Examining the Equilibrium Relationships between Foreign Direct Investment Inflows and Employment in Manufacturing and Services Sectors: Evidence from Malaysia

: The current paper examines the long-run and short-run equilibrium relationships between FDI inflows and employment in Malaysian manufacturing and services sectors using ARDL approach for the 1972-2011 period. It employs ADF and PP tests to detect the stationary levels of above variables. Also, it utilizes the bounds F-statistics test to identify the co-integration among variables. Results of ARDL approach indicate the presence of significant long-run and short-run equilibrium relationships between FDI inflows and employment in manufacturing and services sectors. The paper’s findings are of particular interest and importance to Malaysian policy makers towards increasing FDI inflows and employment in manufacturing and services sectors. enhanced economic growth and increased employment in economic sectors. This paper contributes to the existing literature by examining the long-run and short-run equilibrium relationships among FDI inflows, EM and ES in Malaysia using ARDL Approach. The results of ADF and PP tests suggest that all variables are stationary at I(1) and 1% significance level. The findings of the bounds F-statistics test reveal the existence of co-integration among variables. Besides, EM and ES are positively associated with FDI inflows in the long-run and short-run. In fact, future research could broaden this study by including more economic sectors such as agriculture, forestry and fishing sector; mining and quarrying sector; and construction sector. From statistical perspective, the main limitation of this study is the small sample size of 40 observations which has a limiting factor, since the number of lags that use, consumes the number of observations and leads to specification errors in the analysis (Gujarati and Porter, 2008).


Introduction
Foreign direct investment (FDI) is an important source of capital in host countries that complements domestic private investment, creates employment, enhances technology transfer and boosts overall economic growth (Chowdhury and Mavrotas, 2005). Also, FDI is one of the most major factors leading to the globalization of international economy (Kok and Ersoy, 2009). However, several studies established the relationships between FDI and its determinants (See, for example, Vita and Kyaw, 2007;Andraz and Rodrigues, 2008;Ang, 2008).The results showed that FDI enhanced economic growth and significantly influenced by various determinants. On the other hand, there are few studies conducted to investigate the influences of FDI inflows on employment in economic sectors (Mickiewicz et al., 2000;McDonald et el., 2002;Jenkins, 2006;Tang and Gyasi, 2012). The results revealed that FDI inflows increased economic growth and boosted employment in various economic sectors. The current paper extends the existing literature by analyzing the long-run and short-run equilibrium relationships between FDI inflows and employment in manufacturing and services sectors. However, the significant of this paper that there is no authoritative work analyzed the long-run and short-run equilibrium relationships between FDI inflows and employment in manufacturing and services sectors using autoregressive distributed lag (ARDL) approach and Malaysian annual time-series data for the 1972-2011 period. The paper plan is organized as follows: section 2 provides an overview of FDI and Malaysian economy. Section 3 presents literature review. Data and methodology specification are discussed in section 4. Section 5 reports results analyses. Policy implications are drawn in section 6. Finally, conclusions, future research and limitation are presented in section 7.   (Athukorala, 2010). The goal of NEP was to establish new public corporations to enhance manufacturing sectors and exports in Malaysia. On November 1980, the NEP entered a new phase and the Malaysian government announced to establish the heavy industries corporations of Malaysian (HICOM) projects in many areas. These areas are; petrochemicals, iron, steel, paper products, cement, machinery and equipment, general engineering, building materials, transport equipment and energy projects including petronas's production (Athukorala, 2010). However, FDI flows into Malaysia increased significantly in 1981 valued RM3.794 billion as a result of HICOM projects establishment (Athukorala, 2010). As a result of high levels of FDI inflows, Malaysia has been one of the best performers among largest South-East Asian countries. Figure 2 reveals that Malaysian economic growth rate, real gross domestic product (RGDP), achieved 5.9% for the 1980-2011 period.  However, due to HICOM projects establishment and the high levels of FDI flows into Malaysia, unemployment rates decreased dramatically and achieved an inverse growth rate of -3% for the 1985-2011 period (See, Figure 4).  1972 1975 1978 1981 1984 1987 1990 1993 1996 1999 2002 2005 (Montgomery et al., 2008).
Methodology Specification: The current paper attempts to examine the long-run and short-run equilibrium relationships among Malaysian LFDI inflows, LEM and LES by relaying on Equation (1).
Whereα0 denotes the intercept term; α1 and α2 represent the coefficients of explanatory variables; εt denotes the error term. Equation 1 can be augmented by introducing the error correction terms of ARDL Approach (See, Equations 2-4). (4) Where λij(i,j=1,….3) represents the long-run coefficients of the one lagged variables; Γij(i,j=1,….3) denotes the short-run coefficients of the variables differences; δi(i=1,…3) represents the coefficients of error correction terms; and (ECMt-1) 's indicate the error correction terms that are used to link the long-run with the short-run equilibrium; εit(i=1,…3) represents error terms. The present paper uses ARDL approach to examine the longrun and short-run equilibrium relationships among Malaysian LFDI inflows, LEM and LES. Specifically, it starts by testing the variables stationarity levels using augmented Dickey-Fuller (ADF) and Phillips and Perron (PP) tests. Then, the paper utilizes the bounds F-statistics to test the co-integrating relationships among variables. Finally, we use the results of stationarity and co-integration to analyze the long-run and short-run equilibrium relationships.

Results and Analyses
Stationarity Tests: Table 1 reports the stationarity results of ADF and PP tests which are carried on the logarithms of the variables in levels and using the deterministic components of intercept and trend. However, the results show that all variables are stationary at I(1) and 1% significance level. Therefore, it is reasonably to apply ARDL approach.

Co-integration Test:
The current paper employs the bounds F-statistics test as suggested by Pesaran et al. (2001) to test H0 of no co-integration among the variables. However, the long-run coefficients of the one lagged variables (Equations 2-4) are equal zero, i.e., H0: λij=0against H1:λij ≠ 0. The computed F-statistics value is compared with critical values tabulated in statistical tables at I(1) and I(0) [Pesaran et al., 2001]. If the computed F-statistic is greater than I(1), then, H0 of no co-integration is definitely rejected. Conversely, if the computed F-statistic is less than I(0), then, H0 is accepted (Pesaran et al., 2001). If the computed F-statistic falls between I(1) and I(0), then, the decision becomes inconclusive (Pesaran et al., 2001). Table 2 provides results of F-statistics computed and critical values. 3.24 * Notes: **, * denote the significance at 10% and 5% levels respectively. Source: Output of Micro-fit package, version 4. Table 2 illustrates that H0of no co-integration among variables in LFDIt and LEMt models are rejected at 10% significance level. Moreover, we reject H0 of no co-integration among variables in LESt model, since the computed F-statistics (3.24) falls between I(1) and I(0). Therefore, all the variables in LFDIt, LEMt and LESt models are co-integrated.
Policy Implications: The relationships between FDI inflows and economic growth are two ways directions. FDI inflows could enhance economic growth and similarly economic growth could attract FDI inflows. Metwally (2004) proposed that the economies that achieve economic growth succeed in attracting FDI. Conversely, he argued that FDI inflows enhance economic growth through the expansion of economic productive capacity and reduces the needs for borrowings that increase technological and managerial skills. Also, FDI inflows boost economic growth through the creation of employment. The findings of this paper are consistence with findings of McDonald et el. (2002) study that revealed, FDI inflows increased employment in European economic sectors. Therefore, Malaysian policy makers should be aware to the influences of FDI inflows on employment in Malaysian economic sectors. The benefits of FDI inflows in host country not only create jobs and increase employment in economic sectors, but also, increase capital, production and management techniques. Furthermore, FDI inflows through multinationals corporations upgrade the technology that is used in economic sectors which increases production's quality and the competitiveness in domestic and international markets.

Conclusion, Future Research and Limitations
Previous studies established that FDI inflows enhanced economic growth and increased employment in economic sectors. This paper contributes to the existing literature by examining the long-run and short-run equilibrium relationships among FDI inflows, EM and ES in Malaysia using ARDL Approach. The results of ADF and PP tests suggest that all variables are stationary at I(1) and 1% significance level. The findings of the bounds F-statistics test reveal the existence of co-integration among variables. Besides, EM and ES are positively associated with FDI inflows in the long-run and short-run. In fact, future research could broaden this study by including more economic sectors such as agriculture, forestry and fishing sector; mining and quarrying sector; and construction sector. From statistical perspective, the main limitation of this study is the small sample size of 40 observations which has a limiting factor, since the number of lags that use, consumes the number of observations and leads to specification errors in the analysis (Gujarati and Porter, 2008).