The Journal of International Trade & Economic

Development
An International and Comparative Review

ISSN: 0963-8199 (Print) 1469-9559 (Online) Journal homepage: https://www.tandfonline.com/loi/rjte20

Exports and the shadow economy: Non-linear

effects

Phuc Nguyen Canh & Su Dinh Thanh

To cite this article: Phuc Nguyen Canh & Su Dinh Thanh (2020): Exports and the shadow
economy: Non-linear effects, The Journal of International Trade & Economic Development, DOI:
10.1080/09638199.2020.1759676

To link to this article: https://doi.org/10.1080/09638199.2020.1759676

Published online: 07 May 2020.

Submit your article to this journal

Article views: 13

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=rjte20
THE JOURNAL OF INTERNATIONAL TRADE & ECONOMIC DEVELOPMENT
https://doi.org/10.1080/09638199.2020.1759676

Exports and the shadow economy: Non-linear effects

Phuc Nguyen Canh a and Su Dinh Thanh b

a School of Banking, University of Economics Ho Chi Minh City, Ho Chi Minh City, Vietnam; b School of
Public Finance, University of Economics Ho Chi Minh City, Ho Chi Minh City, Vietnam

ABSTRACT
This study is the first attempt to investigate the non-linear impacts of export diversifica-
tion and export quality on the shadow economy. Using a global sample of 116 countries
over the period 2003–2014 and applying for panel econometric techniques, we find
that the effects of export diversification and export quality are nonlinear on the shadow
economy, respectively. The non-linear effects are also consistently found for low- and
middle-income economies and high-income economies. The results are robust through
different measures of shadow economy and control variables. Our findings show that
this is a tipping point in the relationship of export diversification and export quality
with the shadow economy. That is, when moving beyond the tipping point, both export
diversification and export quality could reduce the shadow economy. This implies that
trade liberalization towards export diversification and export quality is an important
factor for curbing the shadow economy.

KEYWORDS Export diversification; export quality; shadow economy; non-linear effects; global sample

JEL CLASSIFICATIONS F13, F14, F15

ARTICLE HISTORY Received 15 January 2020; Accepted 18 April 2020

1. Introduction
The shadow economy (or informal economy and unofficial economy),1 an unknow-
ing economic sector (Friedrich Schneider and Enste 2000), exists as a pervasive eco-
nomic feature (Blackburn, Bose, and Capasso 2012). Concrete literature documents
that tax evasion and weak institutions are the main causes of the shadow economy
(Dreher, Kotsogiannis, and McCorriston 2009). The difficulty in the measurement of
the shadow economy has constrained empirical research over the past decades (e.g. Alm
and Embaye 2013 with currency demand approach, Medina and Schneider 2018 with
MIMIC method). Several efforts have been made to investigate the determinants of the
shadow economy, supporting the importance of institutions and taxation (see Dreher
and Schneider 2010; Torgler and Schneider 2009). Some recent studies attempt to explore
augmented drivers of the shadow economy. For instance, Huynh et al. (2019) notice the
linkage between FDI inflow and the shadow economy. However, this topic has been
still a matter of continuous debate due to the important role of the shadow economy

CONTACT Phuc Nguyen Canh [email protected] School of Banking, University of Economics

Ho Chi Minh City, 59C Nguyen Dinh Chieu, District 3, Ho Chi Minh City 700000, Vietnam

© 2020 Informa UK Limited, trading as Taylor & Francis Group

2 P. N. CANH AND S. DINH THANH

Figure 1. Export diversification, Export quality, and the Shadow economy

in policy implications (Canh et al. 2019). Interestingly, recent comprehensive database

from Medina and Schneider (2018) generate more opportunities to further explore the
determinants of the shadow economy.
As the important strand of economic integration, export diversification and export
quality are two important features in the process of trade liberalization, which affect
domestic economic activities (Canh, Schinckus, and Thanh 2019; Phuc Nguyen,
Schinckus, and Su 2020) including the shadow economy. Le et al. (2020) document
a non-linear relationship between export diversification and income inequality, while
previous studies indicate the link between inequality and shadow economy (Liévanos,
Greenberg, and Wishart 2018). For instance, Ahmed, Rosser, and Rosser (2007) indi-
cate that the reduction of income inequality and the improvement in social welfare
are important decreasing drivers to the shadow economy. That is, there might be
non-linear linkages between export dynamics with the shadow economy. Interestingly,
Figure 2 likely shows the potential non-linear linkages between export diversification
(Figure 2(a)), export quality (Figure 2(b)) with the shadow economy.
However, the aspect has so far been neglected in the literature. In the context of
recent trade wars, the shadow economy seems to be related to domestic economic tur-
bulence across countries (Fuchs et al. 2019; Li, He, and Lin 2018), while the shadow
economy is referred to as a contributor to CO2 emissions (Canh et al. 2019). There-
fore, understanding the drivers of shadow economy is significant for both literature and
practices.
This study fills this gap by investigating the non-linear effects of export diversifica-
tion and export quality, respectively, on the shadow economy. There are most agreements
among economists on the benefits of export diversification or export quality on domes-
tic economic conditions (Le et al. 2020) and institutional reforms (Gilbert 2004). That is,
export diversification or export quality may boost official economic activities and then
reduce incentives to work in the shadow economy. The literature shows that the differ-
ences in the links between economic factors and shadow economy are attribute to the
differences in institutional quality or taxation (Dreher, Kotsogiannis, and McCorriston
2009; Torgler and Schneider 2009). However, the literature indicates that export diver-
sification has both pros and cons (Le et al. 2020). Export diversification can carry out
benefits for employment (Egger and Etzel 2012) or poverty reduction (Le et al. 2020), but
it is also related to high initial costs for new production or new market’s entry (Aw and
 THE JOURNAL OF INTERNATIONAL TRADE & ECONOMIC DEVELOPMENT 3

Lee 2017; Hoffman, Munemo, and Watson 2016). The improvements in export quality
require a more initial investment. Therefore, if the costs to improve export diversifica-
tion and export quality are higher than their benefits, especially in initial phase, it could
surge negative effects on domestic economy (Le et al. 2020). In addition, export diver-
sification and export quality may lead to higher competition for domestic producers in
comparison to international producers not only in the international market but also in
the domestic one. That is, if the benefits of trade globalization can be only realized at a
certain level of diversification or quality of export to pass by the costs (Le et al. 2020),
the relationship between export diversification or export quality with shadow economy
may be non-linear.
Empirically, a global sample of 116 countries over the period 2003–2014 is employed
for investigation due to data availability. The sample is also divided into two subsam-
ples, including 78 low- and middle-income economies (LMEs) and 38 high-income
economies (HIEs). The size of the shadow economy of each country is collected from
Medina and Schneider (2018). The currency demand measure of shadow economy from
Alm and Embaye (2013) is derived for robustness check as well. The overall export
diversification index and export quality index are collected from IMF to proxy for
export diversification and export quality.2 The squared terms of export diversification
and export quality are added to estimations, respectively, to examine the possibility of
the non-linear relationship between two variables. The panel corrected standard errors
(PCSE) model is used as the main estimator to deal with cross-sectional dependence. The
pooled OLS, robust pooled OLS, pooled OLS with year effects, and feasible generalized
least square (FGLS) are applied for robustness checks to deal with heteroscedastic-
ity and fixed effects. The study also estimates all models by using one-year lags of
all independent variables to deal with endogeneity. The study concerns the sensitiv-
ity of results by including each control variable one by one.3 Moreover, each proxy of
export diversification indexes and export quality index is estimated separately to avoid
the potential multicollinearity. The predictive margins analysis is also applied for the
squared term of export diversification and export quality, respectively, to illustrate these
non-linear relationships. All estimation procedures are applied for full samples and two
sub-samples.
The empirical results are consistent and robustly. First, export diversification and
export quality have non-linear relationships with the shadow economy by inverted-
U curses. Second, these relationships are properly consistent for two sub-samples. The
empirical results hint that the increases of quality/diversification of exports may result in
higher informal activities at the initial level. However, the continuing improvements in
export diversification and quality would benefit in reducing shadow economy after a cer-
tain level. The results shed new lights of the relationship between export dynamics and
informal economic activities, while it revokes governments keep improving the quality
and diversification of exports to highest levels in confronting the shadow economy.
The study is structured as follows. The next section is short literature and hypothe-
sis development. Section 3 is the methodology and data. The results are presented and
discussed in Section 4. The conclusion is in the final section.

2. A short literature review and hypothesis construction

The existence of the shadow economy is one of the most challenges in policy set-
tings (Torgler and Schneider 2009). The weak government in law implementation and
4 P. N. CANH AND S. DINH THANH

enforcement (La Porta et al. 1999), and tax evasion is considered as the main cause
of the shadow economy (Tanzi 1999). The theory of tax evasion explains that eco-
nomic agents are not willing to pay high taxes (Tanzi 1982). As a result, they would
move economic activities to the shadow economy if taxation level and social security
burdens are high (Friedrich Schneider 1994). Dell’Anno and Davidescu (2019) indi-
cate that the shadow economy is a substitute, while tax evasion is a complement to
the Romanian official economy over the period 2000–2017. Institutional economics
argues that weak institution is the main cause of informal economic activities (Fried-
man et al. 2000) because of higher labor costs in the official economy (Su, Nguyen, and
Christophe 2019; Phuc Canh 2018; Nguyen et al. 2018; Thanh and Canh 2019; Schneider
and Enste 2000). Enste (2018) concludes that institutional quality can explain two-thirds
of the variance of the shadow economy in OECD and transition countries. Besides tax
evasion and institutional quality, the empirical literature also notices some other deter-
minants of the shadow economy (Huynh et al. 2019). Salahodjaev (2015) shows the
negative impacts of intelligence on the shadow economy in a sample of 158 countries
over the period 1999–2007. Goel and Saunoris (2016) find the negative impact of fis-
cal decentralization on the shadow economy. Huynh et al. (2019) add that FDI inflows
also have impacts on the shadow economy. This line of literature is still open for further
investigation.
The international trade is argued with positive influences on domestic economic
activities, especially productivity gains and job creations (see Canh, Schinckus, and
Thanh 2019; Dinh Thanh and Canh 2019). The diversification or quality of export is
considered as two important features (Osakwe, Santos-Paulino, and Dogan 2018), which
affect domestic socio-economic activities and structures (see Le et al. (2020)). The liter-
ature shows positive benefits of export diversification through job creation (Egger and
Etzel 2012), poverty reduction (Le et al. 2020), and new economic activities (Osakwe,
Santos-Paulino, and Dogan 2018). These effects may have potential negative impacts on
the shadow economy.
The export diversification is found with a positive impact on domestic entrepreneur-
ship (Clark 2009) and the engagement of domestic economic agents into international
trade (see Bianchi and Wickramasekera (2016)). These activities are likely to favor more
the official economic sector than the informal one. The export diversification is mostly
agreed as a good factor contributing to the reduction of income inequality (Egger and
Etzel 2012) and social welfare (Le et al. 2020). The reduction of income inequality and
the improvement in social welfare are documented as important factors to reduce the
shadow economy (Ahmed, Rosser, and Rosser 2007). And, the increased export diver-
sification with more trading partners or exporting products would surge pressures on
the government to reform their institutional framework (Levchenko 2016) or more
openness in economic freedom (Pitlik 2007). As a result, the institutional framework
would be improved significantly (Feketekuty 2000; Evenett and Hoekman 2004). On the
other hand, increased export diversification could induce the government in implement-
ing policies to support the development of the private sector (Hillman and Ursprung
1996).
From the perspective of quality, there has been limited literature on the linkage
between export quality and socio-economic factors. The literature shows that export
quality is linked with the specialization in production and the improvement in the qual-
ity of production systems and technology (Alcalá 2016). A country that is assumed to
take the comparative advantages of trade liberalization, has advantages in high skilled
 THE JOURNAL OF INTERNATIONAL TRADE & ECONOMIC DEVELOPMENT 5

sectors and production efficiency (Sampson 2016). Thus, the increased export qual-
ity may lead to an increase in the relative wage of high skilled-workers (Meschi and
Vivarelli 2009) and the skill-biased nature of jobs between trading partners (Ander-
son 2005). In such a context, the unskilled-labors may suffer from social disadvantages,
thus increasing income inequality (Castilho, Menéndez, and Sztulman 2012). That is,
we can hypothesize that higher export quality may cause higher informal economic
activities.
The literature shows that export diversification has both pros and cons (Le et al. 2020).
Export diversification can generate benefits for employment (Egger and Etzel 2012) and
poverty reduction (Le et al. 2020), but it is linked with high initial costs for new prod-
ucts and market entry Aw and Lee (2017). This means that export diversification may
not always make domestic producers efficient (Anderson 2005). In other words, if the
costs of export diversification and export quality are higher, especially in initial stages
than their benefits, the expected returns of official economic activities would be reduced
(Le et al. 2020). In addition, the literature points out that the increased trade openness
requires to lower tariffs and loose trading constraints, thus leading to higher interna-
tional competition for domestic producers (Helpman and Krugman 1985; Nguyen, Le,
and Su 2019). This implies that higher export diversification could put competitive pres-
sures on domestic entrepreneurs not only in the international market, but also in their
domestic market. Consequently, domestic producers can be replaced or crowded out by
the international producers (Lin and Yang 2017). For this reason, domestic economic
agents can move from official economic activities to unofficial ones to hide their busi-
ness and then avoid taxation. Regarding export quality, although the improvements in
production quality or technology require a large number of initial investment costs, the
improvements in export quality have an important contribution to enhancing the com-
petitive capability of domestic producers in both domestic and international markets.
That means that the improvements in export quality may generate more incentive to
stimulate domestic firms to work in official sectors and then enhance their international
competition.
As aforementioned, the increased export diversification and export quality may create
both benefits and costs for domestic economic agencies, thus leading to their differ-
ent responses to choosing official or unofficial sectors. That means that the relationship
between export diversification and export quality, and the shadow economy may not be
linear. Empirically, some studies find a non-linear relationship between trade openness
and export diversification, and domestic economic factors. Lim and McNelis (2016), for
example, argue that the trade openness may have different effects on income inequal-
ity according to economic development level and production structure. This is because
each economy may have its optimal threshold of capital intensity and imported inter-
mediate goods (Lim and McNelis 2016). As in Garrido-Prada, Delgado-Rodriguez, and
Romero-Jordán (2018), the authors find evidence to support a U-shaped relationship
between export diversification and firms’ performance in Spanish listed firms. Recently,
Le et al. (2020) find interesting evidence on the inverted-U shaped structure between
export diversification and inequality. Unfortunately, the empirical literature has ignored
the non-linear influences of export diversification and export quality on the shadow
economy.
The next section forms the theoretical model and presents the empirical data and
estimation techniques to investigate this dynamic relationship between export diversifi-
cation, export quality on the shadow economy.
6 P. N. CANH AND S. DINH THANH

3. Methodology and data

The theoretical literature suggests that institutional quality is one of main driver of
shadow economy. There are also other drivers such as income level (e.g. see Dreher,
Kotsogiannis, and McCorriston 2009; Salahodjaev 2015). However, this study recruits
the database of shadow economy measured by with MIMIC method from the study of
Medina and Schneider (2018), which already includes these variables in estimations.4
Therefore, the study follows the strategy in Goel and Nelson (2016) by firstly esti-
mating the non-linear impacts of export diversification (ED) and export quality (EQ),
respectively, on shadow economy (SE). The study then includes a set of different control
variables (Xj ) for robustness check. The empirical equation is specified as follows:

SEit = β0 + β1 EDit + β2 ED2it + ∂j Xj,it + εit (1)
j

in which: i and t are for country i and time t, respectively. β and ∂ are coefficients; and
ε is the residual term. The squared term of ED is used to investigate the non-linear
relationship between export diversification and shadow economy. The EQ is then esti-
mated by replacing for ED to examine the non-linear effects of export quality on shadow
economy as well. In term of control variables (Xj ), the study employs inflation (Inf ),
unemployment (Unem), tax burden (Taxbur), trade openness (Trade), and FDI inflows
(FDI). Furthermore, government expenditures (GovEx), procedures to register a busi-
ness (StartupPro), cost of business start-up procedures (StartupCost), time required to
start a business (StartupTime), time required to register property (ProperTime), time
to prepare and pay taxes (TaxTime), and democracy index (Democracy) for robustness
check.
In term of main variables, the shadow economy’s size is collected from Medina and
Schneider (2018) estimated by MIMIC method including legal economic and productive
activities, which could contribute to official output if recorded. Medina and Schneider
(2018) do not include illegal or criminal activities, do-it-yourself, charitable or house-
hold activities as previous literature (Schneider and Williams 2013). It is, in fact, the
current best available database of the shadow economy in terms of observations and time
period, but it is still complained about its reliability (Canh et al. 2019). Thus, the study
collects an alternative measure of shadow economy (SEm) by employing the database
from currency demand method of Alm and Embaye (2013) for robustness check. In
term of export dynamics, the export diversification index and export quality index are
collected from the Direction of Trade statistics database (IMF).
In terms of other control variables, the ratio of unemployment (% of the total labor
force), inflation ratio (GDP deflator-%), the ratio of total trade to GDP (%), the ratio of
foreign direct investment net inflows (% of GDP) represents for FDI capital flows (FDI),
general government final consumption expenditure (% of GDP), start-up procedures to
register a business (number), cost of business start-up procedures (% of GNI per capita),
time required to start a business (days), time required to register property (days), and
time to prepare and pay taxes (hours) are collected from the World Development Indi-
cators database (WDIs, World Bank). The tax burden index (Taxbur) is collected from
the Heritage Foundation,5 while the democracy index is derived from Polity IV Project:
Political Regime Characteristics and Transitions.6
As the availability of SE in the period 1991–2015, the availability of ED and EQ from
DOT-IMF to 2014, and the data of control variables7 from WDIs are annually from
 THE JOURNAL OF INTERNATIONAL TRADE & ECONOMIC DEVELOPMENT 7

2003, the period of 2003–20148 is chose for empirical investigation. The final sample
includes 116 countries including 78 low- and middle-income economies (LMEs) and 38
high-income economies (HIEs) as the availability of data for all variables (see Table A1,
Appendix, for list of countries). Table 1 presents variables, definitions, measurements,
sources, and statistical descriptions of main variables and control variables (see Table A3,
Appendix for definitions, measurements, sources, and statistical descriptions of other
control variables).
Table 2 reports the unconditional correlation matrix between main variables. It shows
the significant positive correlations between ED with SE. This means that lower ED has a
significant positive relationship with SE. While EQ has a significant negative correlation
with the shadow economy, this means higher quality of exports would have negative
relationship with SE.
From the perspective of the econometrical technique, our sample is large N (116
countries) and relatively short period T (2003–2014). Therefore, the cross-sectional
dependence test (Pesaran 2004) (CD-test) is conducted for each variable. The results
in the last column of Table 1 show the existence of cross-sectional dependence in all
variables. The study then recruits the panel corrected standard errors (PCSE) model
(Beck and Katz 1995) as the main estimator to deal with cross-sectional dependence
since it is suggested as a good estimator in this case (Bailey and Katz 2011). However,
the estimations of equation (1) may face problems as heteroscedasticity, endogeneity,
and multicollinearity. First, the feasible generalized least squares (FGLS) model (Liao
and Cao 2013) is used as a robustness check, which can deal with heteroscedasticity.
Second, the estimation is applied not only for levels of independent variables, but we
also estimate with a one-year lag of all independent variables to deal with endogeneity.
Third, each proxy of export diversification and export quality is added to models, respec-
tively, to deal with multicollinearity. Fourth, all control variables are added one by one to
the main model to check for the robustness and sensitivity of the results. Furthermore, a
bulk of other traditional estimators including pooled OLS, robust pooled OLS, pool OLS
with year effects are recruited to check for the sensitivity of results. At last, the margins
analysis is applied for the square terms to further illustrate the non-linear relationship.

4. Empirical results
Based on equation (1), we first estimate the effects of export diversification and quality
on the shadow economy for the entire sample. The estimation results are reported in
Table 3. Columns (1)-(6) are for export diversification (ED). Columns (7)-(12) are for
export quality (EQ). We conduct an estimation strategy by adding one by one control
variables9 to equation (1) to check robustness and sensitivity.
As shown in Table 3, observations show that the effect of export diversification (ED)
is estimated to be significantly positive on the shadow economy (see Columns 1–6).
Interestingly, the square term of ED is found to be significantly negative on the shadow
economy, suggesting that the association between export diversification and the shadow
economy is nonlinear with an inverted U shaped curve. Similarly, we observe a positive
effect of export quality (EQ) and its square term on the show economy (see Columns
7–12) respectively, suggesting that the linkage between export quality and the shadow
economy is also nonlinear with an inverted U shaped curve. In the same vein, Figure 3
predicts the marginal changes of the square terms of ED and EQ, which re-affirms the
non-linear relationship between export quality or export diversification and the shadow
 8
P. N. CANH AND S. DINH THANH
Table 1. Variables, definitions, calculations, sources, data description, and CD tests.
Variable Definition Calculation Source Obs Mean SD. Min Max CD-test
SE Shadow economy Shadow economy (% GDP) Medina and Schneider (2018) 1392 29.35 12.92 6.16 69.01 146***
ED Export diversification Export diversification index DOT (IMF) 1392 3.17 1.19 1.42 6.06 11.7***
EQ Export quality Export quality index (all DOT (IMF) 1392 0.83 0.16 0.20 1.07 40.2***
products)
Unem Unemployment Unemployment, total (% of WDIs 1392 7.73 5.62 0.32 34.07 50.0***
total labor force) (modeled
ILO estimate)
Taxbur Tax burden Tax burden index Heritage Foundation 1392 73.94 13.11 29.80 99.90 37.6***
Inf Inflation Inflation, GDP deflator (annual WDIs 1392 6.49 8.04 −25.13 100.63 50.7***
%)
Trade Trade openness Trade (% of GDP) WDIs 1380 86.35 53.02 21.58 442.62 49.8***
FDI FDI inflows Foreign direct investment, net WDIs 1392 4.84 6.80 −15.99 87.44 31.5***
inflows (% of GDP)
Notes: DOT-IMF is direction of trade statistics database, IMF; WDIs is world development indicators database (World Bank); WGIs Worldwide Governance Indicators database (World Bank);
the economic freedom related indices are collected from Heritage Foundation. In CD test: Under the null hypothesis of cross-section independence, CD ∼ N(0,1).
*** is significant level at 1% that indicates data are correlated across panel groups.
 THE JOURNAL OF INTERNATIONAL TRADE & ECONOMIC DEVELOPMENT 9

Table 2. Unconditional correlation matrix.

Correlation SE ED EQ Unem Taxbur Inf Trade FDI
SE 1.00
ED 0.43*** 1.00
p-value 0.00
EQ −0.55*** −0.63*** 1.00
p-value 0.00 0.00
Unem −0.01 −0.01 0.10*** 1.00
p-value 0.58 0.69 0.00
Taxbur 0.28*** 0.23*** −0.22*** −0.10*** 1.00
p-value 0.00 0.00 0.00 0.00
Inf 0.21*** 0.24*** −0.28*** −0.04 0.20*** 1.00
p-value 0.00 0.00 0.00 0.16 0.00
Trade −0.22*** −0.04* 0.23*** 0.02 0.18*** −0.09*** 1.00
p-value 0.00 0.10 0.00 0.41 0.00 0.00
FDI −0.11*** 0.01 0.06** −0.03 0.07*** −0.01 0.49*** 1.00
p-value 0.00 0.80 0.02 0.30 0.01 0.64 0.00
Note: *, **, *** are significant levels at 10%, 5%, and 1%, respectively.

economy. For further robustness check, the estimates for other control variables and
the measure of the shadow economy are implemented as presented in Tables A4 and
A5 in Appendix. Table A4 provides robust evidence by using different control variables,
while Table A5 shows evidence with the currency demand measure of the shadow econ-
omy (SEm) even with shorter period. In addition, Figure 1 in Appendix also predicts
the non-linear impacts of export quality and export diversification respectively, on the
size of shadow economy (calculated by currency demand approach). Overall, the non-
linear effect of export diversification or export quality imply that an increase in export
quality or export diversification would give rise to an increase in the shadow economy
until a certain tipping point. After the tipping point, its effect on the shadow economy is
negative.
Next, the question is that whether the effects of export diversification and quality on
the shadow are different between LMEs and HIEs. By separating the entire sample into
LMEs and HIEs, the estimation results are reported in Table 4. The effects of export
diversification are presented in Panel A, in which the effects for LMEs are in Columns
(1–6) and the effects for HIEs are in Columns (7–12). The effects of export quality are
presented in Panel B, in which the effects for LMEs are in Columns (13–18) and the
effects for HIEs are in Columns (19–24). We also conduct an estimation strategy by
adding one by one control variables to check robustness and sensitivity.
In panel A, observations show that export diversification (ED) has a consistently pos-
itive effect on the show economy both LMEs and HIEs. Regarding the square term of
export diversification, we find its effect is negative in all cases of adding control vari-
ables for LMEs (see Columns 1–6). With regard to HIEs, the square term effect of export
diversification is found to be insignificantly negative in three cases (Columns 7, 11, and
12) and with the exception of Columns 8–10, insignificantly negatively. In Panel B, we
find that the effect of export quality (EQ) is also a consistently positive on the show
economy for LMEs and HIEs. Interestingly, the effect of square term of export quality is
consistently negative on the show economy for LMEs and HIEs (see Columns 13–24). In
addition, Figure 3 predicts the non-linear relationships between ED, EQ and the shadow
economy for two sub-groups. This re-affirms as the estimated results in Table 4. The
results that checked by using the currency demand measure of shadow economy, are
 10
P. N. CANH AND S. DINH THANH
Table 3. Export diversification, export quality and shadow economy: non-linear evidence.
Dep. var: SE (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12)
ED 17.83*** 18.11*** 15.08*** 14.63*** 14.88*** 14.87***
[1.061] [1.081] [1.050] [1.029] [1.149] [1.152]
ED2 −1.871*** −1.910*** −1.525*** −1.486*** −1.543*** −1.542***
[0.140] [0.144] [0.140] [0.139] [0.151] [0.152]
EQ 143.1*** 143.1*** 132.9*** 133.2*** 121.4*** 121.0***
[6.094] [6.231] [6.101] [6.258] [6.292] [5.709]
EQ2 −124.8*** −125.0*** −117.5*** −117.8*** −109.3*** −109.0***
[3.744] [3.810] [3.738] [3.926] [3.709] [3.341]
Unem 0.056*** 0.069*** 0.071*** 0.105*** 0.104*** 0.098*** 0.107*** 0.107*** 0.105*** 0.104***
[0.018] [0.019] [0.020] [0.028] [0.028] [0.021] [0.021] [0.021] [0.025] [0.025]
Taxbur 0.129*** 0.119*** 0.170*** 0.170*** 0.057*** 0.058*** 0.077*** 0.077***
[0.014] [0.013] [0.008] [0.008] [0.011] [0.012] [0.010] [0.011]
Inf 0.122*** 0.078*** 0.078*** −0.008 −0.008 −0.008
[0.033] [0.027] [0.027] [0.032] [0.041] [0.041]
Trade −0.060*** −0.059*** −0.013*** −0.011***
[0.002] [0.003] [0.002] [0.003]
FDI −0.010 −0.025
[0.035] [0.020]
Constant −5.784*** −6.651*** −11.099*** −10.17*** −8.934*** −8.916*** −0.326 −0.921 −2.107 −2.141 1.218 1.404
[1.756] [1.774] [1.569] [1.475] [1.812] [1.824] [2.730] [2.722] [2.530] [2.551] [3.034] [2.866]
Observations 1392 1392 1392 1392 1380 1380 1392 1392 1392 1392 1380 1380
R-squared 0.228 0.228 0.243 0.248 0.304 0.304 0.409 0.411 0.414 0.414 0.416 0.416
No. of countries 116 116 116 116 116 116 116 116 116 116 116 116
Notes: standard errors are in []. *, **, *** are significant levels at 10%, 5%, and 1%, respectively.
Table 4. Export diversification, export quality and shadow economy: non-linear evidence in two sub-samples.
Part A: Export diversification and shadow economy in two sub-groups

Dep. var: SE (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12)

Group: Low- and Middle-Income Economies (LMEs) High-Income economies

ED 8.316*** 7.201*** 7.102*** 7.297*** 7.527*** 7.528*** 1.680** 1.870* 0.554 0.839 2.254** 2.142**
[0.706] [0.631] [0.635] [0.632] [0.777] [0.796] [0.797] [0.994] [0.938] [1.018] [0.925] [0.915]
ED2 −0.845*** −0.690*** −0.662*** −0.679*** −0.705*** −0.701*** −0.234** −0.143 −0.080 −0.186 −0.426*** −0.407***

THE JOURNAL OF INTERNATIONAL TRADE & ECONOMIC DEVELOPMENT

[0.092] [0.083] [0.085] [0.084] [0.098] [0.100] [0.118] [0.133] [0.122] [0.172] [0.156] [0.156]
Unem −0.159*** −0.148*** −0.148*** −0.126*** −0.126*** 0.677*** 0.644*** 0.666*** 0.622*** 0.618***
[0.026] [0.022] [0.022] [0.027] [0.028] [0.087] [0.082] [0.073] [0.071] [0.070]
Taxbur 0.058** 0.066** 0.072** 0.076** 0.079*** 0.070*** 0.099*** 0.095***
[0.029] [0.031] [0.031] [0.031] [0.010] [0.009] [0.011] [0.012]
Inf −0.069** −0.064* −0.064* 0.312*** 0.268*** 0.274***
[0.033] [0.033] [0.033] [0.062] [0.056] [0.056]
Trade −0.018*** −0.014** −0.014*** −0.011***
[0.004] [0.006] [0.002] [0.003]
FDI −0.075 −0.040
[0.056] [0.025]
Constant 17.70*** 20.75*** 16.14*** 15.65*** 16.01*** 15.61*** 14.12*** 7.827*** 5.454*** 4.873*** 3.049* 3.381**
[1.273] [1.188] [2.503] [2.572] [3.129] [3.241] [1.030] [2.068] [1.812] [1.685] [1.558] [1.515]
Observations 936 936 936 936 924 924 456 456 456 456 456 456
R-squared 0.054 0.063 0.065 0.069 0.072 0.073 0.002 0.179 0.205 0.265 0.285 0.287
No. of countries 78 78 78 78 78 78 38 38 38 38 38 38
Part B: Export quality and Shadow economy in two sub-groups

Dep. var: SE (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24)

Group: Low- and Middle-Income Economies (LMEs) High-Income economies

EQ 80.26*** 78.31*** 75.59*** 75.52*** 69.12*** 67.36*** 306.1*** 218.8*** 229.0*** 213.6*** 212.7*** 211.6***
[8.715] [8.538] [7.972] [7.867] [7.891] [4.973] [15.74] [15.45] [16.62] [18.29] [17.44] [17.90]
EQ2 −73.55*** −71.89*** −70.79*** −71.15*** −67.13*** −66.21*** −191.1*** −140.4*** −144.6*** −133.2*** −132.5*** −131.9***
[6.085] [5.955] [5.625] [5.590] [5.807] [3.927] [9.289] [9.103] [9.672] [10.57] [9.969] [10.26]
(continued).

11
 12
P. N. CANH AND S. DINH THANH
Table 4. Continued.
Part B: Export quality and Shadow economy in two sub-groups

Dep. var: SE (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24)

Group: Low- and Middle-Income Economies (LMEs) High-Income economies

Unem −0.069*** −0.048*** −0.047** −0.057*** −0.058** 0.543*** 0.541*** 0.565*** 0.564*** 0.560***
[0.024] [0.018] [0.018] [0.022] [0.023] [0.064] [0.065] [0.064] [0.064] [0.065]
Taxbur 0.081*** 0.090*** 0.093*** 0.100*** 0.036*** 0.039*** 0.040*** 0.037***
[0.029] [0.031] [0.035] [0.037] [0.010] [0.009] [0.012] [0.013]
Inf −0.081** −0.079** −0.078** 0.150*** 0.150*** 0.155***
[0.034] [0.036] [0.036] [0.050] [0.050] [0.050]
Trade −0.000 0.006 −0.001 0.002
[0.005] [0.006] [0.002] [0.003]
FDI −0.107** −0.040*
[0.046] [0.021]
Constant 18.97*** 19.99*** 15.03*** 15.23*** 17.58*** 17.81*** −99.61*** −67.18*** −75.45*** −72.22*** −71.99*** −71.35***
[3.365] [3.322] [3.879] [3.918] [4.947] [3.963] [6.695] [6.218] [7.021] [7.918] [7.714] [7.875]
Observations 936 936 936 936 924 924 456 456 456 456 456 456
R-squared 0.137 0.139 0.144 0.149 0.150 0.153 0.210 0.321 0.327 0.338 0.338 0.340
No. of countries 78 78 78 78 78 78 38 38 38 38 38 38
Notes: standard errors are in []; *, **, *** are significant levels at 10%, 5%, and 1%, respectively.
 THE JOURNAL OF INTERNATIONAL TRADE & ECONOMIC DEVELOPMENT 13

Figure 2. Predictive margins of export diversification, export quality on the shadow economy: non-linear evidence

reported in Table A6 (Appendix), showing consistent findings. In general, our findings

indicate that the nonlinear effects of export diversification and quality on the shadow for
LMEs are the same as those for HIEs. From these results, policy implication is that trade
liberalization along with higher level of export diversification and export quality would
reduce the shadow economy across countries over time.
Regarding control variables, We find that the correlation between unemployment and
the shadow economy is found to be significantly positive, suggesting that an increase in
unemployment leads to an increase in the shadow economy for the entire sample (see
Table 3) and HIEs (see Columns 7–12 in Table 4), which is in line with some previ-
ous studies (Mauleón and Sardà 2017; Adriana 2014; Dell’Anno and Solomon 2008).
Interestingly, we observe a negative effect of unemployment on the shadow economy for
LMEs, suggesting that increasing unemployed persons could lead to the diminishing of
the number of persons involving in the informal sector (Tedds and Giles 2002), while
the informal sector growth could be positively associated with to GDP growth which in
turn is negatively associated with unemployment (Saafi and Farhat 2015). Observations
show a positive effect of the tax burden on the shadow economy for the entire sample
and subsamples, which is different from the study of Mazhar and Méon (2017). Our find-
ings suggest that increasing the tax burden could lead to raising government regulatory
activities, which are likely to be main derivers of increasing the growth and size of the
shadow economy (Stankevičius and Vasiliauskaitė 2014). We find that the effect of infla-
tion is relatively heterogeneous for the entire sample as found in the study of Baklouti
and Boujelbene (2019). Regarding subsamples, the effect of inflation is found to be pos-
itive for HIEs, while it is negative for LMEs. For LMEs, a moderate inflation level could
benefit economic growth (Muzaffar and Junankar 2014; Ibarra and Trupkin 2016) and
then the formal sector growth. The formal sector growth in turn is negatively associ-
ated with the informal sector. In contrast, a positive association between inflation and
the growth of the shadow economy is always found in developed countries (Mazhar and
Méon 2017). In this study, we find that the effect of FDI is inconsistent on the shadow
economy, while the negative effect of trade is relative consistent on the shadow econ-
omy. This suggests that trade liberalization could reduce the size of the shadow economy
(Farzanegan, Hassan, and Badreldin 2019).
14 P. N. CANH AND S. DINH THANH

Figure 3. Export diversification, Export quality and the shadow economy: non-linear evidences for two subsam-
ples. (a) The impacts of ED in LMEs, (b) The impacts of EQ in LMEs, (c) The impacts of ED in HIEs, (d) The impacts of
EQ in HIEs

5. Some discussion and conclusion

This study is the first attempt to examine the non-linear influences of export diversi-
fication and export quality on the shadow economy. A global sample of 116 countries
and two subsamples including 78 LMEs and 38 HIEs over the period 2003–2014 are
recruited for empirical analysis due to the availability of data. PCSE model is used as the
main estimator, while several other estimators are applied as a robustness check. Further-
more, different estimation techniques are also applied to address heteroscedasticity, fixed
effects, endogeneity, and multicollinearity. The results are mostly robust and consistent.
First, the study provides evidence that there exists the inverted-U shape associa-
tion between export diversification (and export quality) and the shadow economy. This
means that this is a tipping point of export diversification and export quality. Thus, when
moving beyond the tipping point, both export diversification and export quality could
reduce the shadow economy. Second, the inverted-U shaped effects of export diversifica-
tion and export quality are also consistently found for subsamples, suggesting that higher
export diversification and export quality are, lower the shadow economy is regardless
of whether they are for developed countries or for developing countries. Our findings
shed light on the relationship between trade and economic liberalization policy and the
shadow in the literature. That is, promoting export-driven economic liberalization may
be a right policy in the context of global integration. As in Farzanegan, Hassan, and
 THE JOURNAL OF INTERNATIONAL TRADE & ECONOMIC DEVELOPMENT 15

Badreldin (2019), the authors document that the way of reducing the shadow economy
is centered on economic liberalization. Economic liberalization that is helpful to reduce
the costs of imports and exports as well as doing business, probably bolsters productivity
growth opportunities in the formal economy, thus diminishing the informal economy.
In the same vein, our study suggests that trade liberalization along with export diver-
sification and export quality, it is possible to strengthen international competitiveness,
domestic market expansion and effective allocation of resources in the formal economy.
From the perspective of the shadow, the trade liberalization might move people out of
the informal sector to join the formal economy through job creation (Egger and Etzel
2012), poverty and inequality reduction (Le et al. 2020; Egger and Kreickemeier 2012),
and new economic activities (Osakwe, Santos-Paulino, and Dogan 2018).
In general, to ensure the success of economic policies in reducing the size of shadow
economy, it is wise to revise and improve trade liberalization policies towards export
diversification and export quality. It has been shown that export diversification and
export quality as export dynamics is not only important for productivity growth (Xue-
feng and Yaşar 2016) but also a relevant method for curbing the size of the shadow
economy.

Notes
1. The shadow economy, informal economy, or unofficial economy are used inter-exchangeable.
2. Higher export diversification index (ED) means lower export diversification. Higher export quality
index (EQ) means higher quality of export.
3. Different control variables are collected for robustness check from the light of Goel and Nelson
(2016). Author(s) thanks to helpful comment from anonymous reviewer.
4. Author(s) accredits to helpful suggestion from anonymous reviewer for this point.
5. See https://www.heritage.org/index/
6. See www.systemicpeace.org
7. Most of variables linked to startups and taxation from WDIs are available from 2003 (see more detail
in notes in Table A3, Appendix).
8. In the database of currency demand approach from Alm and Embaye (2013), the shadow economy is
available from 1984 to 2006. To match with database from Medina and Schneider (2018), the study
uses the period of 2003-2006 for this measure in robustness check (detail is explained in Notes,
Table A2, Appendix).
9. The results by other estimation estimators (pooled OLS, robust pooled OLS, and FGLS) or esti-
mation strategy (one-year lags or other estimators) are all consistent. They are provided upon
requests.

Disclosure statement
No potential conflict of interest was reported by the authors.

ORCID
Phuc Nguyen Canh http://orcid.org/0000-0001-8467-5010
Su Dinh Thanh http://orcid.org/0000-0003-2344-6315

References
Adriana, Davidescu. 2014. “Revisiting the Relationship Between Unemployment Rates and Shadow
Economy. A Toda-Yamamoto Approach for the Case of Romania.” Procedia Economics and Finance
10: 227–236. doi:10.1016/S2212-5671(14)00297-4. http://www.sciencedirect.com/science/article/
pii/S2212567114002974.
16 P. N. CANH AND S. DINH THANH

Ahmed, E., J. B. Rosser, and M. V. Rosser. 2007. “Income Inequality, Corruption, and the Non-Observed
Economy: A Global Perspective.” In Complexity Hints for Economic Policy, edited by M. Salzano and
D. Colander, 233–252. Milano: Springer Milan.
Alcalá, Francisco. 2016. “Specialization Across Goods and Export Quality.” Journal of International Eco-
nomics 98: 216–232. doi:10.1016/j.jinteco.2015.09.005. http://www.sciencedirect.com/science/arti
cle/pii/S0022199615001440.
Alm, James, and Abel Embaye. 2013. “Using Dynamic Panel Methods to Estimate Shadow Economies
Around the World, 1984–2006.” Public Finance Review 41 (5): 510–543.
Anderson, Edward. 2005. “Openness and Inequality in Developing Countries: A Review of Theory
and Recent Evidence.” World Development 33 (7): 1045–1063. doi:10.1016/j.worlddev.2005.04.003.
http://www.sciencedirect.com/science/article/pii/S0305750X05000665.
Aw, Bee Yan, and Yi Lee. 2017. “Demand, Costs and Product Scope in the Export Market.” European Eco-
nomic Review 100: 28–49. doi:10.1016/j.euroecorev.2017.07.009. http://www.sciencedirect.com/sci
ence/article/pii/S0014292117301344.
Bailey, Delia, and Jonathan N Katz. 2011. “Implementing Panel Corrected Standard Errors in R: The
PCSE Package.” Journal of Statistical Software 42 (CS1): 1–11.
Baklouti, Nedra, and Younes Boujelbene. 2019. “The Economic Growth–Inflation–Shadow Economy
Trilogy: Developed Versus Developing Countries.” International Economic Journal 33 (4): 679–695.
doi:10.1080/10168737.2019.1641540.
Beck, Nathaniel, and Jonathan N Katz. 1995. “What to Do (and Not to Do) with Time-Series Cross-
Section Data.” American Political Science Review 89 (3): 634–647.
Bianchi, Constanza, and Rumintha Wickramasekera. 2016. “Antecedents of SME Export Intensity in a
Latin American Market.” Journal of Business Research 69 (10): 4368–4376. doi:10.1016/j.jbusres.2016.
02.041. http://www.sciencedirect.com/science/article/pii/S0148296316302661.
Blackburn, Keith, Niloy Bose, and Salvatore Capasso. 2012. “Tax Evasion, the Underground Economy
and Financial Development.” Journal of Economic Behavior & Organization 83 (2): 243–253.
Canh, Nguyen Phuc, Christophe Schinckus, and Su Dinh Thanh. 2019. “Do Economic Openness
and Institutional Quality Influence Patents? Evidence from GMM Systems Estimates.” International
Economics 157: 134–169.
Canh, Nguyen Phuc, Su Dinh Thanh, Christophe Schinckus, Jo Bensemann, and Lai Trung Thanh. 2019.
“Global Emissions: A New Contribution from the Shadow Economy.” International Journal of Energy
Economics and Policy 9 (3): 320–337.
Castilho, Marta, Marta Menéndez, and Aude Sztulman. 2012. “Trade Liberalization, Inequality, and
Poverty in Brazilian States.” World Development 40 (4): 821–835. doi:10.1016/j.worlddev.2011.09.
018. http://www.sciencedirect.com/science/article/pii/S0305750X11002403.
Clark, Julian. 2009. “Entrepreneurship and Diversification on English Farms: Identifying Business Enter-
prise Characteristics and Change Processes.” Entrepreneurship and Regional Development 21 (2):
213–236.
Dell’Anno, Roberto, and Adriana AnaMaria Davidescu. 2019. “Estimating Shadow Economy and Tax
Evasion in Romania. A Comparison by Different Estimation Approaches.” Economic Analysis and
Policy 63: 130–149. doi:10.1016/j.eap.2019.05.002. http://www.sciencedirect.com/science/article/pii/
S0313592619300062.
Dell’Anno, Roberto, and Offiong Helen Solomon. 2008. “Shadow Economy and Unemployment Rate
in USA: Is There a Structural Relationship? An Empirical Analysis.” Applied Economics 40 (19):
2537–2555.
Dinh Thanh, Su, and Nguyen Phuc Canh. 2019. “Dynamics Between Government Spending and Eco-
nomic Growth in China: An Analysis of Productivity Growth.” Journal of Chinese Economic and
Business Studies 17 (2): 189–212. doi:10.1080/14765284.2019.1567069.
Dreher, Axel, Christos Kotsogiannis, and Steve McCorriston. 2009. “How do Institutions Affect Cor-
ruption and the Shadow Economy?” International Tax and Public Finance 16 (6): 773.
Dreher, Axel, and Friedrich Schneider. 2010. “Corruption and the Shadow Economy: An Empirical
Analysis.” Public Choice 144 (1–2): 215–238.
Egger, Hartmut, and Daniel Etzel. 2012. “The Impact of Trade on Employment, Welfare, and Income
Distribution in Unionized General Oligopolistic Equilibrium.” European Economic Review 56 (6):
1119–1135. doi:10.1016/j.euroecorev.2012.03.006. http://www.sciencedirect.com/science/article/
pii/S0014292112000499.
 THE JOURNAL OF INTERNATIONAL TRADE & ECONOMIC DEVELOPMENT 17

Egger, Hartmut, and Udo Kreickemeier. 2012. “Fairness, Trade, and Inequality.” Journal of International
Economics 86 (2): 184–196. doi:10.1016/j.jinteco.2011.10.002. http://www.sciencedirect.com/sci
ence/article/pii/S0022199611001218.
Enste, Dominik H. 2018. “The Shadow Economy in OECD and EU Accession Countries–Empirical
Evidence for the Influence of Institutions, Liberalization, Taxation and Regulation.” In Size, Causes
and Consequences of the Underground Economy, edited by B Christopher and S. Friedrich, 123–138.
Abingdon: Routledge.
Evenett, Simon J, and Bernard Hoekman. 2004. Government Procurement: Market Access, Transparency,
and Multilateral Trade Rules. Washington, DC: The World Bank.
Farzanegan, Mohammad Reza, Mai Hassan, and Ahmed Mohamed Badreldin. 2019. “Economic
Liberalization in Egypt: A Way to Reduce the Shadow Economy?” Journal of Policy Modeling.
doi:10.1016/j.jpolmod.2019.09.008. http://www.sciencedirect.com/science/article/pii/S0161893819
301292.
Feketekuty, Geza. 2000. “Regulatory Reform and Trade Liberalization in Services.” In GATS 2000: New
Directions in Services Trade Liberalization, edited by S. Pierre and M. S. Robert, 225–240. Washington,
DC: Brookings Institution Press.
Friedman, Eric, Simon Johnson, Daniel Kaufmann, and Pablo Zoido-Lobaton. 2000. “Dodging the
Grabbing Hand: The Determinants of Unofficial Activity in 69 Countries.” Journal of Public Eco-
nomics 76 (3): 459–493.
Fuchs, Richard, Peter Alexander, Calum Brown, Frances Cossar, Roslyn C Henry, and Mark Rounsevell.
2019. Why the US–China Trade War Spells Disaster for the Amazon. Berlin: Nature Publishing Group.
Garrido-Prada, Pablo, Maria Jesús Delgado-Rodriguez, and Desiderio Romero-Jordán. 2018. “Effect of
Product and Geographic Diversification on Company Performance: Evidence During an Economic
Crisis.” European Management Journal. doi:10.1016/j.emj.2018.06.004. http://www.sciencedirect.
com/science/article/pii/S0263237318300707.
Gilbert, Niyongabo. 2004. Trade Openness Policy, Quality of Institutions and Economic Growth. Aulnay-
sous-Bois: CERDI.
Goel, Rajeev K., and Michael A. Nelson. 2016. “Shining a Light on the Shadows: Iden-
tifying Robust Determinants of the Shadow Economy.” Economic Modelling 58: 351–364.
doi:10.1016/j.econmod.2016.06.009. http://www.sciencedirect.com/science/article/pii/S026499931
6301729.
Goel, Rajeev K., and James W. Saunoris. 2016. “Government Decentralization and Prevalence of
the Shadow Economy.” Public Finance Review 44 (2): 263–288. doi:10.1177/1091142114545677.
https://journals.sagepub.com/doi/abs/10.1177/1091142114545677.
Helpman, Elhanan, and Paul R Krugman. 1985. Market Structure and Foreign Trade: Increasing Returns,
Imperfect Competition, and the International Economy. Cambridge, MA: MIT press.
Hillman, Arye L., and Heinrich W. Ursprung. 1996. “The Political Economy of Trade Liberalization
in the Transition.” European Economic Review 40 (3): 783–794. doi:10.1016/0014-2921(95)00090-9.
http://www.sciencedirect.com/science/article/pii/0014292195000909.
Hoffman, Richard C., Jonathan Munemo, and Sharon Watson. 2016. “International Franchise Expan-
sion: The Role of Institutions and Transaction Costs.” Journal of International Management 22 (2):
101–114. doi:10.1016/j.intman.2016.01.003. http://www.sciencedirect.com/science/article/pii/S107
5425316000041.
Huynh, Cong Minh, Vu Hong Thai Nguyen, Hoang Bao Nguyen, and Phuc Canh Nguyen. 2019. “One-
Way Effect or Multiple-Way Causality: Foreign Direct Investment, Institutional Quality and Shadow
Economy?” International Economics and Economic Policy 17 (1): 219–239.
Ibarra, Raul, and Danilo R. Trupkin. 2016. “Reexamining the Relationship Between Inflation and
Growth: Do Institutions Matter in Developing Countries?” Economic Modelling 52: 332–351.
doi:10.1016/j.econmod.2015.09.011. http://www.sciencedirect.com/science/article/pii/S026499931
5002576.
La Porta, Rafael, Florencio Lopez-de-Silanes, Andrei Shleifer, and Robert Vishny. 1999. “The Quality of
Government.” The Journal of Law, Economics, and Organization 15 (1): 222–279.
Le, Thai-Ha, Canh Phuc Nguyen, Thanh Dinh Su, and Binh Tran-Nam. 2020. “The Kuznets Curve for
Export Diversification and Income Inequality: Evidence from a Global Sample.” Economic Analysis
and Policy 65: 21–39. doi:10.1016/j.eap.2019.11.004. http://www.sciencedirect.com/science/article/
pii/S0313592619303765.
Levchenko, Andrei A. 2016. The Impact of Trade Openness on Institutions. Washington, DC: World Bank.
18 P. N. CANH AND S. DINH THANH

Li, Chunding, Chuantian He, and Chuangwei Lin. 2018. “Economic Impacts of the Possible China–US
Trade War.” Emerging Markets Finance and Trade 54 (7): 1557–1577.
Liao, Hua, and Huai-Shu Cao. 2013. “How Does Carbon Dioxide Emission Change with the Economic
Development? Statistical Experiences from 132 Countries.” Global Environmental Change 23 (5):
1073–1082.
Liévanos, Raoul S., Pierce Greenberg, and Ryan Wishart. 2018. “In the Shadow of Production: Coal
Waste Accumulation and Environmental Inequality Formation in Eastern Kentucky.” Social Science
Research 71: 37–55. doi:10.1016/j.ssresearch.2018.01.003. http://www.sciencedirect.com/science/
article/pii/S0049089X17300650.
Lim, G. C., and Paul D. McNelis. 2016. “Income Growth and Inequality: The Threshold Effects of Trade
and Financial Openness.” Economic Modelling 58: 403–412. doi:10.1016/j.econmod.2016.05.010.
http://www.sciencedirect.com/science/article/pii/S026499931630133X.
Lin, Xiaohua, and Xiyan Yang. 2017. “From Human Capital Externality to Entrepreneurial Aspi-
ration: Revisiting the Migration-Trade Linkage.” Journal of World Business 52 (3): 360–371.
doi:10.1016/j.jwb.2016.11.001. http://www.sciencedirect.com/science/article/pii/S109095161630
1729.
Mauleón, I., and J. Sardà. 2017. “Unemployment and the Shadow Economy.” Applied Economics 49 (37):
3729–3740. doi:10.1080/00036846.2016.1267844.
Mazhar, Ummad, and Pierre-Guillaume Méon. 2017. “Taxing the Unobservable: The Impact of the
Shadow Economy on Inflation and Taxation.” World Development 90: 89–103. doi:10.1016/j.world
dev.2016.08.019. http://www.sciencedirect.com/science/article/pii/S0305750X1630465X.
Medina, Leandro, and Friedrich Schneider. 2018. “Shadow Economies Around the World: What Did We
Learn Over the Last 20 Years?” IMF Working Paper 18/17. https://papers.ssrn.com/sol3/papers.cfm?
abstract_id = 3124402.
Meschi, Elena, and Marco Vivarelli. 2009. “Trade and Income Inequality in Developing Countries.”
World Development 37 (2): 287–302. doi:10.1016/j.worlddev.2008.06.002. http://www.sciencedirect.
com/science/article/pii/S0305750X08002222.
Muzaffar, Ahmed Taneem, and P. N. Junankar. 2014. “Inflation–Growth Relationship in Selected Asian
Developing Countries: Evidence From Panel Data.” Journal of the Asia Pacific Economy 19 (4):
604–628. doi:10.1080/13547860.2014.920594.
Nguyen, Canh Phuc, Thai-Ha Le, and Thanh Dinh Su. 2019. “Economic Policy Uncertainty and Credit
Growth: Evidence from a Global Sample.” Research in International Business and Finance 51: 101118.
Nguyen, Canh Phuc, Christophe Schinckus, Thanh Dinh Su, and Felicia Chong. 2018. “Institu-
tions, Inward Foreign Direct Investment, Trade Openness and Credit Level in Emerging Market
Economies.” Review of Development Finance 8 (2): 75–88.
Osakwe, Patrick N., Amelia U. Santos-Paulino, and Berna Dogan. 2018. “Trade Dependence, Liberal-
ization, and Exports Diversification in Developing Countries.” Journal of African Trade 5 (1): 19–34.
doi:10.1016/j.joat.2018.09.001. http://www.sciencedirect.com/science/article/pii/S221485151830
0185.
Pesaran, M. Hashem. 2004. “General Diagnostic Tests for Cross Section Dependence in Panels.”
CESifo Working Paper Series No. 1229; IZA Discussion Paper No. 1240. Available at SSRN:
https://ssrn.com/abstract = 572504.
Phuc Canh, Nguyen. 2018. “The Effectiveness of Fiscal Policy: Contributions from Institutions and
External Debts.” Journal of Asian Business and Economic Studies 25 (1): 50–66.
Phuc Nguyen, Canh, Christophe Schinckus, and Thanh Dinh Su. 2020. “Economic Integration and CO2
Emissions: Evidence from Emerging Economies.” Climate and Development 12 (4): 369–384.
Pitlik, Hans. 2007. “A Race to Liberalization? Diffusion of Economic Policy Reform among OECD-
Economies.” Public Choice 132 (1–2): 159–178.
Saafi, Sami, and Abdeljelil Farhat. 2015. “Is There a Causal Relationship Between Unemployment and
Informal Economy in Tunisia: Evidence From Linear and non-Linear Granger Causality.” Economics
Bulletin 35 (2): 1191–1204.
Salahodjaev, Raufhon. 2015. “Intelligence and Shadow Economy: A Cross-Country Empirical Assess-
ment.” Intelligence 49: 129–133.
Sampson, Thomas. 2016. “Assignment Reversals: Trade, Skill Allocation and Wage Inequality.” Journal
of Economic Theory 163: 365–409. doi:10.1016/j.jet.2016.02.001. http://www.sciencedirect.com/sci
ence/article/pii/S0022053116000223.
 THE JOURNAL OF INTERNATIONAL TRADE & ECONOMIC DEVELOPMENT 19

Schneider, Friedrich. 1994. “Measuring the Size and Development of the Shadow Economy. Can the
Causes be Found and the Obstacles be Overcome?” In Essays on Economic Psychology, edited by H.
Brandstätter and W. Güth, 193–212. Berlin: Springer.
Schneider, Friedrich, and Dominik H Enste. 2000. “Shadow Economies: Size, Causes, and Conse-
quences.” Journal of Economic Literature 38 (1): 77–114.
Schneider, F., and C. Williams. 2013. The Shadow Economy. London: The Institute of Economic Affairs,
Edward Elgar.
Stankevičius, Evaldas, and Asta Vasiliauskaitė. 2014. “Tax Burden Level Leverage on Size of the Shadow
Economy, Cases of EU Countries 2003–2013.” Procedia – Social and Behavioral Sciences 156:
548–552. doi:10.1016/j.sbspro.2014.11.238. http://www.sciencedirect.com/science/article/pii/S1877
042814060583.
Su, D. Thanh, P. Canh Nguyen, and S. Christophe. 2019. “Impact of Foreign Direct Investment, Trade
Openness and Economic Institutions on Growth in Emerging Countries: The Case of Vietnam.”
Journal of International Studies 12 (3): 243–264.
Tanzi, Vito. 1982. The Underground Economy in the United States and Abroad. New York, NY: Free Press.
Tanzi, Vito. 1999. “Uses and Abuses of Estimates of the Underground Economy.” The Economic Journal
109 (456): 338–347.
Tedds, Lindsay M., and David E.A. Giles. 2002. “Taxes and the Canadian Underground Economy.” In
Taxes and the Canadian Underground Economy, edited by E. A. G. David and M. T. Lindsay. Toronto:
Canadian Tax Foundation.
Thanh, Su Dinh, and Nguyen Phuc Canh. 2019. “Fiscal Decentralization and Economic Growth of
Vietnamese Provinces: The Role of Local Public Governance.” Annals of Public and Cooperative
Economics 91: 119–149.
Torgler, Benno, and Friedrich Schneider. 2009. “The Impact of tax Morale and Institutional Quality on
the Shadow Economy.” Journal of Economic Psychology 30 (2): 228–245.
Xuefeng, Qian, and Mahmut Yaşar. 2016. “Export Market Diversification and Firm Productivity: Evi-
dence from a Large Developing Country.” World Development 82: 28–47. doi:10.1016/j.worlddev.
2016.01.017. http://www.sciencedirect.com/science/article/pii/S0305750X16000152.
20 P. N. CANH AND S. DINH THANH

Appendix

Table A1. List of countries.

78 Low- and middle-income economies (LMEs)
Albania China Haiti Mali Romania
Algeria Colombia Honduras Mexico Rwanda
Azerbaijan Congo, Rep. India Moldova Senegal
Bangladesh Costa Rica Indonesia Mongolia South Africa
Belarus Cote d’Ivoire Iran, Islamic Rep. Morocco Sri Lanka
Benin Dominican Republic Jamaica Mozambique Tajikistan
Bolivia Ecuador Jordan Namibia Tanzania
Bosnia and Herzegovina Egypt, Arab Rep. Kazakhstan Nepal Thailand
Botswana El Salvador Kenya Nicaragua Togo
Brazil Ethiopia Kyrgyz Republic Niger
Bulgaria Gabon Lao PDR Nigeria Turkey
Burkina Faso Gambia, The Lebanon Pakistan Uganda
Cambodia Georgia Lesotho Paraguay Ukraine
Cameroon Ghana Madagascar Peru Vietnam
Central African Republic Guatemala Malawi Philippines Zambia
Chad Guinea Malaysia
38 High-income economies (HIEs)
Argentina Finland Italy Oman Sweden
Australia France Japan Poland Switzerland
Austria Germany Korea, Rep. Portugal UAE
Belgium Greece Latvia Saudi Arabia United Kingdom
Chile Hong Kong Lithuania Singapore United States
Croatia Hungary Netherlands Slovak Rep. Uruguay
Czech Rep. Ireland New Zealand Slovenia
Denmark Israel Norway Spain

Table A2. Data description for two sub-samples.

Variable Obs Mean SD. Min Max
78 Low- and Middle-Income Economies (LMEs)
SE 936 35.55 10.49 11.74 69.01
ED 936 3.56 1.12 1.76 6.06
EQ 936 0.77 0.15 0.20 1.06
Unem 936 7.70 6.21 0.32 34.07
Taxbur 936 76.96 9.59 46.00 97.60
Inf 936 8.10 8.64 −20.63 100.63
Trade 924 77.62 32.41 22.11 210.37
FDI 936 4.45 5.31 −4.85 55.08
38 High-Income Economies (LMEs)
SE 456 16.61 6.49 6.16 43.18
ED 456 2.38 0.89 1.42 5.47
EQ 456 0.95 0.10 0.53 1.07
Unem 456 7.77 4.16 2.06 27.47
Taxbur 456 67.74 16.70 29.80 99.90
Inf 456 3.20 5.32 −25.13 40.28
Trade 456 104.05 76.94 21.58 442.62
FDI 456 5.65 9.08 −15.99 87.44
Table A3. Variables for robustness check.
Variable Definition Calculation Source Obs Mean SD. Min Max
SEm Shadow economy (currency Shadow economy (% GDP) Alm and Embaye (2013) 364 29.71 9.64 0.00 49.70
demand method)

THE JOURNAL OF INTERNATIONAL TRADE & ECONOMIC DEVELOPMENT

GovEx Governmetn expenditures General government final WDIs 1,358 15.20 5.29 3.46 38.43
consumption expenditure
(% of GDP)
StartupPro Startup procedures Start-up procedures to register WDIs 1,286 8.76 3.36 1.00 20.00
a business (number)
StartupCost Startup cost Cost of business start-up WDIs 1,286 46.64 73.37 0.00 534.8
procedures (% of GNI per
capita)
StartupTime Startup time Time required to start a WDIs 1,286 32.56 30.45 0.50 260.0
business (days)
ProperTime Time to register property Time required to register WDIs 1,183 59.46 80.97 1.00 956.0
property (days)
TaxTime Time to pay taxes Time to prepare and pay taxes WDIs 1,080 304.72 231.58 12.0 2600
(hours)
Democracy Democracy condition Democracy index Polity IV Project 1,343 6.16 3.68 0.00 10.00
Notes: The alternative proxy of shadow economy is collected from Alm and Embaye (2013), which use the currency demand approach. This data is available from 1984 to 2006. Therefore,
the robustness check for alternative proxy of shadow economy is applied for the period 2003–2006 to match with other variables and the period of study. As a result of matching with
other variables, the final sample of alternative proxy of shadow economy includes 91 countries (60 low- and middle-income economies and 31 high-income economies) over the period
2003–2006. WDIs are World Development Indicators database (World Bank); WGIs Worldwide Governance Indicators database (World Bank). The data of startup procedures, startup cost,
start up time, time to register property, time to pay taxes are collected from WDIs, which present data from Doing Business project of World bank as well (http://www.doingbusiness.org/).
The data is available from 2003. The Democracy index is collected from Polity IV Project: Political Regime Characteristics and Transitions (ranking from 0 (least) to 10 (best democracy)).
Available at http://www.systemicpeace.org/polity/polity4.htm.

21
 22
Table A4. Export diversification, export quality and Shadow economy: non-linear evidence – robustness check by various control variables.

P. N. CANH AND S. DINH THANH

Dep. var: SEm (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14)
ED 13.94*** 14.19*** 13.78*** 13.99*** 13.92*** 13.77*** 13.54***
[1.617] [1.623] [1.545] [1.546] [1.596] [1.626] [1.660]
ED2 −1.433*** −1.648*** −1.702*** −1.741*** −1.713*** −1.683*** −1.675***
[0.224] [0.226] [0.215] [0.215] [0.222] [0.226] [0.230]
EQ 131.7*** 126.9*** 136.8*** 143.2*** 141.9*** 139.0*** 136.3***
[12.67] [12.94] [12.3] [12.37] [12.91] [13.54] [14.02]
EQˆ2 −115.1*** −106.0*** −108.9*** −113.4*** −113.0*** −110.4*** −110.5***
[8.440] [8.681] [8.279] [8.286] [8.615] [9.037] [9.433]
GovEx −0.501*** −0.376*** −0.198*** −0.195*** −0.187*** −0.128* −0.162** −0.236*** −0.247*** −0.085 −0.069 −0.059 −0.029 −0.076
[0.063] [0.064] [0.063] [0.063] [0.065] [0.067] [0.071] [0.057] [0.059] [0.058] [0.058] [0.060] [0.062] [0.066]
StartupPro 1.196*** 0.927*** 0.840*** 0.784*** 0.552*** 0.479*** 0.732*** 0.517*** 0.314*** 0.251** 0.119 0.105
[0.094] [0.092] [0.103] [0.110] [0.116] [0.119] [0.091] [0.089] [0.100] [0.106] [0.111] [0.114]
StartupCost 0.053*** 0.052*** 0.052*** 0.060*** 0.057*** 0.048*** 0.044*** 0.044*** 0.052*** 0.050***
[0.005] [0.005] [0.005] [0.006] [0.006] [0.004] [0.004] [0.005] [0.006] [0.006]
StartupTime 0.022* 0.016 0.007 0.006 0.046*** 0.043*** 0.035*** 0.033**
[0.012] [0.013] [0.014] [0.015] [0.011] [0.012] [0.013] [0.013]
ProperTime 0.011*** 0.010** 0.010** 0.007** 0.006 0.007
[0.004] [0.005] [0.005] [0.004] [0.004] [0.005]
TaxTime 0.010*** 0.010*** 0.006*** 0.006***
[0.001] [0.001] [0.001] [0.001]
Democracy −0.134 0.239**
[0.108] [0.102]
Constant 9.109*** −1.495 −2.377 −2.510 −2.733 −4.555 −1.698 5.883 −2.803 −11.74** −13.55*** −12.80** −13.26** −11.62**
[3.079] [3.260] [3.101] [3.098] [3.210] [3.287] [3.505] [4.830] [4.960] [4.795] [4.778] [5.001] [5.219] [5.444]
Observations 1358 1252 1252 1252 1152 1052 1012 1358 1252 1252 1252 1152 1052 1012
No. of countries 116 116 116 116 116 116 114 116 116 116 116 116 116 114
Notes: standard errors are in []; *, **, *** are significant levels at 10%, 5%, and 1%, respectively.
Table A5. Export diversification, export quality and shadow economy: non-linear evidence – robustness check by alternative proxy of Shadow economy (full sample).
Dep. var: SE (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12)
ED 15.43*** 18.60*** 16.32*** 16.29*** 16.74*** 16.64***
[0.506] [0.637] [0.734] [0.742] [0.705] [0.689]
−1.644*** −2.139*** −1.845*** −1.844*** −1.907*** −1.892***

THE JOURNAL OF INTERNATIONAL TRADE & ECONOMIC DEVELOPMENT

ED2
[0.080] [0.096] [0.111] [0.111] [0.105] [0.103]
EQ 140.2*** 150.4*** 125.7*** 125.5*** 140.5*** 138.5***
[4.348] [5.825] [7.175] [7.752] [9.871] [9.398]
EQ2 −110.2*** −116.2*** −98.70*** −98.49*** −108.1*** −106.8***
[2.955] [3.841] [5.089] [5.789] [7.229] [6.936]
Unem 0.511*** 0.508*** 0.508*** 0.509*** 0.508*** 0.430*** 0.442*** 0.441*** 0.455*** 0.458***
[0.033] [0.031] [0.032] [0.032] [0.032] [0.026] [0.026] [0.027] [0.028] [0.029]
Taxbur 0.094*** 0.093*** 0.103*** 0.103*** 0.108*** 0.107*** 0.097*** 0.097***
[0.017] [0.016] [0.015] [0.014] [0.020] [0.018] [0.020] [0.019]
Inf 0.010 0.004 0.004 0.006 0.007 0.006
[0.039] [0.036] [0.036] [0.040] [0.040] [0.040]
Trade −0.021*** −0.020*** 0.007*** 0.012***
[0.002] [0.002] [0.002] [0.003]
FDI −0.018 −0.084**
[0.017] [0.035]
Constant 0.397 −8.127*** −10.98*** −10.91*** −10.52*** −10.41*** −7.861*** −15.57*** −15.35*** −15.33*** −21.01*** −20.21***
[0.684] [1.120] [1.153] [1.009] [1.012] [1.003] [1.498] [2.210] [2.002] [1.980] [2.644] [2.440]
Observations 364 364 364 364 360 360 364 364 364 364 360 360
R-squared 0.316 0.385 0.398 0.398 0.409 0.409 0.294 0.347 0.364 0.364 0.367 0.369
No. of countries 91 91 91 91 90 90 91 91 91 91 90 90
Notes: standard errors are in []. *, **, *** are significant levels at 10%, 5%, and 1%, respectively.

23
 24
Table A6. Export diversification, export quality and shadow economy: non-linear evidence in two sub-samples – robustness check by alternative proxy of shadow economy.
Part A: Export diversification and Shadow economy in two sub-groups

P. N. CANH AND S. DINH THANH

Dep. var: SEm (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12)

Group: Low- and Middle-Income Economies (LMEs) High-Income economies

ED 8.029*** 10.84*** 10.76*** 10.85*** 11.30*** 11.12*** 26.07*** 19.03*** 14.10*** 15.93*** 15.66*** 15.68***
[0.735] [0.863] [0.869] [0.818] [0.821] [0.808] [3.360] [2.321] [2.430] [3.213] [3.290] [3.379]
ED2 −0.840*** −1.244*** −1.238*** −1.233*** −1.289*** −1.262*** −4.891*** −3.315*** −2.624*** −3.239*** −3.206*** −3.210***
[0.103] [0.118] [0.117] [0.114] [0.113] [0.111] [0.665] [0.435] [0.453] [0.656] [0.670] [0.676]
Unem 0.282*** 0.269*** 0.265*** 0.256*** 0.259*** 1.007*** 0.920*** 0.774*** 0.777*** 0.777***
[0.023] [0.030] [0.031] [0.032] [0.032] [0.084] [0.057] [0.056] [0.056] [0.059]
Taxbur −0.037 −0.029 −0.033 −0.030 0.099*** 0.077*** 0.071*** 0.071***
[0.047] [0.043] [0.044] [0.043] [0.019] [0.017] [0.019] [0.019]
Inf −0.060 −0.060 −0.061 0.550*** 0.573*** 0.572***
[0.037] [0.037] [0.038] [0.165] [0.180] [0.185]
Trade −0.007 −0.004 0.004 0.004
[0.006] [0.006] [0.003] [0.004]
FDI −0.043* 0.002
[0.023] [0.032]
Constant 17.60*** 10.83*** 13.88*** 13.54*** 13.69*** 13.65*** −10.53*** −11.12*** −9.754*** −10.20*** −9.887*** −9.922***
[1.208] [1.542] [4.325] [4.149] [4.125] [4.158] [3.956] [3.217] [2.856] [3.558] [3.655] [3.817]
Observations 240 240 240 240 236 236 124 124 124 124 124 124
R-squared 0.089 0.130 0.132 0.139 0.143 0.144 0.128 0.445 0.481 0.544 0.546 0.546
No. of countries 60 60 60 60 59 59 31 31 31 31 31 31
Part B: Export quality and Shadow economy in two sub-groups

Dep. var: SEm (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24)

Group: Low- and Middle-Income Economies (LMEs) High-Income economies

EQ −7.956 10.759 13.347 13.645 4.761 4.532 692.9*** 438.6*** 515.3*** 511.5*** 570.5*** 571.9***
[7.248] [8.092] [9.318] [9.770] [11.662] [11.154] [43.088] [30.271] [42.819] [53.832] [55.719] [57.206]
EQ2 4.749 −8.258 −9.476 −9.909 −4.221 −4.284 −409.5*** −262.1*** −299.1*** −291.9*** −328.7*** −329.5***
[5.599] [6.109] [6.669] [7.113] [8.297] [8.013] [23.28] [16.67] [23.23] [29.29] [31.77] [32.84]
(continued).
Table A6. Continued.
Part B: Export quality and Shadow economy in two sub-groups

THE JOURNAL OF INTERNATIONAL TRADE & ECONOMIC DEVELOPMENT

Dep. var: SEm (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24)

Group: Low- and Middle-Income Economies (LMEs) High-Income economies

Unem 0.238*** 0.216*** 0.217*** 0.204*** 0.210*** 0.699*** 0.650*** 0.649*** 0.637*** 0.637***
[0.022] [0.028] [0.028] [0.030] [0.029] [0.052] [0.046] [0.038] [0.039] [0.038]
Taxbur −0.062 −0.059 −0.058 −0.053 0.111*** 0.108*** 0.074*** 0.075***
[0.047] [0.044] [0.044] [0.041] [0.015] [0.014] [0.014] [0.014]
Inf −0.025 −0.026 −0.028 0.276 0.254 0.252
[0.030] [0.031] [0.032] [0.262] [0.279] [0.282]
Trade −0.000 0.003 0.016*** 0.015***
[0.006] [0.006] [0.002] [0.001]
FDI −0.055 0.009
[0.040] [0.024]
Constant 37.54*** 29.14*** 32.73*** 32.74*** 36.23*** 36.00*** −265.1*** −163.0*** −209.0*** −212.8*** −234.4*** −235.0***
[2.126] [2.548] [2.866] [2.921] [3.972] [3.730] [19.948] [13.496] [20.022] [26.020] [25.579] [25.988]
Observations 240 240 240 240 236 236 124 124 124 124 124 124
R-squared 0.001 0.033 0.038 0.039 0.037 0.039 0.385 0.513 0.565 0.576 0.595 0.595
No. of countries 60 60 60 60 59 59 31 31 31 31 31 31
Note: standard errors are in []; *, **, *** are significant levels at 10%, 5%, and 1%, respectively.

25
26 P. N. CANH AND S. DINH THANH

Figure A1. Export diversification, export quality and shadow economy: non-linear evidence – robustness check
by different proxy of shadow economy (currency demand approach).