Defence and Peace Economics

ISSN: 1024-2694 (Print) 1476-8267 (Online) Journal homepage: http://www.tandfonline.com/loi/gdpe20

Armed Conflict, Military Expenses and FDI Inflow

to Developing Countries

Nusrate Aziz & Usman Khalid

To cite this article: Nusrate Aziz & Usman Khalid (2017): Armed Conflict, Military
Expenses and FDI Inflow to Developing Countries, Defence and Peace Economics, DOI:
10.1080/10242694.2017.1388066

To link to this article: http://dx.doi.org/10.1080/10242694.2017.1388066

Published online: 13 Oct 2017.

Submit your article to this journal

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=gdpe20

Download by: [DUT Library] Date: 14 October 2017, At: 10:09

 Defence and Peace Economics, 2017
https://doi.org/10.1080/10242694.2017.1388066

Armed Conflict, Military Expenses and FDI Inflow to Developing

Countries
Nusrate Aziza and Usman Khalidb
a
Department of Business and Economics, Algoma University, Sault Ste. Marie, Canada; bSchool of Economics,
University of Nottingham (Malaysia Campus), Semenyih, Malaysia

ABSTRACT ARTICLE HISTORY

This paper investigates the relationship between military expenditure and Received 24 June 2017
FDI inflow conditioning on the exposure of a country to armed conflict in Accepted 2 October 2017
Downloaded by [DUT Library] at 10:09 14 October 2017

the long run. We apply the band spectrum regression estimator, and the
KEYWORDS
maximal overlap discrete wavelet transform, to a panel of 60 developing Conflict; military expenses;
countries, for the years 1990 to 2013. The estimated results indicate that FDI; wavelet
military expenditure, in the absence of armed conflict, reduces FDI inflow.
However, the negative effect is mitigated by increased military expenditure, JEL CODES
in the presence of armed conflict. We also show that the effect of military F21; H56; D74; F51
expenditure on FDI is time sensitive, in that it takes time for military
expenditure to affect FDI inflow. FDI inflow in response to higher military
expenditure is higher for the country that faces higher armed conflict than
the country that faces lower armed conflict. The findings are robust in the
case of overall as well as internal conflict. These results are also robust to the
alternative specification, subsample analysis with different armed conflict
thresholds, and the estimation using the time variant long-run models.

Introduction
Foreign Direct Investment (FDI) is seen by policy-makers as a tool to support development, increase
productivity and advance technology in developing and emerging economies. However, to attract a
decent flow of FDI, a country must maintain peace and stability. A country with higher political insta-
bility is more likely to receive lower FDI than a country with lower political instability. Africa, which is
the most conflict-prone region in the world, is consequently the smallest receiver of FDI (Feenstra and
Taylor 2014). On the other hand, Europe, North America, China and most Asian countries that are safe
from major armed conflict, are amongst the highest receivers of FDI in the world.
Figure 1 shows FDI flows between different regions in the world, in the year 2010. The thickest lines
represent the largest stock of FDI. The figure reveals that the flow of foreign direct investment was
smallest from Europe and North America to Africa, while the flow within and between Europe and
North America was recorded to be the largest world over.
A common policy response to curtail conflict and unrest in a country is to increase military expend-
iture. The policy not only helps protect the state from internal and external conflicts but also leads to
other economic and commercial spin-offs, such as secure investor returns (Deger and Sen 1983). The
increase in military expenditure, however, entails opportunity costs such as an increase in government
expenditure, an increase in tax, an increase in the cost of imports (if the imports include weapons

CONTACT Nusrate Aziz [email protected], [email protected]

© 2017 Informa UK Limited, trading as Taylor & Francis Group
 2 N. AZIZ AND U. KHALID
Downloaded by [DUT Library] at 10:09 14 October 2017

Figure 1. Stock of Foreign Direct Investment, 2010 (US$ billions). Source: Feenstra and Taylor (2014).

and training), an increase in borrowing and expansion in the money supply (see, for example, Dunne,
Smith, and Willenbockel 2005). Furthermore, higher military expenditure not only takes away productive
investments from physical, economic and financial infrastructure but it also sends a preemptive warning
to potential multinational FDI investors to cease investment in the country from the fear of a possible
armed conflict. Therefore, if the costs from increasing military expenditure exceed the returns from it,
spending on the military would be economically irrational.
Non-conflicting countries, foresee a decrease in FDI resulting from an increase in military expenditure.
Insufficient investment in infrastructure, human capital and R&D are some of the factors which deter FDI
in the face of an increase in military expenditure. More importantly, the increased expenditure is viewed
by investors as a precautionary measure against a looming military conflict, which inevitably results in
decreased FDI inflow. Moreover, if a country without conflict spends a large amount on the military, it is
not unlikely that, in future, the country may impose a tax or any other monetary restriction on foreign
and domestic investors to maintain country’s military expenditure. This may lead to FDI outflows from
the country that spends more on the military to the country that spends relatively less on the military.
On the other hand, if the country is already under military threat, increasing military expenditure to
stabilise the economy helps boost foreign investors’ confidence in the prospects of a secure investment
return, leading to a higher FDI inflow to the country. In light of this discussion, the study hypothesises
that:
H1: Countries facing no armed conflict will receive a lower FDI inflow if they spend more (compared to other
countries facing no armed conflict) on the military.
H2: Countries facing either internal or external conflict will receive higher FDI inflow if they spend more (compared
to other countries facing either internal or external threat) on the military to stabilise their economy.
Moreover, the effect of military expenditure on FDI is likely to be time sensitive; it may take time for
military expenditure to reduce conflict and hence increase FDI inflow effectively. The effect of political
instability on FDI, therefore, cannot be determined for a point in time. It demands a thorough analysis
of events that have transpired over a span of several years. As such, political stability is a long run phe-
nomenon and it may take some time for the foreign investors to react to the symptoms of stability in
a country (Fatehi-Sedeh and Hossein Safizadeh 1989). We, therefore, hypothesise that:
H3: Peace and stability is a long-term phenomenon. It takes time for military expenditure to generate a secure
environment for FDI inflow.
 DEFENCE AND PEACE ECONOMICS 3

Several studies indicate that military expenditure crowds out investment in economically produc-
tive sectors, thereby lowering economic growth (Rothschild 1973; Smith 1980; Deger and Smith 1983;
Leontief and Duchin 1983; Lim 1983; Landau 1985; Mintz and Huang 1990; Ram 1995; Dunne, Nikolaidou,
and Smith 2002). Other studies show that military expenditure boosts business confidence in conflicting
countries, which facilitates investment and economic growth (Kennedy 1974; Benoit 1978; Whynes 1979;
Barro and Sala-i-Martin 2004; Dunne, Smith, and Willenbockel 2005). In retrospect, several studies sug-
gest that military expenditure has a non-linear effect on economic growth, conditional upon exposure
to conflict (Frederiksen and Looney 1982; Landau 1996; Cothren 2002; Aizenman and Glick 2006; Aziz
and Niaz Asadullah 2017). Economic growth is an outcome of both internal and foreign investment in
a country. According to the above literature, in the presence of conflict, a positive economic growth is
subject to positive growth in military expenditure. This leads to a pertinent question; is FDI inflow also
conditional on an increase in military expenditure in the face of armed conflict?
When planning on investing abroad, foreign investors do not only take the basic features of the host
country into account, such as infrastructure, natural resources, factor price, financial development and
financial capability for debt payments. They are also concerned with the assurance of a secure return
on foreign investment, which in turn leads to concerns about political uncertainty (Nigh 1985; Li 2006),
Downloaded by [DUT Library] at 10:09 14 October 2017

terrorism (Enders and Sandler 1996), armed conflict (Kim 2016; Lee 2017). Some research indicates
that political instability has a negative impact on FDI inflow (Agarwal 1980; Schneider and Frey 1985;
Woodward and Rolfe 1993; Loree and Guisinger 1995; Asiedu 2006). Kim (2016) finds that interstate
(external) conflict negatively affects FDI inflow in those developing countries which support the endur-
ing risk hypothesis – conflict makes conflict-prone countries persistently less attractive. Maizels and
Nissanke (1987) argue that FDI investors make investment decisions on the assumption that the host
country will offer high security backed by sufficient military expenditure. Li and Vashchilko (2010), and
Nigh (1985) find that armed conflicts reduce FDI while security alliances increase FDI. Li (2006) points
out that investors decrease investments ex-ante when anticipating any form of violence. However, they
only make adjustments ex-post when facing unanticipated violence.
However, economists and political scientists have paid limited attention to the relevance of military
expenditure and armed conflict to FDI flow. Nothing is known so far about the effect of military expend-
iture on FDI in the face of (internal and external) military conflict. Our research, therefore, aims to fill this
vacuum. We estimate the long run relationship between FDI inflows, military expenditure and conflict
using panel data consisting of 60 developing countries for the period 1990 to 2013, by applying the
band spectrum regression estimator. We find that military expenses without a military threat reduce
net FDI inflow. However, growth in military expenses increases net FDI inflow in the countries which
are under military threat. Our results are robust in the case of alternative specifications, subsamples
defined using the severity of military threat and estimations using alternative definitions of the long run.
The rest of the paper is organised as follows: Section ‘Theoretical Background Discusses’ the theo-
retical background, Section ‘The Data’ explains the data, Section ‘Methodology’ lays out the empirical
model and estimation procedure, Section ‘Results and Discussion’ reports and discusses the results, and
Section ‘Conclusion’ concludes the study.

Theoretical Background
The goal of multinational firms is to maximise profit (π). FDI from multinational firms is therefore based
on the productivity and costs of capital. The productivity of capital (K) for firm i (AKiα) is positively
affected by firm i’s safe investment. Based on the AK Model, a firm’s profit function can be written as
(also see Lee 2017):
𝜋i = AKi𝛼 −rKi ; A > 0, 𝛼 > 0 (1)
Here, r is the cost of capital. Denoting τ as political risk due to conflict, we can rewrite (1) as,
𝜋i = A(1 − 𝜏)Ki𝛼 −rKi ; 𝜏>0 (2)
 4 N. AZIZ AND U. KHALID

Productivity can be increased by technological innovation (A) and/or by reducing political risk. Control
of political risk through military expenses increases investor’s return.
Other things remaining equal, the profit for multinational firms in conflict-prone countries, therefore,
depends on τ. If τ is positive due to armed conflict, the return from FDI falls, which discourages multina-
tional enterprises to invest in a conflict-prone country. However, if an increase in military expenditure
can reduce τ sufficiently so that the net effect on FDI is positive, the profit for multinational firms would
be positive, and there would be higher FDI inflow in the conflict-prone region. We can show the impact
of an increase in military expenditure in a conflict-prone area as follows.
𝜋i = A(1 + (𝜃 − 𝜏))Ki𝛼 −rKi ; 𝜃>0 (3)
where θ is a security component due to military expenditure that reduces the effect of τ. An increase
in θ can reduce the security risk from military conflict, thus encouraging MNEs to send FDI and exploit
the benefits of other economic resources in a conflict-prone area.

The Data
Downloaded by [DUT Library] at 10:09 14 October 2017

We estimate our empirical model using a balanced panel consisting of 60 developing countries (see
Appendix A1) for the period spanning 1990 to 2013. Our dependent variable is net inflow of FDI1 (as
a percentage of GDP), data for which has been collected from the World Bank (WB) database. Data
on military expenditure and armed conflict have been obtained from the SIPRI Military Expenditure
Database and PRIO database, respectively. Data for all control variables have been obtained from the
World Bank database. Our main explanatory variables are military expenditure/GDP, i.e. military expenses
recorded as a percentage of GDP, and a dummy variable for armed conflict. A country is classified as
having an armed conflict in the PRIO database if there were 25 or more battle deaths in a year. Any
conflict which took place inside the geographical territory of a country by internal people is defined
as an internal conflict. Internal conflict often occurs between the government and opposition groups
within the country. However, if the conflict takes place between two different nations, it is considered
as external conflict. Armed conflict, internal conflict and external conflict have all been accounted for
by dichotomous dummies in our empirical models. Details about the variables have been provided in
Appendix A2. To avoid any country selection bias, both conflicting and non-conflicting countries have
been included in the model.
In general, armed conflict in developing countries has increased after the end of the cold war.
However, the developing world has faced more internal conflict than external conflict during the years
1990 to 2013. Net FDI inflow to developing countries, on average, has increased while military expend-
iture has fallen between 1990 and 2013 (see, Appendix A3(a)). Military expenditure (as a percentage
of GDP) in countries facing conflict was, on average, higher (2.26) than all other developing countries
(2.04), between 1990 and 2013. FDI (as a percentage of GDP) in countries facing conflict was, on average,
lower (1.93) than all developing countries (3.11) (see Appendix A3(b)).
Some notable stylised facts can be observed from the data. First, growth in military expenditure
reduces FDI inflow. If a country chooses to direct its resources to improve its military, it would have less
to invest in physical and financial infrastructure, human capital and R&D. The country fails to attract
high FDI flows as a result of weak infrastructure. A higher military spending, therefore, reduces the pos-
sibility of productive investment. Subsequently, higher military expenses lead to a lower FDI inflow. For
example, Belize faced no conflict between the years 1990 and 2013, and the country’s average military
expense was relatively low (military expenses/GDP = 1.07). As a result, Belize received very high FDI
(FDI/GDP = 5.75). Similarly, Kazakhstan faced no armed conflict during 1990 and 2013, and the country’s
small expenditure on the military (military expenditure/GDP = 1.04) attracted very high amounts of FDI
(FDI/GDP = 6.24) during this period. Similar economies that experienced high FDI when faced with no
conflict and observed low military expenditure are Bolivia, Brazil, Bulgaria, Dominican Republic, Gambia,
Ghana, Guyana, Mongolia, Swaziland and Tanzania, among others (see Appendix A4). Figure 2 confirms
this relationship as it clearly shows that military spending is negatively associated with FDI inflow.
 DEFENCE AND PEACE ECONOMICS 5

10
Average Net FDI Inflows (% of GDP)

8
6
4
2
Downloaded by [DUT Library] at 10:09 14 October 2017

0 2 4 6
Average Military Expenditure (% of GDP)

Figure 2. Relationship between military expenditure and FDI.

Second, if a country increases spending on the military, the foreign investors perceive this increase
as a signal for current or future political threat. Hence, even if the country faces no threat of an armed
conflict, multinational investors are discouraged to invest in the country which spends more on the
military. For example, Morocco faced no conflict for the years 1990 to 2013. However, the country
observed high military spending during this period (military expenditure/GDP = 3.54). Subsequently,
it received a very low amount of FDI (FDI/GDP = 1.56) during this period. Similarly, Zimbabwe faced no
conflict during 1990–2013 but observed high military spending (military expenditure/GDP = 3.55). In
line with our hypothesis, it received very low FDI (FDI/GDP = 1.32).
Third, conflict obstructs FDI inflow to a country as it fails to promise secure investment returns.
Conventional wisdom suggests that armed conflicts disrupt all types of investments, particularly domes-
tic and foreign investments, resulting from a lack of confidence among investors in returns generated
from a conflict-prone area. Armed conflict, therefore, reduces FDI inflow to a country. India, Indonesia,
Guatemala, Nepal, Pakistan, Philippines, Rwanda, Senegal, Sri Lanka and Turkey are all examples of
countries facing high armed conflict, and consequently receiving very low net FDI inflow, during our
sample period (see Appendix A4). Figure 3 confirms the negative relationship between armed conflict
FDI inflow. Countries with a higher number of armed conflicts receive relatively less FDI.
Lastly, the government can encourage foreign investors to invest in a conflict-prone area by spend-
ing more on military hardware. An increase in military spending to stabilise a conflict-prone economy
assures investors of a secure return on foreign investment, thereby encouraging them to increase
investment. Hence, if a country faces armed conflict but increases investment on the military to restore
stability, FDI inflow will increase. For example, Nepal foresaw 11 years of armed conflict between 1990
and 2013, alongside low levels of military expenditure (military expenditure/GDP = 1.30). FDI inflow
to Nepal for this period was recorded to be very low (FDI/GDP = 0.21). On the other hand, Uganda
faced 22 years of armed conflict during 1990–2013, but military expenditure in Uganda was high (mil-
itary expenses/GDP = 2.36). Consequently, Uganda’s FDI inflow was high (FDI/GDP = 3.13) during our
sample period. Colombia and Egypt both faced 24 and 6 years of armed conflict (during 1990–2013),
respectively. Colombian military expenditure (as a percentage of GDP) was 3.22, on average, and the
country’s military expenditure was as high as 4.45 in 1996. FDI inflow, on average, was 2.97 in Colombia,
 6 N. AZIZ AND U. KHALID

10
Average Net FDI Inflows (% of GDP)
8
6
4
2
Downloaded by [DUT Library] at 10:09 14 October 2017

0 .2 .4 .6 .8 1
Average Conflict

Figure 3. Relationship between armed conflict and FDI.

attaining an all-time high of 6.98 in 2005. Egypt also spent a large amount on military hardware (military
expenditure/GDP = 3.16) and the country’s military spending was as high as 4.86 in 1991. FDI inflow
in Egypt was 2.32, on average, going as high as 9.34 in one year. Both Colombia and Egypt faced high
armed conflict, however, owing to high military expenditure, they managed to attract high FDI inflow.
We, therefore, can conclude (from the data analysis) that a country may be able to attract multina-
tional investors due to the availability of economics resources and lower costs of production, it loses
out on investment in the face of conflict. However, if the government manages to build investors’ con-
fidence by assuring high productivity and secure returns in the future, it can maintain high FDI inflow
despite the threat of conflict.

Methodology
Based on the theoretical model, our core hypothesis is that higher military expenditure (ME) increases
FDI inflow to a conflict-prone country. The effect of military expenditure on FDI, however, is likely to
be time sensitive; as such, political stability/instability is a long run phenomenon, and it takes time for
the foreign investors to react to the symptoms of stability/instability in a country. Hence, although we
estimate both the short-term and long-term effects of military expenditure and conflict on net FDI
inflows, our main focus shall be the long run results.
The basic empirical equation that we estimate is the following:
FDILRjt = 𝛽j1 + 𝛽2 MELRjt + 𝛽3 MELRjt × conflictLRjt + 𝛽4 conflictLRjt + 𝛽5 ZLRjt + 𝜀LRjt , (4)
where Z refers to all control variables. We include country fixed effects to control for any unobserved
heterogeneity across countries. The country fixed effects control for differences in factors such as nat-
ural resource endowment and geography. In Equation (4), if β2 < 0 but β3 > 0, we can conclude that in
the long run,

𝜕FDILRjt
= 𝛽2 + 𝛽3 ConflictLRjt > 𝛽2
𝜕MELRjt
 DEFENCE AND PEACE ECONOMICS 7

In this case, the negative effect of military expenditure on FDI is reversed by higher military expenses
in a conflict-prone country, when compared with countries facing no conflict.
We estimate Equation (4) using the band spectrum regression estimator (Engle 1974; Andersson 2016),
as it allows us to distinguish between short-term and long-term effects. The band spectrum regression
estimator is a two-step estimator. In the first step, all variables are decomposed into their short-run
and a long-run components using a band-pass filter. The second step is to estimate the regression
model using the decomposed data to obtain long-run parameter estimates. For the band-pass filter,
we use the Maximal Overlap Discrete Wavelet Transform (MODWT), because of its desirable properties.
MODWT is most suitable for time-series data that possibly contains structural breaks, outliers and other
non-recurring events that may otherwise adversely affect the decomposition of the data (Percival and
Walden 2006).2 We define long-run as persistent changes lasting more than four years.3 We also test the
robustness by defining long-run changes lasting more than eight years.4 This approach has previously
been used in several studies (see, for example, Andersson, Edgerton, and Opper 2013; Andersson and
Karpestam 2013). By construction, the decomposition of the data into short-run and long-run variations
is linear:
Downloaded by [DUT Library] at 10:09 14 October 2017

Xjt = XSRjt + XLRjt

This means that using the decomposition, we also get the short-run component of the variables.
Therefore, as a robustness check, the short-run parameters are also estimated.
Endogeneity is not likely to be an issue in our study for the following reasons. Firstly, military expend-
iture observed by a country is highly unlikely to be influenced by the investment of multinational
enterprises (MNEs). Secondly, although both FDI and armed conflict in resource-rich countries may be
sensitive to the natural resource endowment, investment by MNEs is unlikely to affect armed conflict
directly. Moreover, oil-rich countries facing conflict are not in our sample. Thirdly, we are only looking at
the long run, which means that whenever there is a conflict the FDI will go out of the country because
of insecurity in the long run even though the conflict may have arisen due to the inflow of FDI in the
country. Despite this, we performed an endogeneity test proposed by Baum, Schaffer, and Stillman
(2007). The null hypothesis for this test is that the conflict variable which is treated as endogenous
regressor can actually be treated as exogenous. Based on the result of the test we fail to reject the null
hypothesis concluding that the conflict variable is exogenous.5

Results and Discussion

Table 1 presents the estimated results. The first three columns present the simple model without any
wavelet transformation. As expected, there is a negative effect of military expenditure on FDI, sug-
gesting that as the country increases military expenditure, its FDI decreases. However, the coefficient
on military expenditure is significant only when we add both conflict and its interaction with military
expenditure. The interaction between conflict and military expenditure is positive but insignificant.
Next, we test our hypothesis for the long run by decomposing data into the long-run using MODWT.
Column 4–6 presents the results where the long-run is defined as ‘changes that last for more than four
years’. Overall, each respective model’s explanatory power, as estimated by the R2, is relatively high
(0.25 and 0.27) for the long-run, as compared to the models without any decomposition. The lower
explanatory power for the models without decomposition indicates that fluctuations in short-run FDI
mask the relationship between FDI, military expenditure and conflict.
Looking at our complete model in column 6, the coefficient on military expenditure is negative and
significant, and the size of the coefficient is almost twice as large in the long run as compared to the
model without decomposition in column 3. The result suggests that an increase in military expenditure
would result in a decrease in the net FDI inflows for countries facing no conflict. A one-percentage point
increase in the military expenditure as a percentage of GDP will reduce the net FDI inflows to a country
by almost 0.85 percentage points. This result supports the argument that higher military expenditure
 8 N. AZIZ AND U. KHALID

Table 1. Relationship between conflict, military expenditure, and FDI.

Downloaded by [DUT Library] at 10:09 14 October 2017

Note: Robust clustered standard errors in parentheses.

a
Long-run is defined as a period of more than four years.
*p < 0.1; **p < 0.05; ***p < 0.01.

in the absence of a threat, takes away productive investments in the infrastructure and other sectors
of the economy, hence discouraging foreign investors from investing in the host country.
The relationship between military expenditure and net FDI inflows is likely to be non-linear as coun-
tries that experience conflict are likely to benefit by increasing their military expenditure. This effect
is captured by the interaction term between military expenditure and conflict. The interaction term is
positive and significant, implying that the relationship between military expenditure and FDI inflows
depends on whether the country experiences any conflict or not. This shows that the negative effect
of military expenditure on net FDI inflows is mitigated for a country experiencing perpetual conflict.
Figure 4 helps visualise the size of the mitigation effect. It depicts the marginal effect of military
expenditure on net FDI inflows when the variable conflict includes both external and internal conflict. It
is evident from Figure 4 that as the probability of conflict increases, higher military expenditure attracts
higher FDI. More specifically ceterus paribus, a one percentage point increase in military expenditure
leads to a 27 per cent higher net FDI inflow in a country, with a probability of conflict of 0.2, as compared
to a country facing no conflict. This difference in FDI inflow between a conflicting and a non-conflicting
country supports the view that increased military expenditure in a conflicting country acts as a signal
to foreign investors that the country is safe to invest. The long run significant effect also indicates that
it takes time for the investors to respond to the stable political environment of a country.
The coefficient for the conflict variable also has the expected negative sign indicating the FDI inflows
will be lower for a country experiencing conflict; however, it is statistically insignificant. Moreover, all
the control variables have expected sign. Net FDI inflows will be higher for a country with higher GDP
per capita ceteris paribus. Similarly, a country with high levels of capital (% of GDP) reflecting good
infrastructure quality will attract higher FDI inflows.

Robustness Checks
We perform several robustness checks to scrutinise the results. First, we exclude countries that have
faced external conflict to determine if our results are driven by a certain type of conflict.6 Second, the
long-run model is re-estimated by defining the long run as fluctuations lasting for more than eight years.
 DEFENCE AND PEACE ECONOMICS 9

1
Mean of conflict = 0.23

Marginal Effect of Milex on FDI

.5
0
-.5
-1
-1.5
Downloaded by [DUT Library] at 10:09 14 October 2017

0 .2 .4 .6 .8 1
Probability of Conflict
Dashed lines give 90% confidence interval.

Figure 4. Marginal effect of military expenditure on FDI for different probabilities of conflict.

Table 2. Robustness checks.

Note: Robust clustered standard errors in parentheses.

a
Long-run is defined as fluctuations lasting for more than 8 years in column 1 and 3.
b
Subsample in column 4 include countries which had more than six years of conflict while in column 6 countries with more than
10 years of conflict is considered.
c
Short-run is defined as fluctuations lasting for less than 4 years.
*p < 0.1; **p < 0.05; ***p < 0.01.

Thirdly, we estimate our model for a subsample of countries that have experienced chronic conflict, i.e.
countries facing a conflict for more than six years or more than ten years. Lastly, we estimate our model
for the short-run to confirm if there is a difference in the short-run and long-run results.
Table 2 presents the results of the robustness checks. The results are robust to changes in the defi-
nition of long-run. Column 1 presents the results when we redefine long-run as fluctuations lasting
 10 N. AZIZ AND U. KHALID

1
Mean of internal conflict = 0.22

Marginal Effect of Milex on FDI

.5
0
-.5
-1
-1.5
Downloaded by [DUT Library] at 10:09 14 October 2017

0 .2 .4 .6 .8 1
Probability of internal conflict
Dashed lines give 90% confidence interval.

Figure 5. Marginal effect of military expenditure on FDI for different probabilities of internal conflict.

for more than eight years. The results are similar to the model presented in Column 6 of Table 1. The
coefficient on military expenditure is negative and significant whereas, the coefficient on the interaction
term is positive and significant. Moreover, there is an improvement in the fit of the model indicated by
the value of R2 going up to 0.30. The findings above lend support to our hypothesis that in the long-
run military expenditure leads to higher FDI inflows for conflicting countries, as it ensures safe returns
to investment.
When we exclude the observations for external conflict, the interaction term between military
expenditure and internal conflict) remains positive and significant (Column 2 and 3). These results
confirm that the negative effect of military expenditure on net FDI inflows is mitigated in conflicting
countries irrespective of the type of conflict. The mitigation effect is clearly reflected in Figure 5 as well.
The marginal effect of military expenditure on FDI inflows is less negative for a country facing conflict as
compared to a country with no conflict. Columns 4 and 5 contain the results obtained for a subsample
of countries. In Column 4, the model is estimated using a subsample of countries that have faced more
than 6 years of conflict, whereas Column 5 depicts the results for a subsample of countries which have
faced more than ten years of conflict. The results in both cases are similar to our initial results, providing
further evidence of the robustness of our model and estimates.
In Column 6, the estimated results for the short-run are presented. Military expenditure and its interac-
tion term are both insignificant. Moreover, the sign of the coefficient on the interaction term is now negative,
in contrast to earlier estimate obtained in the long-run model. In addition to this, the R2 is relatively small
for the short-run model, affirming that the effect of military expenditure on FDI is a long-run phenomenon.

Conclusion
This study investigates the hypothesis that military expenditure without a threat negatively influences
the FDI inflow, whereas, in the presence of a threat increased military expenditure may help attract
higher FDI inflow. We estimate an empirical model using a balanced panel consisting of 60 developing
countries for the period 1990 to 2013.We use band spectrum regression, and the maximal overlap dis-
crete wavelet transform for estimation. The study estimates for both the short and the long run, with a
 DEFENCE AND PEACE ECONOMICS 11

focal analysis of the long run results. We also examine the time sensitivity of data by defining the long
run as a period of more than four years as well as a period of more than eight years. We also estimate
the FDI model taking different types of armed conflict (overall conflict or internal conflict) into account.
The estimated results indicate that an increase in military expenditure without any armed conflict
reduces FDI inflow. We posit that the results obtained are such due to the following reasons. Firstly,
higher military expenses reduce investment in countries’ physical, economic and financial infrastructure
that makes investing in the host country an unfeasible option for foreign investors. Secondly, higher
investment in military hardware in a country transmits a negative signal to foreign investors of a risk
of unsafe returns on investment due to a potential military threat.
An increase in military expenditure in the face of armed conflict, on the other hand, increases FDI inflow.
This positive effect emerges due to an increased confidence amongst investors about a safe return from FDI
in the conflict-prone area. We also show that the effect of military expenditure on FDI is time sensitive, i.e. it
takes time for military expenditure to affect FDI inflow. FDI inflow in response to higher military expenditure
is higher to the country that faces higher armed conflict than the country that faces lower armed conflict.
The findings are robust in the case of overall conflict and internal conflict. Our results are also robust to
the alternative specification, subsample analysis and estimation using different definitions of long-run.
Downloaded by [DUT Library] at 10:09 14 October 2017

In light of the findings discussed above, the study suggests policy-makers to abstain from budgeting
high expenditures on a country’s defence in the absence of a significant political threat, so that FDI flows are
encouraged into the country. In case of a significant military threat, however, it is imperative that the gov-
ernment spends a sufficient amount on military resources in order to attract and sustain higher FDI inflow.

Notes
1. 
FDI net inflows are the value of inward direct investment made by non-resident investors in the reporting economy.
2. 
For more information about the MODWT see Percival and Walden (2006), Crowley (2007) and Andersson (2008);
Although it is possible to decompose data using several different methods, such as the Hodrick-Prescott filter or
Fourier transformation, wavelet transformation offers considerable advantages. Wavelet decomposition combines
time and frequency domains. It is localized both in time and in frequency, which preserves the time domain and
frequency domain information of the original series (Maslova, Onder, and Sanghi 2013). Thus, it does not introduce
phase shifts that change the location of events in time and allows for the observation of structural breaks, outliers
and nonlinearities in the data series (Ramsey 1999; Percival and Walden 2006). Unlike the Hodrick-Prescott filter,
in which data are decomposed into short and long runs, the filtered time horizons are known. This data-driven
technique for separating short and long runs ensures that short-term fluctuations arising from measurement
errors in the data do not affect the results.
3. 
This definition of long run is commonly used in the literature (see e.g. Andersson and Karpestam 2013; Khalid 2016).
4. 
We find similar results using eight years as our definition of long-run.
5. 
The test results are available on request.
6. 
The effect of external conflict has not been examined (separately) due to lack of enough observations.

Disclosure statement
No potential conflict of interest was reported by the authors.

ORCID
Usman Khalid http://orcid.org/0000-0001-9879-6948

References
Agarwal, Jamuna P. 1980. “Determinants of Foreign Direct Investment: A Survey.” Review of World Economics 116 (4): 739–773.
Aizenman, Joshua, and Reuven Glick. 2006. “Military Expenditure, Threats, and Growth.” The Journal of International Trade
& Economic Development 15 (2): 129–155.
Andersson, F. N. G. 2008. “Wavelet analysis of economic time series.” diss., Lund Economic Studies, 149.
Andersson, F. N. G. 2016. “Identifying and Modelling Cycles and Long Waves in Economic Time Series.” In Structural Analysis
and the Process of Economic Development, edited by Jonas Ljungberg, 34–55. New York: Routledge.
 12 N. AZIZ AND U. KHALID

Andersson, F. N. G., David Edgerton, and Sonja Opper. 2013. “A Matter of Time: Revisiting Growth Convergence in China.”
World Development 45: 239–251.
Andersson, F. N. G., and Peter Karpestam. 2013. “CO2 Emissions and Economic Activity: Short- and Long-Run Economic
Determinants of Scale, Energy Intensity and Carbon Intensity.” Energy Policy 61: 1285–1294.
Asiedu, Elizabeth. 2006. “Foreign Direct Investment in Africa: The Role of Natural Resources, Market Size, Government Policy,
Institutions and Political Instability.” The World Economy 29 (1): 63–77.
Aziz, M. Nusrate, and M. Niaz Asadullah. 2017. “Military Spending, Armed Conflict and Economic Growth in Developing
Countries in the Post-Cold War era.” Journal of Economic Studies 44 (1): 47–68.
Barro, R. J., and X. Sala-i-Martin. 2004. EconomicGrowth. 2nd ed. Cambridge, MA: The MIT Press.
Baum, F. Christopher, Mark E. Schaffer, and Steven Stillman. 2007. “Enhanced Routines for Instrumental Variables/Generalized
Method of Moments Estimation and Testing.” The Stata Journal 7 (4): 465–506.
Benoit, Emile. 1978. “Growth and Defense in Developing Countries.” Economic Development and Cultural Change 26 (2): 271–280.
Cothren, Richard. 2002. “A Model of Military Spending and Economic Growth.” Public Choice 110 (1/2): 121–142.
Crowley, Patrick M. 2007. “A Guide to Wavelets for Economists.” Journal of Economic Surveys 21 (2): 207–267.
Deger, Saadet, and Ron Smith. 1983. “Military Expenditure and Growth in Less Developed Countries.” Journal of Conflict
Resolution 27 (2): 335–353.
Deger, Saadet, and Somnath Sen. 1983. “Military Expenditure, Spin-off and Economic Development.” Journal of Development
Economics 13 (1–2): 67–83.
Dunne, J. Paul, Eftychia Nikolaidou, and Ron Smith. 2002. “Military Spending, Investment and Economic Growth in Small
Downloaded by [DUT Library] at 10:09 14 October 2017

Industrialising Economies.” South African Journal of Economics 70 (5): 789–808.

Dunne, J. Paul, Ron P. Smith, and Dirk Willenbockel. 2005. “Models of Military Expenditure and Growth: A Critical Review.”
Defence and Peace Economics 16 (6): 449–461.
Enders, Walter, and Todd Sandler. 1996. “Terrorism and Foreign Direct Investment in Spain and Greece.” Kyklos 49 (3): 331–352.
Engle, Robert F. 1974. “Band Spectrum Regression.” International Economic Review 15 (1): 1–11.
Fatehi-Sedeh, Kamal, and M. Hossein Safizadeh. 1989. “The Association between Political Instability and Flow of Foreign
Direct Investment.” Management International Review 29 (4): 4–13.
Feenstra, R. C., and A. M. Taylor. 2014. International Economics, 19–20. New York: Worth Macmillan, 19–20.
Frederiksen, Peter C., and Robert E. Looney. 1982. “Defense Expenditures and Economic Growth in Development Countries:
Some Further Empirical Evidence.” Journal of Economic Development 7 (1): 113–126.
Kennedy, G. 1974. The Military in Third Word. London: Duckworth.
Khalid, U. 2016 Catch-up in Institutional Quality: An Empirical Assessment. CREDIT Research Paper, (No. 2016-04). Nottingham:
University of Nottingham.
Kim, Dongkyu. 2016. “The Effects of Inter-State Conflicts on Foreign Investment Flows to the Developing World: Enduring
Vs Ephemeral Risk of Conflicts.” International Political Science Review 37 (4): 422–437.
Landau, Daniel L. 1985. “Government Expenditure and Economic Growth in the Developed Countries: 1952–76.” Public
Choice 47 (3): 459–477.
Landau, Daniel. 1996. “Is One of the ‘Peace Dividends’ Negative? Military Expenditure and Economic Growth in the Wealthy
OECD Countries.” The Quarterly Review of Economics and Finance 36 (2): 183–195.
Lee, Hoon. 2017. “Does Armed Conflict Reduce Foreign Direct Investment in the Petroleum Sector?” Foreign Policy Analysis
13 (1): 188–214.
Leontief, Wassily W., and Faye Duchin. 1983. Military Spending: Facts and Figures, Worldwide Implications, and Future Outlook.
New York, NJ: Oxford University Press.
Li, Quan. 2006. “Political Violence and Foreign Direct Investment.” Research in Global Strategic Management 12: 231–255.
Li, Quan, and Tatiana Vashchilko. 2010. “Dyadic Military Conflict, Security Alliances, and Bilateral FDI Flows.” Journal of
International Business Studies 41 (5): 765–782.
Lim, David. 1983. “Another Look at Growth and Defense in Less Developed Countries.” Economic Development and Cultural
Change 31 (2): 377–384.
Loree, David W., and Stephen E. Guisinger. 1995. “Policy and Non-Policy Determinants of US Equity Foreign Direct
Investment.” Journal of International Business Studies 26 (2): 281–299.
Maizels, Alfred, and Machiko K. Nissanke. 1987. “The Causes of Military Expenditure in Developing Countries.” In Saadet
Deger and Robert West’s Defence Security and Development. London: Frances Pinter, 129–140.
Maslova, I., H. Onder, and A. Sanghi. 2013. “Growth and Volatility Analysis using Wavelets.” World Bank Policy Research
Working Paper (6578). The World Bank.
Mintz, Alex, and Chi Huang. 1990. “Defense Expenditures, Economic Growth, and the ‘Peace Dividend’.” American Political
Science Review 84 (4): 1283–1293.
Nigh, Douglas. 1985. “The Effect of Political Events on United States Direct Foreign Investment: A Pooled Time-Series Cross-
Sectional Analysis.” Journal of International Business Studies 16 (1): 1–17.
Percival, Donald B., and Andrew T. Walden. 2006. Wavelet Methods for Time Series Analysis. Vol. 4. Cambridge: Cambridge
University Press.
Ram, Rati. 1995. “Defense Expenditure and Economic Growth.” In Handbook of Defense Economics, edited by Todd Sandler
and Keith Hartley, 251–274. Amsterdam: Elsevier.
 DEFENCE AND PEACE ECONOMICS 13

Ramsey, J. B. 1999. “The Contribution of Wavelets to the Analysis of Economic and Financial Data.” Philosophical Transactions
of the Royal Society of London a: Mathematical, Physical and Engineering Sciences 357 (1760): 2593–2606.
Rothschild, Kurt W. 1973. “Military Expenditure, Exports and Growth.” Kyklos 26 (4): 804–814.
Schneider, Friedrich, and Bruno S. Frey. 1985. “Economic and Political Determinants of Foreign Direct Investment.” World
Development 13 (2): 161–175.
Smith, Ronald P. 1980. “Military Expenditure and Investment in OECD Countries, 1954–1973.” Journal of Comparative
Economics 4 (1): 19–32.
Whynes, David K. 1979. The Economics of Third World Military Expenditure. Springer: University of Texas Press.
Woodward, Douglas P., and Robert J. Rolfe. 1993. “The Location of Export-Oriented Foreign Direct Investment in the
Caribbean Basin.” Journal of International Business Studies 24 (1): 121–144.

Appendices

Appendix A1. List of countries (60 countries)

Downloaded by [DUT Library] at 10:09 14 October 2017

Albania India*** Pakistan***

Algeria*** Indonesia*** Papua New Guinea**
Bangladesh** Jordan Paraguay
Belize Kenya** Peru**
Bolivia Kazakhstan Philippines***
Botswana Kyrgyz Republic Romania
Brazil Lesotho Rwanda***
Bulgaria Malawi Senegal**
Burundi** Malaysia Sierra Leone***
Cameroon Mali** South Africa
China Mauritania Sri Lanka***
Colombia** Mauritius Swaziland
Dominican Republic Mexico Tanzania
Egypt** Moldova Thailand**
El Salvador Mongolia Tunisia
Ecuador Morocco Turkey***
Gambia Mozambique** Uganda***
Ghana Namibia Ukraine
Guatemala** Nepal*** Venezuela
Guyana Nicaragua Zimbabwe
** Countries with more than 3 years and less than 10 years of conflict;
*** Countries with more than 10 years of conflict.

Appendix A2. Variables and sources of data

Name Detail Source

FDI FDI inflow and GDP ratio World Development Indicators of the World Bank
Conflict Both internal and external armed The Uppsala Conflict Data Program, UCDP Datasets - Uppsala
conflict University, Sweden
Conflict_in Internal conflict UCDP
Conflict_ex External conflict UCDP
Military expenditure/ Military expenditure and GDP The SIPRI Military Expenditure Database (2015), Stockholm Inter-
GDP ratio national Peace Research Institute, Sweden
Conflict × Military Conflict multiplied by military Same
expenditure expenditure/GDP
Conflict_in × Military Internal conflict multiplied by Same
expenditure military expenditure/GDP
GDP_growth_rate GDP growth rate World Development Indicators of the World Bank
K_GDP Capital (% of GDP) World Development Indicators of the World Bank
Labour force Labour force participation rate World Development Indicators of the World Bank
participation
 14 N. AZIZ AND U. KHALID

Appendix A3(a). Descriptive statistics

Variable Observations Mean Std. dev. Min Max

FDI (% GDP) 1440 3.112 4.506 −9.658 45.273
Δ FDI (% GDP) 1380 0.162 3.170 −32.081 36.144
Military expenditure (% of GDP) 1440 2.044 1.315 0.0845 10.675
ΔMilitary expenditure (% of GDP) 1380 −0.045 0.473 −7.274 6.444
Conflict 1440 0.231 0.421 0 1
Capital (% of GDP) 1440 21.837 7.685 0.293 74.821
Labour force participation 1440 67.734 10.745 41 91.5
GDP Per capita 1440 6077.226 4515.85 434.600 22,589.37
GDP growth rate 1440 3.844 5.065 −50.248 35.224

Appendix A3(b). Summary Statistics for conflicting countries (23 countries)

Variable Obs Mean Std. dev. Min Max

Military expenditure (% of GDP) 552 2.26 1.32 0.085 7.18
Downloaded by [DUT Library] at 10:09 14 October 2017

FDI (% of GDP) 552 1.93 3.52 −5.01 41.81

Conflict 552 13.78 7.355 3 24

Appendix A4. Armed conflict, military expenditure and FDI of conflicting countries,
1990–2013

Country Armed conflict (years) Military expenses (% of GDP) FDI (% of GDP)

Major conflict (≥10 years)
Algeria 23 3.256 0.923
Burundi 16 4.228 0.121
Colombia 24 3.216 2.967
India 24 2.717 1.064
Indonesia 12 0.947 1.043
Nepal 11 1.301 0.213
Pakistan 22 4.853 1.199
Peru 13 1.526 3.704
Philippines 24 1.632 1.486
Rwanda 16 2.965 0.808
Senegal 10 1.597 1.734
Sierra Leone 11 1.898 4.411
Sri Lanka 18 3.601 1.260
Thailand 11 1.809 2.939
Turkey 24 3.225 1.170
Uganda 22 2.359 3.130
Moderate conflict (≥5 years <10 years)
Egypt 6 3.159 2.318
Guatemala 6 0.734 0.721
Mali 7 1.654 2.056
Papua New Guinea 6 0.964 2.105
Minor conflict (<5 years)
Bangladesh 4 1.169 0.538
Cameroon 1 1.381 1.114
China 1 2.036 3.775
Ecuador 1 2.026 1.550
El Salvador 2 1.587 1.816
Kenya 3 1.638 0.540
Kyrgyz Republic 1 2.780 3.581
Lesotho 1 3.127 8.548
Malaysia 1 2.193 4.320
Mauritania 2 2.920 6.768
Mexico 2 0.500 2.417
Moldova 1 0.554 4.216
Mozambique 4 1.480 7.938
Nicaragua 1 1.003 4.341
Venezuela 1 1.447 1.965