Available online at www.sciencedirect.

com

Procedia - Social and Behavioral Sciences 65 (2012) 717 – 722

International Congress on Interdisciplinary Business and Social Science 2012

(ICIBSoS 2012)

Malaysian outward FDI and economic growth

Jen-Eem Chena,*, Shaliza Azreen Mohd Zulkiflib
a
Universiti Teknologi MARA, 40450 Shah Alam, Selangor, Malaysia
b
Universiti Teknologi MARA, 40450 Shah Alam, Selangor, Malaysia

Abstract

The aim of this paper is to investigate the association between outward FDI and economic growth for Malaysia over
the period 1980-2010 within a multivariate framework. The model specification employs general production function
where the main independent variable is outward FDI, and the controlled variables are labour and domestic
investment. The results obtained from vector error-correction model (VECM) reveal that there exists a positive long-
run relationship between outward FDI and growth as well as long-run bi-directional causation between them. In the
short-run, we do not find Granger-causality between outward FDI and growth.

© 2012
2012The Authors.by
Published Published
ElsevierbyLtd.
Elsevier Ltd. and/or peer-review under responsibility of JIBES University,
Selection
Selection and peer-review under responsibility of JIBES University, Jakarta
Jakarta

Keywords: outward FDI; growth; Granger-causality

1. Introduction

Malaysia has experienced several phases in its economy where economic activity at the beginning was
primarily based on the agricultural sector. In fact, foreign investment in Malaysia has started even during
the colonial period as Malaysia has always been a natural-resource enriched economy. After gaining its
independence during the 1950s, Malaysia is now able to become a host country for foreign investment
due to its good infrastructure, educated workforce, supportive government policies, and political stability.
The inflow of foreign direct investment to Malaysia greatly complements domestic investment thus

*
Corresponding author. Tel.: +60- 4-988-2863; fax: +60-4-988-2526
E-mail address: jechen@perlis.uitm.edu.my

1877-0428 © 2012 The Authors. Published by Elsevier Ltd.

Selection and peer-review under responsibility of JIBES University, Jakarta
doi:10.1016/j.sbspro.2012.11.189
718 Jen-Eem Chen and Shaliza Azreen Mohd Zulkifli / Procedia - Social and Behavioral Sciences 65 (2012) 717 – 722

allowing the country to experience economic growth and high employment rates. Despite of Malaysia is
a recipient of FDI, it also involved in venturing investment abroad since 1980s starting with a small
amount of $201 million. From 1982 until 1992, Malaysia s outward FDI (OFDI) indicated an unstable
volatility with the lowest amount of investment recorded at $115 million in year 1992. A significant
increase could be observed in 2002 where the investment climbed up to $1,905 million, although there
was a slight drop in 2003. Nevertheless, the bullish trend emerged again from year 2004 at $2,061 million
until 2008 at its peak $14,059 million (UNCTAD database). The trend of outward FDI in Malaysia was
highlighted when the outflows exceeded inflows in 2008 and 2009, indicating the economy is presently at
stage three of investment development path (Goh & Wong, 2011). According to investment development
path (IDP) paradigm that proposed by Dunning (1981 & 1986), domestic firms in stage three of IDP had
gained ownership advantages and expanded their operations abroad that elevate the country to a higher
level of economic development.
In recent years, the foreign direct investment (FDI) inflows to Malaysia show a large decreasing trend.
One of the major causes behind the declining FDI is due to the competition from emerging economies
such as China and India. With ever declining domestic markets and lower profit margins, many
Malaysian companies have resorted to venture abroad to seek newer markets. While the inflow of FDI is
growing at slower pace, the outward FDI has become an alternative to cater the slow global economic
growth. Given the existing literature on the empirical knowledge about the causality patterns between
outward FDI from Malaysia and its economic growth is limited, thus this prompts us to i) examine the
long-run effects of Malaysian outward FDI on growth, ii) determine the causal relationship between
outward FDI and domestic output that covers the annual data from 1980 to 2010 in multivariate model.

2. Brief literature review

The effects of outward FDI on domestic output have become the central point of debate in developed
countries. Outward FDI could bring positive or negative effect to domestic income depending on the role
of outward FDI in domestic market. In the event that outward FDI is a substitute to domestic production,
when multinational firms relocate their production facilities abroad, it reduces domestic output,
employment, and economic growth as well (Stevens & Lipsey, 1992). On the other hand, if outward FDI
is a complementary to domestic investment, it stimulates local production (Desai et al., 2005). This can
be seen that multinational firms exploring the new markets, importing intermediate inputs from affiliates
abroad and producing final goods in foreign affiliates at lower costs, and accessing foreign technology.
With the combination of foreign and local productions, it lowers the cost of production and increases the
competitiveness of multinational firms. As a result, it would benefit the overall domestic market with the
associated spillover effects to local firms (Herzer, 2010). Nonetheless, the impact of outward FDI on the
development of a country depends on the absorption ability level of the home economy and the rate of
technological gap between MNCs and non-multinational domestic firms (Denzer, 2011).
The empirical studies in identifying the relationship between inward FDI and economic growth have
been studied extensively. In contrast, little attempts have been conducted in examining the relationship
between outward FDI and home country output at aggregate level. The work of Herzer (2008) found that
outward FDI has positive long-run effects on domestic output in 14 industrialized countries over the
period 1971 to 2005 using panel analysis. The results also pointed out that the long-run causality is bi-
directional between outward FDI and domestic output. Later, Herzer (2010) examined the impact of
outward FDI on economic growth by using two regressions, i.e., cross-country and time series. The
cross-country regression was based on 50 countries and the period covered from 1980 to 2000. At the
same time, outward FDI of U.S. was selected as the ground for time series analysis with data spanned
from 1980 to 2004. Both regressions shared the same findings where outward FDI is positively
associated with growth. The Granger-causality test in time series estimator exhibited a bi-directional
causality between outward FDI and growth.
 Jen-Eem Chen and Shaliza Azreen Mohd Zulkifli / Procedia - Social and Behavioral Sciences 65 (2012) 717 – 722 719

Lee (2010) provided empirical study of Japan which is one of the industrialized economies using the
data set from year 1977 to 2006 in bivariate and multivariate models. The results from bivariate model
explained that GDP per capita Granger-cause outward FDI in short-run and bi-directional causality
appears in the long-run. Turning to multivariate model, the results revealed that positive causality
running from outward FDI to GDP per capita in the long-run. Similarly, the empirical studies on
Malaysian outward FDI and growth was explored by Wong (2010). In the study of Wong (2010), he
employed data of Q1 1999 until Q4 2008. The findings from bivariate model suggested a uni-directional
causality where growth runs to outward FDI. In other words, outward FDI does not Granger-cause
growth but vice versa.
Based on the existing literatures, most of the findings suggested that outward FDI and growth is
positively related in advanced countries. Given the study focuses on developing countries is not much
analysed, thus this paper attempts to add some empirical evidences from developing countries in the
existing literatures.

3. Model specification and methodology

The long-run model specification employed in this study is expressed as:

GDPt = + 1OFDIt + 2DIt + 3 LBt+ t (1)

Based on equation (1), 1 and 2 and 3 are the parameters to be estimated while t is error term. All
variables are measured in real term of natural logarithm. The data is ranged from year 1980 to 2010
which was collected from UNCTAD, World Development Indicators (WDI), International Financial
Statistics (IFS) as well as various issues of Malaysian Economic Report.
Economic growth is measured by GDP. OFDI is outward FDI flows and the expected coefficient sign
can be positive or negative. The two controlled variables are DI and LB. DI represents domestic
investment which is measured by the ratio of gross fixed capital formation to GDP. LB denotes labour
that measured by employment. Domestic investment and labour are expected to be positively related with
growth.
Prior to estimating equation (1), we test the stationarity for each of the variable to avoid spurious
regression so that produces a meaningful estimation. We adopt augmented Dickey-Fuller (ADF) unit root
test (Dicky and Fuller, 1979) and Philip-Perron (PP) unit root test (Philips and Perron, 1988) to determine
the order of integration of each of the variables. When all of the variables are integrated of same order
indicating that variables are independent from unit root and allow us to proceed with cointegration test
based on Johansen-Juselius (1990) procedure. The purpose of conducting cointegration test is to examine
the existence of long-run equilibrium relationship between variables in the model. In addition, we further
with Granger-causality test to incorporate short-run dynamic causal interaction among variables in the
equation within vector error-correction model (VECM) framework.

4. Empirical results

4.1. Unit root tests

The results of augmented Dickey-Fuller (ADF) and Philips-Perron (PP) unit root tests are presented in
Table 1. The variables are tested at level and first difference with the inclusion of intercept and trend.
The ADF and PP tests show that all variables are non-stationary at level, but are stationary after first
differencing. Thus, we conclude that all variables are integrated of order one, I(1), and we proceed to
Johansen s cointegaration test.
720 Jen-Eem Chen and Shaliza Azreen Mohd Zulkifli / Procedia - Social and Behavioral Sciences 65 (2012) 717 – 722

Table 1. Unit root tests

Variables Level First Difference
ADF PP ADF PP
GDP -1.134 -1.351 -4.360*** -4.370***
OFDI -3.094 -3.087 -7.255*** -7.255***
DI -1.638 -1.638 -3.745** -3.580**
LB -0.328 -0.328 -4.245** -4.259**
*** and ** denote 1% and 5% significance level, respectively. The lag length selection in ADF test is
based on Schwarz Info Criterion (SIC) while PP test is based on Newey-West Bandwidth.

4.2. Cointegration test

We employ Johansen-Juselius cointegration test to examine should the long run equilibrium exists in
equation (1). Correlogram test was conducted on the residuals generated from equation (1) to identify the
order of lag length. The results of Ljung-Box Q test indicate that lag order 1 is sufficient to render the
error term serially uncorrelated as all residuals cannot reject the null hypothesis of no autocorrelation at
lag 1.

Table 2. Johansen-Juselius cointegration test

Null Hypotheses Test Statistics Critical Values (5%)
Trace Max-Eigen Trace Max-Eigen
( trace) ( max)
None 64.077** 33.900** 47.856 27.584
At most 1 30.176** 15.423 29.797 21.132
At most 2 14.753 12.274 15.494 14.264
At most 3 2.478 2.478 3.841 3.841
** denotes rejection of null hypothess at 5%.
Based on Table 2, trace statistic suggests that there are two cointegrating vectors among the variables.
On the other hand, max-eigen test indicates one cointegrating vector between growth and its
determinants. Given that trace and max-eigen tests provide different cointegrating vector, we assume that
there exists a single cointegrating vector which is based on production function theory where GDP is
normalized variable.

4.3. Long-run estimates

The results of long-run coefficient which is estimated in the VECM framework with lag order one are
presented as follow:

GDPt-1 = 9.745 + 0.011 OFDIt-1** + 0.252 DIt-1*** + 1.869 LBt-1***

(2.419) (16.198) (88.964)

The figure in parenthesis is t-statistic for each variable. **denotes 5% significance level and *** is 1%
significance level. The results demonstrate that all variables are statistically significant at different levels.
We note a positive association between GDP and outward FDI. The result is in line with the empirical
works of Herzer (2008) and Lee (2010) that they found outward FDI gave positive effects to GDP in the
long-run. This postulates that outward FDI is a complementary to domestic production thus stimulates
 Jen-Eem Chen and Shaliza Azreen Mohd Zulkifli / Procedia - Social and Behavioral Sciences 65 (2012) 717 – 722 721

economic growth. Besides that, we observe that domestic investment and labour are also key drivers of
growth.

4.4. Granger-causality

We integrated short-run dynamic interaction among the variables and examined the dynamic
adjustment back to the long-run. Each variable is treated as endogenous within the VECM framework.
The results of Granger-causality test are illustrated in Table 3.

Table 3. Granger-causality test

Independent Variables
2
-statistics of lagged first differenced term ECTt-1
Dependent coefficient
variable [p-value] (t-ratio)
GDP OFDI DI LB
GDP - 0.002 0.811 0.125 -0.678*
[0.964] [0.368] [0.724]
(-1.878)
OFDI 0.914 - 2.449 1.446
-14.244**
[0.339] [0.117] [0.229]
(-2.148)
DI 0.212 0.119 - 0.341 -0.151
[0.645] [0.730] [0.559](-0.129)
LB 0.398 1.758 0.623 - 0.163
[0.527] [0.185] [0.429] (1.670)
Number in squared brackets denotes p-value while the number in parenthesis denotes t-statistics.
The asterisk ** and * denote statistical significant at the 5% and 10% level, correspondingly.

In Table 3, when GDP serves as dependent variable the error correction term (ECTt-1 ) is significant at
10%. This implies that GDP tends to converge to its long-run equilibrium path in response to the changes
in the regressors. The significance of ECTt-1 in GDP equation confirms the existence of long-run
equilibrium between GDP and its determinants which are outward FDI, domestic investment, and
employment. Thus, this suggests that outward FDI, domestic investment, and employment Granger-cause
GDP in the long-run. The error correction term in OFDI equation is significant at 5%. Hence, there
exists a long-run causality from GDP, domestic investment, and employment to outward FDI. Thus, we
can conclude that there is a bi-directional causality between GDP and outward FDI in long-run. The DI
and LB equations are statistically insignificant.
Turning to the short-run dynamics causality, the results show that each variable is independent and
does not have a significant impact on each other in the short-run. Therefore, it implies that the growth of
outward FDI does not affect the GDP growth in the short-run and vice versa.

5. Conclusion

This study discussed the association between outward FDI and economic growth of Malaysia for year
1980 to 2010. The findings reveal that outward FDI is positively and significantly affect GDP in the long-
run which support the existing literatures. In addition, the direction of causality between growth and
outward FDI is positive bi-directional in the long-term. Therefore, we can conclude that outward FDI
benefits the Malaysian economy as a whole by boosting the GDP which in turn will lead to a further
increase in outward FDI in the form of knowledge and technology transfers between MNCs and local
firms. This process can also create higher numbers of skilled workers in Malaysia.
722 Jen-Eem Chen and Shaliza Azreen Mohd Zulkifli / Procedia - Social and Behavioral Sciences 65 (2012) 717 – 722

Acknowledgement

We would like to thank Ministry of Higher Education Malaysia for granting research fund.

References

Dickey, D. A. and Fuller, W. A. (1979). Distributions of the estimators for autoregressive time series
with a unit root. Journal of American Association, 74, 427-431.

Denzer, A. (2011). The effects of outward FDI on economic growth a theoretical and empirical
analysis (Doctoral dissertation). Eberhard Karls University Tuebingen, Germany.

Desai, M. A., Foley, F. & Hines Jr., J. R. (2005). Foreign direct investment and domestic capital stock.
American Economic Review Paper and Proceedings, 95, 33-38.

Dunning, J. H. (1981). Explaining the international position of countries towards a dynamic or

developmental approach. Weltwirtschaftliches Archiv, 117, 30 64.

Dunning, J. H. (1986). The investment development cycle revisited. Weltwirtschafliches Archiv, 122,
667-677.

Goh, S. K. & Wong, K. N. (2011). Malaysia s outward FDI: the effects of market size and government
policy. Journal of Policy Modelling, 33, 497-510.

Herzer, D. (2008). The long run relationship between outward FDI and domestic output: evidence from
panel data, Economic Letters, 100 (1), 146-149.

Herzer, D. (2010). Outward FDI and economic growth. Journal of Economic Studies, 37(5), 476-494.

Johansen, S. and Juselius, K. (1990). Maximum likelihood estimation and inference and cointegration
with applications to the demand for money. Oxford Bulletin of Economics and Statistics, 52, 169-210.

Lee, C. G. (2010). Outward foreign direct investment and economic growth: evidence from Japan.
Global Economic Review, 39 (3), 317-326.

Philips, P. C. B. and P. Perron (1988). Testing for a unit root in time series regressions, Biometrika, 75,
335-346.

Steven, G. V. G. and Lipsey, R. E. (1992). Interactions between domestic and foreign investment.
Journal of Money and Finance, 11, 40-62.

UNCTAD. Database: FDI outflows by Region and Economy, 1970-2010. New York and Geneva:
United Nations.

Wong, K. N. (2010). Outward FDI and home country economic growth: a Malaysian case. Discussion
paper 56/10 Department of Economics, Monash University Malaysia.