# Amicus Curiae Brief — King v. Burwell, 135 S. Ct. 475 (2014) (No. 14-114) > Briefs, arguments, decisions, and more. URL: https://www.frixlaw.com/law-library/documents/brief%3Amicro_IA40385016_0338%3A44 ## Record - **Collection:** Supreme Court brief - **Document type:** Amicus Curiae Brief - **Published:** January 1, 2014 ## Text Supreme Court, US. FILED No. 14-114 | JAN ? 5 2015 pornos JFFICE OF THE CLERK 3n The Supreme Court of the Anited States ° DAVID KING, et ai., Petitioners, Vv. SYLVIA BURWELL, SECRETARY OF HEALTH AND HUMAN SERVICES, et ai., Respondents. ¢ On Writ Of Certiorari To The United States Court Of Appeals For The Fourth Circuit ° BRIEF OF AMICI CURIAE JEWISH ALLIANCE FOR LAW & SOCIAL ACTION (JALSA), JEWISH SOCIAL POLICY ACTION NETWORK (JSPAN), JEWISH COUNCIL ON URBAN AFFAIRS (JCUA), BOSTON ALLIANCE FOR COMMUNITY HEALTH, LAWYERS’ COMMITTEE FOR CIVIL RIGHTS AND ECONOMIC JUSTICE, AND PROFESSORS OF HEALTH LAW AND CONSTITUTIONAL LAW IN SUPPORT OF RESPONDENTS ° ANDREW M. FISCHER Counsel of Record JASON & FISCHER 47 Winter Street Boston, MA 02108 telephone (617) 423-7904 fax (617) 451-3413 afischer@jasonandfischer.com COCKLE LEGAL BRIEPS /800) 2256-6964 WWW COCKLELEGALSRIEFS.COM QUESTION PRESENTED Petitioners’ interpretation of § 36B raises two serious constitutional questions that the parties’ briefs do not address: first, whether a federal statute may impose different substantive rules in different states simply because some fail to establish exchang- es, see Shelby County v. Holder, 133 S.Ct. 2612, 2624 (2013) (prohibiting “disparate geographic coverage” absent “exceptional conditions”), and, second, wheth- er the potentially destructive set of rules that would apply in non-establishing states would be unconstitu- tionally coercive, see Nat Fed’n of Indep. Bus. v. Sebelius, 132 S.Ct. 2566, 2608 (2012). According to Petitioners, Congress’s intent in § 36B was to incentivize states to establish exchanges by threatening to withhold subsidies from non- compliant states. But given the Affordable Care Act’s structure, the loss of subsidies is not the only threat embedded in that interpretation. Petitioners’ reading also threatens non-compliant states with federal imposition of guaranteed-issue and community-rating requirements without federal enforcement of the individual or employer mandates to counteract result- ant adverse selection. Given the mandates’ accepted role as correctives to the ACA’s otherwise-destructive regulations and given that Petitioners’ interpretation would cause non-enforcement of those correctives in non-compliant states, the threat in Petitioners’ inter- pretation is: “Establish an exchange, or the federal government will destroy your individual insurance il QUESTION PRESENTED ~ Continued market.” That regulatory threat plausibly violates both the principle of equal sovereignty and the anti- coercion constraint — two constitutional challenges that the Court will confront in future litigation if it rules in Petitioners’ favor. The question presented is whether this Court should apply the canon of constitutional avoidance to uphold the IRS Rule, even if the Court finds that Petitioners’ understanding comports better with the plain language of § 36B. ill TABLE OF CONTENTS Page QUESTION PRESENTED......................cccseeeeeeees i TABLE OF AUTHORITIES ........................ceeeeeee Vv INTEREST OF AMICI CURIAE ......................... 1 SUMMARY OF THE ARGUMENT ...................... 2 STE, cid Atuitnnicontnnadiandsbindccnibiniebedinantevcanesss 6 I. PETITIONERS’ INTERPRETATION OF §36B RAISES SERIOUS CONSTITU- FIs CSFII iieineierecsissiscctanesesncntociness 6 A. The Lodestar for Constitutionality of a Fiscal Punishment is Its Magnitude as a Percentage of State Expendi- tures — a Test that the Fiscal Threat in Petitioners’ Interpretation Likely HUE ads cnnasiarsidinakspusmdiedunndscecanndapuninaniaunes 8 B. Under Petitioners’ Interpretation, Section 36B Threatens Non-Compliant States With a Perverse Subset of the ACA’s Reforms, Raising Intense Con- stitutional Difficulties......................... 12 1. Petitioners’ Interpretation Gives Rise to “Disparate Geographic Coverage” of the ACA’s Core Provi- SEATS P RS Eee etae scene eee! wor gen aaron 13 2. There is No Plausible — Much Less “Exceptional” — Justification for This Geographic Differentiation .... 17 iv TABLE OF CONTENTS — Continued Page 3. This Geographic Disparity is In- distinguishable from the Disparity that the Court Invalidated in FT IR ccictrrcncivnsacienpioersastan 19 4. Disparate Geographic Coverage as an Incentive for Implementation of Federal Policy Raises Profound Anti-Coercion Concerns.................. 26 Il. THE GOVERNMENT’S INTERPRETA- TION OF § 36B IS AT LEAST “FAIRLY POSSIBLE” AND AVOIDS ANY CON- ee Os Bs BEF EI OPE scdccensicvesccesvsscseniens 36 A. The Government’s Interpretation of § 36B is at Least “Fairly Possible’”...... 36 B. The Government’s Interpretation Avoids the Serious Constitutional Problems that Infect Petitioners’ In- I iiccenacintsatccanvaperocastaamnniaccon 38 EID ciciatciparincsvedssennsticaizostbareusesseveeanenes 4} TABLE OF AUTHORITIES Page CASES Ashwander v. Tenn. Valley Auth., 297 U.S. 288 (Be TM actatatlonet haart alls tent etew aie Mele Be eet d 3,7 Bolling v. Sharpe, 347 U.S. 497 (1954) ................2..42- 25 Crandall v. Nevada, 73 U.S. 35 (1868) .......0...........00. 25 Crowell v. Benson, 285 U.S. 22 (1932) ............. 6, 36, 37 Edward J. DeBartolo Corp. v. Fla. Gulf Coast Bldg. & Const. Trades Council, 485 U.S. 568 (i nil tata ATA a Ee dite a ot ARN dnt Mitel ats aE eS 7 Gade v. Nat'l Solid Wastes Mgmt. Ass’n, 505 RI MOTE UIs cuciirl cto tas cece taetaionbabnaniotesermastasssocoie 24 Garcia v. San Antonio Metro. Transit Auth., en a 38 Gregory v. Ashcroft, 501 U.S. 452 (1991)............. 24, 38 Halbig v. Burwell, 758 F.3d 390 (D.C. Cir. ea aa IEE te Pe WE cere ee gH AIRE RE ae Se 6, 18, 36 Halbig v. Sebelius, 113 AF.T.R.2d 2014-548 RE RGRAY ett TLE NIA SE EE adver een Re 36 Hooper v. California, 155 U.S. 648 (1895) ...0.00. 36 Jacobellis v. Ohio, 378 U.S. 184 (1964).................2.... 29 King v. Burwell, 759 F.3d 358 (4th Cir. 2014) .....6, 12, 36 King v. Sebelius, 997 F. Supp. 2d 415 (E.D. Va. ER ABLE AR Sh RS BESSY ONG AOR PRION OTOL A Oe Sa! 37 Nat'l Fed’n of Indep. Bus. v. Sebelius, 132 S.Ct. RE a SURES ga Re eee 01 Se a passim vi TABLE OF AUTHORITIES — Continued Page Natl League of Cities v. Usery, 426 U.S. 833 (NTA ARSRL ES SS RE SER aia tre Mi RE sea Rem 38 New York v. United States, 505 U.S. 144 AAR SRE Re A Me Bb RRMC CIENT SRA ER ics 2 OF 19, 25, 40 Nw. Austin Mun. Util. Dist. No. One v. Holder, gh Ree Aeron eee eve eet Saahe 13, 17 Peseer CO E008, GET UB. BI CRB IID eniscecvcnicccevicnccecescsosccs 25 Printz v. United States, 521 U.S. 898 (1997) ............. 40 Rust v. Sullivan, 500 U.S. 173 (1991)........................ 37 Shelby County v. Holder, 133 S.Ct. 2612 (2013)....passim South Carolina v. Katzenbach, 383 U.S. 301 (BRU SRISTRR RIERA 8 Sa SAS ESSE Riad sess ha Bene ss ee 18 South Dakota v. Dole, 483 U.S. 203 (1987)........ passim Spallone v. United States, 493 U.S. 265 (1990)......... 24 Texas v. White, 7 Wall. 700 (1869)...................22.2.0000e 21 United States v. X-Citement Video, Inc., 513 Be sic ntanictesercdaganmampincamien iiapasinnduabenesiats 5, 36 Yates v. United States, 354 U.S. 298 (1957)......... 36, 37 CONSTITUTION AND STATUTES con eee ee. Ge as ccpucsusnesaaeqenienemouensidenst 24 of Gy eg f ee eee 25 ee I, SES “TOD... csonbaxnenenenauoisbientenieodenimniuited 19 Health Care and Education Reconciliation Act of 2010, Pub. L. No. 111-152, 124 Stat. 1029 ........... 3 Vu TABLE OF AUTHORITIES — Continued Page National Highway System Designation Act of 1995, Pub. L. No. 104-59, 109 Stat. 568......0.......... 27 National Maximum Speed Law, Pub. L. No. 97- Eh. eee. 27 Patient Protection and Affordable Care Act, Pub. L. No. 111-148, 124 Stat. 119 (2010)...... passim rs eI iaicissisiressscnihatacsiesceenastetencennenininneiecimensietns 27 Se a, SR i ocierinaninceceieieemninteonatancoassapipdadinnnedl passim Se Ria IEEE sitictciinesasnninatasedenubulbioninnaens 3, 15, 33, 38 Be a: We I ertithcsscenicinicessnissosenseoneraneninrveriupnincieas 3, 15, 32 Oe ee iitcicnennsinchnccisnacinacninanichnonnsniesccsvnknieaniiins 14 as ED icitscsiecertcaseceminsiravennrninidanmpniinintintgt 14 itis te INO cctpantcncicitnsdimhanonencniasginnininascndanouitonniete 14 ee Oe ND cetetininertannsicnenseennnicomnsasensessnsennnbties 14 A is eC Rescimanonnecdncesndanteendeiennndasmmnsteheaciesis 14 Se Oe EE Siccncicrepcstecsncrsionieinnennnisantanseinnnnanitnss 14 rs Oe TED vipcncllasisetinitnicncavesdaennapincacniid 14, 33 Oe ee ES ses cap cetseceeenntcencsictenndceincennetncecmantionen 16 a Oe ID setiicn ints nnccsnsesestacsucnereiocinniness 24, 32 i Oe IE wecicntiennsesiscccenncemisteicasonenaandademannia 37 ee i tenceieniasnenisisinsinetininacessencnsvedinsnan 16 I, Oe HII rnentssne icin cheenpentisninenneesinccaranninegeaetes 22 Oe ee, Oe SII encnnects centctenitncnectinenictentinswueeinnss 19, 22 Oe i Te IO ieieectencticcnencepriestindeemsenecniany 11, 38, 39 vill TABLE OF AUTHORITIES — Continued Page Be Re ITD siicteneccesieestinuncenenenhintysiivontatbmodaiintin 37, 39 Ne Oe IEE eetrenseccenciicieneninvesnpenntntebiestenienseeentio 25 MISCELLANEOUS Br. Amici Curiae America’s Health Insurance Plans, Halbig v. Burwell, No. 14-5018 (D.C. Cir. Nov. 3, 2014) (em bamnc)...................0....c.c000e- 7,29 Br. Amici Curiae Economic Scholars, Halbig v. Burwell, No. 14-5018 (D.C. Cir. Nov. 3, 2014) DSRS. OPER AS RITE ROR a aR OT Oe 7, 14, 17, 29 Br. Amici Curiae State of Indiana and 39 Indiana Public School Corporations................. 31, 38 Br. Amici Curiae State of Oklahoma et al................. 31 Br. Amici Curiae States of Kansas and Nebras- ka, King v. Sebelius, No. 14-1158 (4th Cir. ee ee a eiseentneelbeneniaadeesnenanenneninita 23 Br. for United States, Nat7 Fed’n of Indep. Bus. v. Sebelius, No. M-398 (U.S. Jan. 6, 2012)............. 15 Linda J. Blumberg, Matthew Buettgens & John Holahan, The Implications of a Su- preme Court Finding for the Plaintiffs in King v. Burwell: 8.2 Million More Uninsured and 35% Higher Premiums, Urban Institute, Jan. 2015, http://www.urban.org/publications/ BI ett iietdscteneiacedcradsnssencuiaiasecwncumdatiiies 17 ix TABLE OF AUTHORITIES — Continued Matthew Buettgens & Caitlin Carroll, Elimi- nating the Individual Mandate: Effects on Premiums, Coverage, and Uncompensated Care, Urban Institute, Jan. 2012, http://www. urban.org/U ploadedPDF/412480-Eliminating- the-Individual-Mandate. pdf Andrew Cline, How Obama Broke His Promise on Individual Mandates, The Atlantic, June 29, 2012, http://www.theatlantic.com/politics/ archive/2012/06/how-obama-broke-his-promise- on-individual-mandates/259183/ Compl., Florida ex rel. Bondi v. U.S. Dep't of Health & Human Servs., 780 F.Supp. 2d 1256 (N.D. Fla. 2010) (No. 3:10-cv-91) Compl., King v. Sebelius, 997 F. Supp. 2d 415 (E.D. Va. 2013) (No. 13-CV-630) Cong. Budget Office, Updated Estimates of the Effects of the Insurance Coverage Provisions of the Affordable Care Act (2014), https:// www.cbo.gov/sites/default/files/45231-ACA __ IIIT oi daaccantasiccpinenecassastivncncidenackemioociaoets Cong. Research Serv., Federal Funding for Health Insurance Exchanges, Oct. 14, 2014, https://www.hsdl.org/?view&did=759147 Paul Grimes, Practical Traveler: The 55-M.P.H. Speed Limit, N.Y. Times, Dec. 26, 1982, http://www.nytimes.com/1982/12/26/travel/ practical-traveler-the-55-mph-speed-limit. html de hi ee ee ee ee ee | mers 27 x TABLE OF AUTHORITIES — Continued Page Larry Levitt & Gary Claxton, The Potential Side Effects of Halbig, Kaiser Family Found., July 31, 2014, http://kff.org/health-reform/ perspective/the-potential-side-effects-of-halbig/....... 16 Nat Ass’n of State Budget Officers, State Ex- penditure Report: Examining Fiscal 2012-2014 State Spending (2014), http://www.nasbo.org/ sites/default/files/State%20Expenditure%20 Report%20(Fiscal%202012-2014)S. pdf ................00.. 9 8 6 TEE 3, 12, 29 Transcript of Oral Argument, Halbig v. Bur- well, 758 F.3d 390 (D.C. Cir. 2014) (No. 14- 1 INTEREST OF AMICI CURIAE' The amici Jewish organizations — JALSA, JSPAN, and JCUA -— represent a tradition of believ- ing that the community has an essential role in providing for the sick. Preserving life and health is one of the highest of communal duties in the Jewish tradition. These amici represent a minority communi- ty deeply committed to equity in access to healthcare services. Boston Alliance for Community Health is an alliance of neighborhood health providers and com- munity-based organizations. Achieving health equity is an overarching goal of the organization, and health insurance is an important part of that goal. If a significant number of people in the country are de- nied benefits, the whole system might be in jeopardy, impacting residents including the people who are most affected by health inequities. Lawyers’ Committee for Civil Rights and Economic Justice is a non-profit that specializes in law reform litigation to redress race and national origin discrimination. As part of its effort to reduce disparities in health status outcomes, the Lawyers’ Committee formed a partnership with Massachusetts General ' This brief is submitted with the consent of the parties, as lodged with the Clerk per the Docket Sheets. Pursuant to Rule 37.6, counsel represent that this brief was not authored in whole or in part by counsel for any party. Amici have borne their own expenses, without support from any party. 2 Hospital in 2003 to assist eligible patients in obtain- ing benefits. The program focuses on the “social determinants of health,” providing representation to refugees and immigrants who seek disability awards, naturalization, unemployment benefits, public hous- ing and child support. The case at bar could undo the progress in health care access that has been achieved over the past five years. The regulatory confusion that would ensue puts the most vulnerable low- income patients at risk. Professors Abigail R. Moncrieff (Boston University School of Law), Allison K. Hoffman (UCLA School of Law), Sharona Hoffman (Case Western Reserve University Schoo! of Law), Russell B. Korobkin (UCLA School of Law), Joan H. Krause (UNC School of Law), Stephen G. Marks (Boston University School of Law), Kevin Outterson (Boston University School of Law), and Theodore W. Ruger (University of Pennsylvania Law School) teach and write in the fields of healthcare law, feder- alism, and constitutional law, and they have written extensively on the federalism implications of the ACA. They have strong professional interests in the outcome of this case. ¢ SUMMARY OF THE ARGUMENT Petitioners argue that Congress included health insurance subsidies under the Patient Protection and Affordable Care Act (ACA), Pub. L. No. 111-148, 124 3 Stat. 119 (2010),’ as a “coercive” incentive for states to establish exchanges. See Pet’rs’ Br. at 32. Indeed, Petitioners repeatedly analogize the subsidies provi- sion, 26 U.S.C. § 36B, to the Medicaid expansion that this Court invalidated in Natl Fed’n of Indep. Bus. v. Sebelus, 132 S.Ct. 2566, 2601-08 (2012). That analogy alone ought to raise red flags under the canon of constitutional avoidance, which counsels this Court to disfavor any statutory interpretation that provokes constitutional doubt. See generally Ashwander v. Tenn. Valley Auth., 297 U.S. 288 (1936). But the constitutional questions arising from treating subsidies as incentives are relatively minor compared to those that arise from the regulatory consequences of Petitioners’ interpretation. Under Petitioners’ view, non-compliant states will not only suffer significant opportunity costs of forgone subsi- dies; they will also suffer real costs from federal imposition of a perverse subset of the ACA’s substan- tive insurance regulations. As the ACA makes clear, section 36B subsidies are required to trigger federal enforcement of the individual and employer mandates. 26 U.S.C. § 5000A(eXB\ii) (individual mandate); 26 U.S.C. §4980H (employer mandate). If any state were ineligible for subsidies, that state would be subject to the ACA’s intense market controls ~— including the * Amended by the Health Care and Education Reconcilia- tion Act of 2010, Pub. L. No. 111-152, 124 Stat. 1029. 4 guaranteed-issue and community-rating require- ments — without the offset of federally-enforced mandates. The result of that regulatory mix, as this Court recognized in NFIB, could be rampant adverse selection that shrinks or destroys individual insur- ance markets. 132 S.Ct. at 2585; id. at 2626 (Gins- burg, J., concurring in part and dissenting in part). Importantly, the states would not be able to cure that problem by providing their own subsidies because only § 36B subsidies count in the mandates’ enforce- ment formulae. Petitioners’ interpretation of §36B thus causes “disparate geographic coverage” of the individual and employer mandates, Shelby County v. Holder, 133 S.Ct. 2612, 2622 (2013), and it contains an embedded threat of regulatory punishment for a state’s non- compliance with federal policy. Petitioners’ ACA tells states that they must set up an exchange, or the federal government will purposefully impose a perverse subset of the ACA’s regulations within their borders — a subset perversely designed to promote adverse selection. The threat is, “Establish an exchange, or the federal government will destroy your market.” Both the disparate geographic coverage of the mandates and the threat of regulatory punishment arise uniquely from Petitioners’ interpretation, and each alone renders the statute constitutionally prob- lematic. Although disparate geographic coverage of federal funds (like the subsidies) is common in coop- erative federalisr., disparate geographic coverage of substantive rules (like the mandates) is unheard of — 5 and likely unconstitutional under Shelby County. Id. Similarly, although fiscal threats are common and only sometimes unconstitutional, a regulatory threat of the kind embedded in Petitioners’ interpretation looks much more problematic under the anti-coercion constraint. NFIB, 132 S.Ct. at 2601. Congress has never before threatened to impose different, and intentionally destructive, substantive rules in states that refuse to implement federal laws. Because such a threat is likely unconstitutional, this Court should not hold that Congress intended the ACA to operate in this unusual way. See United States v. X-Citement Video, Inc., 513 U.S. 64, 73 (1994) (holding that the Court should not “impute to Congress an intent to pass legislation that is incon- sistent with the Constitution”). Even if the Court concludes that Petitioners’ is “the most natural interpretation” of § 36B, the avoid- ance canon instructs the Court to resort to every reasonable alternative to avoid constitutional defects. NFIB, 132 S.Ct. at 2594. Petitioners therefore bear a high burden in light of their interpretation’s constitu- tional infirmities. They cannot merely show that their reading is superior to the government’s; they must show that the government’s is so unreasonable that the Court should not resort to it to avoid unconstitu- tionality. But the government’s interpretation is, at mini- mum, a “fairly possible” construction of the statute's text and structure, and it successfully avoids 6 constitutional pitfalls. See Crowell v. Benson, 285 U.S. 22, 62 (1932). We urge the Court to affirm the Fourth Circuit in order to avoid further constitutional htigation. « ARGUMENT I. PETITIONERS’ INTERPRETATION OF § 36B RAISES SERIOUS CONSTITUTIONAL DOUBTS. Throughout this litigation, Petitioners have argued that § 36B intends to “coerce” the states into establishing their own exchanges. See King v. Bur- well, 759 F.3d 358, 372, 375 (4th Cir. 2014). Indeed, in his oral argument in Halbig v. Burwell, 758 F.3d 390 (D.C. Cir. 2014), Petitioners’ attorney repeatedly analogized § 36B to the Medicaid expansion, arguing that the only difference is that Medicaid imposed “more draconian consequences if the states says [sic] no.” Transcript of Oral Argument at 20-24, Halbig v. Burwell, 758 F.3d 390 (D.C. Cir. 2014) (No. 14-5018). But that is neither a genuine difference nor the only difference — and the others raise serious constitution- al difficulties that did not arise for Medicaid. If the Court adopts Petitioners’ reading, it will not just hold that Congress intended non-compliant states to forgo subsidies (a traditional fiscal punish- ment that nevertheless raises doubts under NFIB). It will also hold that Congress intended those states to become subject to a federally-enforced set of policies 7 apparently designed to destroy individual insurance markets (a novel kind of regulatory punishment that raises serious doubts under both NFIB and Shelby County). See Br. Amici Curiae Economic Scholars at 3-6, Halbig v. Burwell, No. 14-5018 (D.C. Cir. Nov. 3, 2014) (en banc) (“Economic Scholars Br.”) (predicting dire consequences from Petitioners’ interpretation); Br. Amici Curiae America’s Health Insurance Plans at 24-35, Halbig v. Burwell, No. 14-5018 (D.C. Cir. Nov. 3, 2014) (en banc) (“Insurance Plans Br.”) (same). Never before has this Court confronted a cooper- ative federalism scheme that threatens states with regulatory, rather than fiscal, harm if they refuse to implement federal policy. That structure raises a stickier set of constitutional difficulties than the fiscal incentives that Congress typically uses. It not only plausibly violates equal sovereignty but also complicates the anti-coercion constraint by requiring the Court to assess economic impacts of a regulatory punishment (an extraordinarily difficult task). Applying the canon of constitutional avoidance, the Court should interpret § 36B to avoid this consti- tutional morass. See Ashwander, 297 U.S. 288; Ed- ward J. DeBartolo Corp. v. Fla. Gulf Coast Bldg. & Const. Trades Council, 485 U.S. 568 (1988). 8 A. The Lodestar for Constitutionality of a Fiscal Punishment is Its Magnitude as a Percentage of State Expenditures - a Test that the Fiscal Threat in Petition- ers’ Interpretation Likely Fails. In both South Dakota v. Dole, 483 U.S. 203 (1987), and NFIB, 132 S.Ct. at 2601-08, tl s Court confronted the question of whether Congress may punish a state that refuses to implement federal policies. The Court’s conclusion was that negative incentives are permissible as long as they are not as severe as “a gun to the head,” NFIB, 132 S.Ct. at 2604, but the Court found “no need to fix a line” for distinguishing permissible from impermissible pun- ishments. Jd. at 2606-07. What the Court did appear to “fix” is that the lodestar for constitutionality is a punishment’s mag- nitude as a percentage of state expenditures. As the Court noted, the Dole punishment amounted to “less than half of one percent of South Dakota’s budget,” id. at 2604, while the Medicaid punishment amount- ed to “over 10 pertent,” id. at 2605. Notably, that test applied only to punishments given the plurality’s remedial holding, which allowed Congress to make a new offer of Medicaid expansion grants as long as a state’s refusal did not endanger preexisting Medicaid. Id. at 2607-08. It was presumably on a simplistic application of these rules that Petitioners felt comfortable raising their coercion theory despite its constitutional 9 infirmities. According to the Congressional Budget Office, subsidies would amount to about $2 billion per state per year, which is about 6 percent of 2014 state expenditures.’ See Cong. Budget Office, Updated Estimates of the Effects of the Insurance Coverage Provisions of the Affordable Care Act (2014), https:// www.cho.gov/sites/default/files/45231-ACA Estimates. pdf (estimating a national cost of $1.032 trillion from 2015-2024); Nat'l Ass’n of State Budget Officers, State Expenditure Report: Examining Fiscal 2012-2014 State Spending at 8, Table 1 (2014), http://www. nasbo.org/sites/default/files/State%20Expenditure%20 Report%20(Fiscal%202012-2014)S. pdf (projecting 2014 state expenditures of about $1.78 trillion: about $36 billion per state). That figure looks awfully close to the 10 percent figure’ that the Court deemed “economic dragooning” in NFIB, 132 U.S. at 2605 — and a far cry from the .05 percent figure that the Court upheld in Dole — but, under Petitioners’ interpretation, the subsidies * This figure might overestimate somewhat. It takes the average annual subsidies expenditure from a 10-year period and compares it to estimated state expenditures in 2014 alone. State budgets will likely grow over those ten years, making the subsidies’ percentage slightly smaller. Unfortunately, states do not project budgets over CBO’s timeframe. * The NFIB opinions make reference to 20 percent, which is aggregate spending on Medicaid from state and federal sources. The threat was loss of only federal dollars, which ranges from 50-83 percent of aggregate spending. The relevant amount is thus half or more of 20 percent, which is “over 10 percent,” as the plurality makes clear. 10 constitute a conditional offer of new funding rather than a withdrawal of preexisting funds. The problem is that newness does not assure constitutionality under NFIB. The test for determin- ing whether a condition is a punishment or an offer is not whether it threatens preexisting money. It is whether a condition on funds restricts the use of those funds or instead restricts some other behavior. As Chief Justice Roberts explained, “We have upheld Congress's authority to condition the receipt of funds on the States’ complying with restrictions on the use of those funds, because that is the means by which Congress ensures that the funds are spent according to its view of the ‘general Welfare.’ Conditions that do not here govern the use of the funds, however, cannot be justified on that basis.” Jd. at 2604. Under this rule, the Court allowed the Medicaid expansion as a freestanding offer not because it did not threaten preexisting funds but because, that way, the expan- sion money is an incentive to play by the money’s own rules. It is not an incent..e for states to implement some other federal program. If subsidies are an incentive to establish ex- changes, they violate this limitation on conditional grants. Under Petitioners’ interpretation, subsidies are not an incentive to play by the subsidies’ rules; they are an incentive to take on the separate task of establishing and operating an exchange — a task that has no bearing on the subsidies’ success or failure. The condition on the subsidies, according to Petition- ers, is not that the states use the money consistently 11 with Congress’s conception of the “general Welfare.” It is just that the states do a job that the federal government would rather not do. One could argue that the subsidies and exchang- es are more of a single program than were the two parts of Medicaid, in which case exchange- establishment could arguably be part of Congress’s vision for the proper use of subsidies. But the central factor on which NFIB relied in arguing that the Medicaid expansion was separate from preexisting Medicaid was the programs’ separate administrative features. See id. at 2506. Applying that factor, the subsidies and exchanges are separate. The states administer the exchanges (according to Petitioners) while the Internal Revenue Service administers subsidies. Furthermore, although the subsidies and exchanges arose from the same Act of Congress, they are codified in separate titles. Compare 26 U.S.C. §36B with 42 U.S.C. § 18031. Most importantly, state-based exchanges simply are not necessary for subsidies to function as Congress intended. If Con- gress thought state leadership was necessary for exchanges, it was for separate federalism reasons. Under Petitioners’ interpretation, then, the subsi- dies would be conditioned on states’ implementation of a program other than the one being funded. NFIB holds that this kind of condition — a fiscally significant condition that does not “govern the use of the funds” — falls beyond Congress’s power to condition its grants. 132 S.Ct. at 2604. If the Court follows Petitioners’ interpretation, it can expect a 12 constitutional challenge along these lines — which it could avoid by affirming the Fourth Circuit. There are thus serious constitutional doubts associated with a $2 billion opportunity cost in for- gone subsidies, highlighted by Petitioners’ references to Congress’s “coercive” intent. The fiscal implications of Petitioners’ interpretation toe the line of “economic dragooning,” justifying application of the avoidance canon to disfavor it. Jd. at 2605. These doubts, however, are relatively minor compared to those that arise from the interpretation’s regulatory impacts. B. Under Petitioners’ Interpretation, Sec- tion 36B Threatens Non-Compliant States With a Perverse Subset of the ACA’s Reforms, Raising Intense Con- stitutional Difficulties. As Petitioners acknowledge, see Pet’rs’ Br. at 15, one consequence of their interpretation is that the ACA would impose an odd and potentially destructive subset of its regulations in states that fail to establish exchanges — while applying the full corpus of its regime in compliant states. See King, 759 F.3d at 374- 75. This “disparate geographic coverage” of the ACA’s core provisions might be unconstitutional whether or not it is a condition of states’ acquiescence in federal prerogatives. See Shelby County, 133 S.Ct. at 2622-24 (discussing the “fundamental principle of equal sovereignty” and high bar for justifying “disparate 13 geographic coverage” of a federal law). As means of incentivizing state participation in a federal program, threats of such differentiation raise extraordinary difficulties. 1. Petitioners’ Interpretation Gives Rise to “Disparate Geographic Coverage” of the ACA’s Core Provisions. Under Petitioners’ interpretation, the ACA cre- ates a two-track regulatory regime, differentiated by state. Like the Voting Rights Act (VRA) provisions that this Court invalidated in Shelby County, Peti- tioners’ ACA would give rise to disparate geographic coverage of its core provisions, and like the VRA, Petitioners’ ACA would select states for unusual rules based on states’ choices to follow federally-disfavored policies. See Part I.B.2, infra. Under the VRA, state choices to use exclusionary “tests or devices” trig- gered peculiar federal treatment while under Peti- tioners’ ACA, state choices to rely on the federal exchange would trigger peculiar treatment. The biggest difference between the VRA and Petitioners’ ACA is that Congress unambiguously intended the VRA’s “‘disparate geographic coverage,” Shelby County, 133 S.Ct. at 2627 (quoting Nw. Austin Mun. Util. Dist. No. One v. Holder, 557 U.S. 193, 203 (2009)), while the disparities arising from Petitioners’ interpretation are hidden in a domino effect. But that difference just means that the Court can avoid the constitutional difficulty here, as it did in Northwest Austin but could not in Shelby County. 14 Here’s how disparate geographic coverage arises from Petitioners’ interpretation: The ACA creates several new rules for insurance companies. Most famously, it requires guaranteed issue, 42 U.S.C. § 300gg-1 (issuance); 42 U.S.C. § 300gg-2 (renewability), and community rating, 42 U.S.C. § 300gg, but it also bans preexisting condition exclusions, 42 U.S.C. § 300gg-3, coverage rescissions, 42 U.S.C. § 300gg-12, and coverage caps, 42 U.S.C. § 300gg-11. It also imposes a minimum medical loss ratio that limits insurers’ administrative costs, in- cluding profits. 42 U.S.C. § 300gg-18. Together, these rules create a strong incentive for consumers to wait until they are sick to buy insur- ance. The strategies that insurers once used to avoid sick enrollees — and to avoid paying for particular sicknesses — are now illegal, eliminating any risk that a patient will become “uninsurable” if she waits to buy coverage. In q world with only these rules, insur- ers would expect primarily “bad risks” to buy insur- ance, and that “adverse selection” would undermine insurance’s risk-pooling function. Moreover, given the medical loss ratio and community-rating, insurers cannot simply raise premiums to avoid losses. As a result, many insurers would likely stop writing policies in individual markets if the ACA’s reforms went into effect without mandates. The withdrawal of major insurers would be the culmination of the “death spiral” that economists predict when modeling the effects of eliminating the individual mandate. See generally Economic Scholars Br. If a state reached 15 that point, many of its patients would be unable to find health insurance even if they wanted and could afford it. That's why Congress included the individual and employer mandates. 26 U.S.C. § 5000A (individual); 26 U.S.C. §4980H (employer). As the government argued at length in NFIB and as the Court accepted, the mandates are “integral partis) of [the ACA’s] comprehensive scheme of economic regulation.” NFIB, 132 S.Ct. at 2591 (quoting Br. for United States 24); see also id. at 2585. “Without the individual man- date,” Justice Ginsburg explained, “guaranteed-issue and community-rating requirements would trigger an adverse-selection death-spiral.” Jd. at 2626 (Gins- burg, J., concurring in part and dissenting in part). But both the individual and employer mandates explicitly depend on the subsidies — not for their success but for their applicability. The employer mandate applies only if “at least one full-time em- ployee” has enrolled in a health plan for which a “premium tax credit or cost-sharing reduction is allowed or paid.” 26 U.S.C. § 4980H(a)(2). In a state whose citizens are ineligible for subsidies, that man- date can never be triggered. Likewise, the individual mandate contains an exemption for anyone whose “required contribution” to the cost of insurance ex- ceeds 8 percent of his household income, 26 U.S.C. § 5000A(e)(1)(A), and the “required contribution” for anyone purchasing on the individual market is offset “by the amount of the credit allowable under section 36B,” § 5000A(e)(1)(B Xii). 16 Congress decided on the 8 percent affordability threshold with the subsidies in mind, calibrating the two provisions carefully to give the mandate wide- spread applicability. Absence of the subsidies’ offset therefore relieves most purchasers on the individual market of the obligation to insure. See Larry Levitt & Gary Claxton, The Potential Side Effects of Halbig, Kaiser Family Found., July 31, 2014, http://kff.org/ health-reform/perspective/the-potential-side-effects-of- halbig/ (estimating that 83 percent of the otherwise subsidy-eligible population would cross the affordabil- ity threshold). This, indeed, is Petitioners’ theory for standing. Without subsidies, they will not be required to purchase insurance. See Compl., King v. Sebelius, 997 F. Supp. 2d 415 (E.D. Va. 2013) (No. 13-CV-630). One consequence, then, of Petitioners’ interpreta- tion is that the statute would create disparate geo- graphic coverage of the ACA’s substantive rules.” The market reforms would go into effect everywhere,’ but the offsetting mandates would go into effect only in those states that establish exchanges. Petitioners’ ACA says, “If you get subsidies, then you’re bound by * Disparate fiscal treatment of states is common in coopera- tive federalism. See, e.g. 42 U.S.C. § 1396d(b). Disparate regulatory coverage in a cooperative federalism scheme would be, as far as amici know, entirely novel. * The ACA authorizes the states to enforce the market reforms themselves, but it obligates the federal government to enforce in states that cannot or will not. 42 U.S.C. § 300gg-22. 17 mandates. But only some states get subsidies, so only some states’ citizens are bound by mandates.” As this Court understood in NFIB, 132 S.Ct. at 2585, 2626, and as the best available data show, see Economic Scholars’ Br., the result of the mandate-free ACA would be that many insurers would refuse to write individual plans in mandate-free states. Those states’ individual markets would shrink or collapse, leaving many citizens unable to find or afford insur- ance. See Linda J. Blumberg, Matthew Buettgens & John Holahan, The Implications of a Supreme Court Finding for the Plaintiffs in King v. Burwell: 8.2 Million More Uninsured and 35% Higher Premiums, Urban Institute, Jan. 2015, http://www.urban.org/ publications/2000062.html. 2. There is No Plausible - Much Less “Exceptional” - Justification for This Geographic Differentiation. Even if some states prefer the mandate-free regulatory bundle, “disparate geographic coverage” of a federal statute raises constitutional red flags under the “fundamental principle of equal sovereignty.” Shelby County, 133 S.Ct. at 2622-24 (quoting Nw. Austin, 557 U.S. at 203). This Court held in North- west Austin, 557 U.S. 193, and Shelby County, 133 S.Ct. 2612, that a geographically differentiated regula- tory regime is constitutionally permissible only if the federal government can show that “‘exceptional conditions’ ” justify the disparities. Id. at 2624 (quoting 18 South Carolina v. Katzenbach, 383 U.S. 301, 334 (1966)). Importantly, states’ ability to choose the federal regulatory bundle that applies does not obviate the need for exceptional justification. In terms of states’ power to choose, Petitioners’ ACA is indistinguishable from the VRA; both set uniform ground rules by which states could select in or out of particular regu- latory rules. See Part I.B.3, infra. Shelby County held that any unjustified geographic disparity of a federal law is unconstitutional, even if arising from state actions. In this case, there are no conditions, exceptional or otherwise, that justify disparate geographic cover- age of mandates. The insurance market operates similarly throughout the nation, and the decision to coordinate that market through a federal exchange does not alter the market’s conditions in any way that justifies regulatory differentiation. The only argu- ment Petitioners raise to rationalize the disparity is that Congress wanted the states to establish ex- changes and needed an incentive for them to take on that thankless task. But this argument (in addition to raising serious anti-coercion concerns, discussed below) fails to justify geographic disparity. Senator Ben Nelson’s alleged desire for state-run exchanges, | see Halbig, 758 F.3d at 409 n.11 — no matter how sincerely felt or politically important — is not the kind of “exceptional condition” that can justify state-to- state differentiation in a federal law’s coverage. 19 3. This Geographic Disparity is Indis- tinguishable from the Disparity that the Court Invalidated in Shelby County. There are, of course, differences between the ACA that emerges from Petitioners’ interpretation and the VRA provisions that the Court invalidated in Shelby County, 133 S.Ct. 2612. None, however, eliminates constitutional doubts. Two differences make Petition- ers’ ACA look constitutionally worse than the VRA; one makes it look superficially better, but only at first blush; one is ambiguous. First, while both the VRA and Petitioners’ ACA differentiate among states based on choices that state governments made, the VRA differentiated based on unconstitutional choices: the effectual use of tests or devices to exclude voters. See 42 U.S.C. § 1973b(b); U.S. Const. amend. XV. Petitioners’ ACA differenti- ates based on choices that states were constitutional- ly entitled to make: refusal to implement an exchange that meets federal standards. See New York v. United States, 505 U.S. 144 (1992) (anti-commandeering). This distinction makes Petitioners’ ACA look, consti- tutionally, much worse than the VRA. To whatever extent Congress may create disparate geographic coverage, surely that power ought to be greater when the legislature is attempting to enforce constitutional rights than when it is attempting to circumvent constitutional limitations. 20 The second difference is subtler but similarly makes Petitioners’ ACA look worse than the VRA. The VRA was a remedial statute, designed to ensure adoption of a uniform substantive policy throughout the nation. Congress’s goal was universal suffrage in all states, but since some states were aiming for that goal before Congress spoke, the legislature deter- mined that only certain parts of the country needed to take the VRA’s strong medicine. The VRA thus reflected geographic differentiation in enforcement procedures, not in desired results. Under Petitioners’ ACA, by contrast, Congress wanted different substan- tive policies to govern different states. In some, the regulatory regime would treat health insurance as social insurance, with near-universal mandatory contribution, while in others, it would treat health insurance as an ordinary commodity that is neverthe- less subject to intense price controls. That differentia- tion would serve no policy or constitutional interests of its own; it would just be an attempt to effect the separate policy of state-run exchanges. Under the principle of equal sovereignty, unjustified disparities in the coverage of a substantive provision seems much worse than the VRA’s remedial differentiation. Of course, geographic differences in substantive policy are common, even in federally-coordinated programs, but such differences have always arisen from state statutes that regulate avove a federal floor or pursuant to a federal waiver. In standard coopera- tive federalism, the federal government sets rules. for programs that the states run themselves. What such 21 programs do not do — but what Petitioners argue the ACA does — is require states to pick between two substantively different federally-run programs, with state flexibility limited to the choice of one or another bundle of preemptive rules. If it is unconstitutional for the federal government to use different enforce- ment strategies for a uniform substantive policy, as in Shelby County, then it seems clearly unconstitutional for the federal government to enforce different sub- stantive policies in different states, even if the states get to choose which one applies. Indeed, if Congress has the power to create that kind of disparate coverage, then it is hard to see how the states retain sovereignty at all. See generally Texas v. White, 7 Wall. 700, 725 (1869) (emphasizing the states’ “indestructible” sovereignty). Congress could overtake states’ jobs, passing one set of preemp- tive regulations in Texas and a different set in Massachusetts. Even if both states liked the results — even if both states chose results from an “either-or” federal menu — they could no longer call themselves sovereign states. The third difference between the two regimes makes Petitioners’ ACA look superficially better than the VRA, but a deeper understanding reintroduces doubt. Neither the VRA nor Petitioners’ ACA lists covered and uncovered states on the face of the stat- ute. Both instead set formulae for distinguishing states — voter registration practices under the VRA and establishment of a compliant exchange under Petitioners’ ACA. But while the VRA selected states 22 for historic practices, see 42 U.S.C. § 1973b(b), Peti- tioners’ ACA selects states for an ongoing failure that first arose after the statute’s enactment. The states therefore have a current opportunity to choose their regulatory regime under Petitioners’ ACA; all they have do to trigger the mandates is to establish an exchange. Setting aside the anti-coercion issue that arises from that structure: retention of real-time choice might make Petitioners’ ACA look less consti- tutionally problematic. But the VRA included a similar kind of choice. Section 4 provided a “bailout” procedure for covered states, which allowed them to terminate federal oversight of their election laws if they enforced a ten- year moratorium on exclusionary tests or devices. See 42 U.S.C. § 1973b(a). Section 3 also included a “bail- in” procedure that allowed the federal government to initiate preclearance requirements in any state that engaged in current constitutional violations, regard- less of whether they satisfied the historically-based coverage formula. See 42 U.S.C. § 1973a(c). All states were therefore eligible to choose, through their cur- rent use or nonuse of tests or devices, whether they wouid be subject to federal preclearance.’ ’ The “bailout” requirement might seem more onerous than Petitioners’ requirement of exchange establishment, but maybe not by much. It is unclear whether Petitioners ACA would require states merely to “establish” exchanges that they could then turn over to the federal government or whether they would (Continued on following page) 23 In comparing the two choices (bailout to exchange- establishment), it might seem relevant that some states strongly prefer not to have mandates enforced in their jurisdictions. See, e.g., Compl., Florida ex rel. Bondi v. U.S. Dep’t of Health & Human Servs., 780 F.Supp. 2d 1256 (N.D. Fla. 2010) (No. 3:10-cv-91) (listing states that challenged the individual man- date’s constitutionality); Br. Amici Curiae States of Kansas and Nebraska at 14-19, King v. Sebelius, No. 14-1158 (4th Cir. Mar. 10, 2014) (arguing that amici states chose not to establish exchanges in order to avoid mandates). But that possibility ought not to matter to the constitutionality of disparate geograph- ic coverage. If several of the VRA-covered states had wanted federal oversight of their election laws — to avoid political accountability for racially equal prac- tices — would the outcome in Shelby County have been different? The Court’s holding was that “disparate geographic coverage” of a federal law raises constitu- tional questions independently of whether the states like the regulations they get. Shelby County, 133 S.Ct. at 2622. It is differentiation, not dissatisfaction, that causes constitutional trouble. Furthermore, inverting the ACA choice does not eliminate dissatisfaction; it just turns the tables on states that want mandates enforced. From those states’ perspective, the necessity of establishing an exchange becomes punishment for wanting mandate need to administer their exchanges in perpetuity. If the latter, then the burdens of choice look similar. 24 enforcement, rather than mandate non-enforcement being punishment for failure to establish an ex- change. Either way, Petitioners’ interpretation gives rise to disparate geographic coverage of the man- dates, and either way, the differentiation looks puni- tive from some states’ perspective. The fourth difference has ambiguous constitu- tional implications. Under the VRA, the federal government acted against states themselves while under the ACA, it acts against states’ citizens. The ACA binds insurance companies, employers, and individuals, not state governments. On one hand, this difference makes Petitioners’ ACA look less problem- atic because it does not interfere directly with the states qua sovereigns. Cf. Gregory v. Ashcroft, 501 U.S. 452 (1991) (political functions doctrine). Fur- thermore, if the purpose of both statutes is to incen- tivize states to comply with federal policy, then a strategy that targets citizens rather than politicians might be constitutionally preferred. That way, politi- cians will choose between adopting or refusing federal policy based on public rather than personal interests. See Spallone v. United States, 493 U.S. 265 (1990). On the other hand, federal regulations that bind citizens are not actually less restrictive of states’ powers than those that bind governments. Under the Supremacy Clause, U.S. Const. art. VI cl. 2, federal regulations supersede state power regardless, and the ACA’s market reforms embody specific regulatory choices that preempt contrary state efforts. See 42 U.S.C. § 300gg-23 (preemption); Gade v. Nat'l Solid 25 Wastes Mgmt. Ass’n, 505 U.S. 88, 98 (1992) (implied preemption). Granted, the ACA provides for “state innovation waivers” that might allow the states to circumvent such preemption starting in 2017, 42 U.S.C. § 18052, but the waivers are available only with the federal executive’s approval — which looks awfully similar to VRA preclearance. Furthermore, this Court has long held that federalism limitations exist to protect individuals, not states. New York, 505 U.S. at 181 ‘emphasizing that “federalism secures to citizens the liberties that derive from the diffusion of sovereign power”) (quotations and citations omitted). By that logic, the ACA’s operation against citizens ought to make no difference. The ACA’s direct application also raises a sepa- rate constitutional question: whether the federal government may discriminate against individuals on the basis of state citizenship. See U.S. Const. amend. XIV §1; Bolling v. Sharpe, 347 U.S. 497 (1954). Because Congress has never allowed federal enforce- ment of different substantive policies in different states, this Court has never considered whether that kind of disparate coverage of a federal law ought to trigger strict scrutiny under equal protection. Given citizens’ fundamental right to travel among states, see Crandall v. Nevada, 73 U.S. 35 (1868), strict scrutiny would seem appropriate in this context. See generally Plyler v. Doe, 457 U.S. 202, 216-17 (1982) (“(WJe have treated as presumptively invidious those classifications that impinge upon the exercise of a ‘fundamental mght.’”). If the Court accepts 26 Petitioners’ interpretation, insurers and patients could challenge the resulting statute by arguing that Congress may not impinge their fundamental right to choose where they live by purposefully destroying some states’ individual insurance markets. The Court could avoid that challenge by upholding the IRS Rule. 4. Disparate Geographic Coverage as an Incentive for Implementation of Federal Policy Raises Profound Anti- Coercion Concerns. The most difficult constitutional question arising from Petitioners’ interpretation is whether the ACA’s punitive differentiation among the states, particular- ly when combined with the opportunity cost of subsi- dies, is unconstitutionally coercive. There are two components to this question. First, are regulatory incentives ever permissible? Second, if so, is the particular mix of fiscal and regulatory punishments that arises from Petitioners’ interpretation permissi- ble? Both of those questions would be extremely difficult to resolve under current doctrine. First, assuming that mere regulatory differentia- tion does not violate the principle of equal sovereign- ty, can Congress use a threat of such differentiation as an incentive for states to enact federal policies? This Court’s anti-coercion doctrine would be extraor- dinarily difficult — if not impossible — to apply to regulatory punishments. The core doctrinal question for anti-coercion is whether a federal incentive leaves 27 states genuinely free to choose, and the test for voluntariness turns on the magnitude of the incentive as a percentage of state budgets. See Part I.A, supra. That test might not work at all for regulatory pun- ishments. Consider, for example, an alteration to Dole. Imagine that, instead of threatening to withhold 5 percent of a state’s federal highway funds, Congress had threatened to enforce a lower speed limit in non- compliant states. At the time Dole was decided, a wildly unpopular federal speed limit of 55 MPH was in effect nationwide. See National Maximum Speed Law, Pub. L. No. 97-424, 96 Stat. 2097 (1974) (re- pealed by National Highway System Designation Act of 1995, Pub. L. No. 104-59, 109 Stat. 568); see also Paul Grimes, Practical Traveler: The 55-M.P.H. Speed Limit, N.Y. Times, Dec. 26, 1982, http://www. nytimes.com/1982/12/26/travel/practical-traveler-the- 55-mph-speed-limit.htm] (discussing the statute’s unpopularity). What if, in the statute at issue in Dole, 23 U.S.C. § 158, Congress had agreed to increase the federal limit to 65 MPH in states that raised their drinking ages while leaving the unpopular 55 MPH limit to govern non-compliant states? Congress could have simultaneously created a federal highway patrol to enforce the limits in all states, making the threat credible. Would that incentive have been coercive? What if compliant siates got a 65 MPH limit while non-compliant states got a 35 MPH limit? A 10 MPH limit? At least at some point, such a threat must be 28 unconstitutionally coercive, but where is the thresh- old for a regulatory incentive? The problem, given the doctrinal test, is that the effects of a regulatory punishment are much harder to quantify than those of a fiscal punishment. The Court would need to know the net cost (or benefit) of a lower speed limit: lost time and money from slower commutes offset by benefits from greater highway safety. The Court also ought to ask how much citizens hate lower speed limits — a highly relevant value when considering state politicians’ freedom of choice, but one that is extraordinarily difficult to quantify. Whereas fiscal incentives engage decision-makers in a relatively straightforward willingness-to-pay game, regulatory incentives create a muddled game that would be exceptionally difficult for the Court to assess. Of course, it is theoretically possible to quantify economic effects of a regulatory change — like the cost of slower commutes. But such economic effects are off-budget for the states, rendering irrelevant the doctrinal test of a threat’s magnitude relative to state expenditures. Even if the Court had those data, it would need to compare regulatory effects to states’ economies — their GDPs. But that comparison looks nearly impossible. The Court would be seeking macro- economic effects of a microeconomic policy in a world in which economists are either micro-economists or macro-economists. There are not usually data availa- ble on micro policies’ macro consequences. See, e.g., 29 Economic Scholars Br. (discussing only microeconom- ic effects of Petitioners’ interpretation). Given the doctrinal difficulties that arise from regulatory incentives ~ and the constitutional doubts associated with any regulatory differentiation — it might make sense to hold that regulatory threats are unconstitutionally coercive no matter how trivial they might appear. Otherwise, the Court will be left with an “I know it when I see it” test for regulatory coer- cion in a realm that Supreme Court justices are institutionally ill-equipped to judge: state politics. Cf. Jacobellis v. Ohio, 378 U.S. 184, 197 (1964) (Stewart, J., concurring). All of that said, the regulatory threat in this case is far from trivial. The second question, whether this particular threat is unconstitutionally coercive, might therefore be simple — and Petitioners’ interpretation simply unconstitutional — under an “I know it when I see it” standard. Although it seems impossible to determine whether the costs of adverse selection “death spirals” would approach the NFIB threshold of 10 percent of state budgets, see 132 U.S. at 2605, there is broad consensus — that Petitioners do not dispute — that non-enforcement of the mandates would wreak havoc on states’ individual insurance markets. See Economic Scholars Br.; Insurance Plans Br.; Pet’rs’ Br. at 15. As discussed above, Congress included the indi- vidual and employer mandates — despite their un- popularity and despite President Obama’s professed 30 opposition to the measures — as necessary corrections to adverse selection that would otherwise arise from the ACA’s market reforms. See Andrew Cline, How Obama Broke His Promise on: Individual Mandates, The Atlantic, June 29, 2012, http://www.theatlantic.com/ politics/archive/2012/06/how-obama-broke-his-promise- on-individual-mandates/259183/. As Chief Justice Roberts explained in NFIB: The guaranteed-issue and community-rating reforms do not, however, address the issue of healthy individuals who choose not to pur- chase insurance to cover potential health care needs. In fact, the reforms sharply exac- erbate that problem, by providing an incen- tive for individuals to delay purchasing health insurance until they become sick, re- lying on the promise of guaranteed and af- fordable coverage. The reforms also threaten to impose massive new costs on insurers, who are required to accept unhealthy indi- viduals but prohibited from charging them rates necessary to pay for their coverage. This will lead insurers to significantly in- crease premiums on everyone. The individu- al mandate was Congress’s solution to these problems. 132 S.Ct. at 2585. Indeed, all Members of the Court accepted this account in their NFIB opinions. /d.; id. at 2614 (Ginsburg, J., concurring in part and dissenting in part); id. at 2670-71 (joint dissent). 31 Given the mandates’ accepted role as correctives to the ACA’s otherwise-destructive regulations and given that Petitioners’ interpretation would cause non-enforcement of those correctives in non- compliant states, the threat in Petitioners’ jnterpreta- tion is: “Establish an exchange, or the federal gov- ernment will destroy your individual health insurance market.” The question, then, is whether that threat is coercive. Would a state be willing, in the long run, to endure the consequences of federally- created “death spirals” in order to avoid the federal command to establish an exchange” Because the fiscal magnitude of this threat seems impossible to calculate, consider a slightly altered metric: whether, under Petitioners’ interpretation, it is possible for a state to achieve a functioning * Several states argue that they wanted to avoid the mandates, implying that they would be not only capable of resisting the threat but happy to accept its consequences. See Br. Amici Curiae State of Oklahoma et al. 2, 14 (Oklahoma, Alabama, Georgia, Nebraska, South Carolina, and West Virgin- ia); Br. Amici Curiae State of Indiana and 39 Indiana Public School Corporations. These states’ filings, however, do not embrace the destructive regulatory regime that would emerge from that decision. None admits that the cost of avoiding mandates is, at least plausibly, destruction of insurance mar- kets. Furthermore, it is not clear that some states’ preference for the “punishment” should matter to the constitutional! question. If Congress intended to make a coercive offer but miscalculated some states’ preferences, the offer might nevertheless be uncon- stitutional. Indeed, it is unclear from NFIB and Dole whether an offer is unconstitutional if any state would be unable to refuse, if every state would, or if an objectively reasonable state would. 32 individual market without establishing an exchange. If not, then Petitioners’ ACA puts the states in a “heads I win, tails you lose” game; the state either loses its individual market or complies with federal policy. Indeed, states’ ability to surmount federal punishments seems to be the heart of the Court’s analysis in Dole and NFIB. States could plausibly make up .05 percent budget shortfalls by increasing state taxes — if citizens were willing to pay for their lower drinking age — but they could not plausibly make up 10 percent shortfalls, no matter how much citizens hated the Medicaid expansion. The Court's focus on budget effects seems to be, fundamentally, about whether states have autonomous options for overcoming the federal punishment without obeying the federal command. In this case, the design of the regulatory pun- ishment is such that autonomous state remedies are entirely unavailable: two legally, one practically. Two fixes are legally impossible. First, states cannot excuse their insurance companies from the market reforms, which preempt contradictory state laws. See 42 U.S.C. § 300gg-23. The states cannot simply opt out of the federal program in favor of running their own individual insurance markets — the way that they could have run their own equivalents to Medi- caid in 1965 or paid for their own highways in 1987. Second, states cannot fix their markets by providing state-funded subsidies, making up the $2 billion themselves, because the mandates both explicitly turn on federal subsidies. 26 U.S.C. § 5000A(e)(BXii) 33 (turning on “the credit allowable under section 36B”); 26 U.S.C. §4980H (requiring certification of an employee's eligibility for ACA subsidies). State subsi- dization would not earn non-compliant states the same federal regulatory treatment as compliant states. Indeed, subsidies in the absence of mandates would only exacerbate adverse selection problems by decreasing the cost of purchasing a policy once sick. The states would have three legally possible options to keep their individual markets afloat, only one of which would count as autonomous state action. First and most obviously, states could establish exchanges, obeying the federal command. Second, they could write and enforce their own mandates. That approach, however, would be politically costly and ultimately pointless; for many states, enforcing mandates would be just as much a concession to federal coercion as establishing an exchange. Fur- thermore, establishing an exchange would be no more expensive than enforcing mandates, and the exchange- establishing route would create clearer federal ac- countability for the mandates. Fourth, non-participating states might be able to convince insurers to keep writing policies in their markets, despite rampant adverse selection, by giving them industry-side subsidies. The medical loss ratio limits insurers’ administrative costs, including prof- its, to 20 percent of revenues, but the provision includes only premiums in revenues and excludes “State taxes” from costs. 42 U.S.C. § 300gg-18. It is therefore possible that states could provide tax 34 rebates to insurers, which would count as neither revenue nor profit under the medical loss ratio. But this approach seems fiscally impractical. Insurers might demand more than the $2 billion in forgone federal subsidies to operate in non-compliant states. Those states would not just be trying to com- pete with compliant states (which will have admira- bly predictable markets); they would be trying to convince insurers to work in markets perversely designed to promote adverse selection. Furthermore, non-compliant states’ citizens would end up paying for such industry subsidies in addition to paying, through their federal taxes, for the federal subsidies in compliant states. See NFIB, 136 S.Ct. at 2661-62 (joint dissent). This solution would also fail to avoid the costs of increased uncompensated care that would arise from eliminating the mandate, which might be substantial given that many individ- uals might lose their bets that they can stay healthy between the federal exchange’s annual enrollment periods and given that premiums in adverse selection markets would be too high for many people to afford. See Matthew Buettgens & Caitlin Carroll, Eliminat- ing the Individual Mandate: Effects on Premiums, Coverage, and Uncompensated Care, Urban Institute, Jan. 2012, http://www.urban.org/U ploadedPDF/412480- Eliminating-the-Individual-Mandate.pdf (predicting a $23 billion nationwide increase in uncompensated care). The overall cost of this approach could thus include: more than $2 billion in subsidies plus wasted federal taxes for other states’ subsidies plus economic 35 costs associated with uncompensated care. That’s almost certainly more than 10 percent of an average state’s budget. All told, the only theoretically possible approach to mitigating death spirals looks prohibitively expen- sive in fact. Cf. Dole, 483 U.S. at 211-12 (requiring states to have a choice “not merely in theory but in fact”). The burden of Petitioners’ regulatory punishment may be impossible to cast in precise dollar terms, but it seems awfully big, especially when combined with the $2 billion opportunity cost of forgone subsidies. Cumulatively, the threat embedded in Petitioners’ reading could be as destructive as the Medicaid threat that this Court invalidated in NFIJB or even more so. Fortunately, the Court can avoid confronting that question by upholding the IRS Rule. All of that said, if the Court would like more information on the effects of Petitioners’ interpreta- tion on states’ insurance markets in order to assess the seriousness of the constitutional doubt, amici respectfully suggest that the Court order the parties to be prepared to discuss the question at oral argu- ment. The parties are better situated than amici to evaluate the effects of Petitioners’ interpretation on state economies. 36 Il. THE GOVERNMENT'S INTERPRETATION OF §36B IS AT LEAST “FAIRLY POSSI- BLE” AND AVOIDS ANY CONSTITUTION- AL DOUBT. Under the canon of constitutional avoidance, this Court will not “impute to Congress an intent to pass legislation that is inconsistent with the Constitution as construed by this Court.” United States v. X- Citement Video, Inc., 513 U.S. 64, 73 (1994). In other words, the Court will not hold that Congress intended “to disregard a constitutional danger zone,” Yates v. United States, 354 U.S. 298, 319 (1957), unless that intent is so clear as to be unavoidable. Furthermore, even when a constitutionally problematic construc- tion is “the most natural interpretation” of a statute, “‘every reasonable construction must be resorted to, in order to save a statute from unconstitutionality.’” NFIB, 132 S.Ct. at 2594 (quoting Hooper v. Califor- nia, 155 U.S. 648, 657 (1895)). If “a construction of the statute is fairly possible by which the [constitu- tional] question may be avoided,” this Court favors that interpretation. Crowell, 285 U.S. at 62. A. The Government’s Interpretation of § 36B is at Least “Fairly Possible.” In this case, six of the nine federal judges to have confronted the statute have held that the govern- ment’s interpretation is more consistent with the text and structure of the ACA. See Halbig, 758 F.3d at 412-27 (Edwards, J., dissenting); King, 759 F.3d 358 (unanimous opinion); Halbig v. Sebelius, 113 37 A.F.T.R.2d 2014-548 (D.D.C. 2014); King v. Sebelius, 997 F. Supp 2d 415 (E.D. Va. 2014). Their analyses and conclusions demonstrate that the government’s interpretation is, at minimum, “fairly possible.” Crowell, 285 U.S. at 62. Furthermore, the statutory definition of “exchange,” 42 U.S.C. § 300gg-91(d\(21), and the reference in 42 U.S.C. § 18041 to “such exchange” lend textual plausibility to the govern- ment’s interpretation, and Petitioners agree that nothing in the legislative history contradicts that construction. Indeed, there is no evidence anywhere in the statute’s text, structure, or history of a clear congres- sional intent to induce state compliance through a threat of regulatory differentiation — and thereby “to disregard [two] constitutional danger zone[s].” Yates, 354 U.S. at 319. The language of § 36B provides, at most, evidence of an intent to use subsidies as an incentive; the regulatory consequer.:es of Petitioners’ interpretation are domino e‘fects that do not seem to have been intended at ali. Even if the Court con- cludes that Petitioners’ is “the most natural inierp. »- tation” of § 36B, this Court should disfavor it to avoid the constitutionally problematic dominos.” * Importantly, the government's interpretation does not raise its own constitutional problems. Cf. Rust v. Sullivan, 500 U.S. 173, 191 (1991) (“{I]}t was likely that any set of regulations funder the statute] ... would be challenged on constitutional grounds.”). In an amici brief, the State of Indiana and some of its schools argue that the government’s interpretation causes (Continued on following page) 38 B. The Government’s Interpretation Avoids the Serious Constitutional Problems that Infect Petitioners’ Interpretation. The IRS Rule successfully avoids all of the consti- tutional infirmities of Petitioners’ interpretation. Under the government’s construction, the ACA gives states two incentives to establish their own exchang- es: a grant to assist with exchange “planning and establishment,” 42 U.S.C. § 18031(a), and regulatory flexibility within established exchanges, 42 U.S.C. § 18031. Neither of these incentives raises an anti- coercion concern or an affront to equal sovereignty. constitutional problems by imposing the employer mandate on the states. See Br. Amici Curiae State of Indiana and 39 Indiana Public School Corporations 20-34. Indiana’s constitutional arguments are substantively specious, one even resting on an overruled opinion. Id. at 31-34 (relying on Natl League of Cities v. Usery, 426 U.S. 833 (1976), overruled by Garcia v. San Antonio Metro. Transit Auth., 469 U.S. 528 (1985)). But setting that aside, the biggest problem with Indiana’s arguments is that Petitioners’ interpretation would not cure the alleged constitu- tional defects. If it is unconstitutional for Congress to regulate or tax the states, then that problem cannot be solved by letting the states choose whether they will be bound by the unconstitu- tional provisions. The remedy would be to read the definition of “employer” narrowly to exclude the states from the ACA’s coverage given that the ACA does not explicitly include states in that definition. See Gregory v. Ashcroft, 501 U.S. 452, 460-61 (1991) (applying a “plain statement rule” to regulation of states as employers); 26 U.S.C. § 4980H(cX2XA) (defining “applicable large employer”). The Court cannot avoid the constitutional trouble Indiana identifies by holding in Petitioners’ favor here. 39 The statute’s “planning and _ establishment grants” were for statutorily unspecified amounts, 42 U.S.C. § 18031(a\1), but the actual amounts turned out to be an average of about $270 million per state — an order of magnitude less than § 36B’s premium subsidies. See Cong. Research Serv., Federal Funding for Health Insurance Exchanges, Oct. 14, 2014, https://www.hsdl.org/?view&did=759147. Crucially for constitutional purposes, the start-up grants were temporally limited to the first year of the exchanges’ operation; the statute prohibits the federal govern- ment from distributing grants after January 1, 2015. 42 U.S.C. § 18041(a)(4)(B). All of the fiscal threats and offers that this Court has considered under the anti-coercion principle, including the § 36B subsidies, have been perpetual grants. That distinction makes a big difference to the irresistibility of an offer. The ACA’s time-limited planning and establishment grants, as an incentive to establish exchanges, do not raise anything like the constitutional doubt that the subsidies and regulatory differentiation would raise if treated as incentives. The other incentive that the government’s con- struction embraces — regulatory flexibility within the exchange — might look superficially problematic under the » inciple of equal sovereignty, but it is critically different from Petitioners’ regulatory incen- tives. Regulatory flexibility obviously allows for geographic differentiation, but it is state-created, bottom-up differentiation — the same kind that is pervasive in cooperative federalism, not the novel 40 federally-created, top-down kind that arises from Petitioners’ interpretation. Regulatory differentiation through exchange management would be state- enacted and state-enforced; it would not allow differ- ent federally-enforced policies in different states. It would not, in other words, cause the federal govern- ment to treat differently the equal sovereigns of its union. Under the government’s understanding, states that do not want the burden of establishing and administering exchanges can avoid that burden by relying on the federal government to run the ex- change for them, avoiding any anti-commandeering problem under New York, 505 U.S. 144, and Printz v. United States, 521 U.S. 898 (1997). But non- participating states cannot avoid any element of the substantive regulatory regime that the ACA created for all states. Although some states dislike the sub- stantive policy that Congress chose, the inescapabil- ity of that policy is necessary to avoid the kind of differentiation that this Court found constitutionally troublesome — and ultimately unconstitutional — in the VRA. If this Court is serious about the principle of equal sovereignty, it must apply that principle when some states dislike uniformity just as it has applied it when some states prefer uniformity. The federal government must treat all states the same, even if many states dislike the result. + 41 CONCLUSION For the foregoing reasons, we urge the Court to avoid the two distinct constitutional challenges that would arise from Petitioners’ reading — under equal sovereignty and anti-coercion — by affirming the Fourth Circuit. Respectfully submitted, ANDREW M. FISCHER Counsel of Record JASON & FISCHER 47 Winter Street Boston, MA 02108 telephone (617) 423-7904 fax (617) 451-3413 afischer@jasonandfischer.com --- Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/brief%3Amicro_IA40385016_0338%3A44. Public record. Not legal advice.