The Youth Crime Index:
Toward a More Complete Measure of Juvenile Crime
by Burk Foster and Craig Forsyth
(December 1992)
In a February 3, 1987, New York Times article, Peter Applebome discussed recent trends in juvenile crime. The gist of his article is summed up nicely in his title: "Juvenile Crime: The Offenders are Younger and the Offenses More Serious."
Interviewing juvenile court officials in New York City, Newark, Fort Worth, San Antonio, Orlando and officials of the National Center for Juvenile Justice in Pittsburgh, Applebome found that the ages of juveniles referred to court after committing serious crimes had declined in recent years, that juveniles 13 and under are much more frequently involved in gun crimes and other violent offenses than they were a few years ago, and "that while the number of youthful offenders may not be rising, there is no question that the severity of the crimes is increasing."
Applebome's article drew an immediate response from Barry Krisberg, President of the National Council on Crime and Delinquency. In his February 5, 1987, letter to the editor, Krisberg spoke to the "myth . . . that today's young people are worse than those of previous generations." He pointed out that juvenile arrest figures had declined steadily from the mid-1970s until 1984 (as had the number of teenagers in the population), and that only for the crime of rape had the 1985 arrest rates of those younger than 15 shown a marked increase, in comparison to 1976.
What bothered Krisberg the most was not the juvenile crime problem itself, it was the increased number of juveniles in confinement. He suggested in conclusion that "expanded incarceration of juveniles does not protect public safety and may actually increase rates of serious youth crime."
Dialogues of this sort have occurred often over the past 30 years, as America has tried to come to grips with a serious juvenile crime problem. At the risk of oversimplification, the discussants have often divided themselves into two opposing camps--the "getting worse" and the "not getting worse." Those in the first group, most often concerned citizens (including victims of violent juvenile crime), journalists and some juvenile justice practitioners, tend to argue that the criminality of today's juveniles is much more serious than that of the young people of 30 or 20 or even 10 years ago. Those in the second group, consisting mostly of professors, social workers and some researchers, argue the contrary: that today's juveniles are no more threatening or no more criminal than those of earlier times, and may in fact be even less so. Or, as the "irresistible force/immovable object" proposition was posed by a student in one of the writers' classes recently, "How can we be warned that juvenile criminals are getting much worse, on the one hand, and encouraged that juvenile crime is going down, on the other? It doesn't make sense."
In attempting to provide a sensible answer to this question, we began with a conventional starting point--juvenile arrest statistics of the Uniform Crime Reports. Since 1930 the Federal Bureau of Investigation has collected and disseminated national crime figures. Beginning with a base population of about 20 million in the early 1930s, the population covered in these police-generated statistics had increased to over 200 million by the 1980s.
The reliability of these figures prior to 1960 has been highly suspect, primarily because of wide variations in how police recorded citizens' reports of crime and how these reports were tabulated to be forwarded to the FBI for inclusion in the national figures. Since the early 1960s, however, most authorities believe that the quality of police reporting practices is much improved, in terms of both inclusiveness and accuracy.
We chose 1963 as the starting point of our examination of juvenile arrest statistics. By 1963 over two-thirds of the American public was policed by agencies contributing statistics to the UCR system. Most crime rate experts also believe that by 1963 the post-World War II "baby boom" was beginning to make its impact on reported crime and arrest figures, beginning the sharp increase in the crime rate that would continue on into the 1970s.
We took juvenile arrest figures from Crime in the United States, the annual compilation of UCR reports, every five years from 1963 to 1988. These figures are
shown in Table 1. The FBI has followed a consistent format in reporting juvenile arrest data.
Table 1
Juvenile Arrests, 1963-1988
1963 1968 1973 1978 1983 1988
Reporting Population 125.760 145.306 155.995 207.060 200.692 188.928
(in millions)
Arrests Under Age 15
Total 307,584 564,343 614,716 728,198 564,983 526,421
As % of all arrests 6.6% 10.0% 9.5% 7.4% 5.5% 5.2%
Arrests by offense:
Murder/Manslaughter 69 164 216 244 157 201
Forcible Rape 291 489 813 l,102 1,332 1,372
Robbery 3,657 8,231 11,015 13,086 9,203 6,470
Aggravated Assault 2,848 5,974 8,200 11,508 10,148 11,345
Burglary 39,754 67,267 73,139 93,652 59,400 42,376
Larceny/Theft 83,060 133,897 146,910 194,680 168,095 152,952
Auto Theft 13,791 20,547 17,736 20,146 8,628 15,870
Arrests Under Age 18
Total 788,762 1,457,078 1,717,366 2,279,635 1,725,746 1,634,790
As % of total arrests 17.5% 25.9% 26.4% 23.3% 16.8% 16.1%
Arrests by offense:
Murder/Manslaughter 477 1,027 1,497 1,735 1,345 1,765
Forcible Rape 1,656 2,559 3,772 4,517 4,388 4,118
Robbery 9,963 22,876 34,374 48,088 35,219 24,337
Aggravated Assault 9,473 17,590 26,270 41,253 33,730 38,536
Burglary 85,151 140,229 170,228 250,649 159,192 111,284
Larceny/Theft 160,089 250,503 310,452 454,994 377,435 351,133
Auto Theft 54,417 75,988 66,868 77,534 36,497 61,301
Source: Crime in the United States, 1963, 1968, 1973, 1978, 1983, 1988
Arrests are reported for each Index offense (now called Part I offenses) and broken down by under-15 and under-18 age groups; within these two age categories, juvenile arrests are also described as a percentage of all arrests reported for that year, and the reporting population (which would be the population of the local and state law enforcement jurisdictions participating in the UCR system) is given in millions.
This format has remained consistent for the survey period, except for the addition of arson as a Part I crime in 1979. We left arson out of our tables but counted arrests for the other seven Part I crimes going back to 1963.
A brief look at the figures of Table 1 tends to support the main points made by the critics of the "juvenile crime wave" theory. Arrests went up sharply, in both absolute numbers and as a percentage of all arrests, for all crimes through the late 1960s and on through 1973 and 1978, except for auto theft, which levelled off in the early 1970s before the others did. By the early 1980s, juvenile arrests were in decline for every offense. The arrests of under-18s--as a part of the total--had fallen from 26.4% in 1973 to 16.8% in 1983, a drop of almost 10 points in 10 years.
What was happening, it was explained at the time, was that the bulge of baby-boomers was beginning to be passed along. Crime was not going down, exactly, it was just flattening out, and an increasing number of arrests were being made of people in their late teens through late twenties, those who had been in the big bulge of baby boomers a few years earlier.
These juvenile arrest statistics, covering the period 1963-1988, provide the base for our research. There are two main objections that can be made to our use of these figures:
1. They are incorrect, because of human error in tabulation and reporting to the UCR system. This is a common and no doubt somewhat valid criticism of all UCR figures, based as they are on reports from over 10,000 mostly local law enforcement agencies across the United States.
2. They do not reflect changes in police operational practices, such as handling more juveniles informally--without arresting them--at an earlier time and changing to a more formal approach emphasizing arrest after the 1960s. This criticism is also valid, except that we would point out that the Part I offenses counted in Table I tend to be the more serious crimes (mostly felonies, except for larceny) for which arrests have always been more likely in the first place. We recognize that police practices in dealing with minor delinquent acts and status offenses have changed considerably--and not always in the direction of leniency, as is evident in the much greater percentage of juveniles waived along to the adult criminal court--over the last 30 years; we suggest only that police practices resulting in arrests of juveniles for Part I crimes have not undergone as much change as have police juvenile procedures overall; they have certainly not undergone enough change to invalidate this comparison over time.
Table 2 shows the population distribution of youths in various age group categories from 1963 to 1988. These figures, derived from the Census Bureau's estimates in their annualPopulation Reports, show that the American population was indeed very young at one point in our recent history, and that we've gotten older fast. In 1963, 36.36% of the American population was under 18; 25 years later that ominous number (ominous in terms of its potential impact on culture) had dropped to 25.92%.
The age group we were most interested in was the 10- to 17-year-old category. This category would include those who are legally juveniles, for arrest purposes, in most states, and who are also in their peak arrest years. The Report to the Nation on Crime and Justice(1988), points out that property crime arrest rates peak at age 16, while violent crime arrest rates peak at age 18. Very few arrests, less than 1% of all arrests included in the UCR annually, involve children under the age of 10.
The representation of 10-to-17-year-olds in the American population was fairly stable through the 1960s and 1970s, peaking at 16.03% in our study year of
1973 and beginning to decline after that. Fifteen years later there were 6,000,000 fewer 10- to 17-year-olds in the United States, a percentage decrease of almost
five points to 11.11%.
Table 2
Population Distribution
1963-1988
1963 1968 1973 1978 1983 1988
Total U.S. Population 189.417 201.166 209.600 220.467 234.496 246.329
(as of July 1, in millions)
Population under 15 58.861 59.670 56.160 51.955 51.588 53.111
(in millions)
% of Population under 15 31.08% 29.66% 26.79% 23.57% 22.00% 21.56%
Population under 18 68.863 70.813 68.726 64.752 62.575 63.846
(in millions)
% of Population under 18 36.36% 35.20% 32.79% 29.37% 26.68% 25.92%
Population 14 to 17 13.507 15.052 16.828 16.923 14.633 13.983
(in millions)
% of Population 14 to 17 7.13% 7.48% 7.90% 7.68% 6.24% 5.68%
Population 10 to 17 28.039 31.374 33.596 31.717 28.768 27.362
(in millions)
% of Population 10 to 17 14.80% 15.59% 16.03% 14.39% 12.27% 11.11%
Table 3 reflects changes in the reporting base from which juvenile arrest figures are compiled. During the 1960s and 1970s the number of law enforcement agencies taking part in the UCR system increased steadily. In the early sixties, as we noted earlier, two-thirds of the American population was included in the base from which arrest statistics were drawn. By the end of the seventies, the base had climbed to almost 94%. But just when you thought we were approaching perfection, total participation resulting in a truly all-inclusive crime statistics network,
participation began to fall off. In the decade from 1978 to 1988, the percentage of the public covered in arrests statistics fell by 17 percentage points.
Table 3
Crime Reporting
1963 1968 1973 1978 1983 1988
Total U.S. Population 189.417 20l.166 209.600 220.467 234.496 246.329
(as of July 1, in millions)
UCR Base Population 125.760 145.306 154.995 207.060 200.692 188.928
for Total Arrests
% of Population Reporting 66.39% 72.23% 73.95% 93.92% 85.58% 76.70%
This sudden, sharp decline in the arrest base most likely resulted from the increased effort on the part of the FBI to heighten the credibility of its statistics. Many
municipal police and sheriff's departments, especially those serving smaller towns and rural areas, could not meet stricter reporting requirements and ceased to
send in arrest statistics. In numbers of people not included, the increase was from 13,000,000 not counted in 1978 to over 60,000,000 ten years later. This is a
huge increase, one that by itself would throw off the reliability of any analysis of juvenile arrest trends that failed to take it into account.
Table 4 lists the age group and base-arrest population variables we have already identified and arranges them according to the five-year intervals of our study.
The two key percentages are the percentage of the population covered by UCR arrest statistics for each year, and the percentage of the population in the 10- to
17-year-old age group.
Table 4
Youth Crime Index Variables
1963 1968 1973 1978 1983 1988
Total U.S. Population 189.417 201.166 209.600 220.467 234.496 246.329
(as of July 1, in millions)
UCR Base Population 125.760 145.306 154.995 207.060 200.692 188.928
for Total Arrests
% of Population Reporting 66.39% 72.23% 73.95% 93.92% 85.58% 76.70%
Population 10 to 17 28.039 31.374 33.596 31.717 28.768 27.362
(in millions)
% of Population 10 to 17 14.80% 15.59% 16.03% 14.39% 12.27% 11.11%
Table 5 is what we have been headed toward all along--the Youth Crime Index. The YCI starts with juvenile arrests (which is our term for the UCR category of arrests under age 18) for each Part I crime. We then divide by percentage of population covered by arrest statistics, to make up for what is missing, and by the percentage of persons age 10 to 17 in the population. The result of these manipulations is a more comprehensive picture of juvenile crime: it shows us, over time, what arrests rates would be if all arrests were reported and if all criminals were arrested at the same rate as 10- to 17-year-olds.
Our position is that the YCI is a more realistic way of considering the true state of juvenile crime. Most people commenting on juvenile crime look simply at arrest statistics.
Some make reference to the declining percentage of juveniles in the population, usually without reference to specific numbers illustrating the decline (over 30%
fewer 10- to 17-year-olds in 1988 than in 1973, for example) or without suggesting what this means in understanding the level of juvenile criminality today.
Table 5
Youth Crime Index
1963 1968 1973 1978 1983 1988
Murder/Manslaughter 4,855 9,120 12,628 12,838 12,809 20,713
Forcible Rape 16,854 22,725 31,820 33,422 41,788 48,326
Robbery 101,397 203,150 289,974 355,810 335,398 285,600
Aggravated Assault 96,410 156,208 221,609 305,237 321,217 452,228
Burglary 866,613 1,245,300 1,436,016 1,854,587 1,516,018 1,305,940
Larceny/Theft 1,629,285 2,224,586 2,618,923 3,366,563 3,594,391 4,120,617
Auto Theft 553,822 674,810 564,088 573,685 347,568 719,380
Note: The Youth Crime Index is calculated by taking the raw number of juvenile arrests for each crime in a given year, dividing by the percentage of arrests
reported (to account for that portion of the population not counted in UCR arrest totals), and then dividing by the percentage of the population made up by of 10
to 17 year olds; the result is a crime-specific, age-specific index of per capita arrest rates, indicating what the arrest rate would be if all arrests were reported and
all persons were arrested at the same rate as 10 to 17 year olds.
Few researchers have apparently noticed or commented on the increased and, later, decreased population from which juvenile arrest statistics are drawn. Lamar Empey and Mark Stafford are notable exceptions here. Their book, American Delinquency: Its Meaning and Construction, describes the influence of both youth population fluctuations and changes in the arrest reporting base and uses a formula to compensate for that portion of the population not counted in arrest figures. They stress the need for comparing rates, rather than simple numbers extracted from a very complex social context.
Recent FBI analyses of juvenile crime have begun to employ arrest rates (per 100,000) as a measure of change in juvenile criminality over time. The Youth Crime Index simply re-combines these rates with actual arrest statistics to get what can be called an "extrapolated" picture of juvenile crime--a table that in the end can be compared with composite arrest statistics for all age groups (and FBI estimates that include populations served by agencies not reporting) to see how juveniles "match up" with other offenders.
What does the Youth Crime Index say about juvenile crime in the United States over the last 30 years? It says, first of all, we would be much worse off if all our criminals were as active as those in the 10- to 17-year-old age group. It is one thing for us to know that arrest rates peak out at 16 or 18; it is something more scary to think that our overall Crime Index would be two-and-a-half times higher (6.95 million YCI arrests in 1988 versus 2.88 million estimated arrests in the 1988 Crime in the United States).
The Youth Crime Index shows that the intensity of juvenile crime continues to increase. The murder rate, for instance, which levelled off from 1973 to 1983, jumped abruptly over the next interval, so that by 1988 it was more than four times what it had been in 1963. The forcible rape and robbery rates had tripled, and aggravated assault had more than quadrupled over the same 25 year period.
Of the seven Part I offenses, the 1988 Index was the highest ever for five; only robbery and burglary had peaked out in earlier years, both in 1978, and then gone into decline over the next decade. The most rapid increases in the 1980s were in the two crimes of personal violence, murder and aggravated assault, which tends to reflect what observers of urban life have suggested about the influence of gangs, drugs and guns on crimes of violence in big city neighborhoods, and the property crime of auto theft, which had been stable for a long time, declined, and suddenly shot up again.
How would we answer our student's question, then, the one about juvenile criminals getting worse while juvenile crime is going down. We would say, first, that juvenile crime, as reflected in arrest figures, is not going down at all, except for burglary and robbery. Apparent declines in other offenses are the result of changes in police reporting practices (or FBI validating practices), so that a much smaller percentage of the population is included in arrest figures.
The agencies that continue to contribute juvenile arrest statistics are generally the larger agencies serving urban populations; we recognize that the missing figures are mostly from less urban areas where the crime rate and the arrest rate are both likely to be much lower than in the cities. Our YCI figure is going to be inflated accordingly, though how much as we compensate for unreported arrests we can't say.
We would say, second, that the only reason our juvenile crime problem hasn't gotten a lot worse is that there are so many fewer 10- to 17-year-olds around to be criminals now than there were 15 or 20 years ago. If there were as many now as there were then, and the arrest rate remained constant, we would have 50% more juveniles being arrested each year. This would place an even greater burden on our juvenile justice system and would almost surely result in a further increase of habitual juvenile offenders being waived to the adult criminal courts.
We would say, third, to those practitioners and researchers cited in such articles as "Juvenile Crime Increases," "Violent Crimes by Juveniles Soar," "Street-Gang Violence Epidemic" and "Children Are Killing More Than Ever," that there is nothing wrong with your perceptions: today's juveniles are more violent and more criminal in general than any that have preceded them. Whether that heightened level of criminality and violence, what has come to be termed "intensity," is spread generally across the country or whether it is focused in the most socially-disadvantaged neighborhoods of our big cities is beyond the scope of our inquiry.
We would say, finally, that two years from now we will have another five-year interval to add to our tables. If the statistics from 1993 show a worsening of the
rate of juvenile crime, the consequences may be the cause of more than ordinary concern. At some point, we know, the percentage of young people in the 10 to
17 age group will begin to increase again, and if they in fact turn out to be even more crime-prone than this present generation, their impact on a disorganized
society may be more negative than anything we have seen in modern times.
REFERENCES
Applebome, Peter. 1987. Juvenile Crime: The Offenders Are Younger and the Offenses More Serious. New York Times. February 3.
Bureau of Justice Statistics. 1988. Report to the Nation on Crime and Justice. Washington, D.C.: U.S. Department of Justice.
Butterfield, Fox. 1992. Seeds of Murder Epidemic: Teen-Age Boys with Guns. New York Times. October 19, 1992.
Doleschal, Eugene and Newton, Anne. 1978. The Violent Juvenile: A Review. Criminal Justice Abstracts 10: 539-573.
Empey, LaMar T. and Stafford, Mark C. 1991. American Delinquency: Its Meaning and Construction, 3rd ed. Belmont, CA:
Wadsworth Publishing Company.
Glass, Andrew J. 1992. Violent Crimes by Juveniles Soar. Lafayette Advertiser. August 30.
Hufstedler, Shirley M. 1984. Should We Give Up Reform? Crime and Delinquency 30:415-422.
Krisberg, Barry. 1987. Letter. New York Times. February 6.
Lafayette Advertiser. 1990. Juvenile Crime Increases. March 19.
________. 1984. Street-Gang Violence Epidemic. October 21.
Murphy, Dean E. 1992. Children Are Killing More Than Ever.Baton Rouge Sunday Advocate. August 16.
Schmalleger, Frank. 1991. Criminal Justice Today. Englewood Cliffs, NJ: Prentice-Hall.
Uniform Crime Reports. 1963, 1968, 1973, 1978, 1983, 1988. Crime in the United States. Washington, D.C.: U.S. Government Printing Office.
U.S. Bureau of Census. 1963, 1968, 1973, 1978, 1983, 1988. Population Reports: Estimates of the Population of the United States by Age, Sex, and Race.
Current Population Reports, Series P-25. Washington, D.C.: U.S. Government Printing Office.