SYRIA ECONOMIC MONITOR Lost Generation of Syrians Spring 2022 Syria Economic Monitor Lost Generation of Syrians Spring 2022 Middle East and North Africa Region © 2022 International Bank for Reconstruction and Development/The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org This work is a product of the staff of The World Bank. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy, completeness, or currency of the data included in this work and does not assume responsibility for any errors, omissions, or discrepancies in the information, or liability with respect to the use of or failure to use the information, methods, processes, or conclusions set forth. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be construed or considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions This work is available under the Creative Commons Attribution 3.0 IGO license (CC BY 3.0 IGO) http://creativecommons.org/licenses/by/3.0/igo. All queries should be addressed to the lead author, at: gmele@worldbank.org. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. Cover photos courtesy of: (top) Doctor examines child malnutrition. Credit: Mohammad Bash / Shutterstock (center, left) Town near Palmyra, Syria. Credit: Fly_and_Dive / Shutterstock (center, right) Syrian refugees families from Kobani district. Credit: Orlok / Shutterstock (bottom) Syrian refugees children studying in the classroom. Credit: Tolga Sezgin / Shutterstock Further permission required for reuse. Publication design and layout by The Word Express, Inc. TABLE OF CONTENTS Acronyms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi ‫الملخص التنفيذي‬ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii 1.  The Conflict Context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 2.  Recent Economic Developments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Deteriorating economic situation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Persistent current account deficit and dwindling foreign reserves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Currency depreciations and surging inflation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Weakening fiscal position . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Declining living standards and rising food insecurity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 3.  Outlook and Risks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .25 Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .25 Risks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .27 Special Focus: Demographic and Labor Market Consequences of the Syrian Conflict . . . . . . . . . . . 31 Technical Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .37 Counterfactual GDP Calculations for Syria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .37 Nowcasting Economic Activity Using Nighttime Lights . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .42 Connectedness between the Syrian and Lebanese Pounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 Average Exchange Rates for Syria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .63 The Exchange Rate Pass-Through in Syria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .65 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 iii List of Figures Figure 1 Syria: Economic Development at a Glance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii Figure 2 Conflict-Related Casualties in Syria by Event Year, 2017 and 2021 . . . . . . . . . . . . . . . . . . . . . .2 Figure 3 Change in Population Density, Syria, 2010–2020 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Figure 4 Syrian Refugees by Hosting Countries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3 Figure 5 Internally Displaced Persons (IDPs) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Figure 6 Worldwide Governance Indicators, Syria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Figure 7 Actual and Counterfactual GDP, Syria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Figure 8 GDP Impact of the Syrian Conflict in the Mashreq Region . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Figure 9 Night-Time Lights and GDP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Figure 10 Economic Activity in Syria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 Figure 11 Investment in Syria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11 Figure 12 Dynamics of Syrian Trade . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 Figure 13 Shipping Density Map . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12 Figure 14 Seaborne Trade Volume of Syria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13 Figure 15 Syria’s Balance of Payments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Figure 16 Fuel Prices and Fuel Imports in Selected MENA Countries . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Figure 17 Connectedness between the Syrian and Lebanese Pounds . . . . . . . . . . . . . . . . . . . . . . . . . . .16 Figure 18 World Bank Estimated Average Exchange Rate in Syria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 Figure 19 Inflation in Syria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Figure 20 Rolling Estimates of the Contemporaneous Effect of the Exchange Rate on Inflation . . . . . 21 Figure 21 Syria’s Fiscal Budget . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 Figure 22 Share of Subsidies in Budget Expenditures in Syria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .22 Figure 23 Syria’s Salary Levels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 Figure 24 Food Insecurity in Syria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .24 Figure 25 COVID-19 in Syria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .28 Figure 26 Syria Population Pyramid, 2010 and 2021 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .32 Figure 27 Masculinity Ratio in 2021, by Age Group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 Figure 28 Trends in under-Five Mortality Rate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 Figure 29 Number of Absent Members for Households Currently Residing in Syria, by reason of absence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 Figure 30 Changes in Working Age Population, 2010–2021 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 Figure 31 Relation between Female Labor Force Participation and Masculinity Ratio at the governorate level, 2021 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 Figure 32 Trends in Labor Force Participation and Unemployment Rates, by Gender . . . . . . . . . . . . . .35 Figure 33 Actual and Counterfactual GDP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .40 Figure 34 Actual and Counterfactual GDP, Adjusted for Regional Shocks . . . . . . . . . . . . . . . . . . . . . . . . 40 Figure 35 Map of Syria with NTLs and Known Flaring Sites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Figure 36 Evolution of Flaring, Non-Oil, and Total NTLs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 Figure 37 Evolution of Non-Oil NTLs by Subnational Regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 Figure 38 Per Capita NTLs by Subnational Region . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 Figure 39 Historical NTL Time Series (1992–2021) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 Figure 40 Historical NTLs and Real GDP (1992–2021) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 Figure 41 Model Predictions of Real GDP from NTLs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 Figure 42 Model Predictions of Real GDP by Subnational Region . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 Figure 43 NTL-Based Gini Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 iv SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS Figure 44 The Levels of LBP and SYP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 Figure 45 The Returns on LBP and SYP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 Figure 46 Autocorrelation and Lead-Lag Relation between the LBP and SYP . . . . . . . . . . . . . . . . . . . . .53 Figure 47 Time-Varying Correlation between the Returns on the LBP and SYP . . . . . . . . . . . . . . . . . . . 54 Figure 48 Time-Varying Conditional Correlation between the Returns on the LBP and SYP . . . . . . . . .54 Figure 49 Conditional Volatility of LBP and SYP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 Figure 50 Quantile-Quantile Plot of the Returns on the LBP and SYP . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 Figure 51 Total Volatility Spillover Index: LBP and SYP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 Figure 52 Directional Volatility Spillovers (FROM) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .58 Figure 53 Returns on the TRY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 Figure 54 Volatility of LBP, SYP and LBP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 Figure 55 Total Volatility Spillover Index: LBP, SYP and TRY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 Figure 56 Directional Volatility Spillovers: LBP, SYP and TRY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .61 Figure 57 Gasoline and Diesel Prices in Lebanon are among the Very Lowest in the Region in the pre-Subsidy Termination Era . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Figure 58 Share of Critical Goods Imports in Syria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 Figure 59 Weight Structure of the Average Exchange Rate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 Figure 60 Average Exchange Rate in Syria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 Figure 61 Time Series Dynamics of the Official, Market and Average Exchange Rates . . . . . . . . . . . . .65 Figure 62 Response to a 1% Shock in the Change in the SYP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .67 Figure 63 Forecast Error Decomposition for Inflation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .67 Figure 64 The Price to Exchange Ratio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 Figure 65 Exchange Rates and the CPI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 Figure 66 Rolling Estimates of the Contemporaneous Effect of the Exchange Rate on Inflation . . . . . 70 Figure 67 Rolling Estimates of the Contemporaneous Effect of the Exchange Rate on Inflation from the second Specification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 Figure 68 Rolling Estimates of the Contemporaneous Effect of the Exchange Rate on Inflation from the third Specification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 List of Tables Table 1 Regression Results of Historical NTLs and Real GDP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Table 2 Granger Causality Tests with the Levels and Returns of the LBP and SYP . . . . . . . . . . . . . . .17 Table 3 Estimates of the Exchange Rate Pass-through Using the Simple Approach . . . . . . . . . . . . . 20 Table 4 Cumulative Effect of an Exchange Rate Depreciation Based on VAR Models . . . . . . . . . . . . 20 Table 5 Subsidies by Items . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .22 Table 6 Macro Outlook Indicators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 Table 7 GDP Level Predictor Means – All Pretreatment Outcome Values Used as Predictors . . . . . 39 Table 8 GDP Growth: Panel Data Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .40 Table 9 Regression Results of Historical NTLs and Real GDP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 Table 10 Granger Causality Tests with the Levels and Returns of the LBP and SYP . . . . . . . . . . . . . . .52 Table 11 Estimates of the Exchange Rate Pass-through Using the Simple Approach . . . . . . . . . . . . . 66 Table 12 Cumulative Effect of an Exchange Rate Depreciation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 Table 13 Cumulative Effect of an Exchange Rate Depreciation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 Table 14 Pass-through Regressions: Estimation Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 Table of Contents v ACRONYMS ACLED Armed Conflict Location & Event Data FLFP Female Labor Force Participation Project FTS Financial Tracking Service AER Average Exchange Rate G7 Group of Seven AIS Automatic Identification System GARCH Generalized Autoregressive Conditional BIC Bayesian Information Criterion Heteroskedasticity BoP Balance of Payments GAUL United Nations Global Administrative CBS Central Bank of Syria Unit Layer CGE Computable General Equilibrium GBV Gender-based Violence CoVDP COVID-19 Vaccine Delivery Partnership GDP Gross Domestic Product COVID-19 Corona Virus Disease 2019 GGFRP Global Gas Flaring Reduction CPI Consumer Price Index Partnership DESA Department of Economic and Social GMM Generalized Method of Moments Affairs GNI Gross National Income DMSP US Air Force Defense Meteorological GSCI Goldman Sachs Commodity Index Satellite Program GTAP Global Trade Analysis Project DWT Deadweight Tons HDI Human Development Index EIA Energy Information Administration HeRAMS Health Resources and Services EMODnet European Marine Observation and Data Availability Monitoring System Network HNAP Humanitarian Needs Assessment ERPT Exchange Rate Pass-Through Programme ESCWA Economic and Social Commission for IDPs Internally Displaced Persons Western Asia ILO International Labour Organization EU European Union IMF International Monetary Fund EUR European Monetary Unit IRFs Impulse Response Functions FAO Food and Agriculture Organization LBP Lebanese Pound FDI Foreign Direct Investment MENA Middle East and North Africa FEVDs Forecast Error Variance Decompositions vii PREFACE T he Syria Economic Monitor is a new semi- tools and data, most notably “big data.” For example, annual economic publication on the Syrian Arab in this Syria Economic Monitor issue, we use innova- Republic, produced by the Macroeconomics, tive geospatial and remote-based data sources (e.g., Trade and Investment (MTI) Global Practice of the World nighttime lights and nighttime lights-based output Bank. The aim of this series is: (1) to provide an update estimates, shipping-position data, and population and on key economic developments, outlook, risks, and conflict maps) to help us draw more informed infer- policies, and situate them in the conflict context; and ences about economic developments in Syria. Through (2) to present findings from the World Bank’s recent these various data sources, including household sur- analytical work in Syria (these are found in a Special veys conducted by humanitarian organizations on the Focus section). The Syrian Economic Monitor is part ground, we are able to quantitatively analyze a range of a more general effort by the MTI Global Practice at of issues, such as: (1) the decline in the male working the World Bank to better understand economic and age population and its impact on the labor market for social dynamics in fragile, conflict, and violence (FCV) Syrian women; (2) Syria’s COVID-19 performance in settings, notwithstanding the lack of physical in-country light of its severely degraded health care system fol- access in some cases. Conflicts are the dominant lowing the decade-long war; and (3) exchange rate source of development regression and are projected dynamics and their impact on inflation, where we by the World Bank to account for up to two-thirds of the estimated the exchange rate pass-through effect in extreme poor by 2030 (Corral et al., 2020). Hence, in Syria, as well as checked the connectedness between addition to the launch of the Syria Economic Monitor the Syrian and Lebanese pound. From this analysis, series, the MTI MENA Global Practice is also launching we gained a better understanding of the depreciation two additional FCV Economic Monitor series, the Libya and inflation trends in Syria and their drivers. Economic Monitor and the Yemen Economic Monitor. The Syria Economic Monitor was prepared by Economic monitoring in FCV contexts presents a team comprising Luan Zhao (Senior Economist, unusual challenges, not least of which is the lack of Task Team Leader), Silvia Redaelli (Senior Economist), reliable, timely, and comprehensive data. Syria, for Ibrahim Jamali (Senior Consultant), Sherin Varkey example, ranked last among the 146 surveyed coun- (Senior Health Specialist), Ali Ibrahim Almelhem tries on the World Bank’s Statistical Capacity Indicator (Economist), Ola Hisou (Consultant), Deyun Ou (SCI). To overcome this serious limitation to economic (Consultant), Priyanka Kanth (Health Specialist), monitoring, we made use of previously unavailable and Katriel Friedman (Consultant). The Special ix Focus Chapter, “Demographic and labor market Leader), Harun Onder (Senior Economist), Gianluca Mele consequences of the Syrian conflict,” was prepared by (Senior Economist), Naoko C. Kojo (Senior Economist), Silvia Redaelli. Ibrahim Jamali wrote the background Ha Nguyen (Senior Economist), Nadia Fernanda Piffaretti notes for “Connectedness between the Syrian and (Senior Economist), Wissam Harake (Senior Economist), Lebanese Pounds” and “The exchange rate pass- Naila Ahmed (Senior Social Development Specialist), through in Syria.” Ali Ibrahim Almelhem prepared the Majid Kazemi (Economist), Mohammad Al Akkaoui background note, “Nowcasting economic activity using (Economist), and Cyril Desponts (Economist), for nighttime lights.” Ola Hisou and Luan Zhao co-wrote invaluable discussions and comments during the prepa- the background note, “Average exchange rates for ration and review of this report. Wholehearted thanks Syria.” Deyun Ou and Luan Zhao prepared the analysis go to Robert Tchaidze (Senior Economist, International of “Counterfactual GDP calculations for Syria,” based Monetary Fund) for his advice and peer-review support. on Harun Onder’s background paper prepared for The team is grateful to Zeina Khalil (Senior External the World Bank’s “The Fallout of War” report. Sherin Affairs Officer), who led on report publishing, com- Varkey, Katriel Friedman, and Priyanka Kanth prepared munications, and outreach, and to Barbara Yuill, for the box, “COVID-19 continues to threaten Syria’s health editing the report. Special thanks to Ekaterina Georgieva system.” The World Bank’s Find My Friends statistical Stefanova (Senior Program Assistant) and Muna Abed tool, developed by Faya Hayati (Senior Economist), was Salim (Senior Program Assistant) for administrative applied to benchmark Syria’s investment performance. support, and to Abdullah Alruwaishan for the Arabic The Syria Economic Monitor builds on several translation of the Executive Summary. Last but not analytical works conducted by the World Bank since least, the team is grateful to Robert W. Reinecke, Salem the onset of the conflict in Syria, in particular, a series Massalha, and Shehab El-Dien (design, communication, of one-off studies that aimed at better understanding and videography experts) for their impeccable work on the economic and social impact of the Syrian conflict. the formatting, design and video/infographic contents These are The Toll of War, issued in 2017, which docu- related to the dissemination of this report. mented the economic and social impact of the conflict The findings, interpretations, and conclusions inside Syria, The Mobility of Displaced Syrians, issued expressed in this Monitor are those of World Bank in 2019, which analyzed the spontaneous returns of staff and do not necessarily reflect the views of the Syrian refugees to determine the key factors that influ- Executive Board of The World Bank or the govern- enced their decisions, and The Fallout of War, issued ments they represent. in 2020, which examined the human, physical, social, Although all efforts have been made to improve and economic destruction from the conflict in Syria on the accuracy of the information that was collected and the country’s neighbors in the Mashreq region. The analyzed, the assessment was produced in a quick Syria Economic Monitor also benefits from a Data timeframe to ensure the relevance of the estimations. Corps Strategic Brief that advised on data collection This is a living document and will be updated as new and analysis for Syria’s macro monitoring, prepared information becomes available.” by Holly Krambeck (Program Manager), Benjamin P. For information about the World Bank and its Stewart (Senior Geographer), Oleksandra Postavnicha activities in Syria, including e-copies of this publication, (IT officer), Rochelle Glenene O’Hagan (Data Scientist), please visit https://www.worldbank.org/en/country/ Han Wang (IT officer), Changamire Anderson (Senior syria/overview#1. We very much hope this new series IT officer), and Gabriel Stefanini Vicente (Consultant). will be helpful to a wide range of stakeholders. We wel- The authors are grateful to Saroj Kumar Jha come comments and suggestions on how to improve (Country Director), Eric Le Borgne (Practice Manager), future issues (e.g., on data availability, future topics of Johannes G. Hoogeveen (Practice Manager), Rekha interest). For questions and comments on the content Menon (Practice Manager), Kevin Carey (Adviser), of this publication, please contact Luan Zhao (lzhao1@ Christos Kostopoulos (Lead Economist), Paul Moreno- worldbank.org), or Eric Le Borgne (eleborgne@ Lopez (Acting Lead Economist), Fatima Zehra Shah worldbank.org). Questions from the media can be (Senior Operations Officer), Kamel Braham (Program addressed to Zeina Khalil (zelkhalil@worldbank.org). x SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS EXECUTIVE SUMMARY N ow moving into its twelfth year, the conflict dramatically since the conflict began. With a severely in Syria has inflicted a devastating impact degraded health care system following the decade- on the inhabitants and the economy. The long war, Corona Virus Disease 2019 (COVID-19) conflict accelerated infrastructure depreciation by has only exacerbated the vulnerable situations. The damaging strategic assets (the destruction channel) number of new cases has started to decrease since and deepened demographic aging by displacing many March 2022. Yet, the high cumulative case fatality rate people (the displacement channel). In addition, the indicates the inability of the health system to cope with conflict eroded social cohesion, degraded governance, the needs of COVID-19 patients. COVID-19-associated and led the division of the previously integrated areas deaths are relatively high in Syria, partially due to a in Syria (the disorganization channel). Together, these slow vaccine rollout. As of May 14, 2022, only 9.1 channels have halved the size of economic activity percent of Syrians were fully vaccinated, and another between 2010 and 2019. The report estimates that 5.2 percent were partially vaccinated. in the absence of the conflict (the counterfactual), Beyond the immediate impact of the con- Syria’s Gross Domestic Product (GDP) in 2019 would flict, the economy suffers from the compounding have been US$ 38.6 billion in 2015 constant prices, effects of the pandemic, adverse weather events, compared to US$ 16.3 billion of the realized GDP. regional fragility, and macroeconomic instability. Furthermore, the massive refugee flows triggered by Since 2020, Syria’s external economic ties have the Syrian conflict, combined with other spillovers, been severely restrained by the deepening crisis including through trade and financial channels, have in neighboring Lebanon and Turkey, as well as the imposed a heavy economic and social toll on Syria’s introduction of new United States (US) sanctions neighbors in the Mashreq region and beyond. under the Caesar Act. The market exchange rate of Conflict, displacement, and the collapse the Syrian pound against the US dollar weakened of economic activities have all contributed to by 26 percent year-on-year (yoy) in 2021, following a the decline in household welfare. Extreme poverty 224 percent yoy depreciation in 2020. Given Syria’s has consistently risen since the onset of the conflict, heavy reliance on imports, currency falls quickly fed reflecting deteriorating livelihood opportunities and the into higher domestic prices, causing high inflation. progressive depletion of household coping capacity. The report estimates that annual inflation reached In non-monetary terms, access to shelter, health, 90 percent yoy in 2021, after hitting 114 percent yoy education, water, and sanitation have all worsened in 2020. The war in Ukraine shocked commodity xi markets, pushing food and fuel prices in Syria even Risks to the growth outlook are significant higher. As a net importer of food and fuel, soaring and tilted to the downside. Two major sources of prices have been adversely affecting Syria’s external uncertainty are the COVID-19 pandemic and the balances, inflation, and international reserves. war in Ukraine. In the event of a rapid spread of Syria’s high inflation has affected the poor more transmissible and deadly COVID-19 variants and vulnerable disproportionately. Food prices in Syria, slow vaccination rollouts and inadequate —proxied by the World Food Programme (WFP) health facilities will exacerbate its impact. Owing to minimum food basket price index—rose by 97 percent its heavy reliance on food and fuel imports, Syria during 2021, on top of a 236 percent increase in is particularly vulnerable to the disruptions in the 2020. Driven by the noticeable increase in commodity commodity market and trade-policy interventions prices, government subsidies on essential food and triggered by the war in Ukraine. Despite the growing fuel products have dramatically risen over the past need, there is a risk that donors may shift some years, accounting for over half of the total budgeted aid away from Syria and the Syrian war-affected expenditures for 2021 and 2022. To save its budget, refugees amid the global decrease in humanitarian Syria’s government has tightened rationing, which funding, which will exacerbate the already acute has inevitably deteriorated the already dire living food insecurity of the country. Economic stagnation conditions of the Syrian people. WFP data show that and the deterioration of public services may lead more than half of households surveyed (52 percent) to increased social unrest. On the upside, Syria’s reported inadequate food consumption in February improved trade relations with its Arab neighbors 2022, double the early 2019 share. Syria’s food inse- could reduce its economic isolation. Recently, for- curity has worsened further after the war in Ukraine. eign investment restrictions in northwest Syria were Economic conditions in Syria are projected eased, and non-governmental organizations (NGOs) to continue to be mired by prolonged armed con- were allowed to do more business in Syria. These flict, turmoil in Lebanon and Turkey, COVID-19, measures may potentially facilitate trade, invest- and the war in Ukraine. Subject to extraordinarily ment, and humanitarian operations in Syria. high uncertainty, we project that Syria’s real GDP will contract by 2.6 percent in 2022 (to US$ 15.5 billion in constant 2015 prices) after declining by 2.1 percent in Special Focus: Demographic and 2021. Private consumption will remain subdued with labor market consequences of the continued erosion of purchasing power amid rising Syrian conflict prices and currency depreciation. Private investment is projected to remain weak as the security situation The current demographic profile of the Syrian is assumed to remain volatile and economic and population provides a good illustration of the massive policy uncertainties persist. Government spending, human impact of the conflict. After about a decade of especially capital expenditures, will continue to be war, the demographic profile of the Syrian population constrained by low revenues and the lack of access has dramatically changed, with the male deficit in to financing. The current account of Syria will remain prime-age adult population being its most prominent firmly in deficit because of an extremely high trade feature. The decline in the male working age deficit that will only be partially offset by net current population, together with the progressive deterioration transfer inflows. A persistent twin deficit will further of economic conditions in the country, have pushed drain foreign exchange reserves, putting further pres- more Syrian women to enter the labor market to help sure on the domestic currency. Inflation is projected support their families. However, Syrian women continue to remain elevated in the short term, due to the pass- to face severe challenges in terms of unemployment through effects of currency depreciation, persistent and lack economic opportunities compared to their food and fuel shortages, and reduced food and fuel male counterparts. These challenges have been rationing, which will stress the already struggling poor. further exacerbated as a result of conflict. xii SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS FIGURE 1 • Syria: Economic Development at a Glance The conflict caused a dramatic change of the demographic profile… … affected labor market… A. Syria population pyramid, male, 2010 and 2021 B. Trends in labor force participation and (Million people, by age group) unemployment rates, by gender (Percent) 80+ 90% 75–79 80% 76% 70–74 72% 65–69 70% 60–64 60% 55–59 50–54 50% 45–49 40% 37% 40–44 35–39 30% 26% 22% 30–34 20% 25–29 13% 20–24 10% 6% 4% 15–19 0% 10–14 5–9 2010 2021 2010 2021 0–4 –8 –6 –4 –2 0 2 4 6 8 10 Participation rate Unemployment rate 2010 2021 Male Female … and led to a drastic governance degradation. Economic activity was less than half of what it could have been without the conflict. C. Worldwide Governance Indicators, Syria (Percentile rank among all countries from 0 (lowest) to 100 (highest)) D. Actual and counterfactual GDP, Syria 60 (Billions, constant 2015 US$) 45 50 40 40 35 30 30 20 25 10 20 15 0 10 1996 1998 2000 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 5 0 Voice and accountability Political stability and 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 violence 2019 Government effectiveness Regulatory quality Rule of law Control of corruption Actual Counterfactural (continued on next page) Executive Summary xiii FIGURE 1 • Syria: Economic Development at a Glance (continued) The economy is estimated to have contracted Trade volume likely further moderated in recent years. in 2021 amid multiple shocks. F. Seaborne trade volume, Syria E. Night-time lights, GDP and GDP projections, Syria (Metric tons of cargo, Daily average in a year, all vessel categories) (GDP in constant 2015 US$, billion; light emissions) 9,000 40 2,600 8,000 35 7,000 2,100 30 6,000 1,600 5,000 25 20 1,100 4,000 3,000 15 600 2,000 10 100 1,000 5 0 0 –400 2016 2017 2018 2019 2020 2021 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 Exports Imports GDP Nightlight-based GDP projections Non-flaring nighttime lights (RHS) High inflation and currency depreciation persisted… … affecting the poor and vulnerable disproportionately. G. Inflation and exchange rate in Syria H. Share of households with inadequate food consumption (yoy percent; SYP/$US) (Share in percent) 400 4,000 60 350 3,500 300 3,000 250 2,500 50 200 2,000 150 1,500 40 100 1,000 50 500 0 0 30 –50 500 –100 –1,000 20 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 Average exchange Market exchange 10 rate levels (RHS) rate Average exchange rate CPI 0 Food CPI WFP minimum food Aug-18 Oct-18 Dec-18 Feb-19 Apr-19 Jun-19 Aug-19 Oct-19 Dec-19 Feb-20 Apr-20 Jun-20 Aug-20 Oct-20 Dec-20 Feb-21 Apr-21 Jun-21 Aug-21 Oct-21 Dec-21 Feb-22 basket price (continued on next page) xiv SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS FIGURE 1 • Syria: Economic Development at a Glance (continued) A cut in fiscal subsidies would stress the already struggling poor. The extremely slow vaccine rollout puts Syria at a high risk of future waves. I. Share of subsidies in budget expenditures in Syria (Percent) J. Daily share of the population receiving 70 a COVID-19 vaccine dose 60 (Percent, 7-day moving average) 0.60% 50 40 0.50% 30 0.40% 20 0.30% 10 0.20% 0 0.10% 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 0.00% 03/2021 04/2021 05/2021 06/2021 07/2021 08/2021 09/2021 10/2021 11/2021 12/2021 01/2022 02/2022 03/2022 04/2022 05/2022 Current spending Investment spending Source: United Nations (UN) World Population Prospects 2010; Humanitarian Needs Assessment Programme (HNAP) household survey data (Summer 2021); Syria Labor Force Survey 2010; Worldwide Governance Indicators report (various years); World Development Indicators (WDI); Penn World Table 10.0; Center for Systemic Peace; The Visible Infrared Imaging Radiometer Suite (VIIRS) and Defense Meteorological Satellite Program (DMSP) satellites; Cerdeiro, Komaromi, Liu and Saeed (2020); UN Comtrade Monitor; Central Bureau of Statistics, Syria; WFP Market Price Watch Bulletin; Syria mVAM Bulletin; Ministry of Finance (MOF) of Syria; Our World in Data; World Bank staff estimates and projections. Notes: A. World Bank calculations were based on UN World Population Prospects 2010 and HNAP household survey data (Summer 2021). B. World Bank calculations were based on Labor Force Survey 2010 and HNAP household survey data (Summer 2021). D. The counterfactual analysis is performed using a tool called synthetic control method (SCM). The SCM consists of searching for a weighted combination of countries that resemble as closely as possible the economic characteristics of Syria in the pre-conflict period to create a synthetic economy (the control), to see how this economy performed over Syria’s conflict period. The counterfactual results also account for the regional negative impact that growth in the Middle East and North Africa (MENA) region was lower than the global average in the post-conflict period. E. Nighttime Lights (NTLs) are regressed against real GDP in logarithmic form. The coefficient is applied to convert NTL observations into GDP estimates for 2020 and 2021. F. Automatic Identification System (AIS) shipping-position data collected by MarineTraffic. It first assesses import and export vessel capacity measured in deadweight tons (DWT) by country and ship type, and then uses machine-learning algorithms to estimate metric tons of cargo carried. H. The survey is collected by World Food Programme (WFP) via live telephone interviews from households in locations across Syria. The data reflects the share of the interviewed households reporting poor or borderline food consumption. I. The fiscal data pertains to the Central Government in Damascus and excludes all taxes, transfers, and expenses incurred by the autonomous region in northeastern Syria. Estimates include the off-budget electricity subsidies. J. All doses, including boosters, are counted individually. Executive Summary xv ‫الملخص التنفيذي‬ ‫الحاصلني عىل لقاح كورونا بشكل كامل ‪ 9.1‬يف املئة فقط من اجاميل‬ ‫عدد السكان‪ ،‬يف حني بلغ عدد الحاصلني عىل لقاح كورونا بشكل جزيئ‬ ‫‪ 5.2‬يف املئة‪.‬‬ ‫دخل الرصاع يف سوريا عامه الثاين عرش جالباً معه ما جلب من‬ ‫آثار مدمرة عىل السكان واالقتصاد‪ .‬أدى هذا الرصاع إىل تسارع‬ ‫تدهور البنية التحتية‪ ،‬كونه تسبب يف اإلرضار باملمتلكات واألصول‬ ‫ي‬ ‫إىل جانب التأثري املبارش للرصاع‪ ،‬يعاين االقتصاد من اآلثار‬ ‫االسرتاتيجية (مسار الدمار)‪ ،‬وتعميق شيخوخة السكان نتيجة لنزوح عدد‬ ‫املضاعفة للجائحة‪ ،‬واألحداث املناخية القاسية‪ ،‬وهشاشة املنطقة ككل‪،‬‬ ‫كبري من الناس (مسار النزوح)‪ .‬كامتسبب يف االرضار باللُحمة املجتمعية‪،‬‬ ‫وعدم استقرار االقتصاد الكيل‪ .‬أصبحت العالقات االقتصادية الخارجية‬ ‫وتردي الحوكمة‪ ،‬وأدى إىل انقسام منطقة كانت تتصف بالوحدة واالندماج‬ ‫لسوريا منذ العام ‪ 2020‬تحت قيود شديدة بسبب األزمات املتفاقمة‬ ‫ذات يوم يف سوريا (مسار الفوىض)‪ .‬ساهمت هذه املسارات مجتمعة يف‬ ‫يف البلدين املجاورين لسوريا‪ ،‬لبنان وتركيا‪ ،‬فضالً عن فرض عقوبات‬ ‫خفض حجم النشاط االقتصادي إىل النصف خالل الفرتة املمتدة بني عامي‬ ‫أمريكية جديدة مبوجب قانون قيرص‪ .‬انخفض سعر رصف السوق للرية‬ ‫‪ 2010‬و‪ .2019‬يقدر هذا التقرير – أنه يف حال افرتاض عدم وجود الرصاع‬ ‫السورية مقابل الدوالر األمرييك بنسبة ‪ 26‬يف املئة يف عام ‪ ،2021‬وذلك‬ ‫(أي بافرتاض وضع أو واقع مغاير للواقع الحايل)‪ ،‬لكان الناتج املحيل‬ ‫بعد انخفاض سنوي بنسبة ‪ 224‬يف املئة عام ‪ .2020‬وبالنظر إىل اعتامد‬ ‫اإلجاميل لسوريا لعام ‪ 2019‬ليبلغ ‪ 38.6‬مليار دوالر أمرييك باستخدام‬ ‫سوريا الكبري عىل الواردات‪ ،‬رسعان ما أدى انخفاض العملة إىل ارتفاع‬ ‫األسعار الثابتة لعام ‪ ،2015‬وذلك مقابل ‪ 16.3‬مليار دوالر أمرييك وهي‬ ‫األسعار محلياً‪ ،‬مام تسبب يف ارتفاع معدل التضخم‪ .‬يقدر التقرير أن‬ ‫قيمة الناتج املحيل اإلجاميل املحقق‪ .‬باإلضافة إىل ذلك‪ ،‬تسببت التدفقات‬ ‫التضخم السنوي بلغ ‪ 90‬يف املئة يف عام ‪ ،2021‬وكان قد بلغ ‪ 114‬يف املئة‬ ‫الهائلة لالجئني الناجمة عن الرصاع السوري‪ ،‬إىل جانب التداعيات األخرى‬ ‫عام ‪ .2020‬تسببت الحرب يف أوكرانيا يف صدمة ألسواق السلع األساسية‪،‬‬ ‫التي تتضمن املسارات التجارية واملالية‪ ،‬يف تكبد البلدان املجاورة لسوريا‬ ‫مام دفع أسعار الغذاء والوقود يف سوريا إىل االرتفاع أكرث وأكرث‪ .‬ولكون‬ ‫يف منطقة املرشق خسائر اقتصادية واجتامعية فادحة‪.‬‬ ‫سوريا مستوردا ً صافياً للغذاء والوقود‪ ،‬فقد أثر ارتفاع األسعار هذا سلباً‬ ‫ساهمت عوامل الرصاع‪ ،‬والنزوح‪ ،‬وانهيار األنشطة االقتصادية‬ ‫عىل موازين التجارة الخارجية لسوريا والتضخم واالحتياطيات الدولية‪.‬‬ ‫يف تدهور املسنوى املعييش لألرسة‪ .‬ارتفعت معدالت الفقر املدقع ارتفاعاً‬ ‫أثر التضخم املرتفع يف سوريا عىل الفقراء والضعفاء بشكل‬ ‫مطردا ً منذ بداية الرصاع‪ ،‬وهذا ما يعكس تدهورا ً يف فرص كسب العيش‬ ‫غري متناسب‪ .‬ارتفع معدل تضخم أسعار الغذاء وفقاً ملؤرش الحد األدىن‬ ‫والتآكل التدريجي لقدرة األرسة عىل التكيف‪ .‬أما من الناحية غري النقدية‪،‬‬ ‫لسعر سلة الغذاء التابع لربنامج الغذاء العاملي بنسبة ‪ 97‬يف املئة خالل‬ ‫فقد ساءت فرص الحصول عىل املأوى‪ ،‬وسبل العيش‪ ،‬والصحة‪ ،‬والتعليم‪،‬‬ ‫عام ‪ ،2021‬باإلضافة إىل زيادة بنسبة ‪ 236‬يف املئة يف عام ‪ .2020‬نتيجة‬ ‫واملياه‪ ،‬والرصف الصحي بشكل كبري منذ بدء النزاع‪ .‬يف ظل التدهور‬ ‫للزيادة امللحوظة يف أسعار السلع األساسية‪ ،‬ارتفع الدعم الحكومي للسلع‬ ‫الكبري لنظام الرعاية الصحية يف أعقاب الحرب التي استمرت عقدا ً من‬ ‫الغذائية األساسية والوقود بشكل كبري خالل السنوات املاضية‪ ،‬حيث ميثل‬ ‫الزمان‪ ،‬أدت جائحة كورونا عام ‪ 2019‬إىل تفاقم الضعف والهشاشة‪ .‬منذ‬ ‫هذا الدعم أكرث من نصف إجاميل بند النفقات يف موازنة عامي ‪2021‬‬ ‫شهر مارس ‪ /‬آذار ‪ 2022‬بدأ عدد حاالت اإلصابات الجديدة بكورونا يف‬ ‫و‪ .2022‬وللتوفري يف املوازنة ‪ ،‬ضيقت الحكومة السورية نظام الحصص‬ ‫االنخفاض‪ ،‬ولكن معدل الوفيات الرتاكمي املرتفع يدل عىل عدم قدرة‬ ‫التموينية‪ ،‬األمر الذي أدى بكل تأكيد إىل تدهور الظروف املعيشية‬ ‫النظام الصحي عىل تلبية احتياجات املصابني بكورونا‪ .‬كان عدد الوفيات‬ ‫للشعب السوري والتي كانت مرتدية يف األساس‪ .‬تظهر بيانات برنامج‬ ‫املرتبطة بكورونا مرتفعاً نسبياً يف سوريا‪ ،‬ويُعزى ذلك جزئيًا إىل البطء يف‬ ‫الغذاء العاملي أن أكرث من نصف األرس التي شملها استطالع املنظمة (‪52‬‬ ‫تقديم اللقاحات‪ ،‬حتى تاريخ ‪ 14‬مايو ‪ /‬أيار ‪ ،2022‬بلغ عدد السوريني‬ ‫‪xvii‬‬ ‫الدميوغرافية لسكان سوريا بشكل كبري‪ ،‬وكانت أبرز تلك السامت انخفاض‬ ‫يف املئة) أبلغت عن عدم كفاية الغذاء يف فرباير ‪ /‬شباط ‪ ،2022‬أي ضعف‬ ‫عدد الذكور بني السكان البالغني سن الرشد‪ .‬دفع االنخفاض يف عدد الذكور‬ ‫النسبة أوائل عام ‪.2019‬هذا وقد ازداد تفاقم انعدام األمن الغذايئ يف‬ ‫يف سن العمل‪ ،‬باإلضافة إىل التدهور التدريجي للظروف االقتصادية يف‬ ‫سوريا بعد الحرب يف أوكرانيا‪.‬‬ ‫البالد‪ ،‬باملزيد من النساء السوريات إىل دخول سوق العمل للمساعدة يف‬ ‫من املتوقع أن تستمر الظروف االقتصادية بالتدهور يف سوريا‪،‬‬ ‫إعالة أرسهن‪ .‬ومع ذلك‪ ،‬ال تزال املرأة السورية تواجه تحديات كبرية من‬ ‫املترضرة بالرصاع طويل األمد‪ ،‬واالضطرابات يف لبنان وتركيا‪ ،‬وجائحة‬ ‫حيث البطالة وانعدام الفرص االقتصادية مقارنة بنظريها الرجل‪ ،‬وهي‬ ‫كورونا‪ ،‬والحرب يف أوكرانيا‪ .‬يف ظل حالة عدم التيقن املرتفعة بشكل‬ ‫تحديات زاد الرصاع من تفاقمها‪.‬‬ ‫استثنايئ‪ ،‬نتوقع أن ينكمش الناتج املحيل اإلجاميل الحقيقي لسوريا بنسبة‬ ‫املصدر‪ :‬التوقعات السكانية العاملية لألمم املتحدة ‪2010‬؛ بيانات‬ ‫‪ 2.6‬يف املئة يف عام ‪( 2022‬إىل ‪ 15.5‬مليار دوالر باألسعار الثابتة لعام‬ ‫مسح األرس لربنامج تقييم االحتياجات اإلنسانية (‪())HNAP‬صيف‬ ‫‪ ،)2015‬وذلك بعد انخفاضه بنسبة ‪ 2.1‬يف املئة يف عام ‪ .2021‬ستظل‬ ‫‪)2021‬؛ مسح القوى العاملة يف سوريا ‪2010‬؛ تقرير مؤرشات الحوكمة‬ ‫معدالت االستهالك الخاص ضعيفة مع استمرار تآكل القوة الرشائية يف ظل‬ ‫العاملية (سنوات متعددة)؛ مؤرشات التنمية العاملية (‪)WDI‬؛ جدول‬ ‫ارتفاع األسعار وانخفاض قيمة العملة‪ .‬يُتوقع أن يظل االستثامر الخاص‬ ‫بن وورلد ‪10.0‬؛ مركز السالم املنهجي؛ مجموعة مقياس إشعاع التصوير‬ ‫ضعيفاً أيضاً مع افرتاض استمرار تقلب الوضع األمني​​وتواصل حالة عدم‬ ‫املريئ باألشعة تحت الحمراء (‪ )VIIRS‬وبرنامج األقامر الصناعية الدفاعية‬ ‫اليقني االقتصادي والسيايس‪ .‬سيظل اإلنفاق الحكومي‪ ،‬وخاصة النفقات‬ ‫لألرصاد الجوية (‪)DMSP‬؛ سريديرو‪ ،‬وكومارومي‪ ،‬وليو‪ ،‬وسعيد (‪)2020‬؛‬ ‫الرأساملية‪ ،‬مقيدا ً بانخفاض اإليرادات ونقص الوصول إىل التمويل‪ .‬سيظل‬ ‫برنامج التجارة ‪ COMTRADE‬لألمم املتحدة؛ املكتب املركزي لإلحصاء‪،‬‬ ‫الحساب الجاري لسوريا يف حالة عجز ثابت نتيجة العجز التجاري املرتفع‬ ‫سوريا؛ نرشة برنامج الغذاء العاملي ملراقبة أسعار السوق؛ نرشة تحليل‬ ‫للغاية والذي لن يُعوض عدا جزئياً من خالل صايف تدفقات التحويالت‬ ‫وخارطة الفئات الهشة عرب الهاتف املحمول (‪)mVAM‬؛ وزارة املالية‬ ‫الجارية‪ .‬سيؤدي العجز املزدوج املستمر إىل زيادة يف استنزاف احتياطيات‬ ‫السورية؛ عاملنا يف بيانات؛ تقديرات وتوقعات خرباء البنك الدويل‪.‬‬ ‫النقد األجنبي‪ ،‬مام سيزيد من الضغط عىل العملة املحلية‪ .‬من املتوقع أن‬ ‫مالحظات‪ :‬أ‪ .‬حسابات البنك الدويل بُنيت عىل بيانات األمم‬ ‫يظل معدل التضخم مرتفعاً عىل املدى القصري‪ ،‬وذلك بسبب اآلثار العابرة‬ ‫املتحدة للتوقعات السكانية يف العامل لعام ‪ 2010‬وبيانات مسح األرس‬ ‫النخفاض قيمة العملة‪ ،‬واستمرار نقص الغذاء والوقود‪ ،‬و تقنني الغذاء‬ ‫لربنامج تقييم االحتياجات اإلنسانية (‪( )HNAP‬صيف ‪)2021‬؛ ب‪.‬‬ ‫والوقود‪ ،‬مام سيضغط عىل الفقراء الذين يعانون يف األساس‪.‬‬ ‫حسابات البنك الدويل بُنيت عىل مسح القوى العاملة لعام ‪ 2010‬وبيانات‬ ‫املخاطر عىل توقعات النمو كبرية ومتيل للتوجه سلباً‪ .‬هناك‬ ‫مسح األرس لربنامج تقييم االحتياجات اإلنسانية (‪())HNAP‬صيف‬ ‫مصدران رئيسيان لعدم اليقني وهام جائحة كورونا والحرب يف أوكرانيا‪ .‬يف‬ ‫‪)2021‬؛ ج‪ .‬يتم إجراء تحليل للوضع املغاير (افرتاض وضع مغاير للوضع‬ ‫حالة حدوث انتشار رسيع ملتحور رسيع االنتقال وفتاك يف سوريا‪ ،‬فإن عمليات‬ ‫الحايل) باستخدام أداة تسمى طريقة التحكم الرتكيبي (‪ ،)SCM‬وهي‬ ‫التطعيم البطيئة وعدم كفاية املرافق الصحية ستؤدي إىل تفاقم تأثريه‪ .‬نظرا ً‬ ‫أسلوب احصايئ يتمثل يف البحث عن مجموعة مرجحة من البلدان التي‬ ‫العتامد سوريا الشديد عىل واردات الغذاء والوقود‪ ،‬فإن سوريا معرضة بشكل‬ ‫تشبه خصائصها االقتصادية إىل حد كبري الخصائص االقتصادية يف سوريا‬ ‫خاص آلثار االضطرابات يف سوق السلع األساسية وتدخالت السياسة التجارية‬ ‫يف فرتة ما قبل الرصاع وذلك إلنشاء اقتصاد تركيبي منها ملعرفة كيفية أداء‬ ‫التي تسببت فيها الحرب يف أوكرانيا‪ .‬عىل الرغم من االحتياجات املتنامية‪ ،‬إال‬ ‫هذا االقتصاد خالل فرتة الرصاع يف سوريا‪ .‬تفرس نتائج الواقع املغاير التأثري‬ ‫أن هناك مخاطر قيام الجهات املانحة بتحويل بعض املساعدات بعيدا ً عن‬ ‫السلبي اإلقليمي املتمثل يف أن النمو يف منطقة الرشق األوسط وشامل‬ ‫سوريا والالجئني السوريني املترضرين من الحرب‪ ،‬وذلك يف ظل االنخفاض‬ ‫إفريقيا كان أقل من املتوسط العاملي يف فرتة ما بعد الرصاع‪ .‬ه‪ .‬ترتاجع‬ ‫العاملي للتمويل اإلنساين‪ ،‬مام سيؤدي إىل تفاقم انعدام األمن الغذايئ الحاد‬ ‫اإلضاءة الليلية مقابل الناتج املحيل اإلجاميل الحقيقي يف مستويات أو عىل‬ ‫بالفعل يف البالد‪ .‬قد يؤدي الركود االقتصادي وتدهور الخدمات العامة إىل‬ ‫شكل لوغاريتامت‪ .‬يتم تطبيق املعامل لتحويل مالحظات اإلضاءة الليلية‬ ‫زيادة االضطرابات االجتامعية‪ ،‬مفضياً إىل حالة من عدم االستقرار السيايس‪.‬‬ ‫إىل تقديرات إجاميل الناتج املحيل لعامي ‪ 2020‬و‪ .2021‬و‪ .‬تستخدم املنصة‬ ‫أما من الناحية اإليجابية فيمكن أن يؤدي تحسن العالقات التجارية السورية‬ ‫بيانات مواقع سفن الشحن التي تجمع بواسطة مرشوع ‪.MarineTraffic‬‬ ‫مع جريانها العرب إىل تقليل عزلتها االقتصادية‪ .‬يف اآلونة األخرية‪ ،‬تم تخفيف‬ ‫تقوم املنصة أوالً بتقييم سعة سفن االسترياد والتصدير املقاسة باألطنان‬ ‫وسمح للمنظامت‬ ‫قيود االستثامر األجنبي يف شامل رشق وشامل غرب سوريا‪ُ ،‬‬ ‫الساكنة (الحمولة الساكنة) حسب بلد ونوع السفينة‪ ،‬ومن ثم تستخدم‬ ‫غري الحكومية مبامرسة املزيد من األعامل التجارية يف سوريا‪ ،‬وبالتايل قد‬ ‫خوارزميات التعلم اآليل لتقدير حجم البضائع املنقولة بالطن املرتي‪ .‬ز‪.‬‬ ‫تسهل هذه اإلجراءات من التجارة واالستثامر والعمليات اإلنسانية‪.‬‬ ‫تم جمع املسح من قبل برنامج الغذاء العاملي من خالل مقابالت هاتفية‬ ‫مبارشة مع األرس يف مواقع تشمل جميع أنحاء سوريا‪ .‬تعكس البيانات‬ ‫نسبة األرس التي أبلغت عن ضعف او محدودية استهالك الغذاء من االرس‬ ‫اتركيز خاص‪ :‬تداعيات الرصاع السوري‬ ‫التي متت مقابلتها‪ .‬ح‪ .‬البيانات املالية تخص الحكومة املركزية يف دمشق‬ ‫عىل السكان وسوق العمل‬ ‫وتستثني جميع الرضائب والتحويالت واملرصوفات ملنطقة الحكم الذايت‬ ‫يف شامل رشق سوريا‪ .‬تشمل التقديرات دعم الكهرباء خارج املوازنة ط‪.‬‬ ‫تقدم السامت الدميوغرافية الحالية لسكان سوريا مثاالً جيدا ً عىل‬ ‫تحسب جميع الجرعات‪ ،‬مبا يف ذلك الجرعات املعززة‪ ،‬بشكل منفرد‪.‬‬ ‫التأثر البرشي الهائل (يالرصاع) فبعد نحو عقد من الحرب‪ ،‬تغريت السامت‬ ‫‪xviii‬‬ ‫‪SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS‬‬ ‫الشكل رقم ‪  •  1‬سوريا‪ :‬ملحة عن التنمية االقتصادية‬ ‫وذﻟﻚ أﺛﺮ ﻋﲆ ﺳﻮق اﻟﻌﻤﻞ‬ ‫‪...‬ﺗﺴﺒﺐ اﻟﴫاع ﰲ ﺗﻐﻴ ﺟﺬري ﻟﻠﺴ ت اﻟﺴﻜﺎﻧﻴﺔ‪/‬اﻟﺪ ﻮﻏﺮاﻓﻴﺔ‬ ‫ب‪ .‬ﺣﺮﻛﺔ ﺳﻮق اﻟﻌﻤﻞ وﻣﻌﺪﻻت اﻟﺒﻄﺎﻟﺔ ﺣﺴﺐ اﻟﺠﻨﺲ‬ ‫أ‪ .‬اﻟﻬﺮم اﻟﺴﻜﺎ ﰲ ﺳﻮرﻳﺎ‪ ،‬ذﻛﻮر‪ 2010 ،‬و‪2021‬‬ ‫)ﻧﺴﺒﻪ ﻣﺌﻮﻳﺔ(‬ ‫)ﻣﻠﻴﻮن ﺷﺨﺺ‪ ،‬ﺣﺴﺐ اﻟﻔﺌﺔ اﻟﻌﻤﺮﻳﺔ(‬ ‫‪90%‬‬ ‫‪80+‬‬ ‫‪80%‬‬ ‫‪76%‬‬ ‫‪75–79‬‬ ‫‪72%‬‬ ‫‪70–74‬‬ ‫‪70%‬‬ ‫‪65–69‬‬ ‫‪60%‬‬ ‫‪60–64‬‬ ‫‪55–59‬‬ ‫‪50%‬‬ ‫‪50–54‬‬ ‫‪40%‬‬ ‫‪37%‬‬ ‫‪45–49‬‬ ‫‪26%‬‬ ‫‪40–44‬‬ ‫‪30%‬‬ ‫‪22%‬‬ ‫‪35–39‬‬ ‫‪20%‬‬ ‫‪30–34‬‬ ‫‪13%‬‬ ‫‪25–29‬‬ ‫‪10%‬‬ ‫‪6%‬‬ ‫‪4%‬‬ ‫‪20–24‬‬ ‫‪0%‬‬ ‫‪15–19‬‬ ‫‪10–14‬‬ ‫‪2010‬‬ ‫‪2021‬‬ ‫‪2010‬‬ ‫‪2021‬‬ ‫‪5–9‬‬ ‫‪0–4‬‬ ‫ﻧﺴﺒﺔ اﳌﺸﺎرﻛﺔ‬ ‫ﻧﺴﺒﺔ اﻟﺒﻄﺎﻟﺔ‬ ‫‪–8‬‬ ‫‪–6‬‬ ‫‪–4‬‬ ‫‪–2‬‬ ‫‪0‬‬ ‫‪2‬‬ ‫‪4‬‬ ‫‪6‬‬ ‫‪8‬‬ ‫‪10‬‬ ‫اﻧﺎث‬ ‫ذﻛﻮر‬ ‫‪2010‬‬ ‫‪2021‬‬ ‫اﻟﻨﺸﺎط اﻻﻗﺘﺼﺎدي أﺻﺒﺢ أﻗﻞ ﻣﻦ ﻧﺼﻒ ﻣ‬ ‫وأدى إﱃ ﺗﺪﻫﻮر ﻛﺒ ﰲ اﻟﺤﻮﻛﻤﺔ‬ ‫‪.‬ﻛﺎن ﻜﻦ أن ﻳﻜﻮن ﰲ ﺣﺎﻟﺔ ﻋﺪم وﺟﻮد اﻟﴫاع‬ ‫ج‪ -‬ﻣﺆﴍات اﻟﺤﻮﻛﻤﺔ ﰲ اﻟﻌﺎ ‪ ،‬ﺳﻮرﻳﺎ‬ ‫د‪ .‬اﻟﻨﺎﺗﺞ اﳌﺤﲇ اﻹﺟ ﱄ اﻟﻔﻌﲇ واﻟﻨﺎﺗﺞ اﳌﺤﲇ ﰲ ﺣﺎﻟﺔ اﻟﻮﺿﻊ اﳌﻐﺎﻳﺮ‪ ،‬ﺳﻮرﻳﺎ‬ ‫)اﳌﺮﺗﺒﺔ اﳌﺌﻮﻳﺔ ﺑ ﺟﻤﻴﻊ اﻟﺒﻠﺪان ﻣﻦ ‪) 0‬اﻷد ( إﱃ ‪) 100‬اﻷﻋﲆ((‬ ‫)ﻣﻠﻴﺎر دوﻻر أﻣﺮﻳ ﺑﺎﻷﺳﻌﺎر اﻟﺜﺎﺑﺘﺔ ﰲ ﻋﺎم ‪(2015‬‬ ‫‪60‬‬ ‫‪45‬‬ ‫‪40‬‬ ‫‪50‬‬ ‫‪35‬‬ ‫‪40‬‬ ‫‪30‬‬ ‫‪30‬‬ ‫‪25‬‬ ‫‪20‬‬ ‫‪20‬‬ ‫‪10‬‬ ‫‪15‬‬ ‫‪0‬‬ ‫‪10‬‬ ‫‪1996‬‬ ‫‪1998‬‬ ‫‪2000‬‬ ‫‪2002‬‬ ‫‪2003‬‬ ‫‪2004‬‬ ‫‪2005‬‬ ‫‪2006‬‬ ‫‪2007‬‬ ‫‪2008‬‬ ‫‪2009‬‬ ‫‪2010‬‬ ‫‪2011‬‬ ‫‪2012‬‬ ‫‪2013‬‬ ‫‪2014‬‬ ‫‪2015‬‬ ‫‪2016‬‬ ‫‪2017‬‬ ‫‪2018‬‬ ‫‪2019‬‬ ‫‪2020‬‬ ‫‪5‬‬ ‫‪0‬‬ ‫اﻟﺼﻮت اﳌﺴﻤﻮع واﳌﺴﺎءﻟﺔ‬ ‫اﻻﺳﺘﻘﺮار اﻟﺴﻴﺎﳼ واﻟﻌﻨﻒ‬ ‫ﻓﻌﺎﻟﻴﺔ اﻟﺤﻜﻮﻣﺔ‬ ‫اﻟﺠﻮدة اﻟﺘﻨﻈﻴﻤﻴﺔ‬ ‫‪1995‬‬ ‫‪1996‬‬ ‫‪1997‬‬ ‫‪1998‬‬ ‫‪1999‬‬ ‫‪2000‬‬ ‫‪2001‬‬ ‫‪2002‬‬ ‫‪2003‬‬ ‫‪2004‬‬ ‫‪2005‬‬ ‫‪2006‬‬ ‫‪2007‬‬ ‫‪2008‬‬ ‫‪2009‬‬ ‫‪2010‬‬ ‫‪2011‬‬ ‫‪2012‬‬ ‫‪2013‬‬ ‫‪2014‬‬ ‫‪2015‬‬ ‫‪2016‬‬ ‫‪2017‬‬ ‫‪2018‬‬ ‫‪2019‬‬ ‫ﺳﻴﺎدة اﻟﻘﺎﻧﻮن‬ ‫ﻣﻜﺎﻓﺤﺔ اﻟﻔﺴﺎد‬ ‫اﻟﻮاﻗﻊ اﻟﺤﺎﱄ‬ ‫اﻟﻮاﻗﻊ اﳌﻐﺎﻳﺮ‬ ‫(يتبع يف الصفحة التالية)‬ ‫امللخص التنفيذي‬ ‫‪xix‬‬ ‫الشكل رقم ‪  •  1‬سوريا‪ :‬ملحة عن التنمية االقتصادية (يتبع)‬ ‫و‪ .‬ﺣﺠﻢ اﻟﺘﺠﺎرة اﳌﻨﻘﻮﻟﺔ ﺑﺤﺮاً‪ ،‬ﺳﻮرﻳﺎ‬ ‫ﺗﺸ اﻟﺘﻘﺪﻳﺮات إﱃ أن اﻻﻗﺘﺼﺎد ﻗﺪ اﻧﻜﻤﺶ ﻣﺮة أﺧﺮى ﰲ‬ ‫)ﻃﻦ ﻣﱰي ﻣﻦ اﻟﺒﻀﺎﺋﻊ‪ ،‬اﳌﺘﻮﺳﻂ اﻟﻴﻮﻣﻲ ﰲ اﻟﺴﻨﺔ‪ ،‬ﺟﻤﻴﻊ ﻓﺌﺎت اﻟﺴﻔﻦ(‬ ‫ﻋﺎم ‪ 2021‬ﻧﺘﻴﺠﺔ ﻟﻠﺼﺪﻣﺎت اﳌﺘﻌﺪدة‬ ‫‪9,000‬‬ ‫ﻩ‪ .‬ﺗﻮﻗﻌﺎت اﻹﺿﺎءة ﻟﻴﻼً واﻟﻨﺎﺗﺞ اﳌﺤﲇ اﻹﺟ ﱄ واﻟﻨﺎﺗﺞ اﳌﺤﲇ اﻹﺟ ﱄ اﳌﺘﻮﻗﻊ‪ ،‬ﺳﻮرﻳﺎ‬ ‫‪8,000‬‬ ‫)اﻟﻨﺎﺗﺞ اﳌﺤﲇ اﻹﺟ ﱄ ﺑﺎﻷﺳﻌﺎر اﻟﺜﺎﺑﺘﺔ ﻟﻌﺎم ‪ ،2015‬ﻣﻠﻴﺎر دوﻻر أﻣﺮﻳ ؛ اﻧﺒﻌﺎﺛﺎﺗﻀﻮﺋﻴﺔ(‬ ‫‪7,000‬‬ ‫‪40‬‬ ‫‪2,600‬‬ ‫‪6,000‬‬ ‫‪35‬‬ ‫‪2,100‬‬ ‫‪5,000‬‬ ‫‪30‬‬ ‫‪4,000‬‬ ‫‪1,600‬‬ ‫‪25‬‬ ‫‪3,000‬‬ ‫‪20‬‬ ‫‪1,100‬‬ ‫‪2,000‬‬ ‫‪15‬‬ ‫‪1,000‬‬ ‫‪600‬‬ ‫‪10‬‬ ‫‪0‬‬ ‫‪2016‬‬ ‫‪2017‬‬ ‫‪2018‬‬ ‫‪2019‬‬ ‫‪2020‬‬ ‫‪2021‬‬ ‫‪100‬‬ ‫‪5‬‬ ‫اﻟﺼﺎدرات‬ ‫اﳌﺴﺘﻮردات‬ ‫‪0‬‬ ‫‪–400‬‬ ‫‪1992‬‬ ‫‪1993‬‬ ‫‪1994‬‬ ‫‪1995‬‬ ‫‪1996‬‬ ‫‪1997‬‬ ‫‪1998‬‬ ‫‪1999‬‬ ‫‪2000‬‬ ‫‪2001‬‬ ‫‪2002‬‬ ‫‪2003‬‬ ‫‪2004‬‬ ‫‪2005‬‬ ‫‪2006‬‬ ‫‪2007‬‬ ‫‪2008‬‬ ‫‪2009‬‬ ‫‪2010‬‬ ‫‪2011‬‬ ‫‪2012‬‬ ‫‪2013‬‬ ‫‪2014‬‬ ‫‪2015‬‬ ‫‪2016‬‬ ‫‪2017‬‬ ‫‪2018‬‬ ‫‪2019‬‬ ‫‪2020‬‬ ‫‪2021‬‬ ‫اﻟﻨﺎﺗﺞ اﻻﺟ ﱄ اﳌﺤﲇ‬ ‫اﻟﻨﺎﺗﺞ اﻻﺟ ﱄ اﳌﺤﲇ اﳌﺘﻮﻗﻊ اﺳﺘﻨﺎدا اﱃ اﻻﺿﺎءة اﻟﻠﻴﻠﻴﺔ‬ ‫أﺿﻮاء ﻟﻴﻠﻴﺔ ﻏ ﻣﺸﻌﺔ ) اﳌﺤﻮر اﻷ ﻦ(‬ ‫أﺛﺮ ذﻟﻚ ﻋﲆ اﻟﻔﻘﺮاء واﻟﻀﻌﻔﺎء ﺑﺸﻜﻞ ﻏ ﻣﺘﻨﺎﺳﺐ‬ ‫اﺳﺘﻤﺮار ارﺗﻔﺎع اﻟﺘﻀﺨﻢ واﻧﺨﻔﺎض ﻗﻴﻤﺔ اﻟﻌﻤﻠﺔ ‪...‬‬ ‫ء ﻛﺎﻓ ً‬ ‫ﻴﺎ‬ ‫ح‪ .‬ﻧﺴﺒﺔ اﻷﴎ اﻟﺘﻲ ﻻ ﺗﺴﺘﻬﻠﻚ ﻏﺬا ً‬ ‫ز‪ .‬اﻟﺘﻀﺨﻢ وﺳﻌﺮ اﻟﴫف ﰲ ﺳﻮرﻳﺎ‬ ‫)اﻟﻨﺴﺒﺔ اﳌﺌﻮﻳﺔ(‬ ‫)اﻟﻨﺴﺒﺔ اﳌﺌﻮﻳﺔ ﻋﲆ أﺳﺎس ﺳﻨﻮي؛ ﻟ ة ﺳﻮرﻳﺔ ‪ /‬دوﻻر أﻣﺮﻳ (‬ ‫‪60‬‬ ‫‪400‬‬ ‫‪4,000‬‬ ‫‪350‬‬ ‫‪3,500‬‬ ‫‪300‬‬ ‫‪3,000‬‬ ‫‪50‬‬ ‫‪250‬‬ ‫‪2,500‬‬ ‫‪200‬‬ ‫‪2,000‬‬ ‫‪40‬‬ ‫‪150‬‬ ‫‪1,500‬‬ ‫‪100‬‬ ‫‪1,000‬‬ ‫‪50‬‬ ‫‪500‬‬ ‫‪30‬‬ ‫‪0‬‬ ‫‪0‬‬ ‫‪–50‬‬ ‫‪500‬‬ ‫‪–100‬‬ ‫‪–1,000‬‬ ‫‪20‬‬ ‫‪2011‬‬ ‫‪2012‬‬ ‫‪2013‬‬ ‫‪2014‬‬ ‫‪2015‬‬ ‫‪2016‬‬ ‫‪2017‬‬ ‫‪2018‬‬ ‫‪2019‬‬ ‫‪2020‬‬ ‫‪2021‬‬ ‫‪2022‬‬ ‫‪10‬‬ ‫ﻣﺴﺘﻮﻳﺎت ﻣﻌﺪل ﺳﻌﺮ اﻟﴫف‬ ‫ﺳﻌﺮ ﴏف اﻟﺴﻮق‬ ‫)اﳌﺤﻮر ﻋﲆ اﻟﻴﻤ (‬ ‫ﻣﺆﴍ أﺳﻌﺎر اﳌﺴﺘﻬﻠﻚ‬ ‫‪0‬‬ ‫ﻣﻌﺪل ﺳﻌﺮ اﻟﴫف‬ ‫اﻟﺤﺪ اﻷد ﻟﺴﻌﺮ ﺳﻠﺔ اﻟﻐﺬاء‬ ‫ﻣﺆﴍ أﺳﻌﺮ اﳌﺴﺘﻬﻠﻚ ﻟﻠﻐﺬاء‬ ‫اﻟﺘﺎﺑﻊ ﻟﱪﻧﺎﻣﺞ اﻟﻐﺬاء اﻟﻌﺎﳌﻲ‬ ‫‪Aug-18‬‬ ‫‪Oct-18‬‬ ‫‪Dec-18‬‬ ‫‪Feb-19‬‬ ‫‪Apr-19‬‬ ‫‪Jun-19‬‬ ‫‪Aug-19‬‬ ‫‪Oct-19‬‬ ‫‪Dec-19‬‬ ‫‪Feb-20‬‬ ‫‪Apr-20‬‬ ‫‪Jun-20‬‬ ‫‪Aug-20‬‬ ‫‪Oct-20‬‬ ‫‪Dec-20‬‬ ‫‪Feb-21‬‬ ‫‪Apr-21‬‬ ‫‪Jun-21‬‬ ‫‪Aug-21‬‬ ‫‪Oct-21‬‬ ‫‪Dec-21‬‬ ‫‪Feb-22‬‬ ‫(يتبع يف الصفحة التالية)‬ ‫‪xx‬‬ ‫‪SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS‬‬ ‫الشكل رقم ‪  •  1‬سوريا‪ :‬ملحة عن التنمية االقتصادية (يتبع)‬ ‫ﺗﻀﻊ ﻋﻤﻠﻴﺔ اﻟﺘﻠﻘﻴﺢ اﻟﺒﻄﻴﺌﺔ ﻟﻠﻐﺎﻳﺔ ﺳﻮرﻳﺎ ﰲ ﺧﻄﺮ ﻛﺒ‬ ‫ﻣﻦ ﺷﺄن ﺧﻔﺾ اﻟﺪﻋﻢ اﳌﺎﱄ أن ﻳﻀﻐﻂ ﻋﲆ اﻟﻔﻘﺮاء اﻟﺬﻳﻦ ﻳﻌﺎﻧﻮن ﻣﺴﺒﻘﺎً‬ ‫ﻣﻦ ﻣﻮﺟﺎت ﻛﻮروﻧﺎ ﻣﺴﺘﻘﺒﻠﻴﺔ‬ ‫ي‪ .‬اﻟﻨﺴﺒﺔ اﻟﻴﻮﻣﻴﺔ ﻣﻦ اﻟﺴﻜﺎن اﻟﺬﻳﻦ ﻳﺘﻠﻘﻮن ﻟﻘﺎح ﻛﻮروﻧﺎ‬ ‫ط‪ .‬ﻧﺴﺒﺔ اﻟﺪﻋﻢ ﻣﻦ ﻧﻔﻘﺎت اﳌﻮازﻧﺔ ﰲ ﺳﻮرﻳﺎ‬ ‫)اﻟﻨﺴﺒﺔ اﳌﺌﻮﻳﺔ‪ ،‬اﳌﺘﻮﺳﻂ اﳌﺘﺤﺮك ﻟﻔﱰة ‪ 7‬أﻳﺎم(‬ ‫)اﻟﻨﺴﺒﺔ اﳌﺌﻮﻳﺔ(‬ ‫‪70‬‬ ‫‪0.60%‬‬ ‫‪60‬‬ ‫‪0.50%‬‬ ‫‪50‬‬ ‫‪0.40%‬‬ ‫‪40‬‬ ‫‪30‬‬ ‫‪0.30%‬‬ ‫‪20‬‬ ‫‪0.20%‬‬ ‫‪10‬‬ ‫‪0.10%‬‬ ‫‪0‬‬ ‫‪2012‬‬ ‫‪2013‬‬ ‫‪2014‬‬ ‫‪2015‬‬ ‫‪2016‬‬ ‫‪2017‬‬ ‫‪2018‬‬ ‫‪2019‬‬ ‫‪2020‬‬ ‫‪2021‬‬ ‫‪2022‬‬ ‫‪0.00%‬‬ ‫‪03/2021‬‬ ‫‪04/2021‬‬ ‫‪05/2021‬‬ ‫‪06/2021‬‬ ‫‪07/2021‬‬ ‫‪08/2021‬‬ ‫‪09/2021‬‬ ‫‪10/2021‬‬ ‫‪11/2021‬‬ ‫‪12/2021‬‬ ‫‪01/2022‬‬ ‫‪02/2022‬‬ ‫‪03/2022‬‬ ‫‪04/2022‬‬ ‫‪05/2022‬‬ ‫ﻧﻔﻘﺎت ﺟﺎرﻳﺔ‬ ‫ﻧﻔﻘﺎت اﺳﺘﺜ رﻳﺔ‬ ‫املعهد الوطني لإلحصاء‪.‬‬ ‫امللخص التنفيذي‬ ‫‪xxi‬‬ 1 THE CONFLICT CONTEXT A lthough the intensity of the conflict has conducted by the World Bank covering 15 Syrian declined from its peak in the mid-2010s, cities in June 2018, about one-fifth of all residential Syria remains a top-ranked country buildings suffered damage, about a quarter of which worldwide in terms of violent deaths. According were fully destroyed. Meanwhile, about 40 percent to statistics compiled by the Armed Conflict Location of educational facilities were damaged, destroyed, & Event Data Project (ACLED), Syria ranked among or occupied by parties to the conflict or serve as the top ten countries in the world with 7,465 conflict- shelters to Internally Displaced Persons (IDPs).1 related deaths in 2021. In 2021, active conflicts were According to the World Health Organization (WHO) largely concentrated in the northern regions, with Whole of Syria consolidated Health Resources and armed clashes and military operations frequently Services Availability Monitoring System (HeRAMS) for occurring in the governorates of Idleb, Aleppo, and Q4 2021, just 59 percent of hospitals and 54 percent Deir-es Zor (Figure 2). of primary healthcare centers (PHCs) were estimated The conflict has inflicted extensive damage to be fully functional across Syria. A forthcoming to Syria’s physical infrastructure. During the damage assessment conducted by the World Bank in initial stage of conflict, cities like Homs, Aleppo, and collaboration with the European Union (EU) revealed Damascus, and many smaller towns, have served as that Syria continued to suffer significant physical battlegrounds for government and rebel offensives. damage in selected sectors and cities in late 2021. Over time, the conflict has caused the partial or The conflict has substantially impacted full breakdown of urban systems in many cities by human lives and caused dramatic changes in destroying houses and public service–related infra- Syria’s demography. In the absence of recent structure like roads, schools, and hospitals. Many official census data, knowledge of the population strategic assets, particularly in the energy, water and dynamics since the onset of the conflict comes from sanitation, and transportation sectors, have been destroyed. According to a damage assessment 1 World Bank, “The Mobility of Displaced Syrians,” (2019). 1 FIGURE 2 • Conflict-Related Casualties in Syria by Event Year, 2017 and 2021 Source: ACLED; World Bank staff estimates. Note: Deaths that resulted from both political violence and civilian targeting events are counted. estimates performed by different agencies. However, FIGURE 3 • Change in Population Density, Syria, all estimates suggest that during the conflict, massive 2010–2020 (100-meter resolution) and rapid movements of Syrians took place, both internally and in the direction of other countries. The exodus of Syrians from some places and their influx into others have changed the distribution of Syria’s population dramatically (Figure 3). The demographic structure of the Syrian population has also been severely affected by conflict, with the male deficit in prime-age adult population being its most promi- nent feature. (Special Focus Chapter). According to the latest estimates by the Humanitarian Needs Assessment Programme (HNAP), the total population within Syria, including residents, returnees, and IDPs registered at 21.1 million as of February 2022, slightly Source: WorldPop; World Bank staff estimates. short of the 2010 (pre-conflict) population of 21.4 mil- lion but about one-third short of the 30.7 million that was projected for 2022 in the absence of the conflict, world, according to the Global Internal Displacement given the country’s high fertility rate (about 3.5 births Database. Furthermore, disasters, most resulting per woman).2 from floods, gales, and snowstorms, triggered 79,000 More than half of Syria’s pre-conflict popula- displacements in Syria in 2021.3 tion remains displaced, including 6.8 million IDPs The conflict has also triggered intangible in Syria and 6.9 million Syrian refugees displaced effects that proved detrimental to the economy. abroad (Figure 4 and Figure 5). The scale and extent In addition to damaging strategic assets (the destruc- of the ensuing forced displacement triggered by the tion channel) and triggering forced outmigration and Syrian conflict represents one of the largest displace- ments since World War II. In 2021, conflict and violence 2 Assuming Syria’s population would have continued to triggered 456,000 internal displacements in Syria, the grow at its 2005–2010 average rate (i.e., at 3.07 percent). lowest since 2012. Yet, new internal displacements 3 “Syrian Arab Republic” country profile, Internal in Syria in 2021 ranked among the ten highest in the Displacement Monitoring Centre. 2 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS FIGURE 4 • Syrian Refugees by Hosting Countries FIGURE 5 • Internally Displaced Persons (IDPs) (Million) (Million) 8,000 10 7,000 6,000 8 5,000 4,000 6 3,000 2,000 4 1,000 0 2 2010 2011 2012 2013 2014 2015 2016 20 17 20 18 20 19 20 20 20 21 0 Turkey Lebanon Jordan 2012 2013 2014 2015 2016 20 17 20 18 20 19 20 20 20 21 Germany Iraq Others Source: United Nations High Commissioner for Refugees (UNHCR)’s Refugee Source: United Nations High Commissioner for Refugees (UNHCR)’s Refugee Population Statistics Database. Population Statistics Database. Note: Both registered refugees and asylum-seekers included. Asylum-seekers refers Note: Both registered refugees and asylum-seekers included. Asylum-seekers refers to individuals who have sought international protection but whose claims for refugee to individuals who have sought international protection but whose claims for refugee status have not yet been determined. status have not yet been determined. domestic displacement (the displacement channel), economic networks has halved the size of the the Syrian conflict has also disrupted social and eco- economy compared to 2010. According to Syria’s nomic networks (the disorganization channel).4 These Central Bureau of Statistics, GDP contracted by 52 per- disruptions did not necessarily reduce the stock of cent between 2010 and 2019 in real terms (Figure 7). productive assets directly, but they reduced the rates A more appropriate estimate of the cost of the conflict at which the economy effectively utilizes those assets. as far as GDP is concerned is to compare economic Since the conflict, the division of the previously activity in 2019 with what it would have been in the integrated areas in Syria has cut off connectivity, absence of the conflict. Such an analysis is possible public service delivery systems, and supply chains. using a tool called a synthetic control method (SCM).6 Furthermore, intensified rent-seeking eroded social cohesion, and governance degradation brought addi- 4 See “The Toll of War” (World Bank, 2017) for a description tional obstacles to economic activity. The Syria Center of the “3D” channels. for Policy Research (SCPR) documented a significant 5 SCPR. “Social Degradation in Syria: The Impact of decrease in Syrian social capital, defined over three Conflict on Social Capital,” Syrian Center for Policy broad categories: social networks and community Research, Beirut, (2017). participation, trust, and shared values and attitudes.5 6 The SCM consists of searching for a weighted combination of countries that resemble as closely as While Syria ranked below the Middle East and North possible the economic characteristics of Syria in the Africa (MENA) average on several governance indi- pre-conflict period to create a synthetic economy (the cators before the conflict, they remained low and, control) to see how this economy performed over Syria’s in many instances, further deteriorated during the conflict period and compare it with Syria’s actual growth conflict (Figure 6). Consistent with this trend, Syria performance. We conduct the SCM analysis using a ranked 178th out of 180 countries (worst 1 percent) in pool of all countries in the world, excluding those that were involved in or significantly affected by the conflict. corruption perception index in 2021, down from 127th The vector of predictors that is selected to explain real out of 178 countries (worst 30 percent) in 2010. GDP includes employment (employers per hundred The destruction of physical capital, casual- persons), trade openness (trade as percentage of GDP), ties, forced displacement, and the breakup of industry shares (agriculture and industry as percentage The Conflict Context 3 Worldwide Governance Indicators, FIGURE 6 •  FIGURE 7 • Actual and Counterfactual GDP, Syria Syria (Billions, constant 2015 US$) (Percentile rank among all countries ranges from 0 (lowest) to 100 45 (highest)) 40 60 35 50 30 40 25 20 30 15 20 10 10 5 0 0 1996 1998 2000 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Voice and Political stability and Actual Counterfactural accountability violence Government effectiveness Regulatory quality Rule of law Control of corruption Source: World Development Indicators (WDI); Penn World Table 10.0; Center for Systemic Peace; World Bank staff estimates. Note: The counterfactual results account for the regional negative impact that growth in Source: Worldwide Governance Indicators report (various years). the MENA region was lower than the global average in the post-conflict period. Through an SCM analysis, we estimate that in 2019, 189 countries in 2020, down from 110th out of 169 Syria’s GDP absent the conflict (the counterfactual) countries in 2010.9 would have been US$ 38.6 billion in 2015 constant The conflict in Syria also imposed a heavy prices, compared to US $16.3 billion of the realized economic and social toll on the country’s neigh- GDP (Figure 7) (Technical Appendix A). bors in the Mashreq region. From 2011 to 2018, Conflict, displacement, and the collapse average annual GDP growth rates were reduced by of economic activity have all contributed to the 1.2 percentage points in Iraq, 1.6 percentage points decline in household welfare. Extreme poverty in Jordan, and 1.7 percentage points in Lebanon has consistently risen since the onset of the conflict, reflecting deteriorating livelihood opportunities and a progressive depletion of household coping of GDP), investment share (gross capital formation as capacity. In non-monetary terms, access to shelter, percentage of GDP), physical capital (capital stock per health, education, and water and sanitation have capita), human capital (human capital index from the all worsened dramatically since the conflict began. Penn World Table), and political status (democracy index). The analysis relies on a cross-country panel data In addition, human capital has eroded primarily set for the period between 1995 and 2019. due to the interruption of schooling of the young 7 According to the estimates from the United Nations population.7 Preliminary calculations show that International Children’s Emergency Fund (UNICEF), 2.4 the combined effects of casualties, forced disper- million children have been forced out of school since the sion, and reduced investments in human capital start of the conflict. could add up to a 30 percent permanent loss in 8 Hamilton and Nguyen (2017). 9 The HDI measures each country’s social and economic the country’s human capital stock (compared with development by focusing on the following four factors: the 2010 stock).8 The Human Development Index mean years of schooling, expected years of schooling, (HDI) complied by the United Nations Development life expectancy at birth, and gross national income (GNI) Programme (UNDP) positioned Syria 151st out of per capita. 4 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS FIGURE 8 • GDP Impact of the Syrian Conflict in the Mashreq Region A. Trade channel B. Productivity channel 6 8 7 5 6 4 5 3 4 3 2 2 1 1 0 0 Jordan Lebanon Iraq Jordan Lebanon Average annual GDP growth during 2011–18 (actual) Average annual GDP growth during 2011–18 (actual) Average annual GDP growth during 2011–18 Average annual GDP growth during 2011–18 (no trade shock) (no TFP shock) Source: The Fallout of War: The Regional Consequences of the Conflict in Syria, June 2020, World Bank. Note: The computable general equilibrium (CGE) model underlying the simulations is the Global Trade Analysis Project (GTAP) model. in real terms solely because of the conflict in Syria. in a mix of congestion and fiscal effects. In education Cumulatively, these reductions correspond to and water, adaptations in the provision of services 11.3 percent of the combined pre-conflict GDPs in largely prevented congestion. In transport, health, 2010 across these three countries.10 With decreasing and energy, congestion was observed. Driven by the transit trade through Syria and stalling service refugee-induced demand, the fiscal burden through exports like tourism, the trade shock lowered growth embedded subsidies has increased. In general, Syrian in Syria’s neighboring countries (Figure 8.A). The refugees faced adverse conditions in exile.12 Yet, with productivity shocks were also substantial, reflecting persistent concerns regarding insecurity in Syria, the intangible factors that have depressed the growth return rates for the Syrian refugees are low, leading through stalling total factor productivity (TFP) growth (Figure 8.B). In contrast, refugee arrivals boosted growth by increasing aggregate demand and labor 10 The synthetic control method (SCM) is employed to supply. The net negative economic impact of the assess GDP in Iraq, Jordan and Lebanon in the absence Syrian conflict on Iraq, Jordan, and Lebanon has of a conflict in Syria. For more information, see “The been remarkably high compared to similar situations Fallout of War,” (World Bank, 2020). elsewhere in the world over the last few decades, 11 World Bank, “The Fallout of War,” (2020). 12 For instance, when it came to the war’s impact on school- driven by three factors: (i) the sheer scale of the going children, the school attendance ratio of Syrian Syrian conflict and ensuing forced displacement; children remained higher in Syria than it did in Lebanon (ii) the high exposure of neighboring countries to a or Jordan, mainly because many Syrian refugees had possible fallout; and (iii) the low institutional resilience to adopt adverse coping strategies in their countries in neighboring countries, which propagated the of asylum; Syrian girls dropped out of school to get shock further. married at a younger age and Syrian boys dropped out to bring extra income for their families’ survival. For The prolonged displacement posed pro- these children, human capital accumulation stops when found challenges to the hosting countries in the they leave school, with persistent effects on their lifetime Mashreq (World Bank, 2020).11 Arrivals of refugees well-being. For details, see World Bank, “The Mobility of have boosted demand for public services, resulting Displaced Syrians,” (2019). The Conflict Context 5 to the continued protracted nature of displacement.13 to a survey conducted in April 2020, the pandemic According to the United Nations High Commissioner caused 35 percent of Syrian refugees to lose their for Refugees (UHNCR), only about 4 percent (177,500) jobs, compared to 17 percent of Jordanian citizens.14 of the total Syrian refugee population returned to Syria as of March 2022. The COVID-19 pandemic has had a particularly acute impact on displaced refugees 13 World Bank, “The Mobility of Displaced Syrians,” (2019). who are largely working in the informal sector; this, 14 International Labour Organization (ILO), “Impact of in turn, has also increased the burden of care on COVID-19 on Workers in Jordan: A Rapid Assessment,” host countries. In Jordan, for instance, according (2020). 6 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS 2 RECENT ECONOMIC DEVELOPMENTS Deteriorating economic situation become inaccessible.17 Syria’s overall score on the World Bank’s Statistical Capacity Indicator (SCI),18 Beyond the immediate conflict impact, the Syrian economy suffers from the compounding effects of 15 The Caesar Syria Civilian Protection Act of 2019, also the pandemic, adverse weather events, regional known as the Caesar Act, is US legislation that sanctions fragility, and macroeconomic instability. Since the Syrian government, including Syrian president Bashar 2020, Syria’s external economic ties have been severely al-Assad, for war crimes against the Syrian population. The Act was signed into law by President Trump in restrained by the deepening crisis in neighboring December 2019 and came into force on June 17, 2020. Lebanon and Turkey, as well as the introduction of new 16 According to Food and Agriculture Organization (FAO), US sanctions under the Caesar Act.15 Domestically, wheat production in 2021 is estimated at around 1.05 a severe drought, worsened by a warming climate million tonnes, down from 2.8 million in 2020 and only and shortages of agricultural inputs, has led to a one-quarter of the pre-crisis average of 4.1 million tonnes dramatic reduction in agricultural production.16 With during the period 2002–2011. a severely degraded health care system following the 17 Syria still uses the 1993 System of National Accounts (SNA) methodology with a base year of 2000, suggesting decade-long war, COVID-19 has only exacerbated the the benchmark estimates have a lag of nearly 20 years. vulnerable situation. Since March 2022, soaring prices For comparison, in 2020, the benchmark estimates triggered by the war in Ukraine have been adversely in about 64 percent of economies worldwide had a affecting Syria as a net food and fuel importer. lag of less than ten years. Furthermore, 63 percent In the absence of an official GDP figure, of economies worldwide had reported their national accounts in line with the new versions of 2008 SNA or this report uses nighttime light emissions to 2010 European System of Accounts (IMF, 2022). infer changes in economic conditions in recent 18 The World Bank’s Statistical Capacity Indicator is a years. With the prolonged conflict, Syria’s statistical composite score assessing the capacity of a country’s capacity has taken a heavy blow, and reliable and statistical system. It is based on a diagnostic framework timely information regarding national accounts have assessing the following areas: methodology; data 7 which used to be in line with the average of the a strong correlation with the historical movements MENA region, has declined sharply since the conflict, of GDP. Through a quantitative analysis of the rela- ranking last among the 146 countries surveyed in tionship between NTLs and economic activity, we 2020. To overcome this serious limitation to economic estimate that Syria’s real GDP growth slowed from monitoring, we make use of nighttime lights (NTLs), 3.7 percent in 2019 to 1.3 percent in 2020 and then which provide an important source of information turned negative, to –2.1 percent, in 2021 (see Box 1 about economic activity. In Syria, NTLs have shown and Technical Appendix B for details). BOX 1: NOWCASTING ECONOMIC ACTIVITY USING NIGHTTIME LIGHTS Nighttime lights (NTLs) are an important source of information about economic activity in Syria, significantly expanding and enriching the information set. NTLs data are high frequency, granular, and insulated from human error or misuse (e.g., misinformation). As such, they are particularly welcome in a conflict context as they provide more timely, granular (spatial information is readily available), comprehensive (they cover 100 percent of Syria’s territory), and potentially more reliable information than official national accounts data. In Syria, NTLs have shown a strong correlation with historical movements of GDP. Both NTLs and GDP climbed steadily from 1992 until 2010, a although the growth in GDP was relatively higher in later years (Figure 9.A). NTLs decreased significantly since the start of conflict, corroborating the large decline in real GDP. It was only in 2017 that NTLs began to recover, but they declined again in 2021. The regression of NTLs against real GDP in logarithmic form reveals a positive and significant relationship between the two. Specifically, the coefficient for this regression is 0.726, suggesting that for every 1 percent increase in NTLs, real GDP increases by almost three-quarter of a percent. Although this estimate is slightly higher than the estimates of Henderson (2012), it remains within the 95 percent confidence interval, giving some additional confidence to the estimate. The conflict in Syria has substantially changed the relationship between NTLs and GDP. The post-conflict estimate of the elasticity between NTLs and GDP is only half, approximately, of the magnitude of the pre-conflict estimate (Table 1: Regression results of historical NTLs and real b GDP). The lower elasticity between NTLs and GDP could be partly explained by the sharp decline in oil-GDP since the start of the conflict. It may also indicate that the official GDP may not have adequately reflected the dramatic changes in economic activity after the conflict. Indeed, if applying the pre-conflict estimate of the elasticity between NTLs and GDP to predict Syria’s GDP after the conflict, the economic contraction in Syria between 2010 and 2017 would be worse than Syrian statistics suggested. The evolution of NTLs by subnational regions allows us to understand the spatial dynamics of economic activity in Syria. In general, most areas exhibit a U-shaped pattern beginning in 2014 and bottoming out in 2017 (See Technical Appendix B). However, large variation exists across regions. By 2021, some regions had a level of economic activity that was still far lower than it was pre-conflict, while a few have recovered close to their pre-conflict levels. Overall, there appears to be a sharper decline in economic activity in opposition-controlled areas than in government- control areas since the conflict began. Demonstrating the persistent impact of the conflict, economic activity appears to have fallen more pronouncedly in conflict-intensive regions (Figure 2: Conflict-related casualties in Syria by event year, 2017 and 2021 and Figure 9). TABLE 1 • Regression Results of Historical NTLs and Real GDP (1) (2) (3) Dependent Variable log(GDP) GDP GDP log(NTLs) 0.726*** (0.009) NTLs 0.072*** (0.005) NTLs * Pre-Conflict 0.071** (0.008) NTLs * Post-Conflict 0.038** (0.015) Observations 28 28 28 R2 0.823 0.861 0.882 Adjusted R2 0.879 0.853 0.881 Note: *p<0.1; **p<0.05; ***p<0.01. (continued on next page) 8 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS BOX 1: NOWCASTING ECONOMIC ACTIVITY USING NIGHTTIME LIGHTS (continued) FIGURE 9 • Night-Time Lights and GDP A. Night-time lights, GDP and predicted GDP B. Changes in predicted GDP, by governorate (GDP in constant 2015 US$, billion; light emissions) (2021 levels compared to 2010 levels, percent) 40 2,600 35 2,100 Al-Hasakeh 30 Aleppo 1,600 Ar-Raqqa 25 Lattakia Idleb 20 1,100 Hama Deir-ez-Zor Tartous 15 600 Homs 10 Damascu 100 5 Rular Damascus Quneitra 0 –400 Dar'a 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 As-Sweida GDP Nightlight-based GDP projections < –60 –60 ~ –40 Non-flaring nighttime lights (RHS) –40 ~ –20 >= 0 Source: The Visible Infrared Imaging Radiometer Suite (VIIRS) and Defense Meteorological Satellite Program (DMSP) satellites; WDI; World Bank staff estimates. Note: NTLs are regressed against real GDP in logarithmic form. The elasticity estimates acquired from the regression are applied to convert the NTLs observations into regional GDP. See Technical Appendix B for details. a To carve out the effects of flaring, we estimate the Sum-of-Lights for all areas that have been tagged as flaring sites by the World Bank’s Global Gas Flaring Reduction Partnership (GGFRP) dataset. This carves out these extremely bright lights that otherwise bias the estimates, allowing us to focus only on conventional sources of electrical lights. This is particularly important in the case of Syria, as flaring nearly quadrupled in early 2017 while remaining relatively flat otherwise. b If the non-oil GDP data is available, non-flaring NTLs are more appropriate to be regressed against non-oil GDP rather than aggregate GDP. Non-oil GDP roughly accounted for 10 percent of GDP in Syria before the conflict (IMF, 2016). As such, regressing non-flaring NTLs against aggregate GDP would result in a about 10 percent overestimation of the elasticity (between non-flaring NTLs and non-oil GDP) prior to the conflict. With oil production plummeting after the war, aggregate GDP became closer to non-oil GDP since the conflict began. Hence, the elasticity is overestimated less during the post-conflict period. Economic activity has shrunk significantly government-control areas, and in governorates where in conflict-intensive regions. The spatial dimen- active conflict was concentrated (Box 1). In Lattakia sion of developments takes on a critical dimension and Tartous, two port cities that have experienced in fragile, conflict, and violence (FCV) contexts: only limited conflict or destruction, the decline in NTL- averaging across areas of the country can be highly based outputs were also significant, likely caused by misleading, as conflict intensity varies widely across a collapse in trade activity. In contrast, according to time and space, which in turn generates large hetero- a study conducted by Mercy Corps in 2021 using geneity of economic and social conditions across time more granular NTLs data,19 economic activity appears and space. Statistics on regional economic output, to have been quite robust in some border areas, typically measured as regional GDP, do not exist for Syria. Alternatively, NTL-based outputs are estimated sources; and periodicity and timeliness. Countries are to predict economic activity at the governorate level. scored against 25 criteria in these areas, using publicly As the estimates suggest, since the conflict began, available information and/or country input. there appears to have been a sharper decline in the 19 See Mercy Corps, “Using Night Lights to Measure predicted GDP in opposition-controlled areas than in Economic Output in Syria,” (May 27, 2021)). Recent Economic Developments 9 FIGURE 10 • Economic Activity in Syria A. Petroleum and other liquids production B. GDP by sector (Thousand barrels per day) (Constant national prices, Index, 2010=100) 250 140 120 200 100 80 150 60 40 100 20 50 0 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 0 Agriculture Building and Construction 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 Mining, Manufacturing and Utilities Services Source: Central Bureau of Statistics of Syria; US Energy Information Administration (EIA). especially where significant economic activity is for private consumption expenditures in national taking place on the other side of the border.20 accounts. The assessment of the demand-side Economic activity has contracted across GDP therefore focuses on investment only. Before sectors since the onset of the conflict. Economic the conflict, Syria’s investment as a share of GDP disruption has been acute in the hydrocarbon sector. was comparable to that of neighboring countries. Crude oil production plunged by 80 percent from Investment declined from 19.2 percent of GDP in 2010 to 2021 (i.e., from 416,000 barrels per day 2006–2010 to 14.2 percent of GDP during 2011– (bpd) to 79,000 bpd),21 owing largely to the conflict- 2014, then to 7.6 percent of GDP during 2015–2019, caused damages to energy infrastructure networks an extremely low contribution even among fragile and (Figure 10.A). There were significant losses in agri- conflict-affected economies (Figure 11.A). Investors cultural production as a result of damage to irrigation exited Syria due to insecurity and the poor business systems, adverse weather events, and shortages of environment, causing private investment as a share labor, seeds, fertilizer, and fuel. Industrial production of GDP to decline from 12.3 percent in 2010 to 4.4 also declined, affected by shortages in fuel and power, percent in 2019. During this period, public investment limited access to capital, severe destruction of infra- also fell substantially, from 8.2 percent of GDP to 2.5 structure, and the relocation of major manufacturing percent of GDP, as revenues declined and spending bases. The service sector was disrupted as economic on arms rose (Figure 11.B). fragmentation impeded trade and commerce, security threats prevented tourism, and economic sanctions impacted financial activities (Figure 10.B). 20 Possible explanations could be smuggling, and the From the demand side, the conflict has led income derived from smuggling, which has become to a collapse in both private and public invest- a major part of the economy. Firms may stay closer to foreign suppliers of parts, as the border for exports is ment. The final consumption expenditure data in much more valuable, given that logistics within a conflict real terms in national accounts is less reliable, as country impose huge costs to firms. evidenced by a lack of close correlation between the 21 According to US Energy Information Administration (EIA) Consumer Price Index (CPI) and the price deflator data. 10 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS FIGURE 11 • Investment in Syria A. Gross fixed capital formation B. Public and private investment (Share of nominal GDP, 2015–2019) (Share of nominal GDP) 18 Syria (2015–2019) 16 14 Syria (2011–2014) 12 10 High-intensity conflict (excluding Syria) 8 6 Syria (2006–2010) 4 2 Medium-intensity conflict 0 2010 2011 2012 2013 2014 2015 2016 20 17 20 18 20 19 High institutional and social fragility Public Private 0 5 10 15 20 25 30 Source: Central Bureau of Statistics of Syria; Find My Friends Tool using the IMF World Economic Outlook (WEO); World Bank staff estimates. Note: The classification is from the FY22 list of fragile and conflict-affected countries situations, released by the World Bank. Persistent current account deficit 19.8 billion in 2010 to US$ 7.7 billion in 2019, according and dwindling foreign reserves to the official statistics (Figure 12: Dynamics of Syrian trade.B). In 2020, the top five suppliers of Syria’s imports Conflict-related disruptions and international were Turkey, China, United Arab Emirates, Egypt, and sanctions led to a collapse of Syrian trade after India, according to the mirror statistics from the UN the conflict. According to data released by the Central Comtrade database. These figures should be treated Bank of Syria (CBS), Syria’s exports fell dramatically, with caution. In particular, the introduction of sanctions from US$ 15.7 billion in 2010 to US$ 3.5 billion in in Syria may trigger evasion strategies, causing a larger 2019. Mirror statistics from the UN Comtrade database share of trade unreported since the conflict. Therefore, show a similar declining trend but lower export levels the contraction in trade activity could be smaller than in recent years. Exports plummeted, largely driven the trade statistics suggest. by a dramatic decline in oil and tourism receipts. Estimates from the maritime data suggest Earnings from these sectors, which amounted to about the trade volume may have further moderated US$ 12.8 billion in 2010, are now insignificant due to amid crisis conditions since 2020. In the absence conflict-related disruptions and sanctions (Figure 12: of recent official trade statistics, we apply maritime Dynamics of Syrian trade.A). In 2020, Syria exported data from the Automatic Identification System (AIS), primarily agricultural goods, such as olive oil, seeds, which has emerged as a potential source for real-time and nuts. Top five destination for Syrian exports were information on trade activity.22 Maritime data provides Turkey, Saudi Arabia, Lebanon, Egypt, and United Arab Emirates, according to the mirror statistics from the UN Comtrade database. The prolonged civil war 22 Over 80 percent of global merchandise trade by volume has led to a collapse in domestic industrial output and and more than 70 percent of its value is carried by agricultural supply, making the country heavily reliant the international shipping industry (United Nations Conference on Trade and Development (UNCTAD, on manufactured goods and foodstuffs produced 2018). Cargo ships are equipped with a device that overseas. Consequently, although imports also periodically emits a signal (Automatic Identification contracted since the conflict began, the decline is less System message, or AIS), which contains information on significant than that of exports, which dropped from US$ the vessel’s location, speed, draught, etc. Recent Economic Developments 11 FIGURE 12 • Dynamics of Syrian Trade A. Gross exports B. Gross imports (Billion $US) (Billion $US) 20 30 18 25 16 14 20 12 15 10 8 10 6 5 4 2 0 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 0 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 Textiles Agriculture Stone Minerals Minerals Textiles Agriculture Metals Chemicals Vehicles Machinery Chemicals Services Others Electronics Others Services Source: Central Bureau of Statistics of Syria; Atlas of Economic Complexity, Center for International Development at Harvard University; World Bank staff estimates. Note: The analysis uses mirror data from the UN Comtrade database. a good indication of trade activity in Syria, as the FIGURE 13 • Shipping Density Map country is heavily dependent on maritime transport (June 2021, all vessel categories) for trade.23 The limitation of this data lies in the fact that ships may not be equipped with a device that periodically emits a signal, and ships seeking to evade seizure or engaging in illegal activities may turn off their AIS transporters near Syrian waters. Syria is served by two primary ports along its Mediterranean coastline, in Lattakia and Tartous. Although the conflict has not touched either port, activity in both has declined sharply. By June 2021, Syria’s coastline had a much lower maritime traffic density than any of its neighbors (Figure 13). Estimates from the mari- time data suggest Syria’s import volume in terms of metric tons has more than halved from 2019 to 2021, possibly due to new policies that have restricted the Source: European Marine Observation and Data Network (EMODnet). imports of non-essential goods (Figure 14.A and Note: The platform utilizes AIS shipping-position data collected by MarineTraffic. It assesses import and export vessel capacity measured in deadweight tons (DWT) by Figure 14.B). Mirror statistics from the UN Comtrade country and ship type, and then uses machine-learning algorithms to estimate metric database show a similar decline for imports in value tons of cargo carried. terms in recent years. Despite the negative impact of the COVID-19 outbreak and disruptions in supply chains, maritime data show that the declining trend in exports between 2010 and 2019 appears to 23 According to the World Trade Organization (WTO), have reversed from 2019 to 2021. For 2020, when for Syria, 74.6 percent of imports and 26.8 percent of data from both sources are available, the uptick of exports, respectively, were carried by sea in 2011. https:// seaborne export volume is consistent with the trends www.wto.org/english/res_e/statis_e/daily_update_e/ revealed by mirror statistics from the UN Comtrade trade_profiles/SY_e.pdf. 12 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS FIGURE 14 • Seaborne Trade Volume of Syria A. Number of port calls B. Metric tons of cargo (Number of port visits, total in a year, all vessel categories) (Daily average in a year, all vessel categories) 450 9,000 400 8,000 350 7,000 300 6,000 250 5,000 200 4,000 150 3,000 100 2,000 50 1,000 0 0 2016 2017 2018 2019 2020 2021 2016 2017 2018 2019 2020 2021 Exports Imports Exports Imports Source: Cerdeiro, Komaromi, Liu and Saeed (2020); UN Comtrade Monitor; World Bank staff estimates. database. Syria’s improved trade relations with its which exceeded US$ 2 billion in 2010, disappeared after Arab neighbors may have contributed to a possible 2011. Syria’s reserve losses were partly limited by about increase in exports since 2019.24 US$ 7 billion in credit lines provided by the Iranian and Syria has experienced a persistent current Russian governments.27 Yet, financial assistance from account deficit since the onset of the conflict. these allied countries is insufficient to cover a large Constrained by sanctions and trade embargos, exports remained substantially lower than imports, resulting in 24 Jordanian authorities announced the reopening of the persistent high trade deficits. The losses were partly Jaber border crossing with Syria for cargo and passengers, offset by net transfer inflows, which accounted for as of September 29, 2021. Syria and Pakistan signed a an average of 11 percent of GDP during 2011–2019, memorandum of understanding to boost bilateral trade according to the official balance of payments (BoP) and expand economic ties on October 31, 2021. The statistics. Notably, net remittance inflows increased Syrian president visited the United Arab Emirates (UAE) on March 18, 2022; his first trip to an Arab country since 2011. from US$ 1.1 billion in 2010 to US$ 1.6 billion in 2019, 25 Remittance inflows were reported at US$ 2.1 billion primarily driven by increased remittances inflows from in 2017, according to the official balance of payments refugees and migrant workers.25 According to the (BoP) statistics. Including informal remittance inflows, same official BoP statistics, the increase in transfer the estimates provided by Economic and Social inflows was also driven by international aid to Syria, Commission for Western Asia (ESCWA) were much which climbed to US$ 1.5 billion in 2019 from a negli- higher, at US$ 8.5 billion, over the same year. 26 The reported humanitarian donor funding for Syria was gible amount prior to the conflict (Figure 15.A).26 As a US$ 2.4 billion in 2019, according to data collected by result, Syria’s current account deficit as a percentage the UN Financial Tracking Service (FTS). of GDP decreased from a peak of 32 percent in 2014 27 Syria reportedly received from the Iranian government to single digits in recent years, according to the official lines of credit amounting to US$ 1.9 billion in 2013, US$ balance of BoP statistics (Figure 15.B). 3 billion in 2014, US$ 0.97 billion in 2015, and US$ 0.5 Syria’s foreign exchange reserves are esti- billion in 2017. In addition, according to an agreement signed by the Syrian and Russian ministers of finance mated to have been almost completely depleted. and reviewed by The Syria Report, Russia granted Reflecting the impact of sanctions and conflict, Syria Damascus EUR 240 million in May 2014. Recently, Syria experienced major capital flight since the start of the signed a contract for a Russian loan of US$ 700 million conflict. Meanwhile, net foreign direct investment (FDI), in December 2020. Recent Economic Developments 13 FIGURE 15 • Syria’s Balance of Payments A. Remittances and international aid B. Current account balance (Million US$) (Percentage share in GDP) 3,000 30 2,500 20 10 2,000 0 1,500 –10 1,000 –20 500 –30 –40 0 –50 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 –500 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Trade balance Services balance Net inflows of private remittances Income balance Current transfers Net inflows of of international aid Current account balance Source: Central Bureau of Statistics of Syria; Central Bank of Syria; World Bank staff estimates. Note: “Workers’ remittances” plus “compensation of employees” from the balance of payments are used to estimate remittances, while “government transfers” plus “other transfers” are applied to estimate international aid. cumulative current account deficit that amounted to proven limited, as evidenced by the continuous US$ 34 billion from 2011–2019, according to the official depletion of foreign reserves. statistics. This accumulated deficit in Syria’s balance of The depreciation of the Syrian pound payments since the conflict began suggests its foreign gained momentum with the start of the exchange reserves of US$ 19.5 billion in 2010 may have Lebanese currency crisis in late 2019. Following been almost completely depleted.28 the onset of Lebanon’s financial crisis, the correla- tion between the movements and volatilities of the Lebanese Pound (LBP) and SYP increased (See Currency depreciations and surging Box 2 and Technical Appendix C for details). The inflation close commercial and trade ties between Lebanon and Syria as well as Syrians’ reliance on Lebanese Since the start of the conflict, the Syrian pound banks for their commercial and personal activities has continuously depreciated against the US can explain the tight link between the Syrian and dollar. The official exchange rate of the Syrian Lebanese pounds prior to July 2021.29 Furthermore, pound (SYP) has declined 50-fold against the US Syrian businesses’ reported use of the Lebanese dollar (US$) since 2011, reaching 2,512 SYP/US$ in black market to obtain US dollars and avert the sanc- May 2022. The market exchange rate, on the other tions of the Caesar Act, coupled with the smuggling hand, registered an 80-fold depreciation during the same period, reaching 3,905 SYP/US$ (Figure 17). 28 Judging from the fact that the authorities have The Central Bank of Syria has taken many measures aggressively restricted imports of non-critical goods to ease currency depreciation, including curtailing since 2019, Syria’s current foreign exchange reserves foreign currency demand, tightening import licensing, should be very tight indeed. raising the interest rates on Syrian pound deposits, 29 A study conducted by Professor Ali Kanaan of the University of Damascus estimated that Syrian deposits in and obliging exporters to surrender foreign currency Lebanon worth around US$ 45 billion, roughly the same earnings to the central bank to help raise US dollar size as Syria’s GDP. See Asharq Al-Awsat, “Damascus liquidity. Nevertheless, the ability of the authorities Estimates Syrian Deposits in Lebanese Banks Worth to intervene effectively to support the currency has $45 Billion,” (2020). 14 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS FIGURE 16 • Fuel Prices and Fuel Imports in Selected MENA Countries A. Diesel price, market price B. Gas price, market price (per litre, $US) (per kg, $US) 1.6 1.8 1.4 1.6 1.2 1.4 1.2 1.0 1.0 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0.0 0.0 Jan-19 Mar-19 May-19 Jul-19 Sep-19 Nov-19 Jan-20 Mar-20 May-20 Jul-20 Sep-20 Nov-20 Jan-21 Mar-21 May-21 Jul-21 Sep-21 Nov-21 Jan-22 Mar-22 May-22 Jan-19 Mar-19 May-19 Jul-19 Sep-19 Nov-19 Jan-20 Mar-20 May-20 Jul-20 Sep-20 Nov-20 Jan-21 Mar-21 May-21 Jul-21 Sep-21 Nov-21 Jan-22 Mar-22 May-22 Lebanon Syria Turkey Iraq Lebanon Syria Turkey Iraq Jordan Yemen Jordan Yemen Libya C. Oil price, Syria and Lebanon D. Lebanon imports of petroleum oil (US$ per ton) (thousand tons; million US$, 3-month moving average) 1,400 1,200 1,200 1,000 1,000 800 800 600 600 400 400 200 200 0 0 Dec-18 Feb-19 Apr-19 Jun-19 Aug-19 Oct-19 Dec-19 Feb-20 Apr-20 Jun-20 Aug-20 Oct-20 Dec-20 Feb-21 Apr-21 Jun-21 Aug-21 Oct-21 Dec-21 Feb-22 Apr-22 Jun-17 Sep-17 Dec-17 Mar-18 Jun-18 Sep-18 Dec-18 Mar-19 Jun-19 Sep-19 Dec-19 Mar-20 Jun-20 Sep-20 Dec-20 Mar-21 Jun-21 Sep-21 Dec-21 Mar-22 Petroleum oil price: Lebanon imports Import volume of petroleum oil (thousand tons) Petroleum oil price: Syrian imports Import value of petroleum oil (million US$) Domestic diesel price: Lebanon Brent crude oil price: Global Domestic (transport) diesel price: Syria Source: WFP Price Bulletin, country office reports; UN Comtrade database, World Bank staff estimates. of subsidized goods, gasoline, and diesel from explained, in part, by the lower demand for dollars Lebanon to Syria, have created exchange market in Syria to purchase the smuggled subsidized pressures and simultaneous demand for US dollars goods, gasoline, and diesel from Lebanon after in Syria and Lebanon. This, in turn, led to a tightening the termination of subsidies. Indeed, owing to the in the link between the two currencies. subsidies that were terminated only in September The link between the two currencies 2021, the prices of diesel and gasoline in Lebanon weakened after September 2021, the date of were the lowest among the comparators: Syria, Iraq, the termination of subsidies in Lebanon, which Turkey, Yemen, Libya, and Jordan, some of which indicates the easing of simultaneous exchange are fuel-rich countries (Figure 16.A, Figure 16.B, market pressures. This apparent decoupling in and Figure 16.C). This created a strong incentive the movement of the two currencies is likely to be for smuggling, particularly given that the centers of Recent Economic Developments 15 BOX 2: CONNECTEDNESS BETWEEN THE SYRIAN AND LEBANESE POUNDS Following the onset of Lebanon’s financial crisis, the correlation between the movements in the Lebanese (LBP) and Syrian (SYP) pounds increased (Figure 17.A). Indeed, movements in the LBP and SYP appeared to be closely connected. Further, the two currencies appeared to exhibit common bouts of volatility. The empirical findings suggest that the changes in the LBP drive the changes in the SYP. More specifically, movements in the LBP have predictive power for changes in the SYP but the converse is not true. That is, the informational content of changes in the LBP is useful for predicting changes in the SYP. This is referred to, in technical parlance, as changes in the LBP Granger-cause changes in the SYP (Table 2: Granger causality tests with the levels and returns of the LBP and SYP).a The findings also indicate that movements in the LBP lead changes in the SYP. b The connectedness between the SYP and LBP can be assessed empirically using econometric techniques offered in the literature. More specifically, the correlation between the movements in the two currencies can be gauged via time-varying correlations or using a multivariate conditional heteroskedasticity model while spillovers in volatility can be measured using the total volatility spillover index of Diebold and Yilmaz (2009, 2012). The findings also indicate that there are commonalities in the movements and volatilities of the SYP and LBP. More specifically, in the wake of Lebanon’s financial crisis, the correlation between the movements of the two currencies increased. The correlation also soared in summer of 2021 and in early 2022. However, there appeared to be a decoupling in the movements and volatilities of the two currencies following the termination of the subsidy scheme in Lebanon in September 2021 (Figure 17: Connectedness between the Syrian and Lebanese Pounds.B). The pattern in the volatility spillovers mimics that of the correlation in the two currencies, in that there was a marked increase in volatility spillovers at the onset of the Lebanese financial crisis in October and November of 2019 and during the summer of 2021, and spillovers in volatility are increasing in early 2022. Moreover, the findings suggest a greater degree of connectedness between SYP and LBP than between any of the latter two currencies and the Turkish Lira. FIGURE 17 • Connectedness between the Syrian and Lebanese Pounds A. The market exchange rate B. Time-varying correlation between the SYP and LBP (LBP/USD; SYP/USD) (30-day moving average) 12,000 35,000 0.5 The termination of subsidies 30,000 0.4 in Lebanon 10,000 25,000 0.3 8,000 20,000 0.2 6,000 15,000 0.1 4,000 10,000 0.0 5,000 –01 2,000 –0.2 0 0 12/2018 3/2019 6/2019 9/2019 12/2019 3/2020 6/2020 9/2020 12/2020 3/2021 6/2021 9/2021 12/2021 3/2022 –0.3 –0.4 11/2019 2/2020 5/2020 8/2020 11/2020 2/2021 5/2021 8/2021 11/2021 2/2022 SYP/USD LBP/USD (RHS) Source: The Visible Infrared Imaging Radiometer Suite (VIIRS) and Defense Meteorological Satellite Program (DMSP) satellites; WDI; World Bank staff estimates. Note: NTLs are regressed against real GDP in logarithmic form. The elasticity estimates acquired from the regression are applied to convert the NTLs observations into regional GDP. See Technical Appendix B for details. a A time-series process xt is said to Granger-cause the other process yt if exploiting information in the past, and contemporaneous values of xt lowers the prediction error of the process yt at some horizon h. That is, the informational content in xt is useful for predicting yt. b Autocorrelations, which measure the self-dependence in a time series, as well as cross-correlations, can be used to gauge the lead-lag relation between SYP and LBP. The evidence suggests the existence of a lead-lag relation between the SYP and LBP, and that movements in the LBP lead changes in the SYP. (continued on next page) 16 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS BOX 2: CONNECTEDNESS BETWEEN THE SYRIAN AND LEBANESE POUNDS (continued) TABLE 2 • Granger Causality Tests with the Levels and Returns of the LBP and SYP Panel A: Granger causality Òtests with returns F-statistic p-value LBP does not Granger Cause SYP 1.84 0.000 SPY does not Granger Cause LBP 0.66 0.969 Panel B: Granger causality tests with levels F-statistic p-value LBP does not Granger Cause SYP 1.44 0.025 SPY does not Granger Cause LBP 0.59 0.991 economic activity in Syria are close to the Lebanese 30 The purported increase in smuggling activity along border.30 The smuggling hypothesis is supported by the porous Lebanese-Syrian border is a case in point, evidence that Lebanon has imported a large volume illustrating the downside of untargeted subsidies. of oil derivatives before the termination of subsidies While smuggling can come from other countries, the (Figure 16.D). transportation costs may be more prohibitive and may The less precipitous depreciation in the affect profitability. Iraq, for example, is farther from Syrian pound since September 2021 is likely to major cities in Syria and the transportation costs would, therefore, be higher. be partly attributable to the import restrictions 31 In Syria, the official exchange rate is used for: (i) that were imposed by the Syrian authorities. the state budget and public sector transactions; (ii) These policies, which restrict the imports of non- money transfers from abroad through the official essential goods, are aimed at restricting the use of channels; (iii) fees paid by Syrian men seeking to the limited foreign currency reserves to essential food avoid mandatory military service; (iv) international aid imports, thereby reducing the demand for US dollars. operations; and (v) imports of critical commodities Furthermore, the Syrian authorities have drastically such as sugar, rice, vegetable oil, and selected medical products. The market exchange rate is reduced the list of critical goods that are imported applied when private funds are transferred into Syria at the preferential exchange rates, leading to lower through unofficial channels. Imports of non-critical margins of profitability for importers and, hence, a goods also apply the market exchange rates. There diminished incentive to import. That diminished incen- are other exchange rates in Syria. This includes “the tive, along with higher import prices for consumers, banks and financial institutions” rate, which is used also reduced demand for imports. The import-restric- by private banks and financial institutions to conduct tion policies likely contributed to the lower volatility of transactions, including financing imports and exports of the private sector; the “remittance” rate, which was the SYP and to the weakened links between the SYP set by the CBS in April 2013 and is used by Syrians and LBP since September 2021. sending money from abroad; the “United Nations” In light of a multiple exchange rate system31 rate, which was set by the CBS in December 2011 and the significant gap between various rates, the and is used by UN agencies operating in Syria; the report uses consumption-based weights to esti- ”military service exemption rate”, which is for Syrian mate the average exchange rate effectively in use men who want to pay the required fee to be exempted in Syria. Since 2019, the official exchange rate has from the mandatory military service. The CBS also become less reflective of the exchange rate effectively issues a customs and airline transactions rate in May 2021. These rates are either close to the official in use in Syria, for two reasons. First, the gap between exchange rates, or they are between the official and the official and parallel market exchange rates has wid- market exchange rates. See The Syria Report, “Syrian ened significantly. By May 2022, the market exchange Pound Exchange Rates – Central Bank of Syria and rate was about 55 percent higher than the official Black Market”, (March 31, 2022). Recent Economic Developments 17 FIGURE 18 • World Bank Estimated Average Exchange Rate in Syria A. Exchange rates B. Weight structure of the average exchange rate (SYP/$US) (Share in percent) 4,500 100% 4,000 90% 3,500 80% 70% 3,000 60% 2,500 50% 1,500 40% 1,000 30% 500 20% 0 10% 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 0% 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 Offical exchange rate Market exchange rate Average exchange rate Offical exchange rate Market exchange rate Source: Syrian Pound Today; Central Bank of Syria; UN Comtrade database; Central Bureau of Statistics; World Bank staff estimates. exchange rate (Figure 18.A). Second, the market against the US dollar would increase inflation by exchange rate has been used more in transactions: 30–84 percent at a 12-month horizon (Box 3). This is as described above, owing to severe foreign currency a relatively high level of pass-through from exchange shortages, the authorities have drastically shortened rate movement to inflation. For comparison, Jašová, the list of critical goods that can be imported, applying Moessner, and Takáts (2019), estimates a yearly pass- the preferential (or official) exchange rate (Figure 18.B). through coefficient of 0.222 to 0.231 for the emerging As the Syrian economy is largely consumption-based, market economies and –0.0127 to 0.00592 for the we estimate consumption-based weights to gauge advanced economies, accordingly.32 Two factors the exchange rate effectively in use in Syria. More may explain why Syria’s inflation is sensitive to the specifically, we use the official CPI weights to estimate depreciation of the Syrian pound. First, Syria’s heavily the consumption shares of goods and services of dependence on imports for essential goods implies different categories. The mirror statistics from the UN that currency falls would quickly feed into higher Comtrade database are applied to estimate imports domestic prices. Second, Syria financed its fiscal goods and services in specific categories. Then we deficit primarily through Central Bank borrowing.33 calculate the average exchange rate, taking into con- This undermines the credibility of the Central Bank sideration the preference exchange rates applied by and unhinges inflation expectations. Furthermore, the the Central Bank for imports of critical commodities time-varying estimates suggest that the pass-through and services through time. As of May 2022, Syria’s effect of the exchange rate on inflation has increased average exchange rate was estimated at 3,390 SYP/ US$ (See Technical Appendix D for details). 32 Jasova, Moessner, and Takáts (2019) estimate the Currency depreciation has triggered high ERPT in the post-2008 crisis using data for a panel inflation in Syria. In Syria, the annual CPI averaged of developed and emerging market economies and 38 percent from 2011 to 2020, and it is evident that Generalized Method of Moments (GMM) estimation of a inflation is highly correlated with the exchange rate hybrid New Keynesian Phillips curve. 33 The Syrian authorities have ceased to release data (Figure 19.A). Using various empirical approaches, we on money supply since the start of the conflict. This estimate that the exchange rate pass-through (ERPT) prevents a quantitative analysis of money supply and its coefficients range from 0.3 to 0.84 in Syria. That is, contribution to high inflation and currency depreciation a depreciation of 100 percent in the Syrian pound in Syria. 18 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS FIGURE 19 • Inflation in Syria A. Inflation and exchange rate in Syria B. Contribution to inflation, by major items (yoy percent) (yoy percent, percentage points contribution to annual growth) 400 4,000 180 350 3,500 160 300 3,000 250 2,500 140 200 2,000 120 150 1,500 100 1,000 100 50 500 80 0 0 –50 500 60 –100 –1,000 40 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 20 0 Average exchange Market exchange rate levels (RHS) rate –20 2014 2015 2016 2017 2018 2019 2020 Average exchange rate CPI Food CPI WFP minimum food Food and beverages Clothing Housing basket price Fuel Services CPI Source: Central Bureau of Statistics, Syria; WFP Market Price Watch Bulletin; World Bank staff estimates. markedly following the onset of the Lebanon financial food basket price for Syria in US dollar terms jumped crisis. This increase can likely be attributed to higher by 28 percent month-over-month (mom), or by 69 per- inflation expectations in Lebanon, which fed into cent yoy. Syria’s food prices grew faster than the global inflation expectations in Syria amid a sharp increase food prices after the war in Ukraine: the global Food in currency in circulation, probably in both countries. Price Index, a measure of the monthly change in inter- CPI rose primarily due to price increases of national prices of a basket of food commodities in US essential goods and services, disproportionately dollar terms, rose by 13 percent mom, or by 34 percent affecting the poor and vulnerable. In Syria, essen- yoy, in March 2022. Syria’s fuel prices rose roughly tial goods and services, including food, clothing, in line with the global trends after the war in Ukraine. housing, and fuel, account for about three-fourths The market price of transport diesel in US dollar terms of the consumption basket. In particular, food alone increased by 27 percent mom, or by 141 percent yoy in accounts for about 40 percent of the consumption March 2022, respectively. For comparison, the World basket. In fact, food contributed more than half of Bank energy price index in US dollars terms from the total headline inflation in Syria over the past few years Pink Sheet rose by 24 mom, or 102 yoy in March 2022. (Figure 19.B). While the official inflation statistics for 2021 are not available, food prices, as proxied by the World Food Programme (WFP) minimum food basket Weakening fiscal position price index,34 rose by 97 percent yoy during 2021 on Fiscal revenues dropped with the economic top of a 236 percent yoy increase in 2020. The record- contraction. Estimates using the World Bank’s high increase in WFP minimum food basket prices in average exchange rate suggest Syria’s fiscal revenues 2021 for Syria was driven by higher global food prices, as well as record low agricultural production due to the severe water crisis and drought-like conditions, as 34 The standard food basket is a group of essential food well as a shortage of agricultural inputs.35 commodities. In Syria, the food basket is set at a group of dry goods providing 2,060 kcal a day for a family of five The war in Ukraine further shocked com- for a month. The basket includes 37 kg bread, 19 kg rice, modity markets, pushing food and fuel prices in 19 kg lentils, 5 kg of sugar, and 7 liters of vegetable oil. Syria even higher. In March 2022, adjusted using the 35 North Press Agency, “Syria’s Wheat Production Is Lowest World Bank’s average exchange rate, WFP minimum In 50 Years,” (December 26, 2021). Recent Economic Developments 19 in US dollar terms fell by 85 percent in 2021 compared in 2022, down from 48 percent in 2010. Non-tax to the pre-conflict level in 2010 (Figure 21: Syria’s fiscal revenues accounted for the majority of fiscal revenues, budget.A). Losses in oil and tax revenues, the collapse which consist primarily of net profits from public entities of international trade due to sanctions, a growing and charged fees for government services. informal economy, and weak administrative collection In response to the revenue shortfall, the capacity all contributed to the revenue shortfall. Tax authorities cut spending, especially capital revenues have decreased more than overall revenues, spending. Between 2010 and 2021, fiscal expen- representing only 34 percent of total budgeted revenues ditures declined by 83 percent in US dollar terms. BOX 3: THE EXCHANGE RATE PASS-THROUGH TO INFLATION IN SYRIA The exchange rate pass-through (EPRT) measures the extent to which fluctuations in the exchange rate leads to changes in aggregate prices (i.e., inflation). The coefficient is, therefore, akin to an elasticity coefficient in that it measures the sensitivity of the CPI to the exchange rate. The simplest approach to gauging the ERPT is to estimate the change in the CPI, DCPIt, that is due to a change in the exchange rate, DEt. This estimates the contemporaneous response of the changes in price level to the changes in the exchange rate, DCPIt /DEt, or to the lagged changes in the exchange rate, DCPIt /DEt–1. Using the exchange rate and inflation data from May 2011 to December 2020, the ERPT coefficients are estimated to range from 0.655 to 0.796 when the Average Exchange Rate (AER) is employed, and from 0.307 to 0.735 when the market exchange rate is employed. This suggests a 100 percent depreciation in exchange rate leads to an increase in the inflation rate ranging from 30.7 to 79.6 percentage points. However, these estimates are subject to considerable uncertainty as evidenced by the high standard deviation of the estimates (Table 3: Estimates of the exchange rate pass-through using the simple approach). TABLE 3 • Estimates of the Exchange Rate Pass-Through Using the Simple Approach Average exchange rate Market exchange rate Average Standard Deviation Average Standard Deviation DCPIt /DEt 65.53 302.63 73.53 273.57 DCPIt /DEt–1 79.56 437.14 30.69 357.54 Estimates of the ERPT coefficient can also be obtained from more elaborate econometric models. The existing literature commonly employs well-specified Vector Autoregressive (VAR) models to gauge the response of prices to an exchange rate shock (see Technical Appendix E for details). The advantage of the latter approach is to allow for discerning the effects of exchange rate fluctuations on inflation over several horizons (one, six, or 12 months). In this analysis, the cumulative effect of the exchange rate shock on inflation over a horizon of 12 months can be interpreted as the pass-through. Using the AER and the CPI data from May 2011 to December 2020, the ERPT coefficient is 0.47 when the VAR is estimated in changes in levels (of AER and CPI) and 0.35 when the VAR is estimated in log-levels (of AER and CPI). If the market exchange rate is employed, the corresponding ERPT coefficients are 0.42 and 0.30 percent, respectively (Table 4). TABLE 4 • Cumulative Effect of an Exchange Rate Depreciation Based on VAR Models Average exchange rate Market exchange rate  VAR in differences in levels 0.47 0.42  VAR in log-levels 0.35 0.30 In line with the literature (see, for example, Gopinath, Itskhoki, and Rigobon, 2010), standard exchange rate pass-through regressions are also employed to gauge the ERPT coefficient. The ERPT coefficient is 0.84 when the market exchange rate is employed and 0.80 when the AER is employed. The time-varying estimates suggest that the contemporaneous effect of the exchange rate on inflation appears to have increased significantly following the onset of the Lebanon financial crisis in October 2019 (Figure 20.A and Figure 20.B). This increase is likely to be attributable to higher inflation expectations in Lebanon, owing to a sharp increase in currency in circulation, which fed into inflation expectations in Syria. (continued on next page) 20 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS BOX 3: THE EXCHANGE RATE PASS-THROUGH TO INFLATION IN SYRIA (continued) FIGURE 20 • Rolling Estimates of the Contemporaneous Effect of the Exchange Rate on Inflation A. Rolling estimates with market exchange rate B. Rolling estimates with the AER 1.2 1.6 1.2 0.8 0.8 0.4 0.4 0.0 0.0 –0.4 –0.4 –0.8 2015 2016 20 17 20 18 20 19 2020 2015 2016 20 17 20 18 20 19 2020 Coefficient Lower bounds Upper bounds Coefficient Lower bounds Upper bounds Source: Central Bureau of Statistics, Syria; World Bank staff estimates. Note: The graphs provide (rolling window) estimates of the contemporaneous coefficient measuring the effect of exchange rate changes on inflation in standard pass-through regression à la Gopinath, Itskhoki, and Rigobon (2010) along with the 95 percent confidence intervals. Given the mounting immediate needs of its popula- The authorities have managed to reduce wage bills tion, a vast majority of fiscal expenditures is devoted from 20 percent of total planned expenditures in to current spending. Only 15 percent of fiscal expen- 2010 to 12 percent in 2022. Fiscal subsidies were ditures in 2022 were planned on capital expendi- also cut, temporarily, from a peak of 63 percent of tures, compared to 44 percent in 2010 (Figure 21.B). expenditures in 2015 to 23 percent in 2020. FIGURE 21 • Syria’s Fiscal Budget A. Fiscal budget B. The composition of the budgeted spending (Billion $US; percent of nominal GDP) (share) 30 30 100% 90% 25 25 80% 70% 20 20 60% 15 15 50% 40% 10 10 30% 20% 5 5 10% 0 0 0% 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 Revenues Expenditures Deficit (RHS) Current spending Investment spending Source: Central Bureau of Statistics of Syria; Central Bank of Syria; World Bank staff estimates. Note: The fiscal data pertains to the Central Government in Damascus and excludes all taxes, transfers, and expenses incurred by the autonomous region in northeastern Syria. Recent Economic Developments 21 FIGURE 22 • Share of Subsidies in Budget total expenditures (Figure 22). Subsidies for food Expenditures in Syria and fuel products accounted for a vast majority of (percent) budgeted subsidies. In addition, electricity subsidies, 70 which were not included in the budgeted subsidies allocation, were also significant (Table 5). Subsidies 60 are generally used to provide essential goods at 50 reduced prices to citizens. Hence, the surge in fiscal 40 subsidies mainly reflects the impact of the growing needs of households and a noticeable increase in the 30 costs of essential goods. 20 The government is deeply indebted. From 2011 to 2019, the fiscal deficit averaged about 12 10 percent of GDP (Figure 21.A). The deficit would be 0 even higher if off-budget military expenditures and 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 electricity subsidies were included. Due to a lack Current spending Investment spending of access to domestic and international financing, the vast majority of the deficit has been financed Source: World Bank estimates using various Syria Ministry of Finance (MOF) reports; by Central Bank borrowing, which has worsened World Bank staff estimates. Note: The fiscal data pertains to the Central Government in Damascus and excludes all inflation. taxes, transfers, and expenses incurred by the autonomous region in northeastern Syria. To rein in the subsidy bill, the authorities Estimates include off-budget subsidies for electricity. have, since 2021, increased the administrative price of essential products while tightening the Since 2021, fiscal subsidies have risen rationing system. The authorities increased the dramatically, representing more than half of subsidized price of essential food and fuel products TABLE 5 • Subsidies by Items (Billion SYP) 2017 2018 2019 2020 2021 2022 Subsidies for oil derivatives 177 275 430 11 2,700 2,700 Subsidies for food Subsidies for flour and yeast 398 357 361 337 700 2,400 Provisions for the agricultural support fund 10 10 10 10 50 50 Subsidies for sugar and rice / / / / / 300 Subsidies for electricity (off-budget) 417 701 720 711 1800 3,600 Others Provisions for the social aid fund 15 15 10 15 50 50 Subsidies for the modern irrigation and the drought fund / / / / / 29 Total budgeted subsidies 423 657 811 373 3,500 5,529 Total subsidies (including subsidies for electricity) 840 1,357 1,531 1,084 6,300 9,129 Memorandum items: Total budgeted expenditures 2,660 3,187 3,882 4,000 8,500 13,325 Total expenditures (including subsidies for electricity) 3,077 3,888 4,602 4,711 10,300 16,925 Source: World Bank estimates using various Syria Ministry of Finance (MOF) reports. 22 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS FIGURE 23 • Syria’s Salary Levels A. The ratio of unskilled wage labour to the B. Monthly minimum wage for public sector WFP minimum food basket price ($US) (index, November 2014=100) 350 1.4 300 1.2 250 1.0 200 150 0.8 100 0.6 50 0.4 0 2020 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 0.2 Offical exchange rate Market exchange rate 0.0 Average exchange rate 2014 2015 2016 2017 2018 2019 2020 2021 Source: Central Bureau of Statistics of Syria; Central Bank of Syria; WFP Syria Market Price Watch Bulletin; World Bank staff estimates. to compensate for the depreciation of the currency, on fuel allocations for public institutions and heating supply shortages, and the growing burden of sub- oil for families, cutting heating oil and gasoline alloca- sidies. In addition, the authorities announced the tions by half for every family.38 Media reported that the exclusion of approximately 600,000 families out of authorities also considered raising public health care four million from its subsidy program.36 Among those fees significantly and imposing public school registra- removed from the eligibility pool were doctors and tion fees.39 lawyers; several categories of merchants; owners of commercial, industrial, and tourist establishments; people owning multiple properties; and owners of Declining living standards and rising cars manufactured after 2011. It was reported by the food insecurity Minister of Telecommunication that about 380,000 families submitted requests claiming the government Inflation and currency depreciation have mistakenly excluded them from receiving subsidies, significantly eroded purchasing power, resulting and about 70,000 families were reincluded in the in declining real wages. Despite repeated subsidies system.37 increases, workers’ salaries have not kept up Higher costs of essential goods triggered with inflation. The ratio between non-skilled labor by the war in Ukraine have forced fiscal policies wages40 and the WFP minimum food basket price to become more restrictive in Syria. Immediately after the start of the war, the Syrian government 36 Daher, J., “Expelled from the Support System: Austerity announced a restriction on public spending to cover Deepens in Syria,” (Middle East Directions Programme, only priorities over the next few months. Some con- blog, February 15, 2022). struction projects have reportedly been suspended. 37 See Syria Ministry of Communications; 70,154 objections Meanwhile, the authorities have rationed critical com- were accepted. modities to ensure they can be sustained for a longer 38 ETANA Syria, “Syria Brief – Economic Crisis – 8 March 2022.” period. More specifically, the Syrian government 39 The Syria Report, “Government to Reduce Health and tightened the supply of essential food commodi- Education Subsidies, Media Says,” (March 15, 2022). ties, including wheat, sugar, rice, vegetable oil, and 40 According to the WFP, non-skilled labor wages represent potatoes. In addition, new restrictions were imposed wages in construction and agriculture. Recent Economic Developments 23 FIGURE 24 • Food Insecurity in Syria A. Share of households with inadequate food consumption B. Number of people in acute food (percent) insecurity, top 10 countries 60 (million person, top 10 countries) 50 Congo Afghanistan 40 Nigeria Ethiopia 30 Yemen Syria 20 South Sudan Sudan 10 Haiti 0 Niger Aug-18 Oct-18 Dec-18 Feb-19 Apr-19 Jun-19 Aug-19 Oct-19 Dec-19 Feb-20 Apr-20 Jun-20 Aug-20 Oct-20 Dec-20 Feb-21 Apr-21 Jun-21 Aug-21 Oct-21 Dec-21 Feb-22 0 5 10 15 20 25 30 Source: Syria mVAM Bulletin; WFP Hunger Hotspots: February to May 2022 Outlook; World Bank staff estimates. Note: Estimates of food insecurity across countries are based on information collected in January 2022. has more than halved since 2019, meaning food is food insecure may have risen by another 10 percent becoming increasingly costly to afford (Figure 23.A). after the war in Ukraine.42 At the end of 2021, the minimum wages for public Food insecurity has prompted households sector workers amounted to only US$ 24 per to adopt negative coping strategies, resulting in month (converted using the World Bank’s average a decline in household resilience. The continued exchange rate), equivalent to less than one-tenth of erosion of purchasing power and unstable livelihood the pre-conflict levels (Figure 23.B). sources pushed many Syrian households to incur Food insecurity has been rising due to the more debt. In February 2022, 72 percent of surveyed ongoing conflict, soaring food prices, diminished households in Syria reported having bought food on subsidies, low crop production, and the economic credit due to lack of food and/or money, according consequences of the war in Ukraine. Food inse- to the WFP Vulnerability Analysis and Mapping curity was already severe prior to the war in Ukraine. (VAM). In addition, according to the same VAM WFP data show that more than half of households survey, relying on child labor as a coping strategy surveyed (52 percent) reported inadequate food for the lack of food has become more prominent consumption in February 2022, double the early 2019 among Syrian households over time. Approximately share (Figure 24.A). Food insecurity was particularly 14 percent of surveyed households across Syria prevalent among vulnerable groups. In February reported taking children who are of mandatory 2022, 58 percent of IDPs and 54 percent of returnees education age out of school to have them engage reported inadequate food consumption in October, in income-generating activities and contribute to the compared to 47 percent of residents. In early 2022, household’s income. the UN estimated that nearly 12 million people were severely food insecure, with an additional 1.8 million at risk of falling into food insecurity.41 As such, Syria 41 The acute food insecure figure includes 1.9 million ranked among the ten most food-insecure countries people living in camps and deemed to be 100 percent globally (Figure 24.B). Syria’s food insecurity has food insecure. worsened further in the wake of the war in Ukraine. 42 Chadwick, V., “WFP Says Syrians at ‘Breaking Point,’” WFP estimated that the number of people who are (May 12, 2022). 24 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS 3 OUTLOOK AND RISKS Outlook expected to increase, due to higher global commodity prices of essential food and fuel products, especially Economic conditions in Syria are projected to after the war in Ukraine. Consequently, the trade def- continue to be mired by prolonged armed conflict, icit will remain extremely high, at around 20 percent of turmoil in Lebanon and Turkey, COVID-19, and GDP in 2022. The persistent trade deficit will be partly the war in Ukraine. Subject to extraordinarily high offset by net current transfer inflows. However, growth uncertainty, we project real GDP will drop by 2.6 in remittances to Syria is forecast to moderate in the percent in 2022 after contracting by 2.1 percent in short run, mainly due to a weaker growth outlook for 2021.43 Private consumption will remain subdued, Syria’s major refugee-hosting countries. In particular, with a continued erosion of purchasing power amid the deepening economic crisis in Lebanon and rising prices and currency depreciation. Government Turkey may lead to a decline in remittance flows into spending, especially capital expenditures, will Syria. Overall, the current account deficit is projected continue to be constrained by low revenues and to increase from around 4 percent of GDP in 2021 to the lack of access to financing. Private investment 5 percent of GDP in 2022. is projected to remain weak as the security situation The fiscal deficit is expected to widen is assumed to remain volatile and economic and further in 2022. For Syria, it is estimated that policy uncertainties persist (Table 6: Macro outlook indicators). 43 We first apply a VAR model to predict GDP in 2022 Syria’s current account will remain firmly in without the shock of the war in Ukraine, using the pre- deficit. Although Syria’s bilateral relations with its Arab conflict estimate of the elasticity between NTLs and GDP described in Box 1. We then quantify the relationship neighbors are improving, export earnings will likely between NTLs and WFP minimum food prices and remain low owing to sanctions-related restrictions and analyzes the impact of the additional food price the recently announced export ban on agricultural growth triggered by the war in Ukraine on NTLs and, products to meet domestic demand. Import bills are consequently, GDP. 25 TABLE 6 • Macro Outlook Indicators (Annual percent changes unless indicated otherwise) 2016 2017 2018 2019 2020 2021e 2022f Real GDP growth, at constant prices –5.6 –0.7 1.5 3.7 1.3 –2.1 –2.6 Inflation (Consumer Price Index) 47.7 18.0 1.0 13.4 114.2 89.6 50.9 Fiscal balance (% of GDP) –10.2 –8.9 –8.3 –7.9 –6.5 –6.8 –7.7 Current account balance (% of GDP) –5.2 –3.0 –1.0 –4.0 –3.0 –4.0 –5.0 Source: World Bank estimates. Notes: e = estimate, f = forecast. approximately 45 percent of budgeted expenditures increase in the WFP minimum food basket price in 2022. are related to food and energy. As such, higher Assuming the historical relationship between the WFP commodity prices, triggered by the war in Ukraine, food basket price and food inflation remain stable—as will raise fiscal spending significantly. Efforts by the well as stability between food inflation and overall infla- authorities to tighten fiscal policy are projected to tion—we project CPI in Syria will reach 51 percent yoy in partially, but not fully, offset the cost-driven increase in 2022, down from 90 percent yoy in 2021.47 expenditures. Meanwhile, fiscal revenues are forecast A further rise in food prices will stress the to decline in 2022 due to reduced earnings from already struggling poor. The evolving rise in food state entities and potential tax cuts in a weakening prices and the higher risk of food insecurity are likely economy. Put together, we forecast the fiscal deficit to hurt poor families the most, because they spend a will widen from 6.8 percent in 2021 to 7.7 percent of larger percentage share of their expenditure on food GDP in 2022.44 and energy. In addition, rising food prices will cause Persistent twin deficits would further drain fiscal balances to deteriorate. Pressure on fiscal bal- foreign exchange reserves, leading to a further ances likely will force governments to reduce food depreciation of the local currency. The deprecia- subsidies further, increasing the vulnerability of the tion trend of the Syrian pound may continue in 2022, poor and adding to already elevated levels of food driven by dwindling US dollar liquidity, a spillover insecurity. effect from the financial crisis in Lebanon and Turkey, and the government’s excessive monetization of its 44 The projections are based on the official 2022 budget deficit. In addition, further depletion of the foreign figures. exchange reserves may trigger a downward spiral of 45 The drought situation has been exacerbated by a devaluation and price rises. reduction in water from the Euphrates River arriving in Inflation is projected to remain elevated in Syria from Turkey. the short term, due to the pass-through effects of 46 Prior to 2011, Syria had been self-sufficient in food currency depreciation, persistent food and fuel production. In recent years, more than half of the food supply comes from imports to Syria, according to shortages, and reduced food and fuel rationing. the WFP data. As domestic food production of Syria Poor harvests are likely to persist in 2022, due to collapsed in 2021, food imports accounted for about reduced sowing of cereal crops in late 2021, limited two-thirds of food supply that year. financial resources for purchasing agricultural inputs, 47 The annual CPI is estimated to increase by 90 percent and decreased availability of irrigation water.45 As Syria yoy in 2021 and 51 percent yoy in 2022, respectively, increasingly relies on imports to meet its food and oil if: (1) the relationship between the WFP minimum food basket price and food inflation for 2020, as well as the consumption,46 the surge in global food and oil prices relationship between food inflation and overall inflation triggered by the war in Ukraine will cause domestic over the same year, remains, and (2) WFP minimum consumer prices to rise. Using trends from the first food basket price growth of 144 percent yoy in 2021 and four months of 2022, we forecast an 83 percent yoy projected growth of 83 percent yoy in 2022. 26 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS Risks As a result of the war in Ukraine, donors may divert some humanitarian assistance away Risks to the growth outlook are significant and from Syria and the Syrian war-affected refugees. tilted to the downside. Two major sources of Despite growing needs, reported humanitarian donor uncertainty are the COVID-19 pandemic and the war funding for Syria in 2021 reached its lowest level in Ukraine. since 2014, according to the data collected by the UN Renewed outbreaks of COVID-19 remain a Financial Tracking Service (FTS). Humanitarian assis- major risk, particularly for Syria, given its limited tance will be needed to cushion the blow from surging COVID-19 surveillance. Syria has experienced two commodity prices triggered by the war in Ukraine, par- major COVID-19 waves, in September/ October 2021 ticularly for Syria, which is facing acute food insecurity. and February 2022, when the Delta and Omicron vari- Yet, with the global economy under significant strain ants hit the country. The number of cases started to and donor governments facing increasing domestic decrease after March 2022. However, the high cumu- costs, global aid may decline further. In addition, lative case fatality rate (5.8 percent) indicates the some aid may shift away from Syria and the Syrian inability of the health system to cope with providing war-affected refugees, as donors prioritize funding care to COVID-19 patients. COVID-19-associated for Ukrainian refugees. After several cuts in 2021, the deaths are relatively high in Syria, partially due to a WFP announced that it would further scale down a slow vaccine rollout. As of May 14, 2022, only 9.1 number of items in its monthly emergency food bas- percent of Syrians were fully vaccinated, and another kets to Northwest Syria in May 2022, a move attributed 5.2 percent were partially vaccinated. In the event to funding constraints and rising global food prices.50 of a rapid spread of more transmissible and deadly Some EUR 6.4 billion was pledged toward funding the COVID-19 variants in Syria, slow vaccination rollouts international humanitarian aid effort in Syria during and inadequate health facilities would exacerbate its the 2022 Brussels Conference, held on May 10, 2022. impact (Box 4). Despite registering an increase over the previous Owing to its heavy reliance on imports, year, pledges made by donor countries continued to Syria is particularly vulnerable to disruptions fall short of the EUR 9.31 billion funding requirements in the commodity market and the associated made by UN agencies and other NGOs.51 trade-policy interventions triggered by the Russia- The ongoing war in Ukraine has geopolitical Ukraine war. Syria is heavily dependent on wheat implications that may threaten the delivery of imports from Russia. In 2021, the Syrian government humanitarian assistance to Syria. The existing UN imported 1.5 million tonnes of wheat, mostly from cross-border mechanism that permits the provision Russia. Media reported that the Syrian government of aid into Idleb (a non-government-controlled area) had reached an agreement with Russia at the end is up for a renewal vote in July 2022. If negotiations of 2021 to import another 1 million tons of wheat among stakeholders fail, there is a risk that the last from Russia.48 With the war in Ukraine, it is uncertain UN humanitarian corridor into Syria, which is vital whether the agreement will be enforced. Syria may to delivering life-saving aid to 2.4 million people in face additional obstacles to food imports as a result Northwest Syria, could be forced to close. In general, of the trade-policy interventions triggered by the war since the war in Ukraine, a constructive diplomatic in Ukraine. Since the beginning of the war, more than 30 new export restrictions have been announced worldwide. India, the second-largest wheat producer, 48 North Press Agency, “Syria to import one million tons of banned exports amid food supply concerns in May wheat from Russia,” (December 6, 2021). 2022. Export restrictions alone were estimated to have 49 Ruta, M., “The Impact of the War in Ukraine on Global Trade and Investment,” World Bank, April 25, 2022). added 7 percentage points to the price of wheat and 50 Middle East Monitor, “WFP Reduces Monthly Aid to NW risk igniting a tit-for-tat escalation that could trigger a Syria, Amid Worsening Food Crisis,” (April 17, 2022). food crisis, potentially negatively affecting Syria as a 51 The Syrian Report, “Brussels Conference Pledges net food importer.49 Unexpectedly Rise to EUR 4.1 Billion,” (May 11, 2022). Outlook and Risks 27 BOX 4: COVID-19 CONTINUES TO THREATEN SYRIA’S HEALTH SYSTEM Prior to the conflict, Syria’s health outcomes were comparable to other countries in the region. In 2011, Syria’s infant mortality rate of 16.7 per 1,000 live births was lower than the MENA regional average (23.7 per 1000 live births).a The health system consisted of a mix of government- run hospitals and primary care facilities, with advanced medical care concentrated in major cities such as Damascus and Aleppo.b The number of hospital beds per 1,000 persons in 2011 was 1.6, compared to 1.3 for Iraq, 1.7 for Iran, 3.5 for Lebanon, and 1.8 for Jordan.c The health financing data from 2012d showed gaps in health financing, with decreased levels of public spending on health as compared to previous years (domestic general government health expenditure amounted to 1.6 percent of GDP). At the same time, households faced a high burden of health expenditure with out-of-pocket spending constituting 53.7 percent of current health expenditure. The conflict devastated the health system, which only partially recovered as hostilities slowed. More than 50 percent of the health infrastructure is estimated to have been damaged or destroyed by the conflict.e In addition to destruction of hospital buildings and equipment, water, electricity, medicines, and consumables have become scarce.f Recent reports show that even among undamaged or rehabilitated facilities, many are not operating or providing only limited services; 44 percent of primary health centers and 34 percent of hospitals are either not functioning or partially functioning.g The conflict has had widely heterogenous impacts across the country, and in particular across areas controlled by different actors. In December 2021, 12 percent of public health centers were not functioning in southern Syria, compared to 52 percent in northwest Syria and 55 percent in northeast Syria.h Closures of border crossings in 2020 has interfered with humanitarian delivery of medical supplies, especially in northeast and northwest Syria.i These challenges health care were exacerbated by a mass exodus of health workers, who fled routine attacks against providers.j As a result, health outcomes deteriorated sharply, although the limited available data suggests that some key outcomes had begun to make a recovery by the eve of the COVID-19 pandemic. The COVID-19 pandemic further exacerbated the strain on the health sector. In September/ October 2021 and February 2022, new waves of COVID-19 brought daily infections and deaths to their highest level on record. As shown in Figure 25.A, the number of cases started to decrease since March 2022. However, the high cumulative case fatality rate (5.8 percent)k points toward the inability of the health system to cope with the needs for providing care for COVID-19 patients. Urban health facilities have struggled to serve all COVID-19 patients and sometimes suspended non-emergency surgeries.l Some services, such as childhood vaccination coverage, have not changed significantly between 2020 and 2021, although it is not clear whether this reflects the absence of an effect of COVID-19 on some routine services or is the result of other confounding factors.m The extremely slow vaccine rollout puts Syria at a high risk of future waves. As of May 9, 2022, only 8.4 percent of Syrians were fully vaccinated and another 5.6 percent partially vaccinated.n A recent report by the COVID-19 Vaccine Delivery Partnership (CoVDP)o highlighted Syria as one of the 34 countries with less than 10 percent coverage and persistently low vaccination rates. While availability of vaccines has improved significantly at the global level, Syria will need a sustained and predictable supply to allow effective planning of vaccination strategies and campaigns. Falling demand for COVID-19 vaccinations also remains a major concern. As shown in Figure 25.B, after a brief uptick in vaccine uptake at the end of 2021, currently the number of daily doses delivered is less than 0.1 percent of the population. Based on progress FIGURE 25 • COVID-19 in Syria A. Daily new confirmed COVID-19 cases B. Daily share of the population receiving a (7-day moving average, as of May 25, 2022) COVID-19 vaccine dose 400 (Percent, 7-day moving average, all doses, as of May 25, 2022) 350 0.60% 300 0.50% 250 0.40% 200 0.30% 150 0.20% 100 50 0.10% 0 0.00% 03/2020 04/2020 05/2020 06/2020 07/2020 08/2020 09/2020 10/2020 11/2020 12/2020 01/2021 02/2021 03/2021 04/2021 05/2021 06/2021 07/2021 08/2021 09/2021 10/2021 11/2021 12/2021 01/2022 02/2022 03/2022 04/2022 03/2021 04/2021 05/2021 06/2021 07/2021 08/2021 09/2021 10/2021 11/2021 12/2021 01/2022 02/2022 03/2022 04/2022 05/2022 Source: Our World in Data; World Bank World Bank staff estimates. (continued on next page) 28 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS BOX 4: COVID-19 CONTINUES TO THREATEN SYRIA’S HEALTH SYSTEM (continued) so far, Syria is projected to reach 70 percent vaccination coverage only by February 2026.p The low vaccination coverage represents a serious threat to economic recovery, especially with the risk of new variants. With low levels of immunity, new waves of cases could lead to spikes in severe disease and death, which would quickly overwhelm the fragile health system. Syria requires a focused strategy to restore health and health services along with urgent efforts to scale up COVID-19 vaccination coverage. Additionally, accelerating the delivery of other COVID-19 tools and treatments is a crucial priority to help build up multiple layers of protection against the COVID-19 virus. a World Bank, “World Development Indicators, Infant Mortality Rate for the Syrian Arab Republic,” (May 5, 2022). b Sharara, S. L. and Kan, S. S., “War and Infectious Diseases: Challenges of the Syrian Civil War.” (2014). c World Bank Open Data, Physicians (per 1,000 people), https://data.worldbank.org/indicator/SH.MED.PHYS.ZS. d World Bank, World Development Indicators, (2022). Accessed February 22, 2022. World Development Indicators | DataBank (worldbank.org) e Devi S., “Health in Syria: A Decade of Conflict,” (The Lancet, March 13, 2021). f Abbara et al., “The Effect of the Conflict on Syria’s Health System and Human Resources For Health,” (World Health & Population, 2015). g World Health Organization, “WHO Summary of Key Indicators – Whole of Syria Dashboard,” Accessed February 10, 2022. h World Health Organization, Syria, “Summary of Key Performance Indicators,” (2022). Accessed February 16, 2022. i United Nations Office for the Coordination of Humanitarian Affairs (OCHA), “Humanitarian Needs Overview: Syrian Arab Republic.” (2021). j Fouad, et al., “Health Workers and the Weaponisation of Health Care in Syria: A Preliminary Inquiry for the Lancet–American University of Beirut Commission on Syria,” (The Lancet,2017). k World Health Organization, “Syria, COVID-19 Dashboard, 2022.” Accessed May 5, 2022. l OCHA (2021). m WHO (Whole of Syria Dashboard), 2022. n World Bank, “Coronavirus Pandemic (COVID-19),” (Our World In Data, database). o World Health Organization and World Bank, “Accelerating COVID-19 Vaccine Deployment (2022).” p World Bank, “COVID-19 Vaccine Deployment Tracker.” environment necessary for advancing reforms for trade, investment, and humanitarian operations in Syria has become more challenging. Syria. Nevertheless, given Syria’s worsening eco- Economic stagnation and deterioration nomic conditions, trade and investment are unlikely to of public services may lead to increased social pick up dramatically in the short term, as the private unrest. The removal of subsidy programs amid high sector may continue to pursue de-risking strategies. food prices already triggered protests in government- controlled areas such as Tartus and Latakia, as well as in Northwest Syria in early 2022.52 Existing evidence 52 The Syria Report, “Government removes millions from suggests that rising food inflation in an environment subsidy programme, triggering rare protests,” (February of fragile political stability, and inadequate resources 8, 2022). See also al-Aswad, H. (2022), “Syria: protests to maintain subsidies, may lead to a significant mount amid anger over ‘unfair’ cuts to government subsidies,” (Middle East Eye, February 12, 2022). increase in social unrest (Arezki and Bruckner, 2014; 53 On November 24, 2021, the US Treasury Department Bellemare, 2015). amended a general license for non-governmental There are upside risks to the outlook. Syria’s organizations to allow them to engage in additional improved trade relations with its Arab neighbors transactions and activities in support of non-profit could reduce its economic isolation. In addition, non- activities in Syria, including new investment, the purchase governmental organizations were allowed to carry out of refined petroleum products of Syrian origin for use in Syria, and certain transactions with parts of the Syrian additional transactions and activities, and restrictions government. In May 2022, the Treasury announced the on foreign investments in non-regime held areas authorization of activities in certain economic sectors in of Northeast and Northwest Syria were also eased the non-regime-held areas of Northeast and Northwest recently.53 These measures could potentially facilitate Syria. Outlook and Risks 29 SPECIAL FOCUS: DEMOGRAPHIC AND LABOR MARKET CONSEQUENCES OF THE SYRIAN CONFLICT T he current demographic profile of the Syrian costs. The demographic profile of a country affected population provides a good illustration of the by war can be a good indicator of the scale of human massive human impact of the conflict. After losses brought about by conflict. Epidemiologists dis- about a decade of war, the demographic profile of tinguish between direct and indirect effects of conflict. the Syrian population has dramatically changed, with Direct effects include higher mortality associated with the male deficit in prime-age adult population being war-related casualties and international displacement its most prominent feature. The decline in the male leading to substantial losses in population. Indirect working age population, together with the progressive effects can be even larger. Massive destruction deterioration of economic conditions in the country, brought about by conflict over time impacts services, has pushed more Syrian women to enter the labor infrastructure, and productive systems that are critical market to help support their families. However, Syrian for people’s survival, leading to increases in malnutri- women continue to face severe challenges in terms tion, morbidity and, ultimately, to higher mortality. of unemployment and lack economic opportunities Destruction of health facilities, collapse of health compared to their male counterparts. These systems, damage to agriculture production and food challenges have been further exacerbated as a result systems, as well as destruction of houses and water of conflict. and sanitation infrastructure are among the most Conflicts can have profound impacts on harmful effects of conflicts, with implications often the demographic structure of populations, with enduring well beyond the termination of hostilities. potential long-lasting implications. The develop- Assessing direct and indirect demographic effects of mental consequences of conflict are immense, both conflict is complicated by the lack of data that charac- in terms of human suffering and social and economic terizes conflict-affected countries. In fact, information 31 FIGURE 26 • Syria Population Pyramid, 2010 and 2021 (Million people, by age group) 2010 2021 80+ 80+ 75–79 75–79 70–74 70–74 65–69 65–69 60–64 60–64 55–59 55–59 50–54 50–54 45–49 45–49 40–44 40–44 35–39 35–39 30–34 30–34 25–29 25–29 20–24 20–24 15–19 15–19 10–14 10–14 5–9 5–9 0–4 0–4 –8 –6 –4 –2 0 2 4 6 8 –8 –6 –4 –2 0 2 4 6 8 10 Female Male Source: World Bank calculations based on UN Department of Economic and Social Affairs (DESA); World Population Prospects 2010; and HNAP household survey data (Summer 2021). from civil registration systems is often time lacking or FIGURE 27 • Masculinity Ratio in 2021, by Age unreliable in times of conflict, and it is typically com- Group (Number of males per 100 females, pounded by the limited availability of representative by age group) household surveys. Despite these challenges, the analysis of changes in the demographic structure of 140 a population can provide important insights into the 130 welfare and long-term development challenges of a 120 110 country in conflict. 100 The current demographic profile of the 90 Syrian population embodies the massive human 80 impact of the conflict. The demographic structure 70 of the Syrian population has been severely affected 60 by conflict. As shown in Figure 26, Syria’s population 50 structure in 2021—as estimated using the Humanitarian 40 Needs Assessment Programme (HNAP) household 0–5 5–10 10–15 15–20 20–25 25–30 30–35 35–40 40–45 45–50 50–55 55–60 60–65 65+ survey data collected in May-June 202154—is signifi- cantly different from the one observed in 2010, before Source: World Bank calculations based on HNAP household survey data (Summer the onset of conflict. In 2010, the gender composition 2021). of Syria’s population was relatively balanced across age groups, and children in the 0–4 age category were the largest demographic group, in line with the 54 From mid-May to mid-June 2021, HNAP conducted a profile of a growing population. These features have nationwide demographic household survey across all 14 not persisted after more than a decade of conflict. The governorates of Syria. The survey, with a total sample of demographic structure of Syria in 2021 is one typical 24,573 households, collected data on key demographic and socio-economic indicators, and it is representative of a conflict-stricken country. Similar to other countries at the country, governorate, and sub-district level. affected by conflict,55 the current shape of Syria’s popu- 55 Demographic deficits of prime-age men have been, lation pyramid clearly shows a substantial male deficit for example, documented in Rwanda, Cambodia, and concentrated in people ages 20 to 40 (Figure 27). Darfur (Guha-Sapir and D’Aoust, 2011). 32 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS FIGURE 28 • Trends in under-Five Mortality Rate FIGURE 29 • Number of Absent Members for (Number of deaths per 1,000 live Households Currently Residing in births) Syria, by Reason of Absence (Number of individuals, by reason 50 for absence) 45 Absent household members, (Number of deaths per 1,000 live births) 40 by reasons of absence 35 Jailed / missing 30 25 Movement outside syria 20 15 Movement within syria 10 War-relate death 5 0 0 100,000 200,000 300,000 400,000 500,000 600,000 700,000 800,000 900,000 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 Source: UNICEF, https://data.unicef.org/country/syr/. Female Male Source: World Bank calculations based on HNAP household survey data (Summer 2021). Moreover, compared to its pre-conflict level, estimates of conflict: approximately 850,000 individuals—93 indicate a decline in the share of children in the 0–4 percent of whom are men—have fled Syria and left age category, consistent with evidence indicating an their households behind, while another 420,000 indi- increase mortality rate related to conflict for children viduals—90 percent of whom are men —have died as a under age 5 (Figure 28). A compounding driver for direct consequence of the war.58 These estimates are this change might be a decline in fertility, spurred by based on reports of household members currently the progressive deterioration of living conditions and living in Syria and should therefore not be interpreted high levels of maternal mortality.56 Overall, despite as indicative of the total displacement or death toll these changes, the demographic dependency rate of the Syrian conflict. Entire households have been has remained stable at its pre-conflict level. displaced as a result of conflict; some of these inter- International displacement and war-related nationally displaced households might have suffered deaths have been the two main driving forces behind observed gender imbalances in Syria’s prime-age adult population. Understanding the 56 In low-income countries, conflict has been correlated causes of observed age-sex deficits can provide with increases in maternal mortality due to the decline in access to reproductive health services and female important insights on household welfare and useful education, as well as social insecurity (Urdal and information for policy action.57 Data for Syria indicate Che, 2013). that both war-related deaths and displacement might 57 Male deficits due to conflict-related deaths may result have substantially contributed to the gender imbal- in an increase of vulnerable groups such as women- or ances currently observed in the population. Overall, child-headed households. On the other hand, welfare close to one third of householders currently living in impacts related to gender imbalances brought about by international displacement or migration could be Syria report at least one “absent household member” mitigated through remittances flows. in relation to the conflict. As shown in Figure 29, 58 More than 50 percent of war-related deaths are “absent” household members in 2021 were predomi- concentrated in five governorates: Deir-ez-Zor, Rural nantly males who either left Syria or died because Damascus, Alleppo, Idleb and Homs. Special Focus: Demographic and Labor Market Consequences of the Syrian Conflict 33 FIGURE 30 • Changes in Working Age Population, FIGURE 31 • Relation between Female Labor 2010–2021 Force Participation and Masculinity (Million people) Ratio at the Governorate Level, 2021 (Percent) Conflict and Working age population 7 FLFP & Masculinity ratio WAP 60 6 Tartous 5 50 Lattakia Female participation rate (%) 4 Homs Hama 40 3 As-Sweida 30 Dar'a 2 Aleppo Ar-Raqqa 20 Al-Hasakeh 1 Idleb Rural Damascus Quneitra 0 Deir-ez-Zor 10 Damascus 2010 2021 2010 2021 Male Female 0 70 45 80 85 90 95 100 105 110 Source: World Bank calculations based on Labor Force Survey 2010 and HNAP Male per 100 Female in WAP (%) household survey data (Summer 2021). casualties, and entire households might have lost their increase in the level of economic activity has been lives due to conflict.59,60 Still, these estimates provide particularly dramatic for Syrian women, with female important insights into the scale of the welfare chal- labor force participation doubling from 13 percent in lenges that Syrian households face to sustain their livelihoods in a dramatically deteriorating economic environment. In addition, the estimates shed light on 59 The Syrian conflict has originated one of the largest the critical role that interventions aimed at sustaining episodes of international displacement since World War international remittances (through economic integra- II. According to latest UNHCR data, 5.7 million Syrian tion of Syrian refugees in host countries) could play refugees are currently hosted in neighboring countries and an additional 1 million reside in Europe, mostly for the livelihoods of households in Syria. In 2021, in Germany and Sweden. As of April 30, 3.76 million according to OCHA data, 8.8 percent of households Syrian refugees are living in Turkey, 839,000 in Lebanon, in Syria were female-headed, up from 4.4 percent in 674,000 in Jordan, 258,000 in Iraq, and 141,000 in 2009; at the same time, close to one in four Syrian Egypt. See https://data2.unhcr.org/en/situations/syria. households receive international remittances and, 60 Estimates of the total death toll of the Syrian war vary critically, relied on them for livelihood. depending on the methodology and reporting agency. In 2021, the UN’s human rights office (OHCHR) released The demographic impact of conflict a tally of 350,200 deaths, including both civilians and coupled with deteriorating economic conditions combatants. The count is based on a strict methodology in the country have important implications for the requiring the deceased’s full name, date of death, and Syrian labor market. Compared to its pre-conflict location of the body, and should therefore be interpreted levels, Syria’s working age population has significantly as an under-estimation of the actual number of war- shrunk, particularly in its male component (Figure 30). related deaths. The Syrian Observatory for Human Rights (SOHR) estimates an overall death toll of 610,000 people However, the impact of this demographic shock has over 11 years of conflict, of which 160,681 were civilians been compensated by an increase in labor force (120,158 men, 15,237 women, and 25,286 children). participation, leaving the overall number of employed 61 Between 2010 and 2021, labor force participation in Syrians almost unchanged at around 5.2 million.61 The Syria increased from 43 percent to 50 percent. 34 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS 2010 to 26 percent in 2021, against a more modest FIGURE 32 • Trends in Labor Force Participation increase in male participation from 72 to 76 percent and Unemployment Rates, by Gender over the same period. With deteriorating economic conditions in the country, more Syrian women have Labor market dynamics been pushed to enter the labor market to support their 90% 80% 76% households in order to make ends meet and, possibly, 72% to compensate for the absence of male adult house- 70% 60% hold members. In fact, as shown in Figure 31, the 50% increase in female labor force participation has been 40% 37% particularly strong in governorates more affected by 30% 26% the conflict-induced male demographic deficits, pos- 22% 20% 13% sibly suggesting the substituting role that women are 10% 6% 4% playing in the Syrian labor market.62 0% Poor labor market opportunities add to 2010 2021 2010 2021 the challenges that Syrian women are facing as a result of the conflict. In a region characterized Participation rate Unemployment rate by low levels of female labor force participation, the Male Female observed increase in economic activity among Syrian Source: World Bank calculations based on Labor Force Survey 2010 and HNAP women cannot be welcomed as an unequivocally household survey data (Summer 2021). positive development. In fact, while an increasing number of women are pushed to seek employment to help support their families, they face more severe these constraints. Lack of access to government- labor market challenges compared to their male issued civil documentation, particularly in areas more counterparts. The female unemployment rate in 2021 affected by conflict, represents a major challenge for stood as high as 37 percent, 33 percentage points women, particularly for widowed and divorced ones, higher than male unemployment and 15 percentage as it limits their ability to inherit property, their assets’ points higher than its pre-conflict level (Figure 32). tenure security, and possibly exposes them to the risk Moreover, the massive destruction of physical capital of violence. Moreover, safety and security concerns and productive capacity of Syria’s agriculture and have heightened as a result of conflict, further limiting industry sectors have had a disproportionate impact the physical mobility and employment opportunities on female employment opportunities. Of the 55,000 of Syrian women and adding additional risks for those jobs lost in agriculture, two in three were held by entering the labor market by force of necessity.63 women. In the manufacturing sector, the decline in Demographic and labor market changes female employment was 82 percent, against a 34 per- brought about by the conflict can have structural cent drop in male employment. As a result, women’s employment options are currently mostly restricted to 62 A similar positive relation between conflict and female labor force participation, oftentimes referred to as the the service sector. In 2021, the service sector repre- “additional worker effect,” has been found in other conflict sented 87 percent of female employment compared episodes such those in Peru, Nepal, and Vietnam. See to 69 percent in 2010. Gallegos (2012); Menon and van der Meulen Rodgers Despite the observed increase in labor (2015); and Kreibaum and Klasen (2015). force participation, other factors associated 63 According to most recent reports, the prevalence of with the conflict have limited women’s oppor- gender-based violence (GBV) remains a prominent concern, with one in five households indicating that tunities for socio-economic inclusion. Similar women and girls feel unsafe in their respective locations, to other countries in the region, gender norms and mainly when crossing checkpoints, at markets, and on legal barriers limit Syrian women’s engagement in public transportation. See OCHA, “2022 Humanitarian the public sphere. Conflict has further exacerbated Needs Overview: Syrian Arab Republic,” (February 2022). Special Focus: Demographic and Labor Market Consequences of the Syrian Conflict 35 implications for Syria’s future growth prospects. the cessation of hostilities and progressive reintegra- More than a decade of conflict has caused massive tion of displaced populations, is the improvement of losses in Syria’s human and physical capital, with socio-economic inclusion of Syrian women. destruction of vital assets and infrastructure. Losses in human capital stem from both changes in the size and demographic structure of Syria’s resident population 64 Decline in current and future productivity can be as well as from the decline in its productivity.64 Long- expected as a result of physical and psychological term implications of these losses can be far reaching. trauma, increased malnutrition, and lower investments in A critical factor that might positively influence future children’s education, due to both supply and demand prospects, besides the expected peace-dividend from constraints. 36 TECHNICAL APPENDIX Counterfactual GDP Calculations for and industry as a percentage of GDP), investment Syria share (gross capital formation as a percentage of GDP), physical capital (capital stock per capita), Using the approach pioneered by Abadie and human capital (human capital index), and political Gardeazabal (2003), popularly known as the status (democracy index). For a full description of synthetic control method (SCM), this note estimates the variables used in this analysis, please refer to a counterfactual GDP for Syria. The approach avoids Appendix 1. the arbitrariness of the selection of the control group The selection of the explanatory variables by identifying a combination of comparator countries reflects a common specification for cross-country that best approximate the characteristics of the studied growth simulations (Abadie and Gardeazabal, 2003; country. This combination of comparator countries is Abadie et al., 2015; Adhikari et al. 2016; Campos et likely to produce a better control/comparison group al., 2019). These variables capture, in a broad sense, for the country exposed to the shock than any simple the impact of institutions, demography, and macro- comparison alone. economic conditions in addition to traditional growth accounting variables, such as the stock of physical Specification and data and human capital. The data relied on a cross-country panel data set for the period between 1995 and The outcome of interest is real GDP. Annual real GDP 2019. The data of employment and investment came in constant 2015 US dollars for Syria was obtained from Penn World Table 10.0 database. The data from the World Bank’s World Development Indicators of democracy proxy came from the Polity Project (WDI). The vector of predictors that was selected initiated by Center of Systemic Peace. And all other to explain GDP included employment (employers variables supporting the calculation of the synthetic per hundred persons), trade openness (trade as counterfactuals came from the World Bank’s World percentage of GDP), industry shares (agriculture Development Indicators Database. 37 Before conducting the SCM to construct Syria’s tit = YitI – YitN(1) macroeconomic counterfactual scenario, we needed to ensure that none of the comparators that were Since we cannot observe the outcome vari- selected violated the exogeneity criterion. Accordingly, able of the treated country had the treatment never we complied a donor pool of all countries in the world occurred (when tiT ,…, tiT), to estimate the treated 0+1 excluding those that were involved in or significantly country’s hypothetical outcome tiT (we rely on the impacted by the conflict or other significant shocks general model proposed by Abadie et al. (2010): during the period of analysis. For example, the study dropped those Middle East and North Africa (MENA) YitI = dt + tit + uit(2) countries affected by the Arab Spring, and Japan, which was affected by the Fukushima disaster of tit = ait Dit(3) 2011. For a complete list of all the countries dropped from the dataset, please refer to Appendix 2. YitN = dt + uit(4) The algorithm of SCM requires having countries with complete data for all years in the outcome vari- uit = qt Zi + lt wt + eit(5) able and, in case of the covariates, at least one year of data for the pre-intervention period. Hence, countries without both specifications have been dropped from Where Dit is a binary indicator adopting the value of the sample. 1 when i = 1 and t > T0 and zero otherwise; ait is the effect of the event on the variable of interest; Zi is a Overview of the synthetic control method vector of country-level covariates; qt is a vector of time-specific parameters; lt a vector of unobserved To investigate what would have been the real GDP common factors; wi the country-specific observable level in Syria if the Syrian civil war had not occurred, term; and eit the unobserved transitory shocks with we use the synthetic control method developed by zero mean. Abadie and Gardeazabal (2003) and extended in The counterfactual will be given by an estimate Abadie et al. (2010, 2015). of Yit (when i > 1) as close as possible to Y1t for every This method searches for a weighted combi- t < T0, based on the previously defined country-level nation of other countries (donors) that resemble as covariates (Zi). A series of weights ( W = w2, ..., wN+1) will closely as possible the characteristics of the target be assigned to construct the counterfactuals, where country in the pretreatment period in terms of an Si=2 N+1 wi = 1 and wi ≥ 0. Therefore: outcome variable (real GDP, in this case), and a N+1 set of other specific covariates. This is achieved by minimizing for the pretreatment period the root mean S wY = Y (6) i=2 i it it squared prediction error (RMSPE). and To describe it more formally, let’s use the nota- N+1 tion described by Abadie et al. (2010). Let’s assume that Yit is the outcome variable of interest (GDP level) S w Z = Z (7) i=2 i it i of a country i at a time t. i = 1 represents the country lead to the unbiased estimator of tit: affected by the event, and i = 2, ..., N = 1 those who N+1 were not impacted. t = 1, ..., T0, ..., T, being T0 the year the event of interest occurred. τ ^ = Y – it it S wY i=2 i it (8) Let’s assume also that YitI is the outcome vari- able of interest of the country affected by the event and YitN the outcome variable of interest of the country 65 In order to implement a synthetic control model, we need had the event never occurred.65 The average treat- to assume that the event does not have any impacts on ment effect estimator can be represented as: the outcome variable of the treated country before t = T0 38 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS The optimal set of weights is chosen to mini- TABLE 7 • GDP Level Predictor Means – All mize the root mean squared prediction error, which Pretreatment Outcome Values Used as Predictors is given by: T0 N+1 GDP level (in billions, 2015 US$) Syria Synthetic Syria RMSPE = 1 T0 S S t=1 (Y – it i=2 w Y )2 (9) i it 1995 17.22 17.78 1996 18.91 18.71 Results 1997 19.86 19.92 1998 21.20 20.64 The study conducted the SCM estimation using all 1999 20.45 20.50 pre-treatment outcome values as predictors. The 2000 32.34 31.98 synthetic Syria is made of nine countries, with the weights in the parentheses: Togo (0.441), Gabon 2001 20.80 21.13 (0.194), Uruguay (0.175), Trinidad and Tobago (0.118), 2002 21.63 21.86 Nigeria (0.039), Kyrgyzstan (0.015), Vietnam (0.011), 2003 23.19 23.16 Argentina (0.005), and Greece (0.003). 2004 24.79 24.70 The resemblance between Syria and the syn- 2005 26.33 26.15 thetic Syria is presented in Table 7. The high level of resemblance reached by the counterfactuals under 2006 27.66 27.61 this specification is due to the use of all pretreatment 2007 29.22 29.30 outcome values as predictors. However, because the 2008 30.53 30.83 covariates are receiving weights close to zero, most 2009 32.34 31.98 of the other predictor means do not reach a high level 2010 34.02 34.27 of similitude when compared to the treated countries. Figure 1 depicts the synthetically constructed Employment rate (%) 26.14 38.96 Syrian real GDP estimates. The in-sample estimates Agriculture (as % of GDP) 23.73 19.58 provided a good match with the actual series before Industry (as % of GDP) 29.48 29.81 the conflict. This is reflected in the small RMSPEs of Trade (% of GDP) 16.64 46.06 0.27, which are relatively small compared to the level Investment (as % of GDP) 12.31 16.73 of outcome variables. The estimates showed that Syrian real GDP was forecast to continue growing Capital stock per capita (in logarithm, 9.77 9.81 2017 US$) without the conflict, albeit at a slower rate than during the pre-conflict era. Specifically, the real GDP is esti- Human capital (index) 2.20 2.11 mated to be US$ 43.6 billion at constant 2015 prices Democracy proxy (index) –7.31 1.14 in 2019, compared with US$ 34.3 billion at constant Note: All variables are averaged for the 1995–2010 period. 2015 prices in 2010. The counterfactuals presented for the post- treatment period could be considered optimistic for captured by the synthetic controls, as none of the Syria which is a country located in the Middle East country in the control group (i.e., the 9 countries that and North Africa (MENA) region. Indeed, as can be make up the synthetic Syria) is from the MENA region. seen in Table 8, results for panel regression shows To account for the effect of the regional shocks, we that countries in the MENA region lose around 1.4 incorporated the rate of adjustment suggested by the percentage points of growth per year after 2010 when panel regression into the counterfactuals in the post- compared to the global average. It could be argued treatment period. that the adverse conditions these MENA countries We then compared the actual and counterfac- were facing post-2010 would prevail regardless of tual GDPs. The comparison indicates a persistent the Syrian conflict. These adverse conditions are not negative impact of the conflict on Syria, and the gap Technical Appendix 39 TABLE 8 • GDP Growth: Panel Data Analysis (1) (2) (3) Dependent Variable Real GDP annual growth Real GDP annual growth Real GDP annual growth MENA*Post-Conflict –0.737* –1.376** (0.369) (0.448) MENA*Pre-Conflict 1.64*** (0.282) Constant 3.98*** 3.78*** 2.330* (0.087) (0.089) (1.086) Year Fixed Effects No No Yes Country Fixed Effects No No Yes R-squared 0.0012 0.0012 0.2329 Observations 3,982 3,982 3,982 Source: Authors’ estimates based on the GDP real growth figures of 180 countries globally collected in the WDI. Note: *p<0.1; **p<0.05; ***p<0.01; “post-conflict” covers the period from 2011 to 2019; “pre-conflict” covers the period from 1995 to 2010. FIGURE 33 • Actual and Counterfactual GDP FIGURE 34 • Actual and Counterfactual GDP, (Billions, constant 2015 US$) Adjusted for Regional Shocks (Billions, constant 2015 US$) 50 45 45 40 40 35 35 30 30 25 25 20 20 15 15 10 10 5 5 0 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 0 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Actual Counterfactural Actual Counterfactural Counterfactural, adjusted for regional shocks Source: WDI; Penn World Table 10.0; Center for Systemic Peace; World Bank staff Source: WDI; Penn World Table 10.0; Center for Systemic Peace; World Bank staff estimates. estimates. 40 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS APPENDIX 1 • Variables Indicator Name Definition Data source Real GDP GDP is the sum of gross value added by all resident producers in the economy plus any product taxes and WDI (in billions, constant 2015 US$) minus any subsidies not included in the value of the products. The real GDP is in constant 2015 US dollars. Employment It is calculated by dividing employers by the total population, which includes all residents regardless of legal WDI (employers per hundred status or citizenship. Employers are who, working on their own account or with one or a few partners, hold persons) the type of job defined as a self-employment job, and have engaged, on a continuous basis, one or more persons to work for them as employees. Trade openness It is proxied by total trade as share of GDP. Trade is the sum of exports and imports of goods and services Penn World (% of GDP) measured as a share of GDP. Table 10.0 Agriculture, value added Share of value added of agriculture in GDP. Agriculture includes forestry, hunting, and fishing, as well as WDI (% of GDP) cultivation of crops and livestock production. Value added is the net output of a sector after adding up all outputs and subtracting intermediate inputs. Industry, value added Share of value added of industry in GDP. Industry includes mining, manufacturing, construction, electricity, WDI (% of GDP) water, and gas. Value added is the net output of a sector after adding up all outputs and subtracting intermediate inputs. Gross capital formation Share of gross capital formation in GDP. Gross capital formation includes land improvements; plant, Penn World (% of GDP) machinery, and equipment purchases; and the construction of roads, railways, and the like, including schools, Table 10.0 offices, hospitals, private residential dwellings, and commercial and industrial buildings. Capital stock per capita (in It is calculated by dividing capital stock by total population. Capital stock is the net accumulation of a physical Penn World logarithm, constant 2017 US$) stock of capital goods, including buildings and the plant and machinery of a firm, industry or economy at any Table 10.0 one point in time. The capital stock is at constant 2017 US dollars. Human capital It calculates the contributions of health and education to worker productivity. The final index score ranges WDI (index) from zero to 1, and measures the productivity of a future worker of a child born today relative to the benchmark of full health and complete education. Political status The index is computed in Polity Project initiated by Center of Systemic Peace. The proxy ranges from +10 Center for (democracy proxy). (strongly democratic) to –10 (strongly autocratic). https://www.systemicpeace.org/inscrdata.html Systemic Peace APPENDIX 2 • Countries Affected by an Exogenous Shock Dropped from the Pool of Donor Countries Country Shock Bahrain Arab Spring after 2011 Chad War between 2005 and 2010 Cote d’Ivoire Second Ivorian Civil War in 2011 Egypt Arab Spring in 2011 Guinea Hit by Ebola in 2014 Honduras Coup in 2009 Iran Sanctions in 2012 Israel Affected by refugee movements Japan Fukushima disaster in 2011 Libya Arab Spring after 2011 Morocco Arab Spring 2011 Thailand Floods in 2011 Tunisia Arab Spring after 2011 Turkey Affected by refugee movements West bank and Gaza Israeli conflict Yemen Arab Spring after 2011 Technical Appendix 41 between the actual and counterfactual GDP are wid- (Elvidge, 1997). NTLs data have been used to measure ening. Specifically, we estimate that in 2019, Syria’s the impact of economic policy shifts, such as India’s GDP absent the conflict (the counterfactual) would 2016 demonetization, or geopolitical events, such as have been US$ 38.6 billion in 2015 constant prices, conflict outbreaks in Afghanistan (Beyer, 2018). Other compared to US $16.3 billion of the realized GDP research uses NTLs to estimate subnational estimates (Figure 2). of economic activity in countries such as India (Beyer, 2018), Egypt (Omar, 2019), Turkey (Basihos, 2016), and the United States (Doll, 2006). Nowcasting Economic Activity Using NTL data benefits from being high frequency, Nighttime Lights with high spatial granularity and broad coverage. The high-frequency nature of NTL data is perhaps Introduction the most attractive feature to economists, as it allows researchers to study the immediate impact of a shock. Nighttime lights (NTLs) are a promising big data Satellites capture data from the same location at source that can be used to estimate real GDP and weekly or even daily frequencies, offering substantial economic activity in general at a high frequency temporal coverage. For example, a number of recent and granularity. NTLs represent both a substitute papers have studied the impact of COVID-19 contain- and a complement to standard data sources, which ment measures on NTLs in India (Beyer, 2021), China lack the granularity across space and time to track (Elvidge 2020), and Morocco (Roberts, 2021). These economic developments in a fragile, conflict, and papers benefited from the monthly frequency of NTLs violence (FCV) setting. NTL data also circumvent the to better isolate the shock, from the timeliness of NTLs typical challenges associated with data availability, data to monitor the impact with a shorter time lag, and reliability, comprehensiveness, and timeliness, which from the credibility of the data used. are compounded by conflict and crisis. Another key advantage—relevant in both data- Although the use of NTLs in economics dates poor and data-rich environments—is the high spatial as far back as 2002 (Sutton, 2002), it was not until the granularity of NTLs data over standard data sources, seminal paper by Henderson (2011, 2012) that uncov- which allows researchers to drill down into subnational ered the link between luminosity and economic activity. areas and study a much wider cross-section of areas. Since then, NTLs have become a widely accepted Such units include cities (Storygard, 2016), ethnic proxy for economic activity at both national and subna- homelands (Alesina, 2016), or subnational administra- tional levels and have been shown to correlate strongly tive units (Hodler, 2014). NTLs also benefit from being with annual movements in real GDP (Henderson 2012). collected uniformly and consistently, without regard to Under the reasonable assumption that consumption of natural disasters or political strife, thereby allowing for light increases with income, NTLs can serve as a proxy broad data coverage across space and time. Finally, for economic activity (Donaldson, 2016). In the eco- coming directly from impartial satellites, NTLs data nomics literature alone, over 150 papers have relied is insulated from intervention by national statistical on night lights since 2012, partly owing to platforms agencies or other data collection actors which can such as Google Earth Engine that dramatically ease sometimes be inefficient or biased. researcher access to the data (Gibson et al., 2021). Using nighttime lights as a proxy for economic NTLs have been shown to correlate with various activity would help overcome several limitations in economic and demographic indicators, and therefore Syrian statistics. Currently, the data available from can be used to proxy for economic activity. In an early the national statistical office comes with significant study on the link between luminosity and economic uncertainty, while third-party estimates rely on heavy activity, NTLs have been shown to have a statistically assumptions. Furthermore, typical measurement significant relationship with population, economic issues are associated with the accurate reporting development, and electric power consumption of economic activity, considering the existence of 42 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS a large informal sector. Official statistics are also saturation of images and improve temporal compa- limited in their frequency, as only annual estimates rability (Elvidge, 2013). Unlike the DMSP data that is are produced, and no monthly or quarterly estimates only available at an annual level, the VIIRS data is also of economic activity are available. In addition, those available at monthly and even daily frequencies. estimates are produced with a long lag, which has For the purposes of this study, we use the DMSP been rising during the conflict, making it difficult and VIIRS NTL data to evaluate the Syrian crisis. The to assess the state of the economy at a given time. VIIRS satellite offers a higher temporal and spatial Lastly, there is currently no information on the location resolution than the DMSP satellite, while also providing of economic activity—such as per region or adminis- a number of technical improvements with respect to trative area, preventing any spatial growth analysis, a measurement and accuracy. More importantly, the critical limitation given that the conflict has a highly VIIRS data extends from April 2012 until present, cov- differentiated impact at the spatial level. ering the Syrian crisis entirely, and is available at the monthly frequency. This allows us to study the evolu- Data Description tion of NTLs throughout the crisis at a high temporal and spatial resolution. On the other hand, the DMSP In this study, similar to key papers in the literature, satellite is only available at the annual frequency and we use NTL data from the US Air Force Defense is useful in constructing a historical NTLs time series Meteorological Satellite Program (DMSP) and the that can be used to infer the long-running relationship Visible Infrared Imaging Radiometer Suite (VIIRS). between NTLs and real GDP. Initially, the DMSP was a collection of satellites For our geographic boundaries, we use the launched by the US Department of Defense in the United Nations Global Administrative Unit Layer 1960s to relay weather and climate data for tactical (GAUL) dataset to clip our map and isolate NTLs only and strategic US military operations. These satellites in Syria and within its subnational areas. Although take high-resolution pictures of the earth’s surface the VIIRS data eliminates non-anthropogenic lights at night, providing imagery of vital for weather reflected from clouds, such as moonlight, for example, predictions. Although these satellites were initially the lights emanating from ephemeral fires are not intended to monitor weather conditions for pilots, the scientific community quickly understood the potential of using this data in social sciences. Indeed, after the Map of Syria with NTLs and Known FIGURE 35 •  data was publicly archived in 1973, the first scientific Flaring Sites paper to use this data appeared only five years later (Croft, 1978). However, the DMSP satellite comes with a number of technical limitations, including low radiometric and spatial resolution, saturation of images in urban cores, and no on-board calibration. While the DMSP was discontinued in 2012, the VIIRS satellites were launched in its place by the National Aeronautics and Space Administration (NASA) and National Oceanic and Atmospheric Administration (NOAA). While the DMSP was designed with Air Force pilots in mind, the design of VIIRS reflected the needs of researchers. The VIIRS satellite provides a number of improvements over the DMSP-OLS—including a wider detection range, Source: Visible Infrared Imaging Radiometer Suite (VIIRS); Global Gas Flaring Reduction finer quantization, lower detection limits, and in-flight Partnership (GGFRP). calibration. Overall, these features help correct for Note: The dark circles identify known flaring sites from the GGFRP. Technical Appendix 43 removed. Given the high amounts of gas flaring in possible explanation for this sharp increase in flaring Syria, we use data from the World Bank Global Gas NTLs could be related to the wider conflict in general. Flaring Reduction Partnership (GGFRP) to carve out The time series for total, flaring, and non-flaring (total NTLs that result from oil production (Figure 35). This minus flaring) is shown in the Figure 36 below. step is important since the intensity of flaring can bias Besides aggregating at the national or country the results considerably, especially in the case of a level, we can also aggregate non-oil NTLs by sub- small developing nation that already has low levels of national regions (Figure 37). This allows us to drill brightness and high levels of flaring. down into the spatial dynamics of NTLs and under- To convert NTLs from pixel-radiance values on a stand the sources of economic growth or loss. The map to a single value, we use the well-known “Sum-of- figures below show the evolution of non-oil NTLs by Lights” method. This method sums up all the radiance administrative-1 areas and their respective six-month values for every pixel under the Syrian map, creating moving average. In general, most areas exhibit a a single estimate for aggregate brightness in the U-shaped pattern beginning in 2014 and bottoming country. This can be done at the national, subnational, out in 2017. Some areas recover by 2019 and con- or district level, giving researchers the freedom to fully tinue to grow, while others decline going into 2020 customize the analysis. To carve out the effects of and beyond. Finally, the pace of recovery is uneven flaring, we estimate the Sum-of-Lights for all areas that across regions, as some areas remain relatively flat have been tagged as flaring sites by the World Bank’s throughout the entire period and do not show major GGFRP dataset. Next, we subtract these flaring NTLs swings or changes. from the total estimate, giving us an estimate of non-oil- While the figures above are useful in under- related NTLs. This allows us to remove these extremely standing the evolution of NTLs within each area, they bright lights that otherwise bias our NTLs estimates are not useful in comparing between areas. This is and focus only on conventional sources of electrical because the geographic size or population density lights. Since the existing NTLs literature mainly can vary widely among the different regions, producing focuses on electrical sources of lights, this remains the NTLs estimates that are not directly comparable. One main focus of our study. This is particularly important way to adjust the NTLs estimates so they can be in the case of Syria, as flaring nearly quadrupled in comparable is to create an NTLs-per-capita estimate early 2017, while remaining relatively flat otherwise. A by dividing the total NTLs by population size. Overall, the remaining areas show a large decline in NTLs per capita since 2018, with large variation between the FIGURE 36 • Evolution of Flaring, Non-Oil, and areas (Figure 38). In 2021, some areas recover slightly Total NTLs while others worsen, suggesting that the recovery has not been the same among all regions. 2.5 Regression Analysis 2.0 By exploiting the overlapping year between VIIRS and 1.5 DMSP, we can construct a historical time series of NTLs 1.0 from 1992 through 2021 at the annual level (Figure 5). Using the DMSP dataset, we construct national “Sum- 0.5 of-Lights” estimates annually for 1992–2013. When combining the two annual datasets, we apply annual 0.0 DMSP growth rates for 2012 (the overlapping year) to 2012 2013 2014 2015 2016 20 17 20 18 20 19 20 20 20 21 the VIIRS time series in order to extend the dataset Flaring NTLs Total NTLs Non-Oil NTLs from 2012 to 2011. We repeat this procedure for the following years to continue extrapolating the VIIRS Source: VIIRS; GGFRP; World Bank staff estimates. time series backward using DMSP growth rates until 44 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS 0.0 0.2 0.4 0.6 0.8 1.0 1.2 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 0.0 0.2 0.4 0.6 0.8 1.0 2012 2012 2012 2012 1.2 2012 2012 2012 2012 2013 2013 2013 2013 2013 2013 2013 2013 2014 2014 2014 2014 2014 2014 2014 2014 2015 2015 2015 2015 2015 2015 2015 2015 2016 2016 2016 2016 2016 2016 2016 2016 2017 2017 2017 2017 Ar-Raqqa Damascus NTLs Ar-Raqqa NTLs Deir-ez-Zor NTLs Damascus Deir-ez-Zor Al-Hasakeh NTLs Al-Hasakeh 2017 2017 2017 2017 2018 2018 2018 2018 2018 2018 2018 2018 2019 2019 2019 2019 MA MA MA MA 2019 2019 2019 2019 2020 2020 2020 2020 2020 2020 2020 2020 2021 2021 2021 2021 2021 2021 2021 2021 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.0 0.5 1.0 1.5 2.0 2.5 3.0 2012 2012 2012 2012 2012 2012 2012 2012 FIGURE 37 • Evolution of Non-Oil NTLs by Subnational Regions 2013 2013 2013 2013 2013 2013 2013 2013 2014 2014 2014 2014 2014 2014 2014 2014 2015 2015 2015 2015 2015 2015 2015 2015 2016 2016 2016 2016 2016 2016 2016 2016 Dar'a Dar'a NTLs 2017 2017 2017 2017 Hama Hama NTLs Aleppo Aleppo NTLs As-Sweida NTLs As-Sweida 2017 2017 2017 2017 2018 2018 2018 2018 2018 2018 2018 2018 MA MA MA 2019 2019 2019 2019 MA 2019 2019 2019 2019 2020 2020 2020 2020 2020 2020 2020 2020 2021 2021 2021 2021 2021 2021 2021 2021 Technical Appendix (continued on next page) 45 FIGURE 37 • Evolution of Non-Oil NTLs by Subnational Regions (continued) Homs Idleb 3.5 0.7 3.0 0.6 2.5 0.5 2.0 0.4 1.5 0.3 1.0 0.2 0.5 0.1 0.0 0.0 2012 2012 2013 2013 2014 2014 2015 2015 2016 2016 2017 2017 2018 2018 2019 2019 2020 2020 2021 2021 2012 2012 2013 2013 2014 2014 2015 2015 2016 2016 2017 2017 2018 2018 2019 2019 2020 2020 2021 2021 Homs NTLs MA Idleb NTLs MA Lattakia Quneitra 0.8 0.45 0.7 0.40 0.6 0.35 0.5 0.30 0.25 0.4 0.20 0.3 0.15 0.2 0.10 0.1 0.05 0.0 0.00 2012 2012 2013 2013 2014 2014 2015 2015 2016 2016 2017 2017 2018 2018 2019 2019 2020 2020 2021 2021 2012 2012 2013 2013 2014 2014 2015 2015 2016 2016 2017 2017 2018 2018 2019 2019 2020 2020 2021 2021 Lattakia NTLs MA Quneitra NTLs MA Rural Damascus Tartous 2.5 0.45 0.40 2.0 0.35 0.30 1.5 0.25 1.0 0.20 0.15 0.5 0.10 0.05 0.0 0.00 2012 2012 2013 2013 2014 2014 2015 2015 2016 2016 2017 2017 2018 2018 2019 2019 2020 2020 2021 2021 2012 2012 2013 2013 2014 2014 2015 2015 2016 2016 2017 2017 2018 2018 2019 2019 2020 2020 2021 2021 Rural Damascus NTLs MA Tartous NTLs MA Source: VIIRS; World Bank staff estimates. 1992. We extrapolate the VIIRS time series backward and Weil (2012). Annual Real GDP data for Syria to 1992 instead of forecasting the DMSP series from were obtained the World Bank’s World Development 2013 onward because the VIIRS satellite is considered Indicators (WDI) and are in constant 2015 US dollars. more accurate than the DMSP and does not suffer Figure 40 below plots Syrian GDP against national from issues of top coding.66 NTLs from 1992 until 2020. NTLs and GDP climb To establish whether NTLs have a long-running relationship with economic activity, we regress them 66 Top-coding occurs when pixels in bright areas, such as in against real GDP and real GDP growth rates, using city centers, reach the highest possible digital value (i.e., the procedure outlined by Henderson, Storeygard 63 for DMSP), and no further details can be recognized. 46 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS FIGURE 38 • Per Capita NTLs by Subnational TABLE 9 • Regression Results of Historical NTLs Region and Real GDP 12 (1) (2) (3) 10 Dependent Variable log(GDP) GDP GDP 08 log(NTLs) 0.726*** 06 (0.009) 04 NTLs 0.072*** 02 (0.005) 0 NTLs * Pre-Conflict 0.071** Al-Hasakeh Aleppo Ar-Ra qq a As-Sweida Damascus Dar'a Deir-ez-Zor Hama Homs Id leb Lattakia Rural Damascus Tartous (0.008) NTLs * Post-Conflict 0.038** (0.015) NTL per capita 2011 NTL per capita 2018 NTL per capita 2021 Observations 28 28 28 R2 0.823 0.861 0.882 Source: VIIRS; United Nations World Population Prospects; World Bank staff estimates. Note: Quneitra was found to be a large outlier and was removed from the figure. This is Adjusted R2 0.879 0.853 0.881 likely because of its proximity to the borders of Israel, producing NTL-capita estimates far higher than other regions. Note: *p<0.1; **p<0.05; ***p<0.01 steadily from 1992 until 2010, although the growth specification, NTLs are regressed against real GDP in in GDP was relatively higher than NTLs from 2003 to logarithmic form. In the second, both are regressed 2011. Both NTLs and GDP peak in 2010 and suffer against each other in levels. While this second speci- large losses throughout 2015, and afterward NTLs fication shows the absolute relationship between the slightly recover while GDP remains subdued. two variables, the interpretation is ambiguous due to The regression results show positive and signifi- the large differences in units. However, the first specifi- cant results for all specifications (Table 9). In the first cation reveals the percent-to-percent elasticity between FIGURE 39 • Historical NTL Time Series FIGURE 40 • Historical NTLs and Real GDP (1992–2021) (1992–2021) 2.0 40 1.6 1.8 35 1.4 1.6 30 1.2 1.4 25 1.0 1.2 20 0.8 1.0 15 0.6 0.8 0.6 10 0.4 0.4 5 0.2 0.2 0 0.0 19 92 19 94 1996 1998 2000 20 02 20 04 20 06 20 08 20 10 20 12 2014 2016 2018 20 20 0.0 19 92 19 94 1996 1998 2000 20 02 20 04 20 06 20 08 20 10 20 12 2014 2016 2018 20 20 GDP (constant 2015 US$) Non-oil NTLs Total NTLs Flaring NTLs Non-Oil NTLs Source: VIIRS and DMSP; GDP data comes from the World Bank’s World Development Source: VIIRS and DMSP; World Bank staff estimates. Indicators. Technical Appendix 47 FIGURE 41 • Model Predictions of Real GDP from dynamics of growth. The analysis below shows NTLs regional NTLs as an index (April 2012 = 100) along with its associated GDP estimates expressed also as 40 an index. The relationship between the NTL and GDP 35 indexes is found in the log-log regression specifica- 30 tion (2). The coefficient translates between percent 25 NTL growth and GDP percent growth, namely 0.726, 20 and allows us to estimate the associated change in 15 GDP. This can be done at both the national level and 10 subnational level as Figure 41 and Figure 42 below 5 show. One notable observation is that the model 0 predictions are well below the official GDP estimates 19 92 19 94 1996 1998 2000 20 02 20 04 20 06 20 08 20 10 20 12 2014 2016 2018 20 20 between 2010 and 2017. This indicates that the economic contraction Syria is experiencing could be GDP (constant 2015 US$) Predictions worse than Syrian statistics suggest. To extend the analysis beyond the national Source: World Bank staff estimates. and subnational levels, we can use these regional NTL estimates to study the relationship between NTLs and GDP, which reveal a clearer relationship areas. This is most often done by studying inequality between the two. This coefficient for this regression between groups by using the well-known “Gini Index,” is 0.726, suggesting that for every 1 percent increase which is commonly used with income instead of in NTLs, real GDP increases by almost three-quarters NTLs. However, given the large literature on NTLs of a percent. Although this estimate is slightly higher in economics in addition to the regression evidence than the estimates of Henderson (2012), it remains shown above, we are comfortable using NTLs as a within the 95 percent confidence interval, giving some fair proxy for income instead. To adjust NTLs so that additional confidence to the estimate. they are comparable, we calculate an NTL-per-area In the third specification, we consider if the con- measure that divides absolute NTLs by the geo- flict substantially changes the relationship between graphical spatial extent. This is done in the absence NTLs and GDP. This is done by creating a dummy of consistent annual population data that would have variable equal to 0 pre-conflict (1992 until 2010) and allowed us to do the analysis using NTLs-per-capita equal to 1 post-conflict and continues until present instead. Figure 43 below shows the evolution of the (2011 until 2021). This allows the regression to esti- Gini Index for Syria annually from 1992 until 2021. The mate two unique elasticities between NTLs and GDP results show a sharp decline in inequality during the for both the pre- and post-conflict periods. The results beginning of the 1990s, which steadily declines until show that both elasticities are positive and significant, 2010. Soon after the conflict began in 2011, inequality while the post-conflict estimate is approximately worsened, although there has been some improve- half the magnitude of the pre-conflict estimate. This ment since 2016 when the intensity of the conflict in suggests that NTLs are contributing less to national Syria declined. income during the conflict, as NTL use is diverted to less productive areas of the economy. Moreover, the increase in R-squared (R^2) by adding the conflict Connectedness between the Syrian dummies is marginal, suggesting that information and Lebanese Pounds gain by separating the two periods is limited. These coefficients are useful because they This note analyzes connectedness (or spillovers) allow us to convert our NTL observations into GDP between the Syrian pound (SYP), Lebanese Pound estimates, revealing the regional and temporal (LBP), and Turkish Lira (TRY). 48 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS "! &!! #! $! 0 20 40 60 80 100 120 140 160 0 20 40 60 80 100 120 0 20 40 60 80 100 120 0 20 40 60 80 100 120 140 %! 2012 2012 2012 2012 2012 2012 2012 2012 2013 2013 2013 2013 2013 2013 2013 2013 2014 2014 2014 2014 2014 2014 2014 2014 2015 2015 2015 2015 2015 2015 2015 2015 2016 2016 2016 2016 Damascus NTLs Deir-ez-Zor NTLs Ar-Raqqa NTLs Al-Hasakeh NTLs 2016 2016 2016 2016 2017 2017 2017 2017 Ar-Raqqa Damascus Deir-ez-Zor Al-Hasakeh 2017 2017 2017 2017 2018 2018 2018 2018 2018 2018 2018 2018 2019 2019 2019 2019 2019 2019 2019 2019 Ar-Raqqa GDP 2020 2020 2020 2020 Damascus GDP Deir-ez-Zor GDP Al-Hasakeh GDP 2020 2020 2020 2020 2021 2021 2021 2021 2021 2021 2021 2021 0 20 40 60 80 100 120 0 20 40 60 80 100 120 140 0 20 40 60 80 100 120 140 160 180 200 0 20 40 60 80 100 120 2012 2012 2012 2012 2012 2012 2012 2012 2013 2013 2013 2013 FIGURE 42 • Model Predictions of Real GDP by Subnational Region 2013 2013 2013 2013 2014 2014 2014 2014 2014 2014 2014 2014 2015 2015 2015 2015 2015 2015 2015 2015 2016 2016 2016 2016 As-Sweida NTLs Aleppo NTLs Dar'a NTLs Hama NTLs 2016 2016 2016 2016 Dar'a 2017 2017 2017 2017 Hama Aleppo As-Sweida 2017 2017 2017 2017 2018 2018 2018 2018 2018 2018 2018 2018 2019 2019 2019 2019 Dar'a GDP Hama GDP 2019 2019 2019 2019 Aleppo GDP 2020 2020 2020 2020 As-Sweida GDP 2020 2020 2020 2020 2021 2021 2021 2021 2021 2021 2021 2021 Technical Appendix (continued on next page) 49 FIGURE 42 • Model Predictions of Real GDP by Subnational Region (continued) Homs Idleb 450 120 400 100 350 300 80 250 60 200 150 40 100 20 50 0 0 2012 2012 2013 2013 2014 2014 2015 2015 2016 2016 2017 2017 2018 2018 2019 2019 2020 2020 2021 2021 2012 2012 2013 2013 2014 2014 2015 2015 2016 2016 2017 2017 2018 2018 2019 2019 2020 2020 2021 2021 Homs NTLs Homs GDP Idleb NTLs Idleb GDP Lattakia Quneitra 140 180 120 160 140 100 120 80 100 60 80 40 60 40 20 20 0 0 2012 2012 2013 2013 2014 2014 2015 2015 2016 2016 2017 2017 2018 2018 2019 2019 2020 2020 2021 2021 2012 2012 2013 2013 2014 2014 2015 2015 2016 2016 2017 2017 2018 2018 2019 2019 2020 2020 2021 2021 Lattakia NTLs Lattakia GDP Quneitra NTLs Quneitra GDP Rural Damascus Tartous 120 120 100 100 80 80 60 60 40 40 20 20 0 0 2012 2012 2013 2013 2014 2014 2015 2015 2016 2016 2017 2017 2018 2018 2019 2019 2020 2020 2021 2021 2012 2012 2013 2013 2014 2014 2015 2015 2016 2016 2017 2017 2018 2018 2019 2019 2020 2020 2021 2021 Rural Damascus NTLs Rural Damascus GDP Tartous NTLs Tartous GDP Source: World Bank staff estimates. Granger causality More formally, let Wt denote the information set at time t and y1,t+h,W denote the optimal (i.e., lowest MSPE) t The exposition in this section draws on Dagher et h-step prediction of y1t. Let sy2 (h/Wt) denote the MSPE 1 al. (2020). In a seminal contribution to the literature, of the variable y1t. Kilian and Lütkepohl (2017) note that Granger (1969) introduces a concept of causality that the process y2t is said to Granger-cause the process y1t if: closely ties to the predictive power of one variable for another. Let y1t and y2t denote two time series.67 sy2 (h/Wt) < sy2 (h/Wt{y2s|S ≤ t}), 1 1 The variable y2t is said to Granger-cause y1t when accounting for the information in lowers the Mean 67 Kilian and Lütkepohl (2017) and Lütkepohl (2005) are Square Prediction Error (MSPE) in y1t. excellent references on Granger causality testing. 50 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS FIGURE 43 • NTL-Based Gini Index of interest. A VAR relates yt to p of its lags. A test of Granger causality amounts to a test of zero restric- 0.45 tions on a subset of the coefficients of the VAR. 0.40 Suppose that the data generating process is a 0.35 VAR(p). Granger causality testing can be undertaken 0.30 within a bivariate VAR (Lütkepohl and Kratzig, 2004): 0.25 0.20 Y a a Y u 0.15 Y =S a 1t 2t p i=1 11,i 21,i a 12,i 22,i 1t–1 Y 2t–1 + u , 1t 2t 0.10 0.05 by testing the null that a12,i = 0, which implies that y2t 0.00 does not Granger cause y1t. That is, if the coefficients 19 92 19 94 1996 1998 2000 20 02 20 04 20 06 20 08 20 10 20 12 2014 2016 2018 20 20 associated with the lags of y2t are statistically significant, then y2t Granger-causes y1t. Source: World Bank staff estimates. The analysis begins with the LBP and SYP. The empirical analysis is undertaken with daily data on the where Wt{y2s|S ≤ t}) denotes the information set LBP and SYP for the period November 1, 2019 to March excluding past and present information regarding 18, 2022. The levels of the two exchange rates are pro- the series y2t. In other words, the process y2t is said to vided in Figure 44. The data for the SYP are obtained Granger-cause the process y1t if exploiting information from the website Syrian Pound Today (https://sp-today. on the past and contemporaneous values y2t of lowers com/en/), while the data for the LBP are obtained from the prediction error of the process y1t at some horizon h. the website Lira Rate (https://lirarate.org/#pills-sayrafa). In a bivariate setting, the variable y2t does not The correlation between the levels of the LBP Granger cause y1t if and only if: and SYP is 0.83. However, due to the non-stationarity in the two exchange rates, this correlation is likely to y1t+h|W = y1t+h|W \{y . be spurious. t t 2,S|S q, r > q, 0.5 < q < 1 S a the termination of subsidies in Lebanon. (1–a)T t=1 1 t 2 t Dependence in quantiles day. The observations for the high and low prices are not The dependence between the SYP and LBP may available for LBP and SYP. Therefore, the latent volatility is estimated using conditional heteroskedasticity models. be more pronounced in the quantiles. Therefore, Alternatively, volatility can be estimated as the square of a Quantile-Quantile (QQ) plot, which illustrates the the daily return. However, the latter estimate of volatility is relationship between the quantiles of the two returns, noisy. If intra-daily data were available, realized volatility is provided in Figure 50. could be computed from the high-frequency data. Technical Appendix 55 For 0 < q < 0.5, the sample quantile dependence is where e is a vector of independently and identically 0.92 while it is 0.18 for < q < 1. This suggests that distributed disturbances. The moving average the dependence between the two returns in the lower representation is: quantiles is very high while it is weak in the upper quantiles. Therefore, the downward movements in the xt = SË i=0 A e t t –1 , SYP and LBP are closely associated with each other. Where the N×N coefficient matrices Ai obey the Measuring network connectedness and recursion Ai = φ1Ai −1 + φ2Ai −2 + ! + φp Ai −p , with A0 spillovers being an N×N identity matrix and with Ai = 0 for i < 0. The variance decompositions allow for Measuring connectedness and volatility assessing the fraction of the H-step-ahead error spillovers variance in forecasting x i that is due to shocks to In general terms, connectedness is measurable as the x j , ∀i ≠ j . The variance decomposition is also the relative importance of the exogenous variations of unit basis for computing the variance shares, total, direc- i in explaining variations in unit j. As succinctly noted tional, and net spillovers discussed next. in Ferroni and Canova (2022), this corresponds, in the The own variance share is defined as the frac- context of a VAR, to the fraction of the forecast error tion of the H-step-ahead error variances in forecasting variance of unit j, explained by shocks originating x i that are due to shocks to x i for i = 1 ,2, ... ,N . The from unit i at a given horizon. cross variance shares, or spillovers, are the fraction In two influential contributions, Diebold and of the H-step-ahead error variances in forecasting x i Yilmaz (2009, 2012) propose measures of connect- that are due to x i for j = 1,2, ... ,N , such that i ≠ j . edness, or volatility spillovers, which can be easily Denoting the H-step-ahead error variance constructed from the variance decompositions of a VAR. decomposition by θijg (H ), for H = 1 ,2, ..., we have: More specifically, Diebold and Yilmaz (2009) σ −1ΣH −1 (e ' A Σe ) 2 offer a method to compute a total volatility spillover θ (H ) = jj H −1h =0 ' i h ' i g Σh =0 (ei Ah ΣAhei ) ij index from a VAR model identified using a Cholesky decomposition while, cognizant of the fact that the results obtained a VAR identified using a Cholesky Where Σ is the variance matrix for the error vector e, decomposition are not invariant to the ordering of sjj is the standard deviation of the error term for the jth the variables, Diebold and Yilmaz (2012) build on equation, and ei is the selection vector, with one as the their earlier work to advocate the use of a variance ith element and zero otherwise. It should be noted that decompositions obtained from the generalized VAR ΣN j =1θijg (H ) ≠ 1 . The normalized variance decomposi- framework of Koop, Pesaran, and Potter (1996) and tion is: Pesaran and Shin (1998). ! g (H ) = θij (H ) , g In Diebold and Yilmaz (2012), the authors also θ ij ΣN j =1θijg (H ) extend their earlier work by proposing, in addition to the total spillover index introduced in Diebold and Note that, by construction Yilmaz (2009), measures of directional spillovers and ! g (H ) ≠ 1 and ΣN θ !g net spillovers in volatility. This section introduces the ΣN j =1θ ij i , j =1 ij (H ) = N . measures of total and directional volatility spillovers, and applies them to the volatilities of SYP, LBP, and The total volatility spillover index can be con- the Turkish Lira (TRY).70 structed from the total volatility spillover index as: Starting from a covariance stationary N-variable VAR (p): xt = Sp i=1 f x t t –1 + et , 70 The analysis is based on the variance or volatility obtained from the GARCH (1,1) model for each currency. 56 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS N i , j =1 g ij (H ) N g i , j =1 ij (H ) Figure 51 shows that the volatility spillovers i =j i =j g S (H ) = N g 100 = 100. between SYP and LBP are elevated in June 2020 and i , j =1 ij (H ) N in May and June of 2021. The total volatility spillover The directional volatility spillover received by index reaches a maximum of 16.06 percent on June currency i from all other currencies j is given by: 9, 2021. That is, the volatility spillovers increased N j =1 g ij (H ) N g j =1 ij (H ) at the onset of Lebanon’s financial crisis as well as i =j i =j g S (H ) = N g 100 = 100. during the episodes of fuel shortages in the summer i , j =1 ij (H ) N of 2021. The total spillover has also increased, albeit In a similar vein, the directional volatility spillover more moderately, in March 2022. transmitted by currency i from all other currencies j is In order to discern the drivers of the increase given by: in volatility, the analysis proceeds with an examination N j =1 g ji (H ) N g j =1 ji (H ) of directional and net volatility spillovers. Figure 52 i =j i =j g S (H ) = N g 100 = 100. provides the directional volatility spillovers. i , j =1 ji (H ) N The directional volatility spillover analysis sug- The directional volatility spillovers provide a gests that the increase in the total spillover index in decomposition of the total spillovers to those from or June 2020 and March 2022 originates from shocks to to a particular source. The net spillover from currency the volatility of LBP, while the increase in the spillovers i to all other currencies j is the difference between the in May and June 2021 appears to be ascribable to gross volatility shocks transmitted and those received shocks in the volatility of the SYP. from other currencies while the net pairwise volatility spillover between currencies i and j is the difference Volatility spillovers: SYP, LBP and TRY between the gross volatility shocks transmitted from The volatility spillovers analysis is extended by currency i to currency j. including TRY. Figure 53 provides the volatility of the SYP, LBP and TRY. Volatility spillovers: SYP and LBP The TRY exhibited significant volatility in This section examines the volatility spillovers between November and December of 2021. Figure 53 shows the (returns on) LBP and SYP using a 200-day rolling these large swings. Indeed, the TRY depreciated sample. Figure 51 provides the total volatility spillover significantly following the Central Bank’s unorthodox index constructed in this manner. Figure 51 provides decision to decrease its policy rate on December 16 the total volatility spillovers between SYP and LBP. amid rising inflation. FIGURE 51 • Total Volatility Spillover Index: LBP and SYP Correlation 18 16 14 12 10 8 6 4 2 0 2020:05:21 2020:06:11 2020:07:02 2020:07:23 2020:08:13 2020:09:03 2020:09:24 2020:10:15 2020:11:05 2020:11:26 2020:12:17 2021:01:07 2021:01:28 2021:02:18 2021:03:11 2021:04:01 2021:04:22 2021:05:13 2021:06:03 2021:06:24 2021:07:15 2021:08:05 2021:08:26 2021:09:16 2021:10:07 2021:10:28 2021:11:18 2021:12:09 2021:12:30 2022:01:20 2022:02:10 2022:03:03 Technical Appendix 57 FIGURE 52 • Directional Volatility Spillovers (FROM) Directional Voltality Spillovers (FROM) Variance of LBP 20 15 10 5 0 M J J A S O N D J F M A M J J A S O N D J F M 2020 2021 Variance of SYP 30 20 10 0 M J J A S O N D J F M A M J J A S O N D J F M 2020 2021 FIGURE 53 • Returns on the TRY TRYRET 15 10 5 0 –5 –10 –15 –20 –25 4/11/19 4/01/20 4/03/20 4/05/20 4/07/20 4/09/20 4/11/20 4/01/21 4/03/21 4/05/21 4/07/21 4/09/21 4/11/21 4/01/22 4/03/22 This decision was followed by a deposit scheme, in the TRY, provide an impetus for the de-dollarization announced by Turkey’s President Recep Tayyip of deposits, and stem the TRY’s depreciation. The TRY Erdogan on December 20, that aimed to protect the pur- appreciated as a result of Erdogan’s announcement, chasing power of deposits in TRY, shore up confidence although the appreciation was short-lived. 58 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS FIGURE 54 • Volatility of LBP, SYP and LBP VOL_LBP 18 16 14 12 10 8 6 4 2 0 111/02/19 01/02/20 03/02/20 05/02/20 07/02/20 09/02/20 11/02/20 01/02/21 03/02/21 05/02/21 07/02/21 09/02/21 11/02/21 01/02/22 03/02/22 VOL_TRY 9 8 7 6 5 4 3 2 1 0 11/02/19 01/02/20 03/02/20 05/02/20 07/02/20 09/02/20 11/02/20 01/02/21 03/02/21 05/02/21 07/02/21 09/02/21 11/02/21 01/02/22 03/02/22 VOL_SPY 16 14 12 10 8 6 4 2 0 11/02/19 01/02/20 03/02/20 05/02/20 07/02/20 09/02/20 11/02/20 01/02/21 03/02/21 05/02/21 07/02/21 09/02/21 11/02/21 01/02/22 03/02/22 The GARCH (1,1) model for the TRY is explosive, The volatilities of the LBP, TRY, and SYP are likely due to the large movements (i.e., appreciation provided in Figure 54. and depreciation relative to the US dollar) in the TRY Figure 54 suggests the absence of strong in November and December of 2021. Therefore, the commonalities between the volatilities of the volatility of the TRY is estimated using an Exponential TRY, SYP, and LBP. Indeed, the commonality in GARCH (1,1) model. volatility between SYP and LBP is stronger than the Technical Appendix 59 FIGURE 55 • Total Volatility Spillover Index: LBP, SYP and TRY TOTALSPILL 25 20 15 10 5 0 2020:05:27 2020:06:17 2020:07:08 2020:07:29 2020:08:19 2020:09:09 2020:09:30 2020:10:21 2020:11:11 2020:12:02 2020:12:23 2021:01:13 2021:02:03 2021:02:24 2021:03:17 2021:04:07 2021:04:28 2021:05:19 2021:06:09 2021:06:30 2021:07:21 2021:08:11 2021:09:01 2021:09:22 2021:10:13 2021:11:03 2021:11:24 2021:12:15 2022:01:05 2022:01:26 2022:02:16 2022:03:09 commonalities among the volatility of TRY and LBP More specifically, following the onset of Lebanon’s or TRY and SYP. financial crisis in late 2019, the correlation between The total volatility spillover index is provided in the movements of the two currencies increased. The Figure 55. correlation also increased in the summer of 2021 and in The total volatility spillover index exhibits similar early 2022. However, there appeared to be a decoupling dynamics to that in Figure 51. More specifically, the index in the movements and volatilities of the two currencies is at a maximum of 21.94 percent on June 15, 2021. in the wake of the termination of the subsidy scheme The directional volatility spillover analysis, in Lebanon in September 2021. The pattern in the provided in Figure 56, suggest that directional spill- volatility spillovers mimics that of the correlation in the overs from TRY were the highest in November and two currencies. There is a marked increase in volatility December 2021. The increase in the total spillover spillovers at the onset of the Lebanese financial crisis in index toward the end of 2021 can be ascribed to the October and November of 2019 and during the summer increase in the directional spillover from the TRY. of 2021, and spillovers in volatility are increasing in early In contrast, at the onset of the Lebanese financial 2022. Moreover, the findings suggest a greater degree crisis and in early 2022, the directional spillovers from of connectedness between SYP and LBP than between the LBP were high, while the directional spillovers from any of the latter two currencies and TRY. the SYP were elevated in summer 2021. The directional The close commercial and trade ties between spillovers from SYP and LBP appear to have a larger Lebanon and Syria as well as Syrians’ reliance on bearing on the total volatility spillover index than direc- Lebanese banks for their commercial and personal tional spillovers from TRY. The latter dynamics of the activities explain the tight link between the Syrian and volatility spillovers between the SYP and LBP are similar Lebanese pounds prior to July 2021. Further, Syrian to those uncovered when the analysis was carried out businesses’ reported use of the Lebanese black with only LBP and SYP earlier. This suggests that the market to obtain US dollars and avert sanctions under link (in volatility) between SYP and LBP is tighter than the Caesar Act, coupled with the smuggling of sub- the link between TRY and the latter two currencies. sidized goods, gasoline, and diesel from Lebanon to Syria, have created exchange rate market pressures Summary and interpretation of the and simultaneous demand for US dollars in Syria and findings Lebanon. This, in turn, led to a tightening in the link between the two currencies. The findings indicate that there are commonalities in The link between the two currencies weakened the movements and volatilities of the SYP and LBP. after September 2021, the date of the termination 60 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS FIGURE 56 • Directional Volatility Spillovers: LBP, SYP and TRY Directional Volatility Spillovers (FROM) Variance of LBP 30 25 20 15 10 5 0 2020 2021 Variance of TRY 45 40 35 30 25 20 15 10 5 0 2020 2021 Variance of SPY 30 25 20 15 10 5 0 2020 2021 of subsidies in Lebanon, which indicates the easing Indeed, owing to the subsidies that were ter- of simultaneous exchange market pressures. This minated only in September 2021, the prices of diesel apparent decoupling in the movement of the two cur- and gasoline in Lebanon were the lowest among the rencies is likely to be explained by the lower demand comparators, which are Syria, Iraq, Turkey, Yemen, in Syria to purchase the smuggled subsidized goods, Libya, and Jordan, some of which are fuel-rich countries gasoline, and diesel from Lebanon after the termina- (Figure 57.A, B and C). This created a strong incentive tion of subsidies. Indeed, following the termination for smuggling, particularly given that the centers of of subsidies, purchasing these goods in Lebanon economic activity in Syria are close to the Lebanese and selling them in Syria would not have been less border. The smuggling hypothesis is supported by profitable. evidence that Lebanon has imported a large volume Technical Appendix 61 FIGURE 57 • Gasoline and Diesel Prices in Lebanon are among the Very Lowest in the Region in the pre-Subsidy Termination Era A. Diesel price, market price B. Gas price, market price (per litre, $US) (per kg, $US) 1.6 1.8 1.4 1.6 1.2 1.4 1.2 1.0 1.0 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0.0 0.0 Jan-19 Mar-19 May-19 Jul-19 Sep-19 Nov-19 Jan-20 Mar-20 May-20 Jul-20 Sep-20 Nov-20 Jan-21 Mar-21 May-21 Jul-21 Sep-21 Nov-21 Jan-22 Mar-22 May-22 Jan-19 Mar-19 May-19 Jul-19 Sep-19 Nov-19 Jan-20 Mar-20 May-20 Jul-20 Sep-20 Nov-20 Jan-21 Mar-21 May-21 Jul-21 Sep-21 Nov-21 Jan-22 Mar-22 May-22 Lebanon Syria Turkey Iraq Lebanon Syria Turkey Iraq Jordan Yemen Jordan Yemen Libya C. Oil price, Syria and Lebanon D. Lebanon imports of petroleum oil (US$ per ton) (thousand tons; million US$, 3-month moving average) 1,400 1,200 1,200 1,000 1,000 800 800 600 600 400 400 200 200 0 0 Dec-18 Feb-19 Apr-19 Jun-19 Aug-19 Oct-19 Dec-19 Feb-20 Apr-20 Jun-20 Aug-20 Oct-20 Dec-20 Feb-21 Apr-21 Jun-21 Aug-21 Oct-21 Dec-21 Feb-22 Apr-22 Jun-17 Sep-17 Dec-17 Mar-18 Jun-18 Sep-18 Dec-18 Mar-19 Jun-19 Sep-19 Dec-19 Mar-20 Jun-20 Sep-20 Dec-20 Mar-21 Jun-21 Sep-21 Dec-21 Mar-22 Petroleum oil price: Lebanon imports Import volume of petroleum oil (thousand tons) Petroleum oil price: Syrian imports Import value of petroleum oil (million US$) Domestic diesel price: Lebanon Brent crude oil price: Global Domestic (transport) diesel price: Syria Source: WFP Price Bulletin, country office reports; UN Comtrade database, World Bank staff estimates. of oil derivatives before the termination of subsidies data suggest that these policies were implemented (Figure 57.D). successfully. Indeed, Syria’s import volume has more The less precipitous depreciation (and lower than halved from 2019 to 2021. Further, the Syrian volatility) in the Syrian pound since mid-2021 is likely authorities have drastically reduced the list of critical to be attributable to the import restrictions that were goods that are imported at the preferential exchange imposed by the Syrian authorities. These policies, rates, leading to lower margins of profitability for which restrict the imports of non-essential goods, importers and, hence, a diminished incentive to import. aimed at restricting the use of the limited foreign cur- The import-restriction policies also likely contributed to rency reserves to essential food imports and thereby the lower volatility of the SYP and to the decoupling decelerate their depletion. Estimates from the maritime between the SYP and LBP since September 2021. 62 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS Average Exchange Rates for Syria • Csm denotes ratio of imported services to total services in the consumption basket. Introduction Cg and Cs are derived from the weights for A multiple exchange rate system exists in Syria. different components in the Consumer Price Index More specifically, two prevailing exchange rates are (CPI). Specifically, Cs is calculated by summing up the Central Bank of Syria (or official) and market the weights of CPI components that are assumed to exchange rates. The official exchange rate is used be consumption of services.72 Remaining components for: (i) the state budget and public sector transactions; in the CPI are assumed to be consumption of goods. (ii) money transfers from abroad through the official The weights of CPI components change through time. channels; (iii) fees paid by Syrian men seeking to According to the latest data from the Central Bureau of avoid mandatory military service; (iv) international aid Statistics, in 2020: operations; and (v) imports of critical commodities Cs = 26% such as sugar, rice, vegetable oil, and selected medical products. In contrast, the market exchange Cg = 1− Cs = 74% 73 rate is applied when private funds are transferred into Syria through the unofficial channels. Imports of non- We assume that 80 percent of goods in the critical goods also apply the market exchange rates. consumption basket are imported, whereas only 40 There are other exchange rates in Syria71. This percent of services in the consumption basket are includes “the banks and financial institutions” rate, imported. Hence, which is used by private banks and financial institu- m Cg = 80% tions to conduct transactions, including financing Csm = 40% imports and exports of the private sector; the “remit- m tance” rate, which was set by the CBS in April 2013 We assume Cg and Csm remain unchanged and is used by Syrians sending money from abroad; throughout the conflict period. the “United Nations” rate, which was set by the CBS in December 2011 and is used by UN agencies operating Categories of imports in Syria; the ”military service exemption rate”, which is for Syrian men who want to pay the required fee to There are two categories of imported goods and be exempted from the mandatory military service. The services, critical and others: CBS also issues a customs and airline transactions rate in May 2021. These rates are either close to the 71 The Syria Report, “Syrian Pound Exchange Rates – official exchange rates, or they are between the official Central Bank of Syria and Black Market”, (March 31, and market exchange rates. 2022). 72 Components that are assumed to be focused on the Consumption basket consumption of services and their associated weights are: Dwelling Maintenance and Repair (1.254 percent), Goods and Services for Household Maintenance (1.845 The note uses consumption-based weights to estimate percent); Health (3.821 percent); Transportation (7.056 the average exchange rate in Syria. percent); Communication (4.245%); Recreation and We adopt the following nomenclature: culture (0.953 percent); Education (1.556 percent); Restaurant and hotels (2.071 percent); Miscellaneous • Cg denotes share of goods in the consumption Goods and Services (3.348 percent); and Non-Profit basket; Institutions Serving Households (0.002 percent). 73 While the components that are assumed to be focused • Cs denotes share of services in the consumption on the consumption of services also includes goods basket; (i.e., communications), we can assume that this is offset m • C g denotes ratio of imported goods to total by components that are assumed to be focused on the goods in the consumption basket; consumption of goods but that also include services. Technical Appendix 63 FIGURE 58 • Share of Critical Goods Imports in FIGURE 60 • Average Exchange Rate in Syria Syria (SYP/USD) (Share in percent) 4,500 45% 4,000 40% 3,500 35% 3,000 30% 2,500 1,500 25% 1,000 20% 500 15% 0 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 10% 5% Offical exchange rate Market exchange rate 0% Average exchange rate 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 Source: Syrian Pound Today; Central Bank of Syria; UN Comtrade database; Central Bureau of Statistics; World Bank staff estimates. Source: UN Comtrade database; World Bank staff estimates. FIGURE 59 • Weight Structure of the Average Before 2019, the list of critical goods did not Exchange Rate change significantly. Some major policy moves (Share in percent) include the period from 2012–2013, when the authori- 100% ties tightened imports to save scarce foreign reserves, 90% and inputs for strategic industries, such as textiles, 80% rubber, and chemicals, were excluded from the list. In 70% 2016, some food items were excluded from the list of 60% critical goods, such meat, fish, and cereal. However, 50% a few non-food items were added back on the list, 40% such as textiles, yarns, and plastic, in addition to a few 30% pharmaceutical and medical products. 20% 10% Owing to severe foreign currency shortages, the 0% authorities have vastly reduced the list of critical goods 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 that are imported, applying the preferential exchange rates since 2019. More specifically, the authorities Offical exchange rate Market exchange rate announced the prioritization of only a limited number of food commodities and medicine in imports in Source: Syrian Pound Today for market exchange rate; Central Bank of Syria for official exchange rate; UN Comtrade database fo cacualtion of critical imports that are August 2019. In mid-2020, a few more food items were imported at the official exchange rate; Central Bureau of Statistics for CPI weights in Syira; World Bank staff estimates. excluded from the list. Nevertheless, the list of critical imports was expanded in 2021 to include more food items in an effort to mitigate the negative impact of • Imports of critical goods imports apply the price increases and food shortages on citizens. preferential (or the official) exchange rate (E o ) ; We apply the mirror statistics from the UN • Other goods imports, and all services imports, Comtrade database to estimate imports of goods and are traded at the market exchange rate (E m ) ; services by the 4-digit Standard International Trade c • M g denotes the ratio of the value o f Classification (SITC). The share of the imports of critical c critical goods imported as a proportion of the goods M g in Syria from 2011–2020 is estimated as value of total goods imports. follows: 64 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS Categories of imports FIGURE 61 • Time Series Dynamics of the Official, Market and Average Exchange Rates We assumed that domestic produced components all apply to the official exchange rate. This stems from the 3,000 fact that key contracted prices, such as wages, rents, university and school tuition, and services continue to 2,500 apply the official exchange rate. 1,500 Using the above assumptions that 80 percent of goods in the consumption basket are imported, 1,000 whereas only 40 percent of services in the consump- tion basket are imported, the share of transactions 500 that applies the market exchange rate, is thus: 0 Jan-11 Jun-11 Nov-11 Apr-12 Sep-12 Feb-13 Jul-13 Dec-13 May-14 Oct-14 Mar-15 Aug-15 Jan-16 Jun-16 Nov-16 Apr-17 Sep-17 Feb-18 Jul-18 Dec-18 May-19 Oct-19 Mar-20 Aug-20 g) R m = Cs 40% + Cg 80% (1 M c Offical exchange rate Market exchange rate The consumption-based average exchange Average exchange rate rate (AER) can be calculated as follows: AER = R m ∗ E m + (1− R m ) ∗ E o Direct approach to computing the ERPT The Exchange Rate Pass-Through in The simplest approach to gauging the ERPT is to Syria estimate the change in the CPI , ΔCPIt , which is due to a change in the exchange rate, ΔE t . Estimates Introduction of the contemporaneous response of the changes in price level to changes in the exchange rate, The exchange rate pass-through measures the extent ΔCPIt / ΔE t , or to lagged changes in the exchange to which fluctuations in the exchange rate leads rate, ΔCPIt / ΔE t −1, ΔCPIt / ΔE t −2 . to changes in aggregate prices (i.e., inflation). The Table 11 provides the estimates of the various Exchange Rate Pass-Through (EPRT) coefficient is, pass-through coefficients estimated using data for the therefore, akin to an elasticity coefficient, in that it period May 2011 to December 2020. measures the sensitivity of the Consumer Price Index These estimates are subject to considerable (CPI) to the exchange rate. uncertainty as evinced the high standard deviation. In As described in the previous note, a multiple fact, the standard deviation of each of these estimates exchange rate system exists in Syria. More specifi- is very large. cally, the two prevailing exchange rates are the Central Estimates the ERPT coefficient can also be Bank of Syria (or official) and market exchange rates. obtained from more elaborate econometric models. The official exchange rate is not employed in The existing literature commonly employs well-speci- the empirical analysis, given that it does not exhibit fied Vector Autoregressive (VAR) models to gauge the variation in the post-2016 period (Figure 61). The response of prices to an exchange rate shock. The market and average exchange rates are employed in advantage of the latter approach is to allow for dis- the empirical analysis. The latter is computed in the cerning the effects of exchange rate fluctuations on same manner as Lebanon’s Average Exchange Rate inflation over several horizons (one, six, or 12 months). (AER). That is, the AER uses the consumption-based Existing research also employs a direct method weights derived from the weights of the consumption for estimating the ERPT coefficient using regression basket in the CPI. The time series dynamics of the analysis. Studies that use standard pass-through three exchange rates are provided in Figure 61. regressions include Jašová, Moessner, and Takáts Technical Appendix 65 TABLE 11 • Estimates of the Exchange Rate Pass-Through Using the Simple Approach Average exchange rate Market exchange rate Average Standard Deviation Average Standard Deviation ΔCPI t / ΔE t 65.53 302.63 73.53 273.57 ΔCPI t / ΔE t 79.56 437.14 30.69 357.54 (2019), Bailliu and Fuiji, (2004), Bussiere (2013), Ben y t = B0−1B1y t −1 + ! + B0−1B p + B0−1ω t ,(2) Cheikh and Rault (2016), and Ihrig, Marazzi, and Rothenberg (2006), among others. That is, the reduced form residuals relate to the This note will employ the two approaches to structural residuals via: ut = B0 −1 ωt . gauging the ERPT.74 It is worth noting that differences The vector of variable yt includes the logarithmic in the results between the pass-through regression and change in commodity prices, the exchange rate, and the VAR approach are to be expected at the outset. inflation, as measured by changes in the Consumer In fact, when the VAR approach is used, the pass- Price Index (CPI). through effect is computed from Impulse Response Let Pt c , E t , and Pt denote, respectively, the Functions, which are non-linear functions of the VAR’s levels of the commodity prices, as proxied for using parameters (or more specifically, the Vector Moving the S&P GSCI index,75 the exchange rate, measured Average representation’s coefficients). In contrast, as the market or average exchange rate, as well as the pass-through regressions are linear. Hence, some the CPI. differences in the results are to be expected. The vector of variables used in the VAR is c pt , et , and pt which denote the natural logarithm of Estimating the ERPT coefficient with a VAR the variables. That is, the VAR comprises the variables Existing studies commonly employ Vector Autoregres- yt = ⎡ Δ ⎣ pt , Δet , Δpt ⎤ c ⎦ and is estimated using a sive (VAR) or Vector Error Correction (VECM) models recursive ordering (i.e., Cholesky). The VAR model to gauge the degree of the pass-through from the ex- is estimated using the logarithmic change in the vari- change rate to inflation (Bhundia, 2002; Ha, Stock- ables to circumvent possible non-stationarity, and the er and Yilmazkuday, 2019; McCarthy, 2007; Korhonen optimal number of lags is selected using the Bayesian and Wachtel, 2006;; Leigh and Rossi, 2002; McCarthy, Information Criterion (BIC). The sample period is May 2007). All of the latter studies estimate the ERPT coeffi- 2001 to December 2020 (accounting for lags) and cient using impulse response analysis from a well-spec- the analysis is carried out at the monthly frequency. ified model. The extent to which exchange rate (or de- The VAR is estimated using the market and average valuation/depreciation) shocks drive inflation is also exchange rates. examined using forecast error variance decompositions. VAR in differences 74 For a discussion of other methodologies to estimating The first approach to estimating the ERPT coefficient the ERPT, see Jiménez-Rodríguez and Morales- is to specify and estimate a VAR model. A VAR relates Zumaquero (2016). Ortega and Osbat (2020) offer an a (k×1) vector of variables, Yt, to p of its own lags. A excellent review of the literature on the exchange rate structural VAR model is given by: pass-through. 75 Data on the S&P GSCI and Bloomberg indexes are collected from Datastream. Only the S&P GSCI data B0y t = B1y t −1 + ! + B p y t −p + ω t ,(1) are used, given the high correlation between the two commodity indexes. Monthly observations are obtained The VAR model in reduced form can be written as: from daily data by averaging the daily observations. 66 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS FIGURE 62 • Response to a 1% Shock in the Change in the SYP CPC SYP_M_C INF 0.2 0.20 0.8 0.15 0.0 0.10 0.4 0.05 –0.2 0.00 0.0 –0.05 –0.4 –0.4 –0.10 0 1 2 3 4 5 6 7 8 9 10 11 0 1 2 3 4 5 6 7 8 9 10 11 0 1 2 3 4 5 6 7 8 9 10 11 Responses to SYP The impulse responses and Forecast Error FIGURE 63 • Forecast Error Decomposition for Variance Decompositions (FEVDs) for inflation are Inflation provided in Figures 62 and 63, respectively. 1.00 The Impulse Response Functions (IRFs) in Figure 62 show that inflation responds significantly to 0.75 a shock in the SYP and that the response of inflation peaks one month following the shock. The FEVD also 0.50 suggests that the change in the SYP is an important 0.25 driver of changes in the variance of inflation. The pass-through coefficient (Leigh and Rossi, 0.25 2002) or, more precisely, the Price-to-Exchange Ratio (PERR) coefficient is computed as (Ortega and Osbat, 0.00 2020): 0 1 2 3 4 5 6 7 8 9 10 11 Commodity Price Change in SYP Inflation PTt ,t + j = Pt ,t + j / E t ,t + j ,(3) where ,t + j t = PTtPT = ,t + jPt ,t +P and j t/ E ,t + j t/ E ,t + ,t + j respectively, the cumulative j tare, changes in the price level and the exchange rate TABLE 12 • Cumulative Effect of an Exchange Rate Depreciation between months t and t+j. The PERR is provided in Figure 63. Panel A: Market Rate The cumulative effect, over a horizon of twelve Change in Exchange Rate Change in Inflation months, of the exchange rate shock on inflation, which 1% 0.42% can be interpreted as the pass-through, is reported in 100% 42% Table 12. Panel B: Average Exchange Rate VAR in log level 1% 0.47% In order to assess the robustness of the results, a VAR 100% 47% in levels is estimated. In fact, the VAR in differences will be misspecified in the presence of a cointegrating relation between the variables. This is possible, given nonstationary) and may have a common stochastic that the variables are integrated of order one (i.e., trend (i.e., may be cointegrated) (Figure 62). Technical Appendix 67 FIGURE 64 • The Price to Exchange Ratio PERR PERR2 0.35 0.35 0.30 0.30 0.25 0.25 0.20 0.20 0.15 0.15 0.10 0.10 0.05 0.05 0.00 0.00 1 2 3 4 5 6 7 8 9 10 11 1 2 3 4 5 6 7 8 9 10 11 FIGURE 65 • Exchange Rates and the CPI Johansen (1988) proposes testing for the existence of a long-run cointegrating relation by exam- 3,000 ining the rank of matrix P. The rank of P will not be 2,500 significantly different from zero if the variables are not cointegrated. 1,500 The presence of a cointegration relation in the VAR in log levels is tested using the Johansen (1988) 1,000 approach. More specifically, the existence of cointe- 500 grating vectors can be examined using the trace statistic: 0 ( ˆ λtrace (r ) = −T ∑ig=r +1 In 1− λ) i ,(5) Apr-11 Sep-11 Feb-12 Jul-12 Dec-12 May-13 Oct-13 Mar-14 Aug-14 Jan-15 Jun-15 Nov-15 Apr-16 Sep-16 Feb-17 Jul-17 Dec-17 May-18 Oct-18 Mar-19 Aug-19 Jan-19 Jun-20 Nov-20 CPI SYP_CB where r is the number of cointegrating vectors SYP_M SYP_AER under the null hypothesis and λ ˆ is the estimated i i ordered eigenvalue of the matrix P. The trace th statistic tests the null hypothesis that the number of cointegrating vectors is r or less against the Therefore, estimating the VAR in levels and testing for alternative hypothesis that there are more than r cointegration is necessary. cointegrating vectors. Starting from a VAR in log levels as a data The trace statistic, as well as the Phillips generating process: and Ouliaris (1990) cointegration test suggest the absence of a cointegrating relation. Therefore, the Yt = A1y t −1 + ! + Ap y t −p + ut ,(3) VAR is estimated in log levels. The results from estimating the VAR in log The Vector Error Correction Model (VECM) levels are provided in Table 13. representation can be obtained by subtracting yt–1 from the two sides of the equation: Standard pass-through regressions Δy t = Πy t −1 + Γ1Δy t −1 + ! + Γ p Δy t −p +1 + ut ,(4) The standard pass-through regression: where first specification The second approach consists of using the standard = (I k A1 Ap ) and i = (A i +1 + ) + Ap for pass-through regressions (Gopinath, Itskhoki, and i =1, ..., p 1 Rigobon, 2010): 68 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS TABLE 13 • Cumulative Effect of an Exchange TABLE 14 • Pass-Through Regressions: Estimation Rate Depreciation Results Panel A: Market Rate Market Rate AER Change in Exchange Rate Change in Inflation Constant –2.093 –0.976 (1.307) (1.260) 1% 0.30% 100% 30% Contemporaneous Exchange Rate Change 0.237*** 0.346*** (0.062) (0.077) Panel B: Average Exchange Rate First Lag of Exchange Rate Change 0.088** 0.077* 1% 0.35% (0.042) (0.045) 100% 35% Second Lag of Exchange Rate Change 0.055** 0.051 (0.029) (0.044) pt = a + nj =0 j et 1 + n j =0 j ptf j Third Lag of Exchange Rate Change 0.165*** 0.203*** 3 com + j =0 j pt j + t , (6) (0.031) (0.040) Fourth Lag of Exchange Rate Change 0.072*** 0.036 where p is the CPI in Syria, e is the exchange rate (0.023) (0.031) quoted as units of SYP per US$, ptf is the foreign price Fifth Lag of Exchange Rate Change –0.029 –0.078** level proxied for using the CPI of the United States (0.022) (0.030) and ptcom is a commodity price index. As noted in Sixth Lag of Exchange Rate Change 0.121*** 0.135*** Gopinath, Itskhoki, and Rigobon (2010), the statistic of (0.030) (0.043) interest, which measures the effect of changes in the Seventh Lag of Exchange Rate Change 0.072*** 0.085** exchange rate on inflation, is β (n ) ≡ Σn j =0 β j . Equation (0.024) (0.034) (6) is estimated with n = 12 and using data for the Eighth Lag of Exchange Rate Change –0.025 –0.069 period May 2012 to December 2021 (accounting for (0.025) (0.042) lags).76 Ninth Lag of Exchange Rate Change 0.050 0.034 The above regression is a modified version of (0.034) (0.048) Gopinath, Itskhoki, and Rigobon (2010)’s pass-through Tenth Lag of Exchange Rate Change 0.054** 0.070** regression, and variants of it have been employed in (0.027) (0.034) the literature to measure the ERPT coefficient (Bailliu Eleventh Lag of Exchange Rate Change –0.038 –0.081* and Fuiji, 2004; Bussiere, 2013, among others). (0.023) (0.041) When the market rate is employed, the results Twelfth Lag of Exchange Rate Change 0.018 –0.014 (0.027) (0.041) show that b(12) = 0.84. When the AER is employed, the results suggest that b(12) = 0.79. While the coef- Contemporaneous US Inflation 0.771 –0.709 (2.512) (2.411) ficient associated with the contemporaneous change in the exchange rate, b0, is in line with the results of First Lag of US Inflation 7.329*** 6.295** (2.735) (2.469) the VAR, the latter pass-through effects are larger than those suggested by the VAR. Second Lag of US Inflation (2.663) –2.066 –2.173 (2.218) Figure 66 provides time-varying estimates of b0 Third Lag of US Inflation –0.267 –1.433 obtained from rolling regressions. (2.900) (2.830) The time-varying estimates suggest that the Fourth Lag of US Inflation 5.632*** 5.574*** contemporaneous effect of the exchange rate on infla- (2.019) (1.941) tion increased markedly since 2019. This increase is Fifth Lag of US Inflation –1.547 –0.850 (1.818) (1.615) 76 The data are the same as those used for estimating Sixth Lag of US Inflation 1.978 1.447 the VAR models. More specifically, the sample spans (1.247) (1.063) the period May 2011 to December 2020. The sample Seventh Lag of US Inflation 2.576* 1.720 reported in the text accounts for missing observations (1.337) (1.266) due to lags. (continued on next page) Technical Appendix 69 TABLE 14 • Pass-Through Regressions: Estimation in circulation, which fed into inflation expectations in Results (continued) Syria. Market Rate AER Standard pass-through regression: second Eighth Lag of US Inflation 2.546 2.786* (1.705) (1.457) specification A simpler specification of the standard pass-through Ninth Lag of US Inflation –1.177 –2.363 (1.833) (1.500) regression is estimated: Tenth Lag of US Inflation 1.946 1.920 j =0 β j Δet − j + ∑ j =0 δ j Δpt − j + ε t ,(7) Δpt = a + ∑n 3 com (1.595) (1.619) Eleventh Lag of US Inflation 0.365 0.453 (1.350) (1.217) Equation (7) excludes US inflation, which brings it closer to the specification used in the VAR. Twelfth Lag of US Inflation –1.909 –2.325* When the market rate is employed, the results show (1.239) (1.242) that b(12) = 0.70. When the AER is employed, the Contemporaneous Commodity Price 0.018 0.074 results suggest that b(12) = 0.69. Figure 65 provides Change (0.053) (0.051) the time-varying estimates of the coefficient beta zero. First Lag of Commodity Price Change –0.071 –0.031 In order to better track the variation in the (0.073) (0.069) contemporaneous pass-through coefficient, we plot Second Lag of Commodity Price Change –0.211** –0.169* the time-variation in the coefficient beta zero from the (0.097) (0.092) simplest possible specification: Third Lag of Commodity Price Change 0.068 0.118 Δpt = a + β0 Δet + εt ,(8) (0.090) (0.097) The results are provided in Figure 68. Notes: Newey and West (1987) Heteroskedasticity and Autocorrelation Consistent standard errors are in parentheses. *,**,*** denote, respectively, statistical significance at the 1%, 5% and 10% levels. Summary likely to be attributable to higher inflation expectations The results suggest that a depreciation of 100 percent in Lebanon, owing to a sharp increase in currency in the SYP increases inflation by 30 to 84 percent. FIGURE 66 • Rolling Estimates of the Contemporaneous Effect of the Exchange Rate on Inflation Rolling Estimates of Beta Zero with Market Rate Rolling Estimates of Beta Zero with AER 1.2 1.4 1.0 1.2 1.0 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0.0 0.0 –0.2 –0.2 –0.4 –0.4 –0.6 Aug-15 Feb-15 Nov-16 May-16 Aug-16 Feb-16 Oct-17 Mar-17 Aug-17 Feb-17 Jun-18 Nov-18 Aug-18 Feb-18 Feb-19 Jul-19 Aug-19 Feb-19 Oct-20 Mar-20 Aug-20 Nov-20 Aug-15 Feb-15 Nov-16 May-16 Aug-16 Feb-16 Oct-17 Mar-17 Aug-17 Feb-17 Jun-18 Nov-18 Aug-18 Feb-18 Feb-19 Jul-19 Aug-19 Feb-19 Oct-20 Mar-20 Aug-20 Nov-20 Coeffs(2) Lower Upper Notes: This figure provides rolling estimates of the contemporaneous effect of the exchange rate on inflation. The 95% confidence bands are in blue. 70 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS FIGURE 67 • Rolling Estimates of the Contemporaneous Effect of the Exchange Rate on Inflation from the Second Specification Rolling Estimate of Beta Zero with Market Rate Rolling Estimate of Beta Zero with AER 0.8 1.2 0.7 1.0 0.6 0.8 0.5 0.6 0.4 0.4 0.3 0.2 0.2 0.1 0.0 0.0 –0.2 –0.1 –0.4 –0.2 –0.6 Jun-15 Oct-15 Feb-16 Jun-16 Oct-16 Feb-17 Jun-17 Oct-17 Feb-18 Jun-18 Oct-18 Feb-19 Jun-19 Oct-19 Feb-20 Jun-20 Oct-20 Jun-15 Sep-15 Dec15 Mar-16 Jun-16 Sep-16 Dec-16 Mar-17 Jun-17 Sep-17 Dec-17 Mar-18 Jun-18 Sep-18 Dec-18 Mar-19 Jun-19 Sep-19 Dec-19 Mar-20 Jun-20 Sep-20 Dec-20 Coeffs(2) Lower Upper FIGURE 68 • Rolling Estimates of the Contemporaneous Effect of the Exchange Rate on Inflation from the Third Specification Rolling Estimate of Beta Zero with Market Rate Rolling Estimate of Beta Zero with AER 0.7 1.0 0.6 0.8 0.5 0.4 0.6 0.3 0.4 0.2 0.1 0.2 0.0 0.0 –0.1 –0.2 –0.2 –0.3 –0.4 Jun-14 Oct-14 Feb-15 Jun-15 Oct-15 Feb-16 Jun-16 Oct-16 Feb-17 Jun-17 Oct-17 Feb-18 Jun-18 Oct-18 Feb-19 Jun-19 Oct-19 Feb-20 Jun-20 Oct-20 Jun-14 Oct-14 Feb-15 Jun-15 Oct-15 Feb-16 Jun-16 Oct-16 Feb-17 Jun-17 Oct-17 Feb-18 Jun-18 Oct-18 Feb-19 Jun-19 Oct-19 Feb-20 Jun-20 Oct-20 Coeffs(2) Lower Upper Therefore, the pass-through effect in Syria is high. report short and long-run ERPT coefficients for the The ERPT coefficient appears to have increased G7 countries that range from –0.0138 to 0.00179.77 significantly following the onset of the Lebanon Admittedly, a better assessment of the degree of financial crisis in October 2019. In fact, the upward the pass-through in Syria would entail comparing trend in the ERPT since October 2019 is clearly the ERPT coefficient for Syria to that of developing discernable (in Figures 63, 64 and 65). economies or emerging markets using comparable The ERPT coefficient for Syria can be placed in context by benchmarking it to the findings of 77 The pass-through effect reported by Ihrig, Marazzi, and the literature for the other countries. For instance, Rothenberg (2006) for the Group of Seven (G7) is of a Jiménez-Rodríguez and Morales-Zumaquero (2016) comparable magnitude. Technical Appendix 71 techniques. Jašová , Moessner, and Takáts (2019) Ca’Zorzi, Hahn, and Sánchez (2007) gauge the estimate the ERPT in the post-2008 crisis using data pass-through by accumulating the responses of con- for a panel of developed and emerging market econo- sumer prices to 1 percent exchange rate shock from a mies and Generalized Method of Moments (GMM) well-specified VAR. Therefore, their results are directly estimation of a hybrid New Keynesian Phillips curve. comparable to the results obtained from the VAR for The authors document that the pass-through effect Syria. At a horizon of 12 months, the pass-through declined for emerging market economies following effect for Syria is larger than that reported for China, the 2008 financial crisis. More specifically, Jašová , Hong Kong, Korea, Singapore, Taiwan, Turkey, Poland, Moessner, and Takáts (2019) estimate a yearly pass- Chile, and Argentina in Table 14 of Ca’Zorzi, Hahn, and through coefficient of 0.222 to 0.231 for the emerging Sánchez (2007). Only the Czech Republic, Hungary, and market economies. Mexico exhibit a higher exchange rate pass-through. 72 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS REFERENCES Abadie, A., A. Diamond, and J. Hainmueller. 2015. al-Aswad, H. 2022. “Syria: Protests Mount Amid “Comparative Politics and the Synthetic Anger Over ‘Unfair’ Cuts to Government Control Method.” American Journal of Political Subsidies.” Middle East Eye, February 12, Science 59 (2): 495–510. 2022. https://www.middleeasteye.net/news/ Abadie, A., A. Diamond, and J. Hainmueller. 2010. syria-sweida-anger-mount-protests-cuts-govern- “Synthetic Control Methods for Comparative ment-subsidies. Case Studies: Estimating the Effect of Alesina, A., S. Michalopoulos, and E. Papaioannou. California’s Tobacco Control Program.” Journal 2016. “Ethnic Inequality.” Journal of Political of the American Statistical Association 105 Economy 124 (2): 428–488. (490): 493–505. Arezki, R., and M. Brueckner. 2014. “Effects of Abadie, A., and J. Gardeazabal. 2003. “The Economic International Food Price Shocks on Political Costs of Conflict: A Case Study of the Basque Institutions in Low-income Countries: Evidence Country.”  American Economic Review 93 (1): From an International Food Net-export Price 113–132. Index.” World Development 61 (C): 142–153. Abbara, A., K. Blanchet, Z. Sahloul, F. Fouad, et al. 2015. Bailliu, J., and E. Fujii. 2004. “Exchange Rate “The Effect of the Conflict on Syria’s Health Pass-Through and the Inflation Environment System and Human Resources for Health.” in Industrialized Countries: An Empirical World Health & Population 16 (1): 87–95. Investigation.” Working Paper 2004–21, Bank Adhikari, B., R. A. Duval, B. Hu, and P. Loungani. of Canada, Ottawa. 2016. “Can Reform Waves Turn the Tide? Basihos, S. 2016. “Nightlights As a Development Some Case Studies Using the Synthetic Indicator: The Estimation of Gross Provincial Control Method.” IMF Working Paper 16/171, Product (GPP) in Turkey.” http://dx.doi. International Monetary Fund, Washington, DC. org/10.2139/ssrn.2885518. 73 Bellemare, M. F. 2015. “Rising Food Prices, Food Markets, edited by S. Goutte and K. Guesmi, Price Volatility, and Social Unrest.” American chapter 3. Hackensack, NJ: World Scientific. Journal of Agricultural Economics 97 (1): 1–21. Daher, D. 2022. “Expelled from the Support System: Ben Cheikh, N., and C. Rault. 2016. “The Pass- Austerity Deepens in Syria.” Middle East Through of Exchange Rate in the Context of the Directions Programme Blog. February 15, 2022. European Sovereign Debt Crisis.” International https://blogs.eui.eu/medirections/expelled-from- Journal of Finance & Economics 21 (2): the-support-system-austerity-deepens-in-syria/. 154–166. Devi, S. 2021. “Health in Syria: A Decade of Conflict.” Beyer, R., E. Chhabra, V. Galdo, and M. Rama. 2018. The Lancet 397 (10278): 955–956. “Measuring Districts’ Monthly Economic Activity Diebold, F. X., and K. Yilmaz. 2009. “Measuring From Outer Space.” Policy Research Working Financial Asset Return and Volatility Spillovers, Paper 8523, World Bank, Washington, DC. With Application to Global Equity Markets.” The Beyer, R. C., S. Franco-Bedoya, and V. Galdo. 2021. Economic Journal 119 (534): 158–171. “Examining the Economic Impact of COVID-19 Diebold, F. X., and K. Yilmaz, K. 2012. “Better to Give Than in India Through Daily Electricity Consumption to Receive: Predictive Directional Measurement and Nighttime Light Intensity.” World of Volatility Spillovers.” International Journal of Development 140, 105287. Forecasting 28 (1): 57–66. Bhundia, A. 2002. “An Empirical Investigation of Doll, C. N., J. P. Muller, and J. G. Morley. 2006. Exchange Rate Pass-Through in South Africa.” “Mapping Regional Economic Activity From IMF Working Paper WP/02/165, International Night-time Light Satellite Imagery.” Ecological Monetary Fund, Washington DC. Economics 57 (1): 75–92. Bussiere, M. 2013. “Exchange Rate Pass‐Through to Donaldson, D. and A. Storeygard. 2016. “The Trade Prices: The Role of Nonlinearities and View From Above: Applications of Satellite Asymmetries.”  Oxford Bulletin of Economics Data in Economics.” Journal of Economic and Statistics 75 (5): 731–758. Perspectives 30 (4): 171–98. Ca’ Zorzi, M., E. Hahn, and M. Sánchez. 2007. Elvidge, C. D., K. E. Baugh, E. A. Kihn, H. W. Kroehl, et “Exchange Rate Pass-Through in Emerging al. 1997. “Relation Between Satellite Observed Markets.” ECB Working Paper No. 739, Visible-near Infrared Emissions, Population, European Central Bank, Frankfurt. Economic Activity and Electric Power Campos, N. F., F. Coricelli, L. Moretti. 2019. Consumption.” International Journal of Remote “Institutional Integration and Economic Growth Sensing 18 (6): 1373–1379. in Europe.” Journal of Monetary Economics 103 Elvidge, C. D., K. E. Baugh, M. Zhizhin, and F. C. Hsu. 88–104. 2013. “Why VIIRS Data Are Superior to DMSP Chadwick, V. 2022. “WFP Says Syrians at ‘Breaking Point.’” for Mapping Nighttime Lights.” Proceedings of Devex. May 12, 2022 https://www.devex.com/ the Asia-Pacific Advanced Network 35: 62–69. news/wfp-says-syrians-at-breaking-point-103192. Elvidge, C. D., T. Ghosh, F. C. Hsu., M. Zhizhin, et al. Corral, P., A. Irwin, N. Krishnan, D. Gerszon Mahler, 2020. “The Dimming of Lights in China During et al.. 2020 Conflict and Fragility: On the Front the COVID-19 Pandemic.” Remote Sensing 12 Lines of the Fight Against Poverty. Washington (17): 2851. https://doi.org/10.3390/rs12172851. DC: World Bank. ETANA Syria. 2022. “Syria Brief: Economic Crisis.” Croft, T. A. 1978. “Nighttime Images of the Earth From March 8, 2022. Damascus. https://etanasyria. Space.” Scientific American 239 (1): 86–101. org/syria-brief-economic-crisis-8-march-2022/. Dagher, L., I. Jamali, and N. Badra. 2020. “The FAO (Food and Agriculture Organization of the United Predictive Power of Oil and Commodity Nations). 2021. The State of Food Security and Prices for Equity Markets.” In Risk Factors and Nutrition in the World. Making Agrifood Systems Contagion in Commodity Markets and Stocks More Resilient to Shocks and Stresses. FAO. 74 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS https://www.fao.org/documents/card/en/c/ Henderson, J. V., A. Storeygard, and D. N. Weil. 2012. cb4476en. “Measuring Economic Growth From Outer Ferroni, F., and F. Canova, 2022. “A Hitchhiker’s Guide Space.”  American Economic Review 102 (2): to Empirical Macro Models.” Working Paper 994–1028. 2021–15, Federal Reserve Bank of Chicago, Hodler, R., and P. A. Raschky. 2014. “Regional Chicago, IL. Favoritism.”  The Quarterly Journal of Fouad, F. M., A. Sparrow, A. Tarakji, M. Alameddine, Economics 129 (2): 995–1033. F. et al. 2017. “Health Workers and the Hurn, S., V. Martin, P. C. Phillips, and J. Yu, J. Weaponisation of Health Care in Syria: A 2021. Financial Econometric Modeling. Oxford, Preliminary Inquiry for The Lancet – American U.K.: Oxford University Press. University of Beirut Commission on Syria.” The Ihrig, J. E., M. Marazzi, and A. D. Rothenberg. Lancet 390 (10111): 2516–2526. 2006. “Exchange-Rate Pass-Through in the Gallegos, J. V. 2012. “The Effects of Civil Conflicts G-7 Countries.” FRB International Finance on Women’s Labor Force Participation: A Discussion Paper 851, Federal Reserve Board, Causal Mechanism Approach.” Institute of Washington, DC. Development Studies, University of Sussex. ILO (International Labour Organization). 2020a. Brighton, UK. Facing Multiple Crises: Rapid Assessment of Gibson, J., S. Olivia, G. Boe-Gibson, and C. Li. the Impact of COVID-19 on Vulnerable Workers 2021. “Which Night Lights Data Should We and Small-scale Enterprises in Lebanon. Use in Economics, and Where?” Journal of Geneva: International Labour Organization. Development Economics 149: 102602. https:// ILO (International Labour Organization). 2020b. doi.org/10.1016/j.jdeveco.2020.102602. Impact of COVID-19 on Workers in Jordan: Gobat, J. and K. Kostial. 2016. “Syria’s Conflict A Rapid Assessment. Geneva: International Economy.” IMF Working Paper 16/123, Labour Organization. International Monetary Fund, Washington, DC. Jašová, M., R. Moessner, and E. Takáts. 2019. Gopinath, G., O. Itskhoki, and R. Rigobon. 2010. “Exchange Rate Pass-Through: What Has “Currency Choice and Exchange Rate Pass- Changed Since the Crisis?” International Through.” American Economic Review 100 (1): Journal of Central Banking 15 (3): 27–58. 304–36. Jiang, J., and D. Kim. 2013. “Exchange Rate Pass- Granger, C. W. 1969. Investigating Causal Relations Through to Inflation in China.” Economic by Econometric Models and Cross-Spectral Modelling 33: 900–912. Methods. Econometrica 37 (3): 424–438. Jiménez-Rodríguez, R., and A. Morales-Zumaquero. Guha-Sapir, D., and O. D’Aoust. 2011. “Demographic 2016. “A New Look at Exchange Rate Pass- and Health Consequences of Civil Conflict.” Through in the G-7 Countries.” Journal of Policy World Development Report 2011, Background Modeling 38 (5): 985–1000. Paper. World Bank, 2011. Johansen, S. 1988. “Statistical Analysis of Cointegration Ha, J., M. Stocker, and H. Yilmazkuday. 2019. “Inflation and Vectors.”  Journal of Economic Dynamics and Exchange Rate Pass-Through.” Policy Research Control 12 (2–3): 231–254. Paper 8780, World Bank, Washington, DC. Kilian, L., and H. Lütkepohl. 2017. Structural Vector Hamilton, K. and Nguyen, T. 2017. “Estimating Syrian Autoregressive Analysis. Cambridge, UK: Human Capital Loss Over the Civil War Years.” Cambridge University Press. Mimeo, World Bank, Washington, DC. Koop, G., M. H. Pesaran, and S. M. Potter. 1996. “Impulse Henderson, V., A. Storeygard, and D. N. Weil. 2011. Response Analysis in Nonlinear Multivariate “A Bright Idea for Measuring Economic Models.” Journal of Econometrics 74 (1): 119–147. Growth.”  American Economic Review 101 (3): Korhonen, I., and P. Wachtel. 2006. “A Note 194–99. on Exchange Rate Pass-Through in CIS References 75 Countries.” Research in International Business OCHA (United Nations Office for the Coordination and Finance 20 (2): 215–226. of Humanitarian Affairs). 2021. “Humanitarian Kreibaum, M., and S. Klasen. 2015. “Missing Men: Needs Overview: Syrian Arab Republic.” United Differential Effects of War and Socialism on Nations, New York. Female Labour Force Participation in Vietnam.” Omar, N. S., and A. Ismal. 2019. “Night Lights and Discussion Paper 181, Courant Research Economic Performance in Egypt.” Advances in Centre, Goettingen, Germany. Economics and Business 7 (2): 69–81. Leigh, D., and M. Rossi. 2002. “Exchange Rate Ortega, E., and C. Osbat. 2020. “Exchange Rate Pass-Through in Turkey.” IMF Working Paper Pass-Through in the Euro Area and EU WP/02/04, International Monetary Fund, Countries.” ECB Occasional Paper, European Washington DC. Central Bank, Frankfurt. Lütkepohl, H., 2005. New Introduction to Multiple Our World in Data (database). “Coronavirus Pandemic Time Series Analysis. Berlin: Springer Science (COVID-19).” Accessed on January 3, 2022. & Business Media. https://ourworldindata.org/coronavirus. Lütkepohl, H., and M. Krätzig. 2004. Applied Time Patton, A. 2013. “Copula Methods for Forecasting Series Econometrics. Cambridge, UK: Multivariate Time Series.” In Handbook of Cambridge University Press. Economic Forecasting 2 (B), edited by G. Elliott McArthur, J. W., and G. C. McCord. 2017. “Fertilizing and A. Timmermann, 899–960. North-Holland Growth: Agricultural Inputs and Their Effects Publishing. in Economic Development.” Journal of Pesaran, H. H., and Y. Shin. 1998. “Generalized Impulse Development Economics 127: 133–152. Response Analysis in Linear Multivariate McCarthy, J. 2007. “Pass-Through of Exchange Rates Models.” Economics Letters 58 (1): 17–29 and Import Prices to Domestic Inflation in Some Phillips, P. C., and S. Ouliaris. 1990. “Asymptotic Industrialized Economies.” Eastern Economic Properties of Residual Based Tests for Journal 33 (4): 511–537. Cointegration.”  Econometrica: Journal of the McGovern, M. E., A. Krishna, V. M. Aguayo, and S. V. Econometric Society 58 (1): 165–193. Subramanian. 2017. “A Review of the Evidence Prakash, A., A. K. Shukla, C. Bhowmick, and R. Linking Child Sunting to Economic Outcomes.” Beyer. 2019. “Night-time Luminosity: Does It International Journal of Epidemiology 46 (4): Brighten Understanding of Economic Activity 1171–1191. in India?” Reserve Bank of India Occasional Menon, N., and Y. Van der Meulen Rodgers. 2015. Papers 40 (1) 1–24. “War and Women’s Work: Evidence From Roberts, M. 2021. “Tracking Economic Activity the Conflict in Nepal.” Journal of Conflict in Response to the COVID-19 Crisis Resolution 59 (1): 51–73. Using Nighttime Lights – The Case of Mercy Corps 2021. “Using Night Lights to Measure Morocco.” Development Engineering 6: 100067. Economic Output in Syria.” May 27, 2021. https:// https://doi.org/10.1016/j.deveng.2021.10006. www.humanitarianaccessteam.org/sites/default/ Rose, C. and A. Söderholm. 2022. Intelligence files/reports-docs/Quantitative%20report_ Briefing. “The Captagon Threat: A Profile of Illicit Using%20nighlights%20to%20measure%20 Trade, Consumption, and Regional Realities.” economic%20output_May%202021_0.pdf. New Lines Institute for Strategy and Policy, North Press Agency. 2021. “Syria To Import One April 5, 2022. https://newlinesinstitute.org/ Million Tons Of Wheat From Russia,” December wp-content/uploads/20220404-Captagon_ 6, 2021. https://npasyria.com/en/68780/ Report-NLISAP-final-1.pdf. North Press Agency. 2022. “Syria’s Wheat Production Ruta, M. 2022. The Impact of the War in Ukraine on Is Lowest In 50 Years,” December 26, 2021. Global Trade and Investment. Washington, DC: https://npasyria.com/en/69764/. World Bank. 76 SYRIA ECONOMIC MONITOR: LOST GENERATION OF SYRIANS Sharara, S. L., and S. S. Kanj. 2014. “War and The Syria Report. 2022c. “Government to Reduce Infectious Diseases: Challenges of the Syrian Health and Education Subsidies, Media Civil War.” PLoS Pathogens 10 (11) e1004438. Says.” March 15, 2022. https://syria-report. https://doi.org/10.1371/journal.ppat.1004438. com/news/government-to-reduce-health-and- Silungwe, A., A. Baer, and V. Guerreiro. 2022. “Global education-subsidies-media-says/. Stocktaking of National Accounts Statistics: The Syria Report. 2022d. “Syrian Pound Exchange Availability for Policy and Surveillance.” IMF Rates – Central Bank of Syria and Black Working Paper 2022/029, International Market.” March 31, 2022. https://syria-report. Monetary Fund, Washington, DC. com/library/rate-of-the-u-s-dollar-in-the-syrian- Storeygard, A. 2016. “Farther on Down the Road: forex-market-2011–21–2/. Transport Costs, Trade and Urban Growth in UNCTAD (United Nations Conference on Trade and Sub-Saharan Africa.” The Review of Economic Development). 2018. Review of Maritime Studies 83 (3): 1263–1295. Transport 2018. Geneva: UNCTAD. Sutton, P. C., and R. Costanza. 2002. “Global Estimates Urdal, H., and C. P. Che. 2013. “War and Gender of Market and Non-market Values Derived Inequalities in Health: The Impact of Armed From Nighttime Satellite Imagery, Land Cover, Conflict on Fertility and Maternal Mortality.” and Ecosystem Service Valuation.” Ecological International Interactions 39 (4): 489–510. Economics 41 (3): 509–527. World Bank. 2017. The Toll of War: The Economic and Syrian Center for Policy Research. 2017. “Social Social Consequences of the Conflict in Syria. Degradation in Syria. The Conflict Impact on Washington DC: World Bank. Social Capital.” Damascus. World Bank. 2019. The Mobility of Displaced Syrians: The Syrian Report. 2022a. “Brussels Conference An Economic and Social Analysis. Washington Pledges Unexpectedly Rise to EUR 4.1 Billion.” DC: World Bank. May 11, 2022. https://syria-report.com/news/ World Bank. 2020. The Fallout of War: The Regional brussels-conference-pledges-unexpectedly- Consequences of the Conflict in Syria. rise-to-eur-4–1-billion/. Washington DC: World Bank. The Syria Report. 2022b. “Government Removes Millions World Bank. Forthcoming. Syria Joint Damage from Subsidy Programme, Triggering Rare Assessment of Selected Cities. Washington Protests.” February 8, 2022. https://syria-report. DC: World Bank. com/news/government-removes-millions-from- subsidy-programme-triggering-rare-protests/. References 77 1818 H Street, NW Washington, DC 20433