(concurring in part and dissenting in part). Recognizing that we have neither the authority, nor the ability, nor the will to micromanage education financing, in Campaign for Fiscal Equity v State of New York (100 NY2d 893, 925 [2003] [CFE II]) the Court demarcated standards that must be met, but left it to defendants to come into compliance, affording them more than a year to do so. Regrettably, our trust was misplaced. Today, more than three years later — and more than 13 years after this litigation began — defendants still have failed to fund the New York City public schools adequately. Having failed to satisfy their responsibility, defendants now compel this Court to determine the specific steps that must be taken to remedy the undisputed constitutional violation. Also regret*34tably, I must dissent because the majority does not resolve the inadequate funding of the New York City public schools and reaches a result that is well below what the governmental actors themselves had concluded was required.
Although the dollar differences that separate the majority and dissent are great, our actual points of difference are only two: first, the deference owed, and second, the rationality of two factors used by defendants to calculate the cost of a sound basic education.1
I. The Issue of Deference
The heart of the majority writing is that substantial deference is owed to the executive and legislative branches of government, particularly in matters of budgeting and policymaking (see majority op at 28-30). I agree wholeheartedly. When the Executive and the Legislature have acted together on matters within their particular province, the courts should indeed tread lightly. That, however, is not what happened here. There is no state budget plan for bringing the schools into constitutional compliance; that is precisely the problem. When, as here, the Executive and the Legislature are specifically at odds as to the cost of providing the opportunity for a sound basic education to New York City schoolchildren, the approach of a single branch, rejected by another, cannot legitimately be considered “the State’s estimate” (majority op at 29), and is entitled to no special weight.
In CFE II we directed defendants to ascertain the actual cost of providing a sound basic education in New York City; to ensure that every city school have the resources necessary to provide the opportunity for a sound basic education; and to ensure a system of accountability to measure whether the reforms actually provide the opportunity for a sound basic education. Although there is no dispute that defendants failed to comply with our second directive, I cannot agree with the majority that they complied with our first.
To be sure, the Governor, in undertaking to determine the cost of a sound basic education, and in proposing $4.7 billion in increased annual funding for the New York City public schools, took important steps toward that objective. And in most cases, of course, we can assume that the Governor, in whom the exec*35utive power is vested, speaks for the State. But not so here. The enactment of an appropriation bill that ensures adequate educational funding requires agreement among the Governor and both houses of the Legislature, and plainly that has not occurred.2
As the deadline for compliance approached, defendants advised the trial court, to which we had remitted the case, of the actions they had by then undertaken to satisfy this Court’s order. Based on defendants’ failure to fulfill the CFE II mandate, the court appointed three referees
“to hear and report with recommendations on what measures defendants have taken to follow the [CFE II] directives and bring this State’s school funding mechanism into constitutional compliance insofar as it affects the New York City School System. The referees shall also identify the areas, if any, in which such compliance is lacking.”
Even today, it is hard to see what alternative the court had. The Executive and Legislature were at loggerheads — there was no agreed amount to be implemented.
The majority faults the trial court and referees for undertaking to determine the actual cost of a sound basic education, believing instead that calculations proffered by certain of the defendants should have been beyond question. For more than 200 years, however, it has been the province and duty of the judicial branch to enforce compliance with constitutional norms, including (when necessary) as against the other branches of government. Defendants’ continued failure to cure the violation properly obligated the courts to determine the extent of noncompliance and to direct a remedy.
When courts undertake to resolve a controversy that others have brought before them, they appropriately resort to the tools of the judicial trade — testimony, evidence and fact-finding. The referees thus conducted seven days of evidentiary hearings on the primary question before them — the actual cost of a sound basic education for New York City schoolchildren. Based on the testimony of 15 witnesses, expert evidence, and extensive brief*36ing and argument, they rationally determined that the actual cost of providing a sound basic education in New York City required an annual increase in operational funding of $5.63 billion, calculated in 2004-2005 dollars. The Appellate Division, too, determined that constitutional compliance required additional annual expenditures of between $4.7 billion and $5.63 billion. The majority, however, rejects the findings of the referees, and indeed the very process undertaken by the courts, and instead accepts at face value $1.93 billion as sufficient to satisfy the Education Article.
The origins of that number are instructive. In response to our decision in CFE II, the Governor established the New York State Commission on Education Reform — the Zarb Commission — charged with, among other things, “study[ing] and mak[ing] recommendations regarding . . . [t]he actual cost of providing all children the opportunity to acquire a sound basic education in the public schools of the State of New York” (Executive Order [Pataki] No. 131 [9 NYCRR 5.131]). The Commission, in turn, retained Standard.& Poor’s School Evaluation Services, which produced a lengthy Resource Adequacy Study offering a wide variety of possible amounts of education spending, each based on different assumptions and variables as to the desired level of educational achievement and the costs required to attain it.
Inasmuch as the Commission was asked to study statewide costs, even though this Court’s mandate had been limited to New York City, Standard & Poor’s used each scenario to calculate both statewide amounts and their citywide equivalents. With respect to New York City, Standard & Poor’s calculated resource gaps — defined as the difference between the amount actually spent in the most recently completed fiscal year and the amount that would be required to fund the particular scenario — ranging from $1.93 billion to $7.28 billion in core operating expenditures.3 In so doing, however, Standard & Poor’s itself repeatedly made clear that it “does not recommend any spending level or the adoption of any particular achievement scenario” (Standard & Poor’s School Evaluation Services, Resource Adequacy Study for the New York State Commission on Education Reform, at 23 [2004]). Relying on the Resource Adequacy Study, and specifically endorsing the “cost effectiveness” vari*37able considered by Standard & Poor’s, the Commission similarly reported a wide range to the Governor — that is, from $2.45 billion to $5.57 billion statewide, which translated to between $1.93 billion and $4.69 billion for the City — concluding that “[t]he State’s elected leaders should make a choice of funding within this range” (New York State Commission on Education Reform, Final Report, at 24 [2004]).
At no time did either Standard & Poor’s or the Commission undertake to conclude that any particular amount within the proffered ranges was sufficient to ensure the opportunity for a sound basic education to all New York City schoolchildren (see Standard & Poor’s School Evaluation Services, Resource Adequacy Study for the New York State Commission on Education Reform, at 12-13 [2004] [“Standard & Poor’s does not recommend any particular spending level; nor does it define a ‘sound, basic education’ ”]; New York State Commission on Education Reform, Final Report, at 24 [2004] [“The Commission believes the decision on which educational standard to use should be left to the State’s elected leaders”]). Nevertheless, the majority accepts the calculations performed in the Resource Adequacy Study as sacrosanct, and from there concludes that the lowest number reached using any possible combination of variables must equal the amount needed to provide a sound basic education.
Even defendants did not recommend these numbers. Rather, defendants, through the Attorney General, submitted to the referees an “education plan prepared by Governor Pataki setting forth his proposal for complying with the June 23, 2003 decision of the Court of Appeals in this action.” The Governor’s plan proposed additional annual operating funds for New York City of $4.7 billion, and $8 billion statewide. As the Attorney General wrote in a letter to the referees:
“Governor Pataki is a named defendant in this action, and this office therefore is presenting the plan prepared by the Governor, as his counsel, and a description of that plan as prepared by the Governor’s office. We also note, however, that the decision of the Court of Appeals requires the enactment of legislation by the State of New York, and this office is the institutional counsel for the State.
“Unfortunately, the two houses of the Legislature and the Governor have not been able to agree upon a single unified plan for submission to this panel.”
*38On this record, the $1.93 billion figure is not entitled to special deference.
II. The Issue of Rationality
The $1.93 billion now embraced by the majority is, moreover, based on two unsustainable factors. When those factors are properly adjusted, the resultant cost of a sound basic education, as calculated by Standard & Poor’s, is identical to that reached by the referees and accepted by the trial court — a determination that should therefore be upheld as rationally based.
In conducting its study, Standard & Poor’s, as instructed by the Commission, utilized the “successful schools and districts” model for calculating per-pupil expenditures — one of three distinct methodologies typically used by experts in education policy and finance. No party disputes the legitimacy of the successful schools method as a means for determining the cost of a sound basic education. By using this method, Standard & Poor’s was able to calculate a range of numbers that varied according to choices made as to four distinct factors — first, the standard for measuring a successful school district; second, the additional expenditures necessitated by special needs students; third, the use of a cost filter; and fourth, the manner of converting “standardized” education dollars into New York City dollars.
With respect to the first factor, Standard & Poor’s made a set of calculations using four different academic achievement standards for identifying a successful school district, and again the parties agree, and the referees found, that satisfaction of the Regents Criteria provides an appropriate standard for measuring a successful school. Nor is there any dispute as to the fourth factor — the proper conversion of state education dollars into New York City dollars. Inasmuch as the purchasing power of a dollar varies across the state, and the successful schools method examined all of the school districts in the state, Standard & Poor’s offered two alternative regional cost adjustment indices to determine the cost in New York City dollars: (1) the “New York Regional Cost Index,” provided by the State Education Department; and (2) the “Geographic Cost of Education Index” (GCEI), provided by the National Center for Education Statis*39tics. The referees found, and the parties agree, that use of the GCEI was appropriate.4
The difference — a huge dollar difference — centers on the second and third factors: the proper weighting adjustments for special needs students, and the use of a 50% cost reduction filter.
A. The Low-Income Weighting
A successful schools analysis produces “base expenditures,” which are estimated costs per pupil. However, such base expenditures must then be multiplied by “weightings” for students with special needs, who require such costly accommodations as differentiated curricula, smaller class sizes, assistive technology and classroom aides. Standard & Poor’s assigned a weighting of 2.1 to students with disabilities (special-education students); 1.35 to economically disadvantaged students; and 1.2 to students with limited English proficiency. Although the parties and referees agree that the special-education and English-language weightings were appropriate, the record reflects that the 1.35 low-income weighting applied by Standard & Poor’s— according to which $1.35 must be allocated to students in poverty for every dollar spent on a student not in poverty — was irrational and cannot be sustained.
Defendants’ principal rationale for choosing this weighting — a choice that drew considerable criticism from the witnesses and amici — was that the Standard & Poor’s study had identified 1.35 as the proper adjustment for educating economically disadvantaged students. But Standard & Poor’s had in fact emphasized that 1.35 was simply a figure that it had “drawn from a review of research literature on the coefficients that education agencies tend to use in practice,” and that “insufficient empirical evidence exists in New York to determine how much additional funding is actually needed for different categories of students with special needs to consistently perform at intended achievement levels” (Standard & Poor’s School Evaluation Services, Resource Adequacy Study for the New York State *40Commission on Education Reform, at 8-9 [2004]).5 As a result, Standard & Poor’s made clear that its study “does not explicitly recommend a particular set of weightings.” (Id.)
Unlike Standard & Poor’s, the Regents, in determining that a higher poverty weighting was required, had focused on the specific circumstances of New York City schools, including an especially heavy concentration of high-needs students, very low graduation rates, large classes and a disproportionate number of schools in need of improvement, and thereby determined that the appropriate low-income weighting for New York City was 1.8. With respect to the state as a whole, the Regents recommended weightings for low-income students ranging from 1.5 to 2.0, depending on the concentration of poverty in the district.
Because the Standard & Poor’s weighting of 1.35 for low-income students was not focused on the specific circumstances of New York City schools, its use to determine the actual cost of providing a sound basic education to economically disadvantaged New York City students was irrational, as the referees found. The referees thus properly determined that a poverty weighting of at least 1.5 — the lower end of the range proposed by the Regents — must instead be used.
B. The 50% Cost Filter
Finally, as endorsed by the Zarb Commission, Standard & Poor’s applied a “cost effectiveness filter” of 50% in an alleged effort to screen out successful school districts that either spent money inefficiently or spent more than was necessary to provide the opportunity for a sound basic education. Under that approach, Standard & Poor’s considered the average expenditures of only the lower-spending half of successful school districts in New York State, thereby excluding from the analysis 140 of the 281 successful school districts meeting the Regents Criteria standard.
Multiple witnesses and amici heavily criticized this filter. Indeed, several testifying witnesses criticized the use of any cost reduction filter, and others would have used an approach substantially different from the one adopted by defendants, such as simply eliminating the highest- and lowest-spending 5% *41of districts as “outliers.” Defendants’ own expert, Dr. Robert M. Palaich, testified that his own firm would not use the 50% filter, and there was no evidence offered that this filter is generally accepted by experts in educational finance or, more fundamentally, that the higher-spending districts that were excluded from defendants’ analysis by the cost filter were in fact inefficient. Inasmuch as defendants made no attempt to determine why some successful schools spent less per pupil than others, the assumption that this must have been because the lower-spending schools were more efficient is utterly speculative. Indeed, certain expert amici posited that the lower spending in the selected schools might instead have been due to low wage costs and a low concentration of disadvantaged students, not to efficiency.
Only one decisionmaker anywhere in the country — the New Hampshire Legislature — has ever implemented a 50% cost reduction filter. But as even defendants acknowledge, “in New Hampshire it appears that the efficiency factor was selected to drive costs down” to a predetermined amount; it was not based on the expertise of any education finance experts. The 50% number not only is wholly arbitrary, but also has the effect of eliminating most of the school districts in Westchester and Nassau, the two counties that border New York City and thus most resemble the City in the concentration of students who are not English proficient and in the higher regional costs, particularly in hiring and retaining capable teachers. Accordingly, the 50% cost filter was properly rejected by the referees.
Defendants nevertheless defend the use of the 50% filter on the ground that the State Board of Regents also used it in its own budget proposal, submitted to the referees. But if deference to the Regents is called for with respect to the cost filter, surely it must also be shown with respect to the appropriate poverty weighting. The Regents, as noted, recommended a weighting for New York City low-income students of 1.8. Application of a 1.8 low-income weighting — even with the 50% cost filter — would result in a resource gap of $5.25 billion, almost identical to the $5.26 billion gap found by the referees.6
*42Notably, the Governor proposed additional annual New York City spending of $4.7 billion; the Regents proposed $4.7 billion from the State, plus $0.9 billion from the City, for a total of $5.6 billion; and New York City proposed $5.3 billion. Plainly, every governmental actor knew what the referees and the Appellate Division here concluded: A sound basic education will cost approximately $5 billion in additional annual expenditures. I remain hopeful that, despite the Court’s ruling today, the policymakers will continue to strive to make the schools not merely adequate, but excellent, and to implement a statewide solution.
Judges Rosenblatt, Read and Smith concur with Judge Pigott; Judge Rosenblatt concurs in a separate opinion; Chief Judge Kaye concurs in part and dissents in part in another opinion in which Judge Ciparick concurs; Judge Graffeo taking no part.
Order modified, etc.
. I concur with the majority that no further judicial direction is required with respect to capital improvements or accountability.
. Indeed, our State Constitution places the obligation to provide all the state’s children with a sound basic education squarely on the Legislature: “The legislature shall provide for the maintenance and support of a system of free common schools, wherein all the children of this state may be educated” (NY Const, art XI, § 1 [the Education Article]).
. “Core operations” exclude debt service and capital and transportation spending.
. However, the particular GCEI used by Standard & Poor’s in its Resource Adequacy Study was outdated. Accordingly, as the majority recognizes and defendants concede, the $1.93 billion ordered by the Court must be adjusted using the correct GCEI.
. Evidence in the record indicated that such nationally derived weightings generally “result from guesses or policy decisions based on the amount of available funding” and “are essentially arbitrary and do not reflect the actual costs of providing adequate educational opportunities to students with special needs.”
. The $5.63 billion calculated by the referees reflected both the updated GCEI and an adjustment for inflation from January 2004 to 2004-2005 dollars. Without those corrections, this amount equated to the Standard & Poor’s permutation that resulted in $5.26 billion. Indeed, all figures calculated by Standard & Poor’s and proffered before the referees were measured in Janu*42ary 2004 dollars. As defendants concede, the $1.93 billion endorsed by the majority must be adjusted for inflation.