Texas v. New Mexico, 602 U.S. ___ (2024)

NOTICE: This opinion is subject to formal revision before publication in the United States Reports. Readers are requested to notify the Reporter of Decisions, Supreme Court of the United States, Washington, D. C. 20543, pio@supremecourt.gov, of any typographical or other formal errors.

SUPREME COURT OF THE UNITED STATES

_________________

No. 141, Orig.

_________________

TEXAS, PLAINTIFF v. NEW MEXICO AND COLORADO

on exception to third interim report of the special master

[June 21, 2024]

Justice Jackson delivered the opinion of the Court.

The Rio Grande River begins in Colorado, flows through New Mexico into Texas, and then courses along the Texas-Mexico border. The Rio Grande Compact (Compact)—an interstate agreement between Colorado, New Mexico, and Texas—governs the “equitable apportionment” of the waters of the Rio Grande among those three States. To ensure that Texas receives its share of water, the Compact relies on the United States Bureau of Reclamation to operate the Rio Grande Project, an irrigation system in southern New Mexico.

In 2013, Texas filed suit against the other two signatory States, alleging that, in violation of the Compact, excessive groundwater pumping in New Mexico was depleting supplies of Rio Grande water bound for Texas. The United States sought to intervene, and in a decision we issued six Terms ago, we allowed it to do so. See Texas v. New Mexico, 583 U.S. 407 (2018). In our opinion, we explained that the Federal Government has its own distinct interests in holding New Mexico to its obligations under the Compact, as the Compact is “inextricably intertwined” with the United States’ operation of the Rio Grande Project. Id., at 413.

Now, Texas and New Mexico have agreed to a proposed consent decree that would resolve this case and codify a methodology for determining each State’s allocation of the Rio Grande’s waters. But the United States opposes the proposed consent decree, contending that it would dispose of the Federal Government’s claims that New Mexican groundwater pumping is violating the Compact.

We agree with the United States. “[P]arties who choose to resolve litigation through settlement may not dispose of the claims of a third party.” Firefighters v. Cleveland, 478 U.S. 501, 529 (1986). The United States still advances the same claims as it did in 2018, backed by the same unique federal interests we identified then. Through the consent decree, the States would settle all parties’ Compact claims and, in the process, cut off the United States’ requested relief as to New Mexican groundwater pumping. Because our precedent does not permit that result, the States’ motion to enter the consent decree is denied.

I

A

The Rio Grande springs from the San Juan Mountains just east of the Continental Divide in southwestern Colorado. After tumbling out of the Rocky Mountains, the river cuts south through the deserts of New Mexico before crossing into Texas near the city of El Paso. From there, the river snakes its way southeast, marking the border between the United States and Mexico and eventually spilling into the Gulf of Mexico at the city of Brownsville, Texas.

Of course, when a river touches so many jurisdictions, disputes about water rights are bound to follow. The Rio Grande is no exception. In the late 19th century, Mexico began to voice concerns about water shortages caused by increased use of the Rio Grande’s upstream waters in the United States. See National Resources Committee, Regional Planning: Part VI—The Rio Grande Joint Investigation in the Upper Rio Grande Basin in Colorado, New Mexico, and Texas, 1936–1937, pp. 7–8 (1938). In 1906, the United States and Mexico settled that dispute and entered into a treaty, with the United States promising to provide Mexico 60,000 acre-feet of Rio Grande water each year. See Convention Between the United States and Mexico Providing for the Equitable Distribution of the Waters of the Rio Grande for Irrigation Purposes, May 21, 1906, 34 Stat. 2953, T. S. No. 455. To deliver on that promise, the United States needed to harness the river’s irregular ebb and flow brought on by alternating dry spells and floods. Accordingly, the Federal Government resolved to construct a new dam and reservoir at Elephant Butte in New Mexico, about 100 miles north of the Texas-New Mexico border. Among the first irrigation projects authorized by the Reclamation Act of 1902, the dam and reservoir constituted an essential component of the new Rio Grande Project, an irrigation system implemented by the United States Bureau of Reclamation (Reclamation). See Act of Feb. 25, 1905, ch. 798, 33 Stat. 814.

Thanks to the Rio Grande Project, the United States had harnessed the Rio Grande’s water. But that raised another question: What to do with it? Enter the “Downstream Contracts,” a series of agreements between the United States and two irrigation districts in New Mexico and Texas. First signed in 1906 and later renegotiated in the 1930s, the Downstream Contracts provided that, after allocating Mexico’s share of Rio Grande water under the 1906 Treaty, the United States would deliver apportionments of water to the two political subdivisions—the Elephant Butte Irrigation District in New Mexico (EBID) and El Paso County Water Improvement District No. 1 in Texas (EP1). Specifically, Reclamation agreed to supply water to 88,000 irrigable acres in EBID and 67,000 irrigable acres in EP1, amounting to shares of about 57% and 43% of the reserved water, respectively. Letter from S. Somach to Special Master, p. 36 (May 8, 2018).

That left the competing water-rights claims of Colorado, New Mexico, and Texas. To resolve that dispute, those States looked to the U. S. Constitution’s Compact Clause, which permits States to enter into agreements among themselves, with the consent of Congress. Art. I, §10, cl. 3. While contractual in nature, an interstate compact “ ‘is not just a contract,’ but also ‘a federal statute enacted by Congress’ that preempts contrary state law.” New York v. New Jersey, 598 U.S. 218, 224 (2023) (quoting Alabama v. North Carolina, 560 U.S. 330, 351 (2010)). Once Congress gives its stamp of approval, an interstate compact becomes the law of the land, much like any other federal statute.

In 1938, with Congress’s endorsement, Colorado, New Mexico, and Texas agreed to the Rio Grande Compact, which “effect[ed] an equitable apportionment” of the Rio Grande’s waters among the three States. Act of May 31, 1939, 53 Stat. 785. For the upstream States, the Compact imposed certain delivery obligations. It required Colorado to deliver a particular amount of water to the New Mexican border. Id., at 787–788. “But then, instead of similarly requiring New Mexico to deliver a specified amount of water annually to the Texas state line, the Compact directed New Mexico to deliver water to the [Elephant Butte] Reservoir.” Texas, 583 U. S., at 410–411. That “choice made all the sense in the world in light of the simultaneously negotiated Downstream Contracts that promised Texas water districts a certain amount of water every year from the Reservoir’s resources.” Id., at 411. In other words, the Compact relied on Reclamation to apportion water through its contractual obligations to EBID and EP1.

Although the Rio Grande’s waters were plentiful in the 1930s, drought conditions set in beginning in the late 1940s and early 1950s. As a result, entities in southern New Mexico below the Elephant Butte Reservoir began pumping groundwater at increasing levels to support local agriculture. That groundwater pumping had important hydrological implications for the Rio Grande Project.

Here’s why: When Reclamation releases water from Elephant Butte, the water flows into the bed of the Rio Grande, and then to a series of canals and ditches, eventually reaching irrigated farms, its final destination. Some of the water runs off of the fields or percolates into the ground, returning to the river through drains or seepage. Due to these “return flows,” water trickles back to the Rio Grande riverbed, where it proceeds farther downstream to other irrigation destinations. But groundwater pumping in southern New Mexico interrupts that process, both by drawing water away from the river and by intercepting the return flows that would otherwise replenish it. Put simply, the more groundwater pumping between the Elephant Butte Reservoir and Texas, the more water Reclamation has to release from the reservoir to comply with its delivery obligations.

Reclamation dealt with these changing circumstances by developing an equation known as the D2 Curve. Using Project data from 1951 to 1978—the so-called D2 Period that witnessed New Mexico’s ramped-up groundwater pumping—Reclamation devised a linear regression model to help it predict how much water would be available to EBID and EP1 based on a given release of water from the Elephant Butte Reservoir.

The extent of groundwater pumping in New Mexico nonetheless remained a point of contention, and in 2013, Texas filed an original action in this Court against New Mexico.[1] Among other things, Texas alleged that New Mexico was violating the Compact by permitting local entities to pump groundwater at levels exceeding those contemplated in 1938, intercepting water bound for the Lone Star State. Texas requested declaratory, injunctive, and monetary relief, including an injunction commanding New Mexico to cease all interference with the United States’ operation of the Rio Grande Project.

The United States sought to intervene in Texas’s suit and filed its own complaint in 2014. Like Texas, the Federal Government took issue with New Mexico’s groundwater pumping, explaining that excessive water interception below Elephant Butte could reduce Project efficiency “to a point where 43% of the available water could not be delivered to [EP1], and 60,000 acre-feet per year could not be delivered to Mexico.” Intervening Complaint 4, ¶15. For relief, the United States sought a declaration and an injunction requiring New Mexico to stop in-state entities from interfering with the Project’s delivery of water to EBID, EP1, and Mexico. Id., at 5.

The Special Master appointed to adjudicate this case recommended dismissing the United States’ complaint. But this Court allowed the United States to intervene. Specifically, we held that “the federal government has an interest in seeing that water is deposited in the [Elephant Butte] Reservoir consistent with the Compact’s terms,” as that “is what allows the United States to meet its duties under the Downstream Contracts, which are themselves essential to the fulfillment of the Compact’s expressly stated purpose.” Texas, 583 U. S., at 414.

B

The litigation continued. After the Special Master denied summary judgment and held the first phase of trial, Texas and New Mexico negotiated a proposed consent decree. The consent decree would make “[c]ompliance with th[e] Decree” sufficient to show “compliance with the Compact with respect to the division of Rio Grande water below Elephant Butte Reservoir.” Third Interim Report of the Special Master Addendum 8, ¶7 (Third Interim Report Addendum).

The centerpiece of the proposed consent decree would be the establishment of the Effective El Paso Index (EEPI), a new method of determining the allotment of Rio Grande water New Mexico must deliver downstream into Texas. The EEPI’s calculations of water allocations would be based on conditions during the D2 Period, when New Mexico was actively depleting return flows through groundwater pumping. That is, the EEPI would permit levels of pumping “reflected in the 1951–1978 timeframe rather than [requiring] a strict return to a pumping condition as existed in 1938.” Third Interim Report 75. The EEPI would then rely on the El Paso Gage, a flow indicator near the New Mexico-Texas border, to measure New Mexico’s delivery of water into Texas. Finally, the consent decree would require Reclamation to transfer water between EBID and EP1 as needed to maintain a specified allotment.

The States moved the Special Master to approve the proposed consent decree, but the United States objected. As relevant here, the United States maintained that the consent decree would impermissibly dispose of its Compact claims without its consent. The Special Master disagreed, however, and issued a Third Interim Report recommending that this Court approve the consent decree. The United States filed an exception to the report, and we set the case for argument.

II

A consent decree “embodies an agreement of the parties and thus in some respects is contractual in nature.” Rufo v. Inmates of Suffolk County Jail, 502 U.S. 367, 378 (1992). But it is also “an agreement that the parties desire and expect will be reflected in, and be enforceable as, a judicial decree.” Ibid.

In Firefighters v. Cleveland, 478 U.S. 501, we described the rules that apply when parties wish to settle via consent decree over the objection of a nonconsenting intervenor. “[W]hile an intervenor is entitled to present evidence and have its objections heard . . . on whether to approve a consent decree,” it generally cannot block a decree that would settle the other parties’ claims “merely by withholding its consent.” Id., at 529.

That rule does not apply, however, when the parties’ settlement would also affect the intervenor’s claims. Under those circumstances, parties “who choose to resolve litigation through settlement may not dispose of the claims of a third party . . . without that party’s agreement.” Ibid. In other words, a “court’s approval of a consent decree between some of the parties . . . cannot dispose of the valid claims of nonconsenting intervenors; if properly raised, these claims remain and may be litigated by the intervenor.” Ibid.

Consequently, and as we explained 20 years before Firefighters, “where the Government seeks an item of relief to which evidence adduced at trial may show that it is entitled, the [court] may not enter a ‘consent’ judgment without the actual consent of the Government.” United States v. Ward Baking Co., 376 U.S. 327, 334 (1964).

III

With these legal rules in mind, we must now decide whether to approve the States’ proposed consent decree over the Federal Government’s objection. The relevant questions under our precedents are whether the United States has valid Compact claims and whether the proposed consent decree would dispose of those claims. Because the answer to each of those questions is yes, the consent decree cannot be approved without the United States’ consent.

A

1

Conventional wisdom posits that, because time changes all things, no one can step into the same river twice. This case may be an exception, though, for the same considerations that convinced us to let the United States intervene six Terms ago also lead us to conclude that the United States still has valid Compact claims today.

In 2014, the United States asked to intervene in this action, asserting “essentially the same claims Texas already” pleaded. Texas, 583 U. S., at 409. Namely, the United States alleged that New Mexico was impermissibly “siphon[ing] off water below the Reservoir in ways the Downstream Contracts do not anticipate.” Id., at 411. The Special Master recommended that we dismiss the United States’ complaint, reasoning “that the Compact does not confer on the United States the power to enforce its terms.” Ibid. But in its exception to that report, the United States maintained that “it may pursue claims for violations of the Compact itself.” Id., at 412.

We agreed with the United States. Although interstate compacts are (as the name suggests) agreements between States, “we have sometimes permitted the federal government to participate in compact suits to defend ‘distinctively federal interests’ that a normal litigant might not be permitted to pursue in traditional litigation.” Id., at 412–413 (quoting Maryland v. Louisiana, 451 U.S. 725, 745, n. 21 (1981)). Examining the nature of the United States’ claims and the Rio Grande Project’s unique relationship to the Compact, we ticked through “several considerations” persuading us that the United States “may pursue the particular claims it has pleaded in this case.” 583 U. S., at 413.

First, “the Compact is inextricably intertwined with the Rio Grande Project and the Downstream Contracts,” both carried out by the Federal Government. Ibid. The purpose of the Compact, recall, was to “ ‘effec[t] an equitable apportionment’ ” of the Rio Grande’s waters among the signatory States. Ibid. (alteration in original). But it “can achieve that purpose only because, by the time the Compact was executed and enacted, the United States had negotiated and approved the Downstream Contracts, in which it assumed a legal responsibility to deliver a certain amount of water to Texas.” Ibid. The United States, therefore, “might be said to serve, through the Downstream Contracts, as a sort of ‘agent’ of the Compact,” responsible for ensuring Texas and New Mexico receive their apportionments. Ibid. (some internal quotation marks omitted). Or, put another way, “the Compact could be thought implicitly to incorporate the Downstream Contracts by reference.” Ibid. “However described,” the bottom line was that the “federal government has an interest in seeing that water is deposited in the Reservoir consistent with the Compact’s terms.” Id., at 414. And although running parallel with Texas’s asserted interests, the United States’ interest was “distinctively federal.” Id., at 413 (internal quotation marks omitted). If New Mexico interfered with the Project, for instance, Reclamation might prove unable “to meet its duties under the Downstream Contracts, which are themselves essential to the fulfillment of the Compact’s expressly stated purpose.” Id., at 414 (emphasis added).

Second, along similar lines, we stressed that New Mexico had “conceded that the United States plays an integral role in the Compact’s operation” and so had its own interests in this litigation. Ibid. Specifically, New Mexico had argued that the Federal Government was “an indispensable party” because it was “ ‘responsible for . . . delivery of . . . water’ as required by the Downstream Contracts and anticipated by the Compact.” Ibid. (quoting New Mexico’s Brief in Opposition to Texas’ Motion for Leave to File Complaint 33 (Mar. 11, 2013) (2013 BIO); alterations in Texas). For that reason, the “ ‘entry of a Decree in accordance with Texas’ Prayer for Relief would necessarily affect the United States’ interests.’ ” 583 U. S., at 414 (quoting 2013 BIO 33; emphasis added).

Third, we also took note of the Federal Government’s obligations under the 1906 Treaty. As explained above, the United States must deliver 60,000 acre-feet of water from the Elephant Butte Reservoir to Mexico, but the United States can “fill that Reservoir” only if New Mexico complies with its obligation “to deliver a specified amount of water to the facility.” 583 U. S., at 414. Thus, the United States’ ability to deliver water to Mexico depends on New Mexico’s compliance with “its Compact obligations,” and “a breach of the Compact could jeopardize the federal government’s ability to satisfy its treaty obligations.” Ibid. “Permitting the United States to proceed” with its own Compact claims would “allow it to ensure that those obligations are, in fact, honored.” Id., at 415.[2]

In light of these “ ‘distinctively federal interests,’ ” we held that the United States could validly claim that New Mexico was “effectively breaching its Compact duty to deliver water to the Reservoir.” Id., at 411, 413. Our 2018 decision leads inexorably to the same conclusion today: The United States has its own, uniquely federal claims under the Compact. If it did not, one might wonder why we permitted the Federal Government to intervene in the first place.

2

Our 2018 decision is also all but dispositive of the States’ arguments that the United States lacks valid Compact claims today.

For starters, the States contend that the United States has no valid Compact claims because it does not itself receive an apportionment of water under the Compact. Joint Reply to Exception of the United States by the State of Texas et al. 29–31 (Joint Reply). But the United States did not receive an apportionment of Rio Grande water in 2018 any more than it does now. Rather, as we explained, its claims arise from the Compact’s incorporation of the Downstream Contracts and the attendant risk that New Mexico’s interference with the Project could leave Reclamation unable to meet its contractual and treaty obligations.

The States and the dissent also assert that the United States failed to allege a “1938 baseline”—a shorthand for the claim that New Mexico’s groundwater pumping should be restricted to levels in effect when the Compact was enacted. See Joint Reply 36–37; post, at 18–24, and nn. 2–3 (opinion of Gorsuch, J.). But that argument, too, is foreclosed by our prior decision. There, we explained that Texas had alleged New Mexico was “breaching its Compact duty” by allowing downstream water “users to siphon off water . . . in ways the Downstream Contracts do not anticipate.” Texas, 583 U. S., at 411; see Texas’s Complaint 10, ¶18 (alleging that current pumping “changed the conditions that existed in 1938”). And we recognized that the United States asserted “essentially the same claims Texas already has.” Texas, 583 U. S., at 409; see id., at 411 (United States’ claims “parallel Texas’s”); id., at 415 (United States seeks “substantially the same relief ” as Texas). Whether the United States’ complaint uses the term “1938 baseline” is beside the point. Both Texas and the United States pleaded that New Mexico was violating the Compact by pumping more groundwater than the Compact contemplates, and that is still the claim that the United States wishes to pursue now.[3]

Last, we are not persuaded by the States’ reliance on our decisions in Hinderlider v. La Plata River & Cherry Creek Ditch Co., 304 U.S. 92 (1938), and California v. United States, 438 U.S. 645 (1978). The States maintain that they alone represent EBID’s and EP1’s interests in an allocation of Compact water; accordingly, they say, any interest Reclamation has in fulfilling the Downstream Contracts is strictly derivative of the States’ interest in how the water is apportioned. Joint Reply 31–36; see post, at 14, 18. For support, they rely on Hinderlider, which held that a Colorado ditch company had no right to water that the State of Colorado had agreed to apportion to New Mexico under the La Plata River Compact. 304 U. S., at 106–108. “[T]he States,” we explained, “had power to bind by compact their respective appropriators,” id., at 108, notwithstanding the ditch company’s pre-existing right under Colorado law to a certain apportionment of water, id., at 98.

The States’ argument here fails for at least two reasons. First, our decision in 2018 is incompatible with the suggestion that the Federal Government’s interest is either entirely derivative of the States’ interests (as with the relationship between the Colorado ditch company and the State of Colorado in Hinderlider) or merely a stand-in for the interests of the water districts. See post, at 20. Our reasons for finding that intervention was warranted—(1) the United States’ duties under the Project and the Downstream Contracts, (2) the United States’ integral role in the Compact’s operation, and (3) the United States’ treaty obligations—stemmed from “ ‘distinctively federal interests’ ” the United States has, independent of Texas, “in seeing that water is deposited in the Reservoir consistent with the Compact’s terms.” Texas, 583 U. S., at 413–414 (emphasis added). As it did then, the United States continues to claim that New Mexico’s interference with the Project’s delivery of water violates the Compact. That Texas’s litigation strategy has since changed, such that it is now willing to accept a greater degree of groundwater pumping, does not erase the United States’ independent stake in pursuing claims against New Mexico.

Second, because Hinderlider was based on a compact that is different from the one at issue here, its reasoning is inapposite. Different compacts divide state and federal authority differently. Hinderlider’s analysis of the States’ “conclusive” power to determine their citizens’ shares of water was a function of the specifics of the compact in that case, which gave the States the sole authority over and responsibility for apportionments of the La Plata River. 304 U. S., at 96–98, 107. Here, by contrast, the United States “plays an integral role in the Compact’s operation.” Texas, 583 U. S., at 414. Reclamation’s operation of the Project, and the United States’ obligations to EBID and EP1 under the Downstream Contracts, are the means by which the States chose to effectuate the apportionment of water in the Compact. Rather than “requiring New Mexico to deliver a specified amount of water annually to the Texas state line,” the Compact instead “directed New Mexico to deliver water to the” Elephant Butte Reservoir. Id., at 410–411. That choice made sense only because the “Downstream Contracts . . . promised Texas water districts a certain amount of water” via the operation of the Project. Id., at 411. Accordingly, the Federal Government has its own “interest in seeing that water is deposited in the Reservoir consistent with the Compact’s terms” and not “siphon[ed] off . . . in ways the Downstream Contracts do not anticipate.” Id., at 411, 414.

For similar reasons, our continued recognition of the United States’ valid Compact claims would not, as the States assert, “tur[n] on its head the hierarchy of authorities governing the distribution of water within a federal irrigation project.” Joint Reply 34. Relying on California v. United States, 438 U.S. 645, the States maintain that the Federal Government must “comply with state water laws in operating its federal Reclamation projects.” Joint Reply 34. True, so far as it goes. California held that §8 of the Reclamation Act required the United States to comply with state-imposed permit requirements when impounding water from the Stanislaus River for use in a federal irrigation project. 438 U. S., at 647–650. But the United States is not seeking to skirt any state law here.

Again, the United States’ position is that the Compact itself imposes a duty of noninterference on New Mexico. That claim is not at odds with California’s holding that the Secretary of the Interior must “comply with state laws, not inconsistent with congressional directives, governing use of water employed in federal reclamation projects.” California v. FERC, 495 U.S. 490, 504 (1990) (discussing California v. United States, 438 U. S. 645). The United States’ claims rest on its interpretation of the Compact, and the Compact trumps state water law. See Texas, 583 U. S., at 412 (“[O]nce Congress gives its consent, a compact between States—like any other federal statute—becomes the law of the land”); New York, 598 U. S., at 224.[4]

B

1

Because the United States has valid Compact claims and has not agreed to the proposed consent decree, the only remaining question is whether the consent decree would dispose of those claims. Firefighters, 478 U. S., at 529. We conclude it would.

To start, the States have conceded as much. In their briefing before the Special Master, the States acknowledged that the consent decree would “resolv[e] all of the Compact claims stated by any party.” States’ Joint Motion To Enter Consent Decree 33 (Nov. 14, 2022) (emphasis added). Likewise, in their reply, the States reaffirmed that “upon entry of the Consent Decree, the United States will have no remaining Compact claims.” States’ Joint Reply in Support of Joint Motion To Enter Consent Decree 7 (Feb. 3, 2023) (emphasis added). The Special Master agreed, explaining that the consent decree would “resolv[e] the dispute over the Texas and downstream New Mexico apportionments.” Third Interim Report 2.

And those concessions state an obvious proposition, because the consent decree would in fact resolve the United States’ claims in this action. The United States maintains that New Mexico’s pumping breaches that State’s alleged duty under the Compact not to interfere with the Project. Intervening Complaint 4–5. And the United States seeks an injunction against New Mexico that would prohibit that interference. Id., at 5. The proposed consent decree, however, would dispose of that legal claim and the associated prayer for relief without addressing the United States’ contentions, as it neither imposes the duty of noninterference that the United States seeks nor enjoins New Mexico from allowing groundwater pumping beyond 1938 levels. To the contrary, the consent decree would incorporate New Mexico’s groundwater pumping into the Compact by adopting a new method for apportioning Rio Grande water—the EEPI.

As explained above, the EEPI would establish “an index-based methodology” to assess New Mexico’s compliance with its water delivery obligations “based upon Project operations during the D2 Period,” from 1951 to 1978. Third Interim Report Addendum 9, 23, 25. Those decades coincided with the onset of drought conditions in the Rio Grande Basin and an accompanying increase in groundwater pumping in New Mexico. Measuring New Mexico’s compliance with the consent decree (and, by extension, its compliance with the Compact) against D2 Period conditions would therefore take for granted the very increase in groundwater pumping that the United States maintains violates New Mexico’s duty of noninterference. See Third Interim Report 75 (“Undisputedly, the Consent Decree’s reliance on the D2 period seeks to limit pumping to an average amount as reflected in the 1951–1978 timeframe rather than a strict return to a pumping condition as existed in 1938”).

Were the consent decree adopted, the United States would be precluded from claiming what it argues now—that New Mexico’s present degree of groundwater pumping violates the Compact. Indeed, the consent decree would settle that question by deeming New Mexico compliant with the Compact, even as it allows pumping at the D2 levels. And that legal determination would “be reflected in, and be enforceable as, a judicial decree.” Rufo, 502 U. S., at 378.

The proposed consent decree, therefore, would have the effect of “cutting [the United States] off from a remedy to which” it alleges it is entitled. Lawyer v. Department of Justice, 521 U.S. 567, 579 (1997).

The United States’ argument that groundwater pumping at D2 levels violates the Compact may or may not ultimately prevail at trial. But we “may not enter a ‘consent’ judgment without the actual consent of the Government” when “the Government seeks an item of relief to which evidence adduced at trial may show that it is entitled.” Ward Baking Co., 376 U. S., at 334. Because the consent decree here would have that effect, we cannot approve it over the United States’ objection.

2

The States and the dissent nevertheless argue that rejecting the consent decree would unjustly expand the scope of this original action and that the United States can and should litigate its claims in another forum instead. Joint Reply 38–45; post, at 14–21. Neither argument holds up.

The first objection boils down to the unremarkable fact that the United States’ and Texas’s interests have now diverged. As we explained in 2018, both Texas and the United States at that point asserted “essentially the same claims” and sought “substantially the same relief ”—an end to New Mexico “siphon[ing] off water below the Reservoir in ways the Downstream Contracts do not anticipate.” Texas, 583 U. S., at 409, 411, 415. The United States still asserts that same claim today and seeks that same relief. That Texas has chosen to compromise does not mean that, by staying the course, the United States is expanding this action. What is more, this Court was well aware in 2018 that the States’ interests might diverge from those of the United States. See, e.g., New Mexico’s Reply to Exceptions of the United States and Colorado 25 (July 28, 2017); Reply Brief for United States 18 (July 28, 2017).

The second objection turns on a mischaracterization of the United States’ claims. The States maintain that the Federal Government’s qualms with New Mexico’s groundwater pumping pose only “an intrastate dispute between the United States and New Mexico” that is better left to existing litigation in other courts. Joint Reply 43–45. For the reasons already explained, however, the United States’ claims are not limited to “issues related to reclamation law, Project operations, or the details of New Mexico water administration.” Id., at 43. Rather, the United States maintains that New Mexico’s groundwater pumping contravenes the Compact itself. Nothing in the consent decree prohibits that alleged breach of the Compact; to the contrary, compliance with the consent decree would instead constitute compliance with the Compact. We therefore do not see how the United States could elsewhere vindicate its claim that the Compact itself bars New Mexico’s allegedly excessive groundwater pumping.[5]

*  *  *

Our decision today follows directly from our prior recognition of the United States’ distinct federal interests in the Rio Grande Compact. Having acknowledged those interests, and having allowed the United States to intervene to assert them, we cannot now allow Texas and New Mexico to leave the United States up the river without a paddle. Because the consent decree would dispose of the United States’ Compact claims without its consent, the United States’ exception is sustained, and the States’ motion to enter the consent decree is denied.

It is so ordered.

SUPREME COURT OF THE UNITED STATES

_________________

No. 141, Orig.

_________________

TEXAS, PLAINTIFF v. NEW MEXICO AND COLORADO

on exception to third interim report of the special master

[June 21, 2024]

Justice Gorsuch, with whom Justice Thomas, Justice Alito, and Justice Barrett join, dissenting.

Texas and New Mexico ask us to approve a consent decree resolving their decade-long original jurisdiction dispute over the Rio Grande Compact. The decree would fairly apportion water from the Rio Grande River between those two States and leave federal reclamation operations in the area running the way they have run for decades. A Special Master we appointed to consider the dispute has recommended approving the proposed decree, concluding that it is “difficult to envision a resolution to this matter that might be superior.” Third Interim Report of the Special Master 15 (Third Interim Report). The States’ dispute resolved, and the basis for our original jurisdiction gone with it, the Special Master also recommends dismissing without prejudice any claims the United States, an intervenor in the case, might hold.

Still, the Court denies entry of the consent decree. Why? Because the federal government demands as much. Not content with receiving what it asked for when it intervened in this litigation—the protection of its existing federal reclamation operations—the United States now seeks to advance a theory about how water should be distributed between Texas and New Mexico so aggressive that New Mexico fears it could devastate its economy. In the process, the federal government seeks to prolong this original jurisdiction dispute, a form of litigation usually reserved for disputes between States, over the objection of both Texas and New Mexico. And it does so despite the fact the consent decree would leave the federal government free to pursue any claims it believes it has in the lower courts, where disputes between the federal government and States are normally tried.

The Court’s decision is inconsistent with how original jurisdiction cases normally proceed. It defies 100 years of this Court’s water law jurisprudence. And it represents a serious assault on the power of States to govern, as they always have, the water rights of users in their jurisdictions. The Special Master issued a detailed 115-page report laying all this out. His views were wise, his recommendations sound, and, respectfully, we should have done as he suggested.

I

A

Beginning its journey high in the San Juan Mountains, the Rio Grande runs through Colorado, New Mexico, and Texas before flowing into Mexico and eventually the Gulf of Mexico. Along the way, the river serves as a vital irrigation source for crops as varied as the terrain through which it passes, nourishing everything from pecans to the justly famous green chiles of the Hatch Valley. See El Encanto, Inc. v. Hatch Chile Co., 825 F.3d 1161 (CA10 2016).

To ensure “an equitable apportionment” of the Rio Grande’s waters, Colorado, New Mexico, and Texas entered into the Rio Grande Compact in 1938. 53 Stat. 785. Congress approved it the following year. Ibid.; see U. S. Const., Art. I, §10, cl. 3 (requiring congressional approval for a State’s “Compact with another State”). The Compact directs Colorado to deliver a specified amount of water to the New Mexico-Colorado border. 53 Stat. 787–788. New Mexico must then deliver water to Elephant Butte Reservoir, located about 100 miles north of the Texas line, in order to ensure Texas receives its share of the river’s waters. Id., at 788.

The United States Bureau of Reclamation operates the Reservoir as part of the federal Rio Grande Project. That Project serves two roles relevant here. First, pursuant to contracts with New Mexico and Texas water districts (serving areas around Las Cruces and El Paso), the Project supplies water from the Reservoir to those districts using a roughly 57%–43% split between New Mexico and Texas. Texas v. New Mexico, 583 U.S. 407, 410 (2018) (Texas I ). We have called these the Downstream Contracts, and they essentially work to supplement the Compact, which is silent as to the precise quantity of water owed Texas. Id., at 410–411. Second, the Project ensures the delivery of a set amount of water to Mexico to satisfy treaty obligations to that country. Id., at 410.

Over the better part of a century, this arrangement has worked reasonably well. Yes, disagreements occasionally arise, sometimes leading to the filing of a complaint in this Court. But, invariably, these disputes have settled before the Court reached the merits. See, e.g., Texas v. New Mexico, 308 U.S. 510 (1939); Texas v. Colorado, 474 U.S. 1017 (1985).

B

In the early 2000s, another disagreement arose. The causes? The 100-mile-long journey water must travel from Elephant Butte Reservoir to Texas, and the increase in groundwater pumping along that route. Groundwater and surface water (like the Rio Grande) are often connected, drawing from and feeding back into one another. Because of this connection, pumping by New Mexicans downstream of the Reservoir (that is, between Elephant Butte and Texas) reduces the amount of Project water that reaches Texas’s water district. Texas saw this as a violation of the Compact. So in 2013, it sought to file a bill of complaint in this Court against New Mexico. (Colorado, as a signatory to the Compact, joined as a defendant). We agreed to exercise our original jurisdiction over the case and appointed a Special Master to aid in our consideration of it. Texas I, 583 U. S., at 411.

In brief, here is how Texas framed its claim. It argued that the Compact implicitly guarantees that the State’s water district will receive a certain minimum quantity of Rio Grande water from New Mexico. And, Texas contended, we should calculate the amount of that water based on the “conditions” in and around the river “that existed in 1938 at the time the Rio Grande Compact was executed.” Texas Complaint 5, ¶10. Back in 1938, there was hardly any groundwater pumping. So adopting 1938 conditions as our baseline would have the effect of giving Texas’s water district more water. See id., at 8–10, ¶18.

New Mexico resisted Texas’s claim. Among other things, New Mexico observed that the Compact is silent about how to measure water due Texas. N. M. Brief in Opposition 14–15 (Mar. 11, 2013). And New Mexico stressed that, since approximately 1980, the federal government has relied on data about Rio Grande conditions between 1951 and 1978—the so-called D2 Period, when groundwater pumping was more prevalent—to calculate the amount of water due Texas’s water district under the Downstream Contracts. See N. M. Counterclaims 10–11, ¶¶40–41 (May 22, 2018); Joint Reply to Exception of the United States by the State of Texas et al. 5–6. New Mexico stressed, too, that Texas had not previously objected in this Court to that practice—a sign, New Mexico said, that Texas understood it to be entirely consistent with the Compact. N. M. Answer 10–11, ¶36 (May 22, 2018).

Abandoning decades of practice and mutual understanding, New Mexico continued, would threaten dire consequences for its economy. Farming along the Rio Grande, New Mexico explained, relies in part on groundwater pumping for irrigation. And replacing the D2 Period with a 1938 baseline, when pumping was all but nonexistent, could put at risk nearly 50,000 jobs (in a State of 2 million people) and up to 10% of the State’s gross domestic product. See 1 Tr. of Proceedings before the Special Master 47 (Oct. 4, 2021).

In short order, the United States moved to intervene and “filed a complaint that presented the federal government’s interests.” Tr. of Oral Arg. 4 (Jan. 8, 2018) (2018 Transcript). For its part, the United States agreed with Texas about the bottom line—that New Mexican groundwater pumping below the Elephant Butte Reservoir was “interfering with the equitable apportion[ment of] water to Texas.” Id., at 29–30. But it disagreed with Texas about the appropriate method for calculating the amount of water owed Texas. A holding for Texas that the Compact required the use of a 1938 baseline, the federal government worried, would require it to alter its longstanding use of the D2 Period when assessing what deliveries were due under the Downstream Contracts. Reply Brief for United States 20 (July 28, 2017) (2017 Reply). Intervention, as the federal government put it, would allow it to protect its interest “in the Project’s operation” as well as its interest in ensuring sufficient water reaches Mexico to satisfy its treaty obligations. Id., at 11–12, 18.

The United States also flagged for us a procedural issue: Because the federal government wasn’t a signatory to the Compact, it wasn’t clear on what basis it could press any Compact claims separate from the claims held by the signatory States. This raised the question whether the United States “could go forward” with claims in its own right “if Texas’s complaint were dismissed” or the parties settled. 2018 Transcript 14. The government speculated that it might be able to bring an independent claim as a “third-party beneficiary,” id., at 19, or perhaps had some “implied right of action” under general “equitable” principles, id., at 20. But because Texas’s complaint was “going forward,” the government asked us not to “reach that” issue. Id., at 14.

After hearing argument, in 2018 we “permitted the federal government to participate in [this] compact sui[t].” Texas I, 583 U. S., at 412. In our decision, we accepted the federal government’s suggestion that there was no need to decide whether it had valid, independent Compact claims of its own. Id., at 415. Instead, we held, four “considerations taken collectively persuade[d] us” that the government’s participation was appropriate. Id., at 413. First, we recognized the federal government’s “duties under the Downstream Contracts” afforded it an “interest in seeing that water is deposited in the Reservoir consistent with the Compact’s terms.” Id., at 414. Second, we gave weight to New Mexico’s concession that the Project “plays an integral role in the Compact’s operation.” Ibid. Third, we said that “[p]ermitting the United States” to intervene would “allow it to ensure” its treaty obligations to Mexico were “honored.” Id., at 415. Fourth, we emphasized that we had no reason to decide whether the government could press Compact claims independently of the States because “the United States ha[d] asserted its Compact claims in an existing action brought by Texas, seeking substantially the same relief and without that State’s objection.” Ibid.

At the same time, we expressly warned that “permission” to intervene “should not be confused for license.” Id., at 413. In particular, we stressed, “[t]his case does not present the question whether the United States could initiate [its own] litigation . . . under the Compact or expand the scope of an existing controversy between [the] States.” Id., at 415. And, we added, “[n]othing in our opinion should be taken to suggest whether a different result would obtain in the absence of any of the considerations” we had laid out, “or in the presence of additional, countervailing considerations.” Ibid.

C

Once the case returned to the Special Master, it appeared to be heading in the same direction as prior disputes about the Rio Grande Compact. After completing an “initial phase” of a trial, months-long negotiations followed. Third Interim Report 35. Ultimately, those discussions culminated in a settlement and proposed consent decree in 2022. In the decree, the parties agreed to continue using the D2 Period to measure the amount of water due Texas’s water district. Id., at 42; see Addendum to Third Interim Report 8–11 (Addendum). But they also agreed Elephant Butte—over 100 miles from the Texas border—wasn’t the appropriate place to measure the amounts due Texas in light of the New Mexican groundwater pumping between the Reservoir and state line. Instead, the States resolved to measure water flows into Texas at a federally operated gauge near El Paso, Texas, by the border between the two States. Third Interim Report 7; Addendum 8–9.

In short, as with any settlement agreement, each side gave something up to gain something it wanted. Through the use of the El Paso gauge, Texas received a guarantee that deliveries to its water district would be protected from excessive New Mexican groundwater pumping between Elephant Butte and the state line. And through the continued use of the D2 Period as the baseline, New Mexico won its water users the right to maintain at least some of that pumping. Colorado, as a signatory to the Compact, gave its assent.

For the United States, the consent decree promised business as usual. That’s because “the [c]onsent [d]ecree essentially adopt[ed]” the federal government’s “own method of operating.” Third Interim Report 107. The government would continue to use the D2 Period for measuring the amounts it distributed to Texas’s and New Mexico’s water districts, just as it had sought when it intervened and as it has done “for approximately the last 40 years.” Id., at 42. The federal government would not even have to establish a new water gauge at El Paso, for it already operates one. See id., at 107. It was undisputed, too, that the consent decree would protect water due Mexico under this country’s treaty with that nation. Id., at 94, n. 10.

The federal government objected to the decree’s entry anyway. In an unexpected and still-unexplained move, the United States abandoned its position, held for over 40 years, that its own D2 Period data supply the correct method for measuring the amount of water it must deliver to Texas and New Mexico water districts. Instead, the federal government began advocating for something similar to what Texas had once urged—the “broad elimination of New Mexican [groundwater] pumping through a return to a 1938” baseline. Id., at 14. Unlike Texas, however, the federal government had never alleged in its complaint that the Compact required the use of the 1938 baseline. In fact, it still has not sought to plead such a claim. Perhaps even stranger yet, despite its new litigating position, the United States continued (and still continues) to deliver water to the water districts using the D2 Period as its guide.

D

In a detailed 115-page report, the Special Master recommended we approve the consent decree. He advised that it was “difficult to envision a resolution to this matter that might be superior” to it. Id., at 15. In particular, the Special Master observed that the States and federal government had long used the D2 Period to measure the apportionment of water due each State. And nothing in the voluminous submissions he received suggested that they had to do otherwise. As he put it, no evidence suggested that “the Compacting States believed [in 1938] they were locking in . . . any particular condition of development,” such as a certain amount of groundwater pumping, for determining what water was due Texas or New Mexico. Id., at 76–77.

That left the question what to do with any claims the federal government might believe it has and wishes to pursue as a result of its newfound views. Our decision in Texas I, the Special Master recalled, did not decide whether the government had viable, independent Compact claims of its own. And rather than undertake that assessment himself, he recommended dismissing any claims the government might have without prejudice. Third Interim Report 115. The Court, he reasoned, had taken the rare step of exercising its original jurisdiction because the case involved a dispute between two States. Id., at 11. That dispute was now resolved. And, he said, the federal government could pursue any claims it might have against the States or other water users as it normally does, “in one of several ongoing or any new lower court actions.” Id., at 99. In fact, as the Special Master alluded to, the federal government is already involved in Compact-related litigation with New Mexico in federal district court. See New Mexico v. United States, No. 1:11–cv–00691 (DNM).

Though the States’ agreement and the Special Master’s recommendations promised to bring to an end a decade of litigation, the United States filed an exception to those recommendations. It asked us to reject the proposed decree and order the Special Master to conduct further proceedings yet. We agreed to hear oral argument on the federal government’s request.

II

A

The principles that guide our analysis in original jurisdiction water disputes like this one are long settled. The “power to control navigation, fishing, and other public uses of water,” we have said, “is an essential attribute of [State] sovereignty.” Tarrant Regional Water Dist. v. Herrmann, 569 U.S. 614, 631 (2013) (internal quotation marks omitted). But in our federal system, one State may not exercise its sovereignty in ways that deny another State the capacity to exercise its own. So to prevent upstream States from wholly draining rivers that would otherwise reach their downstream neighbors, this Court many years ago developed the doctrine of equitable apportionment—the notion “that States have an equal right to make a reasonable use of a shared water resource.” Mississippi v. Tennessee, 595 U.S. 15, 24 (2021) (internal quotation marks omitted); see Kansas v. Colorado, 206 U.S. 46 (1907).

Time and again, we have urged States to effect this apportionment “by mutual accommodation and agreement” rather than through litigation. Florida v. Georgia, 585 U.S. 803, 809 (2018) (internal quotation marks omitted) (collecting cases). Agreements of that kind usually take the form of an interstate compact. Once approved by Congress, compacts gain the status of federal law. Texas I, 583 U. S., at 412. And because States’ authority over their waters is an essential attribute of their sovereignty, a compact’s apportionment of water between two or more States “is binding upon . . . all water claimants” in those States, “even where [a] State had granted the water rights before it entered into the compact.” Hinderlider v. La Plata River & Cherry Creek Ditch Co., 304 U.S. 92, 106 (1938). So, for example, a compact between Texas and New Mexico allocating water between them binds their respective water districts that contract for water with the federal government, along with all other water users in their jurisdictions.

Notably, compacts also bind the federal government when it distributes water from its reclamation projects pursuant to agreements like the Downstream Contracts. Compacts do so not only because they are federal law. Texas I, 583 U. S., at 412. They do so as well because Congress has specifically directed federal reclamation projects to “follow state law as to water rights” unless that law conflicts with some other “explicit congressional directive.” California v. United States, 438 U.S. 645, 673 (1978); see 43 U. S. C. §383. As we have put it, Congress has “subject[ed] to the authority of ” the States “[a]ll of the acts of the [federal] Reclamation Bureau in operating [its] reservoirs.” Nebraska v. Wyoming, 295 U.S. 40, 42 (1935) (Nebraska I ). So an interstate water rights compact “necessarily bind[s]” the government as it would “any other appropriator in th[e] [S]tate.” Id., at 43. We have referred to this aspect of congressional water policy as a form of “ ‘cooperative federalism.’ ” California, 438 U. S., at 650.

While compacts provide a highly valuable tool for resolving water disputes, disagreements about the meaning of their terms arise from time to time. The Constitution vests this Court with original jurisdiction to adjudicate these interstate disputes, an “ ‘awkwar[d]’ ” arrangement where we sit, in effect, as a trial court, a court of first (and last) review. South Carolina v. North Carolina, 558 U.S. 256, 267 (2010). Decide though we may, our general “ ‘preference’ ” is for States to negotiate to resolve their differences. Florida, 585 U. S., at 809. When those negotiations bear fruit, the product is often a proposed consent decree containing “detailed mechanisms to promote compliance with the [c]ompact’s terms.” E.g., Kansas v. Nebraska, 574 U.S. 445, 451 (2015).

Because a consent decree in a water rights case seeks simply to provide more “detailed mechanisms” to implement a compact, it bears the same force as one. Just like a compact, a consent decree is binding on all those in the affected States, regardless of their “participation” in the case, Nebraska v. Wyoming, 515 U.S. 1, 22 (1995), or their “assent or dissent,” Hudson County Water Co. v. McCarter, 209 U.S. 349, 355 (1908). And, once more, the same holds true when it comes to federal reclamation projects that distribute water to users in the affected States. They must operate consistently with a decree’s terms unless doing so conflicts with some other explicit congressional directive. See Nebraska I, 295 U. S., at 43; California, 438 U. S., at 674.

Still, our approval of a consent decree is hardly a given. The parties may not use a settlement to rewrite a compact, for a new compact requires new congressional approval. See Kansas, 574 U. S., at 455–456. So, when presented, as we are here, with a request to approve a proposed consent decree, two considerations guide our decisionmaking. First, we ask whether the decree is “consistent with the compact itself.” Id., at 455. In answering that question, we do not require the States’ proposal to be perfect. Rather, we will “give [a settlement] effect” as long as it is not “wholly contrary to relevant evidence, . . . even if we would reach a different conclusion upon the same evidence.” New Hampshire v. Maine, 426 U.S. 363, 369 (1976). Second, because the parties’ agreement is the driving force behind the decree, we consider whether the decree purports to bind third parties the States have no authority to represent. In particular, we confirm that a proposed settlement does not improperly impose duties or obligations on those third parties without their consent or dispose of the valid claims they enjoy. Firefighters v. Cleveland, 478 U.S. 501, 529 (1986).

B

With these rules in mind, I see no sound basis on which we might refuse to adopt the Special Master’s recommendation to approve the States’ consent decree.

First, the decree is consistent with the Compact. All agree the Compact implicitly guarantees Texas some minimum amount of Rio Grande water each year. Third Interim Report 75–76, and n. 6. In their settlement, the States propose to calculate that amount by reference to the D2 Period and measure it at a water gauge at El Paso. Both terms are entirely appropriate. The States have relied on the D2 Period for decades. And in making distributions to those States’ water districts pursuant to the Downstream Contracts, so has the federal government. These longstanding practices are “highly significant evidence of [everyone’s] understanding of the [C]ompact’s terms.” Tarrant Regional Water Dist., 569 U. S., at 636 (internal quotation marks omitted). All agree, too, that the Compact expressly authorizes representatives from each compacting State to choose gauge locations. See Third Interim Report 69–70. So use of the El Paso gauge is consistent with the Compact as well. Indeed, by using that gauging station rather than one 100 miles upstream from the Texas border at the Elephant Butte Reservoir, the decree ensures Texas’s water district is protected from excessive groundwater pumping in New Mexico between the Reservoir and the state line.

Second, the consent decree does not impose any new improper duty or obligation on the federal government or deny it the ability to pursue any valid claim it may have. Yes, under the decree, reclamation authorities must measure water they distribute to Texas’s and New Mexico’s water districts using the D2 Period. And they must use the El Paso gauging station to do so. But, again, the federal government has employed the D2 Period to measure the water it distributes for decades, and it has long maintained the El Paso gauging station. The government cannot sensibly suggest that it would be improper to require it to continue doing as it has long done.

Nor is there anything unusual about any of this. As we have seen, under longstanding federal law, a consent decree between the States “will necessarily bind” “the Reclamation Bureau” because “[a]ll of [its] acts . . . in operating the [Project] so as to impound and release waters of the river are subject to the [States’] authority.” Nebraska I, 295 U. S., at 42–43; see Part II–A, supra. Accordingly, Texas and New Mexico are entitled to decide what water rights their governmental water districts are due, and the federal government’s reclamation project is bound to honor what the States say on the subject. See ibid.; California, 438 U. S., at 675. Of course, a consent decree would be improper if it left the federal government unable to meet some other “explicit congressional directive.” Id., at 673. But the government does not argue anything of the sort here, never suggesting, for example, that the proposed decree would risk its obligations under its treaty with Mexico. To the contrary, it is undisputed that compliance with the decree would “protect the [t]reaty water.” Third Interim Report 94, n. 10 (emphasis added).

Think about it this way. A federal reclamation project may not decide that state water districts are entitled to different water rights than States have specified in their compacts and consent decrees. Those agreements bear the force of federal law, as Congress has directed and our cases have long recognized. And were there any possible remaining room for doubt, the Downstream Contracts themselves dispel it. They direct the contracting parties (the federal government and the water districts) to apply two sources of law when allocating water: that of the States and the federal Reclamation Act of 1902—the same statute that instructs the federal government to defer to the States in allocating water rights among their users and to operate reclamation projects consistent with what state law requires. See 4 Tex. App. in Support of Partial Summary Judgment 593; 2 id., at 911–912; 43 U. S. C. §383.

Nor does the consent decree dispose of any valid claims the federal government may possess in its own right, whether under the Compact or any other source of law. To be sure, to the extent the federal government seeks to pursue a claim “wholly derivative” of the States (or their water districts), those claims necessarily “rise or fall with the claims of the States,” and the federal government has no independent right to press them. Alabama v. North Carolina, 560 U.S. 330, 357 (2010). But, to the extent the federal government thinks it has any independent claims of its own, the Special Master recommends dismissing them without prejudice.

That is “an entirely appropriate”—and our long- preferred—“means of resolving whatever questions remain” after the resolution of an interstate dispute. California v. Nevada, 447 U.S. 125, 133 (1980). After all, once a dispute between the States comes to an end, so does the basis for our exclusive original jurisdiction—jurisdiction we exercise only “ ‘sparingly.’ ” South Carolina, 558 U. S., at 267; see 28 U. S. C. §1251. A dismissal without prejudice allows the federal government to pursue any valid independent claims it may have in the ordinary course in lower courts. And, naturally, should the federal government prevail in that litigation in a way that affects the consent decree, it may return to this Court and seek a modification of that decree.

Our consent-decree decisions outside the water-rights context confirm as much. Take Firefighters v. Cleveland, a Title VII discrimination suit brought by “an organization of black and Hispanic firefighters” against their employer, the city of Cleveland. 478 U. S., at 504. Under the terms of a proposed consent decree, the city sought to revise an allegedly discriminatory promotion exam and otherwise make up for its “assertedly limited minority advancement.” Id., at 505; see id., at 510. An intervenor, the union “represent[ing] a majority of Cleveland’s firefighters,” objected on the ground that the remedy would harm its “ ‘non- minority’ ” members. Id., at 506, 512. Affirming the entry of the decree, we noted that the union remained free to bring its own independent Title VII or Fourteenth Amendment claims in separate litigation. Id., at 530. “[W]hether [those] claims have merit [is a] questio[n] that must be presented in the first instance to the [d]istrict [c]ourt.” Ibid. Until then, city employees, including union members, would be subject to the consent decree’s promotion provisions. If the union members won on their statutory or constitutional challenges, however, the decree would have to be modified to bring it in line with those laws. See id., at 526–528. So too here.[1]

III

Despite reaching a different result, the majority has little to say in response. It does not dispute the above account of our settled water law jurisprudence. Nor does it identify any inconsistency between the proposed consent decree and the Compact’s terms. Instead, bypassing all that, the majority proceeds in two steps. First, it suggests, the United States may have valid, independent Compact claims of its own that the consent decree extinguishes; second, the majority insists, holding otherwise would be inconsistent with our decision in Texas I, where we allowed the federal government to participate to protect its interests. Neither argument is sound.

A

Primarily, the majority contends the decree risks disposing of valid, independent claims that may belong to the federal government. To advance its position, the majority relies on supposed “concessions” by Texas and New Mexico before the Special Master that, if the consent decree were confirmed, the federal government would be left with “ ‘no remaining Compact claims.’ ” Ante, at 16 (emphasis deleted); see also ante, at 19, n. 5 (treating a similar assertion by Texas at oral argument as a “conce[ssion]”). This argument is wrong for a number of reasons.

First, the majority’s telling omits what happened next. Far from “agree[ing]” with the States, ante, at 16, the Special Master recommended we dismiss any claims the federal government might have in its own right “without prejudice to being asserted in other fora,” Third Interim Report 11. This recommendation applied, he said, “regardless of whether the United States bases its claims on Reclamation law, state law, the Compact, or some other source of authority.” Ibid. Because the States did not file an exception to this recommendation, we may treat them as having acceded to it. See Texas v. New Mexico, 592 U.S. 98, 105 (2020). That alone is enough to answer the majority.

Second, the majority does not explain why the usual course of dismissing a third party’s claims without prejudice wouldn’t be “entirely appropriate” here, as it ordinarily is in our original jurisdiction cases. California, 447 U. S., at 133; see ante, at 16–18. The majority does not, for example, explain why the federal government could not press whatever independent Compact claims it believes it has in lower courts and return here, if necessary, to seek modification of the States’ consent decree. See Third Interim Report 99–100. The majority does not offer any such explanation because it cannot. See supra, at 15–16, and n. 1. Until the government had the case stayed to participate in this one, the United States was already involved, as we have seen, in Compact litigation with New Mexico in federal district court. See Memorandum Opinion and Order in New Mexico v. United States, No. 1:11–cv–00691 (DNM, Mar. 29, 2013), ECF Doc. 193, pp. 5–6. Perhaps the government thinks it more convenient to remain here than to return for decree modification should it prevail in that suit or another. But that “do[es] not provide a basis for declining to [approve] a decree.” Idaho ex rel. Evans v. Oregon, 462 U.S. 1017, 1026 (1983).

Third, the majority struggles to spell out how the government might possibly hold Compact claims in its own right—that is, independent of the States’ claims. Yes, the majority describes the government’s present “position”—namely that the Compact imposes a 1938 baseline—and repeats the observation that “the Compact trumps state water law.” Ante, at 15. But the majority does not suggest, as the government once did, that the United States may sue as a third-party beneficiary of the Compact or under some ill-defined equitable cause of action. See ante, at 11, n. 2, 14; Part I–B, supra; 2018 Transcript 19–20. Nor does the majority purport to identify anything in the Compact that might entitle the federal government the right to sue to demand a 1938 baseline. See Tarrant Regional Water Dist., 569 U. S., at 632 (“silence in compacts” must be read in favor of “the States’ authority to control their waters”). In fact, the majority does not dispute that the United States still has yet even to plead such a claim of its own.[2]

Fourth, the majority conspicuously avoids the lessons of our water law jurisprudence. So, for example, the majority expresses surprise that the government might be bound to honor the terms of the consent decree until and unless it prevails in other litigation on its own claims and then returns here to seek revision of the decree. Ante, at 19, n. 5; n. 1, supra. But about that there should be no surprise. Few rules in water law are more settled than that federal reclamation projects must comply with any Compact, state water law, or consent decree term “not inconsistent with clear congressional directives respecting the project.” California, 438 U. S., at 672; see also Part II, supra. And here, no one, the majority included, has identified any congressional directive, much less a clear one, inconsistent with the consent decree before us.

Fifth, the majority’s reasoning doesn’t withstand scrutiny even under ordinary consent-decree principles. Suppose, as the majority does (incorrectly), that approval of the decree would necessarily preclude the United States from claiming in any other forum “that New Mexico’s present degree of groundwater pumping violates the Compact.” Ante, at 17. Even so, the majority is mistaken when it claims that the proposed consent decree “would have the effect of ‘cutting [the United States] off from a remedy to which’ it alleges it is entitled.” Ante, at 17–18 (quoting Lawyer v. Department of Justice, 521 U.S. 567, 579 (1997)). It is undisputed that the government’s present “prayer for relief ” in this case seeks only to “prohibit th[e] interference” with the Project caused by excess groundwater pumping in New Mexico. Ante, at 16–17. All agree, too, that at the time the United States intervened, the government determined how much pumping was too much by reference to the D2 Period; the government did not allege—and still has not alleged—in its complaint that the Compact mandates a 1938 baseline. Part I–B, supra. To complete the majority’s clipped quotation, then, the decree would “dispos[e] of [the government’s] claim not in the forbidden sense of cutting [it] off from a remedy” to which it alleges it is entitled, “but only in the legitimate sense of granting [the government] the very relief [it] had sought.” Lawyer, 521 U. S., at 579; see infra, at 22–24.

Instead of answering any of these problems, the majority changes the subject. It replies by observing that the federal government’s deliveries under the Downstream Contracts play a central role in effectuating the Compact by ensuring certain Rio Grande waters reach New Mexico and Texas water districts. Ante, at 9–10. That may be so, but it is no answer for reasons we have already seen. Those contracts do not promise water districts that the 1938 conditions will be used in measuring the water due them. Nor may the federal government seek to vindicate the contractual rights of the States’ own water districts. Rather, Congress’s instructions, a century’s worth of this Court’s precedents, and the Downstream Contracts themselves teach that the compacting States get to decide what water rights those and other water users in their jurisdictions enjoy. And a federal reclamation project is bound to honor those decisions absent some clear congressional command to the contrary. See Part II, supra.

B

At this point, the majority retreats. Perhaps what I have laid out above would hold true in any other case, it replies, but this one is special. Special, the majority asserts, because in Texas I we allowed the United States to participate in this case. And that ruling, the majority says, necessarily means the United States may pursue, independently of Texas, a claim that the Compact requires use of the 1938 conditions. Ante, at 9–13.

This argument is mistaken, too. Recall that, in Texas I, the government suggested it might be able to sue in its own right under third-party-beneficiary or equitable-cause-of-action theories. See Part I–B, supra. But recall, too, that the federal government asked us not to “reach” the question whether it could independently bring claims of its own under these theories or any others. Ibid.; 2018 Transcript 14. It said answering the question whether it could sue in its own right was needless because Texas’s claims were live. Ibid. We proceeded in express reliance on that representation, stressing that we were not resolving either “the question whether the United States” could sue independently of Texas “under the Compact” or the question whether it could otherwise “expand the scope of an existing controversy between States.” Texas I, 583 U. S., at 415.

Really, there was no way we could have passed on the federal government’s current assertion that it has a right to pursue a claim that the Compact requires the use of a 1938 baseline. As the majority concedes, Texas I “repeatedly” cabined our permission to intervene to “ ‘the Compact claims [the United States] has pleaded in this original action.’ ” Ante, at 11, n. 2 (quoting 583 U. S., at 415). As the majority admits, too, the government has never pleaded the existence of a 1938 baseline. Ante, at 13, and n. 3. Instead, when it sought to intervene, the government took just the opposite view, arguing that its longstanding use of the D2 Period was consistent with the Compact. Allowing the government to reverse course now is not required by anything in Texas I. More nearly, it defies that decision by “expand[ing] the scope” of the parties’ litigation. Ante, at 18. In fact, it is hard to imagine anything that might do more to expand the scope of this dispute than forcing the States to continue to litigate when they have already resolved their differences. Cf. Town of Chester v. Laroe Estates, Inc., 581 U.S. 433 (2017) (intervenor expands the scope of a case when it requests a money judgment different from the one sought by plaintiff ).[3]

The truth is, this Court has “often permitted the United States to intervene” even without a valid claim of its own. Maryland v. Louisiana, 451 U.S. 725, 745, n. 21 (1981). Texas I was simply of a piece with that practice. See 583 U. S., at 413 (citing that portion of Maryland). Far from holding the federal government could pursue a claim in its own right, we permitted it to “participate . . . to defend . . . interests that a normal litigant might not be permitted to pursue in traditional litigation.” 583 U. S., at 412–413 (internal quotation marks omitted); accord, ante, at 9. And allowing intervention in that posture is anything but a holding that the government may pursue an independent claim of its own. Cf. Fed. Rules Civ. Proc. 24(a)–(b) (setting out different Rules for intervention depending on whether an individual has “an interest” or “a claim”).

Beyond that flaw with the majority’s reading of Texas I lie others. In deciding to take the rare step of permitting intervention, we stressed that multiple “factors” “taken collectively persuade[d] us” to do so. 583 U. S., at 413, 415. At the same time, we stressed that “[n]othing in our opinion should be taken to suggest” the same result “would obtain in the absence of any of the[m] . . . or in the presence of additional, countervailing considerations.” Id., at 415. Factors present then, however, are absent now. And additional considerations have indeed arisen. In fact, through the consent decree, the federal government promises to receive everything it initially sought. Cf. Campbell-Ewald Co. v. Gomez, 577 U.S. 153, 178 (2016) (Roberts, C. J., dissenting) (“When a plaintiff files suit seeking redress for an alleged injury, and the defendant agrees to fully redress that injury, . . . there is no longer any necessity to expound and interpret the law” (emphasis deleted; internal quotation marks omitted)).

Take the treaty. One of the factors we cited as favoring intervention concerned the then-live possibility that “a breach of the Compact could jeopardize [the federal government’s] treaty obligations” to Mexico requiring it to deliver certain Rio Grande waters. 583 U. S., at 414. Now, however, everyone agrees the consent decree will do nothing to interfere with those obligations, but will instead “protect the [t]reaty water.” Third Interim Report 94, n. 10.

Next, consider the federal government’s concern in 2018 that litigation over the Compact could ultimately require it to use the 1938 conditions in its distributions to water districts, as Texas then sought. That development, the government worried, could interfere with its longstanding use of the D2 Period in its operations at the Reservoir and the Downstream Contracts. 2017 Reply 20; see 2018 Transcript 30–31 (Texas highlighting this as an example of where “Texas and the United States are not exactly going to be raising the same arguments”). But that, too, is no longer a worry. Under the proposed settlement, operations may continue at the Reservoir as they have for over 40 years.

Finally, recall that, when it intervened, the federal government disagreed with Texas about the use of the 1938 baseline but “substantially” agreed that groundwater pumping in New Mexico below the Reservoir interfered with the Texas water district’s receipt of water to which it was entitled. Texas I, 583 U. S., at 415. The parties’ proposed decree addresses this concern, as well, by ensuring the water due Texas (again, calculated using the government’s D2 Period data) is measured near the state line, at the El Paso gauging station, and not over 100 miles upstream, along a course where New Mexico users pump groundwater from the Rio Grande.

Here’s the bottom line: Texas I did not hold—nor could it have held—that the United States could pursue an independent Compact claim to enforce a 1938 baseline. To the contrary, the government’s disagreement with Texas about the appropriateness of a 1938 baseline was one of the considerations that led us to permit intervention. At the same time, the interests the federal government did assert then have been satisfied now by the States’ agreement. To conclude, as the majority does, that the government at this late hour may assert essentially any Compact-related claims—even unpleaded ones—is to ignore all this and the many caveats that accompanied our decision. Where Texas I warned the United States not to “confus[e]” “our permission” to intervene “for license,” id., at 413, the Court now reverses course and allows the government to exercise squatter’s rights over our original jurisdiction.

IV

“The history of the relationship between the Federal Government and the States,” we once observed, contains a “consistent thread of . . . continued deference to state water law by Congress.” California, 438 U. S., at 653. By “den[ying]” the Special Master’s recommendation to approve the States’ consent decree “without [the] consent” of the federal government, ante, at 20, the Court disregards this long, unbroken practice. Not to ensure the federal government can comply with some statutory directive at odds with the decree. Not to protect the interests the government identified when it entered the case. Certainly not to avoid impermissibly disposing of a valid claim. No, the majority defies Congress’s directions and a century of our precedent all in aid of a position that the federal government has never pleaded, one that works against the government’s decades-old, real-world interests. And the majority does so even when the consent decree would permit the government to raise any valid, independent claims of its own in a different forum.

Where does that leave the States? After 10 years and tens of millions of dollars in lawyers’ fees, their agreement disappears with only the promise of more litigation to follow. All because the government won’t accept a settlement providing it with everything it once sought, and now seeks to promote the use of an alternative 1938 baseline that no party seeks and New Mexico represents could cost it tens of thousands of jobs and a large segment of the State’s economy. “ ‘[C]ooperative federalism’ ” that is not. California, 438 U. S., at 650.

Looking beyond this case to future ones does not brighten the prospect. When the federal government sought to enter the case, it did so “without [Texas’s] objection,” a consideration that carried weight with us. Texas I, 583 U. S., at 415. But in light of the veto power the Court seemingly awards the government over the settlement of an original action, what State in its right mind wouldn’t object to the government’s intervention in future water rights cases? If, as happened here, even heavily caveated permission to intervene may end up federalizing an interstate dispute, what State (or Court) would ever want to risk letting the nose make it under the tent? In that way, too, I fear the majority’s shortsighted decision will only make it harder to secure the kind of cooperation between federal and state authorities reclamation law envisions and many river systems require.

With respect, I dissent.