Social institutions and the crime "bust" of the 1990s.

• Author: Lafree, Gary

1. INTRODUCTION

   While scarcely visible behind the wave of media attention devoted to crime, beginning in the early 1990s, researchers began to observe declining rates in violent and other serious forms of crime in the United States. In this article, I begin by evaluating the size of these declines and comparing crime trends in the 1990s to crime trends in the U.S. since World War II. I also examine trends for different crime types and data sources, and for large cities. My review shows that crime rates in the 1990s have dropped rapidly and that declines have been sweeping, affecting all of the street crimes that are routinely tracked by the two major sources of crime data in the United States (the Uniform Crime Reports and the National Crime Victimization Survey). In short, the period from 1990 to 1997 represents the closest thing to a sustained decline in crime, or a "crime bust" that the United States has experienced in more than fifty years.

   In a recently published book, I argue that changes in the legitimacy of social institutions provide the most promising explanation for the rapid changes in street crime trends observed in America after World War II.(1) In particular, I argue that American crime rates surged in the 1960s as a result of increasing distrust of political institutions, increasing stress produced by economic institutions, and declining strength of family institutions.(2) American society countered this growing legitimacy crisis by investing more in other institutions, most notably criminal justice, education, and welfare. Stabilization in the legitimacy of political, economic, and family institutions, and investments in criminal justice, education, and welfare eventually produced downward pressure on crime rates. In this essay, I consider the applicability of these arguments for the declines in U.S. crime rates that have occurred during the first eight years of the 1990s.

   I concentrate here on the group of offenses popularly known as "street crimes." While the term is imprecise, it has generally come to include the familiar crimes of murder, robbery, rape, aggravated assault, burglary, and larceny. Glaser calls these crimes "predatory" because they all involve offenders who "prey" on other persons or their property, while contrasting these offenses with nonpredatory crimes like prostitution and gambling.(3) Predatory crimes are especially worthy of attention because they generally evoke the greatest popular fear and concern and draw the most universal condemnation from society. Probably as a consequence of these characteristics, we have more complete information on predatory crimes than on any other crime types.

2. STREET CRIME TRENDS IN THE UNITED STATES

   Street crimes are all similar in that they each consist of an offender, a victim, and at least for the crimes that are reported or discovered, the involvement of police and other legal agents. This, in turn, defines the crime data that may be collected: "official" data collected by legal agents, "self report" data collected from offenders, and "victimization" data collected from crime victims. Because self report data are likely to be most reliably collected for the least serious crimes,(4) they are of less value for the street crimes examined here, leaving us to rely almost exclusively on either official or victimization data. Moreover, national crime victimization survey data in the United States have only been collected on an annual basis since 1973. In the next section, I examine street crime trends from official data and victimization survey data for the 1990s.

   1. STREET CRIME IN THE 1990S

      The most comprehensive official data on street crime in the United States come from the Uniform Crime Reports (UCR), collected annually by the Federal Bureau of Investigation since 1930. Table 1 shows rates for the seven street crimes on which data are most reliably compiled by the UCR, as well as totals for violent crimes, and crimes to property, as well as all crimes, for 1990 and 1997.(5) The percentage change column indicates sizeable decreases in all seven crimes. The greatest percentage declines are for murder, robbery, burglary, and motor vehicle theft--all registered at least a 23% drop in the first eight years of the 1990s. Percentage declines for rape, aggravated assault and larceny have been somewhat less, but still sizeable. Taken together, these seven crimes logged a 15.4% drop from 1990 to 1997.

      TABLE 1.

      STREET CRIME RATES PER 100,000 U.S. INHABITANTS, 1990 AND 1996, UCR DATA.

      Crime 1990 1996 Percent Change Murder 9.4 6.8 -27.7 Rape 41.2 35.9 -12.9 Robbery 257.0 186.1 -27.6 Aggravated Assault 424.1 382.0 -9.9 Violent Crimes 731.7 610.8 -16.5 Burglary 1235.9 919.6 -25.6 Larceny 3194.8 2886.5 -9.6 Motor Vehicle Theft 657.8 503.8 -23.1 Property Crimes 5088.5 4311.9 -15.3 Total Crimes 5820.2 4922.7 -15.4 Note: Data from U.S. Federal Bureau of Investigation, Uniform Crime Reports, CRIME IN THE UNITED STATES 1990, 1997.

      National Crime Victimization Survey (NCVS) data have been collected annually by the Bureau of Justice Statistics since 1973.(6) The NCVS includes the same street crimes as the UCR with two exceptions: NCVS data do not report murders, and NCVS data distinguish between aggravated and simple assaults. Table 2 shows rates for seven street crimes collected by the NCVS for 1990 and 1996. Although there are substantial differences by crime type, the NCVS data, like the UCR data, show declines in all seven street crimes reported. Declines from the NCVS data are greatest for the three property crimes--all showing more than a 20% drop. Among violent crimes, rape rates show the greatest percentage decline 17.6%. Simple assaults show the least change, declining by only a little more than 1%.

      TABLE 2.

      STREET CRIME RATES PER 1,000 POPULATION, AGE 12 AND OVER, 1990 AND 1996, NCVS DATA.

      Crime 1990 1996 Percent Change Rape 1.7 1.4 -17.6 Robbery 5.7 5.2 -8.8 Aggravated Assault 9.8 8.8 -10.2 Simple Assault 26.9 26.6 -1.1 Total Violent Crime 44.1 42.0 -4.8 Burglary 64.5 47.2 -26.8 Larceny 263.8 205.7 -22.0 Motor Vehicle Theft 20.6 13.5 -34.5 Total Pro 348.9 266.3 -23.7 Crime Note: Data for 1990 from Michael R. Rand, et al, CRIMINAL VICTIMIZATION 1973-95 (1997); data for 1996 from C. Ringel, CRIMINAL VICTIMIZATION 1996 (1997).

      To assess how broad-gauged declines in national street crime rates in the 1990s have been, I next examined changes in crime rates for ten of the nation's largest cities from 1990 to 1995. Using UCR data, I calculated crime rates per 100,000 citizens for the seven crimes reported in Table 1 for Boston, Chicago, Dallas, Detroit, Houston, Los Angeles, New York, Philadelphia, Phoenix, and San Diego. Crime rates for the seven crimes in these ten cities produced a total of sixty-nine contrasts between 1990 and 1995.(7) Of these sixty-nine contrasts, sixty (87.0%) showed that crime rates had decreased. The size of decreases were typically much greater than the size of increases. For example, from 1990 to 1995, burglary rates declined by 51.0% in Dallas, 45.7% in Houston, and 40.7% in San Diego. By contrast, for the one city that recorded increases in robbery from 1990 to 1995 (Philadelphia), rates increased by 10.2%.(8)

      For all seven crimes, a majority of cities reported declining crime rates from 1990 to 1995. Thus, all ten cities reported declining murder and burglary rates, nine cities reported declining robbery and motor vehicle theft rates, eight cities reported declining rape rates, and seven cities reported declining aggravated assault and larceny rates.

      Taken together, data from both the UCR and the NCVS strongly support the conclusion that there have been substantial, broad-based declines in street crime rates during the first half of the 1990s. The UCR national data suggest that the declines have been greatest for murder, robbery, and burglary data show the largest decreases for motor vehicle theft, burglary, and larceny. City-level data from the UCR confirm that declines can also be observed in most cases for the nation's largest cities, especially for murder, robbery, rape, burglary, and motor vehicle theft. I will now consider how the magnitude of the recent crime decreases compares to earlier crime trends in the post-World War II United States.

   2. PUTTING THE RECENT DOWNTURNS INTO CONTEXT

      NCVS data allow us to examine annual trends only from 1973 to 1996. I use UCR data to examine street crime trends from 1946 to 1996.

      1. NCVS Data, 1973-1996

         Figure 1 shows trends for the four violent crimes tracked by the NCVS from 1973 to 1996. Trends for these four crimes provide considerable but not total support for the idea of a crime "bust" during the 1990s. The best case for a rapid decline in the 1990s can be made for rape cases and aggravated assault cases. Rape rates declined by 36.4% from 1991 to 1996 and aggravated assault rates declined by 27.3% from 1993 to 1996. These were the steepest declines in the years included in the NCVS data. Similarly, from 1994 to 1996, simple assault rates exhibited the greatest three-year decline (14.5%) since data collection began in 1973--although simple assault rates were only slightly lower in 1996 (26.6) than they had been in 1990 (26.9).

         [Figure 1 ILLUSTRATION OMITTED]

         Although robbery rates fell considerably during the 1990s, they declined even more rapidly during the 1980s.

         Taken together, rates for robbery and aggravated assault were lower in 1996 than at any other point spanned by the twenty-four years of NCVS data. The lowest level of reported rapes was recorded in 1995. While the lowest NCVS level of simple assaults occurred in 1986 (25.3), the levels were similar to those recorded for simple assaults in 1996 (26.6).

         Figure 2 shows trends in the three property crimes included in the NCVS data from 1973 to 1996. Data on property crimes are consistently supportive of a 1990s crime bust. In fact, for burglaries, the declines began well before the 1990s, starting in about...