Foreign direct investment, exports and real exchange rate linkages in Vietnam: evidence from a co-integration approach.
The international economic literature has paid special attention to resolving the question of whether foreign direct investment (FD1) is a substitute for or a complement to international trade. The traditional Heckscher-Ohlin-Samuelson (HOS) model suggests that international trade can substitute for FDI by allowing for the free movement of labour and capital across countries. Based on the assumptions of the HOS model, Mundell (1957) finds that international labour and capital mobility constitute a perfect substitute for exports. In another pioneer study, Kojima (1975) concludes that FDI may have a positive impact on trade if production moves towards a country with a shortage of capital.
Other theoretical contributions seek to explain the complementary relationship between trade and vertical FDI. (1) Helpman (1984) suggests that "headquarter" countries tend to export capital equipment and production services to a host country. In return, the host country exports input resources to the home/headquarter country. Therefore, in the case of vertical FDI, there is a complementary link between the trade flows of final goods and intrafirm transfers of intermediate goods. Focussing on the differences in factor endowments and technological capabilities among countries, Markusen and Venables (1995, 1999) argue that trade and FDI between North-North countries can be considered a "substitute" for international trade while FDI and trade between North-South countries are likely to be "complementary" to international trade.
In addition to these direct connections, there is an indirect link between trade flows and FDI because each share a common determinant--the real exchange rate (RER). For these reasons, our paper aims to investigate the possible triangular relationship between exports, FDI and RER using Vietnam as a case study. We use a panel data set covering Vietnam's exports, inward FDI and the bilateral RER between Vietnam and its ten trading country partners. Vietnam's trading partners are not only important destinations of the country's exports but are also important sources of FDI. Our econometric methodology is as follows: we first test for the order of integration or the presence of "unit root" in our data sample. Secondly, we employ the heterogeneous panel co-integration technique developed by Pedroni (1999) to test for possible long-run co-integrated relationships between our variables (FDI, Export, RER). Finally, we employ a fixed-effects estimator to assess the channels through which the variables under consideration interact.
The remainder of this paper is organized as follows. Section II briefly reviews the export performance, inward FDI pattern and exchange rates policies of Vietnam from 1990 to 2010. Section III describes the panel data set tinder examination. Section IV details our methodology followed by an analysis of the empirical results. We end with concluding remarks in section V.
II. Vietnam's Economic Integration: An Overview
11.1 Inward FDI
As is the case with other developing countries, insufficient domestic capital investment and outdated technology greatly hinders Vietnam's economic growth. FDI is therefore considered a shortcut to overcoming this problem. At the beginning of the Doi Moi reform period, the government drafted the 1987 Foreign Investment Law which formed the legal basis for foreign investment into Vietnam. This piece of legislation was revised several times to improve the country's investment climate and incorporate national industrial policies. With its stabilized economy, abundant cheap labour, rich natural resources and an improved legal system, Vietnam grew to become an attractive Southeast Asian destination for FDI.
As displayed in Figure 1, Vietnam has seen an increase in inward FDI since 1990. Officially registered FDI inflows rose from US$189.7 million in 1990 to their peak of US$8.6 billion in 1996. Implemented FDI also grew rapidly from US$180 million in 1990 to US$2.5 billion in 1996. Several reasons inform this robust increase. Foreign investors were attracted to Vietnam's untapped market as well as to the country's abundant and cheap labour force. In addition to these reasons, Freeman (2002) points to three other external factors explaining the huge inflows of FDI into Vietnam. Firstly, in the 1980s and early 1990s, ASEAN countries in particular were key FDI recipients within Southeast Asia. Vietnam became more attractive to foreign investors after it joined ASEAN in 1995. Secondly, as Vietnam began to restructure its socialist economy into a market economy, the promise of new business opportunities attracted inward flows of FDI. Finally, Vietnam also emerged as an important destination of intra-regional FDI particularly from Malaysia, Singapore, and Thailand.
From 1997 to 1999, Vietnam experienced a dip in registered FDI inflows (49 per cent in 1997, 16 per cent in 1998 and 59 per cent in 1999). This is explained partly by the Asian Financial Crisis (AFC). The five major foreign investors of Vietnam were all from Asia and had to face their own challenges during the crisis. To maintain business operations in their local countries these investors had to postpone or cancel overseas expansion plans and FDI projects.
From 2000 to 2002, registered inward FDI to Vietnam increased again to 25.8 per cent in 2000 and to 22.6 per cent in 2001. However, these figures still made up less than two-thirds of the inward FDI figures in 1996. This decline in inward FDI was mostly due to the downfall of the global economy after the U.S. high-tech bubble burst. The years 2000 to 2002 also marked the enforcement of a bilateral trade agreement (BTA) between Vietnam and the United States. However, Parker et al. (2005) argue that the BTA has not had a positive effect on new registered FDI from the United States to Vietnam. In fact, the registered U.S. FDI fell substantially from US$159.5 million in 2002 to US$47.1 million in 2004. Implemented FDI from the United States to Vietnam, however, presents a more positive outlook--growing from US$61.3 million in 2002 to US$162.4 million in 2004. The years 2006 to 2007 witnessed a strong return of inward FDI to Vietnam. In 2006, Vietnam attracted over US$10.2 billion of newly registered capital, signalling an increase of 57 per cent; exceeding even the US$8.6 billion high attained in 1997.
Vietnam's accession to the WTO in 2007 had a positive impact on inward FDI. That year, both registered and implemented FDI grew tremendously to record highs of over US$21.35 billion and US$8.03 billion respectively. Furthermore, the annual growth rate of registered inward FDI almost tripled to 236 per cent from 2006 to 2008. Figure 1 also displays a sharp decline in inward FDI from 2008 to 2009. Total registered inward FDI dropped to 67.8 per cent in 2009 and then to 13.9 per cent in 2010. This would be an outcome of the Global Financial Crisis (GFC).
Turning now to the sources of inward FDI--up until 2010, Vietnam has received FDI from over ninety-two different countries and economies. Table 1 lists the ten most important FDI sources measured by registered and implemented FDI from 1990 to 2011. Japan is the largest foreign investor with US$24.4 billion of registered capital, followed by South Korea, Singapore, Taiwan and the British Virgin Islands. In general, these top ten investors account for about 75 per cent of Vietnam's total registered and implemented FDI inflows.
II.2 Vietnam's Trade Openness
Since the 1986 Doi Moi reform, international trade has become an increasingly important part of the Vietnamese economy. Vietnam aggressively promotes trade via various policies including trade liberalization (undertaken via several bilateral and multilateral commitments), tariff reductions and other measures designed to relax import-export restrictions. Vietnam has signed BTAs with more than sixty countries including the abovementioned BTA with the United States that came into force in December 2001. Vietnam is also implementing reforms to integrate itself into the ASEAN Free Trade Area (AFTA) with detailed plans for reducing tariff rates and removing other non-tariff barriers. As shown in Figure 2, Vietnam's international trade has been rapidly expanding since 1990.
Vietnam's export-led economic growth strategy has been very successful. The government has aggressively promoted exports by setting up export processing zones and devaluing the currency, for example. Vietnamese exports rose from US$2.74 billion in 1990 to US$35.41 billion in 2006. From 1990 to 2006, the annual growth rate of exports averaged 17.74 per cent while the share of exports to GDP increased from 36.04 per cent to 73.61 per cent. Exports have become a principal source of Vietnam's economic growth, employment creation, and poverty reduction. Simultaneously, Vietnam's imports rose from US$3.65 billion in 1990 to US$38.14 billion in 2006. In general, the share of Vietnam's trade to ratio of merchandise trade (the sum of exports and imports) to GDP was only 19 per cent at the end of 1988 but rose sharply to 74.72 per cent in 1995 and to 151.77 per cent in 2006. However, during 1990 to 2006, the share of imports to GDP exceeded that of exports by an average of about 6.45 per cent.
Vietnam's exports increased to US$39.41 billion and US$41.40 billion in 2007 and 2008 respectively. On the other hand, in terms of growth rates, Vietnam's exports decreased from 11.29 per cent to 5.05 per cent from 2007 to 2008. However, Vietnam's exports rebounded and increased from 11.08 per cent in 2009 to 14.65 per cent in 2010. In addition, WTO membership has a visible impact on Vietnam's imports. Vietnam's imports increased from US$48.68 billion to US$63.75 billion from 2007 to 2010, with an annual growth rate of 14 per cent.
Vietnam's rapid growth of exports has also been helped along by foreign invested enterprises (FIEs). Their share of total exports increased from 27.03 per cent in 1995 to 47.02 per cent in 2005 and subsequently to 54.2 per cent in 2010. This trend is an outcome of a boom in export-led FDI projects. For instance, light industries had become the number one exporting sector by 2007. They accounted for 44 per cent of total exports; 18 percentage points higher than 1990. The share of FDI in the light industry sector increased from 4 per cent in 1990 to 38 per cent in 2003, and thereafter decreased to 14 per cent in 2007, making it the second largest sector with a cumulative FDI. The oil and heavy industries attracted the largest amount of FDI. FDI towards these industries increased from US$166.1 million in 1990 to US$4.3 billion in 2007, making up approximately 54 per cent of total FDI into Vietnam. As a result, oil and heavy industry exports accounted for US$17.3 billion in 2007--making them the second largest export sectors after light industries. FDI has generally been concentrated in Vietnam's export sectors. Furthermore, the exports of foreign-invested sectors grew faster than the average rate. We may therefore consider a possible causal relationship between exports and FDI.
II.3 Exchange Rates Management of Vietnam
Up until the Doi Moi reform, Vietnam experienced triple-digit inflation (at 774 per cent in 1986 for example), multiple exchange rates, and a rapidly depreciating currency in the parallel market. Since 1990, the Vietnamese government began focussing on containing the country's inflation and stabilizing the dong (VND). To do so, in the final stage of disinflation, the State Bank of Vietnam (SBV) kept the VND/US$ exchange rate at around 11,000 from late 1991 to early 1997. This "11,000 VND policy" may be interpreted as an attempt to ensure price stability by the dollar peg. This reform allowed Vietnam to reduce its inflation rate to a very low level. However, overvaluation was one side-effect of this policy. From mid-1996, the SBV decided to depreciate the VND by broadening its band within an official central rate. Figure 3 depicts Vietnam's exchange rate movements from 1990 to 2010.
Vietnam had to cope with the impact of the AFC from 1997 to 1998. While Vietnam was not directly affected by speculators, the VND became overvalued relative to regional currencies, which fell sharply. The exchange rate band was further broadened to [+ or -] 5 per cent in February 1997 and to [+ or -] 10 per cent in October 1997. In February 1998, the official central rate itself was devaluated from 11,175 to 11,800 VND/US$. These adjustments brought the actual exchange rate to 12,980 at the most depreciated end of the revised band.
In February 1999, the SBV introduced a new exchange rate mechanism. The central rate was now set according to the daily average interbank exchange rate of each preceding transaction day with a very narrow band of [+ or -] 0.1 per cent. Upon the implementation of this mechanism, the VND started to depreciate very slowly to around 15,600. The subsequent period (2004-07) saw the creation of the Decree Law on foreign exchange management in December 2005. It aimed to: achieve a phased approach to the convertibility of the VND in foreign transactions; ameliorate the foreign exchange management system; improve the payment balance; and boost economic growth. This Law came into effect in June 2006. During this period, the official VND/US$ rate remained mostly unchanged and increased only slightly from 15,664 in 2004 to 16,055 in 2006.
Vietnam's accession to the WTO did not trigger any major change in the official VND/US$ rate. The VND slightly depreciated in nominal terms and was valued at 16,105 by the end of 2007. However, Vietnam's foreign exchange market experienced some turbulence due to the GFC. The SBV had to broaden the VND's trading band against the US$ to [+ or -] 5 per cent in March 2009, in order to make up for the shortage of foreign currency in the formal market. Furthermore, the spread between the official exchange rate and the black market exchange rate increased to more than 15 per cent in November 2009. To tackle this issue, the SBV narrowed the trading band to [+ or -] 3.0 per cent and devaluated the official exchange rate by 5.4 per cent in November 2009 and then by 3.4 per cent in February 2010.
III. Data and Methods
Our analysis is based on annual panel data spanning 1994 to 2010. The data covers Vietnam's inward FDI, exports and bilateral RER with its ten trading country partners. (2) The variables trader consideration, which have been converted into logarithms, are identified as follows:
* [EX.sup.i.sub.t] represents exports from Vietnam to country i at year t in million constant 1995 U.S. dollars.
* [FDI.sup.i.sub.t] represents foreign direct investment flows into Vietnam from country i at year t in million constant 1995 U.S. dollars. We analyse data on implemented FDI inflows because many foreign investors who invested in Vietnam during the period under examination failed to register their projects with the Ministry of Planning and Investment early on. Instead, they registered their investment projects only in the years following the commencement of their respective projects. This explains why the officially registered FDI cannot be used as a consistent and accurate measurement for FDI activities in Vietnam for our work and also for other rigorous researches.
* [RER.sup.i.sub.t], which is calculated as the product of the nominal exchange rate and relative price levels in each country, represents the bilateral real exchange rate between Vietnam and country i at year t. The real exchange rate between foreign country i and Vietnam at time t is therefore:
[RER.sup.i.sub.t] = [e.sub.i,t] x [p.sup.*.sub.i.t] / [p.sup.VN.sub.t] (1)
where, [p.sup.VN.sub.t] is the price level of Vietnam, [p.sup.*.sub.i,t] is the price level in foreign country i, and [e.sub.i] is the nominal exchange rate between the VND and the currency of foreign country i. [e.sub.i] is expressed as the number of VND units per foreign currency unit, so that [e.sub.i] rises with a depreciation in the dong. Equation (1) suggests that we should expect to find a positive coefficient on the real exchange rate in all estimated regressions--where an increase in the bilateral real exchange rate represents a real depreciation of the VND. To construct the bilateral RER between Vietnam and a foreign country, we use consumer price indices (CPI). These are advantageous because they are timely, similarly constructed across countries, and available for a wide range of countries over a long time span.
In addition, we evaluate the potential impact of financial crises on the FDI decisions of donor countries as well as on Vietnam's exports by introducing a binary dummy in each estimated equation. In order to determine the value of this dummy, we adopt the work of Laeven and Valencia (2008), in which the authors present a new database on the timing of systemic financial and banking crises as well as policy responses to resolve them. This binary dummy takes the value of 1 if the FDI source country and/or Vietnam actually suffer from a financial crisis; and takes the value of 0 in the opposite scenario. Inward FDI and exports of a country are typically determined by many other macroeconomic factors. We, however, only add the three abovementioned variables in our estimated equation to keep our empirical analysis simple.
IV. Methodology and Empirical Results
To investigate the possible links between inward FDI, exports and bilateral RER, we apply three consecutive econometric tests: (i) the unit-root test; (ii) the co-integration test; and (iii) the instrumental variable estimator.
IV.1 Panel Unit Root Test
A large number of econometric methodologies have been developed to test for the panel unit root. Of these, the LLC test (Levin, Lin and Chu 2002) and the IPS test (Im, Pesaran and Shin 2003) are most frequently applied. The LLC test assumes homogeneity in the dynamics of autoregressive (AR) coefficients for all panel members. It assumes that each individual unit in the panel shares the same AR (1) coefficient but also allows for individual effects, time effects and possibly a time trend. Lags to the dependent variables may be introduced to allow for serial correlations in the errors. The test may be viewed as a pooled or augmented Dickey-Fuller (DF) test when lags are introduced to the null hypothesis of non-stationarity (I behavior). The t-star statistic becomes a distributed standard normal under the null hypothesis of non-stationarity following a transformation.
The IPS test also assumes that all series are non-stationary under the null hypothesis. However, it is more general than the LLC test because it allows for heterogeneity in a dynamic panel. It is therefore described as a "Heterogeneous Panel Unit Root Test". It is reasonable to allow for such heterogeneity in choosing the lag length in the augmented Dickey-Fuller (ADF) test especially when the imposition of a uniform lag length is deemed inappropriate. In addition, the IPS test allows for individual effects, time trends, and common time effects. The exact critical values of the t-bar statistic are provided in the IPS test. For these reasons, the IPS test is deemed more helpful than other tests including the LLC test. We report the results of the LLC and IPS tests in Table 2.
In the LLC test, the large negative statistics values for each of these variables--bilateral FDI flows and exports--exceed the critical values (in absolute terms). This means that we can reject the null hypothesis of non-stationarity at the 1 per cent and 5 per cent significance levels. As with the LLC test, the IPS results show that the series are stationary at their respective levels and are integrated (order one [I(1)]) in levels of at least 5 per cent significance. Based on the LLC and IPS test results, we may conclude also that all variables are stationary and integrated (order one) in levels of at least 5 per cent significance. Having established that FDI, Exports and RER series are integrated, we employ the panel co-integration technique to determine the nature of the long-run relationship between these variables of interest.
IV.2 Panel Co-integration
Most of the recent empirical studies examining long-run co-integrated relationships employ the heterogeneous panel co-integration test developed by Pedroni (1999). The Pedroni test permits different individual cross-sectional effects by allowing for the occurrence of heterogeneity in the intercepts and slopes of the co-integrating equation. The Pedroni panel co-integration technique makes use of a residual-based ADF test. The Pedroni test for the co-integrated relationship between bilateral FDI, bilateral RER and Vietnamese exports in our panel is based on the estimated residuals from the following long-rim model:
[FDI.sup.i.sub.t] = [[beta].sub.0i] + [[beta].sub.1i][EX.sup.i.sub.t] + [RER.sup.i.sub.t] + [[epsilon].sub.1it] (2)
where, i = 1, ..., 10 countries and t = 1, ..., 18 period observations. The term [[epsilon].sub.1it] = [[rho].sub.1i][[epsilon].sub.i(t-1)] + [zeta][1.sub.it] represents deviations from the modelled long-run relationship. If the series are co-integrated, [[epsilon].sub.1it] should be a stationary variable. The null hypothesis in Pedroni's test procedure is whether [[rho].sub.i] is unity. In addition, the Pedroni technique allows us to test for the co-integrated relationship between bilateral FDI and exports using four different models: a model with heterogeneous trend but ignoring common time effect (M1); a model without a heterogeneous trend or common time effect (M2); a model without a heterogeneous trend but allowing common time effect (M3); a model with a heterogeneous trend and common time effect (M4). All the Pedroni statistics of the different model specifications are reported in Table 3.
Our econometric results include seven different statistics for the null hypothesis test (of the lack of co-integration) in our heterogeneous panels. The first group of tests is termed "within dimension" and includes: (i) the "panel v-stat" and the "panel rho-stat" which are similar to the Phillips and Perron (1988) test; and (ii) the "panel pp-stat" (panel non-parametric) and the "panel adf-stat" (panel parametric) which are analogous to the single-equation ADF-test. The second group of tests termed "between dimensions" is comparable to the group mean panel tests of Im, Pesaran and Shin (2003). The "between dimensions" tests consist of the "group rho-stat", "group pp-stat", and the "group adf-stat" tests. The calculated test statistics reject the null hypothesis of the absence of co-integration at the 1 per cent level for both the FDI and Export regressions. These results allow us to conclude that a long-run relationship does exist between the variables under consideration.
IV.3 The IV Estimator's Estimator
The previous section highlights the presence of a co-integrating relationship among the variables but does not determine the channels through which the variables may affect each other. To do so, we employ another econometric technique to estimate two separate equations: a FDI regression and an Export regression. However, simply including export flows or inward FDI in the FDI or Export equations can potentially induce endogeneity. To counter this, we employ the Instrumental Variables (IV) method--a widely known solution to endogenous regressors. This method also provides a way of obtaining consistent parameter estimates. However, determining the IV in each estimated equation is not an easy task. In this article, the export variable is instrumented by the level of hidden import barriers of the destination countries (3) and Vietnam's export price. (4) Secondly, under the assumption that Vietnam's relatively low labour cost could have an impact on attracting FDI; the FDI variable is instrumented by Vietnam's average monthly wages. (5)
Firstly, we test for the validity of each instrumental variable. In the lower part of Table 4, we report the weak instrument test proposed by Stock and Yogo (2002) and the Hansen/Sargan test of over-identifying restrictions. The Cragg-Donald F-statistics within the weak instrument test are superior to the critical value of 10 per cent of the maximal IV size proposed by Stock and Yogo (2002) and therefore rejects the null hypothesis of weak instruments. On the other hand, the Sargan/Hansen test of over-identifying restrictions, which is reported in the last line, checks the validity of the instruments. According to these empirical results, we cannot reject the null hypothesis of the Sargan/Hansen test which means that the instruments are valid though notably uncorrelated with the error term and that the excluded instruments are accurately removed from the estimated equation.
We now examine the IV estimator's main results which are reported in the upper part of Table 4. In the FDI regression, the dependent variable is Vietnam's inward FDI and the independent variables include Vietnamese exports to FDI home countries and bilateral RER. In the Export regression, the dependent variable is exports from Vietnam to FDI home countries and the independent variables are bilateral RER and FDI flows into Vietnam. In each regression, we have introduced a dummy variable DU which accounts for the possible impact of a financial crisis on Vietnam's inward FDI and exports.
Firstly, from Table 4, we can observe a strongly positive and causal link between Vietnam's inward FDI and exports from Vietnam to the FDI home country. This indicates that the FDI flow into Vietnam is oriented towards exports. In other words, FDI inflow encourages Vietnamese exports. This result is consistent with Helpman's (1984) theory that headquarters tend to export capital equipment and services such as research and development to the host country. In return, the host country exports input resources to the home country. Simultaneously, FDI flows into Vietnam (from a specific country) can also play an important role in promoting exports from Vietnam to other countries. Foreign investors may use Vietnam as a platform for selling their products to third markets worldwide. Secondly, the results show that when the VND depreciates with respect to foreign currency (i.e. when RER increases), there is a relative increase in Vietnam's inward FDI. Specifically, a 1 per cent depreciation of the VND causes an increase of 0.172 per cent in FDI flows into Vietnam. This may be explained by two reasons. The first being that a depreciation of the bilateral RER reduces the cost of domestic labour and other production inputs relative to foreign production costs. As such, a reduction in Vietnam's production costs may attract more foreign investors. The second would be the imperfection of Vietnam's capital markets. Due to this imperfection, a real depreciation of VND improves the capital wealth of foreign investors relative to that of domestic investors thereby increasing FDI flows into Vietnam. The main empirical result also indicates that the bilateral RER depreciation has a positive effect on Vietnam's exports. A depreciation of 1 per cent of RER with respect to foreign currency causes an increase in Vietnam's exports by 0.219 per cent. Finally, in each of our models, the statistical values of the DU suggest that this dummy variable (which represents financial crises) negatively affects FDI and Vietnam's exports at the 5 per cent level of significance, if not more.
V. Concluding Remarks
Relying on annual panel data from 1990 to 2010, we examined the possible connections between bilateral RER, inward FDIs and Vietnamese exports using a panel co-integration analysis. We also expanded our empirical analysis to identify the channels through which these variables affect each other. This was done using the IV estimator. Several conclusions can be drawn from our empirical research.
Firstly, our results reveal a number of statistically significant links between FDI, RER and Exports. For example, we found that a real depreciation of the VND leads to an increase in both inward FDI and exports. We also found evidence of a causal relationship between FDI and Vietnamese exports. Moreover, Vietnamese exports to a specific country have been promoted not only by FDI from the destination country but also by FDI into Vietnam from other source countries. This emphasizes the importance of the role of export growth in attracting inward FDI flows as well as the relevance of the Vietnamese government s export-oriented FDI policy.
Secondly, the set of relationships between bilateral RER and FDI or exports and between FDI and exports allows us to determine two possible channels through which the RER affects exports flows: (i) a direct impact on the relative price of goods; and (ii) an indirect impact through FDI flows.
Finally, although some existing studies argue that Vietnam's exchange management is inefficient (Ohno 2003), (6) we strongly believe, based on our empirical findings, that Vietnam's policy-makers have made commendable efforts to improve the determinant role of exchange rate policies particularly with regard to investment inducement and export expansion.
Dickey, D.A. and W.A. Fuller. "Distribution of the estimators for autoregressive time series with a unit root". Journal of the American Statistical Association 74, no. 366 (1979): 427-31.
Freeman, N.J. "Foreign direct investment in Vietnam: An overview". Presented at Workshop on Globalisation and Poverty in Vietnam, Department for International Development (DFID), Hanoi, Vietnam, 23-24 September 2002.
Helpman, E. "A Simple Theory of international Trade with Multinational Corporations". Journal of Political Economy (1984): 451-71.
Im, K., M.H. Pesaran, and Y. Shin. "Testing for unit roots in heterogeneous panels". Journal of Econometrics 115, no. 1 (2003): 53-74.
Kojima, K. "International trade and foreign investment: substitutes or complements". Hitotsubashi Journal of Economics 16 (1975): 112.
Laeven, L. and F. Valencia. "Systemic banking crises: A new database". IMF Working Paper, No. 08/224. Washington, D.C.: International Monetary Fund, 2008.
Levine, A., Lin C.E and Chu C.S. "Unit root tests in panel data: asymptotic and finite-sample properties". Journal of Econometrics 108 (2002): 1-24.
Markusen, J.R. and A.J. Venables. "Multinational Firms and the New Trade Theory". NBER Working Paper 50, no. 36, Cambridge, Mass., 1995.
Markusen, J.R. and A.J. Venables. "Foreign direct investment as a catalyst for industrial development". European Economic Review 43 (1999): 335-56.
Mundell, R. "International Trade and factor Mobility". American Economic Review 47, no. 3 (1957): 321-35.
Ohno, K. "Exchange Rate Management of Vietnam: Re-examination of Policy Goals and Modality", VDF Working Paper, No. 09/2003.
Parker, S., Phan V.Q. and Nguyen N.A. "Has the U.S.-Vietnam bilateral trade agreement led to higher FDI into Vietnam?". International Journal of Applied Economics 2 (2005): 199-223.
Pedroni, E "Critical Values for Cointegration Tests in Heterogeneous Panels with Multiple Regressors". Oxford Bulletin of Economics and Statistics 61, Special Issue (1999): 653-78.
Stock, J.H. and M. Yogo. "Testing for weak instruments in linear IV regression". NBER Technical Working Paper, No. 284. Cambridge, MA: NBER, 2002.
(1.) Vertical FDI implies that a multinational firm has headquarters located in the home country and several branches located in the host country where production costs and input resources are cheaper.
(2.) Countries/economies in the sample are: Japan, United States, France, Thailand, Singapore, Australia, Malaysia, Hong Kong, Taiwan and Korea.
(3.) Data on the levels of hidden import barriers is collected from the 2012 KOF Globalization Index.
(4.) Data on Vietnam's export price is collected from the MPI.
(5.) Data on average monthly wages in Vietnam is collected from the MPI.
(6.) According to Ohno (2003), the problem with Vietnam's exchange rate policy is that the exchange rate mechanism of setting the central rate according to the daily average interbank exchange rate of each preceding transaction day is imperfect since it is merely a technical procedure without analytical links to economic fundamentals.
Thi Hong Hanh Pham is an Associate Professor at the Institute of Economics and Management, University of Nantes, France.
Thinh Due Nguyen is Head of the Parliament Relations Bureau at the Department of National Economics Synthesis, Ministry of Planning and Investment, Vietnam.
TABLE 1 Vietnam's Sources of FDI (as of 31 December 2011) Registered capital No Country, Territory US$ million % 1 Japan 24,381.7 12.2 2 Korea 23,695.9 11.9 3 Taiwan 23,638.5 11.9 4 Singapore 22,960.2 11.5 5 British Virgin Islands 15,456.0 7.8 6 Hong Kong 11,311.1 5.7 7 Malaysia 11,074.7 5.6 8 USA 10,431.6 5.2 9 Cayman Islands 7,501.8 3.8 10 Thailand 5,853.3 2.9 Implemented capital No Country, Territory US$ million % 1 Japan 1,3486.74 11.6 2 Korea 1,3107.39 11.3 3 Taiwan 1,3075.64 11.2 4 Singapore 1,2700.44 10.9 5 British Virgin Islands 8,549.489 7.3 6 Hong Kong 6,256.737 5.4 7 Malaysia 6,125.972 5.3 8 USA 5,770.242 5.0 9 Cayman Islands 4,149.622 3.6 10 Thailand 3,237.754 2.8 SOURCE: Foreign Investment Agency, MPI. TABLE 2 Unit Root Tests LLC Unit Root test Level EX FDI RER (1) -- -- -- 7.87 *** 6.65 *** 7.89 *** (2) -- -- -- 11.44 *** 11.70 *** 11.43 *** (3) -- -- -0.78 * 4.13 *** 2.53 *** IPS test Level With common time effect EX FDI RER (1) (a) -- -1.90 ** -- 2.13 *** 1.94 ** (2) (b) -- -2.59 ** -- 2.58 *** 2.59 ** IPS test Level Without common time effect EX FDI RER (1) (a) -1.87 ** -- -- 1.96 ** 2.24 *** (2) (b) -2.55 ** -- -- 2.05 ** 2.72 *** NOTES: (1) Model with heterogeneous intercepts. (2) Model with heterogeneous intercepts and heterogeneous trend. (3) Model without heterogeneous intercepts. *** (**) Rejection of the null hypothesis at the 1 per cent (5 per cent) significance level. (a.) The critical value at 1 per cent, 5 per cent and 10 per cent is -2.00, -1.86 and -1.78 respectively. (b.) The critical value at 1 per cent, 5 per cent and 10 per cent is -2.63, -2.49 and -2.42 respectively. TABLE 3 Pedroni Panel Co-integration Tests Test statistics FDI regression M1 M2 M3 M4 panel v-stat -0.41 -1.02 0.85 -1.54 panel rho-stat -5.65 -5.85 -7.11 -5.15 panel pp-stat -11.02 -11.05 -12.45 -13.09 panel adf-stat -6.89 -7.23 -7.89 -11.25 group rho-stat -6.05 -5.89 -0.23 -6.45 group pp-stat -13.24 -12.48 -13.45 -14.01 group adf-stat -11.34 -7.01 -7.35 -7.89 Test statistics Exports regression M1 M2 M3 M4 panel v-stat -0.95 -1.45 -0.79 -1.34 panel rho-stat -5.15 -5.46 -6.03 -6.23 panel pp-stat -8.08 -8.23 -7.45 -7.55 panel adf-stat -5.34 -5.01 -5.65 -0.03 group rho-stat -5.09 -5.86 -6.17 -0.75 group pp-stat -7.89 -8.75 -9.16 -10.24 group adf-stat -0.02 -6.17 -7.34 -7.99 TABLE 4 Instrumental Variable Estimator's Results Explanatory variables Regressions FDI regression Exports regression 1. Inward FDI to Vietnam -- 1.067 (0.174) *** 2. Exports to FDI 1.177 (0.260) *** -- home country 3. Real exchange rate 0.172 (0.055) *** 0.219 (0.073) *** 4. Crisis dummy -0.733 (0.318) ** -0.943 (0.385) *** Constant -7.102 (5.011) 2.652 (2.725) Cragg-Donald Wald 19.349 30.019 F-statistic F-statistic (11.59) (a) (8.96) (a) Sargan Statistic 0.178 0.000 [0.6729] [1.000] P-value of Sargan test (Equation exactly identified) NOTES: Values in brackets are P-values. Values in parentheses are robust standard errors. *** ** *: Significant at the 1 per cent, 5 per cent, 10 per cent levels, respectively. (a.) critical value of the 10 per cent maximal IV size proposed by Stock and Yogo (2002).
|Printer friendly Cite/link Email Feedback|
|Author:||Pham, Thi Hong Hanh; Nguyen, Thinh Duc|
|Publication:||Journal of Southeast Asian Economies|
|Date:||Dec 1, 2013|
|Previous Article:||The early effects of preferential trade agreements on intra-regional trade within ASEAN+6 members.|
|Next Article:||Financial sector development and income inequality in Vietnam: evidence at the provincial level.|