UCLA UNIVERSITY OF CALIFORNIA Los ANGELES

LAW -REVIEW -

COMMUNITY CORRECTIONS AS THE FRONT LINE IN CRIME CONTROL -

Mark A.R . Kleiman ,/

(

/

.

'.

-

/"

Volume 46

~eprinted

•

August1999 .

•

Number 6

with permission of the Regents of the University of California, ©

19~~

,

COMMUNITY CORRECTIONS AS THE FRONT LINE IN CRIME CONTROL

Mark A.R. Kleiman * The inadequacy of supervision practices within the current communitycorrections system (probation and parole) is among both the causes and the results of the overuse of incarceration. In this Article, Professor Mark Kleiman argues that plans of reform need to take into account the decision-making styles of persistent offenders, who tend to be more impulsive and more reckless than the economist's version of normative rationality would dictate. This implies the need for swift and certain, rather than severe, sanctions for violations of program rules, and thus for changes in the technologies used to monitor the behavior of community-corrections clients and in the process by which charges of violations are adjudicated and sanctions imposed. Promising technologies include drug testing, electronic position monitoring, and data links between community-corrections offices and other public and private agencies that routinely interact with communitycorrections clients. Kleiman further argues that sanctions ought, to the extent possible, to be decoupled from the need for judicial processing . Technical violations, and sanctions for them, should be treated as routine incidents within the process of community corrections rather than as occasions for termination of conditional liberty. Serious political and managerial obstacles stand in the way of this proposed direction of reform, but Kleiman concludes that the potential benefits of overcoming those obstacles, in both reduced crime and reduced incarceration, would be substantial.

INTRODUCfION .................................................... ....................................................... 1910

I. THE CHALLENGE ............ ... .................... ...... ....................•................................... 1911 II . THE CURRENT SITUATION ......................... ..................................................... •... 1913 III. THE IMPORTANCE AND BEHAVIOR OF REPEAT OFFENDERS ................................. 1915 IV. REINVENTING PROBATION AND PAROLE .. .... ......•............... ...... ...... ... ..... ............. 1919 CONCLUSION ....•.•.......•................................... ....... .................................. ..... .. ............ 1924

* Professor, School of Public Policy and Social Research, University of California at Los Angeles. Warren Allen, Terri Patchen, and Ryan Wilson provided invaluable research assistance. Douglas Besharov, Jonathan Caulkins, David Cavanagh, Roger Conner, Gloria Danzinger, Jonathan Duke-Evans, Merle Frank, Adam Gelb, Susan Ginsburg, Deborah Harlow, Francis X. Hartmann, Frederick O'R. Hayes, Lowry Heussler, David Hsia, Robert Jesse, Jack Katz, David. Kennedy, Kelly Kleiman, Eric Lotke, Glenn Loury, Robert MacCoun, Sarah Mars, Barbara Maynard, Linda Mills, Daniel Nagin, Jeremy Paretsky, Joan Petersilia, Anne M. Piehl, Peter Reuter, Stephen Shaw, Michael Smith, Eric Sterling, Steven Teles, Douglas Wilson, and Rebecca M. Young all read and commented on one or more drafts. Financial support was rece·ived from the Faculty Grants Program of the UCLA Academic Senate.

1909

1910

46 UCLA LAW REVIEW 1909 (1999)

INTRODUCfION

America overimprisons. The head count behind bars has quadrupled in the last quarter century,1 and its growth shows no sign of letting up. John Dilulio's assertion that "two million prisoners are enough"l is hard to argue with, but current enforcement and sentencing policies will carry us well beyond that figure. On the other hand, while crime has diminished, it remains at unacceptably high levels; changes in correctional policy need to be consistent with the need to further reduce the victimization risks faced by U.S. residents and enterprises, especially in poor urban neighborhoods. The ideal correctional system would have shorter sentences, on average, than the current one. Long sentences should be reserved for those who pose intense and intractable threats to public safety and those whose crimes require very severe punishment on retributive grounds. An ideal system would also spend less on institutional corrections-the "inpatients" of the criminal justice system-and more on the larger number of probationers and parolees, the system's "outpatients." One reason for the current overreliance on prison is that the alternative to prison, community corrections (probation and parole), fails so badly l in terms of punishment, social control, and rehabilitation. That leaves legislators, prosecutors, and judges with no good option for offenders who need and 4 deserve less than imprisonment but more than next-to-nothing. The current system is caught in a vicious cycle. The inadequacy of community corrections helps maintain the prison-building craze, and the expense of incarceration helps keep community corrections starved for resources and therefore inadequate. But just adding resources to probation and parole as now conceived makes little difference, as the experiments with intensively s supervised probation demonstrated. Community-corrections agencies need the ability to effectively require that their clients obey the law, meet their obligations, and do what they need

I. See BUREAU OF JUSTICE STATISTICS, U .S. DEP'T OF JUSTICE, SOURCEBOOK OF CRIMINAL JUSTICE STATISTICS 1997, at 490 tb1.6.35 (Kathleen Maguire & Ann L. Pastore eds., 1998). 2. See John J. Dilulio Jr., Two Million Prisoners Are Enough, WALL ST. J., Mar. 12, 1999, at A14. 3. See generally Joan Petersilia, Parole and Prisoner Reentry in the United States, in UNDERSTANDING PRISONS: PERFORMANCE AND POLICY OPTIONS (Michael Tonry & Joan Petersilia eds., forthcoming 1999) (on file with author). 4. No one deserves the rape that happens to far too many prisoners, but that is another story. 5. See generally Joan Petersilia & Susan T umer, Intensille Probation and Parole, in 17 CRIME AND JUSTICE: A REVIEW OF RESEARCH 281 (Michael Tonry ed., 1993).

Community Corrections

1911

to do to fit themselves for nonoffender social roles. If that were the meaning of probation and parole, community corrections would come to represent a serious alternative to incarceration, rather than being a mere placeholder for absent punishments. To make that idea a reality, community-corrections agencies need the capacity to set and enforce specific rules, including the availability of swift and certain low-intensity punishment for each detected violation of those rules. That in tum requires both technology to observe probationers' behavior and authority and capacity to administer sanctions swiftly and predictably, in the face of the due process requirements established by Morrissey v. Brewer.6 Additional fiscal resources are needed to acquire and employ that technology, authority, and capacity; money by itself is no magic bullet. Three examples of promising technologies are drug testing, electronic position monitoring, and automated data transfer to probation departments from other institutions that serve and supervise probationers. The cost of a community-corrections system employing the full range of supervision technologies would be several times that of existing probation supervision, but it would remain a small fraction-perhaps a quarter-of the cost of • • 7 Impnsonment. The proposed approach is so far from the current understanding of probation and parole that it would be premature to leap forward to implementation, or even to field research, without first creating a national colloquy around the question: "What would it mean to have a real communitycorrections system ?',8 This Article is intended to promote the creation of such a colloquy.

I.

THE CHALLENGE

Crime, especially homicide, has been declining for the past several years, but measured crimes are still twice as common as they were in the 9 1950s. The unmeasured costs of crime avoidance-arming, guarding, locking

6. 408 U.S. 471, 480-84 (1972) (holding that a probationer has a liberty interest in probation status that can be taken away only after a due process hearing). 7. See Mark A.R. Kleiman, Coerced Abstinence: A Neopaumalist Drug Policy Initiative, in THE NEW PATERNALISM: SUPERVISORY APPROACHES TO PoVERTY 182 passim (Lawrence M. Mead ed., 1997). 8. See Michael E. Smith & Walter J. Dickey, What If Corrections Were Serious About Public Safety!, CORRECTIONSMGMT. Q., Summer 1998, at 12, 12. 9. For homicide, see generally JAMES ALAN Fox & MARIANNE W. ZAWITZ, U.S. DEPT OF JUSTICE, HOMICIDE TRENDS IN THE UNITED STATES (1999).

1912

46 UCLA LAW REVIEW 1909 (1999)

up, staying in, shutting down, and moving-must far outweigh the damage done by completed crimes. 1o The costs of victimization fall largely on victims. The costs of victimization-avoidance measures fall both on those who take them and on those who are inconvenienced (or worse) when the public life of street corners and parks withers because of fear of crime, when shops close early or move away, when businesses relocate, taking job opportunities with them, and when residential real estate is allowed to decay in the face of crime-induced reductions in demand. Indeed, the widely believed proposition that the key to crime control is improving economic opportunity for young people in poor neighborhoods might usefully be turned around: The key, or at least a key, to improving economic opportunity in poor neighborhoods is to reduce the criminal riskiness of low-income urban environments for potential employers and customers. II Crime statistics overstate the true gains in public safety in two ways. First, the fear of crime drives the huge, largely unmeasured costs of crime 12 n avoidance. Fear has been increasing, even as the measured crime rate falls. Second, the flow of population from high-crime to lower-crime areas creates a downward bias in overall crime rates; if crime rates in each area were to remain constant, but population were flowing from high-crime to low-crime areas, the nation as a whole would show a decrease in crime even though no individual jurisdiction or neighborhood had actually become safer. Demographic trends, which have been favorable for the past fifteen years, are beginning to turn the other way as the echo baby boomers come into their crime-prone years. Therefore, if age-specific crime rates, or, worse, age-by-race-specific crime rates, remain at current levels, overall crime rates will rise sharply over the next decade.

10. If they did not, it would be hard to understand in rational tenns the political salience of the crime issue, since the human and financial costs of completed crimes are small by comparison with the figures for problems such as automobile accidents, which engage much less public concern. 11. See generally WESLEY G. SKOOAN, DISORDER AND DECLINE: CRIME AND THE SPIRAL OF DECAY IN AMERICAN NEIGHBORHOODS (199O). 12. See generally Philip J. Coole, The Demand and Supply of Criminal Opportunities, in 7 CRIME AND JUSTICE: A REVIEW OF RESEARCH 1 (Michael Tonry & Norval Morris eds., 1986) (arguing that the rates of completed crimes derive from interactions between offenders' willingness to commit crimes and victims' crime-avoidance efforts). 13. See Pamela Wilcox Rountree & Kenneth C. Land, Burglary Victimization, Perceptions of Crime Risk. and Routine ActWities: A Multilevel Analysis Across Seattle Neighborhoods and Census Tracts, 33 J. RES. CRIME & DELINQ. 147 passim (1996).

Community Corrections

1913

At the same time, the costs of crime control, monetary and nonmonetary, have also been growing. The money costs are still relatively modest: $125 billion nationally, compared to twice that amount for national defense, about three times that amount for education, and five or six times that amount for health care. 14 However, at the state level the competition between prisons and universities-the two largest discretionary items in most state budgets-has become intense, threatening both access to public higher education for the nonwealthy and the quality of public higher education for all its students. IS Since public higher education has been a massive engine of social mobility, degradation of that capacity might have far-reaching consequences. The nonmonetary costs of crime control, including both the intrusiveness of policing and the suffering occasioned by imprisonment, are tragically 16 high, especially in poor urban neighborhoods. The politics of crime control have undergone a fundamental change over the past generation. For those of us who regard crime as a central social problem, and especially a central problem in the lives of poor and working-class Americans, this has been a mixed blessing. Crime has a solid place on the public agenda, and crime-control funding is easier to find than it has been in the past, but the nation's political and intellectual leadership has yet to shape that public concern into effective and just crime-control policies. The terror of the American people in the face of seemingly inexorable increases in crime, and especially criminal violence, has abated only modestly with declining crime rates, and support for foolishly cruel policies • 11 remams strong. Thus our challenge is to continue to move crime rates downward despite unfavorable demographic trends, while at the same time reducing the direct and indirect costs of private anticrime precautions and the monetary and human costs of public crime-control efforts.

II.

THE CURRENT SITUATION

There are always many more recently convicted people than prisons can hold. About 70% of the people under criminal justice supervision are

14. See BUREAU OF THE CENSUS, U.s. DEP'! OF COMMERCE, STATISTICAL ABSTRACT OF THE UNITED STATES: 1998, at 308 tb1.500, 342 tb1.543 (1998). 15. See ill. at 308 tb1.500. 16. See The Dark Side of Zero Tolerance, ECONOMIST, Apr. 3,1999, at 13,13-14. 17. See The Case for EmptieT Prisons, ECONOMIST, Dec. 9, 1995, at 25, 25-26.

1914

46 UCLA LAW REVIEW 1909 (1999)

on probation or parole rather than behind bars.IS Yet the prisons, holding only 30% of the offenders, get more than 80% of the budget dollars.19 Limited budgets and underuse of technology lead to inadequate super~ vision of probationers. For example, Los Angeles County runs caseloads in the hundreds per correction officer, even for felony probation, and averages 20 about 1.3 drug tests per probationer per year. Even if the caseloads were much smaller, effective supervision would remain impossible as long as community~corrections officers rely primarily on one~on~one meetings in the probation office as their means of obtaining information about probationer behavior. Thus high rates of continued offending should be no surprise, in fact, 44% of crimes resulting in prison terms (not including probation and parole revocations) were committed by persons on community~corrections 21 status at the time. A little arithmetic points out the problem. Assume community~ corrections case loads of twenty~five offenders per officer, a level typical of the most intensively supervised programs. Assume further a corrections officer workweek of 37.5 hours and that one~third of that officer's time is taken up with meetings, paperwork, and other activities not involving meeting with "clients." Under those implausibly generous assumptions, a community~corrections officer would have twenty~five hours per week for client meetings, or one hour per client per week, leaving each client 167 hours per week of unsu~ pervised activity. Thus high recidivism rates among probationers and parolees should not surprise us; under current conditions, probation and parole are probably best thought of as legal statuses allowing swifter incarceration or reincarceration when fresh offenses are detected, rather than as programs with independent incapacitative or rehabilitative effects. Even when a probationer or parolee is detected in violation of the terms of his or her liberty, any substantial sanction requires an adversarial court (or parole board) hearing and is entirely at the discretion of the judge (or parole board), and therefore unpredictable from the perspective of the 18. In 1997. 69.4% of the population under correctional supervision-almost four million people- were on either probation or parole. See BUREAU OF JUSTICE SfATISTICS. supra note 1. at 464 tb1.6.1. 19. In 1993. correctional institutions received 83 .6% of state corrections funding-$15 .9 billion-and parole and probation received 16.4%. or $3.1 billion. See BUREAU OF JUSTICE SfATISTICS. supra note 1. at 11 tp1.1.9. 20. As of December 1997. the Los Angeles County Probation Department reported 74.249 adult probationers and 98.045 drug tests administered in the previous twelve months. (Of those. 10.779 registered drug use; in addition. there were 14.800 failures to appear for testing.) See Letter from Laurence Naples. Los Angeles County Probation Department. to Mark A.R. Kleiman. Professor. UCLA 1 (Dec. 1. 1997) (on file with author). 21. See Petersilia. supra note 3 (manuscript at 29).

Community Corrections

1915

offender. As a result, many offenders regard a probationary sentence as virtually equivalent to a suspended sentence. This creates at least four unfortunate results. First, handing out merely nominal punishments to those just starting their {adult} offending careers may reduce their fear of arrest and conviction, thus encouraging continued offending. Second, crime victims whose victimization results in communitycorrections sentences are frustrated. Third, judges who want to impose real punishment are driven to overuse incarceration. Fourth, legislatures are driven to tie the hands of judges through mandatory sentencing and to limit or eliminate parole through "truth in sentencing" laws, further increasing the number of persons behind bars.22

III.

THE IMPORTANCE AND BEHAVIOR OF REPEAT OFFENDERS

Most serious offenses are committed by offenders with high personal crime rates {the criminologists' "lambda"} who cycle through probation again and again, both before they go to prison for the first time and in between prison terms. That is not because this is the most typical pattern of offending-most of those who offend have low personal crime rates-but because of the overwhelming contribution of the "right-hand-tail" of the distribution of offenders, consisting of those with very high crime rates. 23 In this, offending resembles the many other behaviors covered by "Pareto's Law," which holds that 80% of the volume of any given activity tends to be 24 concentrated among 20% of the participants. Despite the relatively low probability of being caught and punished for any given offense, the cumulative probability of detection and punishment over a career of persistent serious offending is sufficient to make most forms of criminality available to poor and poorly educated young men thoroughly unattractive, even compared with their other opportunities. The calculations below illustrate this principle. The typical "take" from a domestic burglary, as paid by a receiver of 25 stolen property to the burglar, is on the order of $300. If we imagine a 22. First enacted in 1984, truth-in-sentencing laws require offenders to serve a substantial portion of their prison sentence. Parole eligibility and good-time credits are restricted or eliminated. See PAULA M. OIlTON & DoRIS JAMES WILSON, TRUTH IN SENTENCING IN STATE PRISONS 1 (Bureau of Justice Statistics Special Report No. 170032, 1999). 23. See generaU, JAN M. CHAIKEN & MARCIA R. CHAIKEN, VARIETIES OF CRIMINAL BEHAVIOR (Rand Corp. Doc. No. R-2B14-NIJ, 1982). 24. See generaU, Norris O. Johnson, The Pareto Law, REV. ECON. STAT., Feb. 1937, at 20; D.H. Macgregor, Pareto's Law, ECON. J., Mar. 1936, at BO. 25. The average value of property lost in a burglary is $1344 (total losses of $2.35 billion spread over 1.76 million burglaries). See FEDERAL BUREAU OF INVESTIGATION, U.S. DEP'T OF

1916

46 UCLA LAW REVIEW 1909 (1999)

conscience,free, individually rational actor contemplating the decision whether to undertake such a burglary, that person would weigh the benefit of $300 against the effort involved {not very great} and the risks of the activity, including physical injury and legal punishment. Being caught and punished is, from the burglar's viewpoint, a stochastic 26 process. Each burglary carries a relatively low risk of apprehension. One useful summary statistic used by rational decision,makers in such situations is the "expected value" {also called the "mathematical expectation"}. It is the weighted average of probability times severity, across all possible outcomes. In the case of crime and incarceration, we can approximate the expected value of punishment by dividing the total number of person, years served for a given crime in a given year {approximated in turn by the number of persons in prison for that offense on a given date within that year} by the total number of reported incidents of that crime in that year. 27 For burglary, that expected value works out to approximately twenty days.2s Dividing the expected reward by the expected punishment yields roughly $15 per day behind bars, illustrating the maxim that "the wages of sin are well below the legal minimum." In practice, of course, punishments are unpredictable and come later than offenses. For our notional rational would,be burglar, unpredictability would further reduce the attractiveness of the proposition {because of risk aversion}. On the other hand, the time delay would reduce the value to be placed on the punishment {because of discounting} at the rate of several percent per year, thus making burglary slightly more attractive, but still not nearly attractive enough. That suggests that low,skilled burglary is not a rational choice by the canons of economic reasoning-even for those whose licit earning opportunities are very limited-and conversely that those who engage in it frequently may not be rational actors as economists understand that notion.

JUSTICE, UNIFORM CRIME REPORTS: CRIME IN THE UNITED STATES 210 tbl.23 (1997) . The estimate of $300 reflects the discount between the value as reported and the payment made by receivers of stolen property. In 1995, about 2.1 million reported burglaries-presumably an underestimate of the 26. total-led to fewer than 266,363 burglary arrests, implying an arrest risk of just under one in seven. BUREAU OF JUSTICE STATISTICS, supra note 1, at 330 tb\.3.116, 377 tbI.4.6. 27. See MARK A.R. KLEIMAN ET AL., IMPRISONMENT·TO·OFFENSE RATIOS 2 (BOTEC Analysis Corp. Working Paper No. 89·06·02,1988). 28. There were 109,500 burglars in prison at year.end 1996. See BUREAU OF JUSTICE STATISTICS, U .S. DEP'T OF JUSTICE, EXECUTIVE SUMMARY NO. 171684, CORRECTIONAL PoPULATIONS IN THE UNITED STATES, 1996, at 3 tbl. (1999). Burglaries reported to police totaled about 2.1 million, and 109,500 /2,100,000 = 0.05 years, or 19 days, per burglary.

Community Corrections

1917

Recent advances in behavioral economics and psychology suggest that the normative rational-actor models of risk preference and time preference tend to be relatively poor empirical models for many persons and in many circumstances. "Prospect theory" describes persons who are risk-averse when it comes to gains, as the normative theory would imply, but act as if they were risk-seeking when it comes to losses. This makes activities that yield many small gains and a few large losses more attractive to such persons 29 than would be the case for rational actors. The theory of "hyperbolic discounting" describes behavior that, compared to the orthodox account of time preference, overweights immediate gains and losses compared to even 30 slightly deferred gains and 10sses. Thus, these two theories, taken together, account for reckless and impulsive behavior. Repeat offenders tend to be reckless and impulsive.)! To what extent their behavior in fact conforms to the precise models of prospect theory and hyperbolic discounting is as yet unknown. But to the extent that it does, controlling the behavior of repeat offenders will depend more on the ability to impose swift and predictable sanctions than it does on the severity of the sanctions eventually imposed. That is not to say that the current system, which might be called one of "randomized Draconianism," serves no purpose. On the contrary, it makes offending unattractive to all of those whose decision-making styles even approximate the canonical economic account of rational choice under uncertainty and over time.J2 That leaves the problem of deterring criminality on the part of those whose decision styles depart substantially from that norm. Much of the political discussion of crime control has centered on increasing the severity of punishment, in part because of the relative ease of varying severity by legislative action. But if the problem of crime is largely one of repeat offenders, and if the current system already deters repeat offending by all except those whose decision-making styles react only weakly to severity, then increasing severity {through mandatory sentencing, "truth in sentencing," "three strikes," or other sentencing schemes} will

29. See Amos Tversky & Daniel Kahneman, The Framing of Decisions and the Psychology of Choice, 211 SCIENCE 453,453-54 (1981). 30. See George Ainslie & Nick Haslam, Hyperbolic Discounting, in CHOICE OVER TIME 57, 59-71 (George F. Loewenstein & Jon Elster eds., 1992). 31. See JAMES Q. WILSON & RICHARD J. HERRNSTEIN, CRIME AND HUMAN NATURE 194-98 (1985); see also John J. Dilulio Jr., Defining Criminality Up, WALL ST. J., July 3, 1996, at

A10. 32.

lowe this point to Daniel Nagin.

46 UCLA LAW REVIEW 1909 (1999)

1918

prove to be largely a dead end except insofar as it singles out high-lambda offenders for long imprisonment early enough in their criminal careers to avoid turning the prisons into retirement communities for former offenders.]] But the process of catching offenders and taking them through the courts is, by its nature, too random and too slow to allow great increases in the certainty or swiftness of punishment. Thus all the routes to increased deterrence-through severity, certainty, or swiftness-seem blocked in the case of repeat offenders. The community-corrections system can, in principle, apply sanctions with greater certainty, because of the legal liability of probationers and parolees to intrusive surveillance, and with greater swiftness, because of the absence of the full panoply of rights accorded criminal defendants. Moreover, community-corrections officers can back up what would otherwise be merely good advice for someone planning and hoping to desist from criminality-advice about drug use, time management, choice of associates, treatment for behavior problems, job seeking and job holding, and skills acquisition-with promises and threats of immediate pleasant and unpleasant consequences. Making those possibilities-in-principle realities is the promise of the new community corrections. Even more is at stake here than controlling the criminality of probationers and parolees themselves. Pareto's Law also fits consumption of heroin and cocaine. While most users of these drugs are casual users, many of whom pay for their drugs entirely from licit sources of income, the bulk of total consumption is accounted for by a relatively small number of very heavy users. That heavily using population overlaps substantially the population of persistent offenders. By one estimate, about three-quarters of all frequent, high-dose users of cocaine and heroin are arrested in the course of a given year, suggesting that some 60% of the total volume in those markets is purchased by those who are under the nominal supervision of the . . commumty-correctlOns system. The implication is clear: Our chances of shrinking the hard-drug markets, with all that would mean in terms of violence reduction, neighborhood protection, and reduced need for incarceration, depend largely on managing the behavior of those frequent hard-drug users who are also clients of the community-corrections system. )~

33. See generally PE'TEjl. W. GREENWOOD & ALLAN ABRAHAMSE, SELECTIVE INCAPACI· TATION (Rand Corp. Doc. No. R.2815.Nlj, 1982). 34.

See Kleiman, supra note 7, at 200.

Community Corrections

1921

Existing probation and parole regimes impose many rules, arguably too many, although as noted above they tend not to require, even on paper, that their clients improve their job-related basic cognitive skills in reading and mathematics. Where they fall down is in their capacity to detect violations of those rules and to impose and administer appropriate sanctions. The keys to making community corrections real are better technology, to allow community-corrections officers to monitor various aspects of their clients' behavior, other than by interview, and increased authority and capacity to impose immediate and predictable sanctions in response to violations of terms of conditional liberty. As argued above, face-to-face interviews do not constitute an effective monitoring technology. Three examples of promising technologies are drug testing, electronic position monitoring, and automated data transfer to community-corrections departments from other institutions that serve and supervise the same clients. Of these, drug testing, while it covers the narrowest slice of behavior, is the best developed and least problematic. Urine tests covering up to five drugs at a time can be administered for a total cost of less than $5 per test, even including confirmation tests when a client challenges the validity of a "dirty" (positive) finding!2 A variety of techniques, from the "electronic handcuff' that monitors an offender's proximity to a fixed base unit in the offender's residence, through currently undeveloped applications of cell-telephone or geopositioning satellite technology, to the use of mobile video scanners equipped with face-recognition capacity, can be applied to the problem of detecting whether an offender is in a required (or forbidden) location. This would have value both in enforcing detailed time-and-place requirements and also in accusing (or exonerating) individual probationers of offenses that can easily be pinned down in time and space, such as robberies and assaults. Less ambitious technically, though more demanding organizationally, would be the development of data links between community-corrections offices and other public and private entities-drug, anger-management, and other treatment programs; agencies supervising community service; agencies collecting child support or restitution payments or fines; family-services agencies; schools; and workplaces--that have in their operating records evidence about whether individual probationers and parolees are complying with the terms of their conditional liberty. A community-corrections officer who could look at a computer screen each morning to find out which of his or her clients had been out of compliance the previous day could

42.

See Kleiman, supra note 7, at 210.

1922

46 UCLA LAW REVIEW 1909 (1999)

exercise a degree of control scarcely dreamed of today. Such a system would produce much of the benefit of "day reporting" programs at a small fraction of the cost. The capability to detect violations is of little use without the authority and capacity to punish them. The current system of "seeing the judge" (or the parole board) is too slow, too time consuming for probation officers, too random in its outcomes, and, when it does result in any sanction at all, almost invariably too severe. A properly functioning community~corrections system will detect many technical violations, and it needs to be able to handle them on a routine basis, within probation or parole status, rather than as exceptional events calling for revoking that status and moving the offender from the community to the institutional part of the corrections system. Typical punishments should be hours of unpaid labor, tightening of 43 curfews and other restrictions, and short periods of confinement. Ideally, such sanctions would be assigned administratively, according to a set of published protocols, and with a hearing to determine only the fact of the violation, not the sanction. That creates the sanction certainty essential to effective deterrence and helps shift the client's attention from the likely behavior of others to the links between his own behavior and its results. With sufficient legal ingenuity, it ought to be possible to conform such a program to the due process requirements of Morrissey v. Brewer in any of several ways: by extending "day reporting"-where confinement is legal rather than physical-around the clock; through a consent to administra~ tive discipline in lieu of judicial processing that imposes tougher sanctions; or by having an "on~call" judge, perhaps at the same location as the probation department, available to hold swift hearings and hand out formu~ laic sanctions. Insofar as those sanctions involve confinement-and it should be noted that anything except confinement is essentially voluntary and there~ fore depends in the last analysis on the existence of a credible confinement threat in the background-that confinement should not be in conventional prisons or jails, but in specially designed and operated community~ corrections sanctions facilities optimized for very~short term confinement. This would avoid the problems created for jailers by having large numbers

43. To be effective, assigned labor and tightened restrictions need to be enforceable; they thus need to be easily monitorable. Where confinement is used to sanction technical violations, terms should be short, rypically just a day or two, both to make nonjudicial sanctioning tolerable and to avoid overstretching scarce confinement capacity.

Community Corrections

1923

of very-short-stay inmates to muster in and out. It would also possibly avoid the "liberty interest" issue under Morrissey v. Brewer by making the confinement legal rather than physical, though it would be necessary to have alarms and someone with arrest powers to stop absconding (not, technically, "escaping") offenders on their way out. Moreover, such special facilities could be much cheaper than jails, because fewer services would be required; less brutal, because short-stay prisoners could be kept apart from one another; and, by the same token, more aversive, because of the imposition of solitude and silence (no radio or television, just books). The new system would require more community-corrections officers, drug-testing and position-monitoring equipment, information-management capacity, and staff and facilities for short-term confinement. It would also need the capacity to seek out probationers and parolees who fail to appear: either community-corrections officers with arrest powers or a cadre of police officers who would make probation- and parole-related warrant service a primary activity rather than an afterthought. 44 (Paid overtime details offer one option for attracting police to an activity that, despite its importance, has low professional standing.) While the full-dress version of this approach would cost a multiple of the current cost of probation or parole, it would still be a just a fraction of the cost of incarceration. Nor would every probationer or parolee need all, or even most, of the supervision capacity of the new communitycorrections system. Even those starting out with the entire menu of restrictions would be expected to earn reduced restriction and scrutiny by periods of compliance. Implementing the program described above will require that probation and parole officers be more numerous, more broadly trained, and better equipped with information technology than they now are. In some cases, it will also require that they be armed. Pay, working conditions, and recruiting standards should reflect the increased level of responsibility implied by the idea of community corrections as a front-line crime-control agency. With the new resources and powers proposed here, community-corrections offices, and even individual officers, could reasonably be held accountable for improving measurable aspects of their clients' behavior and condition, including criminal behavior, drug use, and acceptance of family and workplace responsibilities.

44. On the laxity of current warrant service efforts, see general\y Faye S. Taxman & James M. Byrne, Locating AbscondeTs: Results fram a Randomited Field Experiment, FED. PROBATION, Mar. 1994, at 13.

1924

46 UCLA LAW REVIEW 1909 (1999)

In their admirable discussion of another set of ideas about making community corrections work, Michael Smith and Walter Dickey suggest that even the terms "probation" and "parole" should be abandoned, in favor of "community confinement and control," as recommended by the Wisconsin s Governor's Task Force on Sentencing and Corrections: Certainly neither their proposed system, with probation and parole officers getting out from behind their desks to work side-by-side with police in high-crime neighborhoods, nor the system proposed here has much in common with probation and parole as currently established, and a change of label might be useful for both political and managerial purposes. It would be fatuous to ignore either the managerial weakness of current community-corrections agencies or the lack of political standing of which the managerial problems are both a cause and an effect. The most disheartening fact about community corrections, from the viewpoint of their political standing, is that no one even bothers to blame probation and parole agencies for the bad behavior of their charges; their incapacity to do their assigned task is simply taken for granted, and the blame is assigned to whatever individual, institution, or policy was responsible for having those offenders on the street rather than behind bars. Nothing less than a perceived crisis, or the entrepreneurial intervention of powerful players, is likely to provide the energy to break out of such a low-level performance equilibrium trap. The fact that probation departments tend to be located within the judiciary, while the primary political beneficiaries of reduced crime and corrections spending are mayors and governors, only makes the situation grimmer. However, it would not be unreasonable to hope that some governor or aspiring governor, finding the current menu of crime-control and budgetary choices unpalatable, might be moved to gamble on the reinvention of community corrections, under whatever name. CONCLUSION

Community supervision should be converted into a serious means of punishment, monitoring, control, and rehabilitation. This would involve: • the imposition and modification, as required, of specific rules addressing drug use, educational effort and progress, locations to frequent or avoid at specified hours, participation in treatment programs, performance of required community-service labor, and fulfillment of family, school, and workplace responsibilities; 45.

See Smith & Dickey, supra note 8, at 24.

Community Corrections •

1925

the implementation of means for monitoring compliance with those rules; and • the rapid imposition and execution of formulaic and predictable low, intensity sanctions for each detected violation, rather than the current low,probability, deferred threat of possibly lengthy incarceration. Effectively implementing such a system would reduce both crimecurrent and future-and the need for imprisonment, thus reducing the burdens on victims, taxpayers, and offenders. The operational difficulties, while formidable, could all be overcome if the still more formidable political and managerial barriers were first broken through.