Connecticut Department of Children and Youth Services v. Department of Health and Human Services

Opinion for the Court filed by Circuit Judge SILBERMAN. SILBERMAN, Circuit Judge: The State of Connecticut appeals the judgment of the district court rejecting its challenge to a Department of Health and Human Services (HHS) determination that the State had not met statutory requirements for foster care and thus was ineligible for additional funding for Fiscal Year 1985. The State claims that the Secretary’s standards implementing the statutory requirements were improperly applied retroactively and, in any event, were applied in an arbitrary and capricious manner. We affirm. I. Under the Adoption Assistance and Child Welfare Act of 1980, Pub.L. No. 96-272, 94 Stat. 500 (1980), states can qualify for bonus foster care funds if they meet several conditions. They must inventory all children who had been in foster care for the six-month period preceding the inventory. For each of these children, the states must also determine the appropriateness of current foster care placement and ascertain whether the child should be returned to his parents or placed for adoption. 42 U.S.C. § 627(a)(1). In addition, states must implement certain other foster care procedures and programs. *983 They must establish a database of characteristics, status, and goals for each foster child, adopt a program designed to either return children to their families or place them up for adoption, and implement a case review system for each child. 42 U.S.C. § 627(a)(2)(A)-(C). The statute defines a case review system, in part, as “a procedure for assuring that” each child has “periodic reviews no less frequently than once every six months.” 42 U.S.C. § 675(5)(B). Periodic reviews determine the “continuing necessity for and appropriateness of the placement.” Id. A case review system also requires a “dispositional hearing” no later than 18 months after a placement and every 18 months thereafter. 42 U.S.C. § 675(5)(C). These dispositional hearings plan the child’s future status by ascertaining whether the child should return to his parents, continue in temporary foster care, be placed for adoption, or remain in foster care on a permanent or long-term basis. Id. All of these procedures and programs must be implemented to the “satisfaction of the Secretary.” 42 U.S.C. § 627(a)(2). In 1982, the Department issued Program Instruction (PI) 82-06, which notified states of the methods by which HHS would determine whether a state had satisfied the Secretary. The program instruction informed states that if they wished to receive additional foster care funds, they first would have to certify compliance with the statutory requirements. HHS would then assess whether a state had actually fulfilled section 627’s conditions. The Department would review a state’s administrative procedures and then sample individual foster care ease records. Each case in the sample would be deemed “acceptable” or “unacceptable” depending on whether or not it met the statutory requirements — particularly the three critical elements of section 675, namely a written case plan, periodic reviews no less than once every 6 months, and dispositional hearings no later than 18 months after the child’s original placement and every 18 months thereafter. A case record must also meet 13 of 18 other statutory requirements. The program instruction set two levels of acceptable compliance. In the first year a state seeks bonus funds, at least 66% of the cases sampled must be acceptable. Thereafter, the Secretary would be satisfied only if at least 80% of sampled cases are acceptable (so-called “substantial compliance”). Having achieved substantial compliance, a state would not be subject to further review for at least two years. PI 82-06 did not establish levels of compliance for subsequent reviews. In January of 1985, the Department issued a new program instruction, PI 85-02, which provided the compliance rate for reviews conducted after a state met substantial compliance. State foster care agencies were told that to receive additional funds, at least 90% of sampled foster care cases must be acceptable. As before, each case would have to meet the three critical concerns of section 675. But, instead of having to satisfy 13 of 18 statutory conditions, states would have to meet 15 of the 18 prerequisites. The new program instruction also established that reviews would take place in the third year after the previous review and every third year thereafter (so-called “triennial” reviews). In Connecticut’s first triennial review, conducted in April 1986 for FY 1985, HHS formally reviewed 13 cases and found that 8 were not acceptable. Based on its sequential sampling model, 1 HHS determined that it need not review any more cases to conclude that Connecticut had failed the triennial review. The Department, therefore, informed Connecticut that the State should remit approximately $1.5 million to HHS. Connecticut appealed to HHS’ Grant Appeals Board (GAB). Before GAB considered the appeal, the Department provided GAB with the detailed findings for cases 1 through 76 of the random sample. Of these cases, the Department asserted that 34 were unacceptable, more than sufficient to conclude that Connecticut had failed. Connecticut, conceding that 2 cases had failed to satisfy the statutory requirements, *984 challenged the 32 other HHS determinations as arbitrary and capricious. The State contested HHS’ conclusion that a number of eases did not satisfy the periodic review and dispositional hearings conditions. In several cases, the Department had determined that, although reviews and hearings were held, they were not timely. Connecticut also claimed that certain children were exempt from the periodic review and dispositional hearing requirement, but HHS had ignored the exemptions and had adjudged these cases unacceptable. Appellant raised other miscellaneous arguments regarding the supposed lack of substantial evidence supporting HHS determinations in a number of other cases. Rejecting the bulk of Connecticut’s arguments, GAB concluded that Connecticut had failed to meet the statutory requirements in 28 of those 32 contested eases. Given the State’s original concession that 2 cases were unacceptable, Connecticut had 30 unacceptable eases out of 76. The Department’s sequential sampling model indicated that GAB need not consider any other cases; with a 76 record sample, 17 unacceptable cases were enough to determine that a state had failed. Since Connecticut had more than 30 unacceptable cases (around a 58% compliance rate), it had substantially less than the 90% compliance rate sufficient to satisfy the Secretary. Before the district court, Connecticut argued that PI 85-02, which had raised the compliance rate from 80% to 90%, was an invalid legislative rule which should have undergone notice and comment and that HHS had improperly applied this legislative rule retroactively. 2 Alternatively, the State urged that the GAB had arbitrarily and capriciously determined that 28 cases were unacceptable. The government contended that judicial review was wholly unavailable under the Administrative Procedure Act as there was no law to apply, and that even if review were available, PI 85-02 was not a legislative rule but an interpretative rule, which could be issued without notice and comment and which could be applied retroactively assuming retroactivity was even implicated. Recognizing that the statute granted the Department vast discretion, the district court held that the Secretary’s decision to establish a 90% compliance standard was “committed to agency discretion by law.” Connecticut Dep’t of Children and Youth Servs. v. Department of Health and Human Servs., 788 F.Supp. 573, 578 (D.D.C.1992). The “statute’s broad delegation,” moreover, provided no standard to evaluate the GAB’s decision on individual cases. Id. The court did conclude that judicial review was available to determine “whether any potential procedural right of the plaintiff has been violated by the agency’s failure to promulgate rules.” Id. But the district court rejected the State’s assertions that PI 85-02 was a legislative rule, holding it, instead, to be an interpretative rule which could be applied retroactively. Id. at 579-80. II. Appellant would focus our attention on the issue it features — whether the new program instruction is a regulation issued without notice and comment and applied retroactively in contravention of Bowen v. Georgetown University Hospital, 488 U.S. 204, 215, 109 S.Ct. 468, 475, 102 L.Ed.2d 493 (1988). We think appellant’s argument, in the circumstances of this case, is a rather well-crafted diversion. Were it necessary to characterize the new program instruction, we would be inclined, like the district judge, to reject appellant’s claim that it is a regulation under the Administrative Procedure Act, but not because we see it as an interpretative rule; it does not purport to define statutory terms. The program instruction strikes us rather as a policy statement that sets forth something like enforcement criteria, see Attorney General’s Manual On The Administrative Procedure Act 30 n. 3 (1947) (describing policy statements as setting forth “the manner in which the agency proposes to exercise a discretionary power”), but does not seek to change the normative standards under which the statute operates. As the government points out, the statute holds the State re *985 sponsible for handling each foster care case properly, and the program instructions do not purport to relax that obligation. 3 We do not have to decide this issue, however, because the State of Connecticut, in our view, is not aggrieved by the adoption of the new program instruction. The State would be injured only if its compliance rate reached the 80% deemed satisfactory under the old program instruction, yet fell short of 90% required by the new. As we have noted, however, the State’s compliance rate was only 58%; therefore, unless the State could successfully challenge the Department’s judgment as to the individual eases in the sample deemed unsatisfactory (at least enough to bring its compliance rate above 80%), it has no standing to contest the issuance or application of the new program instruction. The State would have us decide the legislative/interpretative issue first, and then, if we agreed with the State and also agreed that the Department’s determination of compliance in individual eases was reviewable, we should remand consideration of these cases to the district court. We do agree with the State (and disagree with the district judge) that the Secretary’s treatment of the sample cases is reviewable, although our scope of review, in light of the broad delegation to the Secretary, is extremely narrow. But, we see no need to remand because the issue— whether the Secretary’s determinations were reasonable — is strictly a question of law. See East Coast Line v. United States, 234 U.S. 167, 185, 34 S.Ct. 867, 872, 58 L.Ed. 1267 (1914). The government argues that because the statute speaks in terms of the Secretary’s satisfaction, we cannot review the Secretary’s action at all, including even the question the district court decided — whether the new program instruction was validly adopted and applied. Under Heckler v. Chaney, 470 U.S. 821, 831-33, 105 S.Ct. 1649, 1656, 84 L.Ed.2d 714 (1985) and Webster v. Doe, 486 U.S. 592, 599-601, 108 S.Ct. 2047, 2051, 100 L.Ed.2d 632 (1988), the government claims there is no law to apply. This jurisdictional claim is, however, vastly overdrawn. If the new program instruction were in fact a regulation— setting forth the substantive standards a state was obliged to meet to gain the Secretary’s satisfaction — the program instruction itself would constitute the law to apply. See Padula v. Webster, 822 F.2d 97, 100 (D.C.Cir.1987). And, if it were applied retroactively, we would face a Bowen problem. We think this case, however, is quite removed from either Heckler v. Chaney or Webster v. Doe. Heckler, holding that an agency’s prosecutorial decision whether to bring an enforcement action is not reviewable, is quite inapposite. Doe v. Webster, involving the CIA Director’s discretion to discharge employees whenever he “shall deem such termination necessary or advisable in the interests of the United States,” 486 U.S. at 600, is distinguishable because there Congress provided no standards, nor did the Director establish any, by which the court could determine what was in the nation’s interests. By contrast, the statute here explicitly sets forth the factors the Secretary must take into account in determining whether a state complies with the law. See, e.g., Kreis v. Secretary of the Air Force, 866 F.2d 1508, 1514 (D.C.Cir.1989) (noting that although statute granted agency discretion, it also provided law to apply). It could not be seriously argued that if the Secretary refused to provide Connecticut with foster care bonus funds because, say, she disapproved of Connecticut’s income tax policy, we would lack authority to review that decision. Nor do we think it could be maintained that the Secretary’s discretion in determining whether she is satisfied by Connecticut’s compliance rate is unbridled — that the Secretary could, for instance, decide that Connecticut had to meet *986 a 90% compliance rate, but Wisconsin was obliged to show only 50% compliance. 4 To be sure, given the extraordinary grant of discretion Congress gave the Secretary, a state would have to present an egregious claim to prevail, but that goes to the scope of review, not to whether judicial review is available at all. See id. ^ * * * * $ Turning, then, to Connecticut’s challenge to the Secretary’s case determinations, we have little difficulty rejecting appellant’s claim. Connecticut argues that the Secretary was unreasonable in deciding that 10 cases in the sample were not reviewed by the State often enough to meet the statutory periodicity standard. The Agency looked back to FY 1984 to determine when Connecticut had last conducted a periodic review and dispositional hearing for these 10 children. Connecticut insists that since the triennial review was for FY 1985, HHS could only consider reviews and hearings held in FY 1985. The State argues that GAB had “acted arbitrarily” by permitting the Department to penalize Connecticut for conduct that took place the previous year. Appellant’s argument is entirely without merit. The Act requires a periodic review every 6 months and a dispositional hearing every 18 months. As the GAB correctly noted, “[tjhere is no way to determine whether the first review held in FY 1985 was timely without knowing when in FY 1984 the last review took place.” The same holds true for the requirement of a dispositional hearing every 18 months. If HHS was to determine whether Connecticut had complied with periodic reviews and dis-positional hearing requirements of the Act in FY 1985, it necessarily had to look back to FY 1984. Appellant also maintains that HHS determinations that the State failed to provide 7 children with dispositional hearings every 18 months were arbitrary and capricious. Connecticut urges that these children fell within HHS regulations providing exemptions from the dispositional hearing requirement. HHS regulations exempt children who are placed in a “court sanctioned permanent foster family home placement with a specific care giver” or children who “are free for adoption and are placed in adoptive homes pending the finalization of the adoption.” 45 C.F.R. § 1356.21(e)(l)-(2) (1992). Insisting that these seven children were either in permanent foster family homes or were waiting for adoption finalization, Connecticut asserts that HHS had unreasonably determined that the seven cases were unacceptable. Again, we disagree. Though these children had resided with foster care families for a prolonged period of time, HHS concluded that no court had ever made a determination to permanently place them with these families. As the preamble to HHS regulations states, a court must determine “that the child should remain permanently in foster care with a specified foster family.” 45 Fed.Reg. 23,109 (1983). None of these children, moreover, were placed in adoptive homes pending finalization of adoption. As these children were not exempt from the dispositional hearing requirement and dispositional hearings either never occurred or were untimely, we agree that these seven cases were unacceptable as well. Connecticut next asserts that HHS had unreasonably decided that nine cases failed because the required periodic reviews had not taken place for FY 1985. The State claims that these children were somehow exempt from the periodic review requirements. There is, however, no statutory or regulatory exemption from the periodic review requirement. Though there is a limited regulatory exemption from the statutory dispositional hearing requirement, no analogous exemption from the periodic review requirement exists. HHS properly found these nine records unacceptable. Though GAB found that 30 cases had failed to satisfy the Act’s requirements, Connecticut only challenged GAB determinations in 28 of those cases in the district court. By *987 our reckoning, Connecticut, in failing to convince us that 23 of the eases were improperly found wanting, 5 and in conceding that 2 cases had failed, has at most achieved a compliance rate of 67%, well below the 80% necessary to challenge PI 85-02. 6 Since Connecticut lacks standing to challenge the program instruction, we dismiss the appeal. 7 Accordingly, we affirm the decision below. So ordered. Footnotes [1] Sequential sampling conserves resources by permitting inspectors to examine fewer case records. Rather than review all records in a sample, inspectors review records until the cumulative results are sufficient to judge whether a state has met or failed to meet the compliance standard. [2] HHS had issued PI 85-02 in the middle of FY 1985 (January 1985) and then had applied the policy expressed therein to FY 1985 case records. [3] It is not clear to us that it is proper to characterize the new program instructions application as “retroactive" within the meaning of Bowen. See 488 U.S. at 208-09, 109 S.Ct. at 472. It is hard to imagine how Connecticut's conduct would have been affected even had it known of the “new" 90% compliance standard earlier. After all, Connecticut could never know which cases would end up in a sample, and, therefore, the level of care necessary to ensure a successful audit requiring a 90% compliance rate does not seem meaningfully different than one that would ensure compliance at the 80% rate. [4] Even a policy statement that does not have "legal effect” may contribute to a court's conclusion that an agency’s inconsistent positions are arbitrary and capricious. [5] Note that some cases were found unacceptable for numerous reasons and thus were challenged on numerous grounds. [6] We recognize that sequential sampling provides states with a "cushion” of unacceptable cases. For instance, with 76 cases, HHS must determine that 17 cases are unacceptable, much below the 90% mark, before concluding that the states have failed. Similarly, when dealing with an 80% compliance rate, the states would have some leeway. In other words, at 76 cases, Connecticut would be able to have more than 15 cases unacceptable, and the sequential sampling would continue. Connecticut, though, challenging the case-specific determinations of HHS and the issuance of PI 85-02, did not raise this issue. Suffice it to say that where 25 of Connecticut's cases were unacceptable, in a sample of 76, we are confident that Connecticut did not reach the 80% standard in PI 82-06. [7] We thus need not consider those cases challenged by Connecticut on miscellaneous grounds. Even were we to decide that HHS determinations were arbitrary and capricious in all of those cases, Connecticut would still fail to establish an 80% compliance rate. Therefore, the State would still lack standing.