The Road Not Taken:

A Typology of the Welfare Regimes in the Middle East and North Africa

Semuhi Sinanoğlu

Koç University

Abstract

The academic scrutiny of the welfare regimes in the Middle East has been subsumed under other avenues

of inquiry such as clientelism and authoritarianism. In the thrall of the concept of rentier state and

globalization as an overarching economic force, these welfare regimes have not been studied yet in their

totality. This paper aims to incorporate the Middle East into the literature on the comparative political

economy of welfare regimes in the developing countries. By applying cluster analysis to 17 countries in

the region for the time period between 1997 and 2007 with respect to three dimensions; labor market

protection, welfare mix, and welfare outcome; the main findings of this paper reveal systemic divergences

among Middle Eastern countries, which can be categorized into four distinct clusters, namely protective

oil-rich, non-protective oil-rich, mixed, and corporatist regimes.

Introduction1

In 1977, just after President Sadat announced that food subsidies would be curtailed,

waves of protest reverberated across the country and eventually the army had to intervene in

order to suppress the “Bread Intifada” of Egypt. Sadat had to annihilate the policy proposal. Since

then, the specter of food riots has been haunting political elites in Egypt. Even in 2013, during the

negotiations with IMF for 22-month-Stand-By agreement, Morsi government could not accept

the terms for cutting back subsidies2. Aish (bread) means “living” in Arabic, which has permeated

into “moral economy consciousness” of the people and has colored the discourses in the heyday

of the so-called Arab Spring. Albeit ambitious, this paper attempts to bring Middle Eastern

welfare regimes in their totality under academic scrutiny, by applying cluster analysis to 17

countries in the MENA for the time period between 1997 and 2007.

In striking contrast with the voluminous and detailed literature on comparative political

economy of the welfare regimes of the developed countries, the complex constellations of non-

state actors, institutions and social policy designs in the developing countries beg further study.

By the mid-1980s, social policy had negative connotations due to its ultimate linkage with

1 This paper was prepared for the graduate course titled “POLS593: Research & Reading in Comparative

Politics” at Boğaziçi University under the supervision of Prof. Mine EDER. 2 As an example, see Heather Stewart, “IMF keen to splash the cash in Egypt”, The Guardian, April 14,

2013. http://www.theguardian.com/world/2013/apr/14/egypt-unrest-over-imf-cash

2

“inefficient” welfare state (Mkandawire 2001, 11). Then especially since the promulgation of the

Millennium Development Goals, scholarly attention has shifted onto the coalescence of social

policy and economic development. The initial “Chinese wall” separating developmental studies

from the study of the welfare state could be eradicated via the idea of welfare developmentalism

(Kwon et al. 2009, 2). This burgeoning literature on the welfare regimes in the developing

countries attempts to shed light on the wide array of issues comprising the impacts of policy

legacies, economic strategies, globalization as well as political regimes and value systems, which

have been explored for Latin America, East Asia and Eastern Europe as regions (See Gough and

Wood 2004, Haggard and Kaufman 2008, Mesa-Lago 2003, Rudra 2002 among others). On the

other hand, the holistic analysis of the welfare regimes in the Middle East has been systematically

neglected, which can be attributed to two main reasons; namely the concept of rentier state and

globalization.

In the thrall of the concept of the rentier state, the content of social welfare systems has

been subsumed under other avenues of academic inquiry, especially clientelism and

authoritarianism. This generic concept could not capture the multilayered and multifaceted

configurations of the welfare systems in the region. On the one hand, the state-society relations

have been relegated to a “ruling bargain” (Brumberg 1990, 120) in which the robustness of

authoritarian regimes is overstated; on the other hand, the idea of “political giveaways”

(Waterbury 1983, 228) in exchange for acquiescence has been prioritized over the long-run and

coherent economic strategies pursued by the governments in the region. Both of these statements

have remained untested.

The impacts of the globalization have exacerbated the current plight of the literature, too.

Due to the expectation for the retrenchment of the state as a consequence of the globalization,

scholarly works have underscored faith-based social welfare provision as well as non-state actors,

which are supposed to “compete” with the retreating state. A few seminal works notwithstanding

(see Cammett and MacLean 2011), those studies could not capture the relationality between the

state and these private networks, and underestimated the “hegemonic control” (see Eder 2009)

that the state exerts over their institutional settings. Some of these approaches have clung to

“culturalist/essentialist” explanations vis-à-vis Islamic law and practices (Roy 1994, 7).

3

By reviewing the existing literature on the developing contexts, this paper endeavors to

incorporate comparative political economy of the Middle East into the general discussion of the

welfare regimes. Hence, on the basis of three dimensions, namely labor market protection,

welfare mix, and welfare outcome; 17 countries in the Middle East will be analyzed for the time

period between 1997 and 2007 with recourse to the “welfare regime approach” with certain

modifications. The main findings suggest that those countries can be clustered into four distinct

groups; namely protective oil-rich, non-protective oil rich, corporatist, and mixed regimes. This

evidence for “systemic divergence” counters the main untested assumptions regarding the politics

of the Middle East as well as the claim for the total convergence among the welfare regimes in

the developing countries. Although the precision of this typology might suffer from the dearth

and unreliability of the available data and might not be comprehensive enough to address all the

shortcomings of the prevailing literature criticized above, it aims to turn around the corner and set

off down “the road not taken” – a quantitative analysis of the welfare regimes in the Middle East.

Applying to the Development Context

The scholars of comparative political economy have been intrigued by a vital question for

a long time: Is the conceptual framework of welfare regime applicable to the development

context? This particular conception was embedded in “ethnocentric western social research” in

which those societies without capitalist economy or western parliamentary democracy were

neglected and excluded from the discussion of welfare states – “exclusive club of mainly OECD

members” (Walker and Wong 1996, 67-69). The study of welfare regimes in the development

context was languishing in obscurity for many years. Thus Carnes and Mares (2009, 882) vividly

stated that “our understanding of variation in social policies outside of OECD economics still

resembles a sixteenth century map, with vast areas of uncharted territories”. As a consequence,

current literature has been marked by an empirical bias and problems of multi-collinearity, which

could be overcome by enlarging the sample size with developing countries and taking other

possible explanatory variables into consideration (Mares and Carnes 2009, 95). Without any

doubt, the burgeoning literature on the welfare regimes in the developing countries tries to cover

different regions and various aspects of social welfare, albeit still far from understanding

systemic divergences. The vicissitudes of globalization have affected this process of scrutiny.

4

The literature on globalization has deemed welfare states “residual” due to alleged

“retrenchment” and “race-to-the-bottom”. Several studies on developing countries tried to

explore this phenomenon (See Rudra 2002, Sing & Zammit 2004, Wood 1997 among others).

However, states have not followed any unidirectional path and by no means have converged

towards a uniform residual state. In the first place, studies have strived to reveal the relationship

between overarching economic force of globalization and size of the public sector and welfare

state (Cameron 1978, Garrett & Mitchell 2001, Iversen and Cusack 2000, Kaufman and Segura-

Ubiergo 2001, Rieger and Liebfield 1998, Rodrik 1997, Rodrik 1998, Wibbels 2006).

Further studies have endeavored to elucidate the question of what kind of domestic

institutional settings mediate this economic push factor (See Burgoon 2001). As Huber and

Stephens (2001, 48) clarified, “decisions about welfare state expansion are politically mediated

rather than automatic reactions to needs for social protection”. Rudra and Haggard (2005)

examined the divergences across different regime types in terms of social spending by delineating

democratic, non-democratic and intermediate authoritarian regimes. On the other hand, Rudra

(2008) justified race to the bottom in developing countries with respect to the collective actions

problems resulting from fragmented labor market institutions. Mares (2005, 625) espoused

another approach: Her analysis hinged upon the question of whether “higher openness leads also

to different types of welfare states”. She made the distinction between high-risk and low-risk

coalition and highlighted the relative power of the states to satisfy the increased demands under

the condition of external risks. As Mares’ work (2005) suggests, the theories of cross-class

alliances have begun to gain currency as a critique of power-resources theories (Korpi 1983). A

business-centered literature (see Hall and Soskice 2001) emerged, which evaluates social policies

as instruments for skill retention (See Mares 2003 and Swenson 2002).

These cross-class alliances could resist the dismantling of the welfare state under the

condition of “permanent austerity” (Pierson 1996) and under the pressures of globalization. This

neo-institutionalist perspective pinpoints the fact that political choices are framed by earlier

choices (Saint-Arnaud and Bernard 2003, 506). There exist path-dependent processes “which

tend to lock existing policy arrangements into place” (Pierson 2001, 414). This policy-feedback

approach has unfolded the idea of critical alignments, in which cross-national differences have

been elucidated with respect to different economic strategies and skill retention is conceptualized

as the linkage between production and welfare regimes (Mares and Carnes 2009, 101). Wibbels

5

and Ahlquist (2007) argued that the preferred type of social spending such as insurance-based

social regimes or export-led industrialization was woven by four structural conditions, namely

domestic market size, the relative abundance of labor, asset inequality, and the openness.

Haggard and Kaufman (2008) reiterated the study of economic strategies and revealed cross-

regional differences, namely for Latin America, East Asia, and Eastern Europe. Rudra (2007)

concentrated on the different development strategies pursued by developing countries. By

applying cluster analysis, she conceptualized three different regimes, namely protective welfare

state, productive welfare state and mixed dual welfare state3.

The main controversial issue has been the tension between the arguments favoring either

convergence or divergence, especially due to the globalization. Academic inquiries into the main

patterns of divergence for the welfare regimes in Latin America, East Asia, and Eastern Europe

have been realized (see Esping-Andersen 1996, Huber 2002, Kwon 2005, Mesa-Lago 2003,

Riesco 2007, Segura-Ubiergo 2007 among others). For instance, Gough and Wood (2004)

specified three meta-welfare regimes; namely welfare state regime, informal security regime, and

insecurity regime to account for cross-regional differences. Notwithstanding this vigor in the

literature, the study of the welfare regimes in the Middle East has been systematically neglected.

Only a few countries such as Tunisia and Egypt have been analyzed in the models with global

scale. We need to incorporate Middle East into this general theoretical framework.

Incorporating the Middle East

One of the reasons for reluctance and failure to incorporate Middle East into the general

discussion of welfare regimes was to espouse a monolithic view of the states in the region – the

concept of “interventionist-redistributive model” was the epitome of this approach (See Yousef

2004), which had its intellectual roots in the concept of “rentier state”. This notion, according to

Anderson (1987, 9) is “one of the major contributions of Middle East regional studies to political

science”. This concept was coined by Mahdavy (1970) and then refined by Beblawi (1987). This

external rent income, especially during the oil-boom era, dominated the economy and granted the

government the leverage to pursue pro-cyclical spending and the ability to counter the principle

of “no taxation without representation”, which undermined the development of citizenship. In

fact, the endowment of significant quantity of petroleum has stunt the economic progress of

3 The most problematic aspect of this strand of the literature was the implicit assumption of policy

coherence. They cannot account for variation across policy areas (Mares and Carnes 2009, 107).

6

women, as Ross’ (2013, 12) comparison of Algeria, Tunisia, and Morocco attests to. All in all, in

the thrall of the concept of rentier state, which cannot capture the prevailing diversity, the

analysis of the welfare regimes has been languishing in obscurity and subsumed under other

avenues of inquiry such as clientelism4, rentierism and authoritarianism.

There are two significant implications of the conception of rentier state. Firstly, it resulted

in a “static image” of state-society relations, in which the robustness of the coercive state (see

Bellin 2012) was overstated (Jawad 2009, 10). Under the caveat of a “ruling bargain” (Brumberg

1990, 120), the state was assumed to provide either social welfare or limited political

representation in exchange for obedience to the state5. In similar vein, the concept of “moral

economy” implies that in the eve of economic liberalization, public sector-dominated economy

would redistribute resources to the working class in exchange for labor acquiescence (See

Posusney 1997, Pratt 2000/2001). Secondly, in the course of economic reform, the state was

considered to implement “political giveaways” (Waterbury 1983, 228), which indicates the

allocation of resources designed to surmount political unrest. In other words, the patterns of

expenditure were not subject to overarching economic strategies, but short-term political

considerations. Due to the “moral economy consciousness,” the specter of foot riots (as occurred

in 1977 in Egypt) haunts politicians (Marfleet 2009, 32). As a corollary, “the politics of

retrenchment is implicitly reduced to instrumental coalition management strategies, by rational

political actors seeking to maintain their hold on power” (El-Meehy 2009, 56).

Another thread in the literature on social welfare provision in the region discerns non-

state actors, particularly faith-based welfare provision, which is one of the abundantly studied

aspects of the literature. Non-governmental organizations have proliferated, especially for social

welfare provision. For instance, in Tunisia, from 1988 to 1991, the number of NGOs increased

from 1886 to 5186 (Kennett and Oakley 2001, 111). As particular features of welfare systems in

East Asia have been problematically attributed to the “Confucian virtues” (see Jones 1993), an

array of issues vis-à-vis the congruence between Islam and welfare provision has been scrutinized

4 Recent wave of scholarship especially focuses on elections as a mechanism for clientelistic distributive

politics. For instance, public infrastructure coverage has been influenced by the voting behavior of the

governorates in Egypt (See Blaydes 2011). 5 In the advent of “permanent austerity”, welfare provision has become a “liability” for the state due to

declining economic resources. Thus, the balance of the “social contract” has been titled towards the

“society” and has created avenues for political participation, yet limited. On the other hand, this change

has disrupted the stability of the regime. For details of this kind of argument, see Kamel (2001).

7

such as the tenants of Islamic law, Islamic institutions and practices, philanthrophy and other

networks of solidarity as well as Islamist associations (see Clark 2004; El-Ghonemy 1998;

Heynemann 2004; Ismail 2006, 66-95; Sullivan 1994). A “parallel Islamic sector” emerged,

whose institutions have different functions and serve different constituencies and are competitive

with the state (Wickham 2002, 95). The Muslim Brotherhood has been operating 22 hospitals and

numerous small clinics in Egypt, as Jam’iyyat al-Shaban al-Muslimin has branches throughout

Egypt (See Harrigan and El-Said 2009, 103-04). The Gülen movement originated from Turkey

exemplifies the rising phenomenon of transnational religious communities, which operates

numerous hospitals, foundations, associations and schools in different countries (See Ebaugh

2010). The underlying presumption for this analysis was the “retreat of the state” on various

layers of social security as a consequence of globalization and neoliberal turn – the power of the

state and the “hegemonic control” it exerts over the economy and other networks, associations,

and institutions, even under the conditions of privatization have been underestimated (See Eder

2009, Tadros 2006). The recent scholarship has struggled for more systemic and holistic

approach towards the complex constellation of state and non-state actors (see Cammett and

MacLean 2011). To that end, Jawad and Yakut-Cakar (2010) classified religious welfare systems

into five categories; namely, the religious order, the elite family, the popular political movement,

international humanitarian relief, and the para-state organization.

To sum up, the examination of the welfare regimes in the Middle East has been wedged

between the concept of rentier state and welfare state retrenchment vis-à-vis mushrooming faith-

based welfare provision. In the former one, we cannot capture the diverse and multi-faceted

nature of welfare systems and overlook the agency of citizens in certain regards. On the other

hand, in the latter one, the capacity of the state has been understated. Therefore we need to

contemplate this multi-layered phenomenon by accentuating its “systemic divergences”.

Accordingly, “welfare regime approach” is the starting point for this appraisal.

8

The Welfare Regime Approach

Esping-Andersen’s The Three Worlds of Welfare Capitalism (1990) is one of the main

pillars of comparative political economy, which entailed “welfare modelling business”

(Abrahamson 1999) to identify different typologies of welfare regimes as well as elicited harsh

critiques in various regards6.For instance, in order to apply to the developing context, Gough and

Wood (2004) relaxed the de-commodification principle and coined “de-clientelisation.”

One important critique of Esping-Andersen’s work points out at misspecification of

Mediterranean welfare states. To situate Mediterranean countries into these typologies entailed a

great debate about whether a distinct “Southern/Mediterranean model” exists or not (see Ferrera

1996). Şener (2006) incorporated Greece, Portugal, Spain, and especially Turkey into de-

commodification index by applying cluster analysis. The main findings suggest that Turkey is

situated in medium de-commodifying welfare state cluster. What makes this study unique is the

fact that it is one of the earliest attempts to include Turkey in a comparative and empirical social

policy analysis (Şener 2006, 81). Be that as it may, it is open to dispute whether Turkey should be

scrutinized within the European or Middle Eastern context. Although contemplating Turkey’s exact

location in these different contexts analytically is beyond the scope of this paper, it aims to provide

another point of view by incorporating Turkey into the comparative analysis of MENA countries.

The analysis of the welfare regimes in the Middle East has been systematically neglected

Be that as it may; there are seminal works which incorporate the Middle East into the general

literature of comparative political economy. Karshenas and Moghadam (2006) provided a

regional comparison, by scrutinizing the systems of social policy in the MENA hand-in-hand

with “social rights-based approach” in Nordic countries and “developmental welfare state” in

East Asia.

One of the earliest attempts to classify MENA countries into distinct clusters was the

pioneering work of Grossman (2007), which strived to shed light on the period between 1950s

and 1980s. He categorized those countries into three groups; namely, redistributive-republican

6 His analysis privileged the state and concentrated on income maintenance and labor market practices

(Gough and Wood 2004, 24; Jawad 2009, 63). Yet other dimensions of social expenditure such as health

and education expenditures should be evaluated to make sense of systemic features of welfare regimes in

the developing countries. Secondly, his conception of “decommodification” might not be applicable to the

developing contexts because of the huge informal labor market and the ongoing process of

“commodification” (See Rudra 2007).

9

model (Egypt, Tunisia, Syria, Algeria, Libya, and Iraq), residual pluralist model (Jordan,

Lebanon, and Morocco) and distributive oil-rich monarchical model (UAE, Kuwait, Saudi

Arabia, Oman, and Bahrain). He tried to decipher the congruence between regime types and

economic strategies in his illuminating article. One of the shortcomings of this work lies at the

methodology. Although Grossman cemented his analysis with quantitative data, with recourse to

qualitative analysis, standardization of available data became heavily problematic. Secondly, he

missed the opportunity to keep the track of those historical legacies and their repercussions and

implications for present systems of social welfare provision.

A Typology of the Welfare Regimes

Methodology

For constructing typologies, the most effective and widely used technique is cluster

analysis7. It is a quantitative method which classifies cases into homogenous groups, which are as

distinct as possible from each other. Two clustering techniques are commonly applied in the

literature: hierarchical cluster analysis and k-means cluster analysis (See Franzoni 2008, Gough

2001, Rudra 2007 among others). In the former one, the number of clusters is data-determined

while in the latter the number of clusters should be specified a priori by the researcher. Gough

(2001, 166) prefers trying different values of cluster and evaluates their empirical adequacy with

respect to the final cluster centers and distance between clusters, while Powell and Barrientos

(2004) utilize hierarchical cluster analysis to identify the number of clusters, which reflect the

structure of the data best, and apply k-means cluster analysis with the information provided. In

similar vein, both techniques will be utilized and applied consecutively in this paper.

Agglomerative hierarchical cluster analysis begins with each case as a cluster on its own. At

successive stages, similar cases are merged into the same cluster and the algorithm ends with

bringing together all cases under the same cluster. The disadvantage of this technique is the fact

that once two cases are joined in a cluster, they cannot be split again. On the other hand, in k-

means cluster analysis, clusters are recombined through iteration and final cluster centers are

recomputed after each case is classified. Hence, it is useful to employ both techniques –

hierarchical analysis will be used to estimate starting values for k-means cluster analysis.

7 For a list of different techniques utilized by studies which tried to construct typologies from different sets

of countries, see Arts and Gelissen (2010, 575-576).

10

There are various linkage methods to form clusters; namely single linkage, complete

linkage, average linkage, centroid method, and Ward’s method. In this paper, Ward’s method is

preferred among the agglomerative linkage methods (see Rudra 2007, Şener 2006). Ward’s

method employs squared Euclidean distance, which corresponds to the sum of the squared

differences over all of the variables. Ward’s method is designed to minimize the increase in the

overall sum of the squared within-cluster distances. In cluster analysis, different scales of

measurement, variables with numerically large values, might skew the results, since this analysis

relies upon the computation of distance scores. In order to tackle this problem, the variables are

standardized (Kangas 1994, 357). Variables are checked for outliers as well as the cases are

reordered to avoid biasedness.

The dendrogam provides a visual representation of which clusters are combined at

different stages. Vertical lines display joined clusters, as horizontal lines denote the distance

between countries in the same cluster. The lengthier lines between merged cases, the greater the

dissimilarity between them. The interpretation of the dendrogam suggests that four clusters

mirror the structure of the data best. Thus, the number of clusters is set to 4 (See Figure 1) for k-

means cluster analysis, which would reveal systemic divergences, if any. Cluster analysis is

sensitive to missing cases. Hence, the lack and unreliability of the available data complicated the

process of estimator specification. To solve this problem and maximize information that is

obtained from the data, cases are “excluded pairwise”, which implies the assignment of cases to

the clusters based on distance that are computed from all variables with non-missing values.

Figure 1: Dendrogam, Hierarchical Cluster Analysis.

11

Variables

In order to elucidate systemic divergences across cases in the developing context,

different dimensions have been scrutinized: commodification and de-commodification indices,

de-familiarization, regime types and level of democracy, generosity of social assistance, level of

social protection expenditure, welfare mix and welfare outcome, openness of the country, gender

stratification and historical legacies (Bibars 2001, Franzoni 2008, Gough 2001, Gough and Wood

2004, Haggard and Kaufman 2008, Huber 1996, Rudra 2007, Rudra and Haggard 2005, Segura-

Ubiergo 2007). This paper endeavors to discern similarities and divergences by selecting

indicators on the basis of three dimensions; namely, labor market protection, welfare mix, and

welfare outcome.

The most challenging part of this scrutiny is to figure out possible indicators, which can

be compiled based on available information. Unfortunately, the available data for the Middle

Eastern countries are not sophisticated as in Eurostat to accomplish cross-national comparisons.

Despite this limitation, 17 countries in the region are analyzed for the time period between 1997

and 2007. Although the temporal framework of the analysis might be criticized that the extent

does not capture the particularities of the historical legacies and previous policies, the theoretical

structure primarily intends to address the era of globalization and quantitative analysis is difficult

to accomplish for time-series analysis for the region due to data limitations. Moreover, in order to

account for the effects of previous government commitments, welfare outcome is specified as a

dimension of the analysis. The temporal framework is not extended to the present as well in order

to preclude the outcomes of the Arab Spring, since I contend that it is still an on-going process,

albeit with its ruptures. In order to ensure consistency across different dimensions of the data;

Palestine, Israel, and Iraq have been omitted from the analysis. Major historical events in the

specified time period such as Iraqi War in 2003 as well as the ambiguity of the Palestinian-Israeli

contestation played a role in this decision.

In order to assess labor market protection, openness of the economies has been measured

in the first place (see Seguro-Ubiergo 2007), by compiling relevant data from Penn World

Tables. Although different regime types, institutional settings and organization of trade unions

response differently to exogenous shocks (see Rudra and Haggard 2005), openness of the

economies should still be taken into consideration for the developing context without any doubt,

12

as a voluminous literature focuses on this topic elaborated in the literature review. Be that as it

may, in the light of the interpretation of F-test, this variable is omitted from the analysis due to its

low contribution to the classification of the cases. Secondly, the different configuration of

pension schemes engenders different outcomes for labor market protection vis-à-vis

unemployment, sickness and maternity benefits. Moreover, informal labor is an exigent issue that

these countries need to cope with. New laws were promulgated especially to address the issue of

foreign migrant workers. For instance, while distinct pension scheme was constructed for these

workers with Law 50 / 1978 in Egypt (see Helmy 2008), Bahrain and Jordan incorporated them

into the general pension scheme (Loewe 2009). Therefore, effective coverage rate of pension

schemes has been employed for analysis (see Loewe 2009). However, in the light of the

examination of F-test, this variable is omitted as well. As a result, four main variables are

identified to capture labor market protection; namely, the number of occupationally distinct

pension schemes (occup), the number of ILO conventions in force including fundamental,

governance, and technical conventions (ilo), de facto labor market standards (defacto), and the

level of democracy (democracy).

The number of occupationally distinct pension schemes has been utilized for

operationalizing welfare state stratification in terms of the degree of corporatism (Esping-

Andersen 1990, 71). The multiplicity of the pension schemes exacerbates the fragmentation of

the labor force as well as income inequality, since the ceiling on the monthly covered wage,

benefit technique and pension eligibility terms are defined separately for each scheme (Helmy

2008; Loewe 2009, 13). Hence, the higher number of distinct pension schemes is expected to

affect labor market protection negatively. Secondly, the number of ILO conventions in effect is

expected to reflect institutional restrictions for promulgation of laws related to labor issues. As

Rudra (2007) claims, low number of ratified ILO conventions might exhibit low level of labor

market protection. Yet convention ratification does not necessarily imply full protection, since it

is open to dispute whether they are really implemented and enforced. For instance, during

decade-long negotiations for new labor law of Egypt that was finally promulgated in 2003, in

order to address the issue of excess workers in the public sector, the early retirement scheme was

introduced, which workers might take voluntarily. However, several complaints have been

reported that workers are “forced” to take that early retirement package (Pacynzska 2006, 57).

Convention 87 exemplifies this point further – Both Syria and Egypt ratified this convention, but

13

they only allow for one trade union confederation, which directly violates the convention

(Cammett and Posusney 2010, 259). As a consequence, in addition to de jure aspect of labor

market standards, their de facto implementation should be emphasized, too. To this aim, scores

for de facto labor standards, measured by Cammett and Posusney (2010), are employed. De facto

labor standards are measured with freedom of association; right to bargain collectively; right to

strike; unfair labor practices; murder of trade unionists; harassment, intimidation, detention,

arrest, or forced exile of trade unionists; violations of rights to union formation and/or collective

bargaining in export processing zones (Stallings 2010, 147). Finally, democratic countries are

expected to be more conducive to the freedom of unionization, formation of opposition and

constitution of cross-class alliances to “lock existing policy arrangements into place” (Pierson

2001, 414). Although some scholars argued that regime type is epiphenomenal to social policy

outcomes (Mares and Carnes 2009, 101), regime types influence the extent of effective coverage

rate. For instance, on average, the effective coverage rate is 22 % in the MENA monarchies while

44 % in the MENA republics (Loewe 2009, 10).

In order to operationalize welfare-mix, six different variables are used; namely public

health expenditure as a percentage of total government expenditure (health), out-of-pocket health

expenditure as a percentage of total expenditure on health (outofpocket), public spending on

expenditure as percentage of total government expenditure (exped), oil and natural gas rents as

percentage of GDP (rent), personal remittances received as percentage of GDP (remittance), and

net development and foreign aid (logaid). Firstly, to envisage public-private mix, government

expenditures on education and health are measured. As robustness check, public health

expenditure as percentage of GDP is used in the alternative specification and final clusters did not

change. Out-of-pocket health expenditure refers to any direct outlay by households, including

gratuities and in-kind payments, which are not reimbursed by insurance. In spite of the existence

of insurance and pension schemes, this part of private health expenditure imposes a great burden

over the segment of the population with low level of income. For example, in Egypt, public

hospitals are supposed to be “free of charge”. In effect, “free of charge” is called “abu balâsh” in

Egypt, with all its negative connotations (Tadros 2006, 248).

It is important to note that the categorization of different welfare regimes cannot be

realized by simply tallying government expenditures (Esping-Andersen 1990, 3). As Esping-

Andersen (1990, 19) rigorously argued, “Expenditures are epiphenomenal to the theoretical

14

substance of the welfare states. (…) By scoring welfare states on spending, we assume that all

spending counts equally”. For instance, despite encompassing efforts for privatization and

declining levels of government expenditure through structural adjustment programs in the region,

the state is not really withdrawn from economy (Ayubi 2000, 161) and still exerts “hegemonic

control” (Eder 2009). In Egypt, Mubarak was notorious for ERSAP (Economic Reform and

Structural Adjustment Program), which actually decreased government expenditures in some

respects by praising neoliberal ideology. Yet Harrigan and El-Said (2009, 90) revealed that

expenditure on health and education as percentage of total government expenditure remained

same in time: They evaluated this as a sign of transformation to productive - developmentalist

state. But the spending measures may not capture effective government spending (Rudra and

Haggard 2005, 1024). Therefore the welfare outcomes should be underlined, too (Gough and

Wood 2004, 48). To measure welfare outcomes, Human Development Index Score (HDI) is

selected. HDI is calculated based on four indicators; namely, life expectancy at birth, mean years

of schooling, expected years of schooling, and GNI per capita. As a robustness check, Non-

income HDI is applied, too. In the alternative specification, the final clusters did not change.

Furthermore, additional indicators are sine qua non for capturing the state “capacity” for

financing its operations as well as the “international” dimension of public-private mix.

Remittances constitute a great portion of revenue for developing countries. Especially expatriates

working in oil-rich countries such as Bahrain, Kuwait, and Qatar send back their remittances to

their home countries. The “international” dimension of this household strategy should be

highlighted for clustering Middle Eastern welfare regimes. In addition, some countries rely

heavily on either natural gas (such as Iran) or oil rents (such as UAE, Bahrain, Kuwait, Libya,

Saudi Arabia, Qatar, and Oman) to finance their expenditures. For instance, in Egypt, Sadat’s

infitah policy (Law No. 43 / 1974) has been underscored as one of the cornerstones of economic

reform in Egypt. The state was supposed to reduce its expenditures. It did not happen especially

in the aftermath of the bread riots. In 1980, that was possible to finance those food subsidies –

They were not financed by general taxes but external rents. Oil rents at that time amounted to

33.06 % of GDP in Egypt8. Capitalizing on oil revenues, Sadat even funded various social benefit

programs and bureaucracy was not curtailed in that time period (Ayubi 1995, 131). Yet the

scrutiny of these countries should not be reduced to the supply of rent. As this paper tries to

8 Source: World Bank.

15

achieve, other dimensions such as labor market protection should be discerned in order to account

for variation in the political economy groupings. The existence of rent does not entail government

commitment for productive public investments (Ross 2013) or its lack does not necessarily imply

retrenchment for Middle Eastern states, as elaborated in the literature review. The last but not

least, foreign aid can be construed as one of the most significant mechanisms for ensuring state

capacity, although foreign aid does not only aim for development but buttressing military

establishments as in the case of Egypt and Saudi Arabia (see Murat 2006). In similar vein,

foreign aid might not be translated into effective government commitment to development but

expended for clientelistic purposes. Therefore it is vital to check for welfare outcomes.

Table 1: Final Clusters and Cluster Centers

Cluster 1 Cluster 2 Cluster 3 Cluster 4

Kuwait Jordan Libya Morocco

Oman Lebanon Saudi Arabia Tunisia

Bahrain Turkey UAE Algeria

Qatar Egypt

Syria

Iran

Yemen

Dimensions Final Cluster Centers

Labor Market Protection ilo -1,09471 ,41400 -,73246 ,56868

defacto ,50196 ,41408 -1,86065 ,21657

occup -,45132 -,53096 -1,00883 ,87988

democracy -,22698 1,65065 -1,15411 -,17088

Welfare-Mix rent ,52257 -1,16227 ,36154 -,34513

remittance -,59693 1,49046 -,60076 ,02895

logaid -,77236 ,56314 -1,34175 ,61770

outofpocket -1,09563 -,08752 -,54254 ,63313

health -,02637 1,32556 -,11282 -,14876

exped -,45490 -,78936 1,21890 ,33151

Welfare Outcome hdi ,88913 ,09213 ,85149 -,73443

16

Table 2: Distances between Final Cluster Centers

Cluster 1 2 3 4

1 4,288 3,217 3,736

2 4,288 5,639 3,579

3 3,217 5,639 4,468

4 3,736 3,579 4,468

Interpretation

17 countries in the region are classified into four distinct clusters by the application of k-

means cluster analysis (See Table 1), which can be labelled as “protective oil-rich” (Cluster 1),

“non-protective oil-rich” (Cluster 3), “mixed” (Cluster 2), and “corporatist regimes” (Cluster 4).

All variables are significant at p=0,05. Final cluster centers show how each variable contributes

to discerning the differences between clusters while distances between final cluster centers

exhibit how distinct these clusters are from each other (See Table 2).

Preliminary examination of the distance between cluster centers suggests that protective

and non-protective oil-rich countries are closer to each other than corporatist and mixed

countries. Corporatist and mixed regimes relatively ratified more conventions than protective and

non-protective oil-rich regimes. Yet protective and non-protective oil-rich countries are

demarcated based on their de-facto labor standards. As Cammett and Posusney (2010, 257)

vividly clarified, oil-rich countries encompass wide variation. While protective oil-rich countries

exhibit good performance even exceeding the level of other clusters in terms of de facto labor

standards, in non-protective oil-rich countries, the virtual absence of both legal and de facto labor

standards is observable.

The number of occupationally distinct pension schemes exhibits the degree of

corporatism. For instance, in Egypt and Tunisia, Egyptian Trade Union Federation and Tunisian

General Labor Union (UGTT) are supposed to represent workers in an encompassing way.

Egpyt’s pension system was fragmented into six subgroups, mainly comprising public sector

employees, private sector employees, professionals, Egyptian migrant workers living abroad,

army and top bureaucracy, and others covered through Comprehensive Social Security System

17

(Loewe 2004, 7)9. On the other hand, mixed regimes display lower level of fragmentation in

terms of the pension schemes. For instance, in Turkey, three distinct social security schemes

(Emekli Sandığı, SSK, and Bağ-Kur) were merged into one encompassing scheme under the

caveat of Social Security Institution in 2006. In Jordan, by 2003, separate pension programs for

military recruits and civil servants vanished as these groups began to be enrolled on equal terms

with private sector workers. In similar vein, protective oil-rich countries exhibit relatively lower

levels of fragmentation of pension schemes as well. Especially the integration of migrant workers

into general public pension schemes should be underscored in this regard, as it was realized in

Bahrain. Yet for Bahrain, and Oman, there exists still a separate pension scheme for military and

civil servants (Loewe 2009, 12). Although the lowest level of fragmentation is observed in non-

protective oil-rich regimes, this outcome should be interpreted in relation with other indicators of

labor market protection. Saudi Arabia widened the scope of its general pension scheme while

Libya, with its effective coverage rate of 80 % (Loewe 2009), incorporated migrant workers into

its pension structure. Yet it does not necessarily imply high level of labor market protection for

all constituencies. In fact, Saudi Arabia scored “zero” in de facto labor standards (Cammett and

Posusney 2010).

The lowest level of democracy is measured for non-protective oil-rich monarchies, which

supports initial hypothesis stated above. In these countries, unionization has been curtailed and

virtually non-existing. The number of ratified ILO conventions and the level of democracy

constitute the demarcation line between mixed regimes and protective oil-rich countries in terms

of the labor market protection, although their records are similar with respect to de facto labor

standards. Although in 2002, the right to organize collectively is granted in Bahrain, only 10 ILO

conventions are in effect. By contrast, although any labor action was responded with security

apparatus in Jordan, labor insurgency clearly shaped domestic politics, which is epitomized by

political actions of day laborers – working directly for government ministries – since 2006 (see

9 The fragmentation of the labor market does not necessarily imply labor acquiescence. For instance, in

Egypt, Unified Labor Law could not be promulgated until 2003 due to pressure from the Egyptian Trade

Union Confederation, which could wring concessions from the state at the end (Pacynzska 2006). In

Egypt, from 1998 to 2004, there were over 1000 workers’ collective actions, of which more than a quarter

occurred in 2004 alone (Beinin 2009, 77). The labor strike initiated by Misr Spinning and Weaving

Company in Mahalla al-Kubra reverberated across the country and challenged both the supremacy of the

legal trade union federation as well as the regime itself (Beinin 2009, 80). Moreover, international Crisis

Group (2011) highlighted the role of the trade unions (UGTT) for ousting Ben Ali in Tunisia.

18

Adely 2013). The number of ILO conventions in effect is 23, 49, and 52 for Jordan, Lebanon, and

Turkey respectively.

The capacity for financing social welfare provision constitutes a major component of the

clustering. Kuwait, Oman, Bahrain, Qatar, Libya, Saudi Arabia, and UAE definitely diverge from

other countries in terms of their rent income, despite the fact that almost half of the Iran’s GDP in

2005 did stem from oil and natural gas rents10

. Be that as it may, it is not the main defining

feature of Iranian social welfare system and its relatively lower welfare outcome and different

pattern of remittance are the main reasons for its specific location in this clustering. Oil-rich

countries record higher levels of HDI in comparison with other clusters and in general they

receive migrant workers, who then send their remittances back to their home countries.

Therefore, low levels of remittances are observed in both types of oil-rich countries. Another

demarcation line between mixed and corporatist regimes and oil-rich countries is the fact that

they rely heavily upon foreign aid and development assistance.

The composition of welfare-mix in terms of health spending distinguished corporatist

regime from mixed regime. In the former one, the relative role of the state has declined in time in

the course of structural adjustment programs. For instance, in Tunisia, the percentage of the

government expenditure in total healthcare spending decreased from 51,8 % in 1985 to 31 % in

2000. By the same token, the share of the households has risen from 35,7 % to 49,1 % in the

same time period. To put in a nutshell, half of all health spending placed burden over the

shoulders of the households (Romdhane 2006, 56). On the other hand, mixed regimes exhibit

patterns of higher level of government expenditure on health and lower level of household

expenditure. Yet it does not necessarily imply either developmentalist state or complex

configuration of public health care system. Lebanese health services exemplify this point. 2006

Lebanon War devastated country’s infrastructure and the provision of public services based on

sectarian divisions was exacerbated. In the aftermath of severe civil war, followed by various

political crisis and military conflicts, public hospitals were either destroyed or closed. As a

corollary, the state has to buy services from public hospitals. In addition, The Ministry of Health

tries to cover the costs for the citizens, who are not covered by any insurance (Sibai and Sen

10

Source: World Bank.

19

2006). On the other hand, for oil-rich countries, the welfare-mix in terms of expenditure on health

has been tilt towards private household strategies.

In effect, to be able to discern the real divergences in welfare-mix in terms of expenditure

on education, data for private expenditure on education should be compiled as well, which could

not be achieved due to data limitations. Yet this variable yields important insight into

classification of these clusters in relation to welfare outcomes. Relatively higher level of

government expenditure on education has not been translated into welfare outcomes in the case

of corporatist regimes. Moreover, the mixed regimes and protective oil-rich countries converge to

each other in terms of the relatively low levels of the government expenditure on education.

Table 3 summarizes the main defining features of the clusters as well as the relative points

of divergence and convergence for the clustering.

20

Table 3: A Summary for the Relative Points of Convergence and Divergence for Clustering

Protective Oil-rich

Regimes

Non-protective Oil-

rich Regimes

Corporatist Regimes Mixed Regimes

Points of Convergence Points of Convergence

Low number of ratified ILO conventions High number of ratified ILO conventions

High level of rent income Low level of rent income

Low level of remittances High level of remittances

Low level of foreign aid High level of foreign aid

High level of private expenditure on health

Points of Divergence Points of Divergence

High level of de-facto

labor standards

Low level of de-facto

labor standards

High degree of

corporatism

Low degree of

corporatism

High level of

government expenditure

on education

Low level of

government

expenditure on

education

High level of private

expenditure on health

High level of public

expenditure on health

Cases

Kuwait

Oman

Bahrain

Qatar

Libya

Saudi Arabia

UAE

Cases

Morocco

Tunisia

Algeria

Egypt

Syria

Iran

Yemen

Jordan

Lebanon

Turkey

21

Concluding Remarks

This paper attempted to realize the confluence of the study of Middle Eastern political

economy and the study of welfare regimes in the developing countries. By revealing the untested

assumptions of generic concept of rentier state and its implications for the literature and

countering the “prospect” for the residual state in the developing contexts due to globalization;

this paper strived to contemplate systemic divergences in the welfare regimes in the Middle East.

By embracing the welfare regime approach and applying cluster analysis, 17 countries in the

region were clustered into four distinct categories, namely protective oil-rich, non-protective oil-

rich, mixed, and corporatist regimes on the basis of three dimensions; namely welfare-mix,

welfare outcome and labor market protection.

By the time this paper was written; Iraq was in political ferment, civil war erupted in

Syria and Libya, and a coup d’état was realized in Egypt. This turmoil has had severe

repercussions for those countries, especially in terms of the quality of governance. These political

circumstances affected the process of data collection as well. In spite of international institutions’

struggle for standardizing data for Middle Eastern countries in order to facilitate cross-national

comparisons; for significant aspects of comparative political economy, this goal could not be

achieved yet. Hence further studies should attempt to elaborate on thorough analysis of social

policy designs and to construct various indexes and models, which capitalize on standardized

data and can be applied to different countries in the region. In addition to the conventional

variables, those models should figure out new unbiased estimators for both labor power and the

extent of non-state actors. Number of legal and illegal strikes, number of NGOs for social and

economic development, the private networks’ expenditure on health and education as a

percentage of total expenditure on health and education, the level of informality epitomize these

possible estimators. Without any doubt, this vigorous study necessitates meticulous fieldwork in

the region.

22

Bibliography

Abrahamson, Peter. 1999. “The Welfare Modelling Business.” Social Policy & Administration 33

(4): 394-415.

Adely, Fida. 2013. “The Emergence of a New Labor Movement in Jordan.” MERIP 42 (Fall).

Anderson, Lisa. 1987. “The State in the Middle East and North Africa.” Comparative Politics 20

(1): 1-18.

Arts, Wil A., and John Gelissen. 2010. “Models of the Welfare State.” In The Oxford Handbook

of the Welfare State, eds. Francis G. Castles et al. New York: Oxford University Press, 569-

85.

Ayubi, Nazih. 1995. Overstating the Arab State: Politics and Society in the Middle East. London:

I.B. Taurus.

———. 2000. “Etatisme versus Privatization: The Changing Economic Role of the State in Nine

Arab Countries.” Economic Transition in the Middle East: Global Challenges and

Adjustment Strategies. Cairo: The American University in Cairo Press, 125-66.

Beblawi, Hazem. 1987. “The Rentier State in the Arab World.” In The Rentier State, eds. Hazem

Beblawi and Giacomo Luciani. New York: Croom Helm.

Beinin, Joel. 2009. “Workers’ Struggles under ‘Socialism’ and Neoliberalism.” In Egypt: The

Moment of Change, eds. Rabab El Mahdi and Philip Marfleet. London: Zed Books, 68-86.

Bellin, Eva. 2012. “Reconsidering the Robustness of Authoritarianism in the Middle East:

Lessons from the Arab Spring.” Comparative Politics 44 (January): 127-49.

Bibars, Iman. 2001. Victims & Heroines: Women, Welfare and the Egyptian State. London: Zed

Books.

Blaydes, Lisa. 2011. Elections and Distributive Politics in Mubarak’s Egypt. New York:

Cambridge University Press.

Brumberg, Daniel. 1990. “An Arab Path to Democracy?” Journal of Democracy 1 (4): 120-25.

Burgoon, Brian. 2001. “Globalization and Welfare Compensation.” International Organization

55 (3): 509-51.

Cameron, David. 1978. “The Expansion of the Public Economy: A Comparative Analysis.”

American Political Science Review 93 (4): 779-90.

Cammett, Melani Claire, and Lauren M. MacLean. 2011. “Introduction: The Political

Consequences of Non-state Social Welfare in the Global South.” Studies in Comparative

International Development 46 (1): 1-21.

Cammett Melani, and Marsha Pripstein Posusney. 2010. “Labor Standards and Labor Market

Flexibility in the Middle East: Free Trade and Freer Unions?” Studies in Comparative

International Development 45 (2): 250-79.

Carnes, Matthew E., and Isabela Mares. 2009. “The Welfare State in Global Perspective.” In The

Oxford Handbook of Comparative Politics, eds. Carles Boix and Susan Stokes. USA: Oxford

University Press, 868-885.

23

Clark, Janine A. 2004. Islam, Charity, and Activism: Middle-Class Networks and Social Welfare

in Egypt, Jordan, and Yemen. Indiana: Indiana University Press.

Ebaugh, Helen Rose. 2010. The Gülen Movement: A Sociological Analysis of a Civic Movement

Rooted in Moderate Islam. New York: Springer.

Eder, Mine. 2010. “Retreating State? Political Economy of Welfare Regime Change in Turkey.”

Middle East Law and Governance 2 (2): 152-84.

El-Ghonemy, M. Riad. 1998. Affluence and Poverty in the Middle East. London and New York:

Routledge.

El-Meehy, Asya. 2009. “Rewriting the Social Contract: The Social Fund and Egypt’s Politics of

Retrenchment.” Ph.D. diss. University of Toronto.

Esping-Andersen, Gøsta. 1990. The Three Worlds of Welfare Capitalism. Cambridge: Polity

Press.

———, ed. 1996. Welfare States in Transition: National Adaptations in Global Economies.

London: Sage Publications.

Ferrera, Maurizio. 1996. “The ‘Southern’ Model of Welfare in Social Europe”. Journal of

European Social Policy 6 (1): 17-37.

Franzoni, Juliana Martínez. 2008. “Welfare Regimes in Latin America: Capturing Constellations

of Markets, Families, and Policies.” Latin American Politics and Society 50 (2): 67-100.

Garrett, Geoffrey, and Deborah Mitchell. 2001. “Globalization, Government Spending and

Taxation in the OECD.” European Journal of Political Research 39 (2): 145-77.

Gough, Ian. 2001. “Social Assistance Regimes: A Cluster Analysis.” Journal of European Social

Policy 11 (2): 165-70.

Gough, Ian, and Geof Wood, eds. 2004. Insecurity and Welfare Regimes in Asia, Africa and Latin

America: Social Policy in Development Contexts. Cambridge: Cambridge University Press.

Grossman, Guy. 2007. “Towards A Typology of Welfare Regimes in the Arab World: The

Middle East and North Africa in Historical and Comparative Perspective.” Presented at

Columbia University Graduate Students Conference.

Haggard, Stephan, and Robert R. Kaufman. 2008. Development, Democracy, and Welfare States:

Latin America, East Asia and Eastern Europe.

Hall, Peter A., and David Soskice, eds. 2001. Varieties of Capitalism: The Institutional

Foundations of Comparative Advantage. New York: Oxford University Press.

Harrigan, Jane, and Hamed El-Said. 2009. Economic Liberalisation, Social Capital and Islamic

Welfare Provision. New York: Palgrave Macmillan.

Helmy, Omneia. 2008. “Toward a More Efficient and Equitable Pension System in Egypt.” The

Egyptian Economy: Current Challenges and Future Prospects, ed. Hanaa Kheir-El-Din.

Cairo: The American University in Cairo Press, 201-34.

Heynemann, Stephen P, ed. 2004. Islam and Social Policy. Nashville: Vanderbilt University

Press.

24

Huber, Evelyne. 1996. “Options for Social Policy in Latin America: Neoliberal vs. Social

Democratic Models.” In Welfare States in Transition: National Adaptations in Global

Economies, ed. Gøsta Esping-Andersen. London: Sage Publications.

Huber, Evelyne, ed. 2002. Models of Capitalism: Lessons for Latin America. Pennsylvania: The

Pennsylvania State University Press.

Huber, Evelyne, and John D. Stephens. 2001. Development and Crisis of the Welfare States:

Parties and Policies in Global Markets. Chicago: The University of Chicago Press.

International Crisis Group (ICG). 2011. “Popular Protest in North Africa and the Middle East:

Tunisia’s Way.” Middle East and North Africa Report 106.

Ismail, Salwa. 2006. Political Life in Cairo’s New Quarters: Encountering the Everyday State.

Minneapolis: University of Minnesota Press.

Iversen, Torben, and Thomas R. Cusack. 2000. “The Causes of Welfare State Expansion:

Deindustrialization or Globalization?” World Politics 52 (3): 313-49.

Jawad, Rana. 2009. Social Welfare and Religion in the Middle East: A Lebanese Perspective.

Bristol: The Policy Press.

Jawad, Rana, and Burcu Yakut-Cakar. 2010. “Religion and Social Policy in the Middle East: The

(Re) Constitution of an Old-New Partnership.” Social Policy & Administration 44 (6): 658-

72.

Jones, Catherine. 1993. “The Pacific Challenge: Confucian Welfare States.” In New Perspectives

on the Welfare State in Europe, ed. Catherine Jones. London: Routledge, 198-217.

Kamel, Maggie Samir. 2001. “Redrawing State-Society Boundaries: Egypt’s Dynamic Social

Contract.” Ph.D. diss. Durham University.

Kangas, Olli. 1994. “The Politics of Social Security: On Regressions, Qualitative Comparisons,

and Cluster Analysis.” In The Comparative Political Economy of the Welfare State, ed.

Thomas Janoski and Alexander M. Hicks. New York: Cambridge University Press, 346-64.

Karshenas, Massoud, and Valentine M. Moghadam, eds. 2006. Social Policy in the Middle East:

Economic, Political, and Gender Dynamics. New York: Palgrave MacMillan.

Kennett, Patricia, and Ben Oakley. 2001. “Development, Social Welfare and Cross-national

Analysis.” In Comparative Social Policy: Theory and Research, ed. Patricia Kennett.

Philadelphia: Open University Press, 92-115.

Korpi, Walter. 1983. The Democratic Class Struggle. London: Routledge Kegan & Paul.

Kwon, Huck-Ju. 2005. Transforming the Developmental Welfare State in East Asia. New York:

Palgrave Macmillan.

Kwon, Huck-Ju, Thandika Mkandawire, and Joakim Palme. 2009. “Introduction: Social Policy

and Economic Development in Late Industrializers.” International Journal of Social Welfare

18: 1-11.

Loewe, Markus. 2004. “New Avenues to be Opened for Social Protection in the Arab World: The

Case of Egypt.” International Journal of Social Welfare 13: 3-14.

25

———. 2009. “Pension Schemes and Pension Reforms in the Middle East and North Africa.”

UNRISD Project: Financing Social Policy: Pension Funds and Economic Development.

Mahdavy, Hussein. 1970. “The Patterns and Problems of Economic Development in Rentier

States: The Case of Iran.” In Studies in Economic History of the Middle East, ed. M. A. Cook.

London: Oxford University Press.

Mares, Isabela. 2003. The Politics of Social Risk: Business and Welfare State Development.

Cambridge: Cambridge University Press.

———. 2005. “Social Protection around the World: External Insecurity, State Capacity, and

Domestic Political Cleavages.” Comparative Political Studies 38 (6): 623-51.

Mares, Isabela, and Matthew E. Carnes. 2009. “Social Policy in Developing Countries.” The

Annual Review of Political Science 12: 93-113.

Marfleet, Philip. 2009. “State and Society.” In Egypt: The Moment of Change, eds. Rabab El

Mahdi and Philip Marfleet. London: Zed Books, 14-33.

Mesa-Lago, Carmelo. 2003. Market, Socialist, and Mixed Economies: Comparative Policy and

Performance; Chile, Cuba, and Costa Rica. Maryland: John Hopkins University Press.

Mkandawire, Thandika. 2001. “Social Policy in a Development Context.” UNRISD Social Policy

and Development Programme 7 (June).

Murat, Ömer. 2006. “The Role of the Egyptian Military as an Economic Actor in Business.”

Master Thesis. Boğaziçi University.

Pacynzska, Agnieszka. 2006. “Globalization, Structural Adjustment, and Pressure to Conform:

Contesting Labor Law Reform in Egypt.” New Political Science 28 (1): 45-64.

Pierson, Paul. 1996. “The New Politics of the Welfare State.” World Politics 48 (2): 143-79.

———. 2001. “Coping with Permanent Austerity: Welfare State Restructuring in Affluent

Democracies.” In The New Politics of the Welfare State, ed. Paul Pierson. New York: Oxford

University Press.: 410-56.

Posusney, Marsha Pripstein. 1997. Labor and the State in Egypt: Workers, Unions, and

Economic Restructuring. New York: Columbia University Press.

Powell, Martin, and Armando Barrientos. 2004. “Welfare Regimes and the Welfare Mix.”

European Journal of Political Research 43 (1): 83-105.

Pratt, Nicola. 2000/2001. “Maintaining the Moral Economy: Egyptian State-Labor Relations in

an Era of Economic Liberalization.” The Arab Studies Journal 8/9 (2/1): 111-29.

Rieger, Elmar, and Stephan Leibfried. 1998. “Welfare State Limits to Globalization.” Politics &

Society 26 (3): 363-90.

Riesco, Manuel. 2007. Latin America: A New Developmental Welfare State Model in the

Making? New York: Palgrave MacMillan.

Rodrik, Dani.1997. Has Globalization Gone too Far? Washington: Institute for International

Economics.

———. 1998. “Why Do More Open Economies Have Bigger Governments?” Journal of

Political Economy 106 (5): 997-1032.

26

Romdhane, Mahmoud Ben. 2006. “Social Policy and Development in Tunisia since

Independence: A Political Perspective.” In Social Policy in the Middle East: Economic,

Political, and Gender Dynamics, eds. Massoud Karshenas and Valentine M. Moghadam.

New York: Palgrave Macmillan, 31-77.

Ross, Michael L. 2013. The Oil Curse: How Petroleum Wealth Shapes the Development of

Nations. Princeton: Princeton University Press.

Roy, Olivier. 1994. The Failure of Political Islam. Cambridge: Harvard University Press.

Rudra, Nita, and Stephan Haggard. 2005. “Globalization, Democracy, and Effective Welfare

Spending in the Developing World.” Comparative Political Studies 38 (9): 1015-49.

Rudra, Nita. 2002. “Globalization and the Decline of the Welfare State in Less-Developed

Countries.” International Organization 56 (2): 411-45.

———. 2007. “Welfare States in Developing Countries: Unique or Universal?” The Journal of

Politics 69 (2): 378-96.

———. 2008. Globalization and the Race to the Bottom in Developing Countries: Who Really

Gets Hurt? Cambridge: Cambridge University Press.

Saint-Arnaud, Sébastien, and Paul Bernard. 2003. “Convergence or Resilience? A Hierarchical

Cluster Analysis of the Welfare Regimes in Advanced Countries.” Current Sociology 2003

51 (5): 499-527.

Segura-Ubiergo, Alex. 2007. The Political Economy of the Welfare State in Latin America:

Globalization, Democracy, and Development. Cambridge: Cambridge University Press.

Sibai, Abla-Mehio, and Kasturi Sen. 2006. “Can Lebanon conjure a public health phoenix from

the ashes?” BMJ 333 (October): 848-49.

Stallings, Barbara. 2010. “Globalization and Labor in Four Developing Regions: An Institutional

Approach.” Studies in Comparative International Development. 45 (2): 127-50.

Sullivan, Denis J. 1994. Private Voluntary Organizations in Egypt: Islamic Development, Private

Initiative, and State Control. Florida: The University Press of Florida.

Swenson, Peter A. 2002. Capitalists against Markets: The Making of Labor Markets and Welfare

States in the United States and Sweden. New York: Oxford University Press.

Şener, Gözde. 2006. “Classification of Welfare Regimes Using Cluster Analysis: Where Does

Turkey Stand?” Master’s Thesis. Boğaziçi University.

Tadros, Mariz. 2006. “State Welfare in Egypt since Adjustment: Hegemonic Control with a

Minimalist Role.” Review of African Political Economy 33 (108): 237-54.

Walker, Alan, and Chack-Kie Wong. 1996. “Rethinking Western Construction of the Welfare

State.” International Journal of Health Services 26 (1): 67-92.

Waterbury, John. 1983. The Egypt of Nasser and Sadat. New Jersey: Princeton University Press.

Wibbels, Erik. 2006. “Dependency Revisited: International Markets, Business Cycles, and Social

Spending in the Developing World.” International Organization 60 (Spring): 433-68.

Wibbels, Erik, and John Ahlquist. 2007. “Development Strategies and Social Spending.”

Estudio/Working Paper 232 (June).

27

Wickham, Carrie Rosefsky. 2002. Mobilizing Islam: Religion, Activism, and Political Change in

Egypt. New York: Columbia University Press.

Wood, Adrian. 1997. “Openness and Wage Inequality in Developing Countries: The Latin

American Challenge to East Asian Conventional Wisdom.” The World Bank Economic

Review 11 (1): 33-57.

Yousef, Tarik M. 2004. “Development, Growth, and Policy Reform in the Middle East and North

Africa since 1950.” Journal of Economic Perspectives 18 (3): 91-116.

Data Sources:

ILO Social Security Department

IMF Government Finance Statistics (various years)

Penn World Table

Social Security Program throughout the World

United Nations Data

World Bank