End the Warehouse: Prison Transformation Plan

Brief written August 1, 2026 from GPS Intelligence System data.

Georgia Prisoners' Speak (GPS) launched the End the Warehouse campaign on a simple premise: the crisis inside Georgia's prisons is not a management problem that one commissioner or one budget cycle can fix. It is a design. The state now spends more than it ever has on a system that holds roughly 50,000 people, that the U.S. Department of Justice has found unconstitutional, and in which GPS has independently tracked 1,860 deaths in custody since 2020 — while the mechanisms that should expose those facts (grievances, audits, death investigations, legislative oversight) have been systematically hollowed out. This page assembles the evidentiary case for that diagnosis and, just as important, the proof that transformation is achievable: court-tested population reduction, costed oversight models from other states, rehabilitation investments with measured returns, and legal tools unique to Georgia that have never been fully used.

The Warehouse by the Numbers

The money has never been the constraint. GPS's budget analysis traces GDC spending from roughly $1.08 billion in FY2010 — essentially flat for eight years — to an all-time peak of $1,913,888,054 in actual FY2025 spending, a 57 percent surge in just four years, per the Governor's Budget Report FY2027. The FY2026 amended budget sits at approximately $1.80 billion, and the approved FY2027 budget at roughly $1.79 billion. GDC operates 38 state prisons; its July 24, 2026 weekly population snapshot shows 50,014 people in custody — 34,979 in state prisons, 8,258 in private prisons, 4,063 in county prisons, 2,714 in transitional centers — with a backlog of 2,572 state inmates waiting in county jails and 5,753 people aged 60 or older.

What that spending bought is documented in the federal record. The DOJ's October 1, 2024 findings report — 93 pages, produced after what DOJ described as an "unnecessarily contentious" records fight in which GDC refused to produce documents until a subpoena was enforced and conducted pre-inspection cleanups — found reasonable cause to believe Georgia violates the Eighth Amendment by failing to protect medium- and close-security prisoners from violence and LGBTI prisoners from sexual abuse. DOJ documented GDC homicides climbing year by year: 7 in 2018, 13 in 2019, 28 in 2020, 28 in 2021, 31 in 2022, 35 in 2023. More than 1,400 reported incidents of violence were logged from January 2022 through April 2023 across 24 prisons. Correctional-officer vacancy rates have exceeded 50 percent systemwide since mid-2021 and topped 70 percent at eight facilities. GDC publicly rejected the findings as reflecting a fundamental misunderstanding; DOJ recommended 82 remedial measures and noted the Attorney General may sue under CRIPA.

The dying has not paused. GPS's mortality database recorded at least eight deaths across six facilities in the first twenty days of July 2026 alone — at Central State Prison, Pulaski State Prison, Valdosta State Prison, Baldwin State Prison, Augusta State Medical Prison, and the Burruss Correctional Training Center. GPS reporting documented that on April 1, 2026, coordinated gang violence locked down at least twelve prisons, with life flights dispatched to two facilities and stabbings at five. Beyond the public record, GPS has received accounts of severe violence at multiple Georgia prisons — assaults on officers left working isolation units alone, hostage-takings, and in-custody attacks attributed to gang control of housing units.

Follow the Money: Contracts, Kickbacks, and Private Beds

The warehouse is also a revenue center. GDC's inmate telephone contract with Securus, dating to September 2016, paid the department a $4 million one-time incentive plus a 59.6 percent commission on call revenue, subject to a $325,000 monthly floor — roughly $8 million a year flowing back to the agency while families paid approximately 13 cents per minute. Such commissions pool in "Inmate Welfare Funds," opaque accounts that function as shadow budgets free of legislative appropriation oversight, and commission-based contracts give corrections officials a direct financial incentive to approve higher prices. An investigation by The Appeal collecting commissary prices from 46 states found markups ranging from 40 to 600 percent above retail. In 2021, the Consumer Financial Protection Bureau penalized JPay for charging fees on prepaid release cards and conditioning "gate money" on JPay debit cards — conduct the CFPB found violated the law in Georgia among other states — ordering $4 million in consumer redress and a $2 million civil penalty.

The state's own budget documents show where the new money goes. The FY2026 budget increased per-diem payments to the four private prisons — CoreCivic's Coffee, Wheeler, and Jenkins facilities and GEO Group's Riverbend — by a combined $13.6 million, and the proposed FY2027 budget adds $4.2 million for 263 new private prison beds at Coffee and Wheeler. The approved FY2027 budget adds $5.5 million for the Over Watch and Logistics surveillance unit; a peer-led programming pilot at Autry State Prison received $150,000. Capital commitments run the same direction: $130 million to purchase the McRae facility in 2022 and $436.7 million approved in February 2024 for a new Washington State Prison designed for 1,500 single cells. Meanwhile the state's litigation exposure — a floor of $50,633,556 across 261 prison death and injury claims, compiled by GPS from Department of Administrative Services risk-management records — is treated as a routine cost of doing business.

Who Really Pays: Families as the Hidden Tax Base

Appropriations are only the visible half of the system's financing. A first-of-its-kind 2025 FWD.us national survey of more than 1,600 people with incarcerated family members, developed with researchers at Duke University and NORC, found that 64 percent of family-incarcerated person pairs incur direct incarceration expenses, at a median of $172 per month — 6 percent of household income. The burden is racially stratified: Black families average $413 per month in contributions, compared to $365 for Hispanic and $252 for white families; spouses and coparents spend a median of $276 monthly, 12 percent of household income, the highest of any relationship category. Sykes and Maroto's 2016 analysis of federal survey data found that having an incarcerated family member reduced household assets by 64.3 percent. Children of recently incarcerated fathers are three times more likely to experience homelessness.

Inside the fence, the extraction is layered onto unpaid labor. Georgia is one of seven states — alongside Alabama, Arkansas, Florida, Mississippi, South Carolina, and Texas — that pay incarcerated people nothing for most prison work, and O.C.G.A. § 42-5-59 mandates that "the cost of the inmate's keep and confinement shall be deducted" from whatever earnings exist. A Science Advances editorial put the arithmetic plainly: a $0.35 package of ramen costs three minutes of work at the federal minimum wage and two hours at prison wages. Forty states and the federal Bureau of Prisons charge medical copays of $2 to $13 — a policy the National Consumer Law Center found inversely related to healthcare utilization, and which a 2024 JAMA study linked to reduced access for pregnant people and those with chronic conditions, in a system where 40 percent of state prisoners report chronic illness.

The policy alternatives are no longer theoretical. The Martha Wright-Reed Act, signed January 5, 2023, gave the FCC authority over all prison communications rates, and the commission's July 2024 order capped prison phone calls at 6 cents per minute and banned site commissions — before a new FCC majority in 2025 suspended the rules, raised interim caps, and postponed compliance to April 1, 2027, a reversal Worth Rises' Bianca Tylek described as the Commission having "bent to the will of the industry." Five states — California, Connecticut, Massachusetts, Minnesota, and Colorado — now mandate free prison communications; at least nine prohibit commission-based telecom contracts outright. On pay-to-stay, 48 states still allow at least one fee category, but California and Illinois have repealed all of them, Missouri followed in 2025, and Dauphin County, Pennsylvania forgave $65.9 million in detainee debt in 2024. Eight reforms across six states lifted $2.6 billion in outstanding fine and fee debt — often at no revenue loss, because the debt was uncollectible.

An Accountability Vacuum by Design

Georgia is not among the roughly 20 states plus the District of Columbia with an independent external prison oversight body — and a GPS search of the legislative record, LegiScan, and secondary sources found no bill proposing a corrections ombudsman, oversight board, or inspection commission for GDC has ever been introduced in the General Assembly. What exists instead is a captured structure: the Board of Corrections' 19 members are all governor-appointed, the board conducts no independent inspections and employs no monitors, and it appoints the very commissioner it notionally oversees. GDC's internal investigation reports are classified "confidential state secrets" under Ga. Comp. R. & Regs. 125-1-2-.11, releasable only at the Commissioner's discretion. The General Assembly itself is exempt from the Open Records Act under the Georgia Court of Appeals' 2019 decision in Institute for Justice v. Reilly.

When the Senate did act, it acted around the problem. SR 570, adopted 53–0 in February 2024, created a study committee chaired by Majority Whip Randy Robertson; its final report, adopted unanimously on December 13, 2024, recommended single-cell conversion, cell-phone and drone jamming, staffing investment, and a contract-cost audit — and explicitly declined to create an independent oversight body, with Robertson characterizing oversight as "redundant," according to Georgia Recorder and WABE reporting summarized in GPS's research. A separate suggestion by Sen. Bearden to have the GBI investigate all in-custody deaths did not make the final list. Georgia's congressional delegation proved the alternative is possible: the Federal Prison Oversight Act, led by Sen. Jon Ossoff and Rep. Lucy McBath and signed July 25, 2024, created an independent inspection regime and ombudsman — but it covers only the 122 federal Bureau of Prisons facilities, not a single Georgia state prison.

The Grievance Machine and the Litigation Wall

The one accountability channel nominally available to incarcerated people is engineered to fail. GDC's SOP 227.02 requires grievances to be filed electronically by JPay kiosk or tablet within 10 calendar days of the underlying facts; mailed grievances "will not be processed except for Good Cause"; no more than two grievances may be active at once; and each filing is limited to a single issue, a single attached page written on one side, and language free of anything deemed profane or insulting — each a documented rejection trigger. Transfers, housing, classification, work assignments, and disciplinary actions are non-grievable, and — in a paradox GPS's grievance-process research highlights — the GOAL tablets through which grievances must be filed are themselves non-grievable because tablet access "is a privilege, not a right." The warden has 40 days to respond; the prisoner then has 7 days to appeal; the Commissioner has 120 days to decide. Retaliation for filing is prohibited on paper, but no SOP specifies any consequence for staff who retaliate, and the semi-annual grievance reports and annual audits the policy requires are never published.

The trap closes in federal court. Under the PLRA, 42 U.S.C. § 1997e(a), no prisoner may sue over conditions until available administrative remedies are exhausted, and the Supreme Court's decisions in Woodford v. Ngo (2006) and Jones v. Bock (2007) mean the prison's own procedural rules — GDC's ten-day window, its format rules — define the exhaustion standard. Eleventh Circuit precedent from Georgia cases shows the attrition in practice: Johnson v. Meadows (2005) made an untimely grievance a fatal default; Major v. Toole (2017) dismissed an excessive-force suit because the prisoner's timely grievance mentioned only confiscated property and "never mentioned excessive force." Escape valves exist — Turner v. Burnside (2008) recognizes retaliation can make remedies "unavailable," Geter v. Baldwin State Prison (2020) extended that to mental disability and staff interference, and the Supreme Court's 2025 Perttu v. Richards guarantees a jury trial when exhaustion intertwines with the merits — but an exhaustion dismissal remains, in Bryant v. Rich's words, "nothing more than a precondition," never a finding that the abuse did not occur.

The measurable result is silence. Margo Schlanger's data show the national prisoner civil-rights filing rate fell from 24.6 per 1,000 incarcerated people in FY1995 to 10.5 by 2012; Georgia's rate fell 57 percent, from 22.6 per 1,000 (1,496 filings) in FY1995 to 9.7 per 1,000 (984 filings) in FY2019. The PLRA's physical-injury bar, three-strikes provision, $405 filing fee, and attorney-fee caps complete the wall. DOJ's findings confirm the reporting culture this produces: incarcerated people told federal investigators they "do not always report incidents because they do not expect staff to take any action," and sexual assaults go unreported "either for fear of retaliation... or because incarcerated people believe GDC will fail to address their complaints." DOJ found GDC failed to protect even people who cooperated with the federal investigation from ongoing retaliation. GPS's own review of its case files reached the same conclusion: retaliation functions not as a series of accidents but as a system whose purpose is silence.

Paper Compliance and Hidden Deaths

The audit record tells the opposite story from the federal record — which is the problem. GPS pulled and analyzed all 273 GDC facility PREA audit reports across five cycles covering 139 facilities: not one standard was ever found "not met." During that same period, Georgia prisons logged 15,542 PREA allegations from 2014 through 2024 and substantiated 543 — a 3.5 percent aggregate rate. In 2022 alone, 456 documented allegations of sexual abuse between incarcerated people produced 35 substantiations. DOJ's certified PREA auditors visited 17 prisons in 2022–2023 and concluded sexual assault is "rampant" and that GDC "does not reasonably protect incarcerated individuals, including LGBTI individuals, from sexual harm." Federal auditors and the state's own auditors examined the same prisons and reached opposite conclusions.

Death investigation is structured the same way. In 155 of Georgia's 159 counties, the official who takes charge of a body and can certify a death is an elected coroner with no medical training requirement whatsoever — a system the National Academy of Sciences recommended replacing with medical examiners in 2009. Georgia law makes a medical examiner's inquiry mandatory for any in-custody death but leaves the autopsy decision to sole discretion; the GBI's authority over deaths on state property is "authorized," not required; and under GDC's SOP 208.03, the department's own Criminal Investigations Division decides whether to notify the GBI at all — GDC gatekeeps outside involvement in its own deaths. In March 2024, as the Atlanta Journal-Constitution first reported, GDC stopped providing cause-of-death information in its monthly mortality reports altogether. The federal backstop is broken too: the GAO found nearly 1,000 deaths potentially unreported under the Death in Custody Reporting Act in FY2021 and 70 percent of state-provided records missing required elements, and BJS stopped publishing detailed prison mortality data after 2019. Even basic health surveillance fails: GDC's own June 2026 statistical profile shows no hepatitis C test result on file for 27,471 people — 51 percent of the custody population — against 87.3 percent HIV testing coverage in the same document, implying roughly 1,945 undetected infections at GDC's own positivity rate.

The Proof of Concept: Court-Ordered Transformation Actually Happened

The strongest evidence that a warehouse can be dismantled is that it has been. California's prisons held approximately 156,000 people in space designed for roughly 85,000 — nearly 200 percent of capacity — when the Coleman and Plata litigation established what overcrowding meant in practice: an average of one unnecessary death per week, a 54.1 percent psychiatrist vacancy rate, clinics without running water, and a medical system Judge Thelton Henderson ruled in 2005 was "broken beyond repair." After more than 70 prior court orders failed, a three-judge panel — the first ever convened under the PLRA's prisoner-release provision — ordered the population cut to 137.5 percent of design capacity, roughly 46,000 people, the largest court-ordered reduction in U.S. history. The Supreme Court affirmed 5–4 in May 2011, with Justice Kennedy writing that releasing prisoners "could even improve public safety" and Justice Scalia reading his dissent from the bench for over nine minutes.

California's response, the 2011 Public Safety Realignment Act, amended approximately 500 criminal statutes, eliminated state prison time for lower-level felonies, and funded counties $400 million in its first year growing to over $1 billion annually. The results: new prison admissions dropped 41 percent in eight months; the population fell 28,300 in that window and 27,400 by September 2012; the Bureau of Justice Statistics attributed 70 percent of the total national state-prison decrease in 2010–2011 to California alone. Studies found Realignment did not increase violent crime, and the total incarcerated population fell because jail growth did not match prison decline. The cautionary chapters matter as much as the blueprint: the state exported more than 10,000 prisoners out of state, full compliance waited on Proposition 47 in November 2014, the Coleman case remains active after 35 years, and in March 2025 a federal judge placed California's prison mental-health system into receivership with $112 million in contempt fines. The strategic lesson GPS draws is sequencing: the California record took over twenty years to build, and the PLRA's seven statutory requirements — crowding as primary cause, failed less-intrusive remedies, narrow tailoring — must be satisfied in that order. Georgia's DOJ findings and GDC's documented non-compliance are that record being built now.

What Transformation Buys

The program evidence is equally concrete. The RAND Corporation's 2013 meta-analysis of 57 studies found correctional education participants had 43 percent lower odds of recidivism and that every dollar invested saves $4 to $5 in reincarceration costs — the finding that anchored the restoration of Pell Grant access for incarcerated students. Landenberger and Lipsey's meta-analysis of 58 studies found cognitive-behavioral therapy reduces recidivism by 25 percent on average, with the largest effects for higher-risk participants. Incarceration-based therapeutic-community drug treatment showed statistically significant effects in 30 of 35 evaluations reviewed. Oregon cut solitary confinement use by 55.7 to 73.9 percent; North Dakota by 74.28 percent. The U.S. Sentencing Commission found offenders over 50 recidivate at 21.3 percent — less than half the 53.4 percent rate of younger offenders — over an eight-year follow-up, which matters in a system holding 5,753 people over 60. Against GDC's reported three-year felony reconviction rates of 26 to 31 percent, the warehouse model is the control group that keeps losing.

Oversight, the reform Georgia has never tried, is cheap and well-specified elsewhere. Washington's Office of the Corrections Ombuds sits in the governor's office with for-cause removal protection, "golden key" access to inspect and photograph all facility areas, five-day mandatory records production for deaths and assaults, and communications privileged like legal mail — at roughly $191 per incarcerated person per year. Minnesota's ombudsperson holds subpoena power at $147 per person; Connecticut's oversight costs about $75. Applied to Georgia's 50,000-person system, those models price out at roughly $4 to $10 million annually — about half of one percent of GDC's budget, and less than the FY2027 surveillance-technology increase alone. The cautionary designs are documented too: Texas houses its ombudsman's budget inside the corrections agency, and Hawaii created its commission in 2019 but saw funding withheld until 2022. New Jersey's ombudsperson conducts unannounced inspections and releases reports without outside approval; Hawaii's five-member commission splits appointments across five separate authorities to resist capture. Virginia's statute directs its oversight body to run a uniform, published data system covering deaths, assaults, staffing, and grievances — precisely the data streams GDC compiles internally and never releases.

Georgia's Dormant Legal Tools

Georgia's own constitution contains a provision no other state has. Article I, § 1, ¶ XVII — adopted at the 1868 Reconstruction Convention on the motion of Richard Whiteley, alongside the abolition of debtors' prison and whipping — prohibits the abuse of any person "in being arrested, while under arrest, or in prison." The Georgia Court of Appeals held in Long v. Jones that the Abuse Provision "provides an independent state ground" for relief, offering at least as much protection as the Fourteenth Amendment, and the Georgia Supreme Court's interpretive canon in Camden County v. Sweatt requires that it mean something beyond the cruel-and-unusual-punishment clause it follows. Yet in 158 years it has been cited only a handful of times and never fully interpreted — a sleeping cause of action in a state where the federal floor has demonstrably failed.

Other tools need only legislative retrieval. Georgia's victim-services statutes are structurally blind to in-custody harm: O.C.G.A. § 17-15-7(c) categorically bars crime-victims compensation, and § 17-17-3's definitions exclude incarcerated victims' surviving relations, even though the victim-offender overlap has been an established empirical regularity since Lauritsen, Sampson, and Laub's 1991 work. On forensic accountability, Georgia has none of the mechanisms comparator states built after the FBI's 2015 hair-comparison review found errors in roughly 90 percent of examined transcripts: seventeen states audited their hair-comparison convictions; Georgia is absent from that list, has no forensic science commission, no junk-science writ, and no innocence inquiry commission — though the Georgia Supreme Court's unanimous October 2025 decision in Smith v. State opened a path by holding that evolved scientific understanding can constitute newly discovered evidence, and the Wrongful Conviction and Incarceration Compensation Act, signed May 14, 2025, now pays $75,000 per year of wrongful incarceration. Neighboring states mandate what Georgia leaves discretionary: Virginia has required an autopsy after every in-custody death at a state prison since January 2024, Wisconsin mandates autopsies for all prison deaths, Maine requires examination and inquest, and California's 2024 FACTS Act forces sheriff-coroner counties to outsource in-custody autopsies to independent agencies.

The End the Warehouse Agenda

The plan follows the evidence. Its planks:

1. Independent oversight with teeth. Enact a statutory corrections ombuds or inspection body on the Washington–Minnesota–New Jersey model: golden-key unannounced access, subpoena power, confidential and privileged communications, a published statewide data stream for deaths, assaults, staffing, and grievances on the Virginia template, capture-resistant split appointments on the Hawaii template, and a budget independent of the agency it oversees. Cost: an estimated $4–10 million a year, against a $1.9 billion budget and a $50.6 million settlement floor.

2. Make death visible again. Restore cause-of-death publication in GDC mortality reports; mandate autopsy and independent investigation for every in-custody death on the Virginia, Wisconsin, Maine, and California models; make GBI jurisdiction mandatory rather than "authorized"; end GDC's gatekeeping of outside investigators under SOP 208.03; and act on the 2010 state audit's unanswered recommendations on medical-examiner training and suicide-referral compliance.

3. End the extraction economy. Ban commission-based telecom contracts and move toward free communication as five states have done; impose upfront fee disclosure and junk-fee prohibitions on the federal Families Over Fees Act template; bring Inmate Welfare Funds under appropriation oversight; eliminate medical copays; repeal the § 42-5-59 wage-deduction mandate and pay for prison labor; and repeal pay-to-stay as California and Illinois have.

4. Shrink the warehouse deliberately. Expand earned-time and parole eligibility now foreclosed by the serious-violent-felony exclusion and the 30-year consideration timeline added in 2006; enact meaningful compassionate release for a population with 5,753 people over 60 whose recidivism risk is less than half that of younger cohorts; and pursue penalty reform on the Proposition 47 model rather than new beds — including the 263 additional private-prison beds in the proposed FY2027 budget.

5. Fund what measurably works. Shift spending from surveillance to correctional education (43 percent lower recidivism odds, $4–5 saved per dollar), cognitive-behavioral programming (25 percent reduction), therapeutic-community drug treatment, and peer-led models like the $150,000 Autry pilot — the only reentry line in the amended FY2026 budget — and adopt solitary-reduction reforms on the Oregon and North Dakota evidence.

6. Litigate the record. Build toward CRIPA enforcement of DOJ's 82 remedial measures and, on the California blueprint, the documented record of failed less-intrusive remedies the PLRA requires; and bring test cases under the Abuse Provision, Georgia's unused constitutional backstop.

7. Close the accountability loops. Commission a retrospective audit of Howard-era forensic casework on the North Carolina Swecker/Wolf model, which reviewed 15,419 files and identified 230 problematic cases; enact a junk-science writ on Texas's Article 11.073 template; and extend victim-compensation and victim-rights statutes to people harmed in custody.

Sources

This analysis draws on GPS's published research compilations — including its policy research on Brown v. Plata and court-ordered population reduction, correctional oversight models, the GDC grievance system and PLRA exhaustion, Georgia's death-investigation structure, family incarceration costs, PREA compliance, the Abuse Provision, forensic accountability, rehabilitation evidence, and victim-status law — alongside the underlying primary sources those compilations document: the U.S. Department of Justice's October 2024 findings report, GDC budget and population data from the Governor's Budget Reports and GDC statistical publications, federal and Georgia court decisions, GAO and Bureau of Justice Statistics records, state statutes, and GPS's own mortality database. Inmate and family accounts received by GPS inform the framing where noted.