Predatory Public Finance and the Origins of the War on Drugs 1984-1989. Part 1

Introduction

The Harrison Act of 1914 is often cited as the law that made consumption of narcotics illegal. In reality, the Harrison Act established very modest "sin taxes" on the sale of narcotics such as opiates. What became illegal as a result of the act was the possession or sale of untaxed narcotics. Policy against marijuana began in a similar fashion with passage of the Marijuana Tax Act in 1937. Sin taxes such as these inevitably lead to crime, however, as individuals attempt to avoid the tax through black markets, smuggling, and the violent forms of competition and contract enforcement that accompany such activities. An excise tax may reduce the level of the sin being taxed, but it simultaneously induces new kinds of sin that are often much more costly for society. This occurred with narcotics and marijuana. However, rather than recognize the source of the crime and eliminate the sin taxes, full-blown criminalization of possession or sale of narcotics and marijuana evolved as bureaucrats who were given the authority to police these markets and collect the taxes propagated the belief that it was the "sin" of drug consumption that produced the crime, rather than the incentives to avoid the taxes imposed on the sin. The purpose of the following is to show that bureaucratic self- interest and predatory public finance in the form of explicit or implicit sin taxes have been and continue to be the primary determinants of public policy in the area of illicit drug control.

When sin-tax-induced crime becomes a serious problem, the government entity imposing the tax has at least these two options: to crack down on the resulting crime or to repeal the sin tax. Consider Quebec's recent experiences with taxes on cigarettes, for instance. A 9 August 1993 article in Maclean's, a leading Canadian newsmagazine, notes that "Tax levels in excess of 60 per cent on cigarettes have convinced many smokers that they are justified in breaking the law." Roughly half the cigarettes in Quebec and some other Canadian provinces were being bought in illegal black markets in order to avoid excise taxes; indeed, one in nine cigarettes consumed in the entire country was illegally purchased. As one buyer of black market cigarettes explained, "Stupid laws are meant to be broken." As noted earlier, however, tax evasion is not the only crime arising from sin taxes. Canada's illegal cigarettes were smuggled across the border. Canada exported roughly 7.6 billion cigarettes to the United States in 1992, for instance, and police estimate that 80 percent were smuggled back into Canada. Cigarette smuggling was so lucrative that organized crime got involved. Rival smuggling gangs exchanged gunfire as they competed for shares of the illegal market. They stole large boats, armed them with mounted guns, and painted them black for nighttime smuggling. Thus, an inevitable cost of high excise taxes was that crime rose, so demands for already-pressed policing, courts, prisons, and other law enforcement services increased. Canadian police were forced to become involved in the same kinds of interdiction efforts against cigarettes the United States is attempting today with illegal drugs. The re- sults were similar: the flow of illegal cigarettes into Canada was unabated. Further, a crackdown on the sin-tax-induced crime means that either taxes have to be raised to support a larger criminal justice system or some of the society's precious criminal justice resources have to be shifted away from efforts to deter burglars, robbers, rapists, and other criminals who threaten lives and property, in order to control black markets, smuggling, and the violence that arises with these activities. Quebec citizens apparently recognized this, and in 1994 the Quebec government announced a massive reduction in cigarette tax in an effort to end uncontrollable smuggling.

The Wages of Sin Taxes

Citizens of the United States did not see the consequences of sin taxes on narcotics and marijuana that Quebec citizens recognized regarding the taxes on cigarettes. Thus, the United States took a different route. Crime control became the focus of public policy. Nonetheless, sin taxes remain extremely relevant to this day in a different form, as asset seizure laws have been revised to allow law enforcement bureaucrats to "tax" those who participate in illicit drug markets. Indeed, this helps explain why so many state and local components of the criminal justice system responded to the federal call for the latest offensive in the "war on drugs" after 1984.

The most recent renewal of the war on drugs in the United States was declared by President Reagan in October 1982 (Wisotsky 1991). Such an offensive in United States had to be waged by state and local "troops," however, and state and local law enforcement agencies generally did not increase their relative efforts against drugs in a dramatic fashion until 1984, when a substantial reallocation of state and local criminal justice system resources toward drug enforcement began. In fact, although drug arrests relative to arrests for reported crimes against persons and property (Index I offenses of murder, manslaughter, sexual assault, assault, robbery, burglary, larceny, and auto theft) remained relatively constant at one to four from 1970 to 1984, the relative effort against drugs increased by roughly 45 percent over the next five years. By 1989, criminal justice resources were being allocated to make only about 2.2 Index I arrests for each drug arrest.

There are a number of possible explanations for the upsurge in drug enforcement. Perhaps local elected officials, representing median-voter preferences across the nation, almost simultaneously demanded that their police departments escalate the war on drugs. There are strong indications that this explanation does not hold, however (Rasmussen and Benson 1994, 122–27). For example, in 1985, public opinion surveys suggested that drug use was not considered an especially significant problem. Indeed, illicit drug policy appears to be a case in which policy changes led public opinion, at least during the escalation of the drug war over the 1984–89 period. Another explanation for the trends in the reallocation of local police resources during this period is that powerful interest groups demanded the war. As Chambliss and Seidman (1971) concluded, "every detailed study of the emergence of legal norms has consistently shown the immense importance of interest-group activity, not the public interest, as the critical variable" (73). Similarly, Rhodes (1977) pointed out that "as far as crime policy and legislation are concerned, public opinion and attitudes are generally irrelevant. The same is not true, however, of specifically interested criminal justice publics" (13). More recent research reaches similar conclusions, and also makes it clear that one of the most important "specifically interested criminal justice publics" consists of law enforcement bureaucrats (e.g., Berk, Brackman, and Lesser 1977; Benson 1990, 105– 26; Rasmussen and Benson 1994, 119–73). In order to establish the argument that police bureaucracies have considerable power in the game of interest-group politics, at least as it pertains to drug legislation, in the next section we will briefly examine law enforcement influences on the historical emergence of illicit drug legislation and subsequent criminalization. We will do so in the context of Breton and Wintrobe's model of bureaucratic efforts to establish policy. The significant role that entrepreneurial bureaucrats have played in the development and evolution of drug policy is emphasized as a prelude to the "Police Interests" section, in which we examine federal legislation during the 1980s from the same perspective.

The "Police Interests" section provides an explanation for state and local police involvement in the 1984–89 drug war. Specifically, state and local policing officials faced what presumably was an exogenous change in bureaucratic incentives (although at least some of these police officials were important sources of the demand for the change) that induced an increase in drug enforcement efforts. In particular, one section of the Comprehensive Crime Act of 1984 established a system whereby any local police bureau that cooperated with federal drug enforcement authorities in a drug investigation would share in the money or property confiscated as part of that investigation. As a result, police in many states whose own laws or constitutions limited confiscation possibilities began to circumvent state laws by having federal authorities "adopt" their seizures. Thus, under the 1984 federal statute, a substantial percentage of these seized properties went back to the agency that seized them, even if the state's laws mandated that confiscations go someplace other than to law enforcement. This asset seizure law not only established a way to tax the sin of involvement in drug markets, but it required that the resulting revenue go to the tax collector—the law enforcement agency—thus creating relatively strong incentives to collect the tax. Demanded by federal, state, and local law enforcement bureaucrats, this legislation largely reflects the bureaucratic competition and cooperation modeled by Breton and Wintrobe (1982).

Perhaps local police bureaucracies advocated such legislation and joined in the drug war because they perceived it to be in the "public interest." Considerable evidence suggests that the opportunity costs of resources allocated to the war on drugs have been very high, however, and a good deal of evidence also indicates that many law enforcement bureaucracies created misinformation in order to exaggerate the potential benefits of a drug war (Michaels 1987, 311–24). This is relevant because if confiscations can be used at the discretion of local police bureaucrats to significantly enhance their own well- being, then this federal statute may explain a substantial portion of the changes in the allocation of local police resources after 1984. Local interbureau competition for resources may lead government decision makers (bureau sponsors) to treat confiscations as a substitute for ordinary appropriations. Therefore, an important component of the "Police Interests" section is the summary of a case study by Benson, Rasmussen, and Sollars (1995) of the budgetary impact of local police confiscations from the drug war. The findings are consistent with the hypothesis that confiscations legislation creates significant incentives to change the allocation of police resources.

Not all states were equal participants in the 1984–89 war on drugs. In a few states drug arrests accounted for a smaller portion of all arrests in 1989 than in 1984. These are the exceptions, however. The largest states, such as California, Florida, Michigan, Illinois, and Pennsylvania, increased drug enforcement to an extent that the U.S. drug arrest/total arrest ratio rose 67 percent between 1984 and 1989—from 5.8 percent to 9.7 percent. As table 1 shows, there is significant variation in drug enforcement activity across states as well as through time. If the "sin-tax/bureaucratic-self-interest" story actually provides a strong explanation of drug enforcement policy, it should help explain cross-sectional variation in enforcement policy as well as time series variation. This question has been explored in a cross-sectional analysis of cities' drug enforcement policies (Rasmussen, Benson, and Mast 1994). Since federally "adopted" seizures are only partially turned back to the local police (the federal authorities extract a 20 percent handling charge), police in states whose own laws allow them to retain seized assets are able to obtain even greater benefits from seizures than police who must involve federal authorities in the process. Thus, a self-interest view of police bureaucrats suggests that in states whose laws allow the police to retain seizures, police will focus relatively greater efforts against drugs than police do in states whose laws take such seizures away from the police. This expectation is supported by the empirical results in Rasmussen, Benson, and Mast (1994). These findings and their implications are explored in the "Differences in Drug Enforcement" section.

The escalation of the U.S. war on drugs ended in 1989. Indeed, after 1989, a substantial reduction in drug enforcement effort occurred. The concluding section offers several potential explanations for this down-cycle in drug policy, all of which are consistent with the sin-tax/bureaucratic- self-interest arguments that explain the 1984–89 escalation of drug enforcement.

Police Bureaucrats, Interest Group Politics, and Drug Policy

There are many models of bureaucratic behavior based on self-interest assumptions. Tullock (1965) saw bureaucratic behavior driven by a desire for security. Chant and Acheson (1972) contended that a desire for prestige drove bureaucratic behavior. Niskanen (1968, 1971) assumed that a bureau manager could be characterized as a budget maximizer. Migué and Belanger (1974) explained that budget maximization unduly limits the range of utility-maximizing efforts, however, and proposed that the bureaucrat seeks discretion reflected by a budget with excess revenues over actual costs (an argument Niskanen accepted [1975]).

Public officials are presumably characterized by the same basic utility-maximizing behavior that motivates people operating in private markets. The institutional framework of public officials may differ from that of private-sector employees, but their fundamental objectives do not. Employment security is a desirable job attribute in the private sector, for instance. The U.S. Department of Health, Education, and Welfare Task Force (1973) found that job security, along with interesting work and opportunity to develop special skills, were considered to be the most important features of job quality. Not surprisingly, civil service bureaucrats are also very concerned about job security (Johnson and Libecap 1989). The same point can also be applied to elected officials, whose desire to be reelected is consistent with this broadly held desire for secure employment. Surveys of private-sector employees identify other sources of worker satisfaction that may be equally applicable to elected and appointed public officials and civil-servant bureaucrats: good pay; discretionary authority, information, and opportunity to get the job done; and seeing the results of one's work. Discretion to control the intensity and pace of work is also an important job characteristic, and some research suggests that part of the wage premiums paid to workers in very large manufacturing plants is explained by the absence of this job attribute (Stafford and Duncan 1980). Thus, discretion itself may be a major source of satisfaction for bureaucrats and public officials (Parker 1992).

Job characteristics people value in a private setting are not likely to lose their allure just because someone is engaged in public-sector employment. In this light, Breton and Wintrobe (1982) noted: "In addition to size, budgets, discretion, prestige, and self- preservation, it has been suggested that security, the avoidance of risk or responsibility, secrecy, complexity, career promotion, leisure, internal patronage, and a bureaucrat's personal conception of the common…good are objectives of bureaucrats, either one at a time or in groups" (27). They suggested that all of these factors may enter a bureaucrat's utility function and that no general theory of bureaucratic behavior can be built by specifying a particular objective. Thus, they assumed general utility maximization and focused on the institutional setting (e.g., the intensity of interbureaucratic competition for budget shares and intrabureaucratic competition for promotions and positions in net- works, the existence of barriers to mobility, the ability of superiors and sponsors to monitor performance, and so on) as the determinant of which particular objective will appear to dominate in a particular bureau. Breton and Wintrobe characterized the bureaucratic institutional process as one dominated by "entrepreneurial competition," wherein individual bureaucrats pursue their subjective goals by selectively seeking and implementing policy innovations. This characterization fits the role played by law enforcement bureaucrats in the evolution of U.S. drug criminalization policy.

Actually, a number of self-interest political motivations for demanding drug legislation have been identified for both bureaucratic and nonbureaucratic interest groups. Some studies have noted the incentives of professional organizations such as the American Pharmaceutical Association to create legal limits on the distribution of drugs (there was significant competition between pharmacists and physicians for the legal right to dispense drugs, for example), whereas others have focused on the strong racial impacts of illicit drug laws and the desire by some groups to control racial minorities through the enforcement of such laws. More important from the perspective stressed here, however, others have emphasized that law enforcement bureaucrats were a major source of demand for criminalization of narcotics after the Harrison Act was passed and for passage of the Marijuana Tax Act and subsequent criminalization of this illicit drug.

Breton and Wintrobe (1982, 146–54) explained that one bureaucratic strategy to compete for resources is to "generate" demand for a bureau's own services through direct lobbying, policy manipulation, and the selective release of information to other interest groups and the media. This occurs because bureaucrats must compete with other bureaucracies for the support and attention of sponsors (and individual bureaucrats must compete with other bureaucrats for benefits within a bureau), and because the control of resources is necessary before bureaucrats can achieve most of their subjective goals. Indeed, Lindesmith (1965) contended that the nation's program for handling the "drug problem" is one "which, to all intents and purposes, was established by the decisions of administrative officials of the Treasury Department" (3). Why would the Treasury Department care about drug criminalization? Because the Harrison Act established federal taxes on narcotics, and the Treasury Department's Federal Bureau of Narcotics was responsible for its enforcement. For several years after its passage in 1914, the Harrison Act remained a rather unimportant source of taxes and regulatory measures (Reinarman 1983, 21). Indeed, its most important consequence may have been that a bureaucracy grew up to enforce the act. Given its regulatory powers, the Federal Bureau of Narcotics chose to instigate criminalization of opiate users with a series of raids on morphine treatment clinics in 1919. King (1957) maintained that "the Narcotics Division launched a reign of terror. Doctors were bullied and threatened, and those who were adamant [about treating addicts] went to prison" (122). Efforts by bureaucrats in the Narcotics Bureau led to a series of court decisions that reinterpreted the Harrison Act and became the pretext for criminalization of drug use (Reinarman 1983, 21). Furthermore, because of pressure from people in the same bureau, the Marijuana Tax Act was passed in 1937. Some writers have stressed moral entrepreneurship by Narcotics Bureau officials (e.g., Becker [1963]), but others have focused on bureaucratic fiscal self-promotion (e.g., Dickson [1968]). The bureau was in need of a new raison d'être for continued funding in 1937, as it faced stiff competition from the FBI for the attention of the public and congress (King 1978), both of whom wanted crime control to be in the domain of the FBI rather than the Treasury Department's Narcotics Bureau.

Breton and Wintrobe (1982, 39) emphasized that bureaucratic release of both true and false information, or "selective distortion," can play significant roles in bureaucratic policy advocacy. This has clearly been the case in the evolution of drug policy. For example, the bureaucratic campaigns leading to the 1937 marijuana legislation "included remarkable distortions of the evidence of harm caused by marijuana, ignoring the findings of empirical inquiries." The "reefer madness" scare traces to the misinformation propagated by the Bureau of Narcotics. Marijuana was alleged to cause insanity, incite its users to rape, and produce a delirious rage in users, making them irresponsible and prone to commit violent crimes. Distortions did not stop there. For instance, the bill was represented as largely symbolic in that it would require no additional enforcement expenditures (Galliher and Walker 1977).

The evolution of drug policy since initial legislation has also been, at least in part, shaped by bureaucratic competition, both between law enforcement and drug treatment bureaucrats over "ownership of the problem"—that is, over shares of federal, state, and local budgets (Gusfield 1980; Morgan 1983)—and between law enforcement bureaucracies themselves (e.g., between the DEA and the FBI [King 1978] at the federal level, as well as among various local, state, and federal bureaucracies). This evolution reflects another aspect of the bureaucratic process emphasized by Breton and Wintrobe (1982). As the perceived responsibility for some social ill (e.g., crime in this case, and inflation in Breton and Wintrobe's) shifts from outside forces to the government, and hence to the bureaucracy, bureaucrats seek to shift the blame elsewhere (Breton and Wintrobe 1982, 149). Blaming crime on people crazed by drugs provides an opportunity to shift blame.

A good deal of false or misleading information emanating from police bureaucrats about the relationship between drugs and crime has clearly characterized the evolution of drug policy. In fact, a primary source of the "information" (much of which was inaccurate and/or unsubstantiated [Michaels 1987, 311–24]) used to justify the 1984–89 war on drugs was the police bureaucracies. Primarily as a result of information promulgated by police (Barnett 1984, 53), it has become widely believed that drug crime is the root cause of much of society's problems (e.g., see the Office of National Drug Control Strategy [1990, 2]). In particular, drug use is claimed to be a leading cause of nondrug crime because, it is contended, property crime is a major source of income for drug users. This claim has served to justify political demands for the criminal justice system to do something about the drug/crime problem, demands that emanate largely from the police lobbies (e.g., see Berk, Brackman, and Lesser [1977]; Barnett [1984]; Benson [1990]; and Rasmussen and Benson [1994]), and in turn, to justify an emphasis on control of illicit drug traffic as a means of general crime prevention.

State and federal legislators have passed increasingly strict sentencing laws for drug offenders, police have shifted resources to make more drug arrests, and judges have sentenced increasing numbers of drug offenders to prison. Such a reallocation of resources would appear to be justified if drugs truly are the root cause of most other crime, but these causal relationships do not actually hold. In particular, increased drug enforcement efforts tend to cause increases in crime, as other types of crime are less effectively deterred. Thus, the opportunity costs of the war on drugs appear to be quite high. This should not be surprising, of course, given the history of failure of drug and alcohol prohibition policies (Thornton 1991; Nadelmann 1993). The question, "why has this reallocation occurred?" would appear to be even more pressing, given the circumstances.

Breton and Wintrobe (1982, 150–51) offer two reasons bureaucrats advocate a policy of direct control of a source of blame (e.g., alcohol prohibition, criminalization and prohibition of various drugs after 1914 and 1937, increased emphasis on drug control in the mid-1960s, and then again in the mid-1980s), even though such policies have a history of failure. First, there is always opposition to such a policy, so when it fails the opposition can be blamed for not allocating enough resources to combat the problem. And second, since the outcome of the policy depends jointly on the inputs of several different groups and bureaus, and the set of possible control methods is very large, when the subset selected fails, the bureaucrats can argue one or both of the following: (1) although they advocated a control policy, they favored a different subset of control tools (e.g., more severe punishment of drug offenders, greater spending on interdiction efforts), so they are not responsible for the failure; (2) the other groups who had to contribute to make the effort successful (e.g., witnesses, judges, legislators who approve prison budgets, other law enforcement agencies) did not do their share. Indeed, a policy can fail completely while at the same time entrepreneurial bureaucrats expand their reputations and end up being substantially better off.

The ongoing competition for a share of the total budget is always an important factor. After all, few of the subjective goals of bureaucrats can be achieved without a budget. Therefore, each bureau must demonstrate that it is doing a good job in serving its constituencies. The function of police in the minds of most citizens is to "fight crime," of course, but how can interest groups, voters, taxpayers, and elected representatives tell if police are doing a good job? The number of crimes prevented cannot be quantified. Therefore, police need statistical indicators of their "productivity" to use in their lobbying efforts for expanded budgets (Sherman 1983, 156). The number of arrests is a natural measure of "effectiveness," and this is a primary "statistic" police focus on in the budget negotiation process. Others include response times following emergency calls and, increasingly, asset seizures. Indeed, drug prohibition provides a source of arrests that does not require waiting for some victim to report a crime, and it provides a new statistical indicator of effectiveness: the value of drug seizures (and of nondrug property seizures, as will be discussed).

Under prohibition, police incentives may be even more "perverted" than suggested so far: there are actually incentives to allocate resources in order to avoid deterring Index I crime. After all, although arrest statistics may be primary indicators of police performance used in the budget bargaining process, they are not the only important statistic used in such bargaining. As Milakovich and Weis (1975) noted, police have a "vested interest" in keeping crime rates relatively high: if crime rates drop too much, then support for more police and larger budgets declines, and "like all bureaucracies, criminal justice agencies can hardly be expected to implement policies that would diminish their importance" (10). The literature on the economics of crime indicates that higher Index I crime rates clearly are correlated with more police resources in "demand for policing" equations, supporting the assumption that political demands for police services rise if reported crime rates are high. But if police do respond to the incentives outlined here, additional funding need not lead to any decrease in reported crime rates. Police can focus resources on drug control, which can lead to both increased arrests and drug seizures as indicators of effectiveness while, simultaneously, increasing Index I crime rates suggest a greater need for police services. But these incentives have been in place since drugs were initially criminalized. Something else must have changed in 1984 to produce the significant reallocation of policing resources documented above. Indeed, something else did change, at least for the police in many states. A bureaucratically motivated policy innovation appears to have created explicit incentives for shifting resources toward drug enforcement. This innovation allowed police agencies to benefit directly through the collection of a sin tax: the police were given the opportunity to collect a tax in the form of confiscations of money and property used in or purchased with profits from the drug trade and to keep the proceeds from this tax.