No Skin in the Game: Prosecutorial Immunity, Declination, and the Institutional Architecture of Non-Accountability

The Doctrine That Makes Declinations Consequence-Free

In 1976, the United States Supreme Court decided Imbler v. Pachtman, 424 U.S. 409, and established what became the foundational rule of prosecutorial accountability in American law: prosecutors cannot be sued in civil court for decisions made in their official capacity. The majority opinion, written by Justice Lewis Powell, was explicit about the trade-off involved. Granting immunity, the Court acknowledged, leaves “the genuinely wronged criminal defendant without civil redress against a prosecutor whose malicious or dishonest action deprives him of liberty.” The Court considered that acceptable because the alternative, subjecting prosecutors to the threat of personal liability, would, in the Court’s assessment, undermine vigorous and fearless prosecution.

The reasoning has a surface logic. Prosecutors who fear civil suits for losing cases might decline to bring difficult charges. Prosecutors who fear liability for every judgment call might second-guess decisions that ought to be made quickly and without hesitation. The doctrine was built to prevent that outcome.

What it actually built was something different.

In 2011, the Supreme Court extended the architecture further in Connick v. Thompson, 563 U.S. 51. John Thompson had spent 18 years wrongfully incarcerated, 14 of them on death row, while prosecutors in the Orleans Parish District Attorney’s Office possessed a blood type report that would have exonerated him. A jury awarded Thompson $14 million after finding that the office had failed to train its prosecutors on their constitutional obligation to disclose exculpatory evidence. The Supreme Court reversed the award. The office could not be held liable for a single Brady violation, the Court held, because one instance of misconduct does not establish a pattern sufficient to constitute deliberate indifference to training obligations. Thompson got nothing. The office owed him nothing. No one was sanctioned.

The Innocence Project reviewed 75 DNA exonerations and found that prosecutorial misconduct contributed to nearly half. None of the prosecutors involved faced serious professional sanction. The Death Penalty Information Center has identified more than 600 instances in which a capital conviction or death sentence was overturned or a death-row exoneree was wrongfully convicted as a result of prosecutorial misconduct. Official misconduct was present in all eight cases in which an exoneration occurred 31 or more years after conviction. Exonerations take longer when official misconduct is involved, in part because prosecutorial offices tend to more aggressively defend cases in which their own conduct is at issue.

The Ballot Box Is Not a Forensic Audit

Defenders of the current immunity framework typically point to electoral accountability as the check that replaces civil liability. Prosecutors are elected. Voters can remove them. An office that persistently fails the community will eventually face a competitive election and lose.

That argument depends on voters having access to information they do not, in practice, possess. Prosecutors decline warrant requests daily. Those decisions are not public records in most jurisdictions. They are not announced. They are not aggregated into publicly accessible databases. They do not appear in the metrics on which prosecutors campaign — conviction rates, violent crime clearances, high-profile case outcomes. A prosecutor who declines a warrant request in 2017 because a victim is unhoused and unlikely to cooperate does not face a 2018 campaign ad about that declination. The victim has no standing to compel disclosure. The investigating officers may not even be informed of the reasoning. The record closes and the candidate runs on the cases they prosecuted, not the ones they declined.

Michigan has five of the most exonerations of any state in the country, according to the National Registry of Exonerations: 169 wrongful convictions in state courts since 1989, averaging nearly eleven years of wrongful incarceration per case. Michigan also passed the Wrongful Imprisonment Compensation Act to provide a remedy for exonerees. A 2024 ProPublica investigation found that that compensation mechanism has been repeatedly denied to exonerees whose convictions were overturned on grounds of prosecutorial misconduct, because the statute’s language conditions compensation on actual innocence being the basis for relief, not the misconduct itself. The office whose conduct produced the wrongful conviction has no obligation to fund or support the compensation claim. The exoneree, having already lost years to a wrongful conviction, must then navigate a compensation system designed to limit rather than provide relief.

Michigan Cases: When the Office Decides Not to Look

The pattern of prosecutorial declination producing downstream harm is not a theoretical problem in Michigan. It is a documented one.

In Genesee County in 2017, a warrant request was submitted to the office of Prosecutor David S. Leyton after a woman fled a rural property and told Michigan State Police that a man named Duane Reynolds had taken her there. Leyton’s office declined. The stated basis was the absence of a complainant and a victim who reported being hurt. Reynolds died in December 2024. In March 2026, a new property owner found human remains on Reynolds’ Willard Road property. At least four sets of unidentified human remains have since been recovered across two properties Reynolds owned. None have been identified. No charges will ever be filed. Reynolds is dead.

It was not Leyton’s first documented declination to draw subsequent scrutiny. In 2005, shortly after taking office, Leyton declined to file charges in the 2003 death of an 11-year-old Flint boy whose mother alleged he had been poisoned before drowning in the Flint River. Leyton’s office declined despite, according to the boy’s mother, witness statements and post-mortem evidence pointing to poisoning. Two years later, then-Attorney General Mike Cox filed charges. The boy’s stepmother and her brother were convicted of poisoning and drowning the child. Leyton’s office was not publicly revisited for the prior declination after the conviction. The immunity doctrine made that revisitation legally unnecessary. The political structure made it personally counterproductive for any successor or opponent to press the point.

In Oakland County, the pattern is longer and better documented. Juwan Deering spent 15 years in prison for the fire-related deaths of five children in suburban Detroit before murder charges were dismissed in 2021 after an investigation found misconduct by both police and prosecutors. Oakland County Prosecutor Karen McDonald dismissed the charges. Her office — not the office that prosecuted Deering — made the call. The institutionally uncomfortable fact is that successor offices are routinely positioned as the accountability mechanism for the decisions of predecessor offices, which means the accountability flows from a different elected official with a different political interest, and only when the misconduct becomes undeniable enough to survive the institutional pressure to defend it.

The Patterson Contradiction: What a Documented Record Looks Like

The Oakland County Child Killer investigation is the most extensively documented Michigan instance of how prosecutorial declination, institutional self-protection, and the immunity architecture interact across time. It is also one of the clearest examples of what happens when a prosecutor’s public statements are later tested against the contemporaneous documentary record — and the record wins.

Christopher Busch was a convicted pedophile and the son of a General Motors executive. In January 1977, three of the four Oakland County Child Killer victims were already dead. Busch’s associate Gregory Greene was arrested in Flint on criminal sexual conduct charges and told investigators that Busch had murdered Mark Stebbins, the first OCCK victim. Busch was brought in. During the pre-test interview for his polygraph examination, he named Ferndale, Royal Oak, and Berkley as locations where he actively sought boys. Those are, in chronological order, the first three OCCK abduction sites. He was administered a polygraph, cleared, and released. Three subsequent analysts have since concluded that Busch either failed the examination or that the results were inconclusive.

L. Brooks Patterson was the Oakland County Prosecutor at the time. His chief deputy, Richard Thompson, drove to Flint through a blizzard in late January 1977 to personally attend Busch’s interrogation at Flint Police Department — a fact documented in investigative file reports from Southfield Detective Lourn Doan. Patterson was then quoted by name in the Detroit Free Press on March 1, 1977, discussing Christopher Busch by name in connection with the OCCK investigation and stating that Busch had been cleared in all six slayings. That article ran fifteen days before Timothy King was abducted from the Hunter-Maple Pharmacy parking lot in Birmingham. He was the fourth and final OCCK victim.

Patterson held the Oakland County Prosecutor’s office for sixteen years before becoming county executive, a position he held until his death. He was succeeded as prosecutor by Richard Thompson — the same deputy who had been present at the Busch interrogation in Flint. Thompson held the office until 2000. The successor offices inherited a prosecutorial record in which the handling of the Busch lead remained a live institutional liability. No Oakland County Prosecutor’s Office in the decades since 1977 has formally acknowledged that the 1977 polygraph clearance was erroneous, that Thompson’s presence at the interrogation was inconsistent with Patterson’s later claim of ignorance, or that the institutional decision to close the Busch file may have contributed to the fourth OCCK abduction. The immunity doctrine makes none of that acknowledgment legally compelled. The electoral structure makes it politically costly. The result is a record that successive offices have managed rather than corrected.

How Successor Offices Become Cover Operations

The mechanism by which future prosecutorial offices end up defending the decisions of past ones is not conspiratorial. It does not require active coordination or deliberate concealment, though both have occurred in documented cases. It operates through institutional incentive, and it operates reliably.

When a prosecutor retires or dies and a successor takes office, the successor inherits the institutional record. That record includes every charging decision, every declination, every plea agreement, and every case that was tried and lost. Some portion of those decisions will, in retrospect, have been wrong. Some will have been catastrophically wrong. The successor’s institutional interest is not in identifying which ones. The successor’s institutional interest is in running the office effectively, building a record that supports reelection, and not creating liability — political or legal — by opening investigations into the prior administration’s work.

Formally acknowledging that a predecessor’s declination enabled ongoing harm requires the successor to do several things simultaneously: investigate a matter the predecessor closed, produce findings that damage the institutional reputation the successor inherited, create a political narrative that attributes responsibility to an office now under different management, and do all of this without a legal mechanism compelling the disclosure or a civil remedy creating consequences for not doing it. The immunity doctrine ensures the last condition. The electoral structure ensures the others. The result is not cover-up in the active sense. It is institutional inertia with predictable directionality.

This is the specific harm that absolute immunity produces that is most invisible to the public and least addressed by reform proposals focused on wrongful convictions: it is not only the person wrongfully convicted who lacks a remedy. It is also the victim of a crime that was not investigated adequately — the person whose report was declined, whose case was closed, who was told by the institutional record that there was nothing to charge — who has no remedy when the subsequent evidence establishes that the declination was wrong. The wrongful conviction victim was in the system. The victim of an inadequate declination was never allowed in.

What Reform Requires

The absolute immunity doctrine established in Imbler is a judicial creation. The Supreme Court built it from common law reasoning and policy concerns, not from constitutional text. Congress can modify or eliminate it by statute under 42 U.S.C. Section 1983. State legislatures can create parallel accountability mechanisms under state law that impose consequences the federal immunity doctrine does not reach. Neither path requires overturning Imbler directly.

The reform conversation in prosecutorial accountability has historically focused on misconduct in active prosecutions: Brady violations, suppression of exculpatory evidence, false testimony, inflammatory trial conduct. That focus is warranted. It is also incomplete. It addresses the cases that produced convictions that can later be challenged. It does not address the cases that never produced a charge — the declinations that closed files, ended investigations, and left the record showing only that the matter was reviewed and found insufficient.

The question the current architecture cannot answer is not whether a given declination was wrong. It is how many equivalent declinations are sitting in closed files across Michigan’s 83 county prosecutor offices, attached to subjects who have not yet died, whose properties have not yet been renovated by someone who found something in the ground. The Reynolds matter is visible now because circumstances forced it into visibility. The immunity doctrine ensures that visibility is the exception. It was designed to.

You Cannot Remove Them, and They Know It

The electoral accountability argument for prosecutorial immunity assumes that voters can remove a prosecutor whose decisions they find inadequate. In Michigan, that assumption runs into a statutory framework that makes removal of a sitting prosecutor considerably harder than the theory suggests, and that sitting prosecutors understand better than most of their constituents.

Michigan provides three formal removal mechanisms for a county prosecutor. The first is defeat at the next general election, which occurs on a four-year cycle. The second is gubernatorial removal under MCL 168.207, which authorizes the Governor to remove a county officer upon “sufficient evidence” of official misconduct, willful neglect of duty, or extortion. The third is citizen recall under Article II, Section 8 of the Michigan Constitution and the Michigan Election Law recall provisions.

Each mechanism is narrower in practice than it appears on paper. Gubernatorial removal requires official misconduct or willful neglect of duty — a standard that prosecutorial discretion decisions almost never satisfy because the immunity doctrine and the broad deference courts extend to charging decisions effectively foreclose the argument that a declination, however inadequate, constitutes willful neglect. A prosecutor who declines a warrant request for reasons the community later finds to have been wrong has not, in the legal sense, neglected a duty. They have exercised discretion. The Governor cannot remove an official for exercising discretion, even if that discretion produced a catastrophic outcome.

The practical consequence is that a Michigan prosecutor who makes a decision the community considers inadequate — a warrant declined, a case closed, a pattern of declining certain categories of cases — has substantial structural protection from removal during most of their term. That protection compounds with the immunity doctrine: not only can they not be sued, they cannot easily be removed. The two layers of insulation operate together to create an officeholder who has no legal liability for charging decisions, limited political liability during most of their term, and the formal authority to exercise discretion in ways that produce documented harm without formal consequence.

When Being Watched Becomes a Charging Decision

The removal problem describes what happens when a prosecutor’s decisions are inadequate and the community wants accountability. The retaliation problem describes what happens when a prosecutor feels threatened by scrutiny and decides to use the charging power to manage it.

That is not a theoretical concern in Michigan. It is a documented pattern with a case currently before the Michigan Supreme Court.

Robert Reeves challenged Wayne County’s civil forfeiture practices and, in February 2020, filed a federal class action lawsuit with the Institute for Justice aimed at the county’s car-forfeiture program. The day after that lawsuit was filed, Wayne County prosecutors revived a long-dormant investigation into Reeves and charged him with concealing stolen property. A judge dismissed those charges for lack of evidence. Prosecutors refiled. A judge dismissed them again, again for lack of evidence. Reeves filed a second lawsuit alleging the prosecutions were a coordinated effort to derail his civil rights case and intimidate others who might raise similar challenges.

The Reeves case is documented and currently litigated. The broader pattern it represents is not limited to Wayne County or to civil forfeiture challenges. The charging power is the most consequential tool any local government official in Michigan wields. It can end a person’s career, destroy their finances, consume years of their life in the criminal justice process, and attach a permanent public record to their name — all before a single conviction. In a system where charges can be filed and later dismissed without consequence to the office that filed them, the threat of prosecution is itself a form of control. An office that understands this, and that operates without meaningful external oversight, has every structural incentive to use that threat when it feels the scrutiny getting too close.

Why Prosecutors Lie Even When the Record Is Right There

The immunity architecture explains why there are no external consequences for getting it wrong. It does not fully explain why prosecutors and their successors actively manage and control the narrative around their decisions, sometimes in direct contradiction of a documentary record that is sitting in a file they can access. That behavior has a different explanation, and it operates alongside the legal structure rather than because of it.

Prosecution is one of the few professional roles in American public life where certainty is the product being sold. Prosecutors do not argue probability. They argue guilt beyond a reasonable doubt. The entire performance of the role — in courtrooms, in press conferences, in campaign materials, in interactions with victims’ families — is built around the projection of confidence in the correctness of every decision the office makes. A charging decision that later turns out to have been wrong does not merely represent an error. It represents a crack in the foundational claim the office makes about itself: that it knows, that it investigated adequately, and that its judgment can be trusted.

That is not a small crack. For an elected official whose entire professional identity is constructed around the certainty of prosecutorial judgment, acknowledging that a warrant request was declined inadequately, or that a suspect was cleared on a polygraph that three subsequent analysts found unreliable, or that a case was closed because a victim was unhoused and unlikely to cooperate — any of those acknowledgments requires the official to say, in public, that the certainty they sold was not real. For most prosecutors, and for most of the institutions they lead, that is experienced not as correction but as annihilation.

Patterson’s 2012 interview is a clinical example of what narrative management looks like when the record contradicts it. He did not say the decision to clear Busch was correct. He said he was not responsible for it. He said his office handled it, not him personally. He said he did not remember the case. He said he never saw the name until 2008. Each of those statements redirects responsibility: from Patterson to subordinates, from active decision to passive ignorance, from documented involvement to claimed non-involvement. The March 1, 1977 Detroit Free Press article, which carries his name and his words about Christopher Busch, was not a secret document. It was a published newspaper record available in any archive. The interview was conducted in 2012, thirty-five years after the article ran. Either Patterson believed the record would not be checked, or he had been telling the same story for so long that it had, for him, become the true one.

Both of those possibilities describe the same phenomenon: narrative management that has become narrative replacement. The story the office told about the Busch investigation — that Busch was cleared, that the investigation was adequate, that there was nothing more to charge — was told so consistently, over so many years, through so many institutional channels, that it displaced the underlying record in the minds of the people who had been responsible for it. This is not unique to Patterson or to the OCCK investigation. It is how institutional self-protection works when the immunity doctrine removes the external pressure that would otherwise force a reckoning with the record.

The result is a class of public officials who have both the legal protection to avoid accountability and the professional and psychological incentive to actively construct an alternative to it. The immunity doctrine ensures they cannot be compelled to tell the truth by a civil court. The institutional culture ensures they are rarely inclined to volunteer it. And the electoral structure ensures that the voters who might otherwise demand it are working from the narrative the office has already supplied.

This is why reform proposals that focus only on the legal architecture are insufficient. Removing absolute immunity would create a civil remedy. It would not change the fact that prosecutors are trained, selected, and culturally reinforced to treat their own judgment as unassailable. Meaningful reform requires both the external mechanism that imposes consequences and the internal mechanism that changes what professional success looks like for a prosecutor. An office measured only on conviction rates will protect its conviction rate. An office measured on case review accuracy, declination pattern equity, and post-closure accountability will protect those things instead. The metric shapes the culture. The current metrics have shaped the culture we have.