Transcript Individual Political Freedoms and Civil Liberties (contd)

Shirkers and Drug Runners:
Why U.S policy fail to induce
crackdowns on drug trafficking
through transit countries
By
Horace Bartilow
Kihong Eom
The University of Kentucky
Department of Political Science
1
Abstract
Do US anti-drug policies, specifically its bilateral interdiction agreements with drug transit
countries, induce foreign country cooperation against drug trafficking? To answer this question
we present a simple formal model of a two player drug interdiction game that demonstrates that
drug transit countries, relative to drug consumer countries, have a greater incentive not to
cooperate in bilateral drug interdiction arrangements. We empirically test our formal model by
analyzing cocaine and marijuana drug seizure data for Caribbean basin countries from 1984 to
2000 using a time series cross-section model. In addition to bilateral interdiction agreements, a
host of other variables that influence the willingness of drug transit countries to cooperate with
interdiction are considered. The theoretical prediction from our formal model found strong
empirical support for cocaine seizures but limited support for marijuana seizures. Since the
primary threat to US security is cocaine and not marijuana, America’s drug policy needs revision
to provide sufficient incentives for drug transit countries to effectively cooperate in cocaine
interdiction.
2
Drug trafficking and the threat to
US Security
• Drug consumption, production, and sales
have become a growing threat to American
security, in particular since the U.S is the
world’s largest consumer of illicit drugs. In
the late 1980s, the U.S Congressional
Research Service reported that “America is
consuming drugs at an annual rate of more
than 6 metric tons (mt) of heroin, 70 to 90 mt
of cocaine, and 6,000 to 9,000 mt of
marijuana - 80 percent of which is imported.
American demand therefore is the linchpin of
one of the fastest growing and most
profitable industries in the world .”
3
Drug trafficking and the threat to
US Security (contd)
• In the early 1990s, the Organization for
Economic Cooperation and Development
estimated that US$122 billion was spent each
year in the United States and Europe on heroin,
cocaine, and cannabis (OECD, 1990). Other
studies estimate that approximately US$50
billion is spent each year on illicit drugs in the
United States (Rhodes, 1995). Former US drug
‘czar’, General Barry McCaffrey, estimated that
about three hundred metric tons of the
approximately 575 metric tons of cocaine
available world wide in 1994 was consumed in
the US (McCaffrey, 1995).
4
The threat to social and
economic security in the US
• Drug trafficking and America’s appetite for illicit drugs poses
a direct threat to the country’s social and economic security.
In the US drug trafficking and drug abuse continues to lower
economic productivity and competitiveness while
contributing to the explosion of health care costs. According
to the Office of National Drug Control Policy, economic
productivity losses in the U.S due to drug abuse have risen
from $69,421 million in 1992 to $110,491 million in 2000.
Health care costs due to drug abuse have risen from
$10,820 million in 1992 to $14,899 million in 2000 (National
Drug Control Policy, 2001a). Individuals incarcerated for
drug offenses in Federal and State prisons have risen from
147,000 in 1989 to 336,400 in 2000 (National Drug Control
Policy, 2000b). And the number of deaths from drug induced
causes increased from 7,101 in 1979 to 19,102 in 1999
(National Drug Control Policy, 2000c).
5
The significance of the Caribbean
Basin – Drug Transit Centers
• The Caribbean basin’s geographic
proximity to South and North America
makes it a major transshipment center for
South American drugs that are destined
for markets in the U.S and Europe.
According to American policy makers, a
primary security threat to the United States
is the transshipment of large amounts of
cocaine through the Caribbean basin
(Perl, 1993-94; Tokatlian, 1994).
6
The significance of the Caribbean
Basin – Money Laundering Centers
• In addition to facilitating drug trafficking, the Caribbean
basin also plays an important role in drug-related money
laundering. Many countries in the region have welldeveloped offshore banking systems and bank secrecy
laws that facilitate money laundering. In countries with less
developed banking systems, cash often is moved in bulk
shipments through the region. In January 2000, the United
Nations (UN) reported that approximately 42 percent of the
world’s offshore banks and 38 percent of the world’s
offshore corporate entities (also known as international
business companies or IBCs) are located in the Caribbean
basin. The Caribbean Financial Action Task Force (CFATF)
recently estimated that US$60 billion in drug trafficking and
organized crime proceeds are laundered through the
Caribbean every year (Griffith, 1997:97-113; The Strategic
South American/Caribbean Unit, 2001)
7
US International Drug Policy
• American anti-drug policy is informed by
the logic that the reduction of the supply of
drugs will increase market prices and as a
result US consumers will forgo drug
consumption and drug abuse. By this
logic, efforts to reduce domestic drug
consumption in the US have led American
policy makers to seek collaboration with
foreign countries (Toro, 1992).
8
US International Drug Policy
(contd)
• Consequently, the goal of American drug
policy has been to secure the cooperation
of foreign governments in three areas in
the ‘war on drugs’: Drug interdiction, crop
eradication and the arrests or
‘immobilization’ of drug traffickers.
However, the primary purpose of this
research is to analyze the effectiveness of
foreign cooperation in drug interdiction.
9
US International Drug Policy: The
Anti-Drug Abuse Act of 1986
• To ensure foreign country cooperation in
the ‘war on drugs’ the Reagan
administration introduced the anti-drug
abuse act which made US financial
assistance, positive votes within
multilateral lending institutions and trade
preferences conditional on foreign
governments’ cooperation against drug
trafficking (Perl, 1989; United States
Senate, 1988b).
10
The Anti-Drug Abuse Act of 1986
(contd)
• The legislation also strengthened the
capacity of the US to impose the
extraterritoriality of its criminal laws
throughout the Western hemisphere. Under
the threat of ‘decertification’ (sanctions) the
US government has been able to secure
over flight and bilateral Maritime Law
Enforcement Cooperation agreements with
various countries in the Caribbean basin.
11
Does the US Drug Policy induce
foreign country cooperation?
• Students of the politics of transnational
organized crime have consistently noted that the
US certification process may have created an
environment that discourages foreign country
cooperation in combating drug trafficking (Toro
1992; Bryan 1997; Granada 2002; Naylor 2002).
• Most governments in the region perceive the
imposition of the extraterritoriality of American
law enforcement agents and criminal justice as a
loss of state control as an infringement on their
national sovereignty and the independence of
their legal systems.
12
The Argument
• Beyond bruised national feelings, the failure to
induce foreign country cooperation to combat
drug trafficking is largely a function of the
structural impediments that are inherent in drug
interdiction cooperation between unequal
partners; namely the US, the world’s only superpower and the largest consumer of illicit drugs,
and Caribbean basin countries who are
politically and economically vulnerable, and
compared to the US, less exposed to illicit drug
consumption.
13
The Argument (contd)
• American power may coerce smaller
countries into accepting drug interdiction
treaties but power alone may be
insufficient to induce their cooperation
(Bartilow, 2000; Bartilow, 2001). Since
American wealth and the demand for illicit
drugs make the US the primary market for
drugs, small countries for which drug
trafficking represents less of a security
threat may be less inclined to cooperate.
14
The Argument (contd)
• Therefore, asymmetries in power, per
capita wealth and the relative demand for
illicit drugs structure the nature of the
strategic interactions between the US and
Caribbean basin countries and provide a
greater incentive for small countries in the
region to shirk their cooperative
responsibilities with the US in the
interdiction of illicit drugs.
15
A simple two-player drug
interdiction game.
In this game between Big state and Little state we
make the following assumptions:
1) Bg > Lg (where Bg is the gain for Big
State and Lg is the gain for Little State).
2) Bc > Lc (where Bc is the cost for Big State and
Lc is the cost for Little State).
3) Bg – Bc > 0 and Lg – Lc > 0
4) The effectiveness of Little State’s drug
interdiction is conditional on Big State’s
interdiction.
16
Both States can choose to:
1. Shirk, or
2. Interdict
17
The Payoffs
1. If both states interdict,
Big state’s payoff is Bg – Bc
Little state’s payoff is Lg – Lc.
2. If Big state interdicts and Little state shirks
Big state’s payoff is Bg - Bc - Lc
Little state’s payoff is Lg.
3. If Big state shirks and Little state interdicts
Big state’s payoff is 0
Little state’s payoff is Lg – Lc – Bc.
4. If both states shirk,
Big state’s payoff is -Bg
Little state’s payoff is –Lg.
18
Who moves first?
1. Big State Moves First
2. Little State Moves First
3. Both States Move Simultaneously
Regardless of who moves first, using
backward induction, the Nash equilibrium
is: “Big state interdict and Little state shirk.”
19
Figure 1.1.
Big State Chooses First.
Figure 1.2.
Little State Chooses First
Big State
Little State
I
S
s
i
Little State
Little State
s
(-Bg, -Lg)
Big State
Big State
i
(0, Lg –Lc – Bc)
i
s
S
(Bg – Bc, Lg – Lc)
(Bg – Bc – Lc, Lg)
I
S
(Lg, Bg – Bc – Lc)
(-Lg, -Bg)
(Lg –Lc – Bc, 0)
I
(Lg – Lc, Bg – Bc)
Figure 1.3.
Both States Move Simultaneously.
Big State
S
I
Little State
Little State
s
(-Bg, -Lg)
i
(0, Lg –Lc – Bc)
s
(Bg – Bc – Lc, Lg)
i
(Bg – Bc, Lg – Lc)
20
Figure 1.1.
Big State Chooses First.
Big State
S
I
Little State
Little State
s
(-Bg, -Lg)
i
(0, Lg –Lc – Bc)
s
(Bg – Bc – Lc, Lg)
i
(Bg – Bc, Lg – Lc)
21
Central Hypothesis
• Our formal model of the two player drug
interdiction game generates the following
proposition: Bilateral drug interdiction
agreement between unequal partners –
namely the US and Caribbean Basin
governments (CB) - is likely to have a
negative effect on narcotic interdiction.
22
Measuring the Effects of Bilateral
Counter-Narcotics Agreements on
Drug Interdiction
• We empirically test this proposition by
estimating cocaine and marijuana seizures
utilizing a time-series cross-sectional
analysis of the data that spans the period
1984-2000 and includes all the countries
in Central America and the greater
Caribbean (except for Cuba).
23
The Primary Explanatory Variable
• The primary explanatory variable measures the
bi-lateral drug interdiction agreements between
the US and the CB countries in our sample. The
US/CB bi-lateral drug interdiction indicator is a
dichotomous variable that measures whether or
not CB countries, in any given year, are
signatories to a maritime interdiction agreement,
a ship rider agreement, an over-flight interdiction
agreement or an extradition treaty with the U.S.
This variable is coded 1 if a country is party to
any of these agreements and 0 otherwise.
24
The Dependent Variables
• The drug seizure dependent variables are
operationalized as the ratio of a country’s
drug seizures as a proportion of a
country’s law enforcement personnel. For
the countries in our sample, it is calculated
by dividing a country’s yearly drug seizure
of cocaine and marijuana in kilograms by
the yearly number of a country’s active law
enforcement personnel.
25
Empirical Limitations
• To test the effect of US/CB bilateral drug
interdiction agreement on the level of drug
seizures in the region, one must first consider
the factors that shape the behavior of drug
traffickers – since the behavior of traffickers
affect drug seizures. However, due to the fact
that drug trafficking is an illicit enterprise, the
behavior of the drug trafficker is unobservable
and therefore cannot be measured directly. If the
trafficker’s behavior could be observed directly
then drug trafficking would cease to be a global
problem.
26
Empirical Limitations: Spatial
Correlation
• A significant limitation of empirical narcotic
interdiction research is that we cannot observe
how traffickers adjust their behavior in response
to the interdiction measures of law enforcement
officials (Crawford et al., 1988). In other words, if
Costa Rican law enforcement is rigorous in the
patrol of their territorial waters and air space,
then traffickers seeking to penetrate the
American market could simply move their
smuggling operations through Haiti where law
enforcement drug surveillance may be less
rigorous
27
Empirical Limitations: Spatial
Correlation (contd)
• In essence, empirical research on drug
interdiction must address the problem of
spatial correlation. This simply means that
the interdiction of drugs across countries
are related. Our estimation model corrects
for spatial correlation by adopting the
statistical solution that is offered by
Nathaniel Beck and Jonathan Katz (Beck,
2001; Beck and Katz, 1995).
28
Data Imperfection
• Another significant limitation of empirical narcotic interdiction
research is that we cannot observe what portion of the total
amount of drugs that is being smuggled is actually seized.
High levels of drug seizures may reflect the reality that more
rather than less drugs are being smuggled into countries
(Bagley, 1989:46-47; Stares, 1996:11). Empirical narcotic
interdiction research is therefore limited by imperfect data,
which will produce random measurement error. But random
measurement error would simply push the coefficients of our
central explanatory variables towards zero - lessening the
magnitude of their coefficients and thus produce
conservative estimates (Kmenta, 1997).
29
Modeling the Incentive Structure of
Drug Trafficking
• Our research design compensates for this data
imperfection by controlling for a package of
theoretically motivated proxy variables that
capture the incentives that shape the underlying
level of drug trafficking. These include:
geography, per capita national income, and the
domestic consumption of illicit drugs, the level of
government corruption, the level of economic
openness, the level of institutional political
openness, and the level of individual political
freedoms.
30
Geographic Distance
• Drug trafficking is as much about acquiring
wealth as it is about covering distance. We
assume that the geographical distance between
drug transit countries in the CB and destination
markets in North America is an important aspect
of the incentive structure that shape the
underlying level of drug trafficking. Traffickers
are more likely to move their drugs through CB
countries that are closer to the US than CB
countries that are further away. All things being
equal, CB countries that are closer to the US will
be exposed to higher levels of trafficking and are
expected to interdict more drugs than countries
that are further away from the American market.31
Coastline
• The length of a country’s coastline is another
aspect of the incentive structure that shapes the
level of smuggling. Drug traffickers are more
likely to move their drugs through countries with
longer coastlines, which reduce the chances for
detection by law enforcement than countries with
shorter coastlines, which increases the
likelihood of detection. All things being equal, CB
countries with longer coastlines will be exposed
to higher levels of trafficking and are expected to
interdict more drugs than countries with shorter
coastlines.
32
National Income Per-Capita
• We assume that the levels of per-capita national income
among CB countries are critical aspects of the incentive
structure of drug trafficking. All things being equal, since
richer countries in the CB are likely to spend more on
illicit drugs and they will be exposed to higher levels of
trafficking and are expected to interdict more drugs than
countries in the region that are poor. This proposition,
however, is conditional on the different demand and
supply elasticities and market constraints under which
cocaine and marijuana traffickers conduct their business
(Caulkins, 1994; Wisotsky, 1986). As a result our
proposition is more likely to hold true for cocaine
interdiction than marijuana interdiction.
33
National Income Per-Capita (contd)
• Since the street value and financial proceeds are higher
for cocaine than marijuana, we expect that CB countries
with high levels of per-capita income will be exposed to
higher levels of cocaine trafficking and will interdict more
cocaine than poorer countries in the region. On the other
hand, since the street value and financial proceeds for
marijuana are low relative to cocaine (Wisotsky, 1986),
we expect that richer countries in the CB will not attract a
high level of marijuana trafficking and therefore high
levels of per-capita income may have a negative effect
on marijuana interdiction. The fact that CB countries
serve as drug transit zones for the US market, we control
for US per-capita income since the level of interdiction
among CB countries is influenced by America’s wealth.
34
Domestic Demand for Illicit Drugs
• We assume that the level of CB countries domestic appetite for illicit
drugs is also an important aspect of the incentive structure of drug
trafficking. Since marijuana is more widely produced and consumed
throughout the CB, we suspect that the available data sources on
illicit drug consumption among users in CB countries may be biased
in favor of marijuana users. The strong cultural acceptability and
preference for marijuana consumption throughout CB countries has
the tendency to decriminalize marijuana while criminalizing cocaine
(Campbell, 1987; Chvannes, 1988; Kleiman and Saiger, 1992;
MacDonald, 1988; Rubin and Comitas, 1976). Therefore, given the
cultural preference for marijuana usage, we expect that illicit drug
consumption among CB countries will have a negative effect on
marijuana interdiction while having a positive effect on cocaine
interdiction. Moreover, the fact that CB countries serve as drug
transit zones for the US market, we control for US drug consumption
since the level of interdiction among CB countries is influenced by
America’s demand for illicit drugs.
35
Government Corruption
• We assume that the level of a country’s corruption is another
aspect of the incentive structure that shapes the level of drug
trafficking. Government corruption can ‘grease the wheels’ of
drug trafficking by improving the efficiency of smuggling enabling traffickers to smuggle drugs in and out of countries
with impunity (Galtung, 2000; Kaufmann and Wei, 1999; Leff,
1964; Lui, 1985). Drug traffickers are, therefore, more likely to
move their drugs through countries where government
corruption is embedded in the normal operations of the state
and its agents, as opposed to countries where government
agents and institutions are relatively clean. Since corruption
undermines the integrity of the state and its officials, high
levels of corruption is expected to have a negative effect on a
government’s ability to interdict drugs.
36
Government Corruption (contd)
• However, this proposition is conditional on the risk
threshold of corruption prone government officials. Since
corruption carries the risk of detection and punishment,
corruption prone government officials are more likely to
accept a bribe payment that exceeds their risk threshold
and are more likely to reject a bribe payment that is
lower than their risk threshold. Therefore, corruption is
expected to have a negative effect on cocaine
interdiction since the street value and proceeds from this
drug are consistently high enough to effectively exceed
the risk threshold of corrupt government officials.
37
Government Corruption (contd)
• On the other hand, corruption may have a positive effect
on marijuana interdiction. Since the street value and
proceeds form marijuana may not be high enough to
effectively meet or exceed the risk threshold of corrupt
government officials, these officials may be more vigilant
in marijuana interdiction in order to project an image of
integrity and seriousness in combating the drug trade. In
short, marijuana traffickers could be disadvantaged in
their ability to buy protection from corrupt government
officials since they may not meet or exceed the expected
bribe payment that is established by wealthier cocaine
traffickers. Therefore, we expect that corruption will have
a negative effect on cocaine interdiction but have a
positive effect on marijuana interdiction.
38
Economic Openness
• Globalization scholars contended that higher
levels of global economic openness in trade and
investment facilitates the expansion of the drug
industry (Andreas, 2000:15-26; Andreas, 2002;
Serrano, 2002; Stares, 1996). We, therefore,
assume that a country’s level of economic
openness is an essential incentive that shapes
the level of drug trafficking. Drug traffickers are
more likely to smuggle drugs through countries
where the state plays a minor role in regulating
the flow of goods, services and people that
move across its borders.
39
Economic Openness (contd)
• As global free markets displace the state in the
regulation of trade and migration, the likelihood of
detecting drug trafficking is reduced. Greater economic
openness attracts drug traffickers since the probability of
detection by the state is much lower than if they
smuggled drugs through countries where the state
played a central role in the regulation of trade and the
monitoring of its borders. All things being equal,
countries with open economies will be exposed to higher
levels of drug trafficking and is expected to interdict more
drugs than countries with relatively closed economic
systems.
40
Individual Political Freedoms and
Civil Liberties
• The spread of economic globalization has also given rise
to the spread of democratization and individual political
freedoms (Drake, 1998; Gill, 1996; Mittleman, 1996;
Monshipouri, 1995; Rueschemeyer and Evans, 1985).
We argue that individual political freedoms, defined in
terms of the level of political and civil liberties of citizens,
is an important aspect of the incentive structure that
shapes the level of drug trafficking. In countries where
political and civil liberties are relatively high, citizens are
protected from arbitrary arrest, intimidation and searches
by the state (Craig, 1980).
41
Individual Political Freedoms and
Civil Liberties (contd)
• While these rights protect the civil liberties
of citizens they may also constrain the
state’s ability to combat the drug trade.
Drug traffickers are, therefore, more likely
to smuggle drugs through countries where
political rights and civil liberties are
relatively high as opposed to countries
where individual political freedoms are low
and the state is not constrained by the civil
liberties of its citizens.
42
Individual Political Freedoms and
Civil Liberties (contd)
• All things being equal, CB countries that
enjoy high levels of individual political
freedoms and civil liberties will be exposed
to a higher level of drug trafficking and will
interdict more drugs than countries where
individual political freedoms and civil
liberties are limited.
43
Individual Political Freedoms and
Civil Liberties (contd)
• Our proposition is conditional on the type of drug in
question. High levels of individual freedoms and civil
liberties in CB countries may shield cocaine traffickers
and users from arbitrary state power and hence may
reduce the capability of law enforcement to interdict
cocaine. On the other hand, since the cultural tendency
among CB countries is to decriminalize marijuana
consumption, in countries where political freedoms and
civil liberties are high, individuals may be encouraged to
take more risks in the consumption and trafficking of
marijuana which would result in higher levels of
marijuana seizures.
44
Institutional Political Openness
• Economic globalization has also facilitated the
decentralization and the openness of political institutions
(Roberts, 1996; Self, 1993; Sheth, 1995). We argue that
the openness of a country’s political institutions is another
important aspect of the incentive structure that shapes the
level of drug trafficking. The more open countries national
political institutions become the easier it is for drug
traffickers to gain access to political leaders and
potentially corrupt the national policy making process.
Drug traffickers are therefore more likely to smuggle drugs
through countries where national political institutions are
relatively open as opposed to countries where political
institutions are closed. All things being equal, countries
with open political institutions will be exposed to a higher
level of drug trafficking and will interdict more drugs than
countries with closed political institutions.
45
Potentially Confounding Variables
• In addition, to the above proxy variables that
captures the incentives that shape the
underlying level of drug trafficking, we control for
potentially confounding influences on the level of
drug interdiction (cocaine and marijuana
seizures) among CB countries. These variables
include: the level of US economic and military
assistance to CB countries and the level of
cocaine and marijuana seizures among the drug
producing Andean countries (Columbia, Bolivia,
and Peru) in Latin America.
46
The Level of US Economic and
Military Assistance
• An important carrot of the US ‘certification processes’ is the
provision of aid to countries that American policy makers
deem to be cooperating with US anti-drug policies. Since
US financial assistance is given to countries as an incentive
to encourage bilateral cooperation in combating drug
trafficking, US aid may have a positive impact on the
amount of drugs that recipient countries actually interdict.
We, therefore, control for the level of US military and
economic assistance to CB countries in our sample.
However, as our general theory predicts, we expect that the
level of US aid to CB countries will not sufficiently induce
their cooperation in drug interdiction and will therefore not
have a significant impact on the level of cocaine and
marijuana seizures among them.
47
The Level of Andean - South
American Drug Interdiction
• Since CB countries serve as transit zones for drugs that
are largely produced in the Andean countries of Latin
America, the level of drug seizures that is made by law
enforcement personnel in Columbia, Bolivia and Peru
also affects the level of CB drug interdiction. We
therefore control for the level of Andean drug seizures.
We operationalize this variable as the ratio of drug
seizures among Andean countries as a proportion of
their law enforcement personnel. It is calculated by
dividing each Andean country’s yearly drug seizure of
cocaine and marijuana by the yearly number of their
active law enforcement personnel.
48
The Statistical Model
• The theoretical model as discussed can be written as a
regression equation in which the independent variables
are total coast line of CB countries (CBCOASTLINE), CB
countries distance from the U.S. (CBDISTANCE), South
American drug seizures (ANDEANSEIZURE), per capita
purchasing power parity in CB countries (CBINCOME),
government corruption in CB countries
(CBCORRUPTION), domestic illicit drug consumption in
CB countries (CBDRUGCON), domestic illicit drug
consumption in the U.S. (USDRUGCON), economic
trade openness of CB countries (CBECONOPENNESS),
individual political freedom and civil liberties
49
The Statistical Model (contd)
• in CB countries (CBPOLRIGHTS), institutional political
openness in CB countries (CBINSTOPENNESS), U.S.
aid to CB countries (USAID), U.S. per capita purchasing
power parity (USINCOME), and U.S. bilateral drug
interdiction agreement with CB countries
(US/CBBIAGREE). The unit of analysis in the statistical
model is the nation state where the analysis of the data
in each case is cocaine and marijuana interdiction in a
given year. In order to prevent negative predicted values
for drug interdiction, the level of drug interdiction is a
natural log transformation.
50
The Statistical Model (contd)
• Ln(DRUGINTERDICTIONit) =  +
1CBCOASTLINEit + 2CBDISTANCE1it
+ 3ANDEANSEIZURESit + 4CBINCOMEit +
5CBCORRUPTIONit + 6CBDRUGCONit +
7USDRUGCONit + 8CBECONOPENNESSit +
9CBPOLRIGHTSit + 10CBINSTOPENNESSit
+ 11USAIDit + 12USINCOMEit
+
13US/CBBIAGREEit + eit.
• where i is country and t is year.
• The estimation method is ordinary least squares
(OLS) method with panel corrected standard errors
(PCSE) (Beck, 2001; Beck and Katz, 1995).
51
The Results
• Table.1 presents the results of our analysis
of the effects of US/CB bi-lateral drug
interdiction agreements on cocaine and
marijuana seizures among CB countries
during 1984 to 2000 using OLS with PCSE
estimator
52