://www.cftc.gov/sites/default/files/idc/ groups/public/%40lrlettergeneral/documents/letter/ 2018-05/18-13.pdf; CFTC Staff Letter No. 12–71, No-Action Relief: U.S. Bank Wholly Owned by Foreign Entity May Calculate De Minimis Threshold Without Including Activity From Its Foreign Affiliates (Dec. 31, 2012), available at https://www.cftc.gov/idc/groups/public/ %40lrlettergeneral/documents/letter/12-71.pdf; CFTC Staff Letter No. 12–61, No-Action Relief: U.S. Bank Wholly Owned by Foreign Entity May Calculate De Minimis Threshold Without Including Activity From Its Foreign Affiliates (Dec. 20, 2012), available at https://www.cftc.gov/sites/default/files/ idc/groups/public/@lrlettergeneral/documents/ letter/12-61.pdf. 47CFTC Staff Letter No. 12–62, No-Action Relief:
Request that Certain Swaps Not Be Considered in Calculating Aggregate Gross Notional Amount for Purposes of the Swap Dealer De Minimis Exception for Persons Engaging in Multilateral Portfolio Compression Activities (Dec. 21, 2012), available at https://www.cftc.gov/idc/groups/public/@ lrlettergeneral/documents/letter/12-62.pdf. 48See SD Definition Adopting Release, 77 FR 30693. 49 83 FR 27444. 50Comments were submitted by the following entities: 360 Trading Networks Inc. (‘‘360 Trading’’); American Bankers Association (‘‘ABA’’) (ABA also attached a report prepared by NERA Economic Consulting); American Gas Association (‘‘AGA’’); Americans for Financial Reform (‘‘AFR’’); Associated Foreign Exchange, Inc. and GPS Capital Markets, Inc. (‘‘AFEX/GPS’’); Association of Global Custodians (‘‘AGC’’); Better Markets, Inc. (‘‘Better Markets’’); Bond Dealers of America (‘‘BDA’’); Capital One Financial Corporation (‘‘Capital One’’); Cboe SEF, LLC (‘‘Cboe SEF’’); Citizens Financial Group, Inc. (‘‘Citizens’’); CME Group Inc. and Intercontinental Exchange, Inc. (‘‘CME/ICE’’); Coalition for Derivatives End-Users (‘‘CDEU’’); Coalition of Physical Energy Companies (‘‘COPE’’); Commercial Energy Working Group (‘‘CEWG’’); Commodity Markets Council (‘‘CMC’’) (CMC also expressed support for the CEWG comment letter); Covington & Burling LLP (‘‘Covington’’); Daiwa Securities Co. Ltd. (‘‘Daiwa’’); Edison Electric Institute and Electric Power Supply Association (‘‘EEI/EPSA’’); Foreign Exchange Professionals Association (‘‘FXPA’’); Frost Bank; Futures Industry Association and FIA Principal Traders Group (‘‘FIA’’); Institute for Agriculture and Trade Policy (‘‘IATP’’); Institute of International Bankers (‘‘IIB’’); International Energy Credit Association (‘‘IECA’’) (IECA also expressed support for the EEI/EPSA comment letter); International Swaps and Derivatives Association and Securities Industry and Financial Markets Association (‘‘ISDA/SIFMA’’); Japanese Bankers Association (‘‘JBA’’); M&T Bank (‘‘M&T’’); Managed Funds Association (‘‘MFA’’); National Council of Farmer Cooperatives (‘‘NCFC’’); National Rural Electric Cooperative Association and American Public Power Association (‘‘NRECA/ APPA’’); Natural Gas Supply Association (‘‘NGSA’’); NEX Group plc (‘‘NEX’’); Northern Trust; Optiver US LLC (‘‘Optiver’’) (Optiver also expressed support for the FIA comment letter); Regions Financial Corp. (‘‘Regions’’); State Street; SVB Financial Group (‘‘SVB’’); Thomson Reuters (SEF) LLC (‘‘TR SEF’’); six U.S. Senators (‘‘Senators’’); Virtu Financial Inc. (‘‘Virtu’’); Western Union Business Solutions (USA), LLC and Custom House USA, LLC (‘‘Western Union’’); and XTX Markets Limited (‘‘XTX’’). Additionally, there were three meetings with Delta Strategy Group, DRW, Jump Trading, and Optiver, and one meeting with Better Markets. The comment letters and notice of the ex parte meetings are available at https://comments.cftc.gov/PublicComments/ CommentList.aspx?id=2885. 51Additionally, in March
2017, Chairman Giancarlo initiated an agency-wide internal review of CFTC regulations and practices to identify those areas that could be simplified to make them less burdensome and costly (‘‘Project KISS’’). See Remarks of then-Acting Chairman J. Christopher Giancarlo before the 42nd Annual International Futures Industry Conference in Boca Raton, FL (Mar. 15, 2017), available at https://www.cftc.gov/ PressRoom/SpeechesTestimony/opagiancarlo-20. The Commission subsequently published in the Federal Register a Request for Information soliciting suggestions from the public regarding how the Commission’s existing rules, regulations, or practices could be applied in a simpler, less burdensome, and less costly manner. A number of responses submitted pursuant to the Project KISS Request for Information supported modifications to the De Minimis Exception. Project KISS, 82 FR 21494 (May 9, 2017), amended by 82 FR 23765 (May 24, 2017). The suggestion letters filed by the public are available at https://comments.cftc.gov/ KISS/KissInitiative.aspx. 52See ICI v. CFTC, 720 F.3d 370, 379 (D.C. Cir.
2013) (‘‘[A]s the Supreme Court has emphasized,
‘[n]othing prohibits federal agencies from moving in an incremental manner.’ ’’) (quoting FCC v. Fox Television Stations, Inc., 556 U.S. 502, 522 (2009)). 53 7 U.S.C. 1a(49)(D). See also 17 CFR 1.3, Swap dealer, paragraph (4)(v). 54 77 FR 30634 n.464 (‘‘We do not interpret the joint rulemaking provisions of section 712(d) of the Dodd-Frank Act to require joint rulemaking here, because such an interpretation would read the term ‘‘Commission’’ out of CEA section 1a(49)(D) (and Exchange Act section 3(a)(71)(D)), which themselves were added by the Dodd-Frank Act.’’). 55As required by § 712(a)(1) of the Dodd-Frank Act. multilateral portfolio compression exercises.47 Further, certain inter￾governmental or quasi-governmental international financial institutions are not included within the term ‘‘swap dealer.’’ 48 B. The Proposal On June 12, 2018, the Commission published for public comment a Notice of Proposed Rulemaking (‘‘NPRM’’) to amend the De Minimis Exception by: (1) Setting the AGNA threshold for the De Minimis Exception at $8 billion in swap dealing activity entered into by a person over the preceding 12 months; (2) adding new factors to the De Minimis Exception that would lead to excepting from the AGNA calculation: (a) Certain swaps entered into with a customer by an IDI in connection with originating a loan to that customer, (b) certain swaps entered into to hedge financial or physical positions, and (c) certain swaps resulting from multilateral portfolio compression exercises; and (3) providing that the Commission may determine the methodology to be used to calculate the notional amount for any group, category, type, or class of swaps, and delegating to the Director of the Division of Swap Dealer and Intermediary Oversight (‘‘DSIO’’) the authority to make such determinations (collectively, the ‘‘Proposal’’).49 In addition, the Commission sought comment on the following additional potential changes to the De Minimis Exception: (1) Adding as a factor a minimum dealing counterparty count threshold and/or a minimum dealing transaction count threshold; (2) adding as a factor whether a swap is exchange￾traded and/or cleared; and (3) adding as a factor whether a swap is categorized as a non-deliverable forward transaction. The various aspects of the NPRM are discussed in further detail below. The Commission received 43 letters and Commission staff participated in four ex parte meetings 50 concerning the NPRM.51
II. Final Rule—$8 Billion Threshold
Given the more complete information now available regarding certain portions of the swap market, the data analytical capabilities developed since the SD regulations were adopted, five years of implementation experience, and comments received in response to the NPRM, in this adopting release the Commission is amending the De Minimis Exception by setting the AGNA threshold at $8 billion in swap dealing activity. The CFTC may in the future separately propose or adopt rules addressing any aspect of the NPRM that is not finalized in this release.52 This change to the De Minimis Exception is being adopted pursuant to the Commission’s authority under CEA
section 1a(49)(D), which requires the
Commission to exempt from designation as an SD an entity that engages in a de minimis quantity of swap dealing in connection with transactions with or on behalf of its customers, and to promulgate regulations to establish factors with respect to the making of this determination to exempt.53 The Commissions issued the SD Definition Adopting Release pursuant to section 712(d)(1) of the Dodd-Frank Act, which requires the CFTC and SEC to jointly adopt rules regarding the definition of, among other things, the term ‘‘swap dealer.’’ The CFTC continues to coordinate with the SEC on SD and security-based swap dealer regulations. However, as discussed in the SD Definition Adopting Release, a joint rulemaking is not required with respect to the De Minimis Exception.54 The Commission notes that it has consulted with the SEC and prudential regulators regarding the change to the De Minimis Exception adopted herein.55 A. Proposal The Commission proposed to amend paragraph (4)(i)(A) of the De Minimis Exception by setting the AGNA threshold at $8 billion. For added clarity, the Commission also proposed

56See 83 FR 27448–58. The data was sourced from data reported to the four registered SDRs:
BSDR LLC, Chicago Mercantile Exchange Inc., DTCC Data Repository, and ICE Trade Vault. The analysis excluded inter-affiliate and non-U.S. transactions. The total size of the swap market that was analyzed, after excluding inter-affiliate and non-U.S. transactions, was approximately $221.1 trillion in AGNA of swaps activity (excluding non￾financial commodity swaps), approximately 4.4 million transactions, and 39,107 counterparties. The Proposal includes additional discussion regarding the methodology utilized to conduct the analysis. 83 FR 27449–50. 57The term ‘‘FX swaps’’ is used in this release to only describe those FX transactions that are counted towards a person’s de minimis calculation. The term ‘‘FX swaps’’ does not refer to swaps and forwards that are not counted towards the de minimis threshold pursuant to the exemption granted by the Secretary of the Treasury. See 77 FR at 69704–05; 77 FR 48253. 58See 83 FR 27449–50; Preliminary Staff Report, supra note 19, at 21–22; Final Staff Report, supra note 17, at 19. 59As discussed in the Proposal, certain data restrictions limited the usefulness of the SDR data to identify which swaps should be counted towards a person’s de minimis threshold, and the ability to precisely assess the current de minimis threshold or the impact of potential changes to the current exclusions. See 83 FR 27449–50. 60See 17 CFR part 45 app.1. 61See supra section I.A.4 (discussing the de minimis threshold calculation). The Commission notes that the entity-based exclusions and transaction filters are not a determinative means of assessing whether any particular entity is engaged in swap dealing. See also 83 FR 27449 n.73. 62See 17 CFR 1.3, Swap dealer, paragraph (6)(i). 63See generally 78 FR 45292. 64The majority of In-Scope Entities are banks, broker-dealers, non-bank financial entities, and affiliates thereof. 65See 83 FR 27449. 66See 83 FR 27449–50. 67See generally 83 FR 27449–58; Final Staff Report, supra note 19; Preliminary Staff Report, supra note 17. to change the term ‘‘swap positions’’ to ‘‘swaps’’ in paragraph (4)(i)(A). Additionally, the Commission proposed to delete a parenthetical clause in paragraph (4)(i)(A) referring to the period after adoption of the rule further defining the term ‘‘swap,’’ and to remove and reserve paragraph (4)(ii) of the De Minimis Exception, which addresses the phase-in procedure and staff report requirements of the De Minimis Exception (discussed above in
section I.A.2), since both of those
provisions would no longer be applicable. The Commission proposed to maintain the AGNA threshold at $8 billion, and also solicited comment on whether to reduce the threshold to $3 billion, or increase the threshold. The Commission cited as relevant an analysis of SDR data from January 1, 2017, through December 31, 2017 (the ‘‘review period’’).56 Given improvements in the quality of data being reported to SDRs since the Staff Reports were issued, Commission staff analyzed the AGNA of swaps activity for interest rate swaps (‘‘IRS’’), credit default swaps (‘‘CDS’’), FX swaps,57 and equity swaps (whereas the analysis of AGNA data in the Staff Reports was limited to IRS and CDS).58 However, given certain limitations discussed below, AGNA data was not available for non-financial commodity (‘‘NFC’’) swaps. In addition to now-available AGNA information for FX swaps and equity swaps, there were also continued improvements in the consistency of legal entity identifier (‘‘LEI’’) and unique swap identifier reporting.59 Generally employing methodologies similar to those used for purposes of the Staff Reports, staff attempted to calculate persons’ swaps activity in terms of AGNA to assess how the swap market might be impacted by potential changes to the current De Minimis Exception. The reason an entity enters into a swap (e.g., dealing, hedging, investing, proprietary trading) is not collected under the reporting requirements in part 45 of the Commission’s regulations.60 Accordingly, staff applied filters to the data to exclude from the analysis certain transactions and entities that were less likely to be connected to potential swap dealing activity. Entities such as funds, insurance companies, cooperatives, government-sponsored entities, most commercial end-users, and international financial institutions were excluded as potential SDs for the purpose of the analysis because these entities generally use swaps for investing, hedging, or proprietary trading, or otherwise enter into swaps that would not be included in determining whether the entity is an SD.61 Further, additional filters allowed for the exclusion of inter-affiliate 62 and non-U.S. to non-U.S. swap transactions.63 With the benefits of improved data quality and analytical tools, staff conducted a more granular analysis (as compared to the Staff Reports) to more accurately identify those entities that, based on their observable business activities, were potentially engaging in swap dealing activity (‘‘In-Scope Entities’’) 64 versus those likely to be engaging in other kinds of transactions (e.g., entering into swaps for investment purposes). Further, for the purposes of the Proposal, a minimum unique counterparty count of 10 counterparties was utilized to better identify the entities that are likely to be engaged in transactions that have to be considered for the SD Definition. Adding this filter to the analysis reduced the likelihood of false positives—i.e., reduced the potential that entities
likely engaged in hedging or other non-dealing activity would be identified as potential SDs.65 With respect to NFC swaps, Commission staff encountered a number of challenges in calculating notional amounts, including: (1) The vast array of underlying commodities with differing characteristics; (2) the multiple types of swaps (e.g., fixed-float, basis, options, multi-leg, exotic); (3) the variety of data points required to calculate notional amounts (e.g., price, quantity, quantity units, location, grades, exchange rate); (4) locality-specific terms; and (5) lack of industry standards for notional amount-equivalent calculations.66 Given the limitations in the AGNA data, counterparty counts and transaction counts were used as proxies to analyze likely swap dealing activity for participants in the NFC swap market. The analysis conducted for the Proposal largely confirmed the analysis conducted for the Staff Reports; 67 however, there is greater confidence in the results given the improved data and refined methodology. Nonetheless, given the lack of a swap dealing indicator for individual swaps, and the lack of an indicator to identify whether a specific swap need not be considered in determining whether a person is an SD or counted towards the person’s AGNA threshold, staff’s analysis was based on a person’s AGNA of swaps activity, as opposed to AGNA of swap dealing activity. To assess the relative impact on the swap market of potential changes to the De Minimis Exception, CFTC staff analyzed the extent to which the swap market was subject to SD regulation during the review period because at least one counterparty to a swap was a registered SD (‘‘2017 Regulatory Coverage’’). Specifically, with regard to 2017 Regulatory Coverage, staff identified the extent to which: (1) Swaps activity, measured in terms of AGNA or transaction count, was subject to SD regulation during the review period because at least one counterparty to a swap was a registered SD (‘‘2017 AGNA Coverage’’ or ‘‘2017 Transaction Coverage,’’ as applicable); and (2) counterparties in the swap market transacted with at least one registered SD during the review period (‘‘2017 Counterparty Coverage’’). Additionally, staff estimated regulatory coverage by assessing the extent to which the swap market would have been subject to SD regulation at different AGNA thresholds because at least one counterparty to a swap was identified as a ‘‘Likely SD’’ (‘‘Estimated Regulatory Coverage’’). For purposes of this analysis, the term ‘‘Likely SD’’

68See ABA, AGA, AFEX/GPS, BDA, Capital One, Cboe SEF, Citizens, CDEU, COPE, CEWG, CMC, EEI/EPSA, FXPA, Frost Bank, FIA, IIB, IECA, ISDA/ SIFMA, JBA, M&T, NCFC, NRECA/APPA, NGSA, Regions, SVB, Virtu, Western Union, and XTX comment letters. 69See ABA, AFEX/GPS, BDA, Capital One, Citizens, FIA, IIB, IECA, JBA, Regions, and SVB comment letters. 70Additionally, CDEU and CEWG referenced the Congressional Directive stating that the Commission should establish a threshold of $8 billion or greater within 60 days of enactment of the Appropriations Act (i.e., by February 16, 2016), while CEWG also cited to the recent recommendation from the U.S. Department of the Treasury to set the threshold at $8 billion. See CDEU and CEWG comment letters; Accompanying Statement to the Consolidated Appropriations Act of 2016, Explanatory Statement Division A at 32 (Dec. 2015), available at http:// docs.house.gov/meetings/RU/RU00/20151216/ 104298/HMTG-114-RU00-20151216-SD002.pdf; H. Rpt. 114–205 at 76 (July 14, 2015), available at https://www.congress.gov/114/crpt/hrpt205/CRPT￾114hrpt205.pdf; U.S. Department of the Treasury, A Financial System That Creates Economic Opportunities—Capital Markets (available at https://www.treasury.gov/press-center/press￾releases/Documents/A-Financial-System-Capital￾Markets-FINAL-FINAL.pdf). 71See ABA, AGA, AFEX/GPS, BDA, Capital One, Citizens, CDEU, COPE, CEWG, CMC, EEI/EPSA, Frost Bank, IIB, IECA, ISDA/SIFMA, JBA, M&T, NCFC, NRECA/APPA, NGSA, SVB, Virtu, and Western Union comment letters. 72See id. 73See AGA, BDA, Capital One, CDEU, CMC, Frost Bank, IECA, M&T, SVB, and Western Union comment letters. 74See Citizens, IECA, NRECA/APPA, NGSA, and SVB comment letters. 75See Citizens comment letter. 76See AFEX/GPS, Capital One, COPE, EEI/EPSA, FXPA, FIA, IECA, ISDA/SIFMA, JBA, M&T, NGSA, and Regions comment letters. 77See Citizens, Virtu, and Western Union comment letters. 78See Citizens and Virtu comment letters. 79See EEI/EPSA and NGSA comment letters. As stated by EEI/EPSA, if NFC prices increase, the same level of swaps activity will potentially have a higher notional amount. 80See EEI/EPSA comment letter. 81See NGSA comment letter. 82See ABA, IECA, and SVB comment letters. Although addressed by ABA and SVB, the costs associated with SD regulatory requirements (e.g., margin, reporting, technology, etc.) are not considered in the cost-benefit analysis below. See infra notes 249 and 286. 83See ABA comment letter. 84See IECA and SVB comment letters. Although outside of the scope of this rulemaking, IECA also Continued refers to an In-Scope Entity that exceeded a specified AGNA threshold level, and traded with at least 10 unique counterparties. With regard to Estimated Regulatory Coverage, staff identified the extent to which: (1) Swaps activity, measured in terms of AGNA or transaction count, would have been subject to SD regulation during the review period, at a specified AGNA threshold, because at least one counterparty to a swap was
identified as a Likely SD at that AGNA threshold (‘‘Estimated AGNA Coverage’’ or ‘‘Estimated Transaction Coverage,’’ as applicable); and (2) counterparties in the swap market would have transacted with at least one Likely SD during the review period, at a specified AGNA threshold (‘‘Estimated Counterparty Coverage’’). B. Summary of Comments

1. Set Threshold at $8 Billion
   Most commenters that addressed this aspect of the Proposal stated that the AGNA threshold should not decrease to $3 billion, and/or supported setting the threshold at $8 billion.68 Some of those commenters also stated that the Commission could or should consider a higher threshold, as discussed in more detail in section II.B.2 below.69 Commenters generally stated that the policy goals for SD registration— reducing systemic risk, increasing counterparty protections, and/or increasing market efficiency, orderliness, and transparency—and the policy goals for a de minimis exception—increasing efficiency, allowing limited ancillary dealing, encouraging new participants, and/or focusing regulatory resources—would be better advanced if the threshold did not decrease to $3 billion.70 Specifically, commenters stated that a reduced AGNA threshold could lead to some entities reducing or ceasing swaps activity to avoid registration and its related costs, which could lead to negative impacts for swap market participants. For example, fewer de minimis dealers could mean that small and mid-sized end-users and commercial entities who utilize swaps for hedging purposes, as well as NFC swap market participants, would have fewer dealers available to them.71 The potential negative impacts could include: (1) Increased concentration in the swap dealing market; (2) reduced availability of potential swap counterparties; (3) reduced liquidity; (4) increased volatility; (5) increased systemic risk; and/or (6) higher fees or reduced competitive pricing.72 Several commenters also noted that the current $8 billion threshold already subjects the vast majority of transactions to SD regulation, or that a reduced threshold would not capture significant additional dealing activity.73 Some commenters stated that the nature of the swaps activity entered into by certain entities poses less systemic risk—e.g., commercial banks that have swap dealing activity below $8 billion and may be subject to prudential banking rules, and entities that primarily enter into NFC swaps.74 More specifically, Citizens noted that prudential regulators examine the safety and soundness of middle-market banks’ swap businesses, and the swaps offered by these banks are structured conservatively to assist customers with hedging activities. Further, with respect to counterparty protections, Citizens stated that many middle-market banks that would potentially have to register at a lower threshold likely already perform, under applicable prudential banking rules, know-your-counterparty and suitability analyses of their counterparties prior to entering into swaps with them.75 Several commenters stated that maintaining the $8 billion threshold provides regulatory stability or alleviates the uncertainty currently experienced by market participants with an AGNA of swap dealing activity between $3 billion and $8 billion.76 Some commenters suggested that maintaining the $8 billion threshold would enable the
   Commission to focus its limited resources on entities whose swap dealing is sufficient in size and scope to warrant oversight.77 Two commenters also noted that Commission regulations not related to SD registration (e.g., part 43 and 45 reporting requirements, and mandatory clearing and swap execution facility (‘‘SEF’’) trading requirements) already apply to unregistered entities, and therefore, many of the policy goals of SD registration are already being advanced with respect to swaps entered into by these unregistered entities.78 With respect to NFC swaps, EEI/EPSA and NGSA expressed concern that a lower AGNA threshold would provide less accommodation for increasing NFC prices, which could lead to market participants reducing their swap dealing activity to remain below the threshold.79 To address concerns regarding volatility in NFC prices, EEI/EPSA also suggested that the AGNA threshold be adjusted annually, consistent with the consumer price index.80 NGSA also stated that the lower regulatory coverage for NFC swaps is appropriate given the characteristics of that market.81 A few commenters addressed the compliance costs associated with SD registration,82 stating that: (1) Establishing an $8 billion threshold results in aggregate recurring compliance costs over a 10-year period, on a net present value basis, of approximately $373 million; 83 and (2) the cost of SD registration (e.g., systems build-out, external advisors, National Futures Association membership dues, compliance with margin rules) is underestimated,84 with one commenter

asserted that the Commission underestimates the negative impact on market development due to its failure to provide a workable capital rule for non￾bank SDs. 85See SVB comment letter. 86See BDA comment letter. 87See ABA, AFEX/GPS, BDA, Capital One, Citizens, FIA, IIB, IECA, JBA, Regions, and SVB comment letters. 88See ABA, AFEX/GPS, BDA, Citizens, IIB, and SVB comment letters. 89See ABA and AFEX/GPS comment letters. 90See JBA comment letter. 91See AFEX/GPS comment letter. 92See AFEX/GPS and Citizens comment letters. 93See AFEX/GPS, BDA, Citizens, and SVB comment letters. 94See ABA comment letter. 95See Better Markets and Senators comment letters. 96See Better Markets comment letter. 97See Senators comment letter. 98See Better Markets comment letter. 99See id. 100See id. 101See id. estimating that the initial cost would be approximately $8 to $10 million per entity, with ongoing costs to meet regulatory requirements of $2 million per year thereafter.85 BDA stated that the CFTC should clarify whether changes to the De Minimis Exception would be applicable to activity that occurred in the preceding 12 months.86
2. Increase Threshold
Some commenters stated that the Commission should also consider a higher AGNA threshold, maintaining generally that the policy goals for SD registration and a de minimis exception would be better advanced if the threshold was higher than $8 billion.87 Specifically, several commenters stated that an increased threshold would not lead to a significant decrease in regulatory coverage of swap dealing activity.88 ABA and AFEX/GPS asserted that a $20 billion threshold would result in a trivial or non-consequential reduction in Estimated Regulatory Coverage,89 and JBA stated that at a $100 billion threshold, Estimated AGNA Coverage would be almost the same.90 AFEX/GPS also asserted that the cumulative swaps activity conducted by SDs between $8 billion and $20 billion does not pose systemic risk, and entities would still be subject to reporting rules and recordkeeping requirements.91 Additionally, AFEX/GPS and Citizens asserted that a decrease in the number of registered SDs would focus the Commission’s resources on SDs whose dealing activity is sufficient in size and scope to warrant greater oversight.92 Further, a few commenters stated that given the costs of SD registration, a higher threshold would encourage new participants to engage in swap dealing activity, which SVB noted as important given the highly concentrated nature of the SD market, where the nation’s largest banks control the vast majority of swap market share.93 Additionally, ABA indicated that an increased threshold would result in aggregate compliance cost savings for market participants. For example, AGNA thresholds of $15 billion and $50 billion would result in potential aggregate savings of $81 million and $170 million, respectively, on a net present value basis, as compared to an $8 billion threshold.94
3. Allow Threshold to Decrease
Better Markets and the Senators stated that the Commission should permit the AGNA threshold to decrease to $3 billion, contending generally that the data insufficiently or misleadingly justifies maintaining the threshold at $8 billion,95 and arguing that the Proposal did not follow necessary administrative procedures or exceeded statutory authority.96 The Senators stated that though notional amount data for NFC swaps was not used in considering the Proposal, the data that was available for NFC swaps shows significantly less regulatory coverage under an $8 billion threshold than in other asset classes. The Senators commented that though the Proposal notes the ‘‘unique characteristics’’ of NFC swaps, the analysis provided to justify the $8 billion threshold indicates a series of assumptions and possibilities rather than concrete data. The Senators also questioned why, given the lack of relevant notional amount data for NFC swaps, it is necessary to maintain the $8 billion threshold for SDs involved with energy-related swaps.97 Better Markets claimed that the regulatory coverage statistics are incomplete, misleading, and irrelevant to the Dodd-Frank Act’s activities-based standard for SD registration, stating that the high AGNA and transaction coverage percentages are not indicative of the absolute level of swap dealing activities relevant to SD registration under CEA section 1a(49)(A). Further, in connection with the 680 additional counterparties that would potentially benefit from SD regulations under a lower $3 billion threshold, Better Markets asserted that expanding counterparty protections to hundreds of market participants would have more than a ‘‘limited’’ effect on counterparty protection once relative statistics are abandoned.98 Better Markets also asserted that the data filtering methodology was flawed and inadequately explained. Better Markets explained that, with respect to the 10 counterparty count filter, if a commodities affiliate of a large firm held itself out as an SD or stood ready to accommodate the demand of nine counterparties, that affiliate should have been treated, for purposes of the analysis, as an SD on account of its swap dealing activities, unless those activities did not exceed the AGNA threshold or otherwise were excluded from the SD registration analysis. Further, Better Markets argued that: (1) The CFTC should have provided an opportunity for public comment on the assumptions that were made in the CFTC’s analysis; (2) there was some ambiguity in the terms used in the CFTC’s analysis; (3) the CFTC’s reliance upon a 10 unique counterparty filter was based on fatally flawed logic; (4) the data limitations demonstrate the benefits of better field-level and affiliate reporting of swaps, which would give the CFTC an informed basis to consider changes to the $3 billion threshold; and (5) the CFTC must first amend its swap data and chief compliance officer reporting regulations to ensure it has sufficient data to provide
an informed basis for administrative action.99 Further, Better Markets commented that the de minimis threshold framework should be revised to focus on strict, observable measures like total notional amount or transactional activities, rather than a subset of such activities that potential registrants are able to interpret for themselves, and are not presently required by regulation to monitor, report, or internally track across the firm.100 Better Markets also asserted that the statutory provision regarding the de minimis exception authorizes the CFTC to issue exemptive orders for individual or similarly-situated legal entities based upon generally applicable factors for determining whether such entities may be involved in a de minimis amount of swap dealing activities. Better Markets noted that it is unreasonable to conclude that Congress intended a wholesale exemption from registration that is divorced from the particular circumstances of any one petitioner. Further, Better Markets argued that the language in the exemptive mandate must be construed in a manner that is faithful to Congress’ intent that the quantity of exempted swap dealing activities be minimal, a concept that has boundaries that can be drawn far short of billions of dollars and thousands of transactions by unregulated entities.101

102See AFR comment letter.
103See id.
104See IATP comment letter.
105See ABA, CMC, Frost Bank, and IECA comment letters. 106See ABA, CMC, and IECA comment letters. 107See Frost Bank comment letter. 108See BDA comment letter. 109See IECA comment letter. 110See AFR comment letter. 111See id. 112See Better Markets comment letter. 113See IECA comment letter. 114See ABA and Citizens comment letters. 115See ABA comment letter. ABA also suggested that the Commission could consider other market risk metrics, such as value at risk and sensitivities, as well as credit risk metrics, such as total swaps current exposure net of collateral received and largest fifteen swap counterparty current exposures net of collateral received. 116See IECA comment letter. 117See JBA comment letter. 118See IATP comment letter. AFR stated that, though the improved data adds weight to the claim that an $8 billion threshold is appropriate for some financial swaps, arguments against the $8 billion threshold are particularly strong in the case of NFC markets. Specifically, AFR asserted that the Commission should be willing to vary the de minimis threshold based on market characteristics, and in particular should reduce the $8 billion threshold in NFC markets where $8 billion in notional amount represents a different level of economic significance than in some other markets. AFR elaborated that the Commission continues to lack data on the notional amount for NFC swaps, making it difficult to draw definitive conclusions on the economic significance of the activity that is not subject to SD regulation, and stated that significant dealing activity in the NFC market is not subject to SD regulation since roughly half of all the entities with 10 or more NFC swap counterparties are not registered as SDs.102 AFR also stated that the AGNA threshold analysis does not account for the numerous other exceptions proposed, which could exclude very large amounts of swaps activity from being considered in the de minimis calculation.103 IATP stated that the data analysis does not support the idea that more ancillary dealing would promote greater competition, and thus more efficient and transparent price discovery. IATP asserted that the Commission’s true motivation for maintaining an $8 billion threshold is the regulatory compliance cost and burden reduction objective of Project KISS, rather than promoting improved price discovery. Further, IATP claimed that the AGNA of activity in the swap market has shrunk due to the clearing of swaps on centralized platforms and the migration of swaps to the futures markets, not because of constraints of the de minimis threshold or because of the lack of exemptions to the calculation of that threshold. IATP also stated that though it did not have a data-based argument for changing the $8 billion threshold, it believed that maintaining the $8 billion threshold because of potential administrative burdens involved in lowering the threshold is a poor, Project KISS-based, rationale that does not consider the benefits of SD
registration for the financial integrity and price discovery of the swap market.104
4. Other Comments
(i) Testing Frequency for Threshold Some commenters addressed the testing frequency for the threshold. Commenters stated that the AGNA threshold calculation should continue to be based primarily on a rolling 12- month test of the AGNA of swap dealing activity.105 Specifically, commenters indicated that: (1) Resources have been spent and systems have been built to comply with the current approach, and additional changes would add costs with no tangible benefit; 106 and (2) the current test is relatively simple to administer, and the 12-month testing period helps to smooth out any short￾term aberrations in activity and allows for moderation of future swap dealing activity to avoid inadvertently triggering an SD registration requirement.107 However, BDA stated that the CFTC should allow entities to test only at the end of every month, which would significantly reduce the compliance testing burdens for small and mid-sized firms.108 (ii) Alternatives to Single AGNA Threshold A number of commenters addressed whether the Commission should consider an alternative to a threshold based on the AGNA of swap dealing activity. AFR and IECA noted that using AGNA as the relevant criterion for SD registration, as compared to other options, is beneficial because: (1) Resources have been expended to comply with the current approach, and changing that approach would add costs for no perceived benefit; 109 and (2) AGNA provides a stable metric of the gross size of swaps commitments that is not reliant on either current market valuations, model forecasts, or institutional arrangements such as bankruptcy procedures.110 AFR stated that controlling operational risk, not simply market risk, is a major reason for SD designation, and AGNA remains a good measure of the total operational risks incurred by an entity,111 and Better Markets maintained that the de minimis exception must require consideration of the quantity of swap dealing, not net exposures or other risk-based measures.112 However, IECA indicated that although using an alternative netting option (e.g., entity-netted notional amounts) is a reasonable idea and could be incorporated into existing analysis, in the NFC markets, netting would need to be done as a measure of credit exposure with physical and bilateral swaps being able to be offset against each other in connection with perceived ‘‘risk exposure’’ to a third party.113 Additionally, ABA and Citizens stated that the Commission should consider a risk-based de minimis exception.114 ABA asserted that a notional amount￾based threshold is not the appropriate metric for the De Minimis Exception because it is not based on risk, and suggested that the Commission consider initial margin as the relevant metric.115 Commenters also stated that a tiered SD registration structure should not be considered, noting that a tiered structure could: (1) Create more uncertainty for situations where legal and regulatory certainty is important; 116 and (2) subject entities to
instability and inefficiency relative to a permanent, single AGNA de minimis threshold.117 On the other hand, IATP asserted that the Commission should propose, after further analytic work, a tiered SD registration for SDs with a certain threshold of NFC swaps activity (e.g., via commodity indexes).118 Several commenters also addressed whether the Commission should consider counterparty count and transaction count as additional metrics to be included in the de minimis threshold, as discussed in section IV.A below. (iii) Additional Calculation Changes Commenters addressed other calculation changes the Commission should consider for the de minimis threshold. Virtu stated that the CFTC should exempt swap transactions where one party is a registered SD or one party holds their account with a registered SD since these transactions are already subject to the existing reporting

119See Virtu comment letter.
120See JBA comment letter.
121See BDA comment letter.
122See Virtu comment letter. Virtu noted that, while in aggregate the number of transactions engaged in by market makers might exceed the $8 billion threshold, the net risk of these trades would not have the same potential impact to overall systemic risk because exempt market makers’ open net positions in otherwise non-exempt transactions would be capped at $1 billion over a rolling 12- month period. Additionally, certain market makers access the market through prime brokers—who are registered SDs—and, as such, these transactions would be included in the prime brokers’ regulatory reports and subject to CFTC oversight. 123See IIB comment letter. 124See JBA comment letter. 125See Western Union comment letter (referring to Cross-Border Application of the Registration Thresholds and External Business Conduct Standards Applicable to Swap Dealers and Major Swap Participants, 81 FR 71946 (proposed Oct. 18, 2016)). Western Union also stated that the proposed application of the foreign consolidated subsidiary definition to SD registration is inconsistent with principles of international comity and would create an unfair competitive disadvantage for certain market participants. 126See IIB comment letter. 127The Commission also notes that the data analysis discussed in this adopting release and the Proposal confirmed the analysis conducted for the Staff Reports. See generally 83 FR 27449–58; Final Staff Report, supra note 19; Preliminary Staff Report, supra note 17. 128See generally 83 FR 27450–58. 129See supra section II.B.1. 130See generally supra section II.B.1; 83 FR 27450–58. See also Final Staff Report, supra note 19; Preliminary Staff Report, supra note 17. 131The actual number of entities without a single transaction with a registered SD was likely lower than 6,440. Of the 6,440 entities, 1,780 had invalid identifiers that staff was unable to manually replace with a valid LEI. It is possible that these 1,780 invalid identifiers actually represented fewer than 1,780 distinct counterparties because one counterparty may be associated with multiple invalid identifiers. See 83 FR 27451. requirements and, as such, Commission oversight.119 JBA stated that the CFTC should specify that the termination and modification of terms and conditions of existing transactions do not count towards the threshold, noting that termination of transactions mitigates counterparty credit risk and reduces the outstanding AGNA of swaps.120 BDA argued that the CFTC should consider increasing the ‘‘special entity’’ threshold to $100 million in order to provide special entities with more access to the marketplace. BDA maintained that the $25 million threshold results in many mid-sized firms deciding not to enter into swaps with special entities, while an increase in that threshold could provide better market access for special entities while having no material impact on the overall regulation of SDs.121 Virtu asserted that transactions by market makers maintaining net open
positions not exceeding $1 billion (over a 12-month period) should be exempted from the de minimis threshold calculation.122 Virtu explained that certain market makers do not hold positions or carry risk for long periods of time, but rather seek to facilitate efficient risk transference to earn a spread and, in doing so, lower costs for investors through increased price competition and more transparency in the market. IIB stated that entities that have discontinued new swap dealing activity should not have to count towards their AGNA threshold certain transactions that modify legacy swaps entered into by those entities, including: (1) Partial or full terminations; (2) modifications that shorten the duration of an outstanding swap; (3) partial or full novations of legacy swap transactions; or (4) swaps submitted for clearing.123 (iv) Cross-Border Issues With respect to cross-border issues, JBA stated that the market has been divided into two groups because non-SD entities outside of the U.S. avoid transactions with U.S. persons, thereby undermining the diversity of U.S. markets.124 Additionally, Western Union suggested that the Commission should also address the foreign consolidated subsidiary rules in the context of the De Minimis Exception rulemaking.125 Further, IIB stated that the Commission should clarify that a swap between a non-U.S. person and a non-U.S. asset manager that is subject to post-trade allocation and submitted for clearing, or given up to a non-U.S. prime broker prior to being allocated, should not count towards the AGNA threshold in certain circumstances.126
C. Final Rule and Commission Response
Upon consideration of the comments,127 the Commission is adopting an amendment to paragraph (4)(i)(A) of the De Minimis Exception to set the AGNA swap dealing threshold at $8 billion over the immediately preceding 12 months, as proposed. The Commission is also adopting the other conforming and clarifying changes as proposed.

1. Rationale for Not Reducing AGNA
   Threshold to $3 Billion
   As discussed in the Proposal,128 as well as by most commenters that addressed this aspect of the Proposal,129 the policy objectives underlying SD regulation—reducing systemic risk, increasing counterparty protections, and increasing market efficiency, orderliness, and transparency—would not be significantly advanced if the threshold decreased to $3 billion. Additionally, the policy objectives furthered by a de minimis exception— increasing efficiency, allowing limited ancillary dealing, encouraging new participants, and focusing regulatory resources—would not be significantly advanced, and may be impaired to some extent, if the threshold decreased. Generally, as discussed in the Proposal and as agreed with by most commenters, analysis of the data indicated that: (1) The current $8 billion threshold subjects almost all swap transactions (as measured by AGNA or transaction count) to SD regulations; (2) at a lower threshold of $3 billion, there would only be a small amount of additional AGNA and swap transactions subject to SD regulation, and there would potentially be reduced liquidity in the swap market, as compared to the $8 billion threshold; and (3) a lower threshold could lead to reduced liquidity for NFC swaps, negatively impacting end-users who utilize NFC swaps for hedging purposes.130 (i) High Regulatory Coverage at $8 Billion Threshold During the review period, almost all swap transactions involved at least one registered SD as a counterparty—greater than 99 percent for IRS, CDS, FX swaps, and equity swaps. For NFC swaps, approximately 86 percent of transactions involved at least one registered SD as a counterparty. Overall, approximately 98 percent of transactions involved at least one registered SD. Further, almost all AGNA of swaps activity included at least one registered SD—greater than 99 percent for IRS, CDS, FX swaps, and equity swaps. The Commission notes that the 2017 Counterparty Coverage was approximately 83.5 percent—i.e., approximately 16.5 percent of the counterparties in the swap market did not transact with at least one registered SD on at least one swap (6,440 counterparties out of a total of 39,107), and therefore potentially did not benefit from the counterparty protection aspects of SD regulations.131 However, given the 2017 AGNA Coverage and 2017 Transaction Coverage statistics, these 6,440 entities had limited overall swaps activity. Accordingly, to the extent these 6,440 entities were engaged in swap dealing activities, such activity was likely ancillary and in connection with other client services, potentially advancing the policy rationales behind a de minimis exception. This data signifies that nearly all swaps already benefited from the policy considerations discussed above (e.g., reducing systemic

132This analysis is discussed in greater detail in the Proposal, and was also addressed by commenters. See supra section II.B.1; 83 FR 27450– 52. 133This analysis is discussed in greater detail in the Proposal, and was also addressed by commenters. See supra section II.B.1; 83 FR 27452– 54. 134See supra section II.B.1; 83 FR 27452–54. See also Final Staff Report, supra note 19. 135See supra section II.B.1; 83 FR 27452–54. 136See supra section II.B.1; Citizens, IECA, NRECA/APPA, NGSA, and SVB comment letters. 137This analysis is discussed in greater detail in the Proposal, and was also addressed by commenters. See supra section II.B.1; 83 FR 27452–
57. See also CMC, IECA, and NGSA comment
letters.
138See 17 CFR 1.3, Swap dealer, paragraph (6)(iii); 83 FR 27456–57. 139See supra section II.B.3. 140See supra section II.B.3; Better Markets comment letter. 141 7 U.S.C. 1a(49)(D). 142SD Definition Adopting Release, 77 FR 30626; SD Definition Proposing Release, 75 FR 80179. risk, increasing counterparty protections, and increasing market efficiency, orderliness, and transparency) at the existing $8 billion threshold.132 (ii) Minimal Additional Regulatory Coverage at Lower Threshold Given the high percentage of swaps that were subject to SD regulation at the existing $8 billion threshold during the review period, a lower threshold of $3 billion would result in only a small amount of additional activity being directly subjected to SD regulation. Specifically, the Estimated AGNA Coverage would have increased from approximately $221,020 billion (99.95 percent) to $221,039 billion (99.96 percent)—an increase of $19 billion (a
0.01 percentage point increase). The
Estimated Transaction Coverage would have increased from 3,795,330 trades (99.77 percent) to 3,797,734 trades (99.83 percent)—an increase of 2,404 trades (a 0.06 percentage point increase). The Estimated Counterparty Coverage would have increased from 30,879 counterparties (88.80 percent) to 31,559 counterparties (90.75 percent)— an increase of 680 counterparties (a 1.96 percentage point increase). These small increases in Estimated Regulatory Coverage indicate that the systemic risk mitigation, counterparty protection, and market efficiency benefits of SD regulation would be enhanced in only a very limited manner if the threshold decreased from $8 billion to $3 billion. Additionally, the limited regulatory and market benefits of a $3 billion threshold should be considered in conjunction with the costs associated with a lower threshold (e.g., costs of implementing policies and procedures, technology systems, and training programs to address requirements imposed by SD regulations).133 Additionally, as discussed by the Commission and most commenters, a $3 billion AGNA threshold could lead certain entities to reduce or cease swap dealing activity to avoid registration and its related costs.134 Generally, the costs associated with registering as an SD may exceed the profits from dealing swaps for entities with limited dealing activities. This could lead to negative impacts for swap market participants, including, but not limited to, small and mid-sized end-users who use swaps for hedging purposes. Reduced swap dealing activity could lead to increased concentration in the swap dealing market, reduced availability of potential swap counterparties, reduced liquidity, increased volatility, increased systemic risk, and/or higher fees or reduced competitive pricing. The end-user counterparties of these smaller swap dealing entities may be adversely impacted by the above consequences and could face a reduced ability to use swaps to manage their business risks.135 Additionally, as noted by some commenters, the nature of the swaps activity entered into by certain entities poses less systemic risk—e.g., commercial banks that have swap dealing activity below $8 billion and entities that primarily enter into NFC swaps.136 Further, although approximately 86 percent of NFC swaps involved at least one registered SD compared to approximately 99 percent for other asset classes, as discussed in the Proposal, the Commission is of the view that lower SD regulatory coverage is acceptable given the special characteristics of the NFC swap market. A reduced threshold likely would have negative impacts on NFC swap liquidity as some entities (e.g., small and mid-sized banks and/or non-financial entities) reduce dealing to avoid registration and its related costs. This would be detrimental to the end￾users who do not have trading relationships with larger, financial￾entity SDs, and who rely on small to mid-sized banks and/or non-financial entities to access liquidity in the
wider swap market. Additionally, even if the threshold decreased, the available data leaves it unclear if or to what extent the 2017 Counterparty Coverage statistic of 86 percent would increase for NFC swaps since several of those entities may already have less than $3 billion in AGNA of swap dealing activity. Further, many of the entities engaged in limited swap dealing activity for NFC swaps appear to have a specialized role in the market, in that their primary business is generally non-financial in nature and the swap dealing activity is ancillary to their primary role in the market.137 Finally, entities that are active in the NFC swap market may utilize the existing physical position hedging exemption, which is more directly applicable to the NFC asset class than to other swaps.138 (iii) Response to Commenters Advocating Lower Threshold The Commission disagrees with the few commenters that stated that the AGNA threshold should decrease to $3 billion.139 Better Markets stated that the high regulatory coverage ratios are not indicative of the absolute level of swap dealing activities relevant to SD registration, and asserted that maintaining an $8 billion threshold would have more than a limited detrimental effect on counterparty protections.140 The Commission notes that the statutory requirements do not dictate a specific methodology for assessing the de minimis exception, such as the focus on the absolute level of swap dealing suggested by Better Markets. Rather, the CEA requires the Commission to promulgate regulations to establish factors with respect to the making of a determination to exempt from designation as an SD an entity engaged in a de minimis quantity of swap dealing, without stating additional requirements.141 Additionally, as stated in the SD Definition Proposing Release and the SD Definition Adopting Release, the de minimis exception ‘‘should be interpreted to address amounts of dealing activity that are sufficiently small that they do not warrant registration to address concerns implicated by the regulations governing swap dealers and security-based swap dealers. In other words, the exception should apply only when an entity’s dealing activity is so minimal that applying dealer regulations to the entity would not be warranted.’’ 142 This decision inherently requires judgment, and for that reason the Commission has considered whether entities that have less than $8 billion in swap dealing activity meet this standard. Given the nature of the swap market and the Commission’s analysis of the data, requiring an entity that has less than $8 billion in swap dealing activity to register as an SD is not warranted because it would not appreciably impact the systemic risk, counterparty protection, and market efficiency considerations of SD regulation, but

143As discussed, the analysis conducted in connection with the Proposal was consistent with the analysis conducted in connection with the Staff Reports. See generally 83 FR 27449–58; Final Staff Report, supra note 19; Preliminary Staff Report, supra note 17. 144 77 FR 30626. See also 75 FR 80179. 145As noted in the SD Definition Adopting Release, ‘‘implementing the de minimis exception requires a careful balancing that considers the regulatory interests that could be undermined by an unduly broad exception as well as those regulatory interests that may be promoted by an appropriately limited exception.’’ 77 FR 30628. 146See Better Markets comment letter. 147 77 FR 30629–30. 148See supra section II.B.3; Better Markets and Senators comment letters. 149See supra section II.B.3; Better Markets comment letter. 150See Better Markets comment letter. 151See supra section II.A; 83 FR 27449–50. 152See 83 FR 27449. 153 77 FR 30632. 154 Id. at 30632–33. would negatively impact the policy considerations underlying the de minimis exception by reducing the amount of swap dealing allowed under the exception.143 Thus, the Commission concludes that the $8 billion threshold is consistent with a key rationale behind the de minimis exception because it would permit ‘‘amounts of dealing activity that are sufficiently small that they do not warrant registration.’’ 144 No individual policy factor was dispositive in the Commission’s analysis. Rather, the Commission considered all of the policy factors when assessing the regulatory coverage ratios.145 As noted above in section II.B.3, Better Markets also asserted that the statutory provision regarding the de minimis exception authorizes the CFTC to issue exemptive orders for individual or similarly-situated legal entities based upon generally applicable factors for determining whether such entities may be involved in de minimis swap dealing activities. Better Markets contends that it is unreasonable to conclude that Congress intended a wholesale exemption from registration that is divorced from the particular circumstances of any one petitioner.146 As noted, however, the CEA states that the Commission shall promulgate factors, through regulation, regarding the De Minimis Exception determination. Nothing in the statutory language prohibits the Commission from establishing a de minimis exception that is self-effectuating. The Commission believes that the $8 billion threshold appropriately excludes entities ‘‘whose dealing activity is sufficiently modest in light of the total size, concentration and other attributes’’ of the swap market and for which SD regulation ‘‘would not be expected to contribute significantly to advancing the customer protection, market efficiency and transparency objectives of dealer regulation.’’ 147 The Commission sees no basis in the record or requirement in the statute to treat entities differently when they are similarly situated in this respect. Also as noted above, with respect to the data
analysis methodology, Better Markets and the Senators stated that the data insufficiently or misleadingly justifies maintaining the threshold at $8 billion.148 Better Markets also asserted that: (1) The CFTC should have provided an opportunity for public comment on alternative assumptions; (2) there is some ambiguity in the terms used in the CFTC’s analysis; (3) the CFTC’s reliance upon a 10 unique counterparty filter is based on fatally flawed logic; (4) the data limitations argue for better field-level and affiliate reporting of swaps, which would give the CFTC an informed basis to consider changes to a $3 billion threshold; and (5) the CFTC must first amend its swap data and chief compliance officer reporting regulations to ensure it has sufficient data to provide an informed basis for administrative action.149 Each of these comments will be addressed in turn. First, with respect to Better Markets’ comment that the Commission should have provided an opportunity for public comment on alternative assumptions for the data analysis, the Commission notes that the methodology used by Commission staff to analyze data in relation to the de minimis threshold was first laid out in the Preliminary Staff Report, on which the public had the opportunity to comment. The Final Staff Report updated that analysis, and then the Proposal explained how the data related specifically to the proposal to maintain the $8 billion threshold. As discussed in the Proposal, the updated analysis largely confirmed the analysis conducted for the Staff Reports. However, there is greater confidence in the results given the improved data and refined methodology. The Commission believes that the public has had an appropriate opportunity to comment on the data, the methodology, the assumptions about the data, and how the data relates to the maintenance of the $8 billion threshold. Second, the Commission cannot assess Better Markets’ comment that the analysis discussed in the Proposal contained ambiguous terms because Better Markets does not state which terms were ambiguous. Third, the Commission disagrees with Better Markets’ comment that ‘‘the fact that CFTC-registered swap dealers, including every major Wall Street bank, tend to have more than 10 counterparties is irrelevant.’’ 150 The Commission notes that staff used the minimum 10 counterparty count only for analytical purposes, as a heuristic to help isolate those entities that appeared to be dealing. Lacking a dealing field in the data, for the reasons set forth above, staff selected a minimum of 10 counterparties as a conservative estimate to improve the analysis and better identify entities likely engaged in swap dealing.151 The Commission also believes that the 10 counterparty filter is appropriate for purposes of this analysis based on its observations of registered SDs and unregistered entities active in the swap market. As noted in the Proposal, data analysis showed that 83 percent of registered SDs had 10 or more
counterparties, without weighting the results.152 In other words, since the analysis was performed using a non￾weighted ranking, SDs with thousands of counterparties did not bias the results. Fourth, the Commission does not believe that the data limitations warrant a delay in setting the threshold at $8 billion. As discussed, the data has improved since the analysis in the Staff Reports. Further, the Commission believes its analysis was appropriately conservative, particularly given that the volume of activity it analyzed was over￾inclusive (since hedging and other non￾dealing activity could not be excluded), and given that its entity-level exclusions were based on an informed assessment of the likely activity of swap market participants. In the SD Definition Adopting Release, the Commission noted that ‘‘comprehensive information regarding the total size of the domestic swap market is incomplete, with more information available with respect to certain asset classes than others.’’ 153 In 2012, the Commission evaluated the appropriateness of the initial $3 billion AGNA threshold using three primary sources of data: (1) Index CDS; (2) the Quarterly Report on Bank Trading and Derivatives Activities issued by the Office of the Comptroller of the Currency (‘‘OCC’’); and (3) public comments to the 2010 SD Definition Proposing Release.154 At the time, granular, transaction-level swaps data across all swap asset classes was not yet available for review by the Commission. The data now available is significantly more detailed than what was available

155Additionally, Commission staff attempted to accurately identify those entities that, based on their observable business activities, are potentially engaged in swap dealing activity versus those likely engaged in other kinds of transactions. See supra
section II.A; 83 FR 27449.
156The Commission also notes that it recently adopted amendments to its chief compliance officer requirements. See Chief Compliance Officer Duties and Annual Report Requirements for Futures Commission Merchants, Swap Dealers, and Major Swap Participants, 83 FR 43519 (Aug. 27, 2018). 157See supra section II.B.3; Better Markets comment letter. 158See supra section II.B.3; Senators comment letter. As noted above, for NFC swaps, approximately 86 percent of transactions involved at least one registered SD as a counterparty, compared to greater than 99 percent for IRS, CDS, FX swaps, and equity swaps. See supra section II.C.1.i. 159See 17 CFR 1.3, Swap dealer, paragraph (6)(iii); supra section II.C.1.ii; 83 FR 27456–57. 160See supra section II.B.3; AFR comment letter. 161See supra section II.B.1. See, e.g., IECA and NGSA comment letters. See also 83 FR 27456–57; Final Staff Report, supra note 19, at 12 (citing comment letters submitted in response to Preliminary Staff Report, supra note 17). 162See supra section II.B.3; IATP comment letter. 163See Final Staff Report, supra note 19; Preliminary Staff Report, supra note 17. 164See 83 FR 27454–56. 165The decrease would be lower at thresholds of $20 billion and $50 billion, at 0.01 percentage points and 0.04 percentage points, respectively. 166The decrease would be lower at thresholds of $20 billion and $50 billion, at 0.05 percentage points and 0.42 percentage points, respectively. 167The decrease would be lower at thresholds of $20 billion and $50 billion, at 2.80 percentage points and 5.71 percentage points, respectively. 168See supra section II.B.2; 83 FR 27455. 169As noted, the decrease in Estimated Counterparty Coverage would be 2.80 percentage points, 5.71 percentage points, 7.61 percentage points, at thresholds of $20 billion, $50 billion, and $100 billion, respectively. to the Commission when the $3 billion threshold was originally established. The data now includes details such as counterparty pairs, product identifiers, transaction-level data for those market participants active in more asset classes than only index CDS, and transaction￾level data (not just quarterly position data) involving market participants beyond banks subject to OCC reporting. In light of the additional, more detailed data, the Commission believes that the $8 billion threshold continues to be appropriately calibrated to the policy goals of SD registration and the de minimis exception.155 Fifth, for similar reasons, the Commission does not believe it should wait to amend its swap data and chief compliance officer reporting regulations before setting the threshold at $8 billion. As noted above, the Commission believes that it does have sufficient data to support this action, so it is not necessary to wait for future changes to the data reporting regime.156 As noted above, Better Markets also commented that the de minimis threshold framework should be revised to focus on strict, observable measures like total notional amount
or transactional activities, rather than a subset of such activities that potential registrants are able to interpret for themselves, and are not presently required by regulation to monitor, report, or internally track across the firm.157 However, the Commission notes that the statutory definition of ‘‘swap dealer’’ itself limits the scope to swap dealing activity, and therefore, using total notional amount would not be appropriate. As noted, the Senators stated that the data that was available for NFC swaps shows significantly less coverage for that asset class under an $8 billion threshold compared to other asset classes.158 In justifying the $8 billion proposal, the Senators commented that though the Proposal noted the ‘‘unique characteristics’’ of NFC swaps, the analysis provided indicated a series of assumptions and possibilities rather than concrete data. The Senators also questioned whether, given the lack of relevant data for NFC swaps, it is necessary to reduce the threshold for SDs involved with energy-related swaps. However, as discussed in section II.C.1.ii, the Commission believes that a reduced threshold would have a negative impact on NFC swap market liquidity as some entities may reduce dealing to avoid registration and its related costs. Additionally, as noted, entities active in the NFC swap market may utilize the existing physical position hedging exemption, which is more directly applicable to the NFC asset class than other swaps.159 Further, AFR stated that, though the improved data adds weight to the claim that an $8 billion threshold is appropriate for some financial swaps, arguments against the $8 billion threshold are particularly strong in the case of NFC swaps.160 The Commission does not believe a lower threshold for NFC swaps would advance the policy goals of SD registration or the de minimis exception. As noted by the Commission and several commenters, the nature of the NFC swap market poses less systemic risk than financial swaps.161 Additionally, the Commission notes the concerns of reduced liquidity if the threshold is reduced for NFC swaps, including an increased concentration in the market, which could adversely affect end-users who rely on small and mid-sized SDs that do not have to register at an $8 billion threshold. Lastly, the Commission disagrees with IATP’s assertion that promoting improved price discovery is not the true rationale for maintaining an $8 billion threshold, and that rather, the motivation is the regulatory compliance cost and burden reduction objective of Project KISS.162 The Commission has laid out above the various policy-related considerations that justify maintaining an $8 billion threshold; these relate to the regulatory goals of both SD registration in general and of the de minimis exception in particular. Additionally, these goals were discussed in the Staff Reports, well in advance of any comments submitted in response to Project KISS.163
2. Rationale for Not Increasing AGNA
Threshold
Although several commenters suggested a higher threshold, the Commission is declining to increase the AGNA threshold from the current $8 billion level. As discussed in the Proposal,164 at a $100 billion threshold:
(1) The Estimated AGNA Coverage would have decreased from approximately $221,020 billion (99.95 percent) to $220,877 billion (99.88 percent)—a decrease of $143 billion (a
0.06 percentage point decrease); 165 (2)
the Estimated Transaction Coverage would have decreased from 3,795,330 trades (99.77 percent) to 3,773,440 trades (99.20 percent)—a decrease of 21,890 trades (a 0.58 percentage point decrease); 166 and (3) the Estimated Counterparty Coverage would have decreased from 30,879 counterparties (88.80 percent) to 28,234 counterparties (81.19 percent)—a decrease of 2,645 counterparties (a 7.61 percentage point decrease).167 As the Commission and commenters have stated, the small decrease in Estimated AGNA Coverage and Estimated Transaction Coverage at higher thresholds potentially indicates that increasing the threshold to up to $100 billion may have a limited adverse effect on the systemic risk and market efficiency policy considerations of SD regulation.168 Additionally, a higher threshold could enhance the benefits associated with a de minimis exception, for example by allowing entities to increase ancillary dealing activity. However, the Commission is of the view that the decrease in Estimated Counterparty Coverage indicates that fewer entities would be transacting with registered SDs, reducing the counterparty protection benefits of SD regulation if the AGNA threshold increased from $8 billion to $20 billion, $50 billion, or $100 billion.169 The

170See 83 FR 27456–57.
171See supra section II.B.4.i. Potentially, month￾end only testing could marginally encourage competition because newly-established swap dealing businesses (as contrasted to the existing businesses that have adapted to current requirements) could set up only month-end testing as opposed to regular testing. However, the Commission believes that maintaining the current requirements is appropriate even in view of any marginal encouragement of competition that could result from the suggested change. 172See supra sections II.B.4.ii and II.B.4.iii. 173See, e.g., supra sections II.B.4.i and II.B.4.ii. 174See supra section II.B.4.iii; supra note 12; 83 FR 27445 n.14. 175See supra section II.B.4.iv. 176See ABA, BDA, Capital One, CDEU, Citizens, Frost Bank, IIB, ISDA/SIFMA, JBA, M&T, and Regions comment letters. 177See Capital One, Frost Bank, M&T, Regions comment letters. 178See Capital One, Citizens, Frost Bank, M&T, and Regions comment letters. 179 17 CFR 1.3, Swap dealer, paragraph (5)(i)(A). 180See BDA, CDEU, and ISDA/SFIMA comment letters. 181See Capital One and Frost Bank comment letters. 182See ABA and Regions comment letters. Commission also notes that increasing the threshold could result in changes in market behavior that could lead to the regulatory coverage decreasing more than the analysis indicated. Further, maintaining the status quo signals long-term stability of the de minimis threshold, and should provide for the efficient application of the SD Definition, as it allows for long-term planning based on the current AGNA threshold.170
3. Response to Other Comments
With respect to BDA’s comment regarding permitting month-end only testing for the de minimis threshold, the Commission notes that several commenters indicated that the market has adapted to the current requirements and that changes would not be beneficial.171 In particular, the Commission agrees with commenters that the current test is relatively simple to administer, and the 12-month testing period helps to smooth out any short￾term variations in activity. The Commission does not believe that allowing month-end only testing would reduce burdens since persons should already have systems in place to regularly track the level of their swap dealing activity. Therefore, the Commission is not adopting this alternative. Additionally, in response to BDA, the Commission notes that for purposes of the $8 billion threshold calculation, an entity must count activity that took place in the immediately preceding 12 months. Similarly, in response to the commenters that recommended alternatives to the single AGNA threshold or other calculation changes,172 the Commission points out that systems and processes have been established for the current requirements,173 and therefore the Commission is not adopting the proposed adjustments at this time. The Commission may take subsequent action or conduct further study with respect to alternative approaches to the single AGNA threshold, including moving toward a risk-based SD registration metric in the future. The Commission would expect that a change could entail costs as market participants adjust their de minimis threshold calculation processes. Additionally, any modification to the special entity threshold is outside of the scope of the Proposal,174 but as with other suggestions, the Commission may consider this in the future. Lastly, with respect to comments asking that the Commission address cross-border issues,175 this issue is also outside of the scope of this rulemaking.
III. Proposed Rule Amendments Not
Adopted
A. Swaps Entered Into by Insured Depository Institutions in Connection With Loans to Customers

1. Proposal
   The Commission proposed adding an IDI loan-related factor in the De Minimis Exception (the ‘‘IDI De Minimis Provision’’) to address concerns that there are circumstances where swaps not covered by the IDI loan-related swap exclusion in paragraph (5) of the SD Definition (the ‘‘IDI Swap Dealing Exclusion’’) should be excluded from the de minimis calculation. Specifically, the Commission proposed to add specific characteristics that an IDI can consider when assessing whether swaps entered into with customers in connection with loans to those customers must be counted towards the IDI’s de minimis calculation. The proposed IDI De Minimis Provision would have encompassed a broader scope of loan-related swaps than the IDI Swap Dealing Exclusion. The proposed IDI De Minimis Provision included: (1) A lengthier timing requirement for when the swap must be entered into; (2) an expansion of the types of swaps that are eligible; (3) a reduced syndication percentage requirement; and (4) an elimination of the notional amount cap. The IDI could exclude qualifying swaps from the de minimis calculation pursuant to the IDI De Minimis Provision regardless of whether the swaps would qualify for the IDI Swap Dealing Exclusion.
2. Summary of Comments
   Almost all commenters that addressed the IDI De Minimis Provision expressed general support for the proposed amendment.176 Commenters often compared the IDI De Minimis Provision to the IDI Swap Dealing Exclusion. In that regard, commenters generally stated that the IDI De Minimis Provision better aligns the regulatory framework with the risk mitigation demands of bank customers.177 Commenters generally supported proposed new paragraph (4)(i)(C)(1),178 which provided that a swap must be entered into no earlier than 90 days before execution of the loan agreement, or before transfer of principal to the customer, unless an executed commitment or forward agreement for the applicable loan exists. In that event, the 90-day restriction does not apply. In comparison, the IDI Swap Dealing Exclusion in paragraph (5) of the SD Definition requires that a swap must be entered into no more than 90 days before or 180 days after the date of execution of the loan agreement (or date of transfer of principal to the customer).179 On the other hand, three commenters recommended removing the 90-day restriction because it would be detrimental to the IDIs and/or borrowers.180 Additionally, two commenters suggested revisions to the ‘‘executed commitment’’ or ‘‘forward agreement’’ exception to the 90-day restriction.181 Proposed new paragraph (4)(i)(C)(2) stated that for purposes of the IDI De Minimis Provision, a swap is ‘‘in connection with’’ a loan if: (1) The rate, asset, liability or other term underlying such swap is, or is related to, a financial term of such loan; or (2) if such swap is required as a condition of the loan, either under the IDI’s loan underwriting criteria or as is commercially appropriate, in order to hedge risks incidental to the borrower’s business (other than for risks associated with an excluded commodity) that may affect the borrower’s ability to repay the loan. Two commenters requested clarification regarding the proposed ‘‘condition of the loan’’ language.182 Proposed new paragraph (4)(i)(C)(3) stated that the termination date of the swap cannot extend beyond termination of the loan. A few commenters stated that circumstances can be anticipated at the time of loan origination that would support permitting the termination date of the swap to extend beyond

183See ABA, BDA, CDEU, Citizens, and M&T comment letters. 184See ABA, BDA, Capital One, CDEU, IIB, and ISDA/SIFMA comment letters. 185See ABA, BDA, Citizens, and ISDA/SIFMA comment letters. 186See Capital One and M&T comment letters. 187 17 CFR 1.3, Swap dealer, paragraph (5)(i)(E). However, as discussed, pursuant to proposed paragraph (4)(i)(C)(4)(ii), if an IDI is a source of less than a five percent of the maximum principal amount of the loan, the notional amount of all swaps the IDI enters into in connection with the financial terms of the loan cannot exceed the principal amount of the IDI’s loan. See also 83 FR 27461. 188See Capital One and M&T comment letters. 189See ABA, Capital One, and M&T comment letters. 190See ABA, BDA, and Capital One comment letters. 191See ABA and Better Markets comment letter. 192See ABA comment letter. 193See Better Markets comment letter. 194See 83 FR 27462–63. 195See ABA, AGA, AFEX/GPS, BDA, Capital One, CDEU, COPE, CMC, EEI/EPSA, Frost Bank, FIA, IIB, IECA, ISDA/SIFMA, JBA, NRECA/APPA, NGSA, Virtu, and Western Union comment letters. 196See AFR and Better Markets comment letters. 197See ABA, AGA, BDA, Capital One, Citizens, CDEU, EEI/EPSA, Frost Bank, FIA, NGSA, NRECA/ APPA, Virtu, and Western Union comment letters. 198See AFR and Better Markets comment letters. 199See Better Markets comment letter. termination of the loan.183 Additionally, in response to a question in the Proposal, a few commenters stated that in order to qualify for the IDI De Minimis Provision, IDIs should not be required to terminate loan-related swaps if a loan is called, put, accelerated, or goes into default before scheduled termination.184 Proposed new paragraph (4)(i)(C)(4)(i) required an IDI to be, under the terms of the agreements related to the loan, the source of at least five percent of the maximum principal amount under the loan for a related swap not to be counted towards its de minimis calculation, and proposed new paragraph (4)(i)(C)(4)(ii) stated that if an IDI is a source of less than a five percent of the maximum principal amount of the loan, the notional amount of all swaps the IDI enters into in connection with the financial terms of the loan cannot exceed the principal amount of the IDI’s loan in order to qualify for the IDI De Minimis Provision. A few commenters stated that the five percent participation requirement should be eliminated from the IDI De Minimis Provision,185 while two commenters generally supported the five percent requirement.186 The proposed IDI De Minimis Provision did not include the requirement in the IDI Swap Dealing Exclusion that the AGNA of swaps entered into in connection with the loan not exceed the principal amount outstanding,187 and two commenters agreed that there are circumstances where the AGNA of loan-related swaps can exceed the outstanding principal amount of the loan.188 In response to a question in the Proposal, three commenters stated that the CFTC should not impose any prior
notice requirement or other conditions on the ability of IDIs to rely on the proposed IDI De Minimis Provision.189 In response to another question in the Proposal, three commenters stated that there should not be a requirement that swap confirmations reference a specific loan because doing so would add operational complexity for little or no benefit.190 Two commenters discussed whether the IDI De Minimis Provision could be promulgated without a joint rulemaking.191 ABA stated that the Commission is not required to promulgate the IDI De Minimis Provision through joint rulemaking with the SEC.192 However, Better Markets asserted that the CFTC’s position that a ‘‘joint rulemaking is not required with respect to changes to the de minimis exception-related factors’’ is invalid and ‘‘would impermissibly enable the CFTC to conduct an end-run around the statutory joint rulemaking requirement.’’ In particular, Better Markets stated that language potentially permitting unilateral action on the de minimis threshold itself does not permit unilateral regulatory actions affecting core definitional issues that must be accomplished through joint rulemaking.193
3. Commission Response
The Commission has determined not to adopt the IDI De Minimis Provision at this time. The Commission continues to consider the issues raised by commenters. For example, the various contexts in which IDIs enter into swaps with their loan customers, and the relation between those swaps and the larger swap market, may merit further consideration. B. Swaps Entered Into to Hedge Financial or Physical Positions

1. Proposal
   The Commission proposed adding a provision in new paragraph (4)(i)(D) of the De Minimis Exception, to include as a factor whether a swap was entered into primarily for the purpose of hedging and met certain related conditions (the ‘‘Hedging De Minimis Provision’’).194 As proposed, to qualify for the Hedging De Minimis Provision, the primary purpose for the swap would need to be to reduce or otherwise mitigate one or more specific risks to which the person is subject. Proposed paragraph (4)(i)(D)(2) provided that the person entering into the hedging swap could not be the price maker of the hedging swap and receive or collect a bid/ask spread, fee, or other commission for entering into the hedging swap (the ‘‘price maker condition’’). In addition, the proposed Hedging De Minimis Provision included in paragraphs (D)(3) through (D)(5) the following conditions that are similar to conditions in the physical hedging exclusion in paragraph (6)(iii) of the SD Definition: (1) The swap must be economically appropriate to the reduction of risks that may arise in the conduct and management of an enterprise engaged in the type of business in which the person is engaged; (2) the swap must be entered into in accordance with sound business practices; and (3) the swap must not be entered into in connection with activity structured to evade designation as an SD.
2. Summary of Comments
   Most commenters supported including an express hedging exception that would clarify which physical and financial hedging swaps do not need to be included in the AGNA threshold calculation.195 These commenters agreed with the Commission that there is currently some uncertainty and confusion among market participants regarding this determination. However, many of these commenters raised issues with the particular conditions identified in the proposed Hedging De Minimis Provision, and two other commenters objected to inclusion of the Hedging De Minimis Provision.196 Among other issues, the two commenters viewed the Hedging De Minimis Provision as a major expansion of the De Minimis Exception. Generally, commenters supported adding the Hedging De Minimis Provision to the De Minimis Exception to provide more certainty and/or clarity regarding the treatment of hedging activity.197 On the other hand, AFR and Better Markets stated that excepting hedges of swap dealing positions from the de minimis threshold could exclude swaps that appear to be hedges, but are actually dealing swaps.198 Furthermore, Better Markets asserted that a hedge of client facing swap is ‘‘inextricably’’ tied to accommodating customer demands.199 Several commenters noted that the price maker condition included in the proposed Hedging De Minimis Provision could be viewed as more

200See CEWG, CMC, FIA, and IECA comment letters. 201See CDEU, EEI/EPSA, IECA, and Western Union comment letters. 202See ABA, BDA, EEI/EPSA, IECA, IIB, NRECA/ APPA, and Western Union comment letters. 203See COPE and NRECA/APPA comment letters. 204See ISDA/SIFMA comment letter. 205See NRECA/APPA comment letter. 206See AFR and Better Markets comment letters. 207See Better Markets comment letter. Better Markets noted that, in October 2012, DSIO addressed whether hedging activity is included in calculating the de minimis amount when it stated that ‘‘a person must consider the swap in light of all other relevant facts and circumstances to determine whether such hedging activity is swap dealing activity. . . .’’ See Frequently Asked Questions (FAQ)—[DSIO] Responds to FAQs About Swap Entities (Oct. 12, 2012) (‘‘DSIO FAQ Guidance’’), available at https://www.cftc.gov/idc/ groups/public/@newsroom/documents/file/ swapentities_faq_final.pdf. 208 77 FR 30619 n.280 (stating that security-based swaps activity for hedging purposes ‘‘unrelated to activities that constitute dealing’’ would not be expected to lead the person to be a security-based swap dealer). 209See AFR and Better Markets comment letters. AFR described the potential need for a swap-by￾swap analysis and the potential for disputes regarding the proposed anti-evasion provision. 210See supra note 207. 211CFTC Staff Letter No. 12–62, supra note 47. 212See ABA, IIB, ISDA/SIFMA, JBA, and NEX comment letters. 213See ABA, ISDA/SIFMA, and NEX comment letters. 214See IIB comment letter. 215See ABA, IIB, ISDA/SIFMA, and JBA comment letters. 216See IIB comment letter. 217See AFR and IATP comment letters. 218See AFR comment letter. limiting than the existing physical swap hedging exclusion.200 Many commenters expressed concern that the proposed condition would be overly prescriptive, ambiguous, and/or could inadvertently require certain hedging activity to be treated as swap dealing activity.201 In particular, commenters asked that the bid/ask spread limitation be deleted or clarified.202 Conversely, two commenters expressed some support for this condition as proposed.203 ISDA/SIFMA was of the view that the requirement that the primary purpose for entering into the swap must be to reduce or otherwise mitigate one or more ‘‘specific’’ risks is unreasonably restrictive.204 ISDA/SIFMA suggested that the Commission should remove the term ‘‘specific’’ from the regulatory text to better achieve the Commission’s policy objective of encouraging greater use of swaps to hedge risks. On the other hand, NRECA/APPA noted that the specific, but non-exclusive, risks identified in paragraph (4)(i)(D)(1) are consistent with the types of commercial risks that an end-user would hedge.205 AFR and Better Markets objected to the Hedging De Minimis Provision, stating that it could allow even large dealers to escape registration, and that the exclusion of anticipatory hedges allows too much discretion to institutional judgment.206
Better Markets expressed concern that the Hedging De Minimis Provision promotes unregulated swap dealing and is therefore ‘‘not a valid statutory objective.’’ Furthermore, Better Markets stated that the Commission does not need to provide clarity for the existing hedging exemption because the existing standard of using facts and circumstances to distinguish dealing swaps is a ‘‘well-settled framework.’’ 207 Better Markets also asserted that the Commission misinterpreted its prior statements about the use of swaps to hedge dealing positions. However, in doing so, Better Markets cited to language in the joint SD Definition Adopting Release that addressed the definition of ‘‘security-based swap dealer,’’ not ‘‘swap dealer.’’ 208 AFR and Better Markets also asserted that the Hedging De Minimis Provision should not be included in the De Minimis Exception because enforcement of the conditions would be impractical.209
3. Commission Response
The comments generally confirmed that nuanced facts and circumstances may be relevant to determining whether a swap that hedges financial risk, but also has dealing characteristics or is connected to dealing activities, should be counted toward the AGNA threshold. However, the comments also raised specific implementation and compliance issues. For these reasons, the Commission has determined not to adopt the Hedging De Minimis Provision at this time. The Commission confirms that the ‘‘relevant facts and circumstances’’ test established in the SD Definition Adopting Release and further discussed in the DSIO FAQ Guidance 210 continues to be in effect. In doing so, the Commission emphasizes that market participants should continue to evaluate such swaps without consideration of the proposed Hedging De Minimis Provision.
C. Swaps Resulting From Multilateral
Portfolio Compression Exercises

1. Proposal
   The Commission proposed new paragraph (4)(i)(E) of the De Minimis Exception, which would add as a factor in the de minimis calculation whether a swap results from multilateral portfolio compression exercises (‘‘MPCE De Minimis Provision’’). Specifically, the Proposal stated that for purposes of determining whether a person has exceeded the AGNA threshold set forth in paragraph (4)(i)(A), the person may exclude swaps that result from multilateral portfolio compression exercises, as defined in § 23.500 of Commission regulations, to the extent the person does not enter into the multilateral portfolio compression exercise in connection with activity structured to evade designation as an SD. The Proposal was consistent with DSIO no-action relief issued on December 21, 2012 (‘‘Staff Letter 12– 62’’).211
2. Summary of Comments
   Most commenters addressing this aspect of the Proposal supported excepting from the de minimis threshold swaps that result from multilateral portfolio compression exercises,212 stating that multilateral portfolio compression: (1) Advances the Commission’s policy goals of reducing counterparty credit risks by allowing swap market participants with large portfolios to net down the size and number of swaps among them, thus lowering the AGNA of outstanding swaps; 213 and (2) does not involve dealing activity, but rather allows market participants to reduce their risk without implicating any of the other considerations related to SD regulation.214 Several commenters also stated that, given the policy-related similarities between bilateral and multilateral portfolio compression, the Commission should also exclude from counting towards the De Minimis Exception swaps that result from bilateral portfolio compression exercises.215 One commenter asserted that reliance on the ‘‘multilateral portfolio compression exercise’’ definition in § 23.500(h) of Commission regulations may be too limiting.216 On the other hand, AFR and IATP expressed concerns with the MPCE De Minimis Provision.217 AFR stated that the definition of portfolio compression appears overbroad since it goes beyond the termination of fully offsetting swaps to include any exercise which would result in the reduction of current market risks for a set of swaps, even if the exercise might actually increase credit exposure or market risk under stressed market conditions.218 IATP noted that entities should be required to document and report the results of multilateral compression exercises to qualify for the exception. Additionally, IATP stated

219See IATP comment letter.
220The Commission notes that Staff Letter 12–62 is not affected by the Commission’s determination not to adopt the MPCE De Minimis Provision at this time. 221See ABA, Citizens, CEWG, CMC, EEI/EPSA, FIA, Frost Bank, IIB, NGSA, and Western Union comment letters. 222See Citizens, EEI/EPSA, FIA, Frost Bank, and Western Union comment letters. 223See Citizens comment letter. 224See FIA comment letter. 225See NGSA comment letter. 226See NRECA/APPA comment letter. 227See AGA, AFR, COPE, EEI/EPSA, FIA, IATP, ISDA/SIFMA, JBA, NRECA/APPA, and Senators comment letters. 228See AFR, AGA, and FIA comment letters. 229See COPE comment letter. 230See Senators comment letter. 231See Citizens, CDEU, CEWG, CMC, Frost Bank, IIB, NGSA, and Western Union comment letters. 232See CDEU, CEWG, CMC, IIB, and NGSA comment letters. 233See COPE, EEI/EPSA, and IECA comment letters. 234See BDA comment letter. 235See ABA, AGA, BDA, CDEU, CMC, EEI/EPSA, FIA, IECA, IIB, ISDA/SIFMA, NRECA/APPA, and NGSA comment letters. 236See AGA, FIA, and ISDA/SIFMA comment letters. 237See ABA, EEI/EPSA, NRECA/APPA, and NGSA comment letters. 238See BDA, CEWG, CMC, EEI/EPSA, and IECA comment letters. that any de minimis exception-related exemption must be in the public interest, and asked questions regarding the legal authority for the Commission to propose the amendments included in the NPRM.219
3. Commission Response
The Commission has determined not to adopt the MPCE De Minimis Provision at this time. The Commission believes that further action on this provision may require additional consideration of the various relevant issues.220 D. Methodology for Calculating Notional Amounts

1. Proposal
   Given the variety of potential methods that could be used to calculate the notional amount for certain swaps, particularly for swaps where notional amount is not a contractual term of the transaction (e.g., certain NFC swaps), the Commission proposed new paragraph (4)(vii) of the De Minimis Exception, which sets out a mechanism for the Commission, on its own or upon written request by a person, to determine the methodology to be used to calculate the notional amount for any group, category, type, or class of swaps for purposes of whether a person exceeds the AGNA threshold. The proposed rule required that such methodology be economically reasonable and analytically supported, and that any such determination be posted on the CFTC website. Further, to ensure timely clarity to market participants, the Commission proposed to delegate to the Director of DSIO the authority to make such determinations.
2. Summary of Comments
   Several commenters generally supported Commission efforts to provide certainty and clarity regarding calculation of notional amounts.221 Some of these commenters supported providing the Commission with the explicit authority to approve or establish methodologies for calculating notional amount.222 Citizens specifically noted that the lack of clarity regarding notional amount interpretations has persisted for too long, and what little guidance that exists does not provide the certainty that market participants need in order to run their businesses efficiently.223 Further, FIA stated that the DSIO FAQ left open a multitude of questions for market participants attempting to calculate notional amount.224 Additionally, NGSA requested that the CFTC provide a safe harbor for reliance on a notional amount calculation methodology that is based on standard industry practice unless and until CFTC publishes notice that invalidates such a methodology or prescribes a different methodology.225 NRECA/APPA suggested that the Commission should not determine the methodology for calculating notional amounts, stating that the word ‘‘determine’’ in proposed new paragraph (4)(vii) of the De Minimis Exception should be changed to ‘‘provide guidance with respect to.’’ 226 Several commenters did not support the proposal to delegate to the Director of DSIO the authority to make notional calculation determinations.227 Specifically, some commenters stated that the Commission, rather than the Director of DSIO, should determine the methodology for calculating notional amounts because the methodology used to determine the AGNA is a critical component of the de minimis threshold, as it impacts which entities will be designated as SDs.228 Commenters also noted that the delegation, as proposed, would permit Commission staff to make substantive, and potentially critical, policy determinations in an informal process,229 and that Commissioners should not remove themselves from that decision-making process, particularly given that one of the challenges related to NFC swaps was lack of a standard for calculation of notional amount.230 On the other hand, several commenters supported the proposal to delegate to the Director of DSIO the authority to make notional calculation determinations.231 However, many of these commenters supported delegation only if determinations were subject to a public notice and comment process.232 A few commenters noted that if the Commission believes that delegation is proper, it should add safeguards, such as an appeal to the Commission, coupled with a stay of any contested staff determination, pending Commission action.233 One commenter suggested that DSIO should be granted authority to respond to individual dealer requests for guidance on how the notional amount would be calculated for a given transaction, and dealers should be able to rely on any response from DSIO.234 Several commenters stated that notional calculation methodologies should be subject to a
   formal public notice and comment process.235 A few commenters also noted that notional calculation methodologies should be evaluated pursuant to a cost-benefit analysis.236 A few commenters suggested that notional calculations be guided by international standards, industry group comment letters, and the DSIO FAQ Guidance.237 Commenters also provided feedback regarding specific notional amount calculation methodologies.238
3. Commission Response
   The comments raised a number of issues with the proposed authority and delegation regarding the methodology for calculating notional amounts. Given the nature and significance of these issues, the Commission has determined to not adopt this provision at this time.
   IV. Other Matters Discussed in NPRM
   In the NPRM, the Commission did not propose, but sought comment on the following additional potential changes to the De Minimis Exception: (1) Adding a minimum dealing counterparty count threshold and/or a minimum dealing transaction count threshold; (2) establishing as a factor in the de minimis determination whether a given swap was exchange-traded and/or cleared; and (3) establishing as a factor in the de minimis determination whether a given swap is a non￾deliverable forward transaction. The Commission did not propose rule text for any of these topics. At this time, the Commission is not adopting final rules regarding any of these three potential changes. The Commission may take subsequent action

239See generally BDA, IIB, and JBA comment letters. 240See generally Citizens, CEWG, EEI/EPSA, IATP, IECA, ISDA/SIFMA, and NGSA comment letters. 241See generally 360 Trading, ABA, BDA, Daiwa, Cboe SEF, Citizens, CME/ICE, EEI/EPSA, FXPA, Frost Bank, FIA, IIB, IECA, JBA, MFA, Optiver, TR SEF, Virtu, and XTX comment letters. 242See generally AFR and Better Markets comment letters. 243See generally 360 Trading, ABA, AFEX/GPS, AGC, BDA, Capital One, Cboe SEF, Citizens, CDEU, CMC, Covington, FXPA, FIA, IIB, IECA, ISDA/ SIFMA, JBA, Northern Trust, Optiver, Regions, State Street, SVB, TR SEF, Virtu, Western Union, and XTX comment letters. 244See Better Markets comment letter. 245 5 U.S.C. 601 et seq. 246 44 U.S.C. 3501 et seq. 247Parties wishing to review the CFTC’s information collections on a global basis may do so at www.reginfo.gov, at which OMB maintains an inventory aggregating each of the CFTC’s currently approved information collections, as well as the information collections that presently are under review. 248 7 U.S.C. 19(a). or conduct further study with respect to any of these issues. The Commission recognizes the public interest in moving forward with the aspects of the NPRM that it is adopting in this release, rather than delaying action on the NPRM as a whole in order to further consider any of these additional topics. A. Dealing Counterparty Count and Dealing Transaction Count Thresholds The Commission sought comment on whether an entity should be able to qualify for the de minimis exception if its level of swap dealing activity is below any of the following three criteria: (1) An AGNA threshold, (2) a proposed dealing counterparty count threshold, or (3) a proposed dealing transaction count threshold. Although a few commenters expressed general support for adding a dealing counterparty or dealing transaction count threshold to the De Minimis Exception,239 most commenters did not support the idea.240 B. Exception for Exchange-Traded and/ or Cleared Swaps The Commission sought comment on whether an exception from the de minimis calculation for swaps that are executed on an exchange (e.g., a SEF or designated contract market (‘‘DCM’’)) and/or cleared by a derivatives clearing organization is appropriate. Most commenters supported including an exception for exchange-traded and/or cleared trades,241 though two commenters were opposed to the idea.242
C. Exception for Non-Deliverable
Forwards
The Commission sought comment on whether an exception from the de minimis calculation for non-deliverable forwards is appropriate. Most commenters generally supported including an exception for NDFs,243 though one commenter was opposed to the idea.244
V. Related Matters
A. Regulatory Flexibility Act The Regulatory Flexibility Act (‘‘RFA’’) requires that agencies consider whether the regulations they propose will have a significant economic impact on a substantial number of small entities.245 As noted in the Proposal, the regulations adopted herein only affect certain entities that are close to the AGNA threshold in the De Minimis Exception. For example, the regulations would affect entities with a relevant AGNA of swap dealing activity between $3 billion and $8 billion. Moreover, they would affect IDIs that enter into loan￾related swaps. That is, the regulations are relevant to entities that engage in swap dealing activity with a relevant AGNA measured in the billions of dollars. The Commission does not believe that these entities would be small entities for purposes of the RFA. Additionally, the Commission received no comments on the Proposal’s RFA discussion. Therefore, the regulations being adopted herein will not have a significant economic impact on a substantial number of small entities, as defined in the RFA. Accordingly, the Chairman, on behalf of the Commission, hereby certifies pursuant to 5 U.S.C. 605(b) that these regulations will not have a significant economic impact on a substantial number of small entities. B. Paperwork Reduction Act The Paperwork Reduction Act of 1955 (‘‘PRA’’) 246 imposes certain requirements on Federal agencies, including the Commission, in connection with their conducting or sponsoring any collection of information, as defined by the PRA. The Commission may not conduct or sponsor, and a person is not required to respond to, a collection of information unless it displays a currently valid Office of Management and Budget (‘‘OMB’’) control number. As discussed in the Proposal, the final regulations will not impose any new recordkeeping or information collection requirements, or other collections of information that require approval of OMB under the PRA. The Commission notes that all reporting and recordkeeping requirements applicable to SDs result from other rulemakings, for which the CFTC has sought OMB approval, and are outside the scope of rulemakings related to the De Minimis Exception.247
C. Cost-Benefit Considerations
Section 15(a) of the CEA requires the
Commission to consider the costs and benefits of its actions before promulgating a regulation under the CEA or issuing certain orders.248
Section 15(a) further specifies that the
costs and benefits shall be evaluated in light of five broad areas of market and public concern: (1) Protection of market participants and the public; (2) efficiency, competitiveness, and financial integrity of futures markets; (3) price discovery; (4) sound risk management practices; and (5) other public interest considerations. In this
section, the Commission considers the
costs and benefits resulting from its determinations with respect to the
Section 15(a) factors.
In this adopting release, the Commission is amending the De Minimis Exception by setting the AGNA threshold at $8 billion in swap dealing activity. The Proposal requested public comment on the costs and benefits of the proposed regulations, and specifically invited comments on: (1) The costs and benefits to market participants associated with each change; (2) the direct costs associated with SD registration and compliance; (3) the indirect benefits to registering as an SD; (4) the indirect costs to becoming a registered SD; (5) whether entities with dealing activity between $3 billion and $8 billion incur similar registration and compliance costs as compared to entities with dealing activity above $8 billion; (6) the costs and benefits to the public associated with each proposed change; (7) how each proposed change affects each of the Section 15(a) factors; (8) whether the Commission identified all of the relevant categories of costs and benefits in its preliminary consideration of the costs and benefits; and (9) whether the costs and benefits of the proposed changes, as applied in cross￾border contexts, differ from those costs and benefits resulting from their domestic application, and, if so, in what ways and to what extent. As part of this cost-benefit consideration, the Commission will discuss the costs and benefits of the adopted change and analyze the amendment as it relates to each of the

249See also SD Definition Adopting Release, 77 FR 30628–30, 30707–08. To achieve these policy objectives, registered SDs are subject to a broad range of requirements which may carry their own costs and benefits. These requirements include, among other things, registration, internal and external business conduct standards, reporting, recordkeeping, risk management, posting and collecting margin on uncleared swaps, and chief compliance officer designation and responsibilities. However, costs associated with regulatory requirements applicable to SDs result from other rulemakings and are outside the scope of rulemakings related to the De Minimis Exception. 250See id. 251See supra sections I.A.3 and II.B.3; 83 FR 27471–72; 77 FR 30628–30, 30703, 30707. 252See supra sections I.A.3, II.B.1, and II.C.1; 83 FR 27448–58, 27471–72; 77 FR 30628–30, 30703, 30707. 253See supra sections II.B.2 and II.C.2; 83 FR at 27454–56. 254See supra section II.B.1. See also ABA, AGA, AFEX/GPS, BDA, Capital One, Cboe SEF, Citizens, CDEU, COPE, CEWG, CMC, EEI/EPSA, FXPA, Frost Bank, FIA, IIB, IECA, ISDA/SIFMA, JBA, M&T, NCFC, NRECA/APPA, NGSA, Regions, SVB, Virtu, Western Union, and XTX comment letters. 255 83 FR 27451. 15(a) factors. The Commission notes that this consideration of costs and benefits is based on the understanding that the swap market functions internationally, with many transactions involving U.S. firms occurring across different international jurisdictions, with some prospective Commission registrants organized outside the U.S., and other entities operating both within and outside the U.S., and commonly following substantially similar business practices wherever located. Where the Commission does not specifically refer to matters of location, the discussion below of the costs and benefits of the regulations being adopted refers to their effects on all subject swaps activity, whether by virtue of the activity’s physical location in the United States or by virtue of the activity’s connection with or effect on U.S. commerce under CEA section 2(i). As discussed above, the De Minimis Exception provides an exception from the SD Definition for persons who engage in a de minimis amount of swap dealing activity. Currently, a person shall not be deemed to be an SD unless swaps entered into in connection with swap dealing activity exceed an AGNA threshold of $3 billion (measured over the prior 12-month period), subject to a phase-in period that is currently in effect, during which the AGNA threshold is set at $8 billion. The Commission is amending the De Minimis Exception to set the AGNA threshold at the current $8 billion phase-in level. There are market-wide costs and benefits associated with setting the AGNA threshold at $8 billion. In addition, setting the threshold at $8 billion would have specific monetary costs and benefits as compared to a lower or higher threshold. The current $8 billion phase-in level threshold, along with the prospect that the threshold
would decrease to $3 billion after December 31, 2019, in the absence of further Commission action, sets the baseline for the Commission’s consideration of the costs and benefits of the proposed alternatives. Accordingly, the Commission considers the costs and benefits that would result from maintaining the current $8 billion phase-in level threshold, or alternatively, a threshold level below or above the current $8 billion threshold. The status quo baseline also includes other aspects of existing rules related to the De Minimis Exception. The analysis also takes into account any relevant no￾action relief, to the extent such relief is being relied upon. As the Commission is of the belief that existing no-action relief related to the De Minimis Exception is being fully relied upon by market participants, the cost-benefit discussion that follows also considers the effects of that relief.

1. General Costs and Benefits
   There are several policy objectives underlying SD regulation and the de minimis exception to SD registration, which have associated with them general costs and benefits depending on the level of the AGNA threshold. As discussed above in section I.A.3, costs and benefits may be associated with the primary policy objectives of SD regulation, which include reducing systemic risk, increasing counterparty protections, and increasing market efficiency, orderliness, and transparency.249 The Commission also considers the costs and benefits associated with the policy objectives furthered by a de minimis exception, which include increasing efficiency, allowing limited ancillary dealing, encouraging new participants to enter the swap dealing market, and focusing regulatory resources.250 As noted by the Commission and a few commenters, generally, the lower the threshold, the greater the number of entities that are subject to the SD-related regulatory requirements, which could decrease systemic risk, increase counterparty protections, and promote swap market efficiency, orderliness, and transparency.251 However, the Commission and most commenters recognize that a lower threshold could have offsetting costs for the market. For example, it is likely that a lower threshold would discourage new participants from entering into the swap market, and reduce the amount of dealing activity in which swap market participants engage in connection with their other businesses.252 On the other hand, and as discussed further below, the higher the threshold, the greater the number of entities that are able to engage in dealing activity without being required to register, which could increase competition and liquidity in the swap market. However, a higher AGNA threshold could potentially decrease the number of registered SDs, which could have a negative impact on achieving the general benefits associated with the policy objectives of SD regulation. This might adversely affect the swap market to some extent.253 (i) Maintaining the $8 Billion Threshold The comments received for this proposed amendment were generally supportive.254 As discussed in section II.C.1.i, at the $8 billion threshold the 2017 Transaction Coverage and 2017 AGNA Coverage ratios indicate that nearly all swaps were covered by SD regulation, generally giving rise to the benefits of SD regulation discussed above. Almost all swap transactions involved at least one registered SD as a counterparty, approximately 99 percent or greater for IRS, CDS, FX swaps, and equity swaps. For NFC swaps, approximately 86 percent of transactions involved at least one registered SD as a counterparty. Overall, approximately 98 percent of all swap transactions involved at least one registered SD. Further, almost all AGNA of swaps activity included at least one registered SD, approximately 99 percent or greater for IRS, CDS, FX swaps, and equity swaps. Further, the Commission notes that the 6,440 entities that did not enter into any
   transactions with a registered SD had limited activity overall. As discussed in the Proposal, the 6,440 entities entered into 77,333 transactions, representing approximately 1.7 percent of the overall number of transactions during the review period.255 Additionally, collectively, the 6,440 entities had $68 billion in AGNA of swaps activity, representing approximately 0.03 percent of the overall AGNA of swaps activity during the review period. The Commission believes that this limited activity indicates that to the extent these entities are engaging in swap dealing activities, such activity is likely ancillary and in connection with other client services, potentially indicating that the benefits associated with the policy objectives of SD registration and the de minimis

256See supra section II.B.1. See also AGA, BDA, Capital One, CDEU, CMC, Frost Bank, IECA, M&T, SVB, and Western Union comment letters. 257See supra section II.B.1. See also Citizens, IECA, NRECA/APPA, NGSA, and SVB comment letters. 258See supra section II.B.3. 259See supra section II.B.3; Senators comment letter. 260 83 FR 27456. 261 Id. 262See supra section II.C.1.ii; 83 FR 27452–54. 263See 83 FR 27456. Hypothetically, if all 42 entities registered, the percentage of all NFC swaps facing at least one registered SD would rise from approximately 86 percent to 98 percent. 264See 17 CFR 1.3, Swap dealer, paragraph (6)(iii); supra section II.C.1.ii; 83 FR 27456–57. 265See supra sections II.B.1 and II.C.1.ii; 83 FR 27452–54. 266See supra section II.B.1. See also ABA, AGA, AFEX/GPS, BDA, Capital One, Citizens, CDEU, COPE, CEWG, CMC, EEI/EPSA, Frost Bank, IIB, IECA, ISDA/SIFMA, JBA, M&T, NCFC, NRECA/ APPA, NGSA, SVB, Virtu, and Western Union comment letters. 267See Citizens and Virtu comment letters. exception are being advanced at the current $8 billion threshold. Additionally, setting the AGNA at $8 billion would foster efficiency and potentially reduce costs by allowing persons to continue to use existing calculation procedures and business processes that are geared towards the $8 billion threshold. Commenters generally agreed with the Commission’s position. For example, many commenters noted that the current $8 billion threshold already subjects the vast majority of transactions to SD regulation, or that a reduced threshold would not capture significant additional dealing activity.256 Some commenters stated that the nature of the swaps activity entered into by certain entities poses less systemic risk (e.g., commercial banks that have swap dealing activity below $8 billion, and entities that primarily enter into NFC swaps).257 However, as discussed above, Better Markets stated that the high regulatory coverage ratios are not indicative of the absolute level of swap dealing activities relevant to SD registration, and noted that maintaining an $8 billion threshold would have more than a limited effect on counterparty protections.258 The Commission believes that while either percentage of the market or absolute level of swaps activity are valid considerations, it is more relevant in this context of achieving a desirable balance of policy goals to consider the level of activity as a percentage of the whole. Additionally, the Senators stated that though notional amount data for NFC swaps was not used in considering the Proposal, the data that was available for NFC swaps shows significantly less coverage for NFC swaps under an $8 billion threshold than in other asset classes.259 The Commission notes that with respect to NFC swaps, registered SDs still entered into the significant majority (86 percent) of the overall market’s total transactions and, as noted in the Proposal, faced 83 percent of counterparties in at least one transaction, indicating that
the existing $8 billion threshold has helped extend the benefits of SD registration to much of the NFC swap market.260 The trading activity of the 42 unregistered entities with 10 or more NFC swap counterparties represents approximately 13 percent of the overall NFC swap market by transaction count. However, as compared to the existing 44 registered SDs with at least 10 counterparties, these 42 In-Scope Entities have significantly lower mean transaction and counterparty counts, indicating that they may only be providing ancillary dealing services to accommodate commercial end-user clients, also potentially indicating that the benefits associated with the policy objectives of the de minimis exception are being advanced at the current $8 billion threshold.261 The Commission believes these market-wide benefits demonstrate that maintaining an $8 billion threshold is also appropriate with respect to the NFC swap asset class. (ii) $3 Billion Threshold The Commission is of the view that the systemic risk mitigation, counterparty protection, and market efficiency benefits of SD regulation would be enhanced in only a very limited manner if the AGNA threshold decreased from $8 billion to $3 billion, as would be the case if the current regulation and the existing Commission order establishing an end to the phase￾in period on December 31, 2019 were left unchanged. As discussed, Estimated AGNA Coverage would increase from approximately $221,020 billion (99.95 percent) to $221,039 billion (99.96 percent), an increase of $19 billion (a
0.01 percentage point increase);
Estimated Transaction Coverage would increase from 3,795,330 trades (99.77 percent) to 3,797,734 trades (99.83 percent), an increase of 2,404 trades (a
0.06 percentage point increase); and
Estimated Counterparty Coverage would increase from 30,879 counterparties (88.80 percent) to 31,559 counterparties (90.75 percent), an increase of 680 counterparties (a 1.96 percentage point increase).262 The effect of these limited increases is further mitigated by the fact that at the current $8 billion phase-in threshold, the substantial majority of transactions are already covered by SD regulation—and related counterparty protection requirements—because they include at least one registered SD as a counterparty. For NFC swaps, as discussed in the Proposal, without notional-equivalent data, it is unclear how many of the 42 In-Scope Entities with 10 or more counterparties that are not registered SDs would actually be subject to SD registration at a $3 billion threshold.263 It is possible that a portion of the swaps activity for some or all of these entities qualifies for the physical hedging exclusion in paragraph (6)(iii) of the SD Definition, and therefore would not be considered swap dealing activity, regardless of the AGNA threshold level.264 As discussed, a lower AGNA threshold could lead to certain entities reducing or ceasing swaps activity to avoid registration and its related costs.265 Although the magnitude of this effect is unclear, reduced swap dealing activity could lead to increased concentration in the swap dealing market, reduced availability of potential swap counterparties, reduced liquidity, increased volatility, higher fees, wider bid/ask spreads, or reduced competitive pricing. Systemic risk could actually increase as a result. The end-user counterparties of these smaller swap dealing entities may be adversely impacted by the above consequences and could face a reduced ability to use swaps to manage their business risks. Most commenters generally agreed with the Commission’s position. For example, commenters indicated that there would be a market-wide costs associated with a lower threshold given that if entities reduced or ceased swaps activity to avoid registration and its related costs, the small and mid-sized end-users and commercial entities who utilize swaps for hedging purposes and NFC swap market participants would have fewer dealers available to them.266 Two commenters indicated that the market-wide benefit of a lower threshold would be limited because Commission regulations not related to SD registration already apply to unregistered entities, and therefore, many of the policy goals of SD registration are already being advanced with respect to swaps entered into by these unregistered entities.267 IATP suggested that contrary to the assumption that small banks may avoid the swap market due to the costs of SD registration at a $3 billion threshold, the costs and obligations of SD registration would not discourage swap dealing

268See IATP comment letter.
269See supra section II.B.3; Senators comment letter. 270See supra sections II.B.1 and II.C.1.ii. 271See supra section II.B.1; EEI/EPSA and NGSA comment letters. As stated by EEI/EPSA, if NFC prices increase, the same level of swaps activity would potentially have a higher notional amount. 272See NGSA comment letter. 273See OCC, Quarterly Report on Bank Trading and Derivatives Activities (Fourth Quarter 2017), available at https://www.occ.gov/topics/capital￾markets/financial-markets/derivatives/dq318.pdf. 274See supra section II.C.2; 83 FR 27454–56. 275See supra sections II.B.2 and II.C.2; 83 FR 27454–56. 276See supra section II.B.2. As discussed, in comparison to an $8 billion threshold, a $100 billion threshold would reduce the Estimated AGNA Coverage from approximately $221,020 billion (99.95 percent) to $220,877 billion (99.88 percent), a decrease of $143 billion (a 0.06 percentage point decrease). In comparison to an $8 billion threshold, a $100 billion threshold would reduce the Estimated Transaction Coverage from 3,795,330 trades (99.77 percent) to 3,773,440 trades (99.20 percent), a decrease of 21,890 trades (a 0.58 percentage point decrease). The decreases would be more limited at higher thresholds of $20 billion or $50 billion. See supra section II.C.2; 83 FR 27455. 277See supra sections II.B.2 and II.C.2; 83 FR 27455. 278As discussed, the data also indicates that at higher thresholds, there is a more pronounced decrease in Estimated Counterparty Coverage. The Estimated Counterparty Coverage would decrease from 30,879 counterparties (88.80 percent) to 28,234 counterparties (81.19 percent), a decrease of 2,645 counterparties (a 7.61 percentage point decrease). The decrease would be lower at thresholds of $20 billion and $50 billion, at 2.80 percentage points and 5.71 percentage points, respectively. See supra section II.C.2; 83 FR 27455. when there is strong market demand for innovative swap market risk management products. IATP stated that the lack of participation in the swap market by smaller banks may be due to the smaller banks preferring the price transparency of the futures and options markets as compared to the swap market.268 However, as discussed, the Commission believes, and most commenters agree, that a lower threshold could lead to certain entities reducing or ceasing swaps activity. However, the Senators questioned why, given the lack of relevant data for NFC swaps, it is necessary to remove the phase-in reduction of the AGNA threshold for energy-related SDs.269 The Commission believes, and commenters generally agreed, that a reduced threshold would have a cost in terms of a decrease in NFC swap market liquidity because some entities may reduce dealing to avoid registration.270 For example, with respect to NFC swaps, EEI/EPSA and NGSA expressed concern that a lower AGNA threshold would provide less accommodation for increasing NFC prices, which could lead to market participants reducing their swap dealing activity to
remain below the threshold.271 Further, NGSA stated that a lower threshold may reduce ancillary swap dealing in commodity markets and reduce counterparty diversity for end-users.272 The Commission notes that although AGNA data was not available for NFC swaps, the OCC publishes the Quarterly Report on Bank Derivatives Activities, including end-of-quarter gross notional amount position data from call reports filed by insured U.S. commercial banks and savings associations. Although point-in-time position data is not directly comparable to the transaction volume calculations that are required for evaluating AGNA threshold calculations, the report does provide outstanding commodity notional amount position totals in comparison with IRS, CDS, FX swaps, and equity swaps. According to the OCC, as of the end of 2017, NFC swaps represented $1,373 billion out of the $171,964 billion total notional amount reported outstanding, or approximately 0.8 percent of the total.273 Although the number of transactions involving at least one registered SD is lower in the NFC swap market than other asset classes (86 percent compared to over 99 percent for the other four asset classes), the Commission believes it would be inappropriate to lower the AGNA threshold to $3 billion only to potentially increase the registered SD coverage rate (as measured by transaction count) for the smallest of the five asset classes as measured by outstanding notional amount per the OCC Quarterly Report on Bank Derivatives Activities. (iii) Higher Threshold Conversely, a higher AGNA threshold would potentially decrease the number of registered SDs, which could have a negative impact on achieving the general benefits associated with the policy objectives of SD regulation. For example, a higher threshold would allow a greater amount of swap dealing to be undertaken without certain counterparty protections.274 This might impact the integrity of the swap market to some extent. However, the Commission is unable to quantify how the integrity of swap market might be harmed. On the other hand, as noted by the Commission and commenters, the higher the AGNA threshold, the greater the number of entities that are able to engage in dealing activity without being required to register, which could increase competition and liquidity in the swap market.275 A higher threshold could also allow the Commission to expend its resources on entities with larger swap dealing activities that warrant more oversight. Some commenters agreed that the small decrease in Estimated AGNA Coverage and Estimated Transaction Coverage at higher thresholds potentially indicates that increasing the threshold to up to $100 billion may have a limited effect on the systemic risk and market efficiency-related benefits of SD regulation.276 Additionally, a higher threshold could enhance the benefits associated with a de minimis exception, for example by allowing entities to increase ancillary dealing activity.277 However, the decrease in
Estimated Counterparty Coverage indicates that fewer entities would be transacting with registered SDs, reducing the counterparty protection benefits of SD regulation if the threshold increased from $8 billion to $20 billion, $50 billion, or $100 billion.278 The Commission also notes that increasing the threshold could result in changes in market behavior that could lead to the regulatory coverage decreasing more than the analysis indicated. Additionally, though it did not conduct an analysis of AGNA activity for NFC swaps, the Commission is of the view that increasing the AGNA threshold could potentially lead to fewer registered SDs participating in in the NFC swap market, similar to its observations with respect to IRS, CDS, FX swaps, and equity swaps discussed above in section II.C.2. This could reduce the number of entities transacting with registered SDs. The cost of reduced protections for counterparties would be realized to the extent that a higher threshold would result in fewer swaps involving at least one registered SD. Additionally, depending on how the swap market adapts to a higher threshold, it is also possible that the reduction in Estimated Regulatory Coverage would be greater than the data indicates to the extent that a higher threshold leads to an increased amount of swap dealing activity between entities that are not registered SDs. In such a scenario, Estimated Regulatory Coverage could potentially decrease more than the data indicates, increasing the general costs associated with the De Minimis Exception.
2. Direct Cost and Benefits
As discussed in the Proposal, for any AGNA threshold, some firms will have AGNA of swap dealing activity sufficiently close to the threshold so as to require analysis to determine whether their activity qualifies as de minimis. Hence, (1) with a $3 billion threshold,

279Commission staff analyzed the swaps activity of market participants over a one-year period to develop this estimate. The estimate includes 22 In￾Scope Entities that had 10 or more counterparties and between $1 billion and $5 billion in AGNA of swaps activity in IRS, CDS, FX swaps, and equity swaps. Entities that were already registered SDs were excluded. The estimate does not account for entities that primarily are entering into NFC swaps because notional amount information was not available for that asset class. See 83 FR 27474 n.191. 280This estimate is based on the following staff requirements for this determination: 25 hours for an OTC principal trader at $695/hour, 40 hours for a compliance attorney at $335/hour, 35 hours for a chief compliance officer at $556/hour, 80 hours for an operations manager at $290/hour, and 20 hours for a business analyst at $273/hour. These individuals would be responsible for identifying, analyzing, and aggregating the swap dealing activity of a firm and its affiliates. The estimates of the number of personnel hours required have been updated from the SD Definition Adopting Release in light of the Commission’s experience in implementing the SD Definition. The estimates of the hourly costs for these personnel are from SIFMA’s Management & Professional Earnings in the Securities Industry 2013 survey, modified to account for an 1800-hour work-year and multiplied by 5.35 to account for firm size, employee benefits, and overhead, which is the same multiplier that was used when the SD Definition was adopted. See 77 FR 30712 n.1347. The Commission recognizes that particular entities may, based on their circumstances, incur costs substantially greater or less than the estimated averages. See 83 FR 27474 n.192. 281The estimate of 11 entities is approximately 50 percent of the 22 entities that would need to undertake an initial analysis. This estimate assumes that many entities would, following the initial analysis, determine that they would either need to register or choose not to engage in enough dealing activity to require ongoing monitoring. See 83 FR 27474 n.193. 282The Commission estimates that the ongoing analysis would be streamlined as a result of the initial analysis, and therefore would be less costly. For purposes of this calculation, the Commission estimates that the cost of the ongoing analysis would be approximately 50 percent of the cost of the initial analysis. See 83 FR 27474 n.194. 283Commission staff analyzed the swaps activity of market participants over a one-year period to develop this estimate. The estimate includes 29 In￾Scope Entities that had between $3 billion and $15 billion, and 4 In-Scope Entities that had between $15 billion and $25 billion, in AGNA of swaps activity in IRS, CDS, FX swaps, and equity swaps, and at least 10 counterparties. The estimate does not account for entities that primarily are entering into NFC swaps because notional amount information was not available for that
asset class. See 83 FR 27474 n.195. 284See supra note 282. 285See ABA comment letter (attaching NERA study). 286Although addressed by the NERA study, the costs associated with SD regulatory requirements (e.g., margin, reporting, technology, etc.) are not considered in this analysis. Costs associated with regulatory requirements applicable to SDs result from other rulemakings and are outside the scope of rulemakings related to the De Minimis Exception. 287See ABA comment letter (attaching NERA study). To estimate activity, NERA applied a 1.5 assumed turnover ratio to swap position data from the Federal Reserve Bank of Chicago’s ‘‘Holding Company Data’’ for bank holding companies with greater than $10 billion in assets on a consolidated basis. The 1.5 adjustment factor was based on NERA’s estimate of the typical turnover/notional holdings ratio to convert periodic position data into an annualized estimate of AGNA transaction volume. 288NERA estimated median, one-time, upfront SD determination costs of $188,095 per entity, significantly lower than the average cost of $657,696. NERA noted that initial SD determination costs were distributed widely, but the variation did not appear related to institution size or magnitude of annual swaps activity. 289NERA estimated median, ongoing, SD determination costs of $83,430 per entity. 290NERA also calculated a 10 year net present value estimate of the ongoing monitoring costs. NERA estimated the present value of ongoing determination costs to be $723,562 per bank holding company using the average estimate. Additionally, NERA’s analysis included 10 year net present value estimates of business conduct and margin costs, which was outside of the scope of the CFTC’s analysis. some set of entities would likely have to incur the direct costs of analyzing whether they would exceed the threshold, (2) with an $8 billion threshold, a (mostly) different set of entities would have to continue to incur costs of analyzing their activity, and (3) with a higher threshold, some entities would no longer need to conduct an ongoing analysis of whether they would be above the new threshold, while other entities may begin conducting such an analysis. Based on the available data, the Commission estimates that if the AGNA threshold were set at $3 billion, approximately 22 currently unregistered entities would need to conduct an initial analysis of whether they would be above the threshold.279 The Commission estimates that the potential total direct cost of conducting the initial analysis for the 22 entities would average approximately $79,000 per entity, or approximately $1.7 million in the aggregate.280 Certain of those entities with ongoing swap dealing activity that is near a $3 billion threshold may also need to conduct periodic de minimis calculation analyses to assess whether they qualify for the exception. The Commission estimates that approximately 11 entities may need to conduct such analyses.281 Further, the Commission
estimates that the potential annual direct cost of conducting these ongoing analyses for those 11 entities would be approximately $40,000 per entity, or $440,000 in the aggregate.282 The projected 11 entities that may conduct periodic de minimis calculations represents a net figure, as some entities may need to conduct a periodic de minimis calculation, while on the other hand, some entities with AGNA near $8 billion might be able to avoid periodic de minimis calculation costs because they will be certain that their AGNA exceeds the $3 billion threshold. Conversely, the Commission assumes that a higher threshold would permit certain entities to no longer incur ongoing costs of assessing whether they are above the threshold. The Commission estimates the savings that would result from a higher AGNA threshold of $20 billion. Based on the available data, the Commission estimates that if the threshold were set at $20 billion, approximately 29 entities would no longer need to conduct an ongoing analysis of whether they would be above the new threshold, while 4 entities may begin conducting such an analysis.283 The Commission estimates that the ongoing cost savings for the net 25 entities that would no longer be conducting periodic de minimis threshold analyses would average approximately $40,000 per entity, or $1 million in the aggregate per year.284 The Commission notes that ABA submitted a study that evaluated the costs and benefits of SD registration for member banks at various AGNA thresholds, prepared by NERA Economic Consulting (‘‘NERA’’).285 NERA’s study provided cost estimates for initial and ongoing testing of whether a bank holding company has exceeded the AGNA threshold, under various scenarios.286 To arrive at aggregate estimates, NERA estimated the per entity costs of initial and ongoing SD registration determination analyses, and also provided its estimates of the number of registrants at various AGNA thresholds, which Commission staff used to estimate the additional costs or cost savings at different AGNA thresholds, as compared to an $8 billion threshold. First, to estimate initial and ongoing SD registration determination costs, NERA sent a survey to 22 bank holding companies that participate in the swap market and received eight responses.287 Based on these responses, NERA estimated average, one-time, upfront SD determination costs of $657,696 per entity 288 (as compared to the Commission’s estimate of approximately $79,000 per entity on average). Further, NERA estimated average, ongoing, SD determination costs of $89,209 per entity 289 (as compared to the Commission’s estimate of approximately $40,000 per entity on average).290 NERA’s survey of banking entities indicates significantly higher initial and ongoing SD determination monitoring costs than the Commission’s cost estimates on a per entity annualized basis. NERA’s per entity cost estimates were based on the eight responses to their survey, while the Commission’s estimates were
based on: (1) Estimates

291See supra note 280.
292This is based on NERA’s ‘‘Number of Banks Required To Register As Swap Dealer’’ estimates at $3 billion compared to $8 billion under the various scenarios. NERA did not explicitly calculate the number of entities that may yet incur initial determination costs, but instead estimated the number of entities that would be required to register at various thresholds. 293This is based on NERA’s ‘‘Number of Banks Required To Register As Swap Dealer’’ estimates at $15 billion compared to $8 billion under the various scenarios. Note that NERA did not provide estimates at a $20 billion threshold, and its estimates at the $15 billion threshold are the closest for relevant comparison with Commission estimates at $20 billion. 294For Tables 1 through 3, aggregate cost or cost savings estimates are calculated using a given scenario’s per entity average cost estimate multiplied by the relevant entity count. For example, in Table 1, $79,000 multiplied by 22 entities equals $1,738,000. 295As discussed, the Commission considered a higher threshold of $20 billion, while NERA considered a higher threshold of $15 billion. of the number of personnel hours required in light of the Commission’s experience in implementing the SD Definition; and (2) modified costs from SIFMA’s Management & Professional Earnings in the Securities Industry 2013 survey.291 Additionally, NERA’s analysis evaluated bank holding companies on a consolidated basis, while the Commission’s analysis included subsidiaries of banks prior to consolidation and firms unrelated to banks. Second, to estimate the number of entities that would be required to register at different AGNA thresholds, NERA evaluated four different scenarios, including various combinations of an AGNA threshold, a risk-based threshold, and amendments to date restrictions related to the IDI Swap Dealing Exclusion. At various AGNA thresholds—including $3 billion, $8 billion, and $15 billion—NERA estimated the number of bank holding companies expected to register as SDs for each scenario it evaluated. To allow for a more direct comparison with the Commission’s estimates, the Commission made an assumption that the difference in the number of entities required to register at $3 billion and $15 billion thresholds, as compared to an $8 billion threshold, would also be the number of entities that would incur ongoing costs or cost savings related to assessing whether they would be required to register as SDs. Depending on the scenario evaluated, the Commission believes that NERA estimated that 13 to 17 additional bank holding companies would conduct ongoing SD registration-related analyses at the $3 billion threshold as compared to the $8 billion threshold.292 Conversely, depending on the scenario, the Commission believes that NERA estimated that 7 to 10 bank holding companies would no longer incur ongoing monitoring costs at a $15 billion threshold compared to an $8 billion threshold.293 In general, the Commission believes that its per
entity estimated costs reflect the broader nature of the types of entities that would need to conduct such an analysis. For example, NERA’s analysis focused on survey responses from consolidated bank holding companies, whereas the Commission’s estimates also account for smaller financial institutions and non-financial entities that may have less operational complexity and therefore may incur lower costs in making determinations. Additionally, the Commission’s estimates of the number of entities that would incur costs related to SD registration analyses are based on non￾public SDR data on AGNA activity, while NERA’s implied estimates are based on publicly available swap position data from the Federal Reserve Bank of Chicago’s ‘‘Holding Company Data’’ for bank holding companies with greater than $10 billion in assets on a consolidated basis. However, given the different methods and sources of information utilized, the Commission is providing a range of estimated costs or cost savings that combine the per entity costs and the counts of the number of entities required to conduct SD registration analyses, as estimated by the Commission and NERA. The tables below summarize the estimates for initial and ongoing SD determination costs. Since NERA conducted estimates using four different scenarios, the tables below include information based on the highest and lowest number of entities estimated by NERA at given thresholds.
TABLE 1—ESTIMATE OF ADDITIONAL COSTS INCURRED FOR INITIAL SD DETERMINATION ANALYSES
[$3 Billion threshold] 294 Per entity average cost estimate CFTC (22 entities) NERA low estimate (13 entities) NERA high estimate (17 entities) CFTC—$79,000 ......................................................................................................... $1,738,000 $1,027,000 $1,343,000 NERA—$657,696 ...................................................................................................... 14,469,312 8,550,048 11,180,832
TABLE 2—ESTIMATE OF ADDITIONAL COSTS INCURRED FOR ONGOING SD DETERMINATION ANALYSES
[$3 Billion threshold]
Per entity average cost estimate CFTC (11 entities) NERA low estimate (13 entities) NERA high estimate (17 entities) CFTC—$40,000 ......................................................................................................... $440,000 $520,000 $680,000 NERA—89,209 .......................................................................................................... 981,299 1,159,717 1,516,553
TABLE 3—ESTIMATE OF COST SAVINGS FOR NOT CONDUCTING ONGOING SD DETERMINATION ANALYSES
[$15 Billion or $20 billion threshold] 295 Per entity average cost estimate CFTC ($20 billion) (25 entities) NERA low estimate ($15 billion) (7 entities) NERA high estimate ($15 billion) (10 entities) CFTC—$40,000 ......................................................................................................... $1,000,000 $280,000 $400,000

296Using a different methodology, NERA estimated $2,623,925 (median estimate) to $9,174,855 (average estimate) in remaining aggregate initial determination costs. The Commission notes that this estimate is within the $1,027,000 to $14,469,312 range calculated above. 297As discussed in section II.C.1.i, the 2017 Transaction Coverage ratio was approximately 98 percent.
TABLE 3—ESTIMATE OF COST SAVINGS FOR NOT CONDUCTING ONGOING SD DETERMINATION ANALYSES—Continued
[$15 Billion or $20 billion threshold] 295 Per entity average cost estimate CFTC ($20 billion) (25 entities) NERA low estimate ($15 billion) (7 entities) NERA high estimate ($15 billion) (10 entities) NERA—89,209 .......................................................................................................... 2,230,225 624,463 892,090 Based on its analysis, and incorporating information provided by NERA, the Commission estimates that for the 13 to 22 entities at a $3 billion AGNA threshold that may need to conduct an initial SD registration analyses, at per entity average costs of $79,000 to $657,696, the estimated aggregate initial determination cost ranges from $1,027,000 to $14,469,312, as indicated in Table 1.296 Additionally, for the 11 to 17 entities at a $3 billion AGNA threshold that may need to conduct ongoing SD registration analyses, at per entity average costs of $40,000 to $89,209, the estimated aggregate annual ongoing monitoring cost ranges from $440,000 to $1,516,553, as indicated in Table 2. Lastly, for the 7 to 25 entities at a $15 billion or $20 billion AGNA threshold that would no longer need to conduct ongoing SD registration analyses, at per entity average cost savings of $40,000 to $89,209, the estimated aggregate annual ongoing monitoring cost savings ranges from $280,000 to $2,230,225, as indicated in Table 3. The Commission notes that the aggregate estimates of initial and ongoing SD determination and monitoring costs, based on either the Commission or NERA’s per entity cost estimates or marginal entity count estimates, buttress the Commission’s decision to adopt an $8 billion threshold and not let it decrease to $3 billion. Additionally, the Commission is of the view that the cost savings at $15 billion or $20 billion thresholds would not sway its decision to maintain the threshold at $8 billion given the general costs and benefits discussed above. Lastly, in light of all the considerations, the Commission would come to the same conclusion, regardless of where the most accurate cost falls in the range of potential initial and ongoing costs.
3. Section 15(a)
Section 15(a) of the CEA requires the
Commission to consider the effects of its actions in light of the following five factors:
(i) Protection of Market Participants and the Public Providing regulatory protections for swap counterparties who may be less experienced or knowledgeable about the swap products offered by SDs (particularly end-users who use swaps for hedging or investment purposes) is a fundamental benefit advanced by registration of SDs. For example, registered SDs are required to provide mid-mark quotes and perform scenario analyses. However, these requirements are not in standard ISDA agreements and are not required of entities that deal a de minimis amount of swaps. The Commission is maintaining the current de minimis phase-in threshold of $8 billion in AGNA of swap dealing activity. As discussed above, the Commission recognizes that a $3 billion threshold may result in more entities being required to register as SDs compared to the proposed (and currently in-effect) $8 billion threshold, thereby extending counterparty protections to a greater number of market participants. However, this benefit is relatively small because, at the current $8 billion phase-in threshold, the substantial majority of transactions are already covered by SD regulation— and related counterparty protection requirements—since they include at least one registered SD as a counterparty.297 On the other hand, as noted above, a threshold above $8 billion may result in fewer entities being required to register as SDs, thus extending counterparty protections to a fewer number of market participants. Although the Estimated Transaction Coverage and Estimated AGNA Coverage would not decrease much at higher thresholds of up to $100 billion, the decrease in Estimated Counterparty Coverage is more pronounced at higher AGNA thresholds, potentially indicating that the benefit of SD counterparty protections requirements could be reduced at higher thresholds. SD registration is also intended to reduce systemic risk in the swap market. Pursuant to the Dodd-Frank Act, the Commission has proposed or adopted regulations for SDs, including margin and risk management requirements, designed to mitigate the potential systemic risk inherent in the swap market. Therefore, the Commission recognizes that a lower threshold may result in more entities being required to register as SDs, thereby potentially further reducing systemic risk. Conversely, a higher threshold may result in fewer entities being required to register an SD and, thus, possibly increase systemic risk. However, the data appears to indicate that the additional entities that would need to register at the $3 billion threshold are engaged in a comparatively smaller amount of swap dealing activity. Many of these entities might be expected to have fewer counterparties and smaller overall risk exposures as compared to the SDs that engage in swap dealing in excess of the $8 billion level. Accordingly, the Commission believes that that the incremental reduction in systemic risk that may be achieved by registering dealers that engage in dealing between
the $3 billion and $8 billion thresholds is limited. The data also indicates that at higher thresholds of $20 billion, $50 billion, or $100 billion, fewer entities would be required to register as SDs, though the change in regulatory coverage as measured by Estimated AGNA Coverage and Estimated Transaction Coverage would be small. Thus, the Commission believes that the increase in systemic risk that may occur due to a higher threshold would not be significant. However, depending on how the market adapts to a higher threshold, the level of regulatory coverage could potentially decrease more than the data indicates. The Commission believes that setting the AGNA threshold at $8 billion will not substantially diminish the protection of market participants and the public as compared to a $3 billion threshold. Further, as discussed, the Commission does not expect that an increase in the threshold would

298 7 U.S.C. 19(b). substantially increase the protection of market participants and the public. (ii) Efficiency, Competitiveness, and Financial Integrity of Markets Another goal of SD registration is swap market efficiency, orderliness, and transparency. These market benefits are achieved through regulations regarding, for example, recordkeeping, reporting, disclosure, and risk management. As compared to a $3 billion threshold, an $8 billion threshold may have a negative effect on the efficiency and integrity of the markets as fewer entities are required to register as SDs and fewer transactions become subject to SD￾related regulations. However, the Commission also recognizes that the efficiency and competitiveness of the swap market may be negatively impacted if the AGNA threshold is set too low, by potentially increasing barriers to entry that may stifle competition and reduce swap market efficiency. For example, if entities choose to reduce or cease their swap dealing activities in response to the $3 billion threshold, the number or availability of market makers for swaps may be reduced, which could lead to increased costs for potential counterparties and end-users. Conversely, a higher threshold may increase market liquidity, efficiency, and competition as more entities engage in swap dealing without SD registration as a barrier to entry. However, a higher threshold may also result in fewer swaps being subject to SD-related regulations, potentially reducing the financial integrity of markets. Considering these countervailing factors, the Commission believes that setting the AGNA threshold at $8 billion will not significantly diminish the efficiency, competitiveness, and financial integrity of markets as compared to a $3 billion threshold. Further, as discussed, an increase in the threshold would potentially have both positive and negative effects to the efficiency, competitiveness, and financial integrity of the markets. (iii) Price Discovery All else being equal, the Commission believes that price discovery will not be harmed and might be improved if there are more entities engaging in ancillary dealing due to increased competitiveness among swap counterparties. The Commission is of the view that, as compared to a $3 billion threshold, an $8 billion threshold would encourage participation of new swap dealing businesses and promote ancillary dealing because those entities engaged in swap dealing activities below the threshold would not need to incur the direct costs of registration until they exceeded a higher threshold. Similarly, raising the threshold above $8 billion could lead to even more entities engaging in ancillary dealing. The Commission notes that some counterparties might be more likely to transact at off-market prices if they trade with an entity that does not provide mid-market quotes or scenario analyses, as would be required if the entity were a registered SD. If so, such transactions might harm post-trade price discovery
since these transactions would occur at off-market prices. (iv) Sound Risk Management The Commission notes that a higher AGNA threshold could lead to impaired risk management practices because a lower number of entities would be required by regulation to: (1) Develop and implement detailed risk management programs; (2) adhere to business conduct standards that reduce operational and other risks; and (3) satisfy margin requirements for uncleared swaps. For the same reason, a lower threshold could positively impact risk management since more entities would be required to comply with the above mentioned risk-related SD regulations. The Commission also notes that to the extent an entity that is not required to register as an SD at a higher threshold is a prudentially regulated bank, that entity would be subject to the risk management requirements of its prudential regulator. (v) Other Public Interest Considerations The Commission has not identified any other public interest considerations with respect to setting the AGNA threshold at $8 billion in swap dealing activity. D. Antitrust Considerations
Section 15(b) of the CEA requires the
Commission to take into consideration the public interest to be protected by the antitrust laws and endeavor to take the least anticompetitive means of achieving the purposes of the CEA, in issuing any order or adopting any Commission rule or regulation (including any exemption under section 4(c) or 4c(b)), or in requiring or approving any bylaw, rule, or regulation of a contract market or registered futures association established pursuant to
section 17 of the CEA.298 The
Commission believes that the public interest to be protected by the antitrust laws is generally to protect competition. The Commission has considered this final rule to determine whether it is anti-competitive and has identified no anti-competitive effects. Because the Commission has determined that the final rulemaking is not anti-competitive and has no anti-competitive effects, the Commission has not identified any less anti-competitive means of achieving the purposes of the CEA. List of Subjects in 17 CFR Part 1 Commodity futures, Definitions, De minimis exception, Insured depository institutions, Swaps, Swap dealers. For the reasons stated in the preamble, the Commodity Futures Trading Commission amends 17 CFR
part 1 as follows:
PART 1—GENERAL REGULATIONS
UNDER THE COMMODITY EXCHANGE ACT

• 1. The authority citation for part 1
     continues to read as follows:
     Authority: 7 U.S.C. 1a, 2, 5, 6, 6a, 6b, 6c, 6d, 6e, 6f, 6g, 6h, 6i, 6k, 6l, 6m, 6n, 6o, 6p, 6r, 6s, 7, 7a–1, 7a–2, 7b, 7b–3, 8, 9, 10a, 12, 12a, 12c, 13a, 13a–1, 16, 16a, 19, 21, 23, and 24 (2012).
• 2. In § 1.3, amend the definition of the
     term ‘‘Swap dealer’’ by revising paragraph (4)(i)(A) and removing and reserving paragraph (4)(ii). The revision reads as follows:
     § 1.3 Definitions.

---

Swap Dealer. * * *
(4) De minimis exception—(i)(A) In general. Except as provided in paragraph (4)(vi) of this definition, a person that is not currently registered as a swap dealer shall be deemed not to be a swap dealer as a result of its swap dealing activity involving counterparties, so long as the swaps connected with those dealing activities into which the person—or any other entity controlling, controlled by or under common control with the person—enters over the course of the immediately preceding 12 months have an aggregate gross notional amount of no more than $8 billion, and an aggregate gross notional amount of no more than $25 million with regard to swaps in which the counterparty is a ‘‘special entity’’ (as that term is defined in section 4s(h)(2)(C) of the Act, 7 U.S.C. 6s(h)(2)(C), and § 23.401(c) of this
chapter), except as provided in
paragraph (4)(i)(B) of this definition. For purposes of this definition, if the stated notional amount of a swap is leveraged or enhanced by the structure of the

1See Swap Dealer De Minimis Exception Preliminary Report (‘‘Preliminary Report’’), http:// www.cftc.gov/idc/groups/public/@swaps/ documents/file/dfreport_sddeminis_1115.pdf. 2See Swap Dealer De Minimis Exception Final Report (‘‘Final Report’’), https://www.cftc.gov/sites/ default/files/idc/groups/public/@swaps/documents/ file/dfreport_sddeminis081516.pdf. 3See Office of the Comptroller of the Currency, ‘‘Quarterly Report on Bank Trading and Derivatives Activities, Second Quarter 2018,’’ available at:
https://www.occ.gov/topics/capital-markets/ financial-markets/derivatives/dq218.pdf. 4For further discussion, see comment letter to CFTC from Financial Services Roundtable dated January 19, 2016 (‘‘We do not see a benefit to requiring an entity that enters into a small number of swaps with a large notional amount but little exposure to choose between exiting the market or registering as a swap dealer, nor should entities that are taking on very large exposures without crossing a notional threshold, or a trade or counterparty count metric, be unregulated because they have concentrated risk in a small number of trades.’’). swap, the calculation shall be based on the effective notional amount of the swap rather than on the stated notional amount.

---

Issued in Washington, DC, on November 6, 2018, by the Commission. Robert Sidman, Deputy Secretary of the Commission. Note: The following appendicies will not appear in the Code of Federal Regulations. Appendicies to De Minimis Exception to the Swap Dealer Definition— Commission Voting Summary, Chairman’s Statement, and Commissioners’ Statements
Appendix 1—Commission Voting
Summary
On this matter, Chairman Giancarlo, and Commissioners Quintenz, Behnam, Stump, and Berkovitz voted in the affirmative. No Commissioner voted in the negative.
Appendix 2—Statement of Chairman J.
Christopher Giancarlo
Today’s final rule on the numeric threshold for swap dealer de minimis will provide the market with certainty that the threshold will not fall from $8 billion to $3 billion. I fully support the proposed final rule. The action before us is without prejudice to all other items in the Commission’s June 2018 NPRM. That includes various proposed rule amendments and other topics for consideration. Those proposals and considerations are clearly of wide ranging interest as evidenced by the public comments received. They remain under staff consideration pending further Commission action. Indeed, I will direct CFTC staff to continue their analysis of the range of matters raised in the June 2018 NPRM and comments submitted by the public. I will specifically ask staff to conduct a study on possible alternative metrics for the calculation of the swap dealer de minimis threshold drawing upon proposals in the June 2018 NPRM, including the feasibility of:
(i) Removing cleared swaps from the current de minimis calculation; (ii) haircutting cleared swaps included in the current de minimis calculation; (iii) adopting a new, bifurcated de minimis calculation that uses initial margin amounts for cleared swaps and entity-netted notional amounts for uncleared swaps; and (iv) applying other risk-based approaches that the staff may recommend. I will be asking the staff for specific deadlines and deliverables for this work. Once staff has reviewed and analyzed the data, I expect that the study will be made public for further discussion and possible Commission consideration. I deliberately decline at this time to express any view on the appropriateness of whether any of the proposals in the June 2018 NPRM not before us today should be addressed by CFTC unilateral rulemaking or joint consideration with the U.S. Securities and Exchange Commission (SEC). Be assured that SEC Chairman Clayton and I—and our fellow CFTC and SEC Commissioners—are committed to working together on robust harmonization where appropriate and working jointly where necessary on these and other matters. With respect to IDIs, staff has informed me that they would consider no-action relief for IDIs pending formal Commission action should they receive a meritorious request. In sum, I am hopeful that we will today provide market certainty that the de minimis threshold will not fall below its current level. Surely, it has taken a while to reach this point. Yet, I am hopeful that we may achieve it with a good degree of consensus across the full Commission. Assuming so, then we have increased market certainty—a very good thing in trading markets. Sometimes it’s worth the wait.
Appendix 3—Statement of
Commissioner Brian D. Quintenz I support today’s final rule to rescind the de minimis threshold’s scheduled reduction to $3 billion of gross notional swap dealing activity. Every iteration of data analysis completed by CFTC staff on this issue, from the 2015 Preliminary Report,1 to the 2016 Final Report,2 to the updated data and analysis in the 2018 June proposed rule, and to the data presented in this final rule, clearly and unequivocally supported eliminating this ill-conceived reduction. I am pleased that today’s action will remove a large source of negative regulatory uncertainty for market participants in managing their swaps business and serving their customers. However, this is just the first of many necessary steps toward correcting what I believe is a flawed swap dealer registration policy. Therefore, it is my hope that today’s final rule should be viewed with finality only in this one regard. The Dodd-Frank Act advanced three main and substantial policy objectives for swap dealer registration: Systemic risk reduction, counterparty protection, and enhanced swap market transparency and efficiency. As I have emphasized on many prior occasions, given the significant costs of swap dealer regulation, it is critical that the de minimis exception be appropriately calibrated to ensure that the correct market group—those best situated to realize the corresponding policy goals of registration—shoulders the burdens of swap dealer regulations. As I have also said repeatedly in the past, notional value is a poor measure of activity, and it is a meaningless measure of risk. Therefore, by itself, notional value is an incredibly deficient metric by which to impose large costs and achieve substantial policy objectives. A one-size-fits-all notional value test for swap dealer registration captures entities that engage in low volume, low risk activity with high notional amounts, and places those firms under the same regulatory regime as the world’s largest, most complex financial institutions that deal in trillions of dollars’ worth of swaps.3 The end result is that smaller firms are disincentivized from engaging in lower risk activity when faced with justifying the cost of swap dealer registration. I have heard anecdotally from certain small to mid-sized players in the swap markets that the breakeven point of the costs of swap dealer registration as measured by a level of notional swap dealing activity is much higher than the $8 billion level in this rule. If that is the case, the current $8 billion notional threshold effectively forces these smaller players to curtail their swap dealing business, thereby limiting competition and further concentrating swaps activity with their larger competitors.4 In my view, an appropriately calibrated de minimis exception would better align the criteria of the de minimisthreshold with the costs of swap dealer regulation, particularly the largest costs tied to mitigating systemic risk, like capital and margin. A de
minimis threshold based on metrics more closely correlated with the risk of the products traded, as opposed to the current risk￾insensitive notional value metric, would better measure dealing activity and more appropriately capture the entities warranting Commission oversight. I am pleased the Chairman continues to recognize this and has directed staff to study many of the alternative risk-based registration metrics that were suggested in the proposed rule. The staff report will provide the Commission with additional data and insights into the impact that alternative approaches may have on swap dealer registration. For example, staff’s analysis should show how removing or haircutting cleared swaps from the de minimis calculation would impact the number and composition of firms required to register as swap dealers. The report will also provide staff with an opportunity to consider, for the first time, how a registration threshold tied to initial margin for cleared swaps could better represent a de minimis quantity of swap dealing activity. For uncleared products, staff can examine the impact of using entity-netted notional amounts, a more accurate measure of a firm’s risk and market size, as a metric of swap dealing activity. The results of the staff report will be critical to any future Commission consideration of a

5CFTC No-Action Letter 18–20 (August 28, 2018), https://www.cftc.gov/PressRoom/PressReleases/ 7775-18. 1 De Minimis Exception to the Swap Dealer Definition, 83 FR 27444, 27481–2 (proposed June 12, 2018). 2 If the proposed IDI Minimis Provision truly better aligns the swap dealer regulatory framework with the risk mitigation demands of bank customers, as commenters suggested, then it would seem that there should be few hurdles in the way of the CFTC and SEC engaging to reconsider the parameters of the IDI Swap Dealing Exclusion. 3 83 FR 27445–6. 4 83 FR 27445. 5 83 FR 27450, 27456–7. 6 83 FR 27457; Final Rule, De Minimis Exception to the Swap Dealer Definition, section II.C.1.ii (to be codified at 17 CFR pt. 1). 1 Dodd-Frank Wall Street Reform and Consumer Protection Act, section 712(d)(1), Public Law 111– 203, 124 Stat. 1376 (2010) (the ‘‘Dodd-Frank Act’’). 2See 17 CFR 1.3, Swap dealer, paragraph (4)(i)(A); see also Further Definition of ‘‘Swap Dealer,’’ ‘‘Security-Based Swap Dealer,’’ ‘‘Major Swap Participant,’’ ‘‘Major Security-Based Swap Participant’’ and ‘‘Eligible Contract Participant,’’ 77 FR 30596, 30633–34 (May 23, 2012) (‘‘SD Adopting Release’’). 3 17 CFR 1.3, Swap dealer, paragraph (4)(ii)(B). more risk-sensitive swap dealer registration threshold. In addition, many of the policy recommendations discussed in the proposed rule, such as better allowing insured depository institutions to assist their customers in hedging loan-related risks and excluding non-deliverable forwards from an entity’s de minimis count—would advance the policy goals of the de minimis exception by encouraging greater participation and competition in the swap markets. I would eagerly anticipate the Commission’s action on these important reforms. As the Commission’s recent no-action letter to a Main Street bank this past August shows, the deficiencies of the current de minimis exception are beginning to squeeze firms’ activity and constrain their ability to serve clients.5 Any de minimis threshold must always be put into context of the broader swaps market regulatory regime. The Commission is not establishing the de minimis exception in a vacuum. Since the swap dealer definition was adopted in 2012, a broad range of rigorous regulatory requirements have gone into effect which also advance the goals of swap dealer registration, such as mandatory clearing, SEF trading, swap data reporting, and margin requirements for uncleared swaps. The Commission’s regulatory framework for the swap market has greatly evolved from its state six years ago; it is only common sense that the swap dealer registration threshold should evolve as well. It will be a great day when financial regulators, including the CFTC, finally move away from gross notional value as any sort of metric or test of derivatives exposure, activity, or risk. I look forward to that day, and I am committed to working with the Chairman, my fellow Commissioners, and our staff to make sure we get the swap dealer de
minimis exception policy right.
Appendix 4—Concurring Statement of
Commissioner Rostin Behnam Today, the Commission acts decisively to set the aggregate gross notional amount (‘‘AGNA’’) threshold for the de minimis exception at $8 billion in swap dealing activity entered into by a person over the preceding 12 months. I am comfortable supporting today’s final rule because it is limited to establishing a clear and certain de minimis threshold. While I was unable to support the proposed rule—which moved the Commission far beyond the task before it towards unilaterally redefining swap dealing activity absent meaningful, congressionally￾required collaboration with the Securities and Exchange Commission (‘‘SEC’’)—I am gratified that the Commission is not moving forward with aspects of the Proposal which would have further complicated the distinction between dealing and non-dealing activities.1 Such action would have been detrimental to market participants. To the extent the Commission continues to consider addressing long standing concerns with the IDI Swap Dealing Exclusion,2 ambiguity regarding the treatment of swaps used for hedging, or relief applicable to swaps that result from multilateral portfolio compression exercises, it should do so jointly with the SEC. NFC Swap Data Today’s decision to maintain the AGNA threshold at $8 billion follows a period of prolonged uncertainty during which Commission staff conducted more complete data analysis regarding the de minimis exception.3 While swap data repository (‘‘SDR’’) data quality has improved, AGNA data was unavailable for non-financial commodity (‘‘NFC’’) swaps.4 Nevertheless, Commission staff used counterparty and transaction counts and a series of assumptions to analyze likely swap dealing activity in the NFC swap market and concluded that reducing the $8 billion AGNA threshold could lead to reduced liquidity in NFC swaps, negatively impacting end-users and commercial entities who utilize NFC swaps for hedging.5 The Commission further relied upon findings and comments that the unique characteristics of the NFC swap market pose less systemic risk than financial swaps.6 It is my hope that Commission staff will continue to examine and monitor data and activities in the NFC swap market to ensure that concentrated activity by unregistered NFC counterparties in segments of that swap market, such as in energy-related swaps, do not present outsized risk or harm to end￾users, and most importantly, the general public.
Appendix 5—Statement of
Commissioner Dan M. Berkovitz I support amending the swap dealer de minimis exception to set the threshold at $8 billion. This limited amendment relies on extensive data analysis to achieve a balance between the policy objectives of the de minimis exception and the registration of swap dealers. At the outset, I would like to acknowledge the leadership of Chairman Giancarlo and the efforts of my fellow Commissioners to achieve consensus on this rulemaking. I look forward to working together to continue to find areas of agreement where it makes sense for our markets and the American people. Data-Driven Rulemaking Title VII of the Dodd-Frank Act directed the Commodity Futures Trading Commission (‘‘Commission’’) and the U.S. Securities and Exchange Commission (‘‘SEC’’) to jointly further define, among other things, the term ‘‘swap dealer.’’ 1 At the same time, Congress enacted Section 1a(49)(D) of the Commodity Exchange Act (‘‘CEA’’), which directed the Commission to exempt from designation as a swap dealer entities that engage in a de minimis quantity of swap dealing. In 2012, the Commission—jointly with the SEC—adopted the further definition of the term swap dealer. In this rulemaking, the de minimis swap dealing threshold was set at $3 billion. However, recognizing that a lack of swap trading data made it difficult to set an appropriate threshold, the Commission implemented a long phase-in period during which the threshold was set at $8 billion.2 The regulation directed Commission staff to study the data on swap dealing activity that would be collected through swap data repositories (‘‘SDRs’’) and publish a report for public comment, enabling the Commission at a later time to make a data￾based judgment regarding the de minimis quantity threshold.3 To this end, the staff built a comprehensive database to aggregate data from all four SDRs. Over several years, the staff developed and refined new techniques to sort and evaluate the data, published two reports on the de minimis exception, and continued to revise its analysis in response to public comments. This process was not without considerable challenges, but the staff worked diligently to produce meaningful, data-driven information to guide the Commission’s decision-making regarding the appropriate de minimis threshold. This effort provided a highly significant data point: Approximately 98 percent of all swap transactions involved at least one registered swap dealer. We now know that at the $8 billion threshold, nearly all swap transactions benefit from swap dealer regulation. The staff’s analysis also showed that reducing the threshold to $3 billion would have a minimal impact on the amount of swaps activity that would be subject to swap dealer regulation. Indeed, based on the analysis, reducing the threshold to $3 billion would only add swap dealer coverage to less than one-tenth of one percent of reported swaps. By the same token, the analysis demonstrated that increasing the threshold quantity
above $8 billion would have almost no impact on the amount of swaps subject to dealer regulation until that threshold reaches a significantly higher level. At those levels, the effect on specific categories of swaps— notably non-financial commodity swaps (‘‘NFC’’)—becomes much more significant. When considering amending a rule, the Commission should consider both the

4Notice of proposed rulemaking, De Minimis Exception to the Swap Dealer Definition, 83 FR 27444 (June 12, 2018). 5 7 U.S.C. 5(b). 6 Dodd-Frank Act, section 712(d)(1). 7 Dodd-Frank Act, section 712(d)(2)(A). 8 Dodd-Frank Act, section 712(d)(2)(D). 9 7 U.S.C. 1a(49)(D) (emphasis added). 10 Id. 11 Id. (emphasis added). 12 In the preamble of the SD Adopting Release, the Commission discussed the factors envisioned by benefits and costs from those rule changes. Here, data analysis has shown that the benefits of changing the current $8 billion threshold are relatively small because nearly all swap activity is already covered by dealer regulation. On the other hand, decreasing the threshold from its current level would impose tangible costs on market participants. If the threshold were lowered to $3 billion, unregistered dealers that are currently under the $8 billion level, but that could exceed the $3 billion threshold, would have to re￾evaluate whether swap dealing in excess of $3 billion would continue to make business sense. The de minimis rulemaking proposal 4 noted that this issue is particularly important in the NFC swap market. The staff’s data analysis showed that many of the smaller swap dealers for physical commodities are physical commodity producers, distributors, consumers, or merchandizers. Swap dealing is an ancillary business for them. Where the costs of registering as a swap dealer exceed anticipated benefits, it is likely that many of these entities would withdraw from providing swap dealing services to their customers. That would leave many end users looking to hedge their risks with either no dealers available, or very few dealers to provide competitive pricing. The Commission should seek to preserve and foster competition for swap dealer services. One of the fundamental purposes of the CEA is to ‘‘promote . . . fair competition among boards of trade, other markets and market participants.’’ 5 American businesses throughout the country that need to use swaps to hedge their risks should not be forced to rely solely on large Wall Street banks. Retaining the de minimis threshold at $8 billion will help preserve competition and choice for American businesses for these swap dealing services. It is important to note that this rulemaking represents one of the first times in which the Commission has relied on SDR data to set policy, and the staff that undertook this principled and thorough analysis should be commended for their efforts. Given the technological advancements in data collection and analysis, effective use of data to inform policy making is critical for the Commission to meet its policy objectives of fostering open, transparent, competitive, and financially sound markets. In sum, the data demonstrates that the current de minimis threshold level is largely accomplishing its intended purposes. Where the current regulations are working, regulatory stability also is an important objective. Accordingly, after considering the results
of the swap data analysis, relevant policy implications, and limited benefits and potential costs of altering the de minimis threshold quantity, I believe that maintaining the threshold at $8 billion is appropriate and sound public policy. Physical Commodity Swaps The proposal noted that Commission staff encountered challenges in measuring the aggregate gross notional amount of NFC swaps. Instead, the staff used counterparty and transaction counts to approximate swap dealing activity for NFC swaps. The staff’s analysis indicated that fewer NFC swap transactions—86 percent—involved at least one registered swap dealer, as opposed to 99 percent for other swap categories. The market participants who use physical commodity swaps to hedge their risks typically include farmers, ranchers, farm product processors, energy producers and consumers, manufacturers, and other end users. These consumer-facing businesses need a properly functioning physical commodity derivatives marketplace to maintain consistent prices for their customers. Ultimately, the American people benefit from stable prices on the products that these businesses produce and distribute. I am therefore calling on the Commission to continue to focus on improving our data collection and analysis for NFC swaps. More robust data collection will help us improve regulation in this space, including considering ways to balance the benefits of de minimis swap dealing in physical commodities with the need for customer protections and the other benefits of swap dealer registration. Joint Rulemaking Required for Swap Dealer Definition I am voting today solely in favor of setting the de minimis exception threshold quantity at $8 billion because it is within the Commission’s authority to do so. Looking forward, however, I will not support other amendments to the swap dealer definition without a joint rulemaking with the SEC, as required by the Dodd-Frank Act. In addition to setting the threshold level, the proposal sought to alter the swap dealer definition by excluding from counting toward that de minimis threshold: (1) Swaps entered into by an insured depository institution (‘‘IDI’’) in connection with originating loans; (2) swaps hedging financial or physical positions; and (3) swaps resulting from multilateral portfolio compression exercises. The proposal also asked questions about excluding from the threshold calculation swaps that are cleared and/or exchange traded and non-deliverable forwards. Although the Commission is not adopting these provisions today, my view is that any such changes would effectively amount to an amendment of the swap dealer definition, not the de minimis exception. Doing so unilaterally and not as a joint rulemaking with the SEC would be contrary to the statutory language and inconsistent with Congressional intent. When Congress enacted Title VII of the Dodd-Frank Act, its intent was clear: ‘‘[T]he [Commission] and the [SEC], in consultation with the Board of Governors,
shall further define the term[] . . . ‘swap dealer,’ ’’ among other terms.6 Congress clarified that the Commission must use the joint rulemaking process to make any other rules regarding these definitions that it and the SEC determine are necessary for the protection of investors.7 To underscore this point, Congress noted that rules prescribed jointly by the Commission and the SEC under Title VII must be ‘‘comparable to the maximum extent possible,’’ and that any interpretation of, or guidance regarding, a provision of the Dodd-Frank Act would be effective only if issued jointly by the Commission and the SEC.8 Pursuant to this statutory directive, the agencies adopted a joint rulemaking to define ‘‘swap dealer’’ and ‘‘security-based swap dealer.’’ Congress created one exception to the joint rulemaking requirement. CEA subsection 1a(49)(D) authorizes ‘‘the Commission’’ to exempt from designation as a swap dealer ‘‘an entity that engages in a de minimis quantity of swap dealing’’ and ‘‘to establish factors with respect to the making of this determination to exempt.’’ 9 The Commission included this de minimis exception in paragraph 4 of the swap dealer definition, notably separate from other provisions in the definition addressing the IDI exclusion (paragraph 5) and the physical hedging exclusion (paragraph 6). By its terms, the de minimis exception relates solely to exempting a numerical quantity of swap dealing activity. Under the statutory structure, the Commission and the SEC must jointly determine which activities are dealing activities and therefore must be counted toward the threshold; the Commission itself may set a numerical quantity of such dealing as a threshold for registration. Put simply, deciding ‘‘which’’ activity gets counted must be done jointly; deciding ‘‘how much’’ of that activity triggers the registration requirement may be done singly. The proposal framed these additional proposed changes to the swap dealer definition as ‘‘factors’’ in the de minimis threshold determination. In doing so, the proposal sought to use the Commission’s unilateral authority to ‘‘establish factors’’ as provided in the second sentence in CEA subsection 1a(49)(D). However, that interpretation is a misreading of the statutory provision. The second sentence in CEA subsection 1a(49)(D) authorizes the Commission to promulgate regulations to ‘‘establish factors with respect to the making of this determination to exempt.’’ 10 The words ‘‘this determination’’ clearly refer to the quantity determination in the first sentence of the subsection: ‘‘[t]he Commission shall exempt from designation as a swap dealer an entity that engages in a de minimis quantity of swap dealing in connection with transactions with or on behalf of its customers.’’ 11 In other words, the ‘‘factors’’ referred to in the second sentence relate to the numerical quantity determination in the first sentence; this sentence does not create a distinct directive authorizing the Commission to
independently determine what constitutes swap dealing.12

Section 1a(49)(D). For example, the preamble
provided that the Commission could consider whether the de minimis exception would ‘‘lead[] to an undue amount of dealing activity to fall outside the ambit of Title VII regulatory framework, or lead[] to inappropriate reductions in counterparty protections (including protections for special entities).’’ SD Adopting Release, 77 FR 30635. This point is clear when we examine what would happen if each of the five categories of swap dealing activity identified in the proposal as ‘‘factors’’ (i.e., IDI, physical hedging, multilateral portfolio compression exercises, cleared and/or exchange traded, and non-deliverable forwards) were removed from the definition of swap dealing through this interpretation of the de minimis exception. Combined, these five categories of swaps likely total more than half of the notional amount traded. There would appear to be no limit to what dealing activity could be excluded from dealer regulation through the de minimis exception by framing whole categories of swaps to be excluded as ‘‘factors.’’ The Commission could effectively determine unilaterally what constitutes swap dealing. The de minimis exception would swallow the swap dealer definition. This result cannot be reconciled with the Dodd￾Frank Act’s joint rulemaking requirement. For these reasons, while I am amenable to considering further refinements to the swap dealer definition and what gets counted as dealing, I am of the view that this cannot be accomplished without joint rulemaking with the SEC. [FR Doc. 2018–24579 Filed 11–9–18; 8:45 am] BILLING CODE 6351–01–P