The term ERD can sound technical enough to drain the moral force out of it. In MDOC's public OTIS language, the Parole Board Jurisdiction Date is the prisoner's earliest release date. MDOC's parole materials also make the limit clear: parole is not automatic. The board considers risk, conduct, program performance, interview information, and other factors. So the metric on this site does not say, and should not be read to say, that every person counted here is being illegally held.
The claim is narrower and stronger: once a person has reached first parole eligibility, every additional day of custody needs a defensible public explanation. When the cumulative total reaches millions of person-days, the question is no longer only individual conduct. It becomes a question about administration, programming access, staffing, parole-board capacity, classification decisions, documentation, institutional delay, and incentives.
Required programming is not just an individual burden.
The phrase "failed to complete required programming" can make responsibility appear simple. But an incarcerated person does not control the program catalog, the waitlist, the staffing level, the transfer decision, the schedule, or the records used to prove completion. If a person is told release depends on a program the institution does not provide in time, then delay is not merely a personal failure. It is a system output.
That distinction matters because the state has extraordinary power here. Courts impose sentences. MDOC controls custody and access to the rehabilitative machinery inside custody. The Parole Board controls release after jurisdiction attaches. When those handoffs become opaque, delay can hide in plain sight: formally lawful, administratively routine, and still morally indefensible.
What the number does, and does not, prove.
The clock measures people in the local MDOC public-record mirror who are currently marked Prisoner and whose numeric ERD/PBJD is on or before the evaluation date. It is aggregate only. It does not publish a person-level over-ERD list. It does not decide anyone's parole case. It does not replace official records.
What it does is convert hidden administrative time into a public accountability surface. Time past eligibility is not abstract overhead. It is birthdays, work histories, family caregiving, medical decline, lost wages, grief, aging, and public money. Every second on the clock is human time the institution should be prepared to justify.
The reform demand is transparency.
A serious reform agenda should not stop at outrage. It should ask for measurements that make preventable delay harder to bury.
- Publish facility-level required-program waitlists and average time-to-placement.
- Report parole-denial reasons in aggregate, including unavailable or delayed programming.
- Track people held six months, one year, two years, and five years past first eligibility.
- Separate individual refusal from institutional non-access in public reporting.
- Require a concrete remedial plan when a release barrier is controlled by the institution.
Michigan law already recognizes the public importance of this category. MCL 791.233e requires reporting on people with a high probability of parole who were incarcerated at least six months past their first parole eligibility date. The clock here extends that instinct into a public-facing research question: if the state can keep counting custody, the public can count delay.