Economic Analysis and Policy 56 (2017) 176–187

Contents lists available at ScienceDirect

Economic Analysis and Policy

journal homepage: www.elsevier.com/locate/eap

Full length article

Foreign direct investment and economic growth in Latin

America
Rafael Alvarado a, *, María Iñiguez b , Pablo Ponce a
a
School of Economics, Universidad Nacional de Loja, Loja, Ecuador
b
School of Economics, Universidad Técnica Particular de Loja, Ecuador

article info a b s t r a c t
Article history: This work examines the effect of foreign direct investment (FDI) on economic growth in
Received 25 November 2016 19 Latin American countries. Using panel data econometrics, we found robust empirical
Received in revised form 9 August 2017 evidence that suggests that the effect of FDI on economic growth is not statistically signifi-
Accepted 18 September 2017
cant in aggregated form. This result varies when we incorporate the levels of development
Available online 25 September 2017
reached by the countries in the region. FDI has a positive and significant effect on product in
high-income countries, while in upper-middle-income countries the effect is uneven and
JEL classification:
F23 non-significant. Finally, the effect in lower-middle-income countries is negative and sta-
O41 tistically significant. Our results show that FDI is not an adequate mechanism to accelerate
O54 economic growth in Latin America, with the exception of high-income countries.
© 2017 Economic Society of Australia, Queensland. Published by Elsevier B.V. All rights
Keywords: reserved.
Foreign direct investment
Economic growth
Panel data
Llatin America

1. Introduction

The region of Latin America is specialized in extracting natural resources aimed at the international market. FDI has
increased substantially in recent years in several countries (UNCTADSTAT, 2016) as a result of increased prices of the
commodities exported from the region. For example, in 2013 the rate of FDI was 12.3% higher than the year before. Despite
this increase, in the academic and political debate it is argued that Latin America needs higher levels of FDI to accelerate
economic growth. This occurs because internal savings do not meet the total demand for investment, particularly in countries
where FDI has stagnated. Theoretically, FDI generates benefits for investors (reduced costs, expanded markets) and receivers
(technology transfers, human capital transfers and generation of employment). Consequently, providing incentives to attract
and increase FDI represents an excellent strategy to reach higher levels of production in developing and emerging countries
(Yao, 2006).
There is ample theoretical and empirical literature that favors FDI as an instrument to increase economic growth. FDI
can act as a mechanism to accumulate physical capital and transfer human capital to the receiving country, which can
increase economic growth rate. Technology transfer increases the efficiency of production factors and this in turn reduces
the technological gap between national and international enterprises (Anwar and Nguyen, 2010). Likewise, FDI acts as
a technology transfer vehicle between developed and developing countries (Borensztein et al., 1998). Along the same
line, Aghion and Howitt (1998) indicated that economic growth is driven by innovation. They argued that technology is
incorporated into an economy mainly through FDI and international trade. However, when the receiving countries have

* Corresponding author.
E-mail addresses: jose.r.alvarado@unl.edu.ec (R. Alvarado), mpiniguez1@utpl.edu.ec (M. Iñiguez), pablo.ponce@unl.edu.ec (P. Ponce).

https://doi.org/10.1016/j.eap.2017.09.006
0313-5926/© 2017 Economic Society of Australia, Queensland. Published by Elsevier B.V. All rights reserved.
 R. Alvarado et al. / Economic Analysis and Policy 56 (2017) 176–187 177

low levels of human capital and consequently a low level of absorption of technology, FDI has weak role in determining
levels of production (Forte and Moura, 2013). Likewise, increased productivity of factors as a result of FDI would be more
significant if it strengthened the national productive fabric and were sectorally diversified. The concentration of FDI in a few
sectors can provoke increased inequality when the FDI exceeds the critical threshold (Suanes and Roca-Sagalés, 2015). Wu
and Hsu (2012) showed that FDI can be prejudicial for income distribution in receiving countries if they have low levels of
absorption capacity, while FDI has little effect on income inequality in countries with better absorption capacity. In addition,
there is recent empirical evidence suggesting that IDF increases income inequality in both the short and long run (Adams
and Klobodu, 2017) and inequality limits economic growth (Babu et al., 2016). Effectively, some of the empirical evidence
indicates that FDI is not beneficial for receiving countries. It can have a negative effect or no effect on growth (Saltz 1992;
Mencinger, 2003; Ang, 2009; Temiz and Gökmen, 2014). It is well known that FDI in Latin America is oriented to raw material
extraction (UNCTADSTAT, 2016), and if there is flexibilization in the labor market, FDI can encourage volatile growth and
precarious employment conditions for poorly qualified workers characteristic of developing countries in the region.
In this context, this research examines the effect of FDI on economic growth in 19 Latin American countries during the
period 1980–2014 using panel data econometrics. We estimated several regressions for Latin America (LA) in aggregated
form, for high-income countries (HIC), upper-middle-income countries (UMIC) and lower-middle-income countries (LMIC).
This grouping uses national per capita income levels and is based on the Atlas Method of the World Bank (2016). The
econometric strategy was divided into two stages. First, we estimated an expanded production function. The dependent
variable is the logarithm of real GDP and the independent variables are the logarithms of physical capital, the labor force and
inflows of FDI into the country i (i = 1, 2, . . . , I) in the period t (t = 1980, 1981, . . . , T ), respectively. Second, we analyzed
the robustness of the results obtained in the first stage. We included control variables that capture the productive structure
of Latin America and the effect of other factors suggested by the theoretical and empirical literature (Keesing, 1967; Krueger,
1985; Barro and Becker, 1989; Henderson, 2003; Kaldor, 1957). In practice, it cannot be expected that the effect of FDI on
economic growth is the same in an economy with a high absorption capacity and in one in the initial stages of development.
Effectively, the results show distinct effects among groups of countries in the region. FDI has a positive but non-significant
effect on growth in Latin America as a whole. However, when we classify countries according to per capita national income,
we found that the effect of FDI on economic growth is greater in high-income countries. In less developed countries, the
volatile effect is barely significant for UMIC countries and negative and significant for LMIC. Our results suggest a relatively
weak role of FDI in reaching higher levels of production in Latin America, except in Chile and Uruguay. Economic policy
decisions should consider that FDI is not necessarily beneficial for growth in Latin America. The differences among countries
in the absorption capacity associated with the level of human capital can explain the results found.
The rest of the present research is structured in four sections. The second section reviews existing literature. The third
section describes the data and the econometric methodology. The fourth section discusses the results found, which are
contrasted with theory and empirical evidence. Finally, the fifth section presents the conclusions and economic policy
implications arising from this research.

2. Literature review

Interest in the nexus between economic growth and FDI has increased in recent years because of the deindustrialization of
developed countries and the internationalization of production processes. In this process, the FDI integrates some countries
into the global market (Popescu, 2014). In the theoretical literature, endogenous growth models offer weak explanations
of the role of FDI in growth, which is associated with increased technological capital and infrastructure and the generation
of employment. In general, investment plays an important role in the accumulation of physical capital and the formation
of human capital. Anwar and Nguyen (2010) indicated that the impact of FDI on economic growth is greater when more
resources are invested in education and training, developing the financial market and reducing the technological gap
between local and foreign enterprises. Adeniyi et al. (2012) conclude that the degree of financial sophistication is important
in increasing FDI’s profits in the economic growth in development countries. Likewise, when FDI is complemented with local
investment it promotes the development of enterprises (Tan and Tang, 2016). FDI can stimulate technology transfer, which
tends to increase the productive efficiency of factors. It is logical to think that increases in technology translate into improved
productivity of the labor force and this in turn results in increased capital yield. If economic growth is driven by innovation
as argued by Aghion and Howitt (1998), the need for FDI to accelerate development is justified given the important roles
that technology and knowledge play in increasing production levels (Barro, 2001; Lucas, 1988).
The empirical literature on the effect of economic growth shows contradictory results. Firstly, there are several studies
that show a positive effect of FDI on the levels of production through externalities and spillover effect. For example, Barrell
and Pain (1997) suggest that FDI is a mechanism for disseminating ideas and technologies among counties. This conclusion
is similar to that obtained by Borensztein et al., (1998), who verified the effect of FDI on economic growth in developing
countries and indicated that FDI acts as a mechanism of technology transfer through increased productivity and acts only
when the receiving country meets a minimal threshold of stock of human capital. Bengoa and Sanchez-Robles (2003) came
to a similar conclusion for Latin America. This implies that FDI contributes to increasing production when there is sufficient
capacity to absorb technology in the receiving countries (Borensztein, et al., 1998; Gomes and Veiga, 2013) and when linkages
are generated with local firms and the export capacity of the receiving country is improved (Anwar and Nguyen, 2011; Ahmad
et al., 2003; Liu et al., 2002). This occurs because when the level of human capital in a FDI receiving country is low, the cost of
178 R. Alvarado et al. / Economic Analysis and Policy 56 (2017) 176–187

technology transfer is high. In this respect, Romero (2012) suggested that FDI stimulates internal investment and emphasized
that the contribution of FDI to growth is strengthened by the interaction with human capital and macroeconomic policies.
Secondly, there have been empirical investigations that show the negative influence of FDI on product (Musibah et al.,
2015; Saltz, 1992; Mencinger, 2003; Ang, 2009). These results suggest that the relationship between the two variables is
negative and that it changed in the period of study and with the productive structure of the countries. Other investigations
have shown that FDI does not have any effect on economic growth (Hermes and Lensink, 2003; Carkovic and Levine, 2002).
Carkovic and Levine (2002) argued that FDI does not have a robust and independent influence on growth, which implies that
FDI does not always accelerate the economic growth. This conclusion is corroborated by Curwin and Mahutga (2014), who
suggested that the penetration of FDI reduces economic growth in the short and long term in socialist countries.
Thirdly, there have been several investigations that estimate the causality between economic growth and FDI in China
and other Asian countries. These countries are among those that have benefitted the most from the entry of external capital
(Preeti and Agrawal, 2014) because FDI has strengthened their industrial capacity and diversified their exports. It is well
known that manufacturing generates more linkages than does the primary sector and that the income and employment
multipliers are high. In contrast, growth based on raw materials is extremely volatile. However, the empirical evidence is
not clear in this respect. For example, Sun (2011) showed that economic growth in China inevitably leads to increased FDI
and not the inverse. This can be explained by the expectations that are formed about an economy that grows at high rates and
that has a high absorption capacity, which can attract higher levels of investment to maximize profits. Liu et al. (2009) found
a two-way relationship between the two variables. The two-way causality between FDI and economic growth is an expected
result and it is logical that the two variables are reinforced over time. Anwar and Sun (2011) indicated that the inflows of
foreign capital increases the stock of domestic capital in Malaysia, which influences production levels. This is corroborated in
a recent work by Solarin and Shahbaz (2015). As well, the trade liberalization and financial development achieved by these
countries can reinforce the positive effects of the inflows of foreign capital (Iamsiraroj and Ulubaşoğlu, 2015). In addition, it
is necessary to remember that the relationship between the product and the FDI can occur in both directions. For example,
in Vietnam, the impact of a shock on GDP on FDI is more significant than the impact of FDI on GDP (Nguyen et al. 2017).
Although the results of research into the effect of FDI on economic growth are not conclusive, several investigations suggest
that receiving countries benefit more when foreign investment increases exports (Liu et al., 2002; Ahmad et al., 2003; Makki
and Somwaru, 2004; Yao, 2006). Other recent empirical investigations have come to similar conclusions in support of the
idea that FDI brings more benefits than costs to receiving countries (Duttaray et al., 2008).
Suanes and Roca-Sagalés (2015) stated that FDI increases inequality based on determined FDI levels. This is corroborated
by Basu and Guariglia (2007) who argued that FDI promotes growth but also inequality. Likewise, in a recent work Lessmann
(2013) showed that FDI increases inequality in low and middle-income countries. This result can be applied in Latin America,
which is the region with the greatest level of inequality in the world. Likewise, this region is specialized in the exploitation of
raw materials aimed at external markets, with FDI directed mainly at this exploitation (UNCTADSTAT, 2016). The primary-
export specialization of the region results in a scarcity of human capital and volatile growth. In practice, if investment is
oriented to commodities, high levels of technology transfer from investor to receiving countries cannot be expected. As
well, it is known that commodity prices are volatile over time, which translates into growth being dependent on the price of
raw materials. In this sense, the productive structure of Latin America suggests the effect of FDI on economic growth is low
or nil. The results of econometric regressions presented in the fourth section verify this hypothesis.

3. Data and methodology

3.1. Data

With the aim of empirically examining the effect of FDI on economic growth, we used statistics compiled by the World
Development Indicators of the World Bank (2016). The investigation covers 19 Latin American countries for the period 1980–
2014. The Atlas Method of the World Bank is a useful tool for classification using national per capita income for every country
for comparison among countries. The HIC group consists of Chile and Uruguay. The UMIC group includes Argentina, Brazil,
Colombia, Costa Rica, the Dominican Republic, Ecuador, Mexico, Panama, Paraguay, Peru and Venezuela. Finally, the LMIC
group consists of Belize, Bolivia, El Salvador, Guatemala, Honduras and Nicaragua. The remaining countries in the region
were excluded because of limitations in available statistical information.
In the econometric regressions, the dependent variable is the GDP logarithm and the independent variables are the
logarithms of physical capital, the labor force and inflows of FDI. Fig. 1 shows the temporal evolution of the logarithm of
the sum of FDI and of the product of the 19 countries. The two variables had dissimilar tendencies throughout the period
analyzed. When we obtained the average of the annual economic growth rate and the participation of FDI as a percentage of
product per country for the period 1980–2014, we found a positive relationship between the two variables. The association
between the two variables is low (R2 = 0.32) and there is a high degree of dispersion among the data, which suggests a low
adjustment between the two variables in aggregated form (see Fig. 2). This result is partly conditioned by the productive
structure of Latin America, which depends strongly on commodity prices and experience rapid growth of the service sector
without passing through a process of large-scale industrialization.
Although the Latin American region is strongly dependent on the prices of the raw materials that it exports (Alvarado
and Toledo, 2017), there are differences in the levels of per capita income among the countries included in this research. For
 R. Alvarado et al. / Economic Analysis and Policy 56 (2017) 176–187 179

Fig. 1. Evolution of FDI and GDP in Latin America.

Fig. 2. Correlation of FDI and GDP Latin America countries (period average).

example, the real per capita income of Chile in 2014 was USD 9853.5, while that of Bolivia for the same year was only USD
1408.9. Consequently, FDI can have distinct effects on Latin American countries owing to differences in the capacity to absorb
technology associated with the level of development. The descriptive statistics of the panel data are reported in Table 1. Real
GDP is more stable over time than among countries. The standard deviation (SD) between countries is 2.3 and within countries
is 0.36. Similarly, FBKF presents greater variability among countries than over time, the SD between is approximately 2 and
the SD within is 0.5, which indicates that the major part of variance come from variations between countries. The variable
proxy representing the labor force is the economically active population (labor) and measures the availability of the labor
factor. This variable also presents greater stability over time than among countries, the SD between is approximately 2.2
and the SD within is 0.2, which implies that the major part of the variance is explained by variations between countries. The
differences in the size of economies and the levels of development explain the statistical results of these three variables. FDI
shows more variation between countries than over time, with an SD among of 1.85 while the standard deviation within is
1.2. However, the differences in the source of variation are minimal. This result confirms the fact that FDI in the region is
low, not exceeding 5% of GDP.
180 R. Alvarado et al. / Economic Analysis and Policy 56 (2017) 176–187

Table 1
Summary statistics of variables.
Variable Mean S.D. Min. Max. Observations
Ln (GDP) Overall 23.11 2.27 18.94 27.82 N = 944
Between 2.29 19.55 27.33 N = 27
Within 0.36 22.24 24.09 T = 34.96
Ln (FBK) Overall 22.31 1.78 17.85 26.31 N = 659
Between 2.07 18.11 25.63 n = 27
Within 0.50 19.98 23.68 T-bar = 24.40
Ln (Labor) Overall 17.75 2.18 10.56 18.77 N = 875
Between 2.21 10.77 18.47 n = 25
Within 0.22 14.19 15.33 T = 35
Ln (FDI) Overall 19.41 2.22 11.46 24.63 N = 894
Between 1.85 16.64 23.27 n = 27
Within 1.26 12.69 22.36 T-bar = 33.11
Ln (GVA Agriculture) Overall 20.49 2.34 15.50 24.72 N = 903
Between 2.32 15.74 24.16 n = 27
Within 0.28 19.70 21.19 T-bar = 33.44
Ln (GVA Manufacturing) Overall 20.84 2.79 15.84 25.86 N = 872
Between 2.77 16.41 25.55 n = 27
Within 0.31 19.98 22.01 T-bar = 32.29
Ln (GVA Services) Overall 22.42 2.15 18.41 27.23 N = 903
Between 2.11 19.02 26.74 n = 27
Within 0.40 21.25 24.54 T-bar = 33.44
Ln (Exports) Overall 22.78 1.53 18.66 26.55 N = 682
Between 1.85 18.68 25.47 n = 27
Within 0.56 21.32 24.15 T-bar = 25.25
Fertility rate Overall 3.14 1.03 1.72 6.31 N = 861
Between 0.80 1.80 5.05 n = 27
Within 0.65 1.66 5.30 T = 31.88
Trade Overall 79.53 44.51 11.55 280.36 N = 918
Between 42.09 21.17 180.60 n = 27
Within 18.50 −5.54 179.30 T-bar = 34
Urbanization rate Overall 57.90 19.55 18.45 95.15 N = 945
Between 19.36 25.57 90.85 n = 27
Within 4.60 41.99 76.24 T = 35

Sectorial gross value added (GVA) captures the productive structure and specialization of the region. The gross value
added of agriculture (GVA) varies much more among countries than over time, given that the SD between is 3.22 versus a
SD within of 0.28. Similarly, the gross added value of manufacturing (GVAM) has a SD between of 2.7 versus a SD within of
approximately 0.3. In this case as well the gross added value of services (GVAS) varies more among countries than over time.
The SD between is 2 compared an SD within of approximately 0.4. In the three cases, the greater part of variance comes from
variation among countries than over time. Finally, exports, the fertility rate, trade (the sum of exports and imports) and the
urbanization rate have higher SDs between than SDs within, indicating that the greater part of variance comes from variations
among countries and that over time the variables are stable.
According to the Export-Led-Growth (ELG) hypothesis, based on Keesing (1967) and Krueger (1985), exports play an
important role in the economic growth of a country. There are Latin American countries strongly oriented to international
trade, because of which this variable can play a key role in determining levels of growth. In general, in countries with a
strong orientation to the international market, the Export-Led-Growth hypothesis has been verified. For example, Tiwari
and Mutascu (2011) show that the FDI and exports increasing the growth in the Asian countries. According to Barro and
Becker (1989), the fertility rate negatively affects economic growth. Fertility rates have declined slightly in recent years in
the region, the average being 3.14 births per woman. Finally, Latin America has been urbanizing very rapidly in recent years,
because of which the inclusion of this factor is justified based on the hypothesis of Henderson (2003), who argued that
urbanization increases productive efficiency.

3.2. Methodology

The neoclassical growth model offers a referential framework for analyzing economic growth, in which the level of
production is determined by physical capital and the labor force (Solow, 1956; Swan, 1956). This model has been discussed
broadly at the theoretical and empirical approach in the literature on economic growth (Mankiw et al., 1992; Klenow and
Rodriguez-Clare, 1997). If i represents the countries and t the time, the initial model can be represented in a Cobb–Douglas
technology function:

Yit = AKitα L1it−α (1)

where Yit represents the gross internal product of the receiving country i(i = 1, 2, . . . , I) in the period t (t =
1980, 1981, . . . , T ), A is the state of technology, Kit is physical capital, Lit is the labor force and e is the distributed error
 R. Alvarado et al. / Economic Analysis and Policy 56 (2017) 176–187 181

Table 2
Results of baseline regression.
AL HIC UMIC LMIC
Ln (Physical Capital) 0.338*** 0.454*** 0.268*** 0.250***
(24.25) (17.09) (15.04) (8.08)
Ln (Labor) 0.815*** 0.328*** 0.984*** 0.993***
(25.25) (3.63) (24.48) (16.63)
Ln (FDI) 0.0000003 0.0421*** −0.00246 −0.0341***
(0.00) (4.62) (−0.33) (−5.11)
Constant 3.828*** 8.348*** 2.876*** 3.165***
(11.22) (7.82) (6.60) (6.63)
Serial correlation test (p-value) 0.639 0.742 0.791 0.621
Fixed effects (time) Yes Yes Yes Yes
Fixed effects (country) Yes Yes Yes Yes
Observations 596 135 298 163
Adjusted R 2 0.917 0.916 0.935 0.938
Notes: ∗ , ∗∗ , ∗∗∗ denote statistical significance at the 5%, 1% and .1% levels, respectively, t-statistics reported in parenthesis.

term with a zero mean and variance σ . The parameters α and 1 − α measure the participation of factors capital and labor
in production, respectively. However, the neoclassical model requires expansion to capture the effect of other factors that
empirical evidence and theoretical arguments suggest influence product, such as FDI, exports, the productive structure,
the fertility rate, trade, the urbanization rate, among other factors associated with the internal and external behavior of an
economy. If we aggregate these variables we obtain an extended production function:

Ln(GDP )it = f (LnFBKFit , LnLaborit , LnFDIit ) (2)

Ln(GDP )it = f LnFBKFit , LnLaborit , LnFDIit , LnExportsit , LnAgricultureit ,

(

LnManufactureit , LnSer v iceit , Fertilityit , LnTradeit , Urbanizationit ) (3)

In order to correctly specify the econometric model, we use an expanded production function that has been widely
discussed in the theoretical and empirical literature that relates FDI flows and the economic growth (Anwar and Sun, 2011;
Iamsiraroj and Ulubaşoğlu, 2015; Iamsiraroj, 2016, among others). From the structure of the panel data of the previous
equations, Eq. (3) incorporates the term αi + ϕt + εit to capture the variability between observations and time, and stochastic
error, respectively. Further, the control variables in Eq. (3) have theoretical and empirical support, there is high degree
of colinearity between exports and trade and among the GVAs. Consequently, we estimated Eq. (2) plus the covariants
individually, one for each regression, while conserving the variables of the model (2). This strategy allows for incorporating
the effect of the productive structure of the region and of other factors that influence economic growth.
A relevant and even controversial aspect in panel data models is the choice between a fixed effects model (FE) or random
effects (RE). In this research, the Hausman test (1978) states that there is a difference between the coefficients obtained by
fixed and random effects (βFE − βRE ) in all regressions, therefore, we estimate with fixed effects, whose estimators tend to
be consistent (Hausman and Taylor, 1981).
Although the incorporation of fixed effects by country and by year captures in part the spatial and temporal heterogeneity,
the estimators may be biased when the error term Eit are not independent. In our investigation, independence can be
violated when the errors of different countries are correlated (contemporary correlation), when errors within each country
are temporally correlated (serial correlation), and when the variance is not constant. In order to ensure that the models are
corrected for autocorrelation and heteroscedasticity, we used the Wooldridge test to detect autocorrelation and the modified
Wald test to detect heteroscedasticity. Following Wooldridge (2002), autocorrelation was corrected by a self-regressive
process and the inclusion of the fixed effects of time and country eliminates heteroscedasticity.

4. Results and discussion

In this section we discuss the results found when we estimated Eqs. (2) and (3) for all of Latin America (LA) and for HIC,
UMIC and LMIC, respectively. Table 2 shows that (1 − α) is greater than α in all the regressions. The of physical capital
and labor force have positive and highly significant effects on the product. However, our variable of interest, the inflows
of FDI has a positive and significant effect on product only in high-income countries. The effect in UMIC is not significant
and in the LMIC it is negative and statistically significant. The results for Latin America and for UMIC are consistent with
results obtained by Lyroudi et al. (2004) for economies in transition. These authors indicate that FDI and economic growth
do not have any relationship in eastern European countries. Carkovic and Levine (2002) found a similar result in a work
on 72 countries for the period 1960–1995, 18 of which are Latin American. The result is consistent after including control
variables. The capacities of absorption and innovation associated with the education levels of countries can explain this
result (Borensztein, et al., 1998; Castellacci and Natera, 2016).
In order to verify that the variance of the error terms of all the estimated models is constant, we used the Modified Wald
test for heteroskedasticity. The null hypothesis of this test is of heteroskedasticity, that is, σi2 = σ 2 for all i = 1 . . . N, where N
182 R. Alvarado et al. / Economic Analysis and Policy 56 (2017) 176–187

Table 3
Results for Latin America with control variables.
M1 M2 M3 M4 M5 M6 M7
Ln (Physical Capital) 0.289∗∗∗ 0.264∗∗∗ 0.261∗∗∗ 0.249∗∗∗ 0.422∗∗∗ 0.336∗∗∗ 0.335∗∗∗
(23.00) (21.24) (18.71) (18.17) (27.74) (24.38) (23.64)
Ln (Labor) 0.546∗∗∗ 0.380∗∗∗ 0.496∗∗∗ 0.627∗∗∗ 0.846∗∗∗ 0.795∗∗∗
(16.11) (9.17) (13.89) (18.98) (25.85) (20.99)
Ln (FDI) −0.0134∗∗ −0.000012 0.00851 0.00500 0.00681 0.00384 −0.000471
(−3.15) (−0.00) (1.91) (1.11) (1.18) (0.80) (−0.10)
Ln (Exports) 0.195∗∗∗
(14.04)
Ln (GVA Agriculture) 0.479∗∗∗
(14.46)
Ln (GVA Manufacturing) 0.336∗∗∗
(13.20)
Ln (GVA Services) 0.225∗∗∗
(11.93)
Ln (Fertility) −0.197∗∗∗
(−16.62)
Ln (Trade) −0.00147∗∗∗
(−4.25)
Urbanization rate 0.00154
(1.02)
Constant 4.960∗∗∗ 1.940∗∗∗ 2.902∗∗∗ 3.418∗∗∗ 15.27∗∗∗ 3.380∗∗∗ 4.115∗∗∗
(16.11) (6.23) (9.49) (11.52) (43.20) (9.51) (9.29)
Serial correlation test (p-value) 0.171 0.193 0.169 0.175 0.223 0.211 0.227
Fixed effects (time) Yes Yes Yes Yes Yes Yes Yes
Fixed effects (country) Yes Yes Yes Yes Yes Yes Yes
Observations 595 582 551 582 586 595 596
Adjusted R 2 0.938 0.946 0.947 0.940 0.880 0.919 0.917
Notes: ∗ , ∗∗ , ∗∗∗ denote statistical significance at the 5%, 1% and .1% levels, respectively, and t-statistics reported in parenthesis.

is the number of countries. On the contrary, when the hypothesis H0 is rejected, the model has homoskedasticity. The results
confirm that the model has heteroskedasticity. Also, the test of Wooldridge (1991) reports the existence of autocorrelation in
the models. Since the estimated models have both problems, we use an estimation method consistent with autocorrelation
and heteroskedasticity. The modified Wald test for heteroscedasticity and the Wooldridge (1991) test for autocorrelation
applied to these estimates confirm that autocorrelation and heteroscedasticity of all models are corrected, respectively. As
well, to make the coefficients reported in Table 2 robust and unbiased, we incorporated the control variables formalized in
Eq. (3). The incorporated variables are the same as or similar to those included in other recent empirical research (Iamsiraroj
and Ulubaşoğlu, 2015; Iamsiraroj, 2016; Barbosa and Eiriz, 2009). We included a control variable in every regression to
avoid the composition effect of GDP with the sectorial GVAs and the high degree of colinearity between pairs of regressor.
The results suggest that the effects of physical capital and the labor force on product are stable although different covariants
are added. The coefficients of exports and of the GVAs of agriculture, manufacturing and services, the fertility rate and the
urbanization rate have the theoretically expected signs and are significant at 0.1%. Latin America relies heavily on exports
of minerals, oil and gas, and rising commodity prices have helped to expand output. A similar result occurs in Australia,
where Shafiullah et al. (2017) find that mining and fuel exports have played a crucial role in boosting Australia’s economic
growth nationally and in three of its regions. The coefficient of physical capital ranges between 0.25 and 0.45 and that of
the labor force between 0.32 and 0.99. The FDI coefficient is unstable, negative and significant in one regression and positive
and non-significant in the remaining six regressions. These results confirm what is shown in Table 2. The effects of FDI on
product in the 19 countries covered in this research are not consistent or consistent. Our results differ from those of Bengoa
and Sanchez-Robles (2003) and partly with those of Duttaray et al. (2008), who found a positive relationship between the
two variables in only 29 of the 66 countries considered in their research (see Table 3).
The coefficient associated with exports verifies the export-led-growth hypothesis for Latin America (see Table 3). The
strong dependence of the region on incomes from the primary-export sector and the increases in commodity prices in recent
years justify these results. Our results concur with the findings of other authors on this hypothesis in the region (Gutierrez
and Cantavella, 2007; Hernández and Razmi, 2014). Along this same line, the coefficient of agricultural GVA is positive and
significant and its effect on economic growth is greater than those of the manufacturing and services sectors. This result is
explained by the economic structure of the region already described. The effect of the fertility rate on economic growth is
negative and statistically significant, which concurs with the theory of Barro and Becker (1989). The parameter associated
with trade is negative and statistically significant, which corroborates the conclusions of Ulaşan (2015), who argues that
open markets do not accelerate economic growth in Latin America. However, trade has a positive effect on output in several
countries. For example, Turnbull et al. (2016) find empirical evidence that trade liberalization is a mechanism to improve
manufacturing productivity in Australia, and the FDI inflows have not had a statistically significant impact on the productivity
of the manufacturing sector. In the same direction, Manwa and Wijeweera (2016) find that South Africa has clearly benefited
from its trade liberalization policies both in the short and long term. The effect of the urbanization rate is positive, which
 R. Alvarado et al. / Economic Analysis and Policy 56 (2017) 176–187 183

Table 4
Results for HIC countries with control variables.
M1 M2 M3 M4 M5 M6 M7
Ln (Physical Capital) 0.333*** 0.296*** 0.388*** 0.127*** 0.451*** 0.458*** 0.475***
(21.26) (13.58) (10.00) (5.35) (18.98) (19.29) (21.92)
Ln (Labor) 0.285*** −0.0878 0.126 0.0406 0.211* −0.368**
(5.92) (−1.45) (1.43) (0.89) (2.54) (−3.30)
Ln (FDI) −0.00334 0.0196*** 0.0381*** 0.0307*** 0.0352*** 0.0272** −0.000568
(−0.61) (3.41) (4.71) (6.91) (3.78) (3.18) (−0.06)
Ln (Exports) 0.274***
(18.00)
Ln (GVA Agriculture) 0.663***
(12.71)
Ln (GVA Manufacturing) 0.270**
(3.07)
Ln (GVA Services) 0.706***
(17.83)
Ln (Fertility) −0.166***
(−4.50)
Ln (Trade) 0.006***
(5.77)
Urbanization rate 0.0605***
(8.26)
Constant 6.325*** 4.472*** 6.948*** 3.620*** 14.21*** 10.11*** 14.56***
(10.91) (6.02) (5.87) (5.99) (23.41) (10.09) (12.70)
Serial correlation test (p-value) 0.432 0.394 0.416 0.411 0.439 0.407 0.406
Fixed effects (time) Yes Yes Yes Yes Yes Yes Yes
Fixed effects (country) Yes Yes Yes Yes Yes Yes Yes
Observations 135 133 133 133 132 135 135
Adjusted R 2 0.976 0.970 0.937 0.981 0.917 0.933 0.945
Notes: t-statistics reported in parenthesis.
*
Denote statistical significance at the 5% level.
**
Denote statistical significance at the 1% level.
***
Denote statistical significance at the .1% level.

verifies the hypothesis of Henderson (2003) for the Latin America context. This author indicates that urbanization accelerates
economic growth.
The regressions per group of countries according to levels of national per capita income allowed for breaking down the
effect of FDI on economic growth of the 19 countries that cover this research. This process is necessary because there are
significant differences in the levels of income among countries. It cannot be expected that the effect of FDI on product is
similar in all the countries when they have different educational levels (Barro and Lee, 2013). The results of this process for
HIC countries are reported in Table 4, which confirm that the effects of FDI on growth in Chile and Uruguay are positive
and statistically significant in the majority of regressions. This result is contrary to theory, but is in accordance with the
context of the receiving country where the research was developed. It is well known that FDI in Latin America is directed
mainly at natural resource exploitation for export to external markets and to a lesser degree at manufacturing or services
(UNCTADSTAT, 2016). In this sense, the productive structure of the region conditions technology transfer from investors to
FDI receivers.
The positive result of the effect of FDI on GDP in HIC differs for the UMIC. In UMIC the coefficient associated with
FDI is negative and in most of the regressions it is statistically non-significant. When the covariants are exports and the
urbanization rate, the coefficient is negative and significant and in the other cases the coefficient it is non-significant at
5%. This reinforces the initial results shown in Table 2 for this group of countries, which includes the four largest and most
industrialized economies in Latin America: Brazil, Mexico, Colombia and Argentina. The ELG hypothesis is also verified with
these countries. However, the elasticity of exports with respect to product is only 0.13, which is lower than the elasticity
found for HIC (0.34). The GVA coefficient for manufacturing is greater than the estimators for the HIC countries, suggesting
that manufacturing plays an important role in the economic growth of this group of countries. Trade is also positive but
non-significant, which is very important from the perspective of fiscal and commercial policies given that several countries
in the region are immersed in free trade processes like NAFTA and the recent trans-Pacific trade agreement. In the ‘‘HIC’’ the
effect of trade on growth is significant at 10%, but in the UMIC is not statistically significant at any level. Finally, the effect of
the fertility rate on product is negative, while that of urbanization is positive in the previous regressions (see Table 5).
Finally, we report the results obtained for the LMIC (see Table 6). The effect of FDI on product is negative and in some
cases this effect is statistically significant. This implies that foreign investment does not play an important role in increasing
production levels in these countries in the period 1980–2014, but rather it could have decreased production. These results
are consistent with the findings of Saltz (1992) for third world countries. The low technology absorption capacity and
the low level of human capital in this group of countries could explain this result. The empirical evidence suggests that
improvements in the absorption capacity can increase the benefits of FDI (Anwar and Nguyen, 2014). A highly consistent
184 R. Alvarado et al. / Economic Analysis and Policy 56 (2017) 176–187

Table 5
Results for UMIC countries with control variables.
[M1] [M2] [M3] [M4] [M5] [M6] [M7]
Ln (Physical Capital) 0.262∗∗∗ 0.262∗∗∗ 0.255∗∗∗ 0.226∗∗∗ 0.356∗∗∗ 0.267∗∗∗ 0.255∗∗∗
(14.48) (15.36) (14.93) (13.25) (16.36) (15.71) (13.81)
Ln (Labor) 0.895∗∗∗ 0.690∗∗∗ 0.710∗∗∗ 0.805∗∗∗ 1.020∗∗∗ 0.920∗∗∗
(14.18) (10.45) (15.73) (18.67) (26.31) (19.68)
Ln (FDI) −0.00383 −0.00742 −0.0117 −0.00843 0.00598 0.00293 −0.00319
(−0.52) (−1.05) (−1.82) (−1.25) (0.61) (0.41) (−0.44)
Ln (Exports) 0.0466
(1.83)
Ln (GVA Agriculture) 0.252∗∗∗
(5.45)
Ln (GVA Manufacturing) 0.263∗∗∗
(9.48)
Ln (GVA Services) 0.159∗∗∗
(7.89)
Ln (Fertility) −0.278∗∗∗
(−15.52)
Ln (Trade) −0.0023∗∗∗
(−5.57)
Urbanization rate 0.00439∗∗
(2.60)
Constant 3.377∗∗∗ 2.259∗∗∗ 1.809∗∗∗ 3.024∗∗∗ 17.02∗∗∗ 2.405∗∗∗ 3.927∗∗∗
(6.58) (5.25) (4.76) (7.64) (34.42) (5.69) (6.64)
Serial correlation test (p-value) 0.218 0.200 0.209 0.194 0.230 0.226 0.204
Fixed effects (time) Yes Yes Yes Yes Yes Yes Yes
Fixed effects (country) Yes Yes Yes Yes Yes Yes Yes
Observations 298 298 274 297 294 298 298
Adjusted R 2 0.935 0.941 0.960 0.946 0.888 0.941 0.936
Notes: ∗ , ∗∗ , ∗∗∗ denote statistical significance at the 5%, 1% and .1% levels, respectively, and t-statistics reported in parenthesis.

result is that physical capital and the labor force have positive and stable effects in all the regressions. The ELG hypothesis
was also verified for this group of countries. An important initial result of these regressions is that the GVA for agriculture
is close to 0.25, which is the highest of all the coefficients of the included control variables. Productive specialization in this
sector supports this result. Another important result in this group of regressions is that the urbanization rate, the fertility
rate and trade are negative. In this case, the effect of trade is statistically significant (see Table 6).
In general, the results indicate that there is not a clear direction to the effects of FDI on economic growth in Latin America.
When the regressions incorporate the level of development reached by countries, FDI has a positive effect on high-income
countries, while in upper-middle-income countries it is not significant and in lower-middle-income countries it is negative.
In general terms, the results found in our research can be justified due to the low absorption capacity of technology that
the region possesses due to its low technological development and human capital deficient. In this sense, the absorption
capacity plays a crucial role in determining whether the effect of FDI on output is positive or negative. It is clear that in an
economy with low technological absorption capacity associated with foreign capital inflows, it cannot generate productive
linkages with small and medium-sized local enterprises. This hypothesis has been widely discussed recently by Liang (2017),
who analyzes the mechanisms of diffusion of foreign direct investment technology and finds that knowledge transfer is more
effective when the recipient has a high absorption capacity and is close of the source of knowledge. Adams and Opoku (2015)
estimate the effect of FDI on the product in 22 Saharan African countries using the generalized method of moments, and
conclude that both FDI and regulations (total regulations, credit market regulations, business regulations and regulations of
the labor market) do not have a significant independent effect on economic growth. These results are similar to those found
by Sánchez-Sellero et al. (2014), who find that R & D, internal innovation and the quality of human capital determine this
absorption. Both investigations are applied in countries with strong industrial capacity. However, R & D and innovation in
the region are at low levels. It is clear that in Latin America the FDI is oriented towards the exploitation of natural resources,
particularly in Brazil, Venezuela, Colombia, Ecuador, Peru, Chile, Mexico, Argentina, Bolivia and Nicaragua. In these countries,
between 45% and 83% of production is associated with the extraction of raw materials (World Bank, 2016).

5. Conclusions and policy implications

FDI flows into LA have increased significantly in the last decade. This has generated a broad debate about the benefits
and costs for receiving economies. On the one hand, it is argued that FDI transfers technology and human capital from the
investing to the receiving countries. However, productive specialization of the region can play a role in attenuating this
process. The productive structure of Latin America suggests that FDI oriented to the exploitation of natural resources leads
to volatile growth and that the effect on product is low or nil. The results of the empirical evidence are contradictory. Part
of the empirical literature indicates that the level of human capital on product is low or nil. The findings of the empirical
literature are contradictory. A part of the literature indicates that the level of human capital can determine the capacities
 R. Alvarado et al. / Economic Analysis and Policy 56 (2017) 176–187 185

Table 6
Results for LMIC countries with control variables.
M1 M2 M3 M4 M5 M6 M7
Ln (Physical Capital) 0.205*** 0.0968*** 0.0469* 0.147*** 0.393*** 0.258*** 0.272***
(7.44) (3.93) (2.49) (4.96) (10.19) (8.35) (9.37)
Ln (Labor) 0.767*** 0.306** 0.701*** 1.225*** 1.018*** 1.118***
(12.92) (2.75) (15.73) (10.81) (17.08) (18.41)
Ln (FDI) −0.0419*** −0.0152* −0.00810 −0.0165* −0.026** −0.0305*** −0.0304***
(−7.03) (−2.27) (−1.59) (−2.15) (−2.77) (−4.39) (−4.87)
Ln (Exports) 0.182***
(7.57)
Ln (GVA Agriculture) 0.859***
(8.41)
Ln (GVA Manufacturing) 0.431***
(13.38)
Ln (GVA Services) −0.0522
(−0.58)
Ln (Fertility) −0.198***
(−8.95)
Ln (Trade) −0.00119*
(−2.57)
Urbanization rate −0.0125***
(−5.06)
Constant 3.785*** −1.563** 2.300*** 2.533*** 16.32*** 2.594*** 1.277*
(8.87) (−2.65) (8.18) (5.25) (17.50) (4.97) (2.20)
Serial correlation test (p-value) 0.273 0.265 0.290 0.251 0.262 0.296 0.284
Fixed effects (time) Yes Yes Yes Yes Yes Yes Yes
Fixed effects (country) Yes Yes Yes Yes Yes Yes Yes
Observations 162 151 144 152 160 162 163
Adjusted R 2 0.954 0.973 0.984 0.961 0.887 0.939 0.947
Notes: t-statistics reported in parenthesis.
*
Denote statistical significance at the 5% level.
**
Denote statistical significance at the 1% level.
***
Denote statistical significance at the .1% level.

of economies for absorption, imitation and innovation. In this context, this research examines the effect of the inflows of
FDI on economic growth in 19 Latin America countries in the period 1980–2014. We ran several models of the random
effects of panel data in which an extended production function was estimated to include FDI and control variables were
added. The econometric regressions of panel data consider the level of development reached by the countries in the region
based on the Atlas Method of the World Bank. The results indicate that FDI does not have a positive effect on growth in Latin
America, with the exception of high-income countries covered in this research: Chile and Uruguay. Our results are consistent
with the conclusions of Carkovic and Levine (2002). Future research should focus on the instrumentation of the capacity to
absorb technology and productive specialization in theoretical formalization. This would clarify whether these factor have
effects different from that of FDI on economic growth. From the results found, we can draw two possible implications of
economic policy. First, high-income countries in the region can promote FDI inflows to increase and stabilize economic
growth. Second, the medium-low income countries of the region should promote fiscal instruments that allow FDI to be
targeted towards more specialized sectors where they generate productive linkages with the local productive structure and
transfer technology, rather than focusing on capital to the exploitation of natural resources.

References

Adams, S., Klobodu, E.K.M., 2017. Capital flows and the distribution of income in sub-Saharan Africa. Econ. Anal. Policy.
Adams, S., Opoku, E.E.O., 2015. Foreign direct investment, regulations and growth in sub-Saharan Africa. Econ. Anal. Policy 47, 48–56.
Adeniyi, O., Omisakin, O., Egwaikhide, F.O., Oyinlola, A., 2012. Foreign direct investment, economic growth and financial sector development in small open
developing economies. Econ. Anal. Policy 42 (1), 105–127.
Aghion, P., Howitt, P., 1998. Endogenous Growth. MIT Press.
Ahmad, M.H., Alam, S., Butt, M.S., Haroon, Y., 2003. Foreign direct investment, exports, and domestic output in Pakistan [with Comments]. Pakistan Dev.
Rev. 42, 715–723.
Alvarado, R., Toledo, E., 2017. Environmental degradation and economic growth: evidence for a developing country. Environ. Dev. Sustainabil. 19 (4), 1205–
1218.
Ang, J.B., 2009. Foreign direct investment and its impact on the Thai economy: the role of financial development. J. Econ. Financ. 33, 316–323.
Anwar, S., Nguyen, L.P., 2010. Foreign direct investment and economic growth in Vietnam. Asia Pac. Bus. Rev. 16, 183–202.
Anwar, S., Nguyen, L.P., 2011. Foreign direct investment and export spillovers: Evidence from Vietnam. International Business Review 20 (2), 177–193.
Anwar, S., Nguyen, L.P., 2014. Is foreign direct investment productive? A case study of the regions of Vietnam. J. Bus. Res. 67, 1376–1387.
Anwar, S., Sun, S., 2011. Financial development, foreign investment and economic growth in Malaysia. J. Asian Econ. 22, 335–342.
Babu, M.S., Bhaskaran, V., Venkatesh, M., 2016. Does inequality hamper long run growth? Evidence from Emerging Economies. Econ. Anal. Policy 52, 99–113.
Barbosa, N., Eiriz, V., 2009. Linking corporate productivity to foreign direct investment: An empirical assessment. Int. Bus. Rev. 18, 1–13.
186 R. Alvarado et al. / Economic Analysis and Policy 56 (2017) 176–187

Barrell, R., Pain, N., 1997. Foreign direct investment, technological change, and economic growth within Europe. Econom. J. 107, 1770–1786.
Barro, R.J., 2001. Human capital and growth. Amer. Econ. Rev. 91 (2), 12–17.
Barro, R.J., Becker, G.S., 1989. Fertility choice in a model of economic growth. Econometrica 57, 481–501.
Barro, R., Lee, J.W., 2013. A new data set of educational attainment in the world, 1950-2010. J. Dev. Econ. 104, 184–198.
Basu, P., Guariglia, A., 2007. Foreign direct investment, inequality, and growth. J. Macroecon. 29, 824–839.
Bengoa, M., Sanchez-Robles, B., 2003. Foreign direct investment, economic freedom and growth: new evidence from Latin America. Eur. J. Polit. Econ. 19,
529–545.
Borensztein, E., De Gregorio, J., Lee, J.W., 1998. How does foreign direct investment affect economic growth? J. Int. Econ. 45, 115–135.
Carkovic, M.V., Levine, R., 2002. Does foreign direct investment accelerate economic growth?. U of Minnesota Department of Finance Working Paper.
Castellacci, F., Natera, J.M., 2016. Innovation, absorptive capacity and growth heterogeneity: Development paths in Latin America 1970–2010. Struct. Change
Econ. Dyn. 37, 27–42.
Curwin, K.D., Mahutga, M.C., 2014. Foreign direct investment and economic growth: New evidence from post-socialist transition countries. Soc. Forces 92,
1159–1187.
Duttaray, M., Dutt, A.K., Mukhopadhyay, K., 2008. Foreign direct investment and economic growth in less developed countries: an empirical study of
causality and mechanisms. Appl. Econ. 40, 1927–1939.
Forte, R., Moura, R., 2013. The effects of Foreign direct investment on the host country’s economic growth: theory and empirical evidence. Singap. Econ.
Rev. 58, 1350017.
Gomes, D., Veiga, F.J., 2013. Financial globalization, convergence and growth: The role of foreign direct investment. J. Int. Money Finance 37, 161–186.
Gutierrez, S., Cantavella, M., 2007. Export-led growth: are the results robust across methodologies and/or data sets? A case study of Latin America. Appl.
Econ. 39, 1475–1500.
Hausman, J.A., 1978. Specification tests in econometrics. Econometrica 46, 1251–1272.
Hausman, J.A., Taylor, W.E., 1981. Panel data and unobservable individual effects. Econometrica 49, 1377–1398.
Henderson, J.V., 2003. The urbanization process and economic growth: the so-what question. J. Econ. Growth 8, 47–71.
Hermes, N., Lensink, R., 2003. Foreign direct investment, financial development and economic growth. J. Dev. Stud. 40, 142–163.
Hernández, G., Razmi, A., 2014. Latin America after the global crisis: the role of export-led and tradable-led growth regimes. Int. Rev. Appl. Econ. 28, 713–741.
Iamsiraroj, S., 2016. The foreign direct investment–economic growth nexus. Int. Rev. Econ. Finance 42, 116–133.
Iamsiraroj, S., Ulubaşoğlu, M.A., 2015. Foreign direct investment and economic growth: A real relationship or wishful thinking? Econ. Modell. 51, 200–213.
Kaldor, N., 1957. A model of economic growth. Econom. J. 67, 591–624.
Keesing, D.B., 1967. Outward-looking policies and economic development. Econom. J. 303–320.
Klenow, P.J., Rodriguez-Clare, A., 1997. Economic growth: A review essay. J. Monetary Econ. 40, 597–617.
Krueger, A.O., 1985. The Experiences and Lessons of Asia’s Super Exporters in Export-Oriented Development Strategies: The Success of Five Newly
Industrializing Countries. Discussion papers (Centre of Policy Studies, Monash University); no.11.
Lessmann, C., 2013. Foreign direct investment and regional inequality: A panel data analysis. China Econ. Rev. 24, 129–149.
Liang, F.H., 2017. Does foreign direct investment improve the productivity of domestic firms? Technology spillovers, industry linkages, and firm capabilities.
Res. Policy 46 (1), 138–159.
Liu, X., Burridge, P., Sinclair, P.J., 2002. Relationships between economic growth, foreign direct investment and trade: evidence from China. Appl. Econ. 34,
1433–1440.
Liu, X., Shu, C., Sinclair, P., 2009. Trade, foreign direct investment and economic growth in Asian economies. Appl. Econ. 41, 1603–1612.
Lucas, R.E., 1988. On the mechanics of economic development. J. Monetary Econ. 22, 3–42.
Lyroudi, K., Papanastasiou, J., Vamvakidis, A., 2004. Foreign direct investment and economic growth in transition economies. South - Eastern Eur. J. Econ. 2,
97–110.
Makki, S.S., Somwaru, A., 2004. Impact of foreign direct investment and trade on economic growth: Evidence from developing countries. Amer. J. Agric.
Econ. 86, 795–801.
Mankiw, G.N., Romer, D., Weil, D.N., 1992. A contribution to the empirics of economic growth. Quart. J. Econ. 107, 407–437.
Manwa, F., Wijeweera, A., 2016. Trade liberalisation and economic growth link: The case of Southern African Custom Union countries. Econ. Anal. Policy
51, 12–21.
Mencinger, J., 2003. Does foreign direct investment always enhance economic growth? Kyklos 56, 491–508.
Musibah, A.S., Shahzad, A., Fadzil, F.H.B., 2015. Impact of foreign investment in the Yemen’s economic growth: The country political stability as a main issue.
Asian Soc. Sci. 11, 102.
Nguyen, D.T.H., Sun, S., Anwar, S., 2017. A long-run and short-run analysis of the macroeconomic interrelationships in Vietnam. Econ. Anal. Policy 54, 15–25.
Popescu, G.H., 2014. FDI and economic growth in central and Eastern Europe. Sustainability 6, 8149–8163.
Preeti, F., Agrawal, G., 2014. Foreign direct investment (FDI) and economic growth relationship among highest FDI recipient Asian economies: A panel data
analysis. Internat. Business Manag. 8 (2), 126–132.
Romero, J., 2012. Inversión extranjera directa and economic growth en México: 1940-2011. Investig. Econ. 71, 109–147.
Saltz, I.S., 1992. The negative correlation between foreign direct investment and economic growth in the third world: Theory and evidence. Rivista
Internazionale Cienze Econom. Commer. 39 (7), 617–633.
Sánchez-Sellero, P., Rosell-Martínez, J., García-Vázquez, J.M., 2014. Absorptive capacity from foreign direct investment in Spanish manufacturing firms. Int.
Bus. Rev. 23 (2), 429–439.
Shafiullah, M., Selvanathan, S., Naranpanawa, A., 2017. The role of export composition in export-led growth in Australia and its regions. Econ. Anal. Policy
53, 62–76.
Solarin, S.A., Shahbaz, M., 2015. Natural gas consumption and economic growth: The role of foreign direct investment, capital formation and trade openness
in Malaysia. Renewable Sustainable Energy Rev. 42, 835–845.
Solow, R.M., 1956. A contribution to the theory of economic growth. Quart. J. Econ. 70, 65–94.
Suanes, M., Roca-Sagalés, O., 2015. Inversión extranjera directa, crecimiento económico y desigualdad en América Latina. Trimestre Econ. 82 (327), 675–706.
Sun, H., 2011. Co-integration study of relationship between foreign direct investment and economic growth. Int. Bus. Res. 4, 226.
Swan, T.W., 1956. Economic growth and capital accumulation. Econ. Rec. 32, 334–361.
Tan, B.W., Tang, C.F., 2016. Examining the causal linkages among domestic investment, FDI, trade, interest rate and economic growth in ASEAN-5 countries.
Int. J. Econ. Financ. Issues 6, 214–220.
Temiz, D., Gökmen, A., 2014. FDI inflow as an international business operation by MNCs and economic growth: An empirical study on Turkey. Int. Bus. Rev.
23, 145–154.
Tiwari, A.K., Mutascu, M., 2011. Economic growth and FDI in Asia: A panel-data approach. Econ. Anal. Policy 41 (2), 173–187.
Turnbull, C., Sun, S., Anwar, S., 2016. Trade liberalisation, inward FDI and productivity within Australia’s manufacturing sector. Econ. Anal. Policy 50, 41–51.
Ulaşan, B., 2015. Trade openness and economic growth: panel evidence. Appl. Econ. Lett. 22, 163–167.
 R. Alvarado et al. / Economic Analysis and Policy 56 (2017) 176–187 187

UNCTADSTAT. 2016. United Nations Conference on Trade and Development. Disponible online en: http://unctadstat.unctad.org/wds/TableViewer/
tableView.aspx?ReportId=96740.
Wooldridge, J.M., 1991. On the application of robust, regression-based diagnostics to models of conditional means and conditional variances. J. Econometrics
47 (1), 5–46.
Wooldridge, J.M., 2002. Econometric Analysis of Cross Section and Panel Data. MIT Press, Cambridge, MA.
World Bank. (Ed.). 2016. World Development Indicators: 2015. Washington DC. http://data.worldbank.org/data-catalog/world-development-indicators.
Wu, J.Y., Hsu, C.C., 2012. Foreign direct investment and income inequality: Does the relationship vary with absorptive capacity? Econ. Modell. 29, 2183–
2189.
Yao, S., 2006. On economic growth, FDI and exports in China. Appl. Econ. 38, 339–351.