Printed Report
44 pages
Price: $6.95

Table of Contents

Crime in the United States has become a concern for much of the population. A major crime law that increases resources avail-able to combat crime has recently been enacted. Various means and techniques, such as increasing reliance on private police, community policing, and problem-oriented policing, are also be-ing utilized to try to reduce crime and increase citizens’ sense of security. This report will begin with an analysis of crime in Kansas and then evaluate the various means that might be used to help ameliorate the crime problem in a cost-effective and equitable manner.


Crime in Kansas generally increased between 1983 and 1992, going from 46 crimes per 1,000 population in 1983 to 52.8 per 1,000 in 1992.[1] In particular, violent crimes increased 62.3 per­cent over that period; property crimes, 18.2 percent; and total crimes, 21.5 percent. Except for murder and manslaughter, all violent crimes, including rape, robbery, and assault/battery, ex­perienced large percentage increases. Burglary and motor vehi­cle theft increased 53.8 percent and 64.9 percent, respectively.

On a comparative basis, Kansas in 1992 had 5,320 crimes per 100,000 population, slightly below the national average of 5,660.[2] It was essentially in the middle range of all states, ranking 22 (where higher numbers signify higher crime rates). The violent-crime rate in Kansas of 511 per 100,000 population was also be-low the U.S. average of 758. In terms of growth in crime, crimes in Kansas increased 9.01 percent between 1988 and 1992, whereas the United States experienced a slight decrease over the same period (0.07 percent). In fact, Kansas had the 13th-worst state performance in this area. Violent crimes in Kansas increased more than twice the U.S. average (39.95 versus 18.92 percent), and murders increased 76.47 percent in Kansas, com-pared to 10.71 percent in the United States. Kansas ranked ninth-worst in violent-crime increases and was the worst state of all in terms of the increase in murder.

Kansas also performed worse in every category of crime than did a comparable state, Minnesota. For example, Kansas’s 1992 overall crime rate and violent-crime rate were 15.9 percent and 51.2 percent, respectively, above those of Minnesota. The in­crease in crime between 1988 and 1992 in Kansas exceeded that of Minnesota by 41 percent; violent crime grew in Kansas 141 percent more than in Minnesota, and murder grew by 444.5 percent more. Clearly, Kansas is experiencing an increasing crime rate relative to both a comparable state, Minnesota, and the United States as a whole.

Police Officers in Kansas and the United States

Sworn officers in Kansas, excluding state agencies, increased from 3,936 to 5,023 between 1983 and 1992,[3] and all personnel increased from 5,314 to 6,863. Crimes per sworn officer went from 27.9 to 24.5. The percentage increase in sworn officers over this period was 27.6, slightly in excess of the reported in­crease in crime, which accounted for the decrease in crimes per sworn officer. If all employees were considered, the decrease would be slightly greater since the percentage increase in all employees was 29.1. Kansas is evidently shifting some small number of activities from sworn to civilian personnel. That point is confirmed by examination of the percentage sworn in, for ex­ample, between 1985 and 1992 in all police departments in Kansas (excluding state agencies). In 1988 75.4 percent of all employees were sworn in, whereas in 1992 the percentage had declined slightly, to 73.2.[4]

On a comparative basis, Kansas in 1992 had slightly fewer sworn officers per 10,000 population than the United States as a whole. The numbers are 22 and 24, respectively.[5] Interestingly, Minnesota had only 16 sworn officers per 10,000 population, and as mentioned previously, had a much lower crime rate than Kansas.

In summary, crime in Kansas has grown rapidly since the mid-1980s relative both to Minnesota and to the United States as a whole. However, the crime level in Kansas in 1992 was still below the U.S. average.

Privatization: Definition and Growth

A major phenomenon in modern law enforcement has been the increase in private policing as opposed to public policing. In 1990 about 2.5 times as many were employed by the private sec­tor to deter and prevent criminal behavior as were employed by the public sector.[6] Specifically, private security employed 1.5 million and had spending of $52 billion in 1990, compared to 600,000 and $30 billion for public policing. Private security is also growing much faster than public policing. For example, the aver­age annual growth in private security employment between 1980 and 1990 was 4 percent, much faster than the approximately 2.5 percent experienced between 1986 and 1992 for public police department employment.[7] One widely accepted definition of private security is the following:

Private security includes those self employed individuals and privately funded business entities and organizations providing security-related services to specific clienteles for a fee, for the individual or entity that retains or em­ploys them, or for themselves, in order to protect their persons, private property, or interests from various hazards.[8]

The term “private police” includes guards, alarm responders, watch people, investigators, private detectives, and store detec­tives, among others. They have tended to focus on activities such as computer fraud and industrial espionage, which the pub­lic police have tended to downplay in comparison to violent criminal activity. Private security employment has grown so rapidly in part because of income growth. As income grows, demand for security, like demand for any normal good, increases. Since public supply has not increased much, because of tight public budgets, households and others have turned to private security services. In addition, the tight public budgets have themselves encouraged governments to examine the fea­sibility and desirability of privatizing some governmental ser­vices, including police services. This report will evaluate the police functions and activities that could be shifted to the pri­vate sector. It will further suggest the extent to which the public sector needs to monitor the private performance.

The shift toward private policing has occurred in part because of the great demands placed upon the public police and the consequent necessity to develop priorities for use of the avail­able resources. Chiefs of police and sheriffs ranked shoplifters 11th (next to last) and employee theft 12th (last) in terms of prior­ity responses, which is indicative of the low priority attached to protection of business assets.[9] Given the prevailing situation, it is not surprising that businesses have greatly increased use of pri­vate security rather than relying upon the public police.

Another factor contributing to growth in demand for private security is the desire of businesses to avoid unfavorable public­ity. For example, even though an employee skimmed $5.7 million in kickbacks and commissions from his employer, the firm ini­tially chose not even to sue but simply to obtain restitution.[10] In general, going public by using the public police could upset stockholders, among other undesirable consequences. Moreover, businesses may be better able to obtain restitution by not prosecuting. In any event, the desire to avoid publicity and simply recover their losses often leads businesses to rely upon private security to investigate and handle instances of employee theft.

Advantages of Markets

In cases where a market could exist, the benefits of compe­tition could accrue. For example, in the case of hotel security or burglar alarm response, enough firms might be able to offer the service that the price would equal marginal and average cost; that is, it would be just high enough to cover the costs of providing the service and still allow the providers a competitive return. Privatization also means that a business could purchase just the services it wanted, instead of having to confront a monopoly supplier, the public police agency, that will largely be immune to market forces. Competition also provides a powerful force en­couraging efficiency, cost reductions, and innovation. Indeed, the rapid growth of private policing may have occurred in part because of the advantages it provides.

Requirements for Competition and Markets to Work

Markets and competition will work and provide optimal results when the following conditions are met: economies of scale are modest; entry barriers are low; public-good conditions are ab-sent, that is, the ability exists to exclude those unwilling to pay; and external benefits and costs are absent. Insignificant economies of scale mean that enough firms can efficiently offer the service to avoid monopoly conditions. Low entry barri-ers result in prices that do not substantially exceed the competi-tive level of marginal and average costs. If prices are higher than costs, new entrants will be attracted to the industry, and their entry will drive prices down to the competitive level. A public good means that consumers cannot be excluded from enjoying its benefits even when they are unwilling to pay. Consumers who do not pay are termed “free riders.” Public goods include a light-house that warns all boats, traffic signals, and national defense. The absence of externalities means that the participants in the market bear all the costs of and receive all the benefits from consumption of the good.

It is important to mention that even though a market may fail to provide optimal results, government intervention will not nec­essarily improve the situation. Suboptimal private provision may still be superior to public provision, which often involves ineffi­ciency and other problems. Public provision normally entails a government monopoly that has a not-for-profit motivation (for example, not operating as a profit center) but is susceptible to bureaucratic and political pressure.

Private providers tend to operate as profit maximizers and avoid social considerations, whereas public agencies are expected to ensure that overall social welfare is attained. In real­ity, however, the theoretical considerations do not necessarily hold. Public agencies, including the police, generally neither maximize social welfare nor attain maximum efficiency. Following are some of the reasons why:

1. Public agencies operate almost like clubs. Interest groups and coalitions made among representatives yield policies that often benefit subsets of the population.

2. Externalities that spill over a jurisdiction’s geographical boundaries are ignored.

3. The monopolistic position of the police enables higher prices to be paid for inputs than market conditions warrant (for example, police officer wages). The non­competitive stance of public providers creates exces­sive power for labor unions that represent government workers.

4. The noncompetitive position of the police enables policy makers and officers to be unresponsive to con­stituent desires. Indeed, studies show that as the popu­lations of political jurisdictions rise, police officers act in a more arbitrary fashion toward local residents.[11]

Police Functions

Let us now examine the functions performed by the police to determine which of them meet the above criteria for private provision. We do not suggest that all police services could be pri­vatized, because of the probable advantages of coordination between, for example, generalized patrol and investigation. Such coordination may yield economies of scope. However, it is quite possible that some or even many current police services could be provided privately.

Police functions can be grouped under the following six cate­gories: law enforcement, public order maintenance, patrol, traf­fic enforcement, general assistance, and administrative/sup­port work.[12] Law enforcement includes such activities as responding to crimes, investigating crimes, arresting suspects, and helping prosecutors in their work. Public order maintenance involves such activities as removal of abandoned cars, controlling crowds, and assisting in emergency and disaster control. Patrol involves such activities as foot and car patrol, being cognizant of wanted criminals and stolen properties, and order maintenance. Traffic enforcement includes such activities as stopping suspicious vehicles and vehicles violating traffic rules, and investigation of vehicular accidents. General assistance involves such activities as helping lost children, informing businesses and residents about how to protect their property, escorting funerals, answer-ing telephone calls for general information, and helping those locked out of cars or homes. Administrative/support work includes investigation of citizen complaints; conducting back-ground investigation of job applicants; taking reports of things such as incidents, missing persons, and confessions; and keep-ing records of things such as alarm registration and associated violations.

Police Functions and Market Failure

Law enforcement has aspects of a public good because apprehension of a criminal provides external benefits to others, resulting in some being unwilling to pay for such a service. The benefits accrue to a large number of people, and no one can be effectively excluded from enjoying the benefits of public goods even when one does not pay.

Economies of scale can also be found in law enforcement. A small police department might not be able to have an expert and sophisticated lab equipment for homicide investigations. In Bucks County, Pennsylvania, the county government provides investigations and lab equipment for investigations of homicides and other personal crimes. Moreover, some communities that are too small to overcome the threshold for entry for police ser-vices rely upon the state police (or in some states, the sheriff’s department). An entity that can achieve economies of scale will be able to provide law enforcement services more cost-effec-tively than entities that are too small. A reasonable funding approach might have each participating jurisdiction contribute according to its relative population size. Further, the fact that public funding must exist does not mean that public supply is required. Public order maintenance also is a kind of public good. Few would be willing to pay for crowd control, since the benefits accrue to everyone. Similarly, removing abandoned cars, unless the cars are on private property, is unlikely to be provided unless funded by the public treasury.

Patrol is a public good since the entire community is pro­tected, even those who are unwilling to pay. Patrol also exhibits economies of scale in terms of density. It is more cost-effective to protect a whole neighborhood, whether per home or per resi­dent, than to protect every third home.

Traffic enforcement has many aspects of a public good since all benefit if dangerous drivers are removed from the highway. However, accident investigation, especially of minor accidents, may well be a service that does not require public provision or funding. Insurance companies already investigate certain aspects of automobile accidents. They could extend their ser-vices to providing private investigation of traffic accidents.

General assistance mainly involves functions that are not public goods. For example, helping people enter locked vehicles is a service that could be priced, and indeed is currently being provided also by private companies. Escorting oversized vehicles or funerals is also a service that could be priced, which means that it could be supplied privately. Services that could be priced could also be provided by the public police for a fee. Such user charges are commonly employed for other government services where exclusion of nonpayers is possible.

Most administrative/support work is not a public good and could be supplied privately. For example, recording of citizen complaints and investigation of job applicants could be handled privately.

In conclusion, the private market will fail to provide the opti­mal amount of law enforcement, public order maintenance, patrol, and traffic enforcement. However, specific services within the above functions could be priced and hence could be supplied privately. Examples include alarm response and patrol within a defined area. Further, even if the market does not pro­vide the optimal amount of services, government provision is not necessarily superior. Government provision can mean ineffi­ciency and the potential for abuse of citizens. Accordingly, reliance on the private market may be desirable even in the case of market failure.


Services that have the potential to be priced should be considered candidates for private provision or user charges.

Physical and Other Requirements for Police Officers

Another approach to determining which police services are candidates for privatization is to examine the physical and intel­lectual requirements for police officers. Public security, or policing, generally involves the full range of tasks of law enforcement. Accordingly, the skill and training required for public policing are generally greater than that required for most categories of private policing. A public police officer is expected to have the intelligence, training, and physical endurance to apprehend a violent, fleeing felon even though the task may occur rarely.[13] Other examples of tasks with high skill require­ments include serving search warrants, investigating undercover, and handling street fights.[14]

Most Tasks of Police Officers Are Not Crime Related or Do Not Require Their Level of Skill

On the other hand, many tasks of the public police are rela-tively simple and do not require highly skilled and intelligent individuals. Serving as an escort for a funeral procession or directing traffic around an accident are fairly simple tasks. Not surprisingly, these tasks are often contracted out to the private sector. Other examples include parking meter enforcement, building security, airport security, and dispatching police vehi-cles. Occasionally such tasks are retained publicly, but they have been relegated to less-skilled, and lower-paid nonsworn employ-ees. Privatization has the advantage of inducing competition into the process. Some tasks that require high intelligence or skill but not physical strength have also been privatized. They include detection of embezzlement using electronic means. Protecting store property from shoplifters is a service that is usually pro-vided privately. The skill level for that task is not as great as for public policing. In addition, that service can be sold because exclusion of free riders is possible.

In some cases, private policing requires essentially the same skills as public policing. Examples include campus and transit security, where the police officers can exercise the same degree of authority as the public police and are subject to the same problems and crimes; they differ only in the territory cov­ered.

Privatization has occurred in part because the public police have undertaken many tasks that are not crime related. In fact, studies of police work in the late 1970s found that only about 20 percent was crime related.[15] Much of their work entailed helping the homeless, handling abandoned vehicles, guarding parking lots, escorting funerals, and searching for runaway children. Given tight governmental budgets, privatizing the handling of such noncriminal functions is attractive. Noninjury, and possibly all, automobile accident investigation is another example of a generally noncrime function that could readily be accomplished by nonsworn personnel.

The Kansas City, Missouri, Police Department has determined that private security could be contracted to perform many non­crime functions. Examples include traffic and crowd control for special events, response to intrusion alarms, and such functions as directing traffic when traffic lights malfunction and providing standby protection for an owner whose property has an open window or door.[16] These functions do not require the highly skilled and well-trained public police, so contracting them out could save on the use of society’s scarce resources. Economizing in the use of public police is especially important in a place like Kansas City, where calls increased 61 percent since the early 1970s while the number of officers decreased 12 per­cent. Part of the problem of great demand for service arises from the lack of pricing for such service. If the demanders had to pay a price reflecting the using up of resources in providing service, only service yielding that amount of satisfaction would be demanded.


Consideration should be given to privatizing tasks that do not require the full range of skills of public police officers. Not only would savings be obtained, but police officers would become more available for performing the tasks only they can perform.

Examples of Privatization

Alarm Systems. An example of a function that can be provided privately is response to alarms. The beneficiaries are distinct, and the benefits accrue directly to the owners of the sites pro­tected by the alarm systems. The absence of externalities also means that the service could be either provided privately or financed by user charges. Demanders could pay for the service as they do for AAA car service. Moreover, 95 to 99 percent of alarm calls are found to be false, and alarm calls constitute from 10 to 30 percent of total police time and calls for service.[17] The public police have discontinued response to alarms in some areas because of the strain on police resources. For example, the Toronto, Ontario, Police Department does not respond beyond the fourth false activation in a year. The police depart­ments of Los Angeles, Chicago, Philadelphia, and New York City effectively respond only to activations at high-risk establish­ments such as banks, jewelery stores, and government facili­ties.[18] In fact, the public police would prefer to shed the service of responding to burglar alarms entirely.

The problem arises in part from incorrect pricing. Alarm owners could be charged for the service of responding to false alarms. Alarm owners are generally allowed three false activa-tions per year before they incur charges. If alarm owners had to pay for all false activations, they would have an added incentive to try to reduce the incidence of false alarms through greater care, and alarm manufacturers would have an incentive to build systems that are more user friendly and less likely to produce false alarms. In effect, the community at large, which includes many people who do not own alarms, is paying for the cost of responding to the false alarms. Fees that are lower than costs, including free responses, encourage an excessive number of alarm systems, since the users are not paying for all the costs they cause.

There has been a substantial increase in the number of alarm systems. The percentage of residences with alarms has been increasing 11 percent annually. The number of residences with alarms grew from about 2 percent in 1980 to 10 percent in 1992. Commercial alarm ownership in 1992 was 14 percent.[19] Moreover, installation for alarm systems has declined to prices as low as $99 and is sometimes even free. Given the prospect of even more rapid increases in the number of installed alarm systems in the future—because of price reductions on alarm systems—the pressure to do some­thing to ameliorate the problem of false activations is likely to become even greater.

An example of private response to alarm activations comes from a firm in southern California that offers such a service.[20] The firm generally responds first to the activation and, if neces-sary, requests police assistance. This approach has reduced response time, freed up public police resources, and improved the quality of alarm services. It also has overcame the liability insur-ance problem. The private response team uses a radio/cellular device to contact the police if there is evidence of an actual burglary. In the absence of confrontations with burglars, the guards do not have to carry guns, so the firm’s insurance premi-ums are low. It should be emphasized that unless drastic actions are taken to reduce the number of false activations, we will wit-ness a virtual elimination of police department response to alarm activations.

The alarm example shows how privatization could be used to make more efficient use of scarce resources when a service does not require government involvement. Private alarm response satisfies the desirable economic condition whereby the direct beneficiaries pay for the service they receive. Another important privatization element in burglar alarms is the substitu-tion of private capital—burglar alarms and associated devices—for public police labor. Alarms enable officers to spend less time patrolling highly alarmed neighborhoods. Also, less police time has to be spent ensuring that commercial and industrial entrances are secure. Indeed the Portland, Oregon, Police Department encourages businesses and households to install alarms. There is even a proposal for the city to subsidize the purchase of alarms by low-income households.


Services such as response to alarms could be provided privately. In any case, the owners of alarms should pay for the services they demand.

Private Policing in Starrett City

An example of privatization of police services within a given area comes from Starrett City, a 46-building apartment complex in Brooklyn, New York, which currently has about 20,000 resi­dents.[21] Starrett City, situated on 153 acres, is surrounded by high-crime neighborhoods. The apartment management created its own unarmed police force, which in the early 1990s included 59 guards. The average security officer is 39 years old, is a high school graduate, and has been employed at Starrett City for five years. As a result of their being considered special patrol offi­cers, the private police of Starrett City have exercised since 1978 full police powers while on duty at Starrett City.

The security unit has emphasized and practiced proactive prevention of crime rather than mainly responding to citizen requests for assistance (being reactive). For example, they check parking lots, test doors and windows, follow suspicious persons, and patrol on foot. One New York City police officer with 21 years of experience complained that the public police merely travel from one incident to another. He admired the work of Starrett City private police and noted that in the distant past, the public police did the same kind of work.[22]

Private policing in Starrett City has yielded quite good results. In 1985, for example, the felony rate was 7.25 per 1,000 resi­dents, compared to 83.5 for all of New York City. The public police have high regard for the security force of Starrett City and refer to them as Starrett City cops. The residents of Starrett City believe that they are safe within the complex. In fact, 90 percent of the 2,862 residents who responded to the questionnaire said they were at least somewhat safe being alone in the daytime in Starrett City, compared to 41 percent feeling safe in the same circumstances in the surrounding area.[23] The residents believe further that the Starrett City private police are responsible for the low crime rate and their feeling of safety. In fact, about 51 per-cent of respondents stated that they would move if Starrett City did not have its private police. Residents made substantial use of the force; in 1985 the security unit answered 13,248 requests for service.

Privatization in the case of Starrett City resulted in some sav­ings. The hourly cost of a Starrett City officer in 1985 was $15.75, versus $22.22 for a New York City police officer.[24] The public treasury obviously was required to expend less because of the private Starrett City police. One estimate of the minimum annual savings to the public treasury is $750,000. In addition, one mea­sure of the benefits the Starrett City residents received could be derived from the comparative rental rates for Starrett City apart­ments versus similar apartments in the area.

Starrett City is susceptible to privatization: the beneficiaries are distinct, and the services accordingly can be sold. In effect, the residents in Starrett City pay for the security services through their rental payments. They are obviously willingly to do so: they could rent other apartments. Relying solely upon the public police is evidently a less satisfactory alternative. The demands placed upon the public police effectively preclude their giving as much attention to the residents of Starrett City as the residents obviously desire. Privatization allows residents in a clearly defined area to determine more precisely the degree of security they want and for which they are willing to pay. In addition to pro­prietary forces such as the one established by Starrett City, there is the option of contracting out or hiring (where the municipality permits) off-duty public police.


Private security can be effective in a distinct geo­graphic area. Therefore, apartment complexes, among others, ought to be encouraged to consider private policing. Competition among apartment complexes to provide safer environments ought to be encouraged. In fact, requiring or encouraging publication of apart­ments’ safety experience might be desirable to permit renters to make informed choices.

Criteria for Contracting Out

Privatization is feasible when:

1. The tasks can be well specified so that a profit-maxi­mizing contractor’s performance can be effectively monitored and evaluated;

2. The area of operation is a defined geographical area, such as an apartment complex or a central business district;

3. Few negative externalities (costs shifted to third par­ties) result from the private security firm’s operations;

4. Competition exists at the contracting stage or at the service provision stage;

5. The tasks are not so skilled as to require the authority and knowledge of sworn officers; and

6. The size of the population served is sufficiently small to avoid “free rider” problems.

Requirements for Achieving Privatization

It Must Be Possible to Specify Service and Monitor Compliance. Privatization, as mentioned previously, permits the competitive process to work, which generally results in costs as low as pos­sible, incentives for innovation, and products and services tai­lored to individual demands. In the case of security services, as for many other services, competition occurs at the stage of bid­ding for the contract. For example, a city may offer a contract to provide security at the courthouse. If competition is to work, the service must be clearly demarcated, and it must be possible to monitor compliance with the terms of the contract. For court­house security, the contract might specify a minimum of four armed guards on duty whenever court is in session and one armed guard on duty 24 hours a day. Firms could then bid on the contract, and the city could review the performance of the contractor.

Competition Must Exist for the Contract. Competition is feasible once the service is specified. For example, in the United States as a whole in 1992, there were 11,000 firms offering contract guard and patrol services.[25] Included in that number are large national firms such as Pinkerton, Burns, Wackenhut, and Wells Fargo. Borg-Warner, which owns Burns and Wells Fargo, is the largest guard firm, with 7 percent of the market, followed by Pinkerton and Wackenhut, with 4 percent each. These firms compete in a large portion of the United States. In a typical local market, regional and small firms also compete. For example, in 1991 in Montgomery County, Pennsylvania, there were 37 estab­lishments furnishing detective and armored car services, and 29 of them had fewer than 20 employees.[26] Philadelphia County had 61 establishments, 31 of which had fewer than 20 employees. The guard and patrol portion of the industry is only a portion of the overall industry, but it is the largest part, account­ing for 73 percent of all security expenses of commercial and industrial establishments. The typical guard and patrol company in 1988 had about 27 employees, suggesting that large size is not required to compete in the industry. The largest guard firm in 1992 had more than 55,000 employees, while the average firm had 10 and the median had 90.[27] On the other hand, the top 10 guard and patrol firms earned 31 percent of the revenue of all guard and patrol firms in 1988.

Entry into the industry is not difficult. There are no patent or natural-resource barriers, and state licensing is not very stringent: for example, liability insurance is required in only 11 states; only 14 states have training requirements for unarmed guards or offi­cers; and 23 require training for armed individuals.[28] Capital requirements are not high: indeed, a firm could be formed using only off-duty police officers. One factor that may hinder entry somewhat is liability insurance. Large firms claim that they can obtain sufficient insurance;[29] small firms may encounter some difficulty obtaining moderately priced coverage, but they can, in effect, simply exit from the industry if they possess inadequate insurance to satisfy a judgment. The large number of firms in the industry suggests that the insurance/liability problem can be overcome. However, guard service is largely absent in sectors where the chances of litigation are highest (for example, response to alarms).

Another factor affecting the ability to have competition is the possibility of vertical integration. In many segments of private security, including patrol and guard services, buyers could simply hire as employees their own security guards. However, there are advantages to contracting out for such services: problems asso­ciated with employee sickness, resignation, and unionization can largely be avoided. Finally, growing demand for private security means both that more firms can be accommodated in the indus­try and that entry is becoming even easier. For example, average annual revenue growth for the private security industry during the 1980s was about 11 percent.[30]

An industry structure with a large number of firms and low entry barriers should yield prices near or at marginal and average cost. Governmental and other buyers should be able to obtain security services at the lowest price possible. For any given ser­vice that requires a certain level of skill, intelligence, and physi­cal strength, competition yields the minimum price. However, in hiring private security officers with skill and ability equal to the public police, privatization and competition are not likely to re­sult in substantial savings. After all, these officers could become public police officers or obtain some other relatively high-paying position.[31] However, privatization probably means greater flexibil­ity in terms of laying off officers if conditions warrant, compared to the ability of governments to do so.

Competition and Privatization

Competition to obtain contracts has meant that security offi­cer wages are just high enough to attract the desired employ­ment level. Moreover, much private security employment fills the lower-skill tasks or satisfies a smaller range of skills than does employment of public police. Accordingly, it is not surpris­ing that the average wage for a police officer in 1990 ($24,000) was about 50 percent higher than that of an average security of­ficer.[32] Overall savings from privatization could be substantial. Also, Oro Valley, Arizona, in 1975 found that contracting pri­vately even for its complete police service was cheaper than establish­ing its own force.[33]

Specialized high-skill services could also be contracted out. Currently, insurance investigators and appraisers examine resi­dential, commercial, and automobile claims. It is possible to con­tract out for similar types of activities involving criminal behavior.

Privatization and Cost Reductions

Privatization has the potential to yield substantial savings. The experience of Starrett City illustrates the possible savings. More generally, the average police officer earns substantially more than the average security officer. For example, median wages for modestly skilled and skilled/specialized security officers in 1992 were $12,480 and $23,192, respectively.[34] On the high end of the earnings spectrum for private security were guards employed by the federal government who had average earnings of $21,700 in 1993. Median 1992 income for police officers and detectives was $32,000; supervisory officers had median earn­ings of $38,100. Sheriffs, bailiffs, and other law enforcement agents had median earnings of $25,800. Experienced federal law enforcement officers had salaries in 1993 that started at $40,200 (Treasury Department) and $47,900 (Federal Bureau of Investigation). Further, a study on wages in American industry showed that the average guard in Wichita, Kansas, in 1990 earned $4.45 per hour, while a Kansas police officer earned $10.69.[35]

A study on police resources for the Portland, Oregon, area points out some potential savings from privatization.[36] For example, privatizing courthouse security by replacing a sworn deputy with a private security guard saves 30 percent of the deputy’s cost. The same study reports that the U.S. Marshal’s Service has privatized all federal court security. Private personnel who provide security for federal courts carry weapons and become deputized U.S. Marshals for their court security duties. The savings exceed 50 percent of the usual salaries.

Privatization can (and has) achieved substantial savings. A conservative estimate of the potential savings would be 30 per­cent of the public salary expenditures.


Governments would probably be well served by facilitat­ing an expansion of private security. Any relatively low-skill or specialized high-skill services that are currently provided publicly could be considered as candidates for transfer to private security.

In any case, businesses and personal consumers are likely to increase their use of private security. The advantages of compe­tition and private provision are many. Consumers through their purchasing decisions could more easily match the amount of security they want to what is available. Competition helps ensure that the level desired is procured as efficiently and inexpensively as possible. Competition is of paramount importance. We do not want to substitute a private monopoly for a public monopoly.

The contract must be written so that compliance can be monitored at acceptable cost. As mentioned previously, court­house security provision whereby the contractor agrees to pro­vide a specified number of armed guards during certain hours can be adequately monitored. Activities like general patrol are more difficult to monitor: officers could simply fail to perform their patrol. Similarly, detective investigations are not susceptible to easy monitoring.


Monitoring contractor compliance and performance must not be so costly as to eliminate the savings from privatization.

Modified Privatization: Hiring the Public Police

One alternative to more complete privatization is allowing the public police to sell their services to businesses and consumers. This permits store owners, hospitals, and others that want more security than they can obtain from the public police to hire off-duty officers. In addition, privatization has enabled the public police to earn extra income without adding to the tax burdens of their communities. For example, 53 percent of Colorado Springs’s sworn officers with work permits in 1986 worked a total of 20,000 hours off-duty in uniform and earned an average of $1333.[37]

In some communities, commercial establishments pay all the expenses associated with particular police services designed for their needs. For example, the King of Prussia shopping mall pays the expenses for a full-time Upper Merion, Pennsylvania, police officer who is assigned to patrol its parking area. In other com­munities, the police deputize private security guards who per­form patrol duties for commercial districts. The use of off-duty police officers raises issues of liability for the government, questions of unfair competition for private security firms, and questions of loyalty and duty if a crime occurs outside the premises where the off-duty officer is employed.

Legal Issues Associated with Privatization

A factor that inhibits the increased reliance on private secu­rity is the previously mentioned liability issue. For example, an armed private security officer might accidentally wound or kill an innocent bystander when responding to an alarm activation. The resulting judgment could bankrupt a small firm with modest lia­bility coverage. However, one firm in southern California suc­cessfully offered armed response to alarm activations and patrol services for 10 years, and none of its seven shootings resulted in injuries to its officers or provoked lawsuits.[38] The firm hired qualified individuals and enforced substantial training require­ments. Evidence of its success is the fact that the police accepted the firm, a number of its officers became public po­lice officers, and the firm was able to compete in the market by offering a high-quality product.[39]

Although the problem of adequate and affordable liability insurance can be overcome, and even armed private security can be successfully provided, the concerns about liability remain. Awards of $100,000 or more against private security are increasingly common.[40] However, as mentioned previously, large firms do not report substantial problems with obtaining adequate insurance. In the case of small firms, it is possible that the indus­try could develop an insurance program to compete with other carriers, possibly making insurance more affordable. Also, reliance on insurance with large deductibles, in effect self-insuring against small claims, is another possibility. Some large guard companies could establish an insurance umbrella where each contributes to a common fund that would be used for all claims below some amount, while larger claims would be underwritten through a commercial insurer. Such a practice is common among universities in large metropolitan areas. An insurer could even be contracted to manage the funds and claims on the mutual fund. Such cooperation among guard companies could encourage the creation of alarm response companies for designated geographi­cal areas. Moreover, consideration ought to be given to tort reform, where proving clear negligence might be required to obtain a judgment.

Other legal issues also arise in the case of private police. One is granting private police expanded legal authority. Security offi­cers in Las Vegas, Nevada, can be given appointments as special deputy sheriffs.[41] The Texas Department of Public Safety can appoint private security officers as special Rangers with all the arrest and firearms authority of official police. In North Carolina and Maryland, private security officers can be appointed special police with full police powers at businesses that employ them. For example, hospital police in Altoona have legal power equiva­lent to that of the city police.[42]

Another issue arises because private security is not subject to all the constitutional safeguards of public law enforcement, for example, search and seizure rules. Thus, it enjoys some special status.[43] Is such an exclusion appropriate? In effect, pri­vate security derives its authority from the right of property own­ers to protect their property. Private security vehicles are not considered emergency vehicles, so they must obey the traffic rules. Is that inability justified when quick response may save lives? Moreover, unless granted special police powers, a private security officer simply makes a citizen’s arrest. If private secu­rity is expected to become an even more important component of overall security, state legislatures may consider whether cur­rent regulations and policies are still appropriate.


State legislatures should consider whether the current legal status and regulations pertaining to private security are appropriate in view of the expanded role expected from private security. Specifically, emergency vehicle status and expanded powers of arrest ought to be examined.

Forms of Privatization

Privatization can take many forms: divestment of public functions through sale or transfer; delegation by contract, fran­chise, or grant; and displacement by default, withdrawal, or deregulation.[44] Some withdrawal from certain activities, such as alarm response, has already largely occurred. Considerable priva­tization has also occurred through contracting. For example, airport and courthouse security are often contracted out to pri­vate firms.

The Role of Government in Achieving Privatization

Government has an important role to play in achieving privati­zation. One might argue that its involvement in activities ought to be as little as possible given the problems inherent in govern­ment provision. Government involvement should be restricted to the cases of market failure, and the level of intervention should be the minimum necessary to assure efficient utilization of resources. For police services where privatization is feasible, government might limit itself to contracting out instead of sup­plying the complete services itself. Government might develop licensing rules and training requirements for private security officers if regulation is deemed appropriate. Setting standards may be an appropriate function for government. Given the bene­fits of competition, government might want to create the condi­tions under which competition will work.


A modern approach to policing is called “problem-oriented policing.” This involves a change in management of police resources from being mainly reactive to being more proactive. The idea is to determine the cause of a problem, such as rob­bery, to enlist resources in addition to police resources to try to solve the problem, and to assess the impact of the response, asking whether the problem has been solved. The objective is to undertake actions to prevent the problem.

An actual example will make the problem-oriented approach clearer. Newport News Shipping in Newport News, Virginia, in 1984 employed 36,000 workers who parked in nearby lots.[45] Thefts from these parked cars accounted for as much as 10 percent of all reported crime in the area. An officer assigned to the problem interviewed patrol officers, detectives, and shipyard security in order to identify causes and assemble suggestions for reducing the thefts. He further studied the incident and arrest records for the most recent three-year period. This inquiry led to a list of possible suspects and eventually to the arrest of one of the suspects, who was caught breaking into a car. After the sus­pect was convicted, the police officer interviewed him and learned that a primary motive of the thieves was to obtain drugs. They broke into “muscle” cars or those with rock-and-roll bumper stickers because they believed that such cars had a greater likelihood of having drugs. With that information, more arrests, convictions, and interviews followed, and the process was repeated. With the arrest and conviction of the most persis­tent thieves, thefts from cars parked at the shipyard dropped by 53 percent just one year after the program was initiated.

The problem-oriented approach differs from normal police work in the sense that a problem is identified and quite often resources other than those of the police are employed to solve the problem. However, much of the approach seems to be sim-ply good police work where problems are solved and initiative is employed, rather than simply maintaining the “traditional” responses and reactions. For example, instead of considering the above robberies as just robberies, they were grouped into the category of drug-related robberies so that a solution could be sought. Unfortunately, the problem-oriented approach does not always lead to results that can be clearly demonstrated, so the approach enjoys modest support among local politicians and police chiefs.

A recent example of problem-oriented policing comes from San Diego, California, where an apartment complex, the Cucas, was the scene of substantial drug-related criminal activity.[46] Accordingly, the police decided to attack some of the facilitating factors that led to the problem, including inadequate street lights, insufficient and ambiguous street signs, lack of commu­nication among police units, and significant numbers of abandoned vehicles.

The police further assisted the management of the apartment complex in evicting the tenants implicated in drug activity. A nearby gas station where drug dealers loitered waiting for tele­phone calls cooperated with the police. Gas station employees called the police when people loitered near the telephone, and the telephone itself was modified so that it could not receive incoming calls. The only factor that was not accomplished was the erection of street signs. The increased police concern with the Cucas drug problem occurred at the same time when a major Cucas drug dealer died in a police chase, making it difficult to determine the impact of the problem-oriented approach. In any case, arrests did not increase in the complex area, and service calls for police did not decrease. However, a survey of residents who had a view of the courtyard, the scene of much of the drug dealing, reported a decrease in drug activity, although residents still complained about disorder in the complex.

A problem-oriented approach to policing is actually an atti­tude on the part of the police that makes sense. As mentioned, it seems like just good police work in the sense that it is proactive in trying to prevent further incidents rather than simply reactive in responding to individual incidents. Problem-oriented policing requires greater initiative by police officers, and it may also improve morale. However, the experience of San Diego sug­gests that the approach does not always produce marked results. It requires more police resources, at least in the short run, pre­sumably reducing enforcement efforts elsewhere; and there is a concern that displacement may occur as the problem moves to another area. However, employing problem solving at least as a part of a police officer’s work makes sense. Two experts state that “police officers can solve problems as part of their routine; they enjoy problem solving; and their efforts are often success­ful.”[47]

Successful implementation of the problem-oriented approach requires a change in officers’ attitudes. Efforts should be made at the police academy to inculcate such attitudes. In addition to regular lectures, actual case studies should be pro­vided where the recruits are expected to evaluate the problem and make suggestions. This approach would be similar to the case study method for MBA classes at the Harvard Business School.


Problem-oriented policing is a method that offers the prospect of improved police performance.


Another approach to policing is the shift to closer involve­ment with citizens, the community policing method. Community policing has been defined as “a philosophical position which holds that the goals of policing, the conditions which it addresses, the services it delivers, the means used to deliver them, and the assessment of its adequacy should be formulated and developed in recognition of the distinctive experience, needs, and norms of local communities.”[48]

Community policing involves such means as use of minista­tions, foot patrol, and organizing or helping to organize neighbor­hood groups (Neighborhood Crime Watch, for example.)[49] Many other means fall within the rubric of community policing, but the major focus is on closer police-community cooperation.

In April 1988 an Experimental Policing District (EPD) was created in Madison, Wisconsin, encompassing about 16 percent of the city’s population, area, and calls for police service, and about 20 percent of the city’s reported crimes.[50] The EPD encouraged active community involvement as well as coordina­tion and cooperation within the police department itself. A small building was utilized and was designed to encourage interaction with the citizens and amongst the officers themselves. Officers responding to calls for service also inquired about other prob­lems that might exist. The police attended community meetings, visited schools and businesses, and made an effort to get to know neighborhood residents.[51] Participatory management was also emphasized so that individual officers could determine, in conjunction with citizen requests and input, what problems should be focused upon.

After the program had been in operation for three years, offi­cers and citizens were surveyed to determine their perceptions. Statistically significant impacts were found for the EPD in offi­cers’ sense of participation in organizational decision making, in investigative activities, and in the department as a desirable place to work.[52] Citizens also perceived an increased police presence, a belief that police were working with the community to solve the neighborhood’s problems, and a belief that crime in general was less of a problem. On the other hand, citizens were increasingly concerned about drugs and parking violations.[53] This experiment showed positive benefits and satisfaction from community policing. It seems difficult to criticize an approach that emphasizes more involvement of the police in the com­munity. However, it does not necessarily reduce crime or increase citizen feelings of security. It has been concluded that in Madison the approach worked.

One factor suggesting the desirability of such an approach is that police resources are currently spread very thin. For example, suppose a police department has 100 sworn officers. Some are specialists such as detectives, data management experts, and commanders. Given such factors as vacation time and court appearances, during some periods only 15 officers may be on patrol. Increasing community involvement could greatly assist the police. For example, those subject to domestic abuse might be more willing to confide in a patrol officer they know before the abuse escalates into more serious violence. The proactive police work could thus economize on the scarce police resources.

Another example of community policing comes from Seattle, in particular from the Rainier Valley section in south Seattle.[54] This section of Seattle had lagged behind the rest of the city in economic development and had a high crime rate. During the 1970s, the community had become increasingly concerned about crime, establishing, among other crime prevention pro­jects, the first Business Watch program in the United States to combat business burglaries. After the drug problem had become serious, contributing to much of the neighborhood’s crime, the Rainier Chamber of Commerce began to take a more active role in encouraging even more community involvement with policing. The chamber urged creation of a Community Advisory Committee to advise the police about community concerns. It requested more police personnel and announced its willingness to provide community support if the police could not provide sufficient funds for the South Precinct area, the Rainier Valley neighborhood.[55]

In particular, the chamber wanted a strong voice in deter-mining use of police resources, for example, in the selection of targets. The chamber in January 1988 created the South Seattle Crime Prevention Council (SSCPC), a small group representing various community organizations. Citizen reports of drug activity were received in person by patrol officers, at the station, or on the community hotline. In one year, 1,219 citizen reports were received. The property owners gave the police authority to arrest and/or remove trespassers and pay telephones were converted to outgoing call status only. An antigraffiti program and a tele-phone tipline were established. A two-person police patrol unit was permanently assigned to two housing projects to help develop police-community relations. Community-police teams were established to deal with problems that might arise.

The SSCPC-police relationship, which increased the police presence in the community, led to strong public support. The crime rate in south Seattle decreased after 1987, whereas other portions of Seattle experienced decreases only after 1988. Calls for service increased in south Seattle but declined in the rest of Seattle. Interestingly, many of the calls did not require immediate action.

Community policing was also employed in New York City. The Community Patrol Officer (CPO) program began in 1984 in the 72nd precinct in Brooklyn. By 1988 it was expanded to all of the city’s 75 precincts. The program involved approximately 750 police officers and 75 sergeants.[56] Each CPO was assigned to a beat (an area of 16 to 60 square blocks), and the CPO was encouraged to walk the beat and become acquainted with the community’s concerns. The officers were not required to respond to 911 calls, and residents were encouraged to contact the officers directly. The program did not noticeably reduce burglaries or robberies; however, assessment was difficult because the analysis had to be done by precincts, and unless most or all of the CPOs concentrated on robberies or burglaries, any impact was difficult to discern.

The CPO program did improve the community’s perception of control over its environment and did generate a sense of coop­eration between the community and the police, and it raised the issue that the police could become susceptible to false charges of abuse or corruption if they were successful in disrupting the operation of drug dealers (the perception among police is that increased involvement with the community increases the likeli­hood of such charges). The program also showed the importance of command sergeant–patrol officer cooperation, since the CPO has much more discretion under the community policing pro­gram. Community control raises the issue that police resources could be directed toward problems that concern mainly com­munity activists unless an effort is made to have broad commu­nity involvement.

Community policing worked in Madison, New York City, and Seattle. Like problem-oriented policing, community policing makes good use of resources by enlisting the community’s sup­port in determining what the community wants from its police. The approach encourages proactive rather than reactive behav­ior. The police officer under community policing handles prob­lems of abandoned buildings and vehicles, trash on the streets, graffiti, and aggressive panhandlers. These problems troubled area residents much more than the police realized. The officer is also engaged in preventing crime.[57] Evaluation of police perfor­mance becomes more difficult under community policing because numerical measures like arrests per year become less important.

Critics of community policing raise the issue of a possible reduction in crime fighting, concerns about civil liberties as the police become more involved with the community, concerns about fairness and equity in the allocation of police resources, and questions about control as decentralization becomes more pronounced.[58] The question of allocation of limited police resources to a program like community policing is valid, but some element of community policing seems desirable. The impact of community policing on other police work—for exam­ple—911 calls, is unclear. Calls may be stimulated as residents become more willing to report crimes, or crimes may be pre­vented through active community involvement.

The question of allocation of resources is important because with the general increase in 911 calls, much of police work is devoted to reacting to incidents or other problems. Community policing may mean a somewhat longer response time or fewer officers available for backup. Empirical evidence needs to be gathered on the impacts of community policing and/or problem-oriented policing on resources available elsewhere. Along those lines, one might want to examine the benefits of reducing the numbers of specialists, such as the robbery detail, and shifting some of those officers to community policing. It is, after all, not clear that the current allocation of police resources is optimal.

An important issue involves implementation of new policing methods. The experiences of New York City and Madison point out the desirability of introducing the change to community policing in one district and later introducing the change in other districts based on the experience of the experimental district. Further, support for the change in police methods should be obtained from command officers to help ensure success of the program.

Another issue concerns the reward structure for police offi­cers and the consequent impact on performance. Criteria must be developed to replace the usual measures of police perfor­mance. Rewards, commendations, and promotions presumably would be given for exceptional community policing work that may not involve major arrests.


Governments should be encouraged to expand use of community policing, because the approach offers hope for improving police performance and the commu­nity’s sense of participation. Like privatization, com­munity policing helps society better determine the use of its scarce police resources. Further, it brings the po­lice “back” to constituents. Successful community polic­ing satisfies the desires of the community.


Three current trends in the provision of public security were analyzed: privatization, problem-oriented policing, and commu­nity policing. Privatization involves the use of markets to deliver services. Police activities were analyzed, including patrol, crimi­nal investigations, public assistance, and response to alarms. The criteria for public intervention in the market (or imperfections in the marketplace) are: pure public good, externalities, economies of scale, and zero marginal cost. Almost all police services do not satisfy these criteria and hence can be contracted out with limited public monitoring. Alternatively, if the police want to retain their monopolistic power, a minimal policy change would be to impose user charges, for example, fees for response to alarm activations.

Indeed, we face two realities. Police budgets rise at an aver-age annual rate of 3 percent—just enough to cover the increased cost of resources. On the other hand, the rises in income and crime lead constituents to search for better secu-rity. Thus, private security should be allowed and encouraged. Households and businesses already better protect their property by installing alarms, locks, and external lighting.

The argument raised against private security is that it further penalizes the poor, who will not be able to afford the same security as the rich. That argument has the following flaws: (1) Private security will release public police resources, which can then be used to protect lower-income groups, (2) The wealthy already have better food and enjoy a greater variety of services than the poor; if they are willing to pay for better security, they are entitled to have it. (3) Studies show that contracting out police services achieves the same security supply at lower costs. These savings could be used to help lower-income groups obtain more security.

Problem-oriented policing (POP) and community policing change the role of the police. In both cases greater initiative is required from police officers on the street. POP changes the role of the police from reacting to crimes that have already occurred to the proactive role of preventing crimes. Clearly, POP requires courses to reeducate existing officers and new courses at police academies. POP is a variation of Total Quality Management, where lower levels in the corporate hierarchy are given more decision-making power.

Community policing essentially means greater community involvement in use of police resources and more interaction on a daily basis by police officers and their constituents. Com­munity policing is aimed at bringing the police back to the community. Effective policing means providing security for the community. Satisfaction of constituents with the police is the ultimate output of the police. However, as city size increases and the number of constituents rises, the police become more detached from the community. In small localities the “distance” of the politician from his or her constituents is small; thus, the police there are more apt to respond to citizens’ needs and desires. In large cities, however, the police can be more arbitrary in dealing with citizens and less concerned about whether citi­zens are satisfied with police operations.

Community policing is aimed at changing certain attitudes of the police, including lack of concern with the desires of the community. Further, it is important for police officers to feel that they are respected and admired by the community. Like POP, community policing is more difficult than “traditional” police work, where the officer is trained to respond in a prespecified manner to every possible situation and is not required to demon­strate much initiative. Community policing provides more “degrees of freedom” for the officer about what should be done. Personal relationships that develop between the officer and the residents in the community require more common sense and thought about the appropriate actions to take. It is far more diffi­cult to prevent a situation from happening than to react once it occurs.

Governments should consider contracting out police services in small localities. Large cities can contract out for defined ser­vices such as parking meter surveillance, school crossing patrol, and data management. Contracting out requires that adequate competition exists at the bidding stage to ensure competitive prices. In the absence of sufficient competition, contracting out should be reconsidered.

A private security force could provide other services that would add to its economic viability. Currently, police depart-ments struggle with false activations. As stated earlier in this report, it is likely that police will stop responding in many U.S. cities to alarm activations that are not classified as high priority. A private security force could take this obligation and be paid for the service. It could even implement fines for repeat activators; if the fines were not paid, response would cease. The force could provide emergency medical response and a variety of ser-vices made available by the new technology of the “smart home”, such as child tracking and smoke alert. Residents might leave keys with the private force for entrance to their homes in case of need. These added services, provided at additional charge, would add to the economic viability of the force and enable increased security vehicle presence in smaller commu-nities; and, the added security presence would provide a higher safety perception for the residents.

One may argue that there is no reason to implement private security in large cities; that the political power of police unions and organizations may make contracting out difficult. It may seem it is easier to contract out police services in small cities and learn from those experiences before moving on to the larger political jurisdictions. The inherent objective is to establish direct ties between the constituents and the providers of ser-vices without the intervention of the political bureaucracy. Patrol services that have few negative externalities could be privately provided and controlled by the residents of local districts. Of course, tax refunds should be given to the local residents or to their civic organizations, who would manage their own security affairs.

Privatization of capital-intensive services that enjoy signifi-cant economies of scale, such as detective services, is in prac-tice more difficult. It means that the public monopoly might be replaced by a private monopoly. There are also more externalities in the case of detective services. Thus, “traditional” public finance theory suggests that detective services should be pro-vided on a large geographical scale. However, district detective companies may wish to cooperate with other jurisdictions in solving crimes in order to reduce incidents in their own jurisdic-tions, satisfy their own constituents, and possibly gain similar business in adjacent districts. Thus, it is likely that contracting out of detective services and other services with similar charac-teristics could be successful even for large political jurisdic-tions. A more in-depth analysis of the consequences of private investigative companies operating in adjacent districts may require game theory modeling.

Community policing is particularly relevant and appropriate for large cities; in small places, officers are already sensitive to constituent desires. Implementation of community policing takes time. It involves changing attitudes of officers. Some changes can be brought about through education. However, most police attitudes are acquired through long-term and daily police work. Accordingly, it is imperative that police support the changes. Finally, constituents must become involved and exert pressure on government officials to ensure that community policing is employed.


[1]. Kansas Bureau of Investigation, Statistical Analysis Center, Crime in Kansas 1992, July 1993, p. 6.

[2]. These statistics are from Kansas Crime Statistics, prepared for the William I. Koch Commission by Lucas McPhee and Company, April 1, 1994, p. 12.

[3]. State agency employment is not included in order to make the comparison on the same basis. State figures were included in total police personnel in 1991.

[4]. Calculated from Crime in Kansas 1992 and Crime in Kansas 1985.

[5]. Statistics are from U.S. Department of Justice, Census of State and Local Law Enforcement Agencies, 1992, Bureau of Justice Statistics Bulletin, July 1993, p. 7.

[6]. William C. Cunningham, John J. Strauchs, and Clifford W. Van Meter, “Private Security Patterns and Trends,” National Institute of Justice Research in Brief, August 1991, p. 1.

[7]. See William C. Cunningham, John J. Strauchs, and Clifford W. Van Meter, The Hallcrest Report II, Private Security Trends 1970–2000 (Stoneham, Mass.: 1990), Butterworth-Honeymoon, p. 176 (hereafter cited as Hallcrest Report II); and U.S. Department of Justice, Census, July 1993, p. 3.

[8]. Hallcrest Report II, p. 123.

[9]. Hallcrest Report II, p. 119.

[10]. S. Stuart Ditzen, “Business Making Worker Theft a Private Matter,” Philadelphia Inquirer, August 21, 1994, p. B4.

[11]. For a challenge to the usual economic approach toward government involvement, see William C. Mitchell and Randy T. Simmons, Beyond Politics (Boulder Colo.: 1994), Westview Press, especially pp. 214–21.

[12]. This classification comes from Marcia Chaiken and Jean Chaiken, “Public Policing—Privately Provided,” in National Institute of Justice, Issues and Practices in Criminal Justice, June 1987, p. 7. The more detailed breakdown comes from their appendix C, pp. 33–41.

[13]. See Chaiken and Chaiken, “Public Policing,” p. 4.

[14]. Ibid.

[15]. Hallcrest Report II, p. 271.

[16]. Ibid., p. 280.

[17]. Simon Hakim and Mary Ann Gaffney, Commercial Security: Burglary Patterns and Security Measures (Washington, D.C.: 1994), Security Industry Association, chap. 6.

[18]. Simon Hakim and Mary Ann Gaffney, “False Alarms: Fines, Effects, and Solutions,” Locksmith Ledger, forthcoming.

[19]. Simon Hakim and Mary Ann Gaffney, “The Structure of the Installers Market,” Locksmith Ledger, May 1994, pp. 44–56. See also John Keller, “Security Companies Are Alarmed by Baby Bell’s Threat,” Wall Street Journal, July 1, 1992, p. B4.

[20]. See Jerry Usher, “Privatization in Criminal Justice: One Perspective in Southern California,” in Gary W. Bowman, Simon Hakim, and Paul Seidenstat, eds., Privatizing the United States Justice System (Jefferson, N.C.: McFarland, 1992), pp. 148–51. Armed response is provided by a firm in Ft. Lauderdale, Florida. The firm uses off-duty police officers.

[21]. This case comes from William P. Walsh, Edwin J. Donovan, and James F. McNicols, “The Starrett Protective Service: Private Policing in an Urban Community,” in Bowman et al., Privatizing the United States Justice System, pp. 157–77.

[22]. Ibid., p. 163.

[23]. Ibid., pp. 166 and 167.

[24]. Ibid., p. 170.

[25]. Frost and Sullivan, The U.S. Commercial and Industrial Security Equipment and Service Markets (Mountain View, Calif.: 1994), p. 199.

[26]. U.S. Department of Commerce, Bureau of the Census, County Business Patterns 1991 (Washington D.C.: December 1993), U.S. Government Printing Office, pp. 198 and 218.

[27]. Hallcrest Report II, p. 218; and Frost and Sullivan, U.S. Commercial and Industrial Security, p. 68.

[28]. Hallcrest Report II, p. 155.

[29]. See Chaiken and Chaiken, “Public Policing,” p. 23.

[30]. Hallcrest Report II, p. 175; and Frost and Sullivan, pp. 199–201.

[31]. See, for example, the discussion in Chaiken and Chaiken, “Public Policy,” p. 12, concerning the New York City Housing Police who perform essentially as public police in their jurisdiction and have the possibility of becoming public police if their compensation is too low.

[32]. Hallcrest Report II, p. 156.

[33]. Philip E. Fixler Jr. and Robert W. Pool Jr., “Can Police Services Be Privatized?” in Bowman et al., Privatizing the United States Justice System, p. 34.

[34]. Figures came from U.S. Department. of Labor Bureau of Labor Statistics, Occupational Outlook Handbook, 1994–95 edition, May 1994, pp. 300 and 303.

[35]. Marlita A. Reddy, American Salaries and Wages Survey, second edition (Detroit: 1993), Yale Research, Inc., pp. 313 and 576.

[36]. Citizens Crime Commission, The Future of Police Services in Multnomah County, January 1993, especially pp. 90 and 91.

[37]. Abert J. Keiss Jr., “Private Employment of Public Police,” National Institute of Justice Research in Brief, December 1988, p. 1.

[38]. Usher, “Privatization in Criminal Justice,” p. 140.

[39]. Ibid.

[40]. Charles P. Nemath, Private Security and the Law (Cincinnati, Ohio: 1989), Anderson, p. 113.

[41]. This discussion comes from Fixler and Poole, “Can Police Services Be Privatized?” p. 36.

[42]. In fact, the hospital police even investigate violent crime. See “Doctor Attacked, Assailant Killed in Car Crash,” Montgomery County Record, Aug. 29, 1994, p. A4.

[43]. Nemath, Private Security, pp. 76–78.

[44]. From E. S. Savas, “Privatization,” in Mary Hawkesworth and Maurice Kogan, eds., Encyclopedia of Government and Politics, vol. 2 (London: 1992), Routledge, pp. 822–26, especially p. 822.

[45]. This example comes from William Spelman and John E. Ecke, “Problem Oriented Policing,” National Institute of Justice Research in Brief, January 1987, pp. 6 and 7.

[46]. This example comes from George E. Capowich and Janice A. Roehl, “Problem Oriented Policing: Actions and Effectiveness in San Diego,” in Dennis P. Rosenbaum, ed., The Challenge of Community Policing (Thousand Oaks, Calif.: 1994), Sage, pp. 135–37.

[47]. John E. Ecke and William Spelman, quoted in Mark Harrison Moore, “Problem Solving and Community Policing,” in Michael Tonry and Norvy Morris, eds., Modern Policing (Chicago: 1992), University of Chicago Press, p. 131.

[48]. Jerome E. McElroy, Colleen A. Cosgrove, and Susan Sadd, Community Policing: The CPOP in New York (Newbury Park, Calif.: 1993), Sage, p. 7.

[49]. See John E. Ecke and Dennis P. Rosenbaum “The New Police Order: Effectiveness, Equity, and Efficiency in Community Policing,” in Dennis P. Rosenbaum, ed., The Challenge of Community Policing (Thousand Oaks, Calif.: 1994), Sage, p. 4.

[50]. The discussion about Madison comes from Mary Ann Wycoff and Wesley K. Shogan, “Community Policing in Madison: Quality from the Inside Out,” National Institute of Justice Research Report, December 1993, p. 21.

[51]. Ibid., p. 29.

[52]. Ibid., p. 57.

[53]. Ibid., p. 81.

[54]. This case comes from “Community Policing in Seattle: A Model Partnership between Citizens and Police,” National Institute of Justice Research in Brief, August 1992.

[55]. Ibid., p. 4.

[56]. The description comes from McElroy, Cosgrove, and Sadd, Community Policing, p. 11.

[57]. See Edwin Meese III, “Community Policing and the Police Officer,” National Institute of Justice Perspective on Policing, January 1993, p. 2.

[58]. See Moore, “Problem Solving,” pp. 139–47.