It is not clear to what extent the three-strikes laws are intended to achieve retribution or crime control. If the latter, it would be through deterrence and incapacitation. For either, this is a relatively inefficient means for achieving those effects. If focused on the most serious violent felonies, the offender with two prior violent convictions should already be facing a long sentence, and it is not clear how much more deterrent effect will be added by proposing that this be extended to a life sentence. In terms of incapacitation, the laws keep people in prison well after their active criminal careers have ended—very likely by age 50, almost certainly by age 60. (Morris 2002) If the choice arises about whom to release — a currently violent offender not under such a mandatory law and an aging one imprisoned under a three-strikes law—the system would have to release the currently serious offender, obviously a choice that contradicts any crime control objective. Once such laws are passed, even when restricted to the most serious offenses, they then provide a platform for expanding the scope of offenses that can be counted as ”strikes.“ (Schlosser 2003) Indeed, the federal law initially was confined to serious violent felonies, but ended up permitting drug convictions to be counted as one of the strikes. Much of the growth in the prison populations since 1980 is attributable to the increase in drug offenders, who comprised only about 6 percent of prison populations in 1979. The adult incarceration rate for drugs increased by a factor of 10 from 1980 to 1996, far in excess of any of the other crime types contributing to prison population. (Morris 2002) Aside from drug offenses, there has been a general upward trend in the general sanction level for the five other crime types that are important for prison: murder, robbery, burglary, aggravated assault, and sexual assault. Blumstein and Beck (1999) analyzed the factors contributing to this incarceration growth. The candidate explanations they considered were 1) more crime, (2) more arrests per crime, (3) more commitments to prison per arrest, and (4) longer time served, including time served on parole recommitment. It turns out that none of the growth was attributable to more crime and there was no change over this period in arrests per crime. The growth was entirely attributable to a growth in punitiveness, about equally to growth in prison commitments per arrest (an indication of tougher prosecution or judicial sentencing) and to longer time served (an indication of longer sentences, elimination of parole or later parole release, or greater readiness to recommit parolees to prison for either technical violations or new crimes) (Blumstein and Beck 1999). One of the most important recent sources of growth in prison population is the level of recommitment of parolees. Parole boards have received the same political message that they too ought to become tougher to keep up with the political tenor of the times. Indeed, they have been more aggressive in delaying release decisions where they still had that option, in performing urinalysis on parolees, and in making recommitments for parole violations. Another significant contributor to the growth in prison populations through the 1980s was the same demographic shift that contributed to the growth of crimes in the 1960s and 1970s—the baby boomers. (Stone 2001) Since the peak age for crime occurs in the late teens, and the peak age for imprisonment is in the early thirties, the population bulge associated with that generation was expected to impact the prison population about a decade later than it impacted the crime rates.
(Morris 2002) The impact of that demographic shift on prison populations was relatively small, however, compared to the shift associated with the drug war, and in the same order of magnitude as the shifts in the punishment of the more common street crimes. In any event, those effects should have passed their peak by the early 1990s as the baby boomers passed out of the high imprisonment ages, and should already have begun working the other way, waiting for the arrival of the echo boom following the demographic trough associated with the 1976 cohort, the smallest cohort in the US population under age 50. (Morris 2002) The growth of incarceration over the past 25 years has been a consequence of public concern about the widespread use of drugs, especially by children, the growing crime rate exacerbated by the growing attention to it by the media, the political appeal of being tough on crime and the corresponding concern about being seen as “soft” on crime. (Morris 2002) The steady decline of crime rates from a 1993 peak to a plateau in 2000, resulting in rates that had not been seen since the mid-1960s, certainly contributed to an easing of the pressure on politicians to do whatever they could, and increasing sentences always seemed to “work,” not necessarily in reducing crime, but in alleviating the pressure and displaying responsiveness. There has also been an easing of the drug problem, particularly in the demand for crack, the drug whose markets have been most marked by violence in the past, so the insistence on incarceration for drug offenders has also diminished. The result of the crime decline would probably not be very effective without some direct pressure to reduce corrections costs, which rose in direct proportion to prison population, augmented by inflation. That pressure has been provided by the serious revenue shortfalls seen in all the states in their 2002 budgets, with a reasonable prospect of things getting worse in the future. For example, California was projecting a $35 billion deficit in its $63 billion budget (Broder 2003). One estimate for the fiscal year 2004 is that all the states collectively will have deficits of between $61 and 85 billion, representing 13 to 18 percent of the total state budgets (Lav and Johnson 2002). Corrections costs, which have been growing without significant concern for fiscal implications, now find themselves on the list of reasonable candidates for reduction. States are proposing a variety of approaches for earlier release and treatment in the community of nonviolent offenders. Perhaps one of the most striking approaches to introducing rationality is that embedded in California's Proposition 36, which requires treatment before incarceration for a defined group of drug offenders. This approach is being picked up by a number of other states. It represents not only a more rational response to the current drug problem, where the resilient market and its recruitment largely negate the incapacitative and deterrent effects of incarceration. Treatment efforts have been shown to reduce drug abuse, at least during the period of treatment. This approach also represents a significant potential for cost savings, since treatment costs are much less than incarceration costs. The most direct indication of the approach to reconsidering the growth of incarceration is the fact that the incarceration rate in state prisons at the end of 2001 dropped for the first time in more than two decades—it dropped from 425 in 2000 to 422 in 2001 (Harrison and Beck 2002) This was a consequence of prison-population reductions in 10 states, including some of the larger states like New York, New Jersey, and Texas. This reduction was not matched by the federal prison system, which continued its steady increase, with the incarceration rate in federal prisons increasing from 44 to 48. As a result, the total prison incarceration rate was essentially steady, with only a slight increase from 469 to 470. (Morris 2002) One approach that might be pursued to reverse the seemingly unbounded growth in punitiveness is to find means to enhance public education so that the naive belief that more punishment will lead to less crime, regardless of the social dynamics of the particular crime (e.g., the substitution in drug markets) or the offenders on whom the punishment is targeted (e.g., offenders with relatively low disutility for prison), becomes less prevalent. Somehow, the subtleties of those differences have not been able to get through the emotion and the rhetoric of sanction policy. It does appear that too much of the policy debate is from the perspectives of the ideological poles—arguing broadly for more incarceration at one extreme (with no concern for where it can be most effectively applied) or, at the other extreme, against incarceration as a broad generality rather than with a sharp focus on where it is least effective and most wasteful or even harmful. In economic policy, in contrast, there still remain sharp ideological differences between the left and the right, but there has emerged a level of theoretical agreement and strong empirical observations that have brought the poles much closer together than has been the case in criminal justice policy. In part, this is attributable to the much more backward state of empirically grounded theory in criminology, which could force resolution on at least some aspects of the debate. A major rise in the level of support for research in the whole area of crime and crime control through the National Institute of Justice (whose annual discretionary budget has hovered at about $25 million for many years—in marked contrast to the $20 billion for the National Institutes of Health) could help move that process forward. (Morris 2002) In policy terms, one could argue for more widespread use of sentencing commissions as appropriate instruments for generating at least some degree of coherence in sentencing structures. Such commissions can represent an antidote to the passionate response to a particularly heinous crime or the mood of the moment and could sometimes serve as an inhibitor of the frequent legislative response of mandatory-minimum sentences. That approach is limited, however, because even sentencing commissions have to respond politically (in most cases, their guidelines must be approved by the legislature), but they have a degree of insulation from the political pressures compared to legislatures. Sentencing commissions that operate under capacity constraints (requiring their sentencing structure to be compatible with the available or committed prison capacity) make even more sense because the capacity constraint imposes an important discipline on the sentencing schedule; even the most conscientious sentencing commissions, however, are sometimes forced to ignore the constraint by the back pressure they get from the legislature, which must approve the guidelines. It is rare that a sentencing commission hears a demand for reduction of any sentences. (Stone 2001) Almost all the communications call for increased sentences, and so the capacity constraint provides a basis for asking for which offenses the sentence should be reduced in order to provide the capacity for the sentence to be increased. Alternatively, when appropriate, the capacity should be increased through the expenditure of public funds, and this budget constraint then becomes an operational constraint. One might also require the development of prison impact statements to be provided to the legislature before any change in sentencing policy is enacted. This is done as a matter of course for other programs that have budgetary implications (and the Congressional Budget Office was established by the Congress for this purpose), but changes in sentencing policy are usually enacted first by a Judiciary Committee and then the legislature without consideration of their costs and prison population impacts. Changes in the legislative process here might force more responsible consideration of sentencing policy. Probably the most likely prospect for change in consideration of sentencing will be through the emergence of a coalition between fiscal conservatives (as they become appalled at the costs of incarceration and the associated inefficiency, spurred on by the growing budget deficits) and liberals (who want to reduce sentences and use the funds for other social programs that might promise to reduce crime in the longer run). (Morris 2002) There are some glimmerings of the emergence of those coalitions, and one might hope to see them coalesce over the next decade. One of the more helpful efforts at restoring rationality would be through reconsideration of the mandatory-minimum sentencing laws introduced over the past quarter century as acts of passion by legislatures. They epitomize the politically rational response to the pressure to “do something” about a particular societal problem, and their functional irrationality is best exemplified by the fact that they most often are targeted at drug offenses, where incarceration is least effective. (Morris 2002) One would hope that the crime decline of the 1990s would provide the political flexibility to enable the legislatures to rethink those laws (Blumstein and Wallman 2000). It is impressive that the Michigan legislature has indeed done just that with the mandatory-minimum laws relating to drug offenses. The new Michigan law also enhances the possibility for earlier parole release for current prisoners who were sentenced under the mandatory-minimum laws. Families Against Mandatory Minimums, which lobbied particularly effectively for the new law, reports that at least eight other states are already considering similar legislative change. (Morris 2002) One might anticipate that many legislatures will be reluctant to consider such laws because of their lingering concern that such an act might be seen as “soft on crime.” A strong alternative would be to pass a “sunset law” that terminates all mandatory-minimum sentencing laws five years after their passage or three years after the enactment of the sunset law. (Morris 2002) The sunset law recognizes that the mandatory-minimum laws were passed in response to immediate concerns that could well pass after some time. Typically, sunset laws call for an analysis of the effects of the sunsetted law — an assessment of its costs and its benefits — in the year before it is scheduled to be terminated. Thus, for those laws whose benefits are found to exceed their costs, the opportunity is preserved to reenact them.