Trade openness and wage inequality: Case of Tunisia
BEN YOUNES HAMZA
Economics
Faculty of Economic Sciences and Management of Tunis, University of Tunis-El Manar
Tunis El Manar
TUNISIA
benyounes.hamza@gmail.com
(Corresponding author)
BEN AMEUR FAYÇAL
Economics
Faculty of Economic Sciences and Management of Tunis, University of Tunis-El Manar
Tunis El Manar
TUNISIA
Abstract: - The 90's have been marked by an increasing globalization which has revealed two major trends, if
on the one hand there has been a more pronounced opening up, especially in the ranks of emerging countries,
on the other hand there has been a considerable rise in inequalities.
Despite an abundant literature on the link between trade openness and wage inequality, the latter remains
relatively ambiguous, particularly in the absence of a consensus, especially for the developing countries.
This article therefore focuses on the impact of trade openness on wage inequality, particularly between skilled
and unskilled workers for the case of Tunisia.
Our analysis took into account a number of factors that influence this relationship, such as labor market
fluctuations, technological transfer and the effect of institutions.
Our contribution to this work is that unlike the majority of work conducted on the Tunisian case, our analysis
has not been limited to the manufacturing industry but we have extended it to the services sector and the whole
economy by including the non-manufacturing sector in order to provide a comparative analysis between these
different sectors
The exploitation of the estimation results over the period 1990 to 2020 shows that, in general, openness has
contributed to the increase in wage inequalities in Tunisia.
Key-Words: - Trade openness, Wage inequality, employment, Panel data, Tunisia
Received: August 29, 2022. Revised: May 21, 2023. Accepted: June 23, 2023. Published: July 12, 2023.
1 Introduction
As a major force of the 20th century, trade openness
has created a new kind of interactions between
nations, economies, and individuals, and has
significantly increased trade across borders at
multiple levels.
It has also contributed to the fragmentation of
production processes, labor markets and political
entities.
However, despite the innovation, dynamism, and
positive spillover effects of trade openness, it is
likely to have some disruptive aspects, which is why
there has been an intense debate between its
supporters and opponents regarding its effects,
especially among the most disadvantaged
populations. While some perceive it as an effective
solution to reduce poverty through the surplus
growth it generates, others argue that it only further
accentuates wealth gaps without generating an
overall increase in economic activity, thus leading to
the continued marginalization of the poorest
populations.
However, the debate has been somewhat
modified in recent years, with a consensus that trade
openness is seen as an important catalyst for growth.
Thus, the classic question of the effects of such
openness on economic growth has given way to a
much more worrying question, namely the
INTERNATIONAL JOURNAL OF APPLIED MATHEMATICS,
COMPUTATIONAL SCIENCE AND SYSTEMS ENGINEERING
DOI: 10.37394/232026.2023.5.7
Ben Younes Hamza, Ben Ameur Fayçal
E-ISSN: 2766-9823
72
Volume 5, 2023
distribution of its gains, given the considerable rise
in internal inequalities at the global level, which
nowadays constitute a phenomenon that affects
almost all countries.
Opponents of this external openness most often
cite the examples of East Asian countries such as
China and India, or those of Latin America, which
are countries that have implemented a pronounced
liberalization process. Despite a surge in their
growth rates, their openness has gone hand in hand
with a sharp rise in inequalities, affecting primarily
the poorest individuals and placing these countries
at the top of the international rankings concerning
inequalities.
Thus, the most salient international dilemma
today is how to take advantage of the supposed
gains from trade openness while reducing its effects
on disparities. Developing countries seem to be the
perfect sample for illustrating this situation,
countries for which inequality has become a major
concern and a political priority and issue.
Therefore, our research problem is to understand
how trade openness affects inequality, particularly
wage inequality, in Tunisia, a developing country
that has long suffered from these inequalities.
2 Literature review:
The first thoughts were those of the Heckscher-
Ohlin-Samuelson (HOS) final goods trade model,
between countries with different factor endowments.
By referring to this model, trade opening that leads
to greater specialization should have positive
incidence on developing countries. A more
simplified version of the HOS model suggests that
each country will specialize and export production
using the relatively abundant factor. According to
the Stolper-Samuelson theorem, trade liberalization
will lead to an adjustment in relative wages. In fact,
when a developing economy whose abundant factor
is unskilled labor begins to open up more to the
international trade, the demand for low-skilled labor
will increase, the wages of these workers will tend
to follow suit. On the other hand, wages for skilled
workers will be lowered, leading to a reduction in
the wage gap.
Seen in this light, openness should naturally
bring more equality and benefit the poorest
population.
However, several studies have subsequently
criticized the HOS model, pointing out that the
situation is much more complex than the
assumptions indicated.
Indeed, the HOS model has been criticized for
not taking into account elements such as
technological progress, which, according to
Attanasio et al (2004) and Goldberg and Pavcnik
(2004), is likely to contribute to the increase in
disparities by requiring more and more qualification
of the workforce.
The impact of new technologies, information and
communication is often considered to be uni-
directional, i.e. in favor of skilled workers who are
better able to master and integrate these skills into
their activities, while conversely leading to a
reduction in the need for low-skilled labor.
Similarly, increased international competition
has led even developing countries to become more
interested in higher-skilled products.
On the other hand, the HOS model does not take
into account the international mobility of the
production factors. According to Lee and Vivarrelli
(2006), the capital mobility to developing countries
can result in increased inequality.
Deardorff (2001) finds that the existence of
transport costs as well as customs duties can be an
obstruction to the equalization of factor
remuneration.
Wood (2009) also shows that some of the HOS
model's claims do not match up with reality.
For example, technological similarities between
nations and the immobile production factor
highlight these irregularities. In this case, it is
important to note that one of the main characteristics
of openness is the capital mobility. Which coincides
with the findings of Feenstra and Hanson (1996),
Wang and al (1992) or Zhu and al (2001) who note
that the HOS model does not take into account
factor mobility at the international level, while in
fact the inflow of FDI can be seen as a driver of
inequality.
In this same context, foreign direct investment
(FDI) does not escape the accusation of being
responsible of rising inequalities.
MacDonald and Majeed (2011) in studying a
panel of 65 developing countries between 1970 and
2008 to highlight the relationship between openness,
inequality and poverty found no significant link
between openness and inequality. However, they
clearly detect a positive link between FDI and
increased disparities, concluding that globalization
has had a negative effect on wage distribution, this
coincides with the findings of David (2011) who
examined the case of Brazil, or Wu and Hsu (2012)
for whom FDI is the main cause of rising inequality
in China.
Lim and McNelis (2014) also conclude that FDI
helps promoting economic growth, however, it
generates an increase in wage inequality. They find
that this effect is more amplified in middle-income
countries.
INTERNATIONAL JOURNAL OF APPLIED MATHEMATICS,
COMPUTATIONAL SCIENCE AND SYSTEMS ENGINEERING
DOI: 10.37394/232026.2023.5.7
Ben Younes Hamza, Ben Ameur Fayçal
E-ISSN: 2766-9823
73
Volume 5, 2023
We can therefore observe that today both
developed and developing countries suffer from
wage inequalities that largely affect the poorest
populations, who are the most vulnerable and the
least able to protect themselves against the hazards
of trade openness.
Indeed, as far as developing countries are
concerned, several empirical studies have
contradicted the HOS model or Stopler-Samuelson
theory predictions. Many of these studies have
concluded that international trade has increased the
demand for skilled labor at the expense of unskilled
labor in these countries.
Shah and Whalley (1991) found that after
opening up, developing countries were unable to
redistribute income effectively.
Robbins and Gindling (1999) emphasize that
demand for skilled labor has increased following
trade openness. Looking at the Chilean case, they
find that the demand for skilled labor is higher for
imports than for exports. Similar observations are
made in the case of Colombia, where they find that
technology transfer and the import of equipment and
new technologies have contributed to the increase in
demand for skilled labor in the country at the
expense of unskilled ones.
Hanson and Harrison (1999) reach similar
conclusions for Mexico. Indeed, all reforms aimed
at greater openness have resulted in an increase in
the wage gap.
Lustig (1998), who also looked at Mexico, found
that the wages of less skilled workers fell by about
25 percent, while the wages of skilled ones rose by
10 percent.
Gorg and Strobl (2002) and Attansio et al (2004),
focusing on the cases of Ghana and Colombia
respectively, have also concluded that international
trade has increased the need for skilled labor as well
as increasing its relative wage to the detriment of
unskilled labor.
Cornia (2003) studying the case of 73 countries
between 1980 and 2000 has observed that wage
inequalities have increased significantly, especially
in countries characterized by a strong deregulation
and external liberalization policy.
Arbache, Dickerson and Green (2004), focusing
on the case of Brazil, were able to show that trade
openness led to a decline in wages of about 16 per
cent in sectors exposed to foreign competition,
while the decline was only 8 per cent in protected
sectors employing mainly skilled workers.
Kahai and Simmons (2005), who used the GINI
index as a measure of wage inequality to study its
interaction with trade openness, concluded that trade
openness succeeds in increasing inequality in
developing countries.
Aradhyula and al (2007) come to the same
findings, namely that trade openness had led to
rising inequalities and that its effect was mostly felt
in the developing countries.
Acar and Dogruel (2012), with a focus on the
MENA region, concluded that openness improves
women's participation in the labor market, which
weighs heavily. It should be noted that in many of
these countries, women receive lower wages than
men.
Bucciferro (2010) who was interested in Latin
American countries as well as Castilho and al
(2012) for the case of Brazil also agreed that trade
opening had succeeded in reducing inequalities.
This is consistent with the results of Yilmaz
Bayar et al. (2017), studying the case of 11 Latin
American countries for which, over the long run, the
combination of trade openness and financial sector
development succeeded in reducing all inequality
and poverty.
For Amerlia Santos-Paulino (2012) who
analyzed the effect of openness on poverty and
inequality and focusing on the case of developing
countries, she was able to highlight the fact that
despite an increase in inequality, the latter results in
a considerable increase in economic growth, which
decreases poverty levels and improves the situation
of the poorest.
Mah (2013) found that international trade also
led to wage inequality in the case of China. The
same goes for Kylie Tank You et al (2015), in their
study of a panel of 6 South American countries
between 1995 and 2012.
Developed countries are not immune to the trend
of rising inequality either.
Borjas (1991), analyzing the case of the United
States, detected that international trade, along with
the wave of immigration, participated in the
decrease of about 25% of the demand for the least
skilled workers. Sachs and Shatz (1994), who also
looked at the United States in the 1970s and 1980s,
showed that there was a considerable rise in
inequality that coincided with the increase in trading
with developing countries.
The same is true of Giraud (1998), who found
that competing with low-wage countries increased
wage inequality by about 20%.
Beyer and al (1999) found that global trade
increased wage inequalities by 15%.
Wood (1995) came to the same conclusions,
pointing out that the growing inequality in Northern
countries is the result of competition with Southern
INTERNATIONAL JOURNAL OF APPLIED MATHEMATICS,
COMPUTATIONAL SCIENCE AND SYSTEMS ENGINEERING
DOI: 10.37394/232026.2023.5.7
Ben Younes Hamza, Ben Ameur Fayçal
E-ISSN: 2766-9823
74
Volume 5, 2023
countries that are characterized by relatively low
wage levels.
Indeed, Leamer (1994) has established that the
increase in intermediate manufactured goods
imports from the South has largely contributed to
the increase in wage inequalities in the North.
Cribb (2013) also explained that the rise in
inequality in the UK was a result of trade openness.
The country has been tempted to import intensive
unskilled labor products from low-wage countries,
which would cost less. However, this has led to a
decline in the demand for unskilled labour within
the country, further fuelling the wage gap.
Baldwin and al (1999), consider international
trade and new technologies as factors that aggravate
wage inequalities.
For Zakilwal (2000), greater integration in
international trade leads to an increase in the wage
gap between the higher and lower skilled workers.
Townsend (2007), who studied the impact of
trade agreements on the evolution of wages in
Canada, demonstrated that a 1% decrease in
customs barriers led to a 0.4% decrease in wages
and that this decrease was felt even more among the
least qualified workers, thus leading to greater wage
inequalities.
Faustino and Vali (2011) by regressing the level
of inequality by a certain number of variables such
as openness, FDI, GDP per capita, the
unemployment rate, the inflation rate as well as the
number of national listed companies in the case of
24 OECD countries between 1997 and 2007, came
to the conclusion that trade openness and wage
inequalities were going in two different directions
while, conversely, inflation and unemployment
positively affected inequality.
Sarah Polpibulaya (2015) concluded in a study of
86 developed and developing countries that a 1%
increase in openness led to a 2.12% increase in
inequalities.
3 Methodology
We will use panel data econometrics to estimate our
model.
This method has the particularity of providing us
with an important source of information, both on the
individual and temporal dimensions of the studied
subjects (Trognon 2003).
The use of this technique relies on primordial tests
to ensure the validity of the model, namely the
verification of the homoscedasticity of the error
terms but also the absence of autocorrelation.
For our estimations, we will use the two-stage
double least squares method, introduced in 1957 by
Robert Basmann and then in 1961 by Henri Theil, in
response to the endogeneity problem posed by one
of our explanatory variables, namely the one
representing the labor demand.
This method, known as instrumental, is carried out
in two stages. On the one hand, we proceed to the
substitution of the endogenous explanatory variables
by variables that will be their representatives, called
instruments.
Then, in a second step, we proceed to the estimation
by ordinary least squares (OLS) of the instrumental
variables and of the exogenous variables initially
presented in the basic equation, thus giving more
relevant and robust estimates.
3.1 Model Specification:
Several authors have used a multiple modeling
approach to apprehend the effect of trade openness
on employment and subsequently on wages in order
to better understand its effects on wage inequality.
One of the most prevalent models is the one
developed by Milner and Wright in 1998, whose
starting point is a Cobb-Douglas production
function. However, authors such as Katz and
Murphy (1992) have based their model on a
production function with a constant elasticity of
substitution called the production function (CES),
which takes into account two labor factors, namely
skilled and unskilled labor.
In our case, we were inspired by the
contributions of authors such as Cortes and Jean
(1997) or Katz and Autor (1999) to define an
equation in the following form:
𝑙𝑛(𝑦)𝑖𝑡 = 𝛽0+ 𝛽1𝑙𝑛𝑑𝑖𝑡 + 𝛽3𝑋𝑖𝑡 + 𝛿𝑖+ 𝜉𝑖𝑡 (1)
y_it: Refers to the relative wage, which is the
ratio of the average annual wage of skilled to
unskilled workers used as a measure of wage
inequality. d_it: is the labor demand.
X is a vector that includes all the variables
explaining the wage gap. Authors such as Katz and
Autor (2008) have used variables such as the
minimum wage (s) to take into account the effects
of institutions, the openness rate (op) as a variable
representing trade openness, unemployment rate
(ch) and growth rate (gr) to take into account
cyclical fluctuations in the labor market.
However, other authors, such as Cortes and Jean
(1997), have added another variable, namely capital
intensity (k), which is the ratio between the capital
stock and skilled employment.
The capital intensity focuses on the degree of
complementarity between capital and skills, such as
the higher the capital-labor ratio, the higher the level
of skills.
INTERNATIONAL JOURNAL OF APPLIED MATHEMATICS,
COMPUTATIONAL SCIENCE AND SYSTEMS ENGINEERING
DOI: 10.37394/232026.2023.5.7
Ben Younes Hamza, Ben Ameur Fayçal
E-ISSN: 2766-9823
75
Volume 5, 2023
We will also add the variable (g) introduced by
Acemoglu and al (2001), which provides a measure
of the way policies and institutions support their
economies, it reveals the institutional quality within
a country.
3.2 Sample presentation:
Our sample will consist first of six sectors of the
Tunisian manufacturing industry that are the food
industry, the industry of building materials and
glass, the mechanical and electrical industry, the
chemical industry, the industry of textile, clothing
and leather and the diverse manufacturing industry.
Then we will consider the service sector, which is
composed of sectors such as transport and
communication, banking and insurance, hotels,
coffee shops and bars, other commercial services,
and finally the trade sector. Thirdly, we are
interested in the whole economy by grouping
together all the sectors mentioned, while also adding
sectors belonging to the non-manufacturing
industry, such as the mining and the hydrocarbons
sectors. Our analysis period will be from 1990 to
2020.
We focused first on the Tunisian manufacturing
industry, which was among the sectors that are the
most exposed to this trade opening and which
suffered from innumerable repercussions, especially
in terms of job demand as well as in terms of wages,
this sector also constituted for a long time the main
pillar of the Tunisian economy. Subsequently, we
added to our sample a set of sectors belonging to the
non-manufacturing industry, and particularly the
services sectors, because it should be noted that
today those sectors are the driving force of the
Tunisian economy, with a 59% contribution to the
GDP and about 62%
1
contribution to the
employment rate of the active population.
We also chose the period from 1990 to 2020, as
far as it represents the starting point for a process of
trade liberalization that is constantly progressing.
Indeed, such a process was triggered by the
adoption of the SAP towards the end of the 1980s,
and then it was accentuated with the accession to the
GATT in 1990 and to the WTO in 1995. Trade
liberalization deepened with the signing of the free
trade area agreement with the European Economic
Union in 1995 and its final entry into force in 2008,
expressing the common wish of both sides to
strengthen their cooperation ties and establish
harmonious and sustainable relations, but also
conditions favorable to the development and
1
The unfinished revolution chap8, Tunisia report, World Bank.
diversification of trade and which prioritize
Tunisia's integration objectives and assist it to this
end.
This period also includes the democratic
transition that followed the events of 2011 and
which were at the origin of a real upheaval not only
at the political and economic level but also at the
social level and whose effects are felt to this day
mainly in terms of perpetuation of the crisis,
aggravation of the precariousness of the most
deprived strata and especially the increase in
unemployment of the most qualified workers.
3.3 Variable description:
Table 1 Variables presentation
Variables
Description
Dependent variable
Wage Gap (y)
The ratio of the average
annual wage of skilled
workers to that of unskilled
workers as a measure of
wage inequality
Explanatory variables
The openness rate (op)
The sum of exports and
imports as a percentage of
GDP by sector
Labor demand (d)
Total employment by sector
Capital intensity (k)
The ratio between capital
stock and employment by
sector
Growth rate (gr)
The change in GDP from
one year to the next
Unemployment rate (ch)
The percentage of
individuals in the labor force
who are unemployed
The minimum wage (s)
The guaranteed
interprofessional minimum
wage
The governance (g)
The Index of Economic
Freedom established by the
Canadian Fraser Institute
Instrumental variable
Average education’s years
(nme)
Average education’s years
by sector
INTERNATIONAL JOURNAL OF APPLIED MATHEMATICS,
COMPUTATIONAL SCIENCE AND SYSTEMS ENGINEERING
DOI: 10.37394/232026.2023.5.7
Ben Younes Hamza, Ben Ameur Fayçal
E-ISSN: 2766-9823
76
Volume 5, 2023
3.4 Descriptive statistics:
Table 2 Descriptive statistics
Mea
n
Stan
dard
devi
ation
Min
Max
Skew
ness
Kurt
osis
50.9
52
42.7
30
2.98
3
226.
066
1.595
241
5.52
487
7
1.09
e+05
1.06
e+05
4796
.000
4.73
e+05
1.257
381
3.94
369
9
2647
.397
999.
966
1320
.480
4743
.936
.6739
23
2.42
317
7
0.87
1
1.12
2
0.02
0
10.4
80
3.623
922
24.5
519
6
7.57
9
1.88
3
3.21
9
13.4
55
.7659
134
3.61
862
8
0.15
1
0.01
4
0.12
4
0.18
3
-
.2118
595
2.92
809
5
0.02
2
0.04
8
0.00
1
0.41
8
.0022
834
3.92
902
1
0.03
8
0.02
1
0.00
4
0.07
9
.3393
945
2.26
283
7
5.91
3
0.58
7
4.88
0
6.49
0
-
.7795
912
1.87
571
1
Source: Author calculation using STATA software.
Table 2 transcribes the descriptive statistics,
which report indicators of position, dispersion and
form in order to verify empirically the symmetry,
the flatness, the normality, the dispersion and the
precision of the information provided by our
model’s variables.
3.5 Correlation matrix:
Table 3 Correlation matrix
d
o
k
ch
gr
g
s
nm
e
d
1.0
0
op
-
0.4
8
1.0
0
k
0.3
8
-
0.2
1.0
0
4
ch
0.0
1
0.0
2
-
0.0
1
1.0
0
gr
-
0.2
8
0.0
9
-
0.1
4
-
0.4
3
1.0
0
g
0.4
0
-
0.1
9
0.2
4
-
0.4
0
-
0.2
8
1.0
0
s
0.0
2
-
0.0
1
-
0.0
1
0.1
2
-
0.0
9
-
0.1
1
1.0
0
nm
e
0.9
4
-
0.3
4
0.2
9
-
0.0
3
-
0.3
0
0.4
8
0.0
2
1.0
0
Source: Author calculation using STATA software.
Table 3 represents the correlation matrix between
our different explanatory variables. We note that
apart a strong correlation between the endogenous
variable and the instrument used, which is already
one of the conditions necessary for the use of the
instrumental method, the correlation coefficient is
well below 0.8 for the other variables, which is, the
limit set by Kennedy (2008), from which there is
reason to be concerned about a possible problem of
multicollinearity.
4 Findings discussion
Table 4 Estimation results using the DMC
method
VARIABLE
S
(1) DMC
(Manufa
cturing
industry)
(2) DMC
(Services
sector)
(3)
DMC
(Whole
econom
y)
y
y
y
d
-
0.554***
0.286***
0.360**
*
(0.0745)
(0.0564)
(0.0531)
op
0.385***
0.558***
0.372**
*
(0.0520)
(0.0377)
(0.0260)
k
0.680***
0.327***
0.517**
*
(0.0830)
(0.0254)
(0.0250)
gr
0.0748
0.159**
0.139**
(0.0580)
(0.0639)
(0.0586)
ch
-
1.203***
-0.824*
-
1.393**
*
(0.447)
(0.488)
(0.449)
INTERNATIONAL JOURNAL OF APPLIED MATHEMATICS,
COMPUTATIONAL SCIENCE AND SYSTEMS ENGINEERING
DOI: 10.37394/232026.2023.5.7
Ben Younes Hamza, Ben Ameur Fayçal
E-ISSN: 2766-9823
77
Volume 5, 2023
g
-
2.320***
-
2.645***
-
4.004**
*
(0.480)
(0.467)
(0.425)
s
-0.354
-2.538*
-1.524
(1.359)
(1.486)
(1.367)
Constant
8.017***
8.424***
10.05**
*
(0.915)
(0.934)
(0.840)
Observation
s
180
150
390
R-squared
0.557
0.844
0.515
sectors
6
5
13
Prob (f-
statistic)
0.000000
0.000000
0.00000
0
Sargan
hansen
0.36450
0.89143
0.18632
Endogeneity
test
0.0003
0.0000
0.0000
Robustness
test
0.0000
0.0000
0.0000
(***), (**) et (*) corresponding to statistical
significance at 1%, 5% and 10% respectively.
Source: Author calculation using STATA software.
Our econometric analysis allowed us to
understand the effect of all the variables used on the
wage gap between skilled and unskilled workers,
first at the level of the Tunisian manufacturing
industry, then at the level of the services sector and
finally by taking into account the whole economy
including the non-manufacturing sectors in order to
carry out a comparative study between these
different sectors.
First, we have defined a set of specifications in
response to the econometric techniques we
employed, allowing us to capture the effect of all the
variables employed on the wage gap between skilled
and unskilled workers.
We used the double least squares method to
estimate our model in response to the endogeneity
problem associated with the variable representing
labor demand.
As an instrumental variable, we used the average
number of education years per sector as a proxy.
We chose this variable in particular because it
affects wage differences and inequality through
labor demand.
Indeed, the level of education allows us to
classify individuals into different categories
according to their qualification levels. These
qualifications have a major influence on wage due
to the fact that salaries are proportional to the
educational level and can therefore affect the
income gap between skilled and unskilled workers.
In turn, the demand for employment reflects the
need for skills. Thus, the average number of years of
education affects wage differentials through job
demand.
Concerning our regressions, it appears that the
effect of most of the explanatory variables was
statistically significant.
We can see that the effect of the trade openness
rate was positive and statistically significant at the
1% level on the wage ratio in our three regressions.
This result can be explained by the fact that the
opening has brought its share of upheavals in the
Tunisian economy with the emergence of a number
of sectors requiring more qualifications including
the services sector, which is now the driving force
of the Tunisian economy with an increased
participation in the national GDP. A sector that also
employs more than 62%
2
of the workforce,
including for example sectors such as banking and
insurance that use more than 76%
3
of skilled
workers, or the trade, education, health and
administration sectors, which employ more than
60%
4
of the workforce, sectors that constantly
require more qualifications and skills.
It also appears that the highest salaries belong to
sectors such as banking, followed by the transport
and communication and the extractive industries,
particularly hydrocarbons. These sectors have seen
salary increases of 8.7%, 6.73% and 6%
respectively. Conversely, regarding the least
remunerative sectors, we find at the top of the list
sectors belonging to the manufacturing industry
such as textiles, clothing and leather, various
manufacturing industries and building materials and
glass whose wage increases do not exceed 3%
5
.
As for the effect of the variable representing
labor demand, it was found to be significant and
negative at the 1% level for the manufacturing
industry.
The observation is that job creation in this sector
tends to respond to the needs of job seekers with
lower education’s level.
This last finding is in line with the evidence
provided by the Tunisian statistical agency and
2
The unfinished revolution chap8, Tunisia report, World Bank
3
Author's calculation based on TICQS and NSI “NATIONAL
STATISTICS INSTITUTE” statistics from employment surveys
4
Note and analysis of TICQS No. 44 - 2016 Assessment of labour
market evolution in Tunisia :
2006-2015.
5
International Labour Office (ILO): Wage Structure Survey, Tunisia
2011
INTERNATIONAL JOURNAL OF APPLIED MATHEMATICS,
COMPUTATIONAL SCIENCE AND SYSTEMS ENGINEERING
DOI: 10.37394/232026.2023.5.7
Ben Younes Hamza, Ben Ameur Fayçal
E-ISSN: 2766-9823
78
Volume 5, 2023
various studies issued by the Tunisian Institute of
Competitiveness and Quantitative Studies (TICQS)
which have confirmed the fact that the labor
demand, following the initiation of the various
processes supporting the trade opening especially in
the Tunisian manufacturing industry, has not met
expectations and predictions with a demand for
quality work quite low, the latter has concerned
much more the unskilled workforce. Indeed, as an
example, the percentage of employees with no
higher education degree in the manufacturing
industry has reached 66.6%
6
while the percentage
of senior technicians and engineers is only 5.3%
7
with sectors such as textiles, clothing and leather
which employs about 90%
8
of unskilled labor.
These results are in line with the findings of
Munshi (2012), who looked at the manufacturing
industry in Bangladesh, and Chaudhry and Imran
(2013) for the case of Pakistan. Indeed, these
authors concluded that trade openness had increased
the need for unskilled labor and subsequently
contributed to the increase in their wages, thus
reducing the wage gap between unskilled and
skilled workers.
However, this trend is reversed when considering
the Tunisian economy as a whole or the services
sector. In fact, it is the latter that requires the most
skilled labor and is among the highest paying
sectors.
It must be noted that in Tunisia, salary levels
highly depend on qualifications and the sectors in
question. The observation of incomes shows that the
net salary increases considerably with the higher
educational level, in fact, the net base salary is
almost quadrupled between a non-graduate and a
higher graduated employee. This difference is even
multiplied by 5 if you take into account salary
supplements such as bonuses and benefits in kind
9
.
Concerning the effect of the variable
representing capital intensity, it was positive and
statistically significant at the 1% level, the latter
went in favor of the remuneration of the most
qualified workers, thus contributing to the wage gap
between the two groups of workers. This is largely
related to the technological development bias in
6
International Labour Office (ILO): Wage Structure Survey, Tunisia
2011
7
Employment survey 2013 NSI
8
Author's calculation based on TICQS data. Author's calculation based
on NSI statistics from employment surveys
9
Study of real wage trends in Tunisia before and after the revolution:
2005-2015.
favor of the most qualified workers who are better
able to master it and integrate it into their activities.
Thus, the higher the rate of capital accumulation,
the greater the demand for skilled workers.
Similarly, number of authors such as Cohen and
Levinthal (1989), Borjas and al (1994), Fuchs and
Perina (1987) and Pissarides (1997) or Jarmotte and
al (2013), have demonstrated that technology
transfer in developing countries is biased in favor of
the most skilled workers and has a destructive effect
on unskilled labor, in addition to increasing wage
inequalities.
Goldberg and Pavcnik (2004) even consider this
technology bias as an endogenous response to trade
liberalization.
As for the variable representing the rate of
economic growth, it had a positive and statically
significant impact at the 1% threshold, however, this
effect was found to be insignificant in the case of
the manufacturing industry.
This can be explained by the fact that although
there is a consensus at the global level supporting
the thesis of an economic growth generating jobs.
For the Tunisian case, this work creation has much
more concerned the service sectors which are
asserting themselves today as being the most
dynamic ones on the job market. This growth has
also increased the need for qualification in a great
number of sectors, mostly belonging to the services
such as the financial sector, the communication
sector or other various services. However, the
weakness of the Tunisian economy lies in the fact
that the creation of quality jobs remains insufficient,
especially in the manufacturing industry.
Indeed, despite the progress made in education,
health, as well as the fight against poverty, with the
objective of creating a sustainable and sustained
growth. Today, the country is experiencing many
difficulties to achieve this goal, especially in the
current post-revolutionary context, with an
economic growth that does not manage to take off
and reach the expected levels.
For its part, the minimum wage has a negative
effect, but it was statistically insignificant in most of
our regressions. It must be pointed out that the
minimum wage indexed to the evolution of prices, is
considerably behind other salaries, for example,
between 1997 and 2012, it increased by only 0.5%
10
.
As for the effect of the unemployment rate, it
turns out to be negative and statistically significant
at the threshold of 1%, the latter leads to reducing
the wage gap between skilled and unskilled workers.
10
TICQS Notes and Analysis No. 13: Wage growth and productivity
INTERNATIONAL JOURNAL OF APPLIED MATHEMATICS,
COMPUTATIONAL SCIENCE AND SYSTEMS ENGINEERING
DOI: 10.37394/232026.2023.5.7
Ben Younes Hamza, Ben Ameur Fayçal
E-ISSN: 2766-9823
79
Volume 5, 2023
This counterintuitive effect is further evidence of
the situation in Tunisia characterized by a
disconnection and mismatch between the labor
market and the educational system with a difficult
insertion of young graduates in the professional
sphere. It is also them who suffer the most from
unemployment, as shown by the rate of unemployed
graduates of higher education which reached in
2020 30%
11
which is double the national average of
15%.
It should be emphasized that the Tunisian labor
market is facing a low creation of quality jobs
especially in the manufacturing industry. Jobs that
concern largely unskilled work with an economy
that keeps facing difficulties and a number of jobs
not exceeding 5%
12
of the total number of
applicants.
Concerning the effect of the governance variable,
it was negative and significant.
Regarding this theme, Tunisia has a lot of efforts
and a long way to go especially in relation with the
institutional quality. Despite the fact that the latter
has gained 10 positions in the world ranking in 2021
standing 119th
13
out of 162 countries, Tunisia is
still at the bottom of the table at the international
level and the statement is not very bright at the level
of the Arab world where Tunisia is ranked 9th out of
12
14
countries.
Acemoglu and al (2001, 2002) consider that
institutional quality is an important factor in
economic performance since the proper functioning
of the market is largely conditioned by the quality of
the institutions. In the context of market activities
involving a large number of agents and institutions,
the role of good governance is to reduce information
asymmetries and the various risks by ensuring that
property rights and contracts are respected and also
to clarify the responsibilities and limits of action of
each party.
5 Conclusion
The 1990s marked a global context where
economies are increasingly integrated and in
constant interaction through the acceleration and
11
NSI data on unemployment among college graduates 2020.
12
Inadéquation des qualifications en Tunisie : quels sont les
déterminants du sous-emploi ? TICQS 2019.
13
https://www.fraserinstitute.org/economic
freedom/dataset?geozone=world&page=dataset&min-year=2&max-
year=0&filter=1&countries=TUN
14
TICQS Economic Freedom Index EFI 2021.
multiplication of trade. These years also saw an
abundance in the studies focusing on the
relationship between openness to foreign trade and
incomes inequality. The majority of these studies
were primarily dedicated to emerging South Asian
and Latin American countries before there were a
few ones covering the Mena countries, including
Tunisia.
Indeed, the latter has long been a highly
integrated country on the international scene,
notably among the most integrated North African
countries. However, the Tunisian state has
accumulated years of maldevelopment by favoring
growth and wealth creation to the detriment of
redistribution and equity, resulting in a significant
divide within the society, as well as a level of
inequality among the highest in the world.
With regard to the effect of trade openness on
wage inequality, our results run against the
theoretical predictions of neoclassicals that trade
openness would reduce the wage gap between
skilled and unskilled workers in developing
countries,
In our case, we found that, on the contrary, it
exacerbated the inequalities in question. This is in
line with a number of studies that have focused on
developing countries, including Santos-Paulino
(2012), Atif and al (2012), Mah (2013), Kylie Tank
You and al (2015) or other studies on the Tunisian
case such as Ghazali (2009) or Mrabet (2010) whose
conclusions were in favor of an increase in wage
inequality following the trade opening.
Finally, this growing openness to international
trade has long been perceived as easier access to
capital and to new technologies, thus increasing the
demand for skilled labor and enabling the
absorption of the excess number of young graduates
who come to the labor market year after year.
However, this phenomenon did not materialize at
the level of all the sectors, especially within the
manufacturing industry, in which there was not the
expected spillover effect of new technologies and
the development of production systems, with a
certain number of sectors that remain rather
rudimentary and an industry where the demand for
unskilled labor still prevails.
However, this trend is reversed in the service
sector, where demand and needs for labor are more
oriented towards skilled workers. This sector also
has higher wage levels than the manufacturing
industry, thus further contributing to the wage gap
between skilled and unskilled workers.
In the end, our study also confirms that the link
between openness and wage inequality may be the
result of factors and phenomena, such as the
INTERNATIONAL JOURNAL OF APPLIED MATHEMATICS,
COMPUTATIONAL SCIENCE AND SYSTEMS ENGINEERING
DOI: 10.37394/232026.2023.5.7
Ben Younes Hamza, Ben Ameur Fayçal
E-ISSN: 2766-9823
80
Volume 5, 2023
emergence of a certain number of sectors requiring
more qualifications, a technological transfer biased
in favor of the most qualified workers, or the effects
and quality of institutions, as well as the
characteristics and fluctuations of the labor market
within the country itself.
References:
[1] Acar, S., & Dogruel, F. (2012). Sources of
inequality in selected MENA countries.
Structural Change and Economic Dynamics,
23(3), 276-285.
[2] Acemoglu, D., Johnson, S., & Robinson, J. A.
(2005). Institutions as a fundamental cause of
long-run growth. Handbook of economic
growth, 1, 385-472.
[3] Cornia, G. A. (2003). The impact of
liberalisation and globalisation on income
inequality in developing and transitional
economies. Available at SSRN 380347.
[4] Aradhyula, S. V., Rahman, T., & Seenivasan,
K. (2007). Impact of international trade on
income and income inequality (No. 381-2016-
22241).
[5] Arbache, J. S., Dickerson, A., & Green, F.
(2004). Trade liberalisation and wages in
developing countries. The Economic Journal,
114(493), F73-F96.
[6] Atif, S. M., Srivastav, M., Sauytbekova, M., &
Arachchige, U. K. (2012). Globalization and
income inequality: a panel data analysis of 68
countries. Munich Personal RePec Archive,
42385.
[7] Attanasio, O., Goldberg, P. K., & Pavcnik, N.
(2004). Trade reforms and wage inequality in
Colombia. In Macroeconomic Policies and
Poverty Reduction (pp. 229-266). Routledge.
[8] Autor, D. H., Katz, L. F., & Kearney, M. S.
(2008). Trends in US wage inequality:
Revising the revisionists. The Review of
economics and statistics, 90(2), 300-323.
[9] Baldwin, R. E., & Forslid, R. (1999). Putting
growth effects in computable equilibrium trade
models. Dynamic Issues in Applied
Commercial Policy Analysis, Cambridge
University Press, Cambridge, 44-84.
[10] Basmann R., (1957), A Generalized Classical
Method of Linear Estimation of Coefficients in
a Structural Equation, Econometrica, 25(1), 77-
83.
[11] Beyer, H., Rojas, P., & Vergara, R. (1999).
Trade liberalization and wage inequality.
Journal of Development Economics, 59(1),
103-123.
[12] Borjas, G. J., Freeman, R. B., & Katz, L. F.
(1992). On the labour market effects of
immigration and trade. Immigration and the
Workforce. Chicago: University of Chicago
Press/NBER, 213-244.
[13] Bresson, G., Kramarz, F., & Sevestre, P.
(1996). Dynamic labour demand models. In
The econometrics of panel data (pp. 660-684).
Springer, Dordrecht.
[14] Bucciferro, J. (2010). Economic restructuring
and poverty/income inequality in Latin
America. Consilience, (4), 1-16.
[15] Carroll, D., & Tani, M. (2013). Over-education
of recent higher education graduates: New
Australian panel evidence. Economics of
Education Review, 32, 207-218.
[16] Castilho, M., Menéndez, M., & Sztulman, A.
(2012). Trade liberalization, inequality, and
poverty in Brazilian states. World
Development, 40(4), 821-835.
[17] Chaudhry, I. S., & Imran, F. (2013). Does trade
liberalization reduce poverty and inequality?
Empirical evidence from Pakistan. Pakistan
Journal of Commerce and Social Sciences
(PJCSS), 7(3), 569-587.
[18] Cohen, W. M., & Levinthal, D. A. (1989).
Innovation and learning: the two faces of R &
D. The economic journal, 99(397), 569-596.
[19] Cortès, O., & Jean, S. (1997a). Quel est
l'impact du commerce extérieur sur la
productivité et l'emploi? (No. 1997-08).
[20] Cortes, O., & Jean, S. (1997c). Commerce
international, emploi et productivité. Travail et
emploi, 70, 61-79.
[21] Cribb J., (2013), Public Sector Pay and
Employment, Institute for Fiscal Studies,
Spending Round Briefing.
[22] Davis, D. R., & Harrigan, J. (2011). Good jobs,
bad jobs, and trade liberalization. Journal of
international Economics, 84(1), 26-36.
[23] Deardorff, A. V. (2001). Fragmentation in
simple trade models. The North American
Journal of Economics and Finance, 12(2), 121-
137.
INTERNATIONAL JOURNAL OF APPLIED MATHEMATICS,
COMPUTATIONAL SCIENCE AND SYSTEMS ENGINEERING
DOI: 10.37394/232026.2023.5.7
Ben Younes Hamza, Ben Ameur Fayçal
E-ISSN: 2766-9823
81
Volume 5, 2023
[24] Deardorff, A. V. (2001). International provision
of trade services, trade, and fragmentation.
Review of International Economics, 9(2), 233-
248.
[25] Acar, S., & Dogruel, F. (2012). Sources of
inequality in selected MENA countries.
Structural Change and Economic Dynamics,
23(3), 276-285.
[26] Faustino, H. C., & Vali, C. (2011). The effects
of globalisation on OECD income inequality: a
static and dynamic analysis. DE/Working
papers nº 12/2011/DE.
[27] Feenstra, R. C., & Hanson, G. H (1996),
Globalization, Outsourcing, and Wage
Inequality, American Economic Review, 86,
240-245.
[28] Feenstra, R. C., & Hanson, G. H. (1997).
Foreign direct investment and relative wages:
Evidence from Mexico's maquiladoras. Journal
of international economics, 42(3-4), 371-393.
[29] Fuchs, R. J., & Pernia, E. M. (2019). External
economic forces and national spatial
development: Japanese direct investment in
Pacific Asia. In Urbanization and urban
policies in Pacific Asia (pp. 88-111).
Routledge.
[30] Ghazali, M. (2009). Trade openness and wage
inequality between skilled and unskilled
workers in Tunisia. Économie internationale,
(1), 63-97.
[31] Giraud, P. N. (1998). Les causes des inégalités
croissantes dans les pays riches. Etudes (1945),
388(1).
[32] Görg, H., & Strobl, E. (2002). Relative wages,
openness and skill-biased technological change.
Openness and Skill-Biased Technological
Change (October 2002).
[33] HANSON, G., & HARRISON, A. (1999),
Trade Liberalization and Wage Inequality in
Mexico, Industrial and Labour Relations
Review, 52, 88-271.
[34] HECKSCHER, E. F. (1919). The effect of
foreign trade theory of international trade.
Readings on The Theory of International Trade,
272-300.
[35] Jaumotte, F., Lall, S., & Papageorgiou, C.
(2013). Rising income inequality: technology,
or trade and financial globalization?. IMF
economic review, 61(2), 271-309.
[36] Kahai, S. K., & Simmons, W. (2005). The
impact of globalisation on income inequality.
Global Business and Economics Review, 7(1),
1-15.
[37] Katz, L. F. (1999). Changes in the wage
structure and earnings inequality. In Handbook
of labour economics (Vol. 3, pp. 1463-1555).
Elsevier.
[38] Katz, L. F., & Murphy, K. M. (1992). Changes
in relative wages, 1963–1987: supply and
demand factors. The quarterly journal of
economics, 107(1), 35-78.
[39] Kennedy, P. (2008), A guide to econometrics.
John Wiley & Sons.
[40] Goldberg, P. K., & Pavcnik, N. (2005). Trade,
wages, and the political economy of trade
protection: evidence from the Colombian trade
reforms. Journal of international Economics,
66(1), 75-105.
[41] Leamer, E. E. (1994). Trade, wages and
revolving door ideas.
[42] Lee, E., & VIVARELLI, E. M. (2006). Les
conséquences sociales de la mondialisation
dans les pays en développement. Revue
internationale du Travail, 145(3), 191-210.
[43] Lim, G. C., & McNelis, P. D. (2014). Income
inequality, trade and financial openness,
Fordham University Schools of Business
Research Paper No. 2425068.
[44] Lustig, N. (1998). La pobreza y el Programa
Progresa. Alivio a la pobreza. Análisis del
Programa de Educación, Salud y Alimentación.
Memoria del Seminario Progresa y Política
Social, CIESAS, México.
[45] Majeed, M. T., & MacDonald, R. (2010).
Distributional and Poverty Consequences of
Globalization: A dynamic Comparative
Analysis for Developing Countries (No.
2010_22).
[46] Mah, J. S. (2013). Globalization,
decentralization and income inequality: The
case of China. Economic Modelling, 31, 653-
658.
[47] Milner, C., & Wright, P. (1998). Modelling
labour market adjustment to trade liberalisation
in an industrialising economy. The Economic
Journal, 108(447), 509-528.
[48] Mrabet, Z. (2010). Impact de l'ouverture
commerciale sur le marché du travail des pays
en voie de développement: le cas de la Tunisie
(Doctoral dissertation, Université Paris-Est).
[49] Munshi, F. (2012). Does openness reduce wage
inequality in developing countries? Panel data
evidence from Bangladesh. The Singapore
Economic Review, 57(2), 1250012.
[50] Ohlin, B. (1933). Interregional and
International Trade Harvard University Press.
Cambridge MA.
[51] Pavcnik, N. (2003). What explains skill
upgrading in less developed countries?. Journal
of Development Economics, 71(2), 311-328.
INTERNATIONAL JOURNAL OF APPLIED MATHEMATICS,
COMPUTATIONAL SCIENCE AND SYSTEMS ENGINEERING
DOI: 10.37394/232026.2023.5.7
Ben Younes Hamza, Ben Ameur Fayçal
E-ISSN: 2766-9823
82
Volume 5, 2023
[52] Pavcnik, N., Goldberg, P., & Schady, N.
(2004), “Trade Policy and Industry Wage
Structure: Evidence from Brazil,” World Bank
Economic Review, 18, 319-344.
[53] Pissarides, C. A. (1997). Learning by trading
and the returns to human capital in developing
countries. The World Bank Economic Review,
11(1), 17-32.
[54] Polpibulaya, S. (2015). Trade Openness and
Income Inequality, Clemson University.
[55] Robbins, D., & Gindling, T. H. (1999). Trade
liberalization and the relative wages for
moreskilled workers in Costa Rica. Review
of Development Economics, 3(2), 140-154.
[56] Sachs, J. D., Shatz, H. J., Deardorff, A., &
Hall, R. E. (1994). Trade and jobs in US
manufacturing. Brookings papers on economic
activity, 1994(1), 1-84.
[57] Santos-Paulino, A. U. (2012). Trade, income
distribution and poverty in developing
countries: A survey (No. 207). United Nations
Conference on Trade and Development.
[58] Shah, A., & Whalley, J. (1991). The
redistributive impact of taxation in developing
countries. Tax Policy in Developing Countries,
19, 166.
[59] Lai, P. L., Tan, K. K. Y., Ong, H. W., & Lee,
V. W. T. (2015). An empirical study on trade
openness and income inequality in Latin
America incorporating with FDI inflows, GDP
growth and inflation (Doctoral dissertation,
UTAR).
[60] Theil, H. (1961). Economic forecasts and
policy.
[61] Townsend, J. (2007). Do tariff reductions affect
the wages of workers in protected industries?
Evidence from the CanadaUS Free Trade
Agreement. Canadian Journal of
Economics/Revue canadienne d'économique,
40(1), 69-92.
[62] Trognon, A. (2003). L'économétrie des panels
en perspective. Revue d'économie politique,
727-748.
[63] Wang, J. Y., & Blomström, M. (1992). Foreign
investment and technology transfer: A simple
model. European economic review, 36(1), 137-
155.
[64] Wood, A. (2009), Heckscher-Ohlin in Theory
and Reality, In: P. Arestis and J. Eatwell (eds),
Issues in Economic Development and
Globalization, London: Palgrave Macmillan,
Longer version available as Working Paper
157, Queen Elizabeth House, Oxford
University.
[65] Wood, A. (1995), How Trade Hurt Unskilled
Labour , Journal of Economic Perspectives,
9(3), 57-80.
[66] Wu, J. Y., & Hsu, C. C. (2012). Foreign direct
investment and income inequality: Does the
relationship vary with absorptive capacity?.
Economic Modelling, 29(6), 2183-2189.
[67] Bayar, Y., & Sasmaz, M. U. (2017). Impact of
foreign direct investments on unemployment in
emerging market economies: A co-integration
analysis. International Journal of Business and
Economic Sciences Applied Research
(IJBESAR), 10(3).
[68] Zakhilwal, O. (2001). The impact of
international trade on the wages of Canadians
(No. 2001156e). Statistics Canada, Analytical
Studies Branch.
[69] Zhu, S. C., & Trefler, D. (2005). Trade and
inequality in developing countries: a general
equilibrium analysis. Journal of international
Economics, 65(1), 21-48.
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
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
The authors received no financial support for the
research, writing, and publication of this article.
Authors declare that they do not have any
conflict of interest.
INTERNATIONAL JOURNAL OF APPLIED MATHEMATICS,
COMPUTATIONAL SCIENCE AND SYSTEMS ENGINEERING
DOI: 10.37394/232026.2023.5.7
Ben Younes Hamza, Ben Ameur Fayçal
E-ISSN: 2766-9823
83
Volume 5, 2023