Sentencing Policies and Practices in Wisconsin

Wisconsin sentencing policy has long emphasized the discretion of the individual sentencing judge. Unlike some other American states, Wisconsin has never had presumptive sentencing guidelines. Two separate experiments with nonbinding, advisory guidelines were resisted by the judiciary and ended in failure. Nor does Wisconsin have many mandatory minimums; the few minimums on the books are narrowly targeted and infrequently applied. Likewise, Wisconsin’s appellate courts have declined to develop a meaningful “common law” of sentencing; appellate review remains highly deferential to the sentencing judge.

Traditionally, the one significant check on judicial sentencing discretion was an equally wide parole discretion. Prior to 1984, parole officials could release inmates after only a few months in prison. However, the state legislature reduced parole discretion through a series of steps in the 1980s and 1990s, culminating in the adoption of what was likely the nation’s most inflexible “truth-in-sentencing” (TIS) law in 1998. Parole was entirely eliminated for all felonies, and no other early release mechanisms were put into place—the prison term imposed by the judge would be served in full, without any exceptions. Since then, a few early release mechanisms have been established, but they are within the control of the sentencing judge. Thus, on the back end of incarceration as well as the front, Wisconsin now gives a preeminent role to the judiciary.

A period of growing judicial power over punishment coincided with an extraordinary explosion in the size of Wisconsin’s prison population. Although all states experienced such an imprisonment boom in the late 20th century, Wisconsin’s rate of growth was higher than average. To some extent, this growth was fueled by violent crime, which increased more steadily and for a longer period of time than in the nation as a whole, and by the war on drugs. However, prison terms also grew longer on average, reflecting both tougher judge-imposed sentences and the constriction and eventual elimination of early release opportunities. As the prison population stubbornly remains near historical highs, sentencing policymakers will continue to debate how length of stay can be reduced without unduly sacrificing either public safety or judicial discretion.

Here is how this essay is organized. Section I describes crime and imprisonment trends in Wisconsin in the era of mass incarceration. Section II summarizes key features of current sentencing policy and practice. Section III focuses on appellate review, discussing the Wisconsin Supreme Court’s abortive efforts to establish a common law of sentencing. Section IV reviews the history of sentencing policy in Wisconsin since 1980, including the failed guidelines experiments. Finally, Section V considers prospects for reform.

Like all American states, Wisconsin experienced dramatic growth in its prison population between the early 1970s and early 2000s. By the time the year-end population peaked in 2004, it had reached nearly 23,000, more than 11 times what it had been in 1973. Since 2004, the population has stabilized, but no clear downward trend has yet been established. As shown in Figure 1, Wisconsin has consistently maintained an imprisonment rate below the overall average for American states. However, Wisconsin’s pace of growth in this area (428 percent since 1978) has been much higher than the national average (248 percent). At the same time that Wisconsin’s imprisonment rate has moved closer to national norms, it has diverged markedly from that of neighboring Minnesota. Based on overall size, crime rate, voting patterns in presidential elections, and many other demographic variables, Wisconsin and Minnesota are very similar. Yet, Minnesota has maintained a lower imprisonment rate since the late 1950s, with the gap growing especially pronounced after 1980 (O’Hear 2013). Since then, Wisconsin’s imprisonment rate has consistently been at least twice as large as Minnesota’s, a fact that has been noted with some dismay in recent discussions of sentencing policy in Wisconsin.

Much of Wisconsin’s imprisonment increase has been associated with a large, long-term increase in the state’s rate of violent crime. As shown in Figure 2, Wisconsin experienced a mostly continuous increase in its rate of reported violent crime from 1962 to 1995; overall, the rate grew nearly 10 times larger over that period. This increase was more than twice the national average, but Wisconsin started from such a low baseline that its rate of violent crime has remained well below national norms. Minnesota also experienced a comparable increase in violent crime, suggesting the influence of broader regional forces affecting both states. Since 1995, the national rate of violent crime has fallen sharply, but Wisconsin has seen little sustained improvement. After a relatively shallow decline between 1995 and 2004, Wisconsin’s violent crime rate spiked to an all-time high in 2007 and has remained at or above mid-1990s levels since then. These trends have generally paralleled those in Minnesota.

Between 1972 and 1989, the increase in Wisconsin’s prison population closely tracked the increase in violent crime. Over that period of time, as depicted in Figure 3, the prison population grew by 234 percent, but violent arrests went up by nearly as much—213 percent. (“Violent arrests” here refers to arrests for the Federal Bureau of Investigation’s violent index crimes: nonnegligent homicide, rape, robbery, and aggravated assault.) Thus, the first half of Wisconsin’s imprisonment boom can be seen in large part as a response to a surge in violent crime. Indeed, an analysis of state corrections data shows that violent offenders accounted for about three-quarters of the imprisonment increase through 1995 (O’Hear 2013). In the 1990s, however, there was a sharp divergence between imprisonment and violent-crime trends. Between 1989 and 2004, the prison population jumped by another 238 percent. However, major violent arrests were up and down erratically over that time period. Even with a temporary bump in arrests in the early 2000s, the figure for 2004 stood only 28 percent higher than it had been in 1989. This contrast between skyrocketing imprisonment and stable arrest rates for the offenses that had been the chief drivers of imprisonment points to real increases in the severity of the criminal-justice system.

Prison population is largely a function of two variables: the number of admissions and the average length of stay per admission. Both variables contributed to continued growth in the prison population between 1989 and 2004, the second phase of the imprisonment boom. Annual admissions quadrupled over that time period (Soldner 1991, Carmichael 2005). A large share of the increased admissions came from drug offenders, who accounted for fully one-quarter of the newly sentenced inmates arriving in Wisconsin prisons in 2003–2004 (Carmichael 2005). Drug-trafficking arrests tripled between the mid-1980s and mid-2000s, which helped to maintain growth in prison admissions even as the number of serious violent offenders stabilized (Wisconsin Office of Justice Assistance 1987, 2007). Some of the effect of the war on drugs can likely be seen in the sharp increase in felony case filings between the mid-1980s and the mid-2000s, which is also depicted in Figure 3. (Unfortunately, Wisconsin’s publicly available court data are less detailed and complete over the relevant time period than the arrest and corrections data.)

Admissions were also driven by a surge in revocations from probation and parole. Revocations increase predictably as the long-term legacy of an imprisonment boom: more prisoners eventually means more former prisoners, who are prone to violate conditions of release and return to prison. (In Wisconsin, about 60 percent of revocations come from postprison supervision, as opposed to probation [Van Stelle and Goodrich 2009].) Revocation increases may also reflect greater toughness along a number of other dimensions: longer periods of probation or postprison supervision, more conditions that must be complied with, and more of a zero-tolerance approach to violations. In the 1997–1998 biennium, revocations amounted to 15 percent of admissions; over the next eight years, they quadrupled in number and grew to account for more than 40 percent of admissions (Bauer 1999, Carmichael 2007).

Admissions, however, were not the only factor driving the imprisonment boom, at least by the late 1990s and early 2000s. The average length of stay also increased, resulting in part from the curtailment and elimination of parole. Average judge-imposed sentences inched upward in the early and mid-1990s, but these trends were more than offset by a liberalization of parole, which actually caused a slight decrease in length of stay through 1997 (Welch 1998). However, a political backlash against parole resulted in a leadership change at the parole commission and the adoption of more restrictive release practices. Thus, for instance, in the mid-1990s, discretionary parole accounted for more than 80 percent of prison releases in Wisconsin, but that figure fell to just one-third in 1999–2000 (Bauer 1997, 2001). At about the same time that the parole commission was tightening up its standards, the legislature adopted a TIS law that eliminated the possibility of discretionary parole for all felonies committed on or after December 31, 1999 (O’Hear 2014). This change would inevitably result in increased lengths of stay unless the judges carefully and systematically altered their sentencing practices to account for the elimination of parole. However, the judges’ adjustment to TIS proved slow and erratic. Expected confinement time initially spiked after TIS, then temporarily returned to pre-TIS levels before resuming an upward climb (Greene and Pranis 2005; Justice Center 2009). Overall, between 2000 and 2007, judges increased the confinement portion of their sentences from an average of 31 months to 40 months (Justice Center 2009). Postprison supervision periods increased even more dramatically, from 23 months on average to 54 months, which helped to fuel the growing tide of returns to prison for violations. Total sentence length (imprisonment plus postprison supervision) for new inmates climbed from a pre-TIS average of 6.9 years to 8.7 years in 2005–2006 (Bauer 1999, Carmichael 2007). In a sense, this just continued the upward drift of judge-imposed sentences that was also apparent before TIS, but now this movement could no longer be offset through the parole process, as it had been in the 1990s.

A sharp drop in admissions finally helped to stabilize the prison population. Admissions peaked in 2006 and then fell for six consecutive years, representing an overall drop of about one-quarter (Milwaukee Community Justice Council 2015). About two-thirds of this drop resulted from reduced revocations from community supervision. (This decrease in revocations was a surprisingly localized phenomenon, with three-quarters of the decrease coming from Milwaukee County alone.) Most of the rest of the overall drop in admissions could be attributed to a 30 percent decline in admissions for drug offenses (Carmichael 2007, 2013), which paralleled a similar decline in arrests for cocaine and heroin trafficking in the same time period (O’Hear forthcoming). An escalation in the war on drugs helped to drive a continuation of the imprisonment boom throughout the 1990s; a de-escalation helped to attenuate the impact of TIS and longer average sentences in the 2000s.

Reflecting long-term increases in violent crime and other developments, Wisconsin’s prison population has changed not only in size, but also in composition. Figures 4 and 5 compare the most serious offenses of conviction of the prison population in 1970 and 2014. A dramatic shift away from property offenders and toward violent and sexual offenders is apparent. The fact that two-thirds of the current prison population falls into the violent or sexual category helps to explain the difficulty reformers have had in achieving significant reductions in the population. (Forcible rape is here counted as a sexual offense.)

Two features of the prison population have remained constant, however. First, since even before the imprisonment boom, African Americans have been dramatically overrepresented in the prison population relative to their share of the general population. In 1970, African Americans were only 3 percent of Wisconsin’s population but 30 percent of the prison population (Wisconsin Division of Corrections 1971). By 2010, African Americans had grown to 7 percent of the general population and 43 percent of the prison population (Carmichael 2011). Wisconsin’s Hispanic community has recently surpassed the African American community in overall size, but Hispanic disparities in imprisonment are negligible. One recent study found that Wisconsin has the nation’s highest black male incarceration rate but a lower-than-average incarceration rate for Hispanic men (Pawasarat and Quinn 2013).

Wisconsin’s racial disparities are related to another enduring feature of the Wisconsin prison population: the outsized representation of Milwaukee County. With close to one million residents, Milwaukee accounts for more than one-sixth of Wisconsin’s population and more than one-quarter of its prison admissions (Milwaukee Community Justice Council 2015). Since Milwaukee is home to nearly 70 percent of the state’s African American population, the county’s heavy prison utilization drives the state’s overall racial disparities. Indeed, the incidence of black male incarceration in Milwaukee has reached epidemic proportions: more than half of Milwaukee’s black males between the ages of 30 and 50 either are currently in a state prison or have previously served time in a state institution (Pawasarat and Quinn 2013). Routine imprisonment is one of an interlocking set of social and economic disadvantages that plague the intensely segregated and impoverished African American neighborhoods that radiate north and west of Milwaukee’s downtown. To a dismaying extent, state sentencing policy is something that is done by the rest of the state to these disadvantaged neighborhoods.

Criminal cases are initiated in Wisconsin’s county-based circuit courts. The locally elected district attorneys and their staffs make the charging and plea-bargaining decisions. Beyond a minimal requirement that charges be supported by probable cause, there are few formal constraints on the prosecutor’s discretion (Kalal v. Circuit Court for Dane County, 271 Wis. 2d 633 [2004]). The locally elected judges review proposed plea bargains—very few are rejected—and determine the sentences. The sentencing judge may order a presentence investigation in felony cases. An investigation is more likely to be ordered in more serious cases, but judges vary widely in their use of presentence investigations. If ordered, the presentence investigation is conducted by a probation officer employed by the state Department of Corrections.

Judges have wide discretion in choosing a sentence. At present, there are very few mandatory minimum sentences in Wisconsin. The most important is the mandatory life term for Class A felonies, a category whose chief offense is first-degree intentional homicide. (Wisconsin has not had the death penalty since 1853.) Most other crimes have been classified into one of eight other classes of felony or one of three classes of misdemeanor. Probation is permissible for all classes except Class A felonies (Wis. Stat. sec. 939.50(3)). Maximum penalties for each class are prescribed by statute. In multicount cases, judges are free to impose sentences consecutively or concurrently.

The judge who chooses to impose a term of imprisonment must set both a term of initial confinement and a period of extended supervision (ES) in the community after release from prison (sec. 973.01(2)(b)). In Wisconsin, this two-part judgment is referred to as a “bifurcated sentence.” The initial term of confinement must be at least one year, with the maximum based on the class of offense. For instance, armed robbery is a Class C felony carrying a maximum initial term of 25 years (sec. 943.32(2)), soliciting a child for prostitution is a Class D felony carrying a maximum initial term of 15 years (sec. 948.08), and residential burglary is a Class F felony carrying a maximum initial term of 7.5 years (sec. 943.10(1m)). The ES portion of the sentence must be at least 25 percent of the term of initial confinement, subject to certain statutory caps (e.g., 15 years for a Class C felony).

Although the basic statutory ranges control most cases, Wisconsin law does recognize a few “enhancers,” which raise the maximum term or, in some instances, require a minimum prison term. For instance, if a violent felony is committed in a school zone, then the maximum initial term of confinement is increased by five years (sec. 939.632(2)). Of course, this is only a maximum; the judge need not increase the sentence he or she would have otherwise imposed by any particular amount. Wisconsin law also identifies a number of statutory “aggravators,” such as the offender’s use of a disguise, which the judge “shall consider” at sentencing (sec. 973.017(3)). Again, however, there is no requirement that the judge actually give any weight to such aggravators.

Under Wisconsin’s TIS law, the full term of initial confinement must be served without the possibility of early release through parole or “good time” (credits earned for good behavior) (O’Hear 2014). However, in certain limited circumstances, the sentencing judge can establish an opportunity for early release based on participation in rehabilitative programming. For instance, the judge may declare certain nonviolent offenders eligible for “earned release”; such offenders are to be released to ES upon completion of a substance-abuse program in prison (sec. 302.05). Inmates convicted of lower-level felonies may petition the sentencing judge for early release after having served 75 percent of the initial term of confinement (sec. 973.195). Similarly, inmates convicted of midrange felonies may petition for release at the 85 percent mark. Grant rates for such petitions vary widely from county to county, with virtually no grants in Milwaukee and some of the other larger counties (Norris 2012).

The judge who chooses a sentence of probation in lieu of imprisonment may either withhold sentence or impose a stayed sentence of incarceration (sec. 973.09(1)(a)). The judge determines the length of probation, subject to statutory limitations, and may impose “any conditions which appear to be reasonable and appropriate.” Such conditions may include confinement in a county jail or elsewhere for up to one year, from which the judge may grant the offender work-release privileges (sec. 973.09(4)(a)). The judge may later extend probation, modify the conditions of probation, or discharge a probationer early (sec. 973.09(3)). Otherwise, probation is administered by the state Department of Corrections (DOC).

Revocation decisions from probation and ES are made administratively at the behest of the DOC. A revoked probationer is either returned to court for sentencing (if a sentence had initially been withheld) or subjected to whatever sentence was imposed and stayed. In cases of ES revocation, the period of reconfinement is determined administratively without any return to court (sec. 302.113(9)(am)).

Like most American states, Wisconsin lacks rigorous appellate review of sentences (O’Hear 2010). A pair of Wisconsin Supreme Court decisions have established the jurisprudential foundation for the development of a “common law” of sentencing. However, neither the Supreme Court nor the intermediate Court of Appeals has displayed any real interest in building on the foundation.

For many years, Wisconsin’s leading case was McCleary v. State, which supplied the extreme circumstances that practically demanded appellate intervention: a Milwaukee trial judge imposed a nine-year prison term on a first-time offender for forging the signature on a $50 check (49 Wis. 2d 263 [1971]). The judge said little by way of explanation beyond quoting from the presentence investigation’s bleak assessment of the defendant’s character. On appeal, the Wisconsin Supreme Court declined to hold the sentence unconstitutional on equal protection or due process grounds but nonetheless found the sentence an abuse of discretion. The Court thus established a subconstitutional, nonstatutory basis for the appellate review of sentences in Wisconsin. In so doing, the Court seemed to contemplate the eventual creation of a common law of sentencing: “The review of sentences on a consistent and rational basis will inevitably lead to the development of criteria to be used by trial judges in the sentencing process.”

From the standpoint of this objective, the Supreme Court’s mistake was to frame the trial judge’s error in procedural terms. McCleary’s sentence was an abuse of discretion because the judge had not really stated any reasoning to justify a nine-year prison term. The Supreme Court’s analysis suggested that appellate review might have little or no substantive content—as long as the sentencing judge articulated some minimally rational justification for the sentence, then the work of the appellate court was finished. Of course, this procedural interpretation of McCleary was inconsistent with the Court’s stated desire for the development of “criteria to be used by the sentencing judge,” but the Court’s interest in such criteria proved ephemeral. In practice, McCleary came to stand for an empty, formalistic requirement that the sentencing judge specifically identify some purpose to be served by the punishment (community protection, retribution, rehabilitation, or deterrence)—in effect, magic words whose incantation would insulate the sentence from further appellate examination.

Wisconsin’s appellate courts thus rested idly on the sidelines while average sentences inched upward and the prison population exploded in the 1980s and 1990s. In defense of the courts’ passivity, though, there may have seemed little point to developing an elaborate law of sentencing for the judges at a time when nearly all prisoners were granted discretionary parole release. Indeed, this had been precisely the government’s argument in McCleary: don’t worry about the absurd sentence, it will be fixed by parole.

The picture changed with TIS, which led directly to the Supreme Court’s second major statement on sentencing, Gallion v. State (2004 WI 42). A Milwaukee judge sentenced Gallion to 21 years in prison in connection with a fatal car accident. Invoking public safety and general deterrence, the judge plainly complied with the prevailing, minimalist interpretation of McCleary. Yet, Gallion argued on appeal, the new realities of TIS demanded more. Since judges now had the final say over length of imprisonment, the standards of appellate review should be tightened. And simply gesturing in the direction of safety and deterrence did not make clear why 21 years was a more appropriate sentence than 10 or 15 years.

The Wisconsin Supreme Court agreed with Gallion’s general points: Wisconsin sentencing had become “mechanical” and unduly focused on the use of “magic words,” even as the elimination of parole created a “greater need to articulate on the record the reasons for the particular sentence imposed.” The Gallion Court thus demanded that sentencing judges explain how the articulated purposes of a sentence are advanced by the sentence.

Gallion seemed to call for more rigorous appellate review, but the focus remained on process, not substance. Moreover, the Court seemed tentative and evasive when it came to explaining how judges were to connect purposes to outcomes. The Justices lamely “encourage[d] courts to refer to information provided by others,” including the recommendations of the lawyers and probation officers, data from other cases, and the recently implemented second Wisconsin sentencing guidelines. The suggestion that data be consulted was intriguing, but the Court provided no guidance about what data precisely would be relevant, how such data could be obtained, and how it should figure into the sentencing determination. None of this was terribly reassuring if one hoped Gallion actually would reenergize appellate review of sentences.

Perhaps even more disappointing was the Court’s application of the new standards to Gallion’s sentence, which was affirmed after a superficial analysis that seemed to demand no more than the articulation of recognized purposes of sentencing and the identification of a few facts that were relevant to those purposes (O’Hear 2010). The Court simply failed to meet Gallion’s challenge that virtually any amount of imprisonment might have been equally well justified by the sentencing judge’s reasoning. No data or other benchmarks were even discussed.

Gallion provoked a great deal of conversation within the judiciary and undoubtedly prompted some judges to provide more extensive explanations of their sentences. There is no evidence, however, that Gallion has led to more rigorous appellate review, and a meaningful common law of sentencing remains as elusive today as it was in the decades following McCleary. Gallion’s subtext (ambivalence) prevailed over its text.

In part, this ambivalence about appellate review results from a principled commitment by some justices to individualization and localization. As the Court said in rejecting a sentencing guidelines proposal in 1984, “While each convicted felon is an individual deserving of individual treatment at sentencing, the interests of the public, too, will vary according to the particular community in which the crime was committed, the capacity of the community to rehabilitate the criminal, and the needs of that community for protection from that type of criminal activity” (In the Matter of Judicial Administration of Felony Sentencing Guidelines, 120 Wis.2d 198 [1984]). At the same time, the Court’s unwillingness to engage with sentencing in a serious way also likely stems from the political realities facing an elected judiciary. Appellate judges who overturn a sentence as excessive face grave electoral peril in a tough-on-crime political environment. As a cautionary lesson, the justices of the Supreme Court need look no further than their former colleague Louis Butler, who was defeated in 2008 after a much-publicized television attack ad accused him of responsibility for the release of a child molester (Marley 2008).

In sum, neither the Wisconsin Supreme Court nor the Wisconsin legislature has displayed any real interest in regulating the sentencing discretion of the state’s circuit court judges. When it comes to sentencing in Wisconsin, the trial judge is king. In the serious and midrange felony cases, in particular, the judge typically has the freedom to choose any sentence from probation to a decade or more of imprisonment, subject to no effective system of review. These arrangements seem hard to square with the rule of law, but, as we will see in the next section, they have attracted little sustained attention from reformers.

Several distinct waves of reform have buffeted Wisconsin sentencing policy since 1980. Each wave can be seen as an effort to accomplish one of three distinct purposes: (1) reduction of unwarranted disparity in the sentencing of similarly situated defendants, (2) containment or reduction of the size of the state prison population and associated fiscal burdens, or (3) lengthening of prison terms, especially for violent, sexual, or drug offenders. The second two purposes—although plainly at odds with one another—have proven considerably more influential than the first. As a result of tensions among the three purposes, reforms have generally been short-lived, narrowly targeted, or largely symbolic in character. Without question, Wisconsin’s most consequential sentencing reform since 1980 was the 1998 TIS law. Most reform since 1998 has been intended, at least to some extent, to contain the impact of TIS on the state corrections budget.

In 1980, Wisconsin had a stable, long-established sentencing system that featured wide discretion for judges and equally wide discretion for the parole board; discretion at each level served as a check and balance on the other. From the judge’s perspective, the system looked much the same as it does today: statutory ranges were wide, with virtually no mandatory minimums and maximums that were quite generous (if somewhat lower than current levels). However, the existence of parole sharply distinguished the systems of then and now. In 1980, most prisoners became eligible for parole after serving just six months of their sentences. Even life-sentenced inmates were eligible after as little as eleven years and three months. Not surprisingly, research at the time indicated that the centralized parole process tended to smooth out disparities in judge-imposed sentences (Haas 1980).

Despite this safeguard, some Wisconsin judges began to express interest in sentencing guidelines in the mid-1970s (Fontaine 2005). This interest reflected both a genuine concern that sentencing disparities were harming the judiciary’s standing with the public and a desire to preempt more draconian restrictions on sentencing discretion by the legislature (O’Hear forthcoming). In 1980, Wisconsin’s Office of Court Operations obtained a federal Law Enforcement Assistance grant to fund research on guidelines. The initiative eventually resulted in the development of eight advisory guidelines that were based on existing state practices. The guidelines were data-intensive, involving the analysis of thousands of cases using 142 offense and offender variables (Fontaine 2005).

The guidelines were initially rolled out in just four counties as a pilot. In 1983, judicial supporters of the initiative petitioned the Wisconsin Supreme Court to adopt the guidelines on a statewide basis. The Court declined, calling the advisory guidelines “an unwarranted intrusion into the sentencing discretion and authority of the trial judge” (In the Matter of Implementation of Felony Sentencing Guidelines (113 Wis.2d 689 [1983]). Undaunted, proponents found support in the legislature, which decided in 1984 to adopt advisory sentencing guidelines (1983 Wis. Act 371 [1984]). The legislature offered the Supreme Court one more opportunity to manage the guidelines system, but the Court again declined. A new sentencing commission was then established, which formally extended the pilot guidelines to the rest of the state in 1985.

Wisconsin’s first commission and guidelines lasted one decade, and there is little evidence that they had much impact on sentencing practice. They were handicapped from the start by the legislature’s express proscription of appeals based on the guidelines. From the perspective of busy trial judges, there was little incentive to pay heed to the guidelines and much resentment of the additional paperwork they required. Indeed, many judges simply ignored the guidelines worksheets they were supposed to fill out for each case: in the first five years, only about 60 percent of worksheets were completed and submitted to the Commission (Wisconsin Sentencing Commission 1990). The numbers slowly improved thereafter but never approached 100 percent. Moreover, the worksheets that were submitted indicated high rates of departures from the guidelines ranges for some offenses. In 1990, for instance, judges departed in more than 70 percent of armed robbery cases. Anecdotally, practitioners from the time recall especially fierce resistance in the state’s two largest counties, Dane and Milwaukee, where it was said that savvy lawyers knew better than to invoke the guidelines in their sentencing arguments (O’Hear forthcoming).

Nor did the commission acquire much policymaking influence with the legislature. As we will see, the commission’s decade, 1985 to 1995, coincided with a wave of tough-on-crime sentencing laws. However, the commission played little role in the debates surrounding these laws. Unwilling to be caught in the middle of political controversies, the commission tended increasingly to withdraw from the policymaking process over time (Shane-DuBow 1995).

In the end, the commission and guidelines were eliminated by the legislature in 1995 as part of a broader wave of government cost-cutting (Fontaine 2005). The question remains, though, why the commission and guidelines had so little support in the legislature that they were vulnerable in a time of short-term fiscal pressure. A variety of factors likely played a role, including the perception that the commission was a “Democratic thing” at a time when Republicans increasingly dominated state government; the admitted lack of political experience and skill of the commission’s executive director; the overt hostility of the Wisconsin Supreme Court; chronically low compliance rates; and the frustratingly slow, data-intensive process of updating the guidelines (O’Hear forthcoming).

At the most fundamental level, though, the problem with the guidelines was that they were essentially a judicial creation intended to address the disparity issue, but few judges (or legislators) really cared much about that issue. The Wisconsin Supreme Court doubtlessly spoke for many judges when it invoked the ideal of highly individualized sentences tailored to fit the particular circumstances of the local community. This ideal would naturally resonate with locally elected trial-court judges, but they had little reason to care about the competing goal of statewide uniformity. Moreover, whatever interest judges might have had in addressing disparity in the late 1970s and early 1980s was swamped by get-tough pressures in later years. “Descriptive” guidelines that are intended to codify past practices are inherently vulnerable at a time when judges perceive a growing demand for longer sentences, especially when the updating of guidelines to reflect changing practice patterns is slow and infrequent. Persistently criticized as overly lenient, and lacking any apparent purpose beyond the questionable goal of statewide uniformity, the guidelines were poorly positioned to survive the increasingly punitive 1990s.

Beginning in 1986, Wisconsin adopted a series of new sentencing laws that were self-consciously framed as tough-on-crime measures. These laws did tend to increase sentence lengths, but until the adoption of TIS in 1998, the effects were rather modest. The bark of these laws was worse than their bite. The overall thrust of reforms between 1986 and 1998 seems to have been the symbolic expression of anger against criminals without significantly diminishing the discretion of judges or parole officials to show lenience.

It was initially cocaine traffickers who drew the ire of legislators. In 1986, as part of a broader national trend toward tougher drug sentences, Wisconsin ratcheted up its penalties for cocaine offenses, which had been among the nation’s most lenient (Wisconsin Cocaine Task Force 1986). Borrowing from federal models, the 1986 cocaine law increased maximum sentences, imposed mandatory minimums for higher-volume trafficking offenses, and adopted a special sentence enhancement for cocaine distribution within 1,000 feet of a school zone (O’Hear forthcoming). However, while Wisconsin copied some of the structure of federal law, the new sentencing provisions were much less restrictive of sentencing discretion. For instance, the toughest volume-based minimum was set at only one year. Moreover, while the federal school-zone provision triggered a mandatory minimum, its Wisconsin counterpart only boosted the maximum. In short, the stated desire to send a tougher deterrent message to cocaine offenders was mediated through Wisconsin’s well-established preference for wide discretion and individualization in punishment.

The 1986 cocaine bill set the stage for a series of new drug-sentencing laws over the next five years. In essence, these reforms took the basic concepts of the cocaine bill (new sentence enhancements for traffickers based on volume and distribution in a protected zone) and extended them to heroin and other drugs, while also adding to the severity of these enhancements. By the end of the tough-on-drugs wave of reforms, Wisconsin’s maximum sentences even for routine powder, crack, and heroin distribution offenses reached well beyond the decade mark, and potential penalties were even further augmented if distribution occurred in any of seven different protected zones (school, public pool, community center, and so forth). Yet, minimums remained low by federal standards. More importantly, in 1990, the legislature quietly adopted a safety valve: all drug minimums were expressly converted from “mandatory” to merely “presumptive.” Under this law, the sentencing court “may impose a sentence that is less than the presumptive minimum … if it finds that the best interests of the community will be served and the public will not be harmed” (1989 Wis. Act 121 [1990]).

The various statutory changes contributed to a surge in prison admissions for drug offenses in the 1990s, but the sentences remained relatively short, which meant that drug offenders contributed only modestly to the overall size of the prison population, peaking at about 15 percent (O’Hear 2013). Sentencing commission data from the early 1990s show that most trafficking defendants were still receiving probation, and those who were sent to prison received an average sentence of only 3.3 years (Wisconsin Sentencing Commission 1992). The average sentence was up only slightly from the 3.1-year average of the early 1980s, before any of the statutory changes (O’Hear forthcoming). Moreover, under then-prevailing rules, a prisoner with a 3.3-year sentence would have been eligible for parole in less than a year and mandatorily released at 2.2 years. The data also revealed that judges were taking advantage of the safety valve. For instance, in 17 percent of the cases subject to a presumptive minimum of three years or more, the judge imposed probation.

After 1991, the legislature shifted its attention from drugs to violent and sexual crimes. The shift was foreshadowed by the 1988 “life means life” law, whose shorthand name promised far more than was actually delivered (1987 Wis. Act 412 [1988]). Rather than eliminating parole eligibility for life-sentenced inmates (i.e., those convicted of Class A felonies), the law simply gave the sentencing judge discretion to set a later date for parole consideration, which at the time would normally have been at about the 13-year mark. This modest transfer of power from the parole commission to the sentencing judge anticipated the much larger transfer that would be accomplished a decade later by TIS.

The key event precipitating a new wave of sentencing laws was the Republican takeover of the state Senate in 1993, which ended two decades of Democratic control of both legislative chambers. A torrent of tough-on-crime legislative activity followed, including the first floor vote on restoring capital punishment in Wisconsin since 1859 (Walters 1993). Moderate Republicans ultimately joined with Democrats to reject the death-penalty proposal, but many other tough-on-crime measures passed. In the sentencing area, these enactments generally took the form of narrowly targeted sentence enhancements. Consistent with what we have seen with drug sentencing, the legislature tended to use increased maximums as much as, if not more than, new minimums. A sampling of the new maximums would include those for gang-related crimes, crimes involving terrorism, and repeat child sex crimes (1993 Wis. Act 97 [1993], 1993 Wis. Act 224 [1994]). New minimums included those for repeat serious sexual offenses, repeat serious violent offenses, and assaults in juvenile detention facilities (1993 Wis. Act 97 [1993], 1993 Wis. Act 98 [1993]).

Without a doubt, “three strikes and you’re out” was the most heavily touted sentencing reform of the pivotal 1993 legislative session (O’Hear forthcoming). Inspired by a similarly titled ballot initiative adopted in Washington state, Wisconsin’s version of three strikes introduced into state law the sentence of life without the possibility of parole (LWOP). The law identified about two dozen serious felonies that would count as “strikes”; upon conviction of a third, a defendant would face mandatory LWOP. The law’s lead Senate sponsor proclaimed that “hundreds of … violent, persistent felons live among our citizens here in Wisconsin”; three strikes, she asserted, “will begin to seal the cracks in our legal system to make sure these criminals are locked up for life no excuses, no deals” (Rosenzweig 1994). As with “life means life,” however, this tough-sounding law actually delivered much less than it promised. When first proposed, the state DOC estimated that three strikes would only apply to about nine defendants per year (St. Paul Pioneer Press 1994). In actuality, the numbers proved even much lower. For instance, one study found that only three offenders were sentenced under three strikes in the law’s first four years (Dickey and Hollenhorst 1998). By comparison, California’s much more expansively targeted version of three strikes—adopted the very same month as Wisconsin’s—had been applied to more than 40,000 defendants. The disparity makes clear that Wisconsin’s three-strikes law was in the same spirit as the life-means-life law and so many other sentencing provisions of the tough-on-crime era: a largely symbolic enactment that expressed anger at criminals without actually doing much to push longer sentences. Notably, Assembly Republicans pressed for a more practically significant version of three strikes, but the Republican governor—concerned about the size of the state’s prison population and corrections budget—worked quietly behind the scenes to ensure that the Senate’s more moderate version became law (O’Hear forthcoming).

In retrospect, the tough-on-crime 1990s produced a remarkable volume of criminal-justice legislation, and some enactments were more truly consequential than three strikes. Wisconsin built a “supermax” prison that held hundreds of unfortunate inmates in austere, windowless cells for 23 hours per day (Bauer 1999). Wisconsin authorized the transfer of inmates to out-of-state, for-profit institutions and soon became the nation’s number-one exporter of prisoners (Jones 1998). Wisconsin adopted a civil commitment law for sexually violent offenders, hundreds of whom would eventually languish in indefinite confinement (Cummings 2013). Wisconsin overhauled its juvenile justice system so as to move more young offenders into adult courts; reforms included lowering the maximum age for juvenile court to 16 (Zimmerman 2003). For all of the significance of these measures, though, the most broadly consequential legislation of the tough-on-crime era was likely TIS.

Wisconsin’s wide judicial sentencing discretion was traditionally counterbalanced by equally wide parole discretion. Parole softened sentencing disparities, encouraged good behavior and rehabilitative efforts by prisoners, and served as a safety valve for prison overcrowding. It was the latter function that catalyzed the backlash against parole. Wisconsin embarked on a massive prison-building spree in the 1990s to accommodate the imprisonment boom, but construction could not keep up with demand. Parole practices were liberalized to prevent a crisis (O’Hear 2014). Over the early and mid-1990s, average length of stay actually decreased, even as nominal sentences increased. By 1998, these trends were attracting considerable media and political attention. One district attorney publicly labeled the parole chair “the most dangerous man in Wisconsin” (Welsh 1998). As noted in Section I, the “parole scandal” led to a tightening of parole in the late 1990s.

At about the same time, the legislature adopted TIS, which formally eliminated parole eligibility for all felonies committed on or after December 31, 1999. Wisconsin was hardly alone in restricting parole in this time period; about 40 other states also adopted TIS laws (Sabol et al. 2002). However, Wisconsin’s TIS law was arguably the nation’s most inflexible and draconian (O’Hear 2014): (1) it applied to all felonies, not just violent crimes; (2) it entirely eliminated the possibility of parole, rather than simply pushing eligibility back to a later point in the prison term; (3) it made no allowance for “good time,” which had already been eliminated by the legislature in 1984; (4) it did not establish presumptive sentencing guidelines or otherwise restrict judicial discretion so as to ensure that judges would reduce their sentences in light of parole’s elimination; and (5) it contained an across-the-board 50 percent increase in maximum sentences for felonies.

TIS broke the established pattern for tough-on-crime sentencing reforms in Wisconsin. For a dozen years, Wisconsin had reliably copied the punitive laws that were sweeping the nation, but each time legislators settled on a relatively moderate version of the law. The state’s TIS law, though, went to an unprecedented extreme, which likely resulted from several distinct causes. First, there was an intensifying political rivalry between Wisconsin’s Republican governor and Democratic attorney general. They competed aggressively for the tough-on-crime high ground, and both ended up providing outspoken support for the TIS law. Second, in 1998, when TIS passed, Republicans controlled both houses of the legislature; by contrast, in 1994, when three strikes passed, Democrats still controlled the Assembly. Although most Democrats could be counted on to vote for tough-on-crime legislation when it reached the floor, liberals in the party could use various procedural devices to block or moderate punitive bills as long as Democrats controlled at least one house. Third, unlike three strikes, which took discretion from both the parole commission and the judiciary, TIS took discretion only from the commission. It could be, and was, framed as a “small-d” democratic reform that simply took power from the unelected corrections bureaucrats and transferred it to the elected judges. Finally, TIS included one potential safeguard: the law’s effective date was delayed for a year and a half so that a blue-ribbon Criminal Penalties Study Committee could develop “trailer” legislation to ease the transition to TIS.

Adopted in 2002, the TIS trailer legislation (TIS II) proved a disappointment to those who hoped for a systematic effort to neutralize the severity increases that were implicit in the original TIS law (O’Hear forthcoming). TIS II reorganized the entire penalty classification system, establishing the current Class A to I framework. The maximums in the new framework were generally set so as to replicate the pre-TIS maximums, taking into account the old mandatory release dates that had been part of the parole system. However, the Criminal Penalties Study Committee also chose to retain higher maximums for dozens of offenses. In effect, the across-the-board, 50 percent increase in maximums was only partially undone by TIS II. The trailer legislation also quietly repealed most of the pre-TIS sentence enhancements, converting them to “aggravators” that judges were to “consider” at sentencing. TIS II authorized some inmates to petition for reduced sentences after serving most of their terms; as noted in Section II of this essay, few of these petitions are actually granted. Finally, TIS II created new sentencing guidelines and a new sentencing commission. Not learning from the failure of the first guidelines, however, the second guidelines were also made merely advisory.

Wisconsin’s second commission and guidelines proved even shorter-lived than the first. Temporary guidelines were developed in haste, based largely on input from 18 judges at a single-day meeting in Milwaukee (Wisconsin Criminal Penalties Study Committee 1999). These guidelines employed an unusual three-by-three matrix. Rather than assigning particular weights to particular sentencing factors, the guidelines simply asked judges to categorize each offense as mitigated, intermediate, or aggravated (y axis) and each offender’s risk as lesser, medium, or high (x axis). Underscoring the high level of discretion that was implicit in this approach, ranges in each cell of the matrix tended to be wide (e.g., probation to four years in prison, or 15 to 30 years in prison).

Appointed in 2003, the new commission’s primary task was to develop permanent guidelines. Governor Jim Doyle specifically instructed the commission to address disparities and improve the cost-effectiveness of sentences, including by developing guidelines for the use of alternatives to incarceration (Wisconsin Sentencing Commission 2006). However, the commission proved incapable of developing a consensus around a new approach, and the permanent guidelines ultimately did little more than to codify the temporary ones (O’Hear forthcoming). Meanwhile, whether denominated “temporary” or “permanent,” the guidelines made little headway with the judges. Between 2003, when the guidelines officially took effect, and 2006, judges submitted worksheets to the commission in only 14 percent of the cases that were nominally covered by the guidelines (Wiseman and Connelly 2008). Observers felt that the second guidelines were never really given a chance because of lingering resentment over the first set (O’Hear forthcoming).

Authorized for only five years by TIS II, the ineffectual commission was allowed to expire in 2007. The advisory guidelines technically remained on the books, but without any agency to promote their use or collect and analyze the worksheets. The legislature finally abolished the guidelines two years later.

With the failure of the commission, the onus was back on the legislature to address the fallout from TIS. In 2009, a Democratic legislature and governor adopted a complex set of “earned release” reforms (Norris 2012). In essence, the earned-release law (1) renamed the parole commission as the Earned Release Review Commission (ERRC); (2) transferred authority over sentence-adjustment petitions from the judiciary to the ERRC; (3) reinstituted good time; (4) gave the DOC new authority to release prisoners on its own initiative in a variety of circumstances; and (5) gave early discharge authority to the DOC for offenders on community supervision.

Republicans were harshly critical of earned release from the start, and they continued the attack through the 2010 election season (Pitsch 2009; Marlaire 2011). They argued that the reforms were resulting in the release of many dangerous offenders and transferred power from elected judges to unaccountable bureaucrats. Perhaps in part because of the negative political attention, ERRC and DOC officials proved cautious in exercising their new release powers, which then provoked a new line of attack—that earned release was not saving nearly as much money as had been promised. When Republicans swept to power after the 2010 elections, they wasted little time in repealing nearly the entirety of the earned-release reforms. In all, during their two-year existence, the reforms resulted in the early release of 666 prisoners, although many were already within a few weeks of their regular TIS release date (Norris 2012). As of the time of repeal, only four of the released inmates had committed new crimes, none of them violent (Fanlund 2011).

To optimistic reformers, the Wisconsin–Minnesota comparison depicted in Figures 1 and 2 seems to demonstrate that Wisconsin might cut its imprisonment rate in half without experiencing any increase in crime rates. As an abstract proposition, it seems clear that Wisconsin could have a sentencing and corrections system that is far less costly, in both fiscal and human terms, and no less effective at protecting public safety. However, it is much less clear what politically viable structural changes might move Wisconsin in such a direction.

Two structural differences between the Minnesota and Wisconsin systems likely explain much of the post-1970 divergence in imprisonment rates between the two states (O’Hear forthcoming). First, there was Minnesota’s Community Corrections Act of 1973. A community corrections act shifts responsibility for community supervision from state to county government. (In Minnesota, this is done at the county’s option.) State subsidies support county supervision programs, but financial penalties may be triggered by a county’s overuse of the state-funded prison system. In principle, a community corrections act should encourage locally accountable prosecutors and judges to make greater use of probationary sentences, both because a locally administered probation program should be better tailored to fit local circumstances than a state program and because probation should become more fiscally advantageous for local government. Wisconsin adopted such a system for juvenile corrections in 1978, but a similar change for the adult system was resisted, in part because of the much larger size of the adult system and the corresponding risk to county finances. With the adult system several times larger now, a full transfer of community corrections responsibility to the counties seems unlikely any time soon. However, more narrowly targeted reforms in the spirit of community corrections are viable. In 2005, for instance, the legislature adopted the treatment alternatives diversion program, which authorizes state grants to local government to support treatment-oriented alternatives to conventional prosecution, especially for drug offenders. The program was regarded as sufficiently successful that its funding was quadrupled in 2014. Continued growth seems possible and may over time effectively alter the system of fiscal incentives that has traditionally encouraged the overuse of imprisonment.

Minnesota’s second major structural reform was the adoption of presumptive sentencing guidelines in 1978. Although presumptive guidelines seem to have helped Minnesota and other states to restrain imprisonment growth (Spelman 2009), Wisconsin has never come particularly close to adopting such a system, and even advisory guidelines have encountered strong resistance from the judiciary. There has been no serious proposal for new guidelines of any type in Wisconsin for many years, and the failure of two prior experiments makes a third attempt unlikely in the near term.

Although Wisconsin cannot be expected to replicate Minnesota’s guidelines or community-supervision system, reformers will continue to call for narrower changes in the hope of bringing Wisconsin’s imprisonment rate closer to Minnesota’s. Greater back-end flexibility remains an obvious need, both to reduce unnecessary imprisonment and to establish stronger incentives for inmates to follow prison rules and engage with rehabilitative opportunities (O’Hear 2014). Although a return to traditional parole seems unlikely, a variety of more modest reforms may be possible, including the reinstitution of “good time.” There may also be support for strengthening existing early release mechanisms. For instance, if the state provided legal representation for prisoners seeking a sentence adjustment, then the very low grant rate would presumably increase.

Excessive postprison supervision remains another important issue to address. The judge-imposed period of ES now routinely exceeds the initial term of confinement, which adds considerably to the risk of technical violations and reincarceration. Tighter statutory caps on the length of ES should be imposed. The legislature actually approved such a reform in 2009, but it was vetoed by the governor (O’Hear forthcoming). In the same spirit, there is also a need for an ES analog to “good time”—a way for offenders who do well on ES to earn early discharge.

None of these changes, or any of the others that are now part of the policy conversation in Wisconsin, would do more than nibble around the edges of mass incarceration. However, current circumstances do not seem favorable for more dramatic reforms. For instance, as shown in Figure 5, the great majority of Wisconsin’s prisoners are serving time for violent or sexual offenses. Reform rhetoric in Wisconsin (and nationally) has focused on getting the “nonviolent” offender out of prison. Although well intentioned, this has probably served to reinforce perceptions that violent offenders are a uniformly and unacceptably dangerous lot. There is an urgent need to develop more nuanced, less “othering” ways of talking about violent (and sexual) offenders, which may be a prerequisite to bolder reform.

The prevailing economic focus of reformers may also have to change. There is little taxpayer money to be saved by cutting the prison population unless whole facilities are closed, and there would undoubtedly be considerable political resistance to this in the mostly rural, economically challenged host communities. Reformers must consider how the interests of such communities can be accommodated, but doing so may require new state spending that would lessen any fiscal benefits from imprisonment reduction.

It may be critical, then, to supplement the economic case for reform with an ethical, humanitarian case (Tonry 2011). In Wisconsin, though, an important challenge facing ethical claims made on behalf of prisoners is the state’s sharp racial and geographical divides. Imprisonment has become shockingly commonplace in some of the state’s urban African American communities but remains quite unusual elsewhere. To succeed, the ethical case for imprisonment reduction may ultimately require politically dominant majority groups to display a sense of kinship and compassion where fear and resentment have been far more in evidence for many years.