284 F.3d 251
Valjeanne Currie, et al., Plaintiff, Appellant,
Group Insurance Commission, et al., Defendants, Appellees.
United States Court of Appeals, First Circuit.
April 1, 2002.
S. Stephen Rosenfeld with whom Suzanne L. Schwartz, Rosenfeld & Associates, and Richard Ames were on brief for appellant.
Ginny Sinkel, Assistant Attorney General, with whom Pierce O. Cray, Assistant Attorney General, and Thomas F. Reilly, Attorney General, were on brief for appellees.
Ronald S. Honberg, Mary Giliberti, Thomas M. Sobol, Lydia Alix Fillingham, and Lieff, Cabraser, Heimann & Bernstein, LLP on brief for The National Alliance for the Mentally Ill and the Judge David L. Bazelon Center for Mental Health Law, amici curiae.
Before Lynch and Lipez, Circuit Judges and Woodlock,1 District Judge.
LYNCH, Circuit Judge.
The ultimate question in this case presents significant issues about a state government's ability to allocate insurance benefits by creating distinctions between different classes of individuals. Valjeanne Currie appeals the district court's entry of summary judgment for the defendant, the Group Insurance Commission (GIC), which provides disability benefits for employees of the state of Massachusetts. Currie v. Group Ins. Comm'n, 147 F. Supp.2d 30 (D.Mass. 2001). Currie challenges an aspect of the GIC long-term disability benefits policy, which limits benefits for noninstitutionalized individuals with mental disabilities to one year; GIC imposes no such time limit on benefits for the institutionalized mentally ill or on noninstitutionalized individuals with physical disabilities. Currie argues that this policy violates the Americans with Disabilities Act (ADA), 42 U.S.C. § 12101-12213 (1994), the Equal Protection Clause of the United States Constitution, and the Due Process Clause of the United States Constitution.
The immediate issue is whether this court should proceed to resolve the merits of this case while an appeal proceeds in parallel litigation in the state courts on an issue of state law which could moot or otherwise inform the federal litigation. The plaintiffs have asked us to stay our hand. The defendants urge us to dispose of the case on a difficult issue of federal statutory construction on which the circuits are split. They generally would prefer a prompter disposition of the federal action, but agree that the state court should decide state law issues. There is then the issue of what form a stay should take, should we decide to stay. The alternatives proposed are a stay under Colorado River Conservation District v. United States, 424 U.S. 800 (1976), or certification of the state law question to the Massachusetts Supreme Judicial Court (SJC). In the interest of comity, we elect to stay on Colorado River grounds. Certification would interrupt the normal state appellate processes. Moreover, it would put the decisions of the state law issue directly to the state's highest court on a record developed to address federal, not state, issues. Finally, it is unclear whether the SJC would accept certification where, as here, the state court's decision on state issues would not be dispositive of the federal issue, but would merely render it moot.
Valjeanne Currie was a Massachusetts state employee for fourteen years, working at the Massachusetts Mental Health Center. She suffers from schizophrenia. In 1999, her illness forced her to take a leave of absence from work and she has not been able to return to work since that time. She receives daily psychiatric care on an outpatient basis. There is no dispute in this case that Currie's illness is severe, genuine, and debilitating.
The GIC is a state agency, established by state law, to provide state employees with medical, dental, life, and disability insurance. Mass. Gen. Laws ch. 32A, §§ 1-4, 10D (2000). The long-term disability insurance (LTD) program, which is the policy Currie challenges, provides income assistance to state employees who become disabled and cannot work. The governing statute charges the GIC with establishing a disability insurance plan "on such terms and conditions as it deems to be in the best interest of the commonwealth and its employees." Id. § 10D. The plan is required to be self-supporting; by statute, the Commonwealth may make no contribution to the support of the plan. Id. The plan is also voluntary — state employees may choose whether or not they wish to participate. Participating employees pay premiums during the course of their employment. Massachusetts state employees are not permitted to participate in the federal social security system, and so Currie does not have access to the federally sponsored social security net available to most Americans.
The Commonwealth initiated the LTD plan in 1988. The GIC accepts bids from private insurers to cover the LTD plan. Prior to 1998, the plan did not provide any benefits for mentally disabled individuals who were not hospitalized. In 1994, the Hartford Life Insurance Company, the private insurer carrying the LTD contract, suggested adding coverage for nonhospitalized mentally disabled individuals. However, after some consideration, the GIC determined that the rate increase required for such coverage was infeasible due to the risk of adverse selection. Adverse selection is a problem confronted by voluntary insurance plans, whereby those individuals who consider themselves to have a low risk opt out of the program. This decreases the amount paid into the program, and increases the percentage of program participants who will eventually receive benefits. Of course, as the cost of coverage rises, more low-risk individuals will choose to opt out.
When the Hartford contract was renewed, effective July 1998, the GIC's outside consultants recommended that the new contract provide for one year's worth of benefits for nonhospitalized mentally disabled individuals. The GIC adopted this recommendation, which is the policy challenged by Currie. After this first year of benefits, the individual may only continue to receive benefits if he or she is confined to a hospital or institution, in which case the benefits continue until the individual is discharged. Plan participants who suffer from physical disabilities have no such limitations on their coverage.
Currie began receiving benefits in June of 1999. In October of that year, she received a letter informing her that the payments would be terminated in June of 2000 unless she entered an institution. In January of 2000, Currie filed suit against the GIC in the federal district court. In May of 2000, Currie filed suit in state court, challenging the same provision of the LTD policy based on Massachusetts state antidiscrimination law, Mass. Gen. Laws ch. 151B (2000). On June 7, 2000, a state superior court judge entered a preliminary injunction, ordering GIC to continue her benefits, and thus necessarily finding some probability of success. Currie v. Hartford Life Ins. Co., No. 00-1831 (Mass.Super.Ct. June 7, 2000). On June 14, 2001, the federal district court denied summary judgment for the plaintiffs and granted summary judgment to the defendants. On January 24, 2002, a state Superior Court judge denied plaintiff's motion for summary judgment, granted defendant's motion for summary judgment, and dismissed the case. Currie v. Hartford Life Ins. Co., No. 00-1831-H (Mass.Super.Ct. Jan. 24, 2002). We cannot tell if the record in the state court case is fuller than or identical to the summary judgment record in the federal case. We can say the issues in the state and federal cases are not identical and therefore the evidence presented may be different. Plaintiffs have appealed that decision.
Currie argues that entering an institution would severely decrease the likelihood that her condition would improve to the extent that she would be able to return to work, and has presented affidavits from her treating doctors to support this argument. She implies that GIC's policy, which allows for unlimited benefits for the hospitalized mentally ill, may therefore cost it more in the long run than would a policy allowing her to continue outpatient treatment. What is at stake, then, she argues, is not the amount of money GIC will pay out, but rather her ability to continue in a noninstitutionalized setting.2
Following oral argument in this case, the GIC informed us that it has negotiated a new LTD policy contract which will take effect when the current contract with the Hartford expires in July 2002. The new policy, carried by C.N.A. Group Benefits, will provide LTD benefits beyond one year for individuals, like Currie, who have mental disabilities and are receiving outpatient care in the form of day treatment, partial hospital treatment, or residential treatment for at least five hours per day, four days per week. Because this new policy will not apply to Currie or to other individuals who stop working before the new policy comes into effect in July 2002, GIC does not suggest that this change moots Currie's claim.
Currie makes three challenges to the LTD policy offered by GIC through the Hartford, one premised on the ADA and two premised on the federal constitution.
A. ADA Claim
First, Currie argues that the LTD policy violates Title II of the Americans with Disabilities Act, 42 U.S.C. § 12131-12165, which states that "no qualified individual with a disability shall, by reason of such disability, be excluded from participation in or be denied the benefits of the services, programs, or activities of a public entity, or be subjected to discrimination by any such entity." Id. § 12132. The district court granted summary judgment to the defendant on this claim, holding that: 1) Title II of the ADA does not encompass employment practices; and 2) even if Title II covered employment, the LTD plan would fall under the "safe harbor" provision established by Congress for certain state insurance programs, id. § 12201(c). In its safe harbor ruling, the district court stated that because there was a rational basis for the distinction in benefits, the classification did not violate the state antidiscrimination statute. The court made no ruling on whether the classification violated the state constitution. Currie, 147 F. Supp.2d at 33-38. The GIC agrees with both of the district court's conclusions. Neither question is easily decided.
1. Title II coverage
The question of whether challenges to employment practices are cognizable under Title II has been considered by two of our sister circuits, and by several trial courts within this circuit, with divergent results. See Zimmerman v. Or. Dep't of Justice, 170 F.3d 1169, 1173-74 (9th Cir.1999) (not cognizable), cert. denied, 531 U.S. 1189 (2001); Bledsoe v. Palm Beach County Soil & Water Conservation Dist., 133 F.3d 816, 820-22 (11th Cir.1998) (cognizable); Downs v. Mass. Bay Transp. Auth., 13 F. Supp.2d 130, 134-36 (D.Mass. 1998) (cognizable); Motzkin v. Trs. of Boston Univ., 938 F. Supp. 983, 996 (D.Mass. 1996) (not cognizable); see also McKibben v. Hamilton County, 215 F.3d 1327 (6th Cir.2000) (per curiam) (unpublished table decision) (noting split, but proceeding on merits where coverage not challenged by defendant). Title I of the ADA, 42 U.S.C. § 12112-12117 (1994), expressly covers the employment practices of both private and public entities.3 Id. § 12112.
The district court believed that the clear language of Title I indicated that Title I was the sole avenue for bringing employment claims, and that the clear language of Title II indicated that Title II was limited to so-called "outputs" of a public agency. Currie, 147 F. Supp.2d at 34-35. Specifically, the court found that the second clause of Title II, which mandates that qualified individuals not "be subjected to discrimination by any [public] entity" was not intended to expand the scope of coverage beyond "services, programs, or activities" (articulated in the previous clause), but simply to clarify that Title II prohibits both intentional discrimination (through the "subjected to discrimination" clause) and disparate treatment (through the "excluded from participation in or ... denied the benefits of" clause). Id. at 34-36 (analyzing 42 U.S.C. § 12132). The court supported its clear language analysis by finding that the overall structure of the ADA, which provides a remedy for employment discrimination under Title I, supported its conclusion. Id. at 36.
The answer is not so plain. While Title I's language clearly covers employment discrimination, and public employers are not exempted from the definition of a covered entity, Title I says nothing about it being an exclusive remedy or avenue for suit. 42 U.S.C. § 12112. It is not unheard of for individuals to have overlapping rights, even within one Act.4 Here, the two Titles grant substantively different rights — for instance, while Title I gives successful plaintiffs the opportunity to obtain compensatory and punitive damages, there is no such right under Title II. Id. § 12133 (referencing 29 U.S.C. § 794a). Nor is the language of Title II clear on this question. The words "public services, programs, or activities" do not necessarily exclude employment,5 and the "subjected to discrimination" clause may broaden the scope of coverage further. Moreover, the Department of Justice has promulgated a regulation stating that Title II does cover employment practices. 28 C.F.R. § 35.140 (2001); see also 28 C.F.R. pt. 35, App. A (2001) (elaborating on § 35.140). This regulation is entitled to deference under the Chevron doctrine if the statutory language is unclear. Chevron U.S.A, Inc. v. Natural Res. Def. Council, Inc., 467 U.S. 837, 842-43 (1984). In addition, Currie cites to legislative history which she says demonstrates that Congress intended Title II to cover employment and to function in the same manner as Section 504 of the Rehabilitation Act.
2. Safe Harbor
Even beyond the difficult statutory interpretation question of whether Title II covers employment, there is a second complicated statutory question — whether the "safe harbor" provision of the ADA, 42 U.S.C. § 12201(c)(2), immunizes the LTD. The safe harbor provision states that the ADA shall not be construed as prohibiting a covered organization "from establishing, sponsoring, observing or administering the terms of a bona fide benefit plan that are based on underwriting risks, classifying risks, or administering such risks that are based on or not inconsistent with State law." Id.
Currie argues that the safe harbor provision does not apply to the challenged aspect of the LTD plan because the hospitalization requirement is not based on any actual data. She cites the legislative history and the regulations as support for the proposition that a risk-based defense must be based on "sound actuarial data and not on speculation." See 28 C.F.R. pt. 36, App. B., at 676 (2001) (internal quotation marks omitted) (citing legislative history and discussing 28 C.F.R. § 36.212 (2000)).
The GIC responds that the safe harbor does not require it to conduct actuarial studies to support its policies. See Rogers v. Dept. of Health & Envtl. Control, 174 F.3d 431, 437 (4th Cir.1999) (actuarial data not required for safe harbor); Ford v. Schering-Plough Corp., 145 F.3d 601, 611 (3d Cir.1998) (insurance company need not justify its policy coverage after a plaintiff's mere prima facie allegation). Instead, the GIC asserts, the policy need only be based on "actual or reasonably anticipated experience," a standard used by some district courts and also articulated in the legislative history. See Currie, 147 F. Supp.2d at 37; Chabner v. United of Omaha Life Ins. Co., 994 F. Supp. 1185 (N.D.Cal. 1998), aff'd on other grounds, 225 F.3d 1042 (9th Cir.2000); Doukas v. Metro. Life Ins. Co., 950 F. Supp. 422, 428-29 (D.N.H. 1996) (citing H.R. Rep. 485(II), at 135-48, reprinted in 1990 U.S.C.C.A.N. 303, 418-21). The GIC has presented evidence that it says supports its policy under this standard. This evidence includes the fact that the industry standard is to impose a durational limit on disability benefits; evidence that under the University of Maine's LTD plan, which is an employer-paid LTD covering all employees, 25% of recipients are mentally disabled; evidence that the costs associated with generous disability benefits offered by the University of Massachusetts's LTD plan had led the insurer to cancel the contract; evidence that adverse selection had been a problem in GIC's Indemnity Health Insurance plan; and an affidavit from a principal with the consulting company that the GIC retained to help it evaluate benefits stating that, in the principal's opinion, no insurance company would agree to underwrite a voluntary, employee pay-all LTD plan that offered unlimited mental disability benefits.
Currie, however, argues that there is a second problem with the application of the safe harbor provision. The safe harbor provision applies only to risk-based policies that are "based on or not inconsistent with State law," 42 U.S.C. § 12201(c)(2). She argues that the LTD policy is inconsistent with the Massachusetts Constitution and the state antidiscrimination law, Mass. Gen. Laws, ch. 151B, which prohibits employers from discriminating on the basis of handicap. Thus, we arrive at the questions that are currently pending before the state courts.
B. Federal Constitutional Claims
Currie also claims that the GIC's use of the Hartford policy violates her federal constitutional rights to equal protection and due process of law.
Currie faces a difficult test under the equal protection clause. Currie does not contest the district court's holding that the policy will survive equal protection scrutiny if it is rationally related to a legitimate governmental purpose.6 In its brief, the GIC asserts three possible justifications for the policy: 1) hospitalization serves as a proxy for determining those with the greatest need, i.e., those with the longest or most severe disability; 2) hospitalization serves as a proxy for verifying total disability; and 3) some limitation is necessary to keep costs at a viable level.7 Currie's response boils down to two main points: 1) that none of these rationales explain why mental disability should be treated differently than other disabilities, such as muscular-skeletal disorders, that are common and difficult to verify, and the distinction is therefore arbitrary; and 2) that the policy will eventually cost the GIC more, because it undermines beneficiaries' attempts to return to work, and therefore is not rationally related to cost concerns.
Currie's due process claim rests on the theory that the policy impermissibly denies her a government benefit on a basis that infringes her constitutionally protected interest (namely, her right to liberty of person). See, e.g., Perry v. Sindermann, 408 U.S. 593, 597 (1972) (holding that it would be impermissible for government to deny renewal of employment contract based on employee's exercise of free speech rights). A state policy that has the effect of penalizing the exercise of a fundamental right must be justified by a compelling state interest in order to survive constitutional scrutiny. Shapiro v. Thompson, 394 U.S. 618, 634-38 (1969) (administrative reasons for denying welfare benefits to recent interstate immigrants not compelling), overruled in part on different grounds by Edelman v. Jordan, 415 U.S. 651 (1974).
As explained below, we decline to decide any of Currie's federal claims at this juncture, due to the pendency of the state court proceedings.
Before oral argument in this case, Currie filed a motion with this court requesting a stay in these proceedings, pending the outcome of the state court proceedings.8 Currie suggested that a stay would save this court from having to decide the federal statutory and constitutional issues, particularly the ADA safe harbor question, which is intertwined with the state law question currently under consideration in the state court system. The GIC opposed this motion, arguing that we could affirm summary judgment on the merits by deciding that Title II of the ADA does not cover employment practices, or by deciding that the "contrary to state law" exception to the safe harbor applies only to matters of state insurance law. Of course, if the highest state court were to determine the plan violates state antidiscrimination law or the Massachusetts Constitution, both questions would most likely be moot, as the GIC would necessarily have to change the plan.
Generally speaking, "in cases where the relief being sought is equitable in nature or otherwise discretionary, federal courts not only have the power to stay the action based on abstention principles, but can also, in otherwise appropriate circumstances, decline to exercise jurisdiction altogether by either dismissing the suit or remanding it to state court." Quackenbush v. Allstate Ins. Co., 517 U.S. 706, 720 (1996).
A question is raised as to whether this case is within the scope of the Colorado River stay doctrine. In Colorado River, the Supreme Court held that "in situations involving the contemporaneous exercise of concurrent jurisdiction ... by state and federal courts" it may be appropriate for the federal court to defer to the state court. 424 U.S. at 817. However, the Court emphasized that "the circumstances permitting the dismissal of a federal suit due to the presence of a concurrent state proceeding for reasons of wise judicial administration are considerably more limited than the circumstances appropriate for abstention" and should be "exceptional" to justify deferral to the state court. Id. at 818; see also Rojas-Hernandez v. P.R. Elec. Power Auth., 925 F.2d 492, 495-96 (1st Cir.1991); Villa Marina Yacht Sales, Inc. v. Hatteras Yachts (Villa Marina I), 915 F.2d 7, 12 (1st Cir.1990); Bath Mem'l Hosp. v. Me. Health Care Fin. Comm'n, 853 F.2d 1007, 1015 (1st Cir.1988). There is a "heavy presumption favoring the exercise of jurisdiction." Villa Marina I, 915 F.2d at 13. There must be some extraordinary circumstances for a federal court to shrink from "the virtually unflagging obligation of the federal courts to exercise the jurisdiction given them." Colorado River, 424 U.S. at 817. The mere pendency of parallel state litigation does not warrant a stay, save for exceptions not pertinent here.
This court has identified six factors, based on the Supreme Court's decision in Colorado River and its subsequent decision in Moses H. Cone Memorial Hospital v. Mercury Construction Corp., 460 U.S. 1 (1983):
(1) whether either court has assumed jurisdiction over a res; (2) the inconvenience of the federal forum; (3) the desirability of avoiding piecemeal litigation; (4) the order in which the forums obtained jurisdiction; (5) whether federal law or state law controls; and (6) whether the state forum will adequately protect the interests of the parties.
Rivera-Puig v. Garcia-Rosario, 983 F.2d 311, 320-21 (1st Cir.1992) (quoting Burns v. Watler, 931 F.2d 140, 146 (1st Cir.1991)). However, this is not an exhaustive list, nor is it a litmus test for Colorado River deference, which must remain a discretionary tool. See Villa Marina I, 915 F.2d at 14.
The first two prongs of the Colorado River/Moses H. Cone test have little bearing on this case. There is no res at issue and the federal forum is equally convenient to the state forum, as both are located in the same city. There is some risk of piecemeal litigation here, which may rise above the "routine inefficiency that is the inevitable result of parallel proceedings." Villa Marina I, 915 F.2d at 16. The Supreme Court has clarified that the fourth prong (sometimes called the "priority" element) "should not be measured exclusively by which complaint was filed first, but rather in terms of how much progress has been made in the two actions." Moses H. Cone, 460 U.S. at 21. In this case, both state and federal cases have been through summary judgment and are on appeal, and it may be that the record in the state case is fuller as to the state law issue. See Colorado River, 424 U.S. at 820 (noting the apparent absence of any federal proceedings other than the motion to dismiss as a factor in favor of dismissal); Villa Marina Yacht Sales, Inc. v. Hatteras Yachts (Villa Marina II), 947 F.2d 529, 535 (1st Cir.1991) (holding that state case was further advanced because of development of record for preliminary injunction hearing).
The concerns implicated by the fifth prong of the test, whether federal or state law controls, are important in this case. Although this case presents exclusively federal law claims, two of the three federal claims are constitutional and therefore should only be adjudicated if we are unable to resolve the case through resolution of the statutory claim. See Ashwander v. Tenn. Valley Auth., 297 U.S. 288, 347 (1936) (Brandeis, J., concurring). If we conclude that Title II encompasses employment practices, we must determine whether the LTD program is protected by the safe harbor provision. That federal statutory question is intertwined with a complex issue of state law, pending before the state courts. Further, the underlying subject matter involves state-provided insurance benefits, a matter in which the state has unusually strong interests.
We have noted that "[c]ourts generally have agreed that rare circumstances exist only when a case presents `complex questions of state law that would best be resolved by a state court.'" Villa Marina I, 915 F.2d at 15 (quoting American Bankers Ins. Co. of Fla. v. First State Ins. Co., 891 F.2d 882, 886 (11th Cir.1990)). Colorado River has special appeal where a state court decision "may substantially, perhaps even fully, answer certain questions of state law in a way that will permit easy answers, relatively speaking, to the federal ones." Kartell v. Blue Shield of Mass., 592 F.2d 1191, 1193-94 (1st Cir.1979) (remanding to district court with order to stay proceedings pending state court outcome or to certify state supreme court).
These comity concerns are the same as those underlying the abstention doctrines that predate Colorado River's discretionary deferral. In Colorado River, the Supreme Court described these abstention doctrines, stating that abstention is appropriate "where there have been presented difficult questions of state law bearing on policy problems of substantial public import whose importance transcends the result in the case then at bar." 424 U.S. at 814 (citing La. Power & Light Co. v. City of Thibodaux, 360 U.S. 25 (1959), and Burford v. Sun Oil Co., 319 U.S. 315 (1943)); see also R.R. Comm'n of Tex. v. Pullman Co., 312 U.S. 496, 500 (1941) (discussing "a doctrine of abstention appropriate to our federal system whereby the federal courts, `exercising a wise discretion', restrain their authority because of `scrupulous regard for the rightful independence of the state governments'") (quoting DiGiovanni v. Camden Fire Ins. Ass'n, 296 U.S. 64, 73 (1935) and Cavanaugh v. Looney, 248 U.S. 453, 457 (1919)). Although this case does not mandate abstention under any of these established abstention doctrines, the problem it presents is similar to the problems presented by the Pullman or Thibodaux cases, which presented federal courts with the prospect of being required to resolve complicated state law problems. For instance, two of the major purposes of Pullman abstentions are to "avoid the waste of a tentative decision" and to "avoid ... needless friction between the federal and state proceedings." Ford Motor Co. v. Meredith Motor Co., Inc., 257 F.3d 67, 73 (1st Cir.2001) (internal quotation marks omitted); see also Bath Mem'l Hosp., 853 F.2d at 1016 ("Pullman-type abstention ... may be appropriate ... because ... plaintiffs are making identical claims in two state court suits, and the state courts may resolve the claims in ways that would moot, or significantly affect, the claims plaintiffs make here....").
Much of the rationale supporting abstention in those cases applies to counsel deference here. The state law question is not clear, nor is it clear how the state ultimately would balance the important policy interests of treatment of the disabled with the financial viability of insurance policies. If we were to decide the ADA claim here, we could be forced to make a ruling on whether the policy violates Massachusetts antidiscrimination law, the question before the Massachusetts courts. If we were to decide the federal issues in the manner that the GIC suggests, a state court ruling that the policy violates Massachusetts law would render our opinion merely advisory — an outcome we seek to avoid in any case.9
Finally, the sixth prong of the Colorado River/Moses H. Cone test also supports deferring to the state court. A stay under Colorado River is appropriate only where the parties may obtain complete relief in the state court proceedings:
When a court decides to dismiss or stay under Colorado River, it presumably concludes that the parallel state-court litigation will be an adequate vehicle for the complete and prompt resolution of the issues between the parties. If there is any substantial doubt as to this, it would be a serious abuse of discretion to grant the stay or dismissal at all.
Moses H. Cone, 460 U.S. at 28. Nonetheless, "perfect identity of issues is not a prerequisite." Villa Marina II, 947 F.2d at 533.
It is very significant to us that it is the plaintiff, the party that initially invoked the jurisdiction of the federal courts, who is now requesting that we stay our hand. Moreover, because the plaintiff is the same in both this case and the parallel state case, there is no danger that the plaintiff will be prejudiced by ineffective prosecution of the state law claim. Nor will the defendant be prejudiced by our staying the action: whatever uncertainty exists as to outcome in this case also exists as to the state court litigation.
The parties have proposed certification as an alternative to a Colorado River stay, and the dissent prefers that course. However, the state law question is currently before the state intermediary appellate court, on appeal from the state Superior Court case. If we were to certify, we would be interfering with that normal state appellate process. If the SJC wishes to provide the parties with a more prompt resolution of this question, it can accelerate the appeal by taking direct appellate review of the state law case. Mass.R.App.P. 11(f). In addition, the record before us has been developed to address the federal issues, not the state law question. Therefore, our certification might not provide the SJC with an adequate record on which to decide the question, a requirement under the SJC's certification rule. The state court case, in which the state law question has been directly litigated, contains a far more appropriate record on which to decide the question. Finally, it is not clear that the SJC would accept certification. The Supreme Judicial Court of Massachusetts Rule 1:03 permits certification "if there are involved questions of law of this state which may be determinative of the cause then pending in the certifying court." If the SJC holds that the distinction in insurance benefits is not in violation of state law, that will not be determinative of the federal claims. If the SJC holds that the distinction is in violation of state law, that may moot the federal claims. Whether this situation meets the certification requirement is itself an issue of Massachusetts law which we should not decide. On more than one occasion, state high courts have returned certified questions unanswered, because the factual record was undeveloped on the state law question or because there was a risk that its opinion would be merely advisory. See, e.g., Cuesnongle v. Ramos, 835 F.2d 1486, 1491 (1st Cir.1987) (discussing Puerto Rican Supreme Court's decision not to answer certified question regarding state constitutional law, where state law followed federal law); Lumbermens Mut. Cas. Co. v. Belleville Ind., 407 Mass. 675, 555 N.E.2d 568, 576 (Mass. 1990) (returning certified question unanswered because factual record was insufficiently developed); see also Canal Elec. Co. v. Westinghouse Elec. Co., 406 Mass. 369, 548 N.E.2d 182, 184 (Mass. 1990) (noting that "if, in the future, the `questions certified to us ... are not accompanied by sufficient nonhypothetical, evidentiary facts to allow us to adequately determine' the answers, we may decline to answer such questions"). Under these circumstances, certification is not the wisest course available to us.
We hold, therefore, that a stay pending the outcome of the state proceedings is the wisest course of action at this time. We emphasize that we are not surrendering federal jurisdiction and we retain jurisdiction to permit us to resolve the federal questions if a decision is ultimately necessary.
Motion for stay granted.
Of the District of Massachusetts, sitting by designation
Currie has also stated that if her benefits are cut off, she will inevitably become homeless and enter an institution and implies that at that time, she will then begin receiving LTD benefits again. It is not clear from the policy whether, once she becomes ineligible, she would later be able to receive benefits even if she did enter an institution. The GIC has not addressed this issue
Currie did not attempt to challenge the LTD policy under Title I of the ADA because of the procedural requirements imposed by Title I. Currie, 147 F. Supp.2d at 33 n. 6
For instance, employers that receive federal assistance may be covered by both Title VI (applying to programs and activities that receive federal funds) and Title VII (applying to employment practices) of the 1964 Civil Rights Act. See Guardians Ass'n v. Civil Serv. Comm'n, 463 U.S. 582, 606-07 (1983) (plurality) (district court relief premised on violations of both Title VI and Title VII affirmed in part and reversed in part because relief awarded under Title VI exceeded permitted scope). Similarly, a disabled individual working for a federally funded entity may be covered by both the ADA and the Rehabilitation Act of 1974. See Phelps v. Optima Health, Inc., 251 F.3d 21, 23 & n. 2 (1st Cir.2001)
There are two long-standing civil rights laws, Section 504 of the Rehabilitation Act of 1973, 29 U.S.C. § 794(a) (1994 & Supp. IV 1998) and Title IX of the Education Amendments of 1972, 20 U.S.C. § 1681(a) (1994), under which the phrase "program or activity" has been held to cover employment practices. See Consol. Rail Corp. v. Darrone, 465 U.S. 624, 632-634 (1984) (Rehabilitation Act); North Haven Bd. of Educ. v. Bell, 456 U.S. 512, 520-35 (1982) (Title IX). In addition, Title VI of the 1964 Civil Rights Act, 42 U.S.C. § 2000d, which also uses the "program or activity" language, contains a specific subsection, id. § 2000d-3, limiting its application to employment discrimination to certain instances
The Supreme Court has held that state discrimination on the basis of mental retardation will survive an equal protection challenge unless the challenged practice is not rationally related to some legitimate governmental purpose. See City of Cleburne v. Cleburne Living Ctr., 473 U.S. 432, 446 (1985). Other precedent indicates that this test applies to state discrimination against disabled individuals generally. See Bd. of Trs. of the Univ. of Ala. v. Garrett, 531 U.S. 356, 367 (2001) ("States are not required by the Fourteenth Amendment to make special accommodations for the disabled, so long as their actions toward such individuals are rational."); Currie, 147 F. Supp.2d at 33 (citing City of Cleburne for the proposition that "it is well established that, for purposes of this analysis, the disabled do not constitute a suspect classification")
Currie argues that only the cost justification was presented to the district court
The question of whether to defer to the parallel state proceedings was not before the district court, and therefore our holding is not an indication that the district court abused its discretion in any way by failing to defer. Cf. Moses H. Cone Mem'l Hosp. v. Mercury Constr. Corp., 460 U.S. 1, 19 (1983) (question of whether to defer is within the discretion of the district court). Nor do we view this issue as having been waived by Currie because she did not move for a stay before the district court — a federal court's discretionary authority to defer to a state court due to comity reasons may be invoked regardless of whether the parties request it
Nothing in this opinion undercuts any res judicata effect of the Superior Court judgment, an argument which the dissent creates sua sponte and on which it relies. By the same token, the res judicata effect of a lower state court judgment does not compel the active exercise of federal jurisdiction, rather than a stay when federal doctrines of restraint counsel otherwise. This is true whether the doctrines of restraint counsel certification of questions "to the State's highest court" to avoid reaching federal questions, see Arizonans for Official English v. Arizona, 520 U.S. 43, 78 (1997), or whether they counsel abstention, see generally Heck v. Humphry, 512 U.S. 477 (1994) ("[I]f a state criminal defendant brings a federal civil-rights lawsuit during the pendency of his ... appeal ... abstention may be an appropriate response...."); Ford Motor Co. v. Meredith Motor Co., 257 F.3d 67 (1st Cir.2001) (Pullman abstention appropriate where state administrative board ruling was on appeal to state superior court); Amerson v. State of Iowa, 94 F.3d 510, 512 (8th Cir.1996) (Burford abstention on claim for interference with parental rights appropriate where state court appeals were pending); Turnbow v. Pac. Mut. Life Ins., 934 F.2d 1100, 1103 (9th Cir.1991) ("The time for Pacific Mutual to oppose Turnbow's abstention argument was before the Nevada Supreme Court reached its decision.")
Here the doctrine of sound judicial administration which underlies Colorado River deferral has even more force because the state proceeding is already on appeal on a fully developed record. See Hearne v. Bd. of Educ., 185 F.3d 770, 778 (7th Cir.1999) (holding that the "principles of sound judicial administration which animated the decision in Colorado River ... require[d] a stay of the federal proceedings" pending outcome of appeal in parallel state court proceedings); Rogers v. Desiderio, 58 F.3d 299, 302 (7th Cir.1995) ("It is sensible to stay proceedings until an earlier-filed state case has reached a conclusion" on appeal); Akins v. Rodriguez, 15 F.3d 883, 887, vacated per stipulation, 26 F.3d 105 (9th Cir.1994) (Colorado River abstention appropriate where state appellate court decision on appeal to state supreme court). The party who invoked jurisdiction here has sought the stay and the defendants have not raised the argument made by the dissent, even in their post-argument filings after the Superior Court judgment, and so it is forfeited. See Soares v. Brockton Credit Union, 107 F.3d 969, 972 n. 1 (1st Cir.1997) (noting that defendant never raised res judicata issue and therefore court will deem argument to be waived); see also Walsh v. Int'l Longshoremen's Ass'n, 630 F.2d 864 (1st Cir.1980) (court may raise res judicata issue sua sponte where parties argued res judicata before district court but not on appeal). Even had it not been forfeited, the interests served by res judicata are better served by the stay.
WOODLOCK, District Judge (Dissenting).
The main current in this appeal flows through a problem of federal statutory construction which the District Court resolved on cross motions for summary judgment. The majority chooses to bypass the federal statutory question presented, at least for the time being, by invoking the discretion to stay recognized under Colorado River Water Conservation District v. United States, 424 U.S. 800 (1976). I believe that to do so here is a misuse of the limited discretion that the Colorado River doctrine reserves for truly "exceptional" circumstances. Id. at 818. There is nothing particularly exceptional in the circumstances of this case and there is no good cause to neglect "the virtually unflagging obligation of the federal courts to exercise the jurisdiction given them." Id. at 817. Consequently, I must respectfully dissent.
A. Deployment of the Colorado River Doctrine is Not Justified
The majority opinion relies upon three Colorado River factors whose confluence it finds sufficient to justify avoiding exercise of our jurisdiction. Those factors cannot support setting this case adrift upon the Colorado River doctrine. In this section, I will address the three factors the majority relies upon and the one Colorado River factor — avoidance of piecemeal litigation — the majority accords negligible weight. A careful consideration of the factors demonstrates that the Colorado River doctrine is by its own terms inapplicable.
1. The first filed and decided federal proceeding is substantially more developed than the later filed and decided state proceeding.
The federal proceeding pending before us went to final judgment in the federal district court on June 14 of last year after resolution of cross motions for summary judgment on all issues. Currie v. Group Ins. Comm'n, 147 F. Supp.2d 30 (D.Mass. 2001). The appeal has been fully briefed and argued.
The state proceeding did not go to final judgment in the state trial court until February 6 of this year, following a ruling on cross motions for summary judgment, an aspect of which concerned a subset of one of the issues before us. Currie v. Hartford Life Ins. Co., Suffolk No. 00-1831-H (Mass.Super.Ct. Jan. 24, 2002).10 The plaintiff has reported an intention to appeal but the record has not yet been assembled and no briefing schedule for the appeal has been established. The plaintiff, who is required to bring the appeal to the Massachusetts Appeals Court in the first instance, tells us review in the Supreme Judicial Court will also be sought. Plainly, the first filed11 federal proceeding fully argued before us is substantially more developed than the later filed state proceeding in which appellate proceedings are in their infancy and may have to mature through two state appellate courts.12
2. This exclusively federal law case is not so intertwined with a particularly complex novel state statutory claim as to justify avoidance.
As the majority notes, "this case presents exclusively federal law claims," slip op. at 18. The only one of these federal claims with force is the statutory ADA claim.13 The majority explores the decision tree for this claim and finds hanging under one alternative branch a "federal statutory question ... intertwined with a complex issue of state law, pending before the state courts...." Id.
The use of the horticultural metaphor "intertwined" to describe the relationship of the state law issue to the federal statutory claim illustrates the force of Cardozo's observation that "[m]etaphors in the law are to be narrowly watched, for starting as devices to liberate thought, they end often by enslaving it." Berkey v. Third Ave. Ry. Co., 244 N.Y. 84, 94 (1926). This is demonstrated most readily by simply recasting the horticultural metaphor to capture more accurately the reality the figure of speech is describing.
Far from being intertwined with the entire federal claim, the state law issue is simply appended to one of two independent branches of federal statutory analysis. It is the choice of the majority to lash the two branches together by staying this case — and not something inherent in the decision tree — that intertwines a secondary state law issue with the federal statutory analysis. It is apparent that we could reach the ADA Title II question and dispose of the case without even encountering the state law question presented by the safe harbor branch. That approach has not been substantively explored by us as yet.
Moreover, merely labeling the state law issue appended to one branch of the decision tree as "complex" does not make it so. The majority properly does not seek to ground its choice to stay on traditional abstention doctrines. These are reserved for genuinely difficult questions of state law and no such question is presented here.
In any event, if we are required to reach the alternative "safe harbor" branch of this case, we will simply encounter a manageable issue of state law which two dispositive opinions, that of the federal District Court that we review, Currie v. Group Ins. Comm., 147 F. Supp.2d at 36-38, and that of the state Superior Court, Currie v. Hartford Life Ins. Co., slip op. at 9-12, have already addressed respectively in a broad and a narrow sense and decided with no apparent strain.
3. The protections of the state forum are undermined unless this court applies accepted state rules for recognition of state judgments between the parties.
The decision of the state Superior Court granting summary judgment on all counts to the defendants is a final judgment to which the Massachusetts courts, following the majority view, Restatement (Second) of Judgments § 13 cmt. f. (1982), must accord res judicata effect despite the pendency of any appeal. O'Brien v. Hanover Ins. Co., 427 Mass. 194, 200-01 (1998). We can do no less.
Under 28 U.S.C. § 1738, "judicial proceedings of any court of any... State ... shall have the same full faith and credit in every court within the United States ... as they have by law or usage in the courts of such State...." When we treat a final judgment of the Massachusetts Superior Court as some sort of provisional order we are not acting in deference to the state court litigation; rather, we act in derogation of Massachusetts judgment rules we are bound by a federal statute to observe. The question whether the LTD policy is inconsistent with Massachusetts law has been answered between the parties before us and § 1738 requires we give that answer preclusive effect. Consequently we are now obliged to resolve the remaining questions consistent with that answer and without regard to whether one or another of the Massachusetts appellate courts to which the state appeal will be taken may ultimately reverse the Superior Court.
Application of res judicata through principles of full faith and credit — unlike the comity concerns of the Colorado River doctrine and traditional abstention approaches — is not discretionary. As Justice Frankfurter observed for the Supreme Court in Williams v. North Carolina, 325 U.S. 226, 228 (1945), "the Full Faith and Credit Clause puts the Constitution behind a judgment instead of the too fluid, ill-defined concept of `comity.'" I have found no case in which a federal court has addressed its full faith and credit obligations under § 1738 and concluded that they may be ignored or deferred.14 Indeed, when courts actually have addressed § 1738 in this context, they have made significant efforts to examine carefully its impact on the case at hand. See, e.g., Marrese v. Am. Acad. of Orthopaedic Surgeons, 470 U.S. 373, 379-86 (1985); Cruz v. Melecio, 204 F.3d 14, 18-21 (1st Cir.2000); Rogers v. Desiderio, 58 F.3d 299, 301-02 (7th Cir.1995).
The majority suggests that the question of the applicability of 28 U.S.C. § 1738 has been forfeited because it was not raised. That is not the rule in this circuit. Res judicata is a question which can be addressed by this court on its own motion. Walsh v. Int'l Longshoremen's Ass'n, Local 799, 630 F.2d 864, 867 (1st Cir.1980). In any event, the basis for invoking § 1738 did not arise until after argument before us in this case and could not properly have been anticipated in the briefing and thereby waived. Under Fed.R.App.P. 28(j), "[i]f pertinent and significant authorities come to a party's attention after the party's brief has been filed — or after oral argument but before decision — a party may promptly advise the circuit clerk by letter ... [but] the letter must state without argument the reasons for the supplemental citations...." (emphasis added). The defendants carefully complied with Rule 28(j) here by sending the clerk a copy of the state Superior Court ruling. They could not properly do more.15 By staying this case we neglect our independent duty to apply § 1738. It bears emphasizing that if we were to meet our duty in this regard we would not be required separately to decide the state law issue appended to the safe harbor claim. That issue has already been decided in a state court judgment to which we must give voice. But we have chosen to stand mute by undertaking to stay.
4. A Colorado River stay will encourage the continuation of otherwise avoidable piecemeal litigation.
The current of this litigation, with its history of and prospects for serial decisionmaking in duplicative venues, strongly draws us on a collision course with a material Colorado River factor — the desirability of avoiding piecemeal litigation. Yet, the majority sails quickly past this factor, conclusorily observing that there "is some risk of piecemeal litigation." Slip op. at 17 (emphasis added). How much emphasis one puts on the adjective "some" is key to the judgment. The direction of the litigation charted by the parties causes them to tack back and forth among the several state and federal trial and appellate courts; the record amply demonstrates quite some past — and the likelihood of continued future — resort to piecemeal litigation here. A stay will only increase that likelihood.
A bit of history will explain how the litigation between these parties came to be presented in bits and pieces. Initially, plaintiff's complaint filed in the federal District Court on January 25, 2000, raised six counts, (1) Title II of the ADA, (2) Amendment CXIV (the prohibition against disability discrimination) under the Massachusetts Constitution, (3) Due Process under the Massachusetts Constitution, (4) Due Process under the United States Constitution, (5) Equal Protection of the Massachusetts Constitution and (6) Equal Protection under the United States Constitution. In response to the defendants' motion to dismiss arguing that the Eleventh Amendment to the United States Constitution precluded a federal suit against state officials on the basis of state law, see generally Pennhurst State Sch. & Hosp. v. Halderman, 465 U.S. 89, 117 (1984), on March 25, 2000 the plaintiffs dismissed the state constitutional claims, Counts 2, 3 and 5, in the federal court and commenced litigation in the state court on strictly state law claims.16
In the federal District Court the parties litigated as a general matter the federal safe harbor provision, 42 U.S.C. § 12201(c), under which Congress directed that Title II of the ADA should not be construed to prevent administration of a bona fide benefit plan based on "underwriting risks, classifying risks, or administering such risks that are based on or not inconsistent with State law." Id. § 12201(c)(1) & (2). Whether Mass. Gen. Laws ch. 151B, the state statutory prohibition against employment discrimination, was such an inconsistent law is an issue plainly within the general scope of the safe harbor provision but its applicability was not specifically pressed in the federal litigation. Rather this one potential form of state law inconsistency became an issue dealt with specifically in the state Superior Court.
After the appellees filed their briefs in this court, the appellants for the first time sought a stay of the federal litigation, contending that "[t]he initial exchange of briefs before this court has made clear a feature of this case that unfortunately remained in the background in the district court." Mot. of Pls/Appellants for Ct. to Stay Proceedings Pending Adjudication in State Ct. ¶ 1. That "feature" was the specific issue whether the LTD plan was inconsistent with state law by virtue of Mass. Gen. Laws ch. 151B. The plaintiff does not contend that the question of inconsistency with state law was withheld from the federal proceeding or was somehow reserved for the state courts. To the contrary, she contends the plaintiffs "have not waived their argument of inconsistency with state law. That claim has been part of this [federal] case from the beginning." Appellants' Reply to Appellees' Opp'n to Appellants' Mot. to Stay Proceeding Pending Adjudication in State Ct. ¶ 4.
The federal and state proceedings in which the parties have engaged present a textbook example of piecemeal litigation. To date, by keeping "in the background" an aspect of a claim that has been a part of the federal case "from the beginning," the parties have fully briefed and argued — in whole and in part — to three courts, two federal and one state, the issue of consistency with state law. The decision to stay insures that two more courts — the Massachusetts Appeals Court and the Massachusetts Supreme Judicial Court — will be offered a bite at the apple hanging from one branch of the decision tree in this case. And looming in the background is the potential for yet another court, the Supreme Court of the United States, faced with a preexisting split in the circuits, to review the other branch of the federal statutory question, the availability of a Title II claim — and hence in this case the availability of the ADA at all — to this type of case. We should not sail by the Colorado River factor directed to the desirability of avoiding piecemeal litigation. Rather we should accept the challenge and decide the case by actively exercising our federal jurisdiction.
B. The Exercise of Federal Appellate Jurisdiction is an Imperative
As the competing treatments of the recognized Colorado River factors separately provided in the majority opinion and in this separate opinion illustrate, there is a quicksilver quality to the Colorado River doctrine. Its multifactor test — in which no weights are assigned until the balancing process is actually undertaken — creates conditions that, at a minimum, invite unpredictability. Lacking some greater prescriptiveness beyond the general adjuration that it is reserved for "exceptional circumstances," the Colorado River doctrine periodically overflows the banks meant to contain it and floods garden variety federal question litigation.
That is a major reason the commentators have not been particularly kind to the Colorado River doctrine. See, e.g., James C. Rehnquist, supra note 2; Linda Mullenix, A Branch Too Far: Pruning The Abstention Doctrine, 75 Geo. L.J. 99 (1986); David A. Sonenshein, Abstention: The Crooked Course Of Colorado River, 59 Tul. L. Rev. 651 (1985). The application of the factors has "reveal[ed] great disparity as to what constitutes exceptional circumstances," Professor Chemerinsky reports. Erwin Chemerinsky, Federal Jurisdiction § 14.3, at 830 (3d ed. 1999). "Despite the Court's statement in Moses H. Cone Memorial Hospital v. Mercury Construction Co. that such abstention is to be rare and limited to `exceptional' circumstances, many lower courts continue to order abstention when there are parallel proceedings pending in state courts. But the other lower federal courts refuse Colorado River abstention unless there are truly exceptional circumstances." Id. at 828-29 (footnotes omitted).
This court has traditionally been among those reluctant to resort to the Colorado River doctrine, even in the more inviting circumstance when the litigation pending in the federal court involves a wholly state law dispute. See, e.g., Burns v. Watler, 931 F.2d 140 (1st Cir.1991). When the issue is one of federal law, the imperative of exercising federal jurisdiction takes on its own special hydraulic force. The Supreme Court has taught that the presence of federal law questions "must always be a major consideration weighing against surrender [of jurisdiction]." Moses H. Cone, 460 U.S. at 26 (footnote omitted).
A stay in these circumstances begs a question of first principles. Chief Justice Marshall long ago stated those principles for purposes of federal appellate jurisdiction.
With whatever doubts, with whatever difficulties, a case may be attended, we must decide it, if it is brought before us. We have no more right to decline the exercise of jurisdiction which is given, than to usurp that which is not given. The one or the other would be treason to the constitution. Questions may occur, which we would gladly avoid; but we cannot avoid them. All we can do is, to exercise our best judgment, and conscientiously to perform our duty. Cohens v. Virginia, 6 Wheat. (19 U.S.) 264, 404 (1821).
Drained of unconvincing references to Colorado River factors, what appears to be generating the choice to stay this litigation in this court is the possibility that a reversal of the state Superior Court decision in the Supreme Judicial Court of Massachusetts would provide an adequate and independent state ground for this court to avoid a problem of federal statutory construction which has created (in only one of its independent branches) a split in the circuits and disagreement among district courts. That possibility is not enough to justify a discretionary technique promising a lengthy delay.17 Whatever our doubts, whatever the difficulties, we should decide the federal question presented to us.
C. The Certification Expedient is a Less Damaging Alternative
Ultimately, resort to the Colorado River doctrine, if proper, can be justified and measured only in terms of "wise judicial administration." Colorado River, 424 U.S. at 818. Even if we are entitled for some period of time to defer our obligations to exercise federal jurisdiction, the most efficient tool of judicial administration to use is not a stay but certification to the state's highest court. Certification of state law questions is a tool the Supreme Court has, particularly of late, Arizonans for Official English v. Arizona, 520 U.S. 43, 75-80 (1997), enthusiastically recommended for courts faced with parallel proceedings. The parties themselves agree on little else but that this would be an efficient means to obtain the benefit of the Supreme Judicial Court's views and agree to its use by us. It is a tool which recently provided prompt clarification regarding the reach of Mass. Gen. Laws ch. 151B, the state disabilities law principally at issue here, Dahill v. Boston Police Dep't, 434 Mass. 233 (2001), and has been used by this court on appropriate occasions. See, e.g., Medical Prof'l Mut. Ins. Co. v. Breon Lab., Inc., 141 F.3d 372, 378 (1st Cir.1998); Protective Life Ins. Co. v. Sullivan, 89 F.3d 1, 4-5 (1st Cir.1996); Pyle v. South Hadley Sch. Comm'n, 55 F.3d 20, 22 (1st Cir.1995).
The majority expresses reticence about making use of the certification tool but its reasons are not compelling, particularly in the face of an agreement between the parties that certification is an efficient way to expedite final resolution of an issue that they have variously been presenting in the federal and state courts. There is little doubt, given the chronic underfunding of the state courts, that prosecution of an appeal in the ordinary course will be a time consuming process. However, the briefing in the state Superior Court, attached to the motion to stay papers, makes clear that the issue was presented there as essentially a legal question for which recourse to an elaborate factual record is unnecessary.
The state law question is precisely the type of issue the certification process was designed to address in the interests of saving "time, energy, and resources and helping build a cooperative judicial federalism." Arizonans for Official English, 520 U.S. at 77 (quoting Lehman Bros. v. Schein, 416 U.S. 386, 391 (1974)). We are presumably staying our hand because the issue is one we think, in the words of the Supreme Judicial Court's certification rule, "may be determinative of the cause" pending before us. SJC Rule 1:03, § 1 (emphasis added). We are thus in the mainstream of the certification rule. While the Supreme Judicial Court has warned of the danger of hypothetical questions posed in the context of the interlocutory proceedings by courts of first instance, see, e.g., Knapp Shoes v. Sylvania Shoe Mfg. Corp., 418 Mass. 737, 738 n. 1 (1994); Lumbermens Mut. Cas. Co. v. Belleville Indus., Inc., 407 Mass. 675,
Because the memorandum of the state Superior Court is not published in any readily accessible reporter system and its substance bears upon the issues in this case, I attach a copy as an appendix to this opinion
I recognize, of course, that under governing law this
Colorado River factor "should not be measured exclusively by which complaint was filed first, but rather in terms of how much progress has been made in the two actions." Moses H. Cone Mem'l Hosp. v. Mercury Constr. Corp., 460 U.S. 1, 21 (1983). A very powerful and compelling critique of Colorado River and other avoidance techniques, arguing in favor of an exclusive first filing standard, has been mounted in James C. Rehnquist, Taking Comity Seriously: How to Neutralize the Abstention Doctrine, 46 Stan. L. Rev. 1049, 1068 (1994) ("A federal court should abstain if, and only if, the federal plaintiff has an adequate opportunity to litigate his federal claim in a duplicative suit already pending in state court."). I merely note that declining to invoke Colorado River in this case is not only consistent with the doctrine's own standards but also with the alternative standard proposed by one of its more thoughtful critics.
The two cases cited by the majority regarding this factor, slip op. at 18, serve only to underscore that the disparity of progress in the two proceedings between the parties is a factor actually favoring timely continued exercise of federal jurisdiction by us. In Colorado River the Supreme Court found an "apparent absence of any proceedings in the [federal] District Court, other than the filing of the complaint, prior to the motion to dismiss" which the District Court granted in an unreported oral opinion. 420 U.S. at 805-06, 820 & n. 25. And in Villa Marina Yacht Sales, Inc. v. Hatteras Yachts, 947 F.2d 529 (1st Cir.1991), we adopted the District Court's determination "that the Commonwealth action had progressed substantially further than the federal case," observing that "in addition to the injunction hearing, a pretrial report had been filed in the Commonwealth action and ten depositions have been completed. In the federal action, in contrast, Hatteras has yet to answer the complaint and little discovery has taken place." Id. at 535
The majority, quite properly in an opinion explaining an intention to stay rather than address the substantive issues, presents the respective contentions — statutory and constitutional — of the parties in a largely disinterested fashion without purporting to resolve them
Because the relative propriety of a stay is affected, however, by whether weighty constitutional issues are truly at issue, I find it necessary, in the interest of limiting my disagreement with the majority to the narrowest grounds, to note that the asserted constitutional claims should play no role in the stay calculus. This is because I am of the view the federal constitutional claims plaintiff raises — unlike the federal statutory claim — are meritless, essentially for the reasons stated in the federal District Court opinion, Currie v. Group Ins. Comm., 147 F. Supp.2d 30, 38-39 (D.Mass. 2001), and that of the State Superior Court, which relied on federal constitutional case law in rejecting the parallel state constitutional claims, Currie v. Hartford Life Ins. Co., Suffolk No. 00-1831-H (Mass.Super.Ct. Jan 24, 2002), slip op. at 4-9. By contrast, I have reached no conclusion on the merits of the federal statutory claim.
The collection of cases gathered in a footnote by the majority, slip op. at 21 n. 8, does not suggest otherwise. All are inapposite to the question whether we must apply 28 U.S.C. § 1738 here
I will address in section C, infra, the certification procedure proposed by the parties and endorsed by the Supreme Court in Arizonans for Official English v. Arizona, 520 U.S. 43 (1997). Heck v. Humphrey, 512 U.S. 477 (1994), falls within the penumbra of traditional Younger abstention and is generally understood as part of a long line of federal cases and legislation that requires finality and exhaustion before a later filed federal case may challenge the validity of a state court criminal proceeding or conviction. Ford Motor Co. v. Meredith Motor Co., Inc., 257 F.3d 67 (1st Cir.2001), and Amerson v. Iowa, 94 F.3d 510 (8th Cir.1996), as the majority candidly observes, involve yet other abstention doctrines. Moreover, in Meredith there was no lower state court judgment but only the ruling of an administrative agency lacking res judicata effect on review in the state Superior Court. 257 F.3d at 70. And in Amerson, the Eighth Circuit held "federal court interference in a domestic relations context where the state courts have entered judgment is ... inappropriate." 94 F.3d at 513.
In Turnbow v. Pacific Mutual Life Insurance Co., 934 F.2d 1100 (9th Cir.1991), the Nevada Supreme Court having already ruled on the parallel state case, the Ninth Circuit concluded that "[b]ecause we hold that res judicata bars federal relief, we need not decide whether the district court properly dismissed the action on abstention grounds." Id. at 1104. In Akins v. Rodrigues, 15 F.3d 883 (9th Cir.1994), the Ninth Circuit found that "[t]he California state courts obtained jurisdiction ... more than four years before the federal courts obtained jurisdiction. .. . There ha[d] been little or no progress in the federal litigation [and] [s]tate law questions predominate[d] in the federal action." Id. at 887.
Because the Illinois judgment rule is uncertain, the Seventh Circuit has prudently chosen not to accord lower state court decisions res judicata effect. Thus, in Rogers v. Desiderio, 58 F.3d 299 (7th Cir.1995), Judge Easterbrook, after reviewing conflicting Illinois case law, said "[t]o be blunt, we have no idea what the law of Illinois is on the question whether a pending appeal destroys the claim preclusive effect of a judgment." Id. at 302. Hearne v. Board of Education, 185 F.3d 770 (7th Cir.1999), simply reviewed the status of the Illinois judgment rule and found that "[t]he clarity of the Illinois law of preclusion ... on the effect of a judgment that is still being appealed has not changed appreciably since we decided Rogers." Id. at 778.
The defendants' compliance with Fed.R.App.P. 28(j)
in connection with the state Superior Court's determination contrasts favorably with the quite argumentative letters the parties have submitted to us — purportedly under Rule 28(j) — with respect to the Supreme Court's post-argument decision in Raygor v. Regents of the University of Minnesota, 122 S.Ct. 999 (Feb. 27, 2002).
We noted some time ago the challenge, in the wake of Pennhurst, to the orderly and efficient resolution of controversies having both federal and state law dimensions and the value of certification of state law questions to meet those challenges. As Judge Coffin observed in Rogers v. Okin, 738 F.2d 1 (1st Cir.1984):
In 1982, the Supreme Court avoided decision of a difficult federal constitutional question by remanding to this court, urging us to dispose of the case on state law grounds. [Mills v. Rogers,] 457 U.S. 291, 306, 102 S.Ct. at 2452. With Pennhurst, the Court removed our power to do so. Now, two years after the Supreme Court returned this case to us in furtherance of the Court's "settled policy" of avoiding unnecessary constitutional questions, Pennhurst requires us to face those questions. Fortunately, as will be indicated below, the Massachusetts Supreme Judicial Court's answers to our certified questions of state law have simplified our current task by changing the variables in the constitutional equation.
Id. at 4 (footnote omitted).
The tortured subsequent history of Kartell v. Blue Shield of Massachusetts, Inc., 592 F.2d 1191 (1st Cir.1979), which the majority cites, slip op. at 19, as illustrative of a circumstance in which "Colorado River has special appeal," is more instructive on the ways in which delay can flow from a stay. The decision to stay in Kartell was made in 1979 by a divided panel in which Judge Aldrich for the majority wrote that a prompter resolution of the state law issues would be had in cases then pending in the Supreme Judicial Court where the federal court plaintiffs were not parties than if questions were certified from the federal court. Id. at 1195. Judge Coffin filed a "dubitante" opinion taking the position that "even though the court deferred certification `in the interest of saving time and procedures' such deferment threatens, if decision on pending state cases proves unhelpful, to involve a needless delay. I would immediately certify." Id. at 1196. Judge Coffin proved prescient. The pending state cases did not resolve the federal issues. Questions were then certified by the District Court and the Supreme Judicial Court answered them in 1981. The case finally returned to this court from the District Court for decision on the merits in 1984 when Judge Breyer for the court wearily observed "[i]n view of these legal and practical problems, and the fact that this case has been pending in the federal courts for more than seven years, we believe it simpler and more appropriate to proceed directly to the antitrust merits, which, on our view of the case, are dispositive." Kartell v. Blue Shield of Massachusetts, Inc., 749 F.2d 922, 924 (1st Cir.1984)