SMART ON CRIME
 
A CITIZEN LOBBY
 
CALDER SQUARE
P. O. BOX 11111
STATE COLLEGE, PA 16805–1111
Telephone: 814–238–8054
Fax: 814–234–4317
E-Mail: jph13@psu.edu
 
 

Title: Crime and Punishment in the United States
 Author: Julian Heicklen
 Date: February 1997
SOC Publication Number: 001

 
 
 
CRIME AND PUNISHMENT IN THE UNITED STATES
 
by Julian Heicklen
February 1997
 
Abstract

Violent crime rates have not changed since 1973 (except for murder and non-negligent manslaughter). For murder and non-negligent manslaughter, the rate more than doubled from 1963 through 1980. Since then, there has been a 23% drop through 1996. Rates for property crime dropped 37% from 1975 through 1992.

Arrest rates have risen dramatically since 1973. This increase, coupled with increased rates of conviction, more and longer prison sentences, delay for parole, and more resentencing for technical parole violations, has led to a burgeoning prison population.

The correctional population has risen dramatically from 1980 to the present, the population in 1994 being 2.94 that of the population in 1980. At the end of 1995, 2.8% of all United States adults were under correctional supervision.

The per capita prison population almost tripled from 1980 through 1995. Currently it is 6.15/1000 residents. The inmate population is disproportionately male and Black.

The cost of the criminal justice system was $93.8 billion in 1992. The real per capita increase (adjusted for inflation) was 65% from 1982.

It is recommended that efforts should be made to reduce the per capita prison population by about 80%. In order to effect this large reduction, two things must be done:

1. Many crimes should be decriminalized.

2. For many crimes, alternative punishments that are cheaper than incarceration should be instituted.
 

There has been a general perception in the country that crime rates have been increasing for several years. The Council on Crime in America (Bell and Bennett, 1996) has issued a report that supports this view. This has induced a "tough on crime" attitude among the general population and politicians. As a result, prison sentences are becoming more common and longer.

On the other hand the National Criminal Justice Commission has issued a report (Donziger, 1996) that contends that crime is not increasing, but that convictions and more and longer prison sentences are increasing.

This report is an attempt to analyze what really is happening regarding criminal behavior. We shall examine crime rates, punishments, incarceration, and costs.
 
 

1. Crime Rates

The yearly rate of violent crimes per 1000 adults in the United States is summarized in Figure 1. For this figure, an adult is any person aged 12 or older. Violent crime includes murder and non-negligent manslaughter, forcible rape, robbery, and simple and aggravated assault.

The data labeled "Victims" show the victimization rates given by Bastian (1994) as corrected by the factors given by Perkins et. al. (1996). These data do not include murder and manslaughter, because the victims were dead and could not be interviewed. However the murder rate in the United States has been between 0.079/1000 adults and 0.101/1000 adults. Thus its omission does not significantly affect the total victimization rate, which is more than 300 times as large. It can be seen that the yearly victimization rate for violent crime has remained nearly constant at about (42–53)/1000 adults per year from 1973–1995.

In contrast the yearly rate of violent crime reported by law enforcement agencies (Maguire and Pastore, 1996) in the Uniform Crime Reports (UCR) has risen steadily from 1.58/1000 adults in 1961 to a peak of 7.58/1000 adults in 1991. Since then, it has dropped slightly to 7.16/1000 adults in 1994. This dramatic increase until 1991 is what has caused the public concern over "rising crime rates."

In actuality, the overall violent crime rate has not increased. The increase in the UCR violent crime rate can be analyzed by looking at the various categories of violent crime.

The yearly reported murder and non-negligent manslaughter rate has more than doubled from 0.046/1000 in 1962–1963 to a high of 0.102/1000 adults in 1980. Since then, it has fallen to 0.079/1000 adults in 1984–1985, then rose to 9.8 in 1991, and has fallen again to 0.079 in 1995. Because almost all murders and manslaughters are reported, these numbers reflect the actual rates.

Forcible rate reports from the UCR were 4.55 times as high in the peak year of 1992 (0.428/1000 adults) as in 1961–1963 (0.094/1000 adults). The yearly reported rate of aggravated assaults from the UCR rose from 0.857/1000 adults in 1961 to a peak of 4.418/1000 adults in 1992. This is a factor of 5.16. The yearly reported rate of robbery from the UCR rose from 0.583/1000 adults in 1961 to a peak of 2.727/1000 adults in 1991. This is a factor of 4.68.

The increase in the UCR rates over the past many years has led to the perception that crime is rising, when in fact, it is not. For example, in 1973 citizens reported 861,000 aggravated assaults to police, but the UCR listed only 421,000. By 1988, citizens reported 940,000 aggravated assaults to the police, and the UCR listed 910,000 (Donziger, 1996). The same pattern occurred for robbery and rape. Donziger (1996) attributes the increase in UCR rates to better reporting methods because of the use of computers. More reports, especially from rural communities, reach the national data banks.
 

Figure 1: Yearly Violent Crime Rates per 1000 Adults Aged 12 Years or Over. Violent Crime includes murder & non-negligent manslaughter, forcible rape, robbery, aggravated assault, and simple assault. In 1993, sexual assault was added, but this has been omitted from the data in the figure. The data for victims does not include murder & non-negligent manslaughter, because these victims could not be interviewed. However the annual rate for this category is less than 0.1 per 1000, so the effect on the total is negligible. The data for victims from 1973–1992 comes from Bastian (1994) pp. 9 and 39 corrected by the factors for individual crimes given in Appendix Table 1 on page 152 of Perkins et. al. (1996). Data from 1993–1995 are from Perkins and Klaus (1996) and Smith (1996). Victimizations reported by the victims are presented with simple assaults included [Victims (total)] and omitted [Victims – Simple Assaults], because simple assaults are not included in the Uniform Crime Reports or in Arrests. Data for the Unified Crime Reports and Arrests come from Maguire and Pastore (1996) pp. 324 and 395, respectively.
 

The yearly rate of arrests for violent crimes also has increased over the years (Maguire and Pastore, 1996) along with the increase in yearly UCR violent crime rates. The yearly rate of arrest for violent crimes rose from 1.76/1000 adults in 1971 to 3.11/1000 adults in 1994.

Perhaps what is most surprising (and alarming) about these data is the small percentage of the crimes that are reported in the UCR. Even though the gap has been narrowing in recent years, there were still 2.52 as many victimizations as UCR of violent crimes in 1992. Coincidently the ratio of UCR to arrests for violent crime also was 2.52.

The yearly rate of property crime per 1000 adults or 1000 households (for household crime) in the United States is summarized in Figure 2. An adult is any person aged 12 or older. Property crime includes, burglary, larceny–theft, household theft, and motor-vehicle theft. The UCR data given in the figure do not include household theft.

 

Figure 2: Yearly Property Crime Rates per 1000 Adults Aged 12 Years or Over or per 1000 Households (for Household Thefts). Property Crime Rates include burglary, personal theft, household theft, and motor-vehicle theft. The data for victims from 1973–1992 comes from Bastian (1994) pp. 9 and 39 corrected by the factors for individual crimes given in Appendix Table 1 on page 152 of Perkins et. al. (1996). Data from 1993–1995 are from Perkins and Klaus (1996) and Smith (1996). Victimizations reported by the victims are presented with household theft included [Victims (total)] and omitted [Victims – Household Theft], because household thefts are not included in the Uniform Crime Reports or in Arrests. Data for the Unified Crime Reports and Arrests come from Maguire and Pastore (1996) p. 324. They include commercial burglary (as well as household) burglary, whereas the data reported by victims does not include commercial burglary. Commercial burglary accounts for about 8% of total property crimes (Rand, 1997).
 

The data labeled "Victims" gives the victimization rates listed by Maguire and Pastore (1996) as corrected by the factors listed in Perkins et. al. (1996). It can be seen that the yearly victimization rate for property crime has decreased from a value of 410/1000 adults in 1975 to a value of 258/1000 adults in 1992.

In contrast the yearly rate of property crime reported in the UCR has risen steadily from 17.3/1000 adults in 1960 to a peak of 51.4/1000 adults in 1991. Since then, it has dropped slightly to 46.6/1000 adults in 1n 1994. Again this dramatic increase in reported crimes can be attributed to the increased efficiency of reporting methods.

The yearly rate of arrests for property crime also has increased over the years (Maguire and Pastore, 1996) along with the increase in yearly reported property crime rates. The yearly rate of arrest for property crime rose from 7.2/1000 adults in 1971 to a peak of 9.1/1000 adults in 1990. Since then, it has dropped to 8.4/1000 adults in 1994. In 1994 the ratio of UCR property crime to arrest was 5.56 in contrast to the value of 2.52 for violent crimes.

What has caused the drop in property crime in recent years (and violent crime as well in 1995)? Many attribute it to increased law enforcement, i. e. the large increase in police personnel in recent years. Lungren (1996), the Attorney General of California, attributes it to the "3 strikes and you are in" law adopted by California in 1994. He points out that in 1995, overall crime rates dropped 8.5%; violent crime rate, 5.5%; and property crime rate, 10.1% from 1994. For the United States as a whole, the overall crime rate dropped 6.8%, violent crime dropped 10.0%, and property crime dropped 6.3% (Smith, 1996). Most of the states do not have "3 strikes and you are in" laws. They had a larger drop in violent crime, but a smaller drop in property crime, than California. Drawing inferences from any of this data probably is meaningless. It is too soon to draw any conclusions about the effect of the "3 strikes and you are in" laws.

It is interesting to compare the percentage of crimes reported to the police by victims and those reported in the UCR. This is done in Table 1 for recent years. Only about 42% of violent crime and 34% of property crime are reported to the police. For violent crime, the UCR rates are 90–94% of those reported by victims to the police for the years 1992-1994. The ratio for property crimes was 94% in 1992. However, the UCR data include commercial burglary, whereas the reports by victims do not. When this is taken into account, the UCR data for property crimes are about 86% of those from victims for 1992.

 

Table 1: Percent of Crimes Reported to the Police
Reported By:
1992
1993
1994
1995
Violent Crimes
Victims1
42.3
41.6
41.6
43.2
Unified Crime Reports
39.9
37.5
37.3
Property Crimes
Victims1
33.8
33.1
33.9
34.2
Unified Crime Reports
31.7
 1From Bastian (1995), Perkins and Klaus (1996), and Smith (1996).
 
2. Punishment

After a person is convicted of a crime, he becomes part of the correctional population. He can be sentenced to probation, county jail, state prison or federal prison. After some time in prison, he may be placed on parole. The historical trends for the number of adults serving each of these punishments is given in Figure 3. Technically, some of the county facilities are prisons rather than jails (they are supervised by wardens, rather than by sheriffs). However all of the county facilities are grouped under jails.
 

Figure 3: Number of Adults Aged 18 or Older under Supervision of the Criminal Justice System. Data through 1994 from Maguire and Pastore (1996) p. 540. Data for 1995 from Bureau of justice Statistics (1996) and Gilliard and Beck (1996).
 

The correctional population has risen dramatically from 1980 to the present, the population in 1994 being 2.79 that of the population in 1980 (Maguire and Pastore, 1996). During the same time period, the populations for probation, jail, prison, and parole have become factors of 2.65, 2.65, 3.10, and 3.13, respectively. At the end of 1995, 2.8% of all United States adults were under correctional supervision (Bureau of Justice Statistics (1996).

The estimated percentage of the appropriate adult population under all forms of correctional supervision is given by race and sex in Figure 4. The offenders are predominantly male. Females accounted for only 13.5 percent of the total adult correctional population in 1985 and 14.9 percent in 1994.

There also is a great disparity in the population by race. By 1994, 1.9% of White and 9.1% of Black adults were under the supervision of the correctional justice system. Most of these are males. The percentage of adult White males was 3.2 under correctional supervision. Fully 15.5% of adult Black males were under correctional supervision in 1994.

Correctional supervision is supposed to be for individuals with behavior that deviates criminally from the norm. Is it reasonable that 15.5% of a given population currently are criminal deviants? Perhaps this percentage reflects the imposition of the views of the majority culture on those of a minority culture. Or is it just plain racism? Since more criminals, present and past, probably are not under the jurisdiction of the correctional system, there may be a majority of adult Black males that fall into the category of criminal deviants. Of course, this is a contradiction in terms. If they are the majority, then they all cannot be deviants from the norm.

Criminals of races other than Black or White have less than 1% of their adult populations under correctional supervision, making them the most law-abiding group (or at least the most successful in avoiding prosecution and conviction).
 

Figure 4: Estimated Percent of Adults age 18 or Older under Supervision of the Criminal Justice System by Race and Sex. Data from Maguire and Pastore (1996) p. 540.
 
 
3. Incarceration

The total number of adults in United States prisons and jails rose dramatically from 501,886 in 1980 to 1,475,329 in 1994, a factor of 2.94 (Maguire and Pastore, 1996). In 1994, federal, state, and local facilities accounted, respectively, for 5.85, 61.4, and 32.8 percent of the total. The results are summarized in Figure 5.

 

Figure 5: Number of Adults aged 18 or older in Prison or Jail. Data through 1994 from Maguire and Pastore (1996) p. 548. Data for 1995 from Gilliard and Beck (1996).
 
 

The per capita population incarcerated by group is given in Figure 6 by sex and race. The fraction for the total population rose from 221/100,000 in 1980 to 615/100,000 in 1996. This is the highest incarceration fraction in the world. Russia is second with 590/100,000. Most western European countries have less than 100/100,000 incarcerated.

In 1994, the per capita population incarcerated for each group is listed in Table 2. Males are much more likely than females to be incarcerated. For Whites, females comprise 6.5% of the incarcerated population; for Blacks, 8.4%. A Black male was 7.85 times as likely than his White counterpart to be incarcerated in 1994. A Black female was 7.25 times as likely than her White counterpart to be incarcerated in 1994.

 

Table 2: Number Incarcerated per 100,000 Adults in each Group in 1994
White Males
White Females
Black Males
Black Females
Total Population
860
60
6,753
435
565
 
  
Figure 6: Fraction of Adults Aged 18 or Older in Prison or Jail by Race and Sex. Data from Maguire and Pastore (1996) p. 548.
  The fraction of people held in federal and state prisons (excluding county facilities) is shown in Figure 7 for the years 1960 through 1995. From 1961 through 1972, the state and federal prison population dropped from 119/100,000 to 93/100,000 resident population (not just adults). These figures are comparable to those in western European countries. However starting in 1973, there has been an ever accelerating increase in the prison population, reaching 389/100,000 resident population in 1994. In 1994, 7% of all males of age 18 or older entered prison or jail (Donziger,1996); for Black males of age 18 or older, it was 25%. From 1973 through 1992 there has been no change in the violent crime rate, and a drop of 37% in the property crime rate.

What has led to this enormous increase in incarceration? Several factors are involved:
 

  1. There has been an increasing rate of UCR, as opposed to crimes committed. This has been accompanied by an increase in arrest and conviction rates. In Pennsylvania the percentage of those arrested that have been incarcerated approximately doubled from 1985 to 1994 (Renninger, 1996).
  2. Sentences are becoming longer. This will continue to be true in the immediate future as states adopt "3 strikes and you are in" laws. In Pennsylvania, the average sentence in state prison increased from 32 months in 1984 to 44 months in 1994, even though the percentage of those incarcerated for violent crimes dropped from 55.7% to 46.3% from 1985 to 1994 (Renninger, 1996).
  3. Parole has been delayed, and more parolees are being returned to prison for technical parole violations.
 
Figure 7: Sentenced Fraction of Resident Population (including juveniles) in Federal and State Prisons. Data through 1994 from Maguire and Pastore (1996) p. 556. Data from 1995 from Gilliard and Beck (1996).
 
 

However there even is a more important phenomenon involved. This is the enormous increase in prison sentences for non-violent narcotics violations. According to the Bureau of Justice Statistics (1995) "Drug offenders accounted for 61% of sentenced inmates in federal prisons in 1993, up from 25% in 1980. The proportion of drug offenders in state prisons increased from 9% in 1986 to 21% in 1991. The proportion of drug offenders in local jails increased from 9% in 1983 to 23% in 1989."

In 1994, 32.8% of new admittances (excluding parole violators) to Pennsylvania state correctional institutions were for non-violent narcotics violations (Renninger, 1996). In 1995, in Pennsylvania state correctional institutions, only 32.6% of new inmates were sentenced for violent crimes compared to 41.2% in 1980 (Horn, 1996). This shift toward housing more non-violent offenders is attributable to a 900% increase in commitments for narcotic drug violations—from 202 in 1980 to 2020 in 1994 in Pennsylvania (Horn, 1996).

We can expect even further increases in the prison populations as the states are passing "Three Strikes and You are In" laws and "Zero-Tolerance" alcohol levels for drivers under 21 years of age. Also the nation is "gearing up" for "war" on domestic violence. If the abortion and assisted suicide laws are enforced, the increase in prison admittances will become truly staggering.

If one just looks at the data, one might conclude that incarceration has no effect on violent crime rates, and only a modest effect on property crime rates. However, there are other factors involved. The family structure changed dramatically in the United states between 1973 and 1992. Single-parent families and illegitimate births rose dramatically. Both of these factors might tend to increase crime rates.

Levitt (1996) performed a statistical study in which he attempted to eliminate these other factors when estimating the effect of incarceration on crime rates. He found that, on average, a convict had committed 11 crimes in the year immediately preceding his incarceration. Thus if he was incarcerated, 11 such crimes should not be committed each year of his incarceration. In addition his analysis showed that an additional 4 crimes would not be committed because of the deterrent effect of incarceration. On average, each year's incarceration should prevent 15 crimes.

Levitt's analysis may be valid for some crimes, such as burglary or theft. However for other crimes, such as narcotics sales, it is likely that if one dealer is incarcerated, he just is replaced by another dealer. If so, incarceration has no effect at all on the drug trade.

 

4. Cost

 

Figure 8: Annual cost of Criminal Justice in the United States (Federal + State + Local). Data from Maguire and Pastore (1996) pp. 2 and 3.
 
 

What does the criminal justice system cost the taxpayer? The costs are summarized in Figure 8. The total annual cost has risen from $35.8 billion in 1982 to $93.8 billion in 1992, a factor of 2.62. Over the same period the U. S. population increased by 10.5% and the Consumer Price Index rose 44% (Bureau of Labor Statistics, 1997). When corrected for the population growth and inflation, the real per capita cost increased by 65%. This cost is split among the police, the judiciary, and corrections. The major cost is for police. This cost was 2.17 times as much in 1992 as in 1982 (Maguire and Pastore, 1996). The cost for corrections has risen from $9,048,947 in 1982 to $31,461,433 in 1992, a factor of 3.48.

On a per capita basis, the cost of prison was $26,900 for 1995 for federal prison, if the cost of prison construction is not included. With the cost of prison construction included, it was $30,850 for 1995. The per capita cost is about $24,000 per year for state prisons, and about $17,000 per year for county jails. Is it cost effective to incarcerate prisoners?

To determine the cost effectiveness, let us determine the saving resulting from reduced crime because of incarceration. Assume that Levitt's (1996) analysis is correct, and that 15 crimes per year are prevented for each year an inmate is in prison. For crimes of theft (robbery, burglary, larceny-theft, and motor-vehicle theft), this is likely to be true. Then the break-even cost for sending a person to federal prison is $2,057 for a crime involving theft. For county jail, the break-even cost is about $1,133. Incarcerating people for crimes of theft involving smaller amounts of money is a financially losing proposition. For such crimes, cheaper alternative punishments should be imposed.
 

5. Conclusions

 From the information presented here, we can conclude the following:

  1. Violent crime rates have not changed since 1973, except for murder and non-negligent manslaughter. For murder and non-negligent manslaughter, the rate more than doubled from 1963 through 1980. Since then there has been a 23% drop through 1996. Rates for property crime dropped 37% from 1975 through 1992.
  2. Arrest rates have risen dramatically since 1973. This increase, coupled with increased rates of conviction, more and longer prison sentences, the delay for parole, and more resentencing for technical parole violations, has led to a burgeoning prison population.
  3. The per capita prison population almost tripled from 1980 through 1995. The inmate population is disproportionately male and Black.
  4. The real per capita cost of the criminal justice system (adjusted for inflation) in 1992 was 1.65 that of 1982.
  5.  
6. Recommendations

The per capita prison population in western Europe is less than 1 per 1000. In the United states prior to 1973 it was slightly above 1/1000. Currently it is 6.15 per 1000. Efforts should be made to return the per capita prison population to close to 1/1000. This means a reduction of the prison population by about 80%.

In order to effect this large reduction, two things must be done:

  1. Many crimes should be decriminalized. This may include all victimless crimes. By itself, the decriminalization of narcotics use and sale would reduce the incarceration rate by about 33%.
  2. For many crimes, alternative punishments that are cheaper than incarceration should be instituted. This is particularly true of non-violent crimes of theft involving less than $1,133.
    1.  
 References
Bastian, Lisa D. (1994) Criminal Victimization in the United States: 1973–92 Trends, Bureau of Justice Statistics Report NCJ–147006, U. S. Department of Justice

Bastian, Lisa D. (1995) Criminal Victimization 1993, Bureau of Justice Statistics Bulletin NCJ–151658, U. S. Department of Justice

Bell, Griffen B. and William J. Bennett, co-chairs (1996) The State of Violent Crime in America, The Council on Crime in America

Bureau of Justice Statistics (1995) Drugs and Crime Facts, 1994 Bureau of Justice Statistics Report NCJ–154043, U. S. Department of Justice

Bureau of Justice Statistics (1996) Probation and Parole Population Reaches Almost 3.8 Million, Bureau of Justice Statistics New Release BJS 202/307–0703 , U. S. Department of Justice

Bureau of Labor Statistics (1997) Consumer Price Index: All Urban Consumers, private communication, January 14

Donziger, Steven R., Editor (1996) The Real War on Crime, The Report of the National Criminal Justice Commission, Harper Perennial, HarperCollins Publishers, Inc. New York, NY

Gilliard, Darrell K. and Allen J. Beck (1996) Prisoners and Jail Inmates, 1995, Bureau of Justice Statistics Report NCJ–161132, U. S. Department of Justice

Horn, Martin F. (1996) Testimony before the Senate judiciary Committee, Pennsylvania Department of Corrections

Levitt, Steven D. (1996) "The Effect of Prison Population Size on Crime Rates: Evidence from Prison Overcrowding Litigation," Quarterly Journal of Economics, May, p. 319

Lungren, Dan (1996) "Three Cheers for 3 Strikes," Policy Review, November-December, p 34.

Maguire, Kathleen and Ann L. Pastore, eds. (1996) Sourcebook of Criminal Justice Statistics 1995, Bureau of Justice Statistics Report NCJ–158900, U. S. Department of Justice

Perkins, Craig A., Patsy A. Klaus, Lisa D. Bastian, and Robyn L. Cohen (1996) Criminal Victimization in the United States, 1993 , Bureau of Justice Statistics Report NCJ–151657, U. S. Department of Justice

Perkins, Craig and Patsy Klaus (1996) Criminal Victimization 1994, Bureau of Justice Statistics Bulletin NCJ–158022, U. S. Department of Justice

Rand, Michael R. (1997), private communication

Renninger, Phillip (1996) Data Presentation to: The Pennsylvania Senate Judiciary Committee, Pennsylvania Commission on Crime and Delinquency

Smith, Stu (1996) Victims Report 9 Percent Fewer Violent Crimes Last Year, Bureau of Justice Statistics News Release 202/633–3047, U. S. Department of Justice