Contributions

The Status and Viability of the Efficiencies Defense in Antitrust Law

by Isaac Wein­gram  1

The “effi­cien­cies” or “economies” defense pro­vides a vehi­cle for merg­ing com­pa­nies to con­test alle­ga­tions that a merg­er will result in anti­com­pet­i­tive effects, which will harm the mar­ket and con­sumers. 2 The prin­ci­pal con­cern of courts with merg­ers is that increas­es in mar­ket con­cen­tra­tion will allow busi­ness­es to use their mar­ket pow­er to restrict out­put and increase prices. 3 An economies defense would allow firms to show that the merg­er would not have these neg­a­tive effects. Instead, firms would have the oppor­tu­ni­ty to show that a merg­er would allow the com­pa­nies to reduce pro­duc­tion costs or pro­vide con­sumers with new goods and ser­vices through inno­va­tion. 4


Suc­cess­ful­ly employ­ing the defense requires com­pa­nies to show that reduc­tions in pro­duc­tion costs or gains in inno­va­tion from a merg­er will ulti­mate­ly ben­e­fit con­sumers in the form of low­er prices or high­er qual­i­ty goods and ser­vices. 5 How­ev­er, the extent to which courts accept the defense, in addi­tion to its over­all work­a­bil­i­ty, remains in doubt. 6 This arti­cle dis­cuss­es each of these issues in turn. First, I exam­ine how courts cur­rent­ly view the effi­cien­cies defense. Sec­ond, I look at whether the effi­cien­cies defense can be oper­a­tional­ized. Ulti­mate­ly, I will argue that because of the sig­nif­i­cant obsta­cles asso­ci­at­ed with eval­u­at­ing the effi­cien­cies defense, the cir­cuit courts have been wise to cre­ate imped­i­ments for com­pa­nies seek­ing to jus­ti­fy merg­ers in high­ly con­cen­trat­ed mar­kets based on pur­port­ed effi­cien­cies.

*****

The avail­abil­i­ty of an effi­cien­cies defense in antitrust law has been in con­sid­er­able doubt since the Supreme Court’s rul­ing in Fed­er­al Trade Com­mis­sion v. Proc­ter & Gam­ble Co. 7 In Proc­ter, the Supreme Court explic­it­ly stat­ed that “[pos­si­ble] economies can­not be used as a defense to ille­gal­i­ty.” 8 How­ev­er, the rejec­tion of the effi­cien­cies defense was not cen­tral to the Court’s hold­ing, and in his con­cur­ring opin­ion, Jus­tice Har­lan force­ful­ly argued that effi­cien­cies should be con­sid­ered in the merg­er con­text. 9 Since the rul­ing in Proc­ter, the Supreme Court has not revis­it­ed the effi­cien­cies defense. Rec­og­niz­ing the ambi­gu­i­ty of the prece­dent in Proc­ter, the Eleventh Cir­cuit opened the door to the effi­cien­cies defense in Fed­er­al Trade Com­mis­sion v. Uni­ver­si­ty Health, Inc. 10 Four oth­er circuits—the Sixth, the Eighth, the Ninth, and the D.C. Circuits—have fol­lowed suit, with the Ninth Cir­cuit being the most recent addi­tion. 11 The Ninth Circuit’s deci­sion to con­sid­er an effi­cien­cies defense is par­tic­u­lar­ly strik­ing because the cir­cuit had reject­ed such a defense near­ly three decades pri­or. 12

How­ev­er, the rever­sal of the Ninth Cir­cuit and the adop­tion of the effi­cien­cies defense by sev­er­al oth­er cir­cuits obscure the true sta­tus of the defense. Near­ly all the cas­es in which the cir­cuit courts have rec­og­nized the pos­si­bil­i­ty of the effi­cien­cies defense—H.J. Heinz being the notable exception—have involved health­care merg­ers. 13 More impor­tant­ly, none of the cir­cuit court cas­es that claim to rec­og­nize the effi­cien­cies defense have cred­it­ed a defen­dant for suc­cess­ful­ly deploy­ing it. 14 Even in the cir­cuits where it is accept­ed, the exis­tence of the defense may there­fore be more the­o­ret­i­cal than prac­ti­cal.

The demand­ing stan­dards of the defense may explain why it faces so many prac­ti­cal lim­i­ta­tions, and there­fore, why it is so chal­leng­ing to employ. First, the effi­cien­cies defense requires a defen­dant to show that the effi­cien­cy gains are merg­er-spe­cif­ic, such that the iden­ti­fied gains were achiev­able only through the merg­er. 15 The rea­son for this restric­tion is clear: if the mar­ket can achieve the increased effi­cien­cies with­out a decrease in com­pe­ti­tion, then there is no rea­son to sanc­tion the merg­er. 16

Sec­ond, a suc­cess­ful effi­cien­cies defense requires that the defen­dant show that the claimed effi­cien­cy gains are not mere spec­u­la­tion. 17While this restric­tion may make con­sid­er­able sense, it is also one of the most restric­tive. Inno­va­tion effi­cien­cies pro­duce the great­est ben­e­fits for the mar­ket and con­sumers, 18 but those effi­cien­cy gains are often the most dif­fi­cult to mea­sure and most spec­u­la­tive. 19 This restric­tion may fore­stall com­pa­nies from pre­sent­ing evi­dence about those gains most like­ly to pro­duce a favor­able result for them, both in court and for con­sumers in the long term. Prac­ti­cal­ly speak­ing, the restric­tion is even more oner­ous, as courts have rec­og­nized that the evi­den­tiary show­ing will vary with the degree of mar­ket con­cen­tra­tion. 20There is good rea­son for courts to be more cau­tious in high­ly con­cen­trat­ed mar­kets, since merg­ers in already con­cen­trat­ed mar­kets raise the specter of collusion—tacit or otherwise—among com­peti­tors. 21 How­ev­er, the trend of increased judi­cial scruti­ny and analy­sis rais­es ques­tions about the actu­al avail­abil­i­ty of the effi­cien­cies defense in sit­u­a­tions where it would be most use­ful.

Final­ly, and most impor­tant­ly, the effi­cien­cies defense requires proof that the gains will be passed on to con­sumers. 22 This restric­tion is imposed to main­tain the focus of the court’s analy­sis on the pro­tec­tion of con­sumer wel­fare. To allow courts to mere­ly weigh the effi­cien­cy gains of a merg­er against harm to con­sumers would unmoor antitrust law from its pur­pose of pro­mot­ing con­sumer wel­fare. 23 While this may be a nec­es­sary restric­tion on the effi­cien­cies defense, courts have incon­sis­tent­ly defined the tim­ing and the extent to which the effi­cien­cies must be passed on to con­sumers. 24 The ambi­gu­i­ty of this restric­tion com­pounds the dif­fi­cul­ties of employ­ing an effi­cien­cies defense to rebut alle­ga­tions that a merg­er is anti­com­pet­i­tive.

Because of the high bar imposed by these caveats, the effi­cien­cies defense remains more of a the­o­ret­i­cal con­struct than a prac­ti­cal option for defen­dants in merg­er cas­es. Sev­er­al cir­cuits may have embraced an effi­cien­cies defense on its face, but in prac­tice those cir­cuits do not rec­og­nize its valid­i­ty. 25The weak­ness of the defense is com­pound­ed by the Supreme Court’s near­ly fifty-year silence on the issue. When the Court last spoke on the issue, it left con­sid­er­able ambi­gu­i­ty that has plagued the low­er courts. 26 Giv­en this con­text of uncer­tain­ty, the avail­abil­i­ty of the effi­cien­cies defense is extreme­ly sus­pect. How­ev­er, the rel­e­ga­tion of the defense to a the­o­ret­i­cal con­struct may not be prob­lem­at­ic giv­en the sig­nif­i­cant dif­fi­cul­ty that courts face when eval­u­at­ing the defense.

*****

Even if courts are will­ing to adopt an effi­cien­cies defense, there remain ques­tions about the abil­i­ty of courts to suc­cess­ful­ly eval­u­ate the defense. Con­cerns about the capac­i­ty of courts to oper­a­tional­ize the defense have ani­mat­ed the dis­course since its pro­pos­al in the late 1960s. 27 The courts that have adopt­ed the defense have also not­ed these issues and their own skep­ti­cism about their abil­i­ty to assess the defense. 28

Two major con­cerns exist about the via­bil­i­ty of the economies defense. First, there is the issue of whether courts can actu­al­ly per­form the wel­fare analy­sis nec­es­sary to deter­mine whether there is a net gain to con­sumers. 29 Sec­ond, and per­haps more impor­tant­ly, there is a con­cern about infor­ma­tion­al asym­me­tries between those try­ing to enforce antitrust laws—in most cas­es, the government—and the com­pa­nies that are seek­ing to merge. 30 These sig­nif­i­cant lim­i­ta­tions on the abil­i­ty of courts to ade­quate­ly eval­u­ate the defense pro­vide a jus­ti­fi­ca­tion for the strin­gent stan­dards that courts have adopt­ed.

A wel­fare analy­sis in the merg­er con­text is the assess­ment of how a par­tic­u­lar pol­i­cy will affect the over­all wealth in a giv­en mar­ket. 31 Most schol­ars, in assess­ing the wel­fare effects of merg­ers, have relied on the “naïve trade­off mod­el” pro­posed by the econ­o­mist Oliv­er Williamson, 32 who pur­pose­ly select­ed the seem­ing­ly pejo­ra­tive name to empha­size the num­ber of lim­it­ing assump­tions nec­es­sary for the mod­el. 33 As with all mod­els, the naïve trade­off mod­el is use­ful in illus­trat­ing some points, but not oth­ers. More specif­i­cal­ly, there are sig­nif­i­cant con­cerns about using the mod­el as a tool in spe­cif­ic cas­es, as opposed to the heuris­tic that is was designed to be. 34 To effec­tive­ly employ the mod­el in a giv­en merg­er case, the court would need to be able to com­pute both the demand and the sup­ply curves for that mar­ket. 35Not only are these com­pu­ta­tions out­side the ken of courts, they are also unknown to the actu­al mar­ket par­tic­i­pants. 36

The inabil­i­ty of courts to accu­rate­ly per­form wel­fare analy­sis for a giv­en merg­er is com­pound­ed by infor­ma­tion asym­me­try con­cerns. Even if courts felt capa­ble of mak­ing fac­tu­al deter­mi­na­tions about the wel­fare effects of a giv­en merg­er, they and the gov­ern­ment would be hand­i­capped in merg­er pro­ceed­ings by their lack of infor­ma­tion about poten­tial effi­cien­cy gains. 37 Defen­dants hold near­ly all of the data about poten­tial effi­cien­cy gains—the par­ty chal­leng­ing the merg­er would, there­fore, be at the mer­cy of the defen­dants. Defen­dants employ­ing an effi­cien­cies defense would face a num­ber of advan­tages, chief among those would be the abil­i­ty to selec­tive­ly dis­close infor­ma­tion to por­tray the effi­cien­cy gains in a light most favor­able to the merg­er. The hypo­thet­i­cal nature of all effi­cien­cy gains fur­ther com­pli­cates this issue. 38 Williamson him­self con­ced­ed that the adver­sar­i­al process of the court­room would do noth­ing to ame­lio­rate the infor­ma­tion asym­me­try prob­lems. 39 The inabil­i­ty of courts to accu­rate­ly assess the poten­tial effi­cien­cies, or worse, to assess them in a biased man­ner because of poor infor­ma­tion, rep­re­sents a severe lim­i­ta­tion to the via­bil­i­ty of the effi­cien­cies defense. Courts are wise to pro­ceed with cau­tion in accept­ing an effi­cien­cies defense when faced with prob­lems of infor­ma­tion asym­me­try. A less judi­cious approach would like­ly lead to the approval of merg­ers that have sub­stan­tial anti­com­pet­i­tive effects, despite the osten­si­ble effi­cien­cies claimed by the merg­ing firms.

While some courts have begun to take steps towards rec­og­niz­ing an effi­cien­cies defense in the merg­er con­text, in prac­tice the door is still large­ly closed for defen­dants. Severe prac­ti­cal lim­i­ta­tions to imple­ment­ing a true effi­cien­cies defense cau­tion against pro­vid­ing such a shield to defen­dants. In par­tic­u­lar, courts are unlike­ly to be able to accu­rate­ly cal­cu­late and eval­u­ate the effi­cien­cy gains that defen­dants are pur­port­ing. These lim­i­ta­tions war­rant the high bar­ri­ers that courts have erect­ed for defen­dants attempt­ing to employ an effi­cien­cies defense.

Notes:

  1. This Con­tri­bu­tion stems from my expe­ri­ence in the 2016 Glob­al Antitrust Insti­tute Invi­ta­tion­al moot court com­pe­ti­tion. One of the ques­tions pre­sent­ed, as part of the com­pe­ti­tion, was whether U.S. antitrust law should move to a total wel­fare stan­dard. As a part of that dis­cus­sion, I explored the effi­cien­cies defense to see if it rep­re­sent­ed a move­ment toward a total wel­fare stan­dard. While I con­clud­ed that the effi­cien­cies defense pre­serves the cur­rent con­sumer wel­fare stan­dard, I also noticed sev­er­al weak­ness­es in the effi­cien­cies defense, which are dis­cussed above.
  2. See Oliv­er E. Williamson, Economies as an Antitrust Defense Revis­it­ed, 125 U. Pa. L. Rev. 699 (1977).
  3. See e.g., Hosp. Corp. of Am. v. Fed. Trade Comm’n, 807 F.2d 1381, 1386 (7th Cir. 1986) (explain­ing the con­cerns about coor­di­na­tion among mar­ket par­tic­i­pants as result of increased mar­ket con­cen­tra­tion).
  4. See Williamson, supra note 2, at 699.
  5. See Fed. Trade Comm’n v. Univ. Health, Inc., 938 F.2d 1206, 1223 (11th Cir. 1991).
  6. See Saint Alphon­sus Med. Ctr-Nam­pa, Inc. v. St. Luke’s Health Sys., Ltd., 778 F.3d 775, 790 (9th Cir. 2015) (dis­cussing the adop­tion of an effi­cien­cies defense by oth­er cir­cuits and the Court’s skep­ti­cism of the defense).
  7. 386 U.S. 568, 580 (1967); see also Saint Alphon­sus, 778 F.3d at 789.
  8. Proc­ter, 386 U.S. at 580
  9. Id. at 597–98. (Har­lan J., con­cur­ring).
  10. 938 F.2d 1206, 1223 (11th Cir. 1991).
  11. See Saint Alphon­sus, 778 F.3d at 789; see also ProMed­ica Health Sys., Inc. v. Fed. Trade Comm’n, 749 F.3d 559, 571 (6th Cir. 2014); Fed. Trade Comm’n v. H.J. Heinz Co., 246 F.3d 708, 720 (D.C. Cir. 2001); Fed. Trade Comm’n v. Tenet Health Care Corp., 186 F.3d 1045, 1054–55 (8th Cir. 1999); Univ. Health, Inc., 938 F.2d at 1223.
  12. See Saint Alphon­sus, 778 F.3d at 789 (cit­ing RSR Corp. v. Fed. Trade Comm’n, 602 F.2d 1317, 1325 (9th Cir. 1979)).
  13. Com­pare Saint Alphon­sus, 778 F.3d 775 (dis­cussing the merg­er of health­care sys­tem in Ida­ho), ProMed­ica Health, 749 F.3d 559 (involv­ing the merg­er of health sys­tems in Ohio), Tenet Health, 186 F.3d 1045 (uphold­ing an a pre­lim­i­nary injunc­tion of the merg­er two hos­pi­tals in Mis­souri), and Univ. Health, 938 F.2d 1206 (revers­ing and grant­i­ng a pre­lim­i­nary injunc­tion of the acqui­si­tion of a hos­pi­tal by a rival health sys­tem), with H.J. Heinz, 246 F.3d 708 (involv­ing a merg­er in the baby food indus­try). See gen­er­al­ly Hosp. Corp., 807 F.2d at 1388 (dis­cussing unique fea­tures of the mar­ket for med­ical ser­vices).
  14. See Saint Alphon­sus, 778 F.3d at 789.
  15. See H.J. Heinz, 246 F.3d at 722.
  16. See id.
  17. See id. at 721.
  18. See Joseph F. Brod­ley, The Eco­nom­ic Goals of Antitrust: Effi­cien­cy, Con­sumer Wel­fare, and Tech­no­log­i­cal Progress, 62 N.Y.U. L. Rev. 1020, 1026 (1987).
  19. See id. at 1028; see also H.J. Heinz, 246 F.3d at 722.
  20. See H.J. Heinz, 246 F.3d at 720.
  21. See Hosp. Corp. of Am. v. Fed. Trade Comm’n, 807 F.2d 1381, 1386 (7th Cir. 1986) (argu­ing that, in eval­u­at­ing claims under sec­tion 7 of the Clay­ton Act, courts are mak­ing a judg­ment call about the prob­a­bil­i­ty of col­lu­sion).
  22. See Univ. Health, Inc., 938 F.2d at 1223.
  23. See Brod­ley, supra note 18, at 1033–35.
  24. See Saint Alphon­sus, 778 F.3d at 789 (“[E]ven in those cir­cuits that rec­og­nize it, the para­me­ters of the defense remain impre­cise.”).
  25. See id.
  26. See e.g., Fed. Trade Comm’n v. Sta­ples, Inc., 970 F. Supp. 1066, 1088 (D.D.C. 1997) (“There has been great dis­agree­ment regard­ing the mean­ing of [the Proc­ter] prece­dent and whether an effi­cien­cies defense is per­mit­ted.”). 
  27. Com­pare Robert H. Bork, The Antitrust Para­dox, 125 (1978) (argu­ing against an effi­cien­cies defence because con­duct­ing wel­fare analy­sis on a case-by-case basis is impos­si­ble), with Oliv­er E. Williamson, supra note 2, at 699 (argu­ing for antitrust law to adopt an effi­cien­cies defense despite its lim­i­ta­tions).
  28. See e.g., Saint Alphon­sus, 778 F.3d at 790 (“We remain skep­ti­cal about the effi­cien­cies defense in gen­er­al and about its scope in par­tic­u­lar. It is dif­fi­cult enough in § 7 cas­es to pre­dict whether a merg­er will have future anti­com­pet­i­tive effects with­out also adding to the judi­cial bal­ance a pre­dic­tion of future effi­cien­cies.”); cf. Univ. Health, 938 F.2d at 1223 (“[I]t is dif­fi­cult to cal­cu­late the anti­com­pet­i­tive costs of an acqui­si­tion against which to com­pare the gains real­ized through greater effi­cien­cy; such a com­par­i­son is nec­es­sary, though, to eval­u­ate the acquisition’s total com­pet­i­tive effect.”).
  29. See Bork, supra note 27, at 125.
  30. See Williamson, supra note 2, at 701.
  31. See id. at 700.
  32. See Her­bert Hov­enkamp, Imple­ment­ing Antitrust’s Wel­fare Goals, 81 Ford­ham L. Rev. 2471, 2477–78 (2013); see also Bork, supra note 27, at 107.
  33. See Williamson, supra note 2, at 701.
  34. See Bork, supra note 27, at 125.
  35. See id.
  36. See id. at 126.
  37. See Williamson, supra note 2, at 703 (“Although the gov­ern­ment and the defen­dant have rough­ly equal access to mar­ket share sta­tis­tics, and can present, inter­pret, and con­test such data equal­ly well, the same is not true with respect to a pur­port­ed economies defense. Here, the data are dis­trib­uted uneven­ly to the strate­gic advan­tage of the defen­dant … ”).
  38. See Univ. Health, 938 F.2d at 1223.
  39. See Williamson, supra note 2, at 703 (“[T]he advo­ca­cy process is poor­ly suit­ed for pur­pos­es of get­ting a bal­anced pre­sen­ta­tion of the evi­dence before the court.”).