NJPP Final Report

The Municipal Cost of
Low-Level Enforcement
in New Jersey
May 2017
A report prepared for New Jersey Policy Perspective by graduate

students of the Edward J. Bloustein School of Planning and
Public Policy at Rutgers, The State University of New Jersey
Jane Allen Yan Huang Arcadia Lee

Holly Low Kate Millsaps Thalya Reyes
Contents
Acknowledgements....................................................................................................................................... 2
Executive Summary....................................................................................................................................... 3
Project Overview........................................................................................................................................... 5
Low-Level Offenses and Enforcement Costs................................................................................................. 5
Trends in Low-Level Offense Enforcement ................................................................................................... 7
Low-Level Marijuana-Related Offense Enforcement.................................................................................... 8
Impacts of Low-Level Enforcement on Municipal Spending ........................................................................ 8
Methods .................................................................................................................................................... 9
Direct Relationships .................................................................................................................................. 9
Low-Level Enforcement and Demographic Impacts on Spending .......................................................... 10
Cost Analysis and Case Study of Large Municipalities ............................................................................ 11
Concerns with Low-Level Offenses in FBI Arrest Data ................................................................................ 13
Conclusions ................................................................................................................................................. 14
Recommendations & Discussion for Further Research .............................................................................. 15
Bibliography ................................................................................................................................................ 17
Appendix A: Statistical Methodology and Findings .................................................................................... 19
Dataset and Variable Creation ................................................................................................................ 19
Demographic Indicators ...................................................................................................................... 19
Low-Level Offense Enforcement Indicators ........................................................................................ 19
Spending Indicators............................................................................................................................. 20
Bivariate Analysis .................................................................................................................................... 20
Limitations with FBI UCR Data ................................................................................................................ 21
Regression Analysis ................................................................................................................................. 23
Code book ............................................................................................................................................... 28
Municipalities Included in Random Sample ............................................................................................ 31
Appendix B: Municipal Cost Analysis .......................................................................................................... 33
Works Cited ............................................................................................................................................. 34
Appendix C: Differences between the Statistical and Municipal Cost Analysis Methodologies ................ 35
Appendix D: Individual Cost Profile for Low-Level Offense Arrest ............................................................. 36
Cost to Individual .................................................................................................................................... 36
Collateral Consequences to an Individual ............................................................................................... 37
Works Cited ............................................................................................................................................. 39
1
Acknowledgements
We thank Professor Cliff Zukin for his guidance and expertise as well as Brandon McKoy and Jon Whiten
of New Jersey Policy Perspective for their direction and insight, and for connecting us to
resources. Many experts in the criminal justice and municipal budgeting fields were generous with their
time and shared helpful resources with the team. Thank you to Alison Accettola, Portia Allen-Kyle, Diana
Houenou, Ami Kachalia, Mark Magyar, Maria Pedraza, Alex Staropoli, Alexi Velez, and Elizabeth Weill-
Greenberg.
2
Executive Summary
This study analyzes the municipal costs associated with low-level offense enforcement by examining (1)
the relationship between low-level offense enforcement and municipal spending and revenues, and (2)
if poverty and race impact spending on low-level offense enforcement. A random sample of New Jersey
municipalities with a population greater than 3,000 was selected for the study. The project utilized
spending and revenue figures for police departments and municipal courts from 2015 municipal
budgets. Data from the New Jersey Judiciarys Municipal Court Statistics on disorderly persons and
petty disorderly persons court filings, which include marijuana possession under 50 grams, possession
of drug paraphernalia, disorderly conduct, simple assault, petty theft, shoplifting, resisting arrest, and
trespassing were used as a measure of low-level offenses.
Findings indicate that:
Low-level enforcement has little impact on municipal spending. For each additional low-level
offense enforced, it costs a resident $0.04 in police and court spending. The research team
believes this low amount is due to the high fixed costs of operating a police department.
Police spending makes up the vast majority (approximately 95 percent) of total justice spending.
Each time an additional low-level offense is enforced per capita, courts generate less than a
penny in revenue on behalf of each taxpayer.
An analysis that includes fixed costs found the median amount municipalities in our sample
spent on low-level enforcement was $17 per capita. Due to data limitations, this does not
include health care and pension costs for police and court staff, or prosecutor and public
defender costs for the courts.
On average, the municipalities spent an estimated 1.3% of their budget on low-level

enforcement, with a range of 0.2% to 3.5%.
A case study of large New Jersey cities and municipalities that have lower median income and a
higher proportion of black and Hispanic residents found they had higher rates of overall low-
level offense enforcement and spending, but lower arrest rates for marijuana possession, than
towns that are more white and affluent.
Though the cost of low-level enforcement is practically low, several important conclusions can be made.
Police spending dominates total spending on enforcement, greatly overshadowing court spending. This
suggests that reform efforts to reduce justice spending should be focused on police departments
specifically. But, due to the minor cost of each additional low-level offense, decreasing enforcement
alone is not enough to make a large budgetary impact. Efforts to reduce spending should instead focus
on high fixed costs. It is important to note that despite a low-budgetary impact, there are additional
costs to communities due to this policing approach that are beyond the scope of this report.
The research team also found the demographic composition of a municipality (race, gender, and age)
does not play a significant role in describing the differences in low-level enforcement rates within the
sample. Court filing data does not include race so it was not possible to compare the rates of arrests,
summons, and citations for blacks, Hispanics, and whites.
3
Additionally, a case study of seven large municipalities was completed using a different method that
includes fixed costs. It showed that these cities, which have more black and Hispanic residents and a
lower median income than the sample, spend more on low-level enforcement. These cities also had
lower arrest rates for marijuana possession than the sample average.
The vast majority of municipal court filings consist of traffic and parking violations (86%). Future
research should more closely examine the relationship between traffic and parking violations and justice
expenditures in New Jersey, which may be more sensitive to being used for revenue generation than
low-level offense enforcement.
The FBI Uniform Crime Report (UCR) data are of questionable validity for a number of reasons. First,
police department reporting is voluntary; there have been reports of discrepancies between police
records and their corresponding FBI report (ACLU 2015). Second, there is an issue of categorization. Half
of all arrests in New Jersey fall into a category called all other offenses. As a result, it is impossible to
ascertain exact accounts for the low-level offenses contempt of court, drug paraphernalia, and trust
passing. Third, the UCR does not require tracking and reporting the number of Hispanic individuals
arrested, making it difficult to conduct research on police activity in Hispanic communities. For increased
transparency in police department activity, a better arrest tracking system is needed than the current
UCR.
4
Project Overview
Numerous stakeholders have questioned the enforcement of low-level offenses, arguing that it is an
expensive practice that has little impact on community safety (ACLU 2015; Henrichson, Rinaldi, and
Delaney 2015; Miron and Waldock 2010; Simms 2014; Subramanian et al 2015). Previous research has
also documented racial and economic disparities in such enforcement (ACLU 2013 and 2015). This study
investigates (1) what is the relationship between low-level offense enforcement and municipal spending
and revenues; and (2) do poverty and race impact spending on low-level offense enforcement.
Those questions are investigated using two approaches. In the first, statistical analysis is used to
examine the influence of low-level arrests and court filings on police and court spending. The second
approach estimates what municipalities spend on low-level enforcement, and makes comparisons to a
case study of seven large, urban municipalities. As an exploratory study of a sample of New Jersey
municipalities, this report examines the relationships between low-level offenses, race, poverty, police
and court spending, and court revenue, discusses implications and recommends further research.
Low-Level Offenses and Enforcement Costs

New Jersey statute categorizes low-level, non-criminal offenses as disorderly persons (DP) and petty
disorderly persons (PDP) offenses (defined in N.J.S.A. 2C:1-4). Often called misdemeanors, a DP
offense can carry a fine up to $1,000 and six months in jail, while a PDP carries a fine up to $500 and
thirty days in jail. DP and PDP offenses include marijuana possession under 50 grams, disorderly
conduct, simple assault, petty theft, shoplifting, resisting arrest, and trespassing. These cases are heard
in municipal court and, between 2010 and 2014, there were 430,429 DP and PDP offenses filed on
average statewide in the Municipal Court system (New Jersey Judiciary 2017a). As shown in Figure 1,
low-level filings (DP and PDP offenses) make up the majority (52%) of criminal court filings in the states
municipal courts compared to higher level indictable crimes that are later sent Superior Court and other
criminal offenses (New Jersey
Figure 1. Criminal Filings in New Jersey Municipal Courts in CY2013 Judiciary 2017a).
The majority of Municipal Court

activity, however, is parking and
traffic violations. As shown in
Figure 2, in Court Year 2013, of
just over 6 million filings in
municipal courts statewide, 5.2
million were parking and non-
DWI traffic Judiciary 2017). This
study does not examine spending
and other resources dedicated to
traffic and parking violation
enforcement, which accounts for
most court filings, and then
presumably most police activity.
5
Figure 2. Municipal Court Filings by Offense Type in CY2013 low-level enforcement requires
municipal spending on police,
courts, and incarceration. Based on
our analysis of municipal budgets,
the majority of municipal spending
on all crime enforcement is
concentrated in police spending as
shown in Figure 3. Therefore,
reductions in police spending would
have a much larger budgetary
impact.
Municipal Courts handle the vast

majority of court matters in the
state. The bulk of court spending
supports judge and staff salaries,
interpreters, equipment, security
services, police overtime,
administrative services, facility
Figure 3. Municipal Justice Spending maintenance, and prosecutor and
public defender payments (New
Jersey Judiciary 2010). One-quarter
of New Jersey municipalities
participate in a shared or joint court
arrangement in an effort to reduce
costs (New Jersey Judiciary 2010). In
a shared court, individual courts
share space, staff and supplies, but
municipalities can still appoint their
own judge or administrators. Under
a joint court, all services are
consolidated and judges are
appointed by the Governor. As of
2010, there were 83 shared
municipal courts in the state, the
majority located in Monmouth, Burlington, Hunterdon and Warren Counties, and 21 joint municipal
courts that serve 59 municipalities (New Jersey Judiciary 2010).
Individuals awaiting trial, sentencing, or conviction in the municipal court are housed in county
correctional facilities. A 2015 Vera Institute of Justice report found that nationwide 75% of the inmate
population was incarcerated for nonviolent traffic, property, drug or public order offenses and local
jurisdictions spent $22.2 billion on corrections (Subramanian, R. et al. 2015, 5). The Bureau of Justice
Statistics Jail Inmates in 2015 survey of local inmate populations, found 27% of inmates were held for a
misdemeanor offense (Minton and Zeng 2016).
6
In New Jersey, a 2013 study found about 7% of the county jail population were individuals sentenced in
a municipal court and about 7% were individuals awaiting trial or sentencing in a municipal court, about
1,780 individuals (Van Nostrand 2013). The most common municipal court charges were contempt of
court, drunk driving, and shoplifting or theft. Under a nationwide median estimate of an $101 average
daily cost to house an inmate at a county facility, New Jersey taxpayers pay $9,006 to incarcerate each
individual who is held the average 89 days in a county facility for Municipal Court pretrial (Henrichson,
Rinaldi, and Delaney 2015; Van Nostrand 2013).
Trends in Low-Level Offense Enforcement

Much of the research on low-level enforcement has examined its roots in broken windows policing
and the potential for profit motive in cash-strapped municipalities. Broken windows policing, based on
the premise that the enforcement of minor offenses will prevent more serious crime, was widely
adopted by police forces in the 1990s (Kelling and Wilson 1982, American Civil Liberties Union 2015;
hereafter ACLU, Subramanian et al. 2015). Opponents argue this has resulted in police practices, such as
stop and frisk and zero tolerance, that increase low-level enforcement (ACLU 2013). Studies of
these tactics have found they do not reduce crime overall and target low-income communities of color.
The National Research Council found there is not strong evidence that aggressively responding to
minor offenses, particularly with arrest, effectively reduces or prevents more serious crime
(Subramanian et al. 2015). The ACLU (2013) found that there is no correlation between these marijuana
possession arrests and crime reduction.
In New Jersey, an ACLU report found black residents were between 2.6 and 9.6 times more likely than
Whites to be arrested for low-level offenses in the four high population, urban municipalities included in
their study (2015, 4). The report also found higher arrest levels for Hispanics compared to Whites. The
authors attributed these racial disparities to officers ability to exercise discretion in low-level
enforcement (ACLU 2015, 9). Its important to note that the analysis presented in this report differs
from the ACLU study in its methodology. The ACLU used open records requests to acquire arrest and
race data from police departments, compared these to FBI arrest records, and analyzed both in relation
to Census data on the racial composition of the communities under study.
The ACLU has also criticized police performance measures that are tied to funding as an incentive to
increase low-level enforcement (2015). The federal government provides grants, such Byrne Justice
Assistance Grants (JAG) for police staff and drug enforcement programs, based on performance
measures which include total number of arrests (ACLU 2013, Simms 2014). In 2014, 46 municipalities in
New Jersey qualified for a total of $2,114,397 in federal JAG grants (Bureau of Justice Assistance
2014). The largest recipients were Newark, Jersey City, Camden, Paterson, Trenton, and Elizabeth. The
state provides Safe and Secure Communities grant funding for additional officers or support staff and
currently supports 345 officer positions and 13 staff members in 160 municipalities (NJ Office of the
Attorney General 2017).
Beyond grants, municipal court revenue has been examined as a potential driver of low-level
enforcement. A US Department of Justice study of the Ferguson Missouri Police Department and
Municipal Court found that minor offense enforcement was being used to increase municipal revenue
(2015; hereafter USDOJ) . In 2016, The Asbury Park Press in New Jersey examined 86 towns in
7
Monmouth and Ocean counties. They found seven municipalities rely on court revenue to fund over 5%
of their budgets, and that the revenues had increased 39% on average between 2010 and 2015
(Kachmar 2016). According to their interviews with lawyers and former judges, there is significant
pressure to generate revenue, and there was evidence that one municipality replaced a judge in an
effort to boost revenue (Kachmar 2016).
Some argue that the spending required to enforce low-level offenses could be used in more productive
ways. The ACLU (2015, 12) found that when individuals were charged with a low-level offense in the
four New Jersey cities they studied, over 70% of the charges were not associated with a more serious
arrest as defined by the FBI. By reducing focus on low-level offense enforcement, police departments
can target resources towards solving serious crimes. Nationally, approximately 50 percent of violent
offenses and 80 percent of property crimes were not solved in 2010 alone (ACLU 2013, 115).
Low-Level Marijuana-Related Offense Enforcement

In terms of the cost of low-level enforcement, much of the research has focused on drug prohibition. In
New Jersey, marijuana possession under 50 grams, drug paraphernalia possession, failure to turn over
marijuana to a police officer, and being under the influence of marijuana are all low-level offenses
(NORML 2017). Possession over 50 grams, distribution, cultivation, and sale of paraphernalia are all
indictable offenses, handled in Superior Court. In New Jersey, about 365,900 people age 21 and over are
estimated to use marijuana regularly, using 2.53 million ounces a year (McKoy and Rosmarin 2016). In
2014, there were 24,689 marijuana and hashish possession arrests statewide; note the number of
arrests for possession under 50 grams alone is not available (NJ State Police).
A 2010 study on cost savings from federal and local government drug legalization found legalizing
marijuana would save $183.039 million in New Jersey (Miron and Waldock 2010, 32-33). In New Jersey,
the researchers estimated marijuana spending accounted for 22% of all state-level expenditures on drug
prohibition (Miron and Waldock 2010, 6). An ACLU (2013) study estimated New Jersey spent $127
million on enforcing marijuana possession prohibition. Both cost figures include police spending on
arrests, judiciary spending on prosecution, and incarceration. At this spending level, New Jersey has the
eighth highest per capita fiscal expenditures for enforcing marijuana possession laws in the country
(ACLU 2013, 23). New Jersey was also eighth nationwide for most marijuana possession arrests with
21,659 arrests, 35.1% of which were black individuals (ACLU 2013, 15).
A report by the Drug Policy Alliance report found that in New Jersey roughly 17% of county inmates
were incarcerated with drug charges as the primary charge, however, the researchers did not
specifically report the number of inmates incarcerated on marijuana charges (Van Nostrand 2013). The
trend in New Jersey's Superior Courts for higher level drug offenses has been toward less prison
sentences; the number of incarcerated drug offenders statewide has dropped from 10,385 in 1999 to
5,224 in 2012 (Chesler 2013).
Impacts of Low-Level Enforcement on Municipal Spending

The research team explored the relationship between low-level enforcement and municipal spending in
three ways: (1) the direct relationship between low-level offenses filed in municipal courts on police
spending, court spending, and court revenue; (2) spending and revenues when demographic factors are
8
also considered; and (3) a per capita estimate of spending on low-level enforcement in a random sample
of New Jersey municipalities, compared with a case study of seven cities with high populations.
Methods
To examine whether reductions in low-level enforcement could result in cost savings for municipalities
in the form of lower police and court spending, the research group created an original dataset. First, a
random sample of New Jersey municipalities was pulled. Then data for demographic, arrest, and budget
indicators at the municipal level was compiled. The New Jersey Judiciarys Municipal Court statistics
were used for court filing data and spending and revenue figures were taken from each municipalitys
2015 budget. Demographic data were taken from the American Community Survey. Arrest data was
also collected from the FBI UCR dataset, but ultimately not used based on concerns outlined in the
findings section and Appendix A.
Using this dataset, the study relied on two analysis techniques. The Figure 4. As court filings increase, police
first analysis used was the statistical modeling technique linear spending does too.
regression, which examines how fluctuations in low-level
enforcement influence municipal budgets, specifically police
spending, court spending, and court revenue. The second
analysis is a cost calculation of low and high estimates of
spending on low-level enforcement using a methodology
developed by Harvard economist Jeffrey Miron (2010). A full
description of the methods used in the dataset creation and
statistical testing can be found in Appendix A. Methodology
used in the municipal cost analysis can be found in Appendix B.
Direct Relationships
The research team relied on scatterplots to visually represent
the relationship between low-level court filings and the budget
indicators of police and court spending and court revenue. Each Figure 5. As court filings increase, court
dot on the scatterplots in Figures 4, 5 and 6 represents a spending does too.
municipality. For example, the red dot 5 in Figure 5 is Salem City.
The line running through the graph shows the direction of the
relationship. When the dots cluster closely around the line, it
signals that the indicators are closely related, meaning changes
in one are a good predictor of what will happen in the other.
The horizontal lines, or x-axes, show the range of low-level

citations, summons, and arrests per 100 residents in New
Jersey municipalities. The vertical lines, or y-axes, represent
police spending and court revenue per 100 residents.
Looking at the red dots position between the two axes, it

makes sense that in 2014 for every 100 residents in Salem, the
city spent $35,689 on police and the courts processed 82 low-
level offenses.
9
Low-level Court Filings have a positive relationship to both police
Figure 6. Court filings and court revenue
spending and court revenue, as shown by the upward slope of the line
have the strongest relationship.
in Figures 4, 5 and 6. This means that as court filings increase, so
does police spending, court spending, and court revenue.
But the spending and court revenue scatterplots differ in one

important way. It is clear that the dots are located much more tightly
around the line in Figure 6, indicating they are closely related and
that changes in one are a good predictor of what will happen in the
other. The research team can say with confidence there is a very
strong relationship between court revenue and low-level filings.
Large upticks in low-level summons, citations, and arrests strongly
predict courts will bring in higher amounts of revenue. The same
fluctuation in low-level crime, on the other hand, may nudge police
spending upward, but its more likely that other factors have a larger
impact.
Low-Level Enforcement and Demographic Impacts on Spending

Across the state of New Jersey, regions, towns, and cities vary greatly in their socio-demographic
profiles. Positing that black neighborhoods experience more intense broken windows policing and
poorer communities have a smaller absolute tax base, the research team wanted to explore what
impacts these demographic factors would have on the budgetary indicators. Therefore, the team
conducted three statistical analyses to examine how low-level enforcement, the proportion of black
residents, and poverty rate impact the three spending and revenue indicators. The research team found
that a quarter of spending differences could be explained by differences in low-level enforcement rates,
race, and poverty rate (see Appendix A for full description of methodology and calculations used to
determine the figures presented below).
The results of those analyses are:
Each time an additional low-level offense is enforced, it costs a resident $0.03 in police
spending.
Each time an additional low-level offense is enforced, it costs a resident less than one cent in
court spending.
Each time an additional low-level offense is enforced, courts earn a penny in revenue on behalf
of each resident.
The analysis also found regardless of poverty rate and low-level enforcement rate, a one percent
increase in the proportion of black residents is associated with:
A $4.84 reduction in police spending

A $0.44 drop in court spending
A $0.27 decrease in court revenue
In addition, when taking race and enforcement rates into account, a one percent increase in the per
capita poverty rates leads to an increase in:
police spending of $3.22
10
court spending by $0.03
court earnings of $0.14
Our analysis found that gender and age do not have a significant influence on municipal justice spending
and revenue. The proportion of Hispanic residents has a strong relationship with total low-level court
filings and court spending. Ultimately the research team opted to leave it out of the advanced statistical
models because the census reports Hispanics of all races as one category and New Jersey has one of the
highest percentages of Hispanics who identify as black. The research team also tested whether the
demographic factors influence the number of court filings in a municipality (without including spending
in the models). None of the demographic indicators had a significant impact. Further research is needed
to determine if other factors such as education, lifestyle, policing behavior, or environmental factors are
driving low-level offense enforcement.
Cost Analysis and Case Study of Large Municipalities

Municipal budgets and low-level offense court filings
Figure 7. Per Capita Municipal Spending on
were also analyzed to estimate how much
Low Level Enforcement
municipalities spent on low-level enforcement in
2014. This method uses the number of low-level
offenses filed in municipal court as a proxy for the
allocation of police and court resources since these
represent all the arrests, summons and citations for
these offenses (see Appendix B for full description of
methodology used). Its important to note that these
estimates include the fixed costs involved with police
departments and courts, which is why it differs so
greatly from the statistical analysis. However, this is a
useful way to compare municipalities to one another.
According to this analysis, the median amount spent on

low-level enforcement is $17 per capita in the random
sample of 54 municipalities. On average, the
municipalities spent an estimated 1.34% of their
Figure 8. Top 10 Spenders on Low-Level
budget on low-level enforcement, with a range of
Offense Enforcement as Percent of Budget in
0.23% to 3.52%.
Our Sample
The top 10 municipalities with the highest percent of
their budget going towards low-level offenses were:
Manville Borough (3.5%), Howell Township (2.9%),
Ocean Township (2.8%), Washington Township (2.6%),
Paramus Borough (2.5%), Plumsted Township (2.5%),
Vernon Township (2.5%), Point Pleasant Borough
(2.5%), Lacey Township (2.2%), and Berkeley Township
(2.1%).
The top 10 spenders per capita were: Paramus

Borough ($55), Rochelle Park Township ($44), Manville
11
Figure 9. Top 10 Per Capita Spenders on Borough ($42), Norwood Borough ($35), Ocean Township
Low-Level Offense Enforcement in the ($34), West Deptford Township ($29), Belmar Borough ($27),
Sample Point Pleasant Borough ($26), Howell Township ($26), and
Vernon Township ($25).
Seven large cities and municipalities were also analyzed

separately as a case study, and compared to the random
sample: Linden, Newark, Paterson, Toms River, Willingboro,
Lakewood, and Vineland. While the random sample had an
average median income of about $88,400 and was on
average 74% White, this group had an average median
income of $52,000 and was 55% White. However, when
Toms River and Lakewood are removed, which are both
about 90% White, the average comes down to 41%
White. The case study municipalities showed a higher average rate of low-level enforcement, using the
ratio of low-level court filings to total court filings as an indicator. In the random sample, the
municipalities had an average of 8.5% of their court filings as low-level offenses, while 13% of the court
filings in the case study cities were low-level. Interestingly, the marijuana arrest rates tell a different
story. The average marijuana arrest rate per 100,000 persons was found to be 268 for the random
sample, and 212 for the case study municipalities.
As noted further in the Recommendations section and in Appendix A, the arrest data from the FBI UCR
has significant limitations. Counting all low-level offenses is problematic, but that is not a problem with
marijuana possession arrests. The validity is questionable since reporting is voluntary, and research has
documented disparities between police department records and their UCR submissions (ACLU, 2015).
However, over time the arrest data has reliability, and can be used to look at difference between
groups, such as in this case study comparison.
Overall the case study municipalities had higher spending in terms of totals, per capita, and as a percent
of budget. Linden, Newark, Paterson, Toms River, Willingboro, Lakewood, and Vineland spent an
estimated total of $17,749,332 on low-level enforcement in 2014, with Newark spending the bulk, at
$6,209,584.
These municipalities spent on average 2.12% of their municipal budgets on low-level enforcement:
Vineland (4.7%), Willingboro (2.8%), Toms River (2.5%), Lakewood (2.9%), Newark (0.8%), Paterson
12
(0.9%), Linden (0.3%). They spent on average $26 per capita: Vineland ($46), Willingboro ($35), Toms
River ($32), Lakewood ($24), Newark ($22), Paterson ($14), and Linden ($7). Linden stood out as a low-
end outlier, with an estimated $7 per capita and 0.3% of its municipal budget going towards low-level
enforcement. Compared to the other municipalities, Linden had a very low percentage of low-level
offenses filed in court in relation to their traffic and parking offenses. In Vineland, which had the highest
per capita spending on enforcement, and portion of municipal spending, 25% of court filings were low-
level offenses.
Though the seven cities analyzed here represent a small case study, these findings suggest that
communities that are more racially diverse and have lower median income experience higher rates of
overall low-level offense enforcement. They also appear to spend more on enforcement, both as a per
capita cost to taxpayers, and as a percent of their municipal budgets. However, these communities
appear to experience lower arrest rates for marijuana possession than more affluent, less diverse
communities, and this may warrant further examination. This finding may indicate that in larger cities,
police don't enforce marijuana possession as much because they are busier enforcing other more
serious offenses, or it could mean they spend more time enforcing the other low-level offenses
associated with broken windows practices, such as disorderly conduct, loitering, etc., or a combination
of both. In wealthier, whiter communities, police may spend more time enforcing marijuana possession
because there is not as much opportunity to do the broken windows type enforcement in less urban
areas.
Concerns with Low-Level Offenses in FBI Arrest Data

Although FBI arrest data were collected and analyzed for each municipality under study, ultimately they
were found to be of questionable usefulness for measuring the volume of overall low-level enforcement
in a community.1 First, police agency reporting is voluntary, so the measure lacks some validity.
Previous work by the ACLU (2015) found discrepancies between the arrest numbers in the FBI dataset
and as reported by municipalities under Open Public Records requests.
Second, the FBI arrest reporting has a large category called All Other Offenses which makes up half of
all arrests in New Jersey and combines some misdemeanors and indictable crimes (USDOJ, 2013). The
low-level offenses from this category, which include possession of drug paraphernalia, contempt of
court, and trespassing, could not be separated out and counted. The only offenses that could be
counted for the analysis were marijuana possession (under and over 50 grams), disorderly conduct,
vagrancy, curfew, liquor law violations, and gambling.
Finally, in comparing the FBI arrest data to the low-level offenses filed in municipal courts, there were
significant differences in the numbers, even when accounting for the fact that the court filings count the
total number of charges, while the FBI UCR counts the number of individuals arrested. As a result, the
statistical analysis showed that the court filing data had a strong relationship to police spending, court
spending and court revenue, while the FBI arrest data did not. Looking at Figures 11 and 12, it is clear
that the dots are clustered much more tightly around the line in Figure 11 than they are in Figure 12.
1
For a more in depth explanation of why the court filing indicator is a better predictor of low-level enforcement
than the FBI-UCR, please see Appendix A: Limitations with FBI UCR Data.
13
This signals that court filings predict court revenue much better than low-level arrests from the UCR
data. The steepness of the line in Figure 11 also indicates that the relationship between court filings and
court revenue is much stronger than low-level arrests. Due to these limitations and discrepancies
between the court filing and FBI UCR data, the FBI dataset was not used for statistical analysis in our
study (see Appendix A for further details).
However, the UCR arrest data is a reliable measure over time, so its possible to glean useful information
from looking at arrest trends. Its also possible to make useful comparisons between groups, as was
done in the case study analysis of large municipalities and cities, with marijuana possession arrests
specifically (see Appendix B for full methodology on marijuana possession arrest analysis).
Conclusions
The following conclusions can be made based on the results of our analysis:
1. Low-level enforcement has little impact on municipal budgets in terms of police and courts.
For each additional low-level offense enforced, it only costs a resident $0.04 in police and court
spending. The research team believes this is due to the high fixed costs of operating a police
department.
2. Police departments have high fixed costs, reducing the potential budget impact of a drop in
low-level offense enforcement. Our analysis found that low-level court filing rates explain 33%
of the difference in police spending between municipalities even though it has little practical
impact. The small dollar amount of low-level enforcement seems to indicate that the high fixed
costs of policing, including healthcare and pensions, drive overall spending, a finding which is
consistent with many budget analyses in New Jersey.
3. Police spending makes up the overwhelming majority (95%) of justice spending for
municipalities, while court spending is less than five percent. Therefore, changes to low-level
crime enforcement policies will not have any significant impact on court system spending, and if
municipalities want to lower justice costs, they should focus on police spending.
14
4. The demographic composition of a municipality (race, gender, and age) did not play a
significant role in describing the differences in low-level enforcement rates within the sample.
Court filing data does not include race so it was not possible to compare the rates of arrests,
summons, and citations for blacks, Hispanics, and whites.
5. Large, racially diverse municipalities and cities have more overall low-level offense
enforcement, but lower arrest rates for marijuana possession based on FBI data. According to
the case study findings, large New Jersey municipalities and cities that have higher percentages
of black and Hispanic residents and lower median income, also appear to spend more per capita
and as a percent of their budget, on low-level offense enforcement.
6. Traffic and parking violations are the vast majority of municipal court filings (86%), and
therefore may be more inviting as a source for revenue generation than low-level offense
enforcement. Conversely, statewide low-level offense enforcement made up only 7% of
municipal court filings. Municipalities in our sample had an average 8.97% low-level offense
court filing rate, but there were municipalities that exceeded the statewide and sample average
by a significant amount and could be investigated further. Low-level offenses accounted for
34.89% of court filings in Millville, 32.19% in Salem City, 20.33% in West Deptford, and 20.15% in
Vernon Township.
7. FBI Uniform Crime Reporting (UCR) data is not the best measure for analyzing overall low-
level offense enforcement in communities. Problems include the fact that reporting is
voluntary, half the arrests in New Jersey are grouped into one category called All Other
Offenses, and the UCR does not report the number of Hispanic individuals arrested. For
increased transparency in police department activity, a better arrest tracking system is needed
than the current UCR.
Recommendations & Discussion for Further Research

This dataset can be expanded by adding more municipalities (cases) to strengthen the findings
and provide more reliability to current findings. The findings presented here are based on a
small sample (less than 50). The sampling error for these municipalities is +/- 12%. A larger
dataset may highlight a stronger relationship between demographic factors, municipal spending,
and low-level enforcement.
The fact that race was not a significant driver of low-level court filings in the most sophisticated
statistical analysis does not mean these results are at odds with the ACLU findings for racial
disparities in low-level enforcements (2015). Since this analysis examined court filings, which
have no data on the race of the defendants, it was not possible to compare these to the racial
composition of the communities being analyzed.
Longitudinal data could be added to the dataset to use a methodology developed by the Vera
Institute of Justice to determine municipal spending on low-level enforcement to develop an
additional measure of per capita spending.
About 25% of municipalities in the state already participate in a joint or shared court in an effort
to reduce their spending. However, within in our dataset, court spending was a very small
portion (<5%) of a municipality's overall justice spending. To achieve real reductions in
15
expenditures, more focus would need to be made on police spending, which was over 95% of
justice spending.
In our sample, shore towns had a very different relationship between low-level enforcement
and police spending than other communities around the state, and this may merit further
examination.
Some of the municipal police departments in our sample received federal Byrne grants and state
Safe and Secure Communities grants. Previous studies have posited that communities that
receive these funds have an incentive to increase arrest rates to continue to qualify for grants as
total arrests are reported on the grant application. Our sample was not large enough to
determine if the municipalities receiving such funds are systematically different than other
communities in New Jersey, but this would be an interesting question to investigate with a
larger dataset.
Additional research could be done to determine how municipal revenue is generated in the
court system. This research group did not have the data available to determine what portion of
court fines resulting from enforcement action went to the municipalities and the amount that
went back to the state. The research team was also unable to determine what low-level
enforcement resulted in municipal ordinance violations; the municipalities keep all the fines
collected for these offenses and do not remit anything back to the state.
16
Bibliography
American Civil Liberties Union (ACLU) of New Jersey. (2015). Selective policing: racially disparate
enforcement of low-level offenses in New Jersey. Retrieved from https://www.aclu-
nj.org/files/7214/5070/6701/2015_12_21_aclunj_select_enf.pdf
American Civil Liberties Union (ACLU). (2013). The war on marijuana in black and white: billions of
dollars wasted on racially biased arrests. Retrieved from https://www.aclu.org/files/assets/aclu-
thewaronmarijuana-rel2.pdf
Bureau of Justice Assistance. (2014). Awards made for "BJA FY 14 Edward Byrne Memorial Justice
Assistance Grant (JAG) program: local". US Department of Justice. Retrieved from
https://external.ojp.usdoj.gov/selector/title?solicitationTitle=BJA%20FY%2014%20Edward%20B
yrne%20Memorial%20Justice%20Assistance%20Grant%20(JAG)%20Program:%20Local&po=BJA
Chesler, C. (2013 Oct 22). New Jersey's drug court program: making the sentence fit the crime. NJ
Spotlight. Retrieved from http://www.njspotlight.com/stories/13/10/22/new-jersey-s-drug-
court-program-making-the-sentence-fit-the-crime/?p=all
Henrichson, C., Rinaldi, J., and Delaney, R. (2015). The price of jails: measuring the taxpayer cost of
local incarceration. Vera Institute of Justice. Retrieved from
https://storage.googleapis.com/vera-web-assets/downloads/Publications/the-price-of-jails-
measuring-the-taxpayer-cost-of-local-incarceration/legacy_downloads/price-of-jails.pdf
Kachmar, K. (2016 Nov 27). Exclusive: inside the municipal court cash machine. The Asbury Park Press.
Retrieved from
http://www.app.com/story/news/investigations/watchdog/investigations/2016/11/27/exclusiv
e-inside-municipal-court-cash-machine/91233216/
Kelling, G. L., and Wilson, J. Q. (1982 March). Broken windows: the police and neighborhood safety.
The Atlantic. Retrieved from https://www.theatlantic.com/magazine/archive/1982/03/broken-
windows/304465/
McKoy, B. and Rosmarin, A. (2016). Marijuana legalization & taxation: positive revenue implications for
New Jersey. New Jersey Policy Perspective and New Jersey United for Marijuana Reform.
Retrieved from https://www.aclu-nj.org/files/7414/6409/3443/2016_05_24_MJRevenue.pdf
Minton, T.D., and Zeng, Z. (2016 Dec). Jail inmates in 2015. Bureau of Justice Statistics, Office of Justice
Programs, US Department of Justice. Retrieved from
https://www.bjs.gov/content/pub/pdf/ji15.pdf
Miron, J. A., and Waldock, K. (2010). The budgetary impact of ending drug prohibition. Cato Institute.
Retrieved from https://object.cato.org/sites/cato.org/files/pubs/pdf/DrugProhibitionWP.pdf
New Jersey Judiciary. (2010). Municipal Court Consolidation Plan. Retrieved from
http://www.judiciary.state.nj.us/mcs/consolidationplan9-10.pdf
17
New Jersey Judiciary. (2017). Municipal court statewide statistics. Retrieved from
http://www.judiciary.state.nj.us/mcs/caseloadstatistics.html
New Jersey State Police Uniform Crime Reporting Unit. (2014). Crime in New Jersey for the year ending
December 31, 2014. Retrieved from
http://www.njsp.org/ucr/2014/pdf/2014_uniform_crime_report.pdf
NORML. (2017). New Jersey laws & penalties. Retrieved from

http://norml.org/laws/item/new-jersey-penalties-2
Office of the Attorney General. (2017). Grants: Safe and secure communities. New Jersey Department of
Law and Public Safety. Retrieved from http://www.state.nj.us/lps/grants_state.htm#sas
Simms, N. (2014). Collateral costs: racial disparities and injustice in Minnesotas marijuana laws.
Minnesota 2020. Retrieved from
http://www.mn2020.org/assets/uploads/article/collateral_costs_web.pdf
Subramanian, R. et al. (2015). Incarceration's front door: the misuse of jails in America. Vera Institute of
Justice. Retrieved from https://storage.googleapis.com/vera-web-
assets/downloads/Publications/incarcerations-front-door-the-misuse-of-jails-in-
america/legacy_downloads/incarcerations-front-door-report_02.pdf
United States Department of Justice, Civil Rights Division. (2015). Investigation of the Ferguson
Police Department. Retrieved 7 February 2017 from
https://www.courts.mo.gov/file.jsp?id=95274
U.S. Department of Justice (2013). Summary Reporting System (SRS) User Manual. Criminal Justice
Information Services (CJIS) Division Uniform Crime Reporting (UCR) Program. Retrieved from
https://ucr.fbi.gov/nibrs/summary-reporting-system-srs-user-manual.
Van Nostrand, M. (2013). New Jersey jail population analysis: identifying opportunities to safely and
responsibly reduce the jail population. Luminosity and the Drug Policy Alliance. Retrieved from
https://www.drugpolicy.org/sites/default/files/New_Jersey_Jail_Population_Analysis_March_2
013.pdf
18
Appendix A: Statistical Methodology and Findings
To examine whether reductions in low-level enforcement could result in cost savings for municipalities
in the form of lower police and court spending, the research group created an original dataset based on
a random sample that included demographic, arrest, and budget indicators at the municipal level.
Using this dataset, the study relied on two analysis techniques. This appendix describes the linear
regression methodology that enabled us to model how fluctuations in low-level enforcement influence
municipal budgets. The research group also calculated low and high estimates of spending on low-level
enforcement based on Harvard economist Jeffrey Mirons methodology (2010), explained in Appendix
B.
Dataset and Variable Creation

The 62 municipalities included in this study were selected randomly. Originally, a random sample of 75
municipalities was drawn. Four had a population of less than 3,000 residents and were removed. The
other nine were removed due to a county supported police department (Camden), joint courts,
correctional facilities housed within their boundaries, or lack of budget data.
Demographic Indicators
Because municipalities across the state vary by population, racial composition, median income, and
poverty rate, the American Community (ACS) Survey 5-Year Estimates were utilized to create
demographic variables. ACS is a national survey that relies on monthly samples to produce annual
updates on the areas formerly surveyed by the decennial census. Unlike the census, sample variability in
the ACS is represented by a margin of error. Since the error caused by our small sample of 62
municipalities eclipses the variability in the ACS, the group did not include the demographic variables
margins of error within our analysis.
Low-Level Offense Enforcement Indicators

Two datasets provided indicators for the enforcement of low-level offenses. The group drew upon arrest
data from the Uniform Crime Report (UCR) compiled by the Federal Bureau of Investigation (FBI) to
create variables that reflect marijuana possession, disorderly conduct, total low-level offenses (based on
a limited number of categories available in the dataset), and total arrest rates. The UCR counts one
arrest for each time a person is arrested, cited, or summoned.
The UCR data has serious limitations. Police agency reporting is voluntary, so the measure lacks some
validity. Previous work by the ACLU found discrepancies between the arrest numbers in the FBI dataset
and as reported by municipalities under Open Public Records Act requests (2015). Categorization
presents another problem. The UCR allows police departments to report within an "all other non-traffic
offenses" category and it was impossible to ascertain how many low-level offenses were included in this
group. As we did not have enough time to file open records requests with each municipality, the team
relied upon UCR data to capture arrest data at the municipal level.
We also used the state Judiciarys Municipal Court statistics dataset for Court Year 2013 to create a
second indicator for low-level enforcement. This dataset includes the number of Disorderly Person (DP)
and Petty Disorderly Person (PDP) offenses and total filings in a Court Year, which spans from July to
June. Court filings are counted per charge, not per defendant, and a defendant can have more than one
charge. As a result, the raw number of DP and PDP court filings is higher than how many interactions
19
individuals actually had with police in our sample of municipalities. Nonetheless, as mentioned
throughout this report, the rate of DP and PDP filings is a more accurate representation of low-level
offense enforcement than the variables created using FBI UCR data.
Spending Indicators
2015 municipal budgets were analyzed to collect data for 2014 expenditures for total appropriations,
police spending, court spending, and court revenues. Budgets vary in how prosecutor, public defender,
and other expenses are documented. These public counsel expenses are often rolled into general legal
department line items, which include expenses irrelevant to our study, such as salaries for corporation
counsel or land use attorneys. As a result, we excluded prosecutor and public defender expenditures for
all municipalities in the sample, and our court spending variables are conservative. The police and court
spending variable should also be viewed as a low estimate because pension and benefit expenses were
not included.
Using our original dataset, we created each independent, dependent, and contextual variable. The
independent variables created using judicial statistics and FBI UCR data included low-level court filings,
low-level arrests, marijuana arrests, and disorderly conduct arrests. Dependent variables relied on
municipal budget data to capture police spending, court spending, and court revenue.
As noted in the literature review, research links low-level enforcement to race and income. In response,
we drew upon the American Community Survey to create variables that control for demographic
differences in the proportion of black residents and the poverty rate of cities and townships across the
state. We did not include Hispanic as a demographic variable because the ACS reports Hispanics of all
races as one category. Since New Jersey has one of the highest percentages of Hispanics who identify as
black, we would not be fully capturing the variance of the two populations compared to white only
residents.
Bivariate Analysis
To begin the analysis, each variable was normalized by total population and standardized to account for
having been originally measured with different scales. Initial scatterplots were run to show the
relationship between the crime and cost indicators and to determine which variables would be included
in the subsequent regression analysis. Severe outliers, defined as data points more than two standard
deviations above or below the mean, were omitted from the data analysis.
Many of our primary indicators have strong positive correlations with one another, meaning as one
variable increases, so does the other. In Figure 7 below, the green cells indicate when the relationship
between indicator in row and the corresponding column are 95% unlikely to be due to chance and are
therefore statistically significant (p < .05). The value within the green cells signals the amount of
variance described by the corresponding row or column. The number of low-level court filings, for
example, describes 22% of the variation or differences in police spending between municipalities.
20
Figure A1: Correlation Matrix
Green cells indicate a positive, statistically significant relationship, and the percentages
represent the variation explained.
Low-Level Low-Level Arrests Marijuana Possession
Poverty Hispanic
Court Filings (FBI UCR) Arrests (FBI UCR)
Black 12%
Hispanic 8%
Police Spending 6% 22%
Court Spending 14% 35% 10%
Court Revenue 76% 8%
The proportion of Hispanic residents has a statistically significant relationship to the total low-level court
filings and court spending. Due to the limitations explained both in the Impacts of Low-Level
Enforcement on Municipal Spending section of this report and in the database creation section of this
appendix we did not include it as a demographic control in our regression analysis.
As shown in Figure A1, low-level court filings were related to our spending indicators more often than
the low-level and marijuana possession arrest variables. The former was created using New Jersey
Judiciary municipal court filings, while the latter two relied on FBI UCR data. To further examine which
variable is a better indicator of low-level crime, we compare how they interact with three spending
variables below.
Limitations with FBI UCR Data

Our analysis concludes that the FBI UCR data is not the most useful measure of low-level arrests at the
municipal level. In addition to the concerns outlined in the section titled Concerns with Low-Level
Offenses in FBI Arrest Data, the following scatterplots highlight the different relationships the two
indicators have with the spending variables under study. Each dot on the scatterplots below represents
a municipality and the line running through the graph shows the direction of the relationship. When the
dots cluster closely around the line, it signals that the indicators are highly correlated, meaning changes
in one are a good predictor of what will happen in the other.
Figures A2 and A3 include the same police spending indicator on the vertical axis, but the horizontal axes
differ. The court filing variable (Figure 8) was statistically significant and explained 22% of the variance in
police spending, while the FBI UCR data was not significant and explained less than 1% of the variation.
21
Figures A4 and A5 compare the differing relationships of low-level court filings and FBI UCR arrests to
court spending. The court filings variable explains 3.5 times more variance in court spending (35%) than
the total low-level arrests variable created from UCR data (10%). In this case, both relationships are
statistically significant, but court spending is still a stronger indicator.
Figure A6 highlights the strongest, statistically significant relationship in our dataset. Court filings explain
76% of the variance in court revenue, while the FBI UCR variable in Figure A7, total low-level arrests,
only explains 2% of changes in court revenue.
22
Variables created using FBI UCR data consistently explained a lower amount of variance in our spending
indicators. Based on this finding, we concluded that New Jersey Judiciary municipal court filings are a
better indicator of low-level offenses.
Regression Analysis
The dataset was then examined using ordinary least squares (OLS) regression. Three models were built
to determine how much variation in (1) court spending, (2) police spending, and (3) court revenue can
be explained by differences in low-level enforcement, as represented by the number of municipal court
filings. To ensure outliers did not skew the results, data points that fell more than two standard
deviations from the mean were omitted from each model.
Figure A8 shows that a one standard deviation increase in low-level court filings is associated with a 0.52
standard deviation increase in police spending. A one standard deviation increase of black residents in a
municipality is associated with a 0.29 standard deviation decrease in police spending. A one standard
deviation increase of those classified as living in poverty is associated with a 0.22 standard deviation
increase in police spending.
23
Figure A8. The Impact of Low-Level Enforcement, Proportion of Black Residents, and Poverty Rate on
Police Spending
Using the low-level court filing B value in the Figure A8, Model 3 above and the average annual court
filings for Court Years 2010-2014 from the State Judiciarys Municipal Court Statistics, we were able to
calculate the marginal cost of enforcement using the equation presented in Figure A9. Through this
equation, we determined that it costs each resident $0.03 cents when police departments enforce a
new low-level offense. This value is surprisingly low, which we hypothesize is a result of high fixed costs
within the departments.
24
Figure A10. The Impact of Low-Level Enforcement, Proportion of Black Residents, and Poverty Rate on
Court Spending
Figure A10 demonstrates how an one standard deviation increase of low-level court filings is associated
with 0.72 standard deviation increase in court spending. A one standard deviation increase in black
residents in a municipality is associated with a 0.36 standard deviation decrease in court spending. A
one standard deviation increase of those classified as living in poverty is associated with a 0.04 standard
deviation increase in court spending.
According to our calculation of the marginal cost of low-level enforcement, presented in Figure A11, we
found each resident pays less than one cent to process a new low-level summons, citation, or arrest in
municipal courts.
25
Figure A12. Impact of Low-Level Enforcement, Proportion of Black Residents, and Poverty Rate on
Court Revenue
Figure A12 shows that a one standard deviation increase of low-level court filings is associated with a
0.96 standard deviation increase in court revenue. A one standard deviation increase in black residents
in a municipality is associated with a 0.17 standard deviation decrease in court revenue. A one standard
deviation increase of those classified as living in poverty is associated with a 0.11 standard deviation
increase in court revenue.
On average, according to our calculation presented in Figure A13, the courts make around one penny in
revenue on behalf of each resident when a low-level offense makes its way through the judiciary
system.
The regression model and marginal cost calculation findings suggest that budget effects alone are not
large enough to spur a large effort towards decreasing low-level offense enforcement by municipal
police departments and court systems. That said, a cost-benefit analysis would capture externalities
outside the scope of this report that warrant further research, such as quantifying community impacts.
26
The constants in A8, A10, and A12 show that fixed police costs, $12,340 are much higher than court
costs, $ 597. This regression analysis focused on marginal cost, and did not include health care or
pension costs, which would likely increase both marginal and fixed costs substantially.
We tested whether demographics, such as gender and age had a significant impact on spending and
revenue indicators in our three models. Specifically, we tested models that included those between the
ages of 18 to 24 and male; neither were found to be statistically significant.
We also tested if our dataset explained variation between low level enforcement rates in municipalities,
specifically looking at demographic drivers. We used the American Community Surveys estimates for
demographic variables (race and poverty) per municipality and ran regressions with the low-level court
filings. However, this analysis did not find anything significant. The same was true when we tried to
explain the driving forces behind marijuana possession arrests by examining race, ethnicity, age and
gender populations; there was little to no significance. Additional research is needed to determine if
factors outside the scope of our study, such as educational, policing behavior, lifestyle, or environmental
motivators are driving arrests.
27
Code book
Variable Name Variable Description Data source
Muni Name of municipality
2014 ACS 5-years

estimate
TPop Total population in each municipality
2014 ACS 5-years

percentage of people living under federal estimate
Pov_Rate poverty line
2014 ACS 5-years

estimate
Black Number of black residents
2014 ACS 5-years

estimate
Hispanic Number of hispanic residents
FBI:UCR 2014
TotArrest Total offense number in UCR dataset
FBI:UCR 2014
MParrests Number of marijuana possession offense
FBI:UCR 2014
DOarrests Number of disorderly conduct offense
FBI:UCR 2014
GbArrests Number of gambling offense
FBI:UCR 2014
LiqArrests Number of liquor law offense
FBI:UCR 2014
VagArrests Number of vagrancy offense
FBI:UCR 2014
CurLArrests Number of curfew and loitering offense
FBI:UCR 2014
Total low level offenses (Marijuana
Possession, Disorderly Conduct, Gambling,
TotLow Liquor, Vagrancy, Curfew, and Loitering)
General Appropriations
CourtRev Municipal court revenue: fines and costs
28
PolSpend Total police spending
Total court spending, including court salary
TotCourt plus other
Total police spending plus total court
Justice spending
CourtRev_Profit Court revenue minus total court spending
New Jersey Courts Data

Jul13-Jun14 DP&PDP Disorderly Persons & Petty Disorderly Persons
MuniCt Filings Filings in Muni Court July 2013 - June 2014
Created from 2014 ACS

and General
Appropriation
DV_PolicePop100 Police spending per 100 people

and General
Appropriation
DV_CourtPop100 Court spending per 100 people

and General
Appropriation
DV_CourtRevPop100 Court revenue per 100 people

and General
Appropriation
DV_JusticePop100 Justice spending per 100 people

IV_BlackPop100 Number of black people per 100 people

IV_HispanicPop100 Number of Hispanic people per 100 people

and UCR
IV_MParrestPop100 Marijuana arrest per 100 people
29
and UCR
IV_DOarrestPop100 Number of disorderly arrest per 100 people

Number of court filing of DP/PDP per 100 and court data
IV_CourtfilePop100 people

and UCR
IV_TLowLevPop100 Low-level arrest per 100 people

Number of people under poverty line per 100
IV_PovertyPop100 people
30
Municipalities Included in Random Sample
Municipality County Municipality County
Bellmawr borough Camden North Caldwell Essex
borough
Belmar borough Monmouth Norwood borough Bergen
Berkeley township Ocean Oaklyn borough Camden
Boonton town Morris Ocean township Ocean
Brielle borough Monmouth Oceanport borough Monmouth
Closter borough Bergen Paramus borough Bergen
Collingswood Camden Perth Amboy city Middlesex
borough
Colts Neck Monmouth Plumsted township Ocean
township
Cranbury township Middlesex Point Pleasant Ocean
Beach borough
Delaware township Hunterdon Point Pleasant Ocean
borough
Dover town Morris Robbinsville Mercer
township
Edison township Middlesex Rochelle Park Bergen
township
Elk township Gloucester Rockaway township Morris
Emerson borough Bergen Roseland borough Essex
Englewood city Bergen Roselle borough Union
Flemington Hunterdon Roxbury township Morris
borough
Franklin township Hunterdon Rutherford borough Bergen
Green Brook Somerset Salem city Salem
township
Hasbrouck Heights Bergen Sayreville borough Middlesex
borough
Hillsborough Somerset Shrewsbury Monmouth
borough
Hillsdale borough Bergen South Orange Essex
Village township
Hopewell township Mercer Trenton city Mercer
Howell township Monmouth Vernon township Sussex
Lacey township Ocean Waldwick borough Bergen
Lawrence township Mercer Washington Gloucester
Township
Lincoln Park Morris Watchung borough Somerset
borough
Lumberton Burlington West Deptford Gloucester
township township
Manville borough Somerset Wildwood city Cape May
31
Milltown borough Middlesex Wood-Ridge Bergen
borough
Millville city Cumberland Woolwich township Gloucester
Montclair township Essex
Netcong borough Morris
32
Appendix B: Municipal Cost Analysis
To estimate municipal spending on enforcing low-level offenses, a methodology was employed based on
the Harvard economist Jeffrey Miron. Miron has estimated the cost of drug prohibition using police,
court and corrections data (2010). Miron describes the findings generated from his method as ballpark
figures and the same applies to this analysis (2010). These cost estimates should be understood not as
precise numbers but as a way for municipalities to begin to understand the degree to which low-level
enforcement affects their spending. Its important to note that unlike the statistical analyses, the costs
generated from this method include fixed costs to police and court systems.
Since this analysis focuses on low-level drug offenses and other offenses that do not pose a threat to
public safety, certain adjustments were made to the methodology. In addition, accurate corrections
data on the number of inmates incarcerated in county jails for the study offenses were not available so
this cost was not included. A Drug Policy Alliance report found that roughly 17% of inmates in county
prisons were incarcerated in the county system under drug charges as the primary charge, however, the
researchers did not report the number of inmates within that category incarcerated on marijuana
charges specifically (Van Nostrand, 2013). So for the purposes of this study, police and municipal courts
were the only cost categories under consideration, and the overall justice spending is a conservative
estimate.
Miron utilized FBI UCR arrest data, which counts one arrest for each time a person is arrested, cited or
summoned (2010). Marijuana possession (including possession over and under 50 grams), disorderly
conduct, gambling, liquor law violations, vagrancy and curfew were the offenses that could be counted
as low-level for this analysis because they are non-indictable offenses in New Jersey, known as
disorderly persons and petty disorderly persons. However, the UCR program codes many other low-
level and indictable offenses together into a category called All Other Offenses (U.S. Department of
Justice, 2013). Due to this limitation with counting low-level arrests in the UCR arrest data, and the fact
that a 2015 ACLU-NJ study of low-level offenses found significant issues with UCR arrest data, this
indicator was not used (2015). Instead, the analysis used the municipal court filings as a proxy for both
police time and court time spent on low-level offenses.
In New Jersey, municipal courts are responsible for adjudicating disorderly persons and petty disorderly
persons cases. The courts publish statistics on the DP and PDP filings. Like the UCR data, the court filings
include arrests, citations and summons. The court statistics count each charge, not each person, and a
defendant can have more than one charge. Since this indicator is transformed into a ratio of DP and PDP
filings to total filings, it seems unlikely that this is an overstatement. But since there is no public data
available on the number of individuals charged with more than one offense, or the amount of time
spent processing a low-level offense takes as compared to traffic violations, DUIs, or other offenses, a
conservative downward adjustment of 50% was applied to the DP/PDP filings to create a low-end
estimate. This adjustment is based on the ACLU study of low-level offense arrests in four New Jersey
cities that showed 56% of arrests were for a stand-alone charge for one of the low-level offenses under
study (2015).
To estimate the spending, first the amount of disorderly persons and petty disorderly persons filings is
divided by the total court filings. This ratio is then multiplied by the police budget to get the police
spending estimate. Its important to note here that municipal police budgets are a conservative
33
estimate because they do not include pensions or health benefits, which are a considerable additional
cost. Next, the low estimate for police spending is estimated by multiplying the high estimate by 50%.
Next, the portion of court budgets devoted to low-level offenses is calculated. Prosecutor and public
defender salaries were omitted from the analysis because many municipal budgets do not provide this
information as a separate line; instead it is lumped into legal department spending which includes
additional municipal legal services. Therefore, this is a conservative estimate of court costs. As with
police spending, the court budget is multiplied by the ratio of disorderly persons and petty disorderly
persons filings to total filings to calculate the high estimate. The low estimate is then found by
multiplying by 50%.
Finally, the total spending on low-level offense enforcement can be calculated by adding together the
police and court spending for each municipality. An average can also be calculated from the high and
low estimates, along with estimates for per capita spending, and percent of municipal budget.
The arrest rates for marijuana possession were calculated from the FBI UCR data for each municipality.
Despite that datas limitations for use in cost analysis, it is useful here. Although the UCR datas validity
is questionable, it is a reliable measure because it is collected annually, which is a common issue with
federal data collection. This means that the data is more useful to look at trends over time or to make
comparisons between groups. Therefore, calculating an arrest rate to compare municipalities is an
appropriate use of this data, unlike the cost analysis or the statistical models. The arrest rate
calculations followed the model used by the ACLU in their study of low-level offenses (2015). For each
community, the number of marijuana possession arrests was divided by the population. This ratio was
multiplied by 100,000 to get the arrest rate per 100,000 people.
Delaware Township, Dover, and Elk were excluded from this sample because they have joint courts.
Works Cited
enforcement of low-level offenses in New Jersey. Retrieved from
https://www.aclu-nj.org/files/7214/5070/6701/2015_12_21_aclunj_select_enf.pdf
Miron, J. A., and Waldock, K. (2010). The budgetary impact of ending drug prohibition. Cato
Institute. Retrieved from
https://object.cato.org/sites/cato.org/files/pubs/pdf/DrugProhibitionWP.pdf
U.S. Department of Justice (2013). Summary Reporting System (SRS) User Manual. Criminal Justice
Information Services (CJIS) Division Uniform Crime Reporting (UCR) Program. Retrieved from
https://ucr.fbi.gov/nibrs/summary-reporting-system-srs-user-manual.
Van Nostrand, M. (2013). New Jersey jail population analysis: identifying opportunities to safely and
responsibly reduce the jail population. Luminosity and the Drug Policy Alliance. Retrieved from
https://www.drugpolicy.org/sites/default/files/New_Jersey_Jail_Population_Analysis_March_2
013.pdf
34
Appendix C: Differences between the Statistical and Municipal Cost
Analysis Methodologies
Both the regression and Miron methodology have strengths and weaknesses. In the absence of having
individual-level data on how police spend their time (outputs), Miron's methodology offers a reasonable
model to estimate costs over a one-year period using aggregate costs. A more robust method would
examine the relationship between workload outputs and costs over several years, controlling for
confounding variables. On the other hand, the regression accounts for fixed costs, where Miron does
not. It also allows us to better understand which relationships could not happen by chance, illuminates
demographic differences, and proves that fluctuations in low-level offenses do impact police spending.
Regression Miron
$0.04 per capita for each additional low-level $21.26 per capita spent in 2014 on
offense enforced based on 2014 data low-level offense enforcement
35
Appendix D: Individual Cost Profile for Low-Level Offense Arrest
If charged with a DP or PDP offense, an individual will appear in the Municipal Court where the offense
occurred, and the defendant is not entitled to a jury trial. If convicted of a DP offense, the municipal
judge can impose a monetary fine of up to $1,000 and/or a jail sentence of up to six (6) months, served
in a county facility. PDP offense conviction can result in a fine up to $500 and/or thirty (30) days in
prison. Additional restitution, community service and probation can be required for both. The imposed
fine and jail sentence is largely arbitrary, determined by the presiding municipal judge (Pedraza 2017).
Cost to Individual
Beyond the fines outlined above, there are additional monetary costs and consequences associated with
a DP or PDP offense conviction.
A DP and PDP conviction carries the following mandatory fees, in addition to the fine imposed by a
municipal court judge:
Victims of Crime Compensation Board (VCCB): $50
Safe Neighborhood Services Fund: $75
Court Costs: $33
Drug Enforcement and Demand Reduction Penalty (DEDR) when conviction is for drug
possession or paraphernalia: $500
Mandatory lab fee for drug related convictions: $50
If an individual does not pay these fines, he or she can be held in contempt of court and
incarcerated. Municipalities can hire private collection firms to collect the fines under certain
conditions1, and these firms can charge the defendant up to a 22% additional fee on the amount to be
collected (New Jersey Judiciary 2011). Therefore, a $1000 fine for marijuana possession under 50 grams
will result in $1708 in total court fines and fees; if the balance is referred to collection, an individual
could end up owing over $2080.
A first-time offender that chooses to participate in a diversionary program (conditional dismissal or

conditional discharge) in municipal court to avoid a criminal record can still be charged fines, up to the
amount a conviction would have yielded, and a program application fee (New Jersey Judiciary
2013). The Asbury Park Press interviewed one New Jersey resident who paid $1200 in fines and
surcharges after participating in the conditional discharge program for a marijuana possession charge
(Kachmar 2016).
Appearance in court can cost an individual in terms of lost wages, transportation expenses, and child
care arrangements. If an individual fails to appear for a court date, it is a fourth degree indictable
offense (crime) and a warrant can be issued for their arrest (ACLU 2015).
Finally, a defendant in municipal court will have attorney costs if they choose to be represented. The
state Office of the Public Defender does not cover municipal courts; instead, each municipality provides
public defender services. The municipal court determines if the defendant qualifies based on need,
including financial ability to afford representation, employment status and salary, liquid assets, and
other factors, and the municipality can charge up to $200 to apply for the public defender. However, a
municipal defense attorney in New Jersey the researchers spoke with stated that defendants had to
essentially be below the federal poverty line to qualify for these services (Pedraza 2017). Therefore,
most individuals who have representation in municipal court hire a private attorney, who typically
36
charge a flat fee for handling the case. This can range depending on the quality of the attorney and the
offense charge from $200-$1500, but the municipal defense attorney interviewed by the researchers
estimated the average cost is $500-$750 (Pedraza 2017). Simms (2014, 21) found the flat fee for
representation in Minnesota was $500-2000 for a petty misdemeanor marijuana charge and $1000-3500
for a misdemeanor marijuana charge.
The inability to afford or other lack of access to a lawyer was cited by the National Association of
Criminal Defense Lawyers (NACDL) in 2009 as a major contributor to the denial of constitutional rights to
individuals accused of misdemeanors, and the organization found people of color were
disproportionately impacted (Boruchowitz, Brink, and Dimino 2009). The NACDL noted that in 2000,
28.3% of inmates incarcerated for misdemeanors reported having had no attorney representation
(Boruchowitz, Brink, and Dimino 2009, 14). The organization has called for the provision of counsel to
all defendants facing possible jail time.
Collateral Consequences to an Individual

Although not considered a crime, a defendant convicted of a DP or PDP offense will have a criminal
record and will need to file an expungement application to clear the charge from his or her record. A
waiting period applies before an individual can file for expungement: 5 years for DP offenses, at least 3
years for PDP offenses, and 2 years for municipal ordinance violations (Legal Services of New Jersey
2017). The filing fee is $75, but it can be waived, and documents have to be sent by certified mail to
multiple agencies and offices, multiple times in the process, the cost of which can mount quickly. An
individual may choose to hire (1) a service to help file the paperwork, which can cost between $100-
$300, or (2) a lawyer, which can cost $700-$2000, as a Petition for Expungement will be denied if any
required steps are completed improperly (Tiziano 2016).
Additionally, an individual that uses a diversionary program (conditional dismissal or conditional

discharge) for a first time offense in municipal court, such as on a drug charge, is barred from using a Pre
Trial Intervention (PTI) program if they are later arrested for an indictable crime (New Jersey Judiciary
2013, Pedraza 2017). The diversionary programs and PTI redirect first time offenders from prosecution
to probation monitoring or a rehabilitative program and upon successful completion, the individual does
not have a criminal record (New Jersey Judiciary 2017).
Some DP offenses require automatic driver license suspension or revocation for not less than six months
and up to two years; marijuana possession in a motor vehicle results in an automatic two year license
suspension (NORML 2017). The NACDL report noted that driving on a suspended license constituted a
large portion of the misdemeanor caseload in many jurisdictions, indicating individuals who lose their
license as part of a low level offense conviction may be involved with the municipal court system again
at a later point (Boruchowitz, Brink, and Dimino 2009, 26). To restore your license in New Jersey, the
base fee is $100 (New Jersey Motor Vehicle Commission 2017). The suspension may also affect
automobile insurance rates, or the policy may be dropped entirely (Simms 2014).
A DP offense may result in professional license suspension as well for doctors, nurses, attorneys,
pharmacists, stockbrokers, real-estate agents, accounting, taxi drivers, day care provision and others
(Simms 2014, 25).
These arrests can result in further long-term impacts or collateral consequences, even without
conviction or incarceration. An arrest can result in eviction from public housing or leased residence
37
(ACLU 2015, NORML 2017, Simms 2014). Simms (2014, 3) found that in Minnesota the loss of public
housing could cost a family at least $14,300 more over three years and at least $23,900 more over five
years for private housing. Arrests can also affect immigration status, federal student aid eligibility,
eligibility for other public assistance programs, and child custody (ACLU 2015, Boruchowitz, Brink, and
Dimino 2009, Simms 2014).
An arrest and conviction for low-level offenses can affect income, employment status, and future job
prospects. Simms (2014) found that over 60% of employers indicate that they probably would not or
definitely would not hire someone with a criminal record, including those with only an arrest record
containing no conviction. More individuals can be impacted as criminal records become easier and less
expensive to access. Beyond the arrest or conviction record, it may be harder to find employment if
your driver license is suspended as well. If incarcerated as a result, the loss is even greater. One study
found that serving time resulted in a 10-30% decrease in earnings for at least a decade (Simms 2014,
24). A 2015 study by the Vera Institute of Justice found that following incarceration, men saw hourly
wages decreased by 11 percent, annual employment by nine weeks, and annual earnings by 40 percent
as a result of time spent in jail or prison (Subramanian et al. 2015).
An individual convicted of a DP offense does not lose the right to vote and is not disqualified from
serving on a jury.
Estimating Costs for an Adult Arrested for Marijuana Possession Under 50 Grams
Fines:
Disorderly Persons Offense Fine $1,000
Victims of Crime Compensation Board (VCCB) $50
Safe Neighborhood Services Fund $75
Court Costs $33
Drug Enforcement and Demand Reduction Penalty (DEDR) $500
Mandatory lab fee $50
Attorney fees: $750
Reinstatement of driver's license: $100
Expungement:
Filing Fee $75
Certified Mail Expenses (at least 5 mailings twice at $6.06 each)2 $60.60
Criminal History Background Check from State Police $40.70
Filing Service $300
TOTAL $3034.30
Incalculable:
Lost wages from time in court, meeting probation requirements
Impact on employment status and future earnings
Transportation costs due to loss of license
38
Car insurance rate changes
Civil forfeiture losses
Works Cited
enforcement of low-level offenses in New Jersey. Retrieved from https://www.aclu-
nj.org/files/7214/5070/6701/2015_12_21_aclunj_select_enf.pdf
Boruchowitz, R.C., Brink, M.N., and Dimino, M. (2009). Minor crimes, massive waste: the terrible toll of
Americas broken misdemeanor courts. National Association of Criminal Defense Lawyers.
Retrieved from
https://www.prisonlegalnews.org/media/publications/Minor%20Crimes%2C%20Massive%20Wa
ste%20-
%20Terrible%20Toll%20of%20Americas%20Broken%20Misdemeanor%20Courts%2C%20NACDL
%2C%202009.pdf
Kachmar, K. (2016 Nov 27). Exclusive: inside the municipal court cash machine. The Asbury Park Press.
Retrieved from
http://www.app.com/story/news/investigations/watchdog/investigations/2016/11/27/exclusiv
e-inside-municipal-court-cash-machine/91233216/
Legal Services of New Jersey. (2017). Cleaning your record: a six-step guide to expunging criminal
records in New Jersey. Retrieved from
https://www.lsnjlaw.org/Publications/Pages/Manuals/CYR.pdf
New Jersey Judiciary, Administrative Office of the Courts. (2013 Sept 9). New law advisory: P.L. 2013, c.
158- establishes a conditional dismissal program in the municipal courts. Retrieved from
https://www.judiciary.state.nj.us/legis/P.L.%202013,%20c.158%20-
%20Establishes%20a%20conditional%20dismissal%20program%20in%20municipal%20courts.pd
f
New Jersey Judiciary. (2017). Pre Trial Intervention program (PTI). Retrieved from
http://www.judiciary.state.nj.us/criminal/crpti.html
New Jersey Judiciary, Municipal Court Services Division. (2011). Supreme court procedures governing th
private collection of municipal court debt under L. 2009, c. 233. Retrieved from
http://www.judiciary.state.nj.us/mcs/collectionagency.pdf
New Jersey Motor Vehicle Commission. (2017). Restoring your license. Retrieved
from http://www.nj.gov/mvc/Violations/suspension_restoring.htm
NORML. (2017). New Jersey laws & penalties. Retrieved from

http://norml.org/laws/item/new-jersey-penalties-2
Pedraza, M. (2017, Feb 22). Personal Interview.
39
Simms, N. (2014). Collateral costs: racial disparities and injustice in Minnesotas marijuana laws.
Minnesota 2020. Retrieved from
http://www.mn2020.org/assets/uploads/article/collateral_costs_web.pdf
Subramanian, R. et al. (2015). Incarceration's front door: the misuse of jails in America. Vera Institute of
Justice. Retrieved from https://storage.googleapis.com/vera-web-
assets/downloads/Publications/incarcerations-front-door-the-misuse-of-jails-in-
america/legacy_downloads/incarcerations-front-door-report_02.pdf
Tiziano, J. (2016). Expungement cost in New Jersey. Legal Crunch. Retrieved

from http://expungementinfo.com/expungement-cost-new-jersey/
40

NJPP Final Report

Uploaded by

Copyright:

Available Formats

NJPP Final Report

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

NJPP Final Report

Uploaded by

Copyright:

Available Formats

The Municipal Cost of

A report prepared for New Jersey Policy Perspective by graduate

Jane Allen Yan Huang Arcadia Lee

Findings indicate that:

On average, the municipalities spent an estimated 1.3% of their budget on low-level

Low-Level Offenses and Enforcement Costs

The majority of Municipal Court

Municipal Courts handle the vast

Trends in Low-Level Offense Enforcement

Low-Level Marijuana-Related Offense Enforcement

Impacts of Low-Level Enforcement on Municipal Spending

The horizontal lines, or x-axes, show the range of low-level

Looking at the red dots position between the two axes, it

But the spending and court revenue scatterplots differ in one

Low-Level Enforcement and Demographic Impacts on Spending

The results of those analyses are:

A $4.84 reduction in police spending

police spending of $3.22

Cost Analysis and Case Study of Large Municipalities

According to this analysis, the median amount spent on

The top 10 spenders per capita were: Paramus

Seven large cities and municipalities were also analyzed

Concerns with Low-Level Offenses in FBI Arrest Data

Recommendations & Discussion for Further Research

NORML. (2017). New Jersey laws & penalties. Retrieved from

Dataset and Variable Creation

Low-Level Offense Enforcement Indicators

Limitations with FBI UCR Data

Variable Name Variable Description Data source

Muni Name of municipality

2014 ACS 5-years

2014 ACS 5-years

2014 ACS 5-years

2014 ACS 5-years

PolSpend Total police spending

New Jersey Courts Data

Created from 2014 ACS

Created from 2014 ACS

Created from 2014 ACS

Created from 2014 ACS

Created from 2014 ACS

Created from 2014 ACS

Created from 2014 ACS

Created from 2014 ACS

Created from 2014 ACS

Created from 2014 ACS

A first-time offender that chooses to participate in a diversionary program (conditional dismissal or

Collateral Consequences to an Individual

Additionally, an individual that uses a diversionary program (conditional dismissal or conditional

NORML. (2017). New Jersey laws & penalties. Retrieved from

Pedraza, M. (2017, Feb 22). Personal Interview.

Tiziano, J. (2016). Expungement cost in New Jersey. Legal Crunch. Retrieved

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.