Manufacturing Productivity and FDI Externalities: is Small
Beautiful?
ELEONORA SANTOS
Centre of Applied Research in Management and Economics,
Polytechnic Institute of Leiria, 2411-901 Leiria,
PORTUGAL
RUI ALEXANDRE CASTANHO
Faculty of Applied Sciences, WSB University
41-300 Dabrowa Górnicza
POLAND
&
College of Business and Economics, University of Johannesburg
PO Box 524, Auckland Park
SOUTH AFRICA
GUALTER COUTO
School of Business and Economics and CEEAplA
University of Azores
9500-321 Ponta Delgada, Portugal
PORTUGAL
ORCID: https://orcid.org/0000-0001-5560-5101
Abstract- The role of FDI as a vehicle for economic growth is debatable in practice. On the other hand,
the size of the company and the technological groups can influence the occurrence and magnitude of FDI
externalities. Thus, this article investigates the impact of firm size on the occurrence of foreign direct
investment externalities in the Portuguese industry from 1995 to 2007, by technology groups, using panel
data at the firm level. To this end, we estimate the TFP and regress it on a set of variables, including the
foreign presence in the same sector, upstream and downstream. The results show that only (small and
large) companies in scale-intensive industries; and small firms in science-based industries benefit from
the positive externalities of FDI. This suggests that firm size can influence the occurrence of FDI
externalities in the manufacturing sector, but only in some technology groups. Based on the results,
investment policy recommendations are made.
Key-Words: Foreign Direct Investment, Firm Size, Externalities, Manufacturing.
Received: March 23, 2022. Revised: July 10, 2022. Accepted: August 9, 2022. Available online: September 7, 2022.
1 Introduction
FDI can be an essential vehicle for
economic growth [1-2]. However, the role of
FDI as a vehicle for economic growth is a less
debatable assumption in theory than in practice
[3]. In particular, the issue of whether FDI
contributes to the increase of the Total Factor
Productivity (TFP) in manufacturing is of
particular importance, since Portugal is a small
open economy facing restrictions arising from
the economic crisis that slowed down the
productivity growth. Thus, we investigate the
existence of externalities from FDI in
Portuguese manufacturing, aiming to assist
industrial policy in choosing the appropriate
measures to promote the occurrence of
externalities, either in the same industry
(horizontal externalities) or in
upstream/downstream industries (vertical
externalities). However, one cannot reject the
possibility that externalities occur because the
initially more productive domestic firms attract
more foreign capital. Therefore, we employ
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2023.20.2
Eleonora Santos, Rui Alexandre Castanho, Gualter Couto
E-ISSN: 2224-2899
9
Volume 20, 2023
panel data from the AMADEUS database for
firms of all sizes over the period 1995-2007 to
ensure that firms with different levels of (TFP)
are evenly distributed in the sample. The time
span of 13 years allows for the study of
dynamic effects since previous empirical
literature finds that externalities need 2 years to
materialize.
The choice of the estimator depends on the data
and the underlying assumptions. Following
Wooldridge [4] and Levinsohn & Petrin's [5]
procedure, we use a control function approach
that employs intermediate inputs as the proxy
for unobserved productivity, this procedure has
the advantage of retaining a higher number of
observations than the Olley and Pakes [6]
approach because intermediate inputs are
always positive (at least in my database). In a
second stage, within a growth-accounting
framework, we estimate the impact of FDI on
the TFP growth, using the system GMM (Sys-
GMM) estimator proposed by Arellano and
Bover [7] and Blundel and Bond [8]. In
addition, we cluster the industries by
technological groups according to an adaptation
of O’Mahony e Van Ark [9] and Bogliacino
and Pianta [10] of Pavitt’s taxonomy. Thus,
firms in scale-intensive industries (NACE rev. 2
codes 10, 11, 12, 19, 22, 23, 24, 25, 29, and 30)
are large and their main source of technology
relies on the production engineering of their
suppliers and R&D; Firms in science-based
industries (NACE rev. 2 codes 20, 21, 26 and
27) are characterized by relatively large size
and produce roughly the same share of process
and product innovations. The sources of process
innovations are internal and external (from
suppliers); In supplier-dominated industries
(NACE rev. 2 codes 13, 14, 15, 16, 17, 18 and
31) firms are characterized by relatively small
size, limited resources regarding engineering
and internal R&D and rely on suppliers to
innovate; finally, in specialized-suppliers
industries (NACE rev. 2 codes 28, 32 and 33),
firms are relatively small and the consumers are
sensitive to their performance. We make several
contributions to the literature on externalities
from FDI. First, we investigate the existence of
both horizontal and vertical externalities from
FDI in Portugal. Second, we use lags in the
measures of foreign presence to account for the
time lapse required for externalities to
materialize. Third, we break down the results
across industries along their trajectories of
technological change which allows uncovering
some interesting patterns. Indeed, the
technological groups that benefit more from
foreign presence are scale-intensive and
science-based industries.
This paper is organized as follows. Section 2
reviews the Literature; Section 3 analyses FDI
flows and stocks in manufacturing; section 4
describes data and methodology; Section 5
reports the results; Section 6 discusses de result;
and section 7 concludes.
2 Empirical Literature
Panel studies, at the firm level, include [11-14].
Farinha and Mata [11] analyzed the 1986-1992
period while Proença et al. [12] focused their
analysis between 1996 and 1998, and Crespo et
al. [13-14] analyzed the period 1996-2001.
Except for Farinha and Mata [11], which use a
random-effects model, all authors use the sys-
GMM to regress the labor productivity on the
level of foreign presence (whose proxy is the
employment in foreign firms, except Proença et
al. [12] that use the capital stock). Data sources
are Dun & Bradstreet and Quadros de Pessoal,
except Farinha and Mata [11] that also use data
from Banco de Portugal.
The present study is the most comprehensive
for Portugal, regarding time (1995-2007) and
sample size (65,585 observations). In addition,
until now, only Crespo et al. [13-14] have
investigated the existence of vertical
externalities in Portugal. Moreover, there are no
studies for 2001-2007 and the results for 1996-
2000 are controversial. Indeed, regarding
horizontal externalities, while Crespo et al. [14]
find negative results, for 1996-2001; Proença et
al. [12,15] find no significant results for 1996-
1998, and Crespo et al. [13] find negative
results for 1996-2000. Finally, Crespo et al.
[14] find evidence of positive vertical
externalities (via backward linkages) for 1996-
2001, but only at a regional level. One possible
cause for these controversial results may be the
underestimation of the externality effects due to
econometric problems associated with
traditional panel data estimation methods, as
highlighted by Proença et al. [15]. In addition to
different results for the same period, hitherto
researchers tested the impact of a few
determinant factors of FDI externalities, i.e., the
technological gap/absorptive capacity and the
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2023.20.2
Eleonora Santos, Rui Alexandre Castanho, Gualter Couto
E-ISSN: 2224-2899
10
Volume 20, 2023
geographical proximity between MNCs and
domestic firms. Given the lack of consensus in
these studies, we analyze the occurrence of both
horizontal and vertical externalities in the
manufacturing firms from 1995 to 2007. The
results will enable policymakers to identify the
industries that benefit more from the foreign
presence and to implement relevant policies to
leverage positive externality effects.
3 FDI in the Manufacturing
During the 1990s, the industrial policy in
Portugal focused on attracting foreign capital,
mostly through privatizations, but also by
offering Government and EU subsidies and
assistance to investors. Although, FDI inflows
represented only 5.7% of GDP over this period,
in 2007, foreign capital stocks represented
nearly 50% of GDP.
Though the Portuguese accession to the
European Economic Community (EEC) in 1986
encouraged the increase of FDI flows; in the
1990s there was a sharp decline compared to
the previous decade due to adverse factors,
namely the instability of interest and exchange
rates, the slowdown of the privatization
program and the end of the full exploitation of
single market investment opportunities. The
value of flows in 1995-1999 ranged from $660
million to $3,005 million. From 2000 to 2007,
there was a large fluctuation in FDI flows.
From 2000 to 2001, the trends in mergers and
acquisitions (M&As) caused a boom in FDI
flows at a global level [16]; with Portugal
attracting over $6 billion. However, in the
following year, the rise of oil prices caused a
drop in the flows of FDI directed to Portugal to
$1,801 million. In 2003 there was an increase to
$7,155 million, and then a sharp decrease in the
following years. According to OECD [17], in
2006 the global economy grew faster and thus,
FDI flows recorded the highest peak of $10,914
million. However, in 2007 this amount was
reduced by more than half, standing at $ 3,063
million. In that year, according to UNCTAD,
Portugal occupied the 29th position worldwide
in terms of FDI attraction, which was above
that of several Eastern European countries such
as the Czech Republic and Hungary. Given the
peripheral location and the weaknesses that
Portugal presents at the aggregate level, namely
the low productivity, low educational level, and
low R&D expenditures, which are prone to
cause disadvantages when competing with other
low-cost labor destinations, this represented a
major achievement. Nevertheless, according to
the OECD database, from 1995 to 2007,
Portugal attracted, on average, only 0.7% of
global FDI flows. In this context, the
importance of the manufacturing concerns the
technology transfer from MNCs to local firms,
due to its high innovation indices and potential
indirect and induced impacts on other sectors
through "pull" and "push" effects.
In 1995, the manufacturing sector ranked first
relative to other economic sectors capturing
40% of FDI flows. In the following years, due
to the domestic economic crisis, its importance
in attracting flows decreased, and there were
disinvestments in 1998-1999, 2001-2002, 2005,
and 2007. This may be attributed to the
worldwide reorganization of labor-intensive
manufacturing industries towards fragmented
production systems taking advantage of cost
differentials of Central and Eastern European
countries (CEECs). However, in 2004, the
flows to manufacturing grew exponentially
compared to 2003, reaching a peak that
represented nearly half of total FDI inflows to
Portugal. In the years 1996-1998; 2002-2003
and 2005-2007, the preferred industries by
foreign investors were chemicals and rubber,
and plastics. From 1999 to 2000, foreign firms
targeted the textiles, clothing, and footwear
industries. In 2001, most foreign capitals was
directed to machinery and equipment industries;
and in 2004, to electric and optical equipment.
Textiles ranked second, in the years 1996-1998
and 2003; and machinery and equipment ranked
second in the years 2002 and 2005- 2007. In
2004 the industries of food, beverages, and
tobacco ranked second in foreign investors’
preferences.
4 Methodology
Data for the Portuguese manufacturing firms in
1995-2007 come from AMADEUS database.
The balanced panel data set includes 5,045
manufacturing firms of all sizes (4,685
domestic and 360 foreign) for the 13 years in a
total of 65,585 observations. The regression
analysis, however, includes only 51,535
observations since the rest of the observations
dropped due to collinearity. Table 1 shows
some basic statistics.
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2023.20.2
Eleonora Santos, Rui Alexandre Castanho, Gualter Couto
E-ISSN: 2224-2899
11
Volume 20, 2023
Firms with foreign capital represent 7% with a
mean share of foreign capital of 58%. There are
12 Greenfield projects in 9 industries. The rest
of the foreign firms correspond to Mergers &
Acquisitions (M&As). Since foreign firms are
more productive than their domestic
counterparts, we will analyze the occurrence of
productivity externalities from foreign firms to
domestic firms. Using the WLP (09) procedure
to address endogeneity in capital and the
possibility of productivity shocks, we estimate
the level of TFP, departing from the following
equation
Yijt = Aijt Kijt βk Lijt βlMijtβm (1)
where Yijt represents the physical output of
firm i in sector j and period t, Kijt, Lijt and Mijt
are the inputs of capital, labor, and materials,
respectively. Aijt is the Hicksian neutral
efficiency level (my concept of total factor
productivity – TFP) of firm i in period t. For a
given level of A, higher output levels demand
higher inputs (K, L and M) levels.
We assume that L =LP+LNP, where LP stands for
production worker (unskilled) labor and LNP
stands for non-production worker (skilled)
labor. we proxy LNP by the sectoral average of
years of schooling since we do not possess
information for individual firms. Aijt is not
observable and needs to be estimated.
Table 1. Summary statistics
Total number of firms
5,045
Fully domestic firms
4,685
Firms with foreign share
360
Mean (domestic firms)
TFP
5,152
Capital
2,657,802
Labor
34
Mean (foreign firms)
TFP
7,691
Capital
1,162,430
Labor
24
Source: Author’s calculations on Stata 13.0
The estimation of Aijt, depends on several
different components such as skills,
knowledge and firm-level capabilities,
including managerial and organizational
competences. we assume that Aijt, or TFP in
logs, is given by:
ln (Aijt ) = β0 + εijt (2)
where β0 measures the mean efficiency level
across firms over time; εijt is the time- and
producer-specific deviation from that mean.
Taking natural logs of (1) and inserting
equation (2) we obtain a linear production
function
yijt = β0 + βkkijt + βlPlPijt + βlNPlNPijt + βmmijt + εijt
(3)
where lower cases refer to natural logarithms.
The error term εijt can be further decomposed
into an observable (or at least predictable); and
an unobservable i.i.d. component, representing
unexpected deviations from the mean due to
measurement error, unexpected delays, or
other external circumstances, i.e, εijt =vij + uqijt.
Hence, equation (3) becomes
yijt = β0 + βkkijt + βlPlPijt + βlNPlNPijt + βmmijt +
vijt + uqijt (4)
Since the firm-level productivity is tfpijt = β0 +
vijt; and rearranging the terms of (2) we obtain
tfpijt= yijt –( βkkijt + βlPlPijt + βlNPlNPijt + βmmijt ) -
uqijt (5)
And the estimated productivity is
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2023.20.2
Eleonora Santos, Rui Alexandre Castanho, Gualter Couto
E-ISSN: 2224-2899
12
Volume 20, 2023
q
ijt ijt
=tfp u
tfp
(6)
Defining the value added as vaijt=yijt-βmmijt, then
it can be estimated through equation (4) as a
residual
v P v NP v
ijt jP ijt jNP ijt jK ijt
ijt ˆ ˆ ˆ
= va - ( l + l + k )
tfp
(7)
The growth of the estimated TFP is regressed on a
set of variables, within a fixed effects dynamic
model, including a time trend. The three sets of
variables are described as follows.
Variables related to foreign presence.
Externalities from FDI may be horizontal or
vertical. Horizontal externalities occur when
the entry of the MNC generates positive
externalities for local competitors. Vertical
externalities occur when the links between
MNCs and their local suppliers/customers
(backward/forward linkages) generate positive
externalities. Hence, we measure the foreign
presence through three variables hor, back and
for defined at sectoral level. Horizontal
technology transfer occurs through contact
with local competitors (via
demonstration/imitation, labor mobility,
exports, competition, consulting and
specialized services, and coordination with
local institutions). hor is a sectoral externality
variable that measures the share of output by
foreign firms in the total output of the
industry, i.e., measures the presence of FDI on
a given industry and is calculated in the
following way
(8)
where foutputit is the output of firms with
foreign capital operating in industry j at time t.
Thus the value of the variable increases with
the output of foreign firms. Hirschman [18]
stated that a lack of linkages in the developing
economy leads to a lack of industrial
development. From a developmental
perspective, it is generally assumed that
linkages between MNCs and domestic firms
are better than no linkages, and the more and
deeper linkages are, the better it is for the host
economy [19-20]. MNCs in other industries
appeared to foster broad linkages in the host
economy by creating industries that supply the
MNC and by inducing forward industries to
use the multinational’s output as inputs, the
crowding-in effect of FDI [21]. The variable
hor measures the presence of FDI in a given
industry, then the higher its value the greater
the increase in domestic firms’ productivity.
Thus, following Barrios and Strobl [22] 2002
we expect a positive effect on domestic firms’
TFP growth. Vertical externalities occur when
an MNC increases the demand for local inputs,
leading to increased specialization in upstream
sectors and, as a result, causing the reduction
of costs in downstream sectors. If the MNCs
are interested in maintaining the quality
standards they are likely to provide technical
support to local suppliers in order to improve
the quality of inputs, or assist them in the
introduction of innovations, training, creation
of productive infrastructure, procurement of
raw materials, as well as the introduction of
new management techniques, among others
[23]. Vertical technology transfer occurs
through linkages with local suppliers
(backward linkages) or local customers
(forward linkages). We define back as
(9)
where δjk is the share of industry j’s output
supplied to industry with foreign presence k.
The variable back is intended to capture the
effect that multinational customers have on
domestic suppliers. Both j and k are two-digit
industries.
Forward linkages occur when the MNCs
provide higher quality and/or cheaper inputs to
their clients that produce final goods [24].
Better quality inputs supplied by foreign firms
may increase the productivity of domestic
firms in industry j. Similarly, we define for as
(10)
where λkj is the share of inputs that industry j
buys from industry k. The variable captures
the contacts between domestic firms and their
foreign suppliers. Parameters δ and λ are
obtained from the OECD Input-Output (IO)
Tables. We exclude the diagonal elements of
the IO tables in the calculation of the weighted
average because intrasectoral effects are
accounted for in the variable hor. Moreover,
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2023.20.2
Eleonora Santos, Rui Alexandre Castanho, Gualter Couto
E-ISSN: 2224-2899
13
Volume 20, 2023
we focus on inputs for intermediate
consumption; therefore, we do not include the
imports, exports, or other components of final
demand in the calculation of the IO
coefficients. As highlighted by [25], the net
effect of linkages can either be positive or
negative when domestic suppliers serve the
MNCs exclusively. Indeed, under these
circumstances, the technology transferred to
domestic suppliers increases but the reduction
of the rivalry among domestic suppliers tends
to reduce the aggregate output level of the
intermediate goods industry. In addition, a
decrease in the cost of inputs compatible with
the foreign technology, while benefiting
foreign firms and the most productive
downstream domestic firms adopting the
foreign technology, it negatively affects firms
using the domestic technology [26]. However,
we assume that the higher the value of back
and for, the greater the magnitude of vertical
externalities and thus the greater the effect on
the TFP growth of domestic firms. The
increase in demand for high-quality inputs by
MNCs or due to the purchase of better-quality
inputs provided by foreign firms [23-24].
Hence, following Markusen and Venables [24]
we expect a positive coefficient for variables
back and for.
Control variables. We include six control
variables; hfd is the Herfindhal index that
measures market concentration, rd is the value
of R&D expenses proxied by firms’ intangible
assets, mrdf is the average value of sectoral
foreign R&D expenditure, s measures the
scale of operations, tg is the technological gap,
and kl measures the capital intensity.
Concentration. The Herfindhal index
indicates market concentration and is
calculated as
(11)
where X represents the output of firm g
(domestic or foreign) belonging to sector j, at
time t. The output is proxied by firm turnover
obtained from AMADEUS database, deflated
by a Producer Price Index. The Herfindahl
index also serves as a proxy of (the lack of)
competition. Indeed, since this variable is
calculated as a share (%), values close to 0
indicate markets under perfect competition,
and a value of 100 denotes the presence of
monopoly rents. If the impact of the variable
hfd on the TFP growth is positive, it means
that the market power can facilitate access to
the necessary resources for domestic firms to
increase their productivity. Indeed, stronger
industry concentration generates larger profits
that can be re-invested, for example, in new
technologies or in the production of more
sophisticated products; however, if the sign is
negative, it implies that the monopolistic
inefficiencies are causing a decrease in the rate
of innovation [27] and, thus, a loss of
productivity. As a result, the expected sign of
this variable is not predefined.
Domestic R&D expenditure. Endogenous
growth theories predict R&D activities to be
an important determinant of TFP growth since
innovations can ultimately raise efficiency
[28-30]. The variable rd is included in our
model to proxy the domestic firms’ absorptive
capacity. A certain level of absorptive capacity
is required to absorb foreign technology [31].
Domestic R&D expenditures influence
domestic TFP in three ways. Firstly, R&D
may be cost reducing, lowering the production
costs. Secondly, firms may create and produce
new products with R&D expenditures by using
the same volume of factors. Finally, R&D
activities increase the capacity of domestic
firms to imitate new technologies and use it as
a proxy for absorptive capacity [32-34]. Thus,
we expect a positive sign for the coefficient of
rd.
Average sectoral R&D expenditure of foreign
firms. The variable mrdf is included in our
model to proxy the average stock of foreign
knowledge in each industry. Liu and Buck
[31] found evidence that foreign R&D
activities had positive impacts on the
innovation performance of domestic firms if
domestic firms possess the absorptive capacity
to learn foreign knowledge. Because
innovations are a source of TFP growth, we
expect a positive sign for the coefficient of
mrdf.
Scale. Small firms have less capacity to
benefit from the foreign presence and are less
capable to face competition [35]. Yet, some
studies [36-38] find that only small domestic
firms (and medium in the latter case) benefit
from positive externalities from FDI. Hence,
the evidence on the impact of scale on firms’
productivity appears to be inconclusive.
Nonetheless, in the presence of increasing
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2023.20.2
Eleonora Santos, Rui Alexandre Castanho, Gualter Couto
E-ISSN: 2224-2899
14
Volume 20, 2023
returns to scale, i.e., if there is an industry-
specific optimal scale, then TFP increases with
scale [39-40] and we expect a positive
coefficient for s.
Technological gap. The determinants of
technology diffusion build on models by [30,
41-43]. Following Gerschenkron [44]
hypothesis, technological progress is an
increasing function of the technology gap (tg).
We define a way to measure the speed of
technology diffusion, i.e., to capture
autonomous technological transfer from
foreign firms to technologically laggard
domestic firms [45-46]. The indicator is a ratio
of labor productivity between domestic firms
and the presumptive foreign leader. Therefore,
the variable tg is constructed as an inverse
measure of the technological gap since values
of this variable close to 1 mean a small gap,
and values close to 0 signify a large gap. Thus,
and according to the catching-up hypothesis, if
the value of tg is close to one, the gap is too
small, which means that domestic and foreign
firms possess similar levels of efficiency and,
thus, the domestic firms are not prone to learn
much from the MNCs. However, according to
the technology-accumulation hypothesis, if the
value of tg is close to zero, the gap is too
large; which means that domestic firms do not
possess the necessary "absorptive capacity" to
incorporate the knowledge of foreign firms
[32, 47-49]. Thus, the expected coefficient of
this variable is not predefined.
Capital intensity. Capital intensity represents a
firm’s commitment to modernization and
upgrading of its productive capacity. In the
long run, capital expenditures typically have a
positive impact on firms’ performance [50-
51]. The higher the capital intensity is, the
higher the expected TFP [52]. Hence, we
expect a positive coefficient for kl.
Interaction variables. These variables are
included in our model to test the impact of
foreign presence on the TFP growth of
Portuguese manufacturing firms, given the
values of concentration, absorptive capacity,
and a sectoral average of foreign knowledge,
scale, and technological gap, and capital
intensity. Thus, we include the interaction
variables labeled F*hfd, F*rd, F*mrdf, F*s,
F*tg, and F*kl, respectively. Where F stands
for the measure of foreign presence in the
same industry (hor), in downstream (back) or
upstream industries (for).
FDI and concentration. If the impact of the
variable F*hfd is positive, it means that the
impact of foreign presence in the TFP growth
of Portuguese manufacturing firms is positive,
given the values of market concentration. In
other words, the influence of concentration on
the referred impact is positive because the
benefits of having market power outweigh the
potential disadvantage of inefficiencies from
monopoly rents; and otherwise, if the value of
F*hfd is negative. Hence, the sign of F*hfd is
not predefined.
FDI and absorptive capacity. From what was
said above about the domestic firms’
absorptive capacity, we assume that the impact
of foreign presence in the TFP growth of
Portuguese manufacturing firms, given a
certain level of absorptive capacity, is positive,
i.e., that the coefficient of the variable F*rd is
positive.
FDI and the average stock of foreign
knowledge in the industry. The empirical
literature provides evidence of the positive
impacts of foreign R&D activities on the
innovation performance of domestic firms, as
described above. Hence, we assume a positive
impact of foreign presence in the TFP growth
of Portuguese manufacturing firms, given a
certain level of foreign R&D activities. The
expected sign for the variable F*mrdf is
positive.
FDI and scale. We assume a positive impact
of foreign presence in the TFP growth of the
Portuguese manufacturing firms, given a
certain level of scale, because the adoption of
an efficient scale of operations is important to
increase the TFP. Consequently, we expect a
positive sign for the variable F*s.
FDI and technological gap. For the
Portuguese economy, Flôres et al. [53]
suggest that the externality effects are
maximized when the technological gap is
between 50%- 80% while Proença et al. [12]
find that tg must be around 60%-95% to
maximize the externalities. Thus, the expected
sign of F*tg is not predefined.
FDI and capital intensity. Foreign firms
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2023.20.2
Eleonora Santos, Rui Alexandre Castanho, Gualter Couto
E-ISSN: 2224-2899
15
Volume 20, 2023
usually use more capital-intensive
technologies [54-55]. The extent to which
local firms benefit from this superior
technology depends largely on their own
technological capabilities as defined by capital
intensity [56-57]. Therefore, we assume a
positive impact of foreign presence in the TFP
growth of Portuguese manufacturing firms,
given a certain level of capital intensity, and
expect a positive sign of F*kl.
Thus, in the second stage of our econometric
strategy, the growth of the estimated TFP is
regressed against a set of variables that measure
the foreign presence, interaction terms, and
other explanatory variables, within a fixed-
effects dynamic model, including a time trend.
The econometric specification is
ij( t 1)
0 2 j(t m) 3 jt ijt 4 jt ijt 5 jt
1 jt
6 jt ijt) 7 jt ijt 8 jt 9 10 11 12
i
ijt
jt ijt ijt jt ijt
13 ijt t it
ijt
2
)
m0
14
d f (f *hfd ) (f *rd (f *mrdf )
tfp tfp
(f *s (f *kl ) (f *tg ) hfd rd mrdf ds
tg kl
(12)
Where the lowercases denote variables in
logarithms and f is the measure of foreign
presence (hor, b1, and f1). we also include year
dummies γt that account for possible changes in
the growth of TFP due to stochastic shocks at a
firm or sectoral level over time and an error
term . If it is expected that the current level
of the dependent variable (DV) is heavily
determined by its past level, then we use a
dynamic specification that includes a lagged
dependent variable ( ). The inclusion of
lagged DVs is necessary to avoid unreliable
results due to an omitted variable bias and
reduce the occurrence of autocorrelation arising
from model misspecification. we include two
lags of the variables that represent the foreign
presence since empirical studies indicate a
period of two years for domestic firms to absorb
the foreign knowledge and externalities to
materialize. For example, Merlevede et al. [58]
find evidence that “the first two years after
entry, domestic firms that supply minority
foreign entrants enjoy a substantial contribution
to productivity growth” (op cit. p.22). we use
the Sys-GMM to estimate equation (12), which
combines the equation in first differences with
the equation in levels. Hence, we use fixed
effects in the equation in levels. In this dynamic
model, the lagged dependent variable ( )
may be correlated with the error term () and
the endogenous variables, causing the OLS
estimator to be inconsistent and biased [59].
Nickell [60] demonstrated that the use of the
within estimator (also known as fixed effects
estimator) in first-order autoregressive models
with fixed effects leads to biased results for the
estimated coefficient of the lagged dependent
variable. However, there is still the
autocorrelation problem since the term is
correlated with the term in
. The independent
variables are endogenous (kl, tg, f, f*hfd);
predetermined (s) and exogenous (hfd, rd, mrdf,
f*mrdf, f*s and f*tg). However, any not strictly
exogenous predetermined variable becomes
potentially endogenous since it can also be
correlated with the error term [61].
Arellano and Bond [52] and Bond [53] suggest
the use of instrumental variables in equation
(12) to deal with the autocorrelation and
endogeneity issues. Considering equation (12),
we use lags of the dependent variable in levels,
lagged two or more periods, as valid
instruments for periods t=3…, T, as in Arellano
and Bond [62] and Bond [63].
Since the variables that proxy for foreign
presence are highly correlated, we regress each
type of externality for domestic firms in a
separate equation. we use the command
xtabond2 in software STATA 13.0 to
implement the System GMM two-step estimator
with the Windmeijer (2005) correction.
Industries of tobacco and petroleum (with codes
12 and 19 according to classification Nace
Revision 2) were dropped due to an insufficient
number of observations.
5 Results
Table 2 (in annex) presents the impact of
company size on externalities, by type of
externality and technological groups.
Examining Table 2, we can conclude that the
large size influences the occurrence of
externalities in scale-intensive industries. There
are significant and positive horizontal
externalities in food industry, and backward and
forward linkages in beverages industry.
However, FDI only has an effect in terms of
externalities with time lags except for backward
linkages in beverages, where externalities are an
immediate effect of foreign presence. The large
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2023.20.2
Eleonora Santos, Rui Alexandre Castanho, Gualter Couto
E-ISSN: 2224-2899
16
Volume 20, 2023
size is also important for externalities to occur
by backward linkages in the repair installation
of machinery and equipment immediately.
With respect to small firms, we find positive
horizontal externalities in scale intensive
industries such as basic metals and other
transport equipment; and in specialized
suppliers (repair and equipment) and science-
based industries (computers and electronics).
Except for basic metals where the externalities
occur immediately, in the remaining
industries, it takes one or two periods for
horizontal externalities to occur.
we find externalities via backward linkages in
scale-intensive industries like basic metals and
other transport equipment, but with a lag; we
also find positive externalities with lags in
science-based industries (computers and
electronics).
We find positive externalities through forward
linkages in scale-intensive industries such as
beverages and metal products; and also in
science-based industries (electrical equipment).
Except for beverages, externalities via forward
linkages occur with lags.
6 Dıscussıon
The validity of the results with system-GMM
depends on the statistical diagnostics. we
started by testing for the presence of second-
order autocorrelation in the error term. The
presence of AR (1) poses no problem because
the differenced residuals are expected to
follow a MA(1) process, however, if there is
AR(2) autocorrelation, then the GMM-
estimator is inconsistent. The reason is that the
Arellano-Bond (AB) orthogonality conditions
are established under the assumption that the
error term in the level’s equation is not auto
correlated. If the error term in the level’s
equation is not autocorrelated, then the error
term in the first-difference equation has
negative first-order autocorrelation and 0
second-order autocorrelation. If one rejects the
hypothesis that there is 0 second-order
autocorrelation in the residuals of the first-
difference equation, then one also rejects the
hypothesis that the error term in the level’s
equation is not autocorrelated.
This indicates that the AB orthogonality
conditions are not valid, no matter which lags
are used as instruments. Thus, we tested for
second-order serial correlation. we also report
the results of Hansen's J test of overidentifying
restrictions but not the Sargan's statistic. The
reason is that Sargan's statistic is a special case
of Hansen's J under the assumption of
homoscedasticity.
Table 2. Impact of firm size on FDI externalities
Scale
Specialized
Suppliers
Science-Based
Supplier-
Dominated
Horizontal
Industry
Food
Basic
Metals
Other
Transport
Equipment
+
-1, -2
Repair and
Installation of
Machinery &
Equipment
Computers &
Electronics
Signal
+
+/-
+
+
Period
-1, -2
0,-1
-2
-1, -2
Size
Large
Small
Small
Small
Small
Backward
Industry
Beverages
Basic
Metals
Other
Transport
Industries
Repair and
Installation of
Machinery &
Equipment
Computers &
Electronics
Wood
Signal
+
+/-
+
+
+
-
Period
0
-1, -2
-1, -2
0
-2
0
Size
Large
Small
Small
Large
Small
Large
Forward
Industry
Beverages
Metallic
Products
Electrical
Equipment
Wood
Signal
+
+/-
-/+
-
Period
0,-2
-1, -2
-1, -2
0
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2023.20.2
Eleonora Santos, Rui Alexandre Castanho, Gualter Couto
E-ISSN: 2224-2899
17
Volume 20, 2023
A reasonable approach is to use robust
standard error estimation to deal with
heteroskedasticity (and thus rendering the
Sargan test unjustified), and then test for
the remaining autocorrelation using the
autocorrelation test which is more
sensitive to such problems than the Sargan
test. Following Roodman [61], we also
report the number of instruments used in
the dynamic panel, since this kind of
models can generate an enormous number
of potentially “weak” instruments that can
cause biased estimates. There are no clear
rules concerning how many instruments is
“too many” [61], but the number of
instruments should not exceed the number
of groups, which is the case. Second, the
p-value should have a higher value than
the conventional 0.05 or 0.10 levels. we
examined the sensitivity of system-GMM
regression results to the number of lagged
instruments and to alternative number of
independent variables.
However, in theses alternative
specifications, Arellano-Bond [AR (2)]
and Hansen tests rejected the null
hypothesis and/or the coefficient of
variables become non-significant. Tables 3
to 20 show that the results for AR (2) and
Hansen's J test support the validity of the
chosen model specification. In this
analysis, we report only the significant
results (p-values are listed in parenthesis,
next to the coefficient values). The Impact
of Size in Scale-Intensive Industries-
Although this research shed light on FDI
externalities, it is not exempt of limitations.
First, it was necessary to obtain disaggregated
data at the firm level from different sources,
that had to be harmonized. Secondly, data
referring to human capital and R&D are
lacking, so we had to use proxies with their
inherent limitations. Thirdly, we do not have
input-output tables for all years and had to
assume that values would not change largely
each year.
7 Conclusıon
The Portuguese manufacturing industry is
characterized by small companies and a
dynamic of innovation that relies heavily on
the so-called traditional industries. These
industries are characterized by small firms,
typically producing low value-added products,
which potentially threatens the growth of TFP.
With this in mind, we investigated whether
firm size is more likely to benefit from FDI
externalities in 1995-2007. Grouping
industries by technological groups, we find
that FDI externalities in the analyzed period
are more likely to occur via horizontal or
backward links, in small scale-intensive
companies with time lags. However, there are
also positive externalities via forward linkages
in scale-intensive, science-based and supplier-
dominated industries.
Negative externalities occur only in basic
metals, metallic products, electrical equipment
and wood products.
Summing-up the externalities by size, we find
that 1% increase in the turnover of foreign
firms increases the TFP of small domestic
firms in nearly 0.71 percentage points and of
large firms in nearly 0.45 percentage points.
Performing the same exercise by period, we
conclude that the magnitude of increases in the
TFP of domestic firms is higher in lagged
periods (0.44 p.p. after 2 year and 0.37 after 1
year of the arrival of the foreign firm),
although the magnitudes are very similar, as
well as the one in the current period (0.34
p.p.). Thus, we confirm that externalities take
time to occur, possibly because domestic firms
need time to absorb the foreign knowledge.
Accordingly, the Portuguese Investment
Promotion Agency (AICEP) should endeavor
to stimulate FDI in scale intensive and
science-based industries. This could be
achieved, in the case of horizontal
externalities, by providing incentives for R&D
cooperation and supporting private sector
training programs. On the other hand, the
government can contribute to the occurrence
of vertical externalities from FDI by
supporting partnerships with foreign firms.
This can be attained by several ways:
providing linkage information in seminars,
Size
Small,
Large
Small
Small
Large
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2023.20.2
Eleonora Santos, Rui Alexandre Castanho, Gualter Couto
E-ISSN: 2224-2899
18
Volume 20, 2023
exhibitions, and missions; sponsoring fairs and
conferences; organizing meetings and visits to
plants; promoting supplier associations; and
providing advice on subcontracting deals.
These results, compared with previous
econometric studies analyzing the
consequences of FDI in Portugal, show that
FDI has a wider range of consequences than
previously assumed. It has been shown in this
study that industries are affected by FDI in
different ways. None of the previous studies
has analyses all the consequences investigated
in this study. First, this study is based on more
recent and previously unexplored datasets, and
we use a large panel of manufacturing firms
which allows us to control for firm fixed
effects and year effects, ruling out main
concerns related to endogeneity. Second, we
are one of the few authors that investigate the
existence of both horizontal and vertical
externalities from FDI in Portugal. Third, we
use lags of the measures of foreign presence to
account for the time lapse required for
externalities to materialize. Fourth, we break
down the results across industries along their
trajectories of technological change which
allow us to uncover some interesting patterns.
Indeed, the technological groups more
positively affected by foreign presence are
scale intensive and science-based industries.
Thus, an important contribution has been made
by providing a more complete picture of the
effects of FDI in Portugal. By and large, the
fact that externalities from FDI are unevenly
distributed across firm sizes and may take up
to 2 years to occur, makes possible to
understand the conflicting results of previous
studies for Portugal. This analysis provides
enough incentive for further research. Avenues
of future research include extending the
research for recent years.
Acknowledgements
This research was financed by National Funds
of the FCT— Portuguese Foundation for
Science and Technology within the project
UIDB/04928/2020 and under the Scientific
Employment Stimulus—Institutional Call
CEECINST/00051/2018. Also, this work is
financed by Portuguese national funds through
FCT—Fundação para a Ciência e a Tecnologia,
I.P., project number UIDB/00685/2020.
Moreover, the project is funded under the
program of the Minister of Science and Higher
Education titled “Regional Initiative of
Excellence” in 2019-2022, project number
018/RID/2018/19, the amount of funding PLN
10 788 423,16”.
References:
[1] Zhou, Xiaoxiao, and Xin Zhao. "Does
diversified environmental regulation make FDI
cleaner and more beneficial to China’s green
growth?." Environmental Science and Pollution
Research 29.3 (2022): 3487-3497.
[2] Zhang, M., & Yang, R. (2022). FDI and
spillovers: New evidence from Malaysia’s
manufacturing sector. Review of Development
Economics.
[3] Keita, O., & Baorong, Y. (2022). Foreign
Direct Investment and Economic Growth
Nexus: Empirical Insight from Guinea. Asian
Economic and Financial Review, 12(1), 58-69.
[4] Wooldridge, J. M. (2009). On estimating firm-
level production functions using proxy
variables to control for
unobservables. Economics letters, 104(3), 112-
114.
[5] Levinsohn, J., & Petrin, A. (2003). Estimating
production functions using inputs to control
for unobservables. The review of economic
studies, 70(2), 317-341.
[6] Olley, G. S., and A. Pakes. 1996. The
dynamics of productivity in the
telecommunications equipment industry.
Econometrica 64: 1263–1297
[7] Arellano, M., & Bover, O. (1995). Another
look at the instrumental variable estimation of
error-components models. Journal of
econometrics, 68(1), 29-51.)
[8] Blundell, R., Bond, S., & Windmeijer, F.
(2001). Estimation in dynamic panel data
models: improving on the performance of the
standard GMM estimator. Emerald Group
Publishing Limited.).
[9] O'Mahony, M., & Van Ark, B. (2003). EU
productivity and competitiveness: an industry
perspective: can Europe resume the catching-
up process? O’Mahony & van Ark eds.
European Commision.
[10] Bogliacino, F., & Pianta, M. (2016). The
Pavitt Taxonomy, revisited: patterns of
innovation in manufacturing and
services. Economia Politica, 33(2), 153-180.
[11] Farinha, L. and Mata J. The impact of Foreign
Direct Investment in the Portuguese Economy.
Banco de Portugal Working Paper.1996: 16-
96.
[12] Proença, I.; Fontoura, M. and Crespo, N.
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2023.20.2
Eleonora Santos, Rui Alexandre Castanho, Gualter Couto
E-ISSN: 2224-2899
19
Volume 20, 2023
Productivity Externalities from Multinational
Corporations in the Portuguese Case: Evidence
from a Short Time Period Panel Data. CEDIN,
CEMAPRE, ISEG/UTL, Universidade
Técnica de Lisboa. Departamento de
Economia Working Paper. 2002
[13] Crespo, N., Fontoura, M. P., & Proença, I.
(2009). FDI spillovers at regional level:
Evidence from Portugal. Papers in Regional
Science, 88(3), 591-607.
[14] Crespo, N., Proença, I., & Fontoura, M.
(2012). The Spatial Dimension in FDI
Externalities: Evidence at the Regional Level
from Portugal. Regional and Sectoral
Economic Studies, Euro-American
Association of Economic Development, 12(1),
115-130.
[15] Proença, we.; Fontoura, M. and Crespo, N.
Productivity Externalities from Multinational
Corporations: Vulnerability to Deficient
Estimation. Applied Econometrics and
International Development. 2006; 6(1): 87–98.
[16] Kar, R. N., Bhasin, N., & Soni, A. (2021).
Role of mergers and acquisitions on corporate
performance: emerging perspectives from
Indian IT sector. Transnational Corporations
Review, 13(3), 307-320.
[17] OECD. FDI in Figures. December 2014
available from www.oecd.org/daf/inv/fdi-in-
figures- dec-2014.pdf.
[18] Hirschman, A. The Strategy of Economic
Development. New Haven, Yale University.
1958. Hsiao, C. Analysis of Panel Data.
Cambridge University Press, Cambridge.
1986.
[19] Pavlínek, P. (2018). Global production
networks, foreign direct investment, and
supplier linkages in the integrated peripheries
of the automotive industry. Economic
Geography, 94(2), 141-165.
[20] Narula, R. (2018). Multinational firms and the
extractive sectors in the 21st century: Can they
drive development?. Journal of World
Business, 53(1), 85-91.
[21] Narula, R., & Pineli, A. (2019). Improving the
developmental impact of multinational
enterprises: policy and research
challenges. Journal of Industrial and Business
Economics, 46(1), 1-24.
[22] Barrios, S. and Strobl, E. Foreign Direct
Investment and Productivity Spillovers:
Evidence from the Spanish Experience.
Weltwirtschaftliches Archiv. 2002;
138(3):459-481.
[23] Lall, S. Vertical Interfirm Linkages in LDCs:
An Empirical Study. Oxford Bulletin of
Economics and Statistics. 1980; 42(3): 203–
226.
[24] Markusen, J. R. and Venables, A.
Multinational Firms and the New Trade
Theory. Journal of International Economics,
Elsevier. 1999; 462: 183–203.
[25] Lin, P. and Saggi, K. Multinational firms,
exclusivity, and the degree of backward
linkages.
[26] Nguyen, P. V., Tran, K. T., Le, N. T. T., &
Trieu, H. D. X. (2020). Examining fdi
spillover effects on productivity growth: firm-
level evidence from vietnam. Journal of
Economic Development, 45(1), 97-121.
[27] Yu, M., Huang, Y., Zhong, H., & Zhang, Q.
(2021). Monopoly and corporate innovation:
evidence from antitrust law. Nankai Business
Review International.
[28] Aghion, P., Howitt, P. A model of growth
through creative destruction, Econometrica.
1992; 60: 323-351.
[29] Jones, C. I. (1995). R & D-based models of
economic growth. Journal of political
Economy, 103(4), 759-784.
[30] Romer, P. Endogenous Technological Change.
Journal of Political Economy. University of
Chicago Press. 1990; 98(5): 71-102.
[31] Duan, Y., Liu, S., Cheng, H., Chin, T., & Luo,
X. (2021). The moderating effect of absorptive
capacity on transnational knowledge spillover
and the innovation quality of high-tech
industries in host countries: Evidence from the
Chinese manufacturing industry. International
Journal of Production Economics, 233,
108019.
[32] Yu, H., Zhang, J., Zhang, M., & Fan, F.
(2021). Cross-national knowledge transfer,
absorptive capacity, and total factor
productivity: the intermediary effect test of
international technology spillover. Technology
Analysis & Strategic Management, 1-16.
[33] Roy, I., & Paul, S. B. (2022). Knowledge
Spillovers and Productivity Growth: Role of
Absorptive Capacity in the Indian
Manufacturing Sector. Journal of Industry,
Competition and Trade, 1-25.
[34] Lam, L. F. L., Law, S. H., Azman-Saini, W. N.
W., & Ismail, N. W. (2022). High Technology
Trade and the Roles of Absorptive Capabilities
in Malaysia. Millennial Asia,
09763996211065565.
[35] Aitken, B. J., & Harrison, A. E. (1999). Do
domestic firms benefit from direct foreign
investment? Evidence from
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2023.20.2
Eleonora Santos, Rui Alexandre Castanho, Gualter Couto
E-ISSN: 2224-2899
20
Volume 20, 2023
Venezuela. American economic review, 89(3),
605-618.
[36] Dimelis, S. and Louri, H. Foreign Direct
Investment and Efficiency Benefits: A
Conditional Quantile Analysis. CEPR
Working Paper 2868. 2001.
[37] Girma, S.; Greenaway, D. and Wakelin, K.
Who Benefits from Foreign Direct Investment
in the UK? Scottish Journal of Political
Economy, Scottish Economic Society. 2001;
48(2): 119-133.
[38] Meyer, K. E., & Sinani, E. (2009). When and
where does foreign direct investment generate
positive spillovers? A meta-analysis. Journal
of international business studies, 40(7), 1075-
1094.
[39] Bucci, A., & Ushchev, P. (2021).
Specialization versus competition: an anatomy
of increasing returns to scale. Journal of
Economic Geography, 21(3), 461-486.
[40] Xie, B. C., Ni, K. K., O'Neill, E., & Li, H. Z.
(2021). The scale effect in China's power grid
sector from the perspective of malmquist total
factor productivity analysis. Utilities
Policy, 69, 101187.
[41] Nelson, R. R., & Phelps, E. S. (1966).
Investment in humans, technological diffusion,
and economic growth. The American
economic review, 56(1/2), 69-75.),
[42] Benhabib, J., & Spiegel, M. M. (2005).
Human capital and technology
diffusion. Handbook of economic growth, 1,
935-966.),
[43] Aghion, P., Howitt, P., & Prantl, S. (2013).
Revisiting the Relationship
between. Advances in economics and
econometrics, 1, 451.
[44] Gerschenkron, A. Economic backwardness in
historical perspective, a book of essays,
Cambridge, Massachusetts: Belknap Press of
Harvard University Press. 1962.
[45] Miao, Y., Salomon, R. M., & Song, J. (2021).
Learning from technologically successful
peers: the convergence of asian laggards to the
technology frontier. Organization
Science, 32(1), 210-232.
[46] Sharma, A., Sharma, R., & Panda, S. (2022).
The role of technological capabilities and gap
in the cross-country patenting: an empirical
investigation. International Economics and
Economic Policy, 19(1), 1-27.
[47] Wang, J. and Blomström, M. Foreign
Investment and Technology Transfer: A
Simple Model. European Economic Review.
1992; 36: 137-155.
[48] Roy, I., & Paul, S. B. (2022). Knowledge
Spillovers and Productivity Growth: Role of
Absorptive Capacity in the Indian
Manufacturing Sector. Journal of Industry,
Competition and Trade, 1-25.
[49] Makieła, K., Wojciechowski, L., & Wach, K.
(2021). Effectiveness of FDI, technological
gap and sectoral level productivity in the
Visegrad Group. Technological and Economic
Development of Economy, 27(1), 149-174.
[50] Ionita, C., & Dinu, E. (2021). The effect of
intangible assets on sustainable growth and
firm value–Evidence on intellectual capital
investment in companies listed on Bucharest
Stock Exchange. Kybernetes.
[51] Lu, Y., Tian, Z., Buitrago, G. A., Gao, S.,
Zhao, Y., & Zhang, S. (2021). Intellectual
capital and firm performance in the context of
venture-capital syndication background in
China. Complexity, 2021.
[52] Zhou, C. (2021). How Does Capital Intensity
Affect the Relationship between Outward FDI
and Productivity? Micro-evidence from
Chinese Manufacturing Firms. Emerging
Markets Finance and Trade, 57(14), 4004-
4019.
[53] Flôres, R., M. Fontoura and Santos, R. Foreign
Direct Investment and Externalities:
Additional Lessons from a Country Study.
Ensaios Económicos da EPGE 455.2002.
[54] Deng, Y., Wu, Y., & Xu, H. (2021). On the
relationship between pollution reduction and
export product quality: Evidence from Chinese
firms. Journal of Environmental
Management, 281, 111883.
[55] Zheng, L., Zhao, Y., Shi, Q., Qian, Z., Wang,
S., & Zhu, J. (2022). Global value chains
participation and carbon emissions embodied
in exports of China: Perspective of firm
heterogeneity. Science of The Total
Environment, 813, 152587.
[56] Wang, S., He, Y., & Song, M. (2021). Global
value chains, technological progress, and
environmental pollution: Inequality towards
developing countries. Journal of
Environmental Management, 277, 110999.
[57] Anand, J., McDermott, G., Mudambi, R., &
Narula, R. (2021). Innovation in and from
emerging economies: New insights and
lessons for international business
research. Journal of International Business
Studies, 52(4), 545-559.
[58] Merlevede, B. and Purice, V. Distance, Time
since Foreign Entry, and Knowledge
Externalities from Foreign Direct Investment,
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2023.20.2
Eleonora Santos, Rui Alexandre Castanho, Gualter Couto
E-ISSN: 2224-2899
21
Volume 20, 2023
Working Papers of Faculty of Economics and
Business Administration, Ghent University,
Belgium 14/896, Ghent University, Faculty of
Economics and Business Administration.
2014.
[59] Hsiao, Cheng [1986]. Analysis of Panel Data.
Cambridge: Cambridge University Press
[60] Nickell, S. Biases in dynamic models with
fixed effects. Econometrica. 1981; 49: 1417-
1426. O’Mahony, M., Van Ark, B. (Eds.) EU
Productivity and Competitiveness: An
Industry Perspective. Can Europe Resume the
Catching-up Process? European Commission,
Luxembourg. 2003.
[61] Roodman, D. A Note on the Theme of Too
Many Instruments, Oxford Bulletin of
Economics and Statistics, Department of
Economics, University of Oxford. 2009; 71(1):
135-158.
[62] Arellano, M. and Bond, S. Some Tests of
Specification for Panel Data: Monte Carlo
Evidence and an Application to Employment
Equations, Review of Economic Studies.
1991; 58(2): 277-97.
[63] Bond, S. Dynamic panel data models: a guide
to micro data methods and practice.
Portuguese Economic Journal. 2002; 1(2):
141–162.
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2023.20.2
Eleonora Santos, Rui Alexandre Castanho, Gualter Couto
E-ISSN: 2224-2899
22
Volume 20, 2023
Contribution of Individual Authors to the
Creation of a Scientific Article (Ghostwriting
Policy)
The authors equally contributed in the present
research, at all stages from the formulation of the
problem to the final findings and solution.
Sources of Funding for Research Presented in a
Scientific Article or Scientific Article Itself
Conflict of Interest
The authors have no conflicts of interest to declare
that are relevant to the content of this article.
Creative Commons Attribution License 4.0
(Attribution 4.0 International, CC BY 4.0)
This article is published under the terms of the
Creative Commons Attribution License 4.0
https://creativecommons.org/licenses/by/4.0/deed.en
_US
This research was financed by National Funds
of the FCT— Portuguese Foundation for
Science and Technology within the project
UIDB/04928/2020 and under the Scientific
Employment Stimulus—Institutional Call
CEECINST/00051/2018. Also, this work is
financed by Portuguese national funds through
FCT—Fundação para a Ciência e a Tecnologia,
I.P., project number UIDB/00685/2020.
Moreover, the project is funded under the
program of the Minister of Science and Higher
Education titled “Regional Initiative of
Excellence” in 2019-2022, project number
018/RID/2018/19, the amount of funding PLN
10 788 423,16”.
