My thanks to Hans Bader, Michael Toth, and Jonathan F. Mitchell for their thoughtful responses to my essay concerning state attorneys general (AGs) and contemporary American federalism. Each raises good points about the AGs’ various roles in the era of executive federalism that has rapidly expanded during the Obama years. As all three authors note, AGs can and have pushed back on federal power in numerous ways that can indeed provide a check on administrative discretion and regulatory expansion. Ultimately, however, I would emphasize again that AGs are at best unsteady allies of federalism.

First, I have a point of agreement with a theme contained in all three responses: AGs are not the main culprits in the demise of the traditional practice of federalism. The trend towards “cooperative” federalism bolstering regulatory power at all levels of government began in full force during the New Deal era and has continued as a central aspect of American policy ever since. Much of the social welfare state established during the 1960s and 70s built on this model, and the Obama Administration’s executive-centered federalism has expanded administrative power through bargaining and negotiations enabled by “cooperative” federalism. This system has been the product of many institutions on both the state and federal levels.

In fact, even when it comes to the AGs themselves, one can point to Congress as the real empowering force. It has been Congress that has consistently bolstered AGs’ power to enforce federal law and provided various grants and incentives for AGs to seize a new activist role in contemporary politics.[1] Furthermore, this has been very much a bipartisan endeavor. Liberals in Congress have liked how AGs act as pro-regulatory “consumer advocates,” while conservatives have favored expanding AG power as an incentive to fight crime as well as “fraud” in federal social programs.[2]

While I agree with the authors that AGs are not single-handedly ruining federalism, I have a considerably less sanguine view of AGs’ overall contribution to contemporary governance. One can certainly point to particular cases in which AGs have fought for federalism principles against an overweening administrative state, as all three authors do. But when set against the full range of AG activity, it becomes apparent that AGs have been far more likely to expand the national regulatory state.

Bader notes the largest and most striking example of this: the tobacco Master Settlement Agreement originally signed in 1998 and enforced by AGs in the years since. Bader agrees that multistate settlement “trampled on federalism,” but highlights other areas in which AGs have challenged federal overreach in the courts. I would argue, however, that Bader’s focus on AG activities in the Supreme Court obscures the bulk of the AGs’ multistate activities, which never even reach the courts and that more resemble the tobacco settlement. It is true that AGs have not entered a settlement approaching the sheer scope of the tobacco deal, which remains the largest civil settlement in American history. But they have conducted thousands of investigations of various industries in the years since that resemble a more patchwork – but no less concerning – attempt at regulating entire national industries.

The AGs’ longstanding investigations of the pharmaceutical industry provide the best example. Starting in the immediate aftermath of the tobacco deal, AGs used regulation-through-settlement strategies against drug firms to fight “corporate greed” and establish numerous new regulatory restrictions on industry pricing and marketing. Unlike the tobacco campaign, the result was not a single, high-profile settlement. Instead, the still-ongoing campaign has resulted in hundreds of settlements with individual firms over several years, but which collectively had a similar effect to the tobacco settlement: the transfer of billions of dollars from private firms to government and the imposition of new regulations that were never approved by Congress.

Many of these raise fundamental concerns similar to the tobacco settlements. For one, they represent similar attempts at extraterritorial regulation outside of the normal policymaking process. AGs have been quite explicit that many of their activities seek to establish nationally applicable corporate restrictions act in the face of congressional inaction to “protect consumers” or “fight fraud.” AG litigation against pharmaceutical industry pricing practices, for example, arose after the repeated failure to advance price control legislation in Congress. Various AG lawsuits against energy firms, some of which were based upon the Clean Air Act and others on common law theories, sought to regulate utilities following the failure of climate change legislation.[3] AGs have used similar regulation-through-settlement strategies against gun manufacturers, banks, insurance firms, and many other industries following alleged congressional “failures to act.”

The nature of global settlements makes them different from the other sorts of state regulations that, as Bader notes, frequently add to federal mandates. State statutes or regulations apply only within their jurisdictions – Wisconsin, for example, cannot enact a law regulating industries in Illinois. Yet AGs’ multistate settlements bypass this restriction by placing nationwide industries under a single set of rules. This is much like Congress might do, except these national regulatory regimes emerge from the states (sometimes even a minority of the states) rather than from any federal institution.

That the vast bulk of the AGs’ form of regulation appears in the form of settlements rather than court judgments further insulates them from constitutional protections. Because they are styled as “voluntary” agreements, most settlements are subject to no or only cursory judicial oversight. This allows settlements to contain regulatory provisions that would not likely pass muster if it came in the form of a traditional state or federal law. The various marketing restrictions in tobacco and pharmaceutical settlements raise significant commercial free speech issues, for example. An AG settlement with the manufacturer of Botox even includes provisions requiring the company to drop pending free speech lawsuits against the FDA as a condition of ending the AG investigation.[4]

Mitchell argues that there are “serious limits” on AGs’ use of these regulation-through-settlement strategies. The first is that AG’s lawsuits cannot begin unless a company has engaged in illegal conduct, so “companies that have done nothing wrong cannot be sued.” Yet one of the AGs’ primary goals has been to shift previously unproblematic corporate conduct and transform it into actionable “fraud” through the very process of their investigations. The tobacco settlement was predicated on new and legally untested theories that framed decades-long tobacco industry practices as conspiracies to defraud state Medicaid programs. Likewise, lengthy investigations by AGs transformed a formula that the government itself had been using for decades to reimburse drug companies – “Average Wholesale Prices” – into a supposedly fraudulent scheme on the part of greedy drug companies. The $25 billion national mortgage settlement with five banks in 2012 contained numerous regulatory provisions covering conduct not explicitly illegal under existing federal laws.

The problem, in part, is that the definition of “fraud” is so ambiguous, and state consumer protection statutes in particular so elastic, that the definitions of “illegal” and “fraudulent” are often whatever AGs like New York’s Eric Schneiderman say they are. Given that settlements are rarely subject to judicial review, these claims of “fraud” are seldom actually tested in court.

Mitchell identifies as a second limit the fact that settlements bind only the parties to the case, as opposed to an entire industry as a whole (as federal regulations might do). While true, I believe this to be less of a limit than Mitchell does. For one, AGs can use a variety of creative strategies to try and lock in as much of the industry as possible into a new regulatory regime. The tobacco settlement, as Mitchell notes, contained numerous complicated provisions pertaining to “Non-Participating Manufacturers” (NPMs). These provisions required settling states to enact model statutes forcing NPMs to pay into state escrow accounts in order to prevent them from undercutting the prices of participating manufacturers. While these statutes depend on the political will of the states to enact them, it is telling that all states have done so (otherwise, they would be giving up the settlement money). Indeed, much of the activity in the past few years has been strengthening state NPM model statutes to make it even more difficult for new entrants to the settlement-sanctioned cartelized industry.

The tobacco deal is admittedly quite unusual in this regard, largely because the high level of market concentration in the tobacco industry made it easier to apply the settlement’s regulatory regime across nearly all manufacturers. It is not surprising that similar attempts to regulate entire industries in single settlements (such as gun manufacturers and banks) have occurred in industries with high market concentration. Yet in other industries, a second strategy seeks to overcome the limitation that settlements apply only to the parties to the case. AGs frequently aim at broader reform in their settlement strategies. This includes creating business precedents through settlements with leading firms, which in turn seek to convince other firms in the industry to voluntarily change their behavior or else risk being sued. (This has been particularly effective in the pharmaceutical context). Another part of the strategy is to try to convince Congress that “something must be done” to address concerns AGs have raised about corporate conduct. Ironically, the very fact that multistate settlements are “hardly an effective substitute for regulations enacted by legislative bodies,” as Mitchell rightly notes, makes it more likely that legislators will step in and regulate after the AGs have acted. Industries themselves will often ask Congress for a single regulatory solution to fix the inevitable problems that arise in the AGs’ patchwork, yet nationalized, method of regulation.[5]

To this point, I have focused on one aspect of the AGs’ contemporary, nationalized activism (though I maintain that it is the largest, most important, and probably the most problematic element of AG activity). But what of the many examples the authors provide of AGs seemingly defending federalism principles, including challenges to Obama Administration policy? Doesn’t this indicate that AGs are indeed friends of federalism?

This is where a fundamental question arises: is the goal to protect “federalism” and “the rights of the states” or to limit the extent of liberal government regulation? This is key because these two elements are far from one and the same, as AG activism amply demonstrates. If the main goal is to challenge regulatory expansions that favor the political left, then AGs can indeed be helpful allies. The examples Toth raises are especially illustrative here: AG challenges to federal regulations concerning transgender restrooms, attorney-client communications in labor disputes, and medical procedures performed by physicians. (Toth’s examples are particularly good, because they illustrate some of the often under-the-radar administrative policymaking that characterizes so much of contemporary governance). Bader and Mitchell likewise note AG challenges to liberal Obama-era policies.

Yet challenging liberal ideological prerogatives is not the same thing as defending the principle of “federalism.” Indeed, the increasingly partisan battles amongst Democratic and Republican coalitions of AGs feature both sides claiming that they are defending federalism principles. Republican AGs have good claims to being the allies of federalism and state power when pushing back on the sorts of federal initiatives Toth raises. Yet who are the real friends of federalism in gun rights cases determining whether states and localities can institute firearm control regulations within their own jurisdictions? Was it the coalition of (mostly, though not exclusively Republican) AGs who argued in D.C. v. Heller and McDonald v. Chicago that the Second Amendment contained an individual right serving to limit state police powers? For several of these AGs, this argument implicitly attacked their own state’s gun laws – the conservative equivalent of liberal AGs’ refusal to defend their own state laws that Bader characterizes as a great risk to federalism.

The recent ExxonMobil case Toth raises is another good example. This case involves several AGs (all Democrats) who announced an “AGs United for Clean Power” initiative this past March. The idea behind this initiative follows the AGs’ general regulation-through-litigation strategies, in this case investigating “whether fossil fuel companies misled investors and the public on the impact of climate change of their businesses.”[6] This is fairly unremarkable (though still problematic), given that AGs have initiated these sorts of investigations quite frequently in recent years. What is remarkable is the response of the opposing coalition of AGs, all Republican, that Toth praises. These AGs have not only openly denounced this effort as “chilling free speech,” but sided with the private utilities in court against their fellow AGs’ investigation. This has operated in tandem with House Republicans’ investigation of the AGs leading the “AGs United for Clean Power” effort – a remarkable example of federal interference with a state investigation. Who is defending “federalism” here? The AGs running their state-based investigations, or the opposing state and federal authorities interfering with those investigations?

One might respond that the Republican AGs are doing a service to federalism by opposing the activist liberal AGs in this case, because they are pushing back on the sort of problematic activity I discussed earlier in this piece. But this response would be more convincing if Republican AGs were not every bit as complicit in the regulation-through-settlement strategies used to expand regulation in other contexts. Even though other areas of AG activities have been highly polarized, regulatory settlements attract bipartisan support from Republican and Democratic AGs alike. The tobacco settlement consisted of forty-six AGs (with the other four settling separately). The $25 billion national mortgage settlement attracted forty-nine AGs working alongside federal agencies in an example of executive federalism in action. Pharmaceutical settlements regularly attract all fifty states and are often led by Republican AGs.

It would also be more convincing if it was just Democratic AGs using their investigatory powers in allegedly abusive ways. A few years ago, for example, former Republican AG Ken Cuccinelli issued a civil investigative demand to University of Virginia climate scientist Michael Mann, claiming that Mann’s “use of manipulated data to apply for taxpayer-funded research grants in Virginia is potentially fraud.”[7] Predictably, liberals denounced this as a smear campaign aimed at “chilling free speech” and urged an end to the state’s investigation. Switch the sides and you have the battle lines in the ExxonMobil suit. Needless to say, any consistency here has little to do with “the state’s right to conduct investigations of fraud” but rather one’s ideological position on climate change policy.

My larger point is if one is looking for a principled commitment to “federalism,” the AGs are going to be unsteady allies given that they privilege their partisan commitments over any principled commitment to traditional constitutional governance. Sometimes this means AG coalitions of both parties calling upon the federal government to prevent their opponents from instituting policies with which they disagree, even if that means interfering with state police powers.[8] When the AGs do work in a bipartisan fashion, it often serves to either expand regulation through settlements or to expand their own powers in front of a sympathetic Supreme Court that often (wrongly) views its actions as bolstering “federalism.”[9] Again, I agree that AGs are far from alone in ruining federalism. But AGs of both parties have played an increasingly significant role in its demise.

[1] Provisions empowering AG litigation in federal law became increasingly common by the early 1990s, particularly in the area of consumer protection. According to my count, Congress enacted eleven new federal provisions specifically authorizing AGs to enforce provisions of federal law in the 1990s and an additional sixteen in the 2000s. Further, federal grants to AGs have also expanded over time. To take just one example, the grants provided as part of the Medicare-Medicaid Anti-Fraud and Abuse Amendments of 1977 enabled AGs to create specialized prosecution teams able to conduct wide-ranging investigations of health care providers and pharmaceutical firms. Today, these federal grants now total over $188 million, enabling these prosecution teams to employ over 1,800 staff members collectively. See Medicaid Fraud Control Units FY 2015 Annual Report, available at https://oig.hhs.gov/oei/reports/oei-07-16-00050.pdf.

[2] Sen. Chuck Grassley (R-IA), for example, has been a major proponent of strengthening the False Claims Act, which among other things incentivizes AG lawsuits against drug companies and other industries. See, e.g., Chuck Grassley, “Grassley Floor Statement on the False Claims Act,” available at http://www.grassley.senate.gov/news/news-releases/grassley-floor-statement-false-claims-act.

[3] Various AG efforts to disrupt the Bush Administration’s New Source Review regulations are a good example of this activity. See, e.g., Darren Samuelsohn, “States, Enviros Seek End Run Around Bush NSR Strategy with New Lawsuits,” Greenwire, July 29, 2004.

[4] Natasha Singer, “Maker of Botox Settles Inquiry on Off-Label Use,” New York Times, September 2, 2010, A1.

[5] This was the strategy AGs employed against energy firms, for example. Throughout the 2000s, AGs sued dozens of utilities claiming that they were contributing to climate change and other alleged environmental harms. Part of the AGs’ hope was that it would prompt the companies to seek a single regulatory solution through Congress rather than face lengthy litigation from numerous states.

[6] Press Release, NY Attorney General’s Office, March 29, 2016, available at http://www.ag.ny.gov/press-release/ag-schneiderman-former-vice-president-al-gore-and-coalition-attorneys-general-across.

[7] Attorney General of Virginia, “Statement Regarding the University of Virginia’s Challenge of the Attorney General’s Civil Investigative Demand,” May 28, 2010, http://www.oag.state.va.us/Media%20and%20News%20Releases/News_Releases/Cuccinelli/52810_UVA_Statement.html.

[8] I might mention another brief example here: the Oklahoma and Nebraska AGs’ various efforts to prevent Colorado from enforcing its own marijuana statute. John Ingold, “Nebraska and Oklahoma’s New Tactic to Overturn Marijuana Legalization,” The Denver Post, April 25, 2016, available at http://www.denverpost.com/2016/04/25/nebraska-and-oklahomas-new-tactic-to-overturn-marijuana-legalization/.

[9] The majority in Massachusetts v. EPA characterized its opinion as advancing federalism values, for example, particularly through its announcement of “state solicitude” in the Court’s standing analysis. Other recent cases have served to expand AG investigative power based upon supposed federalism values. See, e.g. Cuomo v. Clearning House Association, 557 U.S. 519 (2009); Hood v. AU Optronics, 571 U.S. 310 (2014) .