Is foreign direct investment productive?

A case study of the regions

of Vietnam
Sajid Anwar
a,c,
, Lan Phi Nguyen
b,1
a
School of Business, University of the Sunshine Coast, Maroochydore, QLD 4558, Australia
b
Monetary Statistics and Forecasting, State Bank of Vietnam, Hanoi, Viet Nam
c
IGSB, University of South Australia, Adelaide, SA 5001, Australia
a b s t r a c t a r t i c l e i n f o
Article history:
Received 1 February 2013
Received in revised form 1 July 2013
Accepted 1 August 2013
Available online 27 September 2013
Keywords:
Foreign direct investment
Total factor productivity
Technological progress
Panel data analysis
Vietnam
By making use of a recently released dataset that covers a large number of manufacturing rms over the period
20002005, this paper examines the impact of foreign direct investment (FDI) and FDI generated spillovers on
total factor productivity (TFP) in eight regions of Vietnam. Unlike most existing studies, this paper focuses on
the impact of spillovers that take place through both horizontal and vertical linkages. The results presented in
this paper suggest that the impact of FDI spillovers onTFP varies considerably across regions. FDI spillovers generate
a strong positive impact on TFP through backward linkages only in Red River Delta, South Central Coast, South East
and Mekong River Delta while in other regions the impact is negative and mostly insignicant. The paper also
examines the impact of the absorptive capacity on TFP growth in each of the eight geographical regions.
2013 Elsevier Inc. All rights reserved.
1. Introduction
Economic reforms create international business opportunities. Due to
the rising cost of doing business in China, Vietnamhas become a popular
destination for foreign investment. A number of existing studies have
examined the impact of foreign direct investment (FDI) and FDI-related
spillover effects on rm productivity and export behavior in developed
as well as developing countries. This paper focuses on Vietnam, a country
that due to the rising cost of doing business in China is now attracting
signicant FDI. However, due to lack of appropriate data, so far, relatively
few studies have considered the case of Vietnam.
In recent years, the government of Vietnamhas started releasing rm
level data, including data onFDI inows that couldbe usedtoexamine the
impact of FDI-related spillover effects on rmperformance. Most existing
studies, for example Anwar andNguyen(2010a) andAthukorala andTien
(2012), focus only on the direct effect of FDI on rm performance in
Vietnam. In addition, the existing studies (such as Anwar and Nguyen,
2010b) are highly aggregated and hence do not present a clear picture
of the impact of FDI on different regions of Vietnam. An important role
of the government is to take steps to reduce regional economic disparity.
A region by region analysis of the impact of FDI-related spillover effects
(i.e., the indirect effect) on rm productivity can provide useful informa-
tion to domestic policy makers.
By making use of a rm level panel dataset, this paper empirically
examines the impact of FDI generated horizontal and vertical spillovers
on total factor productivity (TFP) of manufacturing rms located in all
eight regions of Vietnam.
The rest of this paper is structured as follows. Section 2 contains a
review of related studies. Section 3 includes a brief description of the
methodology. The empirical results are presented and discussed in
Section 4. Section 5 contains some concluding remarks.
2. Review of related studies
While a number of studies have examined the impact of FDI on rm
productivity and GDP growth, relatively few studies have explicitly
considered the impact of FDI-related spillover effects on productivity.
Liu (2002) and Liu and Wang (2003) consider the effect of FDI on tech-
nology transfer in China but they do not consider the impact of spillover
effects. Pan (2003) outlines source and host country factors that can
affect foreign direct investment in China. Bwalya (2006) examines the
nature of spillover from foreign to domestic rms by using rm level
data on Zambian manufacturing rms for the period 19931995.
Bwalya nds little evidence in support of technology spillovers from
foreign rms to local rms through horizontal channels, suggesting
that the productivity of local rms is negatively affected by the scale of
operation of foreign rms.
Stancik (2007) considers the effect of FDI on the growth rate of
sales of domestic rms in the Czech Republic by using rm level panel
data from1995 to 2003. Stancik focuses on the impact of FDI spillovers.
The empirical results suggest that the presence of foreign rms has
Journal of Business Research 67 (2014) 13761387
Corresponding author. Tel.: +61 7 5430 1222.
E-mail addresses: SAnwar@usc.edu.au (S. Anwar), nplan_dbtk@sbv.gov.vn
(L.P. Nguyen).
1
Tel.: +84 91 775 1123.
0148-2963/$ see front matter 2013 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jbusres.2013.08.015
Contents lists available at ScienceDirect
Journal of Business Research
adversely affected most domestic rms in the CzechRepublic, especially
in upstream sectors.
Bitzer, Geishecker, and Grg (2008) use industry level data for 17 Or-
ganisation for Economic Cooperation and Development (OECD) countries
to investigate the importance of horizontal and vertical spillovers. Their
empirical work shows that spillovers through vertical backward linkages
between multinationals and domestic rms exist in all 17 countries. They
also found evidence to support the existence of positive spillover effects
from horizontal FDI. Using annual data on 44 host countries over the pe-
riod 19832003, Beugelsdijk, Smeets, and Zwinkels (2008) argue that FDI
generatedhorizontal andvertical effects have a signicant positive impact
on host developed countries but those benets to host developing coun-
tries are insignicant. Liu (2008) examines the effect of technology trans-
fer arising from FDI in the Chinese manufacturing sector, nding that
spillovers through backward and forward linkages have a positive effect
onproductivity of domestic rms andbackwardlinkages are the most im-
portant channel through which technology spills over fromforeign to do-
mestic rms. Using an event history technique, Lin (2010), focuses on the
determinants of Taiwanese investment in Chinas IT industry. Du,
Harrison, and Jefferson (2012) examine the impact of FDI-related hori-
zontal and vertical spillovers on the productivity of Chinese manufactur-
ing rms. They nd that horizontal spillovers have a weak effect on
productivity but the impact of vertical spillovers is statistically signicant.
In summary, the empirical evidence regarding the impact of FDI gen-
erated spillovers on host economies is mixed (see Table 1, which includes
only some of the recent studies). A review of earlier studies can be found
in Meyer and Sinani (2009), Table 1. Some existing studies suggest that
the impact of FDI generated spillover effects also depends on the absorp-
tive capacity of host countries. This means that the impact of FDI spillovers
on host countries can vary from country to country and from industry to
industry.
2.1. FDI in Vietnam
Due to lack of data, only a few studies attempt to investigate the
impact of FDI on Vietnam. These studies consider a number of issues.
For example, Jenkins (2006) considers the direct effect of FDI on employ-
ment growth in Vietnam(especially in early years), nding only a limited
positive effect. Athukorala and Tien (2012) suggest that, in recent years,
FDI inows have played an important role, not only in providing invest-
ment capital but also in stimulating export growth. Le (2005) examines
the effect of FDI on labor productivity of 29 Vietnamese manufacturing
industries over the period 19952002. Le found FDI's effect on productiv-
ity to be positive. Nguyen and Nguyen (2007) used rm-level data to in-
vestigate the effect of FDI spillovers. They conclude that FDI has resulted
in improved labor productivity. Nguyen et al. (2008) used rm level
data to investigate the effect of FDI-related horizontal, backward and
forward spillover effects in Vietnam's manufacturing and services sec-
tors. However, they do not consider the effect of FDI-related spillovers
on TFP growth. The same applies to the work of Athukorala and Tien
(2012). Anwar and Nguyen (2010a) found that a two-way mutually
reinforcing relationship between output growth and FDI growth
exists in Vietnam. However, TFP is a better measure of productivity
(Lipsey & Carlaw, 2004). Other studies on Vietnam include Binh and
Haughton (2002), Giroud (2007), Anwar and Nguyen (2010b, 2011a,
2011b), and Nguyen and Sun (2012). These studies have focused on
the impact of FDI on trade and export performance. Unlike the existing
studies, this paper focuses on the impact of indirect effect of FDI on all
eight regions of Vietnam and TFP is used as a measure of productivity.
FDI in Vietnam concentrates mostly in the key economic regions,
namely Red River Delta, North East, South East, and Mekong River Delta,
while remote regions receive only a small fraction of FDI (see Table 2).
Table 2 shows that differences between regional economic growth
rates and per-capita income across regions are signicant. In 2005, the
per-capita income in South East was more than twice the national aver-
age. The per-capita income in Red River Delta was the same as the na-
tional average, but income in North West was approximately 40% of
the national average (see Table 2). These differences can inuence the
magnitude of FDI-relatedhorizontal and vertical spillovers. Accordingly,
a region-wise analysis can present a better picture of the benets from
FDI (Table 3).
The broad hypotheses tested in the following section include (1) the
impact of FDI and FDI-related spillovers on productivity of domestic
rms varies across the regions of Vietnam and (2) absorptive capacity
(as measured by human capital, technology gap with foreign rms and
the level of nancial development) enhances the FDI spillover effects
across the regions of Vietnam.
Table 1
FDI and economic performance A summary of some recent studies.
Bwalya (2006) 19931995 Zambia Panel data FDI spillovers through horizontal channels have a positive impact on rm output.
Abu-Bader and Abu-Qarn
(2008)
19602001 Egypt Time series There is a bi-directional relationship between nancial development and economic growth.
Alvarez and Lopez (2008) 19901999 Chile Panel data Exporting leads to positive spillover effects.
Beugelsdijk et al. (2008) 19942003 44 countries Panel data Vertical FDI has a stronger positive impact on productivity.
Bitzer et al. (2008) 19892003 17 OECD Countries Panel data FDI linked backward spillovers have a positive impact on productivity.
Liu (2008) China 19951999 Panel data FDI can have a negative impact on productivity in the short term but its impact on long term
productivity is positive.
Barbosa and Eiriz (2009) Portugal 19941999 Panel data FDI spillovers do not have a signicant impact on rm productivity.
Suyanto et al. (2009) Indonesia 19882000 Panel data FDI and FDI-related spillovers have a positive impact on productivity.
Anwar and Nguyen (2010) 19902007 Vietnam Panel data There is a bi-directional relationship between FDI and economic growth.
Wang (2010) 19731991 Canada Panel data FDI generates strong positive impact on productivity.
Anwar and Nguyen (2011a) 2004 Vietnam Cross section data FDI has contributed to increase in net exports in the post Asian crisis period.
Anwar and Nguyen (2011b) 20002007 Vietnam Panel data FDI spillovers can have positive impact on export performance of domestic rms.
Nguyen and Sun (2012) 20032004 Vietnam Panel data FDI spillovers improve rm export performance.
Athukorala and Tien (2012) 20002005 Vietnam Time series data FDI has a positive impact on real output.
Du et al. (2012) 19982007 China Panel data FDI benets local rms through both vertical and horizontal linkages.
Fernandes and Paunov (2012) 19922004 Chile Panel data FDI has a positive impact on productivity of rms in both manufacturing and services sectors.
Table 2
FDI in regions of Vietnam (1988 to 2005).
Source: GSO (2013).
Regions of Vietnam Number of
FDI projects
FDI in million
US dollars at
constant prices
FDI projects
in percentage
of the total
Red River Delta 1239 14884.3 20.10
North East 291 1945.5 4.72
North West 23 100.6 0.37
North Central Coast 90 1368.8 1.46
South Central Coast 280 3476.3 4.54
Central Highlands 94 1001.1 1.52
South East 3831 32380.5 62.15
Mekong River Delta 268 1812.9 4.34
1377 S. Anwar, L.P. Nguyen / Journal of Business Research 67 (2014) 13761387
3. The model and data
The empirical model used in this paper is developed in two stages.
Stage one species a standard CobbDouglas model that can be used
to estimate TFP of each region. TFP is the dependent variable, whereas
FDI-related spillovers are the independent variables. While the focus
of the empirical exercise is on the impact of FDI-related spillovers on
TFP, in stage two, we also identify a number of control variables that
can affect TFP. The inclusion of the control variables in our regression
equation serves to reduce the severity of omitted variable bias, which
can reduce the reliability of the estimated results.
Accordingly, stage two, in order to correctly estimate the impact
of FDI-related spillovers on TFP, species an empirical model where
FDI-related spillovers and other variables appear as independent
variables. As all identied independent variables cannot be directly
measured, following the existing literature, we use several proxies.
Similar proxies are widely used in the existing literature (for example
see Wang, 2010).
In order to focus on the impact of FDI-related spillover effects on
productivity and economic growth, in stage one, we remove the impact
of growth in capital and labor on regional production growth, which in-
volves calculation of TFP. In stage one, TFP of Vietnamese rms in
manufacturing industries in all eight regions of Vietnam is estimated
by means of the following production function.
Y
ijt
A
ijt
K

ijt
L

ijt
1
where Y
ijt
is the real output of domestic rmi inindustry j inperiodt; K
ijt
is the real book value of xed assets of rmi (following Blomstrmand
Sjoholm, 1999, the book value is used as a proxy for capital stock); L
ijt
is
the number of workers employedby rmi; A
ijt
is TFP of rmi inindustry
j in period t; and and respectively are the production elasticities of
capital and labor.
TFP (in log form) can be calculated by rewriting Eq. (1) as follows:
ln A
ijt

ln Y
ijt

ln K
ijt

ln L
ijt

: 2
In stage two, the impact of FDI spillovers and other control variables
(i.e., absorptive capacity) on TFP is evaluated by estimating Eq. (3) as
follows:
ln A
ijt

a
0
a
1
ln H
ijt

a
2
S
ijt
a
3
C
jt
a
4
T
ijt
a
5
F
ijt
a
6
H FDI
jt
a
7
B FDI
jt
a
8
F FDI
jt
c
ijt
3
where H is human capital of rm i in industry j in period t, which is
approximated by real expenditure on education and training; S
ijt
is
sales of each of the domestic rms relative to the total industry sales
in period t; C
jt
is the Herndahl index, which is a measure of the size
of rms within industry j; T
ijt
is the technology gap which is measured
by the difference in average productivity of domestic and foreign rms
in percentage terms in the same industry; F
ijt
is the level of nancial
development within industry j, which is measured by working capital
as a percentage of total assets; H _ FDI
jt
,B _ FDI
jt
and F _ FDI
jt
respective-
ly are measures of horizontal, backward-vertical and forward-vertical
FDI spillovers.
By making use of the existing literature, (for example Grima, Grg &
Pisu, 2008 and Wang, 2010), FDI spillovers are measured as follows:
The horizontal spillover in industry i at time t, say H_FDI
it
, is calculat-
ed as the proportion of output accounted for by foreign rms in that
industry.
The vertical backward spillover in industry i at time t is computed as
follows:
B FDI
it

X
ki

ki
H FDI
ki
4
where
ki
is the proportion of industry i's output supplied to in-
dustry k. It is assumed that the greater the proportion of output
supplied to an industry with foreign multinational presence, the
greater the degree of linkages between foreign and domestic
rms.
The vertical forward spillover in industry i at time t is calculated
as follows:
F FDI
it

X
ki

ki
H FDI
ki
5
where
ki
represents the proportion of industry k's output sup-
plied to industry i. This measure captures the extent of forward
linkages between domestic rms in downstream and foreign
rms in upstream industries. The values of and are obtained
from the Inputoutput Tables of Vietnam published by the Gen-
eral Statistic Ofce of Vietnam (GSO).
The Herndahl index, which is a measure of concentration in an in-
dustry j, is dened as follows:
HERF
jt

X
n
i1
x
ijt
X
jt
!
2
i 1; 2; ; n
where x
ijt
is the sales of rm i in industry j; X
jt
denotes the total sales of
industry j at time t.
An increase in the Herndahl index reects a decline in compe-
tition, which negatively affects TFP growth. An increase in technol-
ogy gap is expected to decrease TFP growth, whereas the effect of
nancial development on TFP growth is likely to be positive. TFP
growth is expected to be positively related to the stock of human
capital. An increase in sales is expected to increase TFP. Finally,
the impact of FDI-related spillovers on TFP growth can be either
positive or negative.
The empirical analysis is conductedby means of rmlevel panel data
for 23 manufacturing industries over the period 20002005. Most of the
data are collected from GSO. The number of rms per year varies from
Table 3
Gross regional product per capita in US dollars.
Source: GSO (2013); and authors estimates.
2000 2001 2002 2003 2004 2005
Red River Delta 342 361 399 453 531 615
North East 201 214 234 264 309 352
North West 155 162 177 191 226 262
North Central Coast 200 208 228 252 293 336
South Central Coast 275 289 321 355 418 501
Central Highlands 197 193 211 235 276 332
South East 866 902 963 1156 1388 1632
Mekong River Delta 309 316 355 393 452 517
National Average 363 379 414 477 563 657
Table 4
Number of manufacturing rms in regions of Vietnam (20002005).
Source: GSO (2013).
2000 2001 2002 2003 2004 2005
Red River Delta 2425 2894 3855 4071 4584 5280
North East 534 636 848 980 1469 1890
North West 66 73 94 137 170 210
North Central Coast 444 541 710 775 806 938
South Central Coast 616 756 907 1032 1057 1757
Central Highlands 232 259 309 361 408 496
South East 3672 4456 5792 6487 7965 9457
Mekong River Delta 3672 4456 5792 6487 6965 7450
Total 11,661 14,071 18,307 20,330 23,424 27,478
1378 S. Anwar, L.P. Nguyen / Journal of Business Research 67 (2014) 13761387
approximately 11,661 rms in 2000 to 27,478 rms in 2005 (see
Table 4). This paper utilizes this dataset as it provides all information
needed for the study. Due to condentiality considerations, rm level
data is released with a lag. All nominal variables are measured in 1994
prices.
4. Empirical results and analysis
In stage one, using rm level panel data, we estimate TFP for each of
the eight geographical regions of Vietnam. These regions are Red River
Delta, North East, North West, North Central Coast, South Central Coast,
Central Highlands, South East, and Mekong River Delta. Calculation of
TFP involves estimation of a log-linear version of Eq. (1) by means of
OLS with a correction for heteroskedasticity.
The estimated coefcients are reported in Table 5. The results in
Table 5 suggest that both capital and labor make a statistically signi-
cant contribution to regional output. Using the estimated coefcients
reported in Table 5 in Eq. (2), the logarithm of TFP of each region is
calculated, which is the dependent variable in the rest of this paper.
Instage two, making use of the estimatedvalues of the dependent var-
iable for eachregion, Eq. (3) is estimated. This involves the use of the two-
stage least squares (2SLS) with a correction for heteroskedasticity. In
order to avoid the endogeneity that results from the presence of FDI
and industrial characteristics, dummy variables were added to the right
hand side of Eq. (3). The inclusion of dummy variables tends to reduce
the severity of the endogeneity problem. This problem arises from the
fact that industries that are known to be highly productive, in relative
terms, are likely to attract more foreign and domestic rms. The industry
capital and human capital stocks of foreign and domestic rms may be
correlated with the error term. Accordingly, this paper includes industry
dummies as regressors. This helps to remove the error term's industry-
specic unobservable effect that may be correlated with the industry
stock variables. Inclusion of the dummy variables also allows one to re-
duce the effect of omitted time-varying industry-specic shocks that
may be correlated with the FDI variables and the error termin the regres-
sion. In addition, lagged values of relevant variables were used as instru-
ments to account for the potential endogeneity problem. As indicated by
Nowak-Lehmann, Dreher, Herzer, Klasen, and Martinez-Zarzoso (2012),
in general, lagged variables are not good instruments when persistent au-
tocorrelation is present. In the case of this paper, panel data is used where
the sample period covers only 6 years and hence autocorrelation is not a
serious problem.
The DurbinWuHausman test is used to test for endogeneity. The
null hypothesis is rejected, suggesting that OLS estimates might be biased
and inconsistent and hence OLS is not an appropriate estimation tech-
nique. As a result, Hansen's J-test is used to test for over-identication of
2SLS (i.e., the null hypothesis of correct model specication and valid
over-identifying restrictions is tested and the results are found to be
satisfactory).
Table 6 shows that both the horizontal and the backward linkage
effects on the TFP level of domestic rms vary across regions. The back-
ward linkage effects are positive and statistically signicant only in
Red River Delta, South Central Coast, South East and Mekong River
Delta whereas, in other regions, these effects are either negative
and signicant or positive but insignicant. This reects the fact
that the regions that attract higher volumes of FDI and where most
of Vietnam's exportables are produced will benet more from back-
ward linkages. As a result, the backward linkages impact positively
on TFP level of local rms. Local rms located in relatively developed
regions of Vietnam, where FDI is more productive, experience a rel-
atively rapid technology spillover from foreign to domestic rms as
compared to local rms that are located in remote regions. This
may also result in greater backward linkages for rms that are located
in relatively developed regions. The impact of FDI spillovers on TFP
varies across regions.
Table 6 shows that the impact of the horizontal spillover effect is pos-
itive and signicant only in North East, Central Highland, and Mekong
River Delta regions. The backward spillover effect is positive and statisti-
cally signicant only in Red River Delta, South Central Coast, South East,
and Mekong River Delta, whereas the forward spillover effect is positive
and signicant only in North West and North Central Coast regions. It is
interesting to note that backward linkages are signicant in those areas
that tend to receive a very big proportion of the total FDI. These are also
the areas where most of the exportable goods are produced. The introduc-
tionof domestic content requirements has contributed to stronger back-
ward linkages. For example, rms that export less than 30% of their
production were offered concessionary import duties only if these
rms satised the 30% local content requirement. This policy helped de-
velop stronger backward linkages in automotive, electronics and engi-
neering (Truong & Gates, 1996; Athukorala & Tien, 2012).
The empirical results presented in Table 7 suggest that local rms
located inrelatively more developed regions, where FDI is concentrated,
are in general more efcient and thus possess greater ability to benet
fromadvanced technologies that are introduced inthe regionby foreign
rms. These regions are in the secondstage of Dunning's investment de-
velopment path (IDP).
The presence of FDI-generated technology spillover effects also
depends on absorptive capability (as measured by the technology gap,
human capital, and nancial development) of domestic rms in host
economies (See Wang, 2010 andreferences therein). Inorder to examine
this hypothesis, this paper considers the impact of the interaction be-
tween (i) the technology gap and FDI spillovers, (ii) the stock of
human capital and FDI spillovers, and (iii) the level of nancial de-
velopment and FDI spillovers on TFP of manufacturing rms in
each of the eight regions of Vietnam. The main results of empirical
estimation are summarized in Table 7, whereas detailed results
appear in Appendices 18.
Table 7 shows that the interaction of human capital and horizontal
spillovers has a strong positive effect on TFP of Red River Delta. This is
also true for North East and the Central Highlands. The impact of the in-
teraction between human capital and backward spillovers is positive
and signicant in all regions except South Central Coast and South
East. The interaction of human capital and forward spillovers is positive
and signicant only in South Central Coast. The impact of technology
gap and horizontal spillovers is negative and signicant in Red River
Delta, North Central Coast, South East, and Mekong River Delta regions.
Table 5
Estimated production functions for Vietnamese regions.
Independent variables Red River Delta North East North West North Central Coast South Central Coast Central Highlands South East Mekong River Delta
log (K) 0.6641 0.5403 0.5174 0.6051 0.6882 0.5038 0.7008 0.8546
(67.05)* (32.90)* (13.30)* (37.84)* (51.18)* (19.51)* (58.97)* (77.45)*
log (L) 0.4843 0.7434 0.6403 0.5081 0.4576 0.6207 0.4165 0.3056
(36.46)* (34.53)* (10.96)* (21.64)* (28.12)* (18.57)* (50.01)* (22.13)*
Constant 0.63127 0.2301 0.7193 0.8432 0.5869 1.202 0.7275 0.3234
(14.75)* (3.60)* (4.13)* (12.71)* (8.87)* (9.24)* (21.80)* (6.59)*
Adjusted R
2
number
of observations
0.68 0.77 0.72 0.75 0.77 0.71 0.73 0.64
15,843 4436 465 2987 3885 1354 24,286 11,922
Notes: (i) Robust t-statistics in parentheses; (ii) ***signicant at 10%, **signicant at 5%, and *signicant at 1%.
1379 S. Anwar, L.P. Nguyen / Journal of Business Research 67 (2014) 13761387
The interaction of technology gap and backward spillovers is negative
and signicant in Red River Delta, North East, North Central Coast, and
South East regions. The interaction of technology gap and forward spill-
overs is negative and signicant in all regions except Red River Delta,
North East, and North Central Coast regions. The interaction of the
level of nancial development and horizontal spillovers is positive and
signicant only in Red River Delta, North Central Coast, and South
East regions. The interaction of the level of nancial development
and backward spillovers is positive and signicant only in Red
River Delta and North East regions. Finally, the interaction of the
level of nancial development and forward spillovers is positive and
signicant only in South East region. It is clear that, as far as the level
of nancial development is concerned, Vietnam has a long way to go.
In others words, steps need to be taken to improve the level of nancial
development in Vietnam.
5. Conclusion
This paper focuses on the impact of FDI and FDI-generated spillovers
ontotal factor productivity (TFP) growthof manufacturing rms located
in all eight regions of Vietnam. An increase in TFP can be attributed to
technological advancement. The empirical analysis presented in this
paper is based onpanel data that covers a large number of manufacturing
rms over the period 20002005. Unlike the existing studies, this paper
considers the impact of FDI spillovers on TFP growth via both horizontal
and vertical linkages in different regions of Vietnam. The paper also con-
siders the role of absorptive capacity.
The results presented in this paper suggest that impact of FDI and
FDI spillovers on TFP of Vietnam's manufacturing rms varies across
regions. In other words, the presence of foreign rms is contributing
to technological advancement in Vietnam but the rate of such ad-
vancement varies considerably across regions of Vietnam. Backward
linkages with foreign rms are an important channel of technology
transfer from foreign to domestic rms in Red River Delta, South
Central Coast, South East, and Mekong River Delta. These regions are
well known for (i) better quality infrastructure, (ii) higher level of
human capital stock, and (iii) relatively more advanced technology.
The empirical analysis presented in this paper also suggests that the
effect of horizontal and vertical linkages with foreign rms on TFP
of local rms also depends on Vietnamese regions' absorptive capac-
ity. Regions with better technology, stock of human capital and a rel-
atively higher level of nancial development gain more benets from
FDI spillovers. It appears that North East, North West, North Central
Coast, and Central Highlands regions have not signicantly beneted
Table 6
The impact of FDI on TFP of domestic rms in Vietnam's regions.
Independent variables Dependent variable: log (TFP)
Red River Delta North East North West North Central Coast South Central Coast Central Highlands South East Mekong River Delta
Human capital [log(H)] 1.1017 1.2744 0.4105 0.4797 1.3199 0.7480 0.6994 1.8812
(7.20)* (5.17)* (8.04)* (21.34)* (5.01)* (2.35)* (5.71)* (16.48)*
Scale (S) 6.7832 16.2509 3.1745 0.1383 19.2282 5.0855 1.2087 9.9852
(2.66)* (3.69)* (0.53) (0.13) (3.32)* (0.31) (1.85)*** (6.17)*
Concentration (C) 0.2600 0.3224 0.6395 0.3367 0.4171 0.1886 0.1723 0.3275
(1.92)* (2.42)* (1.67)*** (2.67)* (1.91)** (0.39) (1.30) (1.70)***
Technology gap (T) 0.0003 0.0002 0.0046 0.0021 0.0008 0.0060 0.0003 0.0004
(26.51)* (0.92) (7.12)* (11.53)* (4.54)* (3.94)* (11.54)* (1.15)
Horizontal FDI spillovers 0.0029 0.0044 0.0008 0.0018 0.0019 0.0037 0.0051 0.0115
(5.43)* (3.59)* (0.33) (1.91)** (1.78)*** (1.84)*** (10.79)* (8.14)*
Backward FDI spillovers 0.0003 0.0001 0.0028 0.0060 0.0021 0.0010 0.0022 0.0019
(2.18)** (0.14) (1.36) (7.70)* (2.12)** (0.54) (5.54)* (1.83)***
Forward FDI spillovers 0.0041 0.0086 0.0165 0.0043 0.0087 0.0051 0.0078 0.0411
(2.55)* (2.87) (2.93)* (2.04)** (3.57)** (1.10) (9.01)* (15.45)*
Financial development (F) 0.7444 1.0106 0.8286 0.9349 0.8172 1.0561 1.0828 0.7001
(8.95)* (9.95)* (5.00)* (11.99)* (10.65)* (9.38)* (39.24)* (10.95)*
Time dummies Yes Yes Yes Yes Yes Yes Yes Yes
Industry dummies Yes Yes Yes Yes Yes Yes Yes Yes
Hansen test (p-value) 0.21 0.22 0.45 0.55 0.18 0.22 0.28 0.68
DurbinWuHausman (p-value) 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Number of observations 15,843 4436 465 2987 3885 1354 24,286 11,922
R-squared 0.30 0.27 0.38 0.43 0.25 0.34 0.43 0.27
Notes: (i) Robust t-statistics in parentheses; (ii) ***signicant at 10%, **signicant at 5%, and *signicant at 1%.
Table 7
Absorptive capacity and TFP growth in Vietnamese regions.
Red River Delta North
East
North
West
North Central
Coast
South Central
Coast
Central
Highlands
South
East
Mekong River
Delta
Interaction of H and horizontal spillovers P & S P & S N & IS N & IS N & S P & S P & IS N & IS
Interaction of H and backward spillovers P & S P & S P & IS P & S N & S P & S N & S P & S
Interaction of H and forward spillovers N & S N & S P & IS N & IS P & S N & S P & IS P & IS
Interaction of T and horizontal spillovers N & S N & IS P & IS P & S N & S N & IS N & S N & S
Interaction of T and backward spillovers N & S N & S P & S N & S P & IS P & IS N & S N & IS
Interaction of T and forward spillovers N & IS P & IS N & S P & IS N & S N & S N & S N & S
Interaction of F and horizontal spillovers P & S N & IS N & IS P & S N & IS N & IS P & S N & IS
Interaction of F and backward spillovers P & S P & S P & IS P & IS N & S P & IS N & S N & S
Interaction of F and forward spillovers P & IS N & S N & IS N & IS P & IS N & IS P & S N & IS
Notes: Hstock of human capital; T technology gap; F the level of nancial development; P positive effect on TFP; Nnegative effect on TFP; S signicant and IS insignicant.
1380 S. Anwar, L.P. Nguyen / Journal of Business Research 67 (2014) 13761387
Independent variables Dependent variable: log (TFP)
Estimation without
interaction
Estimation with interaction
between FDI spillover and H
Estimation with interaction
between FDI spillover and T
Estimation with interaction
between FDI spillover and F
Human capital [log(H)] 1.1017 1.1975 2.3708 1.0868
(7.20)* (5.58)* (2.36)** (5.39)*
Scale (S) 6.7832 8.3485 2.5549 6.5548
(2.66)* (2.36)** (2.31)** (1.90)***
Concentration (C) 0.2600 0.2069 0.9492 0.2536
(1.92)* (1.15) (1.15) (1.48)
Technology gap (T) 0.0003 0.0001 0.0003 0.0002
(26.51)* (1.79)*** (5.28)* (5.60)*
Horizontal FDI spillovers 0.0029 0.3163 2.9881 0.5786
(5.43)* (5.24)* (1.75)*** (4.38)*
Backward FDI spillovers 0.0003 0.0002 0.0099 0.0023
(2.18)** (0.40) (3.36)* (1.99)**
Forward FDI spillovers 0.0041 0.0037 0.0543 0.0001
(2.55)* (1.92)** (1.52) (1.20)
Financial development (F) 0.7444 0.6946 0.7977 0.6102
(8.95)* (6.44)* (6.17)* (4.41)*
Technology gap horizontal 0.00001
(1.92)**
Technology gap backward 0.00001
(2.62)*
Technology gap forward 0.00005
(3.21)*
Human capital horizontal 0.0169
(1.94)**
Human capital backward 0.0055
(3.60)*
Human capital forward 0.0316
(1.53)
Financial development horizontal 0.0052
(2.20)**
Financial development backward 0.0049
(2.48)*
Financial development forward 0.0075
(1.56)
Time dummies Yes Yes Yes Yes
Industry dummies Yes Yes Yes Yes
Hansen test (p-value) 0.21 0.12 0.14 0.57
DurbinWuHausman (p-value) 0.00 0.00 0.00 0.00
Number of observations 15,843 15,843 15,843 15,843
R-squared 0.30 0.27 0.20 0.31
Notes: (i) Robust t-statistics in parentheses; (ii) ***signicant at 10%, **signicant at 5%, and *signicant at 1%.
from backward linkages but these regions have beneted from hori-
zontal or forward linkages. It seems that government policy that en-
courages foreign rms to invest in outer regions of Vietnamhas so far
not worked.
Based on the empirical analysis presented in this paper, Vietnam
can gain more benets from FDI by improving its absorptive capacity.
At the central level, Vietnam needs to further develop its nancial
system. Increased transparency is the rst step in this direction
followed by further liberalization. Increase in real spending on ad-
vanced education and training may help to reduce the technology
gap between foreign and local rms and increase the stock of
human capital. This combined with infrastructure improvement can
help to boost manufacturing sector productivity.
5.1. Limitations and areas for further research
The quality of Vietnamese data is questioned in some prior stud-
ies. The General Statistics Ofce of Vietnam has taken steps to im-
prove reliability of the data collection process. More recent data,
when it becomes available, is likely to be relatively more reliable.
Empirical analysis of more recent data when available would allow
one to assess the impact FDI-related spillovers in the post global
nancial crisis period. In addition to the impact of FDI on productiv-
ity in Vietnam, one can also examine the impact of FDI on product
quality. Recent studies such as Chang, Kao, Kuo, and Chiu (2012)
have highlighted the importance of governance quality in the con-
text of rm entry choices. It would be useful to examine the interac-
tion of governance quality and FDI on industrial growth in Vietnam.
Finally, one can also examine the impact of FDI on entrepreneurship
in Vietnam; specically, the impact of FDI on entry and exit of domestic
rms.
Acknowledgments
This paper has greatly beneted from extremely useful com-
ments and suggestions received from two anonymous reviewers.
The authors are also grateful to Dr Robert Alexander and participants
of seminar at Grifth University (QLD, Australia) and Flinders Uni-
versity (SA, Australia). However, the authors are solely responsible
for all remaining errors and imperfections.
Appendix 1. The effect of FDI on TFP in Red River Delta
1381 S. Anwar, L.P. Nguyen / Journal of Business Research 67 (2014) 13761387
Appendix 2. The effect of FDI on TFP in North East
Appendix 3. The effect of FDI on TFP in North West
Independent variables Dependent variable: log (TFP)
Estimation without
interaction
Estimation with interaction
between FDI spillover and H
Estimation with interaction
between FDI spillover and T
Estimation with interaction
between FDI spillover and F
Human capital [log(H)] 1.2744 0.4035 0.3194 0.6056
(5.17)* (1.10) (0.76) (1.83)***
Scale (S) 16.2509 2.9655 2.8065 5.6267
(3.69)* (0.49) (0.91) (0.96)
Concentration (C) 0.3224 0.1418 0.2266 0.1542
(2.42)* (0.73) (0.85) (0.87)
Technology gap (T) 0.0002 0.0034 0.0009 0.0006
(0.92) (2.71)* (3.52)* (2.48)**
Horizontal FDI spillovers 0.0044 0.3061 0.2457 0.5952
(3.59)* (1.98)** (0.41) (2.06)**
Backward FDI spillovers 0.0001 0.0025 0.0088 0.0071
(0.14) (2.08)** (3.99)* (3.62)*
Forward FDI spillovers 0.0086 0.0039 0.0200 0.0072
(2.87) (0.69) (0.93) (1.30)
Financial development (F) 1.0106 1.2495 1.2794 1.3992
(9.95)* (10.25)* (17.76)* (11.59)*
Technology gap horizontal 0.00002
(0.36)
Technology gap backward 0.00018
(2.62)*
Technology gap forward 0.00045
(2.20)**
Human capital horizontal 0.0027
(0.94)
Human capital backward 0.0052
(4.51)*
Human capital forward 0.0101
(0.76)
Financial development horizontal 0.0064
(1.36)
Financial development backward 0.0141
(3.79)*
Financial development forward 0.0209
(2.56)*
Time dummies Yes Yes Yes Yes
Industry dummies Yes Yes Yes Yes
Hansen test (p-value) 0.22 0.52 0.19 0.18
DurbinWuHausman (p-value) 0.00 0.00 0.00 0.00
Number of observations 4436 4436 4436 4436
R-squared 0.27 0.31 0.47 0.46
Notes: (i) Robust t-statistics in parentheses; (ii) ***signicant at 10%, **signicant at 5%, and *signicant at 1%.
Independent variables Dependent variable: log (TFP)
Estimation without
interaction
Estimation with interaction
between FDI spillover and H
Estimation with interaction
between FDI spillover and T
Estimation with interaction
between FDI spillover and F
Human capital [log(H)] 0.4105 0.7991 1.3296 0.6477
(8.04)* (3.19)* (2.24)** (2.81)*
Scale (S) 3.1745 1.4416 1.0360 0.4318
(0.53) (0.51) (0.40) (0.18)
Concentration (C) 0.6395 0.6292 0.9543 0.6638
(1.67)*** (1.64)*** (2.18)** (1.82)***
Technology gap (T) 0.0046 0.0041 0.0039 0.0033
(7.12)* (3.20)* (4.49)* (2.34)**
Horizontal FDI spillovers 0.0008 0.3213 0.2272 0.4882
(0.33) (1.00) (0.26) (0.82)
Backward FDI spillovers 0.0028 0.0048 0.0208 0.0062
(1.36) (2.16)** (2.25)* (1.19)
Forward FDI spillovers 0.0165 0.0160 0.1071 0.0222
(2.93)* (2.67)* (2.15)** (1.65)***
Financial development (F) 0.8286 0.7067 0.7631 0.9694
(5.00)* (3.34)* (4.17)* (2.67)*
Technology gap horizontal 0.00004
(0.59)
Technology gap backward 0.00004
(1.01)
Technology gap forward 0.00011
(0.85)
1382 S. Anwar, L.P. Nguyen / Journal of Business Research 67 (2014) 13761387
Appendix 4. The effect of FDI on TFP in North Central Coast
(continued)
Independent variables Dependent variable: log (TFP)
Estimation without
interaction
Estimation with interaction
between FDI spillover and H
Estimation with interaction
between FDI spillover and T
Estimation with interaction
between FDI spillover and F
Human capital horizontal 0.0032
(0.71)
Human capital backward 0.0108
(2.00)**
Human capital forward 0.0588
(1.82)***
Financial development horizontal 0.0086
(0.75)
Financial development backward 0.0065
(0.63)
Financial development forward 0.0087
(0.33)
Time dummies Yes Yes Yes Yes
Industry dummies Yes Yes Yes Yes
Hansen test (p-value) 0.45 0.39 0.38 0.37
DurbinWuHausman (p-value) 0.00 0.00 0.00 0.00
Number of observations 465 465 465 465
R-squared 0.38 0.32 0.28 0.36
Notes: (i) Robust t-statistics in parentheses; (ii) ***signicant at 10%, **signicant at 5%, and *signicant at 1%.
Appendix 3 (continued)
Independent variables Dependent variable: log (TFP)
Estimation without
interaction
Estimation with interaction
between FDI spillover and H
Estimation with interaction
between FDI spillover and T
Estimation with interaction
between FDI spillover and F
Human capital [log(H)] 0.4797 0.6584 0.5863 0.9498
(21.34)* (1.87)*** (0.93) (2.23)**
Scale (S) 0.1383 1.3827 3.3211 3.0785
(0.13) (0.62) (1.47) (1.09)
Concentration (C) 0.3367 0.0173 0.2098 0.1061
(2.67)* (0.12) (1.32) (0.46)
Technology gap (T) 0.0021 0.0028 0.0037 0.0007
(11.53)* (2.27)** (3.22)* (0.71)
Horizontal FDI spillovers 0.0018 0.0128 2.2669 0.4616
(1.91)** (0.08) (1.85)*** (1.71)***
Backward FDI spillovers 0.0060 0.0075 0.0047 0.0064
(7.70)* (8.18)* (1.15) (3.12)*
Forward FDI spillovers 0.0043 0.0074 0.0501 0.0085
(2.04)** (1.49) (1.81)*** (1.64)***
Financial development (F) 0.9349 0.8135 0.9892 0.7537
(11.99)* (6.81)* (12.46)* (4.27)*
Technology gap horizontal 0.00002
(0.41)
Technology gap backward 0.00012
(2.25)**
Technology gap forward 0.00028
(1.40)
Human capital horizontal 0.0110
(1.69)***
Human capital backward 0.0062
(2.80)*
Human capital forward 0.0346
(1.97)
Financial development horizontal 0.0087
(2.09)**
Financial development backward 0.0005
(0.16)
Financial development forward 0.0131
(1.42)
Time dummies Yes Yes Yes Yes
Industry dummies Yes Yes Yes Yes
Hansen test (p-value) 0.55 0.19 0.16 0.12
DurbinWuHausman (p-value) 0.00 0.00 0.00 0.00
Number of observations 2987 2987 2987 2987
R-squared 0.43 0.43 0.34 0.34
Notes: (i) Robust t-statistics in parentheses; (ii) ***signicant at 10%, **signicant at 5%, and *signicant at 1%.
1383 S. Anwar, L.P. Nguyen / Journal of Business Research 67 (2014) 13761387
Appendix 5. The effect of FDI on TFP in South Central Coast
Appendix 6. The effect of FDI on TFP in Central Highlands
Independent variables Dependent variable: log (TFP)
Estimation without
interaction
Estimation with interaction
between FDI spillover and H
Estimation with interaction
between FDI spillover and T
Estimation with interaction
between FDI spillover and F
Human capital [log(H)] 1.3199 2.27651 9.4182 1.4996
(5.01)* (3.26)* (2.11)** (4.52)*
Scale (S) 19.2282 46.9831 76.0676 23.8597
(3.32)* (2.46)** (2.03)** (2.98)*
Concentration (C) 0.4171 1.1422 6.0229 0.6587
(1.91)** (2.36)** (1.93)*** (2.12)**
Technology gap (T) 0.0008 0.0015 0.0001 0.0006
(4.54)* (0.89) (0.23) (1.98)**
Horizontal FDI spillovers 0.0019 0.5698 18.7057 0.3917
(1.78)*** (1.38) (1.91)*** (1.10)
Backward FDI spillovers 0.0021 0.0001 0.0205 0.0019
(2.12)** (0.04) (1.30) (0.79)
Forward FDI spillovers 0.0087 0.0300 0.4149 0.0017
(3.57)** (2.92)* (1.97)** (0.36)
Financial development (F) 0.8172 0.5706 1.6669 1.4744
(10.65)* (3.18)* (4.19)* (7.37)*
Technology gap horizontal 0.0005
(2.13)**
Technology gap backward 0.00001
(0.24)
Technology gap forward 0.0013
(2.36)**
Human capital horizontal 0.0915
(1.94)***
Human capital backward 0.0121
(1.49)
Human capital forward 0.2405
(2.00)**
Financial development horizontal 0.0105
(1.80)***
Financial development backward 0.0002
(0.06)
Financial development forward 0.0003
(2.22)**
Time dummies Yes Yes Yes Yes
Industry dummies Yes Yes Yes Yes
Hansen test (p-value) 0.18 0.29 0.56 0.21
DurbinWuHausman (p-value) 0.00 0.00 0.00 0.00
Number of observations 3885 3885 3885 3885
R-squared 0.25 0.12 0.16 0.14
Notes: (i) Robust t-statistics in parentheses; (ii) ***signicant at 10%, **signicant at 5%, and *signicant at 1%.
Independent variables Dependent variable: log (TFP)
Estimation without
interaction
Estimation with interaction
between FDI spillover and H
Estimation with interaction
between FDI spillover and T
Estimation with interaction
between FDI spillover and F
Human capital [log(H)] 0.7480 0.5904 2.3187 0.8272
(2.35)* (1.39) (2.99)* (2.40)**
Scale (S) 5.0855 0.4973 5.7480 10.6405
(0.31) (0.03) (0.52) (0.66)
Concentration (C) 0.1886 0.1021 0.0004 0.2474
(0.39) (0.15) (0.11) (0.42)
Technology gap (T) 0.0060 0.0111 0.0083 0.0060
(3.94)* (2.92)* (4.49)* (3.10)*
Horizontal FDI spillovers 0.0037 0.1436 5.1689 0.4214
(1.84)*** (0.49) (2.99)* (0.96)
Backward FDI spillovers 0.0010 0.0035 0.0151 0.0025
(0.54) (1.74)*** (1.42) (0.65)
Forward FDI spillovers 0.0051 0.0004 0.1011 0.0081
(1.10) (0.05) (1.91)*** (0.92)
Financial development (F) 1.0561 1.1004 1.2021 1.5293
(9.38)* (8.97)* (9.47)* (6.25)*
Technology gap horizontal 0.0002
(1.70)***
Technology gap backward 0.0002
(1.88)***
Technology gap forward 0.0004
(1.60)***
1384 S. Anwar, L.P. Nguyen / Journal of Business Research 67 (2014) 13761387
Appendix 7. The effect of FDI on TFP in South East
(continued)
Independent variables Dependent variable: log (TFP)
Estimation without
interaction
Estimation with interaction
between FDI spillover and H
Estimation with interaction
between FDI spillover and T
Estimation with interaction
between FDI spillover and F
Human capital horizontal 0.0238
(2.85)
Human capital backward 0.0072
(1.34)
Human capital forward 0.0597
(2.01)**
Financial development horizontal 0.0023
(0.28)
Financial development backward 0.0008
(0.13)
Financial development forward 0.0260
(1.52)
Time dummies Yes Yes Yes Yes
Industry dummies Yes Yes Yes Yes
Hansen test (p-value) 0.22 0.14 0.29 0.31
DurbinWuHausman (p-value) 0.00 0.00 0.00 0.00
Number of observations 1354 1354 1354 1354
R-squared 0.34 0.39 0.31 0.32
Notes: (i) Robust t-statistics in parentheses; (ii) ***signicant at 10%, **signicant at 5%, and *signicant at 1%.
Appendix 6 (continued)
Independent variables Dependent variable: log (TFP)
Estimation without
interaction
Estimation with interaction
between FDI spillover and H
Estimation with interaction
between FDI spillover and T
Estimation with interaction
between FDI spillover and F
Human capital [log(H)] 0.6994 0.7547 2.1076 1.6723
(5.71)* (1.78)*** (6.99)* (11.32)*
Scale (S) 1.2087 0.9034 1.6491 3.2606
(1.85)*** (0.44) (3.85)* (3.63)
Concentration (C) 0.1723 0.1004 0.5606 0.0765
(1.30) (0.52) (2.44)* (0.43)
Technology gap (T) 0.0003 0.0004 0.0003 0.0001
(11.54)* (1.45) (2.06)** (0.77)
Horizontal FDI spillovers 0.0051 0.5525 4.9724 1.4650
(10.79)* (8.15)* (4.63)* (11.20)*
Backward FDI spillovers 0.0022 0.0027 0.0161 0.0114
(5.54)* (3.46)* (7.48)* (7.28)*
Forward FDI spillovers 0.0078 0.0079 0.0230 0.0163
(9.01)* (5.86)* (2.78)* (6.40)*
Financial development (F) 1.0828 1.0737 1.1141 0.5483
(39.24)* (18.68)* (42.27)* (7.24)*
Technology gap horizontal 0.00004
(1.41)
Technology gap backward 0.00003
(2.07)**
Technology Gap Forward 0.00001
(0.73)
Human capital horizontal 0.0216
(5.07)*
Human capital backward 0.0057
(6.84)*
Human capital forward 0.0154
(3.93)*
Financial development horizontal 0.0152
(7.47)*
Financial development backward 0.0137
(5.81)*
Financial development forward 0.0084
(2.18)**
Time dummies Yes Yes Yes Yes
Industry dummies Yes Yes Yes Yes
Hansen test (p-value) 0.28 0.51 0.92 0.16
DurbinWuHausman (p-value) 0.00 0.00 0.00 0.00
Number of observations 24,286 24,286 24,286 24,286
R-squared 0.43 0.44 0.31 0.26
Notes: (i) Robust t-statistics in parentheses; (ii) ***signicant at 10%, **signicant at 5%, and *signicant at 1%.
1385 S. Anwar, L.P. Nguyen / Journal of Business Research 67 (2014) 13761387
Appendix 8. The effect of FDI on TFP in Mekong River Delta
References
Abu-Bader, S., & Abu-Qarn, A. S. (2008). Financial development and economic growth:
The Egyptian experience. Journal of Policy Modeling, 30, 887898.
Alvarez, R., & Lpez, R. A. (2008). Is exporting a source of productivity spillovers? Review
of World Economics, 144, 723749.
Anwar, S., & Nguyen, L. P. (2010a). Foreign direct investment and economic growth in
Vietnam. Asia Pacic Business Review, 16, 183202.
Anwar, S., & Nguyen, L. P. (2010b). Absorptive capacity, foreign direct investment-linked
spillovers and economic growth in Vietnam. Asian Business & Management, 9,
553570.
Anwar, S., & Nguyen, L. P. (2011a). Foreign direct investment and trade: The case of
Vietnam. Research in International Business and Finance, 25(2011), 3952.
Anwar, S., & Nguyen, L. P. (2011b). Foreign direct investment and export spillovers:
Evidence from Vietnam. International Business Review, 20, 177193.
Athukorala, P., & Tien, T. Q. (2012). Foreign direct investment in industrial transition: The
experience of Vietnam. Journal of the Asia Pacic Economy, 17(3), 446463.
Barbosa, N., & Eiriz, V. (2009). Linking corporate productivity to foreign direct invest-
ment: An empirical assessment. International Business Review, 18, 113.
Beugelsdijk, S., Smeets, R., & Zwinkels, R. (2008). The impact of horizontal and vertical FDI
on host's country economic growth. International Business Review, 17, 452472.
Binh, N., & Haughton, J. (2002). Trade liberalization and foreign direct investment in
Vietnam. ASEAN Economic Bulletin, 19, 302318.
Bitzer, J., Geishecker, I., & Grg, H. (2008). Productivity spillovers through vertical link-
ages: Evidence from 17 OECD countries. Economics Letters, 99(2), 328331.
Blomstrm, M., & Sjoholm, F. (1999). Technology transfer and spillovers: Does local par-
ticipation with multinationals matter? European Economic Review, 43, 915923.
Bwalya, S. M. (2006). Foreign direct investment and technology spillovers: Evidence from
panel data analysis of manufacturing rms in Zambia. Journal of Development
Economics, 81, 514526.
Chang, Y., Kao, M., Kuo, A., & Chiu, C. (2012). Howcultural distance inuences entry mode
choice: The contingent role of host country's governance quality. Journal of Business
Research, 65, 11601170.
Du, L., Harrison, A., & Jefferson, G. H. (2012). Testing for horizontal and vertical foreign
investment spillovers in China, 19982007. Journal of Asian Economics, 23(3),
324343.
Fernandes, A.M., & Paunov, C. (2012). Foreign direct investment in services and
manufacturing productivity: Evidence for Chile. Journal of Development Economics,
97, 305321.
Giroud, A. (2007). MNE vertical linkages: The experience of Vietnam. International
Business Review, 16, 159176.
Grima, S., Grg, H., & Pisu, K. (2008). Exporting, linkages and productivity spillovers from
foreign direct investment. Canadian Journal of Economics, 41, 320340.
GSO (2013). General Statistics Ofce of Vietnam. http://www.gso.gov.vn
Jenkins, R. (2006). Globalization, FDI and employment in Viet Nam. Transnational
Corporations, 15, 115142.
Le, T. (2005). Technological spillovers from foreign direct investment: The case of Vietnam.
Graduate School of Economics, University of Tokyo.
Lin, F. (2010). The determinants of foreign direct investment in China: The case
of Taiwanese rms in the IT industry. Journal of Business Research, 63, 479485.
Lipsey, R., & Carlaw, K. (2004). Total factor productivity and the measurement of techno-
logical change. Canadian Journal of Economics, 37(4), 11181150.
Liu, Z. (2002). FDI and technology spillovers: Evidence from China. Journal of Comparative
Economics, 30, 579602.
Liu, Z. (2008). Foreign direct investment and technology spillovers: Theory and evidence.
Journal of Development Economics, 85, 176193.
Liu, X., & Wang, C. (2003). Does foreign direct investment facilitate technological progress?
Evidence from Chinese industries. Research Policy, 32, 945953.
Meyer, K. E., & Sinani, E. (2009). When and where does foreign direct investment generate
positive spillovers? Journal of International Business Studies, 40, 10751094.
Independent variables Dependent variable: log (TFP)
Estimation without
interaction
Estimation with interaction
between FDI spillover and H
Estimation with interaction
between FDI spillover and T
Estimation with interaction
between FDI spillover and F
Human capital [log(H)] 1.8812 1.9586 1.9339 1.7709
(16.48)* (15.72)* (3.55)* (14.92)*
Scale (S) 9.9852 13.4318 10.3396 11.0623
(6.17)* (2.68)* (1.95)** (2.60)*
Concentration (C) 0.3275 0.4462 2.3368 0.2645
(1.70)*** (1.83)*** (1.58) (1.21)
Technology gap (T) 0.0004 0.0011 0.0049 0.0010
(1.15) (1.86)*** (1.59) (1.72)***
Horizontal FDI spillovers 0.0115 1.1771 1.7551 1.3946
(8.14)* (7.49)* (3.41)* (4.51)*
Backward FDI spillovers 0.0019 0.0009 0.0147 0.0079
(1.83)*** (0.88) (0.95) (3.50)*
Forward FDI spillovers 0.0411 0.0415 0.1651 0.0342
(15.45)* (14.86)* (2.22)** (6.46)*
Financial development (F) 0.7001 0.6999 0.5823 1.2198
(10.95)* (10.79)* (2.53)** (8.03)*
Technology gap horizontal 0.00005
(0.12)
Technology gap backward 0.00008
(1.66)***
Technology gap forward 0.00007
(0.09)
Human capital horizontal 0.3114
(3.37)*
Human capital backward 0.0058
(0.74)
Human capital forward 0.1177
(2.81)*
Financial development horizontal 0.0037
(0.68)
Financial development backward 0.0127
(2.95)*
Financial development forward 0.0137
(1.37)
Time dummies Yes Yes Yes Yes
Industry dummies Yes Yes Yes Yes
Hansen test (p-value) 0.68 0.18 0.36 0.21
DurbinWuHausman (p-value) 0.00 0.00 0.00 0.00
Number of observations 11,922 11,922 11,922 11,922
R-squared 0.27 0.23 0.35 0.32
Notes: (i) Robust t-statistics in parentheses; (ii) ***signicant at 10%, **signicant at 5%, and *signicant at 1%.
1386 S. Anwar, L.P. Nguyen / Journal of Business Research 67 (2014) 13761387
Nguyen, A. N., & Nguyen, T. (2007). Foreign direct investment in Vietnam: An overview and
analysis of the determination of spatial distribution. Hanoi, Vietnam: Development and
Polices Research Center.
Nguyen, A. N., Nguyen, T. L., Dang, T. P., Quang, N., Nguyen, D. C., & Nguyen, D. N. (2008).
Foreign direct investment in Vietnam: Is there any evidence of technological spillover
effects? Development and Policies Research Center Working Paper, Hanoi, Vietnam.
Nguyen, D. T. H., & Sun, S. (2012). FDI and domestic rms' export behaviour: Evidence
from Vietnam. Economic Papers, 31(3), 380390.
Nowak-Lehmann, F., Dreher, A., Herzer, D., Klasen, S., & Martinez-Zarzoso, I. (2012). Does
foreign aid really raise per capita income? A time series perspective. Canadian Journal
of Economics, 45(1), 288313.
Pan, Y. (2003). The inow of foreign direct investment to China: the impact of country-
specic factors. Journal of Business Research, 56, 829833.
Stancik, J. (2007). Horizontal and vertical FDI spillovers: Recent evidence from the Czech
Republic. Center for Economic Research & Graduate Education - Economics Institute
(CERGE-EI) Working Paper No. 340.
Suyanto, S., Salim, R. A., & Bloch, H. (2009). Does foreign direct investment lead to produc-
tivityspillovers? Firmlevel evidence fromIndonesia. World Development, 37, 18611876.
Truong, D., & Gates, C. L. (1996). Vietnam in ASEAN Economic reform, openness and
transformation: An overview. ASEAN Economic Bulletin, 13(15968), 1996.
Wang, Y. (2010). FDI and productivity growth: The role of inter-industry linkages. Canadian
Journal of Economics, 43(4), 12431272.
1387 S. Anwar, L.P. Nguyen / Journal of Business Research 67 (2014) 13761387