Can the states keep secrets from the federal government?

• Author: Mikos, Robert A.
• Position: IV. Restoring State Autonomy: A New Approach to Safeguarding States' Secrets through Conclusion, with footnotes, p. 154-178

IV. RESTORING STATE AUTONOMY: A NEW APPROACH TO SAFEGUARDING STATES' SECRETS

This Part contains the core prescriptive claims of the Article. I suggest that the conventional wisdom has misjudged the commandeering of states' secrets. The courts have made a false distinction between the services commandeered in Printz and the services commandeered by statutory reporting requirements, administrative subpoenas, and grand jury subpoenas. Providing information about federally regulated activity is tantamount to assisting in the administration or enforcement of federal law. Just as importantly, compelling states to provide such information incurs the same structural harms as compelling them to provide other services.

In a nutshell, I propose that courts treat demands for information as constitutionally prohibited commandeering. I make the case for applying the anti-commandeering rule and address the appropriateness of its application to the three forms of commandeering states' secrets. I then address some potential exceptions and objections to my proposal. Lastly, I briefly consider some political solutions as alternatives to my proposal.

1. What It Means to Enforce or Administer Law

   As discussed above, some lower federal courts have summarily dismissed challenges to demands for information on the ground that the Printz Court already approved of them. (244) In Printz, Justice Scalia suggested that federal laws that "require only the provision of information to the Federal Government[] do not involve the precise issue before us here, which is the forced participation of the States' executive in the actual administration of a federal program." (245) And in her concurrence, Justice O'Connor approved of "refrain[ins] from deciding whether other purely ministerial reporting requirements imposed by Congress on state and local authorities pursuant to its Commerce Clause powers are similarly invalid." (246)

   The problem is that the Supreme Court has never actually held that demands for information are constitutionally permissible. The Printz Court's cursory remarks do not instruct the lower courts to exempt federal demands for information from the constitutional prohibition against commandeering. At most, the remarks suggest that the Court may have thought such demands could (somehow) be distinguished from the commands issued under the Brady Act. As a result, the Printz Court may have thought it wise to leave the door open for the lower courts to consider the issue more thoroughly in subsequent cases. But the lower courts never took up that task. Instead, most of them have simply presumed that Printz conclusively decided the matter and have upheld federal demands for information on that basis alone.

   Outside of some exceptions discussed later, Printz invalidated federal commands that require the states to participate in the enforcement or administration of a federal regulatory program. The scope of the anti-commandeering rule thus hinges on what it means to "enforce" or "administer" federal law. Unfortunately, the Printz Court did not expressly elaborate upon those key terms. The lower courts have simply glossed over this important issue and presumed that providing information about regulated activity (somehow) does not amount to participating in the enforcement or administration of federal law.

   The lower courts' presumption appears to rest on a narrow conception of enforce as "to compel obedience." (247) A few courts have employed this particular conception when interpreting the term "enforce" in a handful of congressional statutes ranging from the CSA to the Employee Retirement Income Security Act (ERISA). (248) If accepted for purposes of the anti-commandeering rule, this conception would seemingly forbid Congress only from compelling state officials to compel their own citizens to do (or not do) something. For example, a congressional command to a state official to seize a medical marijuana dispensary and arrest and restrain its proprietor would clearly constitute a command to enforce federal law--i.e., the official would be obliged to compel the proprietor's obedience with the federal CSA.

   It is far from clear, however, that the Supreme Court had such a narrow conception in mind in the anti-commandeering cases. To begin, the Court used the term "enforce" interchangeably with other terms like "administer" (249) (as in, to administer federal law) and "execute" (250) (as in, the executive functions of a state), and these terms do not necessarily have the same restrictive definition as "enforce." In common usage, these terms simply mean to perform a task or duty--and surely, even running a simple search of a state database entails performing a task or duty.

   More importantly, the actual holding of Printz suggests a broader conception of the term "enforce" and of the reach of the anti-commandeering rule. Indeed, Printz easily could be read to condemn rather than to spare demands for information, notwithstanding the dicta noted above. The reason, as Dean Evan Caminker has surmised, is that "the primary duty imposed by the Brady Act itself is a 'reporting' requirement of sorts." (251) The Printz Court even acknowledged that the "central obligation" imposed by the Brady Act upon Chief Law Enforcement Officers (CLEOs) is to "'make a reasonable effort to ascertain within 5 business days whether receipt or possession [of a handgun] would be in violation of the law, including research in whatever State and local recordkeeping systems are available and in a national system designated by the Attorney General.'" (252) In an earlier portion of the opinion, the Court had similarly noted that CLEOs were required only "to provide information that belongs to the State" and "to conduct investigation in their official capacity, by examining databases and records." (253) Importantly, the Court noted that the Brady Act "does not require the CLEO to take any particular action if he determines that a pending transaction would be unlawful; he may notify the firearms dealer to that effect, but is not required to do so." (254) In other words, CLEOs were not required to compel anyone else to do (or not do) anything, e.g., to forego a proposed firearm transaction.

   It is hard to fathom what sort of principled distinction could be drawn between the actual duties imposed by the Brady Act and the duties imposed by congressional reporting requirements, administrative subpoenas, and grand jury subpoenas discussed herein. Consider, for example, the subpoena issued by the DEA requiring a Michigan state agency to search its medical marijuana registry and provide the DEA "copies of any and all documents, records, applications, payment method of any application for Medical Marijuana Patient Cards and Medical Marijuana Caregiver [C]ards and copies of front and back of any cards" pertaining to seven named individuals under federal investigation. (255) Like the background check provisions of the Brady Act, this subpoena required the state to search for and provide information it had gathered in its sovereign capacity to assist federal agents in detecting and sanctioning violations of federal law.

   Further undermining the case for a narrow application of the anti-commandeering rule, the Printz Court rejected the notion that information-reporting requirements could be distinguished from demands for other state services on the basis of historical practice. (256) To be sure, the federal government has demanded information from the states more commonly than it has demanded other services. But the federal government's growing penchant for commandeering states' secrets is no reason to absolve that practice from constitutional prohibition. (257) Most statutory reporting requirements are of distinctly modern origin. They have been adopted as part of The Asbestos School Hazard Abatement Act of 1984, The Health Care Quality Improvement Act of 1986, The Crime Control Act of 1990, The Hotel and Motel Fire Safety Act of 1990, and The Highway Safety Act of 1991, among others. (258) Indeed, in Printz, Justice Scalia downplayed the constitutional significance of such statutes, arguing,

   Even assuming they represent assertion of the very same congressional power challenged here, they are of such recent vintage that they are no more probative than the statute before us of a constitutional tradition that lends meaning to the text. Their persuasive force is far outweighed by almost two centuries of apparent congressional avoidance of the practice. (259) Likewise, statutes authorizing administrative agencies to subpoena records are now commonplace, but those statutes are also of relatively recent vintage. The first notable administrative subpoena statute appeared in 1887, (260) but until 1943, the administrative subpoena power was quite narrow, (261) and it is not even clear when it was first employed to obtain records from an unwilling state.

   The grand jury could pose a special exception. Grand juries have exercised broad investigative powers since well before the founding of the nation, (262) and the Fifth Amendment expressly mandates their use in most federal criminal proceedings. (263) It might be argued that the Fifth Amendment authorizes the grand jury to demand assistance--informational or otherwise--from the states in conducting its criminal investigations. I discuss such an exception below. For now, however, it is worth noting that this exception to the anti-commandeering rule would only apply to demands made by federal grand juries and would not apply to demands made by administrative agencies or Congress.

   In lieu of the lower courts' narrow conception of enforcement, I espouse a broader vision that includes the gathering and reporting of information, even if these tasks do not directly involve compelling a citizen to do (or not do) anything. This conception not only seems more consistent with the actual holding in Printz, but it also reflects a more robust...