Chapter 4

State–Business Relations, Public Policy,

and Industrial Cluster Development in the
Arab World
Mohamed Ismail Sabry

Abstract
This chapter explores how state–business relations (SBR) in the Arab world
influence public policy on industrial clusters and the resulting economic
benefits from these clusters on innovation and productivity. The main SBR
actors are identified as the state, big capitalists (or tycoons), and small and
medium business managers and owners (entrepreneurs). The framework used
here focuses on interactors’ power relations. Such power relations are
reflected in the ability/inability of the state to enforce its will with less
consultation with and/or favorite treatment to nonstate actors. It is also
reflected in the ability/inability of nonstate actors to act individually or
organize in order to defend their interests and influence government policy
formulation and implementation. Such power relations are reflected in the
ability of nonstate actors to organize – especially entrepreneurs – and the
level of favoritism provided by state officials to tycoons. The power
dynamics in SBR lead to different SBR modes and different economic out-
comes, one of which is the development of industrial clusters and how effi-
cient they are in fostering innovation and productivity in Arab countries.
This chapter suggests that the development of industrial clusters in the Arab
world necessitates institutional reform, addressing the power relations gov-
erning SBR in the region. A more open political system allowing the inde-
pendence and growth of broad-based business associations and curbing
favoritism is necessary for realizing the benefits from implementing recom-
mended policies, leading to cluster development and reaping the economic
benefits from such development. Using statistical and regression analysis,
empirical evidence supporting these arguments is provided.

Industry Clusters and Innovation in the Arab World, 99–127

Copyright © 2023 Mohamed Ismail Sabry
Published under exclusive licence by Emerald Publishing Limited
doi:10.1108/978-1-80262-871-520231006
100 Mohamed Ismail Sabry

Keywords: Industrial clusters; state–business relations; Arab world; state;

businesspeople; cronyism; state capture; state dominance; public policy;
innovation; productivity

Introduction
Industrial clusters have been often presented as a viable mean to tackle various
economic problems of many developing countries. The likely positive effect of
industrial clusters on innovation, productivity, and diversification calls attention.
Moreover, these clusters’ perceived success in boosting the growth of small and
medium enterprises (SMEs) and their job creation potential add to their attraction
(see Porter, 1998). Accordingly, there is a growing interest in the Arab world in
industrial clusters. As pointed out by the Arab World Competitiveness Report
(World Economic Forum, 2018), the challenges presented by the Fourth Indus-
trial Revolution urge Arab countries to promote the largely underdeveloped
innovative activities by extending financial services to innovators and risky pro-
jects, diversify, and support the development of the private sector and especially
micro, small, and medium enterprises (MSME). Promoting industrial clusters
provides a promising strategy to meet such challenges.
Such an interest in industrial clusters raises the question on how to foster the
development of these clusters. Is implementing recommended policies sufficient to
realize the anticipated benefits from industrial clusters? Or rather what is needed is
a much deeper investigation to the institutional settings of the Arab countries, an
investigation that would guide our understanding of what could maximize the
economic benefits from industrial clusters. This chapter explores the effect of
state–society relations in the Arab World on public policy targeting the devel-
opment of industrial clusters and the resulting benefits from such development on
innovation and productivity. Given the focus on industrial clusters, the investi-
gation presents a novel framework that places state–business relations (SBR) at
the center of state–society relations. In this framework, the main actors in SBR
are identified as the state, big capitalists (or tycoons), and MSME managers and
owners (entrepreneurs). The framework used here focuses on interactors’ power
relations. Such power relations are reflected in the ability/inability of the state to
enforce its will with less consultation with and/or favorite treatment to nonstate
actors. It is also reflected in the ability/inability of nonstate actors to act indi-
vidually or organize in order to defend their interests and influence government
policy formulation and implementation. Such power relations are better reflected
in the ability of nonstate actors to organize – especially entrepreneurs – and the
level of favoritism provided by state officials to tycoons. The power dynamics in
SBR lead to different SBR modes and different economic outcomes, one of which
is the development of industrial clusters and how efficient they are in fostering
innovation and productivity in Arab economies. The presented framework builds
on the literature on SBRs (Adly, 2010; Evans, 2012; Haggard, Maxfield, &
Schneider, 1997; Hausmann, Rodrik, & Sabel, 2011; Hausmann & Rodrik, 2003;
 Development in the Arab World 101

Schneider, 2005) and political settlements (Behuria, Buur, & Gray, 2017; Khan,
2018).
The discussion supported by the empirical results of this research suggests that
the development of industrial clusters in the Arab world necessitates institutional
reform addressing the power relations guiding SBR in the region. A more open
political system allowing the independence and growth of broad-based business
associations and curbing favoritism is necessary for realizing the benefits from
implementing recommended policies, leading to clusters’ development and reap-
ing the economic benefits from such development. The chapter starts with the
theoretical perspective discussing how SBR influences public policy related to
industrial clusters – in terms of policy formulation, implementation, and effec-
tiveness – and how this affects the development of clusters and the resulting
economic benefits in the Arab world. This is followed by the empirical section
where data obtained from the World Bank and the Global Competitiveness
Indicators (GCI) are used to conduct statistical and regression analysis on the
subject. The chapter ends with a concluding section, summarizing the empirical
findings, bringing them in line with the literature, and providing policy
recommendations.

Theoretical Perspective
Public Policy for Industrial Clusters
As defined by Porter (1998, pp. 78–83), clusters refer to geographical agglomer-
ation of interconnected firms, business associations, government entities, univer-
sities, and other institutions active in a specific field. Interfirm linkages and
synergies are keys for the success of clusters. Industrial clusters are suggested to
foster productivity by allowing better sourcing of inputs, easier access to tech-
nology and specialized information, interfirm coordination, and the use of public
goods such as public professional training and specialized infrastructure. They
should also foster innovation because of the presence of sophisticated buyers,
easiness of sourcing inputs needed for innovation, and the ability to engage in
small-scale and low-cost experimenting.
Different public policies foster the development of industrial clusters. These
policies are often distinguished from industrial policy (Porter, 1998, p. 89), which
refers to any government intervention aiming at improving the business envi-
ronment, restructuring economic activity to foster growth or welfare (Warwick,
2013). Public policies targeting clusters’ development include spending on infra-
structure and human development, enacting competition-fostering legislations
such as intellectual property rights (IPR) protection and antimonopoly laws
(Porter, 1998, p. 89), providing financing and tax incentives, and facilitating
intrafirm knowledge exchange and linkages (Lang, 2009). Rather than being
selective in its intervention, as is the case often with industrial policy, public policy
should aim at fostering all industrial clusters (Porter, 1998, p. 89).
If the development of industrial clusters is regarded as a mean to promote the
growth of SMEs in developing countries, the growth of SMEs within industrial
102 Mohamed Ismail Sabry

clusters could further be strengthened by various policies that create a better

business climate (Yoshino, 2011). This includes credit facilitation (Abe, Troilo, &
Batsaikhan, 2015) and market entry deregulation (Ayyagari, Beck, &
Demirguc-Kunt, 2007). Easing restrictions on venture capital is also important
for SMEs (Koreen & Lucia, 2015) and especially in industrial clusters (Benner,
2012, p. 3). IPR protection should also benefit innovative SMEs that could
temporarily secure a relatively bigger market share in contrast to tycoons that
rather seek higher market barriers to defend their market share (Greenhalgh &
Rogers, 2007, p. 548).
A body of literature points to another important dimension for fostering
cluster development. Given the need for interfirm cooperation and coordination,
the development of social capital becomes essential for industrial clusters. If
well-functioning formal institutions – such as rule of law – create generalized trust
and social capital, the absence of these institutions in developing countries calls
for networking – formal or informal – as an alternative to build this capital.
Business associations are functional in realizing this objective, especially if they
are unrestricted and representative for businesspeople and their interests (Faf-
champs, 2006). Such broad-based and representative business associations are
also needed as policy consultation partners for state officials. It is argued that the
local requirements of industrial clusters raise doubts on the effectiveness of
applying similar policy packages, accounting for local condition calls for
public–private dialogue in order to formulate policies that fit each industrial
cluster (McDonald, Tsagdis, & Huang, 2006; Raines, 2002). The downside risk is
the possible rent-seeking and ineffective allocation of resources to clusters
resulting from such state–business partnership (McDonald et al., 2006).

State–Business Relations and Public Policy for Clusters

Policies are arguably shaped by state–society relations. In this regard, business-
people are more likely to play a big part in guiding and influencing the policy
formulation process. SBR tackle business and government failures (Hausmann
et al., 2011; Hausmann & Rodrik, 2003), but there is often the risk that resulting
policies would lead to rent seeking, corruption, and favoritism benefitting con-
nected firms (Rodrik, 2008).
By the state, the author refers here to officials, bureaucrats, and ruling party
politicians. Businesspeople are subcategorized into two distinctive groups based
on market share: tycoons – or big business managers and owners – and entre-
preneurs – referring here to managers and owners of MSMEs. Hence, the main
actors in SBR are identified here as the state, tycoons, and entrepreneurs (as
defined above).
Tycoons and entrepreneurs have often conflicting goals. For instance, while
tycoons usually benefit from more restrictive markets that allow them a bigger
and less endangered market share (see Peltzman, Levine, & Noll, 1989; Shleifer,
2005; Stigler, 1971), entrepreneurs are more likely to benefit from more
competitive markets with easy entry and exit. The relative power of tycoons and
 Development in the Arab World 103

entrepreneurs fosters their capability to influence public policy. This power

dynamics between tycoons and entrepreneurs, on the one hand, and between each
of them and the state, on the other hand, are crucial for shaping public policy.
Organizational strength is one of the sources of such power. Independent and
broad-based business associations with balanced membership and power distri-
bution between tycoons and entrepreneurs would strengthen both actors’ power
vis-à-vis the state. Yet, imbalance of power within business associations in the
favor of tycoons or the presence of business associations whose membership is
exclusive for tycoons would strengthen tycoons’ power vis-à-vis both the state and
entrepreneurs (see Shadlen, 2002). The opposite is true for business associations
with exclusive membership of MSMEs; such associations should foster entre-
preneurs’ power vis-à-vis the other two actors. Tycoons, however, might not need
organizational strength to influence policy, and they possess individual and
informal ways of influence. They could finance election campaigns, offer signifi-
cant shares in their firms to officials, or assume leading government positions
(Schneider, 2005, 2009, 2015).
As suggested earlier, policies fostering the growth of MSMEs should lead to
industrial cluster development. Thus, business associations representing the
interests of MSMEs would likely push for such policies. However, the enactment
of these policies is dependent on the relative power of these entrepreneur-led
associations vis-à-vis tycoons and the state. Even when policies that are beneficial
for entrepreneurs are enacted, tycoons’ power could subvert these policies and
turn implementation to their favor. They could eradicate their competitors by
utilizing the economic and noneconomic resources under their disposal, with
means such as monopoly pricing, social exclusion, and manipulation of the
judicial system (Shleifer, 2005). Whenever tycoons have strong and unmatched
access to the state officials – facilitated by lower levels of political accountability –
they could capture reforms that are supposedly directed to entrepreneurs
(Claessens & Perotti, 2007). On the other hand, a powerful state vis-à-vis nonstate
actors would likely have a higher power of discretion in implementing policies.
Thus, the gap between policy enactment and actual implementation is expected to
be greater whenever the state is relatively more powerful. In such conditions,
actual policy implementation matters more than policy formulation and
enactment.
The power dynamics between the three considered actors lead to different
modes of SBR. The presence of a powerful state that provides favorable treatment
for connected tycoons is a situation that is often referred to in the literature as
cronyism (Begley, Khatri, & Tsang, 2010; Desai & Olofsgård, 2011). In this SBR
mode, the state is likely to enforce high sectoral regulations and hinder compe-
tition in order to protect tycoons’ interests at the expense of entrepreneurs
(Djankov, La Porta, Lopez-de-Silanes, & Shleifer, 2002; Enderwick, 2005). This is
likely to have negative consequences on economic outcomes, including innovation
and technological upgrading (Desai & Olofsgård, 2011; Enderwick, 2005), but
they should also be expected to be detrimental for industrial cluster development.
Even if the state formulated and enacted policies that foster MSMEs growth, the
power of the state that permits discretion and the power of tycoons vis-à-vis
104 Mohamed Ismail Sabry

entrepreneurs that allows the former to subvert reforms might lead to poor
implementation and insignificant or even a negative impact of these policies on
industrial clusters’ development.
Another possible SBR mode – one referred to in the literature as state capture
– is one where tycoons are the most powerful actor, even more than the state, to
the extent that they dominate policy formulation (Adly, 2010; Enderwick, 2005).
The less powerful state should be less expected to exercise discretion, but the more
powerful tycoons would push for policies that promote their interests. The effect
again might be detrimental to industrial clusters’ development, this time in terms
also of formulated policies rather than only actual implementation.
Another possible mode is one where the state is by far the most powerful actor,
to the extent of preventing any nonstate actor from any significant share of power.
This is here referred to as state dominance. The enactment of policies fostering
MSME growth should not be met by much resistance from the less powerful
tycoons. Moreover, although the uncontested power of the state enables it to
exercise discretion, it has little interest in favoring tycoons and it is more likely
that it fosters MSME growth as a mean for fostering industrial cluster develop-
ment and industrial sector growth.
Different SBR modes would induce the adoption of different public policies
related to industrial clusters and to different levels of implementation of these
policies. A highly dominant and autonomous state is less likely to cooperate with
businesspeople, and the latter would lack the organizational capacity for sup-
porting interfirm networking. It is likely then to have a top-down approach based
on state initiative to construct clusters from scratch. The economic outcomes from
the resulting clusters might suffer from the lack of business involvement in policy
consultation and underdeveloped business networking, which might result in low
gains in terms of innovation and productivity. The approach of building clusters
from scratch and conducting a pure top-down approach to promoting clusters is
often criticized in the literature and perceived as being too costly and leading to
poor economic gains (see Benner, 2012).
The presence of business associations with significant organizational power
would induce state–business partnership in cluster development public policy. The
more broad-based business associations and the more representative to entre-
preneurs they are, the more social capital and networking gains created (Faf-
champs, 2006). The benefits in terms of productivity and innovation should be
maximal. The partnership between the state and businesspeople could also be
realized with informal connections between state officials and tycoons that are
nurturing on favorite allocation of resources. Under poor institutional settings,
such partnership is likely to slide into mere crony allocations to connected cluster
firms (see Benner, 2012; Enright, 2003). The economic benefits from clusters
development could then be minimal, especially in terms of innovation. Interfirm
power imbalances are generally suggested to lead to hierarchical and dependency
relations between dominant firms and smaller ones and to harm cluster stability
and development (Lang, 2009). Arguably, favoritism to tycoons could exacerbate
power imbalances between tycoons and MSMEs within a cluster. Lang (2009)
suggests that MSMEs could diminish such imbalances by innovation. In this case,
 Development in the Arab World 105

it is arguably important to enforce policies that protect innovation, such as IPR

protection.

SBR and Industrial Clusters in the Arab World

SBRs in the Arab countries have some specific common characteristics that seem
to have considerable impact on industrial clusters’ development. A relatively
powerful state and less organizationally powerful and dependent nonstate actors
are distinctive features of the whole region, with some few and short-lived
exceptions (see Ayubi, 2001; Cammett, Diwan, Richards, & Waterbury, 2018;
Hertog, 2013b; Sabry, 2018b, 2019a). Much of the region are additionally char-
acterized by higher levels of favoritism to connected nonstate actors, and espe-
cially tycoons (Adly, 2010; Cammett et al., 2018; Chekir & Diwan, 2015; Rijkers,
Freund, & Nucifora, 2017; Sabry, 2018b). Thus, SBR in the whole region could
only be categorized almost exclusively to either crony or state-dominance modes.
The powerful position of the state is the result of the special conditions of the
region’s political development. One of the reasons for such an outcome was the
colonial legacy and modernization trials preceding and following the colonial
period. European colonial powers that at a time ruled the whole region – with the
exception of Saudi Arabia and North Yemen – were interested in establishing a
powerful bureaucracy to aid them in administrating their colonies. Such an
abnormal, rather than evolutionary, development of the bureaucracy took place
while other nonstate actors, and especially the national bourgeoisie, were still
largely underdeveloped (Ayubi, 2001, pp. 12–13). The second factor is the enor-
mous oil wealth that boosted the economies of much of the region. In the Gulf
region, for instance, the state-controlled oil wealth strengthened the state vis-à-vis
businesspeople in contrast to the preoil boom more balanced relations between
the two actors (Almezaini, 2013; Azoulay, 2013; Valeri, 2013). The adoption of
“socialism” by a number of Arab republics was accompanied by widescale
nationalizations and further weakened the political and economic power of
businesspeople (see Ayubi, 2001; Cammett et al., 2018).
Favoritism to tycoons, on the other hand, evolved with the big economic
changes following structural adjustment programs and more market-oriented
reforms. Privatization offered Arab states with a significant chance to provide
economic favors in the form of privatized assets to connected tycoons (Adly,
2010; Cammett et al., 2018, p. 310; Chekir & Diwan, 2015). However, busi-
nesspeople, whether tycoons or entrepreneurs, remained subservient for a rela-
tively more powerful state. Businesspeople are largely unable to aggregate and
lobby in front of the state (Nabli, 2007, p. 95). They are largely dominated by
tycoons who are interested in defending their privileges and arranging “individual
deals” with the state (Hertog, 2013b, p. 6). Discussing pre-2011 Egypt, Adly
(2012, pp. 91–108) pointed out that public–private dialogues are mere extensions
of informal connections, denied significant autonomy, and mainly advisory.
When they influence policy formulation, this is mainly done by informal channels
(Zovighian, 2013, p. 187).
106 Mohamed Ismail Sabry

The movement toward a more crony SBR mode was more evident in former
“socialist” republics, such as Egypt (Adly, 2010; Sabry, 2019a), Algeria (Cammett
et al., 2018, p. 310), and Syria (Heydemann, 2013; Hinnebusch, 2012). The Gulf
states, on the other hand, had lower levels of favoritism to tycoons and voice and
accountability (V&A), as suggested by the data on the two indicators collected
from the GCI (World Economic Forum, n.d.) and the World Bank’s Worldwide
Governance Indicators (WGI), respectively. These states are, thus, more likely to
be examples of the state-dominance SBR mode. While strong connections with
state officials are crucial for tycoons’ business success (Almezaini, 2013, p. 44), the
rich oil resources ensure that the whole society benefits from state support such
that tycoons’ privileges should not be necessarily provided at great societal sac-
rifice. Business associations are controlled by the state that supervises their
boards’ elections and could appoint their boards’ members, while business rep-
resentatives to economic councils are likely to be selected by the state (Hertog,
2013a, pp. 183–185).
In some Arab countries, however, businesspeople enjoyed some organizational
power. This is remarkably the case in post-Arab Spring Tunisia, but also to some
extent in Morocco and Jordan. The most prominent business association in
Tunisia, UTICA, plays a significant role on the economic and political arenas in
the country (Shiaty, 2014; UTICA, 2015). Morocco has traditionally a relatively
open political system that allowed a degree of state–business partnership and
dialogues and business collective lobbying (Cammett, 2007, pp. 6–13). Collective
lobbying from big business associations has been traditionally active in Jordan,
although it relatively subsided after the 1960s when the state gained more power
because of rents flowing from foreign aid (Moore, 2001, pp. 135, 140).
The three countries, however, have higher levels of favoritism to tycoons, as
suggested by the literature (Hinnebusch, 2015, p. 20; Loewe, Blume, & Speer,
2008, pp. 259–267; Oubenal & Ben Hamouda, 2018; Saadi, 2016) and GCI data.
The development of post-Arab Spring Tunisia suggests that it could be the only
case of state capture SBR in the Arab world. Tycoons in postrevolutionary
Tunisia sought political influence through political party membership and par-
liamentary elections (see Oubenal & Ben Hamouda, 2018). The scores of the
country in V&A and favoritism further support this claim.
Such SBR settings impact on the development of industrial clusters in the
region. Contrasting pre-2011 Tunisia and Morocco, Cammett (2007) argued that
government’s active involvement in Tunisia led to an initially comparatively
higher development for industrial clusters in the country in comparison to
Morocco. However, the weakness of business coordination harmed the further
development of these clusters, especially at a stage where innovation was needed.
In Morocco, on the contrary, business associations eventually helped in the
development of industrial clusters that have emerged from business initiatives and
the collaboration between the state and businesspeople. As pointed out by Benner
(2012), cluster agents cooperate in order to acquire Moroccan state support in the
form of infrastructure investment and other measures. The state’s strategy seeks
promoting collaborative research and development (R&D), innovation-inducive
 Development in the Arab World 107

environment, and global reach, and it is these concerns and similar ones that
guide the choice of which clusters to support.
In Saudi Arabia, public policy for promoting cluster development is a
top-down approach taking the form of creating new “economic cities” with heavy
investment in infrastructure and less investment on social networking – even
informal forms of it (Benner, 2012, p. 12). The United Arab Emirates (UAE)
adopts a strongly state-led cluster promotion policy relying on special economic
zones (SEZ) with inviting business climate, incentives, and infrastructure invest-
ment (Al-Saleh, 2018). Algeria, as Saudi Arabia, constructs a cluster promotion
policy based on building new cities (Benner, 2012, pp. 15–16).
In the previous examples, public policy for promoting clusters are likely
influenced by the SBR modes existing in the mentioned Arab countries. Orga-
nizationally weak businesspeople would not likely be considered as partners in
cluster development especially when the dominant state is more autonomous and
less involved in favoritism to tycoons. In such conditions, a top-down approach is
more expected, as is the case in Saudi Arabia. In countries with relatively more
powerful business associations and/or where more favoritism to tycoons is prac-
ticed, a public policy that involves businesspeople – whether tycoons or entre-
preneurs – would be witnessed, as is the case of Morocco.
The concerns on both state-led top-down and state–business partnership
approaches for promoting cluster formation should apply to the experience of the
Arab world with promoting industrial clusters. Many works found that cronyism
tends to negatively affect innovation in Arab countries (Francis, Hussain, &
Schiffbauer, 2018; Saadi, 2016; Sabry, 2018b), and this could be assumed to apply
to the expected broad economic benefits from industrial clusters’ development. In
this case, public policy for promoting industrial clusters could be diverted into
rent-seeking and favorite treatment for connected firms (see Benner, 2012;
Enright, 2003; McDonald et al., 2006).
Industrial cluster–related public policies in the Arab world seem also to pose
further impediments for clusters’ development. Although major improvement is
witnessed in the infrastructure of many Arab countries, the persistence of a
complex regulatory environment and barriers to competition and the presence of
a less-educated workforce deter private investment in the region. Entrepreneurs
who suffer from the lack of access to finance and venture capital – and especially
angel investment that supports start-ups – should also be promoted (World
Economic Forum, 2018). As argued earlier in this chapter, the interaction of these
policies with SBR leads to the witnessed levels of policy enactment, imple-
mentation, and effectiveness and ultimately the varying levels of cluster devel-
opment and economic benefits from these clusters in the region.

Data and Discussion

The following section depends mainly on data collected from the GCI and the
World Bank’s WGI, Doing Business (DB), and World Development Indicators
(WDI) datasets for the 22 Arab countries between the years 2000 and 2017. A full
108 Mohamed Ismail Sabry

account of the used proxies and the sources from which they are obtained is listed
in the appendix (Table A1).1
The V&A and favoritism indicators are obtained from the WGI and GCI,
respectively. Using the averages of the years 2015–2017 to construct XY coor-
dinates, Fig. 4.1 shows the SBR characteristics of different Arab countries. Both
indicators are rescaled into a scale ranging between 250 and 150 for the lowest
and highest values, respectively.2 The following could be noted:

10

Tunisia
5

0
-30 -20 -10 0 10 20 30 40

-5

Lebanon
-10
Morocco Kuwait
Jordan
-15
V&A

Algeria
Oman
UAE Mauritania
-20

Egypt
Bahrain
-25
Qatar Libya

-30
Yemen

Saudi Arabia
-35

Syria

-40

-45
Favoritism

Fig. 4.1. V&A, Favoritism, and SBR Modes in the Arab World
(2015–2017).
 Development in the Arab World 109

(1) Up-right (favoritism.0, V&A.0): Countries of this part of the graph are
more likely characterized by state capture SBR, where an open political
system allows organizational power, but tycoons are able to secure favorite
treatment at the expense of others. Fig. 4.1 shows that postrevolutionary
Tunisia is the only country in this category.
(2) Down-right (favoritism.0, V&A,0): Countries in this category are likely
characterized by crony SBR, where the state is dominant, nonstate actors
have little organizational power, and tycoons receive favorite treatment.
Most of the Arab countries fall in this category. Fig. 4.1 suggests that this
SBR mode is especially evident in Yemen, Libya, and Syria; modest in
Kuwait, Mauritania, and Algeria; and weakest in Egypt, Jordan, and
Morocco. Lebanon, interestingly, has the highest level of favoritism in this
group of countries (and actually the whole Arab region), but has also the
highest V&A among the countries of this mode.
(3) Down-left (favoritism,0, V&A,0): SBRs here are reflecting the
state-dominance mode with a dominant state, weak nonstate actors, and less
favoritism to tycoons. The countries of this mode are predominantly Gulf
states. UAE and Qatar are the best representatives of this mode. The mode is
less evident in Bahrain and Saudi Arabia and least evident in Oman, whose
position is near the borderline with crony SBR.

Moving to the state of clusters’ development in the Arab world, Fig. 4.2 uses data
obtained from the GCI on cluster development in 2015–2017 in the region. The
indicator is rescaled into a percentage scale. The best performer is UAE followed by
Qatar, then Saudi Arabia, and Bahrain. The four countries were categorized earlier
under the state-dominance SBR mode. The worst performers are Libya, Yemen,
Syria, Mauritania, and Algeria, respectively. This largely corresponds to the strength
of the crony SBR mode in these countries. Thus, the data suggest that there is a high
correlation between state-dominance and higher clusters’ development and between
cronyism and lower clusters’ development. Since the main difference between the two

80
70
60
50
40
30
20
10
0

Fig. 4.2. Clusters’ Development in the Arab World (2015–2017).

110 Mohamed Ismail Sabry

80
70
60
Indust. Clusters

V&A and Clusters

50
40 Favorism and Clusters
30
20 Linear (V&A and Clusters)
10
Linear (Favorism and
0
Clusters)
0 20 40 60 80 100
V&A
Favorism

Fig. 4.3. V&A, Favoritism, and Industrial Clusters’ Development in

the Arab World (2015–2017).

modes is the level of favoritism, it could also be said that a correlation exists between
higher favoritism and lower clusters’ development.
Fig. 4.3 shows the relation in the Arab world between V&A and favoritism on
one the hand and clusters’ development on the other hand. Using the averages of
the period 2015–2017 of each Arab country, best-fit lines are constructed to
suggest the slope of these relations (assuming having simple bivariate relations).
The first best-fit line suggests the presence of a negative relationship between V&A
and clusters with a rather flat slope. The line showing the relation between
favoritism and cluster development, however, has a more steep slope, providing
further evidence about the existence of a negative correlation between the two
indicators.
Further insights could be reached using multivariate regression analysis. Two
sets of regressions are conducted. The first, set {A}, is testing the effect of V&A,
favoritism, and different policies on clusters’ development. The general equation
for set {A} regressions is
C ¼ a 1 b1 ðVAÞ 1 b2 ðFavÞ 1 b3 ðPÞ 1 b4 ðVApPÞ 1 b5 ðFavpPÞ (1)

C stands for industrial clusters and P for a matrix of different policies. The
policies considered here are those suggested by the literature. These are regula-
tions for getting credit, starting business, venture capital, and IPR. The interac-
tion terms of V&A and favoritism with these policies are used also as regressors.
Infrastructure and government spending on education are used as control vari-
ables without considering their interaction terms. Arguably, while the earlier four
policies are more directly related to SBR actors – especially conflicting policy
objectives with regard to tycoons and entrepreneurs – the latter two policies are
more general and should benefit both actors. Thus, studying how the existence of
favoritism and V&A could alter the implementation of the earlier four policies in
 Development in the Arab World 111

a way that reverses the effects of these policies on entrepreneurs and clusters’
development is at the center of this investigation.
The fixed effects panel regression model is used to conduct set {A} regressions.
The choice is guided by the concern on disregarded variables and the recom-
mendations of statistical tests as documented at the appendix.3 Sequential elim-
ination of insignificant regressors, provided that the process did not alter the
regressions’ consistency, is done automatically by the gretl program, and the
results of the test are documented in the appendix. Table 4.1 shows the results of
the conducted regressions for this set. Supporting statistical tables are provided at
the appendix (Tables A2, A3, A4 and A5).
The first regression investigates the effect of the studied policies when favor-
itism and V&A are not accounted for. The regression results show the expected
(strongly significant) effect of four public policy measures on clusters’ develop-
ment, namely venture capital, IPR, infrastructure, and starting business, ordered
in terms of the strength of their effect. In the second regression, more interesting
results are shown. The interaction terms of V&A with venture capital and starting
business have a positive effect, but the effect is negative for the interaction with
IPR. For favoritism, it is largely the opposite picture. The effects of the inter-
action terms of favoritism with getting credit and venture capital are negative, but
the interaction with IPR has a positive effect. Accordingly, V&A is more likely
important for reaping the positive effect of venture capital availability and the
easing of starting business on clusters’ development. Both policies are crucial for
MSME growth, but their potential would only be likely realized, as suggested by
the results, whenever entrepreneurs have some organizational power that would
facilitate networking and coordinate their efforts to defend their interests. To the
contrary, the presence of high favoritism suppresses the perceived positive effect
of credit and venture capital availability on clusters’ development. It is likely that
tycoons sabotage reform in these areas and direct it to their benefit, preventing the
growth of MSMEs and clusters’ development. Nevertheless, IPR regulations seem
to be effective in overcoming the negative effect of favoritism on clusters’
development. Surprising is, however, the negative effect of the interaction term
between V&A and IPR. This raises the question of whether this result is obtained
because IPR manages to overcome the effect of lower V&A – that is to say, higher
levels of authoritarianism. This is a valid claim since lower levels of V&A are
characteristic of state-dominance Gulf countries, which are among the best per-
formers in the Arab world in clusters development, as revealed earlier.
To test the validity of this claim, the third regression of set {A} is run,
replacing V&A and its interaction terms with an indicator for authoritarianism.
The latter is constructed by subtracting the V&A percentage value from 100, and
thus the authoritarianism and V&A indicators provide the opposite image of each
other. The results of the regression support the earlier claim, the interaction of
IPR with authoritarianism is positive, the same as it is for the interaction with
favoritism, and the overall effect of IPR is 0.447.4
Fostering clusters’ development is one thing and the efficiency of these clusters
in boosting innovation and productivity is another. Fig. 4.4 shows the correlation
between clusters’ development and innovation in the Arab world for the years
112 Mohamed Ismail Sabry

Table 4.1. Set {A} Regressions.

Fixed effects
Dependent Variable: Cluster Development
Const 25.467** 21.710 22.623
(2.616) (3.360) (3.440)
Getting Credit – 0.176* 0.180*
(0.092) (0.092)
Starting Business 0.063*** 20.171*** 0.578***
(0.017) (0.057) (0.129)
Venture Capital 0.488*** – 2.069***
(0.052) (0.353)
IPR 0.360*** 0.899*** 21.834***
(0.052) (0.124) (0.371)
Infrastructure 0.236*** 0.206*** 0.210***
(0.060) (0.071) (0.071)
Favoritism*Credit – 20.003** 20.003**
(0.001) (0.001)
Favoritism*Venture Cap. – 20.004** 20.005***
(0.002) (0.002)
Favoritism*IPR – 0.006*** 0.007***
(0.002) (0.002)
V&A*Start Bus. – 0.008*** –
(0.002)
V&A*Venture Cap. – 0.022*** –
(0.003)
V&A*IPR – 20.028*** –
(0.004)
Authoritarian*Start Bus. – – 20.007***
(0.002)
Authoritarian*Venture Cap. – – 20.018***
(0.005)
Authoritarian*IPR – – 0.026***
(0.005)
n 270 269 269
LSDV R-squared 0.876 0.892 0.892
Within R-squared 0.594 0.645 0.647
Note: Standard errors in parentheses, ***p , 0.01, **p , 0.05, *p , 0.1.
 Development in the Arab World 113

70

Qatar
60
UAE

50
Jordan Saudi Arabia

Oman Bahrain
40
Innovation

Lebanon
Tunisia
Morocco
Algeria Kuwait
30
Syria
Egypt
Mauritania
20
Libya Yemen

10

0
0 10 20 30 40 50 60 70 80
Cluster Development

Fig. 4.4. Clusters’ Development and Innovation in the Arab World

(2015–2017).

2015–2017. Referring to the best-fit line, the figure suggests a positive correlation
between the two variables. On top in terms of the two indicators are the UAE and
Qatar and at the bottom is Yemen.
Fig. 4.5 shows the correlation between clusters’ development and (log.) labor
productivity in the Arab world for the same period. The best-fit line reveals also a
positive correlation between the two variables. Disregarding offshoots from the
correlation line, the UAE is on the highest point in the line and Yemen at the
lowest point.
To investigate the relation between clusters’ development – on the one hand –
and innovation and productivity – on the other hand – set {B} regressions are
conducted. The general equation used for this set is
Y ¼ a 1 b1 ðVAÞ 1 b2 ðFavÞ 1 b3 ðCÞ 1 b4 ðVApCÞ 1 b5 ðFav  CÞ (2)

Y in this equation stands for economic outcomes, which are innovation in

regression B-1-1 and B-1-2 and (log.) labor productivity in regressions B-2-1, B-2-
2, and B-2-3. The fixed effects model is used in regressions B-1-1 and B-2-1, the
random effects in B-2-2, and the two-stage least squares (TSLS) in B-1-2 and B-2-
3. The random effect model is used in B-2-2 based on statistical test suggestions
114 Mohamed Ismail Sabry

6
Kuwait Saudi Arabia UAE
Algeria
Oman
5
Mauritania Morocco
Bahrain
Yemen
Log Productivity

4
Jordan
Tunisia Lebanon Egypt
3

0
0 10 20 30 40 50 60 70 80
Cluster Development

Fig. 4.5. Clusters’ Development and Productivity in the Arab World

(2015–2017).

(the results of the tests are reported in the appendix). Nevertheless, the results of
the fixed effects and random effects models are almost identical for the regression
having labor productivity as a dependent variable (regressions B-2-1 and B-2-2).
To check for robustness, the TSLS model is used in regressions of the two
considered dependent variables (regressions B-1-2 and B-2-3). The use of this
model is based on concerns about the presence of reverse causality between
clusters’ development on the one hand and innovation and productivity on the
other hand. Arguably, the concerns are rather minimal, since it is less likely that
innovation and productivity foster clusters’ development rather than the other
way round. A number of geographical, historical, cultural, and other exogenous
variables are used as instruments (a full list of these variables is reported in the
appendix). Sequential elimination is also used for the regressions of this set (the
process is reported in the appendix). The results of the conducted regressions of
this set are reported in Table 4.2.
The results of regression B-1-1 in Table 4.2 shows that clusters’ development
without interaction terms increases innovation, yet the interaction term of clus-
ters’ development with favoritism harms innovation. V&A has a negative effect
on innovation. Calculating the overall effect of favoritism shows that it has also a
negative effect (20.029) on innovation. The overall effect of clusters’ development
is 0.328, and thus clusters’ development boosts innovation, even when its effect is
downplayed by favoritism. Testing for robustness, regression B-1-2 – using TSLS
– confirms the negative effect of favoritism and positive effect of clusters’ devel-
opment on innovation. The effect of V&A is, however, positive in this regression
and none of the interaction terms is significant.
Concerning the effects on productivity, the results of regressions B-2-1 and
B-2-2 show that the overall effect of clusters’ development is 20.003, V&A is
Table 4.2. Set {B} Regressions.

Regression B-1-1 B-1-2 B-2-1 B-2-2 B-2-3

Model Fixed effects TSLS Fixed effects Random effects TSLS
(GLS)
Dependent Variable: Innovation Innovation Log Labor Log Labor Log Labor
Productivity Productivity Productivity
Const 22.077 40.13*** 5.205*** 5.204*** 4.589***
(4.310) (5.488) (0.094) (0.161) (0.244)
V&A 20.188*** 0.145*** 20.015*** 20.015*** 20.127***
(0.046) (0.054) (0.003) (0.003) (0.016)
Favoritism 0.447*** 20.329*** 20.003*** 20.003*** 0.014***
(0.064) (0.038) (0.000) (0.000) (0.004)
Cluster Development 1.005*** 0.253*** 20.014*** 20.014*** –

Development in the Arab World

(0.085) (0.086) (0.002) (0.002)
V&A*Cluster – – 0.000*** 0.000*** 0.002***
(0.000) (0.000) (0.0003)
Favor*Cluster 20.011*** – – – –
(0.002)
n 216 198 162 162 162
R-squared 0.903 0.681 0.994 0.256
Within or Adjusted R- 0.631 0.676 0.360 0.242
squared
Note: Standard errors in parentheses, ***p , 0.01, **p , 0.05, *p , 0.1.

115
116 Mohamed Ismail Sabry

0.0015, and favoritism 20.003. V&A, thus, boosts productivity. The positive
effect of the interaction term of V&A and clusters’ development shows that V&A
is important for boosting productivity gains resulting from clusters’ development.
Clusters’ development without V&A is likely to lead to less productivity. Testing
for robustness, the results of regression B-2-3 using TSLS confirms the effect of
the interaction of clusters’ development and V&A on labor productivity, while
showing different signs for other variables (V&A overall effect is 20.04 and
favoritism is 0.014).

Conclusion
The development of industrial clusters in the Arab world does not depend only on
good policies. While it is true that some recommended policies – such as IPR
protection – can foster the development of clusters, recommended policies are not
sufficient to realize this objective. The characteristics of SBR – and especially the
existing levels of favoritism and businesspeople’s organizational power – play a
crucial role in clusters’ development. The empirical study of this chapter provides
evidence supporting this claim. Favoritism is detrimental to industrial clusters
even when applying recommended policies that normally foster the growth of
MSMEs and clusters’ development. Favoritism also denies countries from
benefiting from industrial clusters in terms of innovation. The ability of busi-
nesspeople and especially entrepreneurs to organize and form strong business
associations that defend their interests is crucial for realizing the benefits of rec-
ommended policies that promote clusters. Such associations would also help Arab
countries in reaping the benefits of industrial clusters in terms of higher
productivity.
Arab countries need policy reform in terms of formulating policies that foster
the growth of MSMEs. The results of this research support the suggestions pro-
vided in many works in the literature that facilitating credit, the availability of
venture capital, easing starting business, and improving infrastructure would lead
to clusters’ development (see Benner, 2012; Lang, 2009; Porter, 1998), which
would then lead to the growth of MSMEs (Porter, 1998). But, more importantly,
they need deeper institutional reform that guarantees the implementation and
effectiveness of these policies. Such institutional reform should target the imbal-
anced and distorted SBR in the region, in a way that would empower business
associations, especially those broad-based ones actively participated by entre-
preneurs, defined here as managers and owners of MSMEs.
Favoritism, which is a main characteristic of SBR in most of the region, should
be significantly curbed. This research provides yet another reason why this should
be set as a priority, adding to the literature that attributes negative economic
outcomes, especially in terms of innovation, to favoritism all over the globe (Ellis,
Smith, & White, 2020; Enderwick, 2005; Huang & Yuan, 2021; Sabry, 2019b,
2021) and in the Arab world (Francis et al., 2018; Rijkers et al., 2017; Sabry,
2018a). Empowering broad-based business associations and engaging them in
public–private dialogues and policy formulation and monitoring could be an
 Development in the Arab World 117

effective way of curbing favoritism (Campos & Root, 1996; Herzberg & Wright,
2005; Sabry, 2017). Such a strategy would also furnish a partnership that would
help the development of industrial clusters and in realizing the benefits of these
clusters in terms of long-run economic growth. As pointed out earlier, the specific
local conditions of each industrial cluster make public–private dialogues essential
to reap the benefits of different policies (see McDonald et al., 2006; Raines, 2002).
The author of this chapter hopes that this research encourages further research
in this topic. While a quantitative empirical approach is important to provide a
general perspective on how SBRs affect the development of industrial clusters and
reaping their economic benefits in the Arab world, studying country-specific
contexts is needed to enrich the discussion and provide more policy recommen-
dations. Thus, comparative case studies on the topic, combining qualitative and
quantitative approaches, would allow more space for a detailed study of SBR
settings in different Arab countries, the mechanism by which various policies
relevant to clusters’ development are enacted and implemented, and the effect of
these settings and policies on the performance in the sectoral and national level on
clusters’ development and their economic benefits.

Notes
1. Imputation is used for missing data, where averages for years lying between two
years for which data are available and repeating the values of an available data for
previous or consecutive years are used to fill missing values. The moderate cor-
relation between the GCI favoritism indicator and the WGI control of corruption
(CCORR) indicator (for the whole sample the correlation is 0.669) is used to
estimate the value of favoritism for the years between 2000 and 2007 for which
information is missing for favoritism. The following equation is used (for country
[i] and time [t]): (FAV)it 5 (CCORR)t * [Avg. (FAV)i/Avg. (CCORR)i]
2. It is to be noted that, as conventional for the GCI indicators, the favoritism
indicator is originally scaled between 1 for “bad” and 7 for “good” institutions.
Thus, in this case, higher levels of favoritism have lower values (the worst of which
is 1) and vice versa. To facilitate analysis, the author rescaled the indicator to
measure favoritism in a way where higher values indicate higher favoritism and
lower values indicate lower favoritism. The equation Fav. 5 (7-X)*(100/6) was
used to reach the rescaled percentage scale for favoritism used in this chapter,
where X is the original score obtained from the GCI. This percentage scale is then
used to calculate the 150 to 250 scale used in Fig. 4.1 by subtraction 50 from the
percentage scale (by calculating Fav250).
3. Endogeneity is considered here to be a minimal concern. Omitted variables bias is
minimized by the use of the fixed effects model. As for reverse causality, the six
chosen policies are of more general kind, and it is very unlikely that clusters’
development lead to changing levels of these policies rather than the other way
round. It is also less likely that clusters’ development lead to varying levels of
favoritism and V&A in a country since both institutional variables are likely to be
more entrenched and long lasting.
4. The equation used to calculate the overall effect of IPR on cluster development is,
using Eq. 1, d(C)/d(IPR) 5 b3 1 b4 (Mean of Authoritarian) 1 b5 (Mean of
118 Mohamed Ismail Sabry

Favoritism), where the mean values of authoritarianism and favoritism for the
whole sample are used in this calculation. The overall effects of other variables are
calculated in a similar fashion.

References
Abe, M., Troilo, M., & Batsaikhan, O. (2015). Financing small and medium enter-
prises in Asia and the Pacific. Journal of Entrepreneurship and Public Policy, 4(1).
doi:10.1108/JEPP-07-2012-0036
Adly, A. I. (2010). Politically-embedded cronyism: The case of post-liberalization
Egypt. Business and Politics, 11(4). doi:10.2202/1469-3569.1268
Adly, A. I. (2012). State reform and development in the Middle East: Turkey and
Egypt in the post-liberalization era. In State reform and development in the Middle
East: Turkey and Egypt in the post-liberalization era. doi:10.4324/9780203100172
Al-Saleh, Y. (2018). Crystallising the Dubai model of cluster-based development.
Place Branding and Public Diplomacy, 14(4). doi:10.1057/s41254-017-0079-1
Alesina, A., Devleeschauwer, A., Easterly, W., Kurlat, S., & Wacziarg, R. (2003).
Fractionalization. Journal of Economic Growth, 8(2). doi:10.1023/A:
1024471506938
Almezaini, K. (2013). Private sector actors in the UAE and their role in the process of
economic and political reform. In S. Hertog, G. Luciani, & M. Valeri (Eds.),
Business politics in the Middle East (pp. 43–66). London: Hurst Publishers.
Ayubi, N. N. (2001). Over-stating the Arab states. In Over-stating the Arab states. doi:
10.5040/9780755612376
Ayyagari, M., Beck, T., & Demirguc-Kunt, A. (2007). Small and medium enterprises
across the globe. Small Business Economics, 29(4). doi:10.1007/s11187-006-9002-5
Azoulay, R. (2013). The politics of Shi’i Merchants in Kuwait. In Business politics in
the Middle East. London: Hurst Publishers.
Begley, T. M., Khatri, N., & Tsang, E. W. K. (2010). Networks and cronyism: A
social exchange analysis. Asia Pacific Journal of Management, 27(2). doi:10.1007/
s10490-009-9137-4
Behuria, P., Buur, L., & Gray, H. (2017). Studying political settlements in Africa.
African Affairs, 116(464). doi:10.1093/afraf/adx019
Benner, M. (2012). Cluster policy as a development strategy case studies from the
Middle East and North Africa. (No. 255; Working Paper Series in Economics).
Cammett, M. C. (2007). Globalization and business politics in Arab North Africa: A
comparative perspective. In Globalization and business politics in Arab North
Africa: A comparative perspective. doi:10.1017/CBO9780511510007
Cammett, M., Diwan, I., Richards, A., & Waterbury, J. (2018). A political economy
of the Middle East. In A political economy of the Middle East. doi:10.4324/
9780429492600
Campos, J., & Root, H. (1996). The key to the Asian miracle: Making shared growth
credible. Washington, DC: Brookings Institution.
Chekir, H., & Diwan, I. (2015). Crony capitalism in Egypt. Journal of Globalization
and Development, 5(2). doi:10.1515/jgd-2014-0025
CIA. (n.d.). The world Factbook. Retrieved from https://www.cia.gov/the-world-
factbook/. Accessed on January 31, 2022.
 Development in the Arab World 119

Claessens, S., & Perotti, E. (2007). Finance and inequality: Channels and evidence.
Journal of Comparative Economics, 35(4). doi:10.1016/j.jce.2007.07.002
Desai, R. M., & Olofsgård, A. (2011). The costs of political influence: Firm-level
evidence from developing countries. Quarterly Journal of Political Science, 6(2).
doi:10.1561/100.00010094
Djankov, S., La Porta, R., Lopez-de-Silanes, F., & Shleifer, A. (2002). The regulation
of entry. Quarterly Journal of Economics, 117(1). doi:10.1162/003355302753399436
Ellis, J., Smith, J., & White, R. (2020). Corruption and corporate innovation. Journal
of Financial and Quantitative Analysis, 55(7). doi:10.1017/S0022109019000735
Enderwick, P. (2005). What’s bad About crony capitalism? Asian Business & Man-
agement, 4(2). doi:10.1057/palgrave.abm.9200126
Enright, M. J. (2003). Regional clusters: What we know and what we should know. doi:
10.1007/978-3-540-24760-9_6
Evans, P. (2012). Embedded autonomy: States and industrial transformation. In
Embedded autonomy: States and industrial transformation. doi:10.2307/2077170
Fafchamps, M. (2006). Development and social capital. Journal of Development
Studies, 42(7), 1180–1198. doi:10.1080/00220380600884126
Francis, D., Hussain, S., & Schiffbauer, M. (2018). Do politically connected firms
innovate, contributing to long-term economic growth? In Do politically connected
firms innovate, contributing to long-term economic growth? doi:10.1596/1813-9450-
8502
Google Developers. (n.d.). Countries.csv j dataset publishing language. Google
Developers. Retrieved from https://developers.google.com/public-data/docs/
canonical/countries_csv. Accessed on November 30, 2021.
Greenhalgh, C., & Rogers, M. (2007). The value of intellectual property rights to firms
and society. Oxford Review of Economic Policy, 23(4). doi:10.1093/oxrep/grm035
Haggard, S., Maxfield, S., & Schneider, B. R. (1997). Theories of business and
business-state relations. In S. Maxfield & B. Schneider (Eds.), Business and the state
in developing countries (pp. 36–60). Cornell University Press. doi:10.7591/
9781501731976-004
Hausmann, R., & Rodrik, D. (2003). Economic development as self-discovery.
Journal of Development Economics, 72(2). doi:10.1016/S0304-3878(03)00124-X
Hausmann, R., Rodrik, D., & Sabel, C. F. (2011). Reconfiguring industrial policy: A
framework with an application to South Africa. SSRN Electronic Journal. doi:10.
2139/ssrn.1245702
Hertog, S. (2013a). State and private sector in the GCC after the Arab uprisings.
Journal of Arabian Studies, 3(2), 174–195. doi:10.1080/21534764.2013.863678
Hertog, S. (2013b). The role of MENA business in policy-making and political
transitions. In S. Hertog, G. Luciani, & M. Valeri (Eds.), (pp. 1–16). Hurst
Publishers.
Herzberg, B., & Wright, A. (2005). Competitiveness partnerships – Building and
maintaining public-private dialogue to improve the investment climate: A resource
drawn from the review of 40 countries’ experience. (Working Paper No. 3683).
Heydemann, S. (2013). Tracking the “Arab spring:” Syria and the future of author-
itarianism. Journal of Democracy, 24(4), 59–73.
Hinnebusch, R. (2012). Syria: From “authoritarian upgrading” to revolution?
International Affairs, 88(1). doi:10.1111/j.1468-2346.2012.01059.x
120 Mohamed Ismail Sabry

Hinnebusch, R. (2015). Change and continuity after the Arab Spring: The conse-
quences of state formation in Arab North African states. British Journal of Middle
Eastern Studies, 42(1), 12–30.
Huang, Q., & Yuan, T. (2021). Does political corruption impede firm innovation?
Evidence from the United States. Journal of Financial and Quantitative Analysis,
56(1). doi:10.1017/S0022109019000966
Khan, M. H. (2018). Political settlements and the analysis of institutions: An Intro-
duction. African Affairs, 117(469). doi:10.1093/afraf/ady051
Koreen, M., & Lucia, C. (2015). New approaches to SME and entrepreneurship
finance: Broadening the range of instruments Cfe/sme (2014)8/final. In OECD
(Issue JT03369708).
Lang, J. C. (2009). Cluster competitiveness: The six negative forces. Journal of Busi-
ness and Management, 15(1), 73–93.
Loewe, M., Blume, J., & Speer, J. (2008). How favoritism affects the business climate:
Empirical evidence from Jordan. The Middle East Journal, 62(2). doi:10.3751/62.2.
14
McDonald, F., Tsagdis, D., & Huang, Q. (2006). The development of industrial
clusters and public policy. Entrepreneurship & Regional Development, 18(6). doi:10.
1080/08985620600884636
Moore, P. W. (2001). What makes successful business lobbies? Business associations
and the rentier state in Jordan and Kuwait. Comparative Politics, 33(2). doi:10.
2307/422375
Nabli, M. K. (2007). Restating Arab economic reform. In M. K. Nabli (Ed.), Breaking
the barriers to higher economic Governance and deeper reforms in the Middle East
and North Africa (pp. 89–102). The World Bank.
Oubenal, M., & Ben Hamouda, H. (2018). The political economy of business elites in
Tunisia: Actors, strategies and identities. (No. 1273).
Peltzman, S., Levine, M. E., & Noll, R. G. (1989). The economic theory of regulation
after a decade of deregulation. In Brookings papers on economic activity. Micro-
economics. doi:10.2307/2534719
Porter, M. E. (1998). Clusters and the new economics of competition. Harvard
Business Review, 76(6), 77–90.
Raines, P. (2002). Cluster development policy and new forms of public—Private
partnership. International Business. doi:10.1057/9781403937766_11
Rijkers, B., Freund, C., & Nucifora, A. (2017). All in the family: State capture in
Tunisia. Journal of Development Economics, 124. doi:10.1016/j.jdeveco.2016.08.002
Rodrik, D. (2008). Normalizing industrial policy. In Commission on Growth and
Development Working Paper; No. 3. World Bank, Washington (No. 3). Washing-
ton, DC: World Bank. Retrieved from https://openknowledge.worldbank.org/
handle/10986/28009
Saadi, M. S. (2016). Moroccan cronyism: Facts, mechanisms and impact. (No. 1063).
Sabry, M. I. (2013). State business relations: Networks, institutions and their effect on
growth and cronyism. Verlag Dr. Hut. Retrieved from https://books.google.com/
books/about/State_Business_Relations_Networks_Instit.html?id5_
TLQnQEACAAJ
Sabry, M. I. (2017). Taming cronyism!: The role of public private dialogues.
International Journal of Social Economics, 44(12). doi:10.1108/IJSE-01-2016-0030
 Development in the Arab World 121

Sabry, M. I. (2018a). Could formalized state business relations lead to economic

growth? Business participation, government representation and institutions.
International Journal of Social Economics, 45(7). doi:10.1108/IJSE-08-2017-0347
Sabry, M. I. (2018b). State business relations and innovation in the MENA region.
Middle East Development Journal, 10(2). doi:10.1080/17938120.2018.1520003
Sabry, M. I. (2019a). Cronyism as an outcome of institutional settings: The case of
pre-2011 Egypt. International Journal of Sociology & Social Policy, 40(1–2). doi:10.
1108/IJSSP-08-2019-0153
Sabry, M. I. (2019b). Fostering innovation under institutional deficiencies: Formal
state–business consultation or cronyism? Economia Politica, 36(1). doi:10.1007/
s40888-018-00137-1
Sabry, M. I. (2021). Innovation versus technology imitation development strategy:
What influences government decision? Innovation and Development, 1–30. doi:10.
1080/2157930X.2021.1954361
Schneider, B. R. (2005). Business politics and policy making in contemporary Latin
America. Political Science.
Schneider, B. R. (2009). Hierarchical market economies and varieties of capitalism in
Latin America. Journal of Latin American Studies, 41(3). doi:10.1017/S002221
6X09990186
Schneider, B. R. (2015). The developmental state in Brazil: Comparative and histor-
ical perspectives. Revista de Economia Politica, 35(1). doi:10.1590/0101-31572015
v35n01a07
Schwab, K., & Sala-i-Martin, X. (2017). WEF: The global competitiveness report
2017/2018. In World economic forum (Vol. 21, Issue 3). Cologny, Geneva: World
Economic Forum.
Shadlen, K. C. (2002). Orphaned by democracy: Small industry in contemporary
Mexico. Comparative Politics, 35(1). doi:10.2307/4146927
Shiaty, M. (2014). Tunisia investment climate reform.
Shleifer, A. (2005). Understanding regulation. European Financial Management, 11(4).
doi:10.1111/j.1354-7798.2005.00291.x
Stigler, G. J. (1971). The theory of economic regulation. Bell Journal of Economics and
Management Science, 2(1), 3–21. doi:10.2307/3003160
The Association of Religion Data Archives (ARDA). (n.d.). Quality data on religion.
Retrieved from https://www.thearda.com/. Accessed on January 31, 2022.
The World Bank. (n.d.). Doing business data. Retrieved from https://www.
doingbusiness.org/en/data. Accessed on January 31, 2022.
United Nations. (n.d.). Ethnocultural characteristics. Retrieved from https://unstats.
un.org/unsd/demographic/sconcerns/popchar/. Accessed on January 31, 2022.
University of Gothenburg. (n.d.). The quality of government. (QoG) Dataset.
Retrieved from https://www.gu.se/en/quality-government/qog-data. Accessed on
January 31, 2022.
UTICA. (2015). Tunisia’s national dialogue. Retrieved from https://ppd.cipe.org/wp-
content/uploads/2016/03/2015-Public-Private-Dialogue-in-Tunisia3.pdf
Valeri, M. (2013). Oligarchy vs. Oligarchy: Bussiness and politics of reform in Bahrain
and Oman. In Business politics in the Middle East. London: Hurst Publishers.
Warwick, K. (2013). Beyond industrial policy: Emerging issues and new trends.
OECD Science, Technology and Industry Policy Papers, 2.
122 Mohamed Ismail Sabry

World Bank. (n.d.-a). World development indicators (WDI). Retrieved from https://
datatopics.worldbank.org/world-development-indicators/. Accessed on September
24, 2021.
World Bank. (n.d.-b). Worldwide governance indicators (WGI). Retrieved from http://
info.worldbank.org/governance/wgi/. Accessed on September 24, 2021.
World Economic Forum. (2018). Arab world competitiveness report 2018. Retrieved
from https://www.weforum.org/reports/arab-world-competitiveness-report-2018
World Economic Forum. (n.d.). Global competitiveness indicators (GCI) dataset
2007–2017. Retrieved from https://view.officeapps.live.com/op/view.aspx?src5http
%3A%2F%2Fwww3.weforum.org%2Fdocs%2FGCR2017-2018%2FGCI_
Dataset_2007-2017.xlsx&wdOrigin5BROWSELINK. Accessed on September 24,
2021.
Yoshino, Y. (2011). In Y. Yoshino (Ed.), Industrial clusters and micro and small
enterprises in Africa. Washington, DC: The World Bank.
Zovighian, D. (2013). The politics of Good Governance in Mubarak’s Egypt: Western
donors and SME politics under authoritarian rule. In S. Hertog, G. Luciani, & M.
Valeri (Eds.), Business politics in the Middle East (pp. 183–209). London: Hurst
Publishers.

Appendix

Table A.1. Proxies, the Sources From Which They Are Obtained, and How
They Are Calculated.
Variable(s) Description Source
Voice and “Perceptions of the extent to World Bank (WB) –
Accountability which a country’s citizens can Worldwide Governance
(V&A) participate in selecting their Indicators (WGI) (World
government, as well as Bank, n.d.-b)
freedom of expression,
freedom of association, and a
free media.”
Favoritism “To what extent do Global Competitiveness
government officials show Indicators (GCI) (Schwab
favoritism to well-connected & Sala-i-Martin, 2017)
firms and individuals when
deciding upon policies and
contracts?”
Getting Credit “The total score for getting WB – Doing Business (DB)
credit is the sum of the Dataset (The world Bank,
strength of legal rights index n.d.)
and the depth of credit
information index, based on
the methodology in the
DB05-14 studies.”
 Development in the Arab World 123

Table A.1. (Continued)

Variable(s) Description Source
Starting “The score for starting a WB-DB
Business business is the simple average
of the scores for each of the
component indicators: the
procedures, time and cost for
an entrepreneur to start and
formally operate a business,
as well as the paid-in
minimum capital
requirement.”
Venture Capital “In your country, how easy is GCI
it for entrepreneurs with
innovative but risky projects
to find venture capital?”
IPR “to what extent is intellectual GCI
property protected?”
Cluster “In your country, how GCI
Development widespread are
well-developed and deep
clusters (geographic
concentrations of firms,
suppliers, producers of
related products and services,
and specialized institutions in
a particular field)?”
Innovation A composite of the following GCI
indicators: capacity for
innovation, quality of
scientific research institutions,
company spending on R&D,
university-industry
collaboration in R&D,
government procurement of
advanced technology,
availability of scientists and
engineers, and PCT patents.
124 Mohamed Ismail Sabry

Table A.1. (Continued)

Variable(s) Description Source
Log. Labor Log. of “Industry (including WB – World Development
Productivity construction), value added Indicators (WDI) (World
per worker (constant 2015 Bank, n.d.-a)
US$)”
Infrastructure A composite of the following GCI
indicators: quality of overall
infrastructure, quality of
roads, quality of railroad
infrastructure, quality of port
infrastructure, quality of air
transport infrastructure,
available airline seat
kilometers, quality of
electricity supply,
mobile-cellular telephone
subscriptions, and
fixed-telephone lines.
Government Government expenditure on WB – WDI
Expenditure on education, total (% of GDP)
Education

Table A.2. Variance Inflation Factor (VIF) Test for Multicollinearity.

Variable VIF
V&A 1.147
Favoritism 3.401
Getting Credit 1.726
Start Business 1.615
Venture Capital 3.732
IPR 6.864
Infrastructure 7.737
Government Expenditure on Education 1.273
Note: A collinearity problem might exist if VIF.10.
Table A.3. Statistical Testing for Choosing the Proper Regression Model.

Regression Test Type Test Result Model Selection

Set {A} Joint significance of differing group means F(14, 238) 5 11.67, p value 5 2.244e-020 Fixed effects
(OLS vs. fixed effects)
B-1-1 and Joint significance of differing group means F(11, 199) 5 19.17, p value 5 6.7e-026 Fixed effects
B-1-2 (OLS vs. fixed effects)
Breusch–Pagan (OLS vs. random effects) LM 5 162.269, p value 5 Random effects
prob(chi-square(1) . 162.269) 5 3.613e-
037
Hausman (random effects vs. fixed effects) H 5 28.849, p value 5 Fixed effects
prob(chi-square(5) . 28.849) 5 2.483e-
005
Asymptotic test statistic: chi-square(3) 5

Development in the Arab World

B-1-3 Hausmann (OLS vs. TSLS) TSLS
72.02, p value 5 1.576e-015
B-2-1 and Joint significance of differing group means F(8, 148) 5 2002.43, p value 5 1.029e- Fixed effects
B-2-2 (OLS vs. fixed effects) 146
Breusch–Pagan (OLS vs. random effects) LM 5 827.56, p value 5 Random effects
prob(chi-square(1) . 827.56) 5 5.499e-
182
Hausman (random effects vs. fixed effects) H 5 10.065, p value 5 Random effects (at 5%
prob(chi-square(5) . 10.065) 5 0.073 of significance)
Fixed effects (at 10%
of significance)
B-2-3 Hausmann (OLS vs. TSLS) Asymptotic test statistic: chi-square(5) 5 TSLS

125
144.632, p value 5 1.853e-029
Table A.4. The Instruments Used in the TSLS Regressions.

126
Variable(s) Description Source

Mohamed Ismail Sabry

Log. Years since Log. (Time – Year of Independence) CIA Factbook (CIA, n.d.)
Independence
Religious Adherence Catholic, Orthodox, Muslim Sabry (2013) depending on data from the UN
“ethnocultural characteristics” (United Nations,
n.d.), CIA Factbook, and the “Association of
Religion Data Archives” (ARDA) (The
Association of Religion Data Archives (ARDA),
n.d.)
Ethnic The Quality of Government Dataset (University
Fractionalization of Gothenburg, n.d.) and Alesina,
Devleeschauwer, Easterly, Kurlat, and Wacziarg
(2003)
Country Size Log. (Population, total) World Development Indicators (WDI) (World
Bank, n.d.-a)
Latitude Absolute of Distance from the Equator Google Developers (Google Developers, n.d.)
Fuel Richness 5-year average of the following calculation: (Fuel WDI
exports [% of merchandise exports]*Merchandise
exports [current US$])*(GDP [current US$]/GDP
[constant 2015 US$])*(1/Population, total)
Metal Richness 5-year average of the following calculation: (Ores WDI
and metals exports [% of merchandise exports]
*Merchandise exports [current US$])*(GDP
[current US$]/GDP [constant 2015 US$])*(1/
Population, total)
Table A.5. Sequential Elimination of Insignificant Variables.

Regression Omitted Variables F-Test

A-1 Getting Credit, Government Expenditure on Education F(2, 249) 5 1.0458
A-2 V&A, V&A*Credit, Favoritism*Start Business F(3, 238) 5 0.2489
Favoritism, Venture Capital, Government Expenditure on Education F(3, 241) 5 1.1378
A-3 Government Expenditure on Education, Favoritism*Start Business, F(4, 238) 5 0.3427
Authoritarian*Credit, Authoritarian
Favoritism F(1, 242) 5 1.31279
B-1-1 V&A*Cluster Development F(1, 199) 5 0.2341
B-1-2 Favoritism*Cluster, V&A*Cluster Development F(2, 192) 5 0.0903
B-2-1 Favoritism*Cluster Development F(1, 148) 5 0.2055
B-2-2 Favoritism*Cluster Development F(1, 156) 5 0.2346

Development in the Arab World

B-2-3 Cluster Development, Favoritism*Cluster Development F(2, 156) 5 1.5011

127