On January 1, 1994, Indiana changed the formulas it uses to pay providers of services under the Medicaid program. For the most part, these rules were short-lived; with the exception of the rules governing outpatient services, all were replaced within a year. But the litigation in their wake is long-lived. Joined by five physicians, the Methodist Hospitals of Gary, Indiana, filed this suit under 42 U.S.C. § 1983, contending that the new rules were invalid. See Maine v. Thiboutot, 448 U.S. 1, 100 S. Ct. 2502, 65 L. Ed. 2d 555 (1980). The district court dismissed some claims under Rule 12(b)(6), holding that the statute does not create a private right of action in plaintiffs' favor, and granted summary judgment to defendants on the rest. Its principal opinion is reported at 860 F. Supp. 1309 (1994); there are several unpublished orders. Victory emboldened Indiana to seek attorneys' fees under 42 U.S.C. § 1988, which the district court declined to award. Collateral disputes also broke out. When plaintiffs sought a temporary restraining order, defendants filed a brief containing confidential information. The district court sealed this brief; defendants want us to unseal it. The parties also disagree about whether the case is moot, which may depend on the meaning of the eleventh amendment. To prevent matters from getting out of hand, we simplify throughout this opinion.

First comes the question whether the parties have a live controversy. Two of the three sets of rules were defunct before the district court decided the case, which neither the judge nor plaintiffs' appellate brief mentions. We learned this for the first time when encountering the state's argument that we should dismiss the proceedings as moot. For its part, the state relegated the outpatient rules to a footnote, as did plaintiffs' reply brief--which contends, despite the eleventh amendment, that the possibility of a monetary recovery forestalls mootness. Neither side tells us anything of value about the outpatient rules--even what they do, let alone whether there is an ongoing controversy about their validity.

According to the district court, until 1994 Indiana reimbursed providers for outpatient services at the provider's customary billing amount, but not to exceed 100% of the provider's actual cost. The new approach pays all providers in the state the same amount for the same service. Following the approach used for Medicare claims, Indiana established nine categories of outpatient services. It pays each provider the sum of 50% of the Medicare rate for that service and 50% of the statewide median amount paid for that service in 1992. Providers with higher-than-average costs (or located where private patients have higher-than-average willingness to pay) thus do worse than before, and providers that have (or can achieve) lower costs of service do well. The Hospitals believe that they are in the former group. An affidavit by the Hospitals' chief financial officer estimates that the outpatient rules will cost the Hospitals $1 million per year compared with the former method. This is only a prediction made before the amendments took effect; the record does not reveal what has happened, and the prediction may be all wet. But because the case was dismissed on the pleadings, we must assume that the Hospitals could prove their loss, see Lujan v. Defenders of Wildlife, 504 U.S. 555, 561, 112 S. Ct. 2130, 2136, 119 L. Ed. 2d 351 (1992), which means not only that they have standing to sue but also that the case survives, for a federal court could enjoin operation of the outpatient rules even if the eleventh amendment forbids an award of damages. Because the Hospitals lose on the merits concerning the outpatient rules and do not contend that they have better arguments for the other sets of rules, it turns out to be unnecessary to decide whether the eleventh amendment would have blocked monetary relief on the other two sets.

Plaintiffs' many avenues of assault on the state's rules are for the most part unsupported by argument. For example, plaintiffs insist that the district court "erred in concluding that the State complied with the regulations mandating consultation with the Medical Advisory Committee before the promulgation and implementation of Medicaid Rules" but their brief does not so much as cite a regulation, let alone tell us why the regulation prohibits Indiana's procedures. Defendants tell us that the regulation in question is 42 C.F.R. § 431.12(e), and that several courts of appeals have held that Indiana's approach to consultation is permissible. E.g., Burgess v. Affleck, 683 F.2d 596, 599-600 (1st Cir.1982); Mississippi Hospital Association v. Heckler, 701 F.2d 511, 520, 523 (5th Cir.1983). We see no need to delve into the subject; plaintiffs have not amplified their arguments sufficiently to present an issue for appellate resolution. Similarly, plaintiffs' contention that Indiana's rules violate 42 U.S.C. § 1396a(a)(10), which they chastise the district court for ignoring, not only is undeveloped but also disregards the fact that the district court did address it--and dismissed the plaintiffs' contentions on the ground that there is no private right of action to enforce § 1396a(a)(10). 860 F. Supp. at 1333. Plaintiffs' reply brief at last gets 'round to this question. Too late.

The only contention plaintiffs have properly presented in this forum is that Indiana's rules do not comply with 42 U.S.C. § 1396a(a)(30). This statute says that every Medicaid plan must

provide such methods and procedures relating to the utilization of, and the payment for, care and services available under the plan ... as may be necessary to safeguard against unnecessary utilization of such care and services and to assure that payments are consistent with efficiency, economy, and quality of care and are sufficient to enlist enough providers so that care and services are available under the plan at least to the extent that such care and services are available to the general population in the geographic area[.]

Plaintiffs stress the state's obligation to offer payments sufficient "to enlist enough providers so that care and services are available under the plan at least to the extent that such care and services are available to the general population in the geographic area". The district court dismissed this claim after finding that the words "geographic area" are so ambulatory that the section does not create a private right of action. 860 F. Supp. at 1331-33. Although Arkansas Medical Society v. Reynolds, 6 F.3d 519, 523-28 (8th Cir.1993), holds that providers have a private right of action to enforce § 1396a(a)(30), the district court declined to follow suit, observing that Arkansas Medical Society had not discussed the ambiguity created by "in the geographic area"--an ambiguity the court thought fatal under Suter v. Artist M., 503 U.S. 347, 112 S. Ct. 1360, 118 L. Ed. 2d 1 (1992).

"Geographic area" could mean many things, depending on what function the boundary serves. If the point of § 1396a(a)(30) is to ensure that every Medicaid recipient can obtain every medical service within walking distance, then the statute places severe constraints on a state plan, but if the law means only that every Standard Metropolitan Statistical Area has to have ample medical services (so that care is available within the reach of public transit systems), then Indiana is home free. Gary, Indiana, is in the Chicago SMSA, which offers medical services galore. How large a geographic area is appropriate may depend on the kind of medical service: emergency care must be swift, and hence close, but longer travel times are tolerable when obtaining outpatient care for chronic conditions. The district judge observed that the Department of Health and Human Services has not issued regulations defining "geographic area" and thought that this is not an appropriate task for a court to tackle unassisted. This position has some appeal, but it is not clear what rule-writers could do if they tried. For more than a century courts and agencies have wrestled with the concept of the "geographic market" in antitrust law without producing a mechanical definition. Defining geographic markets for medical care has proven no more tractable than geographic markets in general, but courts soldier on. Compare United States v. Rockford Memorial Corp., 898 F.2d 1278 (7th Cir.1990), with FTC v. Freeman Hospital, 69 F.3d 260 (8th Cir.1995). Cf. Blue Cross & Blue Shield United of Wisconsin v. Marshfield Clinic, 65 F.3d 1406 (7th Cir.1995); Ball Memorial Hospital, Inc. v. Mutual Hospital Insurance, Inc., 784 F.2d 1325 (7th Cir.1986).

Of course the antitrust laws create express rights of action, while implementation of the Medicaid program is entrusted principally to the Secretary and the states. But Wilder v. Virginia Hospital Association, 496 U.S. 498, 110 S. Ct. 2510, 110 L. Ed. 2d 455 (1990), holds that hospitals have a private right of action to enforce the Boren Amendment, 42 U.S.C. § 1396a(a)(13)(A), a provision that among other things requires state Medicaid plans to "assure that individuals eligible for medical assistance have reasonable access (taking into account geographic location and reasonable travel time) to inpatient hospital services of adequate quality". The Court did not address this source of ambiguity, but Wilder's holding binds us nonetheless. Suter distinguished but did not overrule Wilder. We therefore follow Arkansas Medical Society and hold that providers of medical care have a private right of action, derived through § 1983, to enforce § 1396a(a)(30).

Which does plaintiffs no good, because their substantive arguments do not show any failing in Indiana's plan. Plaintiffs read § 1396a(a)(30) to require comprehensive studies prior to any change in a state's plan of reimbursement--studies that would put an environmental impact study to shame and make it all but impossible for a state to amend its plan. For what plaintiffs demand is nothing less than a complete description of supply and demand schedules for every medical specialty in every part of the state, so that before changing reimbursement rates a state knows exactly what effect the new rate will have on the demand for, and supply of, medical care. How else, plaintiffs ask, can a state assure that every Medicaid recipient can find all essential care nearby? Problem: as the Supreme Court recognized in Illinois Brick Co. v. Illinois, 431 U.S. 720, 97 S. Ct. 2061, 52 L. Ed. 2d 707 (1977), it is exceptionally difficult to determine demand and supply schedules for a single product. See also Kansas v. Utilicorp United, Inc., 497 U.S. 199, 110 S. Ct. 2807, 111 L. Ed. 2d 169 (1990). Doing this for the entire medical segment of the economy would be more than difficult; it would be impossible. A state could send out a survey, but questions such as "Tell us the minimum amount you would accept without withdrawing from the market" would not elicit honest answers. People often do not even know their reservation prices; they do not willingly reveal them.

Nothing in the language of § 1396a(a)(30), or any implementing regulation, requires a state to conduct studies in advance of every modification. It requires each state to produce a result, not to employ any particular methodology for getting there. Section 1396a(a)(30) is in this respect unlike the Boren Amendment, which requires the state to adopt rules that "the State finds, and makes assurances satisfactory to the Secretary, are reasonable and adequate" to achieve identified objectives. The Court concluded in Wilder that a state could not make the necessary findings and assurances without conducting a study. See also Illinois Health Care Association v. Bradley, 983 F.2d 1460 (7th Cir.1993). Under § 1396a(a)(30), however, states may behave like other buyers of goods and services in the marketplace: they may say what they are willing to pay and see whether this brings forth an adequate supply. If not, the state may (and under § 1396a(a)(30), must) raise the price until the market clears.

This is what Indiana has done. On January 1, 1994, it reduced the price it offered for medical services--although not by much. (Recall that it used statewide averages of going prices as benchmarks.) Then it launched studies to find out whether the new prices elicited enough medical care. Plaintiffs deride these as after-the-fact paperwork, which they were from the perspective of January 1, 1994, but they are before-the-fact inquiries for the changes that followed. On October 6, 1994, Indiana modified the reimbursement system for physicians' services, 18 Ind.Reg. 87 (Oct. 1, 1994), and on November 4 it implemented a revised reimbursement system for inpatient hospital services, 18 Ind.Rev. 243 (Nov. 1, 1994). Indiana must have made satisfactory choices based on what it learned, for plaintiffs do not complain about these rules. The reimbursement system for outpatient hospital services was left as is, and although, as we have said, this staves off mootness, plaintiffs do not assert that they have withdrawn from the outpatient market as a result. Indeed, they have not offered to show that any provider withdrew, anywhere in the state. Plaintiffs offered dire predictions, but Indiana used 1994 to check predictions against reality. This approach seems to us sound. Whether or not a better way could be devised, it was a lawful way to proceed.

This conclusion sets up the state's argument that it is entitled to attorneys' fees under § 1988. Indiana acknowledges that the Supreme Court treats § 1988 as slanted in plaintiffs' favor: a prevailing plaintiff recovers legal fees as of course, but a prevailing defendant recovers fees only if the suit was "groundless or without foundation." Hughes v. Rowe, 449 U.S. 5, 14, 101 S. Ct. 173, 178, 66 L. Ed. 2d 163 (1980). See also Christiansburg Garment Co. v. EEOC, 434 U.S. 412, 98 S. Ct. 694, 54 L. Ed. 2d 648 (1978); Newman v. Piggie Park Enterprises, Inc., 390 U.S. 400, 88 S. Ct. 964, 19 L. Ed. 2d 1263 (1968). According to Indiana, Fogerty v. Fantasy, Inc., 510 U.S. 517, 114 S. Ct. 1023, 127 L. Ed. 2d 455 (1994), casts a new light on § 1988. Fogerty held that language in the copyright act materially identical to § 1988 does not permit a court to use one standard for prevailing plaintiffs and a different standard for prevailing defendants. Defendants ask us to adopt a new rule under which plaintiffs who file suit to promote their "commercial interests" must reimburse prevailing defendants. Money that Indiana pays to lawyers to defend litigation such as this is unavailable to provide medical benefits to indigents. But it may not be so easy to distinguish "commercial" from other interests; higher levels of reimbursement mean higher take-home pay for physicians, but also extra care for needy patients--and at all events Fogerty does not support a distinction between commercial and altruistic litigation. We have been told to apply existing precedents while leaving to the Supreme Court the question whether other decisions have undermined their authority. Rodriguez de Quijas v. Shearson/American Express, Inc., 490 U.S. 477, 484, 109 S. Ct. 1917, 1921, 104 L. Ed. 2d 526 (1989), Thurston Motor Lines, Inc. v. Jordan K. Rand, Ltd., 460 U.S. 533, 535, 103 S. Ct. 1343, 1344, 75 L. Ed. 2d 260 (1983), and Hutto v. Davis, 454 U.S. 370, 102 S. Ct. 703, 70 L. Ed. 2d 556 (1982). The district court denied the state's request for attorneys' fees, as Hughes requires. If Indiana thinks that Fogerty calls for a different answer, it must seek relief from the authors of that opinion.

The parties' final dispute, about the sealed brief opposing plaintiffs' motion for a TRO, likewise may be resolved without visiting first principles. To expedite discovery, the parties agreed that sensitive information would be turned over in confidence. A protective order provided that the district judge could order any of this information put in the public record on application of either side. Indiana's brief opposing plaintiffs' motion for a TRO included the five physicians' incomes, which had been revealed under this procedure, so it was sealed pending resolution of the state's request that the judge authorize the disclosures. Indiana later assured the court that it would not apply its new reimbursement rules to plaintiffs until the court decided the suit on the merits; the judge therefore did not rule on the request for a TRO. The principle that materials on which a judicial decision rests are presumptively in the public domain, In re Continental Illinois Securities Litigation, 732 F.2d 1302 (7th Cir.1984); cf. Grove Fresh Distributors, Inc. v. Everfresh Juice Co., 24 F.3d 893 (7th Cir.1994), therefore is not in play.

Defendants stress the common law right of access to documents filed in litigation. See Press-Enterprise Co. v. Superior Court, 478 U.S. 1, 106 S. Ct. 2735, 92 L. Ed. 2d 1 (1986); Nixon v. Warner Communications, Inc., 435 U.S. 589, 597-99, 98 S. Ct. 1306, 1311-13, 55 L. Ed. 2d 570 (1978); In re Reporters Committee for Freedom of the Press, 773 F.2d 1325, 1331-33 (D.C.Cir.1985) (Scalia, J.). Plaintiffs observe that much commercial information is confidential (think of trade secrets) and that access to the courts should not be conditioned on revealing one's secrets to the general public. Both sides are right. Judicial proceedings are presumptively open, but particular information may be withheld. Personal income is among the categories that can be withheld. Disclosure of tax returns is highly restricted, see Hrubec v. National Railroad Passenger Corp., 49 F.3d 1269 (7th Cir.1995), and we do not think that the disclosure of the same information becomes compulsory whenever a state decides to put it in a brief. Its inclusion in the state's memorandum is gratuitous.

Let us suppose that the five physician plaintiffs all enjoy substantial incomes; of what relevance is this? Medicaid is not rate-of-return regulation. Plaintiffs do not allege a "taking" or adopt any other theory that makes their income salient. The substantive question is not, as Indiana professes to believe, whether its reimbursement rules would drive physicians out of the medical profession and induce them to take up architecture or accounting. It is whether a reduction in what they can make from Medicaid patients in Gary would lead them to change the character or location of their practices--perhaps closing their doors to Medicaid patients in favor of those who have private insurance, perhaps moving to Illinois or another state that offers more generous remuneration. The economic question, in other words, is how the rules affect physicians on the margin, and whether they would choose to avoid that effect by altering their practice. This question, about physicians' reservation wages, is not illuminated by information about their annual gross or net incomes.

But it does not follow, as the physicians suppose, that the district court properly ordered the state's memorandum to remain under seal. To say that particular information is confidential is not to say that the entire document containing that information is confidential. Cases under the Freedom of Information Act stress that the confidential material should be redacted and the remainder of the document disclosed, in order to afford access to information in which the public is legitimately interested. The same principle, and for the same reason, applies to information created or turned up in litigation. See In re Krynicki, 983 F.2d 74 (7th Cir.1992) (Easterbrook, J., in chambers); cf. Pepsico, Inc. v. Redmond, 46 F.3d 29 (7th Cir.1995) (Easterbrook, J., in chambers).

The district court therefore should not have denied defendants' motion to unseal their memorandum. The court instead should have ordered defendants to redact the confidential information and then should have placed the remainder of the document in the public record. The case is remanded so that this may be accomplished, but the judgment is otherwise affirmed.