Extent of Life-Disposing Punishment in England & Wales Since 1500

Absence of persons in punishment (persons executed, transported, and in prison) was much greater in early modern England than in England and Wales since the mid-eighteenth century. The penal death rate in England probably was about 20 per 100,000 from 1530 to 1630, and over 25 per 100,000 in 1600.^ Assuming persons executed had about 30 years of additional expected life, persons absent in punishment through execution were over 600 per 100,000 from at least 1560 to 1630. Men in prison for debt in England and Wales probably reached 15,000 in 1670. Given that the total population of England and Wales was then about five million, men in prison for debt amounted to 300 per 100,000 in 1670. In total, the absence of persons in punishment plausibly could have reached 1,000 per 100,000 persons in England (1%) in the 1600s. Since 1750, the highest absence in punishment in England and Wales has been 464 per 100,000 in 1842.

The prevalence of persons absent in punishment in England and Wales has changed significantly over the past two centuries. From 1815 to 1842, that prevalence rose from about 200 per 100,000 to about 450 per 100,000. From 1840 to the early-1920s, the prevalence of persons absent in punishment continually fell. The prevalence stabilized about 30 persons per 100,000 absent in punishment in the 1920s and 1930s. Post World War II, absence in punishment continually rose to reach 153 persons per 100,000 in 2009. That level is roughly equal to the prevalence of persons absent in punishment at the beginning of the nineteenth century. Historiography of English criminal law has described early-nineteenth-century criminal law as a harsh “Bloody Code.” Criminal law in England and Wales today is less literally bloody, but disposes about the same number of persons in proportion to the total population. Aggregate British criminal-justice statistics are difficult to reconcile with basic social facts and a common-sense understanding of human nature and justice.^ Imprisonment today obscures the bloody extent of punishment.

Exceptional Increase in Imprisonment in U.S. Since 1980

The increase in imprisonment in the U.S. since about 1980 is exceptional from both historical and international perspectives. In 1850, the extent of absence in punishment in the U.S. was about 60 persons per 100,000 residents. Punishment prevalence in the U.S. rose from the mid-nineteenth century to 1940, and then remained roughly at the same level through to 1980. In 1980, an estimated 206 persons were in prison per 100,000 residents. After 1980, the extent of punishment increased sharply. In 2010, imprisonment prevalence was 732 persons per 100,000 U.S. residents. That’s more than ten times higher than the corresponding figure in 1850 and more than three times higher than in 1980.

While capital punishment attracts much public attention, the death penalty has accounted for less than 10% of absence in punishment in the U.S. since 1850. About 9% of absence in punishment was from penal death in the U.S. in 1850. That share fell to 3% in 1900 and about 1% in 1960. Since 1970, penal death has accounted for less than 1% of absence in punishment. The U.S. has never used significantly banishment as punishment. Since the mid-nineteenth century, absence in punishment in the U.S. overwhelmingly means absence through the punishment of imprisonment.

The exceptional increase in imprisonment in the U.S. after 1980 is evident in international comparisons. In 1850, punishment prevalence in the U.S. was only a fifth of the corresponding figure for England and Wales. In 2010, imprisonment prevalence in the U.S. was five times that in England and Wales. The growth in imprisonment in the U.S. after 1980 is historically unprecedented.

The U.S. gained international leadership in imprisoning its residents after 1980. A limited set of international data on imprisonment prevalence about 1977 shows the U.S. ranked ninth highest. Imprisonment prevalence (measured in persons in prison per 100,000 residents) was 384 in South Africa and 235 in Poland, compared to 184 in the U.S. Imprisonment prevalence in Finland about 1977 was 111. Hence the U.S. / Finland imprisonment prevalence ratio was 1.7 in 1977. That ratio rose to 12.3 in 2010.

The U.S. now leads the world in persons imprisoned per capita. Across the world in 2010, excluding the U.S., about 120 persons were in prison per 100,000 persons. Within the U.S., imprisonment prevalence is five times that level. Looking across national jurisdictions encompassing nearly all the world’s people, imprisonment in the U.S. is three times higher than the jurisdiction at the 75% percentile of the (increasing) imprisonment prevalence distribution. Imprisonment in the U.S. is extraordinarily high from an international perspective as well as from a historical perspective.

The exceptionally high prevalence of imprisonment in the U.S. has attracted some scholarly concern, but relatively little public concern. From a historical perspective on punishment, the U.S. is exceptional both in its extraordinary increase in punishment after 1980 and in its exceptionally high sex ratio of punishment in the nineteenth century. The U.S. penal history of gender inequality in punishment has attracted almost no concern from scholars or the public. Expansive, coercive criminal control of domestic violence in ways biased against men has been central to the development of mass incarceration in the U.S. since 1980.

Punishment in Convict-Settled Australia in Comparative Historical Perspective

Australia’s criminal justice system in the mid-nineteenth century imposed more punishment than did the U.S. and England & Wales. In Australia in 1858, the first year for which aggregate punishment statistics can be constructed, the overall prevalence of life-disposing punishment in Australia (execution and imprisonment) was 378 persons per 100,000 population. Executions accounted for 24% of persons absence in punishment. In the United States in 1858, punishment prevalence was 75 persons per 100,000. Executions accounted for only 6% of persons absent in punishment in the U.S. The overall prevalence of punishment in England & Wales in 1858 was 232, or about 60% of that in Australia. Execution accounted for only 2% of punishment in England & Wales in 1858. Compared to another overseas area of English settlement, the U.S., and compared to England & Wales itself, mid-nineteenth-century Australia punished more extensively and used execution more frequently. Available data, while incomplete, suggests that this difference was even more pronounced earlier in the nineteenth century.

Convicts transported from the United Kingdom accounted for a large share of European settlement of Australia in the first half of the nineteenth century. Transportation of convicts from the United Kingdom to Australia began in 1787. Transportation peaked at about 6000 convicts per year in the 1830s. After 1852, transportation decreased sharply. It ended in 1867. From 1786 to 1852, about 150 thousand convicts were transported to Australia. Australia’s total population in 1852 was about 500 thousand. Transported convicts and their offspring were a major share of Australia’s population in the first half of the nineteenth century.

The predominance of men among transported convicts and punishment’s bias toward men helps to explain relatively extensive punishment in mid-nineteenth century Australia. From 1787 to 1852, the aggregate sex ratio among convicts transported to Australia was about 5 males per female. In 1830, males were about three times as numerous as females in Australia. In 1852, males outnumbered females by 45%. Males were more likely to be punished than females, and more likely to be punished severely. For example, in the Australian state of Victoria from 1842 to 1967, males were more likely to be charged with a capital offense, more likely to be convicted, and more likely to be executed. A plausible estimate for the sex ratio of punishment in Australia about 1858 is 6 men per woman absent in punishment. That figure, along with the difference in population sex ratios in Australia and the United States, can account for most of the mid-nineteenth-century difference in the prevalence of punishment between Australia and the United States.

Sex Bias in Criminal Justice System of Victoria, Australia, 1842-1967

	charged with capital crime	convicted of crime (% of charged)	executed for crime (% of charged)	persons executed
males	2256	24.6%	7.9%	178
females	335	13.5%	1.5%	5
Source: Douglas and Laster (1991) p. 156, data available in executions Australia dataset.

In a sex-sensitive comparative perspective, convict-settled Australia punished surprisingly little. Today, persons imprisoned are often re-imprisoned. Differences in individual criminal propensities and recidivism are major issues in criminology. Adjusting for population sex ratios makes the prevalence of punishment in Australia and the U.S. nearly the same. That leaves little room for criminal effects of Australia’s distinctive convict settlement. Moreover, adjusting for population sex ratios, both Australia and the U.S. in the mid-nineteenth century had a much lower prevalence of punishment than did England & Wales.

Punishment Sex Ratio in England and Wales Since 1750

From 1750 to 1920, the ratio of men to women absent in punishment in England and Wales changed relatively little despite major societal changes . During this period, England and Wales underwent industrial revolutions, unprecedented population growth, and extensive development of state institutions. Both the prevalence of persons absent in punishment and their punishment positions changed significantly across this period. Nonetheless, with the exception of a brief peak after the American Revolutionary War, the ratio of men to women absent in punishment remained between five to ten men per woman from 1750 to 1920.

From 1780, sex ratios of prisoners show a similar pattern to sex ratios of persons absent in punishment. Absence in punishment accounts for persons punished by execution or transportation, in addition to persons in prison. Accounting for executions and transportation is potentially important because those were predominant forms of punishment in England and Wales prior to 1855. However, like the sex ratio of persons absent in punishment, the sex ratio of persons in prison was typically between five to ten men per woman from 1780 to 1920. Ratios of men to women in prison earlier, particularly in the seventeenth century, were much higher as a result of mass imprisonment of men for debt.

Increased professionalism in the administration of justice coincided with a relatively small nineteenth-century reduction in gender inequality in punishment. In 1835, the British national government appointed national prison inspectors covering local prisons and required those prison inspectors to issue yearly reports. The Prison Act of 1877 nationalized all local prisons in England and Wales. The nineteenth-century growth in Britain’s national penal bureaucracy helps to explain the mid-nineteenth-century shift in punishment from predominately transportation to almost exclusively imprisonment.^

The development in penal bureaucracy probably also contributed to a reduction in gender inequality in punishment. The ratio of men to women absent in punishment fell from about 9 from 1825 to 1845 to about 6 from 1860 to 1910. By giving more weight to procedure and routine, the growth of government bureaucracies probably reduced sex-biased discretion in administering justice. In addition, national penal bureaucracy lessened the personal connections between the administrators of justice and persons disposed in punishment. From the perspective of the professionalizing civil servants administering justice, working-class men and women moving into the growing cities were united more by class and divided less by sex. In the U.S. in the early nineteenth century, the ratio of black men to black women in prison was usually significantly lower than the ratio of white men to white women in prison. In mid-nineteenth-century England and Wales, social class may have had a similar effect on sex ratios in punishment.

A much larger, long-term increase in the ratio of men to women absent in punishment is readily apparent in England and Wales from 1920 to 1970. Absence in punishment in the 1920s almost exclusively meant persons in prison. About 1920, the ratio of men to women in prison was about 7. That’s about the same sex ratio among prisoners as existed among prisoners in England and Wales in 1780. From 1920 to 1970, the prisoner sex ratio rose from 7 to nearly 40 men in prison per woman in prison. The only interruption in this rise was the customary alternative disposal of men in military service during World War II.

The large, long-term rise in men in prison per woman in prison in Britain coincided with a major expansion of democratic deliberation. The Representation of the People Act of 1918 gave voting rights to five million men and nine million women, all previously without the right to vote. That act thus raised the enfranchised share of persons in Britain ages 21 and over from 28% to 76%. The Representation of the People Act of 1928 enfranchised an additional five million women.^ After the Act of 1928, all British subjects ages 21 and over who resided in Britain and met minimal property requirements had the right to vote. Overall, voting rights expanded from 28% of persons ages 21 and over in 1910 to 97% of those persons in 1929.

The expansion of the franchise in Britain occurred along with vigorous political debate and contention. From 1918 through the 1920s, the British people demobilized from total war and participated in intense union activity, including a General Strike in 1926. The Russian Revolution of 1917 intensified arguments over socialism and communism. A new national political party, the Labour Party, rose to control the British government in the 1920s. The Labour Party has had enduring political importance. Extensive new programs of print dissemination and the development of radio and film also intensified public debate.

The ratio of men to women in prison in England and Wales declined sharply from 1970 to 2003, but then increased from 2003 to 2010. A society-wide movement for gender equality may have contributed to reducing the ratio of men to women in prison from 39 to 15 from 1970 to 2003. The suicides of six female prisoners in the English prison at Styal between August, 2002 and August, 2003 prompted high-level political concern about women prisoners. Since then, the prisoner sex ratio has risen from 15.5 men in prison per woman in prison in 2003 to 19.0 in 2010. Both those figures are much higher than the maximum and median ratio of men to women in prison from 1780 to 1920: 8.0 men per woman maximum and 6.1 men per women median across 140 years of criminal justice history. Even with the decrease in gender inequality in punishment from 1970 to 2003, the pattern of punishment by sex is currently extraordinarily unequal.

Recent high-level reports have focused making the criminal justice system better serve women. The Corston Report, published in 2007, was a Home-Office commissioned “review of women with particular vulnerabilities in the criminal justice system.” The reviewer, Baroness Corston, declared:

My first recommendation concerns the treatment of men and women within the criminal justice system. From April 2007 the government will have a statutory duty to take positive action to eliminate gender discrimination and promote equality under the Equality Act. I have seen little evidence that much preparatory work is in hand in respect of the imminent statutory duty or of any real understanding that treating men and women the same results in inequality of outcome.

Inequality in outcomes from government action (imprisonment) that directly affects fundamental life capabilities (personal physical liberty) should be a priority concern in pursuing gender equality.

International development leaders seek to promote equal human capabilities.^ Concern for human capabilities suggests priority for reforming the criminal justice system to promote equal outcomes in imprisoning men and women. If that’s considered an unrealistic ideal of real gender equality, gender equality efforts might aim to reduce gender inequality in punishment to less gender inequality than existed from 1780 to 1920. A reasonable start would be to reduce anti-men gender bias in criminal sentencing.

Yet when the issue is imprisonment, equality doesn’t mean equality. Baroness Corston forthrightly declared:

Women and men are different. Equal treatment of men and women does not result in equal outcomes.

That’s never been a prominent consideration in comparing aggregate labor-market earnings by sex. But at the punishing end of public policy, women and men are regarded as essentially different. With seventeen men in prison per woman in prison in England and Wales, Baroness Corston proposed a separate and even more unequal criminal justice system. She proposed treating women “holistically and individually” so as to free up prison spaces for imprisoning men:

I have, however, concluded that it is timely to bring about a radical change in the way we treat women throughout the whole of the criminal justice system and this must include not just those who offend but also those at risk of offending. This will require a radical new approach, treating women both holistically and individually – a woman-centred approach. …
Equality does not mean treating everyone the same. The new gender equality duty means that men and women should be treated with equivalent respect, according to need. Equality must embrace not just fairness but also inclusivity. This will result in some different services and policies for men and women. …
An additional 8,000 places {in prison) for men are planned and a reported £1.5 billion is being sought to fund them. Unless the current sentencing trend can be reversed, more must follow. A much smaller level of funding would provide an opportunity for government to do something innovative for women. I do not pretend that my proposals will free up hundreds of prison places overnight. It will take time and determination and persistence but I do believe that, if my package of recommendations is implemented, over time the women’s prison population will decrease.^

Gender equality here becomes the abstract social construction of treating men and women “with equivalent respect, according to need.” It apparently implies imprisoning more men and imprisoning fewer women. Increasing the ratio of men to women in prison is also the implied direction of other major reports since 2003 on women offenders in England and Wales. Such policy is consistent with the increasing ratio of men to women in prison since 2003. Apart from power and persuasiveness in public discourse, such policy isn’t consistent with any reasonable understanding of gender equality and justified gender-equality priorities.

Punishment Sex Ratio in Scotland Since 1840

Mid-nineteenth-century Scotland had a relatively equal distribution of punishment across men and women. About 1.9 men were in prison per woman in prison. Less comprehensive statistics from earlier in the nineteenth-century show even more equal sex distribution among Scottish prisoners. In the Glasgow bridewell in 1818, there were 160 men and 160 women.^ In 1820, Glasgow’s tolbooth and bridewell together held 200 male criminals and 151 female criminals.^ The sex ratio among persons transported was even more nearly equal (1.4 men per women) than the prisoner sex ratio. In contrast, the death penalty was administered highly disproportionately to men (about 16 men executed per woman executed). Overall, the sex ratio of Scots absent in punishment from 1840 to 1874 has median by year of 2.3 men per woman.

Life-disposing punishment was applied less unequally in the nineteenth century in Scotland than in England and Wales. The sex ratio of persons executed in the nineteenth-century in total was 16 men per woman in Scotland and 20 men per woman in England and Wales. The sex ratio of persons transported was 2.6 men per woman for Scotland and 7.3 men per woman for England and Wales. The sex ratio of the average number of persons in prison from 1840 (the start of Scottish prison data) to 1899 was 2.3 men per woman for Scotland and 5.0 men per woman for England and Wales. Overall, the median sex ratio of Scottish persons absent in punishment from 1840 to 1899 was less than half that in England and Wales.

The relatively equal sex ratio in punishment in Scotland was a result of a Scottish criminal justice system that differed significantly from England and Wales’ criminal justice system. In contrast to English common law, Scottish law is primarily based on Roman law. Scottish law also encompasses elements of Old Norse law and Celtic law. Under the 1707 Act of Union that created the United Kingdom of Great Britain, Scotland retained its distinctive legal system. In addition, Scotland, particularly Edinburgh, was a center of eighteenth-century Enlightenment thought. Enlightenment thought earlier and more strongly influenced Scottish penal practice than it did penal practice in England and Wales.

In the eighteenth and nineteenth centuries, life-disposing punishment was less prevalent and more rationalized in Scotland than in England and Wales. Enlightenment thinkers generally condemned capital punishment as cruel, unreasonable, and contrary to the public interest.^ In the 1780s, Scotland executed only about 13% as many offenders on a per capita basis as did England and Wales. Across the nineteenth century, Scotland executed on a per capita basis only 53% as many persons as did England and Wales. Transportation had a similar pattern of less prevalent punishment in Scotland than in England and Wales across the eighteenth and nineteenth centuries. Robert Adam’s Edinburgh Bridewell, completed about 1795, was the first prison in Britain that followed Jeremy Bentham’s new rotunda model.^ Bentham was a leading penal rationalizer. Scotland centralized control over its prisons in 1839, more than twenty-five years before England and Wales did.

Despite these differences between the criminal justice systems in Scotland and England and Wales, Scotland paralleled England and Wales in a large increase in the prisoner sex ratio from 1920. By 1875, absence in punishment nearly exclusively consisted of persons in prison. From 1875 to 1920, the sex ratio of prisoners rose gradually, from 2.3 to 6.2 men per woman. The prisoner sex ratio then began rising rapidly. From 1965 to 1970, the prisoner sex ratio was about 33 men in prison per woman in prison. That sex ratio is more than ten times higher than the prisoner sex ratio in Scotland from 1840 to 1900.

From about 30 men in prison per woman in prison from 1960 to 2000, the prisoner sex ratio in Scotland declined rapidly. In 2010, based on the average daily number of prisoners, the prisoner sex ratio in Scotland was 17.1 men in prison per woman in prison. That’s rather different from the movement in the prisoner sex ratio in England and Wales. There the prisoner sex ratio began to decline from historic heights about 1970. In conjunction with influential penal policy advocacy, the prisoner sex ratio in England and Wales began rising in 2005 from 15 men per woman in prison. In 2010, the prisoner sex ratio in England and Wales, 19.0 men in prison per woman in prison, was higher than that in Scotland, 17.1 men in prison per woman in prison. Whether a policy-spurred trend toward more male-biased imprisonment re-emerges in Scotland remains to be seen. Just as in England and Wales, large sex inequality in imprisonment doesn’t seem to be an issue on the gender-equality agenda.

Prisoner Sex Ratio in Australia Since 1858

Like in England and Wales and in Scotland, the ratio of males to females in prison in Australia rose sharply across the mid-twentieth century. Reasonably good yearly statistics for the daily average number of prisoners in Australia are available from 1858. From 1858 to 1890, the ratio of male prisoners to female prisoners averaged 5.6. Australia during this period had 21% more male residents than female residents. Adjusting for this difference in the populations at risk of imprisonment, the ratio of male to females prisoners was 4.7. From 1890 to 1915, the prisoner sex ratio rose to 10. Australian punishment sex ratios, which become available in 1876, parallel these movements in the prisoner sex ratio. A sex ratio of 10 or less males per female in punishment was not unusual internationally prior to the era of intensified concern about gender equality.

The sex disparity in the Australian prison population began increasing sharply about 1920. The ratio of males to females in prison climbed to 33 by 1939. During WWII, that ratio dropped sharply. That’s consistent with a large body of historical evidence indicating that disposal of men in military service substitutes for disposal of men through punishment. From a trough of 13 in 1943, the ratio of male prisoners to female prisoner shot up, reaching 35 by 1960. The peak in the sex disparity in prisoners was about 1975, with an aggregate sex ratio of 42 across 1974 to 1976. Relative to available historical statistics for punishment, that was an extraordinarily high level of disproportionate punishment of men.

Only about three years after that peak in the ratio of men to women in prison, the Australia Institute of Criminology convened a seminar on “Women and Crime.” The introduction to a published volume of papers prepared especially for this seminar observed:

During the 1970s, interest in the study of crimes by women gathered momentum in several countries, a significant result of which was the publication of a number of books and articles on the subject. … While their fundamental impact was welcome in that a grossly and unjustifiably ignored area received the deserved attention, there were some unwelcome results as well; that is, several researchers used the opportunity to present a misconstrued relationship between various women’s movements and crimes by women. … The papers in this volume dispute with cold hard fact the substance of the claim that ‘women’s liberation’ is turning more women to crime.^

The opening article, by the volume’s co-editor, is entitled “Sexism in Criminal Law.”^ It provides a tendentious, misleading, and now conventional discussion of Blackstone’s famous statement and the law of coverture. The article discusses the crimes of domestic violence (imprison more men), rape (imprison more men), and infanticide (imprison fewer women). Men’s liberation, in the direct, factual sense of liberation from highly disproportionate imprisonment, was completely ignored in this volume’s emphasis on disputing myths about women’s liberation.

Despite subsequent, large changes in the prisoner sex ratio, subsequent scholarly literature on Australian prisoners naturalized prisoners as being “almost exclusively male.” The sex ratio among Australian prisoners declined from its peak of 42 men per woman in the mid-1970s to 15 men per woman in prison about 2000. Fifteen men per woman is nearly three times as high as the 5.6 men per woman sex ratio that prevailed on average in Australia from 1858 to 1890. With no definition or justification for “modern,” a 1999 study of sentencing reform and penal change in Australia declared that about 95% of prisoners being male is “modern levels” of sex inequality in punishment.^ Obscuring the historical variation in the sex ratio of prisoners, a 2000 study focusing on female prisoners in Australia began thus:

One of the most fundamental characteristics of incarcerated populations is that they are almost exclusively male. … The reason for this overwhelming preponderance of males over females in prison is fairly straightforward. Crime is predominately committed by young males. Male involvement in criminality of all types is much greater than female involvement, and this tendency is more pronounced for more serious forms of criminality.^

This “explanation” naturalizes men as criminals, obscures the historical variation in the sex ratio of prisoners, and justifies without reason stereotyping criminals as “almost exclusively male.” It ignores the reality that history and politics shape the social construction of criminal acts, the criminal seriousness of particular acts, and imprisonment sentences. With only some light paraphrasing, this introductory criminalization of men was copied into the introduction for a 2003 scholarly paper entitled, “Women in prison – Why is the rate of incarceration increasing?”^ That’s a peculiar scholarly concern in circumstances of 15 men in prison per woman in prison.

The criminological literature has worked to legitimize a highly unequal ratio of men in prison to women in prison. Focusing on females, a study found explanations for the ratio of males to females in prison increasing from 5 to 10 from about 1885 to 1915, but found puzzling the ratio of males to females in prison decreasing from 40 to 17 from about 1975 to 1997:

It is fairly clear that the reduction in female imprisonment in the late 19th and early 20th century was in large part the result of changes in the way that good order offences (especially drunkenness, vagrancy and prostitution) were dealt with. … The increase in female imprisonment since the 1970s has proven rather more difficult to account for.^

The Sentencing Advisory Council of the Australian state of Victoria published in 2010 a study, Gender Differences in Sentencing Outcomes. That study forthrightly recognized inequalities in sentencing:

In the Magistrates’ Court of Victoria, women are less likely to receive a sentence of imprisonment for all of the offence categories examined. For most of the offence types, women receive a shorter term of imprisonment than do men.

The study then justified categorical sex differences with “a constellation of factors that can validly reduce the length of a sentence.” What are the stars in this constellation? One wondrous star unblinkingly reflects effects of historical sex disparity in sentencing, blurred with vague references to seriousness of offenses and “more complex personal histories and situations”:

Women prisoners have less serious criminal histories than do men, with fewer prior convictions and less serious previous and current offending in terms of the type of offences for which they have been imprisoned. They also tend to be sent to prison for shorter periods, likely a reflection of their less serious offending and their more complex personal histories and situations.

The claim that men have less complex personal histories and situations than women do sheds light not on the reality of men’s lives, but on lack of social interest in men’s lives. The reflective reference to shorter sentences leads to women’s more complex personal histories and situations. That leads to extensively discussed, emotive factors:

The biographies of female offenders vary systematically from those of men. Contributing to their blurred status as both victims and offenders, women are more likely than men to have a history of factors that are often causally interrelated, such as mental illness, physical or sexual victimization in childhood or early adulthood, and substance abuse. Women are also more likely than men to have primary caregiver status.

Women being more likely to have primary caregiver status, which does not justify employment discrimination, here is used to justify discriminatory sentencing to prison. The Executive Summary for this study, which was issued by the Sentencing Advisory Council of the Australian state of Victoria, ends with a filigreed justification for sex inequality in imprisonment:

The effect of gender on sentencing is not direct, but travels via two distinct paths: via gender differences in offending behaviour; and via the individual biographies of women that see a greater proportion of women coming before the court with a constellation of characteristics that creates legitimate mitigating circumstances. It is these mitigating factors that lead to disparities in sentencing outcomes for men and women in the criminal courts, disparities that appear warranted and that are not immediately indicative of any pervasive ‘bias’.
Thus the disparities seen in sentencing outcomes for men and women are a reflection of legitimate but gender-linked characteristics: differences are evident because of factors associated with being female, not because of gender per se.^

If there were 3 times as many men as women in prison, or 5 times as many men, one might consider this analysis seriously. But the historical reality is that in that late 1970s, there were about 30 times as many men as women in Australia prisons. Australian scholars then began expressing concern about the number of women in prison. In 2011, Australian prisons held 13 times as many men as women. Scholars straining to justify that level of sex inequality for prisoners in societies formally committed to gender equality indicates not just pervasive anti-men sex bias in punishment. Such scholarly work also indicates pervasive anti-men sex bias in public deliberation.

Sex Ratio of Prisoners in U.S. Since 1880

Two features make the U.S. history of punishment exceptional. First, the U.S. massively increased its prevalence of imprisonment starting from about 1980. Second, the U.S. had a relatively high concentration of punishment on men in the nineteenth century and did not greatly increase its concentration of punishment of men in the two decades after World War II. The post-1980 massive increase in imprisonment is a well-recognized feature of U.S. punishment history. In contrast, the early, exceptionally high U.S. ratio of men to women in prison and the relative stability of the U.S. prisoner sex ratio is hardly known.

Early in the nineteenth century, the ratio of men to women in prisons and penitentiaries was much higher in the U.S. than in European countries. Official French visitors to U.S. state prisons about 1832 noted:

the only certain, incontestable fact which we have remarked in the United States, and which may offer an opportunity for comparison, is the peculiar and extraordinary morality of the women belonging to the white race. Out of one hundred prisoners in the United States, we find but four women, whilst with us there are twenty in a hundred.^

An English official who visited U.S. state prisons about a year later offered a similar observation:

Few circumstances connected with this subject impress a visitor more forcibly than the small number of females to be found in the penitentiaries of the United States. …there prevails a strong indisposition to prosecute, especially if the offender be not a woman of colour. Magistrates are also reluctant to commit women from the circumstances of there not being any suitable prisons for their reception.^

A leading U.S. scholar of crime and punishment in 1833 concurred with these views:

In all countries women commit less crimes than men, but in none is the disproportion of criminals of the two sexes so great as in ours.^

Specific data indicate the U.S. had a much higher ratio of males to females in prison than did other countries. About 1837, the Belgian government sought to construct at Liege a penitentiary capable of holding 300 prisoners – 240 males and 60 females.^ In Britain, the Penitentiary Act of 1779 (19 Geo. 3, c. 74) directed the construction of two national penitentiaries, one for 600 male convicts and one for 300 female convicts. The national penitentiary at Millbank, London, was intended in 1818 to hold 600 males and 400 females.^ In 1840, Millbank held two male prisoners per female prisoner. In contrast, U.S. state prisons in 1840 held twenty-two male prisoners per female prisoners.

The relatively high ratio of males to females in early nineteenth-century U.S. prisons wasn’t an effect of the distribution of life-disposing punishment. In the U.S. from 1825 to 1845, the ratio of males to females executed was 14.7. That’s close to the execution sex ratio of 17.3 in England and Wales during that period. England and Wales during this period was also transporting convicts to Australia. The sex ratio among the transported convicts was 7.4 males to females. Many more persons were transported than were executed, hence substitution of these punishments for imprisonment would in aggregate raise the ratio of men to women in prison. Nonetheless, the ratio of men to women in prison was much lower in England and Wales than it was in the U.S.

Punishment in the U.S. focused increasingly on men through to the mid-twentieth century. Good national statistics for punishment by sex are available for the U.S. only from 1880. From 1880, 97% or more of the persons absent in punishment in the U.S. were in prison. Hence absence in punishment is nearly equivalent to persons in prison. The sex ratio among U.S. persons absent in punishment was about 11 in 1880. That was about twice as high as the sex ratio in punishment in England and Wales at that time. The U.S. sex ratio in punishment rose steadily from 1880 to 22 males per female in punishment in 1933. Execution, while accounting for only a small share of absence in punishment, also focused more extensively on men. The U.S. from 1846 to 2010 executed 90 males per female, a much higher sex ratio than the corresponding statistic of 25 for England and Wales.

The exceptional history of the U.S. sex distribution of punishment may be related to the exceptional U.S. history of communication. Compared to European countries, the United States experienced relatively early development of deliberative democracy.^ Freedom of the press, freedom of speech, and freedom of religion were central to the American colonists’ struggle against the British government and to their relations with each other. The First Amendment to the U.S. Constitution explicitly expressed those values. Moreover, the U.S. Declaration of Independence asserted as self-evident truths that all men are created equal and that governments derive their just powers from the consent of the governed. These principles spurred democratic deliberation. The federal structure of the U.S. government, with non-enumerated powers reserved to the states, allowed considerable political experimentation and political diversity as well as tension between state and national governments. In addition, plentiful land and good opportunities to form new settlements gave persons opportunities to exit from communities where entrenched institutions and established social hierarchies made deliberation unprofitable. The early American republic thus provided propitious circumstances for personally significant discussions of common life. Those discussions included extensive discussions about punishment.

The U.S. electoral franchise was broader before World War I and expanded less in the mid-twentieth century than did the electoral franchise elsewhere. In 1790, about a quarter of adults in the U.S. had the right to vote. In England and Wales, the corresponding figure was 8%. Across the first half of the nineteenth century, U.S. states largely eliminated property and tax-paying qualifications for voting.^ By 1919, 46% of persons in the U.S. ages 21 and over had full voting rights. This figure includes women who lived in states that had fully enfranchised woman (27% of women), but excludes non-citizens (11% of adults) and blacks living in the south (7% of adults). The Nineteenth Amendment to the U.S. Constitution (1920) increased the fully enfranchised share of adults to 82%. These figures do not subtract disenfranchised felons, overwhelmingly male, who increased to 2.5% of adults by 2004. The mid-twentieth-century increase in the U.S. electoral franchise was significantly smaller than the increase from 28% to 97% for the British electoral franchise from 1918 to 1929.

The exceptional history of U.S. imprisonment is associated with an exceptional history of communication and public deliberation. The U.S. was an early-nineteenth-century leader in the development of communication media and democracy. The U.S. also led the early nineteenth-century practice of attempting to suppress completely prisoners’ communication. The U.S. had an exceptionally high ratio of men to women in prison early in the nineteenth century. Sharp increases in the electoral franchise in other countries in the mid-twentieth century are associated with large increases in the ratio of men to women in prison in those countries. Democratic development that does not increase the engagement of prisoners in communication with non-prisoners seems to skew the sex composition of prisoners toward men.

Sex Ratio of Prisoners Across the World: Taking Gender Seriously

Most prisoners are men. Prisoners worldwide numbered 10.1 million persons on a given day about the year 2010. Among those prisoners, about fifteen men were in prison for every woman in prison.

The sex ratio of prisoners is significant. Punishment is a fundamental political responsibility. Prisoners are among the poorest of the poor in human welfare. Prisoners should be an important public concern. Common human groups such as families and communities usually have a roughly equal sex composition. The sex ratio of prisoners is extremely far from equal.

The sex ratio of prisoners isn’t immutable biological destiny. Across 45 international jurisdictions for which data are available and covering 22% of the world’s 2010 population, the aggregate sex ratio of prisoners fell from 20 men prisoners per woman prisoner to 13 men prisoners per woman prisoner from 1977 to 2010. Data are available from 2003 and 2010 for 177 jurisdictions covering about 94% of the world’s population. From 2003 to 2010, the aggregate prisoner sex ratio fell from 15.7 men in prison per woman in prison to 14.5 men per woman.

Considerable variation exists in the sex ratio of prisoners across the world. Among jurisdictions holding more than 1000 prisoners about 2010, Pakistan had the highest sex ratio: 82 men in prison per woman in prison. India, closely related to Pakistan geographically and historically, had only 23 men in prison per woman in prison. Taiwan was among the jurisdictions closest to sex equality: it had only 4.6 men per woman imprisoned. South Korea, in contrast, had 17.5 men in prison per woman in prison. Most persons would readily notice such sex-ratio differences in ordinary human groups.

The sex ratio of prisoners is relatively high in Sub-Saharan Africa and in low-income countries. Sub-Saharan Africa in aggregate had 30 men in prison per woman in prison about 2010. North America, East Asia, and Pacific had in aggregate 12 men in prisoner per woman in prison. These regional differences correlate with income differences. Low-income jurisdictions had in aggregate 21 men per woman in prison, while the corresponding figure for high-income countries is 12. Even controlling for income group, Sub-Saharan Africa had on average a 55% higher prisoner sex ratio than did East Asia and the Pacific.

Large international differences in the sex ratio among prisoners are not easy to explain. Across 209 international jurisdictions about 2010, the median prisoner sex ratio was 21.2 men in prison per woman in prison. Among those jurisdictions, 25% had a sex ratio less than 14.2, while 25% had a sex ratio greater than 34.7. Indicators for region and income group together explain only 20% of the variation in the logarithm of the sex ratio. Adding to those indicators the urban population share, the labor force participation shares of men and women, and the Human Development Index raises the explained variance only to 34%.

Human biological sex differences, gender, and the socio-economic environment allow large variation in prisoner sex ratios. Considerable variation in prisoner sex ratios exists internationally as well as within particular jurisdictions. High ratios of men in prison per woman in prison are a social justice failure.

Prisoners Ignored in International Development Policy

Work on human development has emphasized the capabilities that every human being, male and female, should have. An influential study on development has eloquently declared:

Human beings are the real ends of all activities, and development must be centered on enhancing their achievements, freedoms, and capabilities. It is the lives they lead that is of intrinsic importance, not the commodities or incomes they happen to possess.^

Persons in prison often have little choice in the food they eat and the clothes they wear. They typically do not have the freedom to move beyond the confines of a single building, or even within that building. Their opportunities to meet other persons and to learn about the world are highly constrained. With respect to human capabilities, prisoners are among the poorest of the poor.

The United Nations’ Human Development Index (HDI) is a leading quantitative measure of human development. The HDI measures three dimensions of human development:

leading a long and healthy life, measured by life expectancy at birth; being knowledgeable, measured by literacy and school enrollment; and having a decent standard of living, measured by GDP per capita ^

These measures of human development indicate persons’ capabilities to exercise freedom. As a United Nations Human Development Report has emphasized, “Human development is about freedom.”^ Being confined in prison highly constrains persons’ freedom.

The Human Development Index correlates positively with the prevalence of imprisonment across jurisdictions worldwide. Across 182 jurisdictions covering 98% of the world’s population, the HDI for 2011 has quartiles 0.52, 0.70, and 0.80. A 0.10 increase in the HDI is associated with a 21% increase in the prevalence of imprisonment. The HDI is even more strongly correlated with the prevalence of imprisonment controlling for region and income-group fixed (level) effects on imprisonment prevalence. Imprisonment prevalence might increase human development of the non-imprisoned population (most of the population) through increases in their personal security. More realistically, states with strong public administrative capacity are likely both to foster better human development outside of prison and to hold a larger number of persons in prison.

Despite prisoners’ significance to ideals of human development, prisoners are largely invisible in international development policy. The human development of prisoners is much more directly related to public policies than human development of persons generally. The HDI’s positive correlation with imprisonment prevalence deserves careful consideration. Acknowledging significant human trade-offs in human development is a step toward making those trade-offs as just and humane as possible.

Gender Inequality Index Relative to the Prisoner Sex Ratio

Countries with a higher level of gender inequality tend to have more inequality in the ratio of men in prison to women in prison. The United Nations’ Gender Inequality Index (GII) compares male and female parliamentary representation, labor force participation, and attainment of secondary and higher education. It also includes measures of maternal mortality and adolescent female fertility relative to particular reference norms. The GII, like development policy generally, doesn’t consider male and female imprisonment. It doesn’t recognize large, worldwide inequality in the ratio of men in prison per women in prison.

The United Nations’ Gender Inequality Index correlates positively but imperfectly with inequality in the prisoner sex ratio. Moving from the 2011 median country-value for the GII to the 75th percentile is associated with a 15% increase in the ratio of males to females in prison. The prisoner sex ratio, however, is 56% greater at its 75th percentile relative to its median value. The GII and the prisoner sex ratio are different but correlated measures of gender inequality.

Gender inequality in the sex ratio of prisoners concerns the most directly and personally punishing outcomes of public policy. The GII, in contrast, encompasses elite statuses: the sex ratio in parliamentary representation and in degrees received in higher education. The prisoner sex ratio is a more appropriate focus for welfare concern about gender inequality in human development.

The Gender Inequality Index has more uncertainty in its measurement and is more complicated and difficult to understand than is the sex ratio of prisoners. Measuring labor force participation requires a country-wide survey. Measuring maternal mortality and adolescent female fertility requires compiling nation-wide birth and death data with record-specific information on sex, age, and cause of death. Measuring the level of formal education in the population requires a long-form national survey. Understanding the actual calculation of the GII requires the ability to parse complicated mathematical formulas and some understanding of index-number theory. A specific value of the GII has no specific meaning to most persons. The sex ratio of prisoners, in contrast, requires collecting information only from political bodies authorized to imprison persons. The prisoner sex ratio is simple to calculate and easy to understand.

Just as for inequality in economic growth generally, the weight of welfare concern about gender inequality across population groups is a matter of justice. More concern for the parliamentary sex ratio than for the prisoner sex ratio indicates lack of concern for the poor.

Communicating with Prisoners

Public Interest Analysis