Broadcast Music, Inc. v. North American Concert Promoters Association - Antitrust

23-935 (L) Broadcas t Music, Inc. v. North A merican Conce rt Pro moters Assoc iation In the United States Court of Appe als FOR THE SECOND CIRCUIT A UGUST T ERM 2023 No s. 23-935(L), 23-10 04(XA P) B RO AD C AST M U SIC, I NC., Petitioner-Appellee-Cr oss-Appellant, v. N ORTH A M ERIC A N C ONCERT P RO MOT ER S A SSOCIATION, Respondent-Appellant- Cross-Appellee. * On Appeal from the United S tates District C ourt for the Sou thern District of New York A RGUED: M AY 20, 2024 D ECIDED: F EBRUARY 24, 2026 Before: W ESLEY and M ENAS HI, Circuit Judges. † * The C lerk of C our t is di rected to a mend the capti on as s et forth a bove. † Judge Beth Robi nson, originally a member of the panel, recused and took no part in the resolution of this appeal. The t wo r emaining me mber s of the panel, who are in ag reement, have de termined the m atter. See 28 U.S.C.

2 Broadcast Music, I nc. (“BMI”) license s mus ical works in exchange for a fee. Be cause of BMI’s large s hare of the music l icensing market, i t is subjec t to an antitrus t cons ent d ecree that requires BMI to set reasonable licensing fees and authorizes a court to determine a reasonable fee if BMI and the prospective licensee are unable to agree. North American Co ncert Promote rs Associat ion (“NACPA ”) has historically purchased blanket licenses c ov ering al l the musical works in BMI’s reperto ry so that art ists ma y p erf orm the works at concerts. During the last nego tiation cycle, BMI an d NACPA were una ble to agree on a rate and revenue base, and for the first ti me in the h istory of the parties’ re lation ship, BMI peti tioned the district co urt t o resol ve the impasse. In 2023, the d istric t court accepted BM I’s rate q uote for the 2014 -20 18 period and set a rate of 0.5 pe rcent of NAC PA’s gross revenues for the 2018 -20 22 pe riod. The dis tric t court also expanded the definition of “g ross revenues ” to cove r items not pr eviously included in the revenue base. NACPA appealed b oth de cisions, and BMI cross - appealed t he denial of its motion for prejudgment interest. We conclude that the distr ict court imposed unr easonable rate s. First, t he distr ict cou rt adopted a definiti on of the revenue base that had no precedent in the history of the industry without a compelling reason. The definition included revenue stream s that do not reflect the fair market value of the musi c and that inv olve sign ific an t administrative cos ts withou t a corres ponding benefit. Second, t he distr ict court set a rate that depends disproport ionately on less comparable benchmark agreement s. The distri ct court relied on agreements with u naffiliated indiv idual promoters e ven though NACPA has histor ically obta ined sign ificantly lower r ates from the § 46(d); 2d Cir. IOP E(b); United States v. Desim one, 140 F.3d 457, 458 - 59 (2d Cir. 1998).

3 same counterparties. And it identified no change in economic circumstances that would j ustify a r ate more than doubl e what NACPA has h istorically pa id to BM I and to the Amer ican Soc iety of Composers, Author s, and Publishe rs. Third, w hile we conclude that the dis tr ict c ourt did not a buse its dis cre tion by de ny ing t he motio n for prejudgment inte rest, the denial may have depe nded on the erroneous rate s. Accordingly, w e vacate the judg ment of th e d istric t court and remand for fu rther proceeding s consistent with t his opinion. S COTT A. E DELMA N (Atara Miller, Andrew L. Porter, on the brief), Milbank LLP, New York, New York, for Petitioner-Appellee-Cr oss-Appellant. A NDREW G ASS, Latham & W atkins L LP, S an Francisco, California (Jose ph R. Wetzel, Latha m & Watkins LL P, San Francisco, Califo rnia; Jennifer L. Gior dano, Sarang V. Damle, Blake E. Staffor d, Latham & Watkins LLP, Washington, D.C.; Benjam in E. Marks, S arah Sternlieb, Davi d Yolkut, Weil, Gotshal & Manges LLP, New York, New York; Sami r Deger - Sen, Latham & Watkins LLP, New York, Ne w York, on the brief), for Respondent - Appellant-Cross-Appel lee. Fran k P. Sc ibi lia, Donald S. Zakarin, Katie E. Garber, Pryor Cashman LLP, New York, Ne w Yor k, for Amicus Curiae National Music Publishers’ Ass ociation. Jay Cohen, Hal lie S. Go ldb latt, Pa ul, We iss, R ifk ind, Wharton & Garrison LLP, New York, N ew York; C lara Kim, Richard H. Re imer, Jackson W agener, Am erican Society o f Compose rs, Authors a nd Pu blis her s, New

4 York, NY, for Amicus Curiae A merican Socie ty o f Composers, Authors and Publisher s. Anne M. Voigts, K ing & Spalding LL P, Palo Alto, California; D avid P. Mattern, King & Spalding LLP, Washington, D.C.; K enneth Steinthal, King & Spalding LLP, San Francisco, C alifornia, for Amic i Curiae Motion Picture Association, Inc., Radio Music Lice nse Committee, National Association o f Broadcasters, Digital Media Association, Exhibiti ons & Confere nces Alliance, a nd International Associati on of Venue Man agers. M ENASHI, Circuit Judge: This case involves the licensing of copy righted musical works to concert promoters for live perf ormance. Petitioner - Appellee -Cross- Appellant Broadcast Music, Inc. (“BMI ”) repre sents over one million songwriters, compos ers, and music pub lisher s, and it licenses their works to music user s in exchange for a fee. Be cause of BMI’s large share of the music licensing marke t, it is su bject to an antitrust consen t decree that requires BM I to set reasonable licensin g fees and authorizes a court to de termine a reason able fee if BMI and the prospecti ve licensee are una ble to agree. Respondent - Appe llan t - Cross- Appellee Nor th American C oncert Promo ters Assoc iation (“NACPA”) has hist orically purchased b lanket licenses cover ing all the musical work s in BMI’s repertory so tha t ar tis ts ma y perform the works at conce rts. Durin g th e las t ne go tiat ion c yc le, B MI a nd NACP A we r e unable t o agree on a rate and revenue base, and for the first time in the history of the p arties’ relationsh ip, BMI peti tioned the district court to resolve t he impasse. In 2023, the distric t court accepted BMI’s

5 rate quote for the 2014 -20 18 peri od (the “R etroactive Perio d”) and set a rate of 0.5 percent of NACPA’s gross re venues for the 2018 -2022 period (the “Cu rrent Period”). The distri ct cour t also expand ed the definition of “gross r evenues” to cover items not prev iously in cluded in the rev enue base. NACPA appealed bo th decisions, and BM I cross - appealed the denia l of its motion for p rejudgmen t interest. We conclude that the district court impose d unreasonable rates. First, the district c ourt adopted a definiti on of the revenue b ase that had no precedent in the history of the industry without a compelling reason. The d ef inition inclu ded revenue streams t ha t do not reflect the fair market value of the musi c and that involve si gnificant administrative cos ts withou t a corres ponding benefit. Second, the district court set a rate that depends dispropor tionately o n less compara ble benchmark agreem ents. The district court relied on agreements with u naffiliated indiv idual promoters e ven though NACPA has histor ically obta ined sign ificantly lower r ates from the same counterparties. And the district court identified no change in economic circumstances that would justif y a rate more than double what NACPA has historically paid to BMI and to the A merican Society of Composers, Authors, and Publishers (“ASCAP”). Thi rd, while we conclu de that the district court did not abuse its discr etion by denying the moti on for prejudg ment i nterest, the de nial may have depended on the err oneous rates. Ac cordingly, we va cate the judgment of the distric t court a nd remand for furth er proceedi ngs consistent with this opinion. BACKGROUND The owner of the c opyright in a mus ica l w ork ha s the exclusive right t o perform the work publicly. See 17 U.S.C. § 1 06(4). The c opyright own er may au thorize an oth er to perform the work by

6 granting a license. Se e id. § 201(d)(2). In the Uni ted States, the owner of a musical w ork confers on a Performing Righ ts Organi zation (“PRO”) the r ight to license perfor mances of the work, an d the PRO then issu es licenses, col lects fees, an d distributes ro yalties to the copyright own er. See BMI v. Columbia Br oad. Sys., Inc., 441 U.S. 1, 4 -5 (1979). The purchas ers of performance licenses include “c oncert promoters, commerc ial radio stations, br oadcast televis ion stations, broadcast and cable televis ion networks, intern et websites, commercial music services, conc ert halls, universities, airlines, restaurant s, bars and nightc lubs.” Agreed Findings of Fact ¶ 2, BMI v. N ACPA, No. 18- CV -8749 (S.D.N.Y. Oct. 10, 202 2), ECF No. 124-1. The P RO typ ical ly is sue s a bla nke t licen se tha t a llo w s the licensee to perform any of the songs in the PR O’s repertory. Four PROs con trol t he market for mus ic performanc e licen ses: BMI; ASCAP; the Society of Eu ropean Stage Authors and Composers (“SESAC”); and Global Music Rights (“ GMR”). The distr ict court estimate d that B MI, ASCAP, SES AC, and GM R hold repertories that account for about 45.4 percen t, 46.5 perce nt, 3.6 percent, and 4.5 percent of t he t otal mark et, respectiv ely. See BMI v. NACPA, 664 F. Supp. 3d 470, 474 (S.D.N.Y. 2023). Because of the dom inant shares of the license market that the entit ies control, BMI and ASCAP operat e under antitrust co nsent decrees. See F ina l Jud gment, United States v. BMI, No. 64 - CV -3787 (S.D.N.Y. Nov. 18, 1994) (“B MI Consent Decree”); Secon d Amended Final Judgment, Uni ted States v. ASCAP, No. 41- CV - 1395 (S.D.N.Y. June 11, 2001) (“ASCAP C onsent Decree ”). 1 The BMI C onsent D ecree 1 BMI wa s first sub ject to a consent decr ee in 19 41. See Un ited States v. B MI, 1940- 43 Trade Cas. (CCH) ¶ 56,096, 1941 W L 307851 (E.D. Wis. 19 41). The 1941 consent dec ree was replac ed with the c urr ent consent dec r ee in 1966. See United States v. B MI, 1966 Trade Cas. (CCH) ¶ 74,941, 1966 WL 181780

7 require s BMI “ to offe r any prospect ive music user a license to publicly perform t he songs in it s repertoire.” BMI, 664 F. Supp. 3d at 474. 2 If BMI and a prospect ive licensee canno t agree on a rate, then either party may petition the district court to deter mine a “reasonable” rate. BMI Consen t Decree § XIV(A). I NACPA, formed in 1998, is an association of concert pro moters. “To qualify for NAC PA membership, a p romoter must stage a t least 60 shows in a year, including at least 5 club s hows, 5 theater shows and 5 shows in venue s with more than 1 0,000 seats.” Agreed Findings of Fact ¶ 11. In a ddition to “exe cut[ing] ever y aspect of produci ng [an] artist’s concert,” promoters are responsib le for “negotiat[ing] with an artist or their rep resentative f or their l ive performance right s, either (S.D.N.Y. 1966). The BMI Consent Decree wa s amende d at BM I’s r eque st in 1994. “Among other things, the amendment added to the [BMI] Consent Decre e an au tomatic l icens ing provision and recou rse to a rate c ourt to resolve rate disputes between BMI and license es.” Agreed F indings of Fact ¶ 4. The ASCAP C onsent D ecree was first entered in 1941 and was amended in 1950 to ad d automati c licens ing and rate court pr ovisions. See i d. ¶¶ 21- 22. The ASCAP C onsent Decree was amended in 2001. 2 See BMI Consent Decree § XIV(A) (pro viding that BMI “shall, within ninety (90) days of its receipt of a writte n application from an applica nt for a lic ense f or the right of p ubli c performance of any, some or al l of the compositions in defen dant’s repe rtory, adv ise the applicant in writing of the fee which it deems reasonable for the licens e r equ ested. … Pend ing the compl etion of any su ch negotiation s or proceedi ngs, the applicant shall have the right to us e any, some or al l of the com positions i n defenda nt’s repertor y to which its applic ation pertains, without pa yment of any fee or other c ompensation, but sub ject to th e provisi ons of Subs ection (B) hereof, and to th e fina l order or judgmen t entered by th is Court in su ch proceeding ”).

8 for an individ ual engage ment or a ser ies of en gagements or a tour.” 3 According to NACP A, “because concert promoters typica lly do not know all the songs t he artist wi ll perform, or wh ich PRO (or PROs) administers the lice nse for each of thos e songs, promo ters must in practi ce secure blanket lic enses from each of the four pri ncipal domestic PROs … to avoid the r isk of copyr ight infringe ment.” Appellant’s Br. 13. The struc ture of the concert p romotion industry has changed in the past few decades. Prior to the mid - 1990s, the industry con sisted primarily of small, independent pro moters, each of wh ich “exclusively funct ion[ed] within [its] own market. ” 4 Today, two parent companies — Live Nation Entertai nment, Inc. (“Live N ation”) and AEG Pre sents, LLC (“AE G”) — cont rol approx imately 75 p ercent of NACPA’s membe rs. Live Nat ion, AEG, an d affiliated ent ities pay about 90 percent of the fees u nder NACPA’s licen se with BMI. T he business of Live Nation and AE G includes not only concert promotion but also ticket servicing an d venue ownersh ip. Live Nation and AEG o wn the ticket servicing co mpanies Ticketmaster and AXS, respectively. Ticketmaster and AX S charge servi ce fees on mos t concert tickets sold; the fees typically amoun t to about 20 percent of the f ace value of the tick et. BMI has issued b lanket licenses to NACP A since NAC PA was found ed. The first l icense, issued in 1998, required NAC PA to pay a 3 Trial Transcript (10/31/22) at 587, BMI v. N ACPA, No. 18- CV -8749 (S.D.N.Y. Dec. 23, 2022), ECF No. 182; see also Agree d Fi ndin gs of Fa ct ¶ 12 (“[T ]he industry p ractic e is for promote rs, rather tha n the p erform ing artist, to se cure ne cessary license s.”). 4 Trial Transcript (10/24/22) at 110, BMI v. N ACPA, No. 18- CV -8749 (S.D.N.Y. Dec. 23, 2022), ECF No. 174.

9 fee to BMI equal to 0.3 percent of “Gross T icket R evenues” for concerts in venues with under 10,000 s eats and 0.15 perce nt for concerts in venues with over 10,000 se ats. A pp’x 543. The license defined “Gro ss Ticket Revenues” a s “the total monies received, directly o r indirectly, by [NACPA’s] Licen sed Individual Member(s) or their a uthori zed representati ves from all tic ket sales per attraction,” with exc lusions for taxes, ven ue fees, ticket ser vicing fees, and parking fees. Id. a t 540. The license als o gave NAC PA a 10 percent administrative disco unt provided tha t at least 80 percent of its members s igned on t o the license. See id. a t 539. “B lending the 0.15% and 0.30% rates on a weighted basis ba sed on fees paid to BMI, NACPA paid an effective ra te of 0.21% under the BMI h istorical licenses, before the application o f the adm inistrative d iscount.” Agreed Findings of Fact ¶ 72. BMI and NAC PA entered int o a new licens e in 20 06 on the same te rms as the 1998 li cense, except that the 2006 license added a separate rate for music fest ivals. A ccor ding to BMI, fe st ivals were not covered by the 1998 license. Th e 2006 license was renewed several times bef ore BMI provided notice in August 2013 tha t it wo uld te r mina te effective December 31, 2013. “S ince then, the parti es have been on an interim agreemen t and have engaged i n protracted negot iations for a new license without success.” BMI, 664 F. Supp. 3d at 476. BMI’s cur rent propos al, which it quoted to NACPA in an email dated Decembe r 9, 2020, sets d ifferent terms for the Retroactiv e Period and the Cur rent Period. F or the Retroact ive Period, BMI proposes keepi ng the histo ric al revenue base — face value of tickets sold — an d charging a varying per centage fee based on the size of the venue, as follows:

11 administrative disc ount. N ACPA argue d that the rate sho uld be converted from a t iered s tructure to a simple unitary rate of 0.21 to 0.275 percent of g ross ticket revenues. See BMI, 664 F. Supp. 3d at 477. II When the parties fail ed to reach agreem ent, BMI petitioned the distr ict c our t to determine a reasonable rate. Pursuant to a 2014 agreement, during the pendency of the negotiations and trial, NACPA pa id BM I a n inte rim ra te o f 0.3 pe rcent of the face value o f concert tickets for ve nues with fe wer than 10,000 se ats and 0.15 percent for ven ues with more than 10,000 seats. T he district court issued its opinion on March 28, 2023, follo win g a five - we ek tria l. The d ist rict court f irs t addre ssed the Current Peri od. It ex plained that B MI ha d “attempt[ed] to jus tify its proposed rate through the use of sixteen benchmarks which can be put into t hree grou ps”: (1) licenses issued by BM I or ASC AP to NACPA or non - NAC PA promoters; (2) lic ens e s issued by SESA C and GMR to NAC PA, non- NACPA prom oters, or individual NACPA members (L ive Nation and AEG); and (3) licenses issued by foreign PROs. Id. a t 479. To fac ilitate a comparison between the benchmark agreeme nts and BMI’s proposal, BMI’s economic expert, Professor Cather ine Tucker, ap plied seve ral adjustments to t he rate of each benc hmark agreement. First, Tucker converted a ll of the benchm ar k rates into unitary rates: “[W]h ile BMI’s proposed license stipulates a sing le rate of 0.8 percent for all live concerts and festivals, certa in benchmark licens es contain a fixed fee per tick et, tiered rates base d on venue capacity, tiered ra tes b ased on even t revenue, and/or differe nt licens e

12 rates for festivals and conce rts. ” 5 “ Therefore, t o make t hese different rate struc tures comparable to BMI’s single proposed rate for all live concerts and fe stivals, ” Tuc ker “convert[ed] al l fixed fee, tiered, or event - specific rates into a single, percentage -of- revenue rat e.” 6 She did so “by we ighting … each tier’s or event type’s unique rate by its share of in dustry revenue a nd then calc ulating the weight ed a verage across these ra tes.” 7 Second, because “the pr oposed license cover s songs in B MI’s repertoire, ” Tuck er “convert[ed] t he rates that prom oters negoti ated with other PROs to reflect differences in t he monetized use of BMI’s repertoire i n live concerts relati ve to the monetized use of be nchmark PROs ’ repertoires.” 8 Tucker reas oned that “if a promoter were willing to pay $5 for access to ASCAP’s repertoire for a concert that brought in $100 in re venue, and if the same con cert used twice as many son gs in BMI’s repert oire, ot her things equal, t ha t promot er should be wi lling to pay BMI $10.” 9 Third, because BMI’s propo sed revenue ba se is broader t han the revenue ba se in the benchmark agreement s — wh ich uniformly define gro ss revenue as the fa ce value of ti ckets sold — Tucker adjusted each rate downward to r eflect royalty pa yments as a percentage of the total face value of tickets sold, revenues fr om tickets sold directly into the s econdary market, fee s above the face value of 5 Tucker Affidavit A ppendix C ¶ 2, BMI v. N ACPA, No. 18- CV -8749 (S.D.N.Y. Dec. 23, 2022), ECF No. 123. 6 Id. 7 Id. ¶ 4. 8 Id. ¶ 79. 9 Id.

15 With resp ect to the revenue base, t he district court held th at “sponsorship an d advertis ing contributions must be excluded because the y do not reflect t he music or its performa nce.” BMI, 664 F. Supp. 3d. at 47 9. But th e dis trict co urt inclu ded the other four revenue cat egories that BMI proposed. First, the d istric t cour t concluded that “[d]ef ining the revenue base to inc lude tickets so ld for the first time in a secondary market is a natural c larification of what some prom o ters think the b ase alread y subsumes.” Id. at 484. Second, the dis trict court con cluded th at box suite and VIP pa ckage re venues should be included because, altho ugh those sales “may i nclude charges in add ition t o the seat ticket, li ke for foo d or mer chandise,” the purchaser still “must pay the total price if she wishes to sit in these premium seats and h ear the mus ic.” Id. Third, the d istric t court he ld that ticket ing service fees shoul d be included because the co nsumer pays those fees when sh e purchases a ticket and “[t]here is vir tually no way to opt out o f paying the fees.” Id. The dis trict court acknow ledged NA CPA’s concern that “calculating the [re venue base ] in this man ner is unworkab le because promoters do not have acce ss to all the revenue inform ation, specifically revenues for boxes and VIP packages.” Id. at 485. T he distr ict court believed, however, that “[l] imiting the revenues to those received by the promoters or a co ntractually rel a ted third - party helps to allevia te those concerns, even if it does not produce perfect ly efficient administrat ion.” Id. To determin e the reasonable rate to apply to this exp anded revenue base, the distri ct court first concluded tha t BMI’s p roposed rate of 0.8 percent was unreasonable. Cf. BMI Consent Decree § XIV(A) (“[D]efenda nt shall have the b urden of proof to es tablish the reasonable ness of the fee requested by it.”). The dis tric t court noted that “[w]hen look ing at the secon dary implied rates, on ly the f oreign

16 licensing agree ments have rates above or close to BMI’s propos ed one of 0.8%.” BMI, 664 F. S upp. 3d at 483. For that reas on, “the reasonableness of B MI’s propos ed rate of 0.8% r ests on the foreign licensing agreements.” Id. Th e d istr ict court dec ided tha t “these foreign li censes cannot be relied up on as valid benchm arks because BMI failed to show that the foreign par ties are si milar to BMI a nd NACPA and t hat the agreements were ne gotiated in simila r economi c circumstances.” Id. Instead, th e distr ict court concluded that “[a]ll the BMI and ASCAP licen ses are pro per benchmark s” regardless of whet her those licens es were negotiat ed with NACPA or with non - NACPA promoters. Id. at 486. The d istr ict co urt observed that the l icenses issued to non - NACP A promoters “arose in a market that refle cts the same degree of co mpetition and econ omic circumsta nces, and they cover the same ri g hts.” Id. Mor eover, “[a]lthough NAC PA and non - NACPA prom o ters differ i n the scope of their busi ness operations, they are di rect competitors for shows hosted i n s maller venues.” Id. T he dis tric t court also concluded th at “[t]he SESAC an d GMR licens es are appropriat e benchmarks” because “[c]ompar ed to [the ] proposed license, t he SESAC and GMR licenses are between simi lar parties, for si milar ri ghts, and were nego tiated in s imilar eco nomic circumstances.” Id. a t 487. The d ist ric t co urt explain ed that, beca use SESAC and GMR are not subject to a consent decree tha t prevents those entit ies from refusing a license to any music user, the ir negotiations “approximate[] the dyna mics of a direct licensing negotiation betwee n a music user an d the individ ual music publis her.” Id. The distric t court determined that “the fact that SESAC and GMR oper ate free of a con se nt decree does not mean they are extractin g sup er competitive rates.” I d. at 488.

17 Based on th is analys is, t he distr ic t court consid ered eleven benchmarks, with the “range of the implied rates accounting for the expansion in the gros s revenue base includ[ing] rates of 0.21%, 0.23%, 0.34%, 0.36%, 0.37%, 0.51%, and 0.54%.” Id. Given the benchmark rate s, th e d istric t co urt determined that “[a] rate o f 0.5% of the expanded gross r evenue base is … a reasonable one.” Id. The distr ict court decided that “[i]ncreasing the rate o f a license for l ive concerts better refl ects the fai r market value placed on l icenses in music intensive industries, ” such as “co mmercial radio s tations or virtual live concert st reaming services.” Id. T he d istr ict cour t additionally decided that BMI could n ot eliminate t he 10 per cent administrative discount because N ACPA had already per formed the administrative services that the discount was intended to remunerat e. Id. at 489. The district court a dopted BM I’s proposal for the Retroac tive Period. The distr ict c ourt explained that this “quot ation would al ign the NACPA rates with the rates paid to BMI by non - N ACPA promoters during th at time.” Id. “Hav ing already found that th e BMI - nonNACPA bench mark is valid,” the d istrict court conclude d that BMI’s proposal was reasonable. Id. At th e same time, i t d isal lowed BMI’s proposal to eliminate the 10 per cent administrat ive discount for the Retroactive Period. Id. 12 Following the entry of judgment, BMI moved for an award of pre - and post - judgm ent inter est. The d istr ict court gran ted the motion for post - judgment interest but d enied the motion f or prejudgment interest. The distric t cou rt reasoned that th e interim f ee rate was “correct for the time it [was] in effect” a nd for that reason BMI had 12 On appeal, BMI does not ch allen ge the decision of the distri ct court wi th respec t to t he administrat ive discount for either the Current Period or the Retroac tive Period.

18 not been wron gfully deprived of the t ime value of money. B MI v. NACPA, 674 F. Sup p. 3d 70, 73 (S.D.N.Y. 2023). The distr ict court clarified that its “det ermination of the f inal fee amount was de signed to be the entire embo diment of what const itutes a reasonable fe e.” Id. This appeal followed. DISCUSSION In i ts appea l, NACPA ar gues that the ra te s the district court adopted for the Cur rent Period and the Re troactive Period are unreasonably high. NACPA also contends th at the d istr ict cou rt er red by expandi ng the defini t ion of t he revenue base for th e Current Period. In B MI ’s cross-appeal, BMI argu es that the d istric t co urt abused its discretion by denying its moti on for prejudg ment interest. NACPA maintains that it did n ot. We agree wi th NACPA that the district court erred by expanding the defini tion of the r evenue base and se t unreas onable rates. While we conclude that the dis trict cour t did not abuse its discreti on by denying the motion for prejud gment interest, the denial may have depende d on the rates the di s trict court s et. As a result, we vacate the judgmen t of the district court and remand for further proceedings consistent with this opinion. I “We review the rate set by the Distri ct Court for reasonableness.” Un ited States v. B MI (Mu sic Choice IV), 426 F.3 d 91, 96 (2d Cir. 2005). This stan dard requires us to ask (1) whether the rate is substantively reason able suc h tha t it is not “based on any clearly erroneous findings of fact,” and (2) whether the rate is p rocedurall y reasonable such that “the setting of the rate, including the choice and adjustment of a ben chmark,” is not “ base d on legal err ors.” Id.

19 A The parties disa gree about how we should e valuate the decisions of the distri ct court to rely on the SESAC and GMR licen ses as benchmarks, to expand the revenue base f or the Current Peri od, and to depart from the rates previous ly negoti ated by NACPA with BMI and ASCAP. NACPA argues that such decisions are subject to de novo review, and BMI ma intains that we sh ould review for clear error. According to N ACPA, “th is [c ] ourt has a pplied de novo review to every one of the determin ations chall enged by NACPA in this case,” includ ing the “selection of benchmarks,” the “weigh ing of benchmarks,” an d the “revenue base.” Appe llant’s Reply Br. 6. In United States v. A SCAP, we s aid that “[d ]ete rmination s by the distric t court that particular benchmarks are comparab le and part icular factors a re relevant are questions of law reviewed de novo ” but the “factual findings as to e ach factor under cons ideration or those underlying a proposed benchmark agreement, as we ll as fin ding s with respect to fair market value, are reviewed for clear error.” 627 F.3d 64, 76 (2d Cir. 2010). In United Sta tes v. BMI (Music Choice II), we said that th e dis tric t court “made a fundamen tal error ” in concluding that “ret ail price was not a g ood indica tor of fa ir market value” and sh ould not be us ed as the rev enue base. 316 F.3d 1 89, 195 (2d Cir. 2003). And in ASCAP v. Showti me/The Movie Chann el, Inc. (Showtime II), we said that “whether the price paid b y a buyer may be given le ss weig ht as a ‘comparab le sale’ either b ecause the pr ice was higher than wou ld have obta ined in a more competitive market or was perceived by the buyer to be higher is a matter of law.” 912 F.2d 563, 569-70 (2d C ir. 1990). In response t o these author itie s, BMI argue s that the distr ict court made “factual determ inations” w hen it deci ded to “ adopt[]

20 SESAC and GMR licens es as benchmarks,” to “order[] an e xpansion of the reven ue base to include the full retail cost of atten ding a concert,” and to “set[] a rate at the high end of the benchmark range that does not mirror the ASC AP/ NACPA (and h istorical BMI/NACPA) rate. ” Appellee’s Br. 41. BMI ob serves t hat in Showtime II we said that “[f]air market value is a factual matter, albeit a hypothetical one, ” and that the selection of an appro priate benchmark is “a de ter mination analogou s to an evident iary ruling that would ha ve occurre d if the price of the [relevant ] license had been offered at a jury trial.” Showtime II, 9 12 F.2d at 569-71. Accord ing to BMI, “NACPA does not identify the lega l error s ” in the opinion of the district c ourt “ because there are non e. ” Appellee’s Br. 41. The d istrict court instead “ applied the correct four - factor legal framewor k articu late d in Music Choice IV.” Id. We conclude that w e properly revi ew these issues de novo. In Showtime II, we analogized the de term ination of fair mar ket value to an ev iden tiar y ru lin g but explained that we would nevertheless accord plenary review to the leg al iss ues i nvolved in the determination: [T]he factual compo nent of the issue bef ore the [district judge] does not rende r all aspects of his decision - makin g subject t o rev iew under the “ clearly erron eous” standard. In making a factual determ ination, a dec ision - maker m ight rely on legally impermis sible factors, fail to give considera tion to legally relevan t factors, ap ply incorrect legal stan dards, or m isapply cor rect legal standards. In jury trials, such matters are nor mally resolved by rulings on admiss ibility of evidence and by jury instruc tions. In court trials, where the functions of fact- finding and expos ition of law are perfor med by the same person, the line between the functions is not always

21 distinct. For example, the line bet ween admissibility of evidence (law) an d evaluation of the pers uasive force of evidence (fact) is often blurred. N evertheless, an appellate court is obliged to observe the law/fact distinction as best it can and accord p l enary review to any aspect of a trial court’s de cision tha t can fairly be isolated as deter mining an issue of law. 912 F. 2 d at 569. We h ave articulated lega l standards that gov ern the selection and we ighing of ben chmarks and the definition of an appropria te revenue base. 13 The argu men ts tha t NACPA raises on appeal concern whether the distr ict cour t “fail[ed] t o give consideration to leg ally relevan t factors” and “ misappl[ied] correct legal standards” w he n it adopt ed t he SESAC and GMR licenses as benchmarks, expand ed the revenue base, and set a rate at the high end of the benchmark range. Showtime II, 912 F.2d at 569. B ecause t hese decisions “can f airly be isolated as determin ing an issue of law,” we undertake a plenary review. Id. B If BMI does not meet its burden of proof on the reasonabl eness of its requ ested rate, the distric t cour t mus t “ determin e a reason able fee based upon all t he evidence.” Mu sic Choice II, 31 6 F.3d at 194 (quoting B MI Consent Decree § XIV(A)). It must ma k e that 13 See, e.g., BMI v. DMX I nc., 683 F.3d 32, 45 (2d Cir. 2012) (“In assessing wheth er anothe r agree ment provides a valid benc hmark, the dis trict court mus t cons ider wh ether the othe r agreem ent dealt w ith a compara ble righ t, whether it invo lved similar parties in similar economic circum stances, and wheth er it a rose in a s uff icientl y compe titive ma rket.”) (c iting M us ic Choice IV, 426 F.3d at 95); Music Choice I I, 316 F.3d at 195 (“[A]bsent some valid reason for using a differe nt measure, what ret ail customers pay to receive the product or service in question (in this case, the recor ded music) seem s to us to be an e xcellent ind icator of its f air market valu e.”).

22 determination by “es tablishing the fair market va lue of the m usic rights, in other wor ds, ‘the price that a willing buyer an d a willin g seller would agre e to in an ar m’s length tran saction.’” Music Cho ice IV, 426 F.3d at 95 (qu oting Music Choice II, 316 F.3d at 194). In attempting to apply that precedent, dis tric t courts have lamented “the pr oblematic nature of the analys is envisioned and the enigmati c task thr ust up on a cour t that m ust under tak e this evaluation.” Unit ed States v. A SCAP (ABC/ CBS), 831 F. S upp. 13 7, 143 - 44 (S.D.N.Y. 1993). A s one distr ict co urt h as exp laine d: [T]he market f or blanket licens es appears to be one whose natural consequence is the lack of broad -based competition. To pos tulate wha t prices would pr evail were such a m arket “ comp etit ive ” is p erp lexing in theory, impractica l in practice, and dub ious in ou tcome given the eff iciencies obta ined due to the aggregatin g nature of the serv ice render ed. Little can be adduced as to the expected behav ior of such a market were it populated by multiple sellers. It i s questionable whether such a mar ket coul d sustain m any sellers; whether the market would func tion efficiently and the price leve ls at which its suppl y and demand would con verge are e ven more uncertain. In addition, limite d evidence is discernible as to the [marginal] costs [of i ssuing blanke t licens es] facing [the PROs]. Conseq uently, any rate - setting standard that cal ls, in the abs tract, for a th eoretic construct of a competitive market in blanket licensing must con front the reality that the re e xists m inima l evidence as to what that marke t would look li ke, much less t he pr ices it w ou ld y ield. Id. at 144 (footnotes omitted). S cholars have similar ly argued that, from an economic per spective, there is no “indisputab ly corre ct solution” to the problem of determ ining the fair market va lue of a

23 blanket license because “[i]n the noncompetitive marke t of musical performanc e rights … a competitive price will never emerge, and there is no economically meaningful method of determining a competitive price.” 14 The distr ict court faces three primary obsta cles to comp let ing the task assigned to it under the c onsent decrees. Firs t, because “there is no competitive market i n music rights, ” the part ies and the c our t “ lack any economic d ata that may be readi ly translated into a measure of competitive p ricing for the r ights in q uestion.” Sho wtime II, 912 F.2d at 577. Second, in a co mpetitive market, the price of a product or service should eq ual the margina l cost to the s eller of produc ing an additional unit of that product or service. But “ the margina l cost of additional consump tion of musical composi tions is zero or eve n negative.” Sobel, supra note 14, at 33 -34. 15 As a result, “accord ing to economic theory, the ‘competitive’ price for music [licenses ] wo uld be zero (or less).” Sobe l, supra note 14, at 34; see also Cirace, supra note 14, at 283 - 84 (“Because marginal cost is zero in the marke t of musical 14 John Cirace, CBS v. ASCAP: An Economic Analysis of a Politica l Problem, 47 Fordham L. Rev. 277, 277, 304 (1978); see also Lionel S. Sobel, The Music Business and the Sherma n Act: An Analysis of the Eco nomic Realities of Bla n ket Licensi ng, 3 Loy. L.A. Ent. L.J. 1, 33 (1983) (“ [T]he natur al mar ket fo rces of supply and demand do not operate normally in the music [licensing] busine ss.”). 15 The m arginal cost to the PRO s of issuin g a blanket license could be negative “ because each time they iss ue an additio nal blanket lice nse, a potential copy right infringer is eliminated, and t hus their copyrig ht enforcement expenses are reduced.” Sobel, supra note 14, at 37; see also ABC/CBS, 831 F. Supp. at 144 n.17 (noti ng that a blanket li cens e “ lo wer s [the PRO’s] monitori ng cos t”).

24 performanc e rights … the competit ive price theoret ically woul d also be zero.”). Setting BMI’s price at zero, however, wo uld confl ict with the goa l of co mpensating copyright own ers for the right to u se their musical works. Third, BMI and NA CPA both have subs tantial marke t power, so a negotiation between BM I and N ACPA involve s a bila te ral monopoly. See Cirac e, supra note 14, at 28 1. “O ne distinctive featu re of a bilateral monopo ly is that … price and output are indeterminat e.” Id. at 283; see also id. at 304 (“ Given the wide range of pos sible prices in bilateral monopoly bar gaining, the co urt, if it did set a price, would probably choose the historical price w ith minor adjustments.”). In light of these featu res of the relevant m arket, a distr ict co urt cannot ident ify a uniq ue “market v alue” for a B MI blanket license. “Rather than speak in terms of c ompetitive market pricing, when s uch a term carries vague significance within the context of the bl anket licens e market,” d ist ric t cour ts have cons idered it “more ins truc tive to view the [c] ourt’s role as a m oderating influence on [BMI or ] ASCAP that serve s ‘to minimize the likelihood that [its ] e v ide nt market lev erage may be exerted to obtain unacceptably inflated price levels for it s licen se [s].’” AB C/CBS, 831 F. Supp. at 144 - 45 (quoting Showtime II, 912 F.2 d at 576). The dis trict court will therefore “consid er previou s agreements volu ntarily entered betw een the parties, or those similarly si tuated, as the s tarting po int of its ana lysis” an d then “account for alterations in the econo mic conditions confronti ng the parties, as we ll as appraise var iations in t he nature and v alue of the rights at issue.” Id. The d istric t cour t will not “merely end orse as appropriate for today the ter ms of compromises concluded yesterday, ” but i t also can not “ignore the his tory of the parties’ preferences as exp ressed in thei r prior agreements.” Id. a t 145. In fact, the “prices negot iated voluntari ly in an arms-leng th transac tion offer

25 the only palpab le poi nt from wh ich to pro ceed towards an e stimat ion of fair value for later periods.” Id. II The price of a music license equal s a percentage rate of the revenue deri ved from the licensee’s use of the music. In t his cas e, th e parties disagree as to b oth the rate and the de finition of the gross revenue base. Histori c ally, music lic ense agreements between PROs and concert pro moters have defin ed the gross revenue base a s the face value of concert tickets sold. BMI seeks t o expand the re venue base for the Current Period to include, in addit ion to the face va lue of tickets s old, (1) “revenues received by the promoter from any tickets sold in the f irst instan ce directly onto the secondary marke t (incl uding any amounts abo ve the face va lue of the ticket),” (2) “any ticket service, handling, or other fees above the face value o f the ticket paid by the consu mer if received by the p romoter,” (3) “box s uite and VIP package revenues attribu table to live c oncerts and pa id to the promote r or a venue or arti st with which the prom o ter has a contractual re lationship,” and (4) “spons orship reven ues at tributable to l ive concer ts and paid to the promote r.” BMI, 6 64 F. Supp. 3d at 478. BMI recognizes that “[h]istorical ly, the reve nue base for concer t promoter PRO licen ses included only the face value of the tickets sold, ” b ut it argues th at this “expanded r evenue b ase” is neces sary “[t]o account for changes in the concer t industry.” Appe llee ’s Br. 31. NACPA main tain s t hat the gros s revenue base sh ould co ntinue to be defined as the face value of concer t tickets so ld. Accordi ng to NACPA, “every agreement in the history o f the industry” has adopted that definition and it wo uld b e impr acti cab le f or concert

26 promote rs to calculate gross reven ue under the new definition that BMI proposes. Ap pellant’s Br. 54. The district court began its analysis w ith th e prop ositi on that the gross revenue base should equal th e “fair market val ue of the music” measured by “[w] hat the cons umer pays to at ten d the concert.” BMI, 664 F. Supp. 3d at 479. The district c ourt proceeded from that proposition to the conclusion that the gross rev enue base for the Current P eriod must be “enlarge d to include not o nly the face value of tickets, but also the value of ti ckets sold in the primary instan ce directly onto the secon dary market, ticket serv icing fees received by the promoter, and revenues f rom box suites and VIP package [s] that are attr ibutable to live concerts an d paid to the promote r or a contractu ally related third - party. ” Id. at 485. The district c ourt exp lained that “[e]ach of those categor ies reflect [s] payments to ‘receive the pr oduct or service in que stion,’ a live concert, which is ‘an excellent indica tor of its fair market va lue,’ wh ile controll ing for revenues tha t never flow to the promoters. ” Id. at 484 (citation o mitted) (quoting Music Choice II, 316 F.3d at 19 5). The distri ct court a cknowle dg ed NACP A’s argument “tha t calcula ting the gross in this manner is unworkabl e because promoters do not have access to all the revenue information, specifically revenues for box es and VI P packages. ” Id. at 485. But the dis tric t cour t reasoned tha t “[l]imiting the re venues to thos e received by the promoters or a c ontr actually related third - par ty helps to alleviate those conc erns, even i f it do es not produce perf ectly efficient administration.” Id. We conclude that the decision of the distr ict cour t to expan d the definition of the revenue base for the Current Peri o d was unreasonable for thre e reasons. F irst, the district court did not provide

27 a n adequate justific ation for adopt ing a definition that h as no precedent amon g the benchmark agreements that the parties ident ified. The dis tric t cour t r ecognized that calc ulating the expande d revenue base wou ld create adm inistrative costs for NACP A, but it d id not identify any corresponding benefits to the parties that would offset the costs. The record indicates that there are no such benefit s. Second, even accepting the premise that the revenue base sho uld be measured by what the consumer pays to attend the concer t — rather than by the face va lue of concert ticke ts sold — the d istric t co urt included revenue categories that do not reflect what the consumer pays for the concert. Third, the d istric t cour t did not resolve the objection that the expan ded revenue base would be com mercially imprac tica ble. A The district court considered t welv e ben chmark ag reements — includi n g agreements by bot h regulated and unregu lated domestic PROs with NAC PA and n on - NACPA promoters as we ll as wi th individual members of NACPA such as L ive Nation and AE G. All of the benchm ark agreements defined t he re venue base in t he same way: the face value of con cert tickets sold with “cus tomary deducti ons for ticket servicing fees, taxes, faci lity fees, and par king.” 16 The district court did not addr ess the fact tha t none of t he domestic benchmark agreem ent s the pa rties identif ied u sed a nythin g other tha n the face value of tickets sold to measure the revenue base. Even assuming that the newly defined revenue base captured the “fair market value of the m usic” — wh ich, as explaine d in Part I I.B., it 16 Memorand um in Support of Motion for J udgment as a Matte r of Law at 44, BMI v. N ACPA, No. 18- CV -8749 (S.D.N.Y. Dec. 20, 2022), ECF No. 171; see also App’x 440 -41.

28 did not — the dis tric t court need not define the revenue base according to an abstract no tion of “what retail customers pa y to receive the product or service in question” if there is a “valid reas on for using a different measure.” Music Choice II, 316 F.3d a t 195. In this case, there was not only a v alid but a comp ellin g reason to define the revenue base as the face value of tickets sold: every domestic ben chmark agreement that the district cou r t considered defi ned the revenue base that way. Benchmarks play the paramount r ole in rate - sett ing proc eedin gs b ecaus e it is imp ossib le to id entif y an indisputabl y correct market pri ce based on economic principles. When the dis tric t court adopts a rate, a revenue base, or a substan tive license t erm that ha s no pr ecedent among the benchmark agreements it consider ed, it must have a compelling r eason for doing s o. The primary rol e of the dis tr ict c our t in a r ate - s ett ing proceeding is to serve “as a moderating influence ” on the PRO and to “mini mize the like li hood tha t [it s ] evide nt market leverage may be exerted to o btain unaccept ably inflat ed price levels f or its license [s ].” ABC/CBS, 831 F. S up p. at 144 -45. If the industry has s ettled on a particular defini tion of gros s revenue — or anothe r substantive license term — the role of the d istr ict cour t will pr esumpt ively be confi ned to determi n ing whether the percent age rate quoted by the P RO is reasonable, holding the revenue bas e definition and other license terms fi xed. The economic context supports the conc lusion that a d istric t court wh ich aims to curb the regu lated PRO ’ s excessive bargaining power must ordinar ily decide only the price of a music license while holdin g the substantive terms, such as the definition of gross revenue, fixed. As con tract la w scholars have expl ained:

29 It is widely beli eved that parti es exercise barg aining power by requi ring weaker contra ct ing partners t o take unfavorable [substan tive] terms. … But w hen bargaining power is determi ned prior to contrac t for mation, as is common in business contexts, these v iews are incorrect. Bargaining power ins tead is e xer cised in th e divis ion of the surplus, which is determined by the price term. Parties j ointly ch oo se the contract terms s o as to maximize the surplus, which the price m ay then divide unequally. 17 In other words, both PROs and promoters wi ll prefer the set of substantiv e l icense terms — i nc luding the revenu e base defi nition — that maximizes the total value created by the b lanket license. T he party w ith the greater bargaining power will then seek a greater share of that value by ne gotiating for a higher percentage ra te. I t is in neither pa rty’s i nterest to adopt a rev enue base defi nition that imposes admin istrat ive costs withou t ad ding to the tota l sur plus that the parties share. Consistent with these dynamics, PROs and c oncert promoters have un iformly adopted a sim ple and easily ad ministrable revenue base def inition — the face va lue of concert tic kets sold —and have negotiated va rying royalty ra tes according to the relativ e bargaining power of the parti es. 18 17 Alan Schwartz & Robert E. Scott, Con tract Theor y and th e Li mits of Contr act Law, 113 Yale L.J. 541, 554 (2003). 18 NACPA’s economic expert made a similar point in hi s trial testi mony. See App’x 346 -47 (“ I rea lly don’t think ther e is such a thing as an econom ically idea l revenue b ase. … [F]or an y given amount of royalti es that are going to be paid, you can achieve t hat with a broad base and a relativel y low perc entage or w ith a nar rower bas e and a high er p ercen tage. So the ch oice of the b ase really is just a matter of con venien ce an d practi cali ty in terms of trackin g it. … [I]f the conveni ent and prac tical base someh ow yields too sma ll a reven ue stream, you ca n always deal with that

30 The parti es t hat nego tiated the twe lve benchmark agreements concluded that the benefits, if any, of a broader or more complex revenue base de finit ion did not outweig h the adm inistrative costs. The distri ct court offered no reason to be lieve tha t this neg oti ation between BMI and NAC PA should be the first to rea ch a different conclusion. The dis trict c ourt admi tted that its broad er revenue base definition “does not produce perfectly eff icient administrati on.” BMI, 664 F. Supp. 3d at 485. But the dis tric t court did not identi fy any corresponding benefits to justify the se administra tive costs or any other reason to expect that broa dening the revenue base will increase the surplus from the licens in g agreement. BMI argues on appea l that fai ling to include al l revenue str eams that flow to Live Nation and AEG in the gross revenue base would allow th o se entities to play a “revenu e shell game” to minimi ze payments to PRO s. Appellee’s Br. 3 4. A ccording to BMI and some amic i, L ive Nation an d AEG have increa s ed “junk” ticketing fees — and thereby boost ed revenue to thei r ticketing a ffiliates Ti cketmaster and AXS — w ithout pr oportionately in crea sing th e face valu e of t he ticket s. In other words, Live Nation a nd AEG e ach seeks “to pretend that songwriters s hould only be paid out of its un ilaterally determi ned promoter pocket whil e it stuffs revenu es into its other pockets.” Brief of Am icus Cu riae Na tiona l Mu sic Pu blish ers ’ Association 12. The distr ict co urt did not adopt this “shell game” arg ument or even mention the argument in its opinion. In any event, the argument does not justify the d eparture from the uniform subs tantive ter ms of the benchmark agreem ents. If Li ve Natio n is man ipu lat ing re ven ues in this way, t he expanded revenue base definitio n may preven t Live by inc reasing the royal ty percentag e. ”).

31 Nation f ro m tra nsferring valu e t o itself and awa y from BMI and i ts artists. But i t would not increase the joint surplus from the blanket licens e agreem ent; to the contra ry, it would reduce the surplus because it is costly to administer. BMI could pres erve the la rger surplus by retaining th e simpler de finiti on and still receive compensation for any losses that resul t from the promoters’ “shell game” b y in sis ting on a higher perc entage rate. See supra note 18. Adjusting both the rate and the base, by contrast, w ould allo w another sort of shell game that enables BMI to captur e revenue beyond the appropri ate rate at the cost of an overall destruction of value. BMI argues that the expanded revenue base is n ecessary “[t]o account for changes in the concert indust ry.” Appe llee ’s Br. 31. The primary ch ange is the advent of large and powerful promoters. Even if the emergence of Live Na tion and AEG has increased N ACPA’s bargaining p ower vis -à- vis B MI — wh ich, as exp lain ed in Pa rt III.C., seem s un like ly — the relative bargaining power of the parti es does not ordinarily affect su bs tantive cont ract terms su ch as the rev enue bas e definition; it affect s only t he pri ce term, which in this case is the royalty rate. See S chwartz & Sco tt, supra note 17, a t 554. In a real -world, arm’ s- length negotiation, th e PRO woul d have no incentive to ins ist on a defin ition of the revenue base that incre ases the administrative cost to music users without a corresponding benefit —and that thereby reduc es the total sur plus va lue of th e licensing a greement. Meanwh ile, if an exp anded revenue bas e would increase the total val ue of the deal, then a music user would a gree to it regardl ess of its bargaining power. But there is no a pparent justification for the d istr ict co urt to impos e a n unprecedented revenue base against a pa rty ’s will. T he dis tr ict co urt acted unreasonably by

32 departing from the industry - standard revenue bas e definition without a compe lling reason. B Even if it were appropriat e to define the revenu e base not by reference to hist orical standar ds but acc ording to the pr inciple of “[w] hat the consumer pays to attend the concert,” BMI, 664 F. Supp. 3d at 479, the de finition adop ted here does not c onform to that prin cip le. The dis trict cour t i nclu ded c ategori es of revenue — from box suites an d VIP package s — that do no t reflec t what th e consumer pays to att end the concert. The d istric t court reas oned tha t while “[b]ox s uites and V IP packa ges ma y include char ges in a ddit ion to the seat ticket, like for food or merchandise,” a consumer “ mus t pay the total price if she wishes to sit i n the se premium seats and hear the music.” Id. at 4 84. In reaching that conclusion, the dis tr ict c ourt misinterpreted our de cis ion in Music Choice II. That case involved a compan y in “the so - called residential music service indus try, consisting of com panies providing mus ic to cable and sa tellite TV subs cribers.” Music Choice II, 316 F.3d at 191. The d istr ict cour t had held “tha t r etail revenues of t he cable / satellit e operators did not prop erly reflect the fair ma rke t value of t he music beca use in payi ng the retail pr ice the subscriber w as paying for materials and services not provid ed by the author of th e music, such as the machi nery of tr ansmission and deliv ery to the subscriber’s h ome.” Id. at 194. For that reaso n, the distr ict court “ concluded that the true val ue of the m usic is expressed at the earlier stage wher e it is incorpor ated into Music Ch oice’s programs and sold wholesa le to the cable / satellite op erators.” Id. (internal quotatio n marks o mitted). We vacated the judg ment of the distr ict court and remanded for further proceedings. In doing so, we

33 recognized that “ to make the music av a ilable to its c ustome rs, the retail sell er must incur expenses for various processes and services not provided by the owner of the music, such as t he layin g of cable, the estab lish ment o f sa tellite systems, etc. ” Id. at 195. But we sa id that th ese retail expenses were “in no way i ncompatible with the propositi on that retail revenues derived from the sale of the music fairly measure the value of the music” bec ause “[t]he c ustomer pays the retail price because the cu stome r wants the music, not because the customer wants to finance the laying of cable or the launching of satellites.” Id. In th is c ase, the dis tr ict cour t de cided th at the perks of a b ox suite o r a VI P package resemble “ t he machinery of transm ission and deliv ery to the subscriber’s hom e” that we describe d in Music Choice II. Id. at 194. The distr ict court explained that while the retai l price of a box su ite or a VIP package covers items whose “onl y relat ionsh ip to the mu sic is in conn ec tion wit h its de liver y to m arke t,” th e price s till “reasonably refl ects the value of the music because the customer is willin g to pay it t o hear the musi c.” BMI, 664 F. Supp. 3d at 484. The analogy to Music Choice II is misp lace d. The customer who buys a cable t ele vis ion package with a music channel “pays the retail price because the customer wants the music, not because the customer wants to finance the laying of cable or t he launching of s atellites.” Music Choice II, 316 F.3d a t 1 95. Those retail expen ses represent the cost of delivery to t he consumer. The c ustomer who buys a VIP concert package, by contrast, pays the premium re tail pr ice precisely because the customer does wan t to finan ce the delivery of ad ditional services, such as food and drinks, a ccess to a box sui te, a backstage pass, or ot her benefi ts. The VIP cus tom er pay s more than other custom ers to lis ten to the s ame music bec ause the price of a VIP ticket

34 involves a payment not only to attend th e concert but also to receive these additional benefits. The retail expenses associat ed with producing a concert — which would r eflect the cost of deliver y to the consumer — are already reflec ted in the price o f a regular tic ket. BMI suggests that “[d]efining the revenue base t o explicitly include VIP reven ues is also necessary to en sure consistency i n payment among pr omoters and across concerts” because “[e]ven though NACPA … instructed its memb ers in 2016 that su ch VIP revenues shou ld be included i n the revenue base, concert promoters took inconsistent positions on the inclusion of such reve nues.” Appell ee’s Br. 57 (c itation omitted); see also BMI, 664 F. Supp. 3d at 484 (noting that “[r]even ues from both [box suites and VIP packag es] are in some ins tances already bei ng reported in the base”). 19 BMI and NACPA may freely agree to include VIP revenues in the gross revenue base. But ev en on the assum ptio n tha t the d istric t court was correct to theoriz e about an ideal reve nue base that reflects the princ iple of “[w] hat the consumer pays to attend the concert, ” BMI, 19 It is unclea r whether NACPA issued the inst ruction that BMI describes. BMI identifies an email dated May 18, 2016, in which the sender — presumably a representative of NACPA — states that “if a fan has no choice and must pay for both the concert ti cket + the cos t to acce ss an other functi on, the bundled revenue is par t of the ‘total concert experie nce’ and should be considered for license fee calculations.” S upp. App’x 419 (emphasis i n original). Th e email im plies that if the f an ma y cho ose between a stan dard ticket and a VIP tic ket, t hen the extra cost of the VIP ticke t would not be included in the l icense fee calculatio ns. T he same email explains tha t “[t]he spiri t of our agree ments [with B MI and ASCAP ] is for li cense fees to be paid on the dol lar figure your offic e settles w ith the a ct for the ir m usic a l performance.” Id. at 418. Rev enues fr om bo x suite s and VIP pack ages ar e generally pa id to the venue rathe r than to the arti st.

35 664 F. Supp. 3d at 479, the in clusion of VIP reve nues vio lates that princ iple. NACPA iden tifie s another proble m with the inclusion of box suite revenues: box suites are “sold mos tly through an nual co ntracts for all events hosted at a venue, in cluding sporting events.” Appellant’s Reply B r. 3 5. The d istri ct court purported to include only those box suite rev enues that are “a ttribut able to live concer ts.” BMI, 664 F. Supp. 3d at 485. But NACPA objec ts that “there is no way o f knowing how much of [the annual box suite] revenue is attribu table … to live music events, let a lone to a particular con cert.” Appellant’s Br. 55 - 56. BMI r e sponds that th e trial evidence “ s ho we d that box suites can be sold for indiv idual concerts.” Appellee’s Br. 58. Even if that is possible, howeve r, the t r i al evidence e stablished tha t most box s uite sales are made on a n annual basis. See, e.g., App’x 203 (testimon y of the CEO of AEG tha t “out o f 300 sui tes, probab ly 294 of them are on ann ual contracts”). And the venues rather than the promote rs appare ntly contro l whether b ox suites are sold on an individual or annual basis. The impor tant p oint here is that th e distr ict cour t did n ot eve n address the p otential administrative difficulties in disaggrega ting live concert revenue from annua lized box suite sales. And a rate term cannot be rea sonable if t he d istrict court considered only its correspondenc e to an abstrac t principle withou t regard to whether its imple men tat ion wo u ld be a dmin ist rat ive ly f easib le and e cono mica lly ration al. C NACPA argue s that the expanded revenue base itse lf i s “commercially impracticable.” Appe llant’s Br. 55. Accord ing to NACPA, “[u]nre butted testimony at trial made c lear that N ACPA

36 promoters do not h ave access to t he revenue data that would be necessary to apply all of th ese expanded revenue c ategories t o the rate period at issue.” Id. The distr ict cour t addressed thi s prob lem i n a singl e sentence: “Limitin g the revenues to those recei ved by the promoters or a c ontr actually related third - par ty helps to alleviate those conc erns, even i f it do es not produce perf ectly efficient administration.” BMI, 664 F. Supp. 3d a t 485. That conclusion indicates t hat th e distr ict cou rt sou ght to conform the revenu e base definition to an abstract not ion of what properly c ounts as revenue regard less of whethe r that definiti on would reduc e the surplus value avail ab le to the parties. But there is no sin gle c orrect definition of the revenue base; the primary task of the distr ic t cour t is to re asonably appr oximate the licensing agreement that would result from a n arm’s - length tr ansa ction between a willing seller and bu yer in which the ex ertion of “market levera ge” is “moderat[ed]” to avoid “u nacceptably inf lated price levels.” ABC/CBS, 83 1 F. Supp. at 145. As ex plained above, it would not be ration al for the parti es to agree to incu r a dministrati ve costs that have no corres ponding bene fit, so an agreemen t that i mposes such costs cannot be reasonable. Even on its own terms, moreover, the statement of the dis tri ct court is un responsive to the concern of commercial impr act ica bility. NACPA e xpla ins tha t its promoters do not have acces s to data about revenues recei ved b y contractu ally related thi rd parties rather than by NACPA pro moters. “ The fact tha t the no n - promoter enti ties receivi ng such revenues may hav e a ‘contractual rel a tionship’ with promoters does not solve the promoters ’ absence -of- data problem when the pro moters are not a ctual l y receiving the re venues.” Appellant’s Reply Br. 36. BMI’s econ omic expe rt “opine d that promote rs could include rev enue reporting obligations i n their

37 contracts” to “al leviat[e] the practic al difficulty.” Appellee’s Br. 33 n.8. But that means the new revenue base definition requires the promoters not only to modify indus try - standard contracts but also to ensure that venues compl y with the new reporting obligations and to verify the accuracy of the reven ue figures t he venues r eport. It would not be ratio nal for the parties to agree to a licensing agreement that imposed such extensive costs as t he rewriting of thir d - party contracts and the est ablishment of a new data co llection bureaucracy. It is therefore un reasonab le for a distr ict c ourt to impose it. We conclude that the district c ourt erred by expanding the revenue base. Fi rst, the distr ict cour t did not jus tify its unprecedent ed departure f rom the indust ry - standard de finition of “gross r evenue,” and the recor d indicates that no justifica tion is available. Se cond, the distr ict court prioritized adher ence to an abstr act principle of “[w]hat the consumer p ays t o attend the concert ” but its def inition d id not even conform to that princip le. Third, the dist ric t court did not merely fail to addres s the ob jection that the e xpanded r evenue base would be comme rcia lly i mpra ctic ab le but even acknowledged th at it was imposing inefficiencies without an econo mic justification. III The distr ict court further erred in the benchmark analysis that led it to adopt a ra te of 0.5 percent. Firs t, the distr ict co urt fa iled to assign weights to the b enchmarks and implicitly accord ed greater weight to t he SESA C and GMR benchmarks than to the BMI and ASCAP benchmarks. Second, the d istr ict court erroneously considered license agreements wit h non - NACPA promoter s. Third, the d istr ict co urt failed to identify any significant changes in the l ive concert industry since the BMI and ASCAP benchmark agre ements were negotiated tha t would jus tify a highe r rate.

38 A The distr ict cour t “ must determine the degree of comparability” be tween the parties that negotiated each of the benchmark agreement s, on the one hand, and th e parties b efore the court, on the othe r. Music Cho ice IV, 426 F.3d at 95 (internal quotation mark s o mitted). Th at co mparability an alysis will determine the weight it assigns to each benchmark agree ment. And the distr ict cour t may th ereby “explai n how it reached a p articular rate suff iciently to permit our review of the r ate for reasonabl eness.” Id. at 99. In this case, t he dist rict court d id not indicate th e r ela tive weight s it a ss igne d to the benchmark agreement s it con side re d; nor did it expla in how it derived the rate of 0.5 percent from those benchmarks. It listed the seven diff erent impli ed rates re flecte d i n the twelv e bench marks it co nsidered: 0.21 percen t, 0.23 percent, 0.34 percent, 0.36 per cent, 0.37 per cent, 0.51 per cent, and 0.54 percent. And it “therefore” decided that a rate of 0.5 percent wou ld be “ a reasonable one.” BM I, 664 F. Supp. 3d a t 488. An average of the implied r ates for the twelv e benchmark agreement s yields a rate of 0.4 4 percent. An average of the seven different percentage rates represented am ong t he twelve benchmarks yields a rate of 0.37 percent. The BMI and ASCAP be nchmarks all had implied rates betwe en 0.23 and 0.3 7 percent. T o arr ive at a r ate of 0.5 percent based on the benchmarks, t he distr ict co urt must have implic it ly ass ign ed greater weig ht to the benchmarks wit h higher implie d r ates: the SESAC and GMR benc hmarks. If so, that was un reasonable. T he dis tr ict co urt correctl y recognized that “fairly negotiated prio r agreements are t he proper starti ng point from w hich to deter mine reas onable fees for subsequent periods. ” Id. at 486 (quoti ng United S tates v. A SCAP, 157

39 F.R.D. 173, 195 (S.D.N.Y. 1994)). And it correctl y de termined that “ASCAP is indisput ably BMI’s closest comparat or” and that “[t]he rights in qu estion in the ASCAP agreement are essenti ally the same as those NACPA see ks to license from BMI. ” Id. In the agreements “[f]or those si milar ri ghts, NAC PA has p aid ASC AP and BMI at near parity in prior agreements. ” Id. Based on these conc lusions, one would expect the district c ourt to ass ign greater weight t o th e BMI and ASCAP bench marks. The distr ict cour t, however, expla ined that “unconstrained by a consent decree, GMR and SESA C have been able to preserve the legal monopoly powe r granted by the Copyright Act that rate - setting is intended to incorp orate” and th at “[t] he analogous m arket in which GMR and SESAC oper ate thus reflects the level of compe tition that would be inherent in a direct licensing ne gotiation.” Id. at 487. Those considerations—an d t he a pparent decisi on of the distr ict court to a ssign grea ter weight to the SESAC and GMR be nchmarks — suggest that it adopted the argumen t that the se ben chmarks “provi de the best evidence of the price that a willing licensee and a willing licensor would agree to in a n arm’s - length transaction in a compet itive free market” because “ SESAC and G MR have no antitrust consent decre e mandating th at tho se PRO s grant a l icense to music users upo n request.” Brief of Amicus Curi ae A SCAP 4. This argu ment imp lies that — far from ser ving “to minimize th e likeli ho od” that the “evi dent market leverage” of t he regulated PROs will pr od uce “ una cc eptab ly in fla ted pr ice le ve ls” — the consent decrees have instead kept BM I’s and ASCAP’s r ates artificia lly low. Showtime II, 912 F.2d at 57 6. But we have said that “rate- se tting cour ts must take ser iously t he fact that the y exis t as a res ult of mono polists exercisin g disproportio nate power over the market for music rights.” Music Choice IV, 426 F.3d at 96; see also Showtime II, 912 F.2d at 5 70

40 (“Though the rate court’s ex istence does not mean tha t ASCAP has violated the antitrus t law, the court need not conduct itself without regard to the context in which it was created.”). As long as the consent decrees remai n in place, it is not with in t he d iscre tion o f the distr ic t court to decide that BMI ’s market power no lo nger warrants t he “moderating influe nce” of the rate - setti ng framework. ABC/CBS, 831 F. Supp. at 1 45. The distric t court s ugge sted that the market sh ares of SESAC and GMR “ar e more comparable to t hose of the large music publishers that music users would have to negotiate with directly in the absence of PROs.” BMI, 664 F. Supp. 3d at 487. But a p ublish er with a catalog ue s imilar in size to those of SESAC an d GMR would not necessarily char ge a compe titive pr ice for a b lanket license. NACPA claims that “S ESAC and GM R have repe atedly faced claims that they h ave extra cted supra competitive fees in v iolation of th e antitrust laws.” App ellant’s Br. 40; see als o id. at 11 & n.2. The dist ric t court decid ed that the “realiti es” of extensive “ negoti ations be tween SESAC, GMR, and promoters can not be reconciled with NAC PA ’ s claim th at the PROs a re extr acting rates that are anti - co mpetitive. ” BMI, 664 F. Supp. 3d at 4 88. The fact that a PRO engaged in exte nsive nego tiatio ns, however, does not show that it lacks market po wer. A negotiation might be gin with a deman d for a higher supr acompetitive price and resul t in a lower bu t still supracompet itive price. 20 The district court sa id that “[t]he fact that promoters e lected to b uy a blanket license for their own business goals and convenience does not 20 M arket p ower does n ot de termine bargaining power. See S chwar tz & Scott, supr a note 17, at 553 (explaining that “[i] n stan dard bargaini ng theory, bargaining power is a function ” o f “ the pa rties’ re lative patie nce ” and “ each party’s d isag reement poi nt (or next - best option) ”). And NACPA has its own market power.

41 establish that they di d so under com pulsion.” BMI, 664 F. S upp. 3d at 488. But a lack of comp uls ion does no t e stablis h tha t the selle r is not charging a supracompetitive price. A monopolist can raise its price as long as the alterna tive to its product is sufficiently costly or inconvenient. 21 Other distr ict courts have noted that “[t] he SESAC license has histori c ally been a benchma rk of limited val ue because the pu blic knows little abou t the size of the S ESAC r epertoire ” and therefore “ it is difficult to adjust a SESAC license rate to arrive with confide nce at an implied AS CAP rate. ” In re Pandora M edia, Inc., 6 F. Supp. 3 d 317, 362 (S.D.N.Y. 2014). 22 Moreover, “ SES AC’ s s mall s ize, wh en compared to ASCAP and BM I, not only amp lifies any error in a projecti on, but also reduces t he incentive to resi st SESAC ’ s rate requests ” because “the cost associated with resistance may not be justifi ed when a license fee is relati ve ly small.” Pandora Media, 6 F. Supp. 3d at 3 62. 23 21 Cf. Appella nt’s Br. 39 - 40 (“ [W]hile their r epertories ar e small, their lice nses are still ‘mu st haves,’ m eanin g concert prom oters have no op tion to walk a way fr om any nego tiation with SESAC and GMR wi thout taking a license.”); BMI, 664 F. Supp. 3d at 488 (“[P] romoters opted to not ask performe rs fo r their song l ists before th e show bec ause doi ng so, and the n direct licensing, is, in t he words of Mr. M arciano, ‘ a lot of admi nistrati ve work. N ot impossib le, but it ’ s a lot of admin istrative wo rk. ’ Inst ead promote r s ch ose t o negoti ate h eavily with GMR and SE SAC.”) (citation omitted). 22 In thi s case, NACPA’ s economic expert explained that a differ ence of four pe rcentage points in es timati ng GMR’ s market sha re dou bled the “implied royalt y rate for BMI.” Affidavit of Adam B. Jaffe ¶ 187, BMI v. N ACPA, No. 18- CV -8749 (S.D.N.Y. Oct. 7, 2022), ECF No. 118. 23 See Brief o f Amic i Curiae Moti on Pic ture Assoc iation, I nc., et al. 8 (argui ng that th e “imp act” of SESAC ’s and G M R’s “ supra - compet itive

42 In ligh t of these considerati ons, it was unr easonable for the distr ict court to assig n greater wei ght to t h e SESAC and GMR l icenses than to the prior BMI and ASCA P agreements. B The distr ict cour t addit iona lly erred by treatin g agreements with promoters un affiliated with NACPA as comparable benchmarks. The dis tric t court reasoned that “[c]ompare d to th e BM I - NACPA license,” the agre ements betwee n BMI or ASCAP and non - NACPA promot ers “arose in a market t hat reflects the same degree of competition and economic circumstances, and they cover the same rights.” BMI, 664 F. Supp. 3d at 486. The distr ict court adde d that “[a]lthough NAC PA and non - N ACPA pr omoters differ in the scope of their business op erations, the y are direct compet itors for sh ows hosted in smaller ve nues.” Id. The dis tric t court did not expressly address the licenses between SE SAC or GMR and non - NAC PA promoters, but BMI main tains tha t the distr ict c ourt im plic itly applied the same reasoning to th o se licenses. See Appellee’s Br. 54. That re asoning was inade quate to establ ish comparabi lity. The ASCAP/ NACPA and ASCAP / non- NACPA licenses c overed approximat el y the same peri od —2020-20 2 1 and 2021, respec tively — but the implied r ates differ ed by a s ignificant amoun t: 0.14 percent. See BMI, 664 F. Supp. 3d at 482. Simi larly, the SESAC / NACPA and SESAC/non- NACPA licenses were issu ed around th e same time — covering 2019 -20 24 and 2021, res pectively — but the rates differ ed b y licensing pract ices o n lice nsees ha s bee n cabi ned befo re t he deci sion bel ow, in large part because (a) the actual prices, while inflated, are not so high as to be rui nous to l icens ees given the comparati vely s maller rep ertories involved; and (b) no rate court un til now had reli ed on SESAC or GMR ra tes in se tting rates for the m uch larger BMI an d ASCAP repe rtories ”).

43 0.17 percent. Id. The difference in t he rates indicate s that th e non - NACPA promot ers are not simil arly situa t ed to NAC PA whe n negotia ting w ith PR Os. 24 NACPA explains that BMI has soug ht to “realign” NACP A and non - NAC PA pro moter rates “si nce 2013” b ut “has not be en able to achieve that re sult throu gh arm’s - length negotiations.” Appe llant’s Reply Br. 31. It is no t sur pris in g that NACPA — which negotiates on behalf of a large group of promoter s — has more leverage than othe r promoters in negotiations with PROs. “NACPA is itself an aggregator” an d “represe nts the interests of the largest promoters.” Brief of Amicus Curiae National Music Publishers’ Associ a tion 8. T hat di fference makes t he agreements with non-NACPA promoters inappropriate benchmarks. The parties do not addres s whether, if agreements with non - NACPA promoters a re invalid benchmarks, GMR’s agreements wi th Live Nation and AEG remain valid benchmarks. Li ve Nation, AEG, and their affiliates compose about 75 perce nt of NACPA’s membership and pay about 90 percent of NACPA’s licens ing fees under its agreement with BM I. According to one amic us, “[f]or all practical purposes, N ACPA is Live Nation and AEG.” Id. at 14. All of the GMR ag reements that the di strict court consi dered — including the agreements with Live Nation and A EG — have the same rat es a nd rate structu re, which may s uggest that the agreements are non - nego tiated form licenses. Yet wi tnesses from Live Nation and AE G testified that the GMR li censes were heavily negotiate d. 25 Roux origin ally 24 See also Appellant’s Reply Br. 30 -31 (noting t hat in 2009 “BMI negotiated higher ra tes for n on - N ACPA promoters compared to the preva iling ra tes for NACP A members” a nd that this “ diff er ential trea tment unde rscores that NAC PA and non - NACPA prom oters a re not similarly situated”). 25 The president of U.S. concerts at Liv e Nation, Bob Ro ux, testif ied tha t he “actively neg otiated Live Nat ion’s license with GMR” and “exchange d

44 recommen ded that Live Na tion negotiate a li cense agreement with GMR “through NACPA” an d agree to “GMR[’ s] or iginal as k” with respect to the percentage rate. Trial Transcr ipt (10/31/22) a t 497. On the one hand, the tes timon y sugge sts that the licenses between G M R and Live Nation and AEG re flected a n egotia ted rate and that the rates would have been simil ar if the lic enses had been negotia ted through NACPA. On the other hand, NAC PA nego tiate d a signif ican tly lowe r ra te w ith SE SA C — a PR O of similar size to GMR — than the non-N ACPA promo ters obtained. We leave th is question for the dis trict co urt to address in the first instance. The district court sho uld determin e whether the rat e and rate structure in the GMR license s with Live Nation and AEG reflect what NACP A would have achieved in a ne gotiation with GMR. If NACPA w ould have negotia te d a s ignif ican tly lo w er r ate with G MR, then the two agree ments would not be co mparab le benchmarks. If the rate and rate struc ture in the two li censes reflect what NACPA wo uld have obtained in a negotiation with GMR, then the agreem ents would be com parable benchmarks. This exercise i s admit ted ly hypothetical. A pparent ly, GMR w as un willi ng to negotiate with NAC PA be cause of NACP A’s 6 percent admi nistrative discount. See Agreed Findings of Fact ¶ 62 (“GM R does not license with NACPA. G MR r efused to negoti ate with NAC PA and instead engaged in direct negotiations with ind ividual promo ters.”). But drafts” and “had dozens of phone cal ls” with GMR representatives. Trial Transcript (10/31/22) at 505. Roux said that the resulting deal “was reached after ma ny months of negoti ations” and “ was fair” to both GMR and Live Nation. Id. at 506-07. T he chairman and CEO of AEG, Jay Mar ciano, testified that “AE G Pres ents tr[ied] to negotiate a b etter deal than th e one it eventually signe d with GMR.” Trial Transcript (11/2/22) at 840 -41, BMI v. N ACPA, No. 18- CV -8749 (S.D.N.Y. Dec. 23, 2022), ECF No. 186-87.

45 projectin g the o utco me of a hypo the tic al n egotia tion is the purpose of the r ate - setting proceeding. See Showtime II, 912 F.2d at 5 69 (“Fair market value is a fa ctual matter, albeit a hypothe tical one.”). Assuming that the G MR licen ses with Live Nation and AE G a r e appropriate be nchmarks, the GMR l icenses with non - NACPA promote rs remain inappropriate even if the licenses include the same rate. I n any event, because all six GMR lic en ses cover a simila r p erio d and contain ident ical q uantitative ter ms, it is unnecessary to include all six as ben chmarks. D oing so wo uld ri sk introducing bias into the rate determinat ion. Even the two p ossible GMR b enchmar ks — the agreements with Live N ation and AE G— deserve less weight re lative to the BMI and AS CAP ben chmarks. C The distric t court d id not identify any relevant changes i n economic circumstances that justify its depart ure from the prior rates negotiated by N ACPA w ith BMI an d ASCAP. T he dis tric t court gestured t o ward changes in the live concert in dustry because of the emergence of Live N ation and AE G, but it did no t explain wh y tho se changes justified a higher rate. Moreover, the BMI and A SCAP benchmarks mostly postdate the emerge nce of Live Nation and AEG. T he distr ict court expre ssly found that “[t]he re is no eviden ce in the record that the economic circumstances in 2018 ” when the ASCAP/NACPA license was negoti ated “are signific an tly d iffe ren t from the present d ay.” BMI, 66 4 F. Supp. 3d at 486. On appeal, BM I does not offer a per suasive reas on why th e emergence of L ive Nation and A EG as dominant playe rs in the concert pr omotion i ndustry wo uld just ify a departure from the histor ica l rate. T he argumen t is that “[f ]or all pract ical purposes, NACPA is Live N ation and AEG, an d those entities all have

46 substantial market a nd negotiating powe r.” Brief of Amicus C uriae Nation al M usic Pub lish ers ’ Association 14. But as the b arg ainin g power of BMI ’s counterpart y increas es, the counterp arty sho uld be able to ins is t on a lower rate in an arm’s - length nego tiation. In fact, it appears that the e mer gence of Live Na tion and A EG did not e na ble NACPA to ob tain a lower rate in its last negotiation w ith ASC AP. BMI’s ec onomic expert testi f ied that t he 2018 ASC AP/NACPA l icense reflected “quite a lar ge increase” from pr ior years. S upp. App’x 185. 26 The recor d does not establish that th e consolidati on of the live conce rt industry afforded NACPA additional bargainin g power. It ma y be that, even befor e the emergence of Live Nation and AE G, NAC PA already represented more or l ess all concert promoters, so BMI and ASCAP already had to negotiat e rates with a monop sonist. BMI argues that — in 2018 when the AS CAP/NACPA license was nego tiate d — ASC AP knew “ that BMI w as likely to comm ence a rate proceeding agai nst NACPA. Acc ordingly, ASCA P negotiated for an early terminati on right. B y securing th is right, ASCA P positioned itself to b ene fit if B MI wa s s ucce ssf ul in li tiga tion, w ithou t incu rrin g the sub stantial cost s that BMI would in cur.” A ppe llee ’s Br. 23 (cita tions omit ted). The implicat ion is that ASCAP did not neces sarily believe in 2018 that t he rate to wh ich it agreed represen ted a fair rate. The district c our t did not adopt this theory; to the con trary, it de cided that “[t] he ASCAP - NACPA license is a valid benchmar k.” BMI, 664 F. Supp. 3d at 486. Moreover, BMI conceded bef ore the di strict court that the 2018 AS CAP / NACPA rate reflect ed “ what the parties 26 Accor ding t o NA CPA, “AS CAP’s n ew rat e refle cted a 21% in cre ase ov er the prior eff ective rate for the 2 018 -20 19 peri od and an additional 19.5% increase for the 2020 -20 21 per iod — increas es that transl ate into several millions of additional dollar s paid by NACPA to ASCAP.” Appellant’s Reply Br. 10 n.2.

47 thought Judge C ote would do in a rate court proceed ing.” 27 Th a t concession conflicts with the contention that ASCAP expected BMI to obtain a higher rate in it s rate - settin g pr oceeding. Moreover, the argument does not make sense from an economic perspective. If ASCAP be lieved tha t the rate in the 2018 ASCAP/NACPA a gree ment was unfairly low and that BM I would obtain a higher rate in i ts rate - settin g proceeding, then ASC AP woul d have insisted on —and like ly wo uld hav e obtained — a higher rate in its o wn ne got iatio n. That outcome n ot only wou ld have allowe d ASCAP to collect more revenue; it also would have increase d the like lihoo d that BMI would obtain a highe r rate in the rate - setting proceedin g by providing a higher bench mark rate. The dis trict co urt s uggested that “[i]ncreasing the r ate of a licens e for live concerts bett er reflects the fair market valu e placed on licens es in mus ic int en sive in dust rie s.” B MI, 664 F. Supp. 3d at 488. The distr ict co urt e xplained that “[l]i cen ses in mus ic in tens ive industries, like commer cial radio statio ns or virtual live c oncert streami n g services, have hi gher rates because m usic is at the heart of the product bei ng offered by the busines s.” Id. By contras t, “in a less- music int ensiv e in dus try, like television s porting even ts or talk radio, the music is supplemental to the consumers’ primary purpose and thus the rate owed to BMI’s aff iliates for the ir contribu tions is smaller.” Id. at 488 - 89. Because “it is indisput able that music is essential to a live concert,” the d istr ict court concluded that “[t]hose who contribute to th e musical compositio ns should be compe nsated accordingly.” I d. at 4 89. Yet the commerci al conditions in the “music intens ive indus tr ies ” canno t be compa rable to the l ive concert 27 Post Trial Memorandum at 55, BMI v. N ACP A, No. 18- CV -8749 (S.D.N.Y. Jan. 1, 2023), ECF No. 209.

48 indust ry because music user s such as r adio stat ions and str eaming services pay r ates of up to 2.5 percent — fa r higher than what even the regulated PROs are able to charge for perform ance licenses. 28 Based on the cons iderations we ha ve ident ified, the comparab le benchmark agreements are the BMI/N ACPA license, the ASCAP/NACPA license, the SESAC /NACPA license, and p ossibly th e GM R/Liv e Nation and GM R/AEG li cense s. The p r imar y rates for these agreem ents are 0.21 percent, 0.2 7 pe rcent, 0. 40 percent, and 0.63 percent, respectivel y. A n average of these four rates yields a rate of 0.38 percent. If th e BMI and ASCAP be nchmarks were to receive double t he weight of the SESAC and GMR bench marks, th en the res ultin g rate would be 0.33 percent. Accordingly, a we ll - supported rat e for t he new BMI/NAC PA licens e would be significantly lower than th e rate imposed by the distr ict co urt, but it may be h igher than the rate for the pre exis ting BMI/NACPA licens e. To arrive a t such a rate, the d ist ric t court “ mus t determine the degr ee of comparability ” of the benchmar ks and “explain how it reach ed a partic ular rate sufficiently to perm it our review of the rate for reasonabl eness.” Music Choice IV, 426 F.3d at 95, 99 (inter na l quo tatio n ma rks o mit ted). We re mand for the dis tric t court to do s o. IV The dis tric t court adopted B MI’s rate quo te for the Retr oactive Period that extends from Jan uary 1, 201 4, to June 30, 2018. The parties 28 See Trial Transcript (10/25/22) at 154-55, BMI v. N ACPA, No. 18- CV -8749 (S.D.N.Y. Dec. 23, 2022), ECF No. 176; see also Pandora Media, 6 F. Supp. 3d at 372 (“ The headline rate for the ASCAP - Pandora license for the years 2011 through 2 015 is set at 1.85% of revenue for ev ery year of the l icen se term. ”).

49 do not devote as muc h arg umen t to that d ec ision on appeal, presumably b ecause the parties’ prop osed rates do not differ much. NACPA’s eco nomic expert estimat ed that BMI’s rate quote for the Retroactive Period implied a unitar y rate of 0.28 percent, an d NACPA itself s ou ght a rate of 0.23 per cent. See App’x 372. In additi on, the distr ict court lef t the revenue base for t he Retroactive P eriod “defi ned to include on ly the face value of the tick et (exclusive of all fe es and other charges).” B MI, 664 F. Supp. 3 d at 477. Th e distr i ct court reasoned that BMI’ s rate quote fo r the Retroactive Period was reasonable because it “would alig n the NACPA rates with the rates paid to BMI by non - NAC PA promoters during tha t time,” and the distr ict cour t had “alre ady found that the BMI -n onN ACPA benchmark is val id.” Id. at 489; see also Appellee’s Br. 66 (“For over a decade, BMI licensed N ACPA and non - N ACPA pro moter s at the same rate s. BMI’s rate quote for the Retroact ive Period retu rns that align ment.”) (citatio n omitted). As ex pla ined ab ove, the d istr ict cour t unreas onably relied on licenses with promot ers unaffiliated with NACPA. The comparable benchmarks ind ica te that an arm’s- len gth ne got iatio n wo uld no t result in an alignmen t of NACP A rates with those negotiated b y non- NACPA promoters. And it is unsur prising th at NACPA, which represents ma ny concert promoters, would have the market power to negoti ate a lower rate. It was unr easonab le for the dis tric t c ourt to adopt a rate quote for the Retroactive Pe riod on the ground that it would align N ACPA rates with non-NAC PA rates. V BMI argues in its cross - appeal that the distr ict court erred by denying its moti on for pr ejudgment intere st. We h ave held t hat “ [t] he decision to award prejudgment i n terest is governed b y the equi ties,

50 reflecting ‘considera tions of fairness’ rathe r than ‘a rigid theory of compensation.’” SE C v. Contorinis, 743 F.3d 296, 308 (2d Ci r. 2014) (quoting Bla u v. Lehman, 368 U.S. 403, 414 (1962)). The decis ion to award prejudg ment interest “shoul d be a function of (i) the need to fully compensate the wronge d party fo r actual damages s uffered, (ii) con siderations of fairness and the relative equ ities of the awar d, (iii) the remedi al purpos e of the stat u t e involved, an d/or (iv) such other general pri nc iples as are deemed relev ant by the court.” Wickham Contracting Co. v. Local Unio n No. 3, 955 F.2 d 831, 83 3 - 34 (2d Cir. 1992). “ A wards of prejudgment int erest must not resu lt in over - compensation of the plaintiff,” ho wever, and are disfavore d when “the statute itself fix es damages deeme d fully compensator y as a matter of law.” Id. at 834. The BMI C onsent D ecr ee provides that during the pendency of any rate negotiat ion or rate - se tting proce eding, “the appl icant shall have the right to use any, so me or all of the compositions in [BMI’s] repertory … without payment o f any fe e or other c ompens ation.” BMI Consent Decree § XIV(A). However, “either the applica nt or [BMI] may apply to [the d istric t court] to fix an inter im fee pending final determinat ion of what constit utes a reason able fee.” Id. § XIV(B). If the d ist ric t court fixes an interim f ee, BM I mus t th en issue a lic ense “providing for the pa yment of a fee at such interim rate from t he date the applicant requested a license.” Id. Once the distr ict court make s its decisi on, “the reasonable fee finally determined by [t he dist ric t court] shall be r etroactive to the date the ap plicant re quested a licens e.” Id. The consent decree does n ot specify whether BM I is entitled to prejudgment interest if the reason able fee the dis tr ict court ultim atel y se t s is higher than the int erim fee. Rather than petitioni ng the dis tric t co urt to fix an interim fee, the parties in this case agreed in 2014 to a n interim fee of 0.3 percent

51 of gross revenue — defined as the face value of ticke ts sold —for concerts held in ven ues seating unde r 10,000 and to an interim fee of 0.15 percent for concerts held in venues s eating over 10,000. S ee BMI, 674 F. Supp. 3d at 72. The 2014 agr eement does not specify whe ther BMI would be entitled to prej udgment int erest if t he reasonabl e fee ultim atel y exceeds the interim fee. The dis tric t cour t d ecided that even th ough the parties had negotia te d an interim fee agreement rather than p etition ed t o fix an interim f ee, there was “no rea son to ignore the relat i onship between the int erim fee and final f ee as outlined in the Con sent Decree.” Id. at 73. It exp lained that “Ar ticle XIV clea rl y states that the in terim fee controls until the final fee is deter mined and can be retroa ctively applie d. T hat calls fo r a simp le su bs titu tio n of the inter im r ate by th e new rate, no t a new rate plus interest, as though the new higher rate had been known sin ce back then.” Id. In other words, “[t]he in terim fee rate is correct for the time it is in effect: it is not a down payment on some larger amo unt.” Id. The distr ict cour t concluded t hat prejudgment interest was no t warranted beca use pr ejudgment interest compensates a plaintiff for the lost ti me value o f money a defendant wr ongful ly reta in s, and i t was not wrongful for NACPA to pay onl y the interim fee un til the district court deter mined a re asonable fee. Cf. Water side Ocean Nav. Co. v. Int’l Nav. Lt d., 737 F.2d 150, 154 (2d Cir. 1984) (ex pla inin g t hat an “award of pre - judgment interest” all o ws “a person wrongf ully deprived of his mon ey [to] be made whole for the loss”). The dis tric t court clarif ied that its “d eter mination of the f inal fee a mou nt was designed to be th e entire embod iment of what const itutes a reasonable fee, not merely some fig ure to which add itional embelli shments, such as i nterest, could be attached. ” BMI, 674 F. Supp. 3d at 73. For that reason, “[a]n award of prej udgment

52 interest at this juncture would overcompensate BMI at NACPA ’s expense.” Id. We conclude that the distr ict court di d not abuse its discret ion when it denied the motion for preju dgment interest. Th at the “reasonable fee” the distr ict court u ltim ate ly sets is higher than the interim f ee does not necessarily mean that the interim fee was unreasonable; there may be a range of re asonable fee s. I f the interim fee was reasonab le, then n o money has bee n wrongfully w ithheld from BMI, and it h as not suffered an injury that r equires compensat ion through prejudgment interest. The distri ct court did not expressly state t hat the interim fee was rea sonable in ret r ospect, but it did e xplain that its final fee determinat ion was the “entire embodiment” of a reas onable fee. In other words, the dis trict court determi ned that the final fee, app lied retroactive ly, provided BMI with reasonable compensat ion for NAC PA’s use of the licensed performance rights during the pendenc y of the negot iation and the rate - se tting proceedi ng without the add ition of prejudgment interest. The explanation of t he district court made clear that it denied prejudgment interest because eith er (1) the interim fee w as reasonable in retrosp ect, or (2) some componen t of the f inal “reasonab le fee” would compens ate BMI for the ti me value of money. Either wa y, the addition of prejudgment i nterest would ov ercompensate BMI. Those reasons provided a s ufficient justifica tion for the denia l. We do not ho ld th at the c onsent d ecree requires the distric t court to follow the ap proach of retroa ctively subst ituting the final fee for the i nterim fee. District courts have awarded preju dg ment interest in other cases. See, e.g., ABC/CB S, 831 F. S upp. at 166. But th e distric t court did not abuse its discretion here.

53 Because we vacate the judgment base d on the unreasonableness of the rates the distri ct court adopted, however, the distri c t court may reconsider w hether to award prej udgment interest on remand a f ter it arrives at a final fee determination. CONCLUSION There is no sin gle cor rect way to “est ablish[] the fa ir market value” of a blanket license to the re pertory of a P RO. Music Ch oice IV, 426 F.3d at 95. Nevertheles s, the dis tr ict court mus t exerci se “a moderating influenc e” that avoids “unacce ptably inflated price levels” and “acco unt[s] for altera tions in the eco nomic c onditions confronting the partie s.” ABC/CBS, 831 F. S upp. at 1 45. It must s et a fee that is “r easonable” and “based upon all the evidence.” BMI Consent Decree § XIV(A). To do so, t he di stric t cour t will r e ly on prior agreement s negotiated by the same or similarly sit uated parties unless changed eco nomic circumstances compe llingly sho w th at those agreem ents are obsolete. And th e distric t court will “explain how it reached a particular r ate sufficient ly to perm it our review of the rate for reasonab leness.” Music Cho ice IV, 426 F. 3d at 99. In this case, the distri ct court adopted a reven ue base definition that had no pr ecedent in the history of the industry without a compelling reason. It included revenue st r eams that do not refl ect the fair market val ue of the mus ic an d that involve sig nifica nt administrative cos ts without a corresponding bene fit. The d istric t court did not express ly assign weigh ts t o the benchmark agreement s it considered bu t implic itly accorded great er weight to less comparable benchmarks. It relied on agreements wit h unaf fil iated individual prom oters even tho ugh NAC PA ha s h isto rica lly o btained signif ican tly lo wer ra tes fr om th e same counterparties. And it ident ified no change in econom ic circumstances tha t would justify a

54 rate more than doubl e what NACPA has h istorically pa id to BMI and ASCAP. Under these circumstances, we conclude that the rates the district court ad opted were unreasonab le. The d istric t co urt did n ot abuse it s discretio n by denying the m o tion for prejud g ment interest, but the denia l may have depe nded on its p rior determination of the rates. We vacate the judgment of the district court an d remand for further proceedings consis tent with th is op inion.