Agencies, Polarization, and the States

Agencies, Polarization, and the States

Administrative agencies are strikingly absent from leading accounts of contemporary polarization. To the extent they appear, it is largely as acted-upon entities that bear the fallout from the congressional–presidential confrontations that polar­ization fuels, or as the tools of presidential unilateralism. This failure to incorporate administrative agencies into polarization accounts is a major omission. Agencies possess broad grants of preexisting authority that they can use to reshape governing policy and law, often at presidential instigation, thereby putting pressure on Congress to respond. In the process, they can construct new alliances and arrangements that have the potential to break through partisan divides and alter the political landscape.

This Essay, written in honor of Peter Strauss, aims to deepen current under­standing of the relationship between administrative agencies and political polari­zation. Using Professor Strauss’s insight that agencies exist in a web of control relationships as a launching point, the Essay probes how polarization affects these control relationships and the extent to which agencies can develop policy and potentially reshape partisan divides. Yet focusing simply on the forces empowering or constraining agencies at the federal level misses the critical element of state participation in federal programs and federal regulation. Such state involvement injects a political edge into program implementation and provides a potential mechanism for checking executive branch unilateralism at the same time that it opens up opportunities for bipartisanship. After setting out a conceptual account of the relationship between agencies, polarization, and the states, the Essay examines implementation of the Affordable Care Act (ACA) to assess how these complicated dynamics operate in practice.


    1. Agencies as Acted-Upon Entities: Congressional Stalemate and Presidential Administration
      1. Congressional Sidelining and the Importance of Congressional Inaction
      2. The Further Rise of Presidential Administration
    2. B. Agencies as Actors: Policy Development and Partisan Realignment
      1. Polarization and Agency Policy Development
      2. Judicial and Internal Executive Branch Controls
      3. Agencies’ Impact on Polarization
    3. Incorporating Federalism
    1. Web of Controls: Congress, the Executive Branch, and the Courts
      1. Congressional Absence
      2. Presidential Presence
      3. The Centrality of the Courts
    2. Federalism, Partisanship, and the ACA
      1. The States and ACA Implementation
      2. Reshaping the Political Terrain



Political polarization is all the rage. Both popular and scholarly voices regularly bemoan the depths of partisanship and division to which our national politics have sunk. Assessments of causes and possible cures abound. In Terminal Congressional Dysfunction?, Cynthia Farina offers a comprehensive and insightful analysis of this burgeoning field. 1 Cynthia R. Farina, Congressional Polarization: Terminal Constitutional Dysfunction?, 115 Colum. L. Rev. 1689 (2015). Yet as Professor Farina notes, for all the ink increasingly spilled, key aspects of polarization’s dynamics remain unclear. 2 See id. at 1717–33 (assessing several explanations for political polarization with inconclusive findings). A particularly big hole concerns the relationship between polarization and administrative agencies.

Administrative agencies are strikingly absent from leading accounts of contemporary polarization. 3 For a discussion of recent administrative law scholarship that has begun to explore the role of agencies in today’s polarized politics, see infra text accompanying notes 88–93. The focus instead is on Congress, the President, and voters. 4 See, e.g., Nolan McCarty, Keith T. Poole & Howard Rosenthal, Polarized America: The Dance of Ideology and Unequal Riches 3 (2006) (documenting increased polarization and its relationship to national trends, in particular growing income disparity); Sarah Binder, The Dysfunctional Congress, 18 Ann. Rev. Pol. Sci. 85, 95–96 (2015) [hereinafter Binder, Dysfunctional Congress] (analyzing polarization’s impact on Congress and President); Gary C. Jacobson, Partisan Polarization in American Politics: A Background Paper, 43 Presidential Stud. Q. 688, 691–700 (2013) (examining how changes in voters have affected congressional parties); Nathaniel Persily, Introduction, in Solutions to Political Polarization in America 3, 10, 13 (Nathaniel Persily ed., 2015) (suggesting reforms to voting, political parties, and congressional rules to address growing polarization). For a rare exception that focuses on the policy implications of polarization, see Nolan McCarty, The Policy Effects of Political Polarization, in Transformation of American Politics: Activist Government and the Rise of Conservatism 223, 232–46 (Paul Pierson & Theda Skocpol eds., 2007) [hereinafter McCarty, Policy Effects] (arguing polarization moves policymaking to courts and states). To the extent agencies appear, they surface largely as acted-upon entities who bear the fallout from the congressional–presidential confrontations that polarization fuels. Scholars emphasize how congressional polarization has held up appointments of top agency officials, 5 See infra note 50 and accompanying text. created budget uncertainty for agencies, 6 See infra notes 45–48, 55 and accompanying text. and subjected them to increased investigations. 7 See infra note 52 and accompanying text. Agencies play a somewhat more active role in accounts of presidential unilateralism, but here, too, they feature primarily as tools of the President rather than as policy initiators in their own right. 8 See Kenneth S. Lowande & Sidney M. Milkis, “We Can’t Wait”: Barack Obama, Partisan Polarization and the Administrative Presidency, 12 Forum 3, 5 (2014) (describing recent trend of Presidents using administrative power to serve partisan goals). Hence, the ongoing rulemaking by the Environmental Protection Agency (EPA) on power plant emissions is portrayed and attacked as President Barack Obama’s climate change plan, 9 See, e.g., Matthew Daly, GOP Lawmaker Targets Climate Plan, Wash. Times (Mar. 23, 2015), [] (describing EPA’s proposed rules as “President Barack Obama’s strategy”). while the recent immigration enforcement initiatives promulgated by the Department of Homeland Security (DHS) are commonly referred to as President Obama’s immi­gration executive actions—including by the White House—despite being embodied in memoranda issued by the Secretary of Homeland Security, Jeh Johnson. 10 E.g., Max Ehrenfreund, Your Complete Guide to Obama’s Immigration Executive Action, Wash. Post: Wonkblog (Nov. 20, 2014),
blog/wp/2014/11/19/your-complete-guide-to-obamas-immigration-order/ []; Fixing the System: President Obama is Taking Action on Immigration, White House, [] (last visited Aug. 13, 2015); see also Memorandum from Jeh Charles Johnson, Sec’y, Dep’t of Homeland Sec., on Policies for the Apprehension, Detention and Removal of Undocumented Immigrants to Thomas S. Winkowski, Acting Dir., U.S. Immigration & Customs Enf’t et al.(Nov, 20, 2014), [] (outlining new immi­gration enforcement policies).

The failure to incorporate administrative agencies into polarization accounts is a major omission. Administrative government, and partic­ularly regulatory government, fundamentally transforms the polarization equation. Indeed, the presence of an extensive national administrative state marks a signal difference between the nation’s current situation and prior instances of high polarization, such as in the period from 1890 to 1910, when modern administrative agencies were nascent. 11 See Farina, supra note 1, at 1702–03 (describing earlier period of significant polarization in Congress); see also Robert L. Rabin, Federal Regulation in Historical Perspective, 38 Stan. L. Rev. 1189, 1192–95 (1986) (describing chronological development of national regulatory state). Although agencies are clearly affected by the hyperpartisanship that dominates the political branches, they are still able to act. Agencies possess broad grants of preexisting authority that they can use to reshape governing policy and law, often at presidential instigation, thereby putting pressure on Congress to respond. In the process, they can construct new alliances and arrangements that have the potential to break through partisan divides and alter the political landscape. If nothing else, agencies’ pre­existing powers mean that the policy gridlock produced by polarization at the political level does not forestall policy development altogether.

Importantly, agencies are not simply pawns in a battle between the two parties or institutional struggle among the political branches of national government. To be sure, polarization has reinforced the already strong trend toward presidential administration, as Presidents seek to use agencies to advance partisan policy agendas stymied by congressional stalemate. 12 See infra notes 60–72 and accompanying text; see also Lowande & Milkis, supra note 8, at 3–6, 8–14 (arguing Obama Administration has expanded presidential admin­istration and embraced unilateralism for partisan ends). In turn, Congress increasingly treats executive agencies as presidential surrogates and fair partisan game. 13 See Andrew McCanse Wright, Constitutional Conflict and Congressional Oversight, 98 Marq. L. Rev. 881, 885–89 (2014) (linking government oversight and inves­tigation with partisan disputes). Thus, the increased focus on specifically presidential unilateralism in polarization contexts reflects real-life dynamics. But as Peter Strauss has emphasized repeatedly and powerfully, most recently in Overseer, or “The Decider”? The President in Administrative Law, 14 Peter L. Strauss, Foreword: Overseer, or “The Decider”? The President in Administrative Law, 75 Geo. Wash. L. Rev. 696 (2007). agencies cannot simply be equated with the President. 15 See id. at 700, 712–15 (making this distinction with respect to implications of delegating power to agencies and President); see also Peter L. Strauss, The Place of Agencies in Government: Separation of Powers and the Fourth Branch, 84 Colum. L. Rev. 573, 582–96 (1984) [hereinafter Strauss, Place of Agencies] (describing complex relationships between President and agencies and factors enhancing as well as limiting presidential influence). Agencies have independent stature, responsibilities, and allegiances, and they also have their own policy agendas that they seek to advance on the political branches. 16 See Brigham Daniels, Agency as Principal, 48 Ga. L. Rev. 335, 340–43 (2014) (recognizing circumstances when agencies act as principals capable of “turn[ing] the tables” on political branches); see also Elena Kagan, Presidential Administration, 114 Harv. L. Rev. 2245, 2272–74, 2306–09 (2001) (describing expanding presidential oversight and limits to close presidential control of agencies under President Clinton); Gillian E. Metzger, The Interdependent Relationship Between Internal and External Separation of Powers, 59 Emory L.J. 423, 429–32, 432–37 (2009) (describing internal checks on presi­dential power and interaction between agencies and external institutions).

Whether agencies are in fact able to develop policy in the face of polarization depends on a number of factors. One central consideration is the scope of an agency’s extant authority. 17 See Jody Freeman & David B. Spence, Old Statutes, New Problems, 163 U. Pa. L. Rev. 1, 3 (2014) (“[F]ederal agencies coping with new regulatory challenges often encounter problems of ‘fit’ with older statutes . . . .”). Another is an agency’s internal make-up and character. Agencies are a diverse lot, varying in leadership structure, political independence, institutional capacity, resources, and reputation. 18 See Daniel P. Carpenter, The Forging of Bureaucratic Autonomy: Reputations, Networks, and Policy Innovation in Executive Agencies, 1862–1928, at 23–33 (2001) (describing role of professional networks, reputation, and agency culture in agencies); Kirti Datla & Richard L. Revesz, Deconstructing Independent Agencies (and Executive Agencies), 98 Cornell L. Rev. 769, 786–812 (2013) (describing different structural features among independent agencies as well as between independent and executive agencies). Agencies also contain a range of actors and interests within their midst—political appointees and civil servants; administrators as well as professionals such as scientists and lawyers; front-line personnel and supervisors—often with different responsibilities, priorities, and alle­giances. 19 See Elizabeth Magill & Adrian Vermeule, Allocating Power Within Agencies, 120 Yale L.J. 1032, 1036–38 (2011) (concluding “views and actions of different types of stakeholders shape the agency’s performance of its duties”); Jon D. Michaels, An Enduring, Evolving Separation of Powers, 115 Colum. L. Rev. 515, 538–47 (2015) (describing differences between agency leadership and civil service personnel). Not surprisingly, some agencies are likely to be more able and willing to push a policy agenda than others.

Equally important, however, is the complicated external web of relationships in which an agency operates. These relationships include not only agencies’ interactions with the named national branches at the apex of government—Congress, the President, and the courts—but also their interactions with state and local governments, other administrative entities, regulated parties, beneficiaries, or other interest groups. Again, Professor Strauss’s scholarship provides critical guidance in under­standing this web of agency control relationships, with Strauss empha­sizing the importance of these control relationships to the functioning of the modern administrative state. 20 Strauss, Place of Agencies, supra note 15, at 583–96. Today, these control relationships are equally central to assessing how agencies operate in a polarized world and the extent to which polarization at the apex of government trickles down to the level of administration.

Polarization, or at least polarization combined with divided govern­ment, warps this web of agency controls in significant ways. It hobbles proactive legislative direction, with the result that congressional oversight is exercised most frequently by inaction, delay, and budget constraints. These moves frustrate agencies’ ability to function and meet emerging regulatory challenges, but also create new opportunities for unilateral executive branch action justified by necessity. The decline in legislative controls also serves to heighten the salience of executive and judicial constraints for agencies. Courts in particular face new challenges in monitoring the legality and rationality of agency action when agencies act in the face of legislative stalemate or at the behest of presidential initiative.

Yet focusing simply on the forces empowering or constraining agencies at the federal level misses the critical element of state partic­ipation in federal programs and federal regulation. Such state involve­ment injects a political edge into program implementation. Politically sympathetic states provide a means by which presidents and agencies can advance policy goals over federal opposition. In a polarized world, however, the role of politically opposed states in federal programs may be more significant. Such state involvement can check executive branch uni­lateralism at the same time that it opens up opportunities for bipartisanship.

The aim of this Essay is to deepen the current understanding of the relationship between administrative agencies and political polarization. Its goals are primarily analytic and descriptive. But it is also motivated by the belief that government should address the policy challenges of the day. In the face of congressional gridlock and national political stale­mate, agencies’ continued ability to develop policy is a virtue, not a vice—provided adequate checks exist to ensure, in Professor Strauss’s words, that agencies “will not pass out of control.” 21 Id. at 579.

Part I offers a conceptual account of the relationship between agencies, polarization, and the states. It begins by setting out the back­ground of delegation and control that are the driving forces of the modern administrative state. It then turns to examining how these forces interact with polarization, focusing in particular on how polarization impacts the web of agency control relationships and the extent to which agencies can develop policy and potentially reshape partisan divides. Part II discusses implementation of the Affordable Care Act (ACA) to provide an illustration of how these complicated dynamics operate in practice.

I. Conceptualizing the Role of Agencies in a Polarized World

Two opposing imperatives lie at the heart of the modern admin­istrative state. One is the need for Congress to delegate broad regulatory authority to agencies. 22 See Gillian E. Metzger, Delegation, Accommodation, and the Permeability of Constitutional and Ordinary Law, in The Oxford Handbook of the U.S. Constitution 409, 412–13 (Mark Tushnet, Mark A. Graber & Sanford Levinson eds., 2015) (describing centrality of delegation). Debate surrounds whether this “need” reflects Congress’s lack of the information and expertise necessary to address complicated policy issues, its structural inability to respond quickly and flexibly to new and emergent problems, or its desire to punt politically contentious issues to another body. 23 Id. For a recent article arguing that an additional impetus behind broad delegations is to allow individual members of Congress to influence agencies’ exercises of discretion at the expense of Congress as a whole, see Neomi Rao, Administrative Collusion: How Delegation Diminishes the Collective Congress, 90 N.Y.U. L. Rev. (forthcoming 2015) (manuscript at 15–16) (on file with the Columbia Law Review). Indeed, some deny the consti­tutional legitimacy of broad policymaking delegations altogether. 24 See Gary Lawson, The Rise and Rise of the Administrative State, 107 Harv. L. Rev. 1231, 1237–43 (1994) (“[T]he nondelegation principle, which is textually embodied in the command that all executory laws be ‘necessary and proper,’ constrains the substance of congressional enactments.”); see also Dep’t of Transp. v. Ass’n of Am. R.R.s, 135 S. Ct. 1225, 1242–52 (2015) (Thomas, J., concurring) (stating “Constitution categorically forbids Congress to delegate its legislative power to any other body”). But the Supreme Court has refused to meaningfully police the bounds of congressional delegations since the early New Deal, and as a matter of practical reality, “[a] great deal of national lawmaking has been dele­gated by Congress to administrative agencies.” 25 William N. Eskridge, Jr. & John Ferejohn, The Article I, Section 7 Game, 80 Geo. L.J. 523, 534 (1992) (footnote omitted); see also Whitman v. Am. Trucking Ass’n, 531 U.S. 457, 474–75 (2001) (“[W]e have almost never felt qualified to second-guess Congress regarding the permissible degree of policy judgment that can be left to those executing or applying the law.” (internal quotation marks omitted)). Moreover, these dele­gations to agencies are ongoing until altered through new legislation. Given the significant obstacles to legislative enactment at the national level—in particular, the requirements of bicameralism and presentment combined with supermajoritarianism, with a two-thirds vote in each chamber required to overcome a presidential veto 26 U.S. Const. art. I, § 7. and sixty votes to end a Senate filibuster 27 S. Doc. No. 113-18, Rule XXII, at 15 (2013). —delegations to agencies prove quite durable, even when agencies promulgate policies with which a majority in Congress disagrees. 28 See Eskridge & Ferejohn, supra note 25, at 536–38 (“[T]he agency can (perhaps over time) set policy virtually anywhere it wants, unless Congress would be stimulated to override the agency’s choice by enacting new legislation.”).

The second imperative follows from the first: The delegation of regulatory authority creates a need for mechanisms to control its exercise. Multiple control mechanisms, connecting agencies to a variety of overseers, create the complex structures and interrelationships that characterize the administrative state. Here, debate exists over where such control is most commonly and appropriately wielded: Positive political theorists give pride of place to Congress, arguing that Congress structures delegations and imposes procedural requirements so as to ensure that, in exercising its new powers, an agency does not deviate from the policy preferences of the coalition that got the measure enacted. 29 Matthew D. McCubbins, Roger G. Noll & Barry R. Weingast, Structure and Process, Politics and Policy: Administrative Arrangements and the Political Control of Agencies, 75 Va. L. Rev. 431, 440–41 (1989) [hereinafter McNollgast]; see also Lisa Schulz Bressman, Procedures as Politics in Administrative Law, 107 Colum. L. Rev. 1749, 1767–71 (2007) (describing positive political theory argument). They are countered by executive power theorists who contend that primary control of administration increasingly resides in the President. 30 See Terry M. Moe, An Assessment of the Positive Theory of ‘Congressional Dominance,’ 12 Legis. Stud. Q. 475, 489–90 (1987) [hereinafter Moe, Assessment of Positive Theory] (criticizing congressional dominance approach for, inter alia, not accounting for presidential control of administration through appointments); see also Terry M. Moe & William G. Howell, The Presidential Power of Unilateral Action, 15 J.L. Econ. & Org. 132, 137–38, 141–48 (1999) (arguing Presidents’ ability to act unilaterally and exploit ambiguities in scope of control over executive branch gives them powerful advantages over Congress); Keith E. Whittington & Daniel P. Carpenter, Executive Power in American Institutional Development, 1 Persp. on Pol. 495, 496–99 (2003) (critiquing congressional dominance view and arguing for more dynamic approach viewing congressional and executive power as interactive). More traditional accounts stress the central administrative oversight role of courts 31 See Louis L. Jaffe, Judicial Control of Administrative Action 320 (1965) (“The availability of judicial review is the necessary condition, psychologically if not logically, of a system of administrative power which purports to be legitimate, or legally valid.”). and interest groups, 32 See Richard B. Stewart, The Reformation of American Administrative Law, 88 Harv. L. Rev. 1667, 1670 (1975) (“Increasingly, the function of administrative law is not the protection of private autonomy but the provision of a surrogate political process to ensure the fair representation of a wide range of affected interests in the process of administrative decision.”). as well as the constraining force of agency expertise, professionalism, and the civil service. 33 See id. at 1675 (describing traditional agency expertise view); see also Michaels, supra note 19, at 530–56 (describing role of agency expertise, civil service, and professionalism in maintaining separation of powers). In recent years, scholars have expanded their account of internal executive branch constraints on agency action, highlighting features of admin­istrative structure, such as interagency coordination and consultation requirements. 34 See, e.g., Jody Freeman & Jim Rossi, Agency Coordination in Shared Regulatory Space, 125 Harv. L. Rev. 1131, 1139–45 (2012) (detailing phenomenon of multiple agency delegations and resultant interagency coordination and consultation requirements); Jason Marisam, Duplicative Delegations, 63 Admin. L. Rev. 181, 231–36 (2011) (arguing presence of agencies with duplicative duties enhances separation of powers by affording Presidents more discretion to allocate tasks among agencies than delegation to a single agency would provide, but more constraints and accountability than follows when Presidents are delegated power directly). Put together, these accounts weave a picture of agencies as situated in the midst of a dense and evolving web of relationships that serve to both empower and constrain administrative action.

This web of relationships was the central focus of Professor Strauss’s The Place of Agencies in Government: Separation of Powers and the Fourth Branch. Unlike other scholars, who often prioritized one branch of national government over the others, Professor Strauss insisted that the structural imperative of our separation-of-powers system was that agencies must be subject to control relationships with each of the named branches. The place of agencies was thus at the center of a web of overlapping and crosscutting controls among Congress, the President, and the courts. 35 See Strauss, Place of Agencies, supra note 15, at 577–80 (“What we have, then, are three named repositories of authorizing power and control, and an infinity of insti­tutions to which parts of the authority of each may be lent.”). Equally important was Professor Strauss’s claim that the Constitution left these controls largely unspecified, with the details of agencies’ relationships with the named branches being determined as much by politics as by law. 36 Id. at 592–97, 640–42. According to Professor Strauss, consti­tutional separation of powers requires simply that the three named branches “share the reins of control; means must be found of assuring that no one of them becomes dominant.” 37 Id. at 580. But this imperative allows for a “profusion of forms, each related in significant ways to Congress, President and Court,” with politics controlling how these oversight relationships operate in practice. 38 Id. at 592, 596.

Heightened political polarization has profound effects on the web of controls governing agency actions. This Part analyzes these effects, beginning in section I.A with how polarization affects congressional and presidential constraints on agencies. Congress turns to oversight mech­anisms and strategies of delay in lieu of control through new substantive statutes, while Presidents increasingly direct administration to achieve partisan goals. As section I.B next describes, an additional important effect of polarization is to create room for agency policy development. Such agency initiatives encounter judicial and internal executive branch constraints, and are particularly notable for their potential to realign partisan divides both in Congress and between Congress and the President. Section I.C then discusses the increased role that states play in shaping federal programs, which is a major contributor to the partisan realignment potential of agency action. Together, the varied effects of polarization on agencies demonstrate the continued vitality of Professor Strauss’s construct. At the same time as polarization weakens some strands of the web of controls surrounding agencies, it reinforces others and generates opportunities for new relationships to emerge.

A. Agencies as Acted-Upon Entities: Congressional Stalemate and Presidential Administration

1. Congressional Sidelining and the Importance of Congressional Inaction. — Perhaps the most immediate effect of polarization, combined with divided government and supermajority requirements, is congressional gridlock. Congress becomes unable to direct agencies through enactment of substantive legislation. 39 E.g., Freeman & Spence, supra note 17, at 2, 4, 14–16; McCarty, Policy Effects, supra note 4, at 223–24, 233–36. Divided government may not impede new legis­lation significantly when the parties are ideologically diverse internally and party control is therefore limited. 40 The impact of divided government on congressional performance is a matter of some debate. Compare David R. Mayhew, Divided We Govern: Party Control, Lawmaking, and Investigations, 1946–2002, at xii, 220–26 (2d ed. 2005) (finding divided government did not affect overall volume of major legislation but noting greater conflict and more investigations during period from 1991 to 2002), and Sarah A. Binder, The Dynamics of Legislative Gridlock, 1947–96, 93 Am. Pol. Sci. Rev. 519, 527 (1999) (concluding divided government affects “ability of the political system to address major public problems”), with Stephen Ansolabehere, Maxwell Palmer & Benjamin Schneer, Divided Government and Significant Legislation: A History of Congress from 1789–2010, at 2–6, 17–28 (Apr. 23, 2015) (unpublished manuscript), [] (describing scholarship and concluding divided versus unified government has substantial impact on production of significant legislation, but cannot explain broad trends in legislation throughout Congress’s history). But as the parties become more ideologically pure and bipartisan compromise disappears, lack of single-party control of Congress and the presidency can create a substantial roadblock to legislative enactment. 41 See Binder, Dysfunctional Congress, supra note 4, at 91–96 (“Congress . . . struggles to legislate when partisan polarization rises and when the two chambers diverge in their policy views . . . .”); Jacobson, supra note 4, at 700–02 (describing divided government as “prone to conflict and stalemate”). This gridlock dynamic is further intensified if the party opposing the President lacks a large majority in Congress. 42 See Miranda Yaver, When Do Agencies Have Agency? The Limits of Compliance in the EPA 10–11 (Oct. 29, 2014) (unpublished manuscript), (on file with the Columbia Law Review) (noting “extent to which the party opposing the president controls a substantial share of seats in Congress,” in addition to polarization in Congress, affects whether agencies will face retaliation for developing new policy). Similarly, supermajority requirements are less of an obstacle to legislative enactment when legislators’ preferences are relatively closely aligned. But as polarization or the ideological distance among legislators grows, reaching a supermajority voting threshold becomes increasingly difficult. 43 Thomas E. Mann & Norman J. Ornstein, The Broken Branch: How Congress Is Failing America and How to Get it Back on Track 10 (2006). The net result is a growing range of contexts in which a majority of legislators would prefer to alter the policy status quo but lack the numbers to overcome the objections of the President or a Senate minority. The larger the distance between the point at which Congress will override a presidential veto and the point at which the Senate will invoke cloture to end a filibuster—referred to by political scientists as the gridlock interval—the lower the chances that substantive legislation will be enacted. 44 See Freeman and Spence, supra note 17, at 83–84 (concluding under a gridlock model, “[a]s members of the legislature become more ideologically polarized, the status quo policy can persist even as the median voter’s preferences stray farther and farther from that status quo”); see also Keith Krehbiel, Pivotal Politics: A Theory of U.S. Lawmaking 47–48, 238 (1998) (positing “[p]olicy change requires that the status quo must lie outside the gridlock interval, as defined by the president, filibuster, and [presidential] veto pivots”).

Congress may turn to other means to control agencies, in particular to appropriations. Recent years have witnessed a significant increase in the use of limitation riders on appropriations bills to forestall or require certain agency actions. 45 Jason A. MacDonald, Limitation Riders and Congressional Influence over Bureaucratic Policy Decisions, 104 Am. Pol. Sci. Rev. 766, 767 (2010). The great advantage of appropriations legislation from Congress’s perspective is its must-pass status, dramatically raising the stakes of a presidential veto. 46 See Richard J. Lazarus, Congressional Descent: The Demise of Deliberative Democracy in Environmental Law, 94 Geo. L.J. 619, 635–36 (2006) (noting great political pressure to pass annual appropriations legislation due to threat of government shutdown); MacDonald, supra note 45, at 767 (“[T]hat appropriations must pass implies that the ability of the president to remove provisions to which he objects through veto bargaining . . . is more limited than is the case with ‘normal’ . . . legislation.”). Under the Constitution, “[n]o Money shall be drawn from the Treasury, but in Consequence of Appropriations made by Law.” U.S. Const. art. I, § 9, cl. 7. If Congress fails to pass a measure authorizing funding, the federal government lacks power to spend funds and must shut down. MacDonald, supra note 45, at 767. Congress does appropriations on an annual basis, with the result that “[t]here is enormous political pressure to pass annual appropriations legislation.” Lazarus, supra, at 635. Yet spending measures can encounter the same enactment obstacles as substantive legislation, as demonstrated by congressional Republicans’ recent failure to use spending constraints to prohibit DHS from going forward with its immigration initiatives. 47 See Lauren French, Jake Sherman & John Bresnahan, John Boehner Ends Stalemate, Politico (Mar. 3, 2015), [] (describing Republican leadership’s decision to end effort to overturn immigration policies through appropriations limits). In addition, filibuster threats and fears of another government shutdown have stymied the ordinary appropriations process, replacing the annual budget with a series of continuing resolutions on funding. 48 Nolan McCarty, The Decline of Regular Order in Appropriations: Does It Matter? 1–12 (Feb. 3, 2015) (unpublished manuscript),
/files/research/McCarty_Appropriations.pdf [].

As a result, congressional influence through appropriations is often felt more through budgetary inaction than actual appropriations legi­slation. Sometimes this inaction is simply a reflection of congressional inability to enact legislation. But it may also be a deliberate strategy of obstruction, reflecting the reality that delay and stalling may prove easier tools for congressional opponents of agency action to wield than affirmative congressional enactments. 49 See Note, Independence, Congressional Weakness, and the Importance of Appointment: The Impact of Combining Budgetary Autonomy with Removal Protection, 125 Harv. L. Rev. 1822, 1826–27 (2012) (noting agencies may respond to signals and threats about budgets “even where congressional action appears absent”); see also Jack M. Beermann, Congressional Administration, 43 San Diego L. Rev. 61, 121–39 (2006) (discussing informal measures, including budgetary measures, members of Congress may take to affect agency decisionmaking). Similar use of delay to obstruct executive action is evident in the appointments context, with executive appointment delays growing significantly alongside intensifying polar­ization. 50 See Neal Devins & David E. Lewis, Not-So Independent Agencies: Party Polarization and the Limits of Institutional Design, 88 B.U. L. Rev. 459, 461, 485–88 (2008) (detailing Senate’s increasing reliance on strategy of inaction on nominees, or “holds,” as response to polarization); see also Anne Joseph O’Connell, Shortening Agency and Judicial Vacancies Through Filibuster Reform? An Examination of Confirmation Rates and Delays from 1981 to 2014, 64 Duke L.J. 1645, 1659–81 (2015) [hereinafter O’Connell, Shortening Vacancies] (providing evidence on increasing nomination and appointment delays from 1981 to 2014). A prominent recent example is the 166-day delay for Loretta Lynch to be confirmed as Attorney General in April 2015, the third longest delay to confirm an Attorney General in history. Jennifer Steinhauer, Senate Confirms Loretta Lynch as Attorney General After Long Delay, N.Y. Times (Apr. 23, 2015), (on file with the Columbia Law Review). Interestingly, moreover, delays in agency appointments appear to have grown longer since such appointments were exempted from the filibuster. 51 See O’Connell, Shortening Vacancies, supra note 50, at 1676–81 (presenting data on lack of improvement in agency appointment delays after filibuster reform in November 2013). Congress also resorts more to hearings and investi­gations as tools of control, with divided government and growing parti­sanship significantly increasing use of these forms of oversight. 52 See Douglas Kriner & Liam Schwartz, Divided Government and Congressional Investigations, 33 Legis. Stud. Q. 295, 297 (2008) (providing empirical evidence on increasing use of congressional investigations during divided government); David C.W. Parker & Matthew Dull, The Weaponization of Congressional Oversight: The Politics of the Watchful Eye, 1947–2010, in Politics to the Extreme: American Political Institutions in the Twenty-First Century 47, 52–54, 56–63 (Scott A. Frisch & Sean Q. Kelly eds., 2013) (describing correlation between “divided government and the willingness . . . to charge and investigate the executive branch with malfeasance”). A prime contemporary embodiment of this phenomenon comes from the House Oversight and Government Reform Committee under the recent chairmanship of Representative Darrell Issa, which was marked by strong partisan divides and numerous contentious investigations of the Obama Administration. 53 See Noah Bierman, California’s Darrell Issa Loses Power Along with House Oversight Committee Post, L.A. Times (Mar. 20, 2015), [] (discussing Representative Issa’s investigations of White House and President Obama).

The extent to which congressional inaction impedes agencies depends substantially on the underlying legislative baseline. When agencies need congressional action, either to grant new powers and money or to address emergent issues in existing legislation, congressional inaction can be a serious impediment. Thus, for example, the Supreme Court recently struck down an effort by the EPA to alter the levels at which permits are required under the Clean Air Act to better suit the realities of greenhouse gases, emphasizing that “[a]n agency has no power to ‘tailor’ legislation to bureaucratic policy goals by rewriting unambiguous statutory terms.” 54 Util. Air Regulatory Grp. v. EPA, 134 S. Ct. 2427, 2445 (2014). Even when agencies possess authority to act, moreover, budgetary and appointments delays can impede effective administrative action. Lack of a budget creates great uncertainty for agencies and forces top executive branch personnel to divert their energies to planning for possible shutdowns. 55 Joshua C. Huder & Marian Currinder, Polarization, Budget Failures, and What to Expect Beyond 2014, 43 Pub. Manager 11, 12–13 (2014); see also Office of Mgmt. & Budget, Exec. Office of the President, Impacts and Costs of the October 2013 Federal Government Shutdown 4–5, 15–21 (2013), [] (estimating impact of October 2013 shut­down on programs and services for citizens); Sophie Novack & Clara Ritger, Shutdown Threat Has Federal Agencies in Scramble Mode, Nat’l J. (Sept. 24, 2013), [] (detailing agency contingency plans in pre­paration for government shutdown). Similarly, ongoing vacan­cies in senior positions can limit the extent to which an agency pursues new policy initiatives. 56 See Anne Joseph O’Connell, Vacant Offices: Delays in Staffing Top Agency Positions, 82 S. Calif. L. Rev. 913, 938–39 (2009) [hereinafter O’Connell, Vacant Offices] (determining frequent vacancies harm agency performance and lead to agency inaction). Congressional investigatory scrutiny also can have an inhibitory effect on agencies. 57 See Douglas Kriner, Can Enhanced Oversight Repair “The Broken Branch”?, 89 B.U. L. Rev. 765, 774, 784–87 (2009) (arguing oversight can lead to public pressure on agencies and anticipation of oversight can dissuade agency action).

Yet it is also easy to exaggerate the impact of these indirect congres­sional constraints. Some agencies may be relatively insulated from these measures, either by virtue of independent funding or by having a substantial cadre of high-level career officials who can develop policy. 58 See Rachel E. Barkow, Insulating Agencies: Avoiding Capture Through Institutional Design, 89 Tex. L. Rev. 15, 42–45 (2010) (tracing connection between political pressure on agencies and agency’s funding source); Nina A. Mendelson, The Uncertain Effects of Senate Confirmation Delays in the Agencies, 64 Duke L.J. 1571, 1585–97 (2015) (arguing staffing delays potentially can have beneficial effects on agency performance); see also Moe, Assessment of Positive Theory, supra note 30, at 480–86 (expressing skepticism about impact of congressional oversight on agencies). Moreover, the inhibitory effects of high-level agency vacancies and inves­tigations can be counterbalanced by White House leadership on policy initiatives. 59 See Matthew J. Dickinson, The Executive Office of the President: The Paradox of Politicization, in The Executive Branch 135, 152–54 (Joel D. Aberbach & Mark A. Peterson eds., 2005) (tracing increasing role of White House aides in policy development and suggesting an aide under President George W. Bush was given “preeminent role in the formulation of homeland security policy”); O’Connell, Vacant Offices, supra note 56, at 940 (“To some extent, this inaction effect of agency vacancies can be overcome by other factors. Strong White House involvement in a specific policy area may produce agency action.”). In particular, White House “czars” have emerged as a means by which the White House can coordinate and control policy in areas of key concern to the President. 60 Kathryn A. Watts, Controlling Presidential Control, 114 Mich. L. Rev. (forthcoming 2016) (manuscript at 28–29) (on file with the Columbia Law Review); see also Aaron J. Saiger, Obama’s “Czars” for Domestic Policy and the Law of the White House Staff, 79 Fordham L. Rev. 2577, 2582–83 (2011) (describing czars as White House officials “tasked by the President to oversee policy in a particular substantive area”). As a result, the impact of congres­sional polarization and divided government on agencies will depend on the extent to which an agency and the President share the same policy agenda. When the President and an agency are in sync, indirect congres­sional constraints may be less effective in derailing administrative action.

2. The Further Rise of Presidential Administration. — The importance of the President’s stance highlights a second major consequence of polar­ization: an increase in presidential assertions of policymaking authority and control over agencies. 61 See Neal Devins, Presidential Unilateralism and Political Polarization: Why Today’s Congress Lacks the Will and the Way to Stop Presidential Initiatives, 45 Willamette L. Rev. 395, 411 (2009) (stating political polarization encourages Presidents to take control of administration via budget decisions, signing statements, and appointments); Lowande & Milkis, supra note 8, at 21–24 (“In absence of a politically unified Congress, the administrative action strategy employed by the Obama White House has enabled the president to move closer toward programmatic goals involving climate change and energy efficiency—policy that saw no significant development in Congress.”). This increase in presidential administration is part of a broader trend. Presidentially directed administration expanded significantly starting with the Reagan Administration, reflecting increasing popular focus on the President and the corresponding pressure Presidents face to deliver on policy promises. 62 See Kagan, supra note 16, at 2275–319 (tracing development of presidential administration from Nixon through Clinton); Terry M. Moe, The Politicized Presidency, in The New Direction in American Politics 235, 244–45 (John E. Chubb & Paul E. Peterson eds., 1985) [hereinafter Moe, Politicized Presidency] (explaining feedback cycle between expectations on White House and its institutional competence). But the difficulty of advancing policy through legislation given polarized politics makes presidential resort to administrative measures all the more likely. 63 See Moe & Howell, supra note 30, at 136–38 (noting trend in legacy-conscious Presidents exploiting constitutional ambiguity of agencies to accomplish agendas); see also Charles M. Cameron, Studying the Polarized Presidency, 32 Presidential Stud. Q. 647, 647–48 (2002) (arguing polarization has “pervasive” impact on presidency). Perhaps the clearest acknowl­edgement of this dynamic is the mantra of “We Can’t Wait” that President Obama invoked to justify many of his administration’s initia­tives. 64 See Lowande & Milkis, supra note 8, at 3–4 (describing “We Can’t Wait” campaign); We Can’t Wait, White House, [] (last visited Sept. 14, 2015) (“We Can’t Wait: President Obama is not letting congressional gridlock slow our economic growth.”). Moreover, Presidents assert such policy control in the knowledge that Congress is unlikely to succeed in legislating limits in response. 65 Devins, supra note 61, at 414; see also Lowande & Milkis, supra note 8, at 22–23 (explaining, regardless of disagreement with Obama’s executive actions, Congress was too politically divided to overturn them); Moe & Howell, supra note 30, at 133–38 (arguing presidential power expands because members of Congress focus on individual constit­uencies at expense of defending legislative power). Hence, polarization contributes to the rise of presidential unilateralism as a central governance phenomenon. 66 See Lowande & Milkis, supra note 8, at 5 (emphasizing growth of presidential exploitation of administrative control for partisan gain); see also Abbe R. Gluck, Anne Joseph O’Connell & Rosa Po, Unorthodox Lawmaking, Unorthodox Rulemaking 115 Colum. L. Rev. 1789, 1845 (2015) (“It is the White House, often through OIRA, that now may direct the rulemaking process, instead of the agency . . . .”). But see Dino P. Christenson & Douglas L. Kriner, Political Constraints on Unilateral Executive Action, 65 Case W. Res. L. Rev. 897, 908–12 (2015) (arguing informal political constraints on presidential unilateralism more robust than generally acknowledged).

Such greater presidential instigation and control of agency decision­making manifests in closer White House oversight. In addition to White House czars, a main mechanism for such oversight is centralized regu­latory review through the Office of Information and Regulatory Affairs (OIRA). 67 See Exec. Order No. 12,866, 3 C.F.R. 638, 644–48 (1994), reprinted as amended in 5 U.S.C. § 601 app. at 802, 804–06 (2012) (outlining OIRA oversight of agency regulatory decisions); see also Exec. Order No. 13,563, 3 C.F.R. 215, 215 (2012), reprinted as amended in 5 U.S.C. § 601 app. at 816, 816 (retaining Exec. Order 12,866 regulatory review framework). Scholars have documented an increase in presidential use of the OIRA process to control agency policymaking over the last admin­istrations, with contacts between OIRA and agencies occurring earlier and often not being publicly disclosed. 68 See Lisa Heinzerling, Inside EPA: A Former Insider’s Reflections on the Relationship Between the Obama EPA and the Obama White House, 31 Pace Envtl. L. Rev. 325, 334–41 (2014) (detailing OIRA’s control under Bush II and Obama Administrations); Watts, supra note 60 (manuscript at 15–19, 21–23) (chronicling strong presidential control of agencies during Bush II and Obama Administrations). For an argument by a former head of OIRA that OIRA review under the Obama Administration has focused on coordination rather than preventing rules from going forward, see Cass R. Sunstein, The Office of Information and Regulatory Affairs: Myths and Realities, 126 Harv. L. Rev. 1838, 1840 (2013). Yet in other ways the Obama Administration, following an approach akin to that used by President Clinton, has exercised its oversight role quite publicly, directing agency policy initiatives and using public media to claim agency actions as the President’s own. 69 See Watts, supra note 60 (manuscript at 23–28) (describing President Obama’s use of online videos, social media, and written memoranda posted online and in Federal Register); see also Kagan, supra note 16, at 2281–319 (describing  President Clinton’s use of directives); Press Release, The White House, Year of Action: A Final Progress Report on the Obama Administration’s Actions to Help Create Opportunity for All Americans (Dec. 22, 2014), [] (detailing ex­ecutive actions in 2014 with emphasis on President Obama’s role and involvement). Examples include President Obama’s greenhouse gas and fuel efficiency regulatory directives, instructions to the Secretary of Education to change regulations respecting the repay­ment of student debt, and immediate endorsement of the Clean Water Rule promulgated by EPA and the Army Corps of Engineers. 70 See Watts, supra note 60 (manuscript at 23–28) (listing examples); Jenny Hopkinson, Barack Obama’s Water War, Politico (May 27, 2015, 10:41 am), [] (noting President Obama’s quick endorsement of proposed rule on regu­lating national waterways).

Greater presidential control is not limited to centralized oversight but also seeps into the agencies themselves. Agency politicization is another key mechanism by which Presidents control the executive branch. 71 Moe, Politicized Presidency, supra note 62, at 244–45. Like centralization, politicization is also on the rise, with a par­ticularly marked expansion in the number of agency policy appointees. 72 See David J. Barron, Foreword: From Takeover to Merger: Reforming Administrative Law in an Age of Agency Politicization, 76 Geo. Wash. L. Rev. 1095, 1122–28 (2008) (detailing increased agency politicization); see also David E. Lewis, The Politics of Presidential Appointments: Political Control and Bureaucratic Performance 137, 202–04 (2008) (describing methods of politicization and arguing politicization follows certain predictable patterns, varying across agencies and over time, but also noting increase in politicization under both Bush Administrations compared to Clinton Administration). Polarization plays indirect as well as direct roles in this politicization process; not only does the President face more pressure to advance partisan policy goals, but polarization has increased the President’s ability to identify “a cadre of loyal and competent personnel” who share similar policy views. 73 Michael Livermore, Political Parties and Presidential Oversight, 67 Ala. L. Rev. (forthcoming 2015) (manuscript at 5, 22–23) (on file with the Columbia Law Review).

Again, however, centralization and politicization are imperfect mechanisms of control. Excessive politicization can undermine agency performance and create internal divides between career and political staff. 74 See, e.g., Lewis, supra note 72, at 149–53 (explaining politicization and tension between appointees and careerists caused performance decline in Federal Emergency Management Agency); Barron, supra note 72, at 1124–30 (discussing career–political clashes); Joshua B. Kennedy, “‘Do This! Do That!’ And Nothing Will Happen”: Executive Orders and Bureaucratic Responsiveness, 43 Am. Pol. Res. 59, 67–76 (2015) (noting impact of politicization on performance and providing data on ability of presidential direction to produce agency response). In turn, centralization often operates asymmetrically, with central­ized regulatory review through OIRA working more effectively to delay or prevent agency action than to spur new administrative initiatives. 75 See Nicholas Bagley & Richard L. Revesz, Centralized Oversight of the Administrative State, 106 Colum. L. Rev. 1260, 1262–82, 1312–14 (2006) (describing OIRA’s “antiregulatory bent”); see also Watts, supra note 60 (manuscript at 22–23) (documenting extensive delays from OIRA review during Obama Administration). More­over, OIRA review does not extend to a variety of executive branch actions, including those taken by independent agencies. 76 See Exec. Order No. 12,866, 3 C.F.R. 638, 644–45 (1994), reprinted as amended in 5 U.S.C. § 601 app. at 802, 803 (2012) (excluding independent agencies and limiting OIRA review to significant rulemaking actions). Thus, for example, the Federal Communication Commission (FCC)’s recent and contentious rulemaking on net neutrality was not subject to OIRA review. As a result, although President Obama weighed in through a public statement and video supporting neutrality—and allegedly through behind-the-scenes pressure and influence—the White House could not directly prevent the FCC from issuing a rule with which the President disagreed. 77 See Watts, supra note 60 (manuscript at 43–47) (describing President Obama’s efforts to support net neutrality and acknowledgment that “[t]he FCC is an independent agency, and ultimately [the] decision is theirs alone” (internal quotation marks omitted)); Gautham Nagesh & Brody Mullins, Net Neutrality: How White House Thwarted FCC Chief, Wall St. J. (Feb. 4, 2015), (on file with the Columbia Law Review) (describing internal White House meetings and effort to push net neutrality). As important, the White House lacks the capacity to oversee and control the vast array of agency activities. 78 See Lisa Schultz Bressman & Michael P. Vandenbergh, Inside the Administrative State: A Critical Look at the Practice of Presidential Control, 105 Mich. L. Rev. 47, 70 (2006) (“Nor do advocates of presidential control claim that it reaches every agency action or even every important agency action.”); see also Cynthia R. Farina, False Comfort and Impossible Promises: Uncertainty, Information Overload, and the Unitary Executive, 12 U. Pa. J. Const. L. 357, 399–402 (2010) (asserting full load of agency activity too great for presidential oversight). In Cynthia Farina’s words, “White House control is in fact sporadic, at times cacophonous, and often imperfectly realized” because “[i]t is costly for Presidents and their advisers to monitor the mass of agencies’ policymaking activities, to develop positions on the often complex underlying substantive issues, to communicate those positions to the people formally empowered to decide, and to actually get a decision implementing the President’s policy.” Id. at 412.

Hence, even if presidential administration is on the rise, polarization does not mean that agencies are simply doing the President’s bidding and not also initiating policy. White House direction is particularly likely with respect to politically salient issues. 79 See Jennifer Nou, Agency Self-Insulation Under Presidential Review, 126 Harv. L. Rev. 1755, 1790 (2013) (“[H]igher-cost or -benefit rules are more likely to be politically salient . . . [and] are thus among those rules most likely to gain the President’s attention.”); Frank J. Thompson & Michael K. Gusmano, The Administrative Presidency and Fractious Federalism: The Case of Obamacare, 44 Publius 426, 429 (2014) (describing White House involvement in ACA implementation as reflecting ACA’s high political salience). Yet here, too, agencies still may control significant aspects of policy implementation and development. A recent example is the rollout of the federal health exchange website: White House officials led the way, but day-to-day implementation fell to the Department of Health and Human Services (HHS) and several of its component departments. 80 See Lowande & Milkis, supra note 8, at 18–19 (“The White House was in charge, but on-the-ground work fell largely to the Department of Health and Human Services.”); Sheryl Gay Stolberg, Health Site Puts Agency and Leader in Hot Seat, N.Y. Times (Oct. 28, 2013), (on file with the Columbia Law Review) (noting HHS’s “major responsibility for carrying out the president’s health care overhaul”); see also Ricardo Alonso-Zaldivar, Probe Finds Reasons for Obamacare Website Failure, Inc. (July 31, 2014), (on file with the Columbia Law Review) (stating White House reportedly blindsided by website’s major flaws when it was made available to public). Similarly, officials within the nation’s law enforcement and national security agencies have successfully resisted several presidential initiatives in the War on Terror. 81 See Jack Goldsmith, Power and Constraint: The Accountable Presidency After 9/11, at 36–39 (2012) (describing agency pushback against “aggressive executive unilateralism” post 9/11); see also David Johnson & Scott Shane, Notes Detail Pressure on Ashcroft over Spying, N.Y. Times (Aug. 17, 2007), (on file with the Columbia Law Review) (describing role of top Department of Justice officials in resisting White House effort to obtain authorization for warrantless surveillance). Agencies may also act strategically, providing the impetus behind a policy initiative but shielding their involvement from public view so that it appears to be more of a unilateral presidential action. 82 See Andrew Rudalevige, The Contemporary Presidency: Executive Orders and Presidential Unilateralism, 42 Presidential Stud. Q. 138, 142–44, 153–54 (2012) (reporting departments are major source of executive orders and describing role of EPA in form­ulation of Clinton executive order on children’s environmental health); see also Daniels, supra note 16, at 384–411 (providing historical examples of strategic action by agencies to push policy with President and Congress). When presidential and agency policy goals coincide, independent agency policy development will be harder to identify, but the agency role rises to the fore when a President disagrees with an agency’s mission and priorities. 83 See Freeman & Spence, supra note 17, at 66–67 (“Where the President’s objectives and the agency’s mission are in conflict . . . the result can be turmoil and struggle.”).

More importantly, Presidents lack the ability to undertake policy initiatives successfully without involving agencies. This is in part a result of the White House’s limited institutional capacity and practical dependence on agencies for policy promulgation, implementation, and enforcement. 84 See Bressman & Vandenbergh, supra note 78, at 70 (finding “presidential control, as structured, is selective in its focus” and “[t]he president simply has too many responsibilities, and OIRA and the other White House offices have too few resources, to reach even every major agency decision”); see also Lewis, supra note 72, at 57–61 (“In other cases, presidents have no foothold or means of easy entrée and, as a consequence, agencies have very little interaction with or direction from the White House.”). Equally, it reflects the legal reality that most presidential initiatives rest on preexisting statutory grants of regulatory authority to agencies. 85 On the question of whether such statutory grants are best read to preclude presidential oversight and direction, compare Kagan, supra note 16, at 2326–31 (“An interpretive principle presuming an undifferentiated presidential control of executive agency officials thus may reflect . . . the general intent and understanding of Congress.”), with Kevin M. Stack, The President’s Statutory Powers to Administer the Laws, 106 Colum. L. Rev. 263, 268 (2006) (“[S]tatutory grants of authority to an official (alone) should be read as vesting the official with an independent duty and discretion, not a legal duty to the President.”). Presidents and agencies work in tandem, and growing presidential unilateralism translates into an expansion in agency action at the same time as increased presidential involvement may limit agencies’ indepen­dent policymaking role.

B. Agencies as Actors: Policy Development and Partisan Realignment

Agencies’ broad and ongoing regulatory powers are critical in assessing the impact of polarization. Again, it is the presence of these powers and the vast modern administrative state that distinguishes our current period from all prior instances in which the United States experienced equivalently high levels of polarization. These regulatory powers enable agencies—as a result of presidential instigation or on their own initiative—to push policy in new directions with limited fear of congressional reversal. 86 See Whittington & Carpenter, supra note 30, at 501 (“[D]elegation of administrative discretion within the context of a nonunitary principal creates unavoidable opportunities for the executive to exploit that discretion to alter outcomes and restructure legislative preferences.”); Yaver, supra note 42, at 4, 9–11 (“With higher levels of legislative conflict . . . the likelihood of punishment declines given the difficulty of passing legislation that would curb agency behavior . . . .”). Whether agencies undertake such efforts, and how far they seek to push policy, turns on a variety of factors. These include not just congressional oversight and presidential preferences, as described above, but also judicial controls and internal executive branch constraints. Yet if agencies’ policy initiatives are successful, they have the potential to reshape the political status quo and partisan baselines.

1. Polarization and Agency Policy Development. — Although congressional gridlock and presidential unilateralism are commonly noted effects of polarization, the increased room for agencies to develop policy is less frequently acknowledged. A recent rare exception is Jody Freeman and David Spence’s analysis of how congressional gridlock may prompt agencies to use their authority under preexisting statutes to address newly emerging regulatory challenges. 87 Freeman & Spence, supra note 17, at 17–63. Professors Freeman and Spence focus on environmental and energy regulation, but similar agency policy development is evident in a number of contexts, including education, immigration, national security, and healthcare. 88 See David J. Barron & Todd D. Rakoff, In Defense of Big Waiver, 113 Colum. L. Rev. 265, 277–90 (2013) (providing examples of agency use of waivers to update legislative frameworks in several policy areas, including national security context); Zachary S. Price, The Politics of Nonenforcement, 65 Case W. Res. L. Rev. 1119, 1134–36 (2015) (describing marijuana and immigration nonenforcement initiatives); infra Part II (noting agency efforts to implement ACA); infra text accompanying notes 249–252 (discussing presidential and agency action amid congressional gridlock over No Child Left Behind Act (NCLB)). In a similar vein, Miranda Yaver concludes that partisan conflict in Congress allows agencies to set policies that deviate from authorizing statutes, based on an empirical study of a large scale dataset of cases and bill introductions involving the EPA. 89 See Yaver, supra note 42, at 4–5 (“[T]he multiplicity of oversight actors can expand opportunities for agencies to exploit coordination problems and set policies closer to their preferences.”); see also Michael S. Greve & Ashley C. Parrish, Administrative Law Without Congress, 22 Geo. Mason L. Rev. 501, 511–39 (2015) (analyzing effect of congressional inaction on agencies).

To some extent, this greater room for agency action under political polarization could be viewed as simply an extreme manifestation of the potential for bureaucratic drift, a phenomenon long studied by political scientists. 90 See, e.g., Daphna Renan, Pooling Powers, 115 Colum. L. Rev. 211, 257–58 (2015) (describing bureaucratic drift). But viewing agencies’ policysetting power in polarization contexts in this fashion downplays the extent to which legislative gridlock may create instances in which agencies feel compelled to act on their own initiative, despite recognizing that the regulatory challenges at hand would be better addressed through legislation. 91 Two recent examples are Education Secretary Arne Duncan’s statement that his department was waiving requirements of the NCLB because Congress had not enacted new legislation to address problems in the statute, see Sam Dillon, Overriding a Key Education Law, N.Y. Times (Aug. 8, 2011), (on file with the Columbia Law Review) (discussing NCLB waivers), and Homeland Security Secretary Johnson’s acknowl­edgement of the need for compre­hensive immigration reform through legislation while defending the legality of his department’s grants of deferred status to shield some undocumented immigrants from deportation. See Jerry Markon, Jeh Johnson Defends Obama Immigration Action to House Republicans, Wash. Post (Dec. 2, 2014), [] (describing Homeland Security Secretary’s defense of executive’s immigration policy). Put differently, agency policy development here is not easily classified as surreptitious efforts at policy deviation or as stemming from policy disagreement with governing statutes. Instead, agencies will often publicly call for new legislation and at times even describe their regulatory actions as made necessary by congressional failure to act. 92 See Freeman & Spence, supra note 17, at 67 (describing phenomenon in environmental and energy contexts).

2. Judicial and Internal Executive Branch Controls. — A particularly striking feature of Professors Freeman and Spence’s account is their conclusion that agencies “do not simply ‘go for broke’ . . . . Instead, [agencies] proceed strategically, cognizant of the preferences of their political overseers and the risk of being overturned in the courts.” 93 Id. at 3. This no doubt reflects agencies’ awareness of the ways Congress can still retaliate against them, even absent the practical ability to enact legislation, as well as the reality of greater presidential involvement. 94 See supra notes 49–57, 61–73 and accompanying text (describing congressional and presidential controls). But it also signals how polarization can increase the salience of nonpolitical controls on agencies. Two such controls are judicial review and internal executive branch constraints separate from White House oversight.

Judicial review is a constant factor of administrative life in the United States, with decades of administrative law scholarship debating its impact on agency functioning. 95 See Peter L. Strauss et al., Gellhorn and Byse’s Administrative Law: Cases and Comments 1106–12 (11th ed. 2011) (excerpting scholarship on judicial review of agency reasoning). Its importance is only likely to grow as congressional controls slacken and courts become the more vibrant external constraint on agencies. 96 McCarty, Policy Effects, supra note 4, at 237–40, 246; see also Freeman & Spence, supra note 17, at 68 (“[A]gencies also appear to have been meaningfully constrained by their . . . anticipation of judicial review.”). Some scholars have argued for greater judicial deference to agency action in times of polarization and divided government, given the need for agencies to meet the regulatory challenges that the political branches are incapable of addressing. 97 See, e.g., Freeman & Spence, supra note 17, at 75–76 (urging judicial deference to agency efforts to fit existing statutes to new problems when Congress is unable to act); Cass R. Sunstein, Partyism, 2015 U. Chi. Legal Forum (forthcoming 2015) (manuscript at 15–17) (on file with the Columbia Law Review) [hereinafter Sunstein, Partyism] (arguing for agency power to interpret and adapt statutory terms “as they see fit, so long as their interpretations are reasonable,” to address impact of polarization). But courts may see congressional dysfunction as instead increasing the need for a judicial check to prevent executive branch unilateralism and aggrandizement. 98 See, e.g., Texas v. United States, No. B-14-254, 2015 WL 648579, at *22 (S.D. Tex. Feb. 16, 2015), stay denied, 787 F.3d 733 (5th Cir. 2015) (condemning Obama Administration for complete failure to enforce immigration laws); see also Lisa Heinzerling, The FDA’s Plan B Fiasco: Lessons for Administrative Law, 102 Geo. L.J. 927, 952–58 (2014) [hereinafter Heinzerling, Plan B Fiasco] (describing courts’ willingness to deviate from ordinary administrative law review in response to evidence of political involvement in Food and Drug Administration’s (FDA) decision on whether to allow Plan B to be available over the counter (OTC)). Indeed, the Supreme Court’s recent decisions stand out for their reluctance to grant agencies deference, despite agreeing with the agency on the merits. 99 See, e.g., Tex. Dep’t of Hous. & Cmty. Affairs v. Inclusive Project, Inc., 135 S. Ct. 2507, 2525 (2015) (holding disparate impact claims cognizable under Fair Housing Act without deferring to agency view to that effect); King v. Burwell, 135 S. Ct. 2480, 2488–89 (2015) (refusing to grant deference to IRS interpretation of ACA); see also Michigan v. EPA, 135 S. Ct. 2699, 2712 (2015) (rejecting as unreasonable agency’s reading of statute to preclude consideration of costs in initial decision whether to regulate). Regardless, congressional failure to enact or update legislation means that agencies will need to stretch their existing authority to fit new regulatory challenges. 100 See Freeman & Spence, supra note 17, at 79–81 (“[W]hile Congress absents itself from policymaking, the need to make policy choices continues.”). Even courts that find such agency actions sound as a policy matter may conclude they are simply incompatible with underlying statutory text. 101 See, e.g., Util. Air Regulatory Grp. v. EPA, 134 S. Ct. 2427, 2445–47 (2014) (“The power of executing the laws . . . does not include a power to revise clear statutory terms that turn out not to work in practice.”).

Changes in the ideological composition of the federal courts are also likely to affect the judicial response to polarization. Partisan battles over judicial appointments were a major factor behind the Senate Democrats’ adoption of filibuster reform in 2013. 102 See Jeffrey Toobin, The Obama Brief, New Yorker, Oct. 27, 2014, at 24, 27–28 (describing judicial appointment battles leading to filibuster reform and effect on D.C. Circuit). This in turn allowed appoint­ment of three judges to the D.C. Circuit and gave that court a majority of Democratic appointees, a change particularly relevant for agencies given the high concentration of administrative challenges that the D.C. Circuit hears. 103 Id. Moreover, evidence suggests Democratic and Republican judicial appointees vary in their willingness to uphold agency action based on its liberal or conservative character. 104 See Thomas J. Miles & Cass R. Sunstein, Do Judges Make Regulatory Policy? An Empirical Investigation of Chevron, 73 U. Chi L. Rev. 823, 825–26 (2006) (reporting ideological leaning significantly affects likelihood judge will vote to affirm or overturn agency’s action); Thomas J. Miles & Cass R. Sunstein, The Real World of Arbitrariness Review, 75 U. Chi. L. Rev. 761, 767–69 (2008) (reporting Democratic appointees validate liberal agency actions seventy-two percent of the time compared to fifty-five percent for Republican appointees, with validation rates flipping when agency action is conservative). Nor is the political nature of many administrative challenges hard to spot. Congress itself provides strong clues, bringing high-profile lawsuits on party-line votes and increasingly filing only partisan briefs. 105 See, e.g., United States v. Windsor, 133 S. Ct. 2675, 2684, 2687–88 (2013) (discussing role of House Bipartisan Legal Advisory Group in litigation over consti­tutionality of Defense of Marriage Act); Complaint at ¶¶ 42–50, U.S. House of Representatives v. Burwell, No. 14-cv-01967 (D.D.C. Nov. 21, 2014), 2014 WL 6492097 (challenging delay of ACA employer mandate); see also Neal Devins, Measuring Party Polarization in Congress: Lessons from Congressional Participation as Amicus Curiae, 65 Case W. Res. L. Rev. 933, 933–34 (2015) (documenting increase in partisan briefs and decline in bipartisan filings correlated to growing polarization).

Agencies also face internal checks, both within the agency and in the executive branch as a whole, that continue to have vitality notwith­standing the polarized political climate that dominates at the congres­sional–presidential level. Professional norms, fears of judicial reversal and harm to the agency’s credibility and resources, as well as a longer-term careerist perspective, may lead agency counsel and other executive branch attorneys to take a cautious approach. 106 See Freeman & Spence, supra note 17, at 68 (“It is not hard to imagine then that agencies sometimes pare back or abandon initial proposals deemed too risky.”); see also Rebecca Ingber, Interpretation Catalysts and Executive Branch Legal Decisionmaking, 38 Yale J. Int’l L. 359, 413–14 (2013) (arguing executive constrained by “nature of the events themselves that drive that decisionmaking, and the interplay between those events and the organic reality of internal executive process”). One example of the influence of reputational concerns comes from the efforts to limit the morning-after pill’s availability OTC. The refusal of the FDA to grant the OTC application under the Bush II Administration prompted resignations and complaints from senior scientists connected to the agency. When the Obama Administration then chose to continue OTC limits, FDA Commissioner Margaret Hamburg issued a statement making clear that she and the FDA scientists determined Plan B should be freely available but had been overruled by HHS Secretary Kathleen Sebelius. See Heinzerling, Plan B Fiasco, supra note 98, at 942–46 (recounting agency controversy surrounding Plan B’s OTC availability). These executive branch and judicial controls interact, as fears of judicial reversal lead executive branch officials to greater caution and restraint. 107 See Freeman & Spence, supra note 17, at 68 (describing how “agencies take pains to develop their legal strategies to ensure they are . . . likely to withstand attack”). Additional constraints may be created by requirements of interagency coordination and consultation, given the different institutional structures, concerns, cultures, and personnel that can dominate even agencies performing similar responsibilities. 108 See Eric Biber, Too Many Things to Do: How to Deal with the Dysfunctions of Multiple-Goal Agencies, 33 Harv. Envtl. L. Rev. 1, 41–59 (2009) (analyzing different types of interagency relationships and their impact on agency’s actions); Anne Joseph O’Connell, The Architecture of Smart Intelligence: Structuring and Overseeing Agencies in the Post-9/11 World, 94 Calif. L. Rev. 1655, 1673–84 (2006) (describing how redundancy can check agency excesses). Whether involvement of multiple agencies in fact serves a checking function or instead operates to augment executive branch power is unclear. 109 See Renan, supra note 90, at 212–17 (describing ways in which agency coordination and pooling of powers enhances executive branch authority). Compare Freeman & Rossi, supra note 34, at 1201 (“By seizing control of the interagency process, the President and his staff can play the role of negotiator in chief, helping to broker outcomes that more closely align with his preferences than would the results of an unmediated process.”), and Marisam, supra note 34, at 231–36 (“[D]uplicative delegations alter the balance of powers by affording the Executive significantly more discretion than it usually has to determine which agency performs a task.”), with Neal Kumar Katyal, Internal Separation of Powers: Checking Today’s Most Dangerous Branch from Within, 115 Yale L.J. 2314, 2324–27 (2006) (arguing bureaucratic overlap can serve as important internal check on President). Interestingly, however, evidence suggests that the dispersion of admin­istrative responsibilities is increasing and expands with divided government, implying that Congress may use such arrangements to inhibit independent action by an opposite-party President. 110 See Daniel A. Farber & Anne J. O’Connell, The Lost World of Administrative Law, 92 Tex. L. Rev. 1137, 1155–57 (2014) (describing increase in multi-agency delegations); Sean Farhang & Miranda Yaver, Divided Government and the Fragmentation of American Law, Am. J. Pol. Sci. (forthcoming 2015) (on file with the Columbia Law Review) (providing evidence of increased congressional dispersion of administrative authority during divided government); see also David Epstein & Sharyn O’Halloran, Delegating Powers: A Transaction Cost Politics Approach to Policy Making Under Separate Powers 232–39 (1999) (documenting Congress’s use of more limited delegations under divided government).

Finally, polarization also transforms internal agency dynamics in ways that affect agencies’ ability and willingness to develop policy. This is particularly true today at independent agencies. Often led by multi­member commissions composed of members from both parties, indepen­dent agencies have partisanship baked into their organizational structure. 111 See Ronald J. Krotoszynski, Jr. et al., Partisan Balance Requirements in the Age of New Formalism, 90 Notre Dame L. Rev. 941, 962–82 (2015) (tracing history of statutory partisan balance requirements at independent agencies). Recent years have witnessed an increase in partisan divisions at independent agencies, with commissioners split along party lines and facing political pressure to refuse to compromise. 112 See Devins & Lewis, supra note 50, at 461, 491—93 (detailing increase in partisanship at independent agencies); Andrew Ackerman & Aruna Viswanatha, SEC Bickering Stalls Mary Jo White’s Agenda, Wall St. J. (June 3, 2015), (on file with the Columbia Law Review) (chronicling partisanship dividing SEC); Eric Lichtblau, FEC Can’t Curb 2016 Election Abuse, Commission Chief Says, N.Y. Times (May 2, 2015), http://www.nytimes.
com/2015/05/03/us/politics/fec-cant-curb-2016-election-abuse-commission-chief-says.html (on file with the Columbia Law Review) (describing partisan gridlock at FEC); Floyd Norris, Independent Agencies, Sometimes in Name Only, N.Y. Times (Aug. 8, 2013), http:// (on file with the Columbia Law Review) (tracking increasing influence of partisanship in Congress in selecting SEC commissioners).
Indeed, the situation has become so extreme at the Federal Election Commission (FEC) that the Commission’s Chair and another commissioner filed a petition seeking to force their own agency to engage in rulemaking. 113 See, e.g., Ann M. Ravel & Ellen L. Weintraub, Petition for Rulemaking 1 (June 8, 2015), [] (requesting FEC issue new rules and amend old rules to comply with Citizens United ruling). Internal divisions are also visible at executive agencies, with career officials and civil service personnel some­times publicly resisting the administration’s policy initiatives. 114 See, e.g., Crane v. Johnson, 783 F.3d 244, 253–55 (5th Cir. 2015) (dismissing lawsuit brought by Immigration and Customs Enforcement agents challenging new immigration initiatives). But a more evident effect on executive agencies is the growth in the number of agency political appointees, with Presidents increasingly determined to stock agencies with personnel committed to their party’s policy agenda and able to draw on a more ideologically unified cohort of party sympathizers to do so. 115 See Livermore, supra note 73 (manuscript at 29–33); see also Barron, supra note 72, at 1128–33 (“[L]ayering of political appointments . . . does help to check the resis­tance that might come from within the bureaucracy.”).

3. Agencies’ Impact on Polarization. — The significance of agencies’ enhanced policysetting role goes beyond their ability to fill the regu­latory gaps that polarization can create. By so acting, agencies hold the potential for disrupting the status quo and forcing issues onto the political agenda in ways that may break through, or at least reformulate, partisan divides.

The executive branch’s ability to alter the policy status quo in a lasting fashion is well recognized. As noted above, Congress has difficulty overturning agency action through legislation at the best of times, and polarization serves to significantly increase the window of agency activity that escapes congressional response. 116 See supra notes 39–44 and accompanying text. Terry Moe and William Howell emphasize that this ability to change policy baselines unilaterally carries with it an “agenda power”: When Presidents “take unilateral action to alter the status quo[,] . . . they present Congress with a fait accompli—a new, presidentially made law—and Congress is then in the position of having to respond or acquiesce.” 117 Moe & Howell, supra note 30, at 145; see also Whittington & Carpenter, supra note 30, at 501 (“[T]he executive . . . can play an important role in structuring legislative preferences and their articulation.”). Moe and Howell make this point with respect to presidential initiatives, but the same agenda-setting logic applies to all unilateral executive action. 118 Although in theory Presidents might be more willing to accept and not veto legislation seeking to overturn agency action that was not presidentially initiated, in practice presidential vetoes seem a likely result. Not only do most significant agency regulatory actions emerge with presidential sanction given centralized review, but in a world of polarized politics and divided government, a Congress’s success in overturning agency action taken during the current administration will likely be viewed as a presi­dential political loss. See Moe, Politicized Presidency, supra note 62, at 236–46 (describing popular assignment of responsibility to Presidents for administrative action).

Of course, congressional inaction in the face of unilateral executive branch action need not signal acquiescence and may be accompanied by ongoing congressional threats to overturn a contentious agency action in the future. 119 See, e.g., Nick Timiraos, House Republicans Vow to Continue Battle Against Immigration Measures, Wall St. J. (Mar. 1, 2015, 2:23 pm), (on file with the Columbia Law Review) (noting continued House Republican opposition to Obama Administration’s executive actions on immigration despite legislative failure to bar actions through funding constraints). Moreover, the staying power of executive branch actions turns on control of the White House; a new President from the opposite party has the ability to undo much of the prior administration’s actions—and may run on a promise to do so. 120 See, e.g., Brett Logiurato, Mitt Romney: ‘I Will Repeal Obamacare’, Bus. Insider (June 28, 2012, 12:01 pm), [] (discussing Romney’s campaign promise to repeal ACA). But the passage of time can work towards entrenchment and ultimate acquiescence. 121 See William N. Eskridge Jr. & John Ferejohn, A Republic of Statutes: The New American Constitution 26–27 (2010) (discussing agencies’ role in policy entrenchment).

Agencies can undertake their policy changes in ways that are harder to undo, for example by notice-and-comment rulemaking that can only be altered by a similar rulemaking subject to judicial review. 122 See Perez v. Mortgage Bankers Ass’n, 135 S. Ct. 1199, 1206 (2015) (“[Section] 1 of the APA [mandates] that agencies use the same procedures when they amend . . . a rule as they used to issue the rule in the first instance.”); see also Nina A. Mendelson, Agency Burrowing: Entrenching Policies and Personnel Before a New President Arrives, 78 N.Y.U. L. Rev. 557, 560–65 (2003) (discussing ways agencies entrench policies in anticipation of post-election transitions). In addition, regulatory actions may allow new information and expertise to emerge that diffuses opposition to the agency’s policies. Cass Sunstein maintains that “many disagreements are not really about values or partisan commitments, but about facts, and when facts are sufficiently engaged, disagreements across party lines will often melt away.” 123 Sunstein, Partyism, supra note 97 (manuscript at 16). Obviously, the effectiveness of this mechanism for overcoming partisanship turns on the debatable premise that partisan disagreements center on facts rather than ideology. But even so, the knowledge-generating aspect of agency action represents another means by which agency policy development alters the status quo.

Equally, if not more important, executive branch action can change the political landscape. 124 See Daniels, supra note 16, at 370–77 (describing how agency action can create new political pressures on Congress). Agency actions may create political pressure for Congress to accept the new policy realities. As Keith Whittington and Daniel Carpenter have noted, “[l]egislative proposals emerging from the executive branch, and in particular from the White House, often come attached to a larger political effort . . . which can mobilize public support . . . and crowd other issues off the legislative agenda.” 125 Whittington & Carpenter, supra note 30, at 501. Regulatory actions can create substantial interests in continuity in both regulated parties and beneficiaries, making legislative repeal politically difficult. Utilities that have invested large amounts of capital in building new plants that meet greenhouse gas emis­sions requirements, for instance, will likely oppose efforts to repeal those requirements. 126 Steven Mufson & Tom Hamburger, A Battle is Looming over Renewable Energy, and Fossil Fuel Interests Are Losing, Wash. Post (Apr. 25, 2014), [] (explaining renewable energy companies that invested capital to meet new EPA standards are fighting against repeal); see also About the Alliance, Auto Alliance, [] (last visited Aug. 14, 2015) (arguing stability is best for automobile industry and “ever-changing standards from multiple government bodies” should be avoided). A similar dynamic exists with respect to healthcare reform. See Paul Waldman, Sorry, Republicans: Obamacare Is Getting More and More Entrenched, Wash. Post (May 21, 2015), [] (arguing Republicans will face political pressure and backlash if they take away many popular features and benefits of ACA). But see Eric Patashnik & Julian Zelizer, Five Myths About the Future of Obamacare, Wash. Post: Monkey Cage (Dec. 16, 2013), [] (cautioning against making any predictions about future of Obamacare). Similarly, grants of deferred action and work benefits to undocumented immigrants may prove hard to repeal, as retrenchment may provoke stronger and more public immigrant opposition and immi­grants may have more support from diverse interests such as the business community. 127 See Adam B. Cox & Cristina M. Rodriguez, The President and Immigration Law Redux, 125 Yale L.J. (forthcoming 2015) (manuscript at 9–10 & n.22) (on file with the Columbia Law Review) (hypothesizing Obama immigration measures may encounter entrenchment similar to supposedly temporary measures from Immigration and Nationality Act of 1952); Bianca M. Figueroa-Santana, Note, Divided We Stand: Constitutionalizing Executive Immigration Reform Through Subfederal Regulation, 115 Colum. L. Rev. (forthcoming Dec. 2015) (manuscript at 5, 25–27, 35–37) (on file with the Columbia Law Review) (positing Obama’s immigration reform could become entrenched at subfederal level through state support and mobilization). But see John D. Skrentny & Micah Gell-Redman, Comprehensive Immigration Reform and the Dynamics of Statutory Entrenchment, 120 Yale L.J. Forum 325 (2011), [] (arguing past instances of legalization of unauthorized immigrants have not led to entrenchment of practice). Alternatively, interests that had previously focused on dissuading agency action may now up their pressure on Congress to intervene with legislation. 128 Cf. McNollgast, supra note 29, at 463–66 (arguing Congress responds to lobbying by regulated interests in deciding whether to intervene in agency action). In either case, the net effect is that “[b]y offering new policy approaches to old political issues, the executive can destabilize the status quo and build new legislative majorities.” 129 Whittington & Carpenter, supra note 30, at 502.

To be sure, the question remains whether agency action can break through congressional gridlock and prompt legislative response in our current world of polarized politics. Plainly, polarization makes congres­sional response harder, but the agenda-setting and status-quo-disrupting aspects of agency action still operate. Hence, the potential should exist for agency policy development to foster new legislative and political align­ments even here, particularly over time as the agency-developed policy becomes increasingly entrenched.

C. Incorporating Federalism

A crucial variable remains to be included in this account of agency action and polarization: the states. 130 See generally Jessica Bulman-Pozen & Gillian E. Metzger, Obama and the States: Presidential Unilateralism or Presidential Federalism?, 45-3 Publius (forthcoming 2016) (manuscript at 6–17) (on file with the Columbia Law Review) (discussing central contemporary importance of presidential–state engagement and administrative tools used to facilitate state participation in federal initiatives or allow the executive branch to take advantage of state initiatives). Cooperative federalism represents the reality of U.S. governance. In both regulatory and social welfare contexts, state governments work “cheek to jowl” with federal agencies in enforcing regulations and implementing national programs. 131 Heather K. Gerken, Federalism and Nationalism: Time for a Détente?, 59 St. Louis U. L.J. (forthcoming 2015) (manuscript at 13) (on file with the Columbia Law Review); see also Philip J. Weiser, Towards a Constitutional Architecture for Cooperative Federalism, 79 N.C. L. Rev. 663, 665, 668–73 (2001) (emphasizing political reality of cooperative federalism or joint federal–state implementation). These federal–state relationships vary in their details, but often involve delegation of a significant degree of discretion to state hands. 132 See Gillian E. Metzger, The Constitutional Duty to Supervise, 124 Yale L.J. 1836, 1852–55 (2015) (describing this delegation dynamic). Equally important, the constitutional prohibition on federal commandeering of state government means that Congress cannot force states to play these roles; the states must agree to take them on. 133 See Printz v. United States, 521 U.S. 898, 927–32 (1997) (holding federal government may not compel state executive officials to implement federal regulatory programs); New York v. United States, 505 U.S. 144, 177–83 (1992) (holding federal government cannot force state to legislate pursuant to congressional direction). Congress secures state agreement mainly through conditioned spending grants or the threat of federal preemption and direct regulatory control if the states fail to act. 134 See New York v. United States, 505 U.S. at 165–68 (identifying conditional spending and conditional preemption as “methods, short of outright coercion, by which Congress may urge a State to adopt a legislative program”); see also Erin Ryan, Federalism and the Tug of War Within 326–33 (2012) (describing incentives federal government uses to convince states to participate in federal regulatory programs).

Such joint federal–state implementation significantly impacts not just the shape of national programs, but interbranch dynamics at the national level. As Jessica Bulman-Pozen has noted, states can serve as potent checks on the executive branch through their participation in national programs. 135 Jessica Bulman-Pozen, Federalism as a Safeguard of Separation of Powers, 112 Colum. L. Rev. 459, 486–88 (2012) [hereinafter Bulman-Pozen, Federalism as Safeguard]. States can administer federal law in different ways than federal agencies do, creating competition on the ground. 136 See id. at 477–86 (emphasizing ways in which states diverge from, curb, and goad federal implementation); see also Roderick M. Hills, Jr., Federalism in Constitutional Context, 22 Harv. J.L. & Pub. Pol’y 181, 186–90 (1998) (“Congress can exploit compe­tition between federal agencies and non-federal governments to insure that neither betrays the purposes of federal law.”); David S. Rubenstein, Administrative Federalism as Separation of Powers, 72 Wash. & Lee L. Rev. 171, 175–78 (2015) (urging greater attention to relationship between administrative actions affecting federal–state balance and separation of powers). States also can take advantage of federal agencies’ dependence upon them for implementation to either curb federal administrative efforts with which they disagree or goad agencies into taking actions that they prefer. 137 See Bulman-Pozen, Federalism as Safeguard, supra note 135, at 475, 478–83 (“[S]tates engaging in diverging, curbing, and goading check federal executive authority . . . .”); see also Jessica Bulman-Pozen, From Sovereignty and Process to Administration and Politics: The Afterlife of American Federalism, 123 Yale L.J. 1920, 1939 (2014) [hereinafter Bulman-Pozen, Administration and Politics] (arguing states serve as checks on executive branch). Moreover, partisan divides at the state level mean that at least some state disagreement with federal agency policy is a given. Whether the checking comes from red states or blue states will depend on which party controls the presidency, but partisanship will lead one group or the other to play this role. 138 See Bulman-Pozen, Federalism as Safeguard, supra note 135, at 462–63, 500–03 (“Because there will never be party unity between the federal government and all fifty states, partisan resistance to the federal executive will arise even during periods of unified federal government.”).

Professor Bulman-Pozen emphasizes the ways that states therefore enforce statutes and assert congressional authority against the executive branch. 139 See id. at 488–92 (“Cooperative federalism schemes do not affect the federal executive’s power in isolation, but rather vis-à-vis Congress.”). Yet the dynamic here could be executive branch enhancing as well. The authority delegated to states in cooperative federalism contexts can allow states and federal agencies to work together to expand their powers at Congress’s expense, for example by agreeing to operate programs under different terms and requirements than specified in governing statutes. 140 See Bulman-Pozen, Administration and Politics, supra note 137, at 1944 (acknowledging, albeit discounting, this potential benefit of cooperative federalism). This concern of federal–state collusion at Congress’s expense is fueled by the growing agency use of “big waivers”—agency authorizations that broadly exempt states from statutory requirements. 141 Barron & Rakoff, supra note 88, at 267, 272–90 (describing increased use of waivers and coining “big waiver” term to describe instances in which agencies have broad, discretionary power to dispense with rules Congress has established in statutes). The transfor­mation of the No Child Left Behind Act (NCLB) through waivers negotiated between the states and the federal Department of Education (DOE) is a case in point. In order to obtain the waivers that would exempt them from the NCLB’s accountability requirements, forty-three states agreed to adopt measures that were not required by NCLB but instead represented Obama Administration policy priorities, such as teacher assessment and common standards for student assessment. 142 See Martin A. Kurzweil, Disciplined Devolution and the New Education Federalism, 103 Calif. L. Rev. 565, 606–08 (2015) (describing Elementary and Secondary Education Act (ESEA) Flexibility program); Kenneth K. Wong, Federal ESEA Waivers as Reform Leverage: Politics and Variation in State Implementation, 45 Publius 405, 410 (2015) (noting “[w]aiver applications were generally in alignment with the priorities of the Obama administration”); Sam Dillon, Obama Turns Some Powers of Education Back to States, N.Y. Times (Sept. 23, 2011), (on file with the Columbia Law Review) (describing Republican characterization of NCLB waivers as “power grab”). Partisanship can provide incentive here too, with states controlled by the President’s party being willing to help the Administration find a way to implement new policy initiatives in the face of a recalcitrant opposite-party dominated Congress. 143 This dynamic occurred when Democratic governors supported President Obama’s initiative to increase the minimum wage in the face of congressional inaction. See Obama to Push Minimum Wage Increase Wednesday at Governors’ Conference, N.Y. Daily News (Mar. 2, 2014), [] (describing Democratic governors’ actions). Another recent example involves greenhouse gas regulation: At the Obama Administration’s urging, California ceded its effort to limit new automobile emissions as part of a negotiated agreement under which the auto industry agreed to fuel emission standards set by the EPA. Jody Freeman, The Obama Administration’s National Auto Policy: Lessons from the “Car Deal,” 35 Harv. Envtl. L. Rev. 343, 345–46, 358–66 (2011); see also Jessica Bulman-Pozen, Partisan Federalism, 127 Harv. L. Rev. 1077, 1123–30 (2014) [hereinafter Bulman-Pozen, Partisan Federalism] (detailing how partisanship intersects with federalism to support party not in power at national level).

Reinforcing the executive branch’s ability to exploit state partic­ipation to its advantage is the heavily administrative and negotiated charac­ter of cooperative federalism programs. 144 See Ryan, supra note 134, at 280–314 (describing different forms of federal–state negotiations). As repeat bargainers who are mutually dependent, federal and state officials have both opportunity and incentive to make deals. 145 See John D. Nugent, Safeguarding Federalism: How States Protect Their Interests in National Policymaking 178–93 (2009) (emphasizing importance of federal–state administrative bargaining to states protecting their interests); Ryan, supra note 134, at 211 (discussing efficiency associated with repeat interactions between state and federal governments). The federal government’s dependence on the states can give the states substantial leverage in these negotiations. 146 See Nugent, supra note 145, at 173–75 (noting states use role as implementers of federal policy to promote own interests); cf. John Dinan, Implementing Health Reform: Intergovernmental Bargaining and the Affordable Care Act, 44 Publius 399, 400–01 (2014) (emphasizing importance of prior negotiation experience in federal–state bargaining). Yet federal agencies often hold the strings to substantial benefits for the states, be it federal funds, federal regulatory approval, or the like. 147 See, e.g., Dinan, supra note 146, at 418 (describing federal officials’ leverage with states with respect to implementation of ACA); sources cited supra note 134 (describing use of conditional benefits to secure state participation in federal programs); see also Timothy J. Conlan & Paul L. Posner, Inflection Point? Federalism and the Obama Administration, 41 Publius 421, 425–44 (2011) (describing Obama Administration’s use of conditional and categorical grants, partial preemption, and regulatory devolution with respect to states). And the complicated political economy underlying many national programs can mean that state officials are lobbied hard by in-state interests to reach an agreement with the federal government. 148 See Thompson & Gusmano, supra note 79, at 431 (describing Obama Administration’s strategy of using hospital lobby to pressure governors to expand Medicaid); Margaret Newkirk, Obama’s Emission Plan Splits Coal Lobby from Utilities in States, Bloomberg (Apr. 16, 2015, 5:00 am), [] (discussing how some utilities are convincing states not to pass model legislation that would hinder or delay federal regulations on carbon emissions).

Not surprisingly, political polarization has affected cooperative federalism. Like the national government, state governments are becoming more polarized and partisan. 149 Boris Shor & Nolan McCarty, The Ideological Mapping of American Legislatures, 105 Am. Pol. Sci. Rev. 530, 549–50 (2011) (“At the aggregate level, the states appear to follow the national pattern of high and growing polarization.”); see also Conlan & Posner, supra note 147, at 423 (noting ideological polarization characterizing state participation in federal grant programs). Scholars have identified a rise in ideological federalism at the state level, with states resisting federal initiatives largely out of partisan disagreement with the measures at hand rather than state-specific concerns. 150 See Conlan & Posner, supra note 147, at 444 (“Greater ideological polarization at all levels . . . may very well succeed in shifting the basis for intergovernmental policy formation . . . to an ideological party driven model.”); Thompson & Gusmano, supra note 79, at 429 (describing congressional polarization as creating federalism context in which “partisan identities of actors drive their behavior” and “state policy makers face pressure to act as committed, loyal party members”). States are manifesting this ideological resistance through intergovernmental lobbying, lawsuits, and refusal to implement federal programs. 151 See, e.g., David K. Jones et al., Pascal’s Wager: Health Insurance Exchanges, Obamacare, and the Republican Dilemma, 39 J. Health Pol. Pol’y & L. 97, 127–29 (2014) (describing role of partisanship in state resistance to creating health exchanges); Coral Davenport, Republican Governors Signal Their Intent to Thwart Obama’s Climate Rules, N.Y. Times (July 2, 2015), (on file with the Columbia Law Review) (reporting several governors’ threats not to enforce proposed greenhouse gas regulations); Alyson Klein, Sec. Duncan in Weaker Spot on NCLB Waiver Renewals, Educ. Wk. (Feb. 12, 2015),
.html (on file with the Columbia Law Review) (noting states pushing DOE more in waiver negotiations).
Prime examples include the suit by twenty-six states challenging the legality of the Obama Administration’s immigration actions, as well as state suits challenging proposed environmental regulations and the requirement that states seeking education grants and waivers adopt the Common Core academic standards. 152 See, e.g., Shanna Rose & Cynthia J. Bowling, The State of American Federalism 2014–15: Pathways to Policy in an Era of Party Polarization, 45 Publius 351, 358 (2015) (describing Texas-led lawsuit on immigration); Neela Banerjee, 12 States Sue the EPA over Proposed Power Plant Regulations, L.A. Times (Aug. 4, 2014), [] (describing coal-reliant states’ litigation against EPA’s proposed rule changes on greenhouse gases); Lyndsey Layton, Louisiana Gov. Bobby Jindal Sues Obama over Common Core State Standards, Wash. Post (Aug. 27, 2014),
96fad33_story.html [] (describing suit by Louisiana governor alleging DOE coerced states into adopting Common Core standards as condition of obtaining grants); see also Paul Nolette, State Litigation During the Obama Administration: Diverging Agendas in an Era of Polarized Politics, 44 Publius 451, 452, 464–66 (2014) (suggesting state attorney general litigation has become increasingly polarized and politically driven).
National party elites also lobby state officials in an effort to preserve a solid partisan front across both federal and state levels. 153 See, e.g., Thompson & Gusmano, supra note 79, at 429–32 (explaining how Obama Administration tried to rally support for ACA from state leaders); Davenport, supra note 151 (describing Senate majority leader Mitch McConnell’s effort to convince state governors not to enforce proposed climate rules). Professor Bulman-Pozen argues convincingly that partisanship is the central dynamic of federalism today, with “states function[ing] as important sites of partisan conflict, and partisanship, in turn, shap[ing] state governance . . . . Republican-led states challenge the federal government when it is controlled by Democrats, while Democratic-led states challenge the federal government when it is controlled by Republicans.” 154 Bulman-Pozen, Partisan Federalism, supra note 143, at 1079–80.

Even so, the dynamics of cooperative federalism and intergovern­mental bargaining also may temper polarization’s effects. Partisan coalitions linking national party elites and state officials can prove porous, with state officials more willing than their national counterparts to negotiate and cooperate with an opposite-party presidential admin­istration. 155 See, e.g., infra section II.B (discussing Republican state governors’ negotiations on Medicaid expansion); see also Simon F. Haeder & David L. Weimer, You Can’t Make Me Do It, But I Could Be Persuaded: A Federalism Perspective on the Affordable Care Act, 40 J. Health Pol. & L. 281, 292 (2015) (noting “most voices of ideological opposition usually dissipate” after early implementation stage of federal–state programs). Interestingly, NCLB waivers may represent the opposite dynamic, with resistance at the state level growing over time—although this growing resistance is not as clearly partisan. See Wong, supra note 142, at 412–23 (describing role of internal state politics and partisanship in growing controversies over waivers). Why these vertical relationships would display greater ability to reach across partisan lines than exists in horizontal relationships at the national level is unclear. One factor may be greater heterogeneity in political parties at the state level. Although polarization is growing in state governments, with around half the states more polarized than Congress, there is variation in ideological divides within both the national and state levels. 156 See Louis Jacobson, The Year of Single Party Control and Supermajorities, Governing (Jan. 7, 2013), [] (noting high number of unified governments and increasing polarization at state level but arguing impact on policy is less clear); Boris Shor, How U.S. State Legislatures Are Polarized and Getting More Polarized (in 2 Graphs), Wash. Post: Monkey Cage (Jan. 14, 2014), [] (describing rates and variation in state polarization). In particular, “many states have Republican state legislative con­tingents that are more liberal than the Democratic caucuses of many states.” 157 Shor & McCarty, supra note 149, at 549; see also Jones et al., supra note 151, at 128 (noting role of intra-Republican divisions in state decisions on health exchanges). A further contributor may be the composite nature of state governments. State legislatures can pressure or constrain state governors and other state officials in their interactions with federal agencies, and the presence of multiple elected and unelected state executive officials can provide more routes for federal–state cooperation. 158 See, e.g., Bridget A. Fahey, Health Care Exchanges and the Disaggregation of States in the Implementation of the Affordable Care Act, 125 Yale L.J. Forum 56, 59–64 (2015), [] (describing efforts by HHS to engage state insurance commissioners when state governors were reluctant to create exchanges). On the other hand, state legislators may be more partisan and extreme than state governors, and their influence may work to preserve vertical partisan lines. See Charles Barrilleaux & Carlisle Rainey, The Politics of Need: Examining Governors’ Decisions to Oppose the “Obamacare” Medicaid Expansion, 14 St. Pol. & Pol’y Q. 437, 440, 444–53 (2014) (describing evidence of influence of state legislature partisanship on Medicaid expansion decisions).

Perhaps most significant are the benefits a state gains by participating in a federal initiative, and the costs it incurs by resisting, which create strong political pressures to reach agreements even across partisan divides. 159 Costs have been an important factor in Republican states’ willingness to expand Medicaid under the ACA. See infra text accompanying notes 238–239. As the recent NCLB waivers suggest, congressional gridlock can pressure states to negotiate with the governing national administration even cross-party, as such negotiations may represent the only realistic route to relief from onerous federal requirements. 160 See Kurzweil, supra note 142, at 601–08 (tracing history of NCLB waivers and noting forty-three states now have waivers). The opposite dynamic is also true: Unable to obtain greater resources or grants of power from Congress, federal agencies may prove more accommodating to state demands in such intergovernmental negotiations—which may make the resultant deals even harder for states to resist on partisan grounds. 161 See infra text accompanying note 192 (detailing Obama Administration’s efforts to convince states to participate in ACA’s implementation). Insofar as congressional gridlock is itself a reflection of heightened polarization, this suggests that greater polarization and partisanship at the national level may create an impetus for more bipartisan engagement among state and national officials in the administrative sphere.

Cooperative federalism thus represents a critical means by which agencies and the executive branch can advance policy in a polarized world. It provides a mechanism through which opportunities continually arise for creating cross cutting alliances. National agencies can use state implementation to foster new regulatory and programmatic initiatives, and states can force national agencies to accept new policy approaches as the terms of their participation. As significant, bipartisanship in coop­erative federalism arrangements offers a distinct means by which agency action may be able to reformulate partisan divides at the apex level of government. Such bipartisan federal–state administrative initiatives may create room for national representatives to deviate from standard party lines, if nothing else by changing the on-the-ground reality to which national political leaders then must respond. 162 See infra notes 231–252 and accompanying text (discussing this dynamic in context of ACA and NCLB implementation).

To be sure, cooperative federalism can work in the opposite direction as well, with state level developments reinforcing national partisan divides. State partnership with federal agencies may forestall political pressure being brought to bear on national elected officials to overcome partisan stalemates. This dynamic arguably occurred in the education context, where for a time, administrative waivers diffused pressure for new legis­lation to replace NCLB. 163 See Kristina P. Doan, No Child Left Behind Waivers: A Lesson in Federal Flexibility or Regulatory Failure, 60 Admin. L. Rev. 211, 223–24 (2008) (“If DOE continues to issue waivers to states for NCLB’s larger problems, states and LEAs will have less incentive to challenge NCLB’s provisions. In turn, Congress will have less motivation to reform NCLB’s widespread problems through legislation . . . .”); Maggie Severns, The Plot to Overhaul No Child Left Behind, Politico (Jan. 2, 2015, 5:33 am),
/story/2015/01/the-plot-to-overhaul-no-child-left-behind-113857.html#ixzz3hsdVABqK [] (“The waivers opened a pressure valve that allowed members of Congress to delay rewriting the law . . . .” (internal quotation marks omitted)).
Indeed, state actions may even intensify partisan disagreement, with the effects of state implementation leading national officials to take more extreme positions on particular policy proposals. Here, an example comes from the State Children’s Health Insurance Program (SCHIP). Initially enacted with bipartisan support, state use of the program’s flexibility to provide benefits to more families led to Republican resistance to reauthorization, while simultaneously creating strong Democratic support for the program. 164 See Colleen M. Grogan & Elizabeth Rigby, Federalism, Partisan Politics, and Shifting Support for State Flexibility: The Case of the U.S. State Children’s Health Insurance Program, 39 Publius 47, 48, 60–65 (2009) (describing this dynamic in SCHIP and arguing it is potential feature of block grant programs due to flexibility such programs provide states).

Thus, state participation in federal programs is no sure panacea for polarization and may end up worsening policy divides. Yet in a world of polarized national politics and a gridlocked Congress, simply the potential for federal agencies to move policy by partnering with states is significant.

II. Administration and Polarization in Practice:The Example of the Affordable Care Act

All of this sounds good in theory, but does it hold up in practice? This Part takes up a central contemporary example of administration in polarization’s shadow: implementation of the ACA, otherwise known as Obamacare. 165 Patient Protection and Affordable Care Act, Pub. L. No. 111-148, 124 Stat. 119 (2010) (codified as amended in scattered sections of 26, 42 U.S.C.). Enacted in 2010 through a party-line vote using a special legislative mechanism to avoid a filibuster and subjected to repeated high-stakes litigation, the ACA continues to stand as a flashpoint for polarized politics. 166 See Barbara Sinclair, Unorthodox Lawmaking: New Legislative Processes in the U.S. Congress 210–22 (4th ed. 2011) (describing reconciliation procedure used to enact ACA); Lawrence R. Jacobs & Theda Skocpol, Hard-Fought Legacy: Obama, Congressional Democrats, and the Struggle for Comprehensive Health Care Reform, in Reaching for a New Deal: Ambitious Governance, Economic Meltdown, and Polarized Politics in Obama’s First Two Years 53, 70–76 (Theda Skocpol & Lawrence R. Jacobs eds., 2011) (describing partisan divisions at ACA’s enactment). Although somewhat sui generis given its high political salience, implementation of the ACA offers a useful window on how the dynamics sketched above materialize—or fail to materialize—in practice. Section II.A showcases the effects of polarization on the ACA’s implementation at the national level, while section II.B describes how negotiations between the Obama Administration and the states are creating more bipartisan cooperation on the ground.

A. Web of Controls: Congress, the Executive Branch, and the Courts

The ACA is a massively complex statutory scheme. It is sometimes described as a “three-legged stool”: 167 Eric J. Segall & Jonathan H. Adler, King v. Burwell and the Validity of Federal Tax Subsidies Under the Affordable Care Act, 163 U. Pa. L. Rev. Online 215, 217 (2015), []. The first leg consists of the prohibition on insurers denying insurance or setting premiums based on preexisting conditions; 168 42 U.S.C. § 300gg-3 (2012). the second is the individual mandate or the requirement that all individuals must have health insurance; 169 I.R.C. § 5000A (2012). and the third is the federal provision of tax subsidies to ensure that individuals can afford to purchase the insurance they are required to possess. 170 Id. § 36B. But this description omits several other “legs” that are critical to supporting the ACA regime: the creation of health exchanges on which individuals can purchase insurance, along with the specification of minimum essential health benefits that plans must cover; 171 42 U.S.C. §§ 18022, 18031. the expansion of Medicaid; 172 Id. § 1396a(a)(10)(A)(i)(VIII). the requirement that employers with fifty or more full-time employees provide health insurance (the employer mandate); 173 I.R.C. § 4980H. and the vast number of regulatory and oversight responsibilities needed in order to promulgate and enforce these requirements. 174 See Abbe R. Gluck, Intrastatutory Federalism and Statutory Interpretation: State Implementation of Federal Law in Health Reform and Beyond, 121 Yale L.J. 534, 580–82 (2011) [hereinafter Gluck, Intrastatutory Federalism] (providing brief overview of ACA).

To say implementing the ACA represents a major administrative challenge is an understatement. The sheer volume of necessary regu­latory activity is vast. 175 See, e.g., Maeve P. Carey & Michelle D. Christensen, Cong. Research Serv., R43348, Upcoming Rules Pursuant to the Patient Protection and Affordable Care Act: Spring 2013 Unified Agenda, at Summary, 1–2 (2014), [] (stating ACA contains forty or more grants of rulemaking authority and identifying fifty-six ACA-related regulatory actions during twelve-month period). Moreover, the Act creates a sea of complicated institu­tional structures that entail extensive coordination and negotiation, across both the federal and state governments. At the federal level, two cabinet departments—HHS, in particular its Center for Medicare and Medicaid Services (CMS), and Treasury, through the Internal Revenue Service—bear much of the burden of implementation. 176 C. Stephen Redhead & Janet Kinzer, Cong. Research Serv., R43474, Implementing the Affordable Care Act: Delays, Extensions, and Other Actions Taken by the Administration 1 (2015), []. But any number of other, often structurally independent institutions are delegated key respon­sibilities, from specifying essential medical services to cost containment. 177 See Curtis W. Copeland, Cong. Research Serv., R41315, New Entities Created Pursuant to the Patient Protection and Affordable Care Act app. at 20–38 (2010), (on file with the Columbia Law Review) [hereinafter Copeland, New Entities] (listing entities). Equally notable are the various officials and entities involved at the state level, from state legislatures and governors to state Medicaid agencies, health officials, insurance commis­sioners, and even quasi-state entities such as the National Association of Insurance Commissioners (NAIC). 178 See Gluck, Intrastatutory Federalism, supra note 174, at 589–94 (listing different roles states play in implementing ACA, both through quasi-governmental agencies and state bureaucracies); Timothy Stoltzfus Jost, Reflections on the National Association of Insurance Commissioners and the Implementation of the Patient Protection and Affordable Care Act, 159 U. Pa. L. Rev. 2043, 2045–47 (2011) (detailing role NAIC has played in implementing ACA). Indeed, some entities crucial to the ACA’s implementation—most notably, health exchanges—did not predate the Act and thus had to be created from scratch, which alone represented a massive administrative undertaking. 179 See Copeland, New Entities, supra note 177, app. at 20–38 (listing new entities created pursuant to ACA); Robert Pear, U.S. Cites Rise in Health Plan Signups as Sebelius Testifies, N.Y. Times (Dec. 11, 2013), (on file with the Columbia Law Review) (detailing problems with roll-out of federal health exchange).

1. Congressional Absence. — Implementing the ACA would prove hard at the best of times, but the difficulties agencies face are intensified by ongoing political resistance to the statute, evident in the introduction of over fifty bills seeking to repeal the ACA since its enactment. 180 Ed O’Keefe, The House Has Voted 54 Times in Four Years on Obamacare. Here’s the Full List., Wash. Post: Fix (Mar. 21, 2014), []. This resistance has an overwhelmingly partisan cast, with the ACA closely identified with President Obama and the Democratic Party, and opposition to “Obamacare” being a central Republican rallying cry. 181 See Seth Motel, Opinions on Obamacare Remain Divided Along Party Lines as Supreme Court Hears New Challenge, Pew Res. Ctr. (Mar. 4, 2015), [] (“[N]early nine-in-ten Republicans (87%) [are] against the [ACA] and roughly eight-in-ten Democrats (78%)[are] in support of it.”). Democratic control of the White House and lack of supermajority Republican representation in Congress have foiled congressional efforts at repeal. 182 See Russell Berman, Why Republicans Are Voting to Repeal Obamacare—Again, Atlantic (Feb. 3, 2015), [] (arguing repeal efforts “doomed to fail” because “Republicans have nowhere near the veto-proof majority they’d need to kill Obamacare” and also noting Republican disagreement on appropriate measure to replace it). Instead, Republican opponents in Congress have resorted to other methods to express their dislike of the legislation—such as litigation, funding constraints, hearings, and investigations. 183 See, e.g., Complaint at ¶¶ 25–50, U.S. House of Representatives v. Burwell, No. 14-cv-01967 (D.D.C. Nov. 21, 2014), 2014 WL 6492097 (arguing Obama Administration violated Article I of the Constitution by making certain ACA payments to insurers absent specific appropriation authorization and by postponing activation of ACA’s employer mandate); C. Stephen Redhead & Janet Kinzer, Cong. Research Serv., R43289, Legislative Actions to Repeal, Defund, or Delay the Affordable Care Act 6–9 (2015), [] (discussing appropriations measures taken against ACA); Seung Min Kim, GOP Aims to Sink Obamacare with Reconciliation, Politico (Mar. 18, 2015, 2:31 pm), [] (outlining effort to use budget reconciliation measures to thwart ACA).

The ACA’s implementation thus represents a textbook case of polarization’s first effect on agencies. Congress’s ability to direct and control imple­mentation of the statute is quite limited. 184 See supra notes 39–44 and accompanying text (describing Congress’s limited ability to enact legislation as result of polarization). At the same time, congres­sional gridlock has also worked to constrain agencies by preventing needed legislative fixes. 185 See Gluck, O’Connell & Po, supra note 66, at 1829–30 (noting omnibus bills and bills that bypass complete legislative processes, like the ACA, often come with “unexpected ambiguities, errors, and other complexities” that create implementation problems). The ACA contains several provisions that could benefit significantly from congressional tweaking or alteration. For example, the addition of just a few words could have removed any doubts about the availability of tax subsidies for insurance purchased on a federal exchange without Supreme Court intervention, and revising statutory effective dates could address concerns that employers and insurers need additional time to comply with the Act. 186 See Nicholas Bagley, The Legality of Delaying Key Elements of the ACA, 370 New Eng. J. Med. 1967, 1967–69 (2014) (questioning legality of employer mandate delay); Robert Pear, Four Words that Imperil Health Care Law Were All a Mistake, Writers Now Say, N.Y. Times (May 25, 2015), (on file with the Columbia Law Review) (reporting congressional staff and members involved in enacting ACA viewed statutory text underlying challenge to availability of tax subsidies on federal exchanges as “drafting error”). Particularly in need of a legislative response is the major hole in health insurance coverage for low income individuals created by state decisions not to expand Medicaid. 187 See Nat’l Fed’n of Indep. Bus. v. Sebelius, 132 S. Ct. 2566, 2605–08 (2012) (holding HHS may not withdraw existing Medicaid funds for failure to comply with Medicaid expansion); Sam Dickman et al., Opting Out Of Medicaid Expansion: The Health and Financial Impacts, Health Aff. Blog (Jan. 30, 2014), [] (finding nearly eight million uninsured people with incomes below poverty line could have gotten coverage had their states opted into Medicaid). Given polarization and divided government, however, legislation addressing these issues is not forthcoming. The lack of legislative fixes has created significant litigation risk and other challenges for the agencies charged with implementing the Act. 188 See Robert Pear, Legal Challenges Remain for Health Law, N.Y. Times (June 26, 2015), (on file with the Columbia Law Review) [hereinafter Pear, Legal Challenges] (describing variety of lawsuits brought against ACA). This highlights the point made above, that whether congressional gridlock works to empower or hamper, agencies cannot be assessed separately from the underlying legislation. 189 See supra text accompanying notes 54–57.

2. Presidential Presence. — The ACA’s implementation is equally a poster-child for the second effect of polarization, namely an increase in executive branch and presidential unilateralism. 190 See supra notes 61–73 and accompanying text (discussing increased presidential control and unilateralism resulting from congressional polarization). On numerous occasions, the Obama Administration has taken unilateral action to address implementation challenges in the absence of legislation. Particularly prominent on this score was the Administration’s decision to delay the effective date of the employer mandate and some regulatory requirements on insurers. 191 See Bagley, supra note 186, at 1967–69 (“In the administration’s view, the delays are a routine exercise of the executive branch’s traditional discretion to choose when and how to enforce the law.”). The Administration has also used its waiver-granting authority under Medicaid to encourage states to undertake the Medicaid expansion. 192 See Dinan, supra note 146, at 411–17 (describing bargaining between administration and state officials); Thompson & Gusmano, supra note 79, at 432–35 (“[W]aivers became an attractive tool for enticing state participation.”). Moreover, this unilateralism has a decidedly presidential cast. The ACA’s high political valence has meant that the White House is closely involved in implementation and is publicly charged with responsibility for problems, such as the failed roll-out of 193 See Thompson & Gusmano, supra note 79, at 429 (“Intense partisan polarization surrounding the ACA meant that the administrative presidency, rather than the career bureaucracy, would be front and center during implementation.”); N.C. Aizenman, Peter Wallsten & Karen Tumulty, White House Compromise Still Guarantees Contraceptive Coverage for Women, Wash. Post (Feb. 10, 2012), [] (detailing White House response to complications with ACA and centrality of White House in resolving disputes regarding ACA). White House oversight is further fostered by the fact that implementation is in the hands of executive agencies led by political heads and whose rulemaking activities are subject to centralized OIRA review. 194 See Maeve P. Carey, Cong. Research Serv., R43622, Upcoming Rules Pursuant to the Patient Protection and Affordable Care Act: The Spring 2014 Unified Agenda 5–11 (2014), [] (listing proposed and final rules submitted by agencies for OIRA review).

As a result, the extent to which the ACA’s complicated institutional structure has created robust internal checks against executive branch overreach is open to question. 195 For a description of the potential for internal executive branch constraints to check agency action in lieu of a polarized Congress, see supra text accompanying notes 106–109. Such checking may be occurring, but disagreements among the agencies involved or between these agencies and the White House are not publicly evident. It also seems possible that the ACA’s political aspect and substantial White House involvement have limited the influence of dissenting internal voices.

Reinforcing the perception of limited internal constraints is the mixed procedural record of the ACA’s implementation. The ACA agencies used joint notice-and-comment rulemaking for several key regu­latory decisions, such as rules governing state exchanges and the avail­ability of tax subsidies for individuals purchasing insurance. 196 E.g., Health Insurance Premium Tax Credit, 77 Fed. Reg. 30,377 (May 23, 2012) (to be codified at 26 C.F.R. pts. 1, 602) (detailing final regulations relating to health insurance tax credit enacted by ACA). Some rules have been issued as interim final rules, under which the proposed rule has immediate effect but the agency commits to undertaking a full notice-and-comment rulemaking prior to issuing a final rule. 197 See Curtis W. Copeland & Maeve P. Carey, Cong. Research Serv., R41586, Upcoming Rules Pursuant to the Patient Protection and Affordable Care Act 5–13 (2011), [] (describing ACA rulemakings listed as forthcoming and identifying several uses of interim final rulemaking). But a number of significant decisions were promulgated much more infor­mally, and the heavy reliance on waivers in Medicaid expansion has also limited transparency and public participation in implementation. 198 See Nicholas Bagley & Helen Levy, Essential Health Benefits and the Affordable Care Act: Law and Process, 39 J. Health Pol. Pol’y & L. 441, 442–43 (2014) (“[A]nnouncing the policy [of allowing each state to choose a benchmark plan] through an Internet bulletin . . . allowed the agency to sidestep conventional administrative procedures—including notice and comment, immediate review in the courts, and OIRA oversight . . . .”).

3. The Centrality of the Courts. — The story of the ACA’s imple­mentation strongly supports a third effect of polarization postulated above: the increasing importance of judicial review of administrative action. 199 See supra text accompanying notes 93–105 (discussing impact of congressional polarization on judicial oversight of agency action). Indeed, the real constraints on ACA implementation at the national level have been the courts. The years since the ACA’s enactment have witnessed an endless stream of litigation against it. The most signif­icant to date is the Supreme Court’s decision in NFIB v. Sebelius, which held that failure to expand Medicaid would not cost a state all of its Medicaid funding and paved the way for a large number of states to refuse to expand. 200 Nat’l Fed’n of Indep. Bus. v. Sebelius, 132 S. Ct. 2566, 2605–07 (2012). But NFIB is just one of three Supreme Court cases so far that involve the ACA, a remarkable record given that the Act has been on the legislative books for only five years. In Burwell v. Hobby Lobby Stores, Inc. the Court ruled that regulations requiring for-profit closely held com­panies to cover contraception for their employees violated the Religious Freedom Restoration Act (RFRA). 201 134 S. Ct. 2751, 2759 (2014). And in the most recent decision, King v. Burwell, the Court held that tax subsidies to help individuals cover the cost of insurance are available through health ex­changes run by either the federal government or the states. 202 135 S. Ct. 2480, 2496 (2015). Many more lawsuits challenging administrative implementation of the ACA have been filed, although a large number were dismissed on jurisdictional grounds. 203 See Pear, Legal Challenges, supra note 188 (detailing additional ACA challenges pending in courts after King); Timothy Jost, Implementing Health Reform: ACA Litigation Beyond King v. Burwell, Health Aff. Blog (June 23, 2015), [] (same). Given this continual stream of Supreme Court and broader judicial intervention, the agencies charged with the ACA’s imple­mentation no doubt act under the assumption that their decisions will be subject to legal challenge.

The decisions also stand out for their lack of deference to the agencies involved. 204 Cf. Alex J. Luchenitser, A New Era of Inequality? Hobby Lobby and Religious Exemptions from Anti-Discrimination Laws, 9 Harv. L. & Pol’y Rev. 63, 65 (2015) (noting Hobby Lobby’s lack of deference to government’s determination that no less restrictive alternatives existed); John F. Manning, Foreword: The Means of Constitutional Power, 128 Harv. L. Rev. 1, 42 (2014) (emphasizing NFIB’s lack of deference to Congress). To be sure, the decisions in NFIB and Hobby Lobby involved constitutional and statutory claims outside of the ACA, for which independent judicial judgment is the governing standard. 205 Hobby Lobby Stores, Inc., 134 S. Ct. at 2759, 2789–93 (noting case turns on application of RFRA and refusing to defer to HHS on policy questions such as effect­tiveness and burden associated with alternative approaches); Nat’l Fed’n of Indep. Bus., 132 S. Ct. at 2579–80 (“Our respect for Congress’s policy judgments thus can never extend so far as to disavow restraints on federal power that the Constitution carefully constructed.”). Even so, these cases also involved questions on which the agencies’ expertise seems relevant, such as whether the healthcare markets and health insurance markets are meaningfully separated or the feasibility of extending HHS’s accommodation for non-profit employers to for-profit ones. The lack of deference in King is even more striking: There, the Court expressly stated that even though the relevant statutory language was ambiguous, it was not going to defer to the implementing agency’s interpretation of the statute as the Chevron doctrine would ordinarily instruct it to do. 206 King, 135 S. Ct. at 2488–89 (citing Chevron U.S.A. Inc. v. Nat. Res. Def. Council, 467 U.S. 837, 842–43 (1984)). The Court deemed deference inappropriate because the availability of tax subsidies was “a question of deep economic and political significance that is central to this statutory scheme; had Congress wished to assign that question to an agency, it surely would have done so expressly.” 207 Id. at 2489 (internal quotation marks omitted). Although precedent exists supporting such a major question exception to Chevron, those cases involve instances in which the Court interpreted the statute differently than the agency involved, 208 See, e.g., FDA v. Brown & Williamson Tobacco Corp., 529 U.S. 120, 160 (2000) (“[W]e are confident that Congress could not have intended to delegate a decision of such economic and political significance to an agency in so cryptic a fashion.”); MCI Telecomm. Corp. v. Am. Tel. & Tel. Co., 512 U.S. 218, 231 (1994) (explaining FCC did not have ability to modify Congress’s 1934 law). whereas in King the Court went out of its way to independently interpret the statute to reach the same result as the agency.

As a result, King may signal that the Court is positioning itself as a check against agency efforts to transform statutory schemes in contexts where partisan legislative dysfunction prevents congressional response. 209 Alternatively, King may reflect a developing pullback on agency deference that stems from broader concerns about the legality of deference—a pullback that up until now has been limited to concurring and dissenting opinions. See, e.g., Perez v. Mortg. Bankers Ass’n, 135 S. Ct. 1199, 1211–12 (2015) (Scalia, J., concurring) (discussing Chevron deference’s inconsistency with Administrative Procedure Act and arguing against deference to agency interpretations of agency rules); id. at 1213–22 (Thomas, J., concurring) (arguing deference to agency interpretations of rules is unconstitutional); City of Arlington v. FCC, 133 S. Ct. 1863, 1877–79, 1885–86 (2013) (Roberts, C.J., dissenting) (rejecting Chevron deference for agency jurisdictional determinations and invoking concerns about the “vast power” agencies wield). Although such legality concerns go beyond a focus on polarization, they might well be intensified by polarization’s potential to aggrandize agency powers. Upcoming challenges to the EPA’s clean power regulations may well clarify whether the Court is making such a move. 210 See In re Murray Energy Corp., 788 F.3d 330, 333–34 (D.C. Cir. 2015) (rejecting challenge to proposed clean power rule because rule not yet final). These regulations similarly address matters of “deep political and economic significance” and involve a policy issue—climate change and greenhouse gas regulation—charac­terized by intense political divides. 211 See, e.g., David W. Case, The Lost Generation: Environmental Regulatory Reform in the Era of Congressional Abdication, 25 Duke Envtl. L. & Pol’y Forum 49, 59–61 (2014) (describing increased polarization around environmental issues since 1980s and current stalemate in Congress); Davenport, supra note 151 (describing Republican congressional leadership’s efforts to undermine proposed clean power rules). Thus, if the Court is pulling back from deference to account for the risks of agency aggrandizement in the face of congressional dysfunction, such resistance to deference should manifest here. Indeed, King itself linked the ACA and greenhouse gas contexts by invoking its recent refusal to defer to an initial set of green­house gas regulations as precedent for the rejection of deference in King. 212 See King, 135 S. Ct. at 2488–89 (citing recent Utility Air Regulatory Group decision where Court rejected EPA’s effort to deviate from express statutory text to accommodate practical realities of regulating greenhouse gas emissions).

B. Federalism, Partisanship, and the ACA

In short, the story of the ACA’s implementation fits the expected tale of a web of agency control relationships refashioning itself in light of the polarized realities of the day. This story also exemplifies the importance of the federalism dynamics identified in section I.C. A signal feature of the ACA is the extent to which this web of control spans both national and state levels. The states play numerous critical roles in the ACA regime, including enforcing the ACA’s nondiscrimination provisions, overseeing insurance plans, operating exchanges, and expanding Medicaid. 213 Abbe Gluck has written about this state involvement in particular detail. Gluck, Intrastatutory Federalism, supra note 174, at 589–92; see also Dinan, supra note 146, at 400–01 (describing ACA provisions that depend on state participation). Getting the states on board thus has been a central imperative for the federal agencies charged with implementation. The polarized politics surrounding the ACA, however, have manifested at the state level as well, leading to substantial Republican-state resistance to cooperating with the Obama Administration over the Act. As described below, the resultant need to convince states to take part has been a potent force in shaping how federal agencies, in particular HHS, have approached ACA implementation. More importantly, by convincing red states to expand Medicaid and undertake other ACA-related roles, HHS is giving ACA implementation a bipartisan character. As a result, at the ground level at least, the polarized dynamics long characterizing the ACA are slowly being transformed.

1. The States and ACA Implementation. — Although state participation runs throughout the ACA, the states are particularly critical of the Medicaid expansion and ACA health exchanges. In the ACA, Congress made Medicaid the mechanism for providing health care to low-income individuals and families up to 138% of the poverty line. 214 Medicaid is generally available to those whose “modified adjusted gross income” is at or below 133% of the poverty line. 42 U.S.C. § 1396a(l)(2)(A)(2012). However, modified adjusted gross income is reduced by the difference between the dollar amount of that limit and the dollar amount of that limit increased by five percentage points. Id. § 1396a(e)(14)(I). As a result, the effective upper limit is 138% of the poverty line. The exchanges, in turn, are central to providing affordable insurance above that income level. Exchanges allow individuals to compare and shop among plans that meet certain minimum requirements, benefit from the lower premiums through pooling, and obtain insurance subsidies if qualified. 215 See Bernadette Fernandez & Annie L. Mach, Cong. Research Serv., R42663, Health Insurance Exchanges Under the Patient Protection and Affordable Care Act (ACA) 13–23 (2013), [] (describing operation of health exchanges); Haeder & Weimer, supra note 155, at 301–02 (describing exchanges’ informational value); Jones et al., supra note 151, at 100–04 (describing history of health exchanges and the benefits of pooling to lower the cost of insurance). Although those with incomes between 100% and 400% of the poverty line can qualify for subsidies on ACA-created health exchanges, Medicaid represents the only means for accessing healthcare for those with incomes below the poverty line. 216 See Henry J. Kaiser Family Found., Explaining Health Care Reform: Questions About Health Insurance Subsidies (July 2012), [] (explaining how, in states without expanded Medicaid coverage, individuals with income below 100% of poverty line are ineligible for exchange subsidies). Moreover, the ACA contains no provision for federal expansion of Medicaid if the states fail to do so. As a result, the only way for federal agencies to provide healthcare to this core low-income population is to convince the states to expand their Medicaid programs. Such a federal fallback does exist with respect to the ACA health exchanges, but even here, substantial advantages from state operation—including ACA exchange funding grants as well as state administrative capacity and expertise—make HHS eager to obtain state participation. 217 See Thompson & Gusmano, supra note 79, at 436–38 (explaining Obama Administration’s ongoing efforts to convince states to operate exchanges).

Many states, however, were reluctant to expand Medicaid or create an exchange. 218 See Jones et al., supra note 151, at 98–100, 104–10 (elaborating on history of exchange implementation); Rose & Bowling, supra note 152, at 359 (describing Medicaid expansion over time). Faced with this implementation challenge, HHS has taken a flexible and accommodating stance towards the states. On the Medicaid front, HHS has used its Medicaid waiver authority to approve approaches to Medicaid expansion that differ notably from the tradi­tional Medicaid model. 219 See Thompson & Gusmano, supra note 79, at 432–35 (describing Obama Administration’s use of Medicaid waivers, in particular waivers allowing Medicaid funds to subsidize private insurance premiums, to entice states to expand Medicaid); Christine Vestal, More States Lean Toward Medicaid Expansion, Pew Charitable Trusts: Stateline (Jan. 28, 2015), [] (describing variations and HHS’s responsiveness to Republican proposals). These include allowing states to use Medicaid funds as premium assistance to help newly eligible Medicaid beneficiaries purchase insurance on exchanges, requiring premiums and copayments from Medicaid beneficiaries and temporarily barring re-enrollment if a beneficiary fails to pay premiums, and eliminating certain benefits ordinarily required under Medicaid. 220 See Robin Rudowitz, Samantha Artiga & MaryBeth Musumeci, Kaiser Commission on Medicaid and the Uninsured, Issue Brief: The ACA and Medicaid Expansion Waivers 1, 7–11, (2015), [] (describing Medicaid expansion waivers).

HHS has been similarly flexible with respect to ACA health exchanges. 221 See Dinan, supra note 146, at 403–08 (describing HHS’s flexibility with respect to health exchanges); Thompson & Gusmano, supra note 79, at 436–38 (detailing HHS’s efforts to convince states to run health exchanges); see also Shihyun Noh & Dale Krane, Partisan Polarization, Administrative Capacity, and State Discretion in the Affordable Care Act 9 (Mar. 14, 2014) (unpublished manuscript), [] (describing HHS as pursuing “strategy of negotiation with the states”). Many states initially signaled willingness to run a state-operated exchange but changed course over the following months. 222 Jones et al., supra note 151, at 110–16. This change reflected a rightward shift in state elections in 2010 as well as growing political pressure on Republican state political officials to hold firm in rejecting Obamacare. 223 Id. at 110–28. In response, HHS extended deadlines, designed a variety of roles for states to play in the exchange system short of running an exchange, and defined key regulatory terms in a fashion that incorporated state policy choices. 224 See Dinan, supra note 146, at 403–08, 415 (describing limits of federal accommodation of state requests for flexibility). For a detailed discussion of Medicaid provisions that HHS has not allowed to be waived, see Rudowitz, Artiga & Musumeci, supra note 220, at 11–12. To be sure, HHS refused some state requests. It rejected state proposals to expand Medicaid only partially and denied state efforts to impose work requirements as a condition for Medicaid. 225 See Dinan, supra note 146, at 416–18 (highlighting hard limits on state negotiations). But the agency’s overall attitude has been one of accommodation and compromise. 226 See Thompson & Gusmano, supra note 79, at 439–41 (“The dominant motif of the Obama administration’s exchange strategy involved going the last mile to encourage state participation.”).

These approaches to Medicaid expansion and health exchange operation represent significant policy developments. This is especially true for Medicaid, where federal officials had previously rejected state efforts to impose costs on program participants and other measures that were later accepted as part of expansion waivers. 227 See Henry J. Kaiser Family Found., Policy Brief: Premiums and Cost-Sharing in Medicaid 3–5 (2013), [] (discussing rules about cost sharing in Medicaid, federal denials of waivers relating to cost sharing, and proposed rule changes). But it is also the case with respect to the ACA health exchanges, where HHS’s accommodating stance allows experimentation with different combinations of federal and state expertise. Nor, moreover, is HHS’s flexibility mandated by the statute; if anything, the ACA posits federal and state health exchanges as alternatives rather than partnering entities. 228 See 42 U.S.C. §§ 18031(b)(1), 18041(c)(1) (2012) (mandating each state establish exchange and instructing HHS to establish exchange for state if state chooses not to do so). Similarly, Medicaid and health exchanges stand as alternative pillars of the ACA, with no express statutory authorization for their combination by using Medicaid funds to purchase insurance through an exchange. 229 See id. § 1396a(a)(10)(A)(i)(VIII) (expanding Medicaid); id. § 18031(b)(1) (establishing exchanges). Indeed, statutory requirements that the costs of programs granted waivers be no greater than the cost of traditional Medicaid coverage, § 18051(a)(2)(A), are conditions that premium assistance expansions may have trouble meeting. See Dinan, supra note 146, at 415 (explaining “cost of purchasing private insurance on the exchanges and providing subsidies for Medicaid-eligible persons is expected to be higher than the cost of traditional Medicaid coverage”).

The lack of express statutory basis for these approaches does not mean that HHS exceeded its authority in approving them. HHS has broad implementing authority under the ACA and longstanding waiver power under Medicaid. 230 See id. § 1396n (granting HHS secretary authority to waive certain Medicaid requirements); Curtis W. Copeland, Cong. Research Serv., R41180, Regulations Pursuant to the Patient Protection and Affordable Care Act (P.L. 111-148) 3–14 (2010), [] (identifying provisions in ACA regarding rulemaking by federal agencies to implement legislation). But it highlights the degree to which capacious statutory delegations allow significant federal policymaking and adjust­ment to go forward at the impetus of agencies, even when polarized and divided government produces stalemate in Congress. The Obama Administration has used its administrative powers under the ACA, along with the ACA’s incorporation of the states, to push federal health reform in new directions. At the same time, the states have used their programmatic leverage to obtain executive branch concessions over implementation. The resulting federal–state agreements have also served to sideline a Congress locked in partisan warfare. The ACA may be the most significant social-welfare legislation in decades, but congressional polarization means that the development of national health reform is now firmly an executive branch-, state-, and especially president-led project.

2. Reshaping the Political Terrain. — Partisanship defined the ACA’s adoption and largely controlled its initial implementation. 231 See Jacobs & Skocpol, supra note 166, at 77–78 (noting even after ACA’s enactment, controversies over reform framework remained at “fever pitch”); Thompson & Gusmano, supra note 79, at 429 (claiming “ACA was the poster child for the well documented trend toward partisan polarization in the United States over the last several decades”). Solid blue states embraced the ACA, while solid red states largely refused to participate, and purple states were somewhere in between. 232 See Dinan, supra note 146, at 408–10 (describing how support for state-run exchanges and Medicaid expansion fell along partisan lines); Timothy M. Callaghan & Lawrence R. Jacobs, Dynamic Federalism and the Implementation of the Affordable Care Act 7 (2013) (unpublished manuscript), (on file with the Columbia Law Review) (“The impact of partisanship is amply evident in state decision making on implementation of the ACA’s Medicaid expansion.”);Noh & Krane, supra note 221, at 5–6, 8, 19–20 (noting few exceptions to partisan split in supporting state-based exchanges). Oddly, the Medicaid expansion and creation of state health exchanges have had an opposite trajectory over time. At the outset, Republican state officials signaled willingness to create state exchanges but strongly resisted ex­panding Medicaid. Over time, however, Republican opposition to the exchanges grew significantly, whereas increasing numbers of Republican-led states have agreed to expand Medicaid. 233 See Jones et al., supra note 151, at 110–16 (detailing shift from exchange planning to resistance to implementing ACA in states with Republican legislatures); Sarah Ferris, Red-State Governors Discuss Expansion of Medicaid with Obama, Hill (Jan. 6, 2015, 5:39 pm), [] (indicating expansive federal funding convinced conservative states to begin expanding Medicare); John Tozzi, Why Red States Will Expand Medicaid, Like It or Not, Bloomberg (July 30, 2014), [] (comparing trends of coverage in Democrat- and Republican-controlled states).

The recent Republican move toward expanding Medicaid deserves special note. A steady trickle of states with either Republican governors or Republican-controlled legislatures, or both, have expanded Medicaid since 2012. 234 See Jeffrey M. Jones, Massachusetts, Maryland Most Democratic States, Gallup (Feb. 4, 2015), [] (identifying solid Republican states). This includes some solid-red states like Indiana and Montana, while governors in other red states like Idaho, Utah, Tennessee, and Wyoming have or are discussing expansion waivers with HHS. 235 Data on state expansions is as of October 11, 2015. See Henry J. Kaiser Family Found., Status of State Action on the Medicaid Expansion Decision, [] [hereinafter Status of State Action] (last visited Oct. 11, 2015); see also David Ramsey, Red States Are Reinventing Medicaid to Make It More Expensive and Bureaucratic, New Republic (Jan. 18, 2015),
/article/120781/republican-governors-accept-medicaid-expansion-make-it-costlier [http://] (noting Tennessee, Wyoming, and Utah are in negotiations with HHS); Vestal, supra note 219 (describing expansion in Indiana and expansion efforts in Alaska, Idaho, Montana, Tennessee, Utah, and Wyoming, as well as favorable gubernatorial statements in Alabama, North Carolina, and Texas). The governors of Tennessee and Wyoming reached agreements on waivers with HHS but failed to obtain approval from their state legislatures. Shanna Rose, Opting In, Opting Out: The Politics of State Medicaid Expansion, 13 Forum 63, 76–77 (2015) (chronicling Republican governors’ efforts to create private program option for Obamacare). Montana’s expansion is still awaiting waiver approval from HHS. See Status of State Action supra.
Republican governors have taken the lead on expansion, often over legis­lative opposition and occasionally legislative defeat of expansion plans. 236 See Rose, supra note 235, at 76–77 (exploring defeat of ACA waivers in Tennessee and Wyoming).

Several factors help explain why Republican governors have been willing to break party ranks over the Medicaid expansion, notwith­standing the polarized politics of the ACA. A major contributor is HHS’s accommodating approach detailed above, which has allowed Republican governors to design expansion programs that reflect their conservative policy preferences. 237 See Dinan, supra note 146, at 414 (noting in Arkansas, for example, “federal officials had acquiesced in virtually all the demands state officials had made”); Thompson & Gusmano, supra note 79, at 433 (emphasizing importance of HHS’s flexibility); see also Sarah Kliff, Could Obamacare Make Medicaid More Republican?, Wash. Post (Feb. 25, 2013), http:// [] (noting potential of waivers to address Republican policy preferences).
Pressure from powerful in-state interests, especially hospitals which have a financial incentive to have states expand, has also been significant, as well as public sentiment generally supporting expan­sion. 238 See Thompson & Gusmano, supra note 79, at 431–32 (describing strong support from hospitals for Medicaid expansion); see also Rose, supra note 235, at 65–68, 72, 78 (documenting support for Medicaid expansion from hospitals, business groups, advocacy groups, and other organizations). Another factor is the generous federal match, with the federal government initially picking up 100% of the costs of Medicaid expansion, declining only to 90% in 2020. 239 42 U.S.C. § 1396d(y)(1) (2012); see also Lawrence R. Jacobs & Timothy Callaghan, Why States Expand Medicaid: Party, Resources, and History, 38 J. Health Pol. Pol’y & L. 1023, 1033 (2014) [hereinafter Jacobs & Callaghan, Why States Expand] (“Federal funding during unsteady economic and budgetary times may be especially attractive to states with particularly strained circumstances . . . .”). State budget pressures have also played a role; although expanding Medicaid carries some costs for states, it also allows states to shift to the federal government the costs for care they are already providing to newly eligible individuals. 240 See Rose, supra note 235, at 68, 71–72 (describing cost shifting possibilities); see also Jacobs & Callaghan, Why States Expand, supra note 239, at 1033–35 (noting potential for states to shift costs but concluding economic considerations generally do not drive expansion decisions). Some Republican states have not expanded Medicaid despite the sizable federal funds in play. See, e.g., Wade Goodwyn, Texas Loses Billions to Treat the Poor by Not Expanding Medicaid, Advocates Say, NPR (May 29, 2015, 5:08 am),
/2015/05/29/410470081/texas-didn-t-expand-medicaid-advocates-say-money-is-being-left-on-the-table (on file with the Columbia Law Review) (discussing Texas’s decision to not expand Medicaid and thus forego billions in federal funds).
Scholars have also identified a state’s prior policy on Medicaid eligibility and benefits as an important contributing factor, so that “the gravitational pull of policy history . . . is cross-pressuring states and moderating the effects of polit­ical parties” in expansion decisions. 241 Jacobs & Callaghan, Why States Expand, supra note 239, at 1036–37. Two other potential moderating factors are state administrative capacity and prior experience with intergovernmental bargaining, although these factors likely overlap. See Timothy Callaghan & Lawrence R. Jacobs, Process Learning and the Implementation of Medicaid Reform, 44 Publius 541, 542 (2014) (describing bargaining procedures); Jacobs & Callaghan, Why States Expand, supra note 239, at 1039 (discussing administrative capacity). Moreover, separating capacity and political party influences is particularly difficult as Democratic states tend to have greater administrative capacity. Id. A desire to provide health services to uninsured low-income individuals may also play a role, as well as the recognition that “taxpayers in non-expansion states are paying federal taxes that support the expansion of coverage in other parts of the country.” 242 Rose, supra note 235, at 69, 78; see also Barrilleaux & Rainey, supra note 158, at 440, 447–48, 453 (noting Governor John Kasich’s emphasis on needs of uninsured but concluding “economics and need have little effect” on governors’ expansion decisions).

Whatever the cause, growing red-state willingness to participate in Medicaid expansion represents a significant splintering in the Republican opposition to the ACA. 243 See Rose, supra note 235, at 79 (“[A]dvocates can point to the growing list of participating red states as evidence that expansion is a politically neutral issue.”). This does not mean that the Medicaid expansion will no longer be a source of partisan dispute, but the points of contestation are likely to be more focused on the terms of expansion rather than on whether the expansion occurs at all. Republican officials at the state and national level may push for even greater flexibility than the Obama Administration has granted, whereas Democratic leaders may become concerned about too many concessions, particularly once the ACA’s broad waiver authority comes into effect in 2017. 244 42 U.S.C. § 18052. But these disagreements suggest a less zero-sum debate, one more focused on challenging program details than the program’s existence. These are also issues on which Republican and Democratic officials may not line up as consistently or clearly in oppositional camps. 245 A number of Democratic governors have pushed expansion models containing policy approaches typically favored by Republicans, such as use of private insurance or co-pays, while several Republican governors have sought to expand Medicaid on fairly traditional terms. See Rose, supra note 235, at 71–72, 75–76 (noting traditional Medicaid expansion in Nevada and Arizona, with Republican governors, and premium assistance approach in Arkansas, with Democratic governor). In addition, these future debates will be informed by evidence from the different policy approaches contained in the waivers that HHS and states have negotiated, creating a shared factual basis about which expansion models are the most effective that may undermine existing partisan divides. 246 See Sunstein, Partyism, supra note 97 (manuscript at 15) (“[W]ith imaginable empirical projections, there may be sufficient consensus to ensure agreement on particular outcomes, even amidst significant differences in value and across party lines.”).

In short, the Medicaid expansion represents an instance in which federal agencies acting with and through the states have moved polarized politics on a major policy issue. Whether this will have a broader effect on the deep partisanship surrounding the ACA at the national level remains unclear. As Abbe Gluck has argued, however, state implementation serves an entrenchment function that makes national programs harder to repeal when there is a change in political control. 247 Gluck, Intrastatutory Federalism, supra note 174, at 568–72. If Medicaid’s past is any guide, a similar entrenchment may well occur with respect to Medicaid expansion under the ACA, 248 See Laura Katz Olson, The Politics of Medicaid 4–5, 12–13 (2010) (arguing vast array of stakeholders have assured Medicaid’s expansion and durability over program’s history); Frank J. Thompson, Medicaid Politics 203–32 (2012) (arguing Medicaid expanded between 1993 and 2010 despite polarization and attacks on welfare state and tracing relationship between program’s resilience and federalism). although election of a Republican President in 2016 might lead to significant transformations of the program.


Analyzing polarization’s impact on governance requires probing beyond the phenomena of congressional gridlock and presidential unilat­eralism that currently dominate popular and scholarly accounts. It requires incorporating a nuanced assessment of polarization’s interaction with federal administration and the diverse forces that affect agency action. In a polarized world marked by legislative gridlock, presidential control is a major element in administrative decisionmaking. But agencies can remain subject to notable counterpressures from Congress, the courts, internal agency forces—and, crucially, the states.

Examination of the ACA’s implementation demonstrates the importance of incorporating the states into polarization discussions. The Medicaid expansion in particular is a notable instance of how nego­tiations between federal agencies and the states have led to agree­ments that cross partisan divides. Perhaps this bipartisanship will be limited to the administrative or implementation sphere, and not yield greater bi­partisanship on the ACA in Congress. Or perhaps the Medicaid expansion will be sui generis—a unique context in which strong financial and political incentives overwhelmed partisan ideological opposition. But given agencies’ broad powers and the importance of federal–state programs for both national and state governments, the potential exists for similar instances of agency–state partisan realignment in other contexts. Indeed, a very similar dynamic appears to be occurring in the area of education. There, in the face of congressional failure to address problems with NCLB, the Obama Administration transformed the statute by granting states waivers from the Act’s accountability and performance requirements. 249 See Kurzweil, supra note 142, at 601–08 (tracing history of NCLB waivers). Some of the conditions DOE imposed in exchange—acceptance of teacher-assessment systems and common core standards—are becoming increasingly unpopular and hard for states to imple­ment. 250 See Wong, supra note 142, at 408–18 (documenting difficulties of complying with teacher-evaluation standards, even in states that satisfied most other requirements of NCLB); Klein, supra note 151 (noting “[t]eacher evaluation . . . has been the trickiest area of waiver implementation”). Republican opposition to the NCLB waivers as executive over­reach, combined with growing resistance across the political spectrum to the waivers’ terms, are spurring a bipartisan congressional effort to replace NCLB. 251 See Motoko Rich & Tamar Lewin, No Child Left Behind Law Faces Its Own Reckoning, N.Y. Times (Mar. 20, 2015), (on file with the Columbia Law Review) (detailing ongoing congressional efforts to replace NCLB waivers). This effort may prove unsuccessful; partisan divides on federal education policy remain strong, and the terms of several proposed measures are notably at odds with the administration’s preferences. 252 Id.; see also Emma Brown, Senate’s Effort to Rewrite NCLB Sparks Cautious Optimism, Wash. Post (Apr. 8, 2015), [] (describing reactions of President Obama and DOE Secretary Duncan to proposed bipartisan Senate bill). Still, NCLB may prove to be another context in which a federal agency pushes past polarization and stalemate by partnering with the states, albeit perhaps not achieving the policy outcome that the agency sought.

More broadly, both the Medicaid expansion and NCLB waivers demonstrate the critical roles that administrative agencies play in the world of polarized governance. It falls to agencies to develop policy in the face of political dysfunction, whether acting on the President’s behest, their own initiative, or somewhere in-between. And as Peter Strauss forecast over thirty years ago, the web of controls on agencies will adapt to this new political reality.