(concurring). I concur in the majority’s decision to reverse, but I cannot agree with the majority’s case-specific analysis, which purportedly is premised on this record alone *388and resolves only the enforceability of the agreement at issue here. Like the majority, I conclude that the release is invalid and that the affirmed findings of "voluntariness” by the courts below are not dispositive. I would hold, however, that, as a general proposition, these agreements, in which criminal charges are dismissed in exchange for the defendant’s execution of a release from civil liability, offend public policy and will not be enforced. Accordingly, I write separately to explain my reasons for reversal.
Release/dismissal agreements discharging public officials and government agencies from civil liability as a quid pro quo for the dismissal of pending criminal charges against the releasor have been the subject of considerable debate in recent years (see generally, Fielkow, 42 USC § 1983 — Buying Justice: The Role of Release-Dismissal Agreements in the Criminal Justice System, 78 J of Grim L & Criminology 1119). Although the lower courts of this State have focused principally upon the potentially coercive aspects of these agreements (see, Dziuma v Korvettes, 61 AD2d 677; see also, Matter of City of Cohoes v Spizowski, 72 AD2d 847, 848 [Greenblott, J., concurring]; People v Siragusa, 81 Misc 2d 368), courts of other jurisdictions have looked beyond the private concerns of the litigants and considered their public policy implications (see, Boyd v Adams, 513 F2d 83, 88; MacDonald v Musick, 425 F2d 373, cert denied 400 US 852; Dixon v District of Columbia, 394 F2d 966, 969; Gray v City of Galesburg, 71 Mich App 161, 247 NW2d 338). The leading exposition on the subject to date is Newton v Rumery (480 US 386), in which a closely divided Supreme Court declined to adopt a per se rule of invalidity and upheld the particular release/dismissal agreement before it on the ground that it was voluntary and supported by what the majority deemed "an independent, legitimate reason * * * directly related to [the County Attorney’s] prosecutorial responsibilities” (id., at 398). Significantly, despite the deep divisions within the Rumery court, all of the Justices agreed that, regardless of their voluntary or involuntary character, release/dismissal agreements carry a serious potential for public mischief (id., at 394-395 [plurality opn per Powell, J.]; id., at 400-401 [O’Connor, J., concurring]; id., at 410-415 [Stevens, J., dissenting]).
First, as noted by the United States Court of Appeals and acknowledged by the Supreme Court majority in Rumery, such agreements " 'tempt prosecutors to trump up charges in reaction to a defendant’s [civil] claims, suppress evidence of *389police misconduct, and leave unremedied deprivations of constitutional rights’ ” (480 US, at 394, supra, quoting 778 F2d 66, 69). Second, the availability of such agreements creates a troublesome conflict of interest for prosecutors, who are called upon to balance the private concerns of witnesses and public servants against the legitimate concerns of their primary client, the People of the State of New York, in the enforcement of the criminal laws (see, 480 US, at 412-414 [Stevens, J., dissenting]). Third, they give rise to a serious risk that the societal interest in the prosecution of meritorious criminal charges will be compromised in an effort to protect local law enforcement personnel from the embarrassment and expense that attends civil litigation (see, id., at 400 [O’Connor, J., concurring]). Finally, by providing potential private litigants with a powerful incentive to forgo arguably meritorious claims, these release/dismissal arrangements interfere with an important mechanism for vindicating individual rights and holding public servants accountable, much "to the detriment * * * of society as a whole” (id.).
As is evident even from this very limited list of concerns, the reasons for objecting to release/dismissal agreements extend far beyond their potential for coercion.1 In addition to their potential for impairing the integrity of criminal justice process, such agreements encourage prosecutors to violate ethical rules which forbid attorneys from pressing criminal charges "solely to obtain an advantage in a civil matter” (Code of Professional Responsibility DR 7-105) and caution against using the criminal process to force settlement of private claims (Code of Professional Responsibility EC 7-21).2 Even more seriously, by permitting prosecutors to utilize their *390considerable discretion over the decision to prosecute in the service of goals unrelated to any legitimate law enforcement concerns, they breed cynicism and deprive the public of the independent, disinterested prosecutorial judgment to which it is entitled.
Weighed against this very real potential for harm, the reasons that are most often cited in support of permitting release/dismissal agreements are insubstantial. The benefits of insulating municipalities and law enforcement officers from the burdens of defending against civil claims, however frivolous or lacking in legal merit, are simply not, without more, proper factors for prosecutors to consider in exercising their discretion to prosecute or withhold prosecution of particular crimes. Moreover, while reducing court congestion and promoting judicial economy are generally worthy goals, they cannot justify judicial approval of a quid pro quo arrangement that undermines the fundamental integrity of the criminal justice system. As this court has recently observed, "[t]he first order of business of the criminal courts * * * is justice, not economy or convenience” (People v Ricardo B., 73 NY2d 228, 235). Similarly, the "first order of business” of the State’s prosecutors, who have been accorded considerable discretion within their own sphere of responsibility (see, Matter of Schumer v Holtzman, 60 NY2d 46; People v Zimmer, 51 NY2d 390, 394; People v Di Falco, 44 NY2d 482, 486-487; People v Mackell, 47 AD2d 209, affd 40 NY2d 59), is not to contribute to the improvement of the court system as a whole, but rather to ensure that, within the practical limits of the State’s resources, the people are zealously and ethically represented in the criminal courts.
To be sure, there are many instances in which a legitimate prosecutorial purpose might exist for dismissing seemingly meritorious criminal charges.* *3 For example, dismissal might well be deemed an appropriate course where the District Attorney’s resources are overtaxed, the crime is not a serious one, the defendant does not pose a serious ongoing threat to society or a witness wishes to avoid public exposure. In such cases, however, the prosecutor already has an independent *391motive for seeking dismissal, and the decision to exact a civil release in addition would therefore constitute prosecutorial overreaching. On the other hand, where these or similar considerations are not present, an agreement by the prosecutor to dismiss meritorious charges in exchange for a civil release would be difficult to justify.
For all of these reasons, I conclude that release/dismissal agreements are violative of public policy and should not be given judicial recognition. A promise is ordinarily deemed unenforceable when the interests favoring enforcement are "clearly outweighed in the circumstances” by strong public policies militating against enforcement (see, Restatement [Second] of Contracts § 178 [1]). The substantial potential for corruption of the system that is inherent in these agreements, coupled with the potential for misuse of prosecutorial authority and the dubious societal benefits they confer, counsels against the adoption of any rule that would sanction, or even merely encourage, their use.4
It is on this point that the majority and I part company. Although the majority Per Curiam opinion contains much dictum suggesting its disapproval of these release/dismissal agreements, the majority has pointedly stopped short of holding that release agreements obtained in exchange for dismissal of criminal charges should generally not be enforced. In fact, one must search the majority Per Curiam with great care in order to find a legal principle that has some application beyond "the circumstances presented” (majority opn, at 386). Apparently, and without acknowledging as much, the majority has opted for an approach similar to that adopted by both the plurality and the concurrer in Newton v Rumery (480 US, at 392 [plurality opn per Powell, J.], 399 [O’Connor, J., *392concurring], supra), which involves analyzing the validity of release/dismissal agreements on a case-by-case basis. I have several serious objections to this approach — both in principle and as the majority has applied it here.
First, although it has invoked "the record” and indicated that its finding of unenforceability is limited to the circumstances presented in this particular case, the majority has made factual assumptions that are not actually based on what was established below. The dispositive fact for the majority appears to be that the District Attorney entered into the challenged agreement "solely to protect against the possibility of civil liability” (majority opn, at 387). In fact, although plaintiff has made that claim at various points in this litigation, defendant has disputed it. Moreover, the Assistant District Attorney who actually obtained the release explained at the evidentiary hearing held on defendant’s motion that he did so because it was his "understanding from dealing with [plaintiffs trial attorney] that he [the attorney] would always blame the police” and because he (the Assistant District Attorney) simply "got tired of fighting with [plaintiffs trial attorney].” Thus, it is far from established that the people’s primary goal in trading a dismissal for a civil release was, in reality, to insulate the police or any other government officer from exposure to civil liability.
The majority’s effort to wrestle the disputed facts presented here into manageable form points up some of the more fundamental difficulties with its insistence on a case-specific approach to determining the validity of these release/dismissal agreements.5 Obviously, any inquiry into the motives of parties to a prior agreement is problematic, since it invites post hoc rationalizations and, in some instances, testimony that is tailored to the contours of the case law. Further, the case-by-case approach is unsatisfactory because it would require the courts to second-guess the soundness, legitimacy and probity of the prosecutor’s expressed rationale, bringing the courts into direct conflict with the well-established rule that, as *393elected law enforcement officials, prosecutors have virtually unreviewable discretion to act within the proper sphere of their authority (see, Matter of Schumer v Holtzman, supra; People v Zimmer, supra; People v Di Falco, supra; People v Mackell, supra).
Finally, the majority’s case-specific approach, which provides little guidance for District Attorneys, future criminal defendants and their attorneys, is unsound from the standpoint of judicial policy, because it will lead to uncertainty as to the enforceability of these agreements in other cases. Such uncertainty is a particularly undesirable result in this context, since it creates the risk that a prosecutor entering into a release/dismissal agreement will discharge his side of the bargain by dismissing the criminal charges, only to discover after litigation in a subsequent civil proceeding — to which he is not a party — that the defendant’s side of the bargain, the release, is unenforceable. Manifestly, such a degree of unpredictability would not be tolerated in the law governing commercial contracts. I find it difficult to understand why the majority is willing to tolerate it in this specialized area of release/ dismissal agreements, in which the public interest is so directly implicated.
I do not suggest that there could never be a set of facts in which both parties’ reasons for entering into a particular release/dismissal agreement are so clear and compelling that they would outweigh the strong public policies that generally preclude enforcement of such agreements. However, in this case of first impression in our court, there is no need to speculate about the type of circumstances that might warrant creating an exception to a general rule against enforcement. The common-law process is sufficiently flexible to accommodate and account for special situations as they arise. Accordingly, unlike my colleagues in the majority, I have no reluctance to reach the conclusion that, as a general matter, these agreements violate public policy, while reserving the possibility that some novel fact pattern that is not now readily apparent might subsequently suggest itself as an exception.
In this case, it is sufficient to note that there are no special circumstances and no compelling reasons to give judicial recognition to the release/dismissal agreement that is proffered as a defense to plaintiff’s claims. Accordingly, I agree *394that the release is invalid and that, consequently, the motion to dismiss the complaint should be denied.
Chief Judge Wachtler and Judges Simons, Kaye, Alexander, Hancock, Jr., and Bellacosa concur in Per Curiam opinion; Judge Titone concurs in result in a separate opinion.
Order reversed, etc.
. This is not to suggest that the coercive aspects of these agreements are insignificant. To the contrary, in most instances, the circumstances surrounding these agreements are inherently coercive, since they are made at a time when the civil claimant is facing the prospect of a criminal trial, with all of its accompanying risk, expense and embarrassment. Under such circumstances, even an intelligent and informed defendant, whether innocent or guilty of the charges, would be hard-pressed to resist an offer permitting him to walk away without further consequence in exchange for relinquishing a civil claim which might ultimately prove unsuccessful anyway.
. It would be naive not to recognize that law enforcement personnel occasionally file preemptive disorderly conduct, obstructing governmental administration and harassment charges in order to obtain bargaining leverage when they anticipate that an unpleasant encounter between police and a private citizen will lead to a civil lawsuit. As was noted in Boyd v Adams (513 F2d 83, 89), " 'the danger of concocted charges is particularly great *390because complaints against the police usually arise in connection with arrests for extremely vague offenses such as disorderly conduct or resisting arrest.’ ”
. I note that the majority’s Per Curiam opinion does not seem to take this possibility into account.
. To be distinguished are ordinary plea bargains in which the defendant relinquishes his right to a trial on pending criminal charges in exchange for more lenient penal treatment. In plea bargaining situations, the defendant admits his guilt and the societal interest in punishing wrongdoing is, at least in some measure, fulfilled (see, Newton v Rumery, 480 US 386, 393, n 3 [plurality opn]). In contrast, when criminal charges are disposed of through release/dismissal agreements, the only interest that is satisfied is that of those who wish to see the civil litigation finally resolved. Further, as Justice Stevens noted in his dissenting opinion in Newton v Rumery (supra, at 411), unlike plea bargains in which some penal sanction is imposed, these agreements "exact[] a price unrelated to the character of the defendant’s own conduct.” Thus, conventional plea bargains represent compromises that are consistent with the purposes of the governing penal statutes, while release/ dismissal agreements do not.
. At an earlier point in its Per Curiam opinion, the majority notes that "there remain unresolved factual allegations” and "unanswered questions” regarding defendant’s conduct, the practices of the District Attorney’s office in general and the motives of that office in entering into this particular agreement (majority opn, at 386, 387). These acknowledgements make it even more difficult to understand the basis for the majority’s assumption that the purpose of the challenged agreement was to protect the police.