CHAPTER 1 Economic Consequences of The

FEMISE Euromed Report 2017

Economic Management under Fire

Ahmed Galal

Ishac Diwan Ibrahim Elbadawi Hoda Selim Zafiris Tzannatos Jala Youssef

Funded by the European Union First published in 2017 FEMISE and ERF

This report received financial support from the European Union through the FEMISE project on “Support to Economic Research, studies and dialogues of the Euro-Mediterranean Partnership”. Any views expressed in this report are the sole responsibility of the authors. Contents

Preface ...... iv

Introduction ...... 1 Chapter One: Economic Consequences of the Transition ...... 7 1.1 Introduction ...... 7 1.2 Macroeconomic Balances ...... 10 1.3 Economic Growth: Actual ...... 17 1.4 Economic Growth: Potential ...... 21 1.5 Conclusions ...... 24

Chapter Two: Macroeconomic Policy Response ...... 31 2.1 Introduction ...... 31 2.2 Understanding the Austerity versus Stimulus Debate ...... 32 2.3 Fiscal Policy Response ...... 36 2.4 Monetary Policy Response ...... 44 2.5 Fiscal and Monetary Policy Consistency ...... 56 2.6 Conclusions ...... 58

Chapter Three: Social Policy Response ...... 65 3.1 Introduction ...... 65 3.2 Social Development ...... 67 3.3 Response to Social Demands ...... 72 3.4 Assessment of the Social Response to Date ...... 83 3.5 Conclusions ...... 87

Chapter Four: The Political Transition and Economic Development ...... 91 4.1 Introduction ...... 91 4.2 Transition Challenges, Politics and Institutions ...... 93 4.3 Regime Evolution in the Arab Countries ...... 96 4.4 The Evolution of the State and Fiscal Policy ...... 100 4.5 The Growth Process and the Evolution of State-Business Relations ...... 105 4.6 Macro-Management During and After Regime Changes ...... 111 4.7 Conclusions ...... 114

Chapter Five: Changing Course ...... 121 5.1 Introduction ...... 121 5.2 Restoring and Sustaining Macro-Balances and Economic Growth ...... 122 5.3 Towards More Effective Social Protection and Social Mobility ...... 126 5.4 Twinning Economic Reform and Politics ...... 128 5.5 A Possible Role for External Actors ...... 130 5.6 Conclusions ...... 132

Preface

In selecting the topic of this year’s Euro-Med report, it was difficult not to focus on macroeconomic management in South-Med countries. The uprisings that stormed Tunisia in December 2010 and spread thereafter to other countries in the region were associated with serious economic downturns, worrying macro imbalances and rising poverty. At the same time, aspirations for a better quality of life were intensely high among the populations. Policymakers in transition countries have had to strike a balance between meeting these aspirations and ensuring that their economies fare well in the future. This report is about assessing their efforts in navigating the transition and making suggestions for a better future. The report focuses on five countries: Algeria, Egypt, Jordan, Morocco and Tunisia. It deals with economic management of the transition from different perspectives: macroeconomic management, economic growth and employment, social concerns, and political developments. While the analysis is mostly positive in nature, the report concludes with a normative chapter, prescribing a better way forward. The key findings of the report are summarized in the introduction, and will not be reiterated here. Instead, I would like to take this opportunity to thank the contributors to the report. Without their hard work and insights, it would not have been possible to put this report together. I also want to thank the European Commission for the continued financial support to FEMISE over the years. This support enabled FEMISE to create a community of researchers and policymakers with deep interest in sustainable development across the two shores of the Mediterranean. Finally yet importantly, I want to thank Anita Saxena for her excellent editing of the report, Namees Nabeel for ably handling the copy editing, and Maryse Louis for managing the process. This report received financial support from the European Union through the FEMISE project on “Support to Economic Research, studies and dialogues of the Euro-Mediterra- nean Partnership”. Any views expressed in this report are the sole responsibility of the authors.

Ahmed Galal Former President, FEMISE January 2017

Introduction | 1

Introduction

After a long period of political In Morocco and Jordan, the rul- stagnation, the South-Med region ers have retained their hold on is finally seeing the beginnings of power, but have been compelled political and economic transfor- to adopt certain reforms. Morocco mation. The uprisings began in has amended its constitution to Tunisia, at the end of 2010, and grant more power to the parlia- spread soon thereafter to Egypt, ment and has called new parlia- leading to the ousting of Bin Ali mentary elections. In Jordan, the and Mubarak who had ruled King has dismissed several cabi- their countries for decades. Other nets and has promised economic neighboring countries have seen and social reform. In Algeria, as a gradual political opening (Mo- noted above, no major changes in rocco and Jordan), or have fallen leadership or the political system into mired conflict (Syria, Libya). have occurred. And while Leba- Algeria has retained its previous non has managed to maintain a regime, although the 19 year- tricky balance between different long state of emergency has been sectarian groups, Syria and Libya lifted. Lebanon has suffered the are still struggling to find a politi- spillover effects of the Syrian con- cal settlement. Today, the overall flict, but its political system has picture of the region is far from remained essentially the same. settled, and political instability is More than six years later, prog- manifest in the form of protests ress on the political front has been and sometimes violent clashes. partial and uneven. In Tunisia Beyond the boundaries of and Egypt, the uprisings have led transition countries, the regional to the downfall of their autocratic context is dominated by geopoliti- rulers, giving way to the election cal tension and conflict. In Syria, of new presidents, the adoption the transition is turbulent and has of new constitutions and the elec- been tragic in terms of loss of life, tion of new parliaments. Social the displacement of people and and political tension still reigns the destruction of the country’s in- in both countries, however, and stitutions and physical infrastruc- frequent changes in government ture, not to mention the damage are common. A new political wrought to the social fabric of so- settlement is yet to be reached. ciety. The conflict in Syria has had 2 | Introduction

significant negative spillover ef- political transition, budget deficits fects on Jordan and Lebanon, and and public debts have crept up to even on some EU member coun- averages of 9 and 50 percent of tries. To a lesser extent, the vio- GDP respectively. Also, current lence in Libya has had an adverse accounts have come under signifi- impact on Tunisia. And since cant pressure, rising to 9 percent 2014, the decline in oil prices has of GDP on average, as a result of begun to cast a shadow over the declining tourism receipts and, in economies of oil exporters (such some cases, growing trade defi- as Algeria) and, to a lesser extent, cits. Meanwhile, political instabil- oil importers (such as Jordan and ity has triggered capital outflow, Egypt) with important ties to the which in many countries has led GCC. Beyond the turbulent do- central banks to drain foreign re- mestic and regional contexts, the serves in order to maintain fixed global environment remains sub- exchange rate regimes; subse- dued, as advanced economies are quently, financial buffers have only slowly recovering from the dried-up. During the early years, global financial crisis of 2008. weak macroeconomic imbalances Expectedly, the political tran- were reflected in repeated down- sition, coupled with regional and grades of sovereign ratings and in global adversities, has taken a toll higher sovereign borrowing costs. on the economies of the South- GDP per capita in South-Med Med countries. Macroeconomic economies began to pick up mod- conditions have deteriorated, estly as from 2012, with an aver- most visibly in terms of growing age of 1 percent between 2012 macroeconomic imbalances, slug- and 2014. Yet, despite the passing gish economic growth and rising years since the uprising, growth poverty. South-Med economies, has yet to recover to pre-crisis lev- which had gradually been recov- els, amid claims that the political ering from the global financial crisis may have damaged long- crisis of 2008, experienced an im- term/potential growth. More- mediate drop in their GDP per over, in some South-Med coun- capita growth to 0.4% in 2011, tries, such as Egypt and Tunisia, after having reached around 2 macroeconomic imbalances have percent during the years of the persisted. As these have been global financial crisis and even left unaddressed since 2011, they higher levels of close to 4 percent have considerably eroded finan- during 2003-2008 Unemployment cial buffers and have most likely rose sharply in Egypt and Tunisia undermined macroeconomic re- (to above 12 percent), and more silience. If they persist further, alarmingly among the youth, to they may threaten fiscal and ex- above 35 percent. ternal sustainability. In addition Concerning macroeconomic to these short-term problems, the balances, pressures have intensi- long-term challenges for the re- fied as governments have had to gion remain as pressing as ever, succumb to popular demands in including insufficient job creation, an attempt to calm rising social high unemployment (especially unrest. Due to the erosion of fis- among the youth), low private cal space prior to the onset of the sector development and rising Introduction | 3

levels of poverty and inequality. speak of questioned legitimacy. Throughout the transition, poli- In an attempt to evaluate the cymakers in the South-Med econ- transition experiences to date, omies have faced enormous chal- this report takes stock of the eco- lenges, with the need to reignite nomic and social consequences economic growth while restoring of the political transition/shock macroeconomic balances. They in South-Med countries, assesses have been under pressure to re- the policy responses to ameliorate spond to urgent social demands their effects and offers some rec- and yet have limited resources ommendations for the future. It due to declining tourism receipts focuses on the short term and on and remittances and to increasing how different countries navigat- capital flight. They are required to ed the transition during the Arab build stable political institutions, Spring. In this sense, it comple- while their term in office is too ments a number of recent works short. It was also hoped that they on the economics and politics of would begin to establish a solid the Arab Spring (e.g., Diwan and foundation for a modern, pros- Galal, 2016, Amin, et al, 2012, and perous and politically open state, Muasher, 2014), which tended to and yet they have faced strong focus on long-term development demands from rebels for immedi- issues. The report specifically at- ate gratification. tempts to answer the following The trade-offs involved in questions: pursuing different priorities, and • What were the immedi- the policies required to tackle ate economic consequences them, make the task even more of the political transition in problematic. For example, a pri- South-Med countries? ority on resolving macroeconom- • How well did policymakers ic imbalances would require aus- respond to the economic and terity measures, yet these would social adversities associated likely give rise to contractionary with the political transition? effects on output. Alternatively, • Based on the analysis and adopting expansionary policies the likely political evolution (through increased government of these countries, what can expenditure or tax reductions) be done to promote healthier to reignite growth might seem macroeconomic balances, job reasonable, but fiscal expansion creation and greater social in- would put pressure on the gov- clusion? ernment budget and would increase public debt, which might Coverage of the Report hinder long-term growth. Focus- ing on reducing poverty and in- The report focuses on five South- equality would appear to be con- Med transition countries: Algeria, sistent with rebel aspirations, but Egypt, Jordan, Morocco and Tu- resources during transition are nisia. Libya and Syria were left typically in short supply. Post- out because the first order of busi- transition governments have also ness for these countries is to reach faced the additional problem of a political settlement. Lebanon is very short terms in office, not to a special case among the South- 4 | Introduction

Med countries in that it has not republics. The Maghreb countries experienced the kind of transition were occupied by France, while discussed in the report; it also Egypt and Jordan were occupied has a specific political set up that by Britain. Morocco, Algeria and deserves special treatment in its Jordan have some ethnic diversi- own right. ties, whereas Egypt and Tunisia Nevertheless, the five coun- have more homogenous popula- tries analyzed in this report are tions. quite diverse regarding the size of The economic and political di- their economy and population, as versity of these countries has two well as in their level of economic effects on the analysis. On the one development. Egypt has by far hand, this diversity enriches the the largest population, around discussion and brings out more half of the total population of the insights than would have been five countries, and more than 13 possible had the countries been times the size of the Jordanian very similar. On the other hand, population. All five economies the diversity of the countries’ ini- belong to the middle-income tial conditions suggests that gen- group, but income disparities are eralisations across countries and apparent. As an oil exporter, Al- beyond are not admissible, and geria has the highest GNI per cap- that special attention should be ita, while the per capita income of given to country-specific circum- Jordan and Tunisia (above USD stances. 4,000) surpasses that of Egypt and Morocco. Report Main Findings In addition, while the countries covered in the report share The main findings of the report certain cultural and historical are as follows. characteristics, they differ in the • Firstly, just as in other tran- nature of their political regimes, sition countries, South-Med national histories and ethnic economies have suffered a groups. For example, Jordan and deterioration in economic Morocco are monarchies, while outcomes in the wake of the Egypt, Tunisia and Algeria are political upheavals that began

Key Economic Indicators of 5 South-Med Transition Countries, 2014

GNI per capita, GNI per capita, Exports of goods Atlas method (cur- PPP (current inter- and services (% of Population, total rent US $) national $) GDP) Algeria 38,934,334 5490 13880 30.53

Egypt 89,579,670 3210 10280 14.43

Jordan 6,607,000 5160 11910 43.28

Morocco 33,921,203 3070 7290 34.26

Tunisia 10,996,600 4230 11020 44.62 Source: World Development Indicators data, World Bank. Introduction | 5

in December 2010. The mag- the deteriorating macroeco- nitude of the deterioration in nomic conditions was expan- economic performance, rela- sionary in most South-Med tive to the experience of other countries, in accommodation transition countries, has not of popular demands. How- always been worse. With re- ever, all countries, except spect to growth, South-Med Algeria, eventually turned transition countries have not to the IMF for support. In the suffered the same output interim period, the conduct of losses endured by other coun- fiscal and monetary policies tries in transition. By contrast, was neither consistent, nor other indicators (unemploy- well-coordinated. ment, fiscal balances, current • Fourthly, South-Med coun- account balances) have dete- tries have responded differ- riorated in South-Med tran- ently to the social consequenc- sition countries more signifi- es of the political transition, cantly than elsewhere. So far, but their responses have gen- these imbalances have not erally been inadequate and threatened macroeconomic may not have been consistent stability. However, it is not with long-term stability and clear whether finances would dynamic growth. They have be sustainable if macroeco- all moved to enhance social nomic imbalances were to protection, but have failed to persist and if foreign capital address the risks of serious should fail to come through. illness, old age or loss of jobs. • Secondly, the impact of the More importantly, they have political transition on eco- not made significant prog- nomic performance has var- ress in the delivery of better ied between countries. Egypt and more equitable social and Tunisia have been more services. Finally, other than directly exposed to the dis- Tunisia, and to a lesser ex- ruptive effects of political tent Morocco, they have not unrest than Jordan, Morocco delivered on greater freedom and Algeria. However, the for citizens. key factors in explaining the • Fifthly, from a political stand- severity of the economic situ- point, the analysis reveals ation are related to initial that the five South-Med country-specific conditions. countries are far from attain- For example, large pre-ex- ing a competitive political isting fiscal imbalances in environment, although Tuni- Egypt have increased its fis- sia is striving to move in this cal vulnerability to pressure, direction. It is true that Tuni- however small this may be. sia and Egypt have adopted By contrast, in Tunisia, where new constitutions and have fiscal deficits were initially conducted free Presidential very low, the six-fold deterio- and Parliamentary elections, ration has taken the deficit to while Morocco and Jordan only 6 percent of GDP. have adjusted their constitu- • Thirdly, the initial response to tions. However, the emerg- 6 | Introduction

ing political regimes have References yet to reflect these changes. The importance of political Amin, M., R. Assaad, N. Al-Baha- development is that it defines rna, K. Dervis, RM. Desai, NS. the contours of possible eco- Dhillon, A. Galal, H. Ghanem, nomic reform. C. Graham and D. Kaufmann • Sixthly, looking ahead, the (2012) “After the Spring: Eco- reform agenda of South- nomic Transitions in the Arab Med transition countries is World”, Oxford University extensive, especially if the Press. objectives are to bring about Diwan, I. and A. Galal eds. (2016) shared prosperity, on one “The Middle East Economies hand, and political inclusion, in Times of Transition” Inter- on the other. Chapter 5 elab- national Economic Association orates the elements of this Series, vol. 155-V, Palgrave. agenda, and suggests that Muasher, M. (2014) “The Second external partners could play Arab Awakening: And the a positive role in facilitating Battle for Pluralism”, Yale Uni- the process. versity Press.

Structure of the Report

The report is structured as follows. Chapter 1 reviews the impact of the political transition on the macroeconomic performance of Algeria, Egypt, Jordan, Mo- rocco and Tunisia. Chapter 2 takes stock of the macro-policy response to the political shock and assesses the effectiveness of adopted policies. Chapter 3 focuses on the social response to popular demands in the aftermath of the uprisings. Chapter 4 traces the evolution of political regimes in the Arab countries, focusing on the interaction between politics and different facets of economic development. Finally, chapter 5 outlines a potential reform agenda for South-Med countries and identifies a possible role for external actors. CHAPTER 1

Economic Consequences of the Political Transition

1.1 Introduction under significant pressure as a result of declining tourism and, Before the uprisings began at the in some cases, growing trade defi- end of 2010, South-Med countries cits. Meanwhile, political instabil- were hailed for their economic ity triggered significant capital growth, especially between 2003 outflows, causing central banks and 2008. Since then, they have in many countries to drain for- been hit by two crises. Firstly, the eign reserves in order to maintain global financial crisis, which be- fixed exchange rate regimes. Dur- gan in late 2007 and led to a glob- ing those early years, weakening al slowdown in economic activity macroeconomic balances were re- and trade, causing a temporary flected in repeated sovereign rat- dip in their economic growth. ing downgrades and higher sov- Secondly, the eruption of political ereign borrowing costs. and social uprisings, bringing an Simultaneously, South-Med abrupt end to the first signs of re- economies experienced an im- covery from the former crisis. mediate drop in their GDP per Similar to the effects incurred capita growth to 0.4 percent in by political transition elsewhere, 2011, down from around 2 per- the immediate impact on South- cent during the years of the global Med countries was a deterioration financial crisis and from close to in macroeconomic conditions. 4 percent during the period 2003- Already weakened by the global 2008. Unemployment also rose financial crisis, macro-balances sharply in Egypt and Tunisia, to worsened further, especially dur- above 12 percent and to alarming ing the first few years of unrest, levels among the youth (30 and 40 while fiscal pressure intensified percent, respectively). as most South-Med governments Since then, economic activ- succumbed to popular demands ity in the South-Med economies in an attempt to quench the rising has gained momentum. Between unrest. Due to the erosion of fiscal 2012 and 2015, GDP per capita space prior to the onset of the po- growth rates reached 3 percent on litical transition, budget deficits average. However, these growth and public debts increased sig- rates remain below pre-uprising nificantly. Current accounts came growth levels, raising concerns as

7 8 | Economic Consequences of the Political Transition

to whether the political transition tions have been associated with has permanently damaged long- deteriorating macroeconomic bal- term growth potential. ances (higher deficits, weakened Against this backdrop, the external balances), large output present chapter traces the mac- losses in the short-term and slug- roeconomic effects of political gish growth that generally persists transition in the South-Med coun- for a few years after the unrest. tries, benchmarking their results Moreover, according to Elbadawi, against those of other transition Makdisi and Milante (2011) coun- economies. The chapter also as- tries that have attempted to insti- sesses whether the political tran- tute democratization experienced sition has damaged long-term lower than average growth rates growth potential and whether in the first few years. Where de- there is room for stimulating these mocratization has been success- economies. The chapter seeks to ful, growth rates have been higher answer three specific questions: over longer periods than in coun- • What has been the impact tries that failed the transition, and of transition on macroeco- higher than the average growth nomic balances? rates of countries that have made • What has been its impact on no attempt at democratization. actual short-term economic Khandelwal and Roitman (2013) growth? also show that actual output de- • Finally, has political tran- clines below potential for five sition in South-Med coun- years after the transition. tries permanently damaged While the above findings sug- long-term growth potential, gest that political transition leads or not? to output losses in the short-run, The literature covering the the evidence is inconclusive re- impact of political transition on garding its effect on medium-term macroeconomic performance is growth. On the one hand, sub- substantial. It shows that politi- stantial cross-country literature cal instability generates macro- finds no evidence of a significant economic uncertainty, leading to correlation between democratic reduced investment, lower rates institutions and higher income of physical and human capital growth (Barro 1996 and 1997 and accumulation and, eventually, Przeworski et al. 2000). On the reduced productivity growth other hand, Credit Suisse (2011) (Alesina and Perotti, 1996). Po- finds that growth tends to pick up litical instability tends to shorten eight years after a political shock. policymakers’ horizons, leading Focusing on within country ef- to sub-optimal, short-term, mac- fects, several studies find that only roeconomic policies; it may also successful transitions to democ- lead to more frequent shifts in racy lead to a growth dividend policies, creating volatility and (Rodrik and Wacziarg, 2005; Pers- thus a negative impact on mac- son and Tabellini, 2006; and Papa- roeconomic performance (Aisen ioannou and Siourournis, 2008). and Veiga, 2011). With greater nuance, Freund The empirical literature also and Jaud (2014) identify and ex- shows that past political transi- amine 100 successful, failed, and Chapter 1 | 9

gradual transitions that occurred and catching-up with advanced in 158 countries over 1960–2011 economies. Secondly, because de- and provide empirical evidence termining how the growth of ac- that it is regime change (and not tual and potential output behaves successful democratization per following a political shock is cru- se) that boosts long-term growth cial in determining appropriate (by about 1 percentage point).1 In macroeconomic policies. A cycli- other words, political transitions cal economic downturn would to democracy yield no growth call for expansionary macroeco- dividends compared to other nomic policies (if policy space is types of regime change. More- available) to pick up the slack in over, they also show that coun- aggregate demand. In contrast, tries undergoing a rapid regime lower potential output growth change (taking 3 years or less), from a more permanent shock regardless of whether democracy would mean less scope for stimu- is achieved or not, yielded an an- lus macroeconomic policies and nual growth dividend of 0.5 to would require structural policies. 1 percentage point over 7 to 14 The findings of the present years. Alternatively, countries analysis would suggest that the which undergo long transitions experience of South-Med transi- tend to suffer from slower long- tions is consistent with that of oth- term growth. In other words, the er transitions, including those of short-term cost is rapidly offset Eastern Europe in the 80s and 90s, by higher longer-term growth albeit less radical. The first main when the regime change is rapid, finding is that the political up- whether successful, failed or even risings in the South-Med region reversed. have caused a deterioration in In recent years, further litera- economic performance. The direct ture has shown renewed interest impact, in terms of lower growth in the evolution of potential out- and a deterioration in macroeco- put in the aftermath of a finan- nomic balances, has been exten- cial crisis (European commission, sive in Egypt and Tunisia. With 2009; Furceri and Mourougane, respect to other transition coun- 2012; IMF, 2009 and 2015a). Un- tries, both economies have out- derstanding the dynamics of performed in the area of output potential growth in South-Med losses, while underperforming economies, as we seek to do in with respect to macroeconomic this chapter, is important for two balances. In Jordan, Morocco and reasons. Firstly, because a perma- Algeria, the political transition nent decline in potential growth did not have the same disrup- means lower actual growth, low- tive effects on economic growth er job creation in a region where and macroeconomic balances as unemployment has been persis- in Egypt and Tunisia. The most tently high for decades, lower pronounced adverse effects have incomes, and fewer resources been in growing budget deficits for the government to spend on and public debt, as governments education, healthcare, and so- have had to appease their citizens cial services. Overall, it will also through subsidies and govern- mean slower income convergence ment jobs. These adverse effects 10 | Economic Consequences of the Political Transition

have been exacerbated by other impact of the transition on actual exogenous factors. In Jordan, the economic growth, and section conflicts in Syria and Iraq have 4 on potential growth. Section 5 led to a massive influx of refu- draws some conclusions. gees, putting enormous pressure on Jordan’s limited resources, to disruptions in trade routes and a 1.2 Macroeconomic decline in tourism. The drought Balances in Morocco has significantly shrunk agricultural production This section examines the evolu- and therefore economic activity tion of macroeconomic balances since 2012. In Algeria, declining in the aftermath of the uprisings oil prices have diminished the fi- in 2011. The analysis shows that nancial surpluses of the economy. macroeconomic imbalances have The second main finding of deteriorated with acute pressures this chapter is that the decline in on external and fiscal accounts, actual economic growth in South- although macroeconomic stability Med economies is found to reflect has broadly been maintained. a cyclical downturn and not a permanent decline in output. Given Fiscal deficits that output growth in these coun- According to Khandelwal and tries remains below potential, Roitman (2013), political transi- this means that there is scope in tion tends to be accompanied South-Med economies for macro- by subdued fiscal revenue and economic stimulus (mainly fiscal increased public expenditure, policy) to raise aggregate demand giving rise to a deterioration in and to boost short-term growth. overall fiscal balances by about 1 Third, despite the importance percent of GDP during the event of working to revive growth po- year, and by a cumulative 1.75 tential, South-Med countries percent of GDP for the subse- should not necessarily aim at re- quent two years. Fiscal deficits storing growth to pre-crisis lev- can be expected to return to pre- els (2003-2008), as growth during crisis levels only four years after this period was accompanied by the event. So far, this predication symptoms of economic over- does not seem to hold in the case heating in some cases. Aiming at of the South-Med countries. There higher growth rates to progress has certainly been a deterioration towards economic convergence in macro-balances in these coun- and greater levels of job creation tries, but the magnitude of the de- would require the adoption of terioration has been more signifi- structural reforms to raise long- cant than elsewhere. term growth. In 2009-2010, most South-Med The chapter is, hereafter, economies had weak fiscal posi- structured as follows: the next tions due to the fiscal stimulus section takes stock of the evolu- packages adopted in the aftermath tion of macroeconomic balances, of the global financial crisis. Because both external and internal, in the of increased public spending, aftermath of the political shock; fiscal deficits had increased to section 3 reviews the short-term around 5 percent of GDP in 2009- Chapter 1 | 11

2010, up from around 1 percent 2.5-3 percentage point of GDP in in the period 2003-2008. After the an attempt to preemptively pre- uprisings, fiscal deficits and debt vent potential unrest. Most of the to GDP ratios shot up, particu- increases in expenditure took the larly in Egypt and Tunisia, and form of higher subsidies for food to a lesser extent in Morocco (fig- and energy, higher public-sector ure 1.1). During the same period, wages and pensions, and expand- the fiscal deficit in Jordan went ing unemployment benefits (for down, but net domestic debt in- more details, see chapters 2 and creased. Algeria experienced the 3). Egypt is the only South-Med most dramatic increase in fiscal country that did not increase deficit, but the debt to GDP ratio spending (as a share to GDP) remains very modest (and even until 2013. As will be discussed negative). later, the subsequent increase was The above developments are partially in the form of public in- the product of what happened vestment to stimulate aggregate to government expenditures and demand and economic growth. revenues in the wake of the up- On the revenue side, during risings. On the expenditure side, the first two years of the transi- most South-Med governments tion, fiscal revenues declined in were compelled to increase pub- Egypt relative to 2010, because lic expenditure in an attempt to of a drop in both tax and non-tax appease popular demands in tur- revenues, but picked up in 2014 bulent social and economic envi- as a result of exceptional grants ronments (figure 1.2). The highest (5 percent of GDP) (figure 1.3). increases in public spending oc- To a lesser extent, Jordan experi- curred in Tunisia, by more than enced similar trends. In Morocco 4 percentage points of GDP be- and Tunisia, revenues increased tween 2010 and 2011. In countries during the early years of the tran- that did not witness an uprising, sition but declined afterwards. spending was also increased by In Morocco, this occurred on the

Figure 1.1: Net domestic debt and overall fiscal balance (percent of GDP)

Source: IMF Article IV Consultation data. 12 | Economic Consequences of the Political Transition

Figure 1.2: Government expenditure, 2003-2015 between 3 (Morocco) to 7 (Tuni- (percent of GDP) sia) percentage points of GDP. Fiscal balances improved by 2015 in Jordan and Morocco, on the back of low oil prices and fiscal consolidation efforts. In Egypt and Tunisia, fiscal deficits remain far from their pre-crisis levels by 3 and 5 percentage points of GDP, respectively. The deterioration in fiscal deficits in all South-Med countries surpassed predictions Source: IMF Article IV Consultation data. according to the literature. Fiscal outcomes were worse for back of lower tax receipts, while countries that suffered from high in Tunisia it was a result of lower initial deficits prior to the uprising non-tax revenue. (figure 1.4). Egypt’s fiscal deficit, These developments exerted which was already quite sizeable further strain on the fiscal defi- at 8 percent of GDP in 2010, de- cits. Relative to 2010, all South- teriorated to above 10 percent of Med countries (save Algeria) GDP in 2011, and peaked at 13 witnessed a deterioration in their percent of GDP in 2013. Although budget deficits in the range of 0.2 the deficit has been reduced since (Jordan) to 2.7 (Tunisia) percent- 2014, reflecting a one-off consid- age point of GDP, in 2011. This erable increase in grants in 2014 trend continued until 2013 in followed by fiscal consolidation most South-Med countries (and in 2015 (resulting from lower sub- 2012 for Morocco). The cumula- sidies), it still amounts to around tive deterioration in the deficits 12 percent of GDP. between 2010 and 2013 ranged By contrast, in countries where

Figure 1.3: Revenue components, 2003-2015 (percent of GDP)

Source: IMF Article IV Consultation data. Chapter 1 | 13

initial fiscal conditions were rela- Figure 1.4: Fiscal deficits, central governments, tively stronger, deficits remain 2003-2015 (percent of GDP) below 10 percent of GDP. Tuni- sia’s low fiscal deficit experienced a six-fold deterioration, reaching 6 percent of GDP, although this remains half the level of the deficit in Egypt. Fiscal deficits in Jor- dan and Morocco, which deteriorated to 11 and 5 percent of GDP after the uprising, have improved since 2013 and are currently below 5 percent of GDP. Source: IMF Article IV Consultation data. Algeria’s public finances are overly dependent on hydro- in the aftermath of the global fi- carbon revenues. Revenues in- nancial crisis, and soared even creased in 2011-2012 on the back higher, to more than 82 percent of high oil prices, but have been in 2011 (figure 1.5). It is claimed declining since 2013. Like other that the near complete halt of gas South-Med economies, public flows from Egypt required im- spending also surged in Algeria ports of more expensive fuel for in 2011 and 2012, amid regional electricity generation, which has upheavals. As a result, Algeria’s also added to the debt stock of fiscal balance jumped to above 4 the national electricity company percent of GDP in 2012, up from (IMF, 2015b). However, the debt less than 2 percent of GDP in 2010, to GDP ratio stabilized at 80 per- despite solid oil revenues during cent of GDP since 2013. In Tuni- that year. Although there was a sia and Morocco, debt levels were significant improvement in 2013 initially lower, hovering around (thanks to fiscal consolidation), 40 percent of GDP, but increased the deficit soared to 15 percent of sharply to 64 percent of GDP and GDP in 2015 on the back of lower 55 percent of GDP, respectively, fiscal hydrocarbon revenues. in 2015. In Algeria, thanks to oil Overall, the deterioration in wealth, the level of debt remains fiscal balances and the decline in GDP during the political transition led to the build-up of gov- Figure 1.5: Net domestic debt, central govern- ernment domestic debt, which ments, 2003-2015 (% of GDP) has risen considerably for all South-Med governments, except for Algeria (figure 1.5). The greatest concern lies with countries that entered the transition with a high debt, such as Egypt, where the debt to GDP ratio was around 55 percent of GDP in 2010. In Jor- dan, the debt to GDP ratio was also high in 2010 (71 percent), as a result of the fiscal stimulus Source: IMF Article IV Consultation data. 14 | Economic Consequences of the Political Transition

in single digits as a share of GDP. income as world interest rates plummeted and lower remittances - surpluses have disappeared External balances and deficits have worsened (fig- According to Khandelwal and ure 1.6). Roitman (2013), most transition During the first few years after countries witnessed a deteriora- the uprising, current accounts in tion in their current accounts dur- South-Med economies (except ing the year of political instability, Algeria) have come under pres- with the external current account sure, as a result of declining tour- balances improving in 60 percent ism receipts and in some cases of the countries only five years af- growing trade deficits. Just like ter the event. fiscal deficits, the deterioration in Before both the global finan- current accounts was more pro- cial crisis and the political transi- nounced during the early years of tion, most South-Med countries the unrest. Seven years after the were already running current ac- uprising, current accounts have count deficits between 2003 and stabilized only in Morocco, where 2008 (figure 1.6). The pre-crisis they have returned to pre-crisis deficits were highest in Jordan levels. This suggests that the per- (7 percent of GDP) on the back formance of South-Med external of buoyant energy imports (with balances is worse than endured Jordan importing 96 percent of its by other transition countries. energy needs) but significantly In both Egypt and Tunisia, lower (2 percent of GDP) in Mo- the current account deteriora- rocco and Tunisia. Egypt was tion is primarily the result of a actually enjoying a small surplus drop in tourism revenues follow- of around 2 percent of GDP. In ing ongoing security threats (see the aftermath of the financial cri- box on tourism). Additionally, in sis and the ensuing global slow- the case of Egypt, the fall in Suez down in economic and trade Canal receipts (to less than 2 per- activity - which meant lower ex- cent of GDP in 2015, from above 3 port proceeds, lower investment percent of GDP before the financial crisis), due to the depressed global economic environment Figure 1.6: Current accounts, 2003-2015 (percent of and trade, has also contributed GDP) to the current account dynamics. But certain factors have partially offset some of the pressure on the current account in Egypt, the most significant being the bounce-back in remittances to 6 percent of GDP (which did not occur in Tunisia), proving to be more resilient than other types of inflows during the whole period of the political transition (figure 1.7). In addition, the trade balance has been narrowing Source: IMF Article IV Consultation data. on the back of subdued import Chapter 1 | 15

growth due to sluggish economic Figure 1.7: Remittances, 2000-2014 (percent of activity, and also lower export GDP) revenue as a result of subdued global demand. Finally, in 2014, Egypt received substantial official grants from the GCC (around 4 percent of GDP), a historical high since 1993, which contributed to easing the pressure off external balances for that single year. However, in the case of Tunisia, a growing trade balance (from buoyant energy and capital goods imports and declining oil and Source: World Development Indicators, World Bank. phosphate exports) continues to act as a drag on the current account. As a result of these devel- improved since 2013 because of opments, Egypt’s small deficit (2 lower oil prices, lower food im- percent of GDP in 2010) worsened ports thanks to a good agricultural gradually to 4 percent of GDP by season and high domestic food 2015. However, Tunisia’s large production, in addition to rising pre-crisis current account deficit exports in newly developed sec- (5 percent of GDP) deteriorated tors (IMF, 2015c). to an alarming 9 percent of GDP Finally, Algeria’s current ac- in 2015, its highest level since the count surpluses, which persisted 1980s. Consequently, external im- for over 15 years, started to disap- balances remain higher in Tunisia pear. Lower hydrocarbon exports than in Egypt. (following lower oil prices) and Jordan’s pre-crisis current ac- growing imports (mainly invest- count significantly deteriorated ment goods) led to the emergence to 10 percent of GDP in 2011 and of a current account deficit of 4 then to 15 percent in 2012, follow- percent of GDP in 2015. ing more expensive energy im- In terms of Capital Flows, ports (for electricity generation) South-Med economies, particu- and declining remittances. Natu- larly Egypt and Jordan, attracted rally, Jordan’s external finances strong capital inflows during the have stabilized with lower global period 2003 and 2008, benefit- oil prices. Morocco’s current ac- ing mainly from oil petrodollars. count deficit, which has been in Having already taken a hit dur- continuous deterioration since ing the financial crisis, FDI in- 2002, doubled from 4 to 8 percent flows have declined further over of GDP between 2010 and 2011 the past seven years, as a result of and widened even further in 2012 political uncertainty, which has mainly as a result of higher fuel undermined investor confidence. and food imports (the latter be- FDI inflows, which are usually cause of the drought), and lower considered as longer term, less tourism receipts and remittances reversible and largely non-debt from Europe. However, Moroc- creating flows, came down from 6 co’s current account position has percent of GDP in 2008 (and even 16 | Economic Consequences of the Political Transition

Box: Tourism in South-Med Countries after the uprising

One of the immediate effects of the po- Figure 1.8: Tourist arrivals, 2010-2014 litical uprising and related violent clashes was a sharp decline in tourist arrivals and receipts (figures 1.8 and 1.9). The number of tourist arrivals fell in all countries in 2011, but the decline was most severe in Egypt and Tunisia where arrivals dropped by 30 percent in 2011 and have not yet recovered. Tourism is an important source of foreign exchange earnings in South-Med economies where it ranged from 6 percent of GDP in Egypt and Tunisia and Source: World Tourism Authority data. 12 percent of GDP in Jordan. Tourism receipts were generally resilient during Figure 1.9: Tourism receipts (percent of GDP) the financial crisis, with only Egypt experiencing a decline to 5 percent of GDP. After the political uprising, tourism receipts dropped in both Egypt and Tuni- sia to less than 2 percent of GDP in 2015 from above 6 percent in 2008), following security incidents and terrorist attacks. Given the contribution of this sector to value-added and foreign exchange earnings, its decline has had a significant effect on external balances and on economic growth. Source: Data from national central banks.

higher ratios of 6 and 13 percent 2 percent of GDP in 2011 (figure for Egypt and Jordan, respective- 1.10). Seven years after the up- ly) to 3 percent of GDP after the risings, they remain weak in all financial crisis and to less than South-Med countries. Short-term portfolio investment outflows have also seen a sharp decrease, Figure 1.10: FDI inflows, 2000-2014 (percent of especially those in the form of T- GDP) bills in Egypt. Finally, looking at external debt between 2008 and 2015, the ratios for most countries relative to GDP remained relatively modest. The highest rations were seen in Jordan (83 percent in 2003) and Tunisia (51 percent in 2003). Dur- ing the post-transition period, external debt essentially went down Source: World Development Indicators, World Bank. in all countries. Average gross Chapter 1 | 17

Figure 1.11: External debt Total external debt (% of GDP) Short and long term external debt (as % of total external debt)

Source: World Development Indicators, World Bank. external debt for Morocco, Jordan 1.3 Economic Growth: and Tunisia increased from 45 to Actual 55 percent of GDP (figure 1.11). The strongest increases were in This section examines the evo- 2012 and 2013, when the debt to lution of actual growth in the GDP ratio cumulatively rose by South-Med countries between almost 7 percentage points. More- 2003 and 2015. This period covers over, short-term debt has been different business cycles: a boom rising in Tunisia since 2012, and period that lasted between 2003 in Morocco since 2013, though it and 2008, followed by a mild eco- remains at benign levels of 18 and nomic slowdown after the global 26 percent of GDP respectively. financial crisis (2009-10), ending In Egypt, short-term debt re- with a steeper downturn associ- mains below 10 percent of GDP. ated with the political transition In Jordan, while short-term debt (2011-2014). The performance in has been declining since 2011, it the post-2011 period has partly accounts for almost half of the been the product of prevailing so- total external debt. Standard debt cial and political tensions, but this indicators paint a good picture of period has also coincided with the external vulnerability of the adverse global economic condi- South-Med economies and gener- tions and regional geopolitical ally do not suggest any debt-ser- conflicts. vicing problem in the near future. Prior to the uprisings, the South- Liquidity indicators, such as debt Med transition economies fared service-to-exports ratios, have re- relatively well in terms of growth mained relatively benign over the performance. Egypt and Tunisia last few years (below 15 percent were praised by international fi- in all 5 economies) and raise no nancial institutions for their lead- immediate concerns. ing economic performance in the 18 | Economic Consequences of the Political Transition

Arab region. The 2008 financial after the uprising, in no South- crisis was associated with a short- Med economy (but Algeria) has lived growth dip from close to 5.5 per capita growth recovered to percent between 2003 and 2008 to its pre-crisis level of 2008, amid less than 4 percent, but by 2010 claims that the political crisis may they were on the road to econom- have damaged long-term/poten- ic recovery. In terms of per capita tial growth. growth, the performance was The growth depression has af- more modest, due to high popu- fected all countries, but its mag- lation growth rates (table 1.1). nitude varies from country to In the post-uprisings period, country. Egypt and Tunisia, at the average real GDP growth in the centre of the uprisings (toppling South-Med economies fell from the rulers), have been the hardest 3.7 percent in 2010 to less than 2.2 hit. Both Egypt and Tunisia have percent in 2011, its lowest level in experienced a decline in GDP per over a decade. Per capita growth capita levels and growth rates. rate also dropped to 0.4 percent in During 2011, GDP per capita 2011 but picked-up as from 2012 contracted by 0.4 and 3 percent to an average of 1 percent between respectively and output growth 2012 and 2014. Notwithstanding losses were in the order of 3 and some improvement, seven years 5.5 percentage points. Since 2011, output growth has recovered more quickly in Tunisia (to close Table 1.1: Real GDP Growth in the South-Med to 2 percent) than in Egypt, where Transition Economies, (percent) the growth-population dynamics GDP growth continue to cause contractions in 2003- 2003- 2009- 2012- per capita GDP. Six years later, 2014 2008 2010 2011 2014 both countries are still facing in- Algeria 3.6 4.1 2.6 2.8 3.3 creasing security threats and so- Egypt 4.2 5.5 4.9 1.8 2.2 cial tension, preventing a sustain- Jordan 5.3 7.4 3.9 2.6 2.9 able growth upturn. Meanwhile, growth has been Morocco 4.5 5.0 4.2 5.2 3.4 more resilient in other South-Med Tunisia 3.7 5.3 2.9 -1.9 2.8 economies, remaining positive, Average 4.2 5.5 3.7 2.1 2.9 albeit at a lower level, in the af- Egypt and Tunisia 4.0 5.4 3.9 -0.1 2.5 termath of the uprisings. More importantly, the growth perfor- GDP per capita growth mance may not be solely reflecting Algeria 1.7 2.5 0.6 0.8 1.1 the same disruptive effects related Egypt 2.0 3.3 2.6 -0.4 -0.4 to the uprising as in Egypt and Tu- nisia. In Jordan, GDP growth rates Jordan 2.9 5.0 1.6 0.4 0.6 fell in 2011 (including in per capi- Morocco 3.4 4.0 2.9 4.1 2.3 ta terms) following disruptions of Tunisia 2.7 4.1 2.2 -3.1 2.2 natural gas supplies from Egypt and adverse regional spillovers, Average 2.5 3.8 2.0 0.4 1.2 particularly from the conflict in Egypt and Tunisia 2.3 3.7 2.4 -1.7 0.9 Syria and Iraq with repercussions Source: World Economic Outlook Database, IMF and Tunisia per on security. The Jordanian au- capita GDP from World Development Indicators, World Bank. thorities estimate that they have Chapter 1 | 19

received more than one million The decline is estimated to be in refugees, or about 20 percent the range of 10 and 40 percent, re- of Jordan’s population, putting spectively (Khandelwal and Roit- pressure on economic infrastruc- man, 2013). Moreover, as growth tures and labour markets (IMF, rates tend to remain modest in 2013). Moreover, the emergence subsequent years, both public of the Islamic State in Iraq and the and private investment take at Levant (ISIS) in Iraq, which is Jor- least 5 years to return to pre-crisis dan’s largest trading partner and levels. the destination of some 20 per- A similar pattern occurred in cent of its exports, have caused South-Med countries. In 2011, disruptions to exports as well as investment declined sharply in to tourism in Jordan. Egypt and Tunisia by 2 and 3 Morocco is the only economy percentage points, respectively, in the region where GDP per cap- standing at 17 and 24 percent of ita growth strengthened in 2011 GDP (figure 1.12). By 2014, these but then declined afterwards, figures had deteriorated to 14 and mostly as a result of the drought 22 percent of GDP in Egypt and that led to a poor harvest and a Tunisia, also respectively. The sit- decline in agriculture production. uation was worse in Egypt as the As for Algeria, growth remained investment to GDP ratio had fall- resilient after the regional un- en in 2009-10, following a sharp rest and continued to grow until decline in private investment in 2014. Indeed, between 2003 and the aftermath of the financial cri- 2014 the economy grew on aver- sis. By 2011, low private invest- age at around 3.5 percent. This is ment had not recovered to pre- because Algeria did not witness crisis levels. the same kind of political unrest The disaggregation of in- as other countries and in part, be- vestment data in Egypt further cause GDP growth is primarily reveals that public investment driven by the hydrocarbon sector. declined by close to 20 percent Benchmarking the performance of South-Med countries, Khandelwal and Roitman (2013) Figure 1.12: Gross capital formation, 2003-2014 estimate that declines in real GDP (percent to GDP) ranged between 1 and 7 percent during the first year of political instability. South-Med countries have not only escaped a recession (including Egypt and Tunisia), but also their GDP growth dip induced by the political transition was relatively lower than that seen in other transitions. Turning to investment, it is known that the first year of political instability tends to be associated with a large decline in public and private investment. Source: World Development Indicators, World Bank. 20 | Economic Consequences of the Political Transition

in 2011, down to 6 percent of Savings have been particular- GDP compared with a pre-crisis ly low in South-Med economies. level of 8 percent of GDP (figure Except for Algeria, savings ac- 1.13). Private investment growth counted for around 15 percent of remained positive but also de- GDP, with ratios above 20 percent clined to less than 10 percent in in Morocco and Tunisia. Savings 2011, down from 11.5 percent have declined significantly since during pre-crisis years. Prior to the political uprising with a ratio the uprising, private investment to GDP falling as low as 8 percent had declined (by more than 30 in Egypt (figure 1.14). percent) in 2009 following the At the sectoral level, manufac- global financial crisis, a reaction turing and tourism were among that was much more severe than the most impacted sectors in that which is currently witnessed Egypt and Tunisia (figure 1.15). during the political transition. Non-oil manufacturing growth Because investor confidence has in the two countries declined not been restored, the current by around 6 and 10 percentage share of private investment to points in 2011, respectively. Non- GDP of 8 percent is way below oil manufacturing has grown by the 10 percent of the 2003-2008 less than 3 percent in Egypt since pre-crisis period. 2011 (compared to 5 percent during 2003-2010), largely due to the Figure 1.13: GDP by expenditure in Egypt (aver- severe shortage of foreign cur- age annual growth rates) rency to pay for raw materials and equipment, as will be seen in chapter 2. Non-manufacturing growth has virtually stagnated in Tunisia (compared to 3 percent during 2003-2010). Growth in tourism dropped by close to 18 and 25 percentages in both countries in 2011, and the sector has continued to contract due to sporadic security incidents in both countries. Source: Calculated by the authors based on Egypt’s national accounts, Ministry of Planning. In short, the uprisings caused a slowdown in economic activity, but other factors may have con- Figure 1.14: Gross domestic savings (percent of tributed to the prolonged growth GDP) slump. The economic slowdown in the US, Europe and China meant subdued external demand for South-Med exports. The escalation in geopolitical violence and civil wars (Syria and Libya) have had negative spillovers on neighboring economies that have further disrupted trade and un- Source: World Development Indicators, World Bank. dermined consumer and investor Chapter 1 | 21

confidence in countries like Jor- Figure 1.15: Sectoral performance in Egypt and dan and, to a lesser extent, Tuni- Tunisia after the uprising sia. The recent plunge in oil prices may have reduced the pressure on external finances and aided recovery for oil importers, but Mo- haddes and Raissi (2015) show that such gains have been offset by a decline in external demand and financing by oil exporters (through lower trade, remittances, and investment).

1.4 Economic Growth: Potential Source: Calculated by the authors based on data from national authorities. Given the modest growth performance in the wake of the uprisings, one question is whether slack. Output below potential South-Med economies will be able (a negative output gap) implies to attain the pre-uprising growth that there is underemployment rate of 5.5 percent achieved be- (excess supply) of capital and la- tween 2003 and 2008 (even higher bour. This section describes what for Jordan and Egypt). A similar happened to both potential out- question can be raised about the put and the output gap over the GDP per capita growth of 4 per- period 2003-2015. It does not look cent during the 2003-2008 period. at the medium-term trajectory of This section describes what hap- potential output. pened to both potential output Understanding the dynamics and the output gap over the pe- of potential growth in South-Med riod 2003-2015. The question is economies is important for two whether the political uprisings reasons. The first is that a perma- have caused permanent damage nent decline in potential growth to growth potential and whether means lower actual growth, lower the decline will persist once the job creation in a region where un- cyclical recovery is complete. employment has been persistently Potential output is defined high for decades, lower incomes, as the level of output consis- and fewer resources for the gov- tent with stable inflation (Okun, ernment to spend on education, 1962). In the short term, actual healthcare, and social services. output can deviate temporarily Overall, it will also mean slower from potential output as shocks income convergence and catching- hit the economy, reflecting the up with advanced economies. slow adjustment in wages and Second, the diagnosis of how the prices to shocks (IMF, 2015a). The growth of actual and potential short-term divergence of actual output behaves, following the from potential output is referred political shock, is crucial in es- to as the output gap, or economic tablishing appropriate macroeco- 22 | Economic Consequences of the Political Transition

nomic policies. A cyclical growth tries (EMDC) over the next five slowdown would call for expan- years. sionary macroeconomic policies Complementing this work, (if policy space is available) to this section decomposes actual pick up the slack in aggregate de- growth into cyclical and structur- mand. In contrast, lower growth al components to assess whether in potential output would tend to the current decline in the rate of imply a more permanent shock output growth reflects a cyclical with long-lasting losses, which or structural downturn, and more may prevent potential growth broadly to understand the dy- from returning to pre-crisis rates namics of potential growth. The in the long run. In this case, there methodology used is explained in would be less scope for stimulat- Annex 1.1 at the end of this chap- ing the economy through macro- ter. economic policies and structural The results can be summarized policies would be required. as follows. Actual growth is below The literature addressing the potential in Egypt, Jordan and impact of (financial) crisis on po- Tunisia, suggesting that they are tential growth has thrived in the operating well below potential. context of advanced economies. However, the growth rate dip be- Furceri and Mourougane (2012) low trend does not coincide with find that the crisis lowered po- the start of the political transition tential output by around 1.5–2.4 in all 3 countries (figure 1.16). In percent on average in OECD Jordan and Tunisia, it was the fi- countries. Barrera and others nancial crisis rather than the po- (2009) find that potential output litical uprising that brought actual in the United States is expected to growth below its trend. In Egypt, be about 5.75 percent below the the output gap turned negative in counterfactual level produced by 2011 and has been widening over average growth rates before the time. By 2014, output growth re- global financial crisis. In the case mained 90 percent below poten- of the MENA region, Khandelwal tial. Jordan’s output gap has been and Roitman (2013) estimate that negative since 2010 and has been actual growth rates dipped below widening ever since, with an ac- trend during the year of the event tual of around 40 percent below and in subsequent years. As a re- potential. Just like Egypt, Tuni- sult, output gaps turned negative sia’s actual output took a hard hit the year of the event, and contin- in 2011, thus widening Tunisia’ ued to widen for some time. On negative output gap. However, average, output remained below the output gap has significantly potential for 4 years after the ini- narrowed over time and output tial drop, taking around 5 years is around 40 percent below trend. for the recovery. Mitra and others This suggests that Egypt until (2015) estimate that MENA’s (ex- the political transition, as well as cluding the GCC) per capita GDP Jordan and Tunisia until the glob- will drop from one-half to one- al financial crisis, experienced ex- third of the average of emerging ceptionally high growth rates and markets and developing coun- grew above potential growth, as Chapter 1 | 23

shown by their positive output Figure 1.16a: Egypt actual and potential growth gap. This further suggests that while there is a case for reigniting growth in the short-term, pre- crisis growth (2003-2008) is not an appropriate target, as this level may be accompanied by symptoms of economic overheating such as high inflation, excessive liquidity and credit growth and Figure 1.16b: Tunisia actual and potential growth investment. The short-term target should be to reignite short-term growth to trend rather than to pre-crisis levels. In figure 1.17, per capita growth rates are decomposed into cyclical and trend components for the 5 South-Med coun- Figure 1.16c: Jordan actual and potential growth tries over the period 1991-2014. The figure clearly shows a significant cyclical downturn in per capita growth in 2011, the year of the political uprisings. Being cyclical in nature, this drop is a temporary phenomenon that should not last Source: Calculated by the authors based on World Economic Out- or prevent growth from recover- look Database, IMF. ing again. And as it turned out, the cyclical component of per capita income growth all but sub- Figure 1.17: Cyclical and trend components of per sided in 2014. The other period capita growth in South-Med economies, 1991-2014 that experienced a cyclical change in per capita growth is 2003-2008, reflecting perhaps a boom that was not sustainable. In terms of potential growth, figure 1.18 shows the trend for all 5 South-Med countries over the period 2002-2014. The most notable observation is that there was a mild decline in potential growth in all South-Med economies, except Morocco. The decline in potential per capita growth in Source: Calculated by the authors based on World Economic Egypt and Tunisia was more Outlook Database, IMF. pronounced and started as far back as the mid-2000s. In Egypt, 1.9 percent in 2014, and in Tunisia potential growth declined from it declined from 3 to 2.8 percent about 2.3 percent in 2003 to about during the same period. Owing 24 | Economic Consequences of the Political Transition

Figure 1.18: Potential GDP per capita growth in and Tunisia. However, even if South-Med economies South-Med economies recover to their potential, past performance has not been high enough to enable them to make progress on the convergence with the living standards prevailing in advanced economies, or to create sufficient jobs for new entrants into the labour market. To do so, it would be necessary to carry out deeper and more extensive institutional reforms that would require time to implement and to reap benefits. Source: Calculated by the authors based on World Economic Out- Reigniting short-term growth look Database, IMF. would require the most immediate constraints and opportunities facing these economies to be to a large population drag on directly targeted. Longer term growth, Egypt experienced larger growth would take much more drops in potential growth (by 1/3 substantial reforms. percentage point over the period 2003-2014) than Tunisia (a drop only by 0.1 percentage point). The potential GDP per capita 1.5 Conclusions growth rates in Algeria and Jor- After having barely recovered dan grew at 1.5 and 2.4 percent, from the effects of the global fi- respectively, in the pre-crisis nancial crisis, the political upris- period (2003-2008). These rates ing took the South-Med countries remained fairly constant during by surprise at the end of 2010. the global financial crisis in 2009- The political transition has been 2010. In Jordan, the political tran- accompanied by deteriorating sition has been associated with macroeconomic conditions, as has a decline in potential growth in happened in political transitions the same order of magnitude as elsewhere. The years following Egypt (0.1 percentage point). In the uprisings have been accom- Morocco, there has been an up- panied by a global slowdown and ward trend since the 1980s, with the emergence of geopolitical ten- potential per capita growth rising sion and conflict. from about 2.8 percent in 2003 to Pressing social demands, de- 3.2 percent in 2014. It is the only clining tourism receipts, capital South-Med country to have not outflows and sluggish economic experienced any decline in poten- growth have imposed mounting 2 tial growth since 2003. pressures on fiscal and external Based on these results, one balances. Seven years after the up- would expect that growth in risings, growth remains subdued South-Med countries would start and macroeconomic imbalances picking up as political stability remain persistent, for some coun- is restored, particularly in Egypt tries more than others. Chapter 1 | 25

The review, in this chapter, been worse than in other transi- of the economic consequences of tions. In Egypt, fiscal imbalances the uprisings on the economies of are worrying, as are external im- Algeria, Egypt, Jordan, Morocco balances in Tunisia. In all coun- and Tunisia relative to the find- tries, the persistence of fiscal im- ings of the literature can be sum- balances has led to the build-up marized as follows: of government debt, which is a First, in line with what hap- source of vulnerability in Jordan pened in transitions elsewhere, and Egypt. These developments the first few years (2011 and have markedly heightened con- 2012) were the worst in terms of cerns about the sustainability of economic performance with the public finances. External imbal- sharpest increases in fiscal defi- ances are quite apparent in Tu- cits, current account deficits and nisia with the persistent decline growth and/or output declines. in tourism receipts. So far, these Some countries have been able imbalances have not threatened to reverse these trends, starting macroeconomic stability in the 2013, or at least arrest further de- sense that South-Med economies terioration. The magnitude of the have not encountered any bal- economic impact of the political ance of payments or external fi- transition has varied from one nancing crises. However, should country to another. Egypt and macroeconomic imbalances per- Tunisia have been more directly sist, these countries could become exposed to the disruptive effects more vulnerable. of the political unrest than Jordan Third, the severity of the eco- and Morocco. nomic situation in a country de- Second, if one were to bench- pends on initial conditions. For mark the performance of the example, pre-existing large fis- South-Med economies against the cal imbalances in Egypt have in- literature on the economics of po- creased its fiscal vulnerability to litical transition, the conclusion is further pressure, no matter how that their performance is mixed. slight. In contrast, because Tuni- The output loss (in per capita sia’s fiscal deficits were initially terms) in the aftermath of the up- very low before the uprising risings has been smaller than that (only 1 percent of GDP), a six-fold ensured by other transition coun- deterioration was not disastrous. tries. Moreover, the recovery time The contrast between Egypt and (from negative growth) has also Tunisia with respect to external been shorter. This is more appar- balances is also telling, where ent in the case of Tunisia, which Tunisia’s external balances were experienced negative growth eroded in the aftermath of the fi- of 3 percent in 2011 but positive nancial crisis while Egypt enjoyed growth afterwards. Notwith- some financial surpluses. standing some recovery, growth Fourth, most South-Med econ- has remained below potential and omies are currently operating below pre-crisis levels. below capacity given that their In terms of macroeconomic actual growth is below potential. aggregates, the deterioration in This suggests that there is scope some South-Med economies has for active macroeconomic policy 26 | Economic Consequences of the Political Transition

(mainly fiscal policy) to raise ag- tially but do not result in sustained gregate demand and, in turn, democracy, defined by a high polity score. Gradual ones achieve democ- short-term growth. This is con- racy, but the increase in polity score sistent with the finding that the occurs over a longer period, more weakness in economic activity is than 3 years and up to 15 years. primarily due to cyclical factors; 2. The sensitivity of the results is tested these are temporary effects that through several robustness checks, by applying different smoothing should not prevent a recovery of parameters (the baseline of 6.25 is growth. However, achieving suf- varied to 100 and 200). The main ficient growth to make progress findings do not change significantly on convergence with countries in the robustness tests performed that have higher standards of liv- (table A1.3 in Annex 1.1). ing would require an enhance- ment of growth potential through major structural and institutional References reforms. Fifth, and finally, the political Alesina, A. and R. Perotti (1996) transition is not the sole culprit “Income distribution, political behind weakened economic fun- instability and investment”, damentals in South-Med econo- European Economic Review, vol. mies. The latter have had to face 40, issue 6, pages 1203-1228, two challenging back-to-back April. crises (the global financial crisis Aisen, A. and FJ. Veiga, (2011) followed by the political transi- “How Does Political Instability tion). They have also had to deal Affects Economic Growth?”, with other exogenous shocks, in- IMF Working Paper No. 11/12, cluding the rise (decline) in com- January. modity prices in 2008 and, in 2014 Barrera, N., O. Celasun, M. Es- and beyond, a global economic tevao, G. Keim, A. Maechler, slowdown in the US, Europe and P. Mills, and A. Bhatia (2009) China. Also, many South-Med “United States: Selected Is- countries are very close to con- sues”, International Monetary flict zones and have had to bear Fund. negative spillover effects from re- Barro, RJ. (1997) “Determinants gional conflicts. Nevertheless, the of Economic Growth: A Cross- political uprisings have had their Country Empirical study”, toll on economic performance, NBER Working Paper No. 5698, most notably in Egypt and Tuni- August. sia. Barro, RJ. (1996) “Democracy and Growth”, NBER Working Paper No. 4909, October. Notes Credit Suisse (2011) “From Spring to Revival – Regime change 1. The authors identify democratic and economic transformation” transitions as rapid— occurring Elbadawi, I., S. Makdisi, and G. within 3 years—and large increases Milante (2011) “Explaining in the score on Polity scale. Failed transitions are rapid transitions to the Arab Democracy Deficit: democracy that succeed in chang- the Role of Oil and Conflicts” ing the regime temporarily or par- in “Democracy in the Arab Chapter 1 | 27

World: Explaining the Deficit” Under the Stand-By Arrange- edited by Elbadawi, I. and S. ment, Request for Waivers Makdisi, Routledge and Interna- of Nonobservance of Perfor- tional Development Centre. mance Criteria, Modification European commission (2009) of Performance Criteria, and “Impact of the current eco- Rephasing of Access – Staff nomic and financial crisis on Report; Press Release on the potential output”, Occasional Executive Board Discussion; Papers No.49, June. and Statement by the Execu- Freund, C. and M. Jaud (2014) tive Director for Jordan” “Regime Change, Democracy _____ (2009) “World Economic and Growth”, Peterson Insti- Outlook October 2009: Sus- tute for International Econom- taining the Recovery” ics Working Paper No. 14 – 1, Khandelwal, P. and A. Roitman April. (2013) “The Economics of Po- Furceri, D. and A. Mourougane litical Transitions: Implica- (2012) “The effect of financial tions for the Arab Spring”, crises on potential output: IMF Working Paper No. 13/69, New empirical evidence from March. OECD countries”, Journal of Mitra, P., A. Hosny, G. Abajyan Macroeconomics, vol. 34, issue and M. Fischer (2015) “Esti- 3, September, pages 822-832. mating Potential Growth in Hodrick, R. and EC. Prescott the Middle East and Central (1997) “Postwar US Business Asia”, IMF Working Paper No. Cycles: An Empirical Inves- 15/62, March. tigation”, Journal of Money, Mohaddes, K. and M. Raissi Credit and Banking, vol. 29, no. (2015) “The U.S. Oil Supply 1, February, pages 1-16. Revolution and the Global International Monetary Fund Economy”, IMF Working Paper (IMF) (2015a) “Chapter 3: No.15/259, December. Where are we headed: Per- Okun, AM. (1962) ‘‘Potential spectives on Potential Output” GNP: Its Measurement and in “World Economic Outlook: Significance” in: “Proceedings Uneven Growth – Short and of the Business and Economic Long Term Factors” Statistics Section of the Ameri- _____ (2015b) “Jordan: Seventh can Statistical Association”, and Final Review Under the American Statistical Association, Stand By-Arrangement and pages 89-104. Proposal for Post-Program Papaioannou, E. and G. Siourour- Monitoring – Press Release; nis (2008) “Democratization Staff Report; and Statement and Growth”, The Economic by the Executive Director for Journal, vol. 118, issue 532, Oc- Jordan” tober, pages 1520-1551. _____ (2015c) “Morocco: 2015 Ar- Persson, T. and G. Tabellini (2006) ticle IV Consultation – Press “Democracy and develop- Release; Staff Report; and ment: the devil in details”, Statement by the Executive NBER Working Paper No. Director for Morocco” 11993, February. _____ (2013) “Jordan: First Review Przeworski, A., ME. Alvarez, 28 | Economic Consequences of the Political Transition

JA. Cheibub and F. Limongi Annex 1.1: (2000) “Democracy and Devel- Estimating Potential Out- opment: Political Institutions and Well-Being in the World, put 1950-1990”, Cambridge Univer- sity Press. To estimate potential output Ravn, M. and H. Uhlig (2002) which is unobservable, this paper “Notes on Adjusting the Ho- applies the widely known and drick-Prescott Filter for the common technique of the Ho- Frequency of Observations”, drick Prescott (HP) linear filter The Review of Economics and (1997). This univariate filter esti- Statistics, vol. 84, issue 2, May, mates the series that minimizes pages 371-380. the deviation between actual and Rodrik, D. and R. Wacziarg (2005) potential output while constrain- “Do Democratic Transition ing the rate of change in potential Produce Bad Economic Out- output for the whole sample of T comes”, American Economic observations. Alternatively, it de- Review, vol. 95, issue 2, May, composes or filters raw GDP data pages 50-55. into cyclical and trend or structural components. Hence, the HP filter minimizes the following:

where y is the logarithm of real GDP and y*is the logarithm of potential real GDP. λ, is a weighting factor that determines the degree of smoothness of the trend. A low value of λ will produce potential output that follows actual output more closely. The standard pro- cedure is to set λ equal to 1600 for quarterly data, and 6.25 for annual data, following the Ravn- Uhlig (2002) rule which sets λ to 1600*p^4, where p is the number of periods per quarter. T is the length of the time series. The appeal of this approach is that it is simple, transparent, and can be applied to any country where GDP data exist. However, it also has apparent drawbacks, derived from its limitation as a purely statistical technique. First, the HP filter estimate trends in growth without regard to other macroeconomic variables and therefore is unable to capture Chapter 1 | 29

structural changes. Second, the Table A1.3: Robustness Check of Potential HP filter suffers from the “end- Growth Estimates (in percent) - Variations of point problem”: the instability Smoothing Parameters (lambda) in HP filter (Av- of estimates near the end of the erages) sample period, whereby the end- point-data tend to drive estimates 2003- 2012- of the trend at the end of the sam- 2010 2009 2011 2014 ple. This problem may result in South-Med Economies revisions to potential output es- HP filter baseline, lambda=6.25 2.4 2.4 2.4 2.4 timates at or near the end of the HP filter, lambda=100 2.8 2.4 2.0 1.8 sample as new data is released. A HP filter, lambda=200 2.7 2.5 2.2 2.0 common approach to deal with Algeria the problem is to extend the sam- HP filter baseline, lambda=6.25 1.5 1.5 1.5 1.5 ple by using the forecast data, as we do in this chapter. HP filter, lambda=100 2.1 1.4 1.1 1.1 Finally, it is difficult to iden- HP filter, lambda=200 2.0 1.5 1.3 1.2 tify the appropriate value of the Egypt restriction parameter (λ) which captures the importance of cycli- HP filter baseline, lambda=6.25 2.21 2.13 2.05 1.98 cal shocks to output relative to HP filter, lambda=100 2.6 2.2 1.7 1.5 trend output shocks, and thereby HP filter, lambda=200 2.5 2.2 1.9 1.7 controls the smoothness of the series of potential output: a smaller Jordan value of λ indicates a smaller im- HP filter baseline, lambda=6.25 2.4 2.5 2.5 2.4 portance of cyclical shocks and yields a more volatile series of HP filter, lambda=100 3.4 2.9 2.1 1.6 potential output. To resolve this HP filter, lambda=200 3.2 2.9 2.3 1.9 problem, we test the sensitivity of Morocco the results by estimating potential output for several values of HP filter baseline, lambda=6.25 3.0 3.2 3.2 3.3 λ. For this chapter, the restriction HP filter, lambda=100 3.4 3.4 3.3 3.2 parameter λ is set at 6.25, as sug- HP filter, lambda=200 3.3 3.4 3.3 3.3 gested by the literature for annual time series. STATA is used to ap- Tunisia ply all the filters on real GDP in HP filter baseline, lambda=6.25 2.9 2.9 2.8 2.8 local currency. The data ranges HP filter, lambda=100 2.9 2.3 1.8 1.6 over 1980-2019 and are from the WEO database. HP filter, lambda=200 2.9 2.4 2.0 2.0 The sensitivity of the results Source: Calculated by the authors based on World Economic Out- is tested through several robust- look Database, IMF. ness checks, by applying different smoothing parameters (the baseline of 6.25 is varied to 100 and 200). The main findings do not change significantly in the robustness tests performed (table A1.3).

CHAPTER 2

Macroeconomic Policy Response1

2.1 Introduction sociated with different approaches in dealing with As noted in the previous chapter, macroeconomic imbalanc- the political transition in South- es and sluggish economic Med economies has been accom- growth? panied by two main economic • In the South-Med coun- challenges: deteriorating macro- tries, which approach did economic imbalances and a cycli- policymakers choose? cal economic downturn. In prin- • Finally, how effective ciple, addressing these challenges were the adopted policies involves making one of two mac- in achieving their intend- roeconomic management choices. ed objectives? The first is to give priority to cor- The findings of this chapter recting macroeconomic imbalanc- suggest that South-Med govern- es by adopting austerity measures ments have not implemented to limit their damaging impact on coordinated macroeconomic pol- the economy, but that would be at icies to stimulate aggregate de- the expense of contractionary ef- mand and economic growth. On fects on output. The other is giv- the contrary, these have oscillated ing priority to restoring economic over time. At an earlier stage, fis- growth to its potential through cal management was expansion- expansionary policies, but that ary but had the aim of calming would be at the risk of further social discontent at the expense deteriorations in fiscal deficits of exacerbating fiscal imbalances. and public debt, which may harm At a later stage, fiscal adjustment long-term growth. became inevitable and consolida- This chapter identifies and tion measures were adopted often assesses the fiscal and monetary with the support of the IMF. measures adopted by South-Med Meanwhile, the exchange rate countries in response to macro- stabilization policy, led by the economic challenges in the wake South-Med central banks, was of the political uprisings. To this too costly and ineffective in Egypt end, it will attempt to answer the and Tunisia. The attempt by the following questions: Central Bank of Egypt (CBE) to • What are the trade-offs as- ease the persistent squeeze on

31 32 | Macroeconomic Policy Response

foreign exchange liquidity suc- IMF support. In addition, growth ceeded neither in stabilizing the remains below potential and there exchange rate nor in eliminating is evidence of a crowding out of the black market until the floata- private sector credit. Currently, tion of the pound in 2016. In both not only are these countries work- countries, monetary policy has ing towards improving economic been accommodating, which has performance, but they are also sustained inflationary pressures struggling to configure new and and impeded the effectiveness of more inclusive political institu- monetary policy. In Jordan and tions. Morocco, monetary policy has The rest of the chapter is struc- been restrictive, which has stifled tured as follows. The next sec- credit and economic growth. tion reviews the literature on the In Algeria, the price of oil has debate of austerity versus expan- proved to be the dominant factor sion. Section 3 describes and as- in determining macroeconomic sesses fiscal management during outcomes. the transition, regarding its objec- Two other factors have played tive, composition and impact. Sec- an important role in navigat- tion 4 provides the same analysis ing the transition in South-Med for monetary policy. Section 5 ad- countries. The first is access to dresses the issue of monetary and external finance, which provided fiscal policy consistency. Section 6 temporary relief to plug financ- draws some conclusions. ing gaps. However, this support was not accompanied, in Egypt, by the adoption of the necessary 2.2 Understanding reforms to correct short-term and the austerity versus structural imbalances. The other factor resides in the country’s ini- stimulus debate tial conditions at the time of the The austerity versus stimulus de- uprisings. Tunisia and Egypt pro- bate has gained renewed interest vide opposite examples. In 2010, since the onset of the Great Reces- Tunisia had a low fiscal deficit sion in late 2007 and, later, in the and debt, safe foreign exchange context of the Greek crisis. This buffers and low inflation, all of debate is at the heart of economic which accorded the authorities policymaking in South-Med econ- more space to provide monetary omies given that they face similar and fiscal stimulus. By contrast, policy dilemmas. Policymakers Egypt was not in such an envious are confronted with strikingly dif- position, hence a rapid deterio- ferent choices. Like other coun- ration in macroeconomic condi- tries in this position, policymak- tions. ers in South-Med countries could As of today, all South-Med follow expansionary fiscal policy countries are still in the process to stimulate economic growth, or of restoring macro balances and act prudently by giving priority igniting economic growth. De- to fiscal consolidation and debt spite consolidation efforts, fiscal reduction. But which approach is vulnerabilities remain, leading likely to be more beneficial to the South-Med economies to renew Chapter 2 | 33

economy, both now and later? even increase debt further. It Under what conditions would would, as firmly supported by one approach be superior to the Krugman (2010), reduce short- other? And what role could/ term growth and raise unemploy- should monetary policy play in ment. These findings are empiri- both cases? Unfortunately, the cally supported in a study by the literature is inconclusive with re- IMF (2010). Furthermore, Frankel spect to these questions. et al. (2012) argues that regard- Generally speaking, fiscal pol- less of the economic cycle, fiscal icy can be used to stabilise out- policy should be countercyclical, put through automatic stabiliz- meaning spending hikes and/or ers and/or discretionary actions. tax cuts in response to downturns Automatic stabilizers are the cy- and reduced spending and/ clical component of the revenues or taxes hikes during economic and expenditures that automati- booms. Countercyclical policy cally respond to cyclical changes can dampen the severity of busi- in output and make fiscal policy ness cycle fluctuations, reduce automatically countercyclical. For the risks of overheating (inflation instance, during a downturn, tax- and asset bubbles) during booms es decline because profits and in- and mitigate the losses in output comes fall while unemployment and employment during reces- benefits and other social spending sions. It would also ensure that rise. However, the effectiveness an unexpected decline in revenue of automatic stabilizers depends would not lead to abrupt changes on the existence of a large tax base in government spending with with progressive taxation and/or adverse macroeconomic effects. a good coverage of unemploy- More importantly, it could also ment benefits. Therefore, their prevent overspending during up- impact may be hampered by the turns and therefore help to create absence of proper taxation and and maintain the necessary fiscal unemployment benefits and/or space that would be used during may not be sufficient to restore recessions. growth. In this case, discretion- The effectiveness of stimu- ary or active countercyclical fiscal lus depends on the existence of action may be necessary. But not a slack in the economy and on all economists advocate this view the size of the fiscal multiplier, for countries in recession that which is defined as the short-term are seeking to restore economic change in real GDP caused by growth. one unit increase in government Let’s start with the Keynesian discretionary spending. If fiscal view, which argues for stimu- multipliers are large, then addi- lus policies during downturns to tional spending or tax reductions restore growth over fiscal con- could lead to significant output solidation by increasing public expansion. The size of the fiscal spending (borrowing) or easing multiplier itself depends on the taxes. According to this view, fis- country’s initial conditions.2 Mul- cal consolidation to reduce debt tipliers tend to be large (leading would deprive the economy of to a large output loss in the case its key growth drivers and could of a fiscal consolidation, and vice 34 | Macroeconomic Policy Response

versa) if there is less leakage and, effects of public debt build-up on for example, only a small part of output growth would disappear the stimulus is saved or spent on if public debt is on a downward imports. They are also larger if trajectory (Chudik et al., 2015). there are no automatic stabilizers Alternatively, IMF (2010) shows to partly offset the impact of the that reducing government debt initial fiscal shock on GDP. More- would raise long-term output, as over, the impact of fiscal expan- the lighter burden of interest pay- sion on output would be limited ments would create fiscal room if it is associated with concerns for cutting distortionary taxes. about sustainability, leading the Another part of the literature, private sector to increase savings initiated by a paper by Giavazzi rather than investing or consum- and Pagano (1990), has uncov- ing and therefore offsetting the ered “non-Keynesian” effects of impact of fiscal intervention on fiscal adjustments. It shows that GDP. reducing deficits can stimulate Empirically, fiscal multipliers short-term growth, a notion often are found to be larger in down- referred to as the “expansionary swings than in expansions (IMF, fiscal contractions”. The expan- 2010). In a prominent article, sion would occur as an improve- Blanchard and Leigh (2013) show ment in household and business that some economic conditions in confidence, and therefore private advanced countries (interest rates demand, removing the fear of near the zero lower bound (ZLB), future harsher fiscal adjustments poorly functioning financial sys- and future taxes. tems and the existence of signifi- In terms of how contraction cant slack in the economy) were ought to happen, it has been associated with fiscal multipliers shown that spending-based fis- that were substantially higher cal contractions may have lower than those usually assumed. costs in terms of lost output (alter- At the other end of the spec- natively a smaller multiplier) than trum, advocates of austerity poli- tax increases (IMF, 2010). Alesina cies urge governments to resolve and Perotti (1995) and Alesina fiscal imbalances (through fiscal and Ardagna (2010) had already contractions) because the accu- shown that spending-based con- mulation of the stock of public solidations are associated with debt associated with fiscal stim- above average increases in output ulus, particularly if the reces- and private consumption, while sion is persistent or if the initial revenue-based consolidations levels of deficits and debt are are associated with recessions. high, would hamper long-term Perotti (2012) explains that this growth. Empirical support sug- occurs due to monetary policy ac- gests that a debt-to-GDP ratio of commodation (interest rate cuts, 85-90 percent or more may con- low interest rates and currency strain growth (Kumar and Woo, depreciation/devaluation to pro- 2010 and Reinhart and Rogoff, mote exports) to help cushion the 2010 and Cecchetti, Mohanty, and impact of fiscal contraction on Zampolli, 2011). Other evidence demand, as well as reduced infla- shows that negative long-term tion expectations to make wage Chapter 2 | 35

moderation possible. Alternative- The exchange rate can also ly, fixed exchange rates would play an active role in adjusting not provide any adjustment to to current account imbalances (partly) offset the contractionary which, if large and persistent, effect of fiscal consolidation and could lead to a build-up of exces- can thus be associated with larger sive foreign debt and increase an output contractions (relative to economy’s vulnerability to a sud- flexible regimes). Another possi- den stop in capital flows, which ble reason is that tax-based adjust- might force an abrupt and dis- ments (through indirect tax hikes ruptive adjustment. Gervais et for example) often result in price al. (2016) provide empirical evi- increases that would require a re- dence that flexible real exchange strictive monetary policy stance rates (through nominal flexibility) to curb inflation but would hurt can promote current account re- growth (IMF, 2010). Because not balancing, through expenditure- all countries can increase net ex- switching between domestic and ports at the same time, this find- foreign goods, which can promote ing implies that fiscal contraction exports (Friedman, 1953). More- is likely to be more painful when over, they find the adjustment to many countries adjust at the same be more costly in terms of output time (IMF, 2012a). loss for countries that have a fixed This is essentially what the lit- exchange rate regime, since they erature says about fiscal policy; tend to delay exchange rate de- what does it say about monetary valuations until the occurrence of policy? The commonly accepted a currency crisis. view, derived from Friedman’s In short, there are compelling (1968) quantity theory of money, arguments in favour of each of is that in the long-run, changes the two approaches: expansion- in the money supply only cause ary and contractionary policies. changes in prices but not in out- The course of action most suited put. In the short-run, however, to each country depends on its there is a trade-off between initial and specific conditions that changes in the money supply determine the size of the fiscal (inflation) and output because multiplier or, alternatively, the prices and wages are sticky, as foregone loss in output. Expan- illustrated by the Phillips curve sionary fiscal policy is easier to (Friedman, 1968 and Phelps, adopt when the country has fis- 1968). In practice, this means that cal space, when it can access ad- monetary policy can moderate ditional financing at a reasonable short-term real fluctuations, if it cost and when it has the capacity is autonomous. To the extent that to reorder its existing expendi- fixed exchange rates are managed tures and undertake structural with an independent tool (foreign and institutional reforms. Aus- exchange interventions), there terity policies may be advisable would be some scope for autono- when these conditions are not mous monetary policy using interest met and when there is room for rates to influence domestic objec- monetary policy accommodation tives such as inflation and short- to soften the impact of such poli- term output (Ostry et. Al, 2012). cies. Finally, each policy choice 36 | Macroeconomic Policy Response

involves trade-offs and there is no became evident that South-Med free lunch. countries had to consolidate their Against this brief review of fiscal positions. Thus, by 2016, all the literature, we turn next to the South-Med countries, except Al- macroeconomic responses of 5 geria, concluded agreements with South-Med countries in the after- the IMF (See Table 2.1 for details). math of the political uprisings in The timing and pace of the adjust- 2010. The five countries, in alpha- ment in each country were deter- betical order, are: Algeria, Egypt, mined primarily by three inter- Jordan, Morocco and Tunisia. related factors: (i) the severity of social/political unrest, (ii) initial conditions, and (iii) access to ex- 2.3 Fiscal policy response ternal finance. At one end of the spectrum, The 2011 uprisings triggered im- Jordan and Morocco initiated fis- mediate discretionary fiscal ex- cal consolidation as early as 2012. pansions by most South-Med The consolidation effort came governments, which were rolled sooner than in other South-Med back a few years later. This section countries because the political aims to characterise their fiscal tension was less severe. It was responses in terms of objectives, promoted by the relatively weak stance and composition. It also (pre-crisis) initial fiscal condi- aims to assess the effectiveness of tions, which quickly deteriorated fiscal policy in terms of sustain- in both countries to 9 and 8 per- ability and impact on growth. cent of GDP, respectively. Also, these countries were not the re- Fiscal policy objectives, stance cipient of major financial sup- and composition port from abroad. In both cases, Starting with fiscal policy objec- the consolidation programs were tives, it seems that South-Med accompanied by measures to countries have attempted, in the protect the poor. In Morocco, for past seven years, to strike a bal- example, there was a budget re- ance between easing social ten- allocation in favour of measures sion, requiring an expansionary to strengthen social safety nets policy, and containing the dete- through increases in education rioration of fiscal imbalances, re- allowances, medical insurance quiring fiscal consolidation and and public transportation (IMF, austerity measures. 2015a). During the early years of the At the other end of the spec- uprising (2011-2012), priority was trum, Egypt delayed fiscal con- given to the political objective solidation the most, even though in most countries. Fiscal expan- it had the worst initial fiscal con- sions were used either to ease dition (an overall deficit of 8 per- rising social discontent (Tunisia cent and a primary deficit of 2.5 and Egypt) or, preemptively, to percent of GDP) of all South-Med maintain social peace in coun- countries. This was not for lack of tries that did not experience sig- trying. On the contrary, Egypt ne- nificant unrest (Algeria, Jordan gotiated a program with the IMF and Morocco). At a later stage, it as early as mid-2011 for a USD Chapter 2 | 37

3 billion loan, which met with Table 2.1: IMF programs in South-Med countries public resistance and cancella- in the post-uprising period tion (Hanieh, 2015). A second attempt was made in mid-2012 for Country Date Program Size Duration a loan of USD 4.8 billion, but this Jordan August Stand-By- SDR 1.364 3 years 2012 Arrangement billion (about attempt suffered the same fate as (SBA) US$2 billion) the previous one. Another factor August Extended Fund SDR 514.65 3 years that may have delayed consolida- 2016 Facility (EFF) million (about US$723 mil- tion was having access to external lion) financing from the GCC. With Morocco August Precautionary SDR 4,117.4 2 years these resources, it was possible to 2012 and Liquidity million (about embark on a fiscal stimulus pack- Line (PLL) US$6.21 billion) age (of around 1.2 percent of GDP July Precautionary SDR 3.2351 2 years 2014 and Liquidity billion (about in 2013) with the dual objectives Line (PLL) US$5 billion) of activating the economy and July Precautionary SDR 2.504 2 years achieving social justice (MOF, 2016 and Liquidity billion (about 2014), making Egypt the only Line (PLL) US$3.47 billion Tunisia June Stand-By- SDR 1.146 2 years, South-Med country to have made 2013 Arrangement billion (about 7 months an explicit attempt at stimulating (SBA) US$1.74 billion) extension aggregate demand. By mid-2014, April Extended Ar- SDR 2.04 4 years however, the deterioration in 2016 rangement un- billion (about der Extended US$2.9 billion) public finance (averaging 13 and Facility Fund 7 percent of GDP for the over- Egypt Novem- Extended Fund SDR 8.5966 3 years all and primary deficits, respec- ber 2016 Facility (EFF) billion (about US$12 billion) tively), the easing of social pres- Source: IMF. sure and the curtailment of GCC funds, prompted the government to unilaterally reduce energy sub- employment promotion and cash sidies. By 2016, four years later transfers (IMF, 2013a). The imple- than Jordan and Morocco, the mentation of the program in 2013 government concluded an agree- triggered social protest (Hanieh, ment with the IMF. 2015), which caused some delay. Tunisia falls in the middle. As for Algeria, fiscal manage- Like Egypt, prolonged social un- ment was driven primarily by oil rest made it difficult to reverse prices. And, like other South-Med spending expansions. A healthy economies, fiscal policy was used initial (pre-crisis) condition made initially to quieten discontent. The it possible to sustain expansion- recent sharp drop in hydrocarbon ary policies for a year longer than revenues on the back of falling oil Jordan and Morocco. By 2013, prices led to cuts in spending and however, worsening fiscal bal- an overall deterioration in public ances (overall and primary defi- finance. cits of 7 and 6 percent of GDP, To formalise the above narra- respectively) triggered consoli- tive, the fiscal impulse (FI) was dation plans with IMF support. computed for South-Med coun- This program involved partial tries to capture the stance of fiscal redirection of resources from sub- policy during the period 2011- sidy savings towards such social 2015. The methodology used is programs as education, health, 38 | Macroeconomic Policy Response

explained in Annex 2.1 at the end tionary stance between 2013 and of this chapter and the results are 2015. Meanwhile, higher spend- shown in figure 2.1 in 3 different ing maintained the expansion- ways. The figure shows that, over ary stance in Tunisia until 2013. the period 2011-2015, fiscal policy Spending-based contractions was contractionary in Morocco took place in 2014 though the fis- and Jordan, where consolida- cal stance slightly eased again in tion was implemented less than 2015. Finally, unlike other South- two years after the uprising, but Med economies, Egypt’s fiscal expansionary in Egypt and Tu- expansionary stance continued in nisia, where fiscal consolidation 2013-2014 because of large spend- was undertaken at a later date. In ing expansions. However, the fis- Algeria, fiscal policy was expan- cal impulse slowed down in 2014 sionary throughout. on the back of partial fiscal con- Breaking down the 2011-2015 solidation. period into sub-periods, figure Turning to the composition 2.1 shows that, during the early of the fiscal response, figure 2.2 years of the uprising, the fiscal shows that the spending hikes impulse was positive in all South- during the first two years of the Med economies, suggesting an political uprising were mostly expansionary fiscal policy. The in the form of current spending. FI reached close to 4 percent- The size of fiscal expansions average points of GDP in Algeria, aged slightly less than 3 percent 2.2 percentage points in Tunisia of GDP in South-Med countries, and around 1 percentage point in but there were considerable vari- other South-Med countries. Fiscal ations across countries. Tunisia consolidation in Jordan and Mo- experienced the most significant rocco led to an overall contrac- increase in current spending (by

Figure 2.1: Fiscal impulse (FI) in South-Med countries after the uprisings (percent)

Note: The fiscal impulse measure was calculated once based on overall deficit, once on deficit excluding grants and once on primary deficit excluding grants. Source: Calculated by the authors based on fiscal data extracted from IMF Article IV Consultation central government budget tables, and potential output was calculated using the HP filter, lambda = 6.25. Chapter 2 | 39

Figure 2.2: Primary expenditure components (percent of GDP)

Note: Current spending excludes interest. Source: Calculated by the authors based on fiscal data extracted from IMF Article IV Consultation central government budget tables. around 10 percentage points of 2012 curtailed current spending GDP between 2010 and 2013). by 4 and 5 percentage points of Meanwhile, current spending GDP, respectively. In Morocco, in Jordan and Morocco was in- the removal of subsidies (on both creased by 2 and 3 percentage petroleum products and basic points of GDP. Algeria and Tuni- foodstuffs) and the partial index- sia increased public sector wages ation of domestic retail energy and subsidies, while Jordan and prices to world prices, brought Morocco increased subsidies. subsidies from 6 percent of GDP To finance spending expan- in 2012 to less than 4 percent of sions, Jordan cut capital spend- GDP in 2014. Despite the inflow ing, especially as revenues of some grants to Morocco, fis- decreased (figure 2.3). In fact, Jor- cal consolidation was associated dan’s substantial subsidy increase with reduced capital spending. In was entirely offset by an equal Jordan, fuel pump subsidies were reduction in capital spending. also eliminated in late 2012 and Morocco could increase domestic an automatic pricing adjustment revenues, which also allowed it to was re-established and accompa- increase capital expenditure (by nied by introducing cash trans- 1 percent of GDP), making it the fers to about 70 percent of the only South-Med country to have population. Fiscal consolidation done so. Algeria and Tunisia, was financed by higher revenues with the largest fiscal expansions, (including grants in 2014) so that resorted to both capital spending the government could redirect re- cuts and higher revenues (includ- sources from current spending to ing grants of 1 percent of GDP, in capital spending, which increased 2012, for Tunisia). by 1 percent of GDP between 2012 The austerity measures in Jor- and 2014. dan and Morocco that began in 40 | Macroeconomic Policy Response

Figure 2.3: Revenues Components (percent of GDP)

Source: IMF Article IV Consultation fiscal data.

In Tunisia, the consolidation of GDP). These fiscal expansions plans curtailed public spending were partly financed by substan- in 2014 (from 26 to 21 percent tial grants and a dormant govern- of GDP) by reducing subsidies, ment deposit at the CBE (MOF, which went down from 10 to 6 2014). In 2015, the deterioration percent of GDP. However, on the in public finances forced the gov- back of unceasing tension, Tuni- ernment to curtail current spend- sia was unable to fully implement ing with substantial fuel subsidy the revenue-enhancing measures reductions (down to 6 percent of (higher stamp duties and excises GDP from 9 percent in 2013) and on alcoholic beverages). to impose a freeze on investment In Egypt, declining fiscal rev- spending. enues in 2011-2012 led to a com- During the period 2011-2015, bination of cuts in capital spend- the composition of public spending, which was already low, and ing in all South-Med economies a freeze of (primary) current was biased in favour of wages and spending as a ratio to GDP. Start- subsidies, at the expense of public ing in 2013, current spending ex- investment. In fact, capital spend- pansions (up from 20 percent of ing was brought down from an GDP in 2011-2012 to 22 percent average (excluding Algeria) of 5.5 in 2013) occurred on the back percent, in 2010, to 4.4 percent of of higher subsidies, social ben- GDP, in 2015, while wages and efits and wages. The 2014 fiscal subsidies rose to 15.5 percent of stimulus included current spend- GDP (from less than 14 percent of ing hikes (up by 0.8 percentage GDP). In Egypt and Tunisia, these point of GDP) in the form of sal- trends were more pronounced. ary increases (minimum wage As such, fiscal management was for civil servants) pensions and driven by social concerns at the other social benefits. Moreover, expense of fiscal sustainability. it also included an increase in Only in Egypt was there a short- capital spending on infrastruc- lived attempt to use fiscal policy ture projects (0.4 percentage point to stimulate aggregate demand. Chapter 2 | 41

Sooner or later, consolidation was Table 2.2: Fiscal vulnerability in South-Med inevitable in all South-Med coun- economies tries. 2003- 2009- 2011- 2013- 2008 2010 2012 2015 Fiscal sustainability and growth Algeria Starting with fiscal sustainability, figure 2.4 and table 2.2 show Fiscal impulse -0.7 2.1 3.8 2.6 that the fiscal expansions and/ Primary balance 9.8 -4.0 -2.7 -7.9 or revenue drops in 2011-2012 Primary balance (without grants) 9.8 -4.0 -2.7 -7.9 were accompanied by a substantial increase in primary deficits Debt-to-GDP ratios 61.4 -30.6 -25.3 -16.2 and debt to GDP ratios. Fiscal Egypt consolidation helped contain fis- Fiscal impulse -0.9 0.4 1.1 1.8 cal vulnerabilities and maintain Primary balance -2.5 -2.0 -3.7 -4.5 overall macroeconomic stability, Primary balance (without but concerns remained regarding grants) -3.1 -2.5 -4.1 -6.9 public debt. Debt-to-GDP ratios The fiscal consolidations in 66.2 54.5 59.4 70.8 Jordan and Morocco, in 2012, Jordan improved the situation, so that Fiscal impulse -0.2 -1.6 0.7 -1.0 by the end of 2015 primary bal- Primary balance ances were close to pre-crisis -0.6 -5.0 -4.9 -5.6 ratios. However, public debt in Primary balance (without grants) -6.9 -7.1 -8.7 -9.2 both countries rose beyond pre- crisis levels. Per the IMF, the debt Debt-to-GDP ratios 79.8 69.6 84.8 81.3 to GDP ratios should be reduced Morocco further to 77 percent in Jordan Fiscal impulse 0.1 2.5 1.0 -0.5 and 60 percent in Morocco by 2021 and 2020, respectively (IMF, Primary balance 0.6 -0.7 -4.7 -2.1 2016a and 2015b). To achieve Primary balance (without these targets, both countries em- grants) 0.3 -1.1 -4.8 -3.2 barked on successor IMF arrange- Debt-to-GDP ratios 56.4 48.1 56.2 63.8 ments in 2016. Tunisia Given that Tunisia’s fiscal con- Fiscal impulse solidation program was mindful -0.2 -0.1 2.2 -0.9 of social discontent, it was imple- Primary balance 1.5 0.5 -3.3 -2.6 mented at a slow pace and thus Primary balance (without the primary deficit only marginal- grants) -0.1 0.3 -3.8 -3.0 ly improved in 2014-2015, relative Debt-to-GDP ratios 51.8 41.7 44.9 51.1 to the stimulus years of 2011-2012. Note: The reported fiscal impulse measure excludes both grants However, thanks to the country’s and interest. favorable initial conditions, the Source: Calculated by the authors based on fiscal data extracted debt levels remain slightly below from IMF Article IV Consultation central government budget pre-crisis levels and primary defi- tables, and potential output data for the FI measure using the HP filter, lambda = 6.25. cits are only around 4 percent of GDP. The target for the debt-to- GDP ratio in 2020 is under 50 percent (IMF, 2016b). 42 | Macroeconomic Policy Response

Figure 2.4: Primary deficits, excluding grants (per- finance the expansions without cent of GDP) significant harm to fiscal vulnerability. However, with the drop in oil prices starting 2014, the IMF (2014a) estimates that fiscal policy is currently on an unsustainable path. Algeria’s fiscal breakeven oil price increased from $20/barrel in 2003 to $125/barrel in 2012 and was projected to reach $136/ barrel in 2014, keeping in mind that the fiscal breakeven oil price is the theoretical price of oil that would yield a balanced budget in a given year. Source: Calculated by the authors based on fiscal data extracted from IMF Article IV Consultation central government budget What about economic growth? tables. Although early expansionary policy in South-Med countries was not explicitly motivated by stimulat- Despite the spending freeze ing aggregate demand, it would in the early years of the uprising, still be expected to have some Egypt’s primary deficits (exclud- impact on economic growth. To ing grants) deteriorated from 2.5 assess the macroeconomic effects percent in 2010 to almost 4.4 per- of fiscal management in South- cent of GDP in 2012, on the back Med economies, we attempt to of lower revenues and heavy gov- determine the counter-cyclicality ernment borrowing. The latter re- of different budget components sulted in an increase in the cost of (revenues and spending) to find government borrowing, signal- out how fiscal policy may have ling an elevated risk premium. combated business cycle fluctua- The 91-day T-bill rate jumped tions. The methodology we use is from an average of 10 percent in explained, along with the detailed 2010 to above 18 percent in early results, in Annex 2.2 at the end of 2012, the highest rate since 2008. this chapter. Here we plot the FI The Egyptian government also and economic growth between encountered difficulties on sev- two points: 2010-2011 and 2014- eral occasions in raising the re- 2015 in figure 2.5 and make a few quired amounts for its T-bill auc- broad remarks below. tions during the early years of the Overall fiscal management has uprising, as investors demanded not been very effective in stimu- higher yields. With the accelera- lating economic growth in South- tion of expansion in 2013-2014, Med countries, which remained and despite access to external fi- below potential. Early fiscal ex- nance, fiscal imbalances deterio- pansions took the form of politi- rated further. The partial consoli- cally motivated current spending, dation of 2015 offset some of this which could have at best have deterioration but sustainability contributed to temporary boosts concerns remain. of consumption and investment. In Algeria, oil revenues could But, as shown in figure 2.5, fiscal Chapter 2 | 43

Figure 2.5: Fiscal impulse and growth rates between 2010-11 and 2014-2015 (percent)

Source: Calculated by the authors based on computed fiscal impulse and World Economic Outlook database, IMF. expansions were associated with In addition, the expansion- slightly higher growth in Egypt. ary policy was financed through In other countries, growth essen- extensive borrowing, which re- tially stagnated between 2011- sulted in higher overall costs of 2012 and 2013-2015. borrowing and a crowding out One reason for this result is of private investment. As budget that expansionary policies were deficits soared in the absence of reversed in a relatively short pean effective secondary market for riod of time. And indeed, in the T-bills, governments in Egypt and case of Tunisia, the IMF (2012b) Jordan have increasingly relied shows that the 2011 fiscal im- on commercial banks to finance pulse may have contributed to their deficits (figure 2.6). Credit GDP growth by 0.5 to 0.7 percent to the government outpaced pri- in 2011. The other explanation is vate credit growth in most South- that current spending stimulus Med countries, which more than tends to have lower multiplier doubled in Egypt, tripled in Tu- effects than capital spending. nisia and exponentially increased Preliminary estimates for Tunisia in Morocco. Jordan temporarily find a multiplier on capital spend- suffered from crowding out, but ing of about 0.3, while that on that trend was reversed with fis- current spending is about half of cal consolidation efforts in 2014 that (IMF, 2012b). Most of the ex- and 2015. pansion in South-Med countries Eventually, fiscal manage- took the form of current spend- ment in South-Med countries pri- ing due to the prevailing socio- oritized the preservation of mac- political difficulties, which were roeconomic stability over growth, costly but perhaps unavoidable. for some countries sooner than In other words, the fiscal expan- others. The decision to consoli- sions were not designed to per- date was inevitable for all coun- manently boost investment and tries, including most recently in job-creating growth, but rather to Egypt. The lax fiscal policy pur- accommodate social grievances. sued during the uprisings limited The only exception to this was fiscal space to counteract the de- one short-lived stimulus plan in terioration of economic activity, Egypt, in the second half of 2013. suggesting that the countercycli- 44 | Macroeconomic Policy Response

Figure 2.6: Bank financing of the deficit, 2010-2015 mandate, or actually pursued (percent of total financing) monetary policy goals. It also assesses the impact of monetary 94.7% 92% 96% 100% 88% 89% 81% 83.8% policy on domestic goals such as 80% 64.2% inflation and economic growth. 60% 41% 40% Objectives and exchange rate

20% management Notwithstanding announced ob- 0% -0.5% jectives, all South-Med central -20% banks are exchange rate targeters, -40% at least in de facto terms. Observ- -44.0% -60% ing their actions, central banks 2010 2011 2012 2013 2014 2015 seem to have strived to maintain an official parity between an an- Egypt Jordan chor currency (or basket) and Source: Calculated by the authors based on data from the month - ly financial bulletin from Egypt’s ministry of finance website and their domestic currency (Table Jordan’s IMF Article IV Consultation data. 2.5). Algeria, Jordan and Morocco have had formal peg arrangements, with explicit exchange rate cal capacity of policies was se- targets. Egypt and Tunisia have verely constrained. For most, the de jure more flexible arrange- only way to adjust their fiscal bal- ments, but they implicitly attempt ances was to borrow externally to maintain exchange rate stabil- (through IMF support). Never- ity or, at a minimum, to reduce theless, despite consolidation ef- exchange rate volatility. forts, fiscal vulnerabilities remain, Because foreign exchange in- leading many South-Med econo- terventions were used as an in- mies to renew IMF support. In all dependent tool to manage the South-Med economies, growth exchange rate, there was some has remained below potential and scope for monetary policy in there is evidence of a crowding certain countries, through inter- out of private sector credit. est rates, to pursue domestic objectives, such as price stability, short-term output stabilization 2.4 Monetary policy and financial stability. And some response countries have done so. For example, the central banks in Egypt Among other effects, the 2011 and Tunisia have attempted to al- uprisings were associated with leviate liquidity shortages in the a drop in foreign exchange earn- banking sectors. The central bank ings and short-term capital out- of Egypt (CBE) has supplied for- flows, both of which put signifi- eign exchange liquidity to banks cant pressure on current accounts through regular and exceptional and, in turn, on nominal exchange auctions. The Central Bank of rates. This section explores how Tunisia (CBT) has provided the the central banks in South-Med direct refinancing of banks through countries responded to foreign liquidity injections to alleviate exchange scarcity, given their structural liquidity shortages due Chapter 2 | 45

Table 2.5: Exchange rate regimes in South-Med economies, 2010-2014 Country 2010 2011 2012 2013 2014 Algeria Composite No change No change No change No change exchange rate anchor (managed arrangement) Egypt Crawl-like Other stabilized Other crawl-like Stabilized ar- No change arrangement arrangement stabilized arrangement rangement with- against a com- without an explicit out an explicit posite basket exchange rate anchor exchange rate (de facto peg to US$) anchor (de facto peg to the US$) Jordan Conventional No change No change No change No change Peg to USD Morocco Conventional No change No change No change No change Peg to a composite basket (Euro- USD with 80/20 weights) Tunisia Other conven- n/a Crawl-like arrange- No change No change tional fixed peg ment without an arrangement explicit exchange rate with a composite anchor (de facto peg to anchor a composite basket) Source: IMF. to increased demand for cash and South-Med countries, central sizeable deposit withdrawals. banks have shown notable reluc- These measures aim at preserving tance to let the exchange rate ad- financial stability. just to ever-deteriorating current The pursuit of monetary pol- accounts in the aftermath of the icy has been constrained, how- political shock. As shown in fig- ever, by fiscal dominance in all ure 2.7, there has been very little South-Med economies because of movement in the nominal ex- high fiscal deficits and large pub- change rate versus the US$ or the lic debt. Under a fiscally domi- basket of currencies in almost all nant regime, as in Sargent and South-Med countries, except for Wallace (1981), central banks may Egypt. Central banks have used not be able to effectively counter foreign exchange interventions inflationary pressure. They also to stabilise the currency, interna- tend to succumb to government tional reserves permitting. pressure to finance the deficit or At the same time, following do away with a tighter monetary the reversal of capital inflows, the policy stance to avoid increasing response of South-Med central the cost of public debt service. In banks focused on eroding reserve this context, monetary policy ob- buffers accumulated during pre- jectives are subservient to fiscal vious years. Figure 2.8 shows that objectives. depletion was significant during At any rate, because the ex- the early years of the uprising, change rate has been a central which stood then at healthy lev- component in the conduct of els of 9 (Jordan) and 5 (Tunisia) monetary policy and an impor- months of imports before the po- tant anchor for expectations in litical shock. The safe benchmark 46 | Macroeconomic Policy Response

Figure 2.7: Long-term exchange rates, 1990-2015 scarcity of capital inflows forced South-Med countries to either ini- tiate an exchange rate adjustment, or to borrow to finance the current account. The choices they made depended on (i) the persistence of current account imbalances, (ii) their willingness to adopt a policy mix of reserve depletion and exchange rate flexibility, and (iii) access to external finance. At one end of the spectrum, reserve depletion reached its lim- Note: Bilateral exchange rates (local currency per USD) is plotted its in Jordan and Morocco sooner for Egypt, Jordan and Tunisia. An increase indicates a deprecia- than in Egypt and Tunisia. In tion of the domestic currency against the USD and vice versa. The the former countries, where ex- nominal effective exchange rate is plotted for Algeria and Mo- rocco because their currencies are pegged to a composite basket. change rate regimes are strictly An increase indicates an appreciation of the domestic currency managed, extensive foreign ex- against the anchor basket and vice versa. change interventions were suc- Source: World Development Indicators, World Bank. cessful in maintaining exchange rate stability. However, the pace of reserve depletion worsened in Figure 2.8: Reserves in months of imports, 2010- 2012, relative to 2011, leading to a 2016 fall in reserves below 4 months of imports. The IMF support in 2012 subsequently facilitated access to capital markets. Reserve accumulation was resumed in Jordan by sizeable grants from the GCC countries in 2013 (U$1.25 billion in grants), the World Bank (U$150 million emergency loan to cover the cost of hosting the Syrian refugees), as well as a successful US dollar-denominated domestic bond issuances (IMF, 2014b and Source: data from national central banks World Bank, 2014). Similarly, reserves in Morocco improved on the back of a narrowing current for reserve adequacy is about 3 account deficit and a significant months of imports, which is a increase in foreign direct invest- proxy for current account vulner- ment. By 2014 and 2015, both ability when the balance of pay- countries had recovered their pre- ments is dominated by trade and crisis levels of reserve holdings. when the country has limited ac- At the other end of the spec- cess to capital markets. trum, the central banks of Egypt Once safe reserve buffers and Tunisia combined limited ex- were depleted, persistence of the change rate flexibility and reserve current account imbalances and depletion. By the end of 2012, the Chapter 2 | 47

exchange rates had depreciated also set a ceiling for net foreign respectively by 6 and 9 percent in exchange interventions so that it Egypt and Tunisia, respectively, could only intervene to smooth accompanied by a depletion of “excessively large” fluctuations. around 60 and 25 percent of pre- Despite these measures, the Tuni- crisis foreign exchange holdings. sian currency has also lost half of Among all South-Med econo- its value. mies, Egypt is the country where Both Egypt and Tunisia have official reserves have come down benefited from access to foreign the most. In 2011 alone, it had grants, which provided tempo- lost more than USD 20 billion (or rary boosts to reserves. In Egypt, more than 40 percent of 2010 re- starting 2013, reserve accumula- serve holdings). tion largely reflected sporadic de- As current account pressures posits made by Turkey, Libya and persisted and reserve buffers several GCC states. It is estimated reached a critical minimum, cen- that the total amount of these de- tral banks were compelled to re- posits between May 2012 and No- lax exchange rate management, vember 2014 amounted to USD starting late 2012. In Egypt, the 20 billion. Moreover, the govern- central bank introduced small de- ment could access capital markets valuations paced out over a long through the issuance of foreign period (12 percent in Q1-2013, 6 currency denominated bonds. percent in Q1-2015 and 14 percent In Tunisia, stronger FDI inflows, in March and May 2016). Since the sustained multilateral and bilat- uprisings, the Egyptian currency eral loans (World Bank (US$500 has lost 50 percent of its value. By million) and the African Develop- comparison, the CBT was more ment Bank (US$200 million)), and flexible and more mindful of re- access to capital markets (with serve erosion. By September 2011, US Treasury and Japan Bank for it abandoned the euro-dollar International Cooperation (JBIC) basket as the primary monetary guarantees) boosted central bank anchor and added a monetary ag- reserves (IMF, 2013a). gregate along with the exchange Under increasing pressure, rate anchor. Moreover, it started the CBE and the CBT introduced in late 2012 to base its reference price and capital controls. In 2013, exchange rate on the average in- the CBE imposed restrictions on terbank exchange rate, instead of bid-ask spreads quoted by autho- the fixed currency basket. These rised foreign exchange dealers, measures led to several exchange thus prohibiting them from quot- rate adjustments that helped limit ing their own rates. These restric- CBT interventions to cases when tions became tighter over time. quotations deviated significantly Other capital control measures from the daily interbank rate and were extended to the repatria- overall offset some reserve losses tion of export proceeds of certain (IMF, 2012b). In February 2016, products and limits on transfers the CBT introduced a weekly abroad and, in February 2015 foreign-exchange auction system on foreign exchange cash with- to help promote competitive bid- drawals (and deposits). Later on, ding for foreign exchange. It has limits were imposed on foreign 48 | Macroeconomic Policy Response

exchange currency transactions stock of reserves, respectively. using debit and credit cards. The As exchange rate devaluations CBE has also issued an import were not sufficient for the current priority list for the allocation of account to self-correct, interven- foreign exchange by banks. The tions compromised reserve ad- CBT imposed similar restrictions. equacy. In Egypt, reserves had For instance, it required surplus reached a record low of around funds in foreign currency ac- US$ 13 billion in 2013. Currently, counts to be converted to dinars. the reserve levels in Egypt and However, as of 2016, ex-ante con- Tunisia can only finance 3 and 4 trols on capital account transac- months of imports, respectively, tions were replaced with ex-post and have fallen from triple and controls and ceilings for foreign- double these figures in 2006. ers’ holdings of government se- Secondly, repeated currency curities were eliminated. Also, devaluations never brought the greater powers were granted exchange rate to a value consis- to banks for FX transfers (IMF, tent with long-run fundamentals 2016b). The imposition of partial and therefore never provided ef- capital controls did not preserve fective stabilization. Short-term the stability of the exchange in management (arbitrary and size- both countries. able reserve depletion and un- In a category of its own, the sustainable access to external fi- exchange rate and reserves in Al- nance) in the face of a prolonged geria remained at the mercy of oil shock (owing to political insta- revenues, which only came under bility and/or security incidents) pressure with the decline in oil proved ineffective, ill perceived prices in 2014 and 2015. and costly. Access to foreign What does all this mean? Firstly, grants may have only delayed the reliance on foreign exchange much needed exchange rate de- intervention for exchange rate valuations. At the end, Egypt had stabilization, which was used in to let the exchange rate go in 2016. Jordan and Morocco, was effec- Thirdly, exchange manage- tive in preserving exchange rate ment in Egypt resulted in severe stability with IMF support. In market distortions. As the official Egypt and Tunisia, where a pro- exchange rate was far from its tracted political transition led to a long-term equilibrium value, an more permanent drop in foreign acute foreign exchange liquidity exchange earnings, the combina- squeeze ensued, and eventually tion of exchange rate and reserve led to the emergence of a black ex- mix proved largely ineffective. change rate market in June 2014. This could be traced to several The black-market rate has been factors. First, the policy mix was depreciating at a faster rate than tighter-than necessary and there- the official rate and the black- fore costly in terms of reserve market premium has been widen- depletion. By mid-2016, almost 6 ing over time, reaching more than years after the uprising, the cur- 40 percent in August 2016 (figure rencies of Egypt and Tunisia had 2.9). With the floatation of the lost half of their 2010 values and pound in 2016, the black market around half and a third of the has almost vanished. Chapter 2 | 49

Figure 2.9: Black market premium in Egypt forex market, 2014-2016

Source: Calculated by authors based on market data from HC securities.

Fourthly, ad hoc interventions ly, these tools were not always by the central bank and the lack deployed in a consistent fashion, of a clear nominal target for mon- which makes it difficult to draw etary policy may have fed market conclusive observations about the expectations with negative senti- stance of monetary policy. Never- ments and exacerbated credibil- theless, an attempt is made here ity concerns. In March 2016, for to disentangle this issue. example, the CBE issued a press Starting with reserve require- release to justify an exchange rate ments, most central banks in devaluation, indicating that the South-Med countries in the post policy would become more flex- uprising period eventually opted ible, thus shaping market expec- for monetary easing through re- tations towards less intervention ductions in reserve requirements and future devaluations. This (Table 2.6). The CBE cut the re- step was followed the next day, serve requirement ratio twice, however, by a sizeable (excep- in March and May 2011. It intro- tional) foreign exchange interven- duced a short-term instrument tion that stabilized the exchange (7-day repurchase agreements in rate for the following 6 months. March 2011 and 28-day repurchase agreements in June 2012) Monetary policy stance and mac- to relieve liquidity pressures in roeconomic effects the banking system. The CBT also Prior to the uprisings, South-Med cut the required reserve rate on economies had strengthened their demand deposits 3 consecutive monetary policy frameworks by times in 2011, by a total of 10.5 introducing policy rates that set percentage points to 2 percent. the target for overnight interbank Reserve requirements were fur- rates and other interest rates. In ther lowered in 2013 to 1 percent. addition, they had at their dis- In addition, in 2015, the CBT in- posal other tools to influence mar- troduced foreign exchange swaps ket liquidity, including reserve to support liquidity. The CBM requirements and shorter term eased reserve requirements start- maturity instruments that could ing 2013 before reverting to some ease banking sector liquidity and tightening in 2016. And while spur credit growth. Unfortunate- the CBJ did not adjust reserve 50 | Macroeconomic Policy Response

Table 2.6: Reserve requirements, 2010-2016 (per- in September 2011 and another 50 cent) bps in March 2013) to curb inflation. But this position was re- Year Algeria Egypt Jordan Morocco Tunisia versed into a more relaxed stance 2010 8.0 14 7.00 6.00 12.50 between August and December 2011 9.0 14 7.00 6.00 2.00 2013 (with three cuts in policy 2012 11.0 10 7.00 6.00 2.00 rates, cumulatively adding up 2013 12.0 10 7.00 4.00 1.00 to 150 bps), with the aim of sup- 2014 12.0 10 7.00 2.00 1.00 porting economic growth. The subsidy reduction in July 2014 2015 12.0 10 7.00 2.00 1.00 prompted policy tightening and 2016 12.0 10 7.00 4.00 1.00 the official stance has reverted to Source: IMF. being restrictive since December requirements, it introduced a 2015 (three nominal policy rate in- weekly repurchase agreements creases until June 2016), bringing facility in 2012 to provide liquid- policy rates to historically high ity to banks and minimise vola- levels (of 11.75 and 12.75 percent tility in interest rates in the inter- for overnight deposit and lending bank market. In addition, forward rates, respectively). forex operations with banks were In Tunisia, monetary policy also introduced. Algeria is the has focused on maintaining the only country that increased re- liquidity of the banking sector, serve requirements over time. which had been facing structural With respect to interest rates, shortages since the beginning Central Banks in South-Med of the political uprising, caused countries oscillated between ac- by weak economic activity and commodative and restrictive a deposit drawdown by cash- stances. As shown in Table 2.7, strapped mining companies im- Egypt adopted a restrictive stance pacted by work stoppage (IMF, in the early years of the uprisings 2013a). It was relatively accom- (with interest rate hikes of 100 bps modative in 2011 (interest rates

Table 2.7: Changes in policy rates, 2010-2016 (percent) 2010 2011 2012 2013 2014 2015 2016 Change in nominal policy rates Algeria 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Egypt 0.00 1.00 0.00 -1.00 1.00 0.00 2.50 Jordan -0.50 0.25 1.75 -0.50 -0.75 -1.25 0.00 Morocco 0.00 0.00 -0.25 0.00 -0.50 0.00 -0.25 Tunisia 0.00 -1.00 0.25 0.25 0.75 -0.50 0.00 Change in real policy rates Algeria 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Egypt 0.00 1.00 0.00 -1.00 1.00 0.00 2.50 Jordan -0.50 0.25 1.75 -0.50 -0.75 -1.25 0.00 Morocco 0.00 0.00 -0.25 0.00 -0.50 0.00 -0.25 Tunisia 0.00 -1.00 0.25 0.25 0.75 -0.50 0.00 Source: data from national central banks. Chapter 2 | 51

were cut twice consecutively in the real sector, and, if they persist, June and September 2011 by 100 could also compromise financial bps to their lowest level since stability. 2009), with the aim of easing In Jordan and Morocco, on the credit conditions, but switched back of low-inflation, monetary to tightening between 2012 and policy has been mostly accommo- 2015 (interest rate hikes in August dative in support of growth. In 2012 by 25 bps and in March 2013, Jordan, apart from a short period January and July 2014, cumula- of nominal policy tightening be- tively adding up to 100 bps), in tween June 2011 and November an attempt to reverse excessive 2012 (with the June hikes aimed credit growth. The CBT’s attempt at ensuring a competitive yield to relax the monetary stance once on local currency-denominated more in November 2015 did not assets and the November hikes trigger a decline in real rates, sug- aiming to curb the second-round gesting a restrictive stance. effects of inflation arising from Monetary policy in Egypt and subsidy reform), the CBJ reduced Tunisia has kept real interest rates interest rates between 2012 and in negative territories most of the 2015. Morocco’s central bank time (figure 2.10). The situation is has also attempted to implement more worrying in Egypt because monetary policy expansion in negative interest rates have per- nominal terms. Interest rate cuts sisted over a long period while in 2012, 2014 and 2016, brought nominal interest rates are high. policy rates to historically low Real negative rates suggest that levels of 2.25 percent. However, monetary policy is ineffective in in contrast to Egypt and Tunisia, affecting liquidity since agents the real policy stance in Jordan would prefer to hold cash balanc- and Morocco has been rather re- es rather than hold bank deposits strictive (except for 2012 in Jor- and to spend rather than save. dan) as real interest rates have Of course, negative interest rates increased on the back of deceler- could support a modest expan- ating inflation and even deflation sion in credit, but they also tend since 2015 in Jordan. In December to impede the transmission mech- 2013, the CBM also established a anisms of monetary decisions to new refinancing mechanism to

Figure 2.10: Nominal and real interest rates (percent) (a) Nominal (b) Real

Source: data from national central banks. 52 | Macroeconomic Policy Response

support credit to small enterpris- pharmaceuticals and vaccines). In es, particularly in the industrial or addition, it intervened in the US export-oriented sectors. dollar interbank market in 2015, In Algeria, monetary policy with a total amount of USD 1.2 has been completely inactive billion (CBE, 2015). through interest rates but policy In addition, in 2016 the CBE tightening was implemented to launched an initiative to in- some extent in 2011 and 2012, crease access to finance by SMEs with an increase in the banks’ (at lower interest rates) through minimum reserve requirement commercial banks. In return, par- to control liquidity and inflation. ticipating banks have been per- The CBA introduced, in January mitted to reduce their level of re- 2013, a six-month liquidity man- quired reserves by the CBE with agement instrument. an amount equivalent to their Beyond conventional mea- SME lending. This program aims sures, the CBE and the CBT have to finance 350,000 SMEs and cre- adopted a number of unconven- ate 4 million new job opportuni- tional measures to deal with li- ties over a period of four years. It quidity shortages in the aftermath gives priority to exporting firms of the uprisings. In Egypt, the CBE and those working in innovation. has focused on the provision of In Tunisia, the banking sec- foreign exchange liquidity to the tor’s structural liquidity short- banking sector to ease the severe ages triggered several responses. liquidity squeeze that was caused In 2011, the BCT extended the by the depletion of reserve and maturity of its auctions, which overly tight exchange rate policy. are its principal instrument for in- This has been exacerbated by the tervention on the money market, implementation of major infra- from seven days to one month, structure projects with sizeable and then to three months (IMF, import requirements. Starting 2013b). Through this measure, December 2012, the CBE has held the CBT has injected liquidity regular foreign exchange auctions to refinance commercial banks for banks and, since April 2013, (mainly through auctions and exceptional auctions. Between later through foreign exchange December 2012 and June 2016, swaps), with the aim of maintain- the CBE intervened through regu- ing overall financial stability and lar auctions, with a total amount addressing failures in the mar- of US$ 19 billion. In addition, it ket for interbank liquidity. Un- intervened 9 times through ex- like unconventional measures in ceptional auctions between April advanced economies, these op- 2013 and March 2016, with a total erations cannot be considered as value of US$ 7.4 billion. Commer- quantitative easing because they cial banks were required to apply did not involve large-scale asset with the amount of their clients’ purchases financed by the issu- outstanding import needs for spe- ance of central bank money. Nor cific goods (like staple commodi- did they attempt to directly affect ties, capital goods spare parts; asset prices. intermediate production com- Bank refinancing operations ponents and raw materials; and were quite significant in 2011 and Chapter 2 | 53

2012, leading the CBT to impose cyclical (as indicated by a posi- an additional 50 percent reserve tive correlation between policy requirement on increases in con- rates and inflation), suggesting sumer credit in 2012, though this that central banks have reduced was reduced to 30 percent in policy rates when inflation was March 2013 and removed in 2015. declining and vice versa. How- Nevertheless, banks continue to ever, their attempts at nominal rely on central bank refinanc- tightening have not been suffi- ing. By mid-2016, an amount of ciently transmitted to real interest around TND 25 billion was ex- movements (which matter more), tended to banks through refinanc- which have remained pro-cycli- ing of banks (excluding OMOs in cal. In Egypt and Morocco, mon- the amount of about TND 2.3 bil- etary policy has been pro-cyclical lion). in both nominal and real terms, What was the effect of monetary which reflects CBM’s monetary policy on inflation, credit and stimulus to ease growth in a con- growth? In principle, an accom- text of decelerating inflation. In modative stance (interest rate the case of Egypt, interest rates cuts) would support growth but were increased while inflation could spur inflationary pressures. was declining (November 2011), Meanwhile, a restrictive stance and decreased despite mounting (interest rate hikes) would curb inflationary pressures (second inflation at the expense of harm- half of 2013). ing credit, investment and there- The pro-cyclical response to fore growth. inflation (in real terms) in Jordan A formal approach to evaluate and Morocco could also reflect the monetary policy stance is to the fixed peg arrangements and estimate the central bank reaction thus the inability to raise interest function that relates the interest rates while international interest rate response to inflation and out- rates are kept near the ZLB. This put (Taylor, 1993). However, we would also means their central use a less quantitative measure to banks will soon have almost no get an indication of the cyclicality room to further reduce short- of monetary policy. More specifi- term interest rates, with the rates cally, we measure the correlation standing respectively at 1.5 and between the cyclical component 2.25 percent, respectively. of inflation and the cyclical com- South-Med countries did not ponent of the policy rates. To witness exceptional inflation- get an indication of the response ary pressures during the politi- to output, we also compute the cal transition, a conclusion that average correlation between the echoes with Khandelwal and cyclical component of the policy Roitman (2013). In fact, inflation rate and real GDP. The results are generally eased from previous reported in Table 2.8 and elabo- years (resulting from high com- rated below. modity prices) to lower levels Starting with inflation, the after the uprising (except in Tuni- nominal response of the central sia) on the backdrop of subdued banks in Jordan and Tunisia after economic activity among other the uprising has been counter- factors. 54 | Macroeconomic Policy Response

Table 2.8: Monetary policy management and macroeconomic outcomes Egypt Jordan Morocco Tunisia

2005- 2011- 2005- 2011- 2005- 2011- 2005- 2011- 2010 2015 2010 2015 2010 2015 2010 2015 Counter-cyclicality (inflation) Cyclicality with nominal rates 64.3 -38.9 38.8 81.8 31.5 -36.7 44.6 9.3 Cyclicality with real rates -89.8 -96.3 -91.8 -84.2 -91.6 -69.3 -86.6 -73.4 Counter-cyclicality (output) Cyclicality with nominal rates 11.6 -32.6 1.0 8.4 17.4 -55.8 41.6 80.3 Cyclicality with real rates -43.9 -32.0 -5.9 -8.6 34.5 -10.5 25.2 8.8 Monetary policy outcomes CPI inflation 11.1 9.5 5.4 3.1 2.0 1.2 3.3 4.9 Deposit growth 12.9 15.6 11.7 7.7 Credit growth 8.7 22.7 17.4 8.6 15.3 5.8 11.3 10.3 Loan to deposit ratio 54.2 47.6 66.4 66.7 Note: The cyclical component is derived from the average of the estimated trend using a HP filter with lambda 100 and 6.25. For inflation, a positive correlation is indicative of countercyclical monetary policy, suggesting that policy rates were raised when inflation was high (or vice versa), while a negative correlation indicates pro-cyclical monetary policy, suggesting that policy rates were not used to pre-emptively curb inflation. For growth, a positive correlation is indicative a countercyclical stance while a negative correlation indicates a pro-cyclical one. Source: Calculated by the authors based on data from national central banks and World Development Indicators, World Bank.

In Egypt, where inflation has to rising inflation, reflecting the been in double-digits over the pass-through from devaluation to past decade, monetary policy has domestic prices (figure 2.11). maintained negative real interest In Jordan and Morocco, infla- rates. The CBE’s restrictive stance tion has been below 5 percent, de- in late 2011 subdued inflationary spite expansionary spending and pressures, though it remained at energy subsidy phase-outs (in 7 percent in the following year. November 2012 for Jordan, and Subsequently, the interest rate September 2013 for Morocco). hikes following the elimination Jordan and Morocco’s monetary of energy subsidies in July 2014 accommodating stance, yet re- did not prevent inflation from in- strictive in real terms, has been as- creasing to 11 percent thereafter. sociated with declining inflation. In Tunisia, inflation has remained As for liquidity, credit growth mostly in the low single digits, significantly slowed down in Mo- temporarily edging up to close to rocco and Jordan to well below 10 7 percent in 2013. Inflation spiked percent in the period 2011-2015, once more following the subsidy compared with pre-crisis peri- reform in May 2014. Furthermore, ods (Table 2.8). By contrast, credit exchange rate devaluations in growth in the same two periods both countries have contributed declined very little in Tunisia Chapter 2 | 55

Figure 2.11: Exchange rate change and inflation in Egypt and Tunisia, 2011-2016 (percent) Egypt Tunisia

Source: data from national central banks.

(from 11.3 percent in 2005-2010 to suggests that policy cuts to favour 10.3 percent in 2011-2015), while growth have often led to even going up significantly in Egypt higher inflation. In fact, inflation (from 8.7 percent in 2005-2010 crept up again to 13 percent as the to 22.7 in the period 2011-2015). real monetary stance loosened in The significant increase in credit 2013. In both countries, any po- growth in Egypt in the latter peri- tential positive effects from inter- od was due to a buoyant credit to est cuts to favour investment and the government, which continued growth would be significantly to creep up since 2011 to reach 55 limited by policy uncertainty, se- percent of GDP in 2014. curity concerns (as well as foreign Turning to output stabilization, currency shortages in Egypt). the monetary policy response was Morocco’s monetary policy re- relatively weak after the uprising, sponse appears to be pro-cyclical except in Tunisia where the mon- despite monetary easing. More- etary policy was counter-cyclical over, a low inflation environment as indicated by the positive corre- has led to real rate increases. As lation between the policy rate and for Jordan, nominal attempts at growth. This confirms Tunisia’s monetary expansion have been monetary easing that associated supportive of growth, although the cyclical downturn in 2011, fol- as real rates have been increasing, lowed by policy tightening to ease they may not have supported in- credit conditions while economic vestment and growth. activity picked-up. In contrast, The exchange rate could have Egypt’s pro-cyclical response, in played a major role in demand both nominal and real terms, sug- stimulation in Egypt and Tunisia, gests that monetary policy has on the back of their currency de- not accommodated growth. This valuations. The latter could have could be due to policy tightening helped restore competitiveness taking place several times during and exports and eventually led to the cyclical downturn after the a self-correcting current account uprising. Moreover, the prevail- mechanism. However, devalua- ing high inflation environment tions could have contractionary 56 | Macroeconomic Policy Response

effects to the extent that imported Nominal tightening maintained intermediate goods are exten- negative interest rates, which sively used in the production of suggest a de facto accommodative exports, leading to high inflation monetary policy stance. As a re- that would reduce real income sult, inflation has remained high and consumption. in Egypt (in double-digits) and Moreover, because the CBT has increased in Tunisia. The CBE did not have a formal lender of and CBT also adopted liquidity- last resort (LOLR) facility (law ap- enhancing measures to boost proved in early 2016), these facili- banking sector liquidity. As these ties were provided through CBT measures were not effective, they normal liquidity functions with- engaged in direct liquidity injec- out a formal LOLR framework. tions to the banking sector. The This suggests that such funds CBE did so in the hope of easing were not appropriately secured a persistent foreign exchange li- by acceptable collateral and were quidity squeeze, while the CBT not limited to solvent but illiq- provided direct refinancing to uid banks. These measures may banks in the hope of maintaining also have increased the exposure financial stability. Negative rates of the CBT’s balance sheet to the have also fuelled credit growth in banking system.3 A central bank both countries. In Egypt, the gov- law to this effect was approved in ernment has been the main actor early 2016. to benefit from cheap credit at In summary, monetary policy negative rates, a situation that has management was complicated by contributed to the crowding out the pursuit of domestic objectives of the private sector. In Tunisia, moving in different directions. direct liquidity injections have Meanwhile, the monetary policy underpinned credit growth in a response depended on country- fragile banking sector. Tunisia’s specific circumstances. On the monetary policy response to out- one hand, the response of the cen- put stabilization has been more tral banks in Morocco and Jordan effective than Egypt’s as Egypt’s focused on conventional mea- high inflation environment sug- sures (interest rates and reserve gests that the policy cuts have of- requirements). In the absence of ten led to a deterioration in the in- overt inflationary pressures, their flation outlook. In both countries, central banks have been able to an effective impact of monetary put more emphasis on output expansions on output have been stabilization. Yet, on the back of constrained by the political and decelerating inflation, nominal social unrest and security threats. attempts at loosening monetary policy did not lead to declining real rates and therefore have not 2.5 Fiscal-monetary policy boosted credit conditions, which consistency remain significantly below pre- crisis levels. Having taken stock of fiscal and On the other hand, Egypt and monetary policy responses, we Tunisia resorted to conventional now question the extent to which and unconventional measures. the monetary policy stance has Chapter 2 | 57

been influenced by fiscal policy tion in order to cushion its con- considerations and the extent to tractionary effects. In Egypt 2015 which there has been consistency and Tunisia in 2014, fiscal con- in the policy mix. traction was compounded by a The first step is to characterise restrictive monetary stance. In the the policy mix. Table 2.9 reports case of Morocco, fiscal contrac- the fiscal impulse measure and tions in 2013 and 2015 have been the stance of monetary policy in accompanied by a neutral mon- South-Med economies after the etary policy. uprisings. As may be seen, the Moreover, in Egypt, fiscal policy mix has varied significant- dominance could have con- ly across economies and across strained monetary policy’s deci- time. South-Med authorities have sions to raise interest rates, which not taken coordinated measures was necessary to curb inflation. to stimulate aggregate demand. With lending policy rates of close For example, some fiscal expan- to 12.75 and T-bill rates of close to sions in Egypt (2012), Jordan 13 percent, monetary policy has (2011) and Tunisia (2012) were almost had no scope for further accompanied by a contractionary tightening without this leading to monetary policy. Others were ac- a significant ballooning in interest companied with monetary stimu- payments and, in turn, a deterio- lus: Egypt in 2014, Jordan in 2013 ration in fiscal sustainability. (See and 2014, Morocco in 2014, and figure 2.12 for the trend in over- Tunisia in 2011 and 2015. night CBE’s overnight lending More importantly, with the rates and the cost of T-bills). exception of Jordan in 2015 and In short, there is no evidence Morocco in 2012, fiscal adjust- of coordination between fiscal ment has not been accompanied and monetary policy. Monetary by monetary policy accommoda- policy has not provided the re-

Table 2.9: Fiscal policy response and changes in policy rates Egypt Jordan Morocco Tunisia

Year Change Change Change Change Fiscal in policy Fiscal in policy Fiscal in policy Fiscal in policy impulse rates impulse rates impulse rates impulse rates 2008 0.39 1.75 4.94 -0.75 0.63 0.25 -0.82 0.00 2009 1.08 -1.50 -0.44 -1.50 2.77 -0.25 1.58 -0.75 2010 -0.34 -0.75 -2.72 -0.50 2.15 0.00 -1.88 0.00 2011 0.70 0.00 4.48 0.25 2.26 0.00 2.03 -1.00 2012 1.51 1.00 -3.01 1.75 -0.18 -0.25 2.83 0.25 2013 0.35 0.50 3.09 -0.50 -1.45 0.00 1.73 0.25 2014 3.25 -1.50 0.39 -0.75 0.34 -0.50 -2.88 0.75 2015 -3.15 0.50 -6.53 -1.25 -0.36 0.00 1.00 -0.50 Note: Fiscal impulse: positive means fiscal policy is expansionary and negative means contractionary. Change in policy rates: increase in rates means restrictive monetary policy and a decrease means accommodative monetary policy. Source: Fiscal impulse calculated by the authors based on fiscal data extracted from IMF Article IV Consultation central government budget tables and policy rates data from national central banks. 58 | Macroeconomic Policy Response

Figure 2.12: T-bill rate and overnight CBE lending The objective of this chapter rate in Egypt (percent) was twofold. First, it intended to identify how governments used monetary and fiscal policies to address macroeconomic challenges. Second, it intended to assess the effectiveness of their response against the targeted objectives . In terms of their fiscal policy response, the shock translated into pressing popular demands on the government for jobs, higher wages and other social benefits. In response, governments offered politically-motivated current spending expansions to appease social discontent, thus deterring Source: Central Bank of Egypt fiscal management from the objective of consolidation. The cost of delaying consolidation was high quired easing during fiscal ad- though perhaps unavoidable. It justment in Jordan and Morocco. can be argued that fiscal manage- In Egypt, it has been subservient ment was successful in maintain- to fiscal objectives. ing social peace in some countries more than others. Once this goal was achieved, fiscal sustainabil- 2.6 Conclusion ity concerns rose to the forefront and consolidation measures were South-Med economies have man- implemented. In this context, fis- aged to preserve macroeconomic cal management prioritized the pres- stability despite a turbulent socio- ervation of macroeconomic stability political context, spillovers from over growth. regional conflict, and numerous At the same time, the upris- exogenous shocks, including ter- ings resulted in a sudden drying ror attacks. However, critical eco- up of foreign exchange earnings, nomic challenges remain, mainly which had an impact on the pri- a cyclical economic downturn mary monetary policy target, the and deteriorating macroeconomic nominal exchange rate. In re- imbalances. In this context, mac- sponse, South-Med central banks roeconomic management could relied on foreign exchange inter- only address one of the following ventions to stabilise the exchange challenges: resolving macroeco- rate. This policy could only be nomic imbalances or accommo- pursued as long as there were dating the effects of the shock by safe reserve buffers to finance restoring economic growth. The these interventions. In addition, choice was complicated by in- monetary policy paid attention to tense and turbulent sociopolitical other domestic objectives. Egypt conditions in the wake of the up- attempted to control persistent in- risings that began in 2010. flationary pressures, while Tuni- Chapter 2 | 59

sia targeted banking sector liquid- though insufficiently, thus credit ity. In Jordan and Morocco, where growth remains significantly be- inflation was in low single digits, low pre-crisis levels. The preva- central banks could ease mone- lence of negative interest rates tary policy to stimulate aggregate in Egypt and Tunisia has main- demand. Most central banks used tained a de facto accommodative conventional measures (interest monetary policy stance, which rates and reserve requirements) may have fuelled credit growth, to boost credit conditions. In ad- sustained inflationary pressures dition, the central banks in Egypt and generally impeded the ef- and Tunisia used unconventional fectiveness of monetary policy. In liquidity-enhancing measures to Egypt, the government has been boost banking sector liquidity. the main actor to benefit from The problem is that South- bank credit at negative real rates, Med authorities have not adopted a situation that has contributed to a coordinated effort to stimulate the crowding out of the private aggregate demand. Moreover, sector. In Tunisia, the CBT’s con- their fiscal and monetary policy tinuous refinancing to banks has has not always been consistent. underpinned credit growth in a In terms of the macroeco- fragile banking sector. Finally, the nomic effects, fiscal management CBE’s attempt to ease the persis- (whether consolidation or stimu- tent foreign exchange liquidity lus) has not been very effective ei- squeeze has not been effective. ther in stimulating growth, which Apart from the above issues, it has remained below potential, is worthwhile to note that access or in encouraging private invest- to external finance has provided ment. This is because crisis-driv- a temporary solution to plug fi- en stimulus packages have not nancing gaps, whether internal been designed to boost aggregate or external. However, the support demand, but to contain social dis- has not been accompanied by content. Despite consolidation ef- the necessary reforms to correct forts, fiscal vulnerabilities remain, structural imbalances, and has leading many South-Med econo- not proved effective. The situa- mies to renew IMF support. tion was more favorable for econ- The response by the central omies such as in Tunisia, which banks in Egypt and Tunisia to enjoyed relatively better initial preserve exchange rate stability conditions (low fiscal deficits and has been too costly and ineffec- debt as well as safe foreign ex- tive. It has focused too much on change buffers and low inflation). reserve depletion and too little on The concepts and issues dis- exchange rate flexibility in the face cussed in this chapter are indica- of a prolonged and lasting shock. tive of the complexities involved Otherwise, the effectiveness of in macroeconomic policymaking. monetary policy with respect to The analysis clearly shows that other objectives varies from one there can be a variety of policy country to another. On the one packages to achieve different hand, on the back of decelerating economic policy objectives. Each inflation, monetary policy in Jor- choice will entail policy dilem- dan and Morocco has been eased, mas, trade-offs and costs. The 60 | Macroeconomic Policy Response

real choice, though, is between view H1-FY14” making tough political decisions Batini, N., L. Eyraud, L. Forni and that address the root causes of A. Weber (2014) “Fiscal Multi- the crisis once and for all, with- pliers: Size, Determinants and out postponing—once more—the Use in Macroeconomic Projec- necessary measures with the false tions”, International Monetary hope that these problems will Fund Technical Notes and Man- self-resolve. uals. Blanchard, O. and D. Leigh (2013) “Growth Forecast Errors and Notes Fiscal Multipliers”, IMF Work- ing Paper No. 13/1, January. 1. The authors of this chapter would Cecchetti, SG., MS. Mohanty, and like to thank Diaa Noureldin for F. Zampolli (2011) “The real insightful discussions and HC Re- effects of debt”, BIS Working search for sharing market data on the Egyptian black market exchange Papers No. 352, September. rate. Chudik, A., K. Mohaddes, MH. 2. This section draws on Batini, Ey- Pesaran, and M. Raissi (2015) raud, Forni, and Weber (2014). “Is There a Debt-threshold Ef- 3. This facility should be reserved: (i) for solvent but illiquid banks; (ii) be fect on Output Growth?”, IMF structured so that there is incentive Working Paper No. 15/197, for banks to first seek funding from September. all other sources; and (iii) be more Frankel, J., C. Vegh and G. Vu- expensive than other central bank operations (IMF, 2012). letin (2012) “On graduation from fiscal procyclicality”, Journal of Development Econom- References ics, vol. 100, Issue 1, January, Abdih, Y., P. Lopez-Murphy, A. pages 32-47. Roitman and R. Sahay (2010) Friedman, M. (1953) “The meth- “The Cyclicality of Fiscal Poli- odology of positive econom- cy in the Middle East and Cen- ics” in “Essays in positive eco- tral Asia: Is the Current Crisis nomics” edited by Friedman, Different?”, IMF Working Pa- M., University of Chicago Press. per No. 10/68, March. Friedman, M. (1968) “The role of Alesina, A. and R. Perotti (1995) Monetary Policy”, The Ameri- “Fiscal Expansions and Ad- can Economic Review, Vol. justments in OECD Coun- LVIII, No.1, March. tries”, NBER Working Paper Giavazzi, F. and M. Pagano (1990) Series No. 5730, August. “Can Severe Fiscal Contrac- Alesina, A. and S. Ardagna (2010) tions Be Expansionary? Tales “Large Changes in Fiscal Pol- of Two Small European Coun- icy: Taxes versus Spending” tries” in NBER Macroeconomics in “Tax Policy and the Econ- Annual edited by Blanchard, omy” edited by Brown, JR., O. and S. Fischer, MIT Press. University of Chicago Press, Vol. Gervais, O., L. Schembri and L. 24, pages 35-68. Suchanek (2016) “Current Arab Republic of Egypt Ministry account dynamics, real ex- of Finance (MOF) (2014) “Mid- change rate adjustment, and Year Economic and Fiscal Re- the exchange rate regime in Chapter 2 | 61

emerging-market economies”, Jordan” Journal of Development Econom- ______(2014a) “Algeria 2014 Ar- ics, Vol. 119, March, pages 86- ticle IV Consultation – Staff 99. Report; Press Release; and Hanieh, A (2015) “Shifting Pri- Statement by the Executive orities or Business as Usual? Director for Algeria” Continuity and Change in ______(2014b) “Jordan 2014 Ar- the post-2011 IMF and World ticle IV Consultation, Third Bank Engagement with Tu- and Fourth Reviews Under nisia, Morocco and Egypt”, the Stand-By Arrangement, British Journal of Middle Eastern Request for Waivers of Non- Studies, Vol. 42, Issue 1, De- observance of Performance cember, pages. 119-134. Criterion – Staff Report; Press Heller, PS., RD. Haas and AS. Releases; and Statement by the Mansur (1986) “A Review of Executive Director for Jordan” the Fiscal Impulse Measure”, ______(2013a) “Tunisia: Request Occasional Papers of the Inter- for a Stand-By Arrangement – national Monetary Fund No. 44, Staff Report, Press Release on May. the Executive Board Discus- International Monetary Fund sion; and Statement by the Ex- (IMF) (2016a) “Morocco Re- ecutive Director for Tunisia” quest for an Arrangement Un- ______(2013b) “Monetary Issues der the Precautionary and Li- in the Middle East and North quidity Line and Cancellation Africa Region: A Policy Imple- of the Current Arrangement mentation Handbook for Cen- – Press Release; Staff Report; tral Bankers” and Statement by the Execu- ______(2012a) “Chapter 3: The tive Director for Morocco” Good, the Bad and the Ugly: ______(2016b) “Tunisia Request 100 Years of Dealing with for an Extended Arrangement Public Debt Overhangs” in Under the Extended Fund Fa- “World Economic Outlook: cility – Press Release; Staff Re- Coping With High Debt and port; and Statement by the Ex- Sluggish Growth” ecutive Director for Tunisia” ______(2012b) “Tunisia: 2012 Ar- ______(2015a) “Morocco Ex Post ticle IV Consultation – Staff Evaluation of Exceptional Ac- Report; Public Information cess Under the 2012 Precau- Notice on the Executive Board tionary and Liquidity Line Discussion; and Statement by Arrangement – Press Release; the Executive Director for Tu- Staff Report and Statement by nisia” the Executive Director for Mo- ______(2010) “Chapter 3: Will it rocco” hurt? Macroeconomic Effects ______(2015b) “Jordan Seventh of Fiscal Consolidation” in and Final Review Under the “World Economic Outlook: Stand-By Arrangement and Recovery, Risk, and Rebalanc- Proposal for Post-Program ing” Monitoring- Press Release; Khandelwal, P. and A. Roitman Staff Report; and Statement (2013) “The Economics of Po- by the Executive Director for litical Transitions: Implica- 62 | Macroeconomic Policy Response

tions for the Arab Spring”, Annex 1.2: IMF Working Paper No. 13/69, Fiscal impulse measure March. Krugman, P. (2010) “Myths on Assessing the fiscal policy stance Austerity”, The New York is a challenging task because Times, July 1. there is a two-way relationship Kumar, MS. and J. Woo (2010) between fiscal variables and out- “Public Debt and Growth”, put. Changes in the former can IMF Working Paper No. 10/174, be either the cause or the result July. of changes in economic activity. Ostry, JD, AR. Ghosh, and M. On the one hand, automatic sta- Chamon (2012) “Two Targets, bilizers respond to the business Two Instruments: Monetary cycle. On the other, discretionary and Exchange Rate Policies spending hikes or tax cuts may be in Emerging Market Econo- implemented by governments in mies”, IMF Staff Discussion order to stimulate economic activ- Note No. 12/01, February. ity. In this context, the “cyclically Perotti, R. (2012) “The “Austerity neutral balance” is the budget Myth” Gain Without Pain” in balance when output is close to “Fiscal Policy After the Finan- potential. Any excess of the actual cial Crisis” edited by Alesina, balance from the cyclically neu- A. and F. Giavazzi, University tral balance is considered expan- of Chicago Press, pages 307-354. sionary. This budget is derived Phelps, ES. (1968) “Money-wage from the actual budget by assum- dynamics and labor-market ing that nominal tax revenues equilibrium”, Journal of Po- are unit elastic to actual nominal litical Economy, vol. 76, part 2, GDP and government expendi- August, pages 678-711. tures are unit elastic with respect Reinhart, CM. and KS. Rogoff to nominal potential output. Gov- (2010) “Growth in a Time of ernment expenditure is cyclically Debt”, American Economic Re- contractionary (expansionary) if view, vol. 100, no. 2, May, pag- it increases less than (more than) es 573-578. proportionately with potential Sargent, T. and N. Wallace (1981) output. It can be neutral if it in- “Some Unpleasant Monetar- creases proportionately with ist Arithmetic”, Federal Reserve potential output. Similarly, with Bank of Minneapolis. respect to revenues, a more than Taylor, JB. (1993) “Discretion ver- proportionate change in revenue sus policy rules in practice”, relative to output is considered Carnegie-Rochester Conference contractionary and vice versa. Series on Public Policy 39, pag- Along the lines explained es: 195-214. above, we compute a fiscal im- World Bank (2014) “Jordan Eco- pulse (FI) indicator that measures nomic Monitor - Resilience the excess between the actual and Turmoil” budget and the cyclically neutral. This methodology is described in Heller et al (1986) and computes the FI as follows: Chapter 2 | 63

FI = –∆B + t0∆Y – g0∆Yp tempt to determine the coun- FI is the absolute measure of ter-cyclicality of different bud- the fiscal impulse (FI) get components (revenues and T = Government revenues spending), which relates the abil- G = Government expenditures ity of fiscal policy to combat busi- ∆B = first difference in the ac- ness cycle fluctuations. Accord- tual budget balance (T-G) ing to Abdih et al. (2010), fiscal t0 = T0 /Y0 is the base year policy is: revenue ratio • Pro-cyclical if and only g0 = G0 /Y0 is the base year if government spending expenditure ratio increases (decreases) in ∆Yp = Potential output in good (bad) times. nominal prices (first difference) • Counter-cyclical if and ∆Y = Actual output in nominal only if government spend- prices (first difference) ing decreases (increases) The subscript “0” refers to the in good (bad) times. base year values of any variable. • Acyclical if and only if For the purposes of this study, government spending we have adopted a “rolling” base remains constant across year whereby the figures at time time. t are sequentially taken as base A similar logic applies to the year figures for the FI at time t +1. changes in revenues in the op- An actual deficit in excess of the posite direction. The typical ap- cyclically neutral deficit (defined proaches to estimating the cy- as t0∆Y – g0∆Yp) is considered as clicality of fiscal policy involve expansionary, relative to the base the regression of fiscal variables year fiscal stance, and the FI is (fiscal balance or spending) on positive. Hence, a positive (nega- an indicator of the economic cy- tive) measure of MFI will imply a cle (output gap or growth). We more expansionary (contraction- use a simpler method following ary) fiscal stance compared to Frankel et al. (2012), according to the previous year. This measure which we compute the correla- of fiscal stance does not distin- tions between the cyclical compo- guish whether the stance stems nents of real government expen- from revenue or spending. It also diture and real GDP. A positive does not distinguish whether the (negative) correlation indicates change in results from the impact pro-cyclical (countercyclical) fis- of discretionary actions or auto- cal policy. The results are shown matic stabilizers in the budget. in the table below for each of the five South-Med countries. Beside showing the average correlation Annex 2.2: between the cyclical components Measuring counter-cycli- of real government revenue and spending and real GDP before cality of fiscal policy and after 2010, the table below also reports several other macro- To assess the macroeconomic ef- economic variables. fects of fiscal management in South-Med economies, we at- 64 | Macroeconomic Policy Response

Table 2.10: Fiscal management and macroeconomic outcomes Algeria Egypt Jordan Morocco Tunisia

2002- 2011- 2002- 2011- 2002- 2011- 2002- 2011- 2002- 2011- 2010 2015 2010 2015 2010 2015 2010 2015 2010 2015 Fiscal cyclicality Cyclicality of revenues 0.36 -0.53 0.62 -0.46 -0.31 0.93 0.17 -0.64 -0.26 0.46 Cyclicality of revenues excluding grants 0.36 -0.53 0.64 0.28 0.53 0.42 0.18 -0.57 0.64 0.48 Counter-cyclicality of current primary spending -0.61 0.19 0.24 -0.84 0.07 -0.08 -0.61 -0.36 -0.32 0.37 Counter-cyclicality of capital spending -0.45 0.89 0.57 0.35 0.01 0.76 -0.19 -0.29 0.26 -0.76 Cyclicality of total spending -0.65 0.77 0.39 -0.70 0.15 0.71 -0.57 -0.40 -0.16 0.13 Real Economy Real GDP growth 3.96 3.29 5.09 2.50 6.44 2.73 4.70 3.97 4.36 1.50 Potential growth 3.49 3.14 5.13 2.74 6.43 2.83 4.78 3.72 4.37 1.84 Unemployment rate 15.56 10.53 10.08 12.40 13.84 12.39 10.20 9.36 13.39 16.25 Gross fixed capital formation (% change) 10.24 6.27 7.21 0.53 8.87 1.78 7.27 2.18 6.19 1.86 Gross fixed capital formation, private sector (% change) n/a n/a 17.95 8.93 17.91 8.24 n/a n/a n/a n/a Claims on central government (% change) 37.40 -18.43 15.85 35.56 45.81 16.90 0.15 13.36 7.80 23.53 Domestic credit to private sector (% change) 21.50 14.81 8.28 10.10 12.73 6.42 12.22 3.85 9.42 9.06 Note: All fiscal variables are in real terms. They were deflated by the GDP deflator. Cyclical components were calculated based on the gap between actual and potential measures, the latter calculated using the Hodrick- Prescott filter (lambda=6.25). Source: Calculated by the authors based on World Economic Outlook, IMF and World Development Indicators, World Bank databases. CHAPTER 3

Social Policy Response

3.1 Introduction consumption tells us little about income distribution, let alone In addition to their impact on wealth distribution. Third, no macroeconomic performance, the single indicator captures the com- uprisings have also had adverse plexity of social conditions, which social effects. Poverty, inequal- makes any assessment of social ity and unemployment appear to development difficult, especially have persisted or deteriorated in if individual indicators move in the wake of the uprisings. Ironi- opposite directions. Finally, there cally, social justice was one of the is a time lag between the timing rebels’ main slogans. This chap- of reforms and the time within ter seeks to ascertain whether re- which their effects are felt by the sponses to social demands in the population, for example in such wake of the uprisings were ap- reform areas as education. Not- propriate and sustainable. More withstanding these limitations, specifically, the chapter will at- some insights may still be gained tempt to answer the following from pulling together as many questions: pieces of the puzzle as possible. • What happened to social To place the analysis in a his- development in South- torical perspective, South-Med Med countries in the wake countries experienced a shift in of the uprisings? their social contract in the late • How did different gov- 1980s and early 1990s; from a ernments attempt to im- “statist” to a “pro-market” social prove social conditions? contract. At the expense of over- • Was the response appro- simplification, under the old con- priate and sustainable? tract the state provided citizens This investigation is ham- with various economic benefits, pered by four factors. First, data but denied them political partici- on social outcomes is not read- pation. Under the new contract, ily available in South-Med coun- the state gave way to market forc- tries. Second, the results based on es and allowed a greater role for available data are not always easy the private sector, but only con- to interpret; for example, data ceded a limited degree of politi- on inequality using household cal participation. Across the two

65 66 | Social Policy Response

contracts, however, the relation- conditions, interim governments ship between the rulers and their adopted a variety of measures. citizens retained the main fea- Most notably: tures of the so-called authoritar- • They increased commod- ian bargain (see Desai et al, 2007). ity subsidies and began Protests were suppressed and to phase out energy sub- despite some political liberaliza- sidies, along with the in- tion, the region continued, until troduction of cash transfer 2010, to score poorly on such gov- programs. ernance indicators as voice and • They increased pensions, accountability. The spark of mass minimum wages and in uprisings in Tunisia, in December some cases government 2010, ignited massive uprisings in employment. several neighboring countries. • However, they made Pro-market reforms produced modest progress on initia- decent economic growth rates, tives to empower citizens which were associated with a re- through reforms, for ex- duction in absolute poverty. The ample, of more progres- 5 South-Med countries in our sive taxation and the pro- sample grew on average by 3.7 vision of quality health, percent between 1990 and 2000, education and infrastruc- surging to close to 5 percent (4.8 ture (such as energy, wa- percent) during the decade of ter, and transport) to de- 2000-2010. The problem lies in prived groups. the frequently made claim that While these measures must benefits were not shared widely. have been beneficial to those who This era saw a significant rollback received them, they were partial, of the State from the provision myopic and unsustainable. Par- of social services and the rise of tial because they did not build on preferential treatment in favour a coherent view of the world that of politically connected firms. connects economic growth with Both phenomena led to rising dis- poverty reduction and a more satisfaction among increasingly egalitarian society. They were young and educated populations myopic because they were select- and the rise of crony capitalists ed to produce immediate results, (Diwan et al, 2016 and Reijkers often at the expense of public in- et al, 2014). The share of wages vestment and reforms of health, in some national incomes shrank; education or the tax regime. And equal opportunities diminished they were unsustainable because (e.g., Assaad et al, 2016) and per- they pushed budget deficits and ceptions of inequality worsened. public debts beyond safe bound- Hence the uprisings and the fall aries. As a result, several govern- of Ben Ali and Mubarak in Tuni- ments, including Jordan, Tunisia sia and Egypt, respectively. and more recently Egypt, have In the years following the had to seek IMF support to cover uprisings (2010-2015), economic growing financing gaps, earn in- growth dropped to 2.9 percent vestors’ confidence and carry out and macroeconomic balances short term corrective measures. deteriorated. To improve social More broadly, it is not clear Chapter 3 | 67

whether the rulers in South-Med addition to these opposing forces, countries have changed their bar- social policies cannot be predicted gain with citizens after the upris- while abstracting from the nature ings. Egypt and Tunisia have cer- of the prevailing political regime tainly adopted new constitutions and the distribution of power and conducted multiple elections among different interest groups of presidents and parliaments in in society. the span of a few years. Morocco In this section, an attempt is and Jordan have modified their made to draw a picture of what constitutions to allow more par- has happened to social develop- ticipation. Although the new con- ment before and after the upris- stitutions contain many features ings in the 5 South-Med countries, of liberal democracy, they are data permitting. The focus will be yet to be tested in the day-to-day on three particular dimensions: practice of politics, let alone be (1) the nexus of growth, poverty fully integrated in the body poli- and employment, (2) human de- tics. It is likely that the transition velopment and inequality, and (3) process will take a few rounds of voice and accountability. elections, and even longer for a new bargain to materialise. The rest of the chapter is orga- The nexus of growth, employ- nized as follows. The next section ment and poverty traces the evolution of a number In principle, higher economic of social development indicators growth is necessary for job cre- in South-Med countries. Section 3 ation and poverty reduction. reviews the responses of different Given that economic growth is a governments to social demands. function of factor accumulation Section 4 evaluates the actions and productivity improvements, taken thus far. Section 5 draws an increase in growth means that some conclusions. more labour is being deployed. The intensity of job creation depends, however, on the technol- 3.2 Social development ogy used, the pool of available skills and the characteristics of the The impact of the uprisings on sectors experiencing expansion. social development is hard to pre- All three dimensions can be influ- dict a priori. Two forces seem to enced by policies through incen- operate in opposite directions. On tive structures that may or may the one hand, the deterioration in not promote labour-intensive economic performance associated technology, acquiring the skills with political instability leaves most in demand in the market governments with fewer resourc- place, and the nature and speed es to spend on social develop- of structural transformation to- ment. On the other hand, the upward higher productivity sectors. risings have been driven in part With respect to poverty, even by dissatisfaction with economic moderate increases in average disparities, which means that in- incomes per capita can lead to terim governments face pressure proportionate reductions in pov- to respond to these demands. In erty as long as inequality does 68 | Social Policy Response

not increase and, even more so, earlier period, albeit to varying if inequality decreases. The main degrees. The biggest losses, in or- channel here is employment. In der, were in Jordan, Tunisia and addition, pro-poor policies can Egypt. Morocco and Algeria also play a role in ameliorating pov- experienced a loss in output, but erty in two ways. The first is by amounting to less than one per- establishing effective safety nets centage point. through such programs as cash Taking population growth transfers (conditional or not), into account, figure 3.2 shows the health and unemployment insur- average per capita growth rates ances and pension reforms. The of South-Med countries between second involves the empower- 2002-10 and 2010-14. On average, ment of citizens, for example, per capita growth rates declined through more progressive taxa- from 3.1 percent during the ear- tion and the provision of qual- lier period to 1.0 percent in the ity health care, education and post-uprising period. The decline infrastructures (such as energy, occurred in all countries, but most water, and transport) to deprived alarmingly in Egypt, Tunisia and groups. Jordan. In Egypt, the drop was The above generalizations so significant that Egyptians saw take on distinct characteristics their per capita income decline, during transitions. As noted ear- although very modestly. lier, political instability leads to Correspondingly, the capacity a slowdown in economic activ- of these economies to create new ity and, by association, a rise in jobs diminished. Unemployment unemployment and poverty. Ac- rates shot up in 2011, in Egypt cording to Khandelwal and Roit- and Tunisia, by more than 3 and man (2013), unemployment rates 5 percentage points, respectively tend to rise during political insta- (figure 3.3). Although unemploy- bility, by about 1–1.5 percentage ment among women was much points on average during the first higher than for men in both coun- two years after the start of politi- tries, it was male unemployment cal unrest, and it takes between 4 that bore the brunt of the crisis to 5 years for unemployment to in Egypt. In Tunisia, both male return to pre-crisis levels. With and female unemployment in- the rise of unemployment, pover- creased significantly. On a more ty can be expected to rise as well. positive note, the unemployment To explore what happened to figures appear to have stopped employment and poverty in the rising since 2013 in Egypt and post-uprising period in the South- Tunisia, although they remain Med countries, we first looked at well above pre-crisis levels. In the evolution of economic growth Algeria, Jordan and Morocco, before and after the uprisings. unemployment went down after Comparing the growth rates in 2010, though it remains close to the 2000s with those during the 10 percent. In particular, the fall 5 years after the uprisings, fig- in Algerian unemployment has ure 3.1 shows that all South-Med been outstanding over the past countries experienced a loss in decade, down from 24 percent to economic growth relative to the 10 percent. Chapter 3 | 69

Focusing on youth unemploy- Figure 3.1: Differences in average GDP growth ment, the problem is more acute rates between 2010-15 and 2000-10 in Egypt and Tunisia than in the other countries (figure 3.4). In 2011, youth unemployment increased in these two countries by 7 and 13 percentage points. Sub- sequently, youth unemployment in Egypt crept up to 33 percent in 2013. In Tunisia, it attained 31 percent in 2013. Jordan showed Source: World Economic Outlook Database, IMF. no great differences with the first two countries. In Algeria and Figure 3.2: Average annual growth in per capita Morocco, youth unemployment GDP, 2002-10 and 2010-15 actually went down after 2010, albeit remaining in the vicinity of 10 percent. Given that economic growth declined in the post-uprising period, it is to be expected that poverty would also have risen. Unfortunately, detailed information on the evolution of poverty Source: World Development Indicators, World Bank. in South-Med countries after the uprisings is scarce, if not inex- Figure 3.3: Unemployment rates, 2003-2014 istent. One exception is Egypt, where the percent of poor people rose from 19.6 percent in 2005 to 26.3 percent in 2013. No similar information is available for the other countries. However, if the trend in Egypt were to be used as a guide, poverty might have risen in recent years in Tunisia and Jordan, given their high loss of growth in the latter period. Before the uprisings, the information Source: World Development Indicators, World Bank. provided in table 3.1 suggests that Tunisia managed to reduce Figure 3.4: Youth unemployment rates, 2003-2014 the percent of poor people by more than one half in a single decade (2000s). Morocco also made significant progress, reducing the percent of the poor from 15.3 percent in 2001 to 9.0 percent in 2007. While the discussion of poverty focuses on the least privileged groups in society, inequality focuses on the relative distribution Source: World Development Indicators, World Bank 70 | Social Policy Response

of economic outcomes within or wages relative to GDP over the across countries. Beside inequal- period 2000-2014. The most strik- ity in outcomes, the concept of in- ing observation is the notable equality of opportunity has been difference in the shares of wages gaining grounds in recent times. across countries. Workers in Jor- The basic idea is that differences dan, Tunisia and Morocco enjoy in economic outcomes can be at- a much larger share of GDP than tributed to effort or circumstanc- their counterparts in Egypt and es. Variations due to effort are Algeria. The second observation ethically justified; those attribut- concerns the trend. Here, the big able to circumstances are not. Ac- decline in the share of wages rela- cordingly, policymakers ought to tive to GDP occurred in Egypt take measures to equalise circum- and Morocco. The biggest im- stances if they are to bring about a provements occurred in Algeria, more egalitarian society (for elab- followed modestly by Tunisia. oration, see, for example, Rawls, The data for Jordan is not avail- 1971 and Roemer, 1998). able beyond 2009. In South-Med countries, one The most commonly used crude measure of the evolution measures of inequality are Gini of inequality is the share of wages coefficients and the income in national incomes. A declining shares by different deciles of the share of wages in GDP implies population. These two indicators that capitalists are capturing a are reported in Table 3.1 for dif- larger fraction of national income. ferent years. The Gini coefficients Figure 3.5 shows the share of suggest that Morocco and Tunisia

Table 3.1: Poverty headcount, Gini coefficients, and decile income shares in South- Med countries, selected years Share of income/ consumption by deciles Poverty headcount ratio at national poverty Gini Country Year lines (% of population) index Lowest 10% Highest 10% Algeria 1995 - 35.3 2.9 26.9 Egypt 2005 19.6 32.1 - - 2011 25.2 31.4 3.4 26 2013 26.3 30.2 3.4 24.7 2015 27.8 - 4.2 25.0 Jordan 2006 13 37.7 3.3 27.5 2010 14.4 35.4 3.5 27.5 2013 12 34.8 2.8 26.7 Morocco 2001 15.3 40.6 2.6 32.2 2007 9 40.9 2.6 33.1 Tunisia 2000 32.4 40.8 2.4 31.6 2010 15.5 36.1 2.6 27 Sources: World Development Indicators, World Bank, ILO KILM database, CAPMAS Egypt and Economic Re- search Forum (ERF) data portal for household surveys for 2011 and 2013 for Egypt and 2013 for Jordan. Chapter 3 | 71

are more unequal than Egypt and Figure 3.5: Labor income share in GDP, 2000-2014 Algeria. The trend over time is generally positive, although the changes are relatively modest. As for the income shares of the richest and lowest income deciles, the picture is consistent across countries. In all cases and over time, the income share of the richest 10 percent of the population ranges between one quarter and one third of national incomes, while the poorest 10 percent only receive between 2 and 3 percent of Source: ILO - Global Wages Report data and National Accounts national incomes. data from: Algeria National Office of Statistics, Egypt Ministry The above conclusions must of Planning, Morocco Haut Commissariat au Plan, and Tunisia be taken with a grain of salt, how- National Institute of Statistics. ever. They draw on household surveys, which tend to miss top income data. They also underesti- ties, all studies find that circum- mate the extent of income inequal- stance variables such as parents’ ity because they often measure education and their occupation, expenditure rather than income. area of birth and gender made a More seriously, these are flows of difference in the fortune of their income/expenditure, not wealth, children. Hence, even if Gini coef- as a more appropriate measure ficients are not high in the region, according to Pikkety (2014). citizens of these countries do not With respect to inequality of enjoy equal opportunities in earn- opportunity, the evidence from ings, educational attainment or South-Med countries is scarce health outcomes. but growing. For example, El En- baby and Galal (2015) and Hass- ine (2012) estimated the extent to Voice and accountability which circumstances beyond the Voice and accountability, being control of individuals account- key features of the political sys- ed for variations in earnings in tem, can be viewed as the other Egypt. Salehi-Isfahani et al. (2014) side of the coin of social develop- explored how much circumstanc- ment. Where societies enjoy inclu- es accounted for inequality of op- sive political and economic insti- portunity in education in a num- tutions, the likelihood is high that ber of MENA countries. Assaad they will enjoy high and shared et al. (2012) explored inequality of prosperity (Acemoglu and Robin- opportunity in health outcomes in son, 2012, North, 1990). In South- the Arab world, including Egypt, Med countries, the demand for Morocco and Jordan. While the freedom was one of the main slo- data did not allow researchers gans of the rebels in Tahrir square to take into account a sufficient and other squares. Accordingly, it number of circumstances to ac- is important to find out whether count for variations in opportuni- progress has been made on this 72 | Social Policy Response

front in the post-uprising periods, begin with, all 5 countries, except or not. Tunisia in 2014, enjoyed lower Figure 3.6 shows the evolu- than average levels of voice and tion of voice and accountability accountability. Comparatively, for the 5 South-Med countries for Algeria and Egypt scored less the years 2000, 2010 and 2014. The well than Jordan, Morocco, and indicator reflects perceptions of Tunisia (except in 2010). Over the extent to which a country’s time, only Tunisia and Algeria citizens are able to participate in have done better, while Egypt, selecting their government, as Jordan and Morocco have done well as freedom of expression, worse. freedom of association, and free- If citizens are not satisfied dom of the media. It ranges from with their political systems or approximately -2.5 (weak) to 2.5 economic and social conditions, (strong) performance.1 The re- they have the option of protest- sults give cause for concern. To ing. To get a sense of the evolution of this mechanism over time, Figure 3.6: Voice and accountability indicator figure 3.7 shows the trend in all 5 countries over the period 1996 to 2015. The figure captures strikes and demonstrations, with each event weighted by its intensity (number of protesters) and duration (number of weeks). What the figure shows, unsurprisingly, is that protests peaked in all countries around 2010-2011 during the Arab Spring. By 2015, the level of protests subsided to the levels of the early 2000s. This trend does not necessarily mean that citizens Source: World Governance Indicators, World Bank. are happier with their conditions. It might very well be the result of Figure 3.7: Public protests indicator, 1996-2015 demonstration fatigue, suppres- sion by the state or a position of “wait and see”.

3.2 Response to social demands Interim governments in South- Med countries responded to the social demands of the rebels to varying degrees. Across countries, whether directly witnessing a dramatic political uprising or Source: African Economic Outlook calculations based on news verified by the press agencies (Marchés Tropicaux et Méditer- simply feeling the heat coming ranéens for 1996-2005, AFP and Reuters for 2006 onwards). from their neighbors, govern- Chapter 3 | 73

ments attempted to ameliorate Figure 3.8: Total public social protection expendi- the adverse effects of the transi- ture (percent to GDP) tion and in some cases to appease the public or certain interest groups. This section attempts to capture these responses by docu- menting the main actions taken by different governments under the headings of social protection and empowerment through social services.

Social protection Figure 3.8 shows the total ex- Note: The data for Algeria and Morocco were not available for 2010 and the last available date were for 2009. penditures on social protection Source: ILO Social Security Inquiry Database. in South-Med countries per the ILO definition, which comprises In terms of initiatives to re- 11 functions or policies: health, spond to social concerns in the disability, old age, survivors, ma- post-uprising period, a review of ternity, family/children, unem- the evidence suggests that most ployment, employment injury, countries took actions. Most no- housing, active labour market tably: programs and general protection • They initiated energy sub- against poverty and social exclu- sidy reforms as part of sion. The data compares 2010 and their fiscal consolidation 2011, or one year before the upris- effort, while simultane- ings and one year into the upris- ously adopting cash trans- ings. fer programs to reduce the Figure 3.8 shows that over- burden on the poor. all expenditure on social protec- • They also increased ex- tion was the highest in Egypt, penditure on certain so- although Jordan, Tunisia and Al- cial programs, including geria, to some extent, were not far food subsidies and hous- from Egypt. Morocco stands out ing. For example, the food as having spent only about half of subsidy in Tunisia was the amount spent by Egypt. Over increased by 11 percent in time, between 2010 and 2011, all the wake of the uprising countries increased their expen- (Achy, 2011; Chemingui ditures on social protection, ex- and Sanchez, 2011). cept Algeria. The most notable • Finally, most countries in- increases were in Tunisia and Jor- creased pensions, public dan, followed by Egypt, perhaps sector salaries and govern- reflecting the intensity of politi- ment employment. They cal movements. Public spending also revised the minimum on social protection as a share of wage upwardly, although GDP hardly changed in Morocco, this revision was only ap- while the level of expenditures plied to government em- declined in Algeria. ployees in Egypt. 74 | Social Policy Response

With respect to energy subsi- levels in fuel and electricity pro- dies and cash transfers, all South- duction by 2018/19. In parallel, Med countries were active in the the government introduced two post-uprising period. They initi- cash transfer programs. One for ated cash transfer programs and the elderly and disabled, which is moved gradually towards remov- called Karama. The other is a cash ing energy subsidies. This reform transfer program in support of initiative was consistent with the poor (called Takafol). The lat- the broadly supported empirical ter program is conditional upon finding that universal commod- child enrolment in school and ity subsidies tend to accrue dis- meeting health requirements. The proportionately to the non-poor, two programs were piloted on encourage arbitrage and distort 600,000 families in Upper Egypt, consumption and production. with a plan to cover 1.5 million In contrast, a targeted program households once fully rolled out involving cash transfers tends to (IMF, 2015a). These two programs benefit the poor in a more cost ef- supplement an existing social fective manner. solidarity pension system, which In Egypt, the government ini- covers 1.5 million households. tiated energy subsidy reforms as Similarly, Tunisia increased early as 2012. Between 2011 and petrol and diesel prices and elec- mid-2012, the price of gas was tricity tariffs on average by 7 per- increased by 30 percent for non- cent in 2012. This was followed energy intensive industries and by further increases of around 8 slightly more (33 percent) for percent the year after. In January energy intensive industries. In 2014, energy subsidies to cement January 2013, household electric- companies were reduced by 50 ity prices were increased by 16 percent, with a view to eliminat- percent, while the prices of natu- ing them fully by 2015. At the ral gas and fuel oil for electricity same time, electricity tariffs and generation rose by one-third. In natural gas prices were increased February 2013, the Government for medium and low-voltage con- increased fuel prices for vari- sumers by 10 percent, and by a ous industries from LE1,000 to further 10 percent in May 2014, LE1,500 per ton, and raised the along with the adoption of an au- price of gas for industrial use tomatic petrol price formula. To (with some exceptions, such as mitigate the negative effects of bakeries and electricity). The in- these measures, electricity tariffs crease in electricity prices was not for households consuming less applied to households consum- than 100 kwh/month were kept ing less than 50 kilowatts/month under control. The government (about one quarter of all consum- also introduced a new housing ers). In July 2014, there were fur- program for needy families and ther increases in energy prices, increased the tax deduction for both for households and commer- the poorest households (IMF, cial users. The increase varied be- 2014a). Early in December 2011, tween 40 percent and 80 percent. the government increased the Further reforms were planned monthly benefits for poor fami- to reach close to cost recovery lies and added 50,000 families Chapter 3 | 75

to those already receiving them, poor facing medical expenses), as thus reaching a total of 235,000 well as the introduction of new families (World Bank, 2014a). programs to support low income In Jordan, the government in- widows and physically disabled troduced a cash transfer program, individuals. Limits were imposed in mid-November 2012, to house- on fare increases against rising holds with annual earnings be- costs of fuel (IMF, 2015b). In par- low JD10,000. The amount of the allel, the Moroccan Government transfer was a sizeable JD70 per increased the retail price of diesel, household member/per month petrol and industrial fuel oil by for up to six members. This mea- 14-27 percent, thus saving 0.7 per- sure reached almost 70 percent cent of GDP by the end of 2012. of the population in 2013 (World This was followed in 2013, first, Bank, 2014b). The fiscal savings by partial indexation of the prices from the removal of fuel subsidies for diesel, petrol and industrial and their replacement with cash fuel to world prices and, second, transfers was around 1 percent- by hedging provisions to reduce age point of GDP (World Bank, the volatility arising from fluctua- 2013). With respect to electric- tions in the international price of ity subsidies, Jordan introduced oil. The prices for diesel were in- three sequential increases in tar- creased further by 9 percent, for iffs for richer households, large petrol by 5 percent, and for fuel domestic corporations and select- by 14 percent. Starting February ed economic sectors, including 2014, the subsidies on the domes- banks, telecommunications, ho- tic prices of petrol, diesel and in- tels and mining (IMF, 2014a and dustrial fuel oil were gradually 2014b). Prices at fuel pumps were removed. Transfers to the public liberalized. Subsidies for petrol, electricity company (ONEE) were diesel, kerosene and Liquefied Pe- gradually eliminated by 2017. troleum Gas (LPG) were reduced. The government also considered As a result, subsidies came down ways to gradually reduce subsi- from JD884 million (4 percent of dies on wheat, sugar, and butane. GDP) in 2012 to JD250 million by These efforts reduced the subsidy 2013 (Atamanov et al, 2015; IMF, bill from 6.6 percent of GDP in 2014a and World Bank, 2014b). 2012 to 3.6 percent in 2014 and to In January 2014, electricity tariffs an estimated 2.3 percent of GDP were further increased by up to in 2015 (IMF, 2015b and 2014a). 15 percent for selected consumers Turning to wages, pensions (IMF, 2014a). Further increases and government employment, followed for selected businesses Egypt and Tunisia increased their and households that consume public-sector employment in re- over 600 kilowatt-hours/month sponse to rising unemployment. (World Bank, 2014b). In Egypt, one million persons In Morocco, attempts to al- were recruited by the public sec- leviate the burden of reform on tor in the wake of the uprisings. the poor included the expan- In Tunisia, forty thousand public sion of benefits of existing social jobs were created (Subrahman- programs (such as support for yam and Castel, 2014). In the same school-age children and for the direction, all South-Med coun- 76 | Social Policy Response

tries increased public sector wag- union freedom and activities (See es. Egypt raised civil service sala- Kienle, 2015). ries by 15 percent in the waning In Algeria, the minimum sal- days of the Mubarak regime (and ary was increased by 20 percent, further increases were brought by effective in January 2012 (IMF, the Muslim Brotherhood govern- 2012 and European Training ment). Morocco increased civil Foundation, 2014). In July 2012, service salaries by around $75 per the Tunisian government, after month. Jordan granted a one-off consultation with the unions and cash transfer to civil servants at a employers’ organizations, raised cost of more than $110 million. the minimum salary in the in- By 2012, all five countries had dustrial and agricultural sectors also increased minimum wages. (Kienle, 2015). Jordan also saw In 2011, the Egyptian govern- significant hikes in minimum ment granted public servants on wages in the private sector, most permanent contracts a minimum notably in 2012 (see figure 3.9). monthly wage of LE700 (US$ Finally, the Moroccan gov- 120), effective in July 2012. A ernment increased the minimum similar minimum wage increase wage of workers in the private for private sector workers was sector by 10 percent in 2014. In discussed by the National Wages the public sector, the monthly Council, but no agreement was minimum wage was increased reached. In January 2014, the by nearly 30 percent, although it minimum wage for public sec- was made effective as of 2015 to tor employees was raised again reduce its impact on the public to LE1,200 (US$ 170) (Bank Audi, wage bill in 2014 to only 0.02 per- 2014). A new law on workers’ cent of GDP (IMF, 2014c). rights was approved at the cabinet level after the Uprisings but has yet to be ratified, amid com- Social services plaints about restrictions on trade The swift response by South-Med governments to enhance social protection was not matched by Figure 3.9: Minimum Wages and GDP Per Capita a similar response to improve in Jordan (Index = 100 in 1999) the provision of social services, mainly health and education. Understandably, improving the provision of health and education services is not a short-term project, nor can it be achieved merely by increasing government expenditure. The reform process is much more complex, and involves building schools and hospitals, hiring more teachers and doctors, changing the curriculum and adopting modern teaching Source: Tzannatos, Z. (2015). techniques. In addition, it requires a change in the incentive Chapter 3 | 77

structures for teachers/doctors maintained the same low levels and students/patients, increas- of expenditures on health over ing the autonomy of service pro- the entire period. However, the viders in return for accountabil- Egyptian government initiated ity, and greater transparency and two socially relevant programs: participation in policy making one in March 2015 to cover the (see World Bank, 2008). The ques- poor in Upper Egypt; the other tion remains, however, as to how in 2012 to combat the Hepatitis C much attention governments in Virus (HCV). Only Tunisia and South-Med countries have paid to Algeria increased their public ex- health and education in the post- penditure on health over time, uprising period. especially after the uprisings. Starting with health, it is evi- Notwithstanding the above dent that the health sector has re- declarations and the increase in ceived more attention in national public expenditure on health in debates since the uprisings than some countries, the system con- before. The newly adopted continues to suffer from several short- stitution in Egypt in 2014 guaran- comings. Table 3.2 shows the hos- tees each citizen the right to have pital beds per 1,000 people in 2007 access health care, and sets a floor and 2012. It clearly indicates that for public spending on health at Tunisia is the only country that 3 percent of GDP. The new Tuni- has witnessed an improvement in sian Constitution of 2014 also as- the number of hospital beds per serts that access to health services 1,000 people. The latest available is a human right. It further states data for the percent of the popu- that the government will guar- lation covered (as members of antee preventive health care and health insurance or free access to treatment for every citizen, while health care services provided by ensuring free health care for those the State) predates the uprisings. without adequate means. In Mo- However, the percent was as low rocco, in response to protests in as 51 percent for Egypt (2008) and 2012, the government implemented a Medical Assistance Plan to Figure 3.10: Public Expenditure on Health, 2000- ensure the provision of healthcare 2013 (as % of GDP) to the poor (African Development Bank, 2013). How far was this attention translated into action? Figure 3.10 shows government expenditures on health in South-Med countries over the period 2000-2014. The picture conveys mixed sig- nals. Public spending on health as percentage of GDP declined in Jordan between 2010 and 2013, although Jordan has consistently spent more on this sector than all other South-Med countries. Morocco and Egypt essentially Source: World Development Indicators, World Bank. 78 | Social Policy Response

42 percent for Morocco (2007). In with patients paying more than the other countries, the coverage 50 percent out of pocket. Over was better, 85 percent for Alge- time, the trend of OOP health ex- ria (2005), 75 percent for Jordan penditures has remained relative- (2006) and 80 percent for Tunisia ly stable in all 5 countries, with (2005). (See ILO (2014) for further the exception of Jordan where the details). ratio has decreased since 2008. Another indicator of accessi- Beside issues of access to health bility is out-of-pocket health ex- services, high OOP expenditures penses (OOP), which tend, in the have implications for poverty. In presence of universal healthcare Egypt, Rashad and Sharaf (2015) systems, to account for less than estimated that high OOP expen- 20 percent of total health expenditures have driven 7.4 percent of ditures (Alami, 2014). In the case the households below the poverty of our five countries, figure 3.11 lines. For Tunisia, Elgazzar et al. indicates that the ratio is much (2010) estimates show that 17.8 higher. The problem is particu- percent of households will fall larly acute in Egypt and Morocco, into poverty after accounting for the effect of OOP payments. Ultimately, health sector re- Table 3.2: Health facilities forms, along with concurrent Hospital beds (per 1,000 people) socio-economic developments, ought to be reflected in health 2007 2012 outcomes. By this yardstick, table Algeria* 1.7 - 3.3 indicates that all five countries Egypt 2.1 0.5 have witnessed an improvement Jordan 1.8 1.8 over the past decade. Life expec- Morocco* 1.1 0.9 tancy has increased and maternal Tunisia 1.8 2.1 mortality and mortality rates un- Note: *2004. der the age of 5 have decreased. Source: World Development Indicators, World Bank. However, these positive results are national averages and could conceal significant variations Figure 3.11: Out-of-pocket health expenditures, within society. Therefore, even if 2000-2014 (% of total expenditure on health) health outcomes are improving in general, more benefits may be accruing to certain groups than others. To find out, table 3.4 compares two health outcomes by wealth quintile for Egypt, Jordan and Morocco. The health outcomes are mortality rates under the age of 5 per 1,000 births and the percentage of women delivering in a health facility. The data shows that health outcomes have improved for all groups, irrespec- Source: World Development Indicators, World Bank. tive of their wealth over time. Chapter 3 | 79

However, a gap persists between most countries continue to the highest and lowest quintiles, declare their commitments to especially in Egypt and Morocco. free education, families are re- By comparison, Jordan has done portedly enduring an increasing much better over time. burden of the cost in the form of Turning to education, the re- higher tuition fees and private tu- sponse by South-Med govern- toring. ments after the uprisings was not Regarding the quality of edu- much better than in the health sec- cation, the TIMSS scores (Trends tor. In fact, in the four countries in International Mathematics and for which data is available for Science Study) are shown in Table the period 2002-2013, the share of 3.6. The results consistently show public expenditure on education that all five South-Med countries relative to GDP was lower in 2013 performed less well than the in- than it was on average during ternational average in math and the period 2002-2008 (Table 3.5). science at all levels of educa- During this period, the role of the tion (4th and 8th grades) in 2007 private sector in the provision of and 2011. Within the South-Med education was encouraged, most countries, students scored bet- notably, in Jordan and Egypt (Ga- ter in Tunisia and Jordan than in lal and Kanaan, 2010). Although Egypt and Morocco.

Table 3.3: Health outcomes, 2000-2014 Maternal mortality ratio Life expectancy at birth, (modelled estimate, per Mortality rate, under-5 total (years) 100,000 live births) (per 1,000 live births) 2000 2010 2014 2000 2010 2014 2000 2010 2014 Algeria 70.2 73.8 74.8 170.0 147.0 141.0 39.7 27.3 25.6 Egypt 68.6 70.3 71.1 63.0 40.0 34.0 46.5 29.0 24.8 Jordan 71.8 73.4 74.1 77.0 59.0 59.0 27.7 20.8 18.5 Morocco 68.5 72.6 74.0 221.0 153.0 126.0 50.0 33.1 28.6 Tunisia 72.6 74.6 74.1 84.0 67.0 63.0 31.7 17.4 14.6 Source: World Development Indicators, World Bank.

Table 3.4: Health outcomes by wealth quintiles Under-five mortality rates (per 1000 percentage of women delivering in a births) health facility

lowest highest lowest highest 20% 20% total 20% 20% total Egypt 2005 74.6 25.1 41 40.9 92.0 64.6 2008 49.0 18.9 33 45.4 94.6 71.7 2014 42.0 19.0 27 75.1 97.7 86.7 Jordan 2007 30.0 27.0 22 97.3 99.9 98.7 2012 29.0 11.0 21 96.9 100 98.8 Morocco 2004 78.0 26.0 47 28.3 94 60.8 Source: Demographic and Health Surveys, several years. 80 | Social Policy Response

The TIMMS scores are consis- The responses are shown in tent with the populations’ per- Table 3.7. They indicate that the ception of the population level of quality of the education systems their education systems. Accord- were considered above average ing to the Global Competitiveness in Tunisia and Jordan, but aver- Reports Database, the following age or below average in Algeria, questions were asked: Egypt and Morocco. Compar- • “In your country, how ing the scores for 2015 to those would you assess the qual- for 2010, there was a modest im- ity of primary schools? [1 provement in Algeria, but a stag- = extremely poor; 7 = ex- nation or even a deterioration ev- cellent]” erywhere else. • “How well does the edu- Certainly, South-Med coun- cation system in your tries have improved adult liter- country meet the needs of acy and net enrolment over time a competitive economy? (Table 3.8). The improvements [1 = not well at all; 7 = ex- in adult literacy have been most tremely well]” significant in Morocco and in net • In your country, how enrolment in Egypt and Morocco. would you assess the Nevertheless, adult literacy rates quality of maths and sci- remain at around 70 percent in ence education? [1 = ex- Morocco and Egypt. Moreover, tremely poor; 7 = excel- net enrolment rates came down lent]” from 89.7 percent in 2010 to 87.5

Table 3.5: Public expenditure on education, 2008-2013 (% of GDP) 2002-2008 (average) 2008 2009 2010 2011 2012 2013 Algeria - 4.3 - - - - - Egypt 4.8 3.7 4.2 4.1 3.9 3.7 3.6 Jordan 5.0 3.8 3.7 3.4 3.7 3.8 4.0 Morocco 5.8 5.4 5.9 5.8 5.1 4.9 4.7 Tunisia 6.4 6.3 6.5 6.3 - 6.2 - Source: National Statistics Offices and World Development Indicators, World Bank.

Table 3.6: TIMSS average scores for 4th and 8th grades, 2007 and 2011 Overall Mathematics Overall Science 2007 2011 2007 2011

4th 8th 4th 8th 4th 8th 4th 8th grade grade grade grade grade grade grade grade Algeria 378 387 - - 354 408 - - Egypt - 391 - - - 408 - - Jordan - 427 - 406 - 482 - 449 Morocco 341 - 335 371 297 - 264 376 Tunisia 327 420 359 425 318 445 346 439 International average - - 468 478 - - 461 485 Source: TIMSS Results Reports, International Association for the Evaluation of Educational Achievement. Chapter 3 | 81

percent in 2012 in Jordan. rising overall unemployment, One problem that has per- the situation may have become sisted over time is a mismatch worse. It is worth pointing out between education outcomes and that the sharpest deterioration in labour market demand, especially unemployment among college for graduates with tertiary educa- graduates was in Tunisia. tion. Across countries, unemployment has continued to be high among this group. As per Table Non-economic concerns 3.9, unemployment rates among Beyond economic concerns, the graduates with tertiary educa- uprisings were also an expression tion increased between 2001 and of dissatisfaction with the prevail- 2011, and there are no good rea- ing political systems. Perhaps this sons to suggest that this trend has is so because the two are linked; vanished. On the contrary, with bad political systems fail to meet

Table 3.7: Perceptions of the quality of education, 2007-2015 Algeria Egypt Jordan Morocco Tunisia 2010 2015 2010 2015 2010 2015 2010 2015 2010 2015 Primary 3.1 2.9 2.5 2.1 4.0 4.0 3.1 3.0 5.0 3.6 Education system 2.9 3.3 2.5 2.1 3.9 4.4 3.1 2.8 5.0 3.3 Math and science 3.6 3.3 2.7 2.6 4.4 4.2 4.0 4.0 5.6 4.4 Source: Global Competitiveness Reports Database, World Economic Forum.

Table 3.8: Education outcomes Adult literacy rate, both sexes (%) Net enrolment rate, primary, both sexes (%) Mid 2000s* 2015 2000 2010 2012 Algeria 72.6 79.6 89.2 97.0 Egypt 71.4 75.8 86.3 97.3 98.3 Jordan 91.1 98.0 95.1 89.7 87.5 Morocco 52.3 71.7 75.8 94.4 97.3 Tunisia 74.3 81.1 95.4 98.9 98.8 Note: *Algeria in 2006, Egypt and Jordan in 2005, Morocco and Tunisia in 2004. Source: World Development Indicators, World Bank

Table 3.9: Unemployment among graduates with tertiary education (% of total unemployment) 2001 2008 2011 Algeria 9.2 19.8 23.1 Egypt - 31.7 32.1 Jordan 14.7 30.3 34.4 Morocco 18.7 16.9 18.3 Tunisia 6.7 25.9 30.9 Source: World Development Indicators, World Bank. 82 | Social Policy Response

the aspirations of citizens for so- not free (NF). The results are re- cial mobility and greater equal- vealing. By 2015, Algeria, Egypt, ity, and vice versa. Not surpris- Jordan were considered not free. ingly, one of the common slogans Only Tunisia was considered of the rebels was freedom. The free, while Morocco partially free. question is this: a few years lat- Over time, the changes, if any, er, have these countries become have been very modest. more democratic and less cor- Looking at the corruption in- rupt? Have their political systems dex, the scores for the five South- become more inclusive, or have Med countries in 2003, 2010 and they merely replaced one extrac- 2015 are shown in figure 3.12. tive elite by another? Overall, Tunisia and Jordan are On the positive side, both perceived to be less corrupt than Egypt and Tunisia have adopted Algeria, Egypt and Morocco. The new and more liberal constitu- good news is that people believe tions in the post-uprising period. that corruption has declined over They have run more transparent time in all countries, except in Tu- elections of presidents and par- nisia. The most notable improve- liaments than before. In Morocco ment has been in Jordan. and Jordan, partial political re- In summary, South-Med forms have also been adopted. countries have responded to the Across all countries, the leaders social consequences of the upris- have announced their commit- ings primarily by adopting bet- ment to democracy, good gover- ter targeting mechanisms for the nance and less corruption. poor and by saving resources The problem is that people do through energy subsidy reforms. not seem to have felt that a real They have also attended to labour change has taken place in their demands, mostly in the form of political and social lives. Table government employment and 3.10 shows the Freedom House higher wages. Progress in social index for all five countries at two protection has not been matched different points in time (2010 and by similar progress in empower- 2015). The countries are classified ing citizens through the better as Free (F), Partially Free (PF) and provision of health and education

Table 3.10: Freedom House index, 2000-2015

2010 2016

Political Civil Liber- Political Civil Liber- Rights ties Status Rights ties Status Algeria 6 5 NF 6 5 NF Egypt 6 5 NF 6 5 NF Jordan 6 5 NF 6 5 NF Morocco 5 4 PF 5 4 PF Tunisia 7 5 NF 1 3 F Note: “F” “PF” and “NF” stand for “Free” “Partly Free” and “Not Free”, respectively. Starting 2003, countries whose combined average ratings fall between 3.0 and 5.0 are “Partly Free” and those between 5.5 and 7.0 are “Not Free.” Source: Several “Freedom in the World” annual reports, Freedom House. Chapter 3 | 83

services. And despite some prog- Figure 3.12: Corruption perceptions index (CPI), ress in building de jure political 2003-2015 institutions, citizens are not con- vinced that many de facto changes in political institutions have taken place, except in Tunisia and partially in Morocco.

3.4 Assessment of the social response to date It is difficult to assess the impact Note: The Corruption Perceptions Index aggregates data from different sources of perceptions of business people and country of social policies on citizens be- experts of the level of corruption in the public sector. For 2015, cause of the time lag between the data are standardized to a scale of 0-100, where a 0 equals reforms and their effect and be- the highest level of perceived corruption and 100 equals the low- cause it is not easy to establish a est level of perceived corruption. As for 2010 and 2003, the CPI measures are standardized to a scale from 10 (very clean) to 0 reliable counterfactual. It is pos- (highly corrupt). sible nevertheless to comment on Source: Transparency International data. the reforms in terms of their ap- propriateness and effectiveness in the short-run, under the influ- achieving intended objectives. ence of the politics of the hour. The broad conclusion of this They opted for reforms that have section is that the responses thus immediate rather than delayed far seem to have eased the bur- payoffs. As shown in table 3.11, den of transition and reduced its all five South-Med countries in- hardship for the groups that have creased government expenditures received benefits. Those who during the period 2010 to 2014. have found a job or are receiving The increase in expenditures as higher wages or cash transfers are a percent of GDP between 2010 doubtless better off than if no re- and 2014 was high in hard-hit forms had been implemented. On Egypt and Tunisia. But the same the other hand, the reforms seem was true in Jordan, Algeria and to suffer from several shortcom- Morocco. At the same time, the ings. First, they have been driven opposite trend was held with re- primarily by short-term political spect to public investment. In all gains. Second, they have lacked five countries, public investment, sustainability. Third, they do not as a percent of GDP, was reduced represent a major departure from between 2010 and 2015. Tunisia, the old social contract. Egypt and Jordan went the far- thest, with a reduction of 1-2 percentage points of GDP. Algeria Short-termism and political expe- followed likewise. Only Morocco diency modestly increased the level of The responses to social demands investment relative to GDP dur- surveyed in the previous section ing this period. suggest that interim governments The above pattern of fiscal opted for policies that focus on 84 | Social Policy Response

Table 3.11: Public investment and general government expenditure, 2010-2014 (% of GDP)

Public investment expenditure General government expenditure 2010 2014 change 2010 2014 change Algeria 15.1 14.5 -0.60 37.0 40.7 3.6 Egypt 3.8 2.5 -1.30 31.8 36.7 4.9 Jordan 5.1 4.5 -0.60 30.4 30.9 0.4 Morocco 6.2 7.0 0.80 31.1 33.0 1.9 Tunisia 6.6 4.3 -2.30 25.6 29.2 3.5 Source: IMF Article IV consultations and World Economic Outlook Database.

response suggests that govern- of the political instability experi- ments attempted to appease or- enced during the transition, but ganized groups with political it has proved to be costly. It is influence rather than the public true that aspirations for instant at large. Many of the reforms gratifications at the time were in- addressed organized workers, tense, so was the willingness on either because they have strong the part of many groups to par- trade unions (Tunisia), or because ticipate in riots. At the same time, they work for a large bureaucracy governments were preoccupied (Egypt) whose support is impor- with laying the foundations for tant for the regime. Ironically, more inclusive political institu- when Egypt raised the minimum tions, let alone restoring security. wage, it was only applied to gov- However, the fact remains that ernment employees, as business the responses to social demands associations were opposed to its were myopic and tilted in favour application in the private sector. of certain groups. With some ex- Workers in the informal sector ceptions, those who needed help saw none of the benefits in Egypt, the most were left out. Equally or elsewhere, partly because of important, the responses have enforcement difficulties and part- eroded economic sustainability. ly because they lack the channels Obviously, the adverse eco- to make their voices heard. At the nomic conditions associated with same time, no major actions were political instability made sustain- taken to extract taxes from the ability difficult, but so did the rich to meet the growing expen- unwillingness of interim govern- ditures. The justification was that ments to raise taxes or engage in this would discourage invest- a major restructuring of expen- ment at a time when the economy ditures during the early years of was sluggish. Perhaps, also, the the transition. At the end, South- regimes did not want to antago- Med countries have had to face nize powerful interest groups. the challenge of escalating public debt to risky levels and eventually Sustainability to resort to the IMF for a bailout. The above pattern of behavior To explain what happened may be understandable in light further, we may consider the Chapter 3 | 85

trends in taxation and the govern- Figure 3.13: Limited economic base and taxes ment wage bill over the period against rising wage bills, 2002-2014 (Index 2002 to 2014 (figure 3.13). The re- 2010=100) markable observation from this figure is that the index for tax revenues in all South-Med countries has not kept pace with the increase in public sector wages. The divergence between both trends began around 2010/11. The sharpest divergence occurred in Egypt, Jordan and Algeria. Tu- Egypt 230 nisia was more disciplined than 300 200 other countries. 168 From a social development 100 0 point of view, it would also have been useful for interim govern- 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 ments to begin a process of re- taxes wages structuring public expenditure in favour of education, health, and infrastructure. Several studies have pointed out that publicly provided education and health services have become increasingly ineffective at delivering quality services (World Bank, 2008). Teacher and doctor absen- teeism in schools and clinics are common. A private sector has emerged that makes people pay for the same services to which they were entitled. As an Egyp- tian woman put it, ‘you can go to the private clinic and lose all your money, or go to the public clinic and lose your life’ (Devara- jan, 2015). But not much has been done on this front. Delaying reforms of the social sector can also be destabilizing in the long-run. Exacerbating the problem further is that South-Med countries Source: IMF Article IV Consultation data have increased public spending essentially through debt financ- and has reached alarming levels ing. As a result, public debt posi- of about 90 percent of GDP in tions have deteriorated relative to Egypt and Jordan in recent years. GDP in all countries but Algeria Debt is also high in Morocco and (figure 3.14). The problem has be- Tunisia, but less so than in Egypt come worse over time since 2011, and Jordan. 86 | Social Policy Response

Figure 3.14: General government gross debt, 2008- upon. Broadly speaking, social 2015 (percent of GDP) contracts are liberal and interest- based, or rights-based. There is a further distinction under the rights-based contract, depending on the commitment made by different societies to upholding certain rights and the responsibility of the state in these societies. From the perspective of economic and social development, Neoliberalism gives prominence to individual choice, markets as a mechanism for resource allocation and the fortune of individuals, and opportunity and safety Source: World Economic Outlook Database, IMF. nets as society’s way of dealing with social concerns. At the other A new social contract? end of the spectrum, social de- To place the discussion of social mocracy is built on the premise contracts in a broader context, of citizen empowerment, the cen- Hickley (2011) has put forward tral role of the state in aiding in- a way of differentiating between divisibles and the prominence of two types of social contracts. As universal rights as the foundation shown in Table 3.12, each of these of inclusion. contracts has a key feature per- Against this typology, what taining to its overarching goal, can be said about the South-Med the position of individuals in soci- countries? Unfortunately, the pic- ety, the foundation of society and ture is still unclear, to say the least. the political arrangements agreed Before the uprisings, a ruling elite

Table 3.12: Approaches to social contracts

Liberal/Interest-based Social/Right-based

(Negative) (Positive) Key proponents Hobbes, Hayek, Nozick Rousseau, Kant, Rawls, Gauthier, Buchanan Locke Scanlon Overarching goal of the Protection; maintenance Protecting existing prop- social contract of order erty rights Promoting justice Vision of individual Rational actor; motivated Impartial actor; moti- by subjective ends. Indi- vated by impersonal aims vidual as isolated from Individual in relation to others Rights bearing citizen others Vision of society Individualistic Merit- Commonwealth Equali- based notion of justice Individualistic ty-based social justice Basis for social relations Utilitarian Mutual respect Mutual respect View of political ar- Political arrangements for mutual advantage (although on ethical/moral grounds rangements under a right based approach) Source: Hickley, S. (2011) “The politics of social protection: what do we get from a ‘social contract’ approach?”, Chronic Poverty Research Centre Working Paper No. 216, July. Chapter 3 | 87

captured most of the political and slower investment and economic economic power. Democracy and growth, which has meant that liberties were thin. The liberaliza- governments have had fewer re- tion of the 1980s and 1990s led to sources to meet social demands. rolling back of the role of the state Second, in response to wors- in favour of markets as a mecha- ening social conditions, interim nism for resource allocation and governments opted for short- even the provision of social ser- term measures to appease the vices. Corruption was also wide- streets. They implemented initia- spread. tives to increase social protection, In the post-uprising period, through programs such as cash the changes in the political land- transfers. They also increased em- scape have been modest, or ployment and wages in govern- merely in relation to formal in- ment and allocated more resourc- stitutions. The room for reversal es to commodity subsidies such is not out of the question. And it as food. Under pressure, they will take time for formal institu- acted on energy subsidies to save tions to find their place in reality. resources for their social protec- The next chapter elaborates the tion programs. In the process of evolution of the political systems doing all this, they have tended to in South-Med countries thus far, appease organized groups at the as well as their likely trajectories. expense of unorganized ones. Progress in social protection has not been matched by reforms 3.5 Concluding remarks to improve the provision of social services, such as health and This chapter intends to document education. The two sectors have the social consequences of the up- not seen a greater allocation of risings, analyze the way different resources, nor have they under- governments in South-Med coun- gone reforms to improve qual- tries have responded to social ity and access to these services. demands, and to provide an as- Similarly, although governments sessment of the reforms they have have carried out some reforms of carried out. The analysis is not formal political institutions and limited to the economic dimen- conducted free elections, the po- sions of the consequences, but is litical landscape remains highly extended to include political con- imperfect. Only Tunisia is consid- cerns. The following are the main ered “free” and Morocco “partial- conclusions. ly free”, with the other countries First, the uprisings have led remaining “not free” (Table 3.10). to a deterioration in social con- Corruption may have subsided ditions, especially in Egypt and marginally, but it remains rela- Tunisia, where the political up- tively high. As such, the political heavals were most intense. The bargain that prevailed before the consequences have been seen in uprisings would seem to perdure rising unemployment, poverty to date. and inequality of opportunity. Finally, the assessment of the The uncertainty associated with response to social concerns has political instability has led to been myopic, unsustainable and 88 | Social Policy Response

modest. Myopic because it has Kingdom of Morocco, Ap- focused on improving the current praisal Report” situation, even if the action is to be Alami, R. (2014) “Health, Social costly in the future. This is most Policy and inclusive growth in visible in the increase in current MENA”, SOAS Department of government expenditures, while Economics Working Paper Series cutting public investments. Un- No. 188, the School of Oriental sustainable because the deterio- and African Studies, April. rating fiscal deficits have led to Assaad, R., C. Kraft, J. Roemer an escalation in public debt and and D. Salehi-Isfahani (2016) ultimately to resorting to the IMF “Inequality of Opportunity in for a bail out. Modest in terms of Income and Consumption: the observed changes in the politi- Middle East and North Africa cal bargain between citizens and Region in Comparative Per- their new rulers. Without notable spective”, ERF Working Paper progress on building inclusive No. 1003, May. political institutions, the pros- Assaad, R., C. Krafft, N. Hassine pects for sustainable prosperity and D. Salehi-Isfahani (2012) among all citizens remain distant “Inequality of Opportunity (Acemoglu and Robinson, 2012, in Child Health in the Arab Galal and Selim, 2013). World and Turkey”, ERF Working Paper No. 665, April. Atamanov, A., J. Jellena and U. Notes Serajuddin (2015) “Energy Subsidies Reform in Jordan 1. WGI rely on perceptions from vari- – Welfare Implications of Dif- ous data sources, including surveys ferent Scenarios”, World Bank of firms and households, as well as Policy Research Working Paper the subjective assessments of a variety of commercial business informa- No. 7313, June. tion providers, non-governmental Bank Audi (2014) “Egypt Eco- organizations, and a number of nomic Report – Growing multi-lateral organizations and oth- Hopes for Macro Recovery er public-sector bodies. Following Transitional Years of Economic Sluggishness”. Chemingui, M. and M. Sanchez (2011) “Assessing Develop- References ment Strategies to Achieve the MDGs in the Republic of Tu- Acemoglu, D. and JA. Robinson nisia”, United Nations Depart- (2012) “Why nations fail: the ment for Social and Economic origins of power, prosperity Affairs, October. and poverty”, vol. 4, Crown Desai, R., A. Olofsgard and T. Business. Yousef (2007) “The Logic of Achy, L. (2011) “Tunisia’s Eco- Authoritarian Bargain: A Test nomic Challenges”, Carnegie of Structural Model”, Brook- Papers, December. ings Global Economy and Devel- African Development Bank (2013) opment Working Paper No. 3, “Medical Coverage Reform January. Support Program – Phase 3, Devarajan, S. (2015) “What is the Chapter 3 | 89

social contract and why does Research Centre Working Paper the Arab world need a new No. 216, July. one?”, World Bank Blog Voices International Labor Organization and Views: Middle East and (ILO) (2014) “World Social North Africa, December. Protection Report 2014-2015: Diwan, I., P. Keefer and M. Schiff- Building economic recovery, bauer (2016) “Pyramid Capi- inclusive development and talism: Cronyism, Regula- social justice” tion and Firm Productivity in International Monetary Fund Egypt”, IDB Working Papers (IMF) (2015a) “Arab Republic Series No. IDB-WP-739, Octo- of Egypt 2014 article IV Con- ber. sultation – Staff Report; Press El Enbaby, H. and R. Galal, (2015) Release and Statement by the “Inequality of Opportunity Executive Director for the in Individuals’ Wages and Arab Republic of Egypt” Households’ Assets in Egypt”, ____ (2015b) “Morocco 2014 Ar- ERF Working Paper No. 942, ticle IV Consultation – Staff September. Report; Press Release and Elgazzar, H., F. Raad, C. Arfa, A. Statement by the Executive Mataria, N. Salti, J. Chaaban, Director for Morocco” D. Salehi-Isfahani, S. Feshara- ____ (2014a) “Subsidy Reform in ki, and M. Majbouri (2010) the Middle East and North “Who Pays: Out-of-Pocket Africa – Recent Progress and health spending and equity Challenges Ahead” implications in the Middle ____ (2014b) Jordan 2014 Ar- East and North Africa”, World ticle IV Consultation, Third Bank Health, Nutrition and Pop- and Fourth Reviews Under ulation Discussion Paper. the Stand-By Arrangement, European Training Foundation Request for Waivers of Non- (2014) “Employment Policies observance of Performance and Active Labour Market Criterion – Staff Report; Press Programmes in Algeria” Releases; and Statement by the Galal, A. and H. Selim (2013) “The Executive Director for Jordan” Elusive Quest for Arab Eco- ____ (2014c) “Morocco: Request nomic Development”, Middle for an Arrangement Under the East Development Journal, Vol. Precautionary and Liquidity 05, No. 01, March. Line and Cancellation of the Galal, A. and T. Kanaan (2010) Current Arrangement – Staff “Financing Higher Education Report; Press Release and in Arab Countries”, ERF Policy Statement by the Executive Research Report No. 34, July. Director for Morocco” Hassine, N. (2012) “Inequality of ____ (2012) “Algeria 2011 Article Opportunity in Egypt”, World IV Consultation – Staff Report; Bank Economic Review, vol. 26, Public Information Notice” issue 2, June, pages: 265 – 295. Khandelwal, P. and A. Roitman Hickley, S. (2011) “The politics (2013) “The Economics of Po- of social protection: what do litical Transitions: Implica- we get from a ‘social contract’ tions for the Arab Spring”, approach?”, Chronic Poverty IMF Working Paper No. 13/69, 90 | Social Policy Response

March. Tzannatos, Z. (2015). “Jordan’s Kienle, E. (2015) “Changed Re- National Employment Strat- gimes, Changed Priorities? egy, 2011-2020, an update and Economic and Social Policies future directions”. Internation- After the 2011 Elections in Tu- al Labor Organization. nisia and Egypt”, ERF Working World Bank (2014a) “The Un- Paper No. 928, July. finished Revolution Bringing North, DC. (1990) “Institutions, Opportunity, Good Jobs and Institutional Change, and Eco- Greater Wealth to All Tuni- nomic Performance”, Cam- sians” bridge University Press. World Bank (2014b) “Jordan Eco- Pikkety, T. (2014) “Capital in the nomic Monitor - Resilience Twenty-First Century”, Har- and Turmoil” vard University Press. World Bank (2013) “Jordan Eco- Rashad, A. and M. Sharaf (2015) nomic Monitor - Maintaining “Catastrophic and Impover- Stability and Fostering Shared ishing Effects of Out-of-Pocket Prosperity Amid Regional Health Expenditure: New Evi- Turmoil” dence from Egypt”, ERF Work- World Bank (2008) “The Road ing Paper No. 974 Not Traveled – Education Re- Rawls, J. (1971) “A Theory of Jus- form in the Middle East and tice”, Rev. Ed. Belknap Press North Africa”, MENA Devel- of Harvard University Press. opment Report. Reijkers, B., C. Freund and A. Nucifora (2014) “All in the Family – State Capture in Tu- nisia”, World Bank Policy Re- search Working Paper No. 6810, March. Roemer, JE. (1998) “Equality of Opportunity”, Harvard Uni- versity Press. Salehi-Isfahani, D., N. Hassine and R. Assaad (2014). “Equal- ity of opportunity in educational achievement in the Mid- dle East and North Africa”, The Journal of Economic Inequal- ity, vol. 12, issue 4, November, pages 489-515 Subrahmanyam, G. and V. Castel (2014) “Labor market reforms in post-transition North Afri- ca”, African Development Bank Group Economic Brief CHAPTER 4

Political Transition and Economic Development

4.1 Introduction managed to largely diffuse the political shock. In these countries, This report has so far analyzed the regimes have survived either the economic consequences of the by increasing patronage (esp. in political transitions in South-Med Algeria), or by promising a grad- countries and how successive ual change in politics and eco- governments have attempted to nomics (Jordan, Morocco). restore macroeconomic balances, The countries for which tran- reignite economic growth and sition is more complex are Egypt achieve greater social justice. The and Tunisia (and to some extent unfolding political realities in the Lebanon which is attempting to wake of the uprisings were taken salvage its core political settle- as given, at least regarding how ment), where the political shock they would constrain and shape resulted in regime change. In the choices available to policy- these two cases, the macro and makers, as well as the feedback growth performance have dete- from economic performance to riorated markedly, as discussed politics. In this chapter, an at- in the previous chapters, as have tempt is made to rectify this omis- state services. In these “transi- sion. tioning” countries, the economic Broadening the scope of the reform agenda has centred on analysis beyond pure economic three concerns: the first and most management, it is important immediate is that of economic to point out that six years after stabilization in order to avoid the spark of the Arab uprisings, an economic and financial melt- South-Med countries continue to down, which would further com- face enormous challenges in the plicate the political process; the spheres of security and political second is related to the moderni- development. Iraq, Libya, Syria sation of the state and the reha- and Yemen are at risk of becom- bilitation of public services; and ing failed states. Rebuilding na- the third concerns the business tional coalitions that can preserve environment and job creation. a modicum of stability is now a That much is well known. priority. Other countries – Jor- What is less clear is the extent to dan, Morocco, and Algeria – have which the new political circum-

91 92 | Political Transition and Economic Development

stances allow the new govern- between the changes in political ments to shape and carry out circumstances and the institu- economic reform. Also unclear tions responsible for delivering is how one can build reform economic outcomes, especially if packages that command a mini- political transitions are bringing mum level of political support - a about new political settlements. tricky endeavour in the best of To this end, the chapter at- circumstances and challenging tempts to answer the following in the current hyper-politicized questions: environment. A further compli- • Is there a typology of al- cating factor relates to the quality ternative political/insti- of the institutions inherited from tutional paths in the wake the past, which were already in a of transition? state of partial decay before the • Based on this framework, uprisings of 2011. what is the likely evolu- Additionally, the rulers in the tion of political regimes region now have to reckon with in the Arab Countries in a new reality. In all countries in transition? the Middle East – both where • In comparing Egypt and millions have protested and Tunisia, what can be said managed to oust long-standing about the relationship dictators and where opposition between the evolution of is muted – politics have changed political regimes, on the significantly in the last seven one hand, and the state- years. New political actors have business relationship and taken the reins of state power. the state’s capacity to de- Islamists have won free and fair liver social services and elections in Tunisia, Egypt and macro-manage the econ- Morocco, even if their time in omy, on the other? office was relatively short-lived The analysis takes politics as and their actual strength in so- given, but it describes the feed- ciety is less extensive than many back on politics from economic had long assumed. New politi- performance. It pays special at- cal movements and parties have tention to Tunisia, a polity that been formed and others are in seems to be moving towards formation. Above all, the public more competitive forms, and engages more overtly in politics, Egypt, which by contrast seems making the “street” a force to to be moving toward a more reckon with. dominant path. It explores the The central claim of this chap- implications of these two diver- ter is that in order to develop a gent paths for economic growth, positive grammar of policies social services and macroeco- during transitions, it is essen- nomic management. The analy- tial to understand two things: sis reveals that while the mac- first, the nature of the political ro-situation is likely to remain transition in different countries unstable in the future in both (the starting points and destina- countries, the pressure points tions of each country); second, will be different and will come the likely types of interaction mostly from the state sector in Chapter 4 | 93

Tunisia, and from low private counts do not provide an expla- sector growth in Egypt. nation of the political changes oc- The rest of the chapter is orga- curring through the transition, or nized as follows. Section 2 offers provide a precise account of how a stylized framework that helps economic performance relates to in understanding the dynamic changes in the political regime. political economy path followed The alternative of considering by the post-revolt countries, and each case to be highly specific is discusses the implication for in- not very useful either. What is stitutions - inherited and future. needed is a framework that allows Section 3 applies this framework us to go beyond the every-case-is- in a speculation on the likely evo- unique level of complexity, but lution of political regimes in the does not fall into the one-size-fits- Arab region. Sections 4-6 look in all pitfall while recommending more detail at Tunisia and Egypt, the same best practice as a cure focusing on state capacity (sec- for all ills. We need a simple ty- tion 4), economic growth (section pology of political regimes and 5) and macro-management (sec- how they may evolve over time in tion 6). Section 7 concludes. order to speak of a transition from one regime to another, as well as one that allows for some logical 4.2 Transition challenges, mapping between regime types politics, and institutions and the institutions that matter from a socioeconomic develop- “Transitology”, taken to include ment perspective. Such a typol- both political and economic trans- ogy is offered below. formations, can be confusing. Much about transitions is specific to countries, including start- A typology of political paths ing points and destinations, but A useful political-economy frame- analysts usually look for general work, which integrates gover- rules. Initially, the emphasis is on nance and growth, is provided by the political side, which tends to the work of Khan (2010) and Levy be the leading concern. But if the (2014). It is summarized in Table economy deteriorates, this makes 4.1. In this view of the world, once political change harder, and calls countries manage to ensure a ba- intensify for economic improve- sic level of security, they evolve ments to support political change. into a political space character- This creates a risk of being caught ized by either a dominant party, in a transition trap (EBRD, 2013). or a competitive political regime In previous transitions, especially (the vertical axis). Horizontal in Eastern Europe and the For- movements along the map repre- mer Soviet Union (FSU), growth sent instances where the quality dipped for 4 years on average af- of governance improves or slides ter the political transition before back, and vertical movements it recovered and reverted to nor- represent passages from autocrat- mal rates (Freund and Mottaghi, ic to competitive systems. 2011). The “founding” political set- While useful, the above ac- tlement is a bargain among elites 94 | Political Transition and Economic Development

Table 4.1: A typology of political regimes and regime dynamics

Quality of governance Myopic/personalized Institutionalized Dominant Dominant discretionary Rule by law dominant From conflict to a politi- Arab oil importers af- - developmental state Sustainable cal settlement that binds ter economic liberaliza- Arab states before democracy elites in a rent/ power tion (II) regimes narrowed (I) (VI) sharing coalition, and competitive Rule of law competi- society with a workable Personalized/frag- tive social contract mented competitive Goal post-Arab Spring Arab “transition” for democratising countries after 2011 countries (IV) (III) Source: Inspired by Levy (2014), Khan (2010).

that ends violent conflict – and ter Europe, and Africa after they forms a base for a country to be- moved to a new, more competi- gin its development history. The tive settlement in the 1980s and analysis of “political settlements” 90s. Within each trajectory the is a vast subject (see in particular main developmental challenges the work of Mushtaq Khan and are very different for countries Douglas North). Settlements can in an early-stage than in the later be viewed as deals among elites stages – making for four idealized to create a ruling coalition that categories. can ensure peace (through vio- Democratic/competitive and lence and threats), and that gov- autocratic/dominant party re- erns by restricting organizations gimes develop different types through both formal and informal of institutions, with different rules (Khan, 2010). The quality strengths and weaknesses. But of a settlement can be judged by their economies do not fare dif- how costly it is for the economy ferently on average - the accepted to allow the coalition sufficient fi- wisdom these days being that nance to sustain peace (North et economic growth under auto- al, 2009). cratic and democratic systems The character of a “founding” has similar means but larger vari- settlement sets a country along ance under autocracy (Besley and one of two broad trajectories Kudamatsu, 2007). In this clas- (Levy and Fukuyama, 2010). The sification, all Arab regimes (save resulting political order could be Lebanon) are characterized by a organised around a dominant dominant ruler or party over the political party or leader and top- past 50 years or so, and belong to down nationalistic movements, the upper part of the graph. Their as happened in much of the de- growth rates over time have been colonized world in the 1950s and at least comparable to other de- 60s and in the Arab region. Alter- veloping countries. natively, it could be more compet- The second dimension of itive, as in India and Bangladesh the table relates to the quality of after their independence, and governance in each of these dis- in much of Latin America, Eas- tinct political paths, from basic Chapter 4 | 95

settlements that deliver security, Regimes and institutions to higher quality regimes. Vari- How do institutions evolve with- ous analysts have characterized in this political map? Political sci- low/high “quality” differently. entists, in their discussions, have The duality between “myopic” differentiated between the evolu- vs. “long horizon” rule dates back tion of autocratic and competitive to Olson (1993). Khan focuses paths. In the autocratic regimes, more on the degree of centralisa- authority very much emanates tion, with territorial and within- from the state, and improvements elite fragmentation weakening tend to be top-down, coming from power (Khan, 2010). Levy and central initiatives that roll down Fukuyama, Khan, and North, also the bureaucracy, in an attempt to emphasise a distinction between affect society and the economy. In formal/informal and personal- the early phase of Arab develop- ized/institutionalized relations. ment this took the form, for ex- In these various meanings then, ample, of SOEs, central planning the “good” governance regime and management, mass educa- would be associated with a devel- tion campaigns, iconic infrastruc- opment state that is more institu- ture projects and, in some cases, tionalized, long-lived and central- deeply transformational nation- ised, while a “good” competitive alizations of private interests and system would be close to a rule- land reforms. More broadly, Levy based sustainable democracy. (2014) analyzed 51 dominant We thus obtain, besides the systems, 39 intermediate, and 29 initial settlement, four idealized competitive systems, and found types of regimes, which fall short several key distinctions between of a sustainable democracy. Let the character of those two types us next characterise the character of regimes (Table 4.2). of each regime type and the insti- According to Levy, typical tutions associated with them. autocratic states do not trust the

Table 4.2: Two development trajectories

Dominant party Trajectory Democratic Trajectory Elite bargain organisa- Competitive among at least two long tion Dominant leader/party lived parties/groups Mode of decision- Decentralised, the result of coalitions. making Success on the “edge of chaos”. Islands Centrally coordinated, top-down, pos- of quality can arise if they get protected sibility of a development state from state predation Governance risks All actions emanate from the state – state capacity is central. No trust of the Institutional decay and clientelism. Po- private sector - Cronyism tends to be litical corruption in business. Electoral narrow. Succession risks are high fiscal cycles are initially deep Frontier challenges As economy becomes more complex, ability to open up for more markets and competition. As incomes rise, Stronger capacity of state institutions. ability to respond to new aspirations of Coordination among actors to beat the middle class tragedy of the commons Source: Adapted from Levy (2014) 96 | Political Transition and Economic Development

private sector as the latter tends to Second, the failure of the state- oppose central planning and may led modernisation of the econ- support the opposition. On the omy, and import substitution other hand, competitive systems strategies, pushed all countries to are more open to a competitive implement economic, but not po- private sector, but they make it litical liberalization starting in the difficult to control state machin- 1980s. The roll back of the state ery, which remains undermined was very large by international by political clientelism and rent- standards. Increasingly, rulers seeking, and they tend to be dom- resorted to divide-and-rule strat- inated by uncooperative behavior egies to stay in power, leading to and predatory risk. The question a weakening of the quality of gov- that is discussed next is: where ernance (and in particular in the does the Arab region fit into this rule of law and state capacity). typology and how have their in- So, after the liberalization reforms stitutions evolved over time? of the 1980s-90s, these countries tended to move to the left towards type II, as governance be- 4.3 Regime evolution came more myopic and oriented in the Arab countries to short-term survival in some countries, and more fragmented There are no precise methods for in others. measuring where countries fit in Third, popular frustration the above political space, but vari- with low quality governance, ris- ous heuristic claims can be de- ing repression and unmet social fended.1 First, all Arab countries aspirations among an increasing- (except perhaps Lebanon) ap- ly educated population, produced proached type I after the big state a political shock in 2011, when push post-independence These mass movements throughout the regimes were in some cases led Middle East attempted to shift the by a king (Jordan, Morocco, KSA), course of politics towards more and in other cases by the military competitive systems. Such jumps and/or unique parties, as in Iraq, from autocratic to more competi- Libya, Syria, Algeria. They were tive systems have also occurred organized around charismatic in- elsewhere in the past. Most po- dividuals who became presidents litical scientists seem to believe for life after having successfully that as countries get richer, they consolidated their grip on power will end up in a competitive pol- (Owen, 2014). In this, they were ity with more space for “voice”. similar to other state-dominant This case is powerfully devel- post-independence countries, oped in Acemoglu and Robinson such as Korea in the 1960s and (2012), who argue that without early 1970s, or contemporary political and economic inclusion, Ethiopia and Rwanda, countries societies are unlikely to be able where a flush of authority sets to become prosperous. While a country on the move after its the rise of China (or the GCC leadership has embraced a coher- for that matter) questions such ent and mobilizing strategy for general statements, it is undeni- development. able that countries that success- Chapter 4 | 97

fully traverse the earlier-stages governance challenges during of the ‘dominant state’ trajectory the transition more complex and have, over time, increasingly con- profound. The difficulty of confronted governance-related chal- solidating a competitive system lenges that push them to either is that once the flush of electoral restore lost power, or to open up enthusiasm has worn off, institu- their polity. Such pressures can tions capable of supporting con- arise both from failure and from tinuing competition are unlikely success. Failures have often been to be in place, given the autocratic associated with leadership that past. The result, as in Bangladesh ends-up using its authority for or Kenya, can be a polity in which narrowly self-seeking ends, thus the rules of the game are clientelis- breaking the “autocratic bargain” tic – and which teeters, seemingly and losing legitimacy. In much of indefinitely, on what Levy and the Arab world of the 2000s, there Fukuyama (2010) call the edge of was no ideological foundation chaos. Interestingly, it turns out left to ruling regimes, in sharp that this ‘edge of chaos’ can be contrast to the widely legitimate compatible with significant eco- and mobilizing ideologies of the nomic dynamism. But whether a 1960s. But pressure arises equally country can sustain its high-wire from success. Success in acceler- act for a sufficiently long time for ating economic growth, as hap- these gains to be consolidated – pened in much of the Arab world or whether the ‘edge of chaos’ is in the 2000s as a result of the a prelude to backsliding, is uncer- Structural Adjustment Programs tain. (SAPs) of the 1980s and 90s, re- The cases of Syria, Iraq, Ye- sulted in a more sophisticated men, and Libya illustrate extreme private sector, a growing middle cases, where a move away from class, and an emerging network autocracy has led to the explo- of civil society organizations – all sion of the past settlements. At of which seek institutions capable the other extreme, the GCC coun- of both the impartial resolution of tries have remained stable as they disputes and the provision of a reacted to the uprisings by in- platform for ‘level-playing-field’ creased patronage. In part, their competition. This enhanced social relative stability arises from their and economic complexity tends high per capita resource wealth, to come into increasing conflict which enables rulers to maintain with control-oriented political a modicum of support among and state institutions. their citizen populations. Except Fourth, the reaction of po- for Bahrain, where opposition litical elites to these mass move- mobilisation has been intense ments varied greatly among the (largely in reaction to the fact that countries of the region , setting a Sunni royal family rules over a them on widely different courses. Shi’a majority population in an The extent to which the sudden exclusionary fashion), the politi- political opening, that followed cal situation has remained calm. decades of dominant state poli- But while oil wealth has cush- ties, has unleashed previously ioned these regimes from making suppressed conflicts, makes the major changes in their political 98 | Political Transition and Economic Development

settlements for now, these settle- presidential elections in October ments increasingly pit the inter- and November 2014 – the quar- ests of the local business class tet that brokered this important against those of national labour as deal received the Nobel prize for the size of the oil rent (per capita) peace in 2015. The secular Nidaa falls over time. With youth unem- Tounis movement won both elec- ployment on the rise, opposition tions and formed a government has also risen. The goal of moving of national coalition. In all, there back closer to regime I, to avoid were therefore 3 presidents and 4 backsliding into a type II regime, cabinets between 2011 and 2015. will thus be increasingly harder. Egypt has also undergone In Tunisia and Egypt, the post- a difficult and protracted pe- revolutionary years have marked riod of political instability since a slide towards more competitive, 2011, marked by several major but also chaotic politics, which events: after the fall of Mubarak, have deeply marked the way in the military took charge for sev- which institutions inherited from eral months, before parliamen- the past have been able to func- tary elections were held early in tion. In Tunisia, following Ben 2012 and presidential elections in Ali’s departure from the coun- June 2012, bringing the Muslim try, a state of emergency was de- Brotherhood to power. A new clared. First, a caretaker coalition constitution was passed and the government was created, includ- Parliament was disbanded by ing members of Ben Ali’s party. the Supreme Court. Mass dem- Following more protests, the onstrations in 2013 brought the cabinet was reshuffled to remove army back to power, followed by them. Following further public the adoption of a new constitu- protests, Ghannouchi resigned, tion in January 2014, the election and Essebsi became Prime Min- of Sissi as president in May 2014; ister. On 3 March 2011, the presi- and parliamentary elections held dent announced the elections for from October to December 2015. the Constituent Assembly, which In Egypt too, therefore, there were held on 23 October 2011. were 3 presidents and 4 cabinets The Islamist Ennahda Party won between 2011 and 2015. The po- the plurality of seats and formed litical situation has seemingly a new government, in a coali- stabilized around a young strong tion with two junior partners. military ruler, with rising repres- Tunisia’s National Constituent sion amidst worsening security Assembly approved a new Con- conditions. stitution for the country on Janu- Fifth, as already apparent in ary 25, 2014. Following the assas- the comparison, countries have sination of political leader Brahmi reacted to the new situation in in July 2013, a political deadlock different ways, breaking the uni- ended with a compromise formity of politics that character- reached, between the country’s ized the Middle East in the past ruling and opposition parties, to (Owen, 2014). In the transition form a non-political technocrat countries, the two polar cases are cabinet led by Jomaa to hold the those of Egypt and Tunisia. Egypt fort until the parliamentary and is seeking to adapt to the new cir- Chapter 4 | 99

cumstances through a “reforma- to, since autocratic nationalism tion” of the old system -- moving came right after independence back from III to II and possibly to and decolonization. Unlike the I. Tunisia is seeking to find ways recent transition in South Korea, to adjust its politics and econom- Indonesia, or Mexico, the auto- ics to a more competitive system, cratic rulers had not prepared the as in IV. Political scientists have institutional ground for a more examined the reasons for this inclusive future. Newly formed political divergence – attributing competitive democracies face it variably to the existence of a challenges related to the lack of large army in Egypt, to the lead- adaptation to their new circum- ership of Tunisia’s main parties stances. There are no political and the active role of civil society parties to aggregate preferences in mediating dispute, and to the at the outset of the “jump”, or in- moderate qualities of Tunisia’s stitutions that can generate coop- Islamic opposition and its ability erative solutions. to find compromises (Cammett, Autocratic restorations also 2016). Jordan, Morocco, and Al- face challenges, albeit of a differ- geria have chosen an intermedi- ent type from those experienced ate path, with important political by young and fragile democracies. reforms initiated in order to pre- Examples, from the Former So- serve and gradually change the viet Union’s failed reformation to existing political settlement, but Pakistan’s unstable equilibrium, without major power loss in the point to the risks ahead in Egypt. short term. One can depict such The revolts have unleashed social a strategy as a gradual diagonal activism to a level unseen in the movement from regimes II to IV. past, and while the fear of chaos All these political strategies may have cooled these impulses present challenges and risks. in the short term, it has become ‘Jumps’ from dominant to com- much harder for regimes to stay petitive trajectories could pro- in place if a minimum level of sat- ceed relatively smoothly – as in isfaction is not ensured for impor- South Korea, where ‘later stage tant constituencies in a reformu- state dominance’ evolved fairly lated autocratic bargain. This new seamlessly into a competitive pol- reality can be seen in the dramatic ity. But it could also unfold more shift, over the past few years, of discontinuously – as in the Arab public opinion regarding popular region. Other examples of such values associated with respect for jumps include Eastern Europe authority. In the four Arab coun- color revolutions, and also Indo- tries for which the WVS has data, nesia and Thailand. The peculiari- we observe a steep fall in “respect ties of the Arab region in these re- for authority” between 2008 and spects are several. Unlike Eastern 2013 (see figure 4.1).2 Europe, the delegitimized institu- In the Arab region, therefore, tions of the past could not be re- the risks are high. While newly placed by ready-made new rules. competitive systems lack the in- Unlike the transition in Latin stitutions of checks and balanc- America, there is little in terms of es, the autocratic choice is con- historical tradition to come back strained by heightened fears of 100 | Political Transition and Economic Development

Figure 4.1: Respect for authority, 2008-2013 4.4 The evolution of the state and fiscal policy State capacity is needed to create

0.9 1 public goods – from the protection of property rights, to justice, education, or infrastructure and safety nets. The weaknesses of the state, in weak forms of authoritarianism, are quite different from 0.6 0.7 0.8 its maladies under weak competi-

Authority 2: higher value tive regimes. This section first means more respect for authority characterizes how institutions 4 5 6 fared as Arab regimes moved from type I to II, and the sudden Waves fall to regime III after 2011, before JOR JOR MOR MOR

IRQ IRQ EGY EGY contrasting the future challenges MIC MIC of Egypt and Tunisia in their at- Source: World Value Survey. tempts to move to regimes II/I and IV respectively.

the reaction from their “streets”. Both types of constraints can pre- From type I to type II regimes vent determined efforts to restart The last twenty years have seen economic recovery until political a slow transition in much of the crises are resolved. A downward region, from very large to much spiral may ensue, with polarized smaller and impoverished states. politics exacerbating economic After shooting up in the 1970s, difficulties, leading in turn to government expenditures fell more fractious politics, thus caus- precipitously in the 1980s before ing countries to slide into a transi- stabilizing in the 1990s at much tion trap (EBRD, 2013). lower levels. Many countries had To explore these issues in more just started to increase expen- detail, the analysis in the next 3 ditures again during the last oil sections focuses on Tunisia and boom of the 2000s. To get a sense Egypt, two countries where the of magnitude, one can calculate political path, as argued above, is that from the highest levels (in the likely to diverge. It will be argued, 1970s), by the early 1990s, state ex- as a result of this divergence, that penditures had been cut dramati- they are likely to face different cally – by 36% of GDP in Egypt, constraints on development – in and 16% of GDP in Tunisia, with the case of Tunisia, the main chal- the resultant state expenditures lenges will come from the state reduced to the range of 25-30% sector, while in Egypt, they will of GDP in both countries (Diwan come from the growth process. In and Akin 2015). Such a deep roll- both cases, however, macro-pres- back of state expenditures had, sures are likely to remain high, over time, a dramatic impact on albeit for different reasons. some of the key services offered by the state, especially in Egypt Chapter 4 | 101

where the roll-back was larger, While education has expand- and as a result public institutions ed fast, it is higher education that have declined steadily, and more has particularly expanded in the so in Egypt.3 more recent period. This has es- One of the most deeply affect- pecially been the case in Tunisia, ed items was public investment, where tertiary enrolment rates which collapsed in both coun- jumped from 8% of the age cohort tries, going from a height of 20% in 1980, to 19% in 2000, to a whop- of GDP in Egypt to about 10%, ping 36% by 2010. In Egypt and and in Tunisia, from 15% of GDP Algeria too, the rates of enrolment to about 5% (Diwan and Akin in tertiary education are nearly as 2015). The reduction in public in- high. The expansion of tertiary vestment affected public services, education may have kept the infrastructure, and thus the com- youth out of the street for a few petitiveness of the economy. The years, but at enormous costs for greatest cuts were typically for the long term, producing people groups with little political voice – with high expectations that can- such as the poor - and in lagging not be satisfied given the overall regions with little political weight, low demand for skills generated and from where political tensions in the economy, a rise in the edu- erupted after 2011, such as Sidi cated unemployed and thus high Bouzid in Tunisia’s neglected in- levels of frustrations among these terior, or the Sinai Peninsula in young and their parents, and a Egypt’s neglected periphery. decrease in the overall quality of The civil service wage bill was the education system. squeezed too, but less than pub- One of the consequences of lic investment. The challenge of state disengagement is that social large and underpaid state sectors grievances rose, and with them, is great in many countries, and a broad movement of opposition has negatively affected the ability was unleashed in the 1980s. In this of the state to deliver basic ser- period, and unlike other regions vices to the population. The situ- such as Latin America and sub- ation has become worse in Egypt Saharan Africa in the 1980s and compared to Tunisia – while in 1990s, economic crises in the Arab both countries, the civil service region did not provoke regime employs about 20-25% of the la- change during the near synchro- bour force, Egypt was spending, nous “third wave” of democrati- in 2010, only about 7% of GDP on zation. Rulers in the Arab coun- the wage bill, while Tunisia was tries did not open up the political spending 10%. Reduced expen- space in order to reduce social diture on health and education, pressure stemming from dwin- which were the main levers that dling economic resources. To the allowed the poor to escape their contrary, in 2010, the region was poverty trap in the past, has cre- less politically open than it was in ated a major source of inequality 1980, as the rise in grievances led of opportunity that has slowed governments to tighten their grip down social mobility [see Assaad on the” street”. Mounting op- and Kraft, 2016 for a good sum- position was met with the rising mary of this growing literature]. repression of certain groups, and 102 | Political Transition and Economic Development

the co-optation of others (see, for slow (as opposed by the unions) example, Henry and Springborg and the (formal) labour market 2010; King 2009; Posusney and structures remained rigid (thus Angrist 2005; Schlumberger 2007, slowing the development of ex- Cammett et al 2015). port industries). The “labour aris- Subsidies on consumer goods tocracy” of organized and public- suffered the greatest slash during sector workers was somewhat the adjustment episode. In Egypt, protected, while those forced into they fell from a staggering 23% the informal sector, unemploy- GDP at their height, to less than ment, or emigration have paid 2% of GDP at the end of the ad- the price of the rising dualism justment period, while in Tunisia, in labour markets - few govern- they fell from 11% of GDP to close ments, faced with strikes, street to zero. With the increase in en- protests, and rising opposition ergy prices in the 2000s, however, forces (mainly Islamist), would energy subsidies rose (Diwan and have risked alienating such a key Akin 2015). Unlike the past goal urban constituency as organized of supporting the poor, however, labour (Beinin and Vairel 2013; the more recent rise in subsidies Hibou, 2006). was an attempt by many gov- Thus, by 2011, the inheritance ernments to lure the rich and the of the past included an overstaffed middle class to continue support- and underpaid civil service, poor ing their weakened hold on pow- quality services - especially for er. These expenditures further the poor-, an oversized and poor shrank an already squeezed fis- quality university system, large cal space. Energy subsidies grew subsidies to the middle and up- over time, and by 2011 they were per classes, low levels of pub- much higher in the Middle East lic investment in infrastructure, than in any other region of the impoverished lagging regions, world. In 2010, these subsidies and repression of the opposi- represented about 9% of expendi- tion. Twenty years of decay have tures in Egypt, and 5% in Tunisia. deeply marked the institutions in- Other politically meaning- herited by the regimes that came ful constituencies, and particu- out of the Arab uprisings of 2011, larly powerful labour unions, whether in the education, secu- were also typically protected. rity, or macro-control areas, thus Organized labour has long been severely restricting their ability to a force in Tunisia and in Egypt reform their systems in ways that (as well as in Morocco and Alge- are adapted to the ideologies that ria). Although, trade unions did they embodied. not represent the vast number of workers employed in the informal sector, they were strong in larger enterprises and in the From regimes II to III after 2011 public sector. Keeping the social All of these problems have be- peace requires at least their acqui- come exacerbated by the move escence. This helps explain why to a more competitive political the civil service remained over- regime of type III, just after 2011. bloated, why privatization was In the new competitive environ- Chapter 4 | 103

ment, with political parties that that was immaculately clean un- remain fragmented and weak der dictatorship. governing coalitions, one would expect power to become myopic, and to easily fall prey to rent- What to expect in the future? seeking groups with a bit of voice, State capacity is thought to be of such as public sector unions, several types: vertical, territorial, street movements, and middle and horizontal (or “embedded” class groups unwilling to give up in the sense of Evans, 1995). Orga- “their” subsidies. In Egypt and nizational/vertical accountability Tunisia, these pressures accom- works best under dominant re- modated the fault-lines created gimes, but its territorial reach can by their recent past. Underpaid be low, particularly because of public servants pushed for, and the low political weight attached obtained, higher wages. Hiring to the political periphery. To con- in the public sector also rose. As front the loss of discipline and the a result, the civil servant wage bill parallel deterioration of public rose fast – from 11% of GDP in services brought about by the po- Tunisia in 2011 to 13% of GDP in litical transition, two methods are 2014; and from nearly 8% of GDP possible. The first is what Egypt is in Egypt in 2011 to nearly 9% of attempting, which is a restoration GDP in 2014. With the continua- of vertical accountability by tight- tion of high energy prices world- ening the internal controls in the wide, energy subsidies, which bureaucracy. The second method mainly benefit the middle class, is the development of horizontal also rose – from to 5% of GDP mechanisms, which can comple- in Tunisia in 2011 to nearly 8% ment vertical systems, or if these of GDP in 2013; and from nearly are weak, substitute them to a 9.5% of GDP in Egypt in 2011 to certain extent. The experience 11.5% of GDP in 2014. and pre-conditions for horizon- Although less apparent, both tal accountability mechanisms, countries also experienced a involving civil society and benefi- sharp fall in state capacity due ciary groups, do not exist early in to the inadequacy of inherited transition, and there is no access methods for disciplining state bu- to information to allow such pro- reaucracy, based on the increas- cesses to work. These can howev- ingly unworkable top-down ver- er be built over time as a result of tical accountability systems of the efforts by civil society to increase past, built around discretionary public sector accountability (Levy power, that were supposed to de- and Walton, 2013). ter a lack of performance. While What can be expected in the measures are lacking at this stage future for Egypt and Tunisia? In to document this deterioration, it Tunisia, it is likely that increased was most visible, for example, in political competition will lead to the disappearance of the police a situation where state capacity from the streets of Cairo during will decay further, in the short- 2011-13, and the accumulation to medium-terms, due to the in- of rubbish in the streets of Tunis crease in political clientelism and during the same period, a city the loss of central control. This 104 | Political Transition and Economic Development

is clearly demonstrated with the the education system, no quick continued rise in the civil service changes can be expected on these salary bill – the recent decision to fronts given the low starting point increase wages starting in 2016 levels throughout. In Tunisia, by an equivalent 0.7% of GDP therefore, the task of rebuilding (bringing the wage bill in 2016 to institutions on a different base, in- over 14% of GDP) broke one of cluding the development of new the central tenets of the IMF pro- of horizontal checks and balances, gram, and put external financing is likely to be the main binding and macro-stability in jeopardy. constraint to development in the Moreover, in spite of policy pro- near future. nouncements by all transitional In contrast, one would expect governments, the pressure to that Egypt would increasingly keep up with subsidies did not be able, as it moves (back) to a relent until, luckily, energy prices more dominant party system, to went down in 2014-15. reign over its civil service and On the service delivery front, its state system. Yet, in contrast while this loss of control has re- to Tunisia, the inheritance of the mained limited in Tunisia due to past will not help, as the state the more solid institutions it has bureaucracy is already way over- inherited from the past, one can staffed and way under-paid, and expect that the capacity of the the state will have more political state will continue to fall in the fu- space to resist pressure from lob- ture, because of pressure to con- by groups. This is already visible tinue to increase staffing, which with President Sissi’s early suc- will in turn increase pressure to cess in controlling wages in the underpay public servants. At the public sector (which fell to 8.3% same time, it is also likely that a of GDP in 2015) and in starting to more competitive and clientelis- reduce energy subsidies –energy tic political system will work to subsidies were already cut in half the advantage of more particular in 2015. While vertical controls interest groups who will manage are being re-instated, Egypt has to benefit from the state, at the ex- avoided any move to improve pense of programmatic policies horizontal accountability, (refus- that benefit the whole population. ing, for example, to even consider Therefore, improvements over introducing a freedom of infor- time are unlikely to happen until mation law when prompted by horizontal systems by beneficiary the World Bank to make this a groups become more developed. central condition in a large bud- Given that there is little histori- get support operation in 2012-13), cal experience with such mecha- and recentralising controls as op- nisms, the process is likely to be posed to moving towards a more long. Indeed, the introduction of decentralised system. It remains a freedom of information law, to be seen if this attempt to re- which is proceeding, has been impose top down systems will extremely slow. Moreover, while work in improving the quality of there are increasing discussions services over time. For Egypt, it on decentralization, support for is thus unlikely that control over lagging regions, and reforms of bureaucracy will be the binding Chapter 4 | 105

constraint to development in the these two factors describe the near term. “deals space”, which can be mapped to the political space in the sense that each of its “ideal” 4.5 The growth process cells is more likely to correspond and the evolution of state- to a particular political regime (see table 4.3). In particular: business relations • Type I. Ordered closed Arab States in transition have in- deals. Only those with po- herited hybrid economic systems litical connections get to of moderate performance - part make deals, and they can crony and part informal – a re- be confident that officials flection of a governance system will deliver. Such a situ- based on political and economic ation is likely to accom- exclusion. Regulations and insti- modate medium levels of tutions were tilted towards ac- economic growth. commodating the interests of the • Type II. Disordered close politically connected rather than deals. As political risk ris- serving the purpose of productiv- es, only those with politi- ity gains and job creation. As po- cal clout can make deals, litical regimes weakened, the pa- but with the high prob- tron-client relation became tilted ability that future regimes towards the interest of the clients will not honour deals, rather than put at the service of an deals become disordered. effective industrial policy. Such systems are likely to For private activity to grow have low growth. and be productive, investors need • Type III. Disordered open to know that they can obtain com- deals. Anyone can make mensurate returns for their ef- a deal, but no certainty forts. In low quality governance that officials will deliver. systems, where rules and institu- Firms feel the need to tions are weak, this presupposes make deals, but the return that they have deals with the is highly uncertain as they state that protect their property may face different offi- rights and possibly give them ad- cials in the future. Such vantages over competitors (Hall- systems are likely to have ward-Driemeir et al, 2010). For medium to low levels of investors to invest, therefore, two growth, because they are factors are determining. The first more inclusive than, even is the extent to which these deals though as corrupt as, type are ordered/disordered, i.e., how II situations. far they are respected inter-tem- • Type IV. Open ordered porally once they are made. The deals. Anyone can make a second is the extent to which they deal, and they can be cer- are allowed to enter into deals, tain it will be honoured. which depends on their identity, This is a world close to and how open/closed the deal rules, the only distinction space is. having to do with formal The four combinations of vs. informal rules. Such a 106 | Political Transition and Economic Development

system is likely to support During the move from state- high levels of economic run to market economies in the growth. mid-1980s, done under duress as Below, we describe how Tuni- the Arab state was in crisis, au- sia and Egypt moved from types tocrats faced rising opposition I to II after the liberalization of while simultaneously losing the the 1980s, and then after the Arab levers not just of ideology, but Spring of 2011 to zone III. We also also of ownership of state-owned show how the various institu- enterprises, large investment tions that support private sector budgets, large civil services, large growth fared during these tran- military and security spending. sitions, as well as the challenges Political stability required that the they face now, given the inherited emerging private sector be man- institutions of the past. aged politically in order to ensure its political neutrality, and avoid the risk of its support to the op- A recent history of the private position. Control over the private sector – from regime of type I in the sector was achieved by erecting 1970s and 80s to type II in the 2000s barriers to entry that excluded op- Let us first recount how the move ponents and provided privileges from I to II, which occurred in to a small group of trusted allies. the context of the 1980s’ liberal- While close state-business rela- ization, impacted the economy. tions are also prevalent in type I Essentially, it will be argued that regimes, as with South Korea’s state-business relations moved to General Park, the patron-client the left, from types I to II, as po- relation in type I are more easily litical risk rose and the capitalis- managed towards performance, tic base of the private sector nar- and are akin to successful indus- rowed, due to two inter-related trial policy (Khan, 2010). The dif- factors: first, politicians restricting ference in type II is that the bar- entry into the private sector to re- gaining power shifts to the client, duce opposition risk, and second, who demands higher returns, politically supported firms de- commensurate with higher politi- manding higher returns to offset cal risk. Moreover, in type I, the political risk. exclusionary aspect is not as im-

Table 4.2: The deals space

Disordered Ordered Closed (deals available Only those with political clout can Only those with political connections only to specific indi- make deals, even when they cannot be get to make deals, and they can be con- viduals/organizations) certain that officials will deliver. Type fident that officials will deliver. Type II – Crony base narrows and demands I – patron-client relation well managed, higher returns in the face of rising po- but economy not inclusive -> medium litical risk -> medium- low growth growth Open (deals depend on Anyone can make a deal, but no cer- Anyone can make a deal, and they can actions of agents but not tainty that officials will deliver (petty be certain that officials will deliver. identities) corruption). Type III –generalized cor- Type IV – near-rules world ->high ruption ->low growth. growth Source: Inspired by Khan (2010), Hallward-Driemeier, Khun-Jush, and Pritchett, 2010. Chapter 4 | 107

portant, leading to more inclusion of the labour force in the formal and growth compared to regime private sector. In Tunisia too, Ben II. Ali’s cronies became extremely Mechanisms to bestow privi- powerful and rich – with a wealth leges on allies and to exclude estimated at about one-quarter potential opponents included of Tunisia’s GDP in 2011 (when practices such as protecting mo- their assets were expropriated). nopoly rights, preferential access The confiscation affected 117 indi- to privatization, favoured access viduals and 400 enterprises who to government procurement (and were accused of illegally profiting land), and better treatment by from political connections. These regulators. Privatization tended firms accounted for a whopping to be modest and to advantage 21 percent of all net private-sector insiders. State monopolies were profits in 2010, although they pro- often replaced by private monop- duced only about 3 percent of pri- olies, which, unlike their prede- vate-sector output and employed cessors, did not care about social 1 percent of the labor force (Rijk- welfare and took full advantage ers et al, 2014).6 of their market dominance to reap The banking sector has played monopoly profits. an important role in supporting The networks of privilege the political regulation of mar- became narrower. While these kets. State banks still dominate networks existed in the pre-lib- the financial landscape of Egypt eralization period, they were sec- and Tunisia. Egypt saw a partial ondary to SOEs, and the private liberalization of its banking sec- sector operated as a complement tor, but even private banks en- to the public sector.4 After the lib- gaged in “name lending” to con- eralization period however, these nected interests. Partly because networks became both more im- connected groups developed portant, in terms of their control of their own banks, and partly be- the economy, and also narrower, cause truly private banks chose being concentrated around 20-30 to lend to the connected firms be- families close to the ruler (Owen, cause they were more profitable 2014). In the case of Egypt, Diwan (Diwan and Schiffbauer, 2015). et al (2014) identify around five As a result, the banks’ lending hundred firms controlled by thir- patterns became extremely con- ty-two businessmen with high- centrated, with more than 80% of level political connections, most credit going to connected firms of whom also occupied political in 2010 (Rocha, 2011). In Tunisia, posts in the mid-2000s.5 Between financial liberalization was even 2000 and 2010, these firms came less ambitious. Collusive behav- to dominate the Egyptian corpo- ior among bankers kept compe- rate landscape and monopolized tition levels low and profits high most profitable new economic (Henry and Springborg, 2010). opportunities among themselves. Although there have been some By 2010, these firms earned 60 regulatory changes as well as percent of the overall corporate some privatization, the state has profit in that sector, even though retained a dominant role. Indeed, they employed only 11 percent after three years of stimulus fol- 108 | Political Transition and Economic Development

lowing the Jasmine Revolution be estimated that Egypt’s growth of 2011, one of the main threats could have been 1 % larger per to the Tunisian economy became year, and labour demand 20% the very high number of non-per- higher over the past decade, in forming loans carried by the state the absence of privileges. Similar banks, which required a costly processes were at play in the oth- bail-out, with little in the way of er countries. In Tunisia, the sys- reforms to show for. tem developed a sharp distinction The main channel by which between on-shore and off-shore corruption can hurt growth is that businesses, with the first reserved it reduces the beneficial effect of for the politically connected, and competition – the dynamic forces the second allowed to operate that drive the private sector to in- with a system of low taxation in novate and to invest in processes order to boost exports. In reality, and assets that can help improve however, the growth of export efficiency and labour productiv- businesses was taxed by the low ity. The seminal work of Philippe competitiveness of on-shore ser- Aghion (2001) has shown how vices, and was unable to upgrade this dynamic process is attenu- over time (Cammett et al, 2015, ated when leading firms benefit chapter 9). from exogenous cost advantages The existence of large informal that cannot be overcome, leading sectors is also directly connected to market structures with a few to the regulatory environment. large but inefficient firms, and an Politically, with all formal sup- overly large number of inefficient port going to the loyal formal small firms. In the case of Egypt, private sector, the opposition was there is significant evidence that allowed to “hide” in the informal cronyism led to stagnant econom- sector, but with no access to regu- ic growth and poor job creation. latory support. At the same time, While their advantages allowed the labour market was regulated connected firms to increase their to protect the interest of elite la- profits, they were run less ef- bour, deepening the chasm be- ficiently than their competitors, tween the formal and informal because they were more capital- sectors. The share of employment intensive and were shielded from in micro-firms with less than five competition. This contributed to employees, dominated the pri- the misallocation of scarce na- vate sector, ranging from about tional savings and low growth. 40% in Tunisia to almost 60% in Moreover, Diwan et al (2015) Egypt. Micro-start-ups – firms of find that sectors that were domi- less than 5 years old and with less nated by a few crony firms were than 5 employees – accounted for less dynamic— they saw lower 92% of net job creation in Tunisia firm entry, had a larger “missing from 1996- 2010. Yet, the prob- middle” of medium-size enter- ability that micro-firms might prises, and overall did not create grow beyond 10 employees, 7 as many jobs compared to more years later, is only 3% in Tuni- competitive sectors. These effects sia. This limits the pool of young add up to a sizeable aggregate im- firms and thus the scope for net pact of cronyism on growth. It can job creation, and results in a small Chapter 4 | 109

formal private sector. Small firms duced growth since 2011. In Tu- tend to stay small, and large firms nisia, growth rates were negative stay large, resulting in a “missing in 2010, and below 1% in 2015. In middle”, the component of the Egypt, growth was 1.8% in 2011, private sector that tends to exhibit and 2.2% in 2014, before making a the greatest degree of innovation comeback to 4% in 2015. The low in other countries. growth is reflective of the inabil- The social inheritance of con- ity of governments to further ex- centrated capitalism has also been pand their spending (see next sec- highly problematic. Cronyism tion on macro), but the root cause made “capitalism” unpopular, of the fall in aggregate demand led to the emergence of a very and growth is the unwillingness rich “1 percent,” and the slow of the private sector to invest. In growth of the formal private sec- Tunisia, private investment has tor exacerbated labour market fallen to about 18% GDP since dualism, increasing the inequality 2011. In Egypt however, private of opportunities faced by young investment has fallen more deep- graduates; it also helped keep ly, from the relatively low level of unpopular regimes in power in about 14% of GDP in 2011, to 10- Tunisia and Egypt. For all these 11% in 2014-15. Part of the lower reasons, large swaths of the pop- investment reflects heightened ulation increasingly came to see security risks. But equally, lower cronyism and corruption, both private investment must also re- petty and grand, as the hallmark flect the breakdown in the “order- of economic liberalism and the ing” of deals. In the new regime source of many ills, including the III, investment is fully exposed to job deficit and the rise in inequal- predatory practices, and political ity, and it thus became a driving risk translates into heightened force of popular discontent. Many investment and property right indicators point to the deep polar risks. concerns with corruption – for example, Gallup data during 2009– 2012 in eighteen Arab countries Future challenges reveal that the perception of state The challenges are likely to be corruption was high and highly quite different in Egypt and Tuni- correlated with perceptions of sia, given the different evolution business corruption (Diwan and of their political systems. In the Nabli, 2013). case of Tunisia, the deals space, while disordered, is now more open. Compared to the previous From regime II to regime III after situation, one can thus expect 2011 gains, as the capitalistic base wid- Regime II did deliver some ens, once the security situation growth. In Tunisia and Egypt, improves. This is because unlike economic growth averaged 4 to the situation under Ben-Ali, the 5% a year after 2000, below the new political regime is unlikely to growth rates of successful MICs, be in a position to exclude firms but at about the average global based on their political leanings. growth. The move to III has re- Indeed, an economic reconcilia- 110 | Political Transition and Economic Development

tion legislation, currently under The situation in Egypt is quite debate, would pardon individu- different in this respect, given als who committed economic or that the government of President financial offenses under the Ben Sissi has chosen the route of re- Ali regime, if such individuals rebuilding state dominance, i.e., turn a portion of their illicit gains. moving back from III to II (and Improved regulations to ensure hopefully over time, to I). In such deals are more ordered will surely a situation, the state is unlikely to help in leading to even further in- trust the (unconnected) private creases in security, and thus in in- sector. Indeed, the government is vestments. Such reforms will not itself to implement a large-scale be easy, as reflected in the recent investment program, articulated reforms of the state-owned banks, around mega-projects mostly led which were recapitalized, but by government and SOEs, such as not privatized, mainly because the expansion of the capacity of of pressures from the national the Suez Canal, the construction labour union which was weary of a new mega-industrial zone, a of losing financing for SOEs. On major expansion of housing for 2 the other hand, the recent reform million public sector employees of the investment code, which around a new economic city in offers major simplifications rela- greater Cairo and the attraction tive to the past, is starting to re- of GCC FDI for the construction duce the tax differential between of several refineries and power- on-shore and off-shore systems. plants. Recent offshore oil and These reforms, operated on the gas discoveries will no doubt “edge of chaos”, are a testimony also attract large capital expen- to how competitive political sys- ditures. At the same time, the tems tend to be more attracted new government has taken its to competitive economics. Thus, distance from the old Mubarak Tunisia is likely to fare relatively cronies (Adly, 2014). This is un- well in private sector growth in likely to be sustainable, and is the future, possibly better than likely to be part of a disciplining in the recent past where it was strategy as Sissi develops his own stuck in regime II, with probable group of cronies. But if political further improvement over time risk remains high, a risk inher- as the quality of its private sector ent to this “restoration” strategy management improves from III that so far lacks broad popular le- towards IV. This should also be gitimacy (as reflected in very low the case in Morocco and Jordan, participation rates in the recent albeit to a lesser extent, as rulers 2015 parliamentary elections), it in these countries remain wary is likely that this new group will of truly independent private en- remain as narrow as in the recent trepreneurs and thus, will want past, and growth will remain low. to retain some “closeness” in the Thus, low economic growth is deal space. In the near term, there- likely to be the main constraint fore, progress is likely to continue faced by Egypt in the near future. on the private sector front, which Progress can be achieved if the is thus not likely to be the main government manages to traverse binding constraint to growth. the regime space back from II to I, Chapter 4 | 111

and to develop a larger class of in- that support the private sector do vestors, who can be managed for not adapt, and pressure to spend efficiency, including by fostering more as long as the institutions competition among them, so they supporting state expenditure are are pushed to deliver growth. not overhauled. Initially, governments in transition countries reacted with expansionary policies 4.6 Macro-management to smooth out the downturn, es- during and after regime pecially in the face of rising social demands and the high expecta- change tions generated by the uprisings. In Egypt and Tunisia, the eco- Public sector wages, subsidies nomic costs of the transition, and government investment which were described in chap- were increased in many countries ters 2 and 3, while not as dra- around the region. In the Gulf matic, have been high. In Egypt, countries, budgets expanded GDP annual growth collapsed to massively. Expenditure increased around 2% between 2012-14, as by over one-third in Saudi Ara- political and social turmoil dis- bia. In the oil importing coun- rupted domestic production com- tries, both external accounts and pounded by electricity blackouts budget balances deteriorated. By and energy shortages. Growth 2013, fiscal deficits in Tunisia and rose to 4% in 2015 largely driven Egypt (but also Morocco, Jordan, by rising consumption. Economic and Lebanon) had shot up. growth in Tunisia also slowed In Egypt, the fiscal deficit bal- down considerably following the looned at 13% of GDP in 2013, 2011 Revolution. A short-lived and international reserves plum- rebound in 2012 did not lead to a meted, while in Tunisia, it rose to sustained economic recovery. In 6% of GDP with reserves down 2013 and 2014, real GDP growth to 4 months of import. This was stood at 2.3 percent. Expansion in the result of a combination of in- services and administrative sec- creased spending and lower tax tors were offset by a declining revenues. In Egypt, public ex- private sector and low external penditure rose rapidly after 2011, demand, reflecting stagnation in from 29% GDP in 2011, to 34% in the Eurozone. The social tensions 2013, and 36% in 2014. Right af- and terror attacks that marked ter the revolution, subsidies and the first half of 2015, further nega- the public sector wage bill rose. tively affected activity, as growth By 2013, wages consumed 25% dropped to less than 1% in 2015. of expenditures (up from 18% in Transitions tend to produce larg- 2011). Expenditures on subsidies er deficits for a variety of reasons. reached 32% of total expendi- The slowdown in the economy tures in 2014. Energy subsidies reduces revenues, and spending were the main contributor, ac- tends to rise to reduce social ten- counting for nearly 25 per cent sion. In many ways, therefore, the of public spending (7 per cent of macro-picture is endogenous to GDP), equivalent to the combined the forces described above – low spending on health, education growth, as long as institutions and public investment. Tax rev- 112 | Political Transition and Economic Development

enues fell to 13% of GDP (Tunisia that both Egypt and Tunisia will 19%). Driven by increased spend- experience fiscal pressure in the ing on subsidies, wages and in- years to come, but for different terest payments, combined with reasons: in Tunisia, it will be dif- lower revenues, the fiscal deficit ficult to resist spending that ben- widened from 8.1% of GDP in efits particular high voice groups, 2010, to 10.6% in 2012, 13.7% in at least until institutions that ag- 2013, and 12.8 per cent in 2014. gregate demands, and face trade- In Tunisia too, fiscal offs, are developed (if ever); while policy has been biased towards in Egypt, it will be difficult to re- recurrent spending, while pub- duce macro-tensions due to low lic investment has suffered from economic growth rates. In Egypt, both budget constraints and bot- repression can be used to im- tlenecks in project execution. The prove macro-stability. In Tunisia, bulk of the adjustment was borne on the other hand, this is not pos- by lower capital spending, reach- sible. It will be necessary to build ing a record low in 2013 (1% of institutions that can support good GDP). The challenge of finding a budgets in the presence of elector- balance between fiscal consolida- al competition, as the building of tion and maintaining social peace checks and balances requires the is illustrated in the continued in- existence of various meso-level crease of the wage bill and subsi- institutions, that do not yet exist, dies until 2014, despite the freez- to represent and aggregate pref- ing of salaries, the hiring of public erences, to set the stage and rules servants and the gradual increase for productive bargaining and to of subsidized fuel prices. In 2014, support parties to make credible the wage bill reached 13.8 percent commitments through time. of GDP (against 10.6 percent in Indeed, by 2014-15, under 2010), and subsidies and transfers General Sissi’s stern watch, fiscal represented 7.2 percent of GDP consolidation was starting to be (against 3.6 percent in 2010). The pursued boldly to bring the defi- government’s recent decision cit below 10% GDP (primary be- (September 2015) to increase wag- low 2%). Expenditure fell to 33% es starting in 2016, by an equiva- GDP, with further plans to reduce lent of 0.7 percent of GDP, under- it to 30% in 2016, and to 26% in mines the efforts to slow down 2018 (IMF, 2015). Energy subsidy recurrent spending growth and reforms have been introduced progressively improve expendi- to bring energy prices closer to ture composition. Subsidies have cost recovery levels. The prices become a growing fiscal burden, of key hydrocarbons were raised reaching approximately 24 per- by more than 50 per cent in July cent of total spending. Fuel subsi- 2014, and electricity tariffs have dies reached two-third of the total been rising for both commercial subsidy budget in 2013, against and residential use under a plan less than one third in 2009, equiv- to eliminate power subsidies by alent to almost 90 percent of the 2018. A number of tax increases combined expenditures for health have been promulgated, includ- and basic education. ing wealth tax, a capital gains tax Our discussion above suggests for companies listed on the stock Chapter 4 | 113

exchange, a new property tax, expected for 3-4 elections at least. and an increase in VAT (to 14%). Over time, things improve as In Tunisia, by contrast, the voters learn, and institutions for disappearance of the autocratic checks and balances adapt. discipline of the past, means that In such situations, it is not un- balancing budgets has become usual for distributional flights to a subject of bargaining between push governments to delay stabi- fragile governments afraid of lization all the way to a financial street movements, and new- crisis. There are many examples formed interest groups, or reor- of descents into deep financial ganized old groups. These groups crisis before more responsible de- include young political parties, mocracies take root, for example workers’ unions, regional inter- Turkey 2000-01. Thus, the speed ests, and corporatist interests, all at which deficits can be resolved of which lack long-term strategic depend very much on the exis- horizons and are unable to prop- tence of institutions that distrib- erly access information, which fa- ute losses among social groups vours short term strategies based fairly and effectively. However, on myopia and a lack of coopera- and again, such institutions of so- tion. Exacerbating the situation cial dialogue and bargaining were in early transition, is the lack of not needed under autocratic rule institutions that can commit to a as such decisions were centralised particular sharing of the pie. Such around the ruler’s top circle. a situation is most likely to lead to In both countries, the balance a “grabbing equilibrium” where of payment has deteriorated, re- short-term interest and a lack of quiring increased external sup- trust predominate. port to maintain sustainability. As a result, stabilization will In Egypt, capital outflows, lower involve a war of attrition, espe- foreign direct investment and a cially when important agents worsening current account deficit, refuse to accept to pay for adjust- including a particularly affected ment costs and try to pass them tourism sector, have significantly on to other agents. Such situa- eroded foreign reserves. These tions are especially prevalent un- dropped from 6.8 months of im- der conditions of asymmetric ports end of 2010 to 2.5 months in information over the ability of June 2013. Tourist arrivals have agents to bear such costs, which yet to return to pre-revolution prolongs the attrition war (Ale- levels—9.9 million tourists visited sina and Drazen, 1989). A second in 2014 compared to 14.7 million type of phenomenon, which may in 2010, a 33 per cent decline. As a occur later, is that of fiscal elec- result, public debt has been rising toral cycles, with fiscal deficits rapidly, mainly through internal rising before elections. This effect borrowings. It reached 96% GDP tends to be larger when democra- in 2015. Heavy support from the cies are young, and to weaken in GCC, which is now at about $20 mature democracies, as voters be- billion a year, has made all the come more discriminating (Dra- difference since 2014, but this is zen, 2001). In messy democracies, far from sustainable, especially this electoral budget cycle may be given the recent collapse in oil 114 | Political Transition and Economic Development

prices. And while FDI has picked Tunisia and Egypt will be unable up again since the 2011 lows, to stabilise their macro-situation it remains small at 1.4 per cent before a financial blow-up. Tuni- of GDP and insufficient to pull sia still has a bit more debt space growth. Egypt launched a $1.5 than Egypt. Egypt, however, is billion international bond in June increasingly obtaining its external 2015, marking a return to the mar- support from the GCC at grant ket that is likely to be overused in terms, while Tunisia is borrow- the coming period given the large ing at more expensive terms from BoP deficits and the low level of creditors that are more sensitive foreign exchange reserves, which to issues of credit-worthiness. remain only slightly above the Thus, in the short term, the critical threshold of three months most immediate challenge in all of import-cover. transition countries remains eco- Tunisia’s balance of payment nomic stabilization, in order to has also deteriorated, but to a avoid an economic and financial lesser extent. Current account im- meltdown, which would further balances have risen since 2011, as complicate the political process. a widening trade deficit could not Building a package of measures be offset by net service imports, that reduces expenditures and notably tourism, due to the un- raises revenues, and that com- certain political and security situ- mands a minimum level of popu- ation. A more flexible stance in lar support is a tricky endeavour the exchange rate policy, together in the best of circumstances, and with heavy reliance on external fi- will be especially challenging nancing from international finan- in the current hyper-politicized cial institutions (IFIs), has helped environment of Tunisia. In the preserve an acceptable level of current circumstances, increased international reserves. External external assistance to Egypt can debt has increased since the revo- help smooth a return to lower fis- lution, reflecting the fiscal expan- cal deficit. In Tunisia, however, sion and the widening of the cur- external support of the state bud- rent account deficit. Public debt, get is unlikely to be useful in the mostly external, rose from 40.7% present conditions, as this would of GDP in 2010 to 44.8 percent in only reduce the internal pressure 2013 and 56% in 2015. In 2015, Tu- on the political settlement to be- nisia launched a $1 billion sover- come more efficient, and thus eign bond. A recent debt sustain- delay necessary political adjust- ability analysis shows that in the ments. Instead, support for the absence of reforms, public debt private sector is more likely to be could reach around 75 percent of useful. GDP by 2020. Already, the financing gap for 2016 is creating uncer- tainties, as the heavy reliance on 4.7 Conclusions IFIs confronts an inability by the The transitioning South-Med government to enter into condi- countries stand at a historical tionality agreements that it can crossroad in their development respect. history. In the past 50 years, they It is thus possible that both have been through two distinct Chapter 4 | 115

development phases and one tarian system. Each has structural long transition. State-led growth strengths and weaknesses. delivered fast development in The first area we focused on is the young states before reach- the rehabilitation of public servic- ing its limits in the 1980s. A long es – such as the health, education transition phase of low growth and social protection systems. led to a second phase, which saw Ideally, new governments with the maintenance of the old auto- broad popular support should cratic governance framework, the be able to redirect expenditures roll-back of the state, and a move towards social services and away towards an imbalanced form of from subsidies that benefit the capitalism. The second phase de- better-off, and to make tax sys- livered some growth, but it also tems more progressive while en- generated popular dissatisfaction larging the tax base. But improv- leading to an increasingly vocal ing service delivery and fighting opposition to this system. The petty corruption will require challenge of the third phase will increased public sector wages, be to find a development path which will be complicated by that performs better both eco- the large size of the civil service. nomically and politically. Here, we have argued that in the To avoid getting stuck in a de- short- to medium-terms, this will structive political-economy trap, be easier in Egypt than in Tunisia, it is imperative that economic even though, paradoxically, the performance improves. The coun- political demands will be higher tries that will find a virtuous de- in the latter than in the more re- velopment path are those that can pressed case of the former. The create sufficient political space to use of social accountability mech- implement the kind of reforms anisms, involving more active that can improve social services participation in checks and bal- and boost economic growth. ances by citizens group, munici- Unfortunately, our analysis has palities, and NGOs, will have to revealed that such triumphalist play a central role to support this scenarios are unlikely to arise im- transformation, a development mediately after transition. that is more likely over time in As we have seen, after decades Tunisia than in Egypt. in which the politics of the vari- The other set of concerns dis- ous countries in the region con- cussed, centres on the business verged towards high levels of environment and job creation. authoritarianism, the contours of The ideal road to a more dynamic the emerging political settlement private sector is well known and of the future are now diverging involves moving from a system in important ways in the region. of political regulation of the mar- The cases of Tunisia and Egypt, ket (by granting privileges to re- which we have analyzed in some gime supporters) to a system of detail, represent two polar cases: regulation based on efficiency one that is striving towards com- and job creation. This, we have petitive politics, and the other argued, is unlikely to happen in that is attempting a reformation both Tunisia and Egypt. In Egypt, towards a strengthened authori- a return to a dominant political 116 | Political Transition and Economic Development

regime will revive a mistrust for Notes the independent private sector. In Tunisia, more clientelistic politics 1. It should be possible to operational- are unlikely to allow for the quick ize the typology by measuring the adoption of new rules to govern competition axis using measures such as Polity or Voice in the World private activities, but there will Bank governance indicators, and the be more space for competition, quality of governance using indica- leading to gains in the expansion tors such as Rule of Law and Gov- of private activities in the near fu- ernment Capacity (again from the World Bank governance indicators) ture. – but this task is not pursued here. But while diverging in signifi- 2. This is measured as an index based cant ways, both regimes face high on two questions from the World risks and have certain common Value Survey Poll taken in 2013: “is greater respect for authority a good challenges. To gain popular ap- thing? Is obeying rulers an essential peal, they need to include fewer characteristic of democracy?” favours for elite capital - although 3. This is reflected in declines in the in- the challenge will be to convince dicator of state capacity, which declined in both Egypt and Tunisia af- large capital-holders to invest in ter 1995, in Tunisia from 0.5 to 0, and the future, rather than withdraw, in Egypt from 0 to -0.5 (Kaufman et as happened with the socialist al 2003). revolutions of the 1960s. To sta- 4. Pioneering studies in this field include, on Tunisia, Cammett 2007, bilise the “street”, the interests Hibou 2006; on Egypt, Sfakianakis, of the poor will have to be better 2004, Kienle 2001, 2004, Osman balanced with those of the mid- 2010, and Roll 2010. dle class, which compared to the 5. See also Chekir and Diwan 2012, and Acemoglu et al 2014. poor, benefited under authoritar- 6. See also Chekir and Menard 2012. ian bargains, but only in the sense of receiving a larger share of a small pie. Importantly, in order References to be able to find new “growth- pacts” that enlarge the economic Acemoglu, D., T. Hassan and pie, the main social actors will A.Tahoun (2014) “The Power need to be in a position to rep- of the Street Evidence from resent their interests, make com- Egypt’s Arab Spring”, NBER mitments and reach agreements, Working Paper No. 20665, No- and have the capacity to ensure vember. that commitments by all groups Acemoglu, D. and JA. Robinson are adhered to over time. In this, (2012) “Why nations fail: the the role of politics will be crucial origins of power, prosperity to allow the main social forces to and poverty”, vol. 4, Crown acquire a voice, and to find pro- Business. ductive ways to negotiate coop- Adly, A. (2014) “The Future of Big erative and superior outcomes. In Business in the New Egypt”, short, political transactions will Carnegie Middle East Center, have to play a more central role November. in shaping the future, compared Aghion, P., C. Harris, P. Howitt, to the earlier phases of develop- and J. Vickers (2001) “Compe- ment. tition, imitation and growth Chapter 4 | 117

with step-by-step innovation”, Middle East”, Fourth Edition, Review of Economic Studies, vol. Westview Press. 68, no.3, July, pages 467-492. Cammett, MC. (2007) “Globaliza- Alesina, A. and A. Drazen (1989) tion and Business Politics in “Why are stabilizations de- Arab North Africa: A Com- layed?”, NBER Working Paper parative Perspective”, Cam- No. 3053, August. bridge University Press. Assaad, R., and C. Kraft (2016) Diwan, I. and T. Akin (2015) “Fif- “Inequality of Opportunity ty Years of Fiscal Policy in the in Human Development” in MENA region”, ERF Working “The Middle East Economies Paper No. 914, May. in Times of Transition” edited Diwan, I., P. Keefer, and M. by Diwan, I. and A. Galal, In- Schiffbauer (2015) “Pyramid ternational Economic Association Capitalism: Political Connec- Series, vol. 155-V, Palgrave. tions, Regulation, and Firm Beinin, J. and F. Vairel (2013) “So- Productivity in Egypt” World cial Movements, Mobilization, Bank Policy Research Working and Contestation in the Mid- Paper No. 7354. dle East and North Africa”, Diwan, I., and M. Schiffbauer Second Edition, Stanford Uni- (2015) “Cronyism and Private versity Press. Banks in Egypt”, mimeo. Besley, TJ. and M. Kudamatsu Diwan, I., P. Keefer, and M. Schiff- (2007) “Making Autocracy bauer (2014) “On top of the Work”, Suntory and Toyota Pyramids: Cronyism and Pri- International Centers for Eco- vate Sector Growth in Egypt”, nomics and Related Disci- World Bank, mimeo. plines, London School of Eco- Diwan, I., and M. Nabli (2013) nomics and Political Science, “How Does the Arab Street May. View the Private Sector? – An Chekir, H. and I. Diwan (2012) analysis of the Gallup Polls “Crony Capitalism in Egypt”, 2009-2012 in 18 Arab Coun- Globalization and Develop- tries”, World Economic Forum. ment. Drazen, A. (2001) “The political Chekir, H. and C. Menard (2012) business cycle after 25 years” “Barriers to Private Firm Dy- in “NBER Macroeconomics namism in Tunisia: a Qualita- Annual 2000” edited by Ber- tive Approach”, World Bank nanke, BS. and K. Rogoff, MIT mimeo. Press, vol. 15., pages 75-138. Cammett, MC. (2016) “Transi- European Bank for Reconstruc- tion Experiences and Relevant tion and Development (EBRD) Lessons” in “The Middle East (2013) “Transition report 2013: Economies in Times of Transi- Stuck in Transition?” tion” edited by Diwan, I. and Evans, PB. (1995) “Embedded A. Galal, International Econom- autonomy: states and indus- ic Association Series, vol. 155-V, trial transformation”, vol. 25, Palgrave. Princeton University Press. Cammett, MC., I. Diwan, A. Rich- Freund, C. and L. Mottaghi (2011) ards, and J. Waterbury (2015) “Transition to Democracy”, “A political economy of the World Bank, Mimeo. 118 | Political Transition and Economic Development

Hallward-Driemeier, M., G. tegrating governance and Khun-Jush, and L. Pritchett growth”, World Bank Policy Re- (2010) “Deals versus Rules: search Working Paper Series No. Policy Implementation Un- 5196, January. certainty and Why Firms Hate North, DC., JJ. Wallis, and BR. It”, NBER Working Paper No. Weingast (2009) “Violence 16001, May. and social orders: a conceptual Henry, CM. and R. Springborg framework for interpreting re- (2010) “Globalization and the corded human history”, Cam- Politics of Development in bridge University Press. the Middle East”, Vol. 1, Cam- Olson, M. (1993) “Dictatorship, bridge University Press. Democracy, and Develop- Hibou, B. (2006) “La Force de ment” American Political Sci- L’Obeissance: Economic Poli- ence Review, vol. 87, issue 03, tique de la Repression en Tu- pages 567-576. nisie”, Editions de la Decouverte. Osman, T. (2010) “Egypt on International Monetary Fund the Brink: From Nasser to (IMF) (2015a) “Arab Republic Mubarak.”, Yale University of Egypt 2014 article IV Con- Press. sultation – Staff Report; Press Owen, R. (2014) “The Rise and Release and Statement by the Fall of Arab Presidents for Executive Director for the Life: With a New Afterword”, Arab Republic of Egypt” Harvard University Press. Khan, MH. (2010) “Political Set- Posusney, MP., and MP. Angrist tlements and the Governance (2005) “Authoritarianism in of Growth Enhancing Institu- the Middle East: Regimes and tions.” Unpublished manu- Resistance”, Lynne Rienner script, School of Oriental and Publishers. African Studies (SOAS). Roll, S. (2010) “’Finance Matters!’ Kienle, E. (2001) “A Grand De- The Influence of Financial Sec- lusion: Democracy and Eco- tor Reforms on the Develop- nomic Reform in Egypt”, I.B. ment of the Entrepreneurial Tauris. Elite in Egypt”, Mediterranean King, SJ. (2009) “The new authori- Politics, vol. 15, issue 3, pages tarianism in the Middle East 349-370. and North Africa”, Indiana Rijkers, B., C. Freund, and A. Nu- University Press. cifora (2014) “The Perils of In- Levy, B. (2014) “Working with dustrial Policy: Evidence from the Grain”, Oxford University Tunisia” World Bank, work- Press. ing paper. Levy, B., and M. Walton (2013) Schlumberger, O. ed. (2007) “De- “Institutions, incentives and bating Arab Authoritarianism: service provision: bringing Dynamics and Durability in politics back in” Effective States Nondemocratic Regimes”, and Inclusive Development Stanford University Press. Working Paper No. 18, Febru- Sfakianakis, J. (2004) “The Whales ary. of the Nile: Networks, Busi- Levy, B., and F. Fukuyama (2010) nessmen and Bureaucrats dur- “Development strategies: in- ing the Era of Privatization in Chapter 4 | 119

Egypt” in “Networks of Privi- lege: Rethinking the Politics of Economic Reform in the Mid- dle East” edited by S. Hey- demann, Palgrave Macmillan.

CHAPTER 5

Changing Course

5.1 Introduction development model has failed to meet the population’s aspirations In chapters 2, 3 and 4 of this re- for freedom and shared prosperi- port, the analysis was positive in ty. Seven years on, the model does nature. We attempted to identify not seem to have changed much. the consequences of the 2010 up- To embark on a new course, risings, assess the responses by the new strategy should aim at South-Med governments to the achieving the dual overarching demands of the rebels, and ana- objectives of consolidating the lyze the nature and likely evolu- post-transition democratic polity, tion of the political regimes. In on the one hand, while making this chapter, the analysis is of a the transition to a dynamic and normative nature. Its objective is equitable economy, on the other. to make recommendations that The two objectives are bound may be helpful in putting the together in a symbiotic relation- South-Med countries on the path ship; hence, for a successful out- towards shared and sustainable come, both challenges must be prosperity. addressed simultaneously. To make the analysis tractable, In outlining the features of the the structure of the chapter will new strategy, a few ideas guide mirror the previous chapters. We our thinking. First, we take it as will first deal with restoring and given that high economic growth sustaining macroeconomic bal- is a necessary but not a sufficient ances and economic growth, fol- condition for inclusive develop- lowed by a discussion of social ment. Core issues such as macro- concerns. In Section 5.4, we share stability, social protection, social some thoughts on the relation- mobility, decent employment and ship between economic perfor- quality of life are all important ob- mance and the political regimes, jectives in their own right. Second, followed by Section 5.5 on the it is true that the future of South- role of external partners. Section Med countries partly depends on 5.6 draws come conclusions. external factors, including region- Our starting point for consid- al conflicts, the state of global eco- ering the future of South-Med nomic activity and shifts in politi- countries is that the previous cal powers. However, success or

121 122 | Changing Course

failure is primarily determined ing fiscal deficits and allowing the by the interaction between inter- exchange rate to curb current ac- nal actors and the policies they count deficits. The programs also collectively opt to adopt. Finally, involved the phasing out of dis- short- and medium-term actions tortionary energy subsidies and may not always be compatible. the expansion of social protection And given that short-term mea- programs. By now, all five South- sures determine the contours of Med countries, except Algeria, future economic performance, it have a program with the Fund. is important that current policy The hope is that these programs actions do not jeopardise long- will bring about greater macro- term objectives. economic stability, even if they entail sluggish economic growth and high unemployment in the 5.2 Restoring and sustain- sort-run. Once macroeconomic ing macro-balances and stability has been restored, confidence will return and economic economic growth growth will pick up. The most pressing transitional Considering the choices challenges facing South-Med made, the best South-Med coun- countries include macroeconom- tries can do, going forward, is to ic stabilization, activating the capitalise on the lessons of the economy and addressing youth transition experiences to date. unemployment. In this section, Among these lessons is that fis- we attempt to draw short- and cal expansion, when such a deci- medium-term agendas to deal sion is taken, should be mostly with these challenges. Advocated in the form of greater public in- solutions may not necessarily be vestment, and not expansion of the only viable options, but even current expenditures. There is a then, we hope they will trigger vast literature on the positive im- a healthy debate about the best pact of investment in infrastruc- course of action in these countries. ture on growth and employment, which can be quite sizeable even in the short term. For example, Macroeconomic stabilization the evidence from Latin America The story told in chapters 2 and and the Caribbean suggests that 3 is that the uprisings were asso- every US$1 billion spent on in- ciated with deteriorating macro- frastructure project generates, economic balances, rising public directly and indirectly, about debt, sluggish economic growth 40,000 new jobs annually. This and rising unemployment. Ini- literature also finds that income tially, South-Med countries re- inequality tends to decline with sponded to social demands by higher infrastructure in both following expansionary policies, stock and quality (Calderon and but eventually they came to the Serven, 2004). Extrapolating the conclusion that they needed to analysis to Egypt and Tunisia, consolidate their macro-posi- the IMF (2011) estimates that a tions. The IMF came to the rescue, 1% of GDP spent on infrastruc- with its usual measures of reduc- ture could generate as many as Chapter 5 | 123

87,000 new jobs in Egypt and by central banks elsewhere. An 18,000 jobs in Tunisia in the absence of clarity on monetary short-term. policy objectives, means that Another lesson on the fiscal economic agents have no basis to front is that South-Med countries form their expectations about the would benefit from instituting future, thus raising potential ad- counter-cyclical fiscal rules to verse effects on investment and stimulate the economy at times of growth. slowdown and vice versa. In this On the interaction between vein, Elbadawi and Soto (2012) monetary and fiscal policies, argue that progress on the po- there are two concerns. First, litical front will make it easier to monetary policy in South-Med adopt such rules. But even then, countries have tended to be ac- the learning curve might be long commodative of fiscal policy. and painstaking. The Chilean fis- Fiscal dominance is known to cal rule was a culmination of a have such adverse effects as fu- long history of fiscal prudence, elling inflation and crowding out reflected in the achievement of the private sector in the credit cash surpluses for twelve con- market. It is therefore important secutive years and the creation, not only to establish rules about in 1987, of a fiscal revenue sta- government borrowing from the bilization fund financed by pro- central bank, but also to respect ceeds from copper sales. Until these rules. In Egypt, such a rule the situation is ripe for adopting exists but it is not adhered to by counter-cyclical fiscal rules, it the Central Bank. would still be useful to consider The second concern relates adopting Keynesian policies at to the lack of consistency be- times of economic downturn. tween monetary and fiscal poli- On the monetary policy front, cies. As explained in chapter 2, it was unfortunate that Egypt, South-Med authorities did not and Tunisia to a lesser degree, always adopt a coordinated fis- defended exchange rate tar- cal and monetary policy stance. gets until international reserves The two policies did not always were nearly exhausted. Impos- stimulate aggregate demand ing quantitative restrictions to or tighten economic activities. control exchange rate markets They often moved in opposite reveals to be both ineffective and directions, possibly exacerbat- costly. A more flexible exchange ing rather than resolving the rate at an earlier stage in the tran- macroeconomic problems. Only sition would have saved interna- when South-med countries tional reserves, motivated eco- signed agreements with the IMF nomic actors to adjust and eased did they seem to have followed the imports of much needed in- consistent fiscal and monetary termediate inputs. Subsequently, policies, albeit giving priority to after the central bank of Egypt macro-balances at the expense of had decided to let the exchange growth. rate go, it did not announce a new nominal anchor for monetary policy as commonly done 124 | Changing Course

Reigniting and Sustaining Job- GDP, in 2014. Creating Growth In the medium-term, sustain- Prior to the uprisings, South-Med able economic growth would countries performed relatively require, among other things, a well in terms of growth perfor- much more friendly and neu- mance (averaging 5.4 percent tral business environment than during the period 2003-2008). In South-Med countries have had the post-uprising period, they so far (as evidenced by their low grew only at half that level. How- ranking on the World Bank’s Do- ever, even if we assume that the ing Business Reports). Morocco political transition is a temporary has made progress in this regard shock and that past growth lev- in recent years, but no similar els will once again be achieved, progress has taken place in the that will still not be enough. Past other South-Med countries. On growth suffered from two main the contrary, there is plenty of shortcomings: (1) it was not based pre-uprising evidence suggest- on the contribution of a dynamic, ing that the private sector in the efficient and outward-oriented region flourished because of its private sector, and (2) it was not cosy relationship with govern- sufficiently inclusive, at least ment (World Bank, 2009). Partial in terms of job creation. Future attempts to improve the business growth strategies should avoid environments (e.g., adopting a these shortcomings. In addition, new investment law in Egypt in the strategy should be based on a 2016) may be inadequate to en- two-pronged effort: a short-term tice the private sector to make strategy aimed at stimulating long-term investments. What growth and a medium- to long- South-Med countries need is a term strategy aimed at sustaining package of reforms that dramati- economic growth. cally increases competition, en- In the short run, even in the sures that the regulations are fair presence of an IMF austerity and transparent and provides program, expansionary policy credible commitments to protect could involve frontloaded vi- property rights. able labour-intensive infrastruc- This is not to suggest that the ture projects that are already in role of the state should diminish. the pipeline. It could also target The role of the government as a the elimination of binding bot- regulator should be stronger. So tlenecks, for example, electric- should its role as the provider of ity. Such policies will not only social services, which the private provide wage employment in sector is unwilling or unable to the short run, but also enhance provide. Health and education long-term growth prospects as are prime examples. The distri- well as sustained job creation. butional role of the state through This policy contrasts the trend fiscal and other policies is criti- in South-Med countries with re- cal. More generally, govern- spect to public investment in the ments also provide the rules that past few years, which declined ensure the efficient functioning from 6.6 percent of GDP, in 2012, of markets and the safeguarding to a record low of 4.2 percent of against abuse of market power. Chapter 5 | 125

To play their role effectively, the duration of infant industries however, South-Med govern- until they grow (Galal, 2008, Ga- ments need to deal with their lal et al, 2015, Sekkat, 2010). This bloated and often inefficient bu- policy also seems to have en- reaucracies. couraged rent-seeking behavior In addition to the business and cronyism, two symptoms of environment and a realignment the pre-uprising regimes. In the of the state’s role, an important post-uprising era, South-Med source of economic growth that countries have an opportunity, at remains untapped in South-Med least in countries making inroads countries is that of export and di- into political openness, to revise versification, especially of trade- their industrial policy to truly able goods and services. Here, promote output diversification the real exchange rate1 can be and achieve sustainable growth. an effective policy instrument. Of course, sustained econom- For small and medium size low- ic growth would also require and middle-income developing well-educated human capital, countries, where there are some- higher savings and stable in- what limited opportunities for flation. However, the broader significantly increasing total fac- question is how to make growth tor productivity in the short-run, more inclusive in South-Med a real undervaluation of the ac- countries, especially in terms of tual exchange rate has been pro- creating jobs to absorb youth un- posed, in the recent growth and employment. The reason for sin- development literature, as a via- gling out the youth (15-24 years ble option for an export-oriented old), is that their rates of unem- development strategy. Engineer- ployment are among the highest ing real currency undervalua- in the world. In Egypt, they action would promote exports and counted for close to 30% of total lead to sustained growth transi- unemployment in 1991. By 2014, tions (see for example, Aghion et the figure jumped to 42 percent al, 2006; Aguirre and Calderon, (figure 5.1). In Tunisia and Jor- 2005; Elbadawi et al, 2008; and dan, the ratios hovered around Rodrik, 2008). 30 percent. This phenomenon is In addition to the exchange largely a problem of youth inser- rate policy, South-Med coun- tion into the labour market. tries also need to revise their To address the problem of industrial policies (selective in- youth unemployment in the tervention). Thus far, the evi- short run, South-Med govern- dence points out that this policy ments could launch Active La- has essentially been successful bour Market Programs (ALMPs). in South-East Asia, but not else- They could explore the possibil- where (World Bank, 1993). For ity of supporting employment in South-Med countries, the evi- small and medium enterprises dence suggests that industrial by giving them benefits tied to policy has suffered from being greater employment, especially fundamentally inward looking, for new job seekers. These incen- promoting old rather than new tives could take the form of tax products, with no time limits on exemptions or reductions, credit 126 | Changing Course

Figure 5.1: Youth unemployment as a share of italise on the abundant factor of total unemployment (percent) production in developing coun-

45.0 42.0 tries, which is labour. Accord- 39.5 40.0 ingly, a combination of policies 33.7 that do not favour capital at the 35.0 30.8 31.8 29.4 28.8 30.0 expense of labour, an industrial 25.0 20.0 20.2 policy that favours manufactur- 20.0 17.7 ing and agriculture, and mea- 15.0 sures to promote exports, are 10.0 likely to induce growth in job cre- 5.0 ations. Short of such policies, the 0.0 Algeria Egypt Jordan Morocco Tunisia informal sector becomes the option of last resort. In this sector, 1991 1994 Source: World Development Indicators, World Bank. employment tends to be characterized by low productivity, low wages and certainly disagreeable guarantees, or even contributions working conditions (e.g. Assaad, to social and health insurances. 2002; Galal, 2002; Elbadawi and More broadly, governments Loayza, 2008). could ease the burden of unemployment by introducing unemployment insurance schemes. 5.3 Towards more effective Additionally, they could facili- social protection and social tate the negotiation process on productivity-based wage agree- mobility ments between organized labour In response to social demands, and the private sector. South-Med governments have In the medium to long run, taken measures to improve social growth will create jobs as long protection and, to a lesser extent, as the incentive structure for eco- the delivery of health and educa- nomic agents does not favour tion services. They have started capital at the expense of labour. phasing out energy subsidized In addition, the recent literature and have expanded programs makes a strong link between in- that target the poor. Some coun- dustrialization and job-creation tries have increased public ex- in developing economies. For penditure on health and educa- example, a recent report by the tion, while others have allowed United Nations Industrial Devel- greater private sector provision opment Organization (UNIDO, of these services. On the face of it, 2009) documents the experiences these measures constitute a move of countries adopting labour- in the right direction, especially intensive manufacturing-based as commodity subsidies tend to development. It finds that grow- benefit the rich and middle class ing manufacturing sectors can more than the poor. According to create jobs at a spectacular level the IMF (2012), only 20% of com- and at higher wages. Moreover, modity subsidies reach the poor. manufacturing is usually gender Notwithstanding the merits neutral. of the above measures, they do Similarly, exports tend to cap- suffer from several shortcom- Chapter 5 | 127

ings. The targeting programs terms of the goodness of existing leave out too many poor fami- systems, if any, resource avail- lies. For example, the poor rep- ability and implementation ca- resented about 28 percent of pacity. the population in 2016, whereas With respect to health and the cash transfer program was education, significant reforms to cover only one-fourth of that are necessary to enable citizens percent. Certain social protection to participate in the fruits of eco- measures, e.g., increasing em- nomic growth and to enhance ployment in bloated bureaucra- the possibility of upward social cies, were politically motivated mobility. The challenge facing at the expense of efficient out- South-Med countries is to fig- comes. The measures have also ure out the best way to provide lacked the institutionalization or equal access to good quality upgrading of the critical compo- health care and education. Here nents of social protection, such as the literature is very rich, but it health insurance, social security does not offer a blueprint. For coverage, and unemployment example, Finland is the world’s insurance (Razzaz, 2013). Equal- best performer in education, ally important, the education and though its education system is health services remain of poor highly centralised and govern- quality and are not universally ment driven. In contrast, the US accessible. Accordingly, further system performs reasonably well and more systematic reforms are with a system in which the gov- necessary to strengthen the safe- ernment and the private sector ty nets and to enhance the oppor- operate side-by-side, at a decen- tunity for upward social mobility tralised level. In the health sec- in South-Med countries. tor, both the US and Canada pro- For the next generation of so- vide good quality services, albeit cial protection reforms, two is- following two very different sys- sues stand out. The first relates tems. In Canada, the government to the need to rapidly expand the is a single buyer, while in the US cash transfer programs to cover the citizens are the buyers. What as many eligible families as pos- seems to matter most is that the sible. Although it is not easy to system is internally consistent identify the poor and to continu- and that it responds to the needs ally update the records as con- of a given society. Given the di- ditions change, slow progress versity of options and the varia- on this front can be costly to the tions in the initial conditions in poor themselves, as well as to the South-Med countries, each of level of support for further eco- them should develop its own nomic reform. The second pillar strategy, while recognizing that of social protection concerns cov- increased expenditure may be erage for the elderly, for sickness a necessary, but not a sufficient, and for job loss. Here, South-Med condition to bring about the re- countries need to design or rede- quired improvements. sign their protection schemes in Finally, because social protec- a manner consistent with their tion and social service reforms af- initial conditions, especially in fect all citizens as recipients and/ 128 | Changing Course

or providers, they cannot be car- reaches the conclusion that they ried out successfully without the represent two polar cases: Tuni- wide participation of different sia is striving towards competi- groups in society. Besides able tive politics; while Egypt is at- leadership, South-Med countries tempting a reformation towards would benefit from a social dia- a dominant regime on account of logue to build informed consen- terrorism. Morocco and Jordan sus on society’s objectives, the were not analyzed in detail, but trade-offs associated with differ- gradualism on both politics and ent options and the best instru- economics has been the hallmark ments to achieve the objectives. of their behavior so far. In Al- geria, there has been no notable change in the political regime. 5.4 Twinning economic Under these conditions, the nor- reform and politics mative question is what should be done to move towards a suc- The analysis of the evolution cessful transition and a virtuous of the transitioning South-Med development path. countries in Chapter 4 points out On economic reforms, a list of that these countries stand at a his- suggestions has been drawn up torical crossroad in their devel- in the previous sections. Here, opment history. After decades of we attempt to explore how the convergence towards high levels governments in Egypt and Tu- of authoritarianism, the contours nisia are likely to move forward. of the emerging political settle- Next, we briefly discuss how the ment in these countries is now di- “street” or voice can support the verging. To avoid becoming stuck move towards a successful tran- in a destructive political-economy sition. trap, it is imperative that South- Let us consider first the busi- Med countries work towards im- ness environment and job cre- proving economic performance ation. The ideal road to a more to create sufficient political space dynamic private sector is rela- to implement political reforms. At tively well known and involves the same time, they need to build moving from a system of politi- inclusive political and economic cal regulation of the market (by institutions to empower citizens granting privileges to regime and to make further economic re- supporters) to a system of regu- forms possible. Moving on both lation based on efficiency and tracks would create a virtuous job creation. It was argued in development path, which could Chapter 4 that this kind of re- eventually lead to shared pros- form would be difficult in both perity. Tunisia and Egypt. In Egypt, a Unfortunately, the analysis return to a dominant political re- of chapter 4 was not optimistic gime will bring back the mistrust about the prospects of South- of the independent private sec- Med countries moving directly tor. It will be the responsibility into competitive political sys- of the political leaders to adopt tems any time soon. The in-depth confidence-building measures analysis of Tunisia and Egypt to persuade the private sector to Chapter 5 | 129

engage in large-scale investment withdraw, as happened with the and economic activity. In Tu- socialist revolutions of the 1960s. nisia, the clientelistic politics of To stabilize the “street”, the in- the past are unlikely to support terests of the poor will have to be a quick adoption of new rules better balanced with those of the to govern private activities, but middle class, which compared there will be more space for com- to the poor, benefited under au- petition, leading to gains in the thoritarian bargains, but only expansion of private activities in in the sense of getting a larger the near future. share of a small pie. Importantly, Next, we may consider the re- in order to be able to find new habilitation of health, education “growth-pacts” to enlarge the and social protection systems. economic pie, the main social ac- Ideally, new governments would tors will need to be in a position enjoy broad popular support and to represent their interests, make therefore be in a position to redi- commitments and reach agree- rect expenditures towards social ments, and have the capacity to services and away from com- ensure that commitments by all modity subsidies. They should groups are adhered to over time. also be able to make taxation In this, the role of politics will more progressive, while enlarg- be crucial in allowing the main ing the tax base. However, these social forces to acquire a voice, reforms would be met with op- and to find productive ways to position from the rich elite and negotiate cooperative and supe- bureaucrats. Improving the de- rior outcomes. In short, political livery of social services would transactions will have to play a also require higher civil service more central role in shaping the wages and the end of petty cor- future, compared to the earlier ruption. The question is, which phases of development regime is likely to be able to Also, the future of political achieve this? It was argued in and economic development in chapter 4 that these types of re- transitioning South-Med coun- forms would be easier to imple- tries cannot be divorced from the ment in Egypt, in the short run, “voice” of the “street”. The new than in Tunisia, because of the rulers now have to reckon with divergence of the two political the emerging capacity of citizens regimes. In the long-run, Tunisia to use their newly-earned free- is likely to fare better than Egypt dom to voice their economic and thanks to the emerging social political preferences, and to at- accountability mechanisms and tempt to influence how and who the participation of different citi- should be governing them or zen groups, municipalities and managing their economies. This NGOs. ‘voice’ can arguably shape the In both countries, to have pop- political economy debate, which ular appeal, they need to include will ultimately determine the fu- fewer favours for elite capital - ture of both polity and the econo- although the challenge will be to my in South-Med countries. convince large capital-holders to Thus far, in the early stages of invest in the future, rather than the transition, the deteriorating 130 | Changing Course

conditions in terms of security countries, it is important to es- of individuals and communities, tablish a system capable of de- social services, economic growth livering inclusive and sustain- and unemployment have pro- able development. To this end, duced assertive but uncoordinat- South-Med countries would ed and, often times, contradic- benefit from a trajectory of insti- tory popular demands. Largely, tutionalizing plural voices. Suc- voice was manifest in terms of cessful development strategies ‘demands for more’ subsidies require the understanding and and higher wages for the em- endorsement of the majority of ployed by the most organized so- the population. The participa- cial groups. Realizing the power tion of policy-oriented research, of these groups, the authorities civil society organizations and in the post-transition countries informed public opinion makers have been responsive at the exis key. In turn, these individuals pense of worsening fiscal deficits and organizations can play the and spiralling public debts. Not role of informing various stake- only has this response complicat- holders, offering them a menu ed the prospects for future eco- of choices that reflect their own nomic performance, but it might interests and those pertaining to also have signalled a return to a broader national interests. This new style of bargain, akin to the process will not only help to ill-fated ‘authoritarian bargain’. identify the winners and losers Looking ahead, the impact of various policy options, as well of voice on the future of polity as the trade-offs involved, but it and economics in these societies will also make information avail- hinges on the strategies of the able to the public so that they can social groups involved and the hold governments accountable responses they elicit form the for their commitments. newly elected transitional authorities. Notwithstanding, the underlying risks associated with 5.5 A possible role voice, when it is used as a force for external actors for ‘uninformed populism’ or ‘subnational allocation of rents’, Abstracting from the role that could be a potent driver for plu- external actors can play on the ral democratic participation. This political front, they can help ease could lead to a ‘demand for dif- the transition process in South- ferent’ and transformative new Med countries through different social contracts with radically channels: finance, technical ad- different ‘rules of the game’: the vice and as anchors for reform. constitutions, laws and regula- Thus far, South-Med countries tions, and the underlying public have received some support in discourse on political and eco- all of these areas. The question nomic freedoms, social and eco- is whether the support came in nomic justice, and the future role time, at the levels required and of the state (Razzaz, 2013). whether it was associated with To reap the benefit from po- appropriate reform conditions. In litical transition in South-Med discussing these issues below, we Chapter 5 | 131

take it for granted that it is in the these turned out to be resources interest of external partners to see that IFIs were committed to de- South-Med transition countries liver on their own anyway. The move towards more open and fact remains, however, that ex- prosperous societies. ternal actors kept the South-Med On finance, the uprisings countries afloat at a time when were associated with significant they needed support the most. financing gaps because of grow- Of course, the provision of ing fiscal deficits and diminish- financial support was supposed ing foreign exchange receipts. A to be a temporary measure, un- combination of social pressure, til the economies of South-Med capital outflows and low tour- countries recovered. To ensure ism receipts gave rise to what that recovery was sustainable, seemed to be a temporary need support was sometimes accom- for external borrowing or grants panied by conditionality and from abroad. The euphoric at- sometimes not. For example, mosphere at the time of the Arab grants from the GCC to Egypt Spring prompted the interna- were not conditional on under- tional community to put togeth- taking certain reforms. Rather, er what came to be known as the they were provided in support of Deville Partnership. The GCC the political transformation after also came to the rescue, mostly the ousting of the Moslem Broth- in Egypt and to a lesser extent erhood from office. However, in Jordan and Tunisia. So did the GCC were active in identi- the IFIs. The IMF provided sup- fying certain projects that were port at different times to four out deemed socially worthy of sup- of the five South-Med countries port. At the other end of the spec- studied in this report. The World trum lies the IMF support, which Bank intensified its engagement was conditional on undertaking in South-Med countries, sup- reforms to reduce the budget and porting safety-net programs, current account deficits. Some infrastructure projects, sectoral would argue this supply of soft developments and occasionally money from the GCC may have structural reforms. So did region- delayed the reform process in al development banks, including Egypt, while others would criti- the African Development Bank cise the IMF’ conditionality for and the Arab Fund for Economic squeezing these economies when and Social Development. There economic growth was sluggish was also official support on a bi- and unemployment high. But it lateral basis. is ultimately the responsibility of Whether the level of finan- policymakers in different coun- cial support came in time or at tries to diagnose the problems the level required is question- they have to deal with and to fig- able. There was a discussion at ure out the best course of action. the beginning of the Arab Spring Moreover, the IMF conditional- of a Marshal plan for the region, ity has been the subject of a long which never materialized. The and an intense debate for some Deville Partnership promised time, and this is not the place for more than US$30 billion, but reiterating different views. 132 | Changing Course

Let us now turn to the topic the transition in South-Med of external bindings as an anchor countries and there may be some of reform. The argument here is room for improving their sup- that countries on their own may port. However, the destiny of not be able to make credible com- the transition in the South-Med mitments to reform, and it may countries will be determined at be useful to use external anchors home, through a dynamic pro- as a substitute. Accession to the cess involving different groups/ EU, for example, has served as a institutions and their relative critical basis for reform in aspir- power. ing countries, including countries of the former Soviet Union and Turkey. In contrast, South- 5.6 Conclusions Med countries have not benefited from such a mechanism. This chapter has attempted to The association agreements be- propose a normative road map tween EU and South-Med coun- for South-Med countries to set tries have been imperfect (Galal, their economies and politics on 2015). They have taken the form a trajectory of sustainable and of shallow, rather than deep, shared prosperity for all citizens. Free Trade Agreements (FTAs). It was argued that the prospects They have focused on the liberal- of making inroads in the future ization of barriers at the borders will require a break with previous (through lower or no tariffs), development models, which have with no commitments to reforms failed to meet the aspirations of behind the borders (for example, the population for freedom and competition policy, public pro- shared prosperity. Furthermore, curement, intellectual property the new strategy should aim at rights, etc.). The benefits from achieving the dual overarching shallow FTAs tend to be modest objectives of consolidating the (depending on the magnitude of post-transition democratic pol- trade creation versus trade diver- ity, on the one hand, while mak- sion), while reforms behind the ing the transition towards a dy- borders tend to generate signifi- namic and equitable economy, on cant economic gains. In terms of the other. The two objectives are coverage, the agreements have bound together in a symbiotic re- included neither agriculture nor lationship; hence, both challenges trade in services. Instead, they must be addressed simultaneous- have been limited to manufac- ly in order to achieve a successful tured goods for which Southern outcome. countries have limited compara- To this end, it has been sug- tive advantages. Furthermore, gested that South-Med countries the dicey issues of labour mo- ought first and foremost to deal bility (or migration) and capital with the pressing challenges of flows have not been addressed. the day, including restoring mac- Notwithstanding these res- roeconomic balances, reigniting ervations, the main point is that economic growth and working external actors can play a posi- to reduce unemployment, espe- tive role in easing the pains of cially among the youth. Both fis- Chapter 5 | 133

cal and monetary policies need were available, as if this would better coordination, clarity of cure the problem. What is need- objectives and the more effec- ed is a major rehabilitation of the tive use of counter-cyclical rules. education and health systems In terms of growth, they need to that involves not only the build- significantly improve the busi- ing of schools and hospitals, but ness environment, primarily by a holistic approach centred on getting rid of privileges and cro- students with greater autonomy nyism, providing better protec- and accountability on behalf of tion of property rights and taking the service providers. measures to enhance competi- The above recommenda- tion. They also need to pay more tions are reasonably well known. attention to promoting export, The problem lies in the capacity especially in relation to their and willingness of the emerg- real exchange rate policy, along ing political regimes in Tunisia with a revision of their industrial and Egypt to undertake these policy to ensure that it serves reforms. Here, it was argued well the objective of long-term that Egypt, with an emerging re- diversification. On employment, gime of political dominance on the short-term actions include account of fighting terrorism, is active national labour programs likely to be able, in the short-run, and the support of SMEs. In the to make better progress in the so- medium-term, growth is likely cial agenda than Tunisia, where to create jobs if the incentive the emerging regime is essential- structure does not favour capital ly moving toward competitive at the expense of labour, if the politics. In the medium-term, economy is open and export-ori- Tunisia is likely to do better ented, and if manufacturing and thanks to the plurality of voices agriculture receive the attention and accountability mechanisms. they deserve. Governments in both countries On the social front, it was are likely to face difficulties in noted that South-Med interim getting rid of crony capitalism governments have been respon- in the short run. And both coun- sive to popular demands. They tries may have to strike a bargain have adopted several measures with the private sector if they are to do away with energy subsi- to entice them to invest heavily dies, along with other measures and engage more actively in eco- to strengthen safety nets for the nomic activities. poor and to improve the deliv- Perhaps the biggest change ery of social services. However, in the wake of the uprisings is the measures undertaken to pro- that the new rulers now have tect the poor leave out too many to reckon with the voices of the deserving families. They also “street”. The new rulers now fall short of offering protection have to respond to the emerg- for the elderly, the sick and the ing capacity of citizens to use unemployed. Furthermore, they their newly-earned freedom to seem to have focused too greatly voice their economic and politi- on increasing expenditures on cal preferences and to attempt to social services, where resources influence how and who should 134 | Changing Course

be governing them or manag- low a new development path. ing their economies. This ‘voice’ In line with the slogans during can arguably shape the political the uprisings, the new develop- economy debate, which will ul- ment path should cater for the timately determine the future of majority of the population and both polity and the economy in provide an incentive structure South-Med countries. Initially, that rewards hard work. Beyond these voices asserted their de- economics, it is also a model of sires in what have often been an open society, with more toler- uncoordinated and often contra- ant values and norms. There is dictory popular demands. In the no room in the new development future, the hope is that different model for extending privileges to social groups will be drivers for small elites, tolerating cronyism plural democratic participation. or neglecting social justice. The Also, that they will elicit, from call is for bold changes, not mar- the newly elected transitional au- ginal ones. Are South-Med coun- thorities, a ‘demand for different’ tries willing and able to meet the and a new transformative social challenge? The answer remains contract, with radically different to be seen. rules of the game regarding the underlying public discourse on political and economic freedoms, Notes social and economic justice, and the future role of the state. 1. The real exchange rate concept ad- Finally, on the possible role opted in this chapter refers to the of external actors, they have price of tradeables to non-tradeables. A decline (rise) in the RER provided South-Med countries means appreciation (depreciation); with some support in the form and the RER is overvalued (under- of finance, technical advice and valued) when its lower (higher) to some extent as an anchor for than its corresponding “notional” equilibrium level. reform. However, it is doubtful that this support came in time, at the level required to smooth the transition or that it was suf- References ficiently credible to make reform possible. At any rate, external Aghion, P., P. Bacchetta, R. Ran- actors cannot substitute for local ciere and K. Rogoff (2006) actors who will ultimately de- “Exchange Rate Volatility termine the future of their coun- and Productivity Growth: The tries. Role of Financial Develop- Overall, South-Med coun- ment” NBER Working Paper tries are at a crossroad. They no. 12117, May. hold an enormous potential, Aguirre, Á and C. Calderón that has been held back by mod- (2005) “Real Exchange Mis- est economic performance and alignments and Economic extractive political institutions. Performance”, Central Bank of The uprisings, which began in Chile Working Papers No. 315, December 2010, provide these April. countries an opportunity to fol- Assaad, R. ed. (2002) “The Egyp- Chapter 5 | 135

tian Labor Market in an Era of Working Paper No. 67, March. Reform”, the American Univer- International Monetary Fund sity in Cairo Press. (IMF) (2012) “Subsidy Reform Calderón, C. and L. Servén (2004) in the Middle East and North “The Effects of Infrastructure Africa: Recent Progress and Development on Growth and Challenges Ahead” Income Distribution”, World ____ (2011) “Regional Economic Bank Policy Research Paper No. Outlook – Middle East and 3400. Central Asia”, April. Elbadawi, I., L. Kaltani and K. Razzaz, O. (2013) “The Treacher- Schmidt-Hebbel (2008) “For- ous Path Towards a New Arab eign Aid, the Real Exchange Social Contract”, Issam Fares Rate, and Growth in the Af- Institute for Public Policy and In- termath of Civil Wars”, World ternational Affairs, November. Bank Economic Review, vol. 22, Rodrik, D. (2008) “The Real Ex- no. 1, January, pages 113-140. change Rate and Economic Elbadawi, I. and N. Loayza (2008) Growth”, Brookings Papers on “Informality, Employment Economic Activity. and Economic Development Sekkat, K. ed. (2010) “Market Dy- in the Arab World”, Journal of namics and Productivity in Development and Economic Poli- Developing Countries: Eco- cies, vol. 10, no.2, July. nomic Reforms in the Mid- Elbadawi, I. and R. Soto (2012) dle East and North Africa”, “Resource Rents, Political Springer. Institutions and Economic United Nations Industrial De- Growth”, ERF Working Paper velopment Organization No. 678, May. (UNIDO) (2009) “Industrial Galal, A. (2015) “Rethinking the Development Report 2009 – EU’s Mediterranean policy af- Breaking In and Moving Up: ter the Arab Spring”, Europe’s New Industrial Challenges World. for the Bottom Billion and the Galal, A., H. Selim, I. Atiyas, L. Middle-Income Countries”. Achy, S. Ghali, A. El-Haddad, World Bank (2009) “From Privi- et al. (2015) “Structural Trans- lege to Competition: Unlock- formation and Industrial Pol- ing Private-Led Growth in icy: A Comparative Analysis the Middle East and North of Egypt, Morocco, Tunisia Africa” and Turkey”, FEMISE – ERF World Bank (2004) Report published by the Euro- World Bank (1993) “The East pean Investment Bank. Asian Miracle – Economic Galal, A. Ed. (2008) “Industrial Growth and Public Policy” Policy in the Middle East and North Africa: Rethinking the Role of the State”, the American University in Cairo Press. Galal, A. (2002) “The Paradox of Education and Unemploy- ment in Egypt”, the Egyptian Center for Economic Studies