HomeMy WebLinkAboutCOM 0193.096 2006-2008 •r
t~a~ Jli;;i ~ c,ni ~ .
1 J
A
+A
qll
I~
A
The Budgetary Implications of Marijuana
Decriminalization and Legalization for Hawai i
r
71Y
w
Y
Lawrence W. Boyd, Ph.D.
Economist
University of Hawaii West Oahu
~ Lboyd@hawaii.edu
¦
d q
Cpn1m. No. ~ "l 3
Re}. Toy
R~f• Data
i
Executive Summary
' Economic analysis of current public policies on marijuana reveals that Hawaii state and
county governments could reap up to $33 million annually in new revenues and cost
savings if tax and regulatory policies were to replace law enforcement to control
marijuana distribution. Furthermore, research indicates that enforcement expenditures of
~ up to $10 million each year statewide have failed to reduce the amount of mazijuana
available in Hawaii.
~ This report focuses on the economic effects of two alternative policies:
Decriminalization of marijuana is a policy that reduces the punishment for its possession
to a civil fine rather than criminal penalties or jail time. Trafficking, selling, and
distributing to minors, remain subject to standard criminal punishment.
Leealization is a policy that would eliminate criminal and civil penalties for both
possession and sale of marijuana and replace them with regulation, which would include
restrictions on mazijuana use similar to those applicable to alcohol and tobacco. The
regulation model uses taxes, minimum age requirements, and licensing to control
distribution.
Currently, thirteen states have decriminalized marijuana possession. Spain, Portugal,
Luxembourg, Belgium, and Austria have decriminalized marijuana possession; in
addition, there are seven other countries either considering decriminalization or having a
de facto policy that in essence, decriminalizes or legalizes marijuana (e.g. the
Netherlands).
The primazy cost of the criminalization ofmarijuana is law enforcement. In Hawaii,
possession of less than one ounce of marijuana is a petty misdemeanor. Approximately
65 percent of the cases are dismissed, not prosecuted, or stricken in any given year. First
~ offenses generally receive probation or a deferred acceptance of a guilty plea. Given the
current usage levels, the low risk of arrest, and further risk of punishment, the current
criminalization policy is not deterring marijuana use.
r
The report concludes:
a
• State and county law enforcement agencies spend $4.1 million per year to enforce
marijuana possession laws; an additional $2.1 million is spent by the courts.
~ Enforcement of marijuana distribution laws costs approximately $3 million. The
total costs of enforcing all marijuana laws in Hawaii are approximately $9 to 10
million per year.
a
Between 1994 and 2003, the price of one ounce of high quality marijuana dropped
by 12 percent. The price decline reveals that law enforcement efforts to restrict
• supply have not been effective.
• Reseazch on the effects of decriminalization has tended to find either no
relationship or a weak positive relationship between decriminalization and drug
use. Given the current low prosecution levels and small penalties, it is doubtful
that decriminalization would have much effect on marijuana use in Hawaii.
• Decriminalization of marijuana possession in Hawai' i would save state and
county governments approximately $5 million per year.
Legalizing, taxing and controlling marijuana would save an additional $5 million per year
and would create tax revenues of between $4 million and $23 million.
,r
r
a
w
s
al
•
t
Table of Contents
Introduction 1
Methodology of Estimating Budgetary Implications of Decriminalization and
Legalization 3
Budgetary Effects 5
Implications of Decriminalization and Legalization on Use 12
Decriminalization 12
Legalization 14
Conclusion 1'7
References 1 g
r
r
'r Research for this report was underwritten by the Drug Policy Action Group
rr
Introduction
Decriminalization of marijuana is a policy that reduces the punishment for possession
of marijuana to a civi] fine rather than a criminal offense. Trafficking and se]ling, or
distributing to minors, remain subject to standazd criminal punishments.
Decriminalization is apolicy that has been substantially discussed nationally. A number
of states have decriminalized mazijuana possession.
y' In contrast, legalization would decriminalize both possession and sale of mazijuana
and replace them with a system of regulation and possible taxation. There is also a
substantial literature on legalization of vazious drugs although no state or national
government has actually legalized marijuana.
This study addresses three issues related to marijuana decriminalization, and
" legalization. First, what savings from legalization or decriminalization can be expected to
occur in state and local budgets. Second, in the case of legalization, what tax revenues
could be projected. Third, what would be the impact of these measures on marijuana use.
The United States, like other countries, has chosen to regulate some substances that
aze addictive, or potentially addictive, such as cigarettes and alcohol, and ban others. ~
Regulation uses taxes, minimum age requirements, other restrictions on use, and
education about harmful effects in order to limit the potential damage these goods can do.
Bans involve outlawing the use of certain substances. Taxation and bans both raise the
price of these substances; taxation directly raises the price, while bans limit supply. In
r addition, bans create black markets, encourage illegal activities, and may result in harm
to innocent victims?
,r
~ Potentially addictive means a relatively small part of the population can become
addicted to a substance. Alcoholism, for example, is recognized as a disease, and those
addicted to alcohol represent a small percentage of those that consume alcohol on the
order of 15 percent. The addictive nature of marijuana is questionable. Those who
support its continued ban claim that it is a "gateway drug" whose use leads to more
hazmful drugs; although recent reseazch disputes that theory.
z This discussion follows that of Michael Grossman, "Individual Behaviors and
Substance Use the Role of Price," National Bureau of Economic Research Working
,R Paper 1048.
r
1
s
During the 1970's, eleven states decriminalized mazijuana possession, as have a
number of countries since.3 Currently thirteen states have decriminalized marijuana
, possession. Proponents of decriminalization argue that it can have positive outcomes that
include savings on enforcement for state and local governments, an improved allocation
of criminal justice resources, and expanded funding for prevention education and
treatment for marijuana users. Opponents have claimed that decriminalization produces a
substantial increase in marijuana use along with increased crime and other negative
effects.
Those who favor legalization point to the inefficient use of social resources and azgue
that policies like those involved in the regulation of alcohol and tobacco are far more
effective in limiting the individual and social costs involved. There is also a significant
literature that suggests legalization could also be more efficient in limiting the negative
consequences of marijuana use. Taxation, for example, can ameliorate whatever social
costs occur, and the price effects can significantly reduce its use, especially among
adolescents a Furthermore it can be demonstrated that taxation is significantly cheaper in
terms of enforcement and outcomes than outlawing substances.
This report reviews evidence and literature that suggest marijuana decriminalization
would not lead to a measurable increase in marijuana use. This report does not take a
- stance on whether, or not, mazijuana use is harmful. The conclusion reached below is that
decriminalization would save state and local governments in Hawaii approximately $5
3 Alaska, California, Maine, Minnesota, Mississippi, Nebraska, New York, North
- Carolina, Ohio and Oregon decriminalized. Alaska (1990) voted to recriminalize, but
Alaska's state courts have ruled that privacy rights protected marijuana use in the home.
A twelfth state, South Dakota decriminalized and then recriminalized within a year. In
1996 Oregon recriminalized, but in 1998 voters rescinded recriminalization and returned
to decriminalization. Nevada decriminalized in 2001. Colorado has also decriminalized.
See http://www.norml.or¢/ and http://www.dru~policy.org/ for details. Countries that
~ have decriminalized aze Italy, Spain, Portugal, Luxembourg, Belgium, and Austria.
Several other countries have either de facto decriminalized or are in the process of
decriminalizing. These include Netherlands, Germany, Denmark, France, Switzerland,
United Kingdom and Canada. See http://eldd.emcdda.orel.
n See Becker, G. S., M. Grossman, et al. (1994). "An Empirical Analysis of Cigarette
Addiction." American Economic Review 88(3): 396-418. They find prices have a greater
effect on adolescents.
2
rr
million dollars per year. Legalization would save an additional $5 million per year for a
total of $10 million should legalization be adopted. In addition, legalization would create
tax revenues between $4 and $23 million. The estimates provided here must make use of
approximations in cases where data do not make possible a more detailed analysis.
Wherever possible this report has used detailed information and approximations that bias
estimated budget effects downward.
r
Methodology of Estimating Budgetary Implications of
Decriminalization and Legalization
By and large this report follows the methodology used by Miron (2003) in his
analysis of the effects of decriminalization in Massachusetts. 5 As he suggests, the two
w major budgetary implications of decriminalization are the savings in criminal justice
resources and criminal fines that are shifted to civil fines. The former is the savings that
result to the extent that police, prosecutors, forensic laboratories and court personnel are
not used for marijuana possession offenses. Miron (2003) suggests that the savings on
law enforcement is the predominant one. The amounts and collection of civil as opposed
to criminal fines would likely be at about the same level; therefore the second effect
~ would be small. A look at court statistics, reported below tends to confirm this.
The methodology used by Miron involves the following steps:
1. Determine the percentage of all Hawaii arrests that is for marijuana
possession.
2. Determine the criminal justice budget for Hawaii
3. Multiply the first number by the second
As Miron points out, these steps yield reasonable estimates based on certain
assumptions. First, that average costs equal marginal costs. This means that law
r enforcement is a constant cost industry; increased dollars spent on enforcement leads to
w
'r s See, Jeffrey A. Miron, "The Effect of Mazijuana Decriminalization on the Budgets of
Massachusetts Governments, With a Discussion of Decriminalization's Effect on
Marijuana Use," Drug Policy Forum of Massachusetts. Also Jeffrey A. Miron, "The
~ Budgetary Implication of Mazijuana Decriminalization," June, 2005, The Marijuana
Policy Project.
w
3
w
approximately the same increase in arrests. Police also engage in activities unrelated to
arrests, such as traffic control, but these sorts of activities are minor in terms of costs
.,a
relative to the overall law enforcement and arrests.
ww
Miron (2003) focused primarily on police enforcement of marijuana laws and did
not quantify costs associated with prosecution within the court system. Because court
statistics make it possible to do this in Hawaii, this report adds these costs to
enforcement costs. Court and legal costs, related to prosecution and defense, aze reported
in addition to enforcement costs.
A second question is what exactly does decriminalization mean? Currently under
R Hawaii law marijuana possession is a misdemeanor. Possession of less than one ounce is
a petty misdemeanor punishable by 30 days in jail and or a fine up to $1,000. Possession
of between one. ounce and one pound is a misdemeanor punished by up to one yeaz in jail
and up to a $2,000 fine. One proposal, HB 1751 and SB 1056, introduced in the 2005
state legislative session, decriminalizes possession of less than one ounce. Generally,
however, the data do not break down arrests by weight, but rather record arrests for
possession that include all types of misdemeanors. So the results below are calculated for
all misdemeanor mazijuana possession charges, although the vast majority of these appeaz
to be for under one ounce.
The statistics on arrest are Uniform Crime Statistics reported by law enforcement
agencies to the United State Department of Justice. Other statistics used come from
Reports of the Hawai' i State Judiciary and the U. S. Census of Governments.
- A difficulty raised by Miron (2003, 2004) is that some arrests are the result of an
investigation related to a different crime. Thus these arrests can be broken down into
three categories. The first are "stand alone" arrests, where someone is arrested because an
officer sees them smoking mazijuana. A second type is an arrest made in conjunction
with a traffic stop, also referred to as "civil incidental". A third type is an arrest that
occurs because police have detained a suspect for a crime and then find that the suspect
possessed marijuana, known as a "criminal incidental" arrest. Generally the police
'r resources saved under decriminalization would correspond to the first two categories.
Miron states, "it is useful to know what fraction of arrests are in these first two
4
rr
categories `stand alone' and `civil incidental' as opposed to the criminal incidental."6 As
d
a result, Miron finds the proportion of stand alone arrests and reduces the total of an•ests
.R by this amount. A review of the data definitions used in reporting Uniform Crime
Statistics indicates that in the case of multiple counts the most serious charge is the only
one reported. Given the law in Hawaii and the nature of the arrests, marijuana possession
is only more serious than traffic violations. Thus it is more accurate to use the arrest
statistics reported for Hawaii under the Uniform Crime Information System without
adjustment. In this instance this report differs from Miron.
7 Estimating the budgetary implications of legalization is somewhat more difficult.
In terms of enforcement costs the same procedure used in analyzing decriminalization is
employed. Because additional costs include jail time for offenders, and because actual
numbers exist for persons incarcerated and associated costs, these are added. Taxation,
r, which would play a significant role in legalization, requires assumptions about the public
policy that would be pursued, the social costs of legalization, and estimates about how
much supply would increase. Miron (2004) reports $4 million per year as potential tax
revenue from legalization in Hawaii. A different methodology would be to use per capita
tax revenue from alcohol and tobacco as a basis for estimating tax revenue.
Bud etary Effects
Table 1, below contains possession arrests, total arrests, percentage of possession
arrests, county police expenditures and the cost of enforcement. Arrests are from the
Uniform Crime Statistics of the Department of Justice. County police expenditures are
from the United States Census, State and Local Government. These statistics cover 1998
through 2002. The Census does not provide state and local data for every yeaz. Although
ar
alternative statistics and budgets are available, using Census data combines state and
r
e Ibid. p. 4
~ Miron reduces Massachusetts' arrests by two-thirds as a result of this adjustment. Thus
°'r this means there would be significant differences between the two reports. However the
difference is data driven. Miron finds 1.7 percent of all arrests in Massachusetts in 2000
would fall into these two categories. In 2000, in Hawaii 1.72 percent of all arrests were
'r mazijuana possession arrests. Given similar marijuana use patterns and risk of arrest this
indicates that the results are compazable when the revision is not used.
5
county police expenditures that include many anti-drug prograzns such as the state's drug
interdiction program at the Honolulu airport. It also facilitates national comparisons.8
Following the methodology above, the expenditures on enforcement that would be
reduced by decriminalization are approximately $4.2 million per yeaz. Note that the data
® fluctuate around this number over several yeazs. Because Hawaii does not have the same
sort of statewide law enforcement agencies other states do, this is a burden in Hawaii
primarily for county govermnent.9
Table 1
State of Hawai i Possession Arrests, Total Arrest, Percent of Total Arrests,
Police Bud ets and Cost of Possession Enforcement
Arrests Percent State and Cost of
Marijuana Total Marijuana Local Police Possession
Year Possession Arrests Possession Bud ets Enforcement
W. 1997 1,411 70,060 2.01%$192,287,000 $3,872,637
.e 1998 1,257 63,208 1.99% $207,743,000 $4,131,328
1998 1,232 61,393 2.01% $221,151,000 $4,434,331
r
2000 1,152 64,685 1.72% $221,899,000 $3,951,884
^ 2001 1,142 60,177 1.90%NA NA
2002 1,032 63,021 1.64%$254,636,000 $4,169,790
2003 1,098 58,722 1.87% NA NA
2004 1,054 58,547 1.80%263,768,000 $4,748,518
' Sources: Arrests from HawaP i State Department of the Attorney General, "Crime in Hawaii,'
ham://hawaii.govlag/~ia/main/rs/Folder.2005-12-05.2910; Police Budgets, U. S. Census Bureau,
Federal State and Local Governments, State Government Finances,
http://www.census.gov/govs/www/state.html
Related to the arrest statistics are the criminal proceedings that follow. The U. S.
Census Bureau reports that during the fiscal yeaz 2001-2002 the budget for state and
county judicial and legal services was $213, 854,000.10 Approximately 1 % of the total
r
s For example police budgets are available in county Comprehensive Annual Financial
"r Rye orts. Summing these county numbers does not always produce identical results due
to minor differences in definitions between county reports.
9 The Hawaii State Department of Public Safety participates with local and federal
a~encies in statewide drug enforcement and interdiction efforts.
~ This includes state judiciary expenses, county prosecutors' offices, drug courts and the
r attorney general's office. Probation and investigation related to sentencing is also part of
this budget. For 2001-2002 see:
http://www.census.gov/govs/estimate/0212his1 l.html
6
.r
criminal cases heazd were for marijuana possession. ~ Multiplying the 1% by the $213
million means a total of $2.1 million is used in various court-related possession activities.
a. Also of interest is the actual disposition of these misdemeanor drug cases in
District Court. Approximately 65% are dismissed, not prosecuted, or stricken in any
given year. A very small number is committed to Circuit Court for jury trial and a
relatively small proportion, about 25°10, results in conviction. In addition first offenses
generally get probation. ~z
Table 2
Dis osition of District Court Misdemeanor Dru Cases
r
By By Commitment
Discharge/ By Nole to Circuit Court By
Year Dismissal Prose ui Stricken Ju Trial Conviction Total
i
2000 300 31 3 39 150 523
2001 276 20 4 37 124 461
2002 310 36 2 42 127 517
2003 304 42 12 85 141 584
Source: The Judiciary State of Hawaii Annual Report Statistical Supplement, Various Years,
Table 17.
These statistics tend to confirm that the primary cost of the criminalization of
marijuana criminalization is enforcement. Few are actually prosecuted under the law,
fewer convicted, and virtually none serve jail time. Of those convicted, probation is the
usual sentence for first time offenders. The burden in terms of enforcement costs fall on
county level enforcement efforts.
r
Table 3, below, reports estimates of marijuana use from the National Survey on
Drug Use and Health. Beginning in 1999 this survey was expanded so sample sizes were
large enough to cover states. Combining samples into two-year averages is a means by
r
See The Judiciary State of Hawaii "Annual Report Statistical Supplement." Various
Years; misdemeanor drug cases which are identical to misdemeanor marijuana possession
~ are reported. These numbers are divided by the total criminal cases heard in district and
circuit court to arrive at the 1 percent figure.
12 In addition some are offered deferred guilty pleas, where a defendant is released on
~ good behavior, provided he or she does not re-offend. These outcomes will also be found
r in these statistics since those who do re-offend are found in the convictions category.
s
7
a.
w
T which trends can be measured. A change in the methodology of the survey greatly
expanded it in 2003 and also greatly expanded reported mazijuana use; therefore 2002-
2003 is not directly comparable to previous time periods. The question on past year use
ar
was added in 2003. During 2002-2003 about 7% of the population over twelve used
marijuana within the past 30 days, a measure indicating regulaz users. Almost 12% used
it in the last year, indicative of infrequent users. Adjusting the number of arrests so they
~ match the time periods of the survey, makes it possible to estimate the risk of arrest for
regulaz marijuana users. This was 1.5% in 2002-2003. In terms of punishment it was
effectively zero.
Table 3
Mari uana Use in Hawai i, National Household Sorve on Dru Abuse
Percentage 30 Percentage
Year 30 Da s Use Past Year Use Da s Year Risk of Arrest
2003-2004 66,000 110,000 6.52% 10.80% 1.59%
2002-2003 69,000 115,000 6.95% 11.56% 1.54%
,r 2000-2001 55,000 NA 5.82% NA 2.09%
1999 57,000 NA 5.80% NA 2.16%
Source: httD://oas.samhsa.gov/nhsda, Substance Abuse & Mental Health Services
Administration, Office of Applied Studies
Criminal justice resources used in the enforcement of the current marijuana law
d. on possession in Hawaii aze about $6 million dollazs. Given the usage levels, risk of
arrest, and further risk of punishment it is questionable whether criminalization serves as
+ a deterrent to use. This can be further analyzed by looking at reseazch done nationally on
use.
i
al°
7
i
8
+
Table 4 below reports arrests for distribution and the costs associated with it.
b
Those costs were approximately $1 million in 2002.
1
Apr
Table 4
Mari uana Distribution Arrests
.r
Marijuana- Marijuana- Cost of Manufacture
Year Sale/Manufacture Sale/Manufacture Enforcement
1997 210 0.30% $576,861
1998 166 0.26% $545,697
1999 159 0.26% $568,208
2000 167 0.26% $574,859
2001 125 0.21 % NA
2002 240 0.38% $969,749
2003 159 0.27% NA
2004 110 0.18% $495,576
Sources: Arrests from Hawaii State Department of the Attorney General Uniform Crime Report,
http://www.cpja.ag.state.hi.us/rs/cih/index.shtml. Police Budgets, U. S. Census Bureau, Federal
State and Local Governments, State and Local Government Finances,
http://www.census.gov/povs/www/state.html
Stronger enforcement efforts are directed at suppliers of marijuana. Assuming that
a11240 distribution arrests were tried during fiscal year 2002, and that there were
approximately 56,000 cases terminated that yeaz by the judiciary which had a budget of
approximately $214 million, then the court costs were approximately $850,000.
According to the Department of Public Safety's Budget Office in 2005, each prisoner
costs the state approximately $38,000 dollars per year to incarcerate. There were between
14 and 21 prisoners incazcerated for marijuana distribution over the past several yeazs.
These prisoners cost the state between approximately $582,000 and $800,000 in direct
costs. 13 In addition to enforcement efforts directed towards finding dealers, there are
special units directed at eradicating marijuana plants. The federal government funds
these, in part, with additional funds provided at the local level. Matching three to one
13 It should be noted that Hawaii prisons suffer from severe overcrowding to the point
that mandated court ordered relief has led to the export of prisoners to private prisons on
the mainland. Overcrowding has other indirect costs including early release of prisoners
and the potential return of these prisoners to criminal activity. See Ilyana Kuziemko and
'r Steven Levitt, "An empirical analysis of imprisoning drug offenders, "Journal of Public
® Economics, 88, 2004. Discussed further below.
.r
9
M
grants from the federal government fund various mazijuana eradication programs. These
total $360,900, so the one-third matching funds would be approximately $110,000.
The costs of enforcement of distribution laws appeaz to be approximately $3
million. Thus the total costs for enforcing all of the marijuana laws aze in the $9 to $10
q million range.14
Enforcement efforts should restrict the supply of an illegal substance to such an
,r extent that they affect the price of that product. Therefore it is useful to report prices for
various amounts of marijuana, which is done in Table 5. Figure 1, plots the "real" price
of marijuana in Honolulu, that is the price of marijuana net of inflation. The real price
indicates what its cost is relative to the other prices of goods that consumers buy. The
price series indicates a 12% drop in the real price of one ounce of high quality marijuana
between 1994 and 2003. Given vazious estimates of the relationship between marijuana
prices and consumption, this would suggest a 6% increase in mazijuana consumption.
14 A minimum of $8 million, plus other legal costs. Miron (2005) estimates that the total
cost of marijuana prohibition in Hawaii is $22 million. The differences between his
estimates and these found here are first, I estimate police resources at $4 million while
Miron estimates them at $2.7 million. The reason for this is that Miron reduces these
arrests to stand alone an•ests, a procedure I believe is flawed in Hawaii for reasons
outlined above. Second, Miron was unable to find data on the percentage of possession
r convictions for marijuana and assumes it was equal to the percentage of trafficking
convictions. In turn he estimates trafficking convictions at 10.9 percent, which he then
multiplies by the judicial budget for 2002. In this report it was possible to determine
actual possession hearings. It was further assumed that all trafficking arrests were heard
in the following yeaz. The difference is substantial. Miron estimates judicial costs at
,w $19.6 million, while using the actual numbers found in the Hawaii State Judiciazy
Reports were in the $1 million range. In terms of incarceration, Miron estimates that 1
'r percent of the penal system's budget is used for incazcerating marijuana prisoners. His
estimate is $1.96 million. In this report the actual number of prisoners are used and
multiplied by per capita prisoner costs. The result is $910,000. Although the results aze
~ different from Miron's estimates, it should be noted that the difference lies in the fact that
a large proportion of the court's resources aze not used for marijuana enforcement.
.r
10
w
Table 5
Honolulu Mari uana Prices
Ounce-High
Year Joint Gram Qualit Pound
'w 1994 $3-$5 NA $400-$800 $6000-$9000
1995 $5 $25 $350-$700 $5000-$9000
1996 $5 $25 $350-$700 $5000-$9000
1997 $3-$10 NA $400-$800 $6000-$9000
1998 $3-$10 NA $400-$800 $6000-$9000
1999 $3-$10 N $400-$800 $6000-$9000
a
2000 $3-$10 N $400-$800 $6000-$9000
2001 $5-$20 $25 $400-$800 $6000-$9000
2002 $5-$20 NA $400-$800 $6000-$9000
2003 $5-$20 $25 $400-$800 $6000-$9000
2004 $20-$40 $300-$550 $6000-$9000
Source: Hawaii Community Epidemiology Working Group, National Institute of Health
Figure 1
Real Price of One Ounce of Marijuana in Honolulu Hawaii
E7.000~
,r E900 - - - - -
5600 - - ------_~.-_--_---a_-
.w
E700 _
~ E800
r
E500
w ~ ~ _ . - - - ~ _
E400 -
E300 _
ar
E200
• E100 _ -
1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
E
~LOw Price for One Ounce High Oualiry -High Price for One Ounce High pualiry
r
Legalization can affect the price of mazijuana in two ways. First, supply can
increase and thereby prices will fall. In the Netherlands, where mazijuana possession laws
aze laxly enforced, the price of marijuana is between 50% and 100% of U. S. prices. On
11
the other hand, taxes can raise the price of marijuana. Miron (2005) chooses two tax
regimes; a relatively normal one and one that imposes "sin" taxes on mazijuana that aze
equivalent to those charged for alcohol and tobacco. He then allocates a national number
to each state based on either consumption or population, and divides that total between
federal and state taxes. Miron's estimate for Hawaii tax revenues is in the $4 million
range. By contrast, tax revenues for tobacco in Hawaii during 2003 were $77.5 million
,r and for liquor were $41 million, with 462,000 adults consuming alcohol in the last 30
days and 221,000 consuming cigazettes.~s This provides a range of annual per capita tax
collections of $91.23 for alcohol and $350.59 for tobacco. There were approximately
58,000 adult marijuana users during that year. Assuming that they will pay a per capita
amount of taxes similaz to cigarette and alcohol users, then the range of marijuana tax
collections would be between $5.3 million and $20.3 million.
The Implications of Decriminalization and Le;~alization on Use
y Decriminalization:
Currently in Hawaii, the chance of a marijuana user being arrested and convicted
is approximately 0.4 percent. The appazent lower priority given by law enforcement and
relatively mild penalties for marijuana offenses reflect policy decisions that make the
probability of arrest and punishment of marijuana users insignificant and decreases the
r
potential effects of decriminalization. Gary Becker suggests two efficient means of
allocating enforcement resources. One would be to have lots of police, so law-breakers
r, face a high risk of arrest but the punishments are somewhat mild. A second regime would
be to have fewer police, reducing the risk of arrest but have severe or draconian,
„r punishments for those who are caught. Either can function as a deterrent. Hawaii, it can
be said, meets neither. There is a low risk of arrest and a mild punishment. This means
marijuana users in Hawaii perceive the probability of arrest and punishment as
insignificant.
ar
r 15 Use from 2002-2003 National Surveys on Drug Use. Tax data from Hawaii State
Department of Taxation http://www.hawaii.gov/tax/
~r
12
a
ar
A
Reseazch on the effects of mazijuana decriminalization has tended to find either
W
no relationship or a weak positive relationship between marijuana decriminalization and
drug use. Johnston, O'Malley, and Bachman (1981) use data from Monitoring the Future,
an annual survey of U. S. high school seniors, to see whether there were difference over
time in marijuana use between states that decriminalized and those that did not. They
found little difference. Thies and Register (1993) use data from the National Longitudinal
Survey of Youth between 1984 and 1988 to analyze differences in use of alcohol,
marijuana and cocaine among states who decriminalized and those that did not. They also
find little evidence of any effect. In addition DiNardo and Lemieux (1992) find no effect
of decriminalization on use.
Some studies have found a correlation between decriminalization and use. Model
(1993) finds a statistically significant and positive result from decriminalization, but
because he used hospital emergency room drug mentions these results are difficult to
interpret. It might simply reflect attitudes on the part of the population in decriminalized
states towazd drug use, which could correlate with the establishment of decriminalization.
Other studies that relied on cross-state variation in decriminalization status using recent
data and showing a positive and statistically significant effect were Saffer and Chaloupka
(1999); Chaloupka et al (1999); and Chaloupka, Grossman and Taurus (1999). Again
these are difficult to interpret because they used a measure that assumed decriminalized
state laws were identical and fundamentally different from criminalized states. The
measure used could also reflect attitudes on the part of the population towards drug use.
Pacula, Chriqui and King (2003) found that attitudes on the part of the population
r. towards drug use could play a role. They included actual legal dimensions such as
penalties and found that these did not diminish the association between decriminalization
and recent use. They conclude that their results tend to indicate that attitudes toward drug
~ use simply tended to be more tolerant in decriminalized states leading to the positive
association found in some studies between decriminalization and marijuana use. In other
words, decriminalization did not cause increased drug use, but that it existed
"r independently of the decriminalization statutes. They also demonstrate the extent to
which non-decriminalized states have reduced penalties associated with possession of
~ small amounts of marijuana as early as 1989, and call into question the interpretation of
r
13
y
studies evaluating this policy based on a simple cross sectional use of a variable denoting
* decriminalization.
Miron (2003) concludes, "The result that decriminalization has little impact on
.w
_ ~ marijuana use might seem surprising since standard economic principles suggest that
lowering the penalties for use should increase demand and therefore quantity consumed.
The explanation for this counterintuitive result of little impact on use is that
decriminalization frequently ratifies what has already taken place in the form of reduced
enforcement of marijuana laws."16 Given the limited prosecution and penalties associated
with marijuana possession in Hawaii it is doubtful that decriminalization would have
much effect on marijuana use.
Legalization:
Regulation and legalization means governmental agencies enforce tax, and other
laws, in a way that regulates the use of a product. The economics of substance use and
abuse assumes that the substances in question shaze two properties. First, they are
addictive in the sense that an increase in past consumption of the good leads to an
increase in the current consumption. Second, their consumption harms the consumer and
others. Because of these qualities there are both individual and social costs associated
'r with their use.
Many social costs azise from outlawing the use of some goods, which creates
r
black mazkets. Resulting.social costs range from violence that affects innocent citizens as
illegal gangs compete for markets, to a decline in respect for the law. ~ ~ Often overlooked
is that black market dealers can also lower costs of production in a number of ways. For
r example they do not obey labor laws, thereby reducing their costs of production.18
Standazd economic theory suggests that taxation can be used to offset some of these
s
ib Miron, "The effect of Marijuana Decriminalization the Budgets of Massachusetts
Governments..." p. 8
See Becker,G., Grossman, M., Murphy K., 2001. "The Simple Economics of the Waz
on Drugs, Mimeo, University of Chicago Department of Economics.
is Levitt found almost all members of a drug gang in Chicago were paid less than the
minimum wage forcing most gang members to live with their mothers.
a
14
r
social costs. As a result harmful activities can be reduced through both price effects and
legal sanctions.19
In terms of economics, one could simply assume that addictive goods aze "like"
.,,w
other goods and analyze the effects of prices and incomes on consumption. Or one could
view addiction as "myopic" behavior: past consumption increases future consumption
w.
_ and there is no thought of future consequences. Becker and Murphy (1988) develop a
model of addiction that suggests that addicts in some way also incorporate future
consequences. Grossman suggests that despite its somewhat controversial nature,
"Becker and Murphy's main contribution is to suggest that it is a mistake to assume
addictive goods are not sensitive to price."20
On the other hand there is a growing concern about the efficacy of the waz on
drugs and the use of incarceration as a deterrent. Nationally, in 1980 there were 24,000
drug offenders in state prisons; in 2004 there were 400,000. There was virtually no
increase in other types of offenders. While basic statistics on total mazijuana production
and consumption do not exist in any reasonable form, accurate estimates exist for drugs
like heroin. As a result, studies related to these other drugs can inform our understanding
of the consequences of legalization of marijuana.
For example, out of 700 metric tons of cocaine produced, world wide efforts,
largely by United States authorities, interdict 300 tons. In addition there is the massive
incazceration cited above. Despite these efforts the price of cocaine during this period fell
by more than two-thirds and the consumption of cocaine grew by ten times. Of course the
effects could have been worse without these efforts.
This is the question Kuziemko and Levitt (2004) take up. They analyzed the effect
of incazcerating cocaine drug offenders on cocaine prices. Their results are the highest
found for incarceration. They find that cocaine prices were 5% to 15% higher as a result
of increases in drug punishment since 1985. There is a broad range of estimates related to
the effect of the price of cocaine on use, so this price rise would have resulted in
anywhere from a 5% to 20% drop in cocaine use. They also found that locking up drug
i offenders leads to a crowding out effect in that time served for other offenses dropped by
' 19 Provided that taxes aze not set so high that they encourage a black market to develop.
20 Grossman, "Individual Behavior," p. I S
ar
15
w
r
7.6 months on average. Incarceration had about the same effects on violent and property
crime as locking up other offenders, about 3%. Zi Their results aze the most positive
found for the effects of incarceration.
Their research, while demonstrating that massive incarceration of drug offenders
can deter drug use, also demonstrates how costly such a policy is. They find that such
levels of imprisonment are excessive. In order to justify the level of imprisonment, the
individual and social costs of cocaine consumption would have to be $270 per gram.
Estimates of the economic costs of alcohol and drug abuse in 1992 were approximately
r $30 billion. That is $12.1 billion in health care costs and $17.5 billion in lost
productivity, with half of the lost productivity coming from jail time. For cocaine this
added up to about $50 per gram. Thus the costs were more than five times greater than
w.
the benefits. Therefore even though jail time does affect use, it is an extremely expensive
n„ means of doing it. Further, suppliers have responded in a number of ways that reduce the
cost of production such that the deterrent effect tends to be overwhelmed by these other
effects.
_ Becker, Murphy and Grossman (2004) in a study of cigazette addiction find that in
the short run a 10% increase in cigazette prices resulted in a 4% drop in consumption. In
the long run, however, this increased to 7.5%. Grossman (2004) finds that the 70% rise in
the real price of cigarettes, accounts for almost the entire 12 % drop in cigarette smoking
between 1997 and 2003. Becker et al's results suggest that these effects will be magnified
over time. Grossman also finds the 7% rise in the real price of beer between 1990 and
_ 1992, as a result of federal excise tax hikes, "accounted for 90 percent of the 4 percentage
point decline in binge drinking: '22 Even with illegal substances, such as marijuana, price
swings account for 60 to 70% of the changes in consumption since 1975.23
Legalization would have the effect of probably increasing supply and thereby
reducing the price. This in turn would probably increase consumption. This, however,
could be offset by setting an appropriate tax level. A related question is the social and
r
21 It should be noted that Hawaii and Missouri were excluded from their data due to poor
reporting quality
• zz op cit, abstract
z3 Ibid, marijuana prices rose and fell dramatically over that period
16
individual costs related to marijuana use. Even so the evidence is fairly thin on what the
actual social costs of marijuana are.
Conclusion
The cost savings from decriminalizing marijuana aze approximately $4 million
although this leaves out some additional costs related to legal defense, and some state
programs like drug courts. These would bring the total to $5 million. The reason that
these are so low is that Hawaii appeazs to have a de facto policy of lax enforcement of
this law. Legalization would save an additional $4 to $5 million. Taxes from legalization
would bring in anywhere from $4 million to $23 million depending on tax rates. A large
body of literature suggests that decriminalization would not lead to additional use.
Legalization would increase the supply of mazijuana and thereby reduce the price and
increase use. Should the social costs, and thus public policy wazrant it, this could be
controlled through appropriate tax rates.
r
r
r
•
s
w
17
+Y
References
Becker, G. S., M. Grossman, et al. (1994). "An Empirical Analysis of Cigarette
Addiction." American Economic Review 88(3): 396-418.
w.
Desimone, J. and M. C. Farrely (2003). "Price and Enforcement Effects on Cocaine and
Marijuana Demand." Economic Inguirv41(1): 98-115.
Grossman, M. (2004). Individual Behaviors and Substance Abuse: The Role of Price.
NBER Working Paper Series, NBER: 1-58.
Hawaii State Judiciary (Various Years). "Annual Report Statistical Supplement."
Hawaii State Department of Taxation, (Various Years) "Calendar Year Tax Collection
Report,"
Kuziemko, I. and S. D. Levitt (2004). "An Empirical Analysis of Imprisoning Drug
Offenders." Journal of Public Economics 88: 2043-2066.
Levitt, S. D. and S. A. Venkatesh (2000). "An Economic Analysis of a Drug -Selling
Gangs Finances." Quarterly Journal of Economics: 755-789.
Miron, J. A. (2003). The Effect of Marijuana Decriminalization on the Budgets of
Massachusetts Governments, With a Discussion of Decriminalization's Effect on
Marijuana Use, Boston University: 1-11.
,r, Miron, J. A. (2005). The Budgetary Implications of Marijuana Prohibition. R. t. t. M. P.
Project. Boston, Boston University: 18.
Pacula, R. L., J. F. Chriqui, et al. (2003). Marijuana Decriminalization: What Does It
Mean in the United States? NBER Working Paper Series, National Bureau of
Economic Research: 1-34.
Policy, O. N. D. C. (2004). State of Hawaii; Profile of Drug Indicators. N. C. J. R.
Service: 1-13.
s President, T. E. O. o. t. (2004). The Price and Purity of Illicit Drugs: 1981 through the
Second Quarter of 2003. O. N. D. C. Policy. Washington, D. C, United States
Government. November 2004.
Saffer, H. and F. Chaloupka (1999). "The Demand for Illicit Drugs." Economic Inguirv
Volume 37(No.3): 401-411.
r Saffer, H. and F. Chaloupka (1999). State Drug Control Spending and Illicit Drug
Participation. NBER Working Paper Series, National Bureau of Economic
Research: 1-21.
r
Wood, D. W. (Various Years). Illicit Drug Use in Honolulu and the State of Hawaii. N. I.
H. Community Epidemiology Working Group, National Institute of Health.
r
r
18