HELP US END MASS INCARCERATION The Prison Policy Initiative uses research, advocacy, and organizing to dismantle mass incarceration. We’ve been in this movement for 20 years, thanks to individual donors like you.

Can you help us sustain this work?

Thank you,
Peter Wagner, Executive Director

The Bureau of Prisons is considering a heartless, ineffective policy with far-reaching effects.

by Wanda Bertram, July 29, 2021

The Prison Policy Initiative has campaigned for years to protect incarcerated people’s letters from home, because letters are often their only lifeline to loved ones and the outside world. We’ve been concerned for some time that private companies — which already make it costly for incarcerated people to make phone calls and video calls — would someday partner up with prisons to block people from sending physical mail, too.

Now, our fears are coming true: The telecom company Smart Communications is trying to sell prisons its “MailGuard” service — where the company scans incarcerated people’s letters from home and gives them printed or digital copies instead — and the federal Bureau of Prisons just piloted the service. The Bureau of Prisons isn’t the first to try MailGuard — some jails have been using it for years, and Pennsylvania has been using MailGuard in its prisons since 2018 — but if the federal system starts replacing letters from home with scanned copies, more states will follow.

Indeed, in May, The Orlando Sentinel reported that the Florida Department of Corrections is planning not only to convert incoming mail to scans, but to start charging incarcerated people to access their own mail. The precedent set by the Bureau of Prisons is a dangerous one.

We’ve joined Just Detention International (JDI) and over 40 other civil rights organizations to demand that the Biden Administration stop experimenting with this heartless technology. The MailGuard program is unjust, as JDI’s open letter to Attorney General Merrick Garland explains, and not only because it robs people of the solace of holding a letter from a loved one:

  • When the federal prison system bans mail, it’s endorsing cruelty as a response to a health issue: Its policy implies that punishing all incarcerated people and their families forever is an appropriate response to dangerous items occasionally coming in through the mail. We’ve already seen prisons and jails use this rationale to justify other cruel policies, like banning in-person visits, and it’s an argument that needs to be stopped rather than encouraged.
  • The program is helping a private company get rich, despite President Biden’s campaign pledge to get profiteers out of the criminal justice system. “While scans of letters may be provided for free,” Just Detention International’s letter explains, “the clear intention of Smart Communications’ program is to push incarcerated people toward exorbitantly priced paid services like email or phone calls” — services that Smart Communications also sells.
  • Scanned mail can be very hard to read for people with a visual disability, who will suffer if federal prisons implement MailGuard. People in prison are three times more likely to have a visual impairment.

We’re thrilled to see Just Detention International leading the fight to stop the MailGuard program in federal prisons. And as more advocates get involved in trying to stop this horrible experiment, they may find these other resources helpful:

  • Mia Armstrong’s visually-striking 2018 article in Slate, revealing that when MailGuard scanned a person’s letters and photographs, the scans were often of such poor quality that the recipient couldn’t make out details like their loved ones’ faces.
  • Our previous research on Smart Communications. Before landing its first contract with a state prison system to scan postal mail (in Pennsylvania), Smart Communications mainly contracted with local jails, selling a variety of telecom services often bundled into one deal. These bundled contracts were riddled with exploitative clauses. For example, Smart Communications has offered jails “100% phone commissions” — in other words, offered to charge families high prices to make phone calls and return 100% of the revenue to the jail itself. (Of course, only by bundling phone calls into a deal with other pricey telecom services for families could Smart Communications make this offer profitable to itself.)
  • Our two reports on jails banning letters from home. We’ve been concerned about letter bans ever since some jails began implementing restrictive policies. Our reports Return to Sender and Protecting Written Family Communications in Jails contain some helpful evidence in favor of preserving mail behind bars:
    • Statements from corrections agencies like the American Correctional Association highlighting the benefits of written communication, and even this telling quote from the Los Angeles County Sheriff: “We believe the mail coming to inmates is as important as their phone calls. If we were to limit the mail, we believe we would see a rise in mental challenges, maybe even violence.”
    • A handy state by state chart, showing that many state policies already advise that incarcerated people should have virtually unlimited contact with their families through the mail.

As Just Detention International concludes in the open letter we signed: “Banning physical mail harms the well-being of incarcerated people, while offering no meaningful benefits.” During the COVID-19 pandemic, with prison visits suspended, families with loved ones locked up have had to work even harder to maintain crucial family bonds. The federal government shouldn’t be partnering with a private company to strain these families even further.

We suggest a few questions to ask to assess the strength of your state's public defense system.

by Ginger Jackson-Gleich and Wanda Bertram, July 27, 2021

One of the many reasons mass incarceration persists is because people too poor to afford their own lawyers are denied meaningful representation in court. This injustice happens because public defense systems — the systems tasked with providing attorneys to those in need — are severely underfunded and overburdened.

Public defender offices, assigned counsel, and contract counsel

Three models, similar challenges.

As the Brennan Center explains: In the United States, “there are three main forms of indigent defense delivery: public defender offices, assigned counsel, and contract counsel.” In some states, local governments establish and staff public defender offices. In others, courts assign private attorneys to represent people (who could not otherwise afford an attorney) on a case-by-case basis. In places utilizing contract counsel, the court assigns some or all cases to private attorneys on a flat-fee basis. The questions in this briefing are — by and large — relevant to all three models.

While every state and local public defense system is unique, we’ve identified nine urgent and common problems that plague public defense systems nationwide. Unfortunately, there isn’t enough current data for us to explain how every state stacks up on these issues, but we’ve done the next best thing: We’ve created a list of nine questions you can ask to assess where your state’s public defense system might need help, and we’ve highlighted helpful and detailed resources that can assist reform efforts.

Use these questions as a guide to learn whether your state public defense system is facing common problems and what can help to solve them:

  1. Does an independent agency oversee public defense in your state? As the Sixth Amendment Center has explained, it is critical to have a statewide agency responsible for both “the oversight and [the] evaluation of defender services.” Such agencies must have authority to create and enforce standards for effective representation (such as those related to workloads, supervision, and performance). Alongside these supervisory duties, oversight agencies that have meaningful independence1 can also insulate public defense systems from the whims of political pressure or judicial interference, which enhances the long-term stability of public defense provision.

    Notwithstanding the importance of such agencies, as recently as 2017, 16 states lacked any such entity, and more than half the agencies that did exist lacked independence.

  2. Does your state shoulder the cost of public defense or leave the burden to local governments? As the Brennan Center notes, public defense “has always been and remains an unfunded federal mandate.” Where states don’t assume financial responsibility, local governments are left to cover the cost.2 As recently as a few years ago, 11 states provided “minimal to no state funding” for public defense, leaving their local governments to shoulder the entire financial burden. By contrast, 27 states relieve local governments of “all responsibility for funding right-to-counsel services.”

    As the Sixth Amendment Center succinctly explains: “State funding of indigent defense services has proven to be the most stable for two principal reasons. First, local governments have significant revenue-raising restrictions placed on them by the state while generally being statutorily prohibited from deficit spending. Second, the jurisdictions that are often most in need of indigent defense services are the ones that are least likely to be able to afford it.”3

  3. Does your state increase public defense funding when demands on public defenders increase? Just as important as whether public defense funding is adequate at a moment in time is whether that funding grows when other parts of the criminal legal system are expanded. For example, is funding for police-worn body cameras accompanied by funding to help public defenders review collected camera footage? If new conduct is criminalized, is there an increase in public defense funding to deal with the increased criminal caseload? As the American Bar Association notes: “No part of the justice system should be expanded or the workload increased without consideration of the impact that expansion will have on the balance and on the other components of the justice system.”4

  4. Does the way attorneys are compensated create perverse incentives? In places without public defender offices, counsel is provided in several ways. Sometimes, courts appoint counsel on a case-by-case basis. As the Brennan Center has explained, appointed counsel are frequently compensated at very low hourly rates, and total compensation is often limited by fee caps that restrict how much attorneys can earn per case. Similarly, courts may hire private attorneys to “represent an unlimited number of clients for a set fee.” Under either model, such arrangements “incentivize attorneys to do as little work as possible on each case…because all costs for a case, such as investigation or consulting expert witnesses, come out of the same fee and thus directly eat away at whatever profit the attorney makes.” Accordingly, many legal groups (including the ABA) recommend banning flat-fee contracts and restrictive fee caps–nonetheless, these approaches remain the most common method of providing and funding public defense across the country.

  5. Do fines and fees fund public defense? Some places (like New Orleans) cover the cost of operating the criminal legal system through the collection of fines and fees. Public defenders, prison and jail expenses, probation and parole services, GPS monitors, and court processes can all be funded in this way. However, this “user-funded” approach is both cruel and unstable. Many people saddled with fines and fees are unable to afford them, and being forced to pay can push people deeper into debt, subject people to predatory debt collection practices, and even result in incarceration. Moreover, given widespread inability to pay, revenue from fines and fees is an unstable source of funding for public defense offices.

    Some places have started to abolish such fines and fees and to commit to more reliable funding for public defense. For example, California recently abolished many criminal legal fees, including the fee charged for representation by a public defender.5

  6. Are there meaningful, enforceable workload limits for public defenders? One of the most widespread and pernicious problems facing public defenders is the burden of crushing caseloads. To address such problems, states must adopt realistic and enforceable workload standards. However, few jurisdictions have adopted any such limits.

    Importantly, it is also critical to note that some of the most widely promulgated national standards weren’t actually “based on any empirical research and were set too high” initially, so it is critical that states adopt standards tailored to the reality of criminal practice in their own jurisdictions. As the American Bar Association has explained: “National caseload standards should [never] be exceeded, but the concept of workload (i.e., caseload adjusted by factors such as case complexity, support services, and an attorney’s nonrepresentational duties) is a more accurate measurement.”

  7. Does the state give public defenders access to the information they need to do their jobs effectively? While prosecutors are required to turn over exculpatory evidence to the defense, violations of this rule are not uncommon. In particular, defense attorneys often find it difficult to obtain information about prior police or prosecutor misconduct that may be relevant to their cases. This reality can mean that law enforcement officials with a history of abusing their power and prosecutors who shirk their ethical obligations can continue to do so. However, states can remedy this problem by making information about police misconduct and prosecutorial misconduct available to defense counsel or to the public at large. For example, California recently passed a law intended to make police misconduct records publicly available. And states like Ohio have passed open-file discovery laws, which require the prosecution to share everything in their file with the defense — a practice that can lead to “a more efficient criminal justice system that better protects against wrongful imprisonment and renders more reliable convictions.”

  8. Is counsel appointed before a client’s very first court appearance, or is appointment delayed until later in the process? Increasingly, it is clear that justice is better served when defense counsel is involved in the earliest stages of a criminal proceeding. Ideally, this will mean meeting a client prior to arraignment (the stage at which charges are formally announced) and representing them at that hearing and/or at any prior appearance that might occur. As one project in California demonstrated, meeting clients within hours of an arrest can increase the chance that someone will be released on bail, which means that someone won’t remain incarcerated while their case makes its way through the system. Another California pilot program found that, where public defenders represented clients at arraignment, defendants spent less time in custody, the county benefited financially, and due process was better safeguarded. 6

  9. Does your public defense system provide holistic defense for clients who need other legal (or even non-legal) assistance? For many people summoned to criminal court, the charges pending against them are only one of numerous challenges they face. Collateral consequences like deportation or losing access to social services may flow from a conviction, and may even result from merely facing criminal charges or being detained. Furthermore, issues like a pending eviction, a denial of public benefits, or a need for counseling may also predate the filing of any charges.

    Exemplary public defense systems across the country are finding ways to help their clients and advocate for their rights even beyond representing them in criminal proceedings. For example, in Alameda County, the public defender’s office launched an immigration representation project through which the office now represents noncitizen clients in subsequent removal proceedings and helps them obtain lawful permanent resident status in the United States. And the Bronx Defenders Office commits “to addressing clients’ most pressing legal and social support needs” and pledges to begin representation by “identifying [and supporting] the full range of client needs.” Likewise, some public defender offices have started to embed social work teams within their departments. As one study found, the involvement of social workers has a positive impact on “sentencing, attorney-client relationships, defendant experiences with the court systems, and connecting clients to community services, as well as building court actors’ knowledge of services available in their own community.”


Even an excellent public defense system in every state would not, on its own, end mass incarceration, but ensuring that every person accused of a crime has satisfactory assistance of counsel would certainly help. As many others have noted before us, the constitution’s promise that every criminal defendant has the right to legal counsel has never been a reality in this country.7

Today, at least 4.9 million people are arrested annually, most of them poor, and virtually every public defense system struggles to represent all of the defendants who can’t afford their own lawyer. Until states remove the many barriers to providing adequate public defense, this country will continue to be one where due process and equal protection are imaginary — a place where people are told to believe in a constitutional right that does not actually exist.



  1. To be “meaningfully independent,” an oversight agency might (for example) have members appointed by diverse authorities, such that the agency is not controlled by any one branch of government.  ↩

  2. The cost of a public defense system includes not only compensation for attorneys, but also funding for “technology, facilities, legal research, support staff, paralegals, investigators, and access to forensic services and experts.”  ↩

  3. A report from Gideon at 50 notes: “Without adequate state financial support, counties must rely on property tax generated income to support public defense services. In localities with poor economic bases, the counties are hard-pressed to provide adequate services (and are more likely to have a higher percentage of people qualifying for such services).”  ↩

  4. The ABA and other organizations (as well as vocal public defense leaders) have recommended that jurisdictions ensure funding parity between public defenders’ offices and district attorneys’ offices. While pay or funding parity with district attorneys’ offices does not, on its own, ensure adequate funding for public defense, it is certainly a meaningful place to begin.  ↩

  5. As recently as 2014, 43 states were charging fees for representation by a public defender — representation that, by law, is available only to people who are indigent.  ↩

  6. For a useful discussion of the benefits of having “counsel at first appearance,” see this report by the National Legal Aid and Defender Association and the Safety and Justice Program.  ↩

  7. The constitution requires states to ensure that people facing criminal charges are provided with adequate counsel. See Gideon v. Wainwright, 372 U.S. 335, 344-45 (1963). Despite this constitutional requirement, many people facing criminal charges still lack adequate access to counsel. Along these lines, Professor Paul Butler notes that, at its worst, Gideon “has not improved the situation of most poor people, and in some ways has worsened their plight” because it “provides a degree of legitimacy for the status quo.”  ↩

by Emily Widra, July 12, 2021

Locking up the most medically vulnerable people in our society has created a public health crisis not just inside prison walls,1 but in the outside community and across the country: The health of individuals, families, and entire communities is clearly associated with incarceration.

That’s according to a recent study from a team of researchers from the University of California – Los Angeles, University of Massachusetts – Amherst, Yale School of Medicine, University of Cincinnati College of Medicine, Duke University, and Washington University in St. Louis, which reveals how strongly incarceration is associated with family member health and well-being, as well as with racial disparities in health and mortality.

In the study, Exposure to family member incarceration and adult well-being in the United States, researchers found that people who have an incarcerated or formerly incarcerated family member2 consistently rate their health and well-being lower than those without a family history of incarceration, and have an estimated 2.6 years shorter life expectancy than those with no incarcerated family members, even when adjusted for demographic characteristics like race, household income, gender, and age.

This groundbreaking study was based on a nationally representative sample3 of individuals who rated their own physical health, mental health, social well-being, and spiritual well-being on a scale from zero to ten. Their responses were then scored to categorize the respondents as “thriving,” “surviving,” or “suffering” in each area, as well as in a comprehensive and holistic category of “overall well-being.” Based on the health differences reported by people with and without incarcerated family members, the researchers then estimated the changes in life expectancy associated with family member incarceration.

How exactly did researchers measure well-being and estimate life expectancy?

Some of the tools used in this study are specific to the public health world, so we’ve done our best to outline their purpose and explain how they are used:

  • The Cantril Self-Anchoring Striving Scale asks respondents to imagine a ladder with steps numbered from zero at the bottom (indicating the worst possible life for the respondent) to 10 at the top (indicating the best possible life for the respondent) and to identify where on the ladder they currently identify themselves and where they think they will be on the latter in five years. This is a widely used and validated tool for measuring population health and well-being as a part of the 100 Million Healthier Lives Adult Well-Being Assessment.
  • To calculate the changes in life expectancy reflected by the results of the self-reported health measures in the Cantril scale, the researchers used the validated Life Evaluation Index. For a subset of the population (in this case, people with incarcerated family members), the Life Evaluation Index can be calculated by subtracting the percentage of respondents who were classified as “suffering” from the Cantril Scale from the percentage of respondents who were identified as “thriving” from the scale and multiplying that result by 100. An increase of 1 standard deviation in the Life Evaluation Index reflects a 1.54-year longer life expectancy at the population level.6

The researchers broke down their results by demographic categories, allowing us to see how the impact of family incarceration on well-being varies by age, gender, race and ethnicity, education level, household income, housing type, marital status, family size, and the respondent’s own history of incarceration or substance use disorders. The results were further broken down to how the effects of family incarceration differ depending on whether it’s an immediate family member4 or extended family member5 that was incarcerated, on the length of their incarceration, and on the number of incarcerated family members (the impacts were greatest for those with multiple incarcerated immediate family members).

In this briefing, we discuss two types of analyses conducted by the researchers. First, the researchers analyzed the demographic differences between people with incarcerated family members and people without incarcerated family members, giving a clearer picture of just who is most likely to experience family member incarceration. Their analysis shows stark racial disparities in family incarceration, as well as differences in physical and mental health between those who have incarcerated family members and those who do not.

Second, the researchers isolated the impact of incarceration on life expectancy, by presenting a statistical analysis of their findings that accounted for differences in age, gender, race and ethnicity, education level, household income, housing type, marital status, family size, and the respondent’s own history of incarceration or substance use.

People with incarcerated family members have poorer health outcomes

We already know that having a family member incarcerated can cause financial instability, reduce social support, and increase the burden on caregivers. Studies have also shown that women with incarcerated partners experience depression, hypertension (high blood pressure), and diabetes at higher rates than those with non-incarcerated partners, and children with incarcerated parents are at increased risk for mental health problems and substance use disorders, which have long-term impacts on health.

This new study, however, is the first to look at how the incarceration of a family member is associated with overall well-being, using a comprehensive, holistic measure of well-being (see sidebar for more details on measuring well-being). And the findings show that people with incarcerated immediate family members – such as a parent, partner, child, or sibling – are less likely to report good physical and mental health and less likely to rate as “thriving” in terms of overall well-being, all else being equal.7 A “thriving” score overall or in any subcategory (like physical health or mental health) is significant, because it is strongly correlated with fewer health problems, fewer sick days, fewer negative emotions (worry, stress, sadness, anger) and more positive emotions (enjoyment, interest, happiness, respect).8

Bar graph showing that of those with more incarcerated family members, report poorer physical and mental health People with three or more incarcerated immediate family members are less likely to report “thriving” mental and physical health as compared to people with fewer incarcerated family members and those without any incarcerated family members. Those who do not meet the criteria for “thriving” scores are more likely to report difficulty accessing basic food and shelter, more likely to report health concerns and physical pain, and have less access to health insurance and care. These findings do not adjust for demographic characteristics like race, gender, or age.

A demographic analysis of people with incarcerated family members

This study provides further evidence of the strong overlap between overcriminalized communities and people experiencing disparities in health care. The study found that individuals with any immediate or extended family member incarcerated were more likely to be Black, live in a low-income household, have a history of a substance use disorder, and have a personal history of incarceration.

That Black families are disproportionately likely to have had a family member locked up is unsurprising, given the well-documented racism in the criminal justice system and the rampant racial disparities in prisons and jails. Compared to white respondents, Black respondents were more than twice as likely to have multiple family members incarcerated and over eight times more likely to have had an immediate family member incarcerated for at least a decade.9

Bar graph showing that of those with incarcerated family members, most were black In the 2018 Family History of Incarceration Survey, Black people were overrepresented in the number of people who reported having an incarcerated immediate or extended family member, having a family member incarcerated for over 10 years, and having 3 or more incarcerated family members. Black people make up 40% of the incarcerated population, but only 13% of the general United States population and Black families, regardless of individual involvement with the criminal-legal system, lose their loved ones to prisons and jails at higher rates than their white counterparts. These findings do not adjust for demographic characteristics like gender, income, or age.

Incarceration reduces life expectancy of family members, not just those locked up

A 2016 study from Professor Christopher Wildeman found that the sheer magnitude of mass incarceration in the United States has shortened the overall U.S. life expectancy by 2 years, and that each year in prison reduces an individual’s life expectancy by about 2 years.

This new study from 2021 shows that not only does incarceration impact the population-level life expectancy and the life expectancy of individuals behind bars, but it also impacts families and communities. People with one immediate family member who has ever been incarcerated lose 2.6 years of life expectancy, and people with more than 3 immediate family members with incarceration histories lose almost two times as many years (4.6 years) off their life expectancies, regardless of age, race/ethnicity, gender, income level, housing type, family size, or personal history of incarceration or substance use.

This study is the first to identify the extent to which incarceration is associated with the overall well-being and lifespan of an incarcerated person’s children, spouses, parents, and grandparents. Nationally, an estimated 45% of people have an immediate family member who has been incarcerated and 35% of people have an extended family member with a history of incarceration–suggesting that mass incarceration harms the health and well being of tens of millions of Americans, regardless of their personal involvement in the justice system. The study’s authors conclude that decarceration efforts could have wide-ranging outcomes in terms of national health and life expectancy. Their findings confirm what people with incarcerated loved ones already know: far from keeping people safe, prisons and jails are public health disasters, and closing them is imperative to public health.



  1. People incarcerated in prisons and jails are more likely to have health conditions that shorten their life expectancy, like lung disease, heart conditions, diabetes, renal or liver disease, and other immunocompromising conditions.  ↩

  2. For the purposes of discussing this study, “an incarcerated family member” refers to an immediate or extended family member that was incarcerated at any time. These data do not differentiate between currently or formerly incarcerated.  ↩

  3. The sample used in this study was a cross-sectional, nationally representative sample of respondents from the 2018 Family History of Incarceration Survey. 2,815 individual respondents were included in the analysis of this study.  ↩

  4. Immediate family members are defined as parents, partners, siblings, or children.  ↩

  5. Extended family members are defined as grandparents, aunts or uncles, nieces or nephews, cousins, parents- or siblings-in-law, or godparents.  ↩

  6. For more details about the Cantril Scale and the Life Evaluation Index, see Arora, A., et al., 2016.  ↩

  7. The researchers note that the association between immediate family member incarceration and “thriving” remained statistically significant after adjusting for age, gender, race/ethnicity, education level, household income, housing type, marital status, family size, and the individual respondents’ history of incarceration or substance use disorders. The association between extended family member incarceration and “thriving” was no longer statistically significant after adjusting for respondents’ own incarceration history and immediate family incarceration history.  ↩

  8. For more details about the scoring system for the Cantril Scale, see Understanding How Gallup Uses the Cantril Scale.  ↩

  9. These findings discussed here are not adjusted to account for age, gender, ethnicity, education level, household income, housing type, marital status, family size, and the respondent’s own history of incarceration or substance use disorders. Rather, these findings are a comparison between two groups of respondents: the percentage of white respondents with incarcerated family members and the percent of Black respondents with incarcerated family members.  ↩

How law enforcement and jail architects almost duped taxpayers into approving a new jail far bigger than the county needs, by offering biased analysis and misleading arguments.

by Wendy Sawyer, July 6, 2021

We’ve called on counties before to ask the right questions before deciding to build a costly new jail, because history has shown that expanded jails are quickly filled with people who wouldn’t have been jailed before, with serious personal and community consequences. A recent jail expansion proposal in Otsego County, Michigan now shows us what happens when a county either doesn’t ask those key questions about how its jail is used – or doesn’t offer the public good answers.

Correctional, law enforcement, and court officials in Otsego County, a northern Michigan county with a population of about 25,000, have been pushing to expand their 34-bed jail for fifteen years. They’ve solicited at least three studies to evaluate local correctional systems and recommend solutions to reduce jail overcrowding. So why aren’t voters convinced of the need for a $30 million, 120-170 bed jail? We looked at the arguments offered to the public and quite simply, the arguments aren’t convincing to anyone who doesn’t build new jails or lock people up for a living.


The worst jail assessment we’ve ever seen

In its most recent attempt to justify a much-bigger jail, the county enlisted an architectural firm to conduct a “feasibility study.” The architects were not criminal justice experts – and it shows.1 The 13 pages of the report dedicated to “analysis” rationalizing the need for a bigger jail are heavy on charts and light on analysis; the 30 pages of architectural and construction plans that follow show where the authors’ expertise and priorities lie. The “Analytics” section of the report includes many charts and arguments that nevertheless fail to advance the case, including:


Sample image from the architects’ report. Why does the scale of this graph appear to change in 2019, so that 24,985 appears lower than 24,665? And why does the chart show the population at 10-year intervals except for 2018 and 2019? What is the community supposed to take away from this inscrutable chart?

  • Four charts on county population projections through 2049, ostensibly meant to tie the need for a 253% increase in jail space to the county’s current 0.11% annual growth rate (and some of these charts are nonsensical, as shown in the screenshot here);
  • A summary of annual court caseloads from 2008-2018, which shows a trend of decreasing caseloads over time, and also concludes “no apparent relationship to population growth” and “projection modeling does not indicate probability/necessity for future court [sic] within the study milestone periods”;
  • A summary of jail bookings, average daily populations, and average length of stay from 2009-2012 and 2014-2018, showing that the jail typically operated at between 100-115% capacity before 2013, but has operated at 70-88% capacity for four out of five years since;23
  • A confusing array of jail population “projection models” based on average daily populations from different years of historical data: The model that includes all nine years of past data predicts an average daily population of 29 people or less through 2049; the model based on only the three most recent years of data projects a maximum average daily population of 66 people in the jail in 2049. There is no discussion of why the county should base its plans on one model over any other, and you can probably guess what projection the authors base their recommendation on (yes, the biggest one).
  • Scant attention to the composition of the jail population, making it hard to see how the jail is currently used: There is one chart about the gender breakdown of the daily population. There is no analysis of the population by offense type, sentence length, classification levels, or conviction status. The closest the report gets to evidence-based analysis of how the jail is currently used is one chart showing that, of all the people booked into the jail between 2009-2018, about half were sentenced and half were unsentenced.

The most damning sentence in the architects’ report (our highlighting added) calls the jail population data “unreliable,” even though that data makes up most of their own Analytics section.

The most telling parts of the consultants’ report are its conclusions that the jail population data are “unreliable” and “invalid,” even though those data make up most of the Analytics section. If the data are really “unreliable” or “invalid,” why did the consultants include them and make so many charts about them? Was it simply to give their recommendations the appearance of impartiality and rationality? Instead of basing their recommendations on the data available and presented to the public, the final recommendation for the new jail size is based largely on “interviews with justice and law enforcement personnel” and the “experience of the study team.” And, of course, both groups – justice and law enforcement personnel and the team of architects and engineers – stand to benefit from building a bigger jail.

Law enforcement and judicial officials complained to the consultants that, with the jail often near maximum capacity, they “have one hand tied behind their back,” and have been forced to come up with alternative sanctions instead of jailing people for all of the offenses they’d like to. Ironically, many of the alternatives officials complained about being forced to use are widely considered “best practices” today:


Local officials often dodge the question asked by community leaders: “If we let you build a bigger jail, will you just fill it?”

In Otsego County, instead of expanding upon the alternative sanctions already in place, law enforcement and judicial officials estimated that if they had a bigger jail, they “could now have 90-100” people in jail instead of 30-40. And just who are these residents that they “can’t” lock up now, but would, if only they had a bigger jail? According to the feasibility study, these include 45 “potential probation violators” and 20 people who would – with the current system – be sentenced to the alternative work release program. Their intention is clear: if taxpayers let them build a bigger jail, they can and will fill it.

an analysis shows the expansion would lock up individuals otherwise sentenced to alternative programs

Community members have received mixed messages from local officials in response to their concerns that the new jail will immediately be filled once police have a place to book more people and judges have a place to detain and incarcerate them. In a local newspaper series about the need for a new jail, the county jail administrator said that’s “not necessarily true”: “We have great judges and I don’t see them being overzealous and automatically going to the jail commitment.” But in the same article, one of those local judges commented:

“What it comes down to is it hampers my ability to do my job, which is to protect our community by creating this deterrent effect and this punishment. If I’ve got to give this person work camp or probation, I’m not accomplishing that goal because it doesn’t have the same effect as sitting in jail for 30 or 60 days.”

It’s true that sitting in jail has a different effect than sentences that allow people to remain in the community, such as probation – but the evidence is clear that the costs of locking people up are much greater than the negligible (if any) deterrent effect of incarceration:


Another bad-faith argument from jail proponents: “But we have outstanding warrants!”

In addition to the seriously flawed “feasibility study,” Otsego County law enforcement also point to “over 1,100 outstanding warrants” as evidence that they need a bigger jail. However, a closer look at those warrant details shows that only a fraction are clearly related to public safety. Additionally, many of the underlying offenses are no longer criminal or jailable under Michigan’s recent jail reforms – meaning that even with a bigger jail, very few are likely to actually lead to jail time as our closer look at the warrant list reveals:

”few Our analysis of the county’s backlog of warrants. We took a close look at the list of outstanding warrants jail proponents point to as evidence that more jail space is needed. In our analysis, we paid particular attention to how old the warrants are, whether or not they are “bench warrants” meant to bring someone back to clear up court matters, whether or not the underlying offenses would likely lead to jail time under Michigan’s recent jail reforms, and whether the offenses are related to public safety. The center image shows just a sample page of the 36-page “warrant details” document, and the highlighted lines of text are illustrative examples of our overall findings, explained in the surrounding text boxes.

Proponents of the Otsego jail expansion rely on some thin and out-of-date arguments to make their case. For example, they point to a summary of drug cases initiated by a regional undercover drug task force that includes Otsego County, which shows that the task force initiated an increasing number of cases in Otsego County from 2016-2019, with a high of 56 cases in 2019. That summary does not offer a detailed breakdown of these cases, but a local news report shows that in 2018, the task force’s work – across all seven counties – most typically involved prescription opiates and marijuana, which accounted for 44% of all cases that year.4 Now that Michigan has legalized recreational marijuana use, possession, and licensed sale, the county can expect few, if any, new marijuana arrests. And of course, in 2021, no one would argue that jail is any kind of solution for the prescription opioid problem.


How the county’s jail assessment went wrong and how other counties can get it right

What a national jail planning expert has to say

We asked David Bennett, one of the authors of the Jail Capacity Planning Guide, to review Otsego County’s recent “feasibility study,” and he, too, concluded that it fell far short of a comprehensive assessment. He writes:

We asked David Bennett, one of the authors of the Jail Capacity Planning Guide, to review Otsego County’s recent “feasibility study,” and he, too, concluded that it fell far short of a comprehensive assessment. He writes:

Jail planning can no longer use forecasting exercises based on current practices of a criminal justice system. Using this process assumes that the criminal justice system is operating as effectively as possible and that no changes will occur in the future. Both of these assumptions are false.

The system procedures needs to be analyzed and the analytics prepared to measure system effectiveness. New and innovative additions need to be considered to address the various populations in a county jail—pretrial inmates, post trial prisoners, and those in jail on holds. The mentally ill and substance abusers need unique programs designed for their specific needs and to minimize the threat of their returning to the system. Front-end diversion programs need to be implemented. A Pre-Trial Services program is critical to assessing front end decision making. Specific programs for the sentenced population to ensure coordination with out of custody programs is necessary to prevent released inmates returning to custody.

The efficiency with which the system processes cases has a lot to do with the number of jail beds needed. If changes can be made that reduce the length of time it takes to resolve cases fewer beds will be required and better outcomes will occur without impacting public safety.

All of the above needs to be included in the development of a criminal justice system master plan that includes that need for the number of jail beds. Relying upon incarceration rates to determine the number of jail beds needed for the future will guarantee that a new jail will fill up and simply exacerbate the problems of jail overcrowding.”

The National Institute of Corrections – part of the U.S. Department of Justice – has a Jails Division for the express purpose of offering technical assistance, information resources, networks and training to local jail systems – including help with jail and justice system assessment and planning. Among the agency’s numerous publicly available resources is a Jail Capacity Planning Guide, which emphasizes “a systems approach.” The authors emphasize the need to look at the ways various parts of the local justice system impact a jail’s population and how the jail is used.

A “systems approach” looks closely at all the moving parts that impact jail populations, instead of focusing on those populations alone. This makes sense: Jails are just one facet of a community’s criminal justice system, and any of those facets can be changed – not just jail size:

  • Laws determine which offenses are “jailable” and which should result in a citation or summons at most.
  • Law enforcement use their discretion to arrest, issue citations, connect people with services, or simply let people go.
  • Prosecutors make charging decisions and can offer diversion programs or decline to prosecute.
  • Judges and magistrates order people detained or released pretrial, impact court processes, and make sentencing decisions.
  • Court systems, including data and information sharing systems, affect how long a case takes to process, and therefore how long people are detained pretrial or before sentencing.
  • Finally, jail administrators can also influence who is held in the jail, under what conditions, and for how long.

Unfortunately, the county’s “jail first” approach looked at the jail population alone as evidence of whether the jail is “big enough.” A community should get a chance to understand how all of these different parts of the system are contributing to the jail situation before taking the huge and irreversible step of sinking tens of millions of taxpayer dollars into a bigger jail that will lock up more of their own.


Conclusion and recommendations: Separating what law enforcement wants from what the community needs

The good news in Otsego County is that despite the aggressive public relations campaign launched by its proponents, taxpayers were not convinced and voted down a tax increase to pay for the new jail. But the argument will surely be back, and the whole experience is a warning to other communities.

Advocates working to prevent criminal justice system expansion should anticipate that local criminal justice officials may frame their professional wants as community needs. But law enforcement have a singular perspective on community needs, a view colored by the particular anxieties, frustrations, and mandates of their job. The power to arrest and detain is also a key tool of the trade. Police and sheriffs are bound to favor solutions that enhance those powers – including more jail space to lock people up for lower-level offenses, probation violations, failure to appear at court hearings, and “quality of life” issues (which Otsego County officials seem to hint at with a mention of the “transient” population).

But most community members do not stand to benefit from a bigger jail or the increase in arrests and detention that are certain to come with it. Yet they are the people expected to pay for the multimillion-dollar project and the hundreds of thousands of dollars in additional annual operational costs.

Of course, the community does need improvements to its justice system. It needs court matters to be resolved rather than being converted to bench warrants that hang over residents’ heads, threatening to send them to jail for offenses as minor as old traffic tickets. It needs more non-jail sanctions to respond to probation violations and low-level offenses, and yes, it needs safety improvements for jail staff, incarcerated people, and their visitors. But none of these needs requires a big, new jail. The community has also made clear, through its vote, that it wants to keep jail-related costs as low as possible, now and moving forward. Ultimately, what the community needs is an unbiased, evidence-based, affordable plan to improve their justice system, starting with an open conversation about what that means to residents versus law enforcement. In all likelihood, the community isn’t looking to lock up more people, but rather for a more robust system of alternative responses and community resources.

We recommend that any county considering a new jail should first:

  • Engage the community in conversation about their public safety priorities, including a broad cross-section of community members, not just public safety and administrative officials. This conversation should consider the human costs as well as the economic costs of increasing arrests and jail detention. Any information cited by justice system officials should be provided to community members – with as much detail as possible – to analyze independently.
  • Build upon the jail alternatives the county has already implemented to keep people out of jail. For example, Otsego County already operates a drug court program and an alternative to jail, the sheriff’s “work camp” (day reporting program). At least as recently as 2013, the county also used alternative sentencing such as:
    • Community service in lieu of jail or fines and fees
    • Frequent drug testing in lieu of jail
    • Deferred or delayed sentences
    • Residential treatment
    • Maintaining employment or schooling in lieu of jail
    • “Tether” (electronic monitoring) in lieu of jail, and
    • Probation in lieu of jail

    In its public relations materials advocating for a bigger jail, the county government claims that these programs are already “at capacity,” but if the county is willing to spend taxpayer money to expand the jail’s capacity, it should consider first expanding these existing programs.

  • Change pretrial policies and practices that result in unnecessary jail detention. Pretrial policies have been driving jail growth for decades. Nationwide, about three-quarters of people held in jails for local authorities are not convicted of any crime, very often because they can’t afford the money bail amount required for release before trial. Pretrial detention is rarely necessary to ensure appearance in court and it causes lasting harm to defendants and their families. Counties across the country have implemented an array of reforms to reduce their pretrial populations without putting community safety at risk.
  • Revisit past efforts and recommendations that have been ignored but might be effective under current circumstances. In Otsego County, past jail studies and the Master Plans developed in 2008-2013 by the Criminal Justice Coordinating Committee have recommended further solutions and alternatives that were rejected or abandoned for one reason or another, but which are worth reconsidering now. These include:
    • Resurrecting a restorative justice program that was used successfully to divert prosecution in the 1970s, until the death of the key volunteer
    • Trying new restorative justice programs
    • Seeking additional funding for alternative programs, including by hiring a grant writer
    • Establishing a mental health court, a juvenile drug court, and a peer-led teen court
    • Improving court notifications
    • Educating physicians about available programs to monitor prescription drugs, and
    • Investing in crime prevention programs, such as community-based parenting programs, mental health counseling in schools, affordable after school programs, etc.
  • Implement new programs to minimize missed court appearances and the resulting bench warrants: Many counties have begun using electronic and automatic court reminder programs that defendants can opt-in to for text message or email reminders as their court dates approach. Hennepin County, Minnesota, introduced such a program in 2017, and after the first 2.5 years of the program, defendants who received reminders were 35% more likely to appear as scheduled.

    Counties can also create a bench warrant clearing process that encourages defendants who have missed court dates or outstanding fines to appear in court without fear of jail detention, and that cleans out backlogs of old warrants. “Warrant amnesty” programs operated by local courts around the country designate specific time periods during which individuals with eligible outstanding warrants can make arrangements – without arrest – to have their cases go straight to disposition by paying the original fines, participating in diversion programs, etc. The Hennepin County District Court operates a Warrant Hotline for people with outstanding warrants to schedule a hearing over the phone to take care of their case and to get more information about what options are available. In the first six months of the Hotline program, court staff fielded over 500 calls, with 85% of callers who scheduled hearings over the phone appearing in court to clear up their warrants.

  • Consult a criminal justice system expert who will evaluate the jail as part of the greater local criminal justice system, not just the historical use of the jail – or worse, simple population projections. Many examples of local jail assessments are available online that demonstrate what a “systems approach” to analysis looks like, including those from Hennepin County, Minn. (2018), Macomb County, Mich. (2016), and Monroe County, Ind. (2021), which demonstrate a range of studies in terms of depth and scope. And again, the National Institute of Corrections’ Jail Division offers local jurisdictions technical assistance – including jail planning services and information.



  1. A consultant on the Otsego study conducted by Byce & Associates, Joe Mrak, is the owner and principal of Securitecture, “an analytics company specializing in public safety buildings,” according to a 2019 article. However, a close study of Mr. Mrak’s qualifications shows that the only credentials implying any criminal justice system expertise is a CPTED certification. CPTED, which stands for Crime Prevention Through Environmental Design, is an “approach to crime prevention that uses urban and architectural design and the management of built and natural environments,” according to the International CPTED Association. For example, CPTED practitioners might recommend additional lighting or changes to landscaping in parking lots to increase visibility and thereby reduce the likelihood of assault or robbery outside of a shopping center. While CPTED certification is certainly a useful credential for architects working in the criminal justice space, CPTED is not a comprehensive approach to crime prevention and has nothing at all to do with assessing justice system processes.  ↩

  2. The average daily population has shifted in tandem with the average length of stay, declining from 2014-2017 and then jumping back up in 2018. These numbers are related because jail populations build up as individuals are forced to stay longer. In 2018, the average daily population jumped from 24 up to 39 people, at the same time that the average length of stay rose from 13 days to 39 days. The report does not explain the dramatic change in 2018.  ↩

  3. For context, in 2018, U.S. jails nationwide operated at about 81% capacity, but small jails of less than 50 people were only about 58% occupied.  ↩

  4. Data from the first quarter of 2019 (in the same article) showed an uptick in meth-related cases, but again, the data aren’t detailed enough to show whether these are cases that made a real dent in the availability of drugs or whether the task force was often targeting addicts with overly aggressive policing tactics, as residents of nearby Emmet County – covered by the same task force – have claimed and protested against.  ↩


We thank the John D. and Catherine T. MacArthur Foundation Safety and Justice Challenge for their support of our research into the use and misuse of jails in this country. We’re also grateful to the residents of Otsego County who brought their experiences and the relevant documents provided by the County to our attention.

Stay Informed

Get the latest updates:

Tweet this page Donate Now hiring:
Digital Communications Strategist