The past three decades have witnessed a stormy and controversial debate about the possible merits to society that might be brought about by decriminalizing or legalizing marijuana. Beginning in 1973 with Oregon, 12 states (Alaska, California, Colorado, Maine, Minnesota, Mississippi, New York, Nebraska, Nevada, North Carolina, Ohio, and Oregon) have in some manner altered their existing laws to reduce the penalties for marijuana possession.1 A number of local cities have also modified their local ordinances and criminal justice practices to decriminalize pot (Berkeley, Oakland and San Francisco, California; Breckenridge, Colorado; Amherst, Massachusetts; Madison and Milwaukee, Wisconsin; Urbana and Carbondale, Illinois; and Colombia, Missouri; among others).

There are three central arguments supportive of the decriminalization movement which have been advanced in these and other jurisdictions. Perhaps the most powerful and appealing argument for marijuana decriminalization (and/or decriminalizing other drugs) is that it would save a huge amount of government money now being spent on the enforcement of such laws. The basic tenets of the cost saving argument can be summarized as follows:

1. The criminal justice system, ranging from police to corrections, now allocates a significant portion of its budgets arresting, prosecuting, sentencing and incarcerating marijuana users, dealers and others involved in the illegal drug infrastructure (e.g., transporters, manufacturers of drug paraphernalia, etc.).

2. If these behaviors would no longer be labeled as criminal, criminal justice agencies would reduce the enforcement and processing tasks now associated with such crimes.

3. There is a direct relationship between the proportion of arrests or cases processed for marijuana crimes by the criminal justice system and the amount of money expended by these same agencies.

4. By reducing or eliminating these marijuana related events, there would be a proportionate decrease in the agency expenses.

This perspective has led to some fairly substantial claims regarding the amount of money to be saved by taxpayers if marijuana were decriminalized. For example, a recent study by Scott Bates (2004) claimed that Alaska was spending $25-30 million per year enforcing marijuana prohibition laws. Further, since there is no link between marijuana use and criminal behavior, there would be no impact on crime. And, if the purchase of marijuana were to be taxed as a legal commodity, tax revenues would increase by about $10-20 million per year. So, voters were promised that a net swing of $35 - $50 million per year would appear in the state’s coffers if marijuana were decriminalized.2

Jeffrey A. Miron from MIT made a similar claim in his assessment of drug laws in Massachusetts.3 Applying the same assumptions used by Bates, he estimated that the state would reduce its criminal justice expenditures by $120.6 million per year. The largest savings would be in the courts ($68.5 million), followed by police ($40.3 million) and corrections ($13.6 million). And, Michael Aldrich, Tod Mikuriya, and Gordon Bronwell have claimed that California’s pioneering decriminalization law (SB 95) was generating over $30 million per year in reduced police costs.4

On a national level, a recent report by Jon Gettman estimated that national criminal justice expenditures for enforcing marijuana laws is $7.6 billion per year with $3.7 billion being allocated to police, $853 million to the courts, and $3.1 billion to corrections.5 To his credit he noted that his estimates were “maximum” costs recognizing that actual costs and savings may be significantly lower. Similarly, the 2005 report by the Sentencing Project states that $2.1 billion is being spent by police on enforcing marijuana laws.6

All of these studies use a proportionate cost model. This approach was adopted by the Office of Drug Control Policy (ONDCP) on the Economic Costs of Drug Abuse.7 This report totaled all of the criminal justice costs as well the major activities, events and/or people processed by the criminal justice system. Next, the proportionate numbers of activities or persons that could be related to marijuana crimes (possession and sales) were totaled. The costs associated with these events or persons were based on their proportionate size to the total criminal justice events or persons. This report did not separately analyze marijuana related costs but did claim that the nation was spending $12.1 billion in police and court costs and another $16.9 billion in corrections costs.

In making such estimates for marijuana crime, the methodology would work as follows. If the number of arrests for marijuana in a given jurisdiction reflects 10% of the total arrests, it is argued that the legalization or decriminalization of marijuana would produce a 10% drop or savings in the total law enforcement costs. The primary problem with these estimates is that while they accurately reflect the proportionate level of costs they are not useful in estimating the savings to be realized if marijuana sales and possession were no longer criminalized. In fact, this somewhat simplistic and static cost benefit model generates highly misleading and exaggerated cost savings claims because it fails to recognize that government agency budgets are relatively fixed and operate independent of the level of activities or events (arrests, prosecutions, and sentencing) reported by the agency.

Indeed, it will be argued in this paper that decriminalization will have only a marginal impact on criminal justice costs. This is not to say that decriminalization would have a trivial effect on costs or that it should not be aggressively pursued by state and federal policy makers. The major benefit of decriminalization, in addition to eliminating the needless arrest, prosecution, and court disposition of over 700,000 people each year, would be the ability of the criminal justice system to focus on more important public safety activities.

Finally, the author would like to acknowledge that there are may other “prohibition costs” that are not covered by this report. These are “costs” incurred by marijuana users such as civil forfeiture, driver license suspensions, drug tax stamp violations, loss of access to federal student loans, removal or suspension from public housing, loss of welfare payments, denial of employment opportunities with government and private agencies, loss of professional licenses and the costs of mandatory drug treatment.

Impact of Decriminalization on Marijuana Use

One concern is that decriminalization will result in greater use of the drug. If true, there could be associated increases in the non-criminal justice costs associated with medical, mental health and even losses in worker productivity as access to and use of the drug rises. One would be hard pressed to suggest that decriminalization would result in a decrease in use, but what is the evidence that decriminalization would result in a significant increase in use?

First it must be emphasized that marijuana, despite its criminalization, is widely used by a large proportion of the US population. The most recent national survey conducted by the US Government reported that 25.8 million people or nearly one out of ten US residents use marijuana at least once a year and over 6% of the population uses the drug on a monthly basis. The lifetime use estimate of marijuana is a hefty 95 million, meaning that 47% of all adults have tried marijuana at least once.8

Unlike alcohol and tobacco use which follows more of a daily pattern, marijuana smoking follows more of an occasional use pattern (once a week or once a month). The demographics of the user population show that it matches the US population with respect to race but that users are disproportionately younger males which also suggests the use of the drug reflects more of youthful experimentation pattern rather than an enduring or long-term life pattern of steady use. The point is that since marijuana is at best a low risk to addiction and that most “users” wean themselves off the drug, its unlikely that making it more available would have a significant increase on these well-established use patterns.

If one looks at selected studies of jurisdictions that have decriminalized the drug, the evidence is either no increase or a slight increase among those segments most likely to use the drug. A study by Rosalie Pacula, Jamie Chriqui, and Janna King9 examined several states where marijuana had been decriminalized by reducing the penalties for simple possession. They found that living in a decriminalized state increases use among high school students but by only 2%. Other studies have found either no increases among juveniles or as much as 4% increase.

Impact of Decriminalization on Crime

Another area of concern is that decriminalization will lead to an increase in crime and thus increase the need for criminal justice resources to combat growing crime rates. This view is based on what some have referred to as correctional analysis. It begins with the well known fact that a high proportion of the 7 million plus persons under the control of the adult correctional system (jail, probation, parole and prison) have recently used an illegal drug. The logic of the “drug use causes crime” argument is that since so many “offenders” have used illegal drugs than it must be the cause of criminal behavior. One could also argue that the consumption of milk is a cause of crime since all criminals have consumed milk – some just before they committed the crime.

The problem with this premise is that there are many other factors associated with criminal behavior that are also associated with drug use. These other demographic factors (e.g., age, gender, etc) and socio-economic factors are more powerful causal factors of criminality. As such drug use, either because it predates criminal behavior or is more associated with the lifestyle of lower socio-economic male life-styles, may simply represent a spurious relationship with crime.

On the narrower question of marijuana, there is little evidence of a direct link between marijuana use and criminal behavior. Marijuana is neither a gateway drug (smoking marijuana necessarily leads to the use of more serious drugs like cocaine and heroin) nor one that necessarily propels one to criminal activity with the obvious exceptions of either possession or distribution of the drug. As noted earlier, over 25 million people consume marijuana each year with nearly 15 million using the drug in the past month. By way of comparison, very small percentages of Americans (under 4 % in total) have used cocaine, crack, heroin or other inhalants in the past 30 days.10 So somewhere along the line, the vast majority of the marijuana drug users do not graduate to the more dangerous drugs.

David Boyum and Mark Kleiman in their review of drug control policies note that of three major illicit drugs (marijuana, cocaine, and heroin), marijuana is the least likely to generate criminal activities. This is due to the method of dealing (discreet), amount of drug required to get high (small), the fact that high itself is highly unlikely to trigger violence, and because social circles surrounding marijuana are largely white and from the middle and upper socio economic classes. They conclude that:

“ Making marijuana legally available to adults on more or less the same terms as alcohol would tend to reduce crime…”11

Finally, a proponent of the marijuana use equals crime premise would have to explain why although marijuana use has been increasing at a modest but steady rate since the early 1990s, the crime rate has plummeted. If there was a marijuana-crime link, one would have expected crime rates to increase and not decline.

Marijuana’s Impact on Arrests and Court Processing

Despite the large swell of legislative activity at both the state and local level, the number of persons arrested for marijuana possession and sale has grown significantly. In 1970 there was an estimated 188,682 arrests for the drug – by 2003 the number had increased to 755,000. So it’s clear that despite the decriminalization effort, the chances of a marijuana user being arrested have significantly increased.

Although a large and increasing number of persons are arrested each year for marijuana violations, as a proportion of total number of criminal justice arrest and felony court convictions, marijuana cases are relatively low percentages of police and court’s workload. For example, In 2003 there were 13.6 million arrests made by police agencies meaning that marijuana arrests represent about 5-6% of the total arrests (see Table 1). This also means that based on the large number of marijuana users (nearly 26 million) the probability of a marijuana user actually being arrested is even lower. Gettman calculates the overall arrest rate per 100,000 population at 250 which is twice the rate it was in the early 1970s, but remains relatively low compared to the overall arrest rate for all crimes (in 2001 it was 4,800 per 100,000 population).

These data show that marijuana and marijuana possession is a relatively small piece of the arrest pie (about 4 %). Eliminating the marijuana arrest pool would have only a marginal effect on the universe of arrests and the workload of the police.

The same can be said about the impact on the courts. There are an estimated one million felony convictions by state and federal courts. Of this number only 69,500 convictions were for marijuana violations of which 44,200 are possession convictions. These data do not include the work of the lower or municipal courts where most of police arrests are handled. Obviously the remaining 600,000 plus marijuana arrests must be handled by some court agency. For example, in California of the 1.2 million adult arrests, over half (801,506) are misdemeanor level crimes of which 100,000 are drug violations -- all of which are processed through the Municipal Courts. Unfortunately, there are no national estimates of municipal or lower criminal court activity. But one can assume that marijuana cases do not constitute a major portion of the lower court’s dockets.