Assessing the impact of economic and demographic change on property crime rates in Western Canada
Added on 2023-04-21
7 Pages6916 Words76 Views
Test PrepStatistics and Probability
|
|
|
52
© 2018 Author. Open Access. This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-nd/4.0/. For commercial re-use, please contact marketing@multi-med.com.
ORIGINAL RESEARCH
Assessing the impact of economic and
demographic change on property crime rates
in Western Canada
Stuart Wilson*
ABSTRACT
Western provinces have experienced tremendous change over the last few decades, with oil booms and busts, with large
international and interprovincial movement of workers and families, and with rising and declining property crime
rates. What are the links between these economic, demographic, and crime rate changes? I investigate these links for
Western Canada over the period from 1968 to 2015. Empirical results suggest that increases in household incomes and
alcohol sales per capita, and decreases in unemployment rates, all signs of improved economic prosperity, coincided
with decreases in rates of property crime. Increases in migration turnover (both inward and outward migration) put
upward pressure on rates of property crime. In addition, changes in police reporting methods and categorization have
had dramatic effects on official rates of property crime: the change from UCR1 to UCR2 reporting methodology caused
rates of property crime to rise by between 18 per cent and 30 per cent in 1998, and changes in police reporting methods in
2003 caused property crime rates to rise again, by between 5 per cent and 16 per cent in 2003, for the western provinces.
The recent rise in rates of property crime in the west is closely linked to the economic slowdown following the drop in
oil and resource prices, and should migration turnover rates remain high as people move seeking better opportunities,
property crime rates will remain high. Policymakers and criminal justice professionals may be advised and reminded
of the effect of these economic and demographic changes, as well as the effect of reporting changes, on official rates
property crime.
Key Words Crime rate trends; economic growth; demographic change; police reporting.
Journal of CSWB. 2018 Oct;3(2):52-58 www.journalcswb.ca
INTRODUCTION
Crime rates trended downwards from 2003 to 2013 in western
Canada, coinciding with an international resource boom.
In particular, Saskatchewan experienced sharp increases
in resource development, household income, and large in-
migration movements, from both other provinces and nations,
during a period termed the “Saskaboom”. Rapid economic
and demographic change can have various effects on crime.
Newcomers may have different behaviours and propensities
to commit certain types of crimes. Improved employment
outcomes of young males may reduce criminal activity.
Inequalities in employment, incomes, and housing may lead
to increased crime. Over the past few years with the decline
in oil prices and revenues, some western Canadian provinces
have experienced poor economic growth and increasing
unemployment rates. What influence have demographic and
economic changes had on crime rate patterns in the western
Canadian provinces?
The economic factors that have been found to influence
crime include changes in unemployment, in incomes, in
inflation, in inequality, and in alcohol consumption (Cook
& Zarkin, 1985; Raphael & Winter-Ebmer, 2001; Savoie, 2008;
Pernanen, Cousineau, Brochu et al., 2002; Bunge, Johnson,
& Balde, 2005, Andresen, 2013 among others). Research
investigating the demographic factors that influence crime
have tended to concentrate on the youth and migrant popu-
lations (Butcher & Piehl, 1998; Bunge et al., 2005; Kitchen,
2007; Stevens, Odynak, Brazil et al, 2011; Plecas, Evans, &
Dandurand, n.d.). For more extensive reviews of the influ-
ences on crime, readers are encouraged to refer to Levitt
(2004), Albertson & Fox (2012), Farrell, Tilley & Tseloni (2014),
Tonry (2014), and Wilson, Sagynbekov, Pardy et al. (2015).
Tonry (2014) wrote that overall, there may be a role for
economic and demographic forces in explaining property
crime rate patterns over time, but little support for their influ-
ence on violent crime rate patterns. Recent work by Wilson
(2017) supported this assessment by showing that economic
Correspondence to: Stuart Wilson, Department of Economics, University of Regina, 3737 Wascana Parkway, Regina, SK S4S 0A2, Canada.
E-mail: Stuart.Wilson@uregina.ca
Journal of
COMMUNITY SAFETY & WELL-BEING
© 2018 Author. Open Access. This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-nd/4.0/. For commercial re-use, please contact marketing@multi-med.com.
ORIGINAL RESEARCH
Assessing the impact of economic and
demographic change on property crime rates
in Western Canada
Stuart Wilson*
ABSTRACT
Western provinces have experienced tremendous change over the last few decades, with oil booms and busts, with large
international and interprovincial movement of workers and families, and with rising and declining property crime
rates. What are the links between these economic, demographic, and crime rate changes? I investigate these links for
Western Canada over the period from 1968 to 2015. Empirical results suggest that increases in household incomes and
alcohol sales per capita, and decreases in unemployment rates, all signs of improved economic prosperity, coincided
with decreases in rates of property crime. Increases in migration turnover (both inward and outward migration) put
upward pressure on rates of property crime. In addition, changes in police reporting methods and categorization have
had dramatic effects on official rates of property crime: the change from UCR1 to UCR2 reporting methodology caused
rates of property crime to rise by between 18 per cent and 30 per cent in 1998, and changes in police reporting methods in
2003 caused property crime rates to rise again, by between 5 per cent and 16 per cent in 2003, for the western provinces.
The recent rise in rates of property crime in the west is closely linked to the economic slowdown following the drop in
oil and resource prices, and should migration turnover rates remain high as people move seeking better opportunities,
property crime rates will remain high. Policymakers and criminal justice professionals may be advised and reminded
of the effect of these economic and demographic changes, as well as the effect of reporting changes, on official rates
property crime.
Key Words Crime rate trends; economic growth; demographic change; police reporting.
Journal of CSWB. 2018 Oct;3(2):52-58 www.journalcswb.ca
INTRODUCTION
Crime rates trended downwards from 2003 to 2013 in western
Canada, coinciding with an international resource boom.
In particular, Saskatchewan experienced sharp increases
in resource development, household income, and large in-
migration movements, from both other provinces and nations,
during a period termed the “Saskaboom”. Rapid economic
and demographic change can have various effects on crime.
Newcomers may have different behaviours and propensities
to commit certain types of crimes. Improved employment
outcomes of young males may reduce criminal activity.
Inequalities in employment, incomes, and housing may lead
to increased crime. Over the past few years with the decline
in oil prices and revenues, some western Canadian provinces
have experienced poor economic growth and increasing
unemployment rates. What influence have demographic and
economic changes had on crime rate patterns in the western
Canadian provinces?
The economic factors that have been found to influence
crime include changes in unemployment, in incomes, in
inflation, in inequality, and in alcohol consumption (Cook
& Zarkin, 1985; Raphael & Winter-Ebmer, 2001; Savoie, 2008;
Pernanen, Cousineau, Brochu et al., 2002; Bunge, Johnson,
& Balde, 2005, Andresen, 2013 among others). Research
investigating the demographic factors that influence crime
have tended to concentrate on the youth and migrant popu-
lations (Butcher & Piehl, 1998; Bunge et al., 2005; Kitchen,
2007; Stevens, Odynak, Brazil et al, 2011; Plecas, Evans, &
Dandurand, n.d.). For more extensive reviews of the influ-
ences on crime, readers are encouraged to refer to Levitt
(2004), Albertson & Fox (2012), Farrell, Tilley & Tseloni (2014),
Tonry (2014), and Wilson, Sagynbekov, Pardy et al. (2015).
Tonry (2014) wrote that overall, there may be a role for
economic and demographic forces in explaining property
crime rate patterns over time, but little support for their influ-
ence on violent crime rate patterns. Recent work by Wilson
(2017) supported this assessment by showing that economic
Correspondence to: Stuart Wilson, Department of Economics, University of Regina, 3737 Wascana Parkway, Regina, SK S4S 0A2, Canada.
E-mail: Stuart.Wilson@uregina.ca
Journal of
COMMUNITY SAFETY & WELL-BEING
IMPACTS OF ECONOMIC/DEMOGRAPHIC CHANGE ON PROPERTY CRIME RATES, Wilson53© 2018 Author.OpenAccess.ThisworkislicensedundertheCreativeCommonsAttribution-NonCommercial-NoDerivatives4.0 InternationalLicense.
T
o
v
i
e
w
a
c
o
p
y
o
f
t
h
i
s
l
i
c
e
n
s
e
,
v
i
s
i
t
h
t
t
p
:
/
/
c
r
e
a
t
i
v
e
c
o
m
m
o
n
s
.
o
r
g
/
l
i
c
e
n
s
e
s
/
b
y
-
n
c
-
n
d
/
4.0/. For commercial re-use, please contact marketing@multi-med.com.
and demographic change could help explain only minor
changes in rates of violent crime, but may help explain up to
40 per cent of the changes in rates of property crime in west-
ern Canadian provinces over the last five decades. This article
further focuses on assessing the economic and demographic
influences on property crime in western Canada.
METHODS
Police-reported crime data are available for Canadian prov-
inces starting in 1962. The rates of property crime from 1962
to 2015 are depicted in Figure 1. Bunge et al. (2005) conducted
time-series analyses on the influences on crime over 1962 to
2003 for Canada as a whole. This study serves as an extension
of their work, with a larger time period, and with a larger
dataset by pooling together data for the five provinces west
of Quebec in a panel analysis.1 This study also serves as an
extension of Wilson (2017) in three ways: (i) some explanatory
variables, including inflation, were dropped due to statistical
insignificance; (ii) the model allows for a singular focused
effect of migration turnover on property crime, motivated
by results which suggested that increases in interprovincial
and international migrant movement, both inward and out-
ward, had similarly positive impacts on property crime, and;
(iii) the analysis incorporates data from the more recent
period characterized by low resource prices.
The relationship between rates of property crime and
the economic and demographic variables for the panel of
five provinces (British Columbia, Alberta, Saskatchewan,
Manitoba, and Ontario) is investigated using a linear panel
model. The set of provincial economic and demographic
variables used in the analysis are described in Table 1. These
variables are real household disposable income per capita,
the unemployment rate, per capita alcohol sales, the youth
share of the population, and the migration turnover rate
(the sum of both in- and out-migration, as a proportion of
the total population). Provincial data for unemployment
rates are only available starting in 1967. These variables all
exhibit the property of nonstationarity, and therefore require
transformation using logarithmic first differences to allow for
estimation to overcome the problems of stochastic trends and
spurious regressions; as such, this study may not be directly
compared to that of Andresen (2013), which presented results
from a panel study with variables in their undifferenced
logarithmic forms.2 In addition, a set dummy variables was
used to account for additional province-specific trends and
shifts in rates of property crime. The effective time period of
the study is from 1968 to 2015.
RESULTS
The results from the panel regressions for the growth in the
rates of property crime over the study period are presented in
Table 2. The first regression model includes contemporaneous
and lagged independent variables, and a common constant
term but with no other dummy variables. The estimates
from that model are presented in Column 2, are identified as
Model 1, and indicate how much variation may be accounted
for solely by variation in the independent variables. The
estimated fit of this first model is relatively low, with an
overall R2 of 0.18, indicating that the explanatory variables
used in this regression may account for a small portion of
the variation in the growth of the rates of property crime
over the sample. Only one variable, the unemployment rate,
was statistically significant in the regression. The second
regression model includes province-specific constant terms,
and province-specific one-period and trend shift dummies
which allow for trend changes that occurred around 1990, for
an upward shift in 1998 with the movement to UCR2 report-
ing, and also for changes around 2003 (see notes to Table 2).
The estimates from this model are identified in Column 3
as Model 2. This second regression performed much better
in terms of the R 2 statistic. The inclusion of these province-
specific dummy variables allows the model to measure the
impact of economic and demographic changes on property
crime rates with better precision, and all of the economic and
demographic variables have statistically significant effects on
property crime rates, either contemporaneously, with lagged
effects, or both. This model indicates how important it is to
include these province- and time-specific effects to better
identify the relationship between economic and demographic
change, and changes in rates of property crime.
Influence of Economic and Demographic Change on
Rates of Property Crime
The estimates provided in Table 2 for Model 2, of the impact
of economic and demographic changes on rates of property
crime, will first be explained using a few illustrative exam -
ples. The coefficient estimates of -0.1214 and -0.2232 for Δyt
and Δyt-1 suggest that for every one percentage point increase
in the growth rate of real disposable household income per
capita, the growth rate of the rate of property crime decreased
by 0.12 percentage points, and the growth rate of the rate of
property crime decreased by 0.22 percentage points in the
following period, all else equal. The coefficient estimates of
1.1038 and 0.1322 for Δyouth-sharet and Δyouth-sharet-1 suggest
1 Panel dataset estimation provides several advantages over individual
time series estimation, including “more informative data, more variabil-
ity, less collinearity among the variables, more degrees of freedom and
more efficiency” (Baltagi, 2005, p. 5).
2 Nonstationary series, including the undifferenced series in this analy-
sis, exhibit stochastic trends which may lead such series to appear to
be related over the long run when they aren’t, resulting in a spurious
regression. For more information on issues regarding nonstationar-
ity and spurious regressions, please refer to Stock and Watson (2003,
pp. 460–462).
FIGURE 1 Rates of property crime, four western provinces and Ontario,
1962–2015.
-‐
1,000.00
2,000.00
3,000.00
4,000.00
5,000.00
6,000.00
7,000.00
8,000.00
9,000.00
10,000.00
1962
1965
1968
1971
1974
1977
1980
1983
1986
1989
1992
1995
1998
2001
2004
2007
2010
2013
Incidents per 100,000 residents
British Columbia
Alberta
Saskatchewan
Manitoba
Ontario
T
o
v
i
e
w
a
c
o
p
y
o
f
t
h
i
s
l
i
c
e
n
s
e
,
v
i
s
i
t
h
t
t
p
:
/
/
c
r
e
a
t
i
v
e
c
o
m
m
o
n
s
.
o
r
g
/
l
i
c
e
n
s
e
s
/
b
y
-
n
c
-
n
d
/
4.0/. For commercial re-use, please contact marketing@multi-med.com.
and demographic change could help explain only minor
changes in rates of violent crime, but may help explain up to
40 per cent of the changes in rates of property crime in west-
ern Canadian provinces over the last five decades. This article
further focuses on assessing the economic and demographic
influences on property crime in western Canada.
METHODS
Police-reported crime data are available for Canadian prov-
inces starting in 1962. The rates of property crime from 1962
to 2015 are depicted in Figure 1. Bunge et al. (2005) conducted
time-series analyses on the influences on crime over 1962 to
2003 for Canada as a whole. This study serves as an extension
of their work, with a larger time period, and with a larger
dataset by pooling together data for the five provinces west
of Quebec in a panel analysis.1 This study also serves as an
extension of Wilson (2017) in three ways: (i) some explanatory
variables, including inflation, were dropped due to statistical
insignificance; (ii) the model allows for a singular focused
effect of migration turnover on property crime, motivated
by results which suggested that increases in interprovincial
and international migrant movement, both inward and out-
ward, had similarly positive impacts on property crime, and;
(iii) the analysis incorporates data from the more recent
period characterized by low resource prices.
The relationship between rates of property crime and
the economic and demographic variables for the panel of
five provinces (British Columbia, Alberta, Saskatchewan,
Manitoba, and Ontario) is investigated using a linear panel
model. The set of provincial economic and demographic
variables used in the analysis are described in Table 1. These
variables are real household disposable income per capita,
the unemployment rate, per capita alcohol sales, the youth
share of the population, and the migration turnover rate
(the sum of both in- and out-migration, as a proportion of
the total population). Provincial data for unemployment
rates are only available starting in 1967. These variables all
exhibit the property of nonstationarity, and therefore require
transformation using logarithmic first differences to allow for
estimation to overcome the problems of stochastic trends and
spurious regressions; as such, this study may not be directly
compared to that of Andresen (2013), which presented results
from a panel study with variables in their undifferenced
logarithmic forms.2 In addition, a set dummy variables was
used to account for additional province-specific trends and
shifts in rates of property crime. The effective time period of
the study is from 1968 to 2015.
RESULTS
The results from the panel regressions for the growth in the
rates of property crime over the study period are presented in
Table 2. The first regression model includes contemporaneous
and lagged independent variables, and a common constant
term but with no other dummy variables. The estimates
from that model are presented in Column 2, are identified as
Model 1, and indicate how much variation may be accounted
for solely by variation in the independent variables. The
estimated fit of this first model is relatively low, with an
overall R2 of 0.18, indicating that the explanatory variables
used in this regression may account for a small portion of
the variation in the growth of the rates of property crime
over the sample. Only one variable, the unemployment rate,
was statistically significant in the regression. The second
regression model includes province-specific constant terms,
and province-specific one-period and trend shift dummies
which allow for trend changes that occurred around 1990, for
an upward shift in 1998 with the movement to UCR2 report-
ing, and also for changes around 2003 (see notes to Table 2).
The estimates from this model are identified in Column 3
as Model 2. This second regression performed much better
in terms of the R 2 statistic. The inclusion of these province-
specific dummy variables allows the model to measure the
impact of economic and demographic changes on property
crime rates with better precision, and all of the economic and
demographic variables have statistically significant effects on
property crime rates, either contemporaneously, with lagged
effects, or both. This model indicates how important it is to
include these province- and time-specific effects to better
identify the relationship between economic and demographic
change, and changes in rates of property crime.
Influence of Economic and Demographic Change on
Rates of Property Crime
The estimates provided in Table 2 for Model 2, of the impact
of economic and demographic changes on rates of property
crime, will first be explained using a few illustrative exam -
ples. The coefficient estimates of -0.1214 and -0.2232 for Δyt
and Δyt-1 suggest that for every one percentage point increase
in the growth rate of real disposable household income per
capita, the growth rate of the rate of property crime decreased
by 0.12 percentage points, and the growth rate of the rate of
property crime decreased by 0.22 percentage points in the
following period, all else equal. The coefficient estimates of
1.1038 and 0.1322 for Δyouth-sharet and Δyouth-sharet-1 suggest
1 Panel dataset estimation provides several advantages over individual
time series estimation, including “more informative data, more variabil-
ity, less collinearity among the variables, more degrees of freedom and
more efficiency” (Baltagi, 2005, p. 5).
2 Nonstationary series, including the undifferenced series in this analy-
sis, exhibit stochastic trends which may lead such series to appear to
be related over the long run when they aren’t, resulting in a spurious
regression. For more information on issues regarding nonstationar-
ity and spurious regressions, please refer to Stock and Watson (2003,
pp. 460–462).
FIGURE 1 Rates of property crime, four western provinces and Ontario,
1962–2015.
-‐
1,000.00
2,000.00
3,000.00
4,000.00
5,000.00
6,000.00
7,000.00
8,000.00
9,000.00
10,000.00
1962
1965
1968
1971
1974
1977
1980
1983
1986
1989
1992
1995
1998
2001
2004
2007
2010
2013
Incidents per 100,000 residents
British Columbia
Alberta
Saskatchewan
Manitoba
Ontario
IMPACTS OF ECONOMIC/DEMOGRAPHIC CHANGE ON PROPERTY CRIME RATES, Wilson54© 2018 Author.OpenAccess.ThisworkislicensedundertheCreativeCommonsAttribution-NonCommercial-NoDerivatives4.0 InternationalLicense.
T
o
v
i
e
w
a
c
o
p
y
o
f
t
h
i
s
l
i
c
e
n
s
e
,
v
i
s
i
t
h
t
t
p
:
/
/
c
r
e
a
t
i
v
e
c
o
m
m
o
n
s
.
o
r
g
/
l
i
c
e
n
s
e
s
/
b
y
-
n
c
-
n
d
/
4.0/. For commercial re-use, please contact marketing@multi-med.com.
that for every one percentage point decrease in the growth
rate of the share of youth aged 15 to 24 years to the total
population, the growth rate of the rate of property crime
decreased by 1.10 percentage points, and the growth rate of
the rate of property crime decreased by 0.13 percentage points
in the following period, all else equal. While the other specific
estimates themselves are of importance, it is appropriate and
convenient to continue the analysis focusing on the direction
of the influences of these variables on rates of property crime
for the purpose of simplicity and ease of presentation.
As noted in one of the illustrative examples above, the
empirical results indicate that there were negative contem-
poraneous and lagged relationships between changes in
disposable income and changes in rates of property crime.
In addition, there was a strong positive contemporaneous
link between unemployment rates and property crime rates;
however, that impact was partially muted with a counter-
acting negative effect of lagged unemployment rates. These
results strongly suggest that improvements in the economy
coincided with declines in rates of property crime, and that
increases in property crime rates occurred during poor
economic conditions, supporting a motivational effect of the
economy on crime. Rates of property crime were also found
to be negatively linked with contemporaneous and lagged
changes in per capita alcohol sales. This result runs counter
to the hypothesis that alcohol consumption and crime are
positively linked, given a high proportion of crime is com-
mitted by individuals dependent and under the influence of
alcohol and drugs (Pernanen et al., 2002). However, alcohol
sales were positively correlated with disposable income, and
are more likely a sign of increasing wealth and responsible
alcohol consumption, so that these results reinforce the infer-
ence that increasing prosperity leads to declines in the rate
of property crime.
Demographic change also had an influence on rates of
property crime. Changes in the proportion of youth in the
population and in the rate of migration turnover (both of
inward and outward migration) were positively linked to
changes in property crime. These results reinforce the age–
crime link with adolescents committing a high proportion
of crime in society (Gannon, Mihorean, Beattie et al., 2005;
Brennan, 2012), and over the past decade while property
crime rates have been falling, the proportion of youth in the
population has also been in decline. While much international
work has focused on the impact of immigrant populations on
crime rates, in the Canadian context there is only suggestive
TABLE I Data description
Variable
Symbol for
Logarithmic Form Description Sources
Property Crime Rate pcr Provincial rate of property crime, incidents per
100,000 residents as of July 1.a
CANSIM Tables 252-0001 (1962-1997);
252-0051 (1998-2015); 051-0026 (1962-
1970),- 0001 (1971-2015)
Per Capita Income y Real provincial household disposable income
per capita.
CANSIM Tables 384-5000 (1962-2013);
384-0040 and 051-0051 (2014-2015)b
Unemployment Rate ur The provincial unemployment rate. CANSIM I CHASS I CHASS Labels
D45076,45097,45118, 45139, 45160
(1967-1975)
CANSIM Table 282-0002 (1976-2015)
Per Capita Alcohol
Sales
alcohol_sales Real value of provincial alcohol sales per capita,
by March 31 fiscal year endc
CANSIM Tables 183-0006; 183-0023
(2014-2015 - see Note 3); 051-0026
(1962-1970), 051-0001 (1971-2015)
Youth Population
Share
youth_share The provincial ratio of the population aged 15 to
24 as of July 1 that year divided by the
population as of July 1 of that year.
CANSIM Tables 051-0026 (1962-1970),
051-0001 (1971-2015)
Migration Turnover
Rate
mig_turnover Migration Turnover Rate: The provincial ratio of the
number of international immigrants plus emigrants
plus interprovincial in-migrants plus interprovincial
out-migrants that year divided by the
population as of July 1.
CANSIM Tables 051-0017 and 051-0037
(1962-2015); 051-0026 (1962-1970),
-0001 (1971-2015)
a Police-reported crime data are available starting in 1962, using the Uniform Crime Reporting Survey UCR1 definitions and crime categories. These data are
made available by Statistics Canada from 1962 to 2003. Statistics Canada currently releases police-reported crime data under UCR2 coding. Changes
from the UCR1 coding include more detailed offense categories, expanding from the three-digit coding in UCR1 to the four-digit coding in UCR2. Many
offenses which were previously categorized in the “Other Criminal Code violations” category were re-categorized under UCR2 as violent or property
crimes. The UCR2 data go back to 1998 and are considered as the official crime statistics. Statistics Canada has also released more comprehensive
data using UCR1 methodology for the period from 1977 to 1997, with series similar to those released after 1997. For the purpose of this article, the
UCR1 data was used from 1962 to 1997, and the UCR2 data was used from 1998 to 2014. The change in methodology at 1998, with a substantial
jump in many police-reported crime rates in 1998 compared to 1997, is important to note and to treat appropriately in the econometric analysis.
b The two different data sources resulted in series discontinuities in 2014. Since the data for the 2014-15 period included earlier data, the series in first
differences was spliced starting in 2014 rather than the series in levels.
c Data is for fiscal years ending on March 31 of the given year. Since the data contain nine months of data from the previous year, regressions were
conducted using the lead (t+1) which would contain nine months of data from that calendar year along with data from three additional months from the
following year, but did not yield statistically significant coefficients for these leads.
T
o
v
i
e
w
a
c
o
p
y
o
f
t
h
i
s
l
i
c
e
n
s
e
,
v
i
s
i
t
h
t
t
p
:
/
/
c
r
e
a
t
i
v
e
c
o
m
m
o
n
s
.
o
r
g
/
l
i
c
e
n
s
e
s
/
b
y
-
n
c
-
n
d
/
4.0/. For commercial re-use, please contact marketing@multi-med.com.
that for every one percentage point decrease in the growth
rate of the share of youth aged 15 to 24 years to the total
population, the growth rate of the rate of property crime
decreased by 1.10 percentage points, and the growth rate of
the rate of property crime decreased by 0.13 percentage points
in the following period, all else equal. While the other specific
estimates themselves are of importance, it is appropriate and
convenient to continue the analysis focusing on the direction
of the influences of these variables on rates of property crime
for the purpose of simplicity and ease of presentation.
As noted in one of the illustrative examples above, the
empirical results indicate that there were negative contem-
poraneous and lagged relationships between changes in
disposable income and changes in rates of property crime.
In addition, there was a strong positive contemporaneous
link between unemployment rates and property crime rates;
however, that impact was partially muted with a counter-
acting negative effect of lagged unemployment rates. These
results strongly suggest that improvements in the economy
coincided with declines in rates of property crime, and that
increases in property crime rates occurred during poor
economic conditions, supporting a motivational effect of the
economy on crime. Rates of property crime were also found
to be negatively linked with contemporaneous and lagged
changes in per capita alcohol sales. This result runs counter
to the hypothesis that alcohol consumption and crime are
positively linked, given a high proportion of crime is com-
mitted by individuals dependent and under the influence of
alcohol and drugs (Pernanen et al., 2002). However, alcohol
sales were positively correlated with disposable income, and
are more likely a sign of increasing wealth and responsible
alcohol consumption, so that these results reinforce the infer-
ence that increasing prosperity leads to declines in the rate
of property crime.
Demographic change also had an influence on rates of
property crime. Changes in the proportion of youth in the
population and in the rate of migration turnover (both of
inward and outward migration) were positively linked to
changes in property crime. These results reinforce the age–
crime link with adolescents committing a high proportion
of crime in society (Gannon, Mihorean, Beattie et al., 2005;
Brennan, 2012), and over the past decade while property
crime rates have been falling, the proportion of youth in the
population has also been in decline. While much international
work has focused on the impact of immigrant populations on
crime rates, in the Canadian context there is only suggestive
TABLE I Data description
Variable
Symbol for
Logarithmic Form Description Sources
Property Crime Rate pcr Provincial rate of property crime, incidents per
100,000 residents as of July 1.a
CANSIM Tables 252-0001 (1962-1997);
252-0051 (1998-2015); 051-0026 (1962-
1970),- 0001 (1971-2015)
Per Capita Income y Real provincial household disposable income
per capita.
CANSIM Tables 384-5000 (1962-2013);
384-0040 and 051-0051 (2014-2015)b
Unemployment Rate ur The provincial unemployment rate. CANSIM I CHASS I CHASS Labels
D45076,45097,45118, 45139, 45160
(1967-1975)
CANSIM Table 282-0002 (1976-2015)
Per Capita Alcohol
Sales
alcohol_sales Real value of provincial alcohol sales per capita,
by March 31 fiscal year endc
CANSIM Tables 183-0006; 183-0023
(2014-2015 - see Note 3); 051-0026
(1962-1970), 051-0001 (1971-2015)
Youth Population
Share
youth_share The provincial ratio of the population aged 15 to
24 as of July 1 that year divided by the
population as of July 1 of that year.
CANSIM Tables 051-0026 (1962-1970),
051-0001 (1971-2015)
Migration Turnover
Rate
mig_turnover Migration Turnover Rate: The provincial ratio of the
number of international immigrants plus emigrants
plus interprovincial in-migrants plus interprovincial
out-migrants that year divided by the
population as of July 1.
CANSIM Tables 051-0017 and 051-0037
(1962-2015); 051-0026 (1962-1970),
-0001 (1971-2015)
a Police-reported crime data are available starting in 1962, using the Uniform Crime Reporting Survey UCR1 definitions and crime categories. These data are
made available by Statistics Canada from 1962 to 2003. Statistics Canada currently releases police-reported crime data under UCR2 coding. Changes
from the UCR1 coding include more detailed offense categories, expanding from the three-digit coding in UCR1 to the four-digit coding in UCR2. Many
offenses which were previously categorized in the “Other Criminal Code violations” category were re-categorized under UCR2 as violent or property
crimes. The UCR2 data go back to 1998 and are considered as the official crime statistics. Statistics Canada has also released more comprehensive
data using UCR1 methodology for the period from 1977 to 1997, with series similar to those released after 1997. For the purpose of this article, the
UCR1 data was used from 1962 to 1997, and the UCR2 data was used from 1998 to 2014. The change in methodology at 1998, with a substantial
jump in many police-reported crime rates in 1998 compared to 1997, is important to note and to treat appropriately in the econometric analysis.
b The two different data sources resulted in series discontinuities in 2014. Since the data for the 2014-15 period included earlier data, the series in first
differences was spliced starting in 2014 rather than the series in levels.
c Data is for fiscal years ending on March 31 of the given year. Since the data contain nine months of data from the previous year, regressions were
conducted using the lead (t+1) which would contain nine months of data from that calendar year along with data from three additional months from the
following year, but did not yield statistically significant coefficients for these leads.
End of preview
Want to access all the pages? Upload your documents or become a member.
Related Documents
Canadian Economy: Macro & Alcohol Policylg...
|14
|2522
|53