Military Expenditure and Economic Growth: Evidence from India and Pakistan

 

Rajeshwari. UR

Assistant Professor, Department of Economics, CHRIST (Deemed to be University), Bangalore-29.

*Corresponding Author E-mail: rajeshwari.ur@christuniversity.in

 

ABSTRACT:

This study explores the relationship between military spending, education expenditure and health expenditure in India and Pakistan over the time period of 2000-2018. The data has been collected from SIPRI and World Bank data base. Augmented Dickey Fuller Test has been used for unit root. VAR leg length criteria have been used to identify the ideal lag for the model. As all the variables are either I(0) or I(1), Auto Regressive Distributed Lag model (ARDL) has been used to analyze the relationship between the variables. Bounds test is used to analyze the cointegration among the variables. The result shows that there exists long run relationship between the variables in both countries. But, in India the causality runs from military spending to GDP where as in Pakistan the causality runs from GDP to military spending in the long run. Further, in India there is no evidence of short run relationship between the variables where as in Pakistan there exist positive relationship between the GDP and military expenditure in the short run.

 

KEYWORDS: Defense, Education, Health, military spending, India, Pakistan.

 

 


INTRODUCTION:

Military expenditures are one of the most important concerns of developed and developing countries as they spend a huge amount of money on defense sector.  According to SIPRI 2019-year book, global military expenditure in 2018 has been estimated to $1822 billion which accounts for 2.1% of world’s Gross Domestic Product. At global level, five biggest spenders in 2018 were USA, China, Saudi Arabia, India and France (SIPRI, 2019). These countries accounted for nearly 60% of the global spending.

 

Military Keynesianism considers that military expenditure stimulates the economic growth because it is a part of fiscal policy and on the other hand many economists argue that the military expenditure crowds out more productive expenditure and reduces the civilian investment which converts into negative impact on economic growth. It is also argued that military expenditure can reduce resources for productive sectors like education, health and other development projects. Therefore, a potential problem of trade off exists between military expenditure and other type of developmental expenditures. Though a large literature before early 1970 showed negative relationship between these variables, Emile Benoit showed positive effects of higher defense spending on economic growth. Therefore, there is no consensus find among the researchers on the economic outcome of the military spending. The works which shows positive effect mainly focus on conduit of security, aggregate demand, investment and labor where as the studies that shows negative effect focus on crowding out effect, inefficiency of resource allocation and increased political power.

 

 

Like many developed and developing nations India and Pakistan also spends a huge part of their revenue on defense in order to maintain a credible level of security due to its geopolitical position and a long run dispute over territory of Kashmir. Since 1947, India and Pakistan shares rivalry relationship which led to an arm race. This implies that both countries military expenditure is determined in an action reaction framework. Facts tell us that there exists a long run arms race between India and Pakistan. With continued tension in Kashmir, military confrontation between nuclear armed neighbors India and Pakistan remain high. In this context, it is important to study the relationship between military spending and economic growth as it has opportunity cost in terms of spending on education and health.

 

REVIEW OF LITERATURE:

There are number of literatures available in examining the relationship between military spending and economic growth considering mainly two important variables i.e. Gross Domestic Product and military expenditure. Hasan Raju and Ahmed (2019) evidences for India, Pakistan and China that there is positive long run relationship and no short run relationship between military spending and economic growth. The study also shows unidirectional long run causality in each of the cases. Azmair, Hussain, Abbassi and Gohar (2018) show short term relationship between the economic growth and military spending in Pakistan. Ali and Ather (2014) investigated defense burden on economic growth in Pakistan over the period 1980-2013. The results show that military burden has direct as well as indirect negative effect on the economic growth.  Qureshi and Khan (2017) demonstrate negative relationship between GDP and military expenditure during low economic growth and positive economic growth during higher economic growth.

 

Tiwari, Shahbaz and Muhammad (2011) reinvestigated the relationship between defense spending and economic growth in India which showed long run relationship between the variables and positive effect of defense spending on economic growth. The result also points that there exist bidirectional relationship between the variables. Tiwari and Tiwari (2010), Jariwala (2017)   also confirm bidirectional relationship between GDP and defense expenditure.

 

From the review of literature shows continuing argument about the impact of military expenditure on economic growth and therefore results cannot be generalized across different countries over different time period. This present study tries to contribute to the existing literature by providing empirical evidence from India and Pakistan.

Research Gap:

Though numerous studies available, most of the studies tried to find out the relationship between two major variables GDP and military expenditure. This study tries to explain the relationship between military expenditure, GDP along with two additional variables education expenditure and health expenditure.

 

OBJECTIVE:

The main objective of the study is:

1.     To analyze the relationship between economic growth, military spending, education expenditure and health expenditure in India and Pakistan

 

HYPOTHESIS:

H0: There is no relationship between economic growth, military expenditure, education expenditure and health expenditure.

H1: There is a relationship between economic growth, military expenditure, education expenditure and health expenditure.

 

METHODOLOGY:

This study uses secondary data which is collected from Stockholm International Peace Research Institute (SIPRI) and World Bank Data Base. The variables considered are GDP, military spending as percentage of GDP, General Government Expenditure on Education as percentage to GDP, Health Spending as Percentage of GDP and HDI values. For the econometric models the data is used from 2000-2018. Augmented Dickey Fuller Test (ADF) has been used for unit root. VAR lag length criteria are used to identify ideal lag lengths. Since the variables are combination of I(0) and I(1), ARDL model is used for the analysis purpose.

 

Empirical Results:

Unit Root Test:

Augmented Dickey Fuller (ADF) Test has been used to test unit root. Following Table 1 shows the results of ADF test for India and Pakistan.  According to the result, GDP and health expenditure has unit root, but it becomes stationary after the first difference. This implies that GDP and health expenditure is I(1). All other variables found to be significant at level which implies that other variables are I(0). Similarly, the ADF result for Pakistan has been presented in Table 2.  According to the result, Education expenditure and HDI values has unit root and other variables are significant at level. As the results shows, the variables are either I(0) or I(1) we can use ARDL model to analyze the relationship between these variables in both the countries. 

 

Table 1: ADF test Results India

Variable

Level

First Difference

Decision about Integration

 

Intercept

Trend and Intercept

None

Intercept

Trend and Intercept

None

LNGDP

0.90

(0.99)

-2.70

(0.24)

18.35

(0.99)

-3.64

(0.01)*

-3.55

(0.06)***

-0.39

(0.52)

I(1)

LNME

-2.05

(0.26)

-3.94

(0.03)**

-0.96

(0.28)

-4.34

(0.00)*

-4.19

(0.02)**

-4.22

(0.00)*

I(0)

LNEDU

-3.97

(0.00)*

-1.28

(0.85)

-0.45

(0.50)

-2.51

(0.12)

-2.76

(0.22)

-2.57

(0.01)*

I(0)

LNHEA

-1.42

(0.54)

-1.31

(0.85)

-0.76

(0.37)

-3.93

(0.00)*

-4.18

(0.02)**

-3.83

(0.00)*

I(1)

*denotes significant at 1% ** denotes significant at 5% *** is significant at 10%

 

 

Table 2: ADF test Results Pakistan

Variable

Level

First Difference

Decision about Integration

 

Intercept

Trend and Intercept

None

Intercept

Trend and Intercept

None

 

LNGDP

-0.23

(0.91)

-4.79

(0.00)*

3.55

(0.99)

-3.77

(0.01)*

-3.54

(0.07)*

-0.08

(0.64)

I(0)

LNME

-2.80

(0.08)***

0.73

(0.99)

-0.21

(0.57)

-0.92

(0.75)

-3.64

(0.06)***

-1.0.5

(0.24)

I(0)

LNEDU

-1.84

(0.34)

-2.43

(0.35)

0.94

(0.90)

-4.89

(0.00)*

-4.73

(0.00)*

-4.83

(0.00)

I(1)

LNHEA

-2.94

(0.06)***

-2.81

(0.21)

-0.16

(0.60)

-2.93

(0.06)***

-2.95

(0.17)

-3.04

(0.00)*

I(0)

*denotes significant at 1%          ** denotes significant at 5%      *** is significant at 10%

 

After the unit root test, for further analysis VAR lag length criteria has been used to identify ideal lags for each of the variables. In both the cases, minimum Akaike Information Criterion is considered. The results are mentioned in table 3.

 

Table 3: Result of VAR Lag Order Selection Criteria

 

India                               

 

Pakistan

 

Variable

Lag

AIC

Lag

AIC

LNGDP

1

-5.60

1

-5.78

LNME

1

-2.88

1

-3.49

LNEDU

2

-3.02

2

-1.92

LNHEA

1

-3.66

1

-2.50

AIC: Akaike information criterion

 

After identifying the ideal lag length, Bounds test has been conducted to see whether there exists any cointegration between the variables. The result of Bound’s test is reported in the table 4 and table 5.

 

Table 4: Bounds Test Result India

Estimated Models

GDP=f(ME, Edu, Hea)

ME=f(GDP, Edu,Hea)

Optimal lag

(1,1,2,1)

(1,1,2,1)

Wald Test statistic

1.36

4.76**

Sig Level

Lower Bounds I (0)

Upper Bounds I (1)

 

1%

4.29

5.61

 

5%

3.23

4.35

 

10%

2.72

3.77

 

 

Table 5: Bounds Test Result Pakistan

Estimated Models

GDP=f(ME, Edu,Hea)

ME=f(GDP, Edu,Hea)

Optimal lag

(1,1,1,1)

(2,2,1,2)

Wald Test statistic

4.71**

3.18

Sig Level

Lower Bounds I(0)

Upper Bounds I(1)

 

1%

4.29

5.61

 

5%

3.23

4.35

 

10%

2.72

3.77

 

 

The results for India shows that calculated F statistic is i.e ME= f (GDP, Edu, Hea) =4.76 is greater than Pesaran’s upper critical bound at 5% level of significance. This indicates that there is one cointegrating vector that confirms the existence of long run relationship between military expenditure, GDP, education expenditure and health expenditure in the context of India. Where as in case of Pakistan, the F statistics i.e GDP=f (ME, Edu, Hea)=4.71 is more than upper bound value at 5% level of significance.  This implies that there is one cointegrating vector which consist the existence of long run relationship between the variables in Pakistan.

 

Further, the results clearly show that in India causality runs from GDP to military spending where as in Pakistan the causality runs from military spending to GDP in the long run.  

 

This helps us to find out the impact of all the variables on military expenditure. Results are reported below in table 6 and table 7.

 

Table 6: Long run results India

Dependent Variable: LnME

Variable

Coefficient

t-statistic

Prob

 

Constant

  6.44

1.79

0.12

 

lnGDPt

-0.14

-2.46

0.05**

 

lnEdut

 0.07

0.26

0.79

 

lnHeat

-0.12

0.80

0.81

 

Diagnostic Tests Statistic Prob

R2                  0.74

LM test         2.720.                    17

White            0.15                   0.99

ME- Military Expenditure

               

Table 7: Long run results Pakistan

Dependent Variable: LnGDP

Variable

Coefficient

t-statistic

Prob

Constant

-16.02

-4.09

0.01

lnMEt

-0.07

-2.41

0.05**

lnEdut

-0.06

-0.42

0.66

lnHeat

2.81

1.17

0.22

Diagnostic Tests Statistic                        Prob

R2                                    

0.93

 

 

Serial

1.40 

0.32

 

White

3.07

0.08

 

ME- Military Expenditure

 

It’s evident from the table 6 that in India military expenditure is negatively related with economic growth where as education expenditure and health expenditure is independent of military expenditure as they are statistically insignificant.  This result implies that in India, 1% increase in GDP will bring down the growth of military expenditure as share of GDP by 14%.

 

Table 7 shows very interesting result. In Pakistan GDP is negatively related to military spending which implies that 1% increase in military expenditure will bring down the GDP by 7%. This also implies that in Pakistan military expenditure is purely non developmental expenditure and therefore does not contribute towards GDP.

 

To examine the short run impact of independent variables Error Correction Model (ECM) model is used. The results are reported below in table 8 and table 9.

 

Table 8: Short run Results India

Dependent Variable: LnME

Variable

Coefficient

Std.Error

T Statistic

Constant

0.04

0.06

0.50

D(lnGDP)

-0.86

0.98

-0.88

D(lnEdu)

0.98

0.34

0.98

D(lnHea)

-0.40

0.45

-0.90

ECT(-1)

-0.26

0.27

-2.67**

R2

0.48

 

 

F Statistic

4.08

 

 

DW statistic

1.86

 

 

Diagnostic Checks

 

 

 

B-G LM test

3.11

 

 

White Test

0.45

 

 

CUSUM Test

Stable**

 

 

** denotes significant at 5%

 

Table 9: Short run Results Pakistan

Dependent Variable: LnGDP

Variable

Coefficient

Std.Error

T Statistic

Constant

0.006

0.007

  0.46

D(lnME)

 0.21

0.08

  2.44**

D(lnEdu)

0.03

0.03

  0.97

D(lnHea)

-0.08

0.04

-0.41

ECT (-1)

-0.01

0.03

-2.46**

R2

0.73

 

 

F Statistic

6.62

 

 

DW statistic

1.90

 

 

Diagnostic Checks

 

 

 

B-G LM test

2.97

0.10

 

White Test

0.42

0.82

 

CUSUM Test

Stable**

 

 

** denotes significant at 5%

 

In India there is no evidence of short run relationship between the variables.  ECT (-1) indicates the speed of adjustment and also further validates long run relationship between the variables. In case of India, our empirical results show that coefficient of ECT(-1) is -0.26 and significant at 1% level of significance which implies that 26% of disequilibrium from the current years shock seems to be corrected in the next year.

 

Similarly, in case of Pakistan, the ECT (-1) is -0.01 and significant at 5% level of significance. This implies that only 1% of disequilibrium will be corrected in the next year.  The results also show that in the short run there exist positive relationship between GDP and military expenditure. This implies that 1% increase in military spending increases the GDP by 21%.

 

The stability of all these models is tested by applying Cumulative Sum (CUSUM) test. The CUSUM test result reveals that both the models are stable and reliable. For both long run and short run models, diagnostic tests indicate that there is no evidence of autocorrelation and heteroskedasticity.

 

CONCLUSION:

The main objective of this paper is to analyze the relationship between military spending and other variables like GDP, education expenditure and health expenditure. There exists long run relationship between the variables. This result is consistent with some recent empirical studies like Hassan Raju, Ahmed (2019), Tiwari, Shahbaz (2011), Tiwari, Tiwari (2009).  The result reveals that in India the there exist negative relationship between military expenditure and GDP which implies that as GDP increases the military expenditure decreases in the long run. This also means that in India military spending is not dependent on economic factors rather it may be affected by political situation inside and outside the country. However, in this, result is in contrary with the above mentioned studies. These studies claimed that there exists positive relationship between GDP and military spending.  Education and health expenditure are insignificant which also means that these two are independent of military expenditure.  Results also show that there is no short run relationship between the variables. One of the possible explanations for the contradictory results might be due to different time periods considered by these different studies and also the different variables considered for the analysis.

 

Interestingly, in case of Pakistan there exists long run relationship between the variables. This result is consistent with the results of Anwar, Rafiq, Joiya (2012). The results reveal that there exists negative relationship between GDP and military spending which implies that as military expenditure increases GDP decreases. This also means that the military expenditure is absorbing resources which could otherwise used for developmental sectors like education, health etc. This result supports the view of Ali and Ather (2014). This means that the government should direct the resources towards more productive civilian investments.

 

REFERENCE:

1.      Ali, Ather (2014), Impact of Defense Expenditure on Economic Growth: Time Series Evidence from Pakistan, Global Journal of Management and Business Research: B Economics and Commerce, Vol 14(9).

2.      Azmair, Hussain, Abbassi and Gohar (2018), The Impact of Military Expenditures on Economic Growth of Pakistan, Applied Economics and Finance, Vol 5(2).

3.      Hasan Raju, Ahmed (2019), Effect of Military Expenditure on Economic Growth: Evidences from India, Pakistan and China using Cointegration and Causality Analysis, Asian Journal of German and European Studies, 4(3), 2019.

4.      Hirnissa, Habibullah, Baharom (2009), The Relationship between Defense, Education and Health Expenditures in Selected Asian Countries, International Journal of Economics and Finance, Vol 1(2).

5.      Hou, Chen (2013), Military Expenditure and Economic Growth In Developing Countries: Evidence From System GMM Estimates, Defense and Peace Economics, Vol 24(3).

6.      Khalid, Razaq (2015), The Impact of Military Spending on Economic Growth: Evidence from the US Economy, Research Journal of Finance and Accounting, Vol 6(7).

7.      Khilji, Mahmood (1997), Military Expenditures and Economic Growth in Pakistan, The Pakistan Development Review, Vol 36(4).

8.      Obreja, Laura (2010), The Impact of Defense Expenditure on Economic Growth, Journal for Economic Forecasting, Vol 1(4), 148-167.

9.      Qureshi, Khan (2017), Revisiting the Relationship between Military Expenditure and Economic Growth in Pakistan, Global Social Sciences Review, Vol 2 (2).

10.   Taspinar, Sadeghieh (2015), Military Expenditure and Economic Growth: The Case of Turkey, Procedia Economics and Finance, Vol 25.

11.   Tiwari, Aviral & Tiwari, A. (2010). Defense Expenditure and Economic Growth: Evidence from India. 5. 10.17863/CAM.1363.

12.   Yildirim, Sezgin, Ocal (2005), Military Expenditure and Economic Growth in Middle Eastern Countries: A Dynamic Panel Data Analysis, Defense and Peace Economics, Vol 16(4).

 

 

 

Received on 19.02.2021         Modified on 16.07.2021

Accepted on 10.01.2022      ©AandV Publications All right reserved

Res.  J. Humanities and Social Sciences. 2022;13(1):12-16.

DOI: 10.52711/2321-5828.2022.00002