Ohio Republican Party v. Brunner

KAREN NELSON MOORE, Circuit Judge,

dissenting from hearing en banc and from reinstating the district court’s Temporary Restraining Order.

I dissent from the decision of this court to grant the motion for initial hearing en banc and the “Renewed Petition for Immediate En Banc Review,” thereby reinstating the district court’s Temporary Restraining Order (“TRO”) that imposes burdens on the Ohio Secretary of State (“Secretary”) that are contrary to federal and state law. This case is wholly inappropriate for initial or immediate hearing en banc. Not only have the criteria for initial hearing or rehearing en banc not been satisfied, but also the en banc court is particularly ill-suited to consider in the first instance swiftly developing election-law issues in the compressed time period now available. The majority’s action is judicial activism in the extreme.

I. OPPOSITION TO GRANT OF EN BANC REVIEW

The panel assigned to this case pursuant to Sixth Circuit rules has considered carefully the materials filed on behalf of both sides in this dispute and has issued an order staying the temporary restraining order improperly entered by the district court. The motion for hearing en banc filed by the Ohio Republican Party and Larry Wolpert (collectively “ORP”) at this time is nothing more than a blatant attempt of ORP to overturn the duly authorized panel’s decision to stay the district court’s order that required the Secretary of State to likely violate both the Help America Vote Act (“HAVA”) and the National Voter Registration Act (“NVRA”). By granting this motion and reinstating the district court’s TRO, the majority of this court acts contrary to our own precedent, fails to follow the Supreme Court’s precedent, and creates a new circuit split.

The Secretary should not be required by various federal judges to violate those federal statutes, nor should she be required to reconfigure and reprogram the state’s computers and practices this close to an election. Because the granting of en banc *725hearing and the simultaneous reinstatement of the district court’s temporary restraining order will throw the election process in Ohio into total chaos, I dissent.

On October 10, 2008, the Sixth Circuit panel stayed the district court’s temporary restraining order. In our opinion, we explained our reasoning and concluded as follows:

It is clear that the district court’s four specific orders insert the federal court into the delicate balance struck by HAVA. We have expressed our concern that under established law ORP does not have a private right of action under 42 U.S.C. § 15488(a)(5)(B)(i). Assuming ORP has the right to bring such an action, we believe that the Secretary is likely to succeed on the merits of the issue of the proper interpretation of HAVA; HAVA does not mandate that the Secretary undertake the particularized matching required by the district court’s TRO. Further, the irreparable harm caused by the district court’s TRO is significant. With less than a month until the election, and less than two weeks until the beginning of counting absentee ballots, the Secretary cannot be required to undertake the extensive reprogramming and other changes to the election mechanics without complete disruption of the electoral process in Ohio. The irreparable harm to the voting public caused by the district court’s order is equally clear. Finally, the intrusion into the state’s processes by the federal courts with the ensuing confusion regarding the applicable process weighs heavily against the district court’s order.
As the Supreme Court wrote in Purcell, “Given the imminence of the election and the inadequate time to resolve the factual disputes, our action today shall of necessity allow the election to proceed without an injunction” altering the state’s established practice. Purcell [v. Gonzalez], 549 U.S. [1,] 8, [127 S.Ct. 5, 166 L.Ed.2d 1] [2006], We hereby stay the district court’s TRO....

Sixth Circuit panel opinion of October 10, 2009 at 11-12.

ORP asserts that hearing this case initially en banc and bypassing the duly assigned panel (or rehearing en banc) is appropriate because of the importance of the issues presented and the press of time. The time pressures in this case are entirely caused by ORP’s last-minute challenges to the procedures initiated by the Secretary of State’s office before the current Secretary began her position in 2007. While the Secretary’s procedures have been known and in effect for a considerable time, ORP waited to file its suit in the district court until September 26, 2008, and did not file its motion for this temporary restraining order until October 5, 2008. If ORP had truly wanted to have review of the methods and procedures used by the Secretary, it should have brought its action much earlier. More importantly, the NVRA specifically requires that “any program the purpose of which is to systematically remove the names of ineligible voters from the official lists of eligible voters” must be completed at least 90 days prior to a federal election. 42 U.S.C. § 1973gg-6(c)(2)(A). HAVA provides for removing voters in accordance with the NVRA, and the latter does not permit registered voters to be removed based on computer matching. 42 U.S.C. § 15483(a)(2)(A)©; 42 U.S.C. § 1973gg-6.

Thus ORP has brought its action too late to obtain its requested relief; ORP’s claimed press of time is both self-created and self-defeating. The press of time is not a valid basis for obtaining hearing en banc in this case.

The other reason asserted by ORP for hearing en banc is that the case “is an extraordinary case, involving the integrity *726of the election for President of the United States ... in a state vital to both candidates’ chances at victory.” The Federal Rules of Appellate Procedure allow hearing or rehearing en banc where “the proceeding involves one or more questions of exceptional importance, each of which must be concisely stated.” Fed. R.App. P. 35(b)(1)(B) (emphasis added). Yet ORP completely fails to comply with the requirements of this Rule: it has not specified in its motion for hearing en banc ANY specific question of exceptional importance. ORP’s utter failure to comply with the Federal Rules of Appellate Procedure suggests that its effort to obtain hearing en banc concerning this temporary restraining order is based on ORP’s dissatisfaction with the panel’s composition or with the merits of the decision of the duly assigned Sixth Circuit panel to stay the previous temporary restraining order issued by the district judge in this case. ORP is simply trying to get a different result by having a different forum within the Sixth Circuit.

In summary, a hearing en banc in this case is wholly unwarranted and entirely unjustified. Granting a hearing en bane on this flawed motion and in light of the circumstances of this litigation is unsound intellectually and without any valid justification. Any lawyer reading the plaintiffs/appellees’ motion for initial hearing en banc or its renewed motion can see that substantively it is baseless and procedurally it violates the Federal Rules of Appellate Procedure.1 I dissent from the majority’s rash and meritless decision to grant en banc review and to reinstate the district court’s TRO.

II. ORP HAS NO PRIVATE RIGHT OF ACTION

In addition to disagreeing with this court’s decision to grant en banc review, I must dissent from its decision to reinstate the district court’s temporary restraining order. On its face, HAVA does not create a private right of action. Supreme Court precedent makes it absolutely clear that the provisions of HAVA at issue here, 42 U.S.C. §§ 15483(a)(1)(A) & (a)(5)(B)®, do not create federal rights enforceable by private plaintiffs under 42 U.S.C. § 1983. For a statute to be enforced via § 1983, Congress must have unambiguously intended to create an individual right. Gon-zaga Univ. v. Doe, 536 U.S. 273, 283-85, 122 S.Ct. 2268, 153 L.Ed.2d 309 (2002). However, these provisions of HAVA regulate the conduct of officials involved in the *727voting process; they do not create rights enforceable by individual voters. In sharp contrast to the statutory provisions that the Supreme Court and this circuit have recognized as establishing privately enforceable rights, the HAVA provisions at issue in this case contain absolutely no rights-creating language. It is not surprising, therefore, that no court has interpreted these provisions as conferring a right enforceable by private plaintiffs under § 1983. Instead HAVA provides its own enforcement mechanisms, specifying that the U.S. Attorney General may bring suit to enforce HAVA. 42 U.S.C. § 15511. The U.S. Attorney General has not brought suit.2

A statute is enforceable via a private right of action under § 1983 only if contains “an unambiguously conferred right.” Gonzaga, 536 U.S. at 283, 122 S.Ct. 2268. “[I]t is rights, not the broader or vaguer ‘benefits’ or ‘interests,’ that may be enforced under the authority of that section.” Id. If a “statute by its terms grants no private rights to any identifiable class,” then Congress did not intend to create a private right of action. Id. at 283-84, 122 S.Ct. 2268 (internal quotation marks omitted). This inquiry begins with the text of the statute. “For a statute to create such private rights, its text must be ‘phrased in terms of the persons benefited.’ ” Id. at 284, 122 S.Ct. 2268 (quoting Cannon v. Univ. of Chicago, 441 U.S. 677, 692 n. 13, 99 S.Ct. 1946, 60 L.Ed.2d 560 (1979)). By contrast, statutes that “speak only in terms of institutional policy and practice” or that “have an ‘aggregate’ focus” do not confer a right that may be enforced via § 1983. Id. at 288, 122 S.Ct. 2268.3 Similarly, “[statutes that focus on the person regulated rather than the individuals protected” do not confer enforceable rights. Alexander v. Sandoval, 532 U.S. 275, 289, 121 S.Ct. 1511, 149 L.Ed.2d 517 (2001) (internal quotation marks omitted).

The provisions of HAVA at issue here contain absolutely no rights-creating language. Instead, they impose an obligation on state officials to establish a voter-registration system meeting certain criteria. Section 15483(a)(1)(A) requires each state to “implement ... a ... computerized statewide voter registration list ... that contains the name and registration information of every legally registered voter in the State and assigns a unique identifier to each legally registered voter in the State....” 42 U.S.C. § 15483(a)(1)(A). Section 15483(a)(5)(B)® further requires that “[t]he chief State election official and the official responsible for the State motor vehicle authority of a State shall enter into an agreement to match information in the database of the statewide voter registration system with information in the database of the motor vehicle authority to the extent required to enable each such official to verify the accuracy of the information provided on applications for voter registration.” 42 U.S.C. § 15483(a)(5)(B)®. These provisions focus on the administration of federal elections and the duties of state officials to establish institutional mechanisms meeting certain criteria. *728Nothing in these provisions contains the sort of rights-focused language that the Supreme Court has required to establish a privately enforceable right.

The paradigmatic statutes containing the requisite rights-creating language are Title VI of the 1964 Civil Rights Act, 42 U.S.C. § 2000d, and Title IX of the Education Amendments of 1972, 20 U.S.C. § 1681(a). Those provisions contain similar explicit rights-creating language. 42 U.S.C. § 2000d (“No person in the United States shall, on the ground of race, color, or national origin, be excluded from participation in, be denied the benefits of, or be subjected to discrimination under any program or activity receiving Federal financial assistance.”); 20 U.S.C. § 1681(a) (“No person in the United States shall, on the basis of sex, be excluded from participation in, be denied the benefits of, or be subjected to discrimination under any education program or activity receiving Federal financial assistance.... ”). Because each of these provisions has an “unmistakable focus on the benefited class,” the Supreme Court has recognized that each creates privately enforceable rights. Gonzaga, 536 U.S. at 284, 122 S.Ct. 2268 (internal quotation marks omitted).

In Gonzaga, by contrast, the Supreme Court found that a provision of the Family Educational Rights and Privacy Act of 1974 (“FERPA”) prohibiting federal funding of educational institutions with a policy or practice of releasing records to unauthorized individuals did not create a privately enforceable right. That provision of FERPA provided that “[n]o funds shall be made available under any applicable program to any educational agency or institution which has a policy or practice of permitting the release of educational records ... of students without the written consent of their parents to any individual, agency, or organization.” Gonzaga, 536 U.S. at 279, 122 S.Ct. 2268 (quoting 20 U.S.C. § 1232(g)(b)(l)). Distinguishing this FERPA provision from the rights-creating language of Title VI and Title IX, the Court noted that it spoke only to government officials rather than to individuals potentially benefited under the statute. Id. at 287-88, 122 S.Ct. 2268. The Court concluded that the FERPA provision “entirely lack[ed] the sort of ‘rights-creating’ language critical to showing the requisite congressional intent to create new rights.” Id. at 287, 122 S.Ct. 2268.

Unlike the individually focused language of Title VI and Title IX (“No person ... shall ... be subjected to discrimination .... ”), and like the FERPA provision at issue in Gonzaga, the HAVA provisions in this case speak only to government officials. HAVA directs that state election officials “implement ... a computerized statewide voter registration list,” 42 U.S.C. § 15483(a)(1)(A), and that they “enter into-an agreement to match information in the database of the statewide voter registration system with information in the database of the motor vehicle authority to the extent required to enable each such official to verify the accuracy of the information provided on applications for voter registration,” 42 U.S.C. § 15483(a)(5)(B)(i). In other words, these provisions require state officials to maintain a centralized statewide computerized voter registration list and to maintain that list according to certain standards. Nothing in these provisions refers to the “rights” of “individuals” or “voters.” Thus, unlike Title VI and Title IX and like the FERPA provision at issue in Gonzaga, these HAVA provisions focus only on government officials, who are directed to take certain actions, rather than on individuals who may benefit from the provisions. Consequently, like the FER-PA provision at issue in Gonzaga, these HAVA provisions “lack the sort of ‘rights-creating’ language critical to showing the requisite congressional intent to create *729new rights.” Gonzaga, 536 U.S. at 287, 122 S.Ct. 2268.

The HAVA provisions at issue in this case also differ dramatically from the rights-creating language contained in HAVA § 302(a)(2), which we have held creates a private right enforceable via § 1983. See Sandusky County Democratic Party v. Blackwell, 387 F.3d 565, 572-73 (6th Cir.2004). The provision at issue in Sandusky, HAVA § 302(a)(2), provides that upon making the required affirmation, an “individual shall be permitted to cast a provisional ballot.” 42 U.S.C. § 15482(a)(2) (emphasis added). In San-dusky, we concluded that “[t]his language mirrors the rights-creating language of Title VI of the Civil Rights Act of 1964 and Title IX of the Education Amendments of 1972.” 387 F.3d at 572. We further noted that “this language markedly differs from the statutory language found by the Supreme Court in Gonzaga to be insufficiently focused on the benefited class to create an individually enforceable right.” Id. at 573. We also stated that § 302 of HAVA “refers explicitly to the ‘right of an individual to cast a provisional ballot,’ ... and requires states to post information at polling places about this right along with ‘instructions on how to contact the appropriate officials if these rights are alleged to have been violated.’ ” Id. at 573 (emphasis in original) (citing 42 U.S.C. § 15482(b)(2)(E)). By contrast, the HAVA provisions at issue in this case do not so much as mention “individuals” or “rights.” Instead, these provisions are directed solely at regulating the conduct of state election officials.

Finally, nothing in HAVA’s legislative history indicates that Congress intended to create a private right of action to enforce the provisions at issue in this case, 42 U.S.C. §§ 15483(a)(1)(A) & (a)(5)(B)®. If anything, the legislative history weighs against the recognition of privately enforceable rights under HAVA, because Congress apparently considered, and rejected, the creation of an express private right of action under HAVA. Senator Christopher Dodd, a sponsor of HAVA and principal Senate author of the conference report, stated that “[wjhile [he] would have preferred that we extend [a] private right of action ..., the House simply would not entertain such an enforcement provision[ ]. Nor would they accept federal judicial review of any adverse decision by a State administrative body.” 148 Cong. Rec. S10505 (daily ed. Oct. 16, 2002) (statement of Sen. Dodd).

Despite the fact that HAVA does not create a private right of action, it imposes binding duties on the Secretary of State which are enforceable through alternative means. In states receiving certain federal funding, individuals may challenge actions by the election officer through an administrative complaint system established by the state. See 42 U.S.C. § 15512; Sandusky, 387 F.3d at 573 (acknowledging this administrative remedy). Regardless of funding, HAVA permits the U.S. Attorney General to bring a civil action against any state to enforce HAVA’s requirements. See id. § 15511. Given these mechanisms, it is absurd to assert that HAVA does not impose binding, enforceable duties on the Secretary; Congress has simply determined that private lawsuits are not the best method of enforcing HAVA. Although we have previously held that these enforcement provisions do not foreclose private suits to enforce the right of an individual to a provisional ballot, see Sandusky, 387 F.3d at 573, these provisions suggest that Congress did not intend for private suits to be the means of enforcing the HAVA provisions now at issue that are directed toward duties of state officials and are not creating individual rights.

*730Therefore, pursuant to well-established Supreme Court and Sixth Circuit precedent, ORP has no explicit or implied private right of action to bring this lawsuit or to seek the relief granted by the district court and the en banc majority. Therefore, I would dismiss this action because plaintiffs have failed to state a claim on which relief can be granted. The majority’s contrary finding that there is an implied private right of action here violates Supreme Court precedent such as Gonza-ga, 536 U.S. at 283-85, 122 S.Ct. 2268, and violates Sixth Circuit precedent that members of the majority have previously joined.4 Such judicial intervention in conflict with the intent of Congress in writing the Help America Vote Act is inexplicable and improper. I dissent from the majority’s decision to permit ORP to proceed with its claims.5

III. THE MOTION TO VACATE THE TRO SHOULD BE GRANTED

Even assuming that a private right of action exists under HAVA, I believe that Northeast Ohio Coalition for Homeless v. Blackwell, 467 F.3d 999, 1009 (6th Cir.2006), demands that this court issue a stay and vacate the district court’s temporary restraining order.

A. Success on the Merits

The Help America Vote Act, known as HAVA, requires each state to:

implement, in a uniform and nondiscriminatory manner, a single, uniform, official, centralized, interactive computerized statewide voter registration list defined, maintained, and administered at the State level that contains the name and registration information of every legally registered voter in the State and assigns a unique identifier to each legally registered voter in the State (in this subsection referred to as the “computerized list”)[.]

42 U.S.C. § 15483(a)(1)(A). Further:

The chief State election official and the official responsible for the State motor vehicle authority of a State shall enter into an agreement to match information in the database of the statewide voter registration system with information in the database of the motor vehicle authority to the extent required to enable each such official to verify the accuracy of the information provided on applications for voter registration.

42 U.S.C. § 15483(a)(5)(B)®.

Under the plain language of the statute, it is clear that all that this section requires *731is that a state: (1) have an agreement between the election official and the motor vehicle authority official to share and compare information in the statewide voter registration system with information in the motor vehicle authority database; and (2) give the state election official and the motor vehicle official the ability to check voter registration information in the system. There is absolutely no language that imposes an affirmative duty on any state official to enable a purge of voters by either scouring the database to locate mismatches or by providing a system that would allow a person to make such a search. “It is clear from the language of the statute and by looking at legislative history that HAVA’s matching requirement was intended as an administrative safeguard for ‘storing and managing the official list of registered voters,’ and not as a restriction on voter eligibility.” Washington Ass’n of Churches v. Reed, 492 F.Supp.2d 1264, 1268 (W.D.Wash.2006). In fact, the two federal courts to consider the provision in question here, Reed and Florida NAACP v. Browning, 522 F.3d 1153 (11th Cir.2008), both rejected the argument asserted by ORP that HAVA requires matching to be used to verify eligible voters. Reed, 492 F.Supp.2d at 1268; Browning, 522 F.3d at 1172. The Reed court stressed that the matching system is simply one way that a new voter who registered by mail could be verified. Reed, 492 F.Supp.2d at 1268. Reed went on to explain that a system that made matching the only way to verify voter identity would violate HAVA. Id. at 1269 (“[Requiring a match prior to registration directly conflicts with 42 U.S.C. 15473(b).”). In Browning, the Eleventh Circuit acknowledged that, while HAVA would allow a state to require that an individual have a match to be eligible to vote, HAVA did not compel such a requirement. Browning, 522 F.3d at 1172.

Though Browning and Reed disagree as to whether a state can choose to base voter eligibility solely on matching, they agree on the issue before this court: HAVA does not compel states to use a certain type of match system to verify voters. All that HAVA requires is that election officials have some method by which to verify registrations. The language of HAVA does not mandate what system the states must use to match or verify votes, leaving such decisions to the discretion of the states. The majority of this circuit thus conflicts with the Eleventh Circuit and creates a circuit split that should be resolved by the Supreme Court.

It boggles the mind that the majority can distort this simple language to force a state to implement a specific type of match system. This conclusion flies in the face of HAVA’s own directive that the State gets to choose how to implement HAVA. 42 U.S.C. § 15485 (“The specific choices on the methods of complying with the requirements of this subchapter shall be left to the discretion of the State.” (emphasis added)). I cannot comprehend why the majority is going to such lengths to force this strained interpretation into the case law, and I dissent from the majority’s violation of the plain statutory language.

Moreover, the majority’s view would require the Secretary to violate the National Voter Registration Act (“NVRA”) and HAVA. HAVA specifically provides for the removal of voters “in accordance with the provision of the National Voter Registration Act of 1993.” 42 U.S.C. § 15483(a)(2)(A)®. NVRA, however, does not permit registered voters to be removed based on computer matching. See 42 U.S.C. § 1973gg-6. Indeed, the NVRA requires that “any program the purpose of which is to systematically remove the names of ineligible voters from the official lists of eligible voters” must be completed *732at least 90 days prior to a federal election. § 1973gg-6(c)(2)(A). The stated purpose of the TRO is to allow for an “effective way to access and review the mismatches” located in the current database, because, allegedly, “some mismatches will reflect voter fraud.” Dist. Ct. Ord. at 11. Thus, the TRO demands a “program the purpose of which is to systematically remove the names of ineligible voters from the official lists of eligible voters.” § 1973gg-6(c)(2)(A). Therefore, the TRO, now sanctioned by the majority’s forced interpretation of HAVA, mandates that the Secretary violate federal law. Congress cannot have intended HAVA to be read in such a way.

Applying the correct interpretation of this statute to the facts in this case, I believe Ohio is likely in compliance with HAVA. Currently, the Secretary has a database which matches the information a voter provided when registering to vote with the information contained in the Ohio Bureau of Motor Vehicles database. If there is a mismatch, the database notes the mismatch. This record of mismatches in the database is accessible by both the Secretary and the individual boards of election in Ohio. The individual boards of election can query a voter’s match status in real time in order to verify registration.6 This is all HAVA requires. It is true that the statewide voter database does not provide for a list of these mismatches, but HAVA does not require that level of user-friendliness. To be sure, it might be nice if the system printed out a list of individuals within, for example, a particular precinct, that did not match, but I refuse to manipulate the clear language of Congress in HAVA to require Ohio to comply with a particular interested private party’s view of what type of match system would be best. For all these reasons, I conclude that the likelihood that the Secretary will succeed on the merits is great.

B. Irreparable Harm

The district court found that there would be no harm to others, including the Secretary, if it issued the TRO that ORP sought. This determination is wholly unsubstantiated and contrary to the realities of the TRO issued. In a week, the database that the district court has ordered to be reprogrammed “will be used to generate the Election Day poll books.” Secretary Br. at 19. The district court made its reprogramming order without hearing any evidence pertaining to how such a reprogramming would affect the existing election system and, more importantly, how long it would take to test the reprogrammed system, retrain personnel, and ensure that the new reprogramming did not cause problems. The Supreme Court has exhorted courts to exercise “proper judicial restraint” before making “precipitate changes” to election procedures and policies when “an impending election is imminent and a State’s election machinery is already in progress.” Reynolds v. Sims, 377 U.S. 533, 585-86, 84 S.Ct. 1362, 12 L.Ed.2d 506 (1964); see also Purcell v. Gonzalez, 549 U.S. 1, 7-8, 127 S.Ct. 5, 166 L.Ed.2d 1 (2006). Election procedures are matters of state law, and federal courts should hesitate to interfere, particularly *733when operating on limited information in the weeks preceding an election. See Purcell, 549 U.S. at 7-8, 127 S.Ct. 5.

The Secretary must handle all problems related to the reprogramming that is required under the district court’s novel order embraced by the majority of the en banc court. The most alarming possible side-effect of reprogramming would be that a programming glitch could cause validly registered voters to be inadvertently purged from poll books. Such a glitch would leave the Secretary with a database that can no longer be trusted for use in generating the poll books required for the election, causing “delays and inaccuracies in the creation of the poll books.” Secretary Br. at 19. The short time-frame available to comply with the TRO amplifies the risk of programming errors, one of numerous significant problems presented by the requirements imposed by the district court on short notice.

Additionally, even if the court-ordered reprogramming works perfectly, the Secretary will be faced with a large number of mismatches to contend with at the last minute. See Brennan Ctr. Amicus Br. at 12-13; Ohio Democratic Party (“ODP”) Amicus Br. at 23-24. Because of the time limitations, voters whose information does not match may not be aware that there is any question about their registration and may not have or be able to obtain the documents necessary to further verify their registration. It is unlikely that the state can properly investigate all of the mismatches created by the TRO, and as a result, properly registered voters will likely be forced to cast provisional ballots, will believe that they cannot vote, or will be turned away at the polling places. See Purcell, 549 U.S. at 7, 127 S.Ct. 5 (demanding “careful consideration” of any legal challenge that involves “the possibility that qualified voters might be turned away from the polls”). It is worth noting that the procedures each county uses to deal with these mismatches may not be uniform and could result in disproportionate disenfranchisement. See Bush v. Gore, 531 U.S. 98, 110, 121 S.Ct. 525, 148 L.Ed.2d 388 (2000) (per curiam) (noting importance of statewide standards for resolving questions of voting procedure). This confusion over what HAVA requires in terms of matching would affect not only Ohio, but every state in the country. ODP Amicus Br. at 15-16.

Moreover, at a time when the Secretary is already busy with routine election procedures, the TRO will require the Secretary to expend significant scarce resources on the reprogramming and matching effort. This diversion of resources could leave the state unable to respond to routine election issues such as broken voting machines and lines at polling places. If significant energies are devoted to matching issues, the orderly administration of the election will suffer. For all these reasons, I conclude that the Secretary would suffer irreparable injury if we do not grant a stay and vacate the district court’s TRO.

C. Substantial Harm to Others and Public Interest

Given the fact that hurriedly reprogramming the database could lead to the purging of validly registered voters, a stay of the district court’s intrusive TRO is necessary to protect the voters of Ohio. Though I agree with the district court’s conclusion that “safeguarding the legitimacy of the election of the President of the United States,” Dist. Ct. Ord. at 15, is of great public importance, the TRO is not required to protect that interest. It is striking that both the district court and the majority of this court have determined that the TRO is the proper remedy without any factfind-ing. The district court stressed that the public, as well as ORP, would be injured *734by voter fraud if a TRO were not granted. Dist. Ct. Ord. at 12-14. This is the kind of inquiry that demands extensive factfind-ing. See Purcell, 549 U.S. at 7-8, 127 S.Ct. 5. However, rather than undertake such factfinding, the district court, citing two newspaper articles, merely assumed that there will be widespread voter fraud absent the issuing of a TRO. Numerous other newspaper articles quote election officials who distinguish voter-registration fraud from actual voter fraud, and who indicate that actual improper voter registrations are already being detected. See, e.g., Joe Guillen, Multiple Registrants Tell Cuyahoga County Elections Board ACORN Workers Begged for Signatures, Cleveland Plain Dealer at A-l (Oct. 14, 2008). Given the shaky ground on which the district court’s suggestion of voter-fraud rests, I believe that this unsubstantiated fear does not warrant the district court’s intrusion into the established state practice.

Additionally, available evidence indicates that this intrusive TRO will cause more harm than it seeks to address. Computer matching is not a reliable way to verify voter eligibility. Brennan Ctr. Amicus Br. at 12-13; ODP Amicus Br. at 19-20; ACLU Amicus Br. at 7-8. Indeed, evidence from other states indicates that human error, not fraud, causes most mismatches. ODP Amicus Br. at 19. Data suggests that when the database match is conducted, anywhere from 15 to 30 percent of registered voters will fail to match. Brennan Ctr. Amicus Br. at 12-13. Disturbingly, mismatches have been shown to bar non-white voters more frequently than white voters. ACLU Amicus Br. at 7-8. Data from the Secretary of State shows that there are at least 485,000 new registered voters in Ohio this year. Rich Ex-ner, Ohio Voter Registration Surges, available at http://www.cleveland.com/ datacentral/index.ssf/2008/10/ohio_voter_ registration_surges.html#more (Oct. 6, 2008). Given these numbers, the TRO issued by the district court and revived by today’s result, could result in anywhere between 72,750 to 145,500 registered voters being removed from voting rolls, being forced to cast provisional ballots, or being otherwise wrongly disenfranchised.

In contrast to these concrete projections, ORP has failed to present evidence that any voters, including those who have registered in the last year, have committed actual voting fraud. Indeed, data collected by the Brennan Center, the same nonpartisan organization that studied the failure rate of data matching, indicates that actual voter fraud is extremely rare. Justin Levitt, The Truth About Voter Fraud, available at http://www.brennancenter.org/ content/resource/truthaboutvoterfraud/ (Nov. 09, 2007). The Brennan Center noted that “claims of voter fraud are frequently used to justify policies that do not solve the alleged wrongs, but that could well disenfranchise legitimate voters.” Id. at 3. In fact, this report noted that “[t]he most common source of superficial claims of voter fraud, and the most common source of error, probably involves matching voter rolls against each other or against some other source to find alleged double voters, dead voters, or otherwise ineligible voters.” Id. at 8. Similarly, the League of Women Voters of Ohio and the Coalition on Homelessness and Housing in Ohio studied the 9 million votes cast in Ohio between 2002 and 2004, and found only four fraudulent ballots. Let the People Vote, available at http://www.cohhio. org/alerts/ElectionReformfieport.pdf (Jun. 14, 2005). Therefore, the matching policy mandated by the TRO purportedly to eliminate voter fraud is actually likely to cause mismatches which will erroneously be la*735beled voter fraud, thereby disenfranchising individual voters and undermining public confidence in election results.7 The likelihood that this last-minute TRO will undermine confidence in the election results and will adversely affect, and perhaps even disenfranchise, a large number of voters is greater and more disturbing than the possibility that fraudulent votes will dilute the strength of legitimate ballots. Cf. Purcell, 549 U.S. at 7, 127 S.Ct. 5.

On the eve of the presidential election, ORP asks this court to derail election procedures that have been months in the making. The stay that the majority has vacated would simply have preserved the status quo and allowed Ohio to conduct its elections in the orderly method it had planned. The majority’s ruling today upends all order, injects the potential for erroneous disenfranchisement of qualified voters into the election, and creates confusion surrounding voting rights on the unverified specter of potential fraud. In a case involving legislative apportionment, an issue which raises the same questions of vote dilution that prompted the district court to issue a TRO, the Supreme Court highlighted the importance of not making hasty, unconsidered changes to election plans on the eve of major elections:

under certain circumstances, such as where an impending election is imminent and a State’s election machinery is already in progress, equitable considerations might justify a court in withholding the granting of immediately effective relief in a legislative apportionment case, even though the existing apportionment scheme was found invalid. In awarding or withholding immediate relief, a court is entitled to and should consider the proximity of a forthcoming election and the mechanics and complexities of state election laws, and should act and rely upon general equitable principles. With respect to the timing of relief, a court can reasonably endeavor to avoid a disruption of the election process which might result from requiring precipitate changes that could make unreasonable or embarrassing demands on *736a State in adjusting to the requirements of the court’s decree.

Reynolds, 377 U.S. at 585, 84 S.Ct. 1362.

The TRO perverts the clear language of HAVA, and makes a provision meant to “ensure that every American who is eligible to vote can vote” into a tool to effect widespread voter disenfranchisement. 148 Cong. Rec. S2527 (daily ed. Apr. 11, 2002) (statement of Sen. Daschle). The harm that the TRO will inflict on the Secretary, the voters, and the public is clear: elections in Ohio will be chaotic at best, blatantly unfair at worst, and doubt will fall on the validity of election results across the state. The mere specter of voter fraud, often raised but seldom proven, is the only injury that ORP asserts, and this unfounded threat cannot be used to coerce courts into acceding to the eleventh-hour demands of a partisan group that is intimately concerned with the results of the election it seeks to control. Accordingly, I would hold, as I did previously, that we should stay the TRO originally issued by the district court.

IV. CONCLUSION

It is clear that the district court’s four specific orders insert the federal court into the delicate balance struck by HAVA. I believe that under established law, there is no private right of action under 42 U.S.C. § 15483(a)(5)(B)®. Even if ORP had the right to bring such an action, which it does not, I believe that the Secretary is likely to succeed on the merits of the issue of the proper interpretation of HAVA; HAVA does not mandate that the Secretary undertake the particularized matching required by the district court’s TRO. Further, the irreparable harm caused by the district court’s TRO is significant. With less than three weeks until the election, and less than twelve days until the beginning of counting absentee ballots, the Secretary cannot be required to undertake the extensive reprogramming and other changes to the election mechanisms without complete disruption of the electoral process in Ohio. The irreparable harm to the voting public caused by the district court’s order is equally clear. Finally, the intrusion into the state’s processes by the federal courts with the ensuing confusion regarding the applicable process weighs heavily against the district court’s order.

As the Supreme Court wrote in Purcell, “Given the imminence of the election and the inadequate time to resolve the factual disputes, our action today shall of necessity allow the election to proceed without an injunction” altering the state’s established practice. Purcell, 549 U.S. at 8, 127 S.Ct. 5. Accordingly, I would hold that: (1) granting of en banc is inappropriate; (2) there is no private right of action and thus ORP’s claims must be dismissed; and (3) on the merits, we should GRANT the Secretary’s motion to vacate the district judge’s intrusive and invalid temporary restraining order. I dissent.

BOYCE F. MARTIN, JR., Circuit Judge.

I join Judge Moore’s well-reasoned dissent and would deny the petition to rehear this case en banc and would grant the motion to vacate the TRO. In a case full of irony, we have a majority of this court inferring a private right of action — on the basis of, well, nothing — to create utter bedlam in a state’s election. But Judge Moore has already covered that issue exceptionally well. What I want to highlight is something legitimately troubling, especially compared to what Judge Griffin found so “troubling” in his panel dissent.

What I find troubling is the fact that Judge Batchelder did not recuse herself from voting for rehearing this case en banc, while her husband stands for reelection this year as a state representative in Ohio, whose election will no doubt be sub*737stantially altered by the way the en banc majority ultimately decides this case. At stake here is the public’s confidence not only in the outcome of its elections, but also in the impartiality of its judges who must, from time to time, review the procedures which govern those elections.

The standards on this issue are clear. 28 U.S.C. § 455 requires federal judges to disqualify themselves “in any proceeding in which [their] impartiality might reasonably be questioned.” The Supreme Court has told us that the standard is objective, and what matters “is not the reality of bias of prejudice but its appearance.” Liteky v. United States, 510 U.S. 540, 548, 114 S.Ct. 1147, 127 L.Ed.2d 474 (1994). “Quite simply and quite universally, recusal [is] required whenever impartiality might reasonably be questioned.” Id. (quotations omitted).

When Congress enacted this language as an amendment in 1974, it was not semantics. The previous version of the statute only required recusal when a judge had a “substantial interest” in a case; now judges must recuse themselves when their impartiality “might reasonably be questioned.” 28 U.S.C. § 455(a) (emphasis added); Liteky, 510 U.S. at 547, 114 S.Ct. 1147. The upshot is that judges should be more cautious.

And § 455 does not stop at this catchall provision. We are also told that a judge “shall” disqualify himself or herself when he or she “has a personal bias or prejudice concerning a party,” 28 U.S.C. § 455(b)(1), and when his or her spouse “[i]s a party to the proceeding, or an officer, director, or trustee of a party” or “[i]s known by the judge to have an interest that could be substantially affected by the outcome of the proceeding.” 28 U.S.C. § 455(b)(5). If to quote this language is not to decide whether Judge Batchelder ought to have recused herself, I will elaborate. First, it is arguable that her husband, being a Republican candidate and long-serving member of the Ohio House of Representatives, is a “party” in this case as an “officer” of the Ohio Republican Party, a named plaintiff. Obviously, the language referring to “officers” and “directors” contemplates corporate entities, but there is nothing in the language to indicate that it should not include political parties. The Deputy Speaker of the House no doubt wields as much power within his own party as, say, a vice-president of a corporation.

Even more significant is the language from § 455(b)(5): there can be little argument that Judge Batchelder’s husband will not be “substantially affected by the outcome of [this] proceeding.” Although the actual result of Ohio’s elections is not directly before us, no one would argue that Tipper Gore or Laura Bush, were they to have sat on the U.S. Supreme Court in 2000, would not have had to recuse themselves in Bush v. Gore, 531 U.S. 98, 121 S.Ct. 525, 148 L.Ed.2d 388 (2000). It blinks reality to say that the outcome of this case will not have a “substantial” effect on Ohio’s elections and thus on Representative Batchelder’s re-election prospects.

Finally, even assuming that, under some tortured reading of the statute, one could argue that the outcome of this case would not “substantially affect” Judge Batchel-der’s interests here, what cannot be argued is that her vote was proper under the ABA Model Code of Judicial Conduct. ABA Model Code of Judicial Conduct Rule 2.11 (2007) states that a “judge shall disqualify himself or herself’ when “the judge’s spouse or domestic partner” “has more than a de minimis interest that could be substantially affected by the proceeding.” ABA Model Code of Judicial Conduct R.2.11 (2007). “De minimis” is defined as an “insignificant interest that could not raise a reasonable question re*738garding the judge’s impartiality.” Id. at “Terminology.” Judge Batchelder’s husband, like plaintiff Larry Wolpert, is a member of the Ohio Republican Party, they both currently serve as a Representative to the Ohio General Assembly, and both are up for reelection. Representative Batchelder thus unquestionably has “more than a de minimis interest” in the outcome of a proceeding which considers Ohio’s voting procedures in a rapidly approaching election.

So I respectfully submit that Judge Bat-chelder should not have voted in this case, whether or not the Ohio Republican Party’s claim actually tugs at her heartstrings. But I find this is all particularly ironic because Judge Griffin has said nothing about this, while he decided to publicly “object” to the panel majority’s procedures in his panel dissent. Specifically, he objected to the fact that Judge Moore and Judge Bright, a distinguished member of the Eighth Circuit, “decided to ignore the en banc petition and issue their order” in “violation of the practices of our court.” Indeed. The Secretary of State’s motion for a stay was properly before the panel when it was filed and panels generally ought to act first rather than defer every significant decision to the possibility that the full en banc court will later decide to hear the ease. And whether the panel acted when it did or waited, we would still be in the exact same position we are in now.

But if we are going to talk about what actions create an appearance of impropriety and undermine faith in our democratic system and in the judiciary’s independence, then the above speaks for itself. And if we are going to talk about what violates the practices of our court, then I can think of no better example than Judge Griffin’s decision to impugn a colleague— along with a visiting Judge! — on essentially no basis whatsoever. That said, this is not the first time this sort of thing has happened, so maybe he is right, and that kind of behavior can be considered one of the “practices of our court.” I dissent.

. The "Renewed Petition for Immediate En Banc Review" consists of one paragraph reiterating that plaintiffs/appellees "continue to seek immediate en banc review of” the Secretary's motion to vacate or stay the district court's temporary restraining order. This one-paragraph document does not add any substantive information. Instead, this document wrongly asserts that the panel improperly issued a stay while the petition for initial en banc hearing was pending before the court. Under long-standing Sixth Circuit rules, the panel has full authority to act until the court has decided to grant en banc hearing. The panel simply preserved the status quo that existed before the district judge entered his improper temporary restraining order. If the panel were not permitted to act once a party filed a request for en banc hearing, we would have parties running — indeed hijacking — the court. Any party disliking a panel to which it was assigned could attempt to stop that panel in its tracks by filing a motion for en banc hearing, just as plaintiffs/appellees have done here. In every instance of which I am aware, panels of this court routinely issue orders concerning cases while petitions for en banc are pending. Moreover, a petition to review en banc a decision of a Sixth Circuit motions panel to issue a stay is explicitly prohibited by our Sixth Circuit Rules. Sixth Circuit Internal Operating Procedures ("IOP") 35(e) (prohibiting en banc review petitions regarding panel decisions staying TROs and issuing other non-final orders). Thus, ORP is completely wrong in its procedural posture and analysis.

. As the Supreme Court observed in Gonzaga, 536 U.S. at 280, 122 S.Ct. 2268, " ‘In legislation enacted pursuant to the spending power, the typical remedy for state noncompliance with federally imposed conditions is not a private cause of action for noncompliance but rather action by the Federal Government to terminate funds to the State.’" (quoting Pennhurst State School & Hosp. v. Halderman, 451 U.S. 1, 28, 101 S.Ct. 1531, 67 L.Ed.2d 694 (1981)).

. The majority contends that it is an open question whether a statute creates a privately enforceable right when it "benefits everyone but no one in particular.” Sutton Op. at p. 720. But the Supreme Court has repeatedly answered this question in the negative. See Gonzaga, 536 U.S. at 288, 122 S.Ct. 2268.

. See, e.g., Westside Mothers v. Olszewski, 454 F.3d 532 (6th Cir.2006) (Boggs, C.J., joining in holding no implied private right of action); Hughlett v. Romer-Sensky, 497 F.3d 557, 563-64 (6th Cir.2006) (McKeague, J., holding no private right of action); Thomas M. Cooley Law Sch. v. Am. Bar Ass’n, 459 F.3d 705, 710-11 (6th Cir.2006), cert. denied, 549 U.S. 1116, 127 S.Ct. 985, 166 L.Ed.2d 710 (2007) (Batchelder, Gibbons, J.J., joining in holding no private right of action); Johnson v. City of Detroit, 446 F.3d 614 (6th Cir.2006) (Griffin, Gibbons, J.J., joining in holding no private right of action); Caswell v. City of Detroit Housing Comm’n, 418 F.3d 615, 618-20 (6th Cir.2005) (Batchelder, J., joining in holding no private right of action); Barton v. Summers, 293 F.3d 944, 951-55 (6th Cir.2002) (Boggs, J., holding no implied private right of action); Cline v. Rogers, 87 F.3d 176, 182-84 (6th Cir.), cert. denied, 519 U.S. 1008, 117 S.Ct. 510, 136 L.Ed.2d 400 (1996) (Batchelder, J., holding no implied private right of action).

. Judge Sutton faults the panel for having assumed that a private right of action could be implied when the panel issued its October 10 opinion. It was perfectly appropriate for the panel to assume a private right of action and then to deny relief to the plaintiffs. What Judge Sutton and the majority do, however, is completely illogical: they assume a private right of action can be implied and then grant the plaintiffs relief based on the assumed right to bring a claim. As this dissent has explained, there is no explicit or implied private right of action here.

. Judge Sutton's/the majority’s opinion misleadingly suggests that there has been an “apparent 'turn[ing] off of this voter-registration-verification process.” Sutton Op. at p. 714-15. However, although the precise procedures outlined in section 15.4 of the manual appear to have been discontinued, the computerized verification process has never been "turned off.” According to Gus Maragos, who has served as Ohio's Statewide Voter Database Coordinator since July 23, 2007, during his tenure the computerized verification process "has never been intentionally 'turned off but rather has operated without intentional interruption.” Secretary Br., Ex. 3 (Maragos Aff. at 2).

. Members of the majority lightly assume that reinstating the TRO will not harm individual voters. Sutton at ¶ 22 ("Nothing about this case or the relief plaintiffs seek will allow them to prevent a single voter from casting a ballot in the November election. At most ... it could require an individual to cast a provisional ballot."). This short-sighted assessment ignores the reality that voters do not always know they have a right to cast a provisional ballot and even when they do, may not understand the procedure or be able to wait for necessary assistance. See Ohio Rev.Code § 3505.181 (explaining the complex Ohio provisional ballot requirements). Notably, some who cast provisional ballots in Ohio are required to bring verifying information to the board of elections within ten days or else their votes will not be counted. Id. This is a burden that many individuals cannot bear. See Ari Z. Weisbard, Placebo Ballots: Will “Fail-Safe” Voting Fail? at 2, available at http:// www. demos. org/pubs/HAVA-PlaceboB allotsb wl01904.pdf (Oct.2004). Additionally, there is no guarantee that such ballots will be counted in time to affect the election. Daniel P. Tokaji, An Unsafe Harbor: Recounts, Contests, and the Electoral College, 106 Mich. L.Rev. First Impressions 84, 87 (2008) ("One of the major steps before a final vote total can be ascertained is determining which provisional ballots should be counted, a process that would almost surely become heated in a tight election. It is difficult to see how the process of verification of ballots, let alone any judicial proceedings that might take place over the canvassing and recounting of ballots, could be completed by the safe harbor date.”); Chandler Davidson, et al., Vote Caging as a Republican Ballot Security Technique, 34 Wm. Mitchell L.Rev. 533, 541 (2008) ("A survey of provisional ballots nationwide in 2004 found that only sixty-eight percent of the 1.6 million [provisional] ballots cast were counted.”).