Bill Clinton memorably entered the White House eight years ago over the body of poor, lobotomized Ricky Ray Rector, whose execution he had hurried back to Arkansas to attend. As he departs, the American prison population stands at two million, an all-time high, up from 1,429,000 in 1992, with a disproportionately soaring rate of incarceration among young black men. His administration saw the introduction at Federal level of the ‘three strikes and you’re out’ sentencing policy (imposing life prison-terms, without parole, on a third conviction) and increased penalties for drug-related crimes in the Sentencing Commission’s mandatory guidelines. It actively promoted ‘truth-in-sentencing’ provisions (prisoners forced to serve at least 85 per cent of their sentences before parole), pumping Federal funds into prison-building projects in states where such practices prevailed. Small wonder that Gore and Clinton failed to protest at the 700,000 or so (predominantly black) Florida voters disenfranchised as a result of previous felony convictions; these were policies they had been conniving at for the past eight years.

American incarceration rates are now proportionately six times higher than those of Britain, Canada or France. In addition to this, a further 3.2 million Americans are on probation, and 685,000 on parole. This huge increase in the prison population has been heavily racialized: between 1984 and 1997, the proportion of adult white men in prison rose from 0.5 per cent to 0.9 per cent, whereas the percentage of incarcerated adult black men rose from 3.3 per cent to 7.2 per cent. By the end of this period there were 758,000 black men in prison, along with 274,000 on parole and a further 902,000 on probation. Altogether, more than 18 per cent of all adult black men were under some form of correctional supervision in 1997.footnote1 Almost a third (32 per cent) of black men between the ages of 20 and 29 are currently ‘under some type of correctional control’—incarceration, probation or parole—compared to 1 in 15 whites, or 1 in 8 latinos.

This surge in prison numbers has not been the result of a sudden crime boom but of deliberate changes in US criminal justice and sentencing practice.footnote2 The introduction, across state after state, of ‘three strikes’, ‘truth-in-sentencing’ and ‘zero tolerance’ (suspects arrested and charged for the most minor offences) has hugely increased the number of arrests and prison sentences, and the length of time served. The number of prisoners doing time for relatively minor, non-violent offences has also soared: these accounted for 70 per cent of all new committals to US state prisons in 1996—over 400,000 inmates are held for drug offences alone.footnote3 While the ‘three strikes’ policies are usually assumed to refer to convictions for violent felony, in some states, including California, only the first two offences need come from a specifically enumerated list of ‘serious’ crimes—a list which, astonishingly, includes burglary, although burglaries by definition involve no victim contact, and the amount stolen is usually worth less than $500. The third ‘strike’ can be any felony, no matter how trivial, committed at any subsequent time. Juveniles have no right to trial by jury, yet their offences can also be counted as ‘strikes’. A sixteen-year-old who steals from two neighbours’ garages in the same afternoon can get two ‘strikes’ with one guilty plea.

The only justification for these brutalizing sentencing policies, as propounded by Clinton, Bush and Gore—and parroted on the other side of the Atlantic by Jack Straw and Ann Widdecombe—is that ‘prison works’: that high levels of imprisonment will reduce crime rates and deter serious drug abuse. But do America’s harsh new incarceration practices actually achieve this? The most reliable source of inter-country difference in crime rates, the International Crime Victimization Survey (ICVS) fails to confirm the relationship between high levels of imprisonment and low crime rates (see Table 1).footnote4 The latest survey shows that America’s overall victimization figures remain around average for the sample shown: despite the swollen size of the US prison population, American citizens are just as likely to be victimized as the inhabitants of other countries with far fewer prison inmates and actually run a greater risk of homicide and ‘aggressive contact crime’ (robbery, sexual assault and other violent attacks).

A closer examination of US Justice Department statistics reveals, in fact, an extraordinary absence of correlation between prison population and crime rates.Between 1977 and 1996, there were two spells (1980–84 and 1991–96) when the rise in incarceration rates did coincide with a fall in crime rates; but there were also periods (1977–80 and 1984–91) when the crime rate rose, despite the growth in the prison population (see Figure 1).

Another illustration of the overall failure of America’s prison policy to reduce crime can be found by looking at inter-state variations in the relationship between crime and incarceration rates. Here, one might expect states with the steepest increases in imprisonment to have the slowest growth in crime. Instead, as Franklin Zimring has shown, incarceration rates were poor predictors of change in crime rates during the 1980s.footnote5 Applying Zimring’s technique to the available state-level data for 1990–96 shows that, if anything, the correlation was even weaker. The correlation coefficient between the percentage changes in crime and incarceration rates actually fell, from 0.32 during the period 1980–90, to 0.15 for 1990–96.footnote6 Thus, inter-state variations in prison expansion explain less than 3 per cent of the variation in inter-state crime rates—confounding any attempt to establish a straightline relationship between the two. Of the wide range of social and economic forces that may affect the crime rate, the deterrent effect of tough sentencing policies clearly plays, at best, a limited role.

If prison growth has had little effect in reducing ‘victim’ crimes, it has made even less of an impact on the rate of serious drug-use. The number of prisoners incarcerated annually for drug offences rose more than twelvefold between 1979 and 1997—from 18,000 to 227,000—without any demonstrable effect on the availability of illicit narcotics or the prevalence of ‘hard-core’ use.footnote7 The percentage of high-school seniors who thought it ‘fairly easy’ (88 per cent) or ‘very easy’ (89 per cent) to get hold of marijuana remained unchanged between 1975 and 1995, while the number who thought it easy to get hold of ‘hard’ drugs actually increased.footnote8 The rising purity of the drugs entering the US market has contributed to a fourfold increase in drug-related deaths over the past twenty years, and the government’s irresponsible prohibition of needle possession has helped the expansion of HIV and Hepatitis C epidemics.footnote9 ‘We can’t incarcerate ourselves out of this problem,’ Barry McCaffrey, the four-star general who heads the Office of National Drug Control Policy, has admitted. ‘We have a failed social policy and it has to be re-evaluated.’footnote10