U.S. Department of Justice Office of Justice Programs Bureau of Justice Statistics Special Report Police Departments in Large Cities, 1990-2000 May 2002, NCJ 175703 By Brian A. Reaves, Ph.D., and Matthew J. Hickman, BJS Statisticians ---------------------------------------------------------- This file is text only without graphics and many of the tables. A Zip archive of the tables in this report in spreadsheet format (.wk1) and the full report including tables and graphics in .pdf format are available from: http://www.ojp.usdoj.gov/bjs/abstract/tsp00.htm ---------------------------------------------------------- Highlights From 1990 to 2000, in cities with 250,000 or more residents, the number of UCR violent crimes decreased 34%, the number of UCR property crimes decreased 31%, and the number of full-time local police officers increased 17%. Among large city police departments, 1990-2000, changes included -- *The number of residents served increased by 10%, accompanied by a 7% increase, from 289 to 310, in the number of full-time sworn personnel per 100,000 residents. *The percentage of full-time sworn personnel who were members of a racial or ethnic minority increased from 30% to 38%. *Hispanic representation among officers increased from 9% to 14%, blacks from 18% to 20%, and women from 12% to 16%. *The percent of departments requiring new officers to have at least some college rose from 19% to 37%, and the percent requiring a 2-year or 4-year degree grew from 6% to 14%. *Annual operating costs per resident rose 10%, from $242 to $266, but annual per officer costs increased by just 2%, from $83,814 to $85,786. *The percent of departments using bicycles rose from 39% to 98%. The average number of bicycles in use went from 8 to 95. *The percent of departments using in-field computers increased from 73% to 92%, the percent using automated fingerprint ID systems from 60% to 97%, and the percent with enhanced 9-1-1 from 76% to 97%. *The percent of departments with full-time domestic violence units rose from 50% to 81%; with full-time victim assistance units, from 32% to 47%. ---------------------------------------------------------- From 1990 to 2000, 62 local police departments served cities with a population of 250,000 or more. During this time, the number of residents served by these agencies increased by 10%, from 45 million to 49.4 million; their number of full-time employees by 20%, from 166,823 to 199,627; and their number of full-time sworn personnel by 17%, from 130,242 to 152,858. In contrast to the increases noted above, the volume of serious crime reported in these cities was substantially lower in 2000 than in 1990. According to the FBI Uniform Crime Reports program, the number of violent Crime Index offenses declined by 34%, and the number of property Crime Index offenses by 31%. In addition to employment and crime trends, this report presents other data comparisons based on the 1990 and 2000 Law Enforcement Management and Administrative Statistics (LEMAS) surveys. Topics include staffing levels, race and ethnicity of officers, officer education and training requirements, operating budgets, officer salaries and special pay, types of special units operated, drug enforcement activities, sidearm and armor policies, types of vehicles operated, and computerization. Staffing levels From 1990 to 2000, the average number of full-time personnel in police departments serving cities with a population of 250,000 or more increased by 20%, from 2,691 to 3,220. New York City's police department (NYPD), the Nation's largest, had about 53,000 full-time employees as of June 2000. In 2000, police departments in large cities employed 404 full-time personnel per 100,000 residents. This represented an increase of 9% over the 1990 level of 370 per 100,000. On a land area served basis, employment increased from 85 full-time personnel per 10 square miles in 1990 to 101 per 10 square miles in 2000. From 1990 to 2000 the average number of full-time sworn personnel among police departments in large cities increased by 17%, from 2,101 to 2,465. The NYPD had more than 40,000 full-time sworn personnel as of June 2000. This was about 3 times the next largest department in Chicago, with 13,466. In terms of land area, employment increased from 66 full-time sworn personnel per 10 square miles in 1990 to 78 per 10 square miles in 2000. In 2000 these agencies employed 310 full-time sworn personnel per 100,000 residents. This represented an increase of 21 per 100,000, or 7%, compared to 1990. The Washington (DC) Metropolitan Police Department continued to have the highest ratio, with 631 officers per 100,000 residents in 2000, although this was down 15% from the 1990 ratio of 742 per 10,000. Other large city police departments with at least 500 full-time officers per 100,000 residents in 2000 were the Newark (NJ) Police (536) and the NYPD (505). Minority and female representation Minority representation among local police officers in large cities increased from 29.8% in 1990 to 38.1% in 2000. Hispanics recorded the greatest increase, from 9.2% to 14.1%. Black representation also increased, from 18.4% in 1990 to 20.1% in 2000. Asian and Pacific Islander representation was 2.8% in 2000, up from 2.0% in 1990. The percentage of female officers rose from 12.1% in 1990 to 16.3% in 2000. Using a ratio based on the percentage of sworn personnel who were members of a racial or ethnic minority relative to the percentage of city residents who were members of that minority group indicates that, on average, police departments in large cities were slightly more representative of the cities they served in 2000 than in 1990. From 1990 to 2000, the average ratio increased from .59 to .63 for minorities overall. That is, on average, police departments in large cities had 63 minority police officers for every 100 minority residents in 2000, compared to 59 for every 100 in 1990. For blacks or African Americans, the average ratio increased from .64 in 1990 to .74 in 2000, for Hispanics or Latinos from .54 to .56, and for other minority groups (such as Asians and American Indians) from .26 to .37. Education and training requirements The percentage of police departments in large cities that required a 4-year degree of new officers rose from 1.6% in 1990 to 4.8% in 2000. The percentage requiring a 2-year degree increased from 4.8% to 9.7% during this time, and the percentage with a non-degree college requirement went from 12.9% to 22.6%. Overall, about twice as many departments had some type of college education requirement for new officers in 2000 (37.1%) as did in 1990 (19.3%). From 1990 to 2000, the median number of classroom training hours required of new officer recruits in police departments in large cities increased from 760 to 880. The median field training requirement also increased during this time -- from 520 hours to 600 hours. In 2000, police departments in large cities had a median annual in-service training requirement for officers of 40 hours (data were not collected in 1990). Operating budgets The operating budgets of police departments serving cities with a population of 250,000 or more totaled about $13.1 billion in fiscal 2000. Controlling for inflation (see methodological note on page 15), this was 20% more than in 1990. The per agency average was nearly $212 million in 2000 compared to about $176 million in 1990. Operating budgets for 2000 totaled $266 per resident, $24 higher than in 1990. Per employee operating costs were $64,323 for 2000, about the same as in 1990 ($64,493). The overall operating budget per sworn officer increased 2% during this period, from $83,814 to $85,786. In 2000, departments serving 1 million or more residents had the highest per resident operating budget, $290. Departments serving a population of 350,000 to 499,999 had the highest per employee ($73,811), and per officer ($102,539) budgets. Salaries and special pay From 1990 to 2000, the average base starting salary for police chiefs in cities with a population of 250,000 or more increased by about 2%, from $95,393 to $97,215. The average starting salary for chiefs was highest in departments serving 500,000 or more residents -- about $105,500. Average starting salaries for sergeants or equivalent first-line supervisors increased by 3% from 1990 ($49,081) to 2000 ($50,541), while starting salaries for entry-level officers declined by about 1%, from $35,002 to $34,556. From 1990 to 2000, the percentage of police departments in large cities that offered shift differential pay to officers increased from 66% to 76%. Nearly all (93%) of the departments serving 250,000 to 349,999 residents offered this type of special pay in 2000. About 3 in 5 departments offered education incentive pay in 1990 (60%) and 2000 (61%). Those serving 250,000 to 349,999 residents (71%) were the most likely to offer it in 2000. From 1990 to 2000, the percentage of departments offering hazardous duty pay declined slightly, from 56% to 50%. In 2000, a majority of the departments serving 500,000 to 999,999 residents (57%) offered this type of special pay. The percentage of departments with merit pay in 2000 (34%) was about the same as in 1990 (35%). Those serving 1 million or more residents (50%) or 350,000 to 499,999 residents (47%) were the most likely to offer merit pay. UCR violent crimes From 1990 to 2000, according to the FBI's Uniform Crime Reports (UCR), the number of violent Crime Index offenses (murder, forcible rape, robbery, and aggravated assault) reported to police departments in cities with 250,000 or more residents declined 34% -- from an average of 13,091 per department to 8,686. The drop in the rate of violent crimes per 100,000 residents was even larger, falling from a rate of 1,802 violent crimes per 100,000 residents in 1990, to 1,091 per 100,000 in 2000 -- a reduction of 39%. Cities with 1 million or more residents had the largest decrease (46%), led by a 60% drop in New York City. A large reduction in violent crime also occurred in the number reported per officer employed. In 2000 there were 353 violent crimes per 100 sworn personnel, 43% fewer than the 1990 rate of 623 per 100. Among cities with 1 million or more residents, the decline was 51%, including 57% in New York. By specific type of violent crime, the number of murders per 100,000 residents declined from an average of 21 per city in 1990 to 13 in 2000; forcible rapes, from 84 to 52; robberies, from 671 to 399; and aggravated assaults, from 786 to 603. UCR property crimes From 1990 to 2000, the number of UCR property Crime Index offenses (larceny-theft, burglary, motor vehicle theft) in large cities decreased 31%, from an average of 60,563 per city to 41,571. In 2000, there were 5,221 property crimes per 100,000 residents compared to 8,352 in 1990, a decrease of 37%. In cities with 1 million or more residents, the drop was 46%, led by New York (64%). There were 41% fewer reported UCR property crimes per 100 sworn personnel in 2000 (1,692) than in 1990 (2,887). Departments serving 1 million or more residents reported a drop in property crimes per 100 officers that exceeded 50% (from 2,425 to 1,186), with a 69% drop in New York. The burglary rate in cities with a population of 250,000 or more declined from an average of 2,122 per 100,000 residents in 1990, to 1,167 per 100,000 residents in 2000. For larceny-theft, the average rate declined from 5,082 to 3,798, and for motor vehicle theft from 1,529 to 1,005. Special units Police departments in large cities operated various types of special units to address crime-related problems and to provide resources to those affected by crime. In 1990 and 2000, more than three-fourths of departments had full-time special units or part-time personnel assigned to general crime prevention, child abuse, juvenile crime, drug education in schools, missing children, drunk drivers, or gangs. In 2000 a majority of departments had personnel assigned full time to special units for gangs (84%), domestic violence (81%), child abuse (77%), crime prevention (76%), drug education in schools (73%), juvenile crime (68%), and missing children (66%). Seventy-one percent of departments operated a full-time unit or had part-time personnel assigned to victim assistance in 2000 compared to 45% in 1990. The percentage with full-time victim assistance units rose from 32% to 47% during this time. Ninety-seven percent of departments operated either a full-time domestic violence unit or had personnel assigned part-time to deal with the problem in 2000, compared to 61% in 1990. The percentage with full-time units increased from 50% to 81% during this time. From 1990 to 2000, the percentage with personnel assigned on at least a part-time basis to address gang-related problems rose from 89% to 98%, and the percentage with a full-time gang unit increased from 69% to 84%. The percentage of departments with personnel assigned at least part-time to handle bias-related crimes increased from 58% in 1990 to 71% in 2000; however, the percentage with a full-time unit dropped from 34% to 26% during this time. ---------------------------------------------------------- Community policing initiatives of police departments in large cities, 2000 Since the enactment of the 1994 Crime Act and the subsequent creation of the Department of Justice's Office of Community Oriented Policing Services (COPS), most police departments have taken steps to implement community policing. The COPS office facilitated this effort by providing funding to deploy community policing officers, facilitate problem solving efforts, encourage interactions with communities by officers, promote innovations in policing, and enhance existing technologies. Although such data were not collected in 1990, the 2000 LEMAS survey provided several measures of large city police departments' community policing efforts. For example, 94% of the departments provided all new officer recruits with at least 8 hours of community policing training during the 12-month period ending June 30, 2000. Also, nearly all departments had full-time sworn personnel serving as community policing officers with a mean of 21% and a median of 5% of all officers so assigned. Seventy-one percent of departments had a formal, written community policing plan, while 29% had an informal plan. Other indicators of community policing in large cities: Type of community policing activity during 12-month period ending June 30, 2000 Percent of agencies Met at least quarterly with citizen groups to discuss crime-related problems 100% Gave patrol officers responsibility for specific geographic areas/beats 90% Conducted a citizen police academy 87% Assigned detectives to cases based on geographic areas/beats 84% Trained citizens in community policing techniques such as community mobilization and problem solving 79% Actively encouraged patrol officers to engage in SARA-type problem-solving projects on their beats 77% Upgraded technology to support community policing 76% Conducted or sponsored a survey of citizens on crime or police-related topics 63% Formed problem-solving partnerships through specialized contracts or written agreements 61% ---------------------------------------------------------- Drug enforcement All but 1 of the 62 local police departments serving cities with a population of 250,000 or more reported in the 2000 LEMAS survey that they had receipts from a drug asset forfeiture program during the prior year, as did 60 departments in the 1990 survey. Although data on the value of the goods, money, and property received were not collected in 1990, it totalled an estimated $108.2 million in 1999, or about $708 per sworn officer. Estimated value of drug asset forfeiture receipts, police departments serving cities with a population of 250,000 or more, 1999 Population Total Per served (in millions) officer All sizes $108.2 $708 1,000,000 or more $58.7 $674 500,000 - 999,999 25.0 657 350,000 - 499,999 14.0 845 250,000 - 349,999 10.5 939 Departments had an average of 15 full-time officers each assigned to a multi-agency drug enforcement task force in 2000, compared to 13 in 1990. As a percentage of all officers, the average decreased slightly, from 1.0% in 1990 to 0.9% in 2000. From 1990 to 2000, departments serving a population of 1 million or more increased their average number of officers assigned to a task force from an average of 21, or 0.3% of all officers, to an average of 48, or 0.7%. The average number of officers per department assigned to a special unit for drug enforcement increased from 86 in 1990 to 123 in 2000; however, the average percentage of all officers so assigned in 2000 (3.5%) was lower than in 1990 (4.4%). The largest drop was among departments serving 350,000 to 499,999 residents -- from an average of 6.3% of officers in 1990 to 3.4% in 2000. Equipment In 1990 nearly all police departments serving 250,000 or more residents authorized the use of both semiautomatic (98%) and revolver (97%) sidearms; however, the percentage authorizing revolvers had dropped to 65% by 2000. All departments authorized semiautomatic sidearms in 2000. From 1990 to 2000, the percentage of police departments in large cities requiring all patrol officers to wear protective body armor increased from 21% to 48%. During this time, the percentage of departments that required at least some patrol officers to wear armor rose from 31% to 69%. Departments serving 250,000 to 349,999 residents (43%) were the least likely to have a body armor requirement for patrol officers during 2000. Vehicles In 2000, 29% of police departments in large cities operated airplanes, about the same percentage as in 1990 (26%). Nearly all departments with planes had just one, with no more than four operated by any department. In 2000 those serving a population of 1 million or more (50%) were the most likely to operate an airplane. Sixty-six percent of departments operated at least one helicopter in 2000, compared to 55% in 1990. The median number of helicopters operated was 2 with a maximum of 19. Ninety percent of departments serving a population of 1 million or more operated helicopters during 2000. From 1990 to 2000, the percentage of departments using boats increased from 48% to 58%. Among those using boats in 2000, the median number operated was 3, and the maximum was 27. About three-fourths of the departments serving a population of 500,000 or more used boats during 2000. Among land vehicles, motorcycle use increased slightly, from 90% of departments in 1990 to 95% in 2000. A much more substantial increase was observed for bicycles, with 98% of departments using them in 2000, compared to 39% in 1990. On average, departments operated 44 bicycles per 1,000 sworn personnel in 2000 compared to 3 per 1,000 in 1990. Police departments in large cities operated 307 marked cars per 1,000 sworn personnel in 2000, 17% more than in 1990 (263). The ratio of un- marked cars also increased, from 188 per 1,000 sworn personnel in 1990 to 207 per 1,000 in 2000. The percentage of departments allowing officers to drive marked vehicles between work and home increased from 39% in 1990 to 55% in 2000. However, the percentage of departments allowing officers to use these vehicles for personal errands decreased from 24% to 13%. Computers and information systems The increase in computerization among police departments in large cities from 1990 to 2000 was evident in a variety of areas. For example, all departments were using computer-aided dispatch systems in 2000, compared to 90% in 1990. Participation in enhanced 9-1-1 emergency systems, capable of pinpointing a caller's location automatically, increased from 76% of departments in 1990 to 97% in 2000. In 2000, all departments serving 250,000 to 999,999 residents were using enhanced 9-1-1. Increases in the percentage of departments using in-field computers or terminals were also observed. Ninety-two percent of police departments in large cities used them in 2000, compared to 73% in 1990. All departments serving 500,000 or more residents were using in-field computers or terminals during 2000. The use of Automated Fingerprint Identification Systems (AFIS) rose from 60% of departments in 1990 to 97% in 2000. In 2000 all departments serving 350,000 or more residents had AFIS access. The percentage of departments with exclusive or shared ownership of an AFIS system increased from 57% to 71% during this time. In terms of the technological measures provided by LEMAS, nearly all police departments in large cities had advanced capabilities in 2000. This included the use of computer-aided dispatch, enhanced 9-1-1, automated fingerprint identification, and in-field computers. Methodological notes A large part of the data used in this report are from the Law Enforcement Management and Administrative Statistics (LEMAS) surveys conducted in 1990 and 2000. Population data are from the U.S. Census Bureau decennial census, and crime data are from the FBI's Uniform Crime Reports (UCR). Certain methodological issues arose during analysis and are discussed below. Jurisdictional issues The 62 cities included in this report were chosen because they had a population of at least 250,000 in both the 1990 and 2000 decennial censuses. In most cases the data used represent a city and the local police department that serves it. For certain cities, special circumstances existed and are noted below. Anaheim (CA) The Anaheim (CA) Police did not respond to the 1990 LEMAS survey. Although personnel counts were obtained from other sources, other values for 1990 were based on the agency's responses to the 1987 and 1993 surveys. Where data values for categorical variables differed, the data for 1993 were used. Continous variables such as number of personnel, operating budget, number of vehicles, and asset forfeiture receipts were averaged for the two years to arrive at an estimate for 1990. Charlotte (NC) The Charlotte Police and Mecklenberg County Police merged on October 1, 1993. To provide comparability with 2000 data for the Charlotte-Mecklenberg Police, the data reported by the two original agencies for 1990 were aggregated for continuous variables. For categorical variables in 1990, data from the Charlotte Police were used. Honolulu (HI) The jurisdiction of the Honolulu Police covers the City and County of Honolulu. Therefore, county population data were used for this report. Indianapolis (IN) The city of Indianapolis is served by both the Indianapolis Police and the Marion County Sheriff's Department. Therefore, data from these two agencies were combined for both 1990 and 2000. For categorical variables, data from the Indianapolis Police were used, and for continuous variables, the responses of the two agencies were aggregated. Las Vegas (NV) The city of Las Vegas is under the law enforcement jurisdiction of the Las Vegas Metropolitan Police Department. (LVMPD). The jurisdiction of the LVMPD extends beyond the city of Las Vegas to include significant portions of suburban Clark County. General population counts were available for the actual LVMPD jurisdictional area, however; race and ethnicity were not. Therefore, the population race and ethnicity data used are for all of Clark County. This includes the city of Las Vegas, and county areas inside and outside of LVMPD law enforcement jurisdiction. New York (NY) The New York City transit and housing police agencies were merged into the New York City Police Department (NYPD) on April 30, 1995. To provide comparability with the 2000 data reported by NYPD, data reported by the transit and housing police for 1990 were combined with NYPD data. For categorical variables, the responses of the NYPD were used. For continuous variables, data for the three original agencies were aggregated. Comparability issues All items that could be compared between the 1990 and 2000 surveys were included for analysis. Certain items asked in both surveys could not be included because of comparability issues. In some cases, items were included when the design of the questions had changed because it was determined to not affect comparability. These items included the following: Training requirements The 1990 LEMAS survey asked for the number of classroom and field training hours required while the 2000 survey asked for the number of academy and field hours. Operating costs The 1990 survey asked for the annual operating expenditure while the 2000 survey asked for the operating budget. All monetary data were converted to 2000 dollars by multiplying them by 1.3393. This factor was derived from annual Consumer Price Index averages published by the Bureau of Labor Statistics, Bureau of Economic Analysis. UCR Crime data Changes in reporting methods at the State or local levels may affect the comparability of crime data for 1990 and 2000. This has been documented for Tennessee, including the cities of Nashville and Memphis, which appear to have underreported certain crimes in 1990. Comparability issues may exist in other jurisdictions as well, documented or not. Data on the number of forcible rapes in Chicago were not available for 1990 or 2000. Estimates based on the number of other violent crimes were calculated for this report. When comparing crime rates between jurisdictions, the FBI suggests considering factors such as population density, degree of urbanization, demographic composition of the population, stability of the population, transportation modes and systems, economic conditions, cultural factors, family conditions, climate, effective strength of law enforcement agencies, administrative and investigative emphasis of law enforcement, criminal justice system policies, citizen attitudes, and citizen crime reporting practices. For more information on the Uniform Crime Reports see . Special units The 1990 survey asked if the agency operated a special unit with personnel assigned either full-time or part-time. The 2000 survey asked if the agency operated a unit with personnel assigned full-time or had specially designated personnel who dealt with the issue on a part-time basis. In-field computers The 1990 survey included the categories of laptop, car-mounted digital terminal, hand-held digital terminal and "other". The 2000 survey included the four specific categories of laptop, mobile digital/data computer, mobile digital/data terminal, and "other" within two general categories of vehicle-mounted and portable. Minority representation ratio The officer-to-resident ratios used to measure minority representation in figure 1 and appendix table B were calculated by dividing the percent of an agency's full-time sworn personnel who were members of a racial or ethnic group by the percentage in the population served who belonged to that group. In some instances this resulted in a value greater than 1 (that is, a group was over represented). Such values are included in table B; however, for the purposes of calculating the average ratios used in figure 1, these values were truncated to 1.00. This was done so that average ratios would not be artificially inflated by overrepresentation in certain agencies. By definition the overrepresentation of one group is accompanied by the under representation of one or more other groups. Ideally, all groups would have an officer-to-resident ratio of 1. ---------------------------------------------------------- The Bureau of Justice Statistics is the statistical agency of the U.S. Department of Justice. Lawrence A. Greenfeld is acting director. Brian Reaves and Matthew Hickman, BJS statisticians, prepared this report. The Office of Community Oriented Policing Services (COPS) provided partial funding for the 2000 LEMAS survey. Debra Cohen and Matthew Scheider of COPS assisted with questionnaire development. More information on COPS can be obtained at their website . Data were collected and processed by the U.S. Census Bureau under the supervision of Latrice Brogsdale-Davis and Charlene Sebold. Project staff included Theresa Reitz, Martha Greene, Patricia Torreyson, Bill Bryner, and Paula Kinard. Jayne Robinson, BJS, administered final report production. May 2002, NCJ 173703 End of file th 5/09/02