But aren’t youth committing more serious crimes today?

NO. They are committing LESS than previous generations.

Let’s look at the three major ways to measure crime:

1. FBI Uniform Crime Reports of arrests by age and offenses solved by the arrest of a youth. These annual reports are available in modern form from 1964 through 2005.
2. The Bureau of Justice Statistics’ National Crime Victimization Survey surveys a representative sample of more than 100,000 Americans annually on their victimizations by various types of crimes, including the perceived age of the offender. The NCVS, available from 1973 through 2005, is considered our best measure of crime because it captures offenses that are not reported to police.
3. The Institute for Social Research’s Monitoring the Future, surveys a representative sample of high school seniors every year on victimizations and delinquent activities such as fighting, robbery, and weapons offenses. MTF is available from 1975 through 2005.

 

These three very different measures of crime—offenses reported to police and cleared by arrest, arrest rates by age for serious crimes, and anonymous surveys of crime by victims—yield very similar results: violence and major property crime by young people has plummeted to record low levels over the last generation. Add this to record low levels of homicide by and against younger people shown in FBI and vital statistics reports, and it is hard to escape the conclusion that today’s youth are the least criminal and violent in many a decade, and perhaps ever.

Let’s look at the leading measures of crime:

I. The FBI’s best composite measure of crime by youth is the percentage of offenses “cleared” (that is, solved) by the arrest of a youth under age 18 (shown in Table 28 of the 2005 Uniform Crime Reports, and corresponding tables for previous years). Beginning in 1964, the FBI began collecting reports from thousands of law enforcement agencies in a dozen population categories, from cities over 250,000 people in size to small towns and rural agencies.

Better than raw arrests (which overestimate youth crime, since teenagers tend to be over-arrested and then released more than adults are), crime clearance measures correctly estimate the impact of juvenile crime on society. Since clearances are expressed as the percentage of the total cleared offenses committed by youth, they have to be adjusted by the percentage of the total population comprised by youths.

Table 1 But aren’t youth committing more of our crime today? NO!

The percent of America’s total crime estimated by the FBI to have been committed by, and the percent of the population, age 10-17: Youth vs. Adults

Annual Average	Major Crime*	Population	Odds Ratio*
1964-69	32.1%	21.9%	1.69
1970-74	29.3%	21.8%	1.49
1975-79	28.7%	19.8%	1.63
1980-84	21.3%	17.3%	1.29
1985-89	18.6%	15.6%	1.24
1990-94	20.2%	15.3%	1.40
1995-99	20.2%	15.8%	1.35
2005	16.6%	15.4%	1.09
Change	-44%	+35%

 

Source: FBI Uniform Crime Reports, crimes cleared by the arrest of persons under 18, 1964-2005. Population is percent of population ages 10-64 that is age 10-17.

*Major, or Part I crime consists of the four major violent offenses (murder, rape, robbery, aggravated assault), and the four major property offenses (burglary, larceny/theft, motor vehicle theft, and arson) used by the FBI as the “index of crime.”

**The odds ratio is a standard statistic measuring, in this case, the odds of a youth versus an adult committing a serious crime. It shows (a) the percentage of crime committed by youth divided by the percentage of the population comprised by youth, divided by (b) the percentage of crime committed by adults divided by the percentage of the population comprised by adults.

Table 1 Arranged by multiple years, shows a startling trend:

• In 1964-69, youths ages 10-17 comprised 21.9% of the US population ages 10-64 (middle right column) and committed 32.1% of all major violent and property crime (middle left column). That is, youths were 1.69 times, or 69%, more likely to inflict a serious crime on society (left column) than were adults ages 18-64.
• Yet, in 2005, youths comprised 15.4% of the population and committed 15.4% of all major crime—a crime level 1.09 times, or just 9% higher, than for adults ages 18-64.

 

Another FBI measure, rates of arrests for criminal offenses, is a less reliable way to measure crime trends. Only a fraction of crimes are reported to police, and only a fraction of these result in arrests. However, many authorities use arrests by age, shown in Table 38 of the Uniform Crime Reports, as a rough measure of trends in crime by age group.

If we examine real trends in the largest, most serious category of crime, Part I, or “index,” violent and property felonies—consistently defined crimes that more accurately depict larger trends—we see a surprising trend.

Table 2 But aren’t youth today getting into more serious crimes than ever while adults are less criminal? Just the opposite!

Part I felony arrest rates per 100,000 population for ages:

Annual Average	<-18*	18-24	25-34	35-49	50+*	All ages
1960-64	1,673.3	1,445.5	650.7	301.9	85.6	687.2
1965-69	2,160.3	1,637.4	742.5	333.7	92.6	861.8
1970-74	2,633.9	2,269.8	954.5	418.7	117.9	1,125.6
1975-79	2,918.4	2,534.4	1,078.0	486.1	141.9	1,239.7
1980-84	2,710.3	2,727.9	1,336.9	597.26	170.6	1,267.0
1985-89	2,827.6	2,881.2	1,603.7	705.1	165.2	1,319.6
1990-94	3,049.8	3,014.4	1,760.6	823.4	160.3	1,383.2
1995-99	2,581.0	2,699.0	1,490.0	823.3	136.6	1,191.5
2000-04	1,763.2	2,269.1	1,172.2	733.1	127.1	933.6
2005	1,541.7	2,158.1	1,188.0	756.5	141.6	892.3
2005 rate versus:
1960s	-20%	40%	71%	138%	59%	15%
1970s	-44%	-10%	17%	67%	9%	-25%
1980s	-44%	-23%	-19%	16%	-16%	-31%
1990s	-45%-	-24%	-27%	-8%	-5%	-31%
2000-04	-13%	-5%	1%	3%	11%	-4%

 

*The rate for <18 is all arrests for persons under age 18 divided by the population age 10-17; for 50+, all arrests of persons 50 and older divided by the population age 50-69.

Source: FBI Uniform Crime Reports, 1960-2005. Prior to 1960, crime reports covered only a fraction of the country and no jurisdictions outside of cities. Annual arrests are adjusted to reflect the report’s coverage of the population and divided by the population by age to produce rates.

Young people are the only age group to show lower arrest rates for serious crimes today than 40 years ago (down 20% since the 1960s)—while all other age groups, particularly 35-49 (up 138%), show large increases. No matter what eras are compared, youth today show crime declines that are larger than for grownups.

While youth of 30 to 40 years ago were five to six times more likely than adults of age to be their parents (ages 35-49) to be arrested for serious crime, today that ratio has dropped to 2-to-1. This despite the fact that on average, youths age 10-17 are two to three times more likely to live in poverty than are middle-aged adults, and poorer populations of all ages suffer higher arrest rates.

These youthful improvements do not result from changes in the definitions of Part I crimes (these have remained consistent) or lesser policing. In fact, crime reports are more complete, victims are more likely to report crimes to police, and reported crimes are more likely to result in arrests today than in the past, which makes the young’s decline all the more remarkable.

Conclusions from FBI Uniform Crime Reports:

1. The volume and impact of serious youth crime dropped dramatically—from nearly one-third of total offenses in the 1960s to one-sixth today.
2. The average youth raised in the 1990s and 2000s is substantially less likely to commit a serious offense than a youth raised in the 1950s and 1960s.
3. Rates of youthful arrest for serious offenses are much lower today than in past decades, while rates for middle-aged adults have risen substantially.

 

If you want to examine trends in arrest and arrest rates by age and specific crime for the 1960-2005 period, see the detailed crime data tables on this site.

II. Most crime authorities consider the National Crime Victimization Survey (NCVS) to more reliably capture all crimes—at least, those committed against teenagers and adults. The NCVS typically finds around many hundreds of thousands more violent offenses than are captured in FBI reports. Victims are also asked to state or guess the age of the offenders, which may or may not be accurate.

The Bureau of Justice Statistics has adjusted the violent victimization totals over the last 30 years to be as consistent as possible and expressed them as rates per 100,000 population ages 12 and older. Using victim reports of offenders’ ages, the percent, number, and rate of violent crimes that involve offenders under age 21 can be estimated.

Table 3 But aren’t more Americans being victimized by violent youth? NO!

Victimizations/1,000 people 12+

Year	Murder*	Total	Property	Violent	Violent Offender under age 21
1973-74	0.12	583.6	535.7	47.9	17.7
1975-79	0.11	591.1	541.3	49.8	17.2
1980-84	0.11	506.9	457.8	49.1	16.7
1985-89	0.10	421.6	377.9	43.7	14.4
1990-94	0.12	379.7	331.5	48.2	15.0
1995-99	0.08	280.6	241.9	38.9	14.7
2000-04	0.07	189.5	165.7	23.8	7.8
2005	0.06	175.2	154.0	21.2	6.9
Change, 2005 vs. 1973	-53%	-70%	-71%	-56%	-61%

 

Source: National Crime Victimization Survey. Property victimization includes burglary, theft, and motor vehicle theft. Violent victimization includes rape, robbery, aggravated assault, and simple assault. Victims’ estimates of offender age are not available for property crimes. Age is 12 and older. Age group under 21 is the one used by NCVS.*Murders (homicide deaths) are available from the National Center for Health Statistics.

Note that by our best measure, the United States is a dramatically safer country today than it was 30 years ago, with massive drops in both violent and property crime.

• In 1973, more than half—583.6 per 1,000—of Americans were victimized by a property or violent offense, including minor assaults.
• By 2005, that percentage had plunged to 175.2, or around one-sixth. The biggest drop was in property offenses such as burglary and theft, which traditionally are those youths are most likely to commit.

 

Why are we safer today? Because crime by young people has dropped dramatically:

• In 1975, 17.7 per 1,000 Americans ages 12 and older reported being violently victimized (including by rape, robbery, aggravated assault, and simple assault) by an offender or group of offenders under age 21.
• By 2005, that proportion had fallen to 7 per 1,000.

 

Adjusting offending rates for changes in the teenage population over the period, the average teenager was 45% less likely to violently victimize someone else in 2005 than in 1973. Even this drop understates the decline in violence by youth, since victimizations involving multiple offenders (counted as one victimization in the above table, even though there were several assailants) have fallen more rapidly than have single-offender attacks.

Conclusion from National Crime Victimization Survey findings: The average American’s odds of being violently victimized by a teenager or young adult fell by 61% over the last 30 years. (The ages of most property offenders are not known.)

III. What about self-reported delinquency? Self-reporting surveys contain a number of weaknesses that are hard to factor out, including changes in definitions of crime (for example, 30 years ago, family violence was not considered the criminal behavior it is today), inconsistent personal definitions (i.e., what is a “serious fight”?), and subject understatement or exaggeration.

Monitoring the Future’s survey of high school seniors, the only consistent measure over the last generation, presents a mixed picture. Students report little change in delinquent activities, either violent or property offenses, over the last 30 years. The only significant change is a decline in shoplifting.

Table 3 But don’t youths admit to more violence and crime today?

Question: High school seniors reporting at least one incident in previous 12 months

Year	1975-76	1980	1990	2000	2005
Got into serious fight at work or school	15%	16%	19%	13%	12%
Got into a group fight	17%	18%	21%	20%	19%
Used weapon to commit robbery	3%	3%	4%	3%	4%
Hit instructor/supervisor	3%	3%	3%	3%	3%
Stole something (of any value)	38%	40%	42%	43%	36%
Shoplifted from a store	35%	32%	32%	29%	25%
Drove a car without permission	4%	5%	7%	5%	5%
Committed arson	2%	2%	2%	3%	3%
Damaged property at school or work	18%	20%	20%	21%	19%

 

*Source: Monitoring the Future, 1975-2005.

From self reports, one could conclude that youth today are no more, or less, criminal than they were in their parents’ generation 30 years ago. Nearly nine in 10 youths will not be in a serious fight even once in a year’s time, and most fights (including group fights) are not seen as serious. A bit more than one in three youths say they stole something, one-fourth from a store, and one fifth damaged property at school or work (the question does not ask whether they did it on purpose).

Conclusion from self-reported delinquency: A large majority of youth do not report any serious delinquency (especially violence) in a year’s time. A fraction of youths engage in various, low-level delinquencies whose rates have not changed much over time. About 3% to 4% say they committed more serious violence or property damage. It is not known how consistent, understated, or exaggerated these self-reports may be.