Sector level analysis of FDI-growth Nexus: a case study of Pakistan.
Iram, Somia ; Nishat, Muhammad
The main objective of the study is to empirically investigate the
differential impact of services and manufacturing Foreign Direct
Investment (FDI) on economic growth over the period of 1972 to 2008. The
study further examines the role of FDI in presence of macroeconomic
instability and privatisation. For the investigation of long run,
Autoregressive distributed lag model (ARDL) has been used. For short run
results, we used Error correction method (ECM). Our empirical results
show that FDI inflow in both, service sector as well as manufacturing is
contributing to economic growth positively. But it is apparent from the
results that contribution of services FDI to growth is greater than that
of manufacturing FDI to growth. Furthermore, the results provide
coherent and sound policy recommendations for further policy adaptation
regarding sectors.
JEL classification: F23, F36, F43, C32
Keywords: Foreign Direct Investment, Economic Growth, Manufacturing
Sector, Service Sector, Co-integration
1. INTRODUCTION
The most stable nature of Foreign Direct Investment (FDI) among the
all capital inflows has provoked its importance especially in case of
capital scarce developing economies. Economic growth is the indicator of
the health of economy and capital is one of the prerequisites to
maintain and enhance the momentum of growth. In the current scenario,
growth of Pakistan economy has gone considerable changes and regarding
these changes, Pakistan has adopted different policies concerning
different sectors of economy.
Currently, the services sector share to GDP has improved
considerably against the dismal performance of manufacturing sector over
the last ten years. In order to enhance the economic growth effectively
and efficiently, government has taken several steps to attract foreign
capital. One of these measures is the adaptation of highly liberalised
policies to attract most needed financial capital along with its
spillovers. Even after suffering from obnoxious economic ailments,
Pakistan in current decade is able to attract massive capital inflow
perhaps highest in the history. However, the massive inflow of FDI is
directed towards services sector.
Even after conducting a handful empirical research, the impact of
FDI on growth is countervailing. Some early studies [Singer (1950);
Griffin (1970)] recognised the negative impact of FDI on economic growth
in developing countries. Aitkin and Harrison's (1999) in case of
Venezuela, Jhon and Athanasios (2004) in case of US and Western European
countries, and Katerina, et al. (2004) in case of transition countries
found that FDI do not significantly affect economic growth. However,
Blomstrom, et al. (1992), Caves (1974) and Kokko (1994) showed a
positive effect of FDI inflows on economic growth. Findlay (1978)
highlighted the positive effect through technology spillovers, which has
the strongest potential to enhance economic growth in the host country.
Borensztein, et al. (1998), Xu (2000) and Alfaro, et al. (2003)
suggested the positive impact of FDI in presence of the sound
educational level, development of local financial markets, and other
necessary conditions to absorb spillovers. Blomstrom and Kokko (2003)
explained that positive effects of FDI are not automatic but the local
conditions influence firms' adoption of foreign technologies and
skills. Borensztein (1995, 1998) explained the growth enhancing effect
of FDI through the channel of technology. Borensztein (1995, 1998)
emphasised the presence of at least threshold level of initial human
capital for the diffusion of technology spillovers. Romar (1986, 1990),
Helpman and Grossman (1990) emphasised the importance of knowledge
capital, coming through research and development in the long run
economic growth.
According to Chudnovsky and Lopez (1998), FDI may boost economic
growth through the improvement of manufacturing export and improved
balance of payment. However, in the long run, due to the control of
foreigners over the local production resources, profit outflow
deteriorate the balance of payment condition. In case of developing
countries, FDI mostly work through the channel of externalities.
However, there is no definite conclusion related to spillovers of FDI.
Benefits and cost associated with FDI is not disseminated homogenously
across all countries and even across all sectors. Therefore, different
countries, regions and even sectors react differently to same FDI
inflow.
Alfardo (2003) examined the effect of FDI on growth in the primary,
manufacturing and services sectors. The author suggested that total FDI
exerts an ambiguous effect on economic growth. FDI in the primary sector
tends to have a negative effect on growth. However, investment in
manufacturing tends to have a positive effect on growth. Moreover,
evidence from the service sector is ambiguous. Zaman, et al. (2008)
investigated the factors effecting FDI in case of Pakistan using data
over the period of 1971-2003, and found that variables used for market
size and trade balance are significant, whereas, variable used for
service sector has negative effect on the growth of economy. Importance
of FDI can not be denied, especially in case of developing economies.
FDI not only allows overcoming the financing and liquidity constraints,
but also provide new capital, allowing additional investment in both
human and physical capital, which can be very beneficial for developing
countries.
To the best of my knowledge, in case of Pakistan only few studies
have investigated FDI-growth nexus. FDI-growth nexus is not being
investigated with respect to services sector and manufacturing sector.
Overall, the impact of FDI on growth can be misleading. This study is
first attempt to investigate the impact of manufacturing sector and
services sector FDI on growth in the presence of macroeconomic
instability and privatisation regime in case of Pakistan.
The main objective of the study is to investigate the impact of
services and manufacturing Foreign Direct Investment (FDI) on economic
growth over period of 1972 to 2008. We decompose the FDI into services
sector and manufacturing sector and examine their impact on economic
growth. To further evaluate the role of FDI on growth in presence of
privatisation policy, interaction terms of services as well as
manufacturing FDI with privatisation dummy is introduced.
This paper is organised in to following sections: Section II
regards the data, model and methodology; Section III contains results
and conclusion, whereas, last section includes conclusion and policy
recommendation.
II. MODEL SPECIFICATION AND DATA SOURCES
To estimate the relationship between FDI and economic growth at
sector level, we estimate the following model:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (1)
Where Y represents the real GDP per capita, MFDI and SPFDI
represent FDI in manufacturing and service sectors respectively, INV
represents public sector investment and INF represents Inflation rate
proxied for macroeconomic instability and t represents time period
(1972-2008). The Dummy of privatisation is also included in this model
due to its significance
Data used in this paper is obtained from the electronic database of
International Financial Statistics (IFS), Annual report state bank of
Pakistan. The FDI of selective sectors is chosen for the sample because
of the unavailability of data in case of Pakistan. Data on services and
manufacturing FDI is obtained from foreign liabilities and assets and
investment in Pakistan (Various Issues).
Before estimating the long run and short run results, ADF and PP
unit root tests in order to check the unit root of all variables. We use
the robust technique Autoregressive Distributed Lag model (ARDL)
introduced by Pesaran, Pesaran, and Smith (1998), Pesaran and Shin
(1999) and Pesaran, et al. (2001).
The error correction version of ARDL model is given below:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (2)
Where [[beta].sub.0] component and [mu] trend is the error term.
The term with summation sign represent the short run dynamics. While,
the second part of the equation represent the long run dynamics.
In order to estimate the long run coefficients, the following
long-run model is estimated:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (3)
After finding the long run relation we use the following equation
to estimate the short run coefficients:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (4)
[eta] is the error correction term in the model indicates the pace
of adjustment reverse to long.
III. ESTIMATION RESULTS
III.1. Testing of the Unit Root Hypothesis
We applied ADF test and PP unit root test in order to check the
unit root of all variables. The results suggest that all variables are
having mix order of integration. A summary of the results of ADF and PP
unit root tests is given in Table 1.
From the results of unit root tests, it is apparent that the
variables have different order of integration. In the next step, we
proceed to apply the ARDL approach.
III.2. Autoregressive Distributed Lag Model (ARDL) Lag Selection
The first step of bound testing approach is to select the order of
the lag length. On the basis of lag length, we found the F-statistics
value. The estimated F-value selected on the basis of lag length is
given below:
We found that the optimum lag length is two, which is selected by
using Akaiake Information Criterion (AIC) as shown in Table 2. At lag
length two, the F-statistics is significant at 5 percent level. The
significant F-value indicates that there is co integration among the
variables. After finding a long run relationship we estimated the long
run and short run parameters. The results of long run coefficients are
given in Table 3.
The results of the Table 3 show that all variables are significant
and have expected signs. The positive coefficient of FDI shows that real
GDP per capita is largely depends on inward flow of FDI. However, it is
apparent from the results that FDI inflow contributes to economic growth
mainly when the economy practices the privatisation. The results show
that FDI inflow in the service sector accelerates economic growth by a
high speed. FDI inflow in the manufacturing sector significantly affects
economic growth. However, the magnitude of SFDI is much greater than
MFDI. Moreover, it is apparent from the results that FDI inflow
contributes to economic growth mainly when the economy practices the
privatisation. The reason behind this result is that privatisation
policy in most countries has proved to be investor's friendly.
Privatisation reduces the management bottlenecks. It enhances the
efficiency by introducing new and advance management practices. The
coefficient of private investment (excluding foreign investment) is
positive and significant at 5 percent level of significance, which is
depicting the positive effect of private sector participation.
In case of Pakistan both services as well as manufacturing sectors
are contributing but services sector is contributing much more than
manufacturing sector. In the current decade major surge of FDI was
towards the telecommunication sector that has strengthened the
infrastructure as well as increased job potential and in return caused
increased contribution to economic growth.
The error correction version of ARDL is presented in Table 4.
The sign of estimated lagged error correction term ECt-1 is
negative and significant at 9 percent level of significance. There is 42
percent speed of adjustment towards long run equilibrium. In the short
run, MFDI and SFDI do not significantly affect economic growth. In the
case of developing countries FDI is important because of its spillover
affects that are not instantaneous rather time consuming that's why
FDI do not contributes the growth in the short run but it takes time to
influence the growth patterns of economy. The short run results show
that inflation and investment significantly affect economic growth. The
sign of inflation is positive in the short run. This is due to the fact
that an increase in the prices increases the profit margin of the
producers in the short run. However, in the long run due to decreased
real income of the general masses, economic growth hampers.
IV. CONCLUSION AND POLICY RECOMMENDATIONS
This study empirically investigated the impact of services and
manufacturing Foreign Direct Investment (FDI) on economic growth in the
presence of macroeconomic instability and privatisation over period of
1972 to 2008. In order to find out the order of integration, we used ADF
and PP unit root tests. Autoregressive distributed lag model (ARDL) is
used for the robustness of long-run relationship between the variables.
We found that in the long run MFDI and SFDI significantly affect
economic growth. However, in the short run, both MFDI and SFDI do not
significantly affect economic growth. Moreover, the magnitude of
services sector FDI is greater than manufacturing sector FDI. Whereas,
the variable of macroeconomic instability significantly affect economic
growth both in long run as well as in short run.
The coefficient of inflation is positive in the short run. This is
due to the fact that an increase in the prices increases the profit
margin of the producers in the short run. However, in the long run due
to decreased real income of the general masses, it hampers economic
growth. Private investment is also helping to boost the economic growth.
The results also show that in the presence of privatisation policy, FDI
contribute to economic growth. However, this contribution is more in
services sector as compared to manufacturing sector.
As regards the policy recommendation, proper attention should be
paid proper to strengthen manufacturing sector that is real sector of
economy. In case of services sector government should attract FDI toward
infrastructure base services sector so that it may help to contribute
the growth of manufacturing sector in the long run. FDI should be
encouraged to amplify economic growth, to amplify benefit of innovative
technology to curtail poverty and unemployment, to lift up living
standards but at the mean time proper attention should also be paid to
save sovereignty and profit outflow of the country. In order to enhance
growth, policies should device to attract export oriented FDI instead of
domestic demand oriented.
Appendix
Data Description and Sources
Expected
Sign in Data Sources
S. N. Variables Literature and Description
1. Market Growth (GDP) + GDP growth is used
as dependent
variable. Data is
obtained from SBP
annual report.
2. Investment (INV) +/- Investment is
proxied as GFCF.
Data is obtained
from hand book on
statistics on
Pakistan economy
(Various issues).
3. Macroeconomic - Macroeconomic
Instability (INF) instability is
proxied by inflation
variable. Data on
inflation is taken
from WDI (2008)
electronic database.
4. Manufacturing FDI + Manufacturing FDI is
(MFDI) calculated by adding
the FDI coming
towards all the
manufacturing units.
Source: Foreign
liabilities and
assets and
investment in
Pakistan (Various
Issues).
5. Services FDI (SFDI) - Services FDI is
including FDI in
infrastructure based
services sector.
Source: Foreign
liabilities and
assets and
investment in
Pakistan (Various
Issues).
6. Privatisation (Dummy) + Privatisation is
taken as dummy .1
for years
privatisation was
taken, 0 otherwise.
REFERENCES
Aitken, B. and A. E. Harrison (1999) Do Domestic Firms Benefit from
Direct Foreign Investment? Evidence form Venezuela. American Economic
Review 89:3, 605-618. Alfaro, L. (2003) Foreign Direct Investment and
Growth: Does the Sector Matter? Harvard Business School, Boston, MA.
(Mimeographed).
Atique, Z., M. H. Ahmad, and U. Azhar (2004) The Impact of FDI on
Economic Growth under Foreign Trade Regimes: A Case Study of Pakistan.
The Pakistan Development Review 43:4.
Bengoa, M. and B. Sanchez-Robles (2003) Foreign Direct Investment,
Economic Freedom and Growth: New Evidence from Latin America. European
Journal of Political Economy 19:3, 529-543.
Blomstrom, M. and A. Kokko (2003) Human Capital and Inward FDI. The
European Institute of Japanese Studies. (EIJS Working Paper Series. No.
167).
Blomstrom, M., R. E. Lipsey, and M. Zejan (1994) What Explains
Developing Country Growth? (NBER Working Paper No. 4132).
Borensztein, E., J. De Gregorio, and J. W. Lee (1998) How Does
Foreign Direct Investment Affect Economic Growth? Journal of
International Economics 1:45, 115-135.
Caves, R. (1974) Multinational Firms, Competition and Productivity
in Host- Country Markets. Economica 41:162, 176-93.
Enderwick, P. and M. Akoorie (1996) Fast Forward: New Zealand
Business in World Markets. Longman Paul, Auckland.
Findlay, R. (1978) Relative Backwardness, Direct Foreign Investment
and the Transfer of Technology: A Simple Dynamic Model. Quarterly
Journal of Economics 62:1, 1-16.
Griffin, K. B. (1970) Foreign Capital, Domestic Savings and
Development. Oxford Bulletin of Economics and Statistics 32, 99-112.
Grossman, G. and E. Helpman (1990) Innovation and Growth in the
Global Economy. Cambridge, MA.: MIT Press.
Hunya, G. (1998) Integration of CEEC Manufacturing into the
European Corporate Structures by Direct Investments. (MOSTMOCT, No. 1).
Hunya, G. (1998) Relationship between FDI, Privatisation and
Structural Change in CEECS. Paper prepared for the Conference on
Privatisation, Corporate Governance and the Emergence of Markets in
Central Eastern Europe, FIT, Berlin, May 22-23. (Mimeographed).
Katherina, L., P. Jhon, and V. Athanasios (2004) Foreign Direct
Investment and Economic Growth in Transition Economies. 1, 97-110.
Khan, A. H. (1997) FDI in Pakistan: Policies and Trends. The
Pakistan Development Review 36:4, 959-985.
Kokko, A. (1994) Technology, Market Characteristics and Spillovers.
Journal of Development Economics 43, 279-93.
Ozturk, I. and H. Kalyoncu (2007) Foreign Direct Investment and
Growth: An Emprical Investigation Based on Cross Country Comparison.
Econimia Internazionale 60:1.
Pesaran, H. M. (1997) The Role of Economic Theory in Modeling the
Long-run. Economic Journal 107, 178-191.
Pesaran, H. M. and Y. Shin (1995) Autoregressive Distributed Lag
Modeling Approach to Cointegration Analysis. Department of Economics,
University of Cambridge. (DAE Working Paper Series, No. 9514).
Pesaran, H. M. and Y. Shin (1998) An Autoregressive Distributed Lag
Modeling Approach to Cointegration Analysis. In S. Storm (ed.)
Econometrics and Economic Theory in the 20th Century: The Ragnar Frisch
Centennial Symposium. Cambridge University Press.
Pesaran, M. H., Y. Shin, and R. Smith (1996) Testing for the
Existence of a Long-run Relationship. Department of Applied Economics,
University of Cambridge. (DAE Working Papers 9622).
Pesaran, M. H., Y. Shin, and R. Smith (2001) Bound Testing
Approaches to the Analysis of Level Relationships. Journal of Applied
Econometrics 16, 289-326.
Romer, P. (1986) Increasing Returns and Long-run Growth. Journal of
Political Economy 94, 1002-1037.
Singer, H. W. (1950) The Distribution of Gains between Investing
and Borrowing Countries. American Economic Review 40, 473-485.
UNCTAD (Various Issues) World Investment Report. New York: UNCTAD.
Xu, B. (2000) Multinational Enterprises, Technology Diffusion, and
Host Country Productivity Growth. Journal of Development Economics 62,
477-493.
(1) Critical values are obtained from Pesaran, et al. (2001) and
Narayan (2005).
Somia Iram <somia_irum@yahoo.com> is MPhil Student at Applied
Economics Research Centre (AERC), University of Karachi, Karachi.
Muhammad Nishat <mnishat@iba.edu.pk> is Professor of Business and
Finance at Institute of Business Administration (IBA), Karachi.
Table 1
Unit Root Results
Variables ADF (Drift and Trend) P-P (Drift and Trend)
Level 1st diff Level 1st diff
Y -2.08 *** -3.45 ** -2.19 *** -3.08 **
MFDI -0.98 -2.92 * -1.15 -3.12 *
SFDI -1.08 -2.98 ** -1.44 -3.17 **
INV -2.89 ** -3.76 * -2.78 *** -3.94 **
INF -1.79 -4.58 * -2.01 -4.35 *
Notes: *(**) Shows significance at 1 percent (5 percent) level.
Table 2
Lag Length Selection and Bound Testing for Co-integration
Lags Order AIC HQ SBC F-test Statistics
1 12.27 11.45 12.54 2.34
2 12.04 * 11.21 12.37 * 4.95 **
Short-run Diagnostic Test-Statistics
Serial Correlation LM, F = 0.78 (0.19)
Hetroscdasicity Test F= 1.92 (0.21)
Ramsey RESET Test F= 0.65 (0.35)
Normality J-B Value = 25.03 (0.07)
* Significant at 5 percent level according to Pesaran, et al.
(2001) and Narayan (2005). (1)
Table 3
The Long run Results
Dependent
Variable Y ARDL(1, 1,2, 2.2)
Variables Coefficients
MFDI 0.28 (0.02)
SFDI 0.74 (0.03)
INF -0.45 (0.06)
INV 0.18 (0.04)
MFDI*DUM 0.28 (0.08)
SFDI*DUM 0.47 (0.04)
R2 = 0.97
Adjusted R2 =0.95
F-statistics = 35.04 (0.00)
Dh Stat = 1.97
The values to the parenthesis are the probability values.
Table 4
Error Correction Representation of ARDL Model
Dependent Variable [DELTA]Y ARDL (1. 1,2, 2, 2)
Variables Coefficients
[DELTA]MFDI 0.16 (0.17)
[DELTA]SFDI 0.36 (0.13)
[DELTA]INV 0.42 (0.07)
[DELTA]INF 0.24 (0.18)
EC(-1) -0.42 (0.09)
Adjusted R-= 0.89 F-statistics = 20.03 (0.01)
The values in the parenthesis are the probability values.
