Executive Order (Cuomo) No. 38 (EO38), issued by the Governor in 2012, purportedly sought to curb “abuses involving public funds” by limiting the use of state monies for “excessive administrative costs and outsized compensation for . . . senior executives” (9 NYCRR 8.38). “[T]he commissioner of each Executive State agency that provides State financial assistance or State-authorized payments to providers of services,” including the Commissioner of Health, was instructed to promulgate regulations, that, at a minimum, (1) limited allocation of state funds to administrative costs to 15% by 2015; and (2) prohibited the use of state funds for executive compensation in an amount greater than $199,000 per year (id.). This latter *272 figure was based on the highest salary paid to an executive on the federal pay scale (id. § 8.38 [2] [b]; Matter of LeadingAge N.Y., Inc. v Shah, 153 AD3d 10, 14 [3d Dept 2017]).

In response to the Governor's order, DOH promulgated 10 NYCRR part 1002, placing restrictions on “covered providers” that derive over $500,000 from state funds and earn at least 30% of total annual in-state revenue from those funds for the covered reporting period and year prior (see 10 NYCRR 1002.1 [d]). Excluded from this definition, and the regulations' restrictions, are “[s]tate, county, and local governmental units in New York State, and tribal governments for the nine New York State recognized nations, and any subdivisions or subsidiaries of the foregoing entities” (id. § 1002.1 [d] [6] [i]).

The regulations track EO38 in requiring the same ratio of program service expenses and administrative expenses (see 9 NYCRR 8.38 [2] [a]; 10 NYCRR 1002.2 [a]). They also adopt the salary figure taken from the federal pay scale in denying state funds for executive compensation in an amount greater than $199,000 (10 NYCRR 1002.3 [a]). DOH added a provision, the so-called “soft cap,” which subjects a covered provider to penalties if executive compensation exceeds $199,000 per year from any source (majority op at 255-256).

Section 1002.1 defines “administrative expenses” as those expenses “incurred in connection with the covered provider's overall management and necessary overhead that cannot be attributed directly to the provision of program services,” and includes the “portion of the salaries and benefits of staff performing administrative and coordination functions that cannot be attributed to particular program services” (10 NYCRR 1002.1 [a]).

As the majority notes, the relevant statutes provide DOH with the authority and the responsibility to ensure “that the limited public funding available be directed as efficiently as possible toward high-quality services for New Yorkers in need” (majority op at 262; see Matter of Medicon Diagnostic Labs. v Perales, 74 NY2d 539, 545 [1989]; Public Health Law §§ 201 [1] [o], [p]; 206 [3]). As a means to that end, DOH chose to limit the amount of administrative expenses that may be reimbursed from public funds to ensure those funds are spent directly on program services. By limiting administrative expenses only where they divert state funds away from the actual provision *273 of health care services, the regulation advances the goals set forth in the relevant statutes. As a result, I agree with the majority that the limit on administrative expenses is constitutional.

I also agree with the majority that DOH, in attempting to restrict a covered provider's use of private funds through the soft cap on executive compensation, “ventured beyond legislative directives relating to the efficient use of state funds and into the realm of broader public policy concerns” (majority op at 268). A broad grant of authority is not a license to resolve matters of social or public policy reserved to legislative bodies (Boreali v Axelrod, 71 NY2d 1, 9 [1987]). For the same reasons, the hard cap on executive compensation should be struck down.

Although the majority combines the administrative expenses cap with the state funding limitation on executive compensation under the “hard cap” heading, each of those regulations must be considered individually. *13 Indeed, the restrictions are found in separate regulatory provisions (10 NYCRR 1002.2 [limits on administrative expenses], 1002.3 [limits on executive compensation]). Once the administrative expenses cap passes muster, it becomes clear that the “hard cap” on salary has no effect on whether “limited public funding . . . [is] directed as efficiently as possible toward high-quality services for New Yorkers in need” (majority op at 262).

Under the administrative expenses cap, a covered provider must spend at least 85% of funding on “program services expenses,” rather than “administrative expenses” (10 NYCRR 1002.2 [a]). “Administrative expenses” include, by definition, “that portion of the salaries and benefits of staff performing administrative and coordination functions that cannot be attributed to particular program services, including but not limited to the executive director or chief executive officer, financial officers such as the chief financial officer” (id. § 1002.1 [a] [1] [i] [emphasis supplied]). Executive compensation is defined to include “all forms of cash and noncash payments or benefits given directly or indirectly to a covered executive, including but not limited to salary and wages, bonuses, dividends” (id. § 1002.1 [g] [emphasis supplied]). State funds cannot be used to pay a covered executive more than $199,000 (id. § 1002.3 [a]). Accordingly, once the administrative cap controls, the executive compensation cap has no effect on money spent on program services. For example, assume a covered provider is meeting and exceeding the expense ratio at 90% of state *274 funds going directly to program services. But that same company, perhaps as a reward for achieving such performance, is paying a covered executive $249,000. That executive spends half of his or her time on program services and half on administrative expenses. As a result of the regulation, the company cuts $50,000 from the executive's compensation. That company is free to direct all of that “saved” money to administrative expenses, assuming it remains under the 15% administrative cap—effectively diverting state funds from program services. Put another way, as the Coalition of New York State Public Health Plans petitioners point out,

Nevertheless, DOH argues that “excessive executive compensation and administrative expenses will divert scarce state resources away from the needy New Yorkers who are the intended beneficiaries of such funding.” But the hard cap on compensation does not require that any additional funds be used for program services—it is not concerned with directing money toward “needy New Yorkers” but away from “overcompensated” executives. This is precisely the type of social policymaking prohibited by this Court's prior decisions (see Matter of Campagna v Shaffer, 73 NY2d 237, 242-243 [1989]; Council for Owner Occupied Hous. v Abrams, 72 NY2d 553, 558 [1988]; Under 21, Catholic Home Bur. for Dependent Children v City of New York, 65 NY2d 344, 359 [1985]; Subcontractors Trade Assn. v Koch, 62 NY2d 422, 428 [1984]; Matter of Fullilove v Beame, 48 NY2d 376, 379 [1979]; Matter of Broidrick v Lindsay, 39 NY2d 641, 646-647 [1976]).^*

This is not a new rule. In Boreali, we concluded that the agency had “stretched [the] statute beyond its constitutionally valid reach when it used the statute as a basis for drafting a code embodying its own assessment of what public policy ought to be” (71 NY2d at 9). And we prefaced our discussion of the four “coalescing circumstances” by noting that the analysis of those factors would “persuade us that the difficult-to-define line between administrative rule-making and legislative policy-making has been transgressed” (id. at 11). Indeed, prior to Boreali, we noted that “the general power to enter into contracts which is bestowed upon the executive branch of government ordinarily cannot serve as a basis for creating a remedial plan for which the executive never received a grant of legislative power” (Subcontractors Trade Assn., 62 NY2d at 428; see also Under 21, Catholic Home Bur. for Dependent Children, 65 NY2d at 359 [“(A)n executive may not usurp the legislative function by enacting social policies not adopted by the Legislature”]). It was not conflict with the authorizing legislation that doomed the executive policymaking in those cases, but rather an unlawful assumption of power by the executive in setting that policy (see Under 21, Catholic Home Bur. for Dependent Children, 65 NY2d at 358).

Although the majority characterizes the hard cap as a permissible “balancing of costs and benefits” (majority op at 265), the decision to cap executive compensation is a policy choice made by DOH from competing views of a hotly debated topic (compare Richard A. Posner, Are American CEOs Overpaid, and, if So, What if Anything Should be Done About it?, 58 Duke LJ 1013, 1045 [2009] [“Placing a ceiling on CEO salaries and other compensation would be a mistake. Apart from the infeasibility of a government agency's determining the amount of water in an executive's pay, capping salaries by government fiat, like other regulatory price controls, would incite wasteful activities that would be more costly to society than overcompensation”], *276 with Charles C. Pak, Toward Reasonable Executive Compensation: Outcry for Reform and Regulatory Response, 1994 Ann Surv Am L 633, 659 [1995] [“The simplest response to excessive executive compensation would be to prohibit remuneration above a predetermined level or to impose a penalty for pay in excess of a specified level”]). This debate gained momentum following the 2008 financial crisis (see e.g. William O. Fisher, To Thine Own CEO Be True: Tailoring CEO Compensation to Individual Personality and Circumstances, 2017 Colum Bus L Rev 599 [2017]; Steven A. Bank et al., Executive Pay: What Worked?, 42 J Corp L 59 [2016]; Robert E. Wagner, Mission Impossible: A Legislative Solution for Excessive Executive Compensation, 45 Conn L Rev 549 [2012]; Stuart Lazar, The Unreasonable Case for a Reasonable Compensation Standard in the Public Company Context: Why It Is Unreasonable to Insist on Reasonableness, 59 Buff L Rev 937 [2011]; Omari Scott Simmons, Taking the Blue Pill: The Imponderable Impact of Executive Compensation Reform, 62 SMU L Rev 299 [2009]).

A policy choice about reasonable executive compensation aimed at influencing corporate behavior is law-making beyond DOH's regulatory authority (see Matter of New York Statewide Coalition of Hispanic Chambers of Commerce v New York City Dept. of Health & Mental Hygiene, 23 NY3d 681, 698-699 [2014] [the regulation “evidenced a policy choice relating to the question of the extent to which government may legitimately influence citizens' decision-making . . . . By choosing between public policy ends in these ways, the Board of Health engaged in law-making beyond its regulatory authority”]). DOH's legislative authority related to fiscal oversight cannot justify these regulations. We have made clear that “[e]ven under the broadest and most open-ended of statutory mandates, an administrative agency may not use its authority as a license to correct whatever societal evils it perceives” (Boreali, 71 NY2d at 9). No agency, through regulation, can “effect its vision of societal policy choices” (Matter of New York State Health Facilities Assn. v Axelrod, 77 NY2d 340, 346 [1991]). DOH, with its caps on compensation that have no effect on money directed to program services, has done just that.

In addition to its constitutional infirmity, the hard cap is arbitrary and capricious. As noted above, DOH fails to demonstrate how the hard cap will ensure that public funds are *277 directed to program services rather than any other form of administrative expenses. Further, the benchmark for determining “excessive” compensation is itself arbitrary and capricious. The $199,000 cap was not the result of any compensation study of the health care industry, but instead was lifted from EO38, which for unknown reasons identified the highest salary paid to federal employees at Level I of the Executive Schedule as the benchmark for “outsized” executive compensation for all “providers of services” (see 9 NYCRR 8.38). The majority posits that the “research” conducted by the Governor's task force, convened to investigate nonprofit spending, provides “a rational basis for the hard caps” (majority op at 258, 267). However, as noted *14 in the record, the task force never produced any report supportive of a $199,000 cap (see testimony of Doug Sauer, Chief Executive Officer, New York Council of Nonprofits, Inc., Feb. 6, 2012, at 3 [“The Governor's Executive Order later established a $199,000 threshold presumably based on the recent survey. The data compiled in this survey still has not been publically released. I do not know where the $199,000 figure came from”]). The majority's statement that “[t]he task force's discovery of excessive executive compensation within the health care industry provided a rational basis for the hard caps” (majority op at 267) is likewise unsupported by any citation or authority.

In an attempt to rationalize the caps on executive compensation, DOH argued: “[i]f a provider spends extravagantly on executive compensation—even out of private funds—it may be a red flag that the provider will spend state funds irresponsibly as well” (emphasis supplied). Yet, the regulation exempts certain entities from the cap, including state health care providers (10 NYCRR 1002.1 [d] [6] [i]). As noted in the testimony before the Senate Committee on Investigations and Government Operations, state executives earn upwards of seven figure salaries (see Sauer at 3). It is unclear why salaries greater than $199,000 for executives of state-run providers would not raise any “red flags.” Moreover, the “caps have not been demonstrated through ‘rational, documented, empirical determination’ to have any connection to reducing state health care expenditures or to the more efficient use of taxpayer funds in the delivery of services” (LeadingAge, 153 AD3d at 31 [Mulvey, J., concurring in part and dissenting in part], citing New York State Assn. of Counties v Axelrod, 78 NY2d 158, 168 [1991]).

*278 We have never before allowed an agency, by way of the power of the purse, to direct social policy (see Campagna, 73 NY2d at 242-243; Council for Owner Occupied Hous., 72 NY2d at 558; Under 21, Catholic Home Bur. for Dependent Children, 65 NY2d at 359; Subcontractors Trade Assn., 62 NY2d at 428; Matter of Fullilove, 48 NY2d at 379; Matter of Broidrick, 39 NY2d at 646-647). Under the majority's rule, the executive is now able to do so through the contracting authority of the State. Whoever the executive, whatever the policy, regulation is permissible if implemented under the guise of oversight of public funds—or “getting the biggest bang for the buck” to use the majority's phrase—regardless of whether the regulations do in fact further fiscal responsibility.

Governor Clinton was not one to take the separation of powers lightly: his inaugural address to the legislature dwelt on the 1777 Constitution's “marking the line between the Executive, Legislative, and Judicial powers;”² he called upon each of the branches of government to “remain within the several departments in which the constitution has placed us, and *279 thereby preserve the same inviolate,” promising that it “shall always be my strenuous endeavor, on the one hand, to retain and exercise for the advantage of the people, the powers with which they have invested me; on the other, carefully to avoid the invasion of those rights which the constitution has placed in other persons” (George Clinton, No. 752, Governor's Address to the Legislature, dated Sept. 10, 1777, 2 Public Papers of George Clinton 297, 300 [1900]).

I dredge up this history to highlight just how far we have drifted from our Constitution's text and history in today's decision, a decision applying the unique four-factor separation of powers test laid out in Boreali v Axelrod (71 NY2d 1, 9 [1987]) to one of the most fundamental components of “the executive power”: the choice of which private persons will receive state funding to provide public services. Boreali's separation of powers analysis has been followed by no other jurisdiction, and for good reason: it is unhelpful in cases to which it applies, and this is not even one of those cases.

Today, ostensibly following Boreali, the majority holds that the Constitution renders one part of a regulation valid but another, virtually identical, part of the same regulation invalid. Worse still, the majority performs this strange regulatory surgery to an executive order on a subject that has been part of the executive power since the dawn of an independent New York: selecting vendors to provide government services to execute the legislature's command, implicit in every appropriation, of giving the State “the biggest bang for the buck” (Matter of LeadingAge N.Y., Inc. v Shah, 56 Misc 3d 594, 604 [Sup Ct, Albany County 2015]). The present Governor, like his 1777 predecessor, seeks by the regulation at issue in this case “to retain and exercise for the advantage of the people, the powers with which they have invested me.” By misapplying our dubious Boreali framework to a wholly inapplicable circumstance, we have deprived him of that power.

I agree with most of the majority's exposition of the background of the case. I also agree that the “hard” cap does not violate the separation of powers doctrine. However, I do not agree with the majority that the soft cap is *15 invalid, or that Boreali applies to this case. Indeed, it is difficult to even articulate my disagreement with the majority on the “soft” cap using the Boreali framework, because Boreali has become so encumbered with imprecise and inaccurate verbiage that it no longer provides a useful framework even for disagreement. Judge Bellacosa, *280 dissenting in Boreali, warned that Boreali's “dramatic[ ] change[ ]” to “the principles of ordinary statutory construction . . . will come back to haunt us” (71 NY2d at 19). It has.

Even on its own terms, Boreali does not apply to this case. Boreali is invoked when the Court applies limiting constructions to statutes that, read literally, would give the executive branch unconstitutionally excessive power. Here, both the majority's opinion and much of the parties' arguments boil down to a non-constitutional question: is the challenged executive action (here a regulation) authorized by the cited statutes? That question cannot trigger a Boreali analysis, or we would apply Boreali to every statutory interpretation case to which the government is a party.³ This very session, we considered whether the Department of Labor's executive actions (there, a letter explaining the considerations governing enforcement actions the Department might undertake at public construction sites) were consistent with its enabling statute, and no one mentioned Boreali (International Union of Painters & Allied Trades, Dist. Council No. 4 v New York State Dept. of Labor, 32 NY3d 198 [2018]).

I am not even sure the majority believes Boreali applies here. After dutifully walking through the Boreali factors to uphold the hard cap, the majority largely abandons Boreali thereafter. Its soft cap analysis contains not one peep about the Boreali factors except for the bald statement that “the first two Boreali factors are the most instructive” and a cryptic footnote declaring Boreali factor four is “at best . . . neutral” (majority op at *281 268, 270 n 10).⁴ Factor three is *16 AWOL. Instead, the majority accuses the Department of Health (DOH) of embracing policy considerations “not clearly connected to the objectives outlined by the legislature” (majority op at 268), because the soft cap “imposes a restriction . . . not sufficiently tethered to the enabling legislation” (id.). Even if that accusation were well-founded, it is not connected to concerns about separation of powers but rather the run-of-the-mill question of whether an agency's actions are contrary to law—the same question we ask in most agency-related cases we hear.

Boreali was never intended to be the touchstone for any challenge to the lawfulness of any executive action. It originated from a quite different problem than the one the majority identifies, namely ambiguous statutes with the potential to be read as engaging in a forbidden “delegation” of the legislative power to the executive by adopting laws empowering the executive with insufficient particularity.⁵ In those circumstances, we held in Boreali and its progeny, the canon of constitutional avoidance requires us to select a narrower construction of the statute—perhaps even artificially narrow (see generally Matter of Jacob, 86 NY2d 651, 680 [1995, Bellacosa, J., dissenting])—to avoid declaring the legislation unconstitutional. Thus, in Boreali, the State's Public Health Council (PHC) acted pursuant to Public Health Law § 225 (5) (a), which permitted it to promulgate a sanitary code that “may . . . deal with any matters affecting the security of life or health or the preservation *282 and improvement of public health in the state of New York.” Recognizing that this grant, read literally, could permit the PHC to regulate essentially every aspect of life, the Court explained that “[h]owever facially broad, a legislative grant of authority must be construed, whenever possible, so that it is no broader than that which the separation of powers doctrine permits” (71 NY2d at 9). We held that the anti-smoking regulations, even if authorized by one construction of the statute in question, would not be permitted by the construction of the statute necessary to save the statute from constitutional doubt.

Here, the majority never claims that the relevant statutes would authorize the hard or soft cap but for the separation of powers problem. Instead, the majority explains that the statutes in this case grant “to DOH the authority to [among other things] contract with private entities that receive state money—all with the goal of ensuring that the limited public funding available be directed as efficiently as possible towards high-quality services for New Yorkers in need” (majority op at 262). The majority tacitly recognizes that the authorization by the legislature cannot be plausibly interpreted to create a nondelegation problem: the statutes, read together, obey our command in Packer Coll. Inst. v University of State of N.Y. that the legislature must, in enacting any law that the executive is called upon to execute, “set bounds to the field, and . . . formulate the standards which shall govern the exercise of [executive] discretion within the field” (298 NY 184, 189 [1948]). So Boreali is irrelevant: what matters, at least in the eyes of *17 the majority, is whether the legislature authorized the caps. That is the standard, day-in, day-out judicial question of whether the executive followed the law.

In a situation where we must narrow a statute to meet non-delegation requirements, Boreali was intended to provide some guidance as to how narrowly the statute ought to be read. To detach Boreali's factors from that role of statutory construction, and instead apply them alfresco to the regulations themselves, is to rob them of the context that might give them a chance of coherence. “The tendency of a principle to expand itself to the limit of its logic may be counteracted by the tendency to confine itself within the limits of its history” (Benjamin Cardozo, The Nature of the Judicial Process 51 [1922]). If we do not exile Boreali, we should make sure its cage is small.

Exile may be the better course. The trouble with Boreali's four “coalescing circumstances” is that they are stated at such a high level of generality as to be functionally in permanent flux. Even if they could be stated with precision, they are “not mandatory, need not be weighed evenly, and are essentially guidelines” (Greater N.Y. Taxi Assn. v New York City Taxi & Limousine Commn., 25 NY3d 600, 612 [2015]). Any doctrine so fluid will have difficulty providing either a meaningful constraint on executive power or sensible guidance for the courts. Although the parties do not ask us to reconsider Boreali, today's case illustrates why the Boreali doctrine “while longstanding, warrants serious reconsideration” (cf. People v Cummings, 31 NY3d 204, 213 [2018, Rivera, J., concurring]).

Consider the majority's analysis of Boreali factor three: whether the legislature has “unsuccessfully tried to reach agreement on the issue” (Greater N.Y. Taxi Assn., 25 NY3d at 611), which we have previously measured by examining legislative measures that were introduced but failed.⁶ That factor is omitted entirely from the majority's discussion of the soft cap, and is adorned with a novel exception in the majority's discussion of the hard cap. “[S]ome 40 bills” on the subject of smoking weighed against the regulations at issue in Boreali (71 NY2d at 7), but the failure of six City Council resolutions and five bills in the legislature was enough to weigh against the regulations in Matter of New York Statewide Coalition of Hispanic Chambers of Commerce v New York City Dept. of Health & Mental Hygiene (23 NY3d 681, 692-693, 700 [2014], citing 110 AD3d 1, 14-15 [1st Dept 2013]). Here, the majority instructs us to disregard all such introduced bills unless they “made it out of committee”; it seems that only if a bill succeeds in committee *284 but fails later in the legislative process can it count as an “unsuccessful legislative effort[ ]” (majority op at 265-266). That's new. The Constitution nowhere mentions the committees of the legislature except in connection with the executive budget, which is not at issue in this case (NY Const, art VII, §§ 1, 3). Now, it seems, a committee chairperson can change the lawfulness of an executive act by arranging for proposed legislation to precipitate out of committee—at least until we engraft some other qualifications.⁷

Or consider the majority's treatment of Boreali factor four, the agency expertise factor. The DOH has just as much expertise on the appropriate salary level for persons employed by health care providers regardless of whether the source of the salaries is taxpayer dollars (the hard cap) or elsewhere (the soft cap). Indeed, the majority cites DOH's expertise in “the health care industry,” reflected in crafting the waivers to the hard cap, as supporting its constitutionality. However, it then performs an about-face and declares that while DOH is expert in “the health care industry,” because the soft cap exceptions are a “policy choice” (and the waiver is not?) the expertise factor is now, for unexplained reasons, “neutral.” Because DOH could recreate the blanket soft cap exception by applying the case-by-case waiver universally, the majority leaves us with the unsupportable conclusion that an expert agency that makes policy choices in one way, ceases to be expert when it makes the same choices, even with the same ultimate result, another way—or, more likely, that the expertise factor is wholly irrelevant in the majority's decision.

Inconsistencies like the above have plagued our Boreali jurisprudence. In Matter of NYC C.L.A.S.H., Inc. v New York State Off. of Parks, Recreation & Historic Preserv. (27 NY3d 174, 182 [2016]), a challenge to the New York State Office of Parks, Recreation and Historic Preservation's approval of anti-smoking regulations in public parks, we held that even though the agency knew little about smoking, it was an expert in the “operation” of a park and thus all things within the park were *285 within the agency's expertise (27 NY3d 174, 185 [2016]). Yet in Greater N.Y. Taxi Assn., we held the expertise factor was neutral—while the Taxi Commission had formidable expertise on taxi designs, having dictated them minutely for decades, it was not expert on the question of whether the selection ought to be limited to one model (25 NY3d 600, 612 [2015]). The pièce de résistance is our abandonment of factor four in Statewide Coalition, “[i]n light of Boreali's central theme” (23 NY3d at 700-701).

Our present and past inability to address the “expertise” factor stems from a confusion between statutory and administrative law analyses, in which agency expertise is relevant, and constitutional law, where it is not. Thus, when discussing DOH's expertise, what I believe the majority is trying to get at—and what Boreali has hopelessly muddled—is that the legislature is precise about which agency gets to exercise which power, and so legislation empowering the Secretary of State to prohibit blockbusting, for example, should not be read as empowering the Department of Economic Development from doing the same thing (cf. Matter of Campagna v Shaffer, 73 NY2d 237, 242 [1989] [“Agencies, as creatures of the Legislature, act pursuant to specific grants of authority conferred by their creator”]).⁸ Expertise also matters for the arbitrary-and-capriciousness analysis in administrative law, CPLR 7803 (3), because inexpert agency action is likely to lack a “foundation in fact . . . [and] sound basis in reason” (cf. Matter of Pell v Board of Educ. of Union Free School Dist. No. 1 of Towns of Scarsdale & Mamaroneck, Westchester County, 34 NY2d 222, 231 [1974]).

However, as a matter of constitutional law, we have already held that the Governor need not be an expert on a given subject to issue a valid executive order on that subject, when his orders are directed to officers who serve at his pleasure (Rapp v Carey, 44 NY2d 157, 165 [1978]). Because everyone acknowledges that this DOH regulation, including the $199,000 figure, came directly from the Governor (majority op at 254), DOH's application of its expertise here does not affect the constitutional analysis, unless the majority is prepared to reverse Bourquin v Cuomo (85 NY2d 781, 786 [1995]) and Clark v Cuomo (66 NY2d 185, 189 [1985]), in which the Governor relied on no expertise but his own in creating new administrative bodies to supplement the agencies created by the legislature (and, in Clark, the Constitution itself).

The Governor, in issuing the order, was acting according to his duty to execute the laws—specifically, the laws regarding the control and supervision of expenditure of public funds, and the laws empowering him to pick and choose (according to certain legislatively-prescribed criteria) which health care providers would be beneficiaries of public funds. “[T]he executive power,” after all, is vested in him and not the Commissioner of Health. It would be bizarre, as a constitutional matter, to hold that an order of DOH is a lawful exercise of the executive power but the Governor's is not because DOH is more expert than the Governor. Mired in the Boreali muck, the majority can only reach its chosen conclusion in the face of these inconvenient constitutional facts by jumping outside the Boreali factors, stating that the real problem is that the exceptions “appear to reflect a policy choice” by the agency outside of that authorized by the legislature. I disagree that DOH has done so here. That aside, to say experts employed by the State do not make policy choices is simply incorrect; to say they should not make policy choices is folly. Why should the State have experts at agencies if this Court forbids them from making decisions within their field of expertise that constitute “policy”?

On the subject of policymaking, consider the majority's treatment of the first Boreali factor, whether “the agency did more than balance costs and benefits according to preexisting guidelines, but instead made value judgments entailing difficult and complex choices between broad policy goals to resolve social problems” (NYC C.L.A.S.H., 27 NY3d at 179-180 [internal quotation marks and brackets omitted]). As we said in Statewide Coalition, “the promulgation of regulations necessarily involves an analysis of societal costs and benefits” (23 NY3d at 697). In reality, there is only a difference of rhetoric between “balancing costs and benefits” and “difficult . . . choices,” between “preexisting guidelines” and “broad policy goals” (at least in terms of most enabling legislation, which sets out guidelines in broad terms), and thus Boreali allows courts to uphold or void regulations entirely by dialing up or down their characterization thereof.

The majority's treatment of the “soft” and “hard” cap exceptions exhibits this tension. The majority claims that the *287 Department “did not make impermissibly complex policy decisions” in adopting the hard cap by pointing to the Department's specialized knowledge “of New York's health care industry to develop a multi-factored analysis” (majority op at 264) in designing the waiver provisions of the challenged rule. One of the waiver factors is “the extent to which the executive compensation that is the subject of the waiver is comparable to that given to comparable executives” (10 NYCRR 1002.4 [a] [2] [i]). Apparently this waiver factor either did not spring from a policy choice (majority op at 265), or did so permissibly because it involved agency expertise. However, the same consideration, made more concrete in the soft cap's automatic waiver for salaries outside the 75th percentile of comparable executives (10 NYCRR 1002.3 [b] [1]), is deemed as an impermissible policy choice (majority op at 268). Boreali thus leads the majority to hold that we must strike down as unconstitutional a regulation that says a provider salary cannot be greater than that of comparable executives—but uphold a regulation that says a provider salary cannot be greater than $199,000. Why is the selection of the 75th percentile exclusion less valid than a direct cap on salaries, at least under Boreali? The majority does not say.

Finally, I note that the majority does not explain how it weighs those Boreali factors it chose to consider. What began as “coalescing circumstances” morphed into considerations that “are not mandatory, need not be weighed evenly, and are essentially guidelines for conducting an analysis of an agency's exercise of power” (Greater N.Y. Taxi Assn., 25 NY3d at 612), and are now measured according to some new and unknowable formula applied by the majority in today's opinion: the soft cap pursues a policy consideration contemplated by the legislature (factor one)—but “the connection between the legislative aims and the regulatory means is simply too attenuated” (majority op at 269). Exactly what Boreali factor(s) this corresponds to the majority never makes clear. Is the agency writing on a blank slate (factor two)? Surely not, if a hard salary cap is sufficiently “interstitial.” Perhaps the agency made “value judgments entailing difficult and complex choices between broad policy goals to resolve social problems,” but the hard and soft caps both do that. The majority invalidates the soft cap without regard to the Boreali factors at all, except for a throwaway line that the first two Boreali factors are “most instructive” and a lonely footnote describing factor four as “neutral.” *19

*288 This case aside, examples of the vagaries of Boreali could easily fill a volume of the New York Reports, and, indeed, they have (see e.g. Statewide Coalition, 23 NY3d at 715 [Read, J., dissenting] [recapitulating the many infirmities of Boreali]). Boreali aimed to protect liberty through a vigorous policing of the separation of powers, but “[l]iberty finds no refuge in a jurisprudence of doubt” (Planned Parenthood of Southeastern Pa. v Casey, 505 US 833, 844 [1992]). Boreali has proliferated confusion, confusion the majority today amplifies by applying Boreali outside of its intended ambit. If nothing else, the majority's opinion powerfully demonstrates that Boreali “leads to an unworkable rule, [and] . . . creates more questions than it resolves” (People v Taylor, 9 NY3d 129, 149 [2007]) such that it “no longer . . . withstands the cold light of logic and experience” (Policano v Herbert, 7 NY3d 588, 604 [2006] [internal quotation marks omitted]).

Judge Garcia wisely does not apply the Boreali framework in his concurring and dissenting opinion. (Whether we agree about the infirmities of Boreali, he and I agree that the hard and soft caps stand or fall together.) Instead, he proposes a simpler test, which he argues can be derived from our pre- and post-Boreali decisions: if a regulation is “social policymaking” then it is “an unlawful assumption of power by the executive in setting that policy” and must be struck down (Garcia, J., dissenting op at 274-275).

While that test benefits from concise exposition, I do not think it is workable or consistent with our law. Every act of the Governor or any executive branch agency is a balancing act between different societal interests. Indeed, the legislature wisely forbade major, binding executive actions in the form of regulations without notice and comment (see State Administrative Procedure Act § 202), precisely so those interests could be weighed. That is why the DOH solicited, and took seriously, the comments of industry and others in crafting the regulations at issue here—particularly, as the legislature directed, their impact upon employment, small businesses, and rural areas (State Administrative Procedure Act §§ 201-a, 202-b, 202-bb).

It is also why both the majority and Judge Garcia cite no authority, other than our own cases going back only to the 1970s and a bare citation to the executive or legislative vesting *289 clauses of the Constitution, for the proposition that “policy” is the exclusive domain of the legislature. This Court's past jeremiads against a governor's advancement of his policy goals through his executive actions are oddly justified, even though the results reached in those cases may have been correct.⁹ Governors are elected in part because of their expressed policy preferences. It would be peculiar to hold that the Constitution obliges a governor, once elected, to abandon those goals altogether. Instead, voters expect governors to pursue their stated policies with the caveat that those policies must yield, when governors are exercising the power to “expedite . . . the laws” (NY Const, art IV, § 3) to the policies of the laws they are called upon to “expedite.”

A better view of the cases Judge Garcia cites for his “no policymaking” position (Garcia, J., dissenting op at 275) is to say that while the executive may consider and advance its own policy goals in executing the laws, it may not execute the laws in a manner, or with a result, objectively inconsistent with the policy goals articulated, expressly or implicitly, in the legislation that purportedly authorizes that action. Thus in Matter of Campagna v Shaffer (73 NY2d 237, 245 [1989]), for example, we invalidated an effort by the executive to enforce an anti-blockbusting law by banning, wholesale, real estate agent solicitations on a certain part of the Bronx. It was not the policy-making activity of the executive that *20 rendered the ban unlawful, but the inconsistency between the ban and the legislation that purportedly authorized it; legislation that “explicitly enumerated the types of solicitation it views as forbidden blockbusting activity” (id. at 243) which necessarily implied that the legislature intended for solicitation—but only the non-discriminatory kinds of solicitation—to continue. In Subcontractors Trade Assn. v Koch (62 NY2d 422, 427 [1984]), it was likewise not the executive's policy-making activities (there, creating a quota for the allocation of city contracts flowing from a mayoral policy of promoting small businesses) that rendered its actions ultra vires, but the lack of congruity between *290 those actions and the governing legislation (General Municipal Law § 103) which, as the Appellate Division noted in that case (96 AD2d 774, 775 [1st Dept 1983]), required a competitive bidding process that looked only to “best value,” defined by State Finance Law § 163—a statute that plays an important role in this case. The problem, in sum, in past cases where we invalidated acts of “contracting” was and is not that policymaking was happening through that contracting activity. It was rather whether the policies the executive made were congruent with the policies encoded in law—for when executive policy and legislative policy conflict, it is the legislative policy that governs.

The constitutional question in this case, then, concerns the limits of “[t]he executive power” (NY Const, art IV, § 1). The Governor, and those persons exercising power as gubernatorial appointees, possesses the power and duty to “expedite all such measures as may be resolved upon by the legislature, and shall take care that the laws are faithfully executed” (NY Const, art IV, § 3). This is a part, indeed the principal part, of “[t]he executive power” vested in the office (NY Const, art IV, § 1). The question in this case is solely whether the Governor, and his appointed commissioner, exceeded that “executive power” by promulgating 10 NYCRR part 1002.

The “executive power” has been vested in the governor from the first Constitution of New York State, adopted in 1777, in which the governor was vested with “the supreme executive power” (1777 NY Const art XVII), and obliged “to take care that the laws are faithfully executed to the best of his ability; and to expedite all such measures as may be resolved upon by the Legislature” (1777 NY Const art XIX), wording that has essentially survived until the present day.¹⁰ It descends directly *291 from the colonial governors, and to them from the British Crown. The framers of the 1777 Constitution, and the early 19th century constitutions where the executive power clauses assumed the forms they have today, would have been most familiar with Blackstone's formulation of what the law-execution power entailed:

In his exercise of the power and duty to “expedite all such measures as may be resolved by the legislature,” or “putting those laws in execution,” per Blackstone, the governor and his commissioners are called on to do three sets of tasks routinely: hire people, spend money appropriated by the legislature and, oftentimes in a large and complex government administration like New York's, regulate how those people the governor hired spend the money appropriated by the legislature. Indeed, in describing the federal executive power as congruent with the powers of the Governor of New York under the 1777 State Constitution, Alexander Hamilton described the “executive details” of the “administration of government” as including “the preparatory plans of finance . . . [and] the application and disbursement of the public moneys in conformity to the general appropriations of the legislature” (Hamilton, Federalist No. 72).

The power to spend money appropriated by the legislature is at the very core of the power to execute the laws, and at the heart of the power to spend money is the power to contract. Suppose, for example, that the legislature, worried about New *292 York's growing cheese surplus,¹¹ appropriated $10 million “for services and expenses of a program, to be established by the Agriculture and Markets Commissioner, to increase the consumption of cheese in the State.” The executive branch must spend that money, as we held in Matter of County of Oneida v Berle (49 NY2d 515, 523-524 [1980] [no executive branch power to “impound” lawful appropriations]), but the only way it can do so is by contracting, whether that be contracting for employment, or for some other services. Contracting, in turn, is impossible unless the executive branch can select amongst contractors based on its own conception, bounded of course by relevant legislation, of how one might best increase cheese consumption given the $10 million appropriated. When it comes to “expedit[ing] all such measures as may be resolved upon by the legislature,” and the measure in question is about spending public money, it's contracts all the way down.

The executive branch's power to contract is not unlimited, because the executive power is not unlimited. The executive cannot contract except as authorized by law. If the law forbade the making of a certain contract, or forbade choosing contractors on a certain basis, the executive could not make that contract or one substantially at odds with what the legislature forbade. As Blackstone explained, executive actions were effective only “where they do not either contradict the old laws, or tend to establish new ones; but only enforce the execution of such laws as are already in being.”¹² Those laws must, we have held in connection with cases concerning the legislative power (NY Const, art III, § 1), be sufficiently definite so as to allow assessment of whether the governor or commissioner is enforcing “the execution of such laws as already in being”; the legislature must “set bounds to the field, and . . . formulate the standards which shall govern the exercise of [executive] *293 discretion within the field” (Packer Coll., 298 NY at 189). The executive cannot contract except to execute the existing laws made by the legislature, which necessarily means the executive cannot use the contracting power to “establish new ones.” *22

We can see this principle in operation, if not in description, in Under 21, Catholic Home Bur. for Dependent Children v City of New York (65 NY2d 344 [1985]). There, we held that a regulation issued by the Mayor of New York banning discriminatory employment practices on the part of contractors exceeded the Mayor's executive power, because it attempted to achieve a policy goal not contemplated in the relevant legislation from which the contracting itself sprang. However, we noted, if the executive “feels that the price of a particular contract is too high due to a contractor's employment policies, it is of course free to choose some other contractor” (Under 21, 65 NY2d at 359 n 5 [emphasis added]). In other words, because the authorizing legislation set bounds to the field—the executive could pick and choose contractors, but only on certain bases, like price—the executive was free to select an indicia of prices, there discrimination in employment, and pick and choose contractors on that basis as an indicia of price. (And doing so would of course be informed by policy—a policy judgment that the State gets better deals from non-discriminatory contractors.) But in Under 21, the Mayor was plainly not doing this; rather, he was considering a factor outside the bounds of the legislation from which the contracting power sprang. That, we held, was impermissible (accord Subcontractors Trade Assn. v Koch, 62 NY2d 422, 427 [1984]; Matter of Fullilove v Carey, 48 NY2d 826 [1979]; Matter of Fullilove v Beame, 48 NY2d 376 [1979]; Matter of Broidrick v Lindsay, 39 NY2d 641 [1976]).

Once we understand the constitutional question is whether the regulation exceeds the executive power, the analysis is simpler. As discussed above, the executive power, including the power to contract, exists in reference to existing legislation. Thus, to determine the lawfulness of executive action, we must determine whether it objectively advances the policy goals articulated in the legislation that purportedly authorizes that action. If those goals are undiscernible then the legislation is itself unconstitutional under the nondelegation doctrine as outlined in Packer Coll. (298 NY at 189). If the legislation is susceptible of an interpretation that articulates determinate policy goals and another that does not (the original Boreali question), then it must be interpreted to be sufficiently *294 determinate to have the policy goals against which an executive action can be judged.

This far simpler framework is, in fact, largely consistent with our existing jurisprudence in outcome if not in form.¹³ It is also consistent with the majority's decision on the hard cap. The majority concludes, correctly, that the *23 hard cap regulations advance the legislature's policy goal “that state health care funds should be expended in the most efficient and effective manner possible to maximize the quality and availability of public [health] care” (majority op at 266-267). Where the majority founders, however, is in its refusal to pursue that principle when it turns to the soft cap.

To understand why the “soft cap” is constitutional, it is helpful to explain what the soft cap is doing and the precise statutory goals it is attempting to advance: preventing circumvention of the hard cap, and advising providers that the State may deem them undesirable or inefficient vendors and allocate taxpayer monies away from them and to competitors whom, in the State's judgment, provide superior value.

Much of the majority's analysis of the soft cap turns on its application to private as well as public funds (see e.g. majority op at 268). But any effective regulation involving the use of public funds by private providers needs to take private funds into account. Recall that the hard cap puts a ceiling of $199,000 on the executive compensation certain covered employees can receive from state coffers (10 NYCRR 1002.3 [a]). Many private health care providers do not receive all their revenues from the State. Instead, they treat some patients for whom the State reimburses them, and some patients that pay directly out of pocket or are covered by private health insurance.

Suppose you were a covered executive at such a provider and were facing the hard cap only. You want to pay yourself a salary of $1,000,000. The State says if you use more than $199,000 of the state monies you receive to pay your salary (leaving aside waiver for the moment), the State may cease doing business with you and turn to other providers. So you simply adjust your organization's budget: from here on out, you will get a salary up to the cap from state dollars, but use private dollars to get to the $1,000,000 figure. Perhaps you will not even report that as salary, but as a perquisite of college tuition for your children, use of an automobile, country club membership, etc., funded entirely by private monies (see Russ Buettner, Reaping Millions in Nonprofit Care for Disabled, NY Times, Aug. 2, 2011, https://www.nytimes.com/2011/08/02/nyregion/for-executives-at-group-homes-generous-pay-and-little-oversight.html). Of course, that $801,000 is coming from somewhere—either from private funds paying salaries of other employees or other needed administrative expenses or, of course, patient services for which the State is paying, freeing up the private funding to go to salaries, thus indirectly costing the taxpayers more money in the long run.

All of this circumvents the central goal of the regulation, and the goal which the Members of this Court unanimously agree was written into the authorizing statute: to maximize the “bang for the buck” the State receives when it chooses among service providers (cf. Matter of Medicon Diagnostic Labs. v Perales, 74 NY2d 539, 545-546 [1989] [understanding the relevant statutes to mean that “the agency charged with the responsibility of administering the medicaid program has inherent authority to protect the quality and value of services rendered by providers in that program”]). It is both rational and appropriate for DOH *296 to attempt to ensure that the providers with which it contracts invest in certain administrative expenses (say, a functioning patient portal that allows for more convenient and efficient appointments with doctors) rather than others (compensation for executives), so as to maximize the impact of the state dollars spent on direct service provision.

Circumvention of the hard cap is possible because money is fungible, regardless of its source. Once the State has paid a provider, as a practical matter the only way the State can ensure it is being spent properly is to directly *24 order those spending any money to spend any of their money on some things and not others (and, I stress, we unanimously agree such an order is appropriate in this case, as regards “administrative” versus “direct services” caps). And even if the soft cap were a belt-and-suspenders provision, and the hard cap can stand on its own, “that a regulatory provision may be duplicative or . . . unnecessary in relation to the achievement of the legislative goal does not make it ultra vires” (majority op at 267 n 9).

Understood in this way, the soft cap suffers from none of the infirmities that the majority suggests it possesses. It is not pursuing a policy consideration “not clearly connected” from the objectives outlined by the legislature; it is pursuing the exact same policy consideration as the hard cap—maximizing the impact and power of every state dollar reaching a provider—to avoid circumvention of the hard cap. In tandem, the caps maximize the benefit received by the State while ensuring the private health care provider sector would not circumvent the regulations. Circumvention, whether unscrupulous, negligent, or inadvertent, was surely possible: the most recent annual report for the Medicaid Inspector General (Office of the Medicaid Inspector General, 2016 Annual Report, https://omig.ny.gov/images/stories/annual_report/2016_OMIG_Annual_Report.pdf, cached at http://www.nycourts.gov/reporter/webdocs/2016OMIGAnnualReport.pdf) reveals $418 million in Medicaid recoveries and $1.9 billion in cost-avoidance initiatives, all responding to a combination of bad practices and simple error at the hands of private health care providers.

Accordingly, just as the hard cap objectively advanced the legislatively-circumscribed goal of getting the “biggest bang for the buck” in its Medicaid services, the soft cap, which is really *297 just an anticircumvention provision, does the same.¹⁴ Indeed, the majority concedes this very point in its opinion when it explains that “DOH's intentions were to advance the same interests underlying the hard cap regulations” (majority op at 268). If DOH's “choices” are the critical question here, as the majority would have us believe, then its “intentions” should determine the result, because DOH's policy choice was consistent with the legislature's grant.

Even if the soft cap were pursuing a different policy consideration than the hard cap, it would still be well within the confines of the executive power because it objectively advances another legislatively-decreed policy: ensuring the State does business only with responsible contractors. State Finance Law § 163 (2) (a) declares it the policy of the State to “promote purchasing from responsive and responsible offerers,” defined in part to require that the State examine those offerers' “financial ability, legal capacity, integrity, and past performance” (State Finance Law § 163 [1] [c]).¹⁵ It objectively advances the legislature's demand that the State do business only with those persons with “financial ability, legal capacity . . . [and] integrity” for the executive to require that those it does business with pay salaries that are reasonable, as measured by the oldest standard of reasonableness there is—what everyone else (or at least the 75th percentile of everyone) is doing. It also objectively advances that goal to insist that the nonprofits to which the State will entrust over $150 billion in taxpayer dollars be governed responsibly, with independent or at least attentive boards of directors. Between the centrality of contracting to the basic functioning of the executive, the clarity of the legislative policy command, and the close connection of this regulation to those goals, the soft cap falls squarely within the executive power and is thus constitutional.

The majority concedes that State Finance Law § 163 is “an important aspect of any contracting agency's mandate” but *298 holds the soft cap is still unconstitutional because DOH “has not shown a connection between a provider's decision to use private funds to compensate its executive staff handsomely or even excessively” and the “absence of any of these essential contractor characteristics” (majority op at 270). There are at least three problems with this contention. *25

First, it is wrong as a matter of fact. DOH explained in some detail, both in the briefing of this case and in its discussion of the regulation in the record, exactly why payment of excessive compensation from private sources was a “red flag” indicating the provider's fiscal responsibility and integrity as those terms have been long understood in the public procurement process.

Second, the majority that just told us that administrative agencies “may be vested with considerable discretion to flesh out a policy broadly outlined by legislators . . . [and] have flexibility in determining the best methods for pursuing objectives articulated by the legislature” (majority op at 260) now says that an agency cannot determine the best methods of pursuing the broad (but certainly not unbounded) objectives of State Finance Law § 163 based on its admitted expertise. That flies in the face of elementary principles of administrative law, the same elementary principles the majority vaunts just a few pages into its own opinion.

Third, the majority misquotes State Finance Law § 163 by omitting a key phrase: the various essential contractor characteristics, the legislature tells us, are defined in accordance with how “such terms have been interpreted relative to public procurements” (State Finance Law § 163 [1] [c]). Who conducts public procurements? The executive. By what goals? These are well established in State Finance Law § 163 itself, and include, among other things, the command that “[i]t shall be the responsibility of the head of each state agency to periodically sample the results of the procurement process to [among other things] ensure that the results withstand public scrutiny” (State Finance Law § 163 [11] [emphasis added]). Surely it was reasonable to conclude that the “results” of the pre-soft cap contracting rules, which allowed the State to plow millions of dollars into nonprofits that were paying their own executives lavish compensation, would not “withstand public scrutiny.” Thus DOH's view that State Finance Law § 163 required it to look, among many other factors, at the governance of a potential contractor and the salaries it pays its top executives before giving them taxpayer dollars is simply obedience to a *299 different mandate, a general contractor mandate, as applicable to DOH as any part of the Public Health Law.

The two exceptions in the soft cap illustrate exactly how the soft cap allowed the agency to test the “responsibility” of a given contractor. First, relying on its understanding of the health care sector, DOH concluded that the surest guide of “legal capacity” and “integrity” in an organization is good governance practices—and certainly not executives setting their own salaries without any review. Hence, DOH required that salaries more than the $199,000 benchmark from any source be reviewed by a board, and that it either contain independent directors or at least assess “appropriate comparability data” in signing off on that decision (10 NYCRR 1002.3 [b] [2]). The point of requiring an examination of comparability data was to invoke a provider's own governance structures to encourage efficiency and responsibility, not secure a particular dollar figure salary as the majority supposes (majority op at 269). Second, if a contractor decided not to implement these very basic integrity-based reforms, DOH created an exemption for salaries that were 75% of the norm for the provider and the region (10 NYCRR 1002.3 [b] [1]), which would allow it to concentrate its limited enforcement resources on only the “red flag” actors that DOH might choose to avoid when deciding how to spend taxpayer dollars.

Finally, the majority's analysis of State Finance Law § 163 raises as many questions as it answers. The parties, as well as the majority, do not appear to object to the many and varied ways in which DOH measures “integrity” in its regular contractor accreditation process set out in 18 NYCRR 504.1 et seq., such as not doing business with entities that make false statements to licensure agencies (18 NYCRR 504.5 [a] [4]). Surely DOH's identification of lying to a licensure agency as an “integrity” factor is as much a “value judgment” (majority op at 270) as its finding in 10 NYCRR 1002.3 that “excessive compensation,” minus certain safeguards, is a sign of “irresponsibility”? The majority cannot mean what it says—that DOH's determinations about the integrity of contractors are illegitimate “value judgments” beyond the agency's “health care funding mandate.” Were that so, it would cripple DOH's ability to regulate fraud and malfeasance. Decertifying, or refusing to deal in the first place with, contractors who lack “integrity” is indeed a “public policy determination,” but it is one that the legislature has expressly told the executive branch to make for *300 over a century. Unless discounted as throwaway dicta, the majority's statement places the entirety of 18 NYCRR part 504, in which DOH requires that only “qualified and responsible persons may be enrolled as providers of care, services, and supplies” (and to which no parties object) (18 NYCRR 504.1 [a]), in serious constitutional doubt. *26

This would be a different case, of course, if the legislature had not specified “responsibility” as a permissible criteria by which the Department could pick and choose amongst vendors and had declared other criteria to be the exclusive basis on which the executive could make that choice. But the legislature decreed “responsibility” as a criterion, and DOH is appropriately expediting the matter resolved by the legislature. One way to look at it is that, everything else equal, the State would prefer to contract with providers who use private funds to improve common goods from which state-funded patients can benefit, rather than those providers who pay their executives lavishly. This is perfectly proper behavior on the State's part—just as a construction contractor might be preferred because it owns equipment, financed by a private job but useful for a public building site, or has a skilled workforce that gains experience on private jobs.

Even if State Finance Law § 163 did not give the agency authority to select only responsible contractors to participate in Medicaid and related programs, the appropriations themselves give rise to the executive power to select (consistent with other laws and parts of the Constitution, of course) from competing contractors to spend that money. The executive has had the power to select among contractors since the dawn of an independent New York, provided the selection is made in a manner consistent with the laws or appropriations giving rise to the creation of the contract (basic authority to do so, the majority correctly notes, was vested in the Department by Public Health Law § 206). Inherent in an appropriation are two considerations the executive may lawfully consider: the quality of the service provided, and the cost of that service. Here, once the legislature appropriated money for private providers to give public services, the executive was free to determine that it would direct that money to entities that were frugal and not spendthrift, because doing so objectively advanced the legislatively-circumscribed policy goal of the appropriation—to *301 take the least amount of taxes and generate, if you will, the most amount of health care.

Thus, hard or soft, the regulation at issue here was entirely constitutional. Accordingly, I would uphold the regulation in its entirety.

In each case: Order affirmed, without costs.

Opinion by Chief Judge DiFiore. Judges Stein, Fahey and Feinman concur; Judge Garcia dissents in part and votes to modify in favor of petitioners in accordance with an opinion; Judge Wilson dissents in part and votes to modify in favor of respondents in accordance with an opinion; Judge Rivera dissents in part and votes to modify in favor of respondents for the reasons stated in parts I, III and IV of Judge Wilson's opinion.