Beijing Law Review, 2011, 2, 74-87
doi:10.4236/blr.2011.22008 Published Online June 2011 (http://www.SciRP.org/journal/blr)
The Social Costs of Crime and Crime Control
Klara Kerezsi*, József Kó, Szilvia Antal
National Institute of Criminology, Budapest, Hungary.
Received February 28th, 2011; Revised May 11th, 2011; Accepted May 18th, 2011.
ABSTRACT
There is currently no generally accepted method of estimating the costs of crime. After presenting the most commonly
used methods of estimating crime , the authors attempt to explore the situa tion in Hungary. Taking 2009 as a base year,
they recon the crime-related social expenditure accounts. The authors, with the help of other Hungarian research data
and databases, have also taken into account the costs of the secondary social effects. The results of the calculations
depend on the applied app roach to crime and the interpreta tions of the socia l impacts of the delinquen cy. According to
the authors calcu lations the socia l cost caused by crime was about 2.17 billion USA d ollar: ($) (1.6 billion euro (EUR)
in 2009. The authors d educt the sum that was drawn by the offen ders as a benefit/profit from committing crime; there-
fore the crime caused 1.17 billion $ as a net social damage in 2009 in Hungary. The amount of 1.63 billion $ was spent
on the crime control (e.g. law enforcement, judiciary, prison and crime prevention) in 2009. The results show that de-
linquency caused a total of 3.8 billion $ as a damage, or as an expenditure spent by the government in 2009. Th e co s t of
crime control wa s about 500 million $ higher (1.63 billion $) than the amount of damage caused by crim e (1.17 billion
$). The offenders benefit/profit from committing crime (= 1 billion $) was only 15% less than the damage they caused to
the state and to the citizens (= 1.17 billion $). In other words, the half of the criminal damage shall never be repaid: it
will remain at the criminals!
Keywords: Cost of Crime, Victimization Costs, Methodology
1. Introduction
The concept of crime embraces an extremely diverse
range of activities. Even as regards the functioning of
society and the economy there may be differing conse-
quences from specific types of criminal acts. The taking
of a life is clearly an irreplaceable loss not only for the
individual, but also for society. Where offences against
property are concerned the situation is not so obvious. If
somebody steals or embezzles somebody else’s money,
or robs them of their possessions through the use of force,
then the victim suffers an unquestionable loss, which is
often far greater than the value of the property lost during
the crime. For society as a whole, however, the loss of
one member of society is a significant gain for another
one of its members. Damage beyond the particular mate-
rial losses is also caused by an increase in crime. If the
state is unable to guarantee protection for possessions
that people own, if they can be taken away from them at
any time by using force or other illegal means, then peo-
ple lose their motivation for acquiring assets.
If more and more economic players, instead of choos-
ing market solutions, opt for illegal solutions, which
promise similar or even greater profits, then the well-
being of society may be drastically reduced. Market eva-
sion and the illegal transfer of wealth is not effective in
the economic sense that there is no increase in economic
productivity; on the contrary, there is a reduction in so-
cial prosperity, in spite of the resulting visible wealth that
some individuals who choose the illegal path may enjoy.
It is therefore in the fundamental interests of societies, or
rather, of the state, to make a separation between legal
market behaviour, which brings economic benefits, and
illegal solutions, which serve merely the transfer of
wealth. It is therefore reasonable for a society to pass
laws which punish bad behavio ur, and to operate a ju stice
system to enforce these laws, even if this incurs signifi-
cant costs.
It is precisely these costs and expenditures which we
investigated in our analysis. We were interested in find-
ing out what the costs of crime are to the community, at a
social level. We employed a dual approach:
1) As a first step we sought to define the damage
caused by crime;
2) In the second phase we attempted to summarise the
*Klara Kerezsi, Corresponding Author.

The Social Costs of Crime and Crime Control75
social outlay spent on eliminating crime and keeping it
under control.
2. Calculating the Costs of Crime
International practice shows that there are serious differ-
ences among various countries in the way the costs of
crime are calculated, and even the American and the
British measuring techniques are different from each
other [1].1
Knowledge of the methods used for estimating the
costs of crime is important for scientific experts and for
those working in the field. In its consideration of crime-
related costs, this overview starts takes the costs of trad i-
tional “street” crime as its starting point, and does not
focus on costs related to white-collar crime or organised
crime.
Crime-related costs can be systemised in different
ways. Regarding where the costs arise, we make a dis-
tinction between (1) costs affecting the victim (such as
lost income, medical costs, pain and suffering) and (2)
costs affecting the system of government (such as the
costs of organising the police and criminal justice ser-
vices).
The repercussions of crime do not, however, only
concern the victims and the system of government, but
also have an impact on numerous other segments of soci-
ety.
The professional literatures on the econometrics of
crime traditionally make a distinction between three types
of costs [2]:
(1) The direct costs caused by the criminal offence
(such as external2 costs caused by the perpetrator);
(2) Costs related to the reaction to crime (which target
the suppression of crime, futu re crime prevention, or spe-
cific punishments);
(3) Costs incurred by the perpetrator (such as the costs
of the perpetrator’s time, in other words the time spent on
criminal activity instead of on work or some other kind of
productive activ ity).
Cohen and Bowles [3], in answer to the question “Who
bears the costs?”, make a distinction between three cate-
gories of costs: the costs of (1) foresight, (2) conse-
quences and (3) reactions to crime.
Arising costs are grouped in the following way by
Cohen [4]:
(1) Victimisation costs (including losses of cash, pain,
suffering and a lowering of the quality o f life);
(2) Crime prevention outlay by individuals and com-
panies;
(3) Changes in human behaviour (such as the devel-
opment of individual avoidance behaviour);
(4) The costs of the criminal justice service system;
(5) Governmental prevention and rehabilitation pro-
grams;
(6) Residual effects for the individual (such as a fear of
crime);
(7) Residual effects for the community (such as the
loss of a tax payer);
(8) Costs of overinsurance (such as non-provision of
an activity not undertaken by innocent people for fear of
being accused of committing a crime);
(9) The demonstrative costs of “serving justice” (such
as costs that arise in order to show clearly that justice is
“at your service” but are regardless to e.g. cost-benefit
aspects);
(10) The burdens that arise on the side of the perpetra-
tor and their family.
Methods of Estimating Costs
According to sources in the professional literature there
are two approaches to calculating the costs of crime:
1) The one most generally applied to this day is the
“bottom-up” method of estimation, where estimates are
based on the costs of crime incurring by single individu-
als. The “bottom-up” approach is used in the Home Of-
fice, where the British Crime Survey (BCS), the annually
repeated victim study, estimates the costs of victimisation
on the basis of answers to the survey’s questions. For this
method of calculation it is naturally essential to have
regularly repeated victim surveys [5].3 Research shows
that the number of crimes against natural persons of adult
age and against households is four times greater than the
number reported to the police. Besides the BCS the
number of victims in the business sphere is measured
regularly using the Commercial Victimisation Survey.
The “bottom-up” method of calculation takes the fol-
lowing costs into account:
(1) Costs incurred by the victim;
(2) Costs of avoiding victimisation;
(3) Community expenses;
1In Australia, Canada, the Netherlands and the USA, crime-related
expenditure exceeds 5% of annual GDP. The price of crime and its
costs to society are therefore earning increasing attention when restruc-
turing the criminal justice services and when defining the direction o
f
criminal policy. Decision makers also need to be aware of how the
consequences and the damage caused by crime compare with the
amounts that are spent on dealing with crime.
2In the economics, the term externality (external economic effect) for
effects (benefits or costs)—that may be connected to production o
r
consumption—incurring by those without pursuit of those. External
cost is made by the difference of social and individual marginal cost.
(4) Costs of the fear of crime;
(5) Costs of the criminal justice service;
3The last round of the International Crime Victim Survey (ICVS) was
held in 2005 and the next was planned for 2010. Hungary also took part
in the 2005 survey, which was conducted by Gallup Europe. (The in-
ternational research team made their original results for Hungary
available to us.) Plans call for the creation of a European Standardized
Victimization Survey (EU CVS). The pilot study started in 17 EU
Member States in 2009/2010 and the first full survey will be in 2013.

The Social Costs of Crime and Crime Control
76
(6) Costs arising on the side of the perpetrator.
The correlation and interpretation of the large amount
of data and the cost-implications by each single element
of the above categories present significant difficulties.
2) The other possible approach is the elaboration of
“top-down” estimation. There are typically two solutions
applied with this method of calculation:
(1) Using the available macroeconomic indices, or the
budgetary data and institu tional expenses, to calculate th e
social costs of crime;
(2) Based on a survey of the population, calculating th e
“price” of crime by including every expense that the
population is read y or willing to pay for reduced crime.
There are also several approaches to employing this
latter method:
1) One method of calculation takes actual market prices
as its basis, and concentrates on changes in the value of
assets caused by the frequency of crime. This is what a
house-buyer takes into consideration when they account
for the possible risk of victimisation in an area where
property prices are cheaper.
2) With the second method, respondents are asked to
evaluate public wealth in a subjective way. Typically,
population surveys are used to estimate how much people
are willing to pay to avoid victimisation.
3) A relatively new approach for calculating the price of
crime is when the costs of crime are assumed on the basis
of satisfaction with life. The survey does not ask for an
evaluation but f or the subjective opinion of how satisfied
the respondent is with th eir life. This calculation is a me-
thod suitable for individual evaluation of the feeling of
security in connection with the crime situation.
3. The Costs of Crime in Hungary
There follows an experiment to estimate the costs of
crime in Hungary. Since there is no annual (or at least
regularly conducted) survey of the frequency of victimi-
sation, we are unable to use the “bottom-up” method of
calculation. We encountered, however, several stumbling
blocks concerning the “top-down” method of calculation
when gathering the data required for making the calcula-
tion. Not all the data required for making the calculation
are available, or they are not collected in a format or to
the level of detail in domestic statistical sys tems that they
should have been.
We have indicated the missing elements in our calcula-
tion. To estimate the damage caused by criminal offences
we used the data from the Unified Criminal Statistics of
the Investigative Authority and the Public Prosecutor
(UCS) for the year 2009 as our star ting point. These data
do not correspond precisely with the calendar year, as
43% of the offences contained in the UCS statistics were
not committed in the given year. Data for a given calen-
dar year only become “complete” after about 4-5 years.
In spite of this we rejected the po ssibility of performing a
calculation using the “complete” data of an earlier year,
because in the year 2010, data that refer to a year 4 or 5
years ago are not so informative.
In the case of offences for which Unified Criminal Sta-
tistics does not record a value for the damage we relied
primarily on the results of victimological studies that
were conducted in the past, but for preparing the analysis
we also utilised data from other registers (such as insur-
ance statistics and social insurance registers). Some of-
fences were not considered, either because their effect is
negligible or because there is no reliable information
about the damage caused by these offences. Table 1 pro-
vides an overview of the data sources and the groups of
crimes taken into account.
3.1. Calculating the Damage Caused by Crime
1) Damage recorded in the Unified Criminal Statistics
C1 = Dr – Hrp (1)
where:
C1 = the amount of damage caused by crime;
Dr = recorded damage (the amount of damage caused
by crimes, for which the value of the damage is recorded
in the Unified Criminal Statistics);
Hrp = repaid harm (the value of repaid harm registered
Table 1. Data sources and the groups of crimes taken into
account.
Chapter Included in the
calculation? Sources of data
Crimes against the
state no
Crimes against hu-
manity no
Crimes against the
person yes Unified Criminal Statistics, victimization survey
Traffic offence s yes Unified Criminal Statistics,
insurance statistics
Crimes against mar-
riage, the family,
youth and sexual
morals
yes Unified Criminal Statistics,
victimization survey
Crimes against state
administration, the
course of justice and
the integrity of public
life
yes Unified Criminal Statistics
Crimes against public
order yes Unified Criminal Statistics
Economic crimes yes Unified Criminal Statistics
Crimes against prop-
erty yes Unified Criminal Statistics, victimization survey
Crimes against mili-
tary obligations no

The Social Costs of Crime and Crime Control77
in the Unified Criminal Statistics).
The Unified Criminal Statistics collects data on the
amount of damage caused by crimes in four categories of
crime (crimes against property, economic crimes, crimes
against state administration and those against public or-
der). The reliability and validity of the collected data dif-
fer. The value of damage registered for crimes against
property is regularly below the actual damage caused. If
somebody has their purse or wallet stolen, together with
their documents and other personal effects, the report
regularly only includes the value of the purse or wallet
and the money inside it as damage. The costs of replacing
the documents (charges, time, travel expenses) or, for
example, changing the locks of the house if the key was
in the bag, are not included in the statistics. In summary,
the statistics for crimes against property underestimate
the amount of material damage caused.
In the international literature it is becoming increas-
ingly clear that so-called moral or emotional damage
should also be accounted for in the costs of crime. This
covers all the emotional and mental injuries, and the
changes that take place in the way of life as a conse-
quence of these, which can be considered as resulting
from a crime. The costs of psychological and emotional
support and care in the face of a crime are the costs that
are most frequently included.
Hungarian statistics do not include any damage other
than material harm. There is no tradition of this type of
calculation in Hungarian practice, so we were also unable
to take these into consideration. Nevertheless we must
note that economists do work with some similar index
when calculating national welfare indices (such as the
Index of Sustainable Economic Welfare [ISEW]), where
the fall in general well-being that is attributable to crime
is assessed as an emotional burden. No valid research has
yet been carried out, however, in connection with pro-
viding actual figures.
There is a problem as regards registration; in that data
is only included in the Unified Criminal Statistics after
the investigation has closed. In instances where the per-
petrator was not found, the value of the damage enters the
statistics at the amount determined by the investigative
profession, and in instances where the perpetrator is
known, the amount registered is the one that appears on
the indictment, although if this is later amended during
the judicial process there is no subsequent follow-up in
the register. In summary we are able to state that in cases
where the perpetrator is known the statistics show a lar-
ger amount of damage than that established in final court
judgments. In cases where it was not possible to identify
the perpetrator, it is normal to determine the amount in a
more simplified process (with the involvement of fewer
experts), so we necessarily have to reckon with a degree
of uncertainty in such cases.
For all types of crime we have deducted the value of
repaid harm indicated in the register from the value of the
damage caused. The values that form the basis of the
calculation are summarised in the Table 2.
Based on the data recorded in the Unified Criminal
Statistics, the amount of damage caused by these crimes
in 2009 was:
C1 = 573.5 – 52 = 521.5 million $
2) Correction of the amount of damage recorded in the
Unified Criminal Statistics
C2 = Dr * Corr – Hrp (2)
where:
C2 = the amount of damage caused by crime;
Dr = recorded damage (the amount of damage caused
by crimes, for which the value of the damage is recorded
in the Unified Criminal Statistics);
Hrp = repaid harm (the value of repaid harm registered
in the Unified Criminal Statistics).
In determining the correction factor we relied on the
results of our previous research. The research was con-
ducted earlier, but we have no reason to assume that there
have been since then any significant shifts in the adjust-
ment in the area under investigation, therefore we con-
sider the earlier results to form the basis of relevant con-
clusions on the relationship between the actual value of
damage and that recorded in the system of statistics. The
nationwide victimology research carried out in 2003 [6]
by the National Institute of Criminology (OKRI) involv-
ing a representative sample of 10,000 people is consid-
ered to be the most important source. Among the ques-
tions were those evaluating the damage caused to victims.
The questions on damage were put not only with regard
to crimes against property but in every case.
However, when determining the correction factor K1,
only the values for crimes against property were reck-
Table 2. The basis of calculating the costs of crime.
Damage caused
($) Compensated damage ($)
Crimes against
property 508,285,572 37,194,904
Economic crimes 207,879,145 15,059,929
Crimes against
state administra-
tion, the course of
justice and the
integrity of public
life
5,775,024 3,035,835
Crimes against
public order 1.506.965 15.850
Total 723.446.706 52.285.862

The Social Costs of Crime and Crime Control
78
oned with. This was necessary because the Ve index does
not include a value for damage arising from crimes
against the person.
The following procedure was followed in determining
correction factor K1. For the purpose of comparison we
used the crime statistics for 2003, the time when the vic-
timology questionnaires were completed. As a first step
we separated the damage resulting from offences com-
mitted against natural persons from the Unified Criminal
Statistics data. This was necessary because we were only
able to examine crimes against natural p ersons during the
victimological study. We also filtered out from the victi-
mological survey data those cases where the respondents
themselves admitted they had not reported the crime.
Only data relating to reported cases can be compared
with the data of the official crime statistics. Following
this separation we calculated the value of damage per
criminal offence in both databases (the victimological
survey and the Unified Criminal Statistics data). The d if-
ference between the two average figures provides the
value of the correction factor, which in this case was 1.17.
As far as present knowledge is available, the best ap-
proximation of the value of K1 is as follows:
Corr = 1.17
C2 = (573.5 * 1.17) – 52 = 619 million $
3) Estimation of the value of damage resulting from
crimes for which there is no recorded damage in the Uni-
fied Criminal Statistics
C3 = Dr * Corr – Hrp + Dunr (3)
where:
Dr = recorded damage by the Unified Criminal Statis-
tics;
Corr = correction factor, to adjust the value shown in
the Unified Criminal Statistics to the actual amount of
damage caused;
Hrp = the value of repaid harm registered in the Unified
Criminal Statistics;
Dunr = unregistered damage (the amount of material
damage caused by crimes for which there is no recorded
damage in the Unified Criminal Statistics).
This type of damage must be split into two groups:
(1) The first group covers the material damage that was
caused. One part of these relates directly to the criminal
offence, for example if, during an attack of grievous bod-
ily harm the victim’s mobile phone gets damaged beyond
repair, and the other part represents costs that arise as a
consequence of the offence, such as hospital treatment, or
in extreme cases, the costs of a funeral;
(2) The second group includes costs which do not rep-
resent direct expenses, but which should definitely be
taken into con sideration as social costs. The loss of a life,
for example, in addition to the individual tragedy, may
also represent a loss for society as it has been forced to
part with a productive member of society (depending on
the age of the victim). Time taken off work is also con-
sidered as a loss and a cost from society's point of view.
For every crime that involves a personal injury, therefore,
we need to calculate the cost which depends on the sever-
ity of the injury. Medical treatment and care, as well as
the costs of social insurance are taken into consideration
in the previous point as well as the extra expenses of the
victims, medicines, costs of doctors and the value of any
required therapy.
Unregistered damage to assets arises in relation to
many criminal offences. The largest group is for traffic
offences. Calculating the material damage incurred in this
group of offences is made harder by the fact that some of
the damage is compensated for to the injured persons by
insurance companies under compulsory and voluntary
insurance policies, while the perpetrator of the offence
cannot expect any compensation unless they have taken
out separate insurance specifically for this. The basis of
our investigation, however, does not focus on estimating
the damage caused to individuals, but is rather designed
to assess the damage caused to society as a whole as a
result of crime. From this point of view, the role of insur-
ers is negligible, for although the insurers recompense
individuals for damage they have suffered, this compen-
sation is paid out from the insurance premiums that are
paid in. This means that the sums spent on paying com-
pensation for damage cannot be used in another produc-
tive way, but have to be used for restoring assets which
were created previously. In this sense the part of the
common wealth that has to be spent to compensate for
damage that has been caused shows up as a cost to soci-
ety. This approach leads to unity when calculating dam-
age, because no distinction need be made between dam-
age caused to insured assets and that caused to uninsured
assets. From the point of view of society all damage
represents a loss, regardless of whether the person di-
rectly suffering the loss receives compensation or not.
Insurance companies do play an important role in the
evaluation of damage because their compensation statis-
tics form the starting point for determining the actual
amount of damage caused by criminal offences. The Uni-
fied Criminal Statistics does not contain any data about
the damage caused by traffic offences, so we need to look
for other accident statistics. The problem lies in the fact
that the insurance companies carry out procedures for
compensation claims of many different degrees of seri-
ousness, but they do not record whether a compensation

The Social Costs of Crime and Crime Control79
claim is made in relation to a criminal offence or not. So
we had a set of statistics about compensation claims
which contained a mix of cases that were crime related
and other cases that were not, and we had a set of crime
statistics which also contained traffic offences. To fit the
two sets of statistics together we used a method of pro-
portional weighting. To do this we took out of the total of
traffic offences (17,664) those offences where there was a
high probability that neither personal injury nor material
damage was caused. Mainly offences committed for
drunk driving and driving in a state of inebriation were
listed here (10,363 + 439), provided there were no other
qualifying circumstances. The remaining offences were
ranked according to their seriousness. We created five
groups: the first group included offences where it was
presumed only slight damage would have been caused,
and the highest group featured those offences where there
must have been a high degree of damage, typically acci-
dents resulting in serious injury or fatal accidents. We
also conducted a similar grouping of claims for insurance
compensation, and starting from the average amount of
damage we formed two below-average and two above-
average categories, adjusted to the amount of payouts for
compensation.
As a final step we aligned both of the proportioned and
weighted scales, and multiplied the number of cases by
the value of damage to reach the final result, the amount
of damage caused to assets by traffic offences.(Table 3)
According to our estimates, therefore, 17,226,250 $ of
material damaged was caused directly by traffic accidents.
This value for damage, however, still needs to be sup-
plemented by taking the damage caused by personal inju-
ries into consideration, but we shall return to this in the
calculati on of our next index.
When determining the social costs of crimes involving
personal injuries, the first step was to summarise the di-
rect material damage. Part of this damage is directly re-
lated to the criminal offence and was caused at the same
time as the offence was committed, such as damage
caused to clothing or other assets. Targeted investigations
would be required to estimate these costs, but there are
Table 3. Material damage caused by traffic accidents.
value of
damage ($) number of cases damage caused ($)
slight damage 250 2930 732,500
moderate damage 1250 1027 1,283,750
average damage 2500 865 2,162,500
heavy damage 5000 1247 6,235,000
significant damage 12,500 545 6,812,500
Total 6614 17,226,250
unfortunately no domestic research data available. Judg-
ing from the experiences of studies carried out abroad,
these direct costs represent only a small pr oportion of the
total losses suffered by the victims.
Depending on the type of crime, they would represent
between 50 and 1% of the total costs. The other—and
typically larger—part of the costs consists of those for
nursing, rehabilitation and treatment as a result of physi-
cal or mental injury. These costs will be taken into ac-
count under point 4. In the present category we only
count the costs of care directly incurring: e.g. in case that
someone breaks his/her leg in connection with a crime,
the direct cure costs of the injury (plastering, X-ray, etc.)
will be counted h ere, while the time taken off work, costs
related to the period of illness and losses stemming from
the decrease of life quality will be taken into account
under poi nt 4.
There are two possible ways of estimating direct costs:
1) One is by examining the documents made in relation
to the crimes. Much information can be gained from the
report or from the description of the case, but regrettably
these documents cannot be regarded as having a unified
aspect. It is often very circumstantial which kind of in-
formation makes it into the documents.
2) The second way is to ask the victims in person.
There has not, however, been any targeted survey of this
kind in Hungary yet. There is, though, a partial area
where information about the assessed damage is available:
the clients of the Justice Authority Victim Support Ser-
vice. The Victim Support Service acted in the interests of
19,317 victims in 2009, and provided assistance (infor-
mation, victim support services, state compensation) in
22,951 instances.
This number means that the Victim Support Service
offered some kind of help to a national average of 9% of
the known victims who are natural persons.4 In the year
in question the Service determined an average sum of
compensation of 1050 $. The sum of compensation rarely
covers the entire value of the damage caused, so if we
calculate on the basis of the average sum of compensa-
tion, we will certainly not be overestimating the actual
amount of losses suffered. The Unified Criminal Statis-
tics data records, for the period in question, a total of
32,760 crimes against the person involving aggression or
disorderly behaviour. Using the average sum of compen-
sation for calculating the results, the total figure for
damage is 34,406,517 $.
The amount of damage caused by crimes for which the
Unified Criminal Statistics does not record such a value
is as follows:
4Annual report 2009 on the activities of the Victim Support Service.

The Social Costs of Crime and Crime Control
80
Dunr = 3.4 + 6.9 = 51.5 million $5
C3 = Dr * Corr – Hrp + Dunr
C3 = 158.8 + 10.3 = 845.5 million $
4) The social costs resulting from personal injuries
C4 = Dr * Corr – Hrp + Dunr + Pinj (4)
where:
C4 = the amount of damage caused by crime;
Dr = the amount of damage caused by crimes for
which the va lue of the damage is record ed in the Unified
Criminal Statistics;
Corr = correction factor, to adjust the value shown in
the Unified Criminal Statistics to the actual amount of
damage caused;
Hrp = the value of compensation registered in the Uni-
fied Criminal Statistics;
Dunr = unregistered damage (the amount of material
damage caused by crimes for which there is no recorded
damage in the Unified Criminal Statistics);
Pinj = social costs arising from injury caused by crimes
resulting in personal injury, DALY * twelve times
monthly average earnings.
In estimating the damage resulting from personal inju-
ries we included all crimes which could involve physical
injury, health damage, or in the most serious cases, death.
In the estimate of the social costs the DALY index is
used, which is formed from two other indices: by adding
together the number of years of life lost (YLL) as a result
of premature death, and the number of years lost due to
disability (YLD). The use of both indices originates in
health economics, but it is beginning to spread to surveys
into losses caused by injuries related to criminal offences.
One DALY can be regarded as one year lost from a
“healthy” life.
DALY = YLL + YLD
YLL = N * L
where:
N = number of fatalities;
L = standard life expectancy in years in the year of
death, taking the gender of the deceased into considera-
tion, minus the age of the decease.
We have amended the method of calculating the origi-
nal index from the aspect of accounting for social damage
by counting not the full life expectancy, but only the ac-
tive working life. Although the missing part of a full life
expectancy is also a loss from a social aspect, only the
active working lifespan can be given a monetary value as
regards calculating the costs. The remaining part of the
life expectancy is counted as a significant moral loss. In
our study, for victims who died as a result of a crime we
took the difference between the actual age and the re-
tirement age as the basis, and multiplied the number of
years by average annual earnings.
Identifying the number of victims who died as a result
of a crime is, however, no easy task. In cases where there
might have been more than one victim, we are unable to
form a precise picture because of deficiencies in crime
statistics. The Unified Criminal Statistics is still unab le to
provide data on all victims. The former ERÜBS system
always limited the data it recorded to those of a single
victim. This means that in cases of multiple murder or
manslaughter, only the data for one of the victims ap-
peared in the statistics. To calculate the YLL index,
however, it would be necessary to have data about every
individual victim. To get a precise figure would also re-
quire knowing the age and gender of the deceased. To
overcome this problem, in cases where it was possible to
determine the precise number of victims, the data for the
unrecorded victims was generated and treated as the av-
erage of the other cases. A clear loss of data can be
avoided in this way, and by using average data, the dis-
crepancy when dealing with large numbers is not too
significant.
The next problem was posed by the stage at which the
crimes were carried out. Where possible, we only con-
sidered completed offences, and did not include at-
tempted or planned offences in our calculations. Out of a
total of 38,375 offences involving personal injury, 3322
cases were ignored because the offence was only at-
tempted.
The resulting ca lc ul at i on:
YLL = 735 (the number of fatalities taken in to ac-
count) * L (the number of “productive” years of remain-
ing until the retirement age) 27.75
YLL = 20,883.75
The social costs of fatalities (YLL * average monthly
earnings) 20,889 * 1019 [7] * 12 = 255,440,423 $
Calculating the social costs of personal inju ries:
YLD = I * DW * L
5We did not calculate with the damage caused for a few groups of crime
p
artly because they were committed in negligible numbers and woul
d
have no substantive influence on the results, and partly because no
valid information was available about the possible amount of damage,
for example, crimes against public trust, against military obligations
and against the course of justice. There is one group of crimes which
should appear in the summary, encompassing crimes against the integ-
rity of public life. Significant damage which may influence the results
may arise from these crimes, but there is unfortunately no reliable
source of data for uncovering this type of damage.
where:
I = the number of incidents taken into account;
DW = the degree of disability, weighted with the de-
gree of health damage (The weighting figure is between 0
and 1. 0 refers to complete health, and 1 refers to damage
with fatal consequences.);
L = the average duration of the health damage, until

The Social Costs of Crime and Crime Control81
recovery or death, or in our case the age of retirement
(years).
The YLD index enables us to consider both temporary
injuries and permanent health damage. If the health
damage did not last longer than one year then the value of
“L” will be less than 1, if it lasted several years then it
will be the number of years, and in cases of permanent
disability then the figure for L is given by the difference
between the age of the victim at the time the injury was
received and the age of retirement. We established five
categories for the degree of personal injury:
(1) Slight;
(2) Minor injuries;
(3) Major injuries;
(4) Serious injuries;
(5) Permanent damage.
“Slight” injuries included injuries that took less than 8
days to recover, that required minimal medical interven-
tion generally consisting of writing a medical report and
letting the patient go home. “Minor injuries” covered
injuries requiring more than 8 days to recover without
needing hospital treatment. “Major injuries” are those
which required hospital treatment as part of the cure, but
no significant surgical intervention. The “serious inju-
ries” group consisted of cases needing major surgery
and/or a longer stay in hospital. The costs of the medical
treatment are not included here, but in the calculation of
the “Tsz” index. For the YLD index only the time taken
to recover was included in the calculation. Time lost fro m
work and from living a full life counts. For cases result-
ing in permanent damage, as in the case of fatalities, we
calculated on the basis of the difference between the age
at the time the offence was committed and the age of re-
tirement. Using the DW weighting adjusted for the de-
gree of the injury and the time taken to make a recovery
produces the following result:
YLD = 16.513 * 0.483 * 0.836 = 6.668
The social cost of crimes resulting in a non-fatal per-
sonal injury = YLD * average monthly earnings * 12 =
81,559,908 $
The combined total social cost of crimes involving
personal injury (from slight injuries to fatal injuries)
Cost = DALY * average monthly earnings * 12
Cost = (YLL + YLD) * average monthly earnings * 12
Cost = (20.889+ 6.668) * 1019 * 12 = 337 million $
Pinj = 67.4
C4 = Dr * Corr – Hrp + Dunr + Pinj
C4 = 1182.5 million $
5) The value of damage for dark figure of crime in case
of natural persons
C5 = Dr * Co rr – Hrp + Dunr + Pinj + Lnp (5)
where:
Dr = the amount of damage caused by crimes for
which the va lue of the damage is record ed in the Unified
Criminal Statistics;
Corr = correction factor, to adjust the value shown in
the Unified Criminal Statistics to the actual amount of
damage caused;
Hrp = the value of compensation registered in the Uni-
fied Criminal Statistics;
Dunr = unregistered damage (the amount of material
damage caused by crimes for which there is no recorded
damage in the Unified Criminal Statistics);
Pinj = social costs arising from injury caused by crimes
resulting in personal injury, DALY * twelve times
monthly average earnings;
Lnp = damage caused by latent crimes to victims who
are natural persons.
Data for estimating dark figure of crime are regularly
provided by the results of victimological studies. There is,
however, a problem in the f act that 40% of known crimes
are not committed against natural persons. It is not possi-
ble to draw conclusions about crimes concerning compa-
nies, foundations and other organisations from population
survey data.
This makes it reasonable to separate the two areas.
Trends in crimes concerning the population may be esti-
mated from victimological studies, although it is not ideal
to make a numerical comparison of just a couple of years,
so data covering several years would be needed to for-
mulate a proper picture of the tendencies. A better
grounded version of latent crime estimates originates
from a comparison of the data surveyed within the study.
During the studies, data are also collected on whether the
victim reported the crime or not. The ratio between re-
ported and unreported crimes forms the basis for esti-
mating dark figure of crime. Unfortunately, the only vic-
timological study of this type in Hungary, using a na-
tional sample and providing usable data, was carried out
in 2003. The actual ratio may have changed somewhat in
the intervening period, but there is no indication that any
significant change has taken place in the scale of latency.
Lnp value was calculated by multiplying the value of loss
incurring on the side of natural persons injured by re-
ported crimes with the ration of latency.6
Lnp = 416,794,169 $
C5 = Dr * Corr – Hrp + Dunr + Pinj + Lnp
C5 = 236.5 + 83.3 = 1599.5 million $
6This may lead to overestimation because of the greater proportion o
f
less severe crime remaining unreported, but this may be counterbal-
anced by the fact that the estimation of dark figures is likely to show a
lower value than the real one.

The Social Costs of Crime and Crime Control
82
6) The value of damage for dark figure of crimes in the
case of non-natural persons
C6 = Dr * Co rr – Hrp + Dunr + Pinj + Lnp + Lle (6)
where:
Dr = the amount of damage caused by crimes for
which the va lue of the damage is record ed in the Unified
Criminal Statistics;
Corr = correction factor, to adjust the value shown in
the Unified Criminal Statistics to the actual amount of
damage caused;
Hrp = the value of compensation registered in the Uni-
fied Criminal Statistics;
Dunr = unregistered damage (the amount of material
damage caused by crimes for which there is no recorded
damage in the Unified Criminal Statistics);
Pinj = social costs arising from injury caused by crimes
resulting in personal injury, DALY * twelve times
monthly average earnings;
Lnp = damage caused by latent crimes to victims who
are natural persons;
Lle = damage caused by latent crimes against legal en-
tities.
According to the Unified Criminal Statistics data, the
victims of around 40% of total recorded crime are com-
panies and organisations with legal entity status. The
majority of these are economic companies or corpora-
tions. There is very little reliable data from this area
available to us for analysis. Unfortunately no comprehen-
sive Hungarian research has yet been carried out in this
area. Even abroad, only scattered attempts have been
made in this regard [8], and even in countries with dec-
ades of experience in traditional victimological research
(USA, England) there is no regular research concerning
crimes against companies and institutions . The only body
that conducts regular studies into this field is Pricewater-
houseCoopers. They prepared their fifth Global Eco-
nomic Crime Survey in 2009, which examines trends in
economic crime. The analysis is international: “In the
course of our surv ey we interviewed senior executives of
more than 3000 companies in 54 countries, including the
senior executives of 53 major corporations in Hungary.”
[9] Unfortunately participation in Hungary is declining,
as 77 Hungarian corporations took part in the 2007 sur-
vey. The main shortcoming of the analysis, besides the
small number of elements, is the fact that the data pro-
viders only consulted companies who are clients of
PricewaterhouseCoopers, so the research may not be
considered fully representative. In spite of its shortcom-
ings, the research allows us a valuable insight into the
rather closed area of business life. “In Hungary 63% of
the companies that have fallen victim to economic crime
reported that in the last 12 months, their losses as a result
of economic crime have increased. 13% of the companies
suffered a loss in excess of 0.9 million $ from cases of
abuse, and 25% suffered a loss above 0.45 million $.”
From the point of view of our analysis, however, the
greatest significance lies with latent crime. During the
survey the company executives were asked about the
kinds of measures they have implemented to uncover
crime. There was a fundamental difference in the fre-
quency and composition of the sanctions applied, ac-
cording to whethe r the crime was committed by an in ter-
nal employee or by a third party.
In the case of third parties not employed by the com-
pany, a total of 21% resulted in legal sanctions, and only
a part of these were criminal complaints. Where internal
employees committed the crime, almost half of the inci-
dents (48%) resulted in legal measures. Overall, then, the
injured companies sought legal recourse in less than half
of the cases of economic crime, and the proportion of
crimes reported cannot therefore be more than 50%.
There was no relationship between the amount of damage
caused and whether a complaint was made or not. It is
not the value of the damage, but other factors that deter-
mine whether the affected company takes legal measures
or not. If we proj ect the result we have obtain ed onto the
total of the crimes committed ag ainst non-natural persons,
and assume that there is no significant difference in the
amount of damage concerning reported and unreported
cases, then the result is that the extents of the damage
caused in discovered and in latent cases correspond to
each other.
The value Lle in the equation therefore amounts to the
double of the damage caused by registered “corporate”
crimes.
Lle = 573 million $
C6 = Dr * Corr – Hrp + Dunr + Pinj + Lnp + Lle
C6 = 1,599.5+ 573 = 2,172.5 million $
3.2. The Social Cost of Crime
The above equations show the extent of damage suffered
by the victims. If we wish to determine the social cost of
crime, then we must deduct from this amount the profit
gained by the perpetrators, since from an accounting
point of view it does not matter who is in possession of
the assets of a society. In this case we have to supplement
the equation with a profit member “H”, symbolising the
profit enjoyed by the perpetrators.
The “final” equation showing the social costs of crime
is as follows:
C7 = Dr * Corr – Hrp + Dunr + Pinj + Lnp + Lle – Pp (6)
where:
C7 = the amount of damage caused by crime;

The Social Costs of Crime and Crime Control83
Dr = the amount of damage caused by crimes for
which the va lue of the damage is record ed in the Unified
Criminal Statistics;
Corr = correction factor, to adjust the value shown in
the Unified Criminal Statistics to the actual amount of
damage caused;
Hrp = the value of compensation registered in the Uni-
fied Criminal Statistics;
Dunr = the amount of material damage caused by
crimes for which there is no recorded damage in the Uni-
fied Criminal Statistics;
Pinj = social costs arising from injury caused by crimes
resulting in personal injury, DALY * twelve times
monthly average earnings;
Lnp = damage caused by latent crimes affecting victims
who are natural persons;
Lle = damage caused by latent crimes affecting victims
that are legal entities;
Pp = the profit to the perpetrators deriving from crime.
This also denotes the transactional co sts of crime. This
is the cost to society when certain members choose illega l
methods of acquiring wealth over the legal path.
The calculation of the index is only possib le in th e case
of crimes where the perpetrator has derived an actual
profit from the crime. The damage caused by, for exam-
ple, crimes against the person or traffic offences, were
not taken into consideration. In practice it is relevant to
calculate with the profit to the perpetrator in the case of
economic crimes and crimes against property. According
to the official crime statistics, 6% - 8% of the damage
caused by these crimes is recovered. The perpetrators
therefore definitely do not profit from this portion. Ac-
cording to police experts who work in this area, th e com-
bined profit to the actual perpetrator (and to the traffick-
ers/fence organisations, if involved) does not exceed 60%
of the value of the crime.
Pp = 996.5 million $
C7 = Dr * Corr – Hrp + Dunr + Pinj + Lnp + Lle – Pp
C7 = 2172.5 – 996.5 = 3167 million $
The amount of damage deriving as a result of crime in
Hungary for the year 2009 totals 2172.5 million $. A por-
tion of the damage, 996.5 million $, was retained by so-
ciety, such that it remained as the profit of the perpetra-
tors.
In summary, the total amount of damage suffered by
the country as a result of crime in the year under investi-
gation was 1176 million $.
4. The Costs of the Criminal Justice Service
We consider it important to remark in advance that the
following figures and calculations are the results of the
first Hungarian examination into the “price of crime”.
There is as yet no uniform practice in the methodology
the different bodies use for analysis, mainly due to the
fact that the same types of data are not always available
at the different institutions. In order to obtain a compara-
ble set of results, it was often necessary, therefore, to use
different methods to analyse the statistical data made
available to us. We frequently encountered replies to the
effect that a particular body does not collect the data we
requested, or not broken down in the way we would need
it to be, but more than once we were not even sent the
figures we requested.
At first sight it seems an easy task to calculate the costs
of the criminal justice service, since pursu ant to the regu-
lations governing the publication of data of public inter-
est, every state body is obliged to maintain a register of
their annual budget and headcount.
Every state body also registers internal statistics on the
cases they deal with, withou t a legal requiremen t to do so,
for their own information and for measuring changes in
the amount, severity and internal composition. In practice
the situation is much more complicated. For instance,
among state-financed expenses not only the typical play-
ers connected to the criminal justice service are included.
Nor, with certain state bodies, must we consider all items
in their budget, as these institutions often perform tasks
which are not exclusively related to the criminal justice
service.
4.1. The Police
We examined the larger part of the data on the police
force based on the budget report prepared with the assis-
tance of the Hungarian State Treasury (MÁK) program,
which was concluded on 18 May, 2010. We sorted sepa-
rately the total headcount of employees belonging to
Function Group7, broken down for the police headquar-
ters of each county, and the number of vacant and
long-term vacant positions. Following this, with the help
of the charts sent by the National Police Headquarters
containing the income and expenditure of professional
tasks of the police, we collected the total in come and ex-
penditure for individual counties for the year 2009. We
calculated figures for the entire police organisation of
35,401 people and expenditure of 1160.5 million $.
Based on the calculations, the national average social cost
of every senior or ranking police officer with basic, sec-
ondary or tertiary level education, dealing with crime
prevention, law enforcement and the investigation of
7According to the explanatory notes of the MÁK program: the head-
count performing strictly professional duties (as defined by the statutes
and the organisational and operational regulations) that are closely
linked to the professional profile(s) of the institution. In other words, all
staff performing duties related to crime prevention, crime investigation
and law enforcement belong to the headcount for Function Group I.

The Social Costs of Crime and Crime Control
84
crime, was 32,500 $ for the year 2009.
4.2. Public Prosecutor
Based on the latest data (November 2010 ) the total head-
count for the public prosecution organisation is 1708
people, of whom 1348 work in the criminal division.
78.92% of the employees currently work in the criminal
division, and if we assume this ratio to be constant, then
as of 31 December 2009, out of 1660 employees of the
public prosecution service, 1310 were working in the
criminal division of the organisation. 83.8% of the cases
brought in 2 009 were processed within the criminal divi-
sion. The criminal division, however, should also be sup-
plemented by cases of minor infringements, which are
also considered at other state bodies. Providing a figure
for these is not easy, however, neither as regards esti-
mating the number of cases, nor the number prosecutors
involved. In the public prosecution organisation prosecu-
tors supervising legal compliance deal with minor in-
fringements at the first level, while public administrative
departments conduct procedures within the supervisory
scope. 13.95% of the total staff of the public prosecution
service work in the public administrative departments,
although they do not deal exclusively with tasks related
to infringements, nor even with tasks that belong to the
criminal division. Since the available statistics only give
a combined figure for the number of public administra-
tion prosecutors, and it is not possible to draw assump-
tions about this figure from other accounts either, we
calculated using an artificial ratio (for the sake of sim-
plicity, and in the absence of knowledge that would allow
a more precise approach, the ratio is 50%) when we de-
termined that 7% of the total headcount are prosecutors
dealing with duties supervising legal compliance. Since
the prosecutors dealing with the supervision of legal
compliance do not deal exclusively with infringements,
we once more took a 50% ratio (simply for want of better
data) of the prosecutors supervising legal compliance,
ending up with a figure of half the prosecutors supervis-
ing legal compliance (3.5% of the total) dealing with in-
fringements at the first and second level.8
For the year in question, the ratio of criminal and leg al
compliance supervision cases in the organisation was
90.8%, while the ratio of employees working on such
cases stood at around 82.3% of the total public prosecu-
tion staff. Although the data are only implied, it can be
seen clearly that the combined workload of criminal and
legal compliance supervision prosecutors is quite high (as
shown by the comparison of the ratios of cases and
prosecutors) (Table 4).
Expenditure for the year in question is contained in the
central budget report prepared with the assistance of the
MÁK program. According to the report total expenditure
of the public prosecution in 2009 was 149,869,595 $.
While the major part of the activities of the organisation
concern the investigation and legal enforcement of crimes
that have been committed, we would not obtain a precise
figure unless we deduct from the total expenditure the
amount of civil, public administration and other cases
(and expenses). We therefore reckoned with a ratio of
90.8% for the expenditure on the combined crime and
infringement cases. (Table 5) We calculated using the
ratio of case and not with the ratio of employees, because
the structure of the organisation and the number of
prosecutors conducting procedures in individual divisions
are adjusted according to the number of cases brought.
4.3. The Courts
When calculating the costs of the courts we separated
criminal cases from civil cases, as we are only concerned
with the costs arising in connection with crime. The fig-
ures for the year 2009 on criminal trial judges were pro-
vided to us by the Office of the National Council of Jus-
tice. The following step for us was to calculate the per-
centage of the total expenditure that was “used up” by
judges in the criminal courts. Among the main data on
court cases for the year 2009, all local, county, High
Court and Supreme Court criminal, infringement and
military trials and non-trial cases were collected. For all
practical purposes we worked with the data of cases
Table 4. Ratio of cases and prosecutors.
area number of
cases ratio of cases number of prosecutors ratio of prosecutors
Criminal law 948,00683.8 1310 78.9
Legal
compliance
supervision 157,93913.9 113 6.8
of which,
infringements 78,969 7.0 56 3.4
Combined
criminal and
infringements 1,026,97590.0 1366 82.3
Total 1,131,436 1660
Table 5. Costs of crime and infringement cases.
Expenditure
(million $) Ratio (%)
According to the budget report 149.5 100
Combined crime and infringement cases 136 90.8
8The data is further distorted by the fact that in locations with fewe
r
cases, a single public prosecutor deals with both legal compliance and
p
ublic administration cases. This makes it impossible to give a precise
figure on the ratio of employees dealing with infringements.

The Social Costs of Crime and Crime Control85
brought in 2009, and not the cases that were finalised or
still in progress in 2009. With the help of the breakdown
of criminal cases, we then calculated the amount of ex-
penditure, out of total expenditure, which related to
criminal trials.
If we add together the social costs of criminal judges
working at all the vertical bodies and divide it by the
number of judges working in criminal cases (1658) then
we arrive at a figure of 39,500 $ per criminal judge for
the year 2009. The figure is, however, distorted, if we
exclude the expenditure of the Office of the National
Council of Justice (OITH). The total costs of the Office
cannot be included, as they equally cover the amounts
spent on civil and criminal cases. We are unable to say
with precision what percentage of the work of the Office
consists of tasks relating to criminal cases, so we calcu-
lated the expenses using the national average of the dis-
tribution of criminal and civil cases. Thus, we took
23.8% of the total expenditure of the OITH for the year
2009, and arrived at an estimate of the amounts spent
during the operations of the criminal division of the Of-
fice in 2009.
Calculations show that in the base year, the mainte-
nance and operation of the court system represented an
expenditure of 375 million $. In this year, 1658 judges
presided over criminal cases, and the domestic court sys-
tem spent 66 million $ on fulfilling tasks related to
criminal jurisprudence.
4.4. Prison Service
Total expenditure by the prison authorities in 2009 was
226.5 million $. This was made up of the following items:
Of the total annual expenditure (which for the year 2009
was much higher in practice than the original 182 million
$ that was planned in line with regulations, especially as
it is even higher than this figure) there was (also) an
amount of more than 7.5 million $ in overdue debt for the
year 2009 that was regrouped to 2010. In addition the
revenues of economic companies also have to be indi-
cated in the expenses, as they “generated a loss” of
44,000 $, which the organisation had to save from their
own budget. On 31 December 2009 there were 7786 peo-
ple employed by the organisation (94.71% of the total
headcount permitted by the budget). By dividing the
amount of annual expenditure by the total headcount we
obtain a figure for state expenditure per employee of the
prison services of 29,000 $ for the year 2009.
4.5. Office of Justice
The Office of Justice (IH) began operations on 1 January
2006, and was formed by bringing together the former
National Office of Probation and Legal Aid Services
(PJSZ) and the Central Office for Compensation (KKI).
According to its Rules of Organisation and Operation, the
IH is a central office that comes under the professional
direction of the Minister for Justice and Law Enforce-
ment (now superseded by the Minister for Public Ad-
ministration and Justice). Starting 1 January 2007 the IH
has six areas of responsibility. These are (1) probation
officer duties, (2) mediation duties in criminal cases, (3)
legal aid activities, (4) victim support services, (5) com-
pensation and restitution activities, and (6) enforcement
of the law on lobbying activities. Of these, we will ex-
clude the last two from our further calculations.
In summary, with 553 employees working in criminal
cases, the Office of Justice spent 18.5 million $ in 2009
on assisting victims, reducing the number of crimes
committed, probation services, mediation and providing
legal aid. Thus, the state spent 34,000 $ per officer of the
Office of Justice dealing with criminal cases in the period
in question.
4.6. Ministry of Public Administration and
Justice
It is not by chance that the legislative b ody ap p ears last in
the analysis of institutions dealing with criminal justice.
In 2010 it was impossible to obtain data on the economic
events of the year 2009, so it was here that we had to
operate with the greatest proportion of estimated figures.
The Ministry of Public Administration and Justice
(IRM, witch is the former acronym of the Ministry, used
because the data in 2009 still refers to this body) spent a
total of 32.5 million $ according to the official report
prepared for the State Treasury9. As with the majority of
state bodies analysed so far, however, the entire amount
cannot be used here either, because the Ministry does not
only deal with criminal legislation. We were given no
information at all on which departments or how many
employees dealt with the tasks that concerned this study,
so we used their Rules of Organisation and Operation for
the year 2009 as our starting point. In the first round we
were able to establish that the Ministry functioned with
69 departments in the year in question. The next task was
to investigate tho se units which definitely do not perform
tasks related to criminal law. From the titles of the or-
ganisational units and the Rules of Organisation and Op-
eration we identified 36 departments of the IRM that car-
ried out tasks related to criminal law. We do not know
how many people worked in the different departments, so
we attempted to reach a proportionate figure from the
data available to us.
The total headcoun t of the Ministry was 533 people as
9The data was provided to us by the Budgetary and Financial Depart-
ment of the IRM (Ministry of Justice and Law Enforcement).

The Social Costs of Crime and Crime Control
86
of 31 December 2009. Of these, 247 performed a crimi-
nal law codification activity, which means that 46.3% of
the employees worked in this area. In the absence of
more precise data we have to assume that 46.3% of the
total expenditure was also used during the work of these
247 employees, in other words these employees are re-
sponsible for 15 million $ of the total of 32.5 million $.
Comparing the calculated sum (15 million $) with the
estimated figure (247 people) we conclude that the state
spent 61,000 $ per em ployee dealing with criminal justice
activities in the year 2009.
4.7. Civil Guard
In Hungary “the National Assembly recognises the ac-
tivities of the Civil Guard formed as a result of self-
organisation among the citizens, and appreciates the
support given by the Civil Guard to local self-govern-
ments and to local and regional law enforcement organi-
sations in carrying out public safety tasks”, according to
the introduction to Act LII (2006) on the Civil Guard.
The amount spent by the state and by local governments
on the Civil Guard in 2009 was 3.5 million $. The state
and the local governments paid this amount to the Civil
Guard Alliance by virtue of the fulfilment and partial
assignment of crime prevention tasks. In such a way, the
organisation fulfilled state duties from state finances.
5. A Comparison of the Costs of Crime and
State Expenditure
The results of the calculation of expenditure cannot be
summarised in a single figure. The calculable outcome
varies depending on the approach and on the interpreta-
tion of the social impacts of crime. If we take just the
social costs, then the damage caused by crime has a value
of 2172.5 million $, and if we deduct the profit earned by
the perpetrators of the crimes then we reach a figure of
1176 million $. To this must be added the amounts spent
by the state on fighting crime, on the course of ju stice, on
detention and on crime prevention. Total state expenses
of this kind in 2009 were 1631 million $.
Crime caused damage, or required state expenditure,
worth 2807 million $ in the single year 2009.
At the same time, from the comparison of the costs of
crime and the amount spent by the state, we can establish
that state expenditure in dealing with crime (163 million
$) is around 512.8 million $ higher than the amount of
damage caused by crime (1175 million $). The amount of
expenditure society devotes to law enforcement and
crime prevention cannot be analysed simply by compari-
son with the costs of crime, as it is not possible to dem-
onstrate the effect of crime which is not committed be-
cause of the threat of punishment and because of crime
prevention measures. It is in the fundamental interests of
the state and all members of society to act against those
members of society who follow the path of crime, even if
the actions are extremely costly. This basic “zugzwang”
in practice means that there is no upper limit to spending
on crime. The amounts spent by individual states on
fighting and punishing crime are not decided on the basis
of cost-efficiency analysis, but on social and political
factors. This does not in any way mean, however, that
efficiency can be ignored in this area. The calculations
we have made of the costs could represent the first step
on the path leading to better awareness of the importance
of the role of crime in society. The calculations and
methods used are, even in our opinion, open to debate,
and should be debated too, but we are convinced that
there is a place in this field for calculating and deliberat-
ing economic costs.
6. Acknowledgements
We would like to acknowledge for the support of the
European Union Crime Prevention Network (EUCPN).
We would also like to thank Mr Richard Dubourg (Eco-
nomic Adviser, Home Office, UK) and Professor Roger
Bowles (University of York, Centre for Criminal Justice
Economics and Psychology, UK) for their helpful discus-
sion at the beginning and in the course of the project.
REFERENCES
[1] B. C. Welsh and D. P. Farrington, “Monetary Costs and
Benefits of Crime Prevention Programs Crime & Justice,”
The University of Chicago, Vol. 27, 2000, pp. 305-361.
[2] H. G. Demmert, “Crime a nd Crime Control: Wha t Are the
Social Costs?” Stanford University, Hoover Institution,
Center for Econometric Studies of the Justice System,
1979, pp. 2-5.
[3] M. A. Cohen and R. Bowles, “Estimating Costs of
Crime,” In: A. R. Piquero and D. Weisburd, Eds., Hand-
book of Quantitative Criminology, 2010, pp. 143-162.
doi:10.1007/978-0-387-77650-7_8
[4] M. A. Cohen, “The Costs of Crime and Justice,” Rout-
ledge, New York, 2005.
doi:10.4324/9780203313145
[5] K. Kerezsi and J. Kó, “Victimisation in Hungary in the
Light of the Victimology Survey of 2005,” In: Gy. Virág,
Ed., Criminological Studies, Vol. 47, OKRI, Budapest,
2010, pp. 113-130.
[6] F. Irk (Ed.), “Victims and Opinions,” Vol. 1-2, OKRI,
Budapest, 2004.
[7] Monthly Average Earnings, National Economic Average,
3.1.12.,” In: E. Nemeth, Ed., Statistical Yearbook of
Hungary 2009, KSH (Central Statistical Office), Budapest,
2010, p. 78.
[8] “Monitor Crime in the Business Sector. 2007,” Insight
into the Victimisation of Companies in the Netherlands
and Prevention Measures, Amsterdam, 2008.

The Social Costs of Crime and Crime Control
87
http://english.wodc.nl/onderzoeksdatabase/monitor-crimin
aliteit-bedrijfsleven-2007.aspx
[9] D. Wake, M. Fekete and L. Coles, “Economic Crime in a
Downturn. Global Economic Crime Survey. Hungary
Report. 2008,” Pricewaterhouse Coopers.
http://www.pwc.com/en_HU/hu/publications/assets/PwC-
GlobalEconomicCrimeSurvey-HU2009-HU-OK.pdf