Justice Stevens
delivered the opinion of the Court.
The Arkansas River rises in the mountains of Colorado just east of the Continental Divide, descends for about 280 miles to the Kansas border, then flows through that State, Oklahoma, and Arkansas and empties into the Mississippi River. On May 20, 1901, Kansas first invoked this Court’s original jurisdiction to seek a remedy for Colorado’s diversion of water from the Arkansas River. See Kansas v. Colorado, 185 U. S. 125, 126 (1902) (statement of case). In opinions written during the past century, most recently in Kansas v. Colorado, 514 U. S. 673, 675-678 (1995), we have described the history and the importance of the river. For present purposes it suffices to note that two of those cases, Kansas v. Colorado, 206 U. S. 46 (1907), and Colorado v. Kansas, 320 U. S. 383 (1943), led to the negotiation of the Arkansas River Compact (Compact), an agreement between Kansas and Colorado that in turn was approved by Congress in 1949. See 63 Stat. 145. The case before us today in[5] volves a claim by Kansas for damages based on Colorado’s violations of that Compact.
The Compact was designed to “[s]ettle existing disputes and remove causes of future controversy” between the two States and their citizens concerning waters of the Arkansas River and to “[ejquitably divide and apportion” those waters and the benefits arising from construction and operation of the federal project known as the “John Martin Reservoir.” Arkansas River Compact, Art. I, reprinted in App. to Brief for Kansas A-l, A-2. Article IV-D of the Compact provides:
“This Compact is not intended to impede or prevent future beneficial development of the Arkansas River basin in Colorado and Kansas by Federal or State agencies, by private enterprise, or by combinations thereof, which may involve construction of dams, reservoir, and other works for the purpose of water utilization and control, as well as the improved or prolonged functioning of existing works: Provided, that the waters of the Arkansas River, as defined in Article III, shall not be materially depleted in usable quantity or availability for use to the water users in Colorado and Kansas under this Compact by such future development or construction.” Id., at A-5.
It is the proviso to that paragraph that is of special relevance to this case.
In 1986, we granted Kansas leave to file a complaint alleging three violations of the Compact by Colorado. See 514 U. S., at 679-680. After taking evidence in the liability phase of the proceeding, Special Master Arthur L. Little-worth filed his first report, in which he recommended that two of the claims be denied, but that the Court find, that post-Compact increases in groundwater well pumping in Colorado had materially depleted the waters of the river in violation of Article IV-D. See id., at 680. We overruled [6] Colorado’s exceptions to that recommendation, including an argument that Kansas was guilty of laches. Id., at 687-689. We remanded the ease to the Special Master to determine: an appropriate remedy for the violations of Article IV-D. Id., at 694. ,
After further proceedings the Special Master filed a second report recommending an award of damages. Colorado filed exceptions to that report, arguing that the Eleventh Amendment barred an award based on losses incurred by Kansas citizens, and that the report improperly recommended the recovery of prejudgment interest on an un-liquidated claim. We overruled those exceptions without prejudice to their renewal after the Special Master made a more specific recommendation for a remedy. 522 U. S. 1073 (1998). He did so in his third report, and we are now confronted with exceptions filed by both States.
In the third report, the Special Master recommends that damages be measured by Kansas’ losses, rather than Colorado’s profits, attributable to Compact violations after 1950; that the damages be paid in money rather than water; and that the damages should include prejudgment interest from 1969 to the date of judgment. Colorado has filed four objections to the report. It contends (1) that the recommended award of damages would violate the Eleventh Amendment to the United States Constitution; (2) that the damages award should not include prejudgment interest; (3) that the amount of interest awarded is excessive; and (4) that the Special Master improperly credited flawed expert testimony, with the result that Kansas’ crop production losses were improperly calculated. On the other hand, Kansas has filed an objection submitting that prejudgment interest should be paid from 1950, rather than 1969. The United States, which intervened because of its interest in the operation of flood control projects in Colorado, submits that both States’ objections should be overruled.
[7]*71 — 1
We have decided that a State may recover monetary damages from another State in an original action, without running afoul of the Eleventh Amendment.1 See, e.g., Texas v. New Mexico, 482 U. S. 124, 130 (1987) (“The Court has recognized the propriety of money judgments against a State in an original action, and specifically in a case involving a compact. In proper original actions, the Eleventh Amendment is no barrier, for by its terms, it applies only to suits by citizens against a State” (citations omitted)); see also Maryland v. Louisiana, 451 U. S. 725, 745, n. 21 (1981); South Dakota v. North Carolina, 192 U. S. 286, 317-321 (1904). Colorado contends, however, that the Eleventh Amendment precludes any such recovery based on losses sustained by individual water users in Kansas.
It is firmly established, and undisputed in this litigation, that the text of the Eleventh Amendment would bar a direct action against Colorado by citizens of Kansas. Moreover, we have several times held that a State may not invoke our original jurisdiction when it is merely acting as an agent or trustee for one or more of its citizens. For example, in New Hampshire v. Louisiana, 108 U. S. 76 (1883), we refused to assume jurisdiction over an action to recover payment, on defaulted bonds that had been formally assigned to the state plaintiffs but remained beneficially owned by private individuals. And, in North Dakota v. Minnesota, 263 U. S. 365 (1923), we held that, while the plaintiff State could obtain an injunction against the improper operation of Minnesota’s drainage ditches, the Eleventh Amendment precluded an award of damages based on injuries to individual farmers, where the damages claim was financed by [8] contributions from the farmers and the State had committed to dividing any recovery among the farmers “in proportion to the amount of [their] loss.” Id., at 375.
Those cases make it clear that a “State is not permitted to enter a controversy as a nominal party in order to forward the claims of individual citizens.” Maryland v. Louisiana, 451 U. S., at 737; see also New Hampshire v. Louisiana, 108 U. S., at 89 (Eleventh Amendment applies and acts to bar jurisdiction where “the State and the attorney-general' are only nominal actors in the proceeding”). The “governing principle” is that in order to invoke our original jurisdiction, “the State must show a direct interest of its own and not merely seek recovery for the benefit of individuals who are the real parties in interest.” Oklahoma ex rel. Johnson v. Cook, 304 U. S. 387, 396 (1938).
Kansas has unquestionably made such a showing. Indeed, the present proceeding is but one of several in which Kansas’ own interest in preventing upstream diversions from the Arkansas River has justified an exercise of our original jurisdiction. In Cook we even offered as an example of proper original jurisdiction one of the prior original suits between Kansas and Colorado, see id., at 393-394 (citing Kansas v. Colorado, 206 U. S. 46 (1907)), and in Texas v. New Mexico we held that enforcement of an interstate water compact by means of a recovery of money damages can be within a State’s proper pursuit of the “general public interest” in an original action, 482 U. S., at 132, n. 7.
Moreover, the record in this case plainly discloses that the State of Kansas has been in full control of this litigation since its inception. Its right to control the disposition of any recovery of damages is entirely unencumbered. The injury to individual farmers is but one component of the formula adopted by the Special Master to quantify the damages caused by Colorado’s violation of its contractual obligations. In short, there is simply nothing in the record to suggest that the State of Kansas is merely a “nominal party” to this [9] litigation or that the individual farmers are “the real parties in interest.”
When a State properly invokes our jurisdiction to seek redress for a wrong perpetrated against it by a sister State, neither the measure of damages that we ultimately determine to be proper nor our method for calculating those damages can retrospectively negate our jurisdiction. Nor would our jurisdiction to order a damages remedy be affected by Kansas’ postjudgment decisions concerning the use of the money recovered from Colorado. As we have previously recognized, it is the State’s prerogative either to deposit the proceeds of any judgment in “the general coffers of the State” or to use them to “benefit those who were hurt.” Ibid.
We overrule Colorado’s first exception.
II
Colorado next excepts to the Special Master’s conclusion that the damages award should include prejudgment interest despite the fact that Kansas’ claim is unliquidated.2 At one point in time, the fact that the claim was unliquidated would have been of substantial importance. As a general matter, early common-law cases drew a distinction between liquidated and unliquidated claims and refused to allow interest on the latter. See, e.g., Comment, Prejudgment Interest: [10] Survey and Suggestion, 77 Nw. U. L. Rev. 192,196, and n. 26 (1982) (discussing history and collecting sources). This rule seems to have rested upon a belief that there was something inherently unfair about requiring debtors to pay interest when they were unable to halt its accrual by handing over to their creditors a fixed and unassailable amount. See, e. g., id., at 196.
This common-law distinction has long since lost its hold on the legal imagination. Beginning in the early part of the last century, numerous courts and commentators have rejected the distinction for failing to acknowledge the compensatory nature of interest awards.3 This Court allied itself with the evolving consensus in 1933, when we expressed the opinion that the distinction between cases of liquidated and unliquidated damages “is not a sound one.” Funkhouser v. J B. Preston Co., 290 U. S. 163, 168 (1933). The analysis supporting that conclusion gave no doubt as to our reasoning: ‘‘Whether the case is of the one class or the other, the injured party has suffered a loss which may be regarded as not fully compensated if he is confined to the amount found to be recoverable as of the time of breach and nothing is added for the delay in obtaining the award of damages.” Ibid. Our cases since 1933 have consistently acknowledged that a monetary award does not fully compensate for an injury unless it includes an interest component. See, e. g., Milwaukee v. Cement Div., National Gypsum Co., 515 U. S. 189, 195 (1995) (“The essential rationale for awarding pre[11] judgment interest is to ensure that an injured party is fully compensated for its loss”); West Virginia v. United States, 479 U. S. 305, 310-311, n. 2 (1987); General Motors Corp. v. Devex Corp., 461 U. S. 648, 655-656, n. 10 (1983).
Relying on our cases, the Special Master “concluded that the unliquidated nature of Kansas’ money damages does not, in and of itself, bar an award of prejudgment interest.” Second Report, § XV, reprinted in App. to Third Report 43. In reaching that conclusion, the Special Master was fully cognizant of both the displaced common-law rule and the subsequent doctrinal evolution. In addition, he gave careful consideration to equitable considerations that might mitigate against an award of interest, concluding that “considerations of fairness,” Board of Comm’rs of Jackson Cty. v. United States, 308 U. S. 343, 352 (1939), supported the award of at least some prejudgment interest in this case.
We find no fault in the Special Master’s analysis of either our prior cases or the equities of this matter. While we will deal with the amount of prejudgment interest below, to answer Colorado’s second objection it is sufficient to conclude that the Special Master was correct in determining that the unliquidated nature of the damages does not preclude an award of prejudgment interest.4
[12] Colorado’s second exception is overruled.
III
Colorado’s third exception takes issue with both the rate of interest adopted by the Special Master and the date from which he recommended that interest begin to accrue. As to the second of these two concerns, Colorado submits that, if any prejudgment interest is to be awarded, it should begin to accrue in 1985 (when Kansas filed its complaint in this action), rather than in 1969 (when, the Special Master concluded, Colorado knew or should have known that it was violating the Compact). On the other hand, Kansas has entered an exception, arguing that the accrual of interest should begin in 1950. We first address the rate question, then the timing issue.
A
The Special Master credited the testimony of Kansas’ three experts who calculated the interest rates that they thought necessary to provide full compensation for the damages caused by Colorado’s violations of the Compact in the years since 1950. As a result of inflation and changing market conditions those rates varied from year to year. In their calculation of the damages suffered by Kansas farmers, the experts used the interest rates that were applicable to individuals in the relevant years rather than the (lower) rates available to States.
[13] Colorado argues that the lower rates should have been used because it is the State, rather than the individual farmers, that is maintaining the action and will receive any award of damages. But if, as we have already decided, see Part I, supra, it is permissible for the State to measure a portion of its damages by losses suffered by individual farmers, it necessarily follows that the courts are free to utilize whatever interest rate will most accurately measure those losses. The money in question in this portion of the damages award is revenue that would — but for Colorado’s actions — have been earned by individual farmers. Thus, the Special Master correctly concluded that the economic consequences of Colorado’s breach could best be remedied by an interest award that mirrors the cost of any additional borrowing the farmers may have been forced to undertake in order to compensate for lost revenue.
B
Although the Special Master rejected Colorado’s submission that there is a categorical bar to the award of prejudgment interest on unliquidated claims, he concluded that such interest should not “be awarded according to [any] rigid theory of compensation for money withheld,” but rather should respond to “ ‘considerations of fairness.’ ” Third Report 97 (quoting Jackson Cty., 308 U. S., at 352). Kansas argues that our decisions subsequent to Jackson County have effectively foreclosed the equities-balancing approach that the Special Master adopted. There is some merit to Kansas’ position. See National Gypsum Co., 515 U. S., at 193 (affirming a decision of the Court of Appeals that had read our cases as “disapproving of a ‘balancing of the equities’ as a method of deciding whether to allow prejudgment interest”).
However, despite the clear direction indicated by some of our earlier opinions, we cannot say that by 1949 our case law had developed sufficiently to put Colorado on notice [14] that, upon a violation of the Compact, we would automatically award prejudgment interest from the time of injury. Given the state of the law at that time, Colorado may well have believed that we would balance the equities in order to achieve a just and equitable remedy, rather than automatically imposing prejudgment interest in order to achieve full compensation. See Jackson Cty., 808 U. S., at 352 (prejudgment interest award limited by “considerations of fairness”); Miller v. Robertson, 266 U. S. 243, 258 (1924) (“[W]hen necessary in order to arrive at fair compensation, the court in the exercise of a sound discretion may include interest or its equivalent as an element of damages” on un-liquidated claims); Restatement of Contracts §337, p. 542 (1932) (prejudgment interest on unliquidated claims “may be allowed in the discretion of the court, if justice requires it”). While we are confident that, when it signed the Compact, Colorado was on notice that it might be subject to prejudgment interest if such interest was necessary to fashion an equitable remedy, we are unable to conclude with sufficient certainty that Colorado was on notice that such interest would be imposed as a matter of course. We, therefore, believe that the Special Master acted properly in carefully analyzing the facts of the case and in only awarding as much prejudgment interest as was required by a balancing of the equities.
We also agree with the Special Master that the equities in this case do not support an award of prejudgment interest from the date of the first violation of the Compact, but rather favor an award beginning on a later date. In reaching this conclusion, the Special Master appropriately considered several factors. In particular, he relied on the fact that in the early years after the Compact was signed, no one had any thought that the pact was being violated. Third Report 106. In addition, he considered the long interval that passed between the original injuries and these proceedings, as well as the dramatic impact of compounding interest over many years. Id., at 99-101; see also n. 3, supra.
[15] In its exception, Kansas argues that the Special Master’s reasoning would be appropriate if damages were being awarded as a form of punishment, but does not justify a refusal to provide full compensation to an injured party. Moreover, Kansas argues, a rule that rewards ignorance might discourage diligence in making sure that there is full compliance with the terms of the Compact. Kansas’ argument is consistent with a “rigid theory of compensation for money withheld,” but, for the reasons discussed above, we are persuaded that the Special Master correctly declined to adopt such a theory. The equitable considerations identified by the Special Master fully justify his view that in this case it would be inappropriate to award prejudgment interest for any years before either party was aware of the excessive pumping in Colorado.
In its third exception, Colorado argues that, if prejudgment interest is to be awarded at all, the equities are best balanced by limiting such interest to the time after the complaint was filed, rather than the time after which Colorado knew or should have known that it was violating the Compact. Specifically, Colorado suggests that prejudgment interest should begin to accrue in 1985 rather than 1969. The choice between the two dates is surely debatable; it is a matter over which reasonable people can — and do — disagree. After examining the equities for ourselves, however, a majority of the Court has decided that the later date is the more appropriate.5
[16] When we overruled Colorado’s objections to the Special Master’s first report, we held that Kansas was not guilty of inexcusable delay in failing to complain more promptly about post-Compact well pumping. 514 U. S., at 687-689. In saying that the delay was not inexcusable, we recognized that the nature and extent Of Colorado’s violations continued to be unclear even in the years after which it became obvious that the Compact was being violated. Id., at 688-689. That conclusion is something of a two-edged sword, however. While Kansas’ delay was understandable given the amorphous nature of its claims, there is no doubt that the interests of both States would have been served if the claim had been advanced promptly after its basis became known. Once it became obvious that a violation of the Compact had occurred, it was equally clear that the proceedings necessary to evaluate the significance of the violations would be complex and protracted. Despite the diligence of the parties and the Special Master, over 15 years have elapsed since the complaint was filed. Given the uncertainty over the scope of damages that prevailed during the period between 1968 and 1985 and the fact that it was uniquely in Kansas’ power to begin the process by which those damages would be quantified, Colorado’s request that we deny prejudgment interest for that period is reasonable.
For these reasons, we overrule Kansas’ exception. We also overrule Colorado’s third exception insofar as it challenges the interest rates recommended by the Special Master, but we sustain that objection insofar as it challenges the award of interest for the years prior to 1985.
IV
Colorado’s final objection challenges the Special Master’s determination of the value of the crop losses attributable to the Compact violations, the largest component of Kansas’ damages claim. The Special. Master accomplished the calculation by estimating the amount of farmland affected by [17] Colorado’s violations, the crops planted on that farmland, the price of those crops, and the difference in yield between what the affected land would have produced with the additional water and what the land actually produced with the water it received.
The parties were in agreement concerning most of the facts bearing on the Special Master’s calculation. They agreed that water was in short supply in the affected area each year, 178 Tr. 127-128; they agreed on the amount of the shortage that resulted from Colorado’s violations, ibid.; and they generally agreed on which crops were planted on the affected farmland as well as the prices of those crops in the relevant years. See Third Report 46.
The only issue on which the parties disagreed was the exact effect of the diverted water on the crop yields for the farmland in question. On that score, both Kansas and Colorado accepted the general notion that “[u]p to the point where crops no longer can make use of additional water, more water produces more crop yield.” Id., at 47. But they parted ways on the question of precisely how much additional yield would have been produced with the missing water.6 Kansas’ experts relied upon the hypothesis of a generally linear relationship between water available for [18] use and increased crop yields. With figures drawn from a number of studies, Professor Norman Whittlesey, Kansas’ principal expert,7 developed quantitative estimates of the lost yield, per unit of water, for the various crops grown on the affected farmland. Although Colorado’s expert initially attempted to propose his own model, he ultimately abandoned his position when confronted with flaws in his data. 197 Tr. 44-46.
Its own expert having recanted his alternative proposal for calculating the effects of the diverted water on crop yield, Colorado attempts to poke holes in Kansas’ methodology through a speculative application of abstract economic theory. Kansas’ numbers (for crop losses due to diverted water) cannot be correct, Colorado argues, because if they were, it would have been economically profitable for the affected farmers to drill wells and obtain water from underground sources rather than suffer the reduced yield from the shortage of surface water. Brief for Colorado 41-49. Because Kansas farmers did not install wells, Colorado concludes, we can know that the diverted water was not as valuable as Kansas’ experts claim.
The Special Master did not question Colorado’s assertion that digging wells would, in retrospect, actually have been profitable for Kansas farmers, but he declined to employ Colorado’s argument as a basis for rejecting Kansas’ expert testimony on the extent of crop losses. His thoughtful analysis is worth quoting in full:
“Given the hindsight of present day economists, it might have been profitable for everyone to drill supplemental wells .... However, there are many reasons why this may not have been done, and the failure to drill wells does not by itself indicate that Kansas’ [19] estimate of crop losses is too high. The favorable economics of drilling wells may not have been understood at the time. Quality information regarding costs and returns was not readily available. [211 Tr.] 31. Some farmers, for reasons of age or otherwise, may not have wanted to go into long-term debt. Some farmers may not have had the available capital, or the credit to borrow. Many farmers were ‘cash poor.’ Id. at 32. Some farmers may have been averse to risk. Some farmers may have been tenants, and the landlord may not have been willing to undertake the necessary investment. Some farms may have been small in terms of total acreage, or the acreage spread out over space, so that it was not feasible or practical to consider a well investment. [208 Tr.] 37-39. Capital for well investments, with three to ten year repayment periods, was less available than for long-term investments. [211 Tr.] 32.” Third Report 60-61.
We agree with the Special Master that accepting Colorado’s argument requires a good deal of speculation, not only about the comparative advantages of wells as opposed to irrigation, but also about the ability of the farmers fully to understand or to implement different choices without the benefit of expert hindsight. Given Colorado’s inability to mount an effective challenge to Kansas’ experts on their own terms and its complete failure to provide a plausible alternative estimate of the crop damage that resulted from its violations of the Compact, we conclude that its attack on Kansas’ conclusions is unpersuasive.8
[20] Colorado’s fourth exception is overruled.
V
We remand the ease to the Special Master for preparation of a final judgment consistent with this opinion.
It is so ordered.