Content uploaded by Iranian Economic Review
Author content
All content in this area was uploaded by Iranian Economic Review on Mar 25, 2023
Content may be subject to copyright.
Iran. Econ. Rev. Vol. 22, No. 1, 2018. pp. 105-120
Attitudes of Iranians toward the Becker Proposition
Narges Hajimoladarvish*
1
Received: January 28, 2017 Accepted: May 6, 2017
Abstract
his paper evaluates attitudes of Iranians toward crossing red traffic
lights and their sensitivity to fines. Economic theory of crime under
expected utility predicts that because of the possibility of severe
punishments, risk adverse individuals would not cross red lights. This is
implied by the Becker proposition. However, among 262 individuals
surveyed, more than half of the sample has previous records of
conviction with respect to traffic laws. The result indicates that the
effect of introducing a new fine on pedestrians is about twice the effect
of increasing the existing fine on drivers by 150%. The elasticity of
crossing red lights with respect to fine hike is -0.25. Regression analysis
shows that previous record of breaking traffic laws, being single and
crossing red lights by cars are significant explanatory variables for
decision to do jaywalking.
Keywords: Becker Proposition, Crossing Red Traffic Lights,
Jaywalking, Expected Utility.
JEL Classification: D81, K42.
1. Introduction
A celebrated proposition from Becker (1968) states that the most
efficient way to deter crime is to impose the severest possible penalty
with the lowest possible probability. This is called the Becker
proposition by Dhami and Al-Nowaihi (2006). In other words, to
economize on costs of enforcement such as policing and trial costs, we
should impose the severest punishment with the lowest probability of
detection and conviction.
Becker is dominantly known for applying standard toolkits of
economics to new areas. Becker (1968) is the application of the
1
. Faculty of Social Sciences and Economics, Alzahra University, Tehran, Iran
(Corresponding Author: n.moladarvish@alzahra.ac.ir).
T
106/ Attitudes of Iranians toward the Becker Proposition
standard model of decision making into criminology. This approach is
in sharp contrast with other crime theories asserting that crime is the
result of economic stress, political legitimacy and family
disorganization.
Lea (2006) states that the theory of rational decision making is a
characteristic of an economic approach to any problem. This means
individuals will do the best they can. According to Lea (2006), on the
contrary, most psychologists believe that humans faced with real
economic problems will not in fact do their best.
Nevertheless, looking at crime through a cost-benefit framework is
not a dismal activity as the most important incentive of criminals is
more and easier lucrative economic benefits. Moreover, crime is a
major activity with estimates indicating that it constitutes up to one-
fifth of global gross domestic product (Glenny, 2008). On the other
hand, because of lack of reliable data and difficulties in estimating
crime costs, empirical analysis of crime is difficult. For example,
criminals’ opportunity cost is correlated with unemployment that, in
turn, depends on other factors. Thus, there is not even a consensus on
the existence of positive correlation between crime and some
determinant variables.
As stated by Dhami and Al-Nowaihi (2013) if the Becker
proposition holds, then its central insight should apply to all human
behavior that involves taking some action that with some probability
leads to severely costly outcomes. The work of Bar-Ilan and Sacerdote
(2004) shows how red lights running decreases in response to higher
fines. Dhami and Al-Nowaihi (2013) consider this evidence against
the Becker proposition.
If the Becker proposition holds, given the severe self-inflicted
punishments of running red lights, there should be no change in
behavior when one varies the external factors like the monetary
punishments or probability of an accident. Hence, a designed survey
asked whether people would cross red lights as drivers and pedestrians
if in a rush. The question is repeated when there is an increase in fine
for crossing red lights by cars or introducing a new fine for
jaywalking.
The elasticity of violation with respect to 150% increase in fine is -
0.25. This is consistent with the elasticity of conviction found in the
Iran. Econ. Rev. Vol. 22, No.1, 2018 /107
literature. Introducing a fine on pedestrians decreased jaywalking by
64%. People’s attitude towards jaywalking is best explained by their
previous record of breaking traffic laws and if they would cross red
lights by cars. It seems single people are crossing more red lights as
pedestrians as compared to drivers. As compared to other people,
single people are more sensitive to the imposition of a new fine on
pedestrians. These findings indicate that the Becker proposition does
not hold.
The plan of the paper is as follows. Section 2 investigates crossing
red traffic lights as a crime under expected utility framework. Section
3 describes the survey design and data. Section 4 reports the results.
Some remarks are made in conclusion.
2. The Model
Considering running red traffic lights as a crime, the benefit is
supposed to be the amount of money an individual could lose as a
result of being late. While this benefit can vary across individuals, it
should be a positive amount. Hence, the benefit of crossing red lights
is where denotes income from crossing the red
light, and is the income from not crossing the red light.
Figure 1: The Decision Tree
A decision maker is supposed to face a decision tree depicted in
Figure 1. If the individual crosses the red light, with some
probability she would have an accident, and with
probability she would not. In case of no accident, the
108/ Attitudes of Iranians toward the Becker Proposition
individual is caught with probability 0 . If caught, the
individual is asked to pay a monetary fine denoted by that refers to
public fine. In the crime literature, it is common to assume that the
punishment is a function of probability of detection. Hence, it is
assumed that is a hyperbolic function given by
, where is
a constant (Dhami and Al-Nowaihi, 2013). In case of an accident, the
individual will face a serious injury with probability, ,
and she will not with probability. In case of an accident, the
punishment and the loss of utility are coming from two different
sources. One is the self-inflicted punishment that includes injuries,
higher insurance premium and car repair costs. The self-inflicted
punishment is donated by . The other is the public policy
punishment.
If the individual does not face a serious injury the outcome is
, and if she does, the outcome is where
refers to the loss of utility from the serious injury. Given the
enforcement parameters and , individuals decide whether to cross
the red light or not. Under expected utility, if the individual does not
cross the red light, her payoff from no-crime ( condition is given
by . On the other hand, her expected utility from
crossing the red light is given by
(1)
The individual does not cross the red light if the no-crime condition
() given by is satisfied. is clearly satisfied
for . The continues to hold as declines from 1, if and
only if
(2)
Iran. Econ. Rev. Vol. 22, No.1, 2018 /109
Dividing both side of the above equation by, we get the
same equation that Dhami & Al-Nowaihi (2006) used to prove the
Becker proposition under expected utility;
They have proved that by the assumptions of hyperbolic
punishment function and concave utility function, the will hold
for all when there are severe punishments. Thus, the
expected utility predicts that individuals with concave utility functions
(risk adverse individuals) will not cross red lights in the presence of
severe punishments.
3. Data Collection Procedure and Its Description
Participants were asked whether they would cross red lights while
driving in a rush, and if yes, what if the fine increases up to 150%.
The same question was repeated as if they were pedestrians. However,
since there is no fine on jaywalking, in this case it was a matter of
introducing a fine rather than increasing it.
The sample is made of three distinct groups of people. Group one
consists of Iranians who are living abroad. This group was chosen to
make a comparison with the result of Bar-Ilan and Sacrdote (2004)
indicating that foreigners’ sensitivity towards fine hike is less as
compared to residents of the particular place. This is because
foreigners do not see themselves as part of the society. Group two
consists of Iranians residing in Iran. To compare the attitude of
Iranians with other nationalities, group three involves only Indian
subjects.
Participants were found through two methods. In the first method,
responses were collected through free online survey software (Survey
monkeys). The link of the survey was sent either directly to my
friends and acquaintances or it was shared on Facebook. In the second
method, the questionnaire was distributed where a group of people
could be found: i.e. insurance offices, banks, companies, and so on.
Only group two data was gathered by this method.
110/ Attitudes of Iranians toward the Becker Proposition
3.1 Complete Information of Aggregate Data
Table 1 indicates the description of collected variables for the whole
sample. The sample consists of 262 individuals with complete
information on ten variables.
Table 1: Definitions of Variables and Summary Statistics of Aggregate Data
Variables
Description
Values
Mean
Age
Age
18-64
30.874
Gender
Dummy
Female = 0, Male = 1
0.6297
Mart
Marital status
Single = 1, Other = 0
0.6641
Country
Country of birth
Iran = 1, India = 0
0.8396
Education
Years in full time
education
8-24
16.3454
Record
Previous records with
respect to traffic laws
Yes = 1, No = 0
0.5670
D.C.R
Crossing red lights while
driving
Yes = 1, No = 0
0.2366
If D.C.R
Crossing red lights while
driving after the fine
increase
Yes = 1, No = 0
0.1526
W.C.R
Crossing red lights while
walking
Yes = 1, No = 0
0.6870
If W.C.R
Crossing red lights while
walking after imposing a
fine
Yes = 1, No = 0
0.2519
One of the prominent features of this data is that it consists of well-
educated subjects with average age of 31 years old. The sample has 165
male respondents. 62% of males have a previous conviction with respect
to traffic laws. Among 97 females, only 46% of them have a previous
record. Hence, as compared to females, males break more traffic laws.
66% of respondents are single. As compared to others, singles
declared to cross fewer red lights by cars and have less previous
records with respect to traffic laws. However, single people declared
to do more jaywalking than the rest. Interestingly, the result after the
fine hike shows the exact opposite; as compared to other people,
single subjects declared to cross more red lights as drivers but less so
as pedestrians. It seems single people are more sensitive to the new
fine imposition as compared to the fine hike.
Iran. Econ. Rev. Vol. 22, No.1, 2018 /111
While 57% of the sample declared that they have previous records
of breaking traffic laws, only 24% said that they would cross red
lights if in a rush. The elasticity of violation with respect to the fine
hike is -0.25. This is consistent with Bar-Ilan and Sacerdote (2004)
results indicating that the elasticity of violation with respect to the fine
hike is around -0.33 to -0.26. Approximately, 69% of the sample said
that they would cross red lights as pedestrians. The figure reduced to
25% after imposing £30 sized fine in group one and three and 200,000
Rials in group two.
If we assume that the probability of an accident is equal in the case
of driving and walking, then the question is how these educated
individuals could have more importance for their cars than
themselves? Moreover, the punishment in the case of driving contains
self-imposed injury, costs of car repair, increased insurance premium,
receiving a fine and feeling of guilt in case of fatal accidents.
However, in the case of walking, the punishment is just of a self-
imposed nature. While this evidence shows the positive effect of
monetary punishment, it may also indicate that people believe that
they have more control over themselves than their cars.
I used McNemar tests to check the significance of change in behavior
after the fine increase. According to Conover (1999), when the data
consists of observations on n independent bivariate random variables and
the measurement scale for each variable is nominal with two categories
(yes and no), it's possible to use McNemar test to get informed about
significance of change. In the McNemar test, the data is summarized in a
2*2 contingency table. The null hypothesis is that there is no significant
difference between responses before and after the fine hike.
After
T=17.7894
Before
No
Yes
Yes
190
6
No
32
34
112/ Attitudes of Iranians toward the Becker Proposition
Since
.84 at 0.05 significance level and T is greater than
3.84, the null hypothesis is rejected. This means that there is a
significant difference between responses to crossing red lights before
and after the fine hike. For the case of pedestrians, the McNemar test
is applied to check the significance of change before and after the
introduction of a new fine.
After
T=116.0333
Before
No
Yes
Yes
81
1
No
119
61
Since the test statistic is 116.0333, the null hypothesis is strongly
rejected. This implies that the effect of imposing a new fine on
pedestrians' behavior is statistically significant.
3.2 Complete Information of Group One
This group consists of 100 young Iranians who are living abroad.
Because of sample selection, the average years of full time education
is 17.5 years. Table 2 shows descriptive statistics of this group.
Table 2: Summary Statistics of Group One
Variables
Mean
Standard deviation
Min
Max
Age
26.55
3.6608
21
40
Gender
0.61
0.4902
0
1
Mart
0.85
0.3588
0
1
Education
17.56
2.2141
8
24
Record
0.38
0.4878
0
1
D.C.R
0.16
0.3684
0
1
If D.D.R
0.14
0.3487
0
1
W.C.R
0.79
0.4093
0
1
If W.C.R
0.24
0.4292
0
1
The prominent feature of this group is that 79% of well-educated
people declared that they would cross red lights while walking. The
figure falls to 24% after the imposition of £30 sized fine. Only 16% of
this group would cross red lights while driving, and the elasticity of
Iran. Econ. Rev. Vol. 22, No.1, 2018 /113
violation with respect to the fine hike is -0.08. Thus, it seems
monetary fines have already had high deterrence for this young
educated group.
3.3 Complete Information of Group Two
Group two includes 120 Iranians residing in Iran. For this group, there
were 6 extra questions designed to consider the effect of increase in
probability of an accident happening on jaywalking. The considered
probabilities range from 1/1,000,000 to 1/10. Moreover, for this
group, I was able to get information about their monthly incomes.
Three level of income was defined; the first level is approximately
equal to the poverty line indicated by below ten million Rials
(approximately equivalent to £610 at the time of this study, 2010), the
second level is between ten and thirty million Rials and the last one
are above thirty million Rials. Table 3 demonstrates their responses.
Table 3: Summary Statistic of Group Two
Variables
Values
Means
Standard
devastation
Min
Max
Age
18-64
35.95
9.8402
18
64
Gender
0 or 1
0.64
0.4815
0
1
mart
0 or 1
0.42
0.4964
0
1
Education
12-19
15.15
2.1341
12
19
Income
Below
10m to
upper 30m
15923617
6757712
10,000,000
30,000,000
Record
0 or 1
0.85
0.3585
0
1
D.C.R
0 or 1
0.29
0.4564
0
1
If D.C.R
0 or 1
0.15
0.3665
0
1
W.C.R.A
0 or 1
0.65
0.4789
0
1
If W.C.R.A
0 or 1
0.25
0.4348
0
1
W.C.R.B
0 or 1
0.59
0.4935
0
1
If W.C.R.B
0 or 1
0.25
0.4348
0
1
W.C.R.C
0 or 1
0.51
0.5018
0
1
114/ Attitudes of Iranians toward the Becker Proposition
Table 3: Summary Statistic of Group Two
Variables
Values
Means
Standard
devastation
Min
Max
If W.C.R.C
0 or 1
0.20
0.4078
0
1
W.C.R.D
0 or 1
0.37
0.4861
0
1
If W.C.R.D
0 or 1
0.15
0.3665
0
1
W.C.R.A.E
0 or 1
0.24
0.4298
0
1
If W.C.R.E
0 or 1
0.11
0.3223
0
1
W.C.R.A.F
0 or 1
0.14
0.3501
0
1
If W.C.R.F
0 or 1
0.07
0.2644
0
1
W.C.R.A denotes crossing red traffic lights while walking when the
probability of an accident is 1/1,000,000. This is a dummy and equals
to 1 if yes and 0 otherwise. Again If W.C.R.A refers to the same
variable after the introduction of the fine. In the same manner,
W.C.R.B to W.C.R.F denote crossing red traffic lights while walking
when the probability of an accident is 1/100,000 and decreases to 1/10
by a factor of 10.
Similar to group one, this group has an average age of 36 years old.
Interestingly, 85% of this group declared to have previous records of
breaking traffic laws. However, quite similar to other groups, only
29% of the group declared that they would cross red lights while
driving. This figure declines to 16% after the fine hike. Hence, the
elasticity of crossing red lights with respect to the fine increase is -
0.30. Since the same elasticity is equal to -0.08 for Iranians living
abroad, our finding is consistent with the result of Bar-Ilan and
Sacrdote (2004) indicating less sensitivity of foreigners to an increase
in fine.
Figure 2 illustrates average responses to the questions of
jaywalking with different probabilities of an accident before and after
the introduction of the fine. The horizontal axes show different
probabilities of an accident happening, while the vertical axes shows
the percentage of people who would cross red lights. The blue line
(upper one) shows the sample behavior before the introduction of the
Iran. Econ. Rev. Vol. 22, No.1, 2018 /115
fine and the red line indicates their responses after introducing the
fine. Intuitively, the deterrence effect of monetary punishment is less
than the increase in the probability of an accident happening.
Figure 2: Responses to Increase in the Probability of an Accident Happening
3.4 Complete Information of Group Three
We got only 42 subjects for the last group who are all Indians. Table 4
summarizes their responses.
Table 4: Summary Statistic of Group Three
Variables
Mean
Standard
devastation
Min
Max
Age
26.64
5.7586
20
55
Gender
0.64
0.4849
0
0
Mart
0.90
0.2971
0
0
Education
16.83
2.2837
10
24
Record
0.21
0.4152
0
0
D.C.R
0.35
0.4849
0
0
If D.D.R
0.16
0.3771
0
0
W.C.R
0.54
0.5037
0
0
If W.C.R
0.19
0.3974
0
0
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.000001 0.00001 0.0001 0.001 0.01 0.1
Average crossing
Probability of an accident
116/ Attitudes of Iranians toward the Becker Proposition
As mentioned earlier, because of the selection procedure, the
average age of this group is approximately 27 years. 36% of this
group said they would cross red lights by cars. The figure decreases to
16% after raising the existing fine by 150%. Hence, the elasticity of
violation with respect to the fine hike is -0.37. This shows the highest
absolute elasticity. For pedestrians, the effect of introducing the £30
sized fine is shown by 65% reduction in the numbers of jaywalking.
4. Regression Analysis
Logit regressions are used to estimate the probability of jaywalking
based on a few factors that influence the decision to cross red lights.
These explanatory variables are Age, Gender, Mart, Education,
Income, Country, Record and D.C.R collected in . Let denotes the
decision under study for individual, where = 1 if the individual
declares to cross red lights and = 0 otherwise.
= Pr [= 1| ] = F (
)
The Logit model assumes F (
) has the logistic distribution given by:
F (
) =
The result of this model using aggregate data is given in Table 5.
Table 5: Estimates of the Logit Model with the Data in Table 1
Variables
Coefficients
Standard
error
Z
Odds
ratio
D.C.R
1.0341
0.4119
2.65
0.008
2.8154
Record
1.0027
0.3348
2.99
0.003
2.7256
Education
0.1967
0.0629
3.13
0.002
1.2173
Country
0.8603
1.3722
-2.97
0.003
2.3640
Mart
0.6977
0.3530
1.98
0.048
2.0092
Gender
0.0348
0.3051
0.11
0.909
1.0354
Age
Constant
-0.0086
-4.0703
0.0189
0.3905
-0.46
2.09
0.649
0.037
0.9914
0.0170
Note: Number of obs= 261, Log likelihood = -142.47197, LR chi2(7) = 39.95,
Prob > chi2 = 0.0000
Iran. Econ. Rev. Vol. 22, No.1, 2018 /117
Intuitively, D.C.R is the most important explanatory variable that
would affect the decision to do jaywalking. The same people who
would cross red lights while driving are more likely to do so as
pedestrians. If we increase D.C.R by one unit, the odds ratio of
jaywalking will be multiplied by 2.8154.
We can also expect people with previous records of conviction to
do more jaywalking. Such people cross red lights approximately three
times more than the ones who do not have a previous record. This is
consistent with the work of Bar-Ilan and Sacerdote (2004) indicating
that criminals cross more red lights than those without a record.
The positive coefficient of Mart indicates that single people run
more red lights as compared to other people. The odds ratio indicates
that running red lights by single pedestrians is almost twice as likely
as running red lights by others. Finally, as compared to Indians,
Iranians are more likely to cross red lights as pedestrians.
Contrary to the common belief that educated people will commit
less crime and have greater respect for the rule of law, this sample
shows the opposite. Although the odds ratio for Education is close to
1, the positive significant coefficient indicates that people with more
years of education would do more jaywalking.
Negative coefficients lead to an odds ratio of less than one. For
example, the negative coefficient of Age leads to the odds ratio of
0.9914. This means that a one unit increase in Age leads to the event
being less likely to occur. However, because of the selection
procedure, Age was an insignificant explanatory variable. We have a
young sample with 75% of subjects in the range of 19-34 years old. It
is worth mentioning that if we rule out the insignificant variables the
odds ratios will not change significantly as in the logit regression the
insignificant variables are not counted.
As discussed in Cameron and Trivedi (2010) assessing the fit of the
model enables researchers to measure how effectively the model can
describe the outcome variable. One approach to evaluate the fit of the
model is measuring the percentage of correctly classified observations
by comparing fitted and actual values. As a result of applying this
method, it seems 73.66 % of the values are correctly specified.
For robustness checks, the model was estimated for each group
data. The results are similar with the aggregate data. More
118/ Attitudes of Iranians toward the Becker Proposition
specifically, the model with group one data has the highest odds ratio
of 6.7949 for D.C.R. This means people who would run red lights by
cars would do jaywalking approximately seven times more than
others. Group two with the highest number of observations gives
similar results to the ones obtained from the aggregate data. However,
the income variable was an insignificant explanatory variable. The
model with group three data also indicates that Record and D.C.R are
the best explanatory variables with odds ratio of 2.0773 and 3.6854
respectively. Because of high level of education in all three groups,
we have positive insignificant coefficients for education.
5. Conclusion
A conducted survey data is used to investigate Iranians’ behavior
towards traffic laws and their sensitivity to fines. 262 individuals were
asked whether they would cross red lights by cars or as pedestrians
while varying the associated fines. There is a considerable difference
between responses to an increase in fine, and to the introduction of a
new fine. The result shows that the effect of a new fine on pedestrians
is approximately twice the effect of increasing fine up to 150% for
drivers.
The analysis shows that males are just as sensitive to fines as
females, whereas singles are more responsive than others. Educated
people declared to do more jaywalking than others. Individuals with
previous records of breaking traffic laws would cross more red lights
as pedestrians, while their sensitivity to a new fine is approximately
similar to those without such records.
In comparison with people who do not have previous records with
respect to traffic laws, people with such a record would cross more red
lights by cars. These people are more sensitive to the fine hike, which
contradicts the result from Bar Ilan and Sacerdote (2004) indicating
that criminals are just as sensitive as noncriminal to changes in fine.
In all three groups, we observe a substantial decrease in the
numbers of declared jaywalking after the introduction of the fine.
Since jaywalking has a severe self-inflicted punishment with low
probability, the Becker proposition predicts no jaywalking. As the
data shows, increase in monetary punishment changes behavior
significantly. Thus, on the whole, the conducted survey contradicts the
Iran. Econ. Rev. Vol. 22, No.1, 2018 /119
Becker proposition. The elasticity of crossing red lights with respect
to 150% increase in fine is -0.25, which fits in the range of found
elasticities in the crime literature.
While the introduction of a new monetary fine on pedestrians
decreased the violation by 64%, increasing the existent fine on drivers
only dropped the infringement up to 37%. This can explain the trend
of traffic laws. Traffic laws were initially focused on increasing
associated penalties to economize on enforcement costs. However, as
the effect of monetary punishments was declining, other kinds of
punishments were implemented. The reason why the point system
(when one's driving license is suspended for some period depending
on the received points) has become popular relies on the efficiency
brought by introducing new non-monetary fines.
Furthermore, sever monetary punishments result in public’
perception that policies are designed with other objectives than
deterrence. Moreover, having the severest monetary punishment is
impossible as it depends on offenders’ ability to pay. Hence, the
maximum penalty should not be just the monetary one. People will be
more sympathetic with policies targeting criminals' behavior rather
than their wallets. Thus, policies on deterrence should be creative.
Acknowledgement
I am sincerely thankful to Ali Al-Nowaihi and Sanjit Dhami for their
continuous suggestions and criticism that has improved this paper
substantially. This works is based on my master dissertation under
their supervision. I was supported by the Iranian National Elites
Foundation and Alzahra University during the manuscript writing and
I am grateful for that.
References
Dhami, S., & Al-Nowaihi, A. (2006). Hang ’Em With Probability
Zero: Why Does It Network? University of Leicester Working Papers,
Retrieved from
http://www.le.ac.uk/economics/research/RePEc/lec/leecon/dp06-
14.pdf.
120/ Attitudes of Iranians toward the Becker Proposition
Dhami, S., & Al-Nowaihi, A. (2013). An Extension of the Becker
Proposition to Non-Expected Utility Theory. Mathematical Social
Sciences, 65(1), 10–20.
Bar Ilan, A., & Sacerdote, B. (2004). The Response of Criminals and
Noncriminals to Fines. Journal of Law and Economics, 47, 1–17.
Becker, G. (1968). Crime and Punishment: An Economic Approach.
Journal of Political Economy, 76, 169–217.
Cameron, A. C., & Trivedi, P. K. (2010). Micro-econometrics Using
Stata (2nd Ed.). Texas: Stata Press.
Conover, W. J. (1999). Practical Nonparametric Statistics (3rd Ed.).
Michigan: John Wiley & Sons, Inc.
Glenny, M. (2008). McMafia: A Journey through the Global Criminal
Underworld. New York: Vintage books.
Lea, S. E. G. (2006). How to Do as Well as You Can: The Psychology of
Economic Behavior and Behavioral Ecology, in Altman, M. (Ed.).
Handbook of Contemporary Behavioral Economics Foundations and
Developments. New York: Armonk.
