In recent years, in significant part following the impressive lead of New York Southern District Judge Jed S. Rakoff, U.S. district court judges around the nation have begun to take it upon themselves — as some argue, “arrogated” to themselves — the role of ensuring that when the government settles a case with a private litigant, often “a too big to fail” financial institution, it is gaining for the citizenry in general (or, perhaps, the specific victims of the wrongdoing) the best bang for its buck. And, to boot, they seek accountability for alleged wrongdoing even when the government is willing to settle. Yes, lest there be any mistake about it, some judges — very distinguished judges — are holding litigants to account by refusing to accept settlements.

There is some question as to whether these interventions are acceptable – whether it is proper for judges to reject settlements arrived at by experienced parties with capable counsel. It is important, perhaps essential, to have a public conversation about whether judges should engage in this kind of intervention. Indeed, Judge Rakoff published a controversial article challenging the vitality of the Justice Department’s criminal enforcement policies in a non-legal publication (“The Financial Crisis: Why Have No High Level Executives Been Prosecuted?,” The New York Review of Books, January 9, 2014), seemingly in an effort to prompt such conversation in the public square. I’ve laid out some of the arguments in order to promote that conversation.

No question — judges have a duty to ensure that litigants who settle cases before them don’t effectively pervert justice by settling. If litigants are able to get away with trumped-up litigations by bullying a settlement where there is simply no case, the presiding judge should step in, rather than sitting on his hands, saying: “If represented parties mutually agree to a result, why should I stand in their way?” We need judges to make sure that the “little guy,” or the inadequately-represented-guy, not only has his day in court, but that the day is meaningful where are no bus tire marks on him caused by an incompetent or uncaring advocate.

But what of the government’s settlements with financial institutions? In effect, some judges who have rejected these settlements have taken the legal position that they can’t simply rely on the capacity or willingness of the government. Whether, for example, in the person of the U.S. Justice Department itself, or the Securities and Exchange Commission — to exact a quantum level of punishment and/or restitution from the defendant that the judge would demand if he himself (having stepped into the shoes of prosecutor or regulator) were calling the shots for Uncle Sam as the government’s litigating quarterback, or the official(s) within the agency empowered to make the ultimate settlement decision .

What is the appropriate role for a presiding judge when two sophisticated parties have tirelessly negotiated to reach a settlement? Is the judge necessarily in the best position to determine whether the settlement is reasonable, given the policy considerations and extensive factual circumstances of the particular case, with which he may or may not be as familiar as the litigants themselves? Is it the role of the judge, who might admittedly not know the defects in a case, to basically second guess a government party’s lawyer and say that it hasn’t demanded enough?

In certain kinds of cases, for example, a class action lawsuit where the plaintiffs’ lawyer may never have actually met or even interacted with a client, or where the lawyer represents an infant — a judge has an affirmative legal duty to step in to ensure that the client’s interests are being adequately addressed. Indeed, we are not here talking about the role of a judge when there is a duty under law to speak for those who are essentially “unrepresented,” such as Judge Anita Brody’s refusal to endorse a proposed settlement with the National Football League, or Judge Alvin Hellerstein’s decision to reject the first 9/11 tort settlement proposed by counsel. Nor are we discussing instances where the litigator for the government (or the government employee in charge of the matter) is incompetent, or actually corrupt, exchanging a settlement for a political or personal favor. What we are discussing here is judicial intervention when the litigants are the government and a powerhouse financial institution and whether, given those litigants, the judge should get a veto.

Or maybe judges should be asked to assume (maybe, rebutably presume) that there are sufficient protocols in place within an agency litigating on behalf of the government to ensure that the government has proceeded in good faith and has adequately assessed the ability of gaining a settlement and the risk of losing the case if the settlement eludes it, bearing in mind the multiple “intra-agency” approvals required? Aren’t judges effectively required to give deference to an agency’s policy determinations to presume that the government sought and settled for a level of punishment or restitution that is to commensurate with the wrongdoing and victimization in question? Shouldn’t judges accept the agency’s determination as to its ability to prove the government’s case taking into consideration the resources (financial and otherwise) that would have to be expended to do so?

Federal judges, most of them remarkable, have to deal with plenty on their plates, and they’re not the litigators in the cases before them. They simply can’t possibly know the myriad of reasons, factual and policy, why a government litigator might accept a settlement that might seem on its face. To the judge and, likely, the public — somewhat less than should ideally be demanded to settle a criminal, enforcement or regulatory matter. Yes, judges can and do cajole, persuade and coax but outright rejecting a settlement to force a defendant to pay an amount not demanded by the government can seem like an overstepping of bounds.

But let’s look at the other side of the argument. Institutions are being prosecuted and are often settling for a minute percentage of their earnings (“pocket change,” the view of some judges). Until recently, when the SEC began to require that admission of wrongdoing accompany settlements, no acknowledgement of responsibility was required. In fact, the settlements contained the mantra, “neither admit nor deny wrongdoing.” Some judges have argued, Judge Rakoff most vocal among them, that, in such settlements as these when the penalty is inconsequential to the institution and there is no admission of culpability, not only is there no corporate accountability, but the executives who called the shots, who made the decisions and who turned a blind eye to what was obvious have not actually been prosecuted.

When there is no admission of liability, coupled with payment of an insignificant sum, what is the public expected to think? Put another way, perhaps the settlements allow the government and the institutions to hide information from the public that may inflame or incite, but also explain. Or is it that the information might also cause others to ask questions that no litigant — including the government — might want answered?

This is not a simple issue, but it’s a conversation worth having with the public at large. The courts, particularly the Second Circuit, will surely soon decide under what circumstances judges can reject settlements. But, as I suspect Judge Rakoff intended when he published in the New York Review of Books — such questions should consider the views of the real world, too.