Jordan: Strategy for Adjustment and Growth

OCCASIONAL PAPER 136

Jordan Strategy for Adjustment and Growth

Edited by Edouard Maciejewski and Ahsan Mansur, with contributions from Patricia AIonso-Gamo, Jean-Pierre Chauffour, Etienne de Call at ay, and Christopher McDermott

INTERNATIONAL MONETARY FUND Washington DC May 1996

Cataloging-in-Publication Data

Jordan : strategy for adjustment and growth / edited by Edouard Maciejew- ski and Ahsan Mansur, with contributions from Patricia Alonso-Gamo . . . [et al.]. — Washington, DC : International Monetary Fund, [1996] p. cm. — (Occasional Paper, ISSN 0251-6365 ; 136) ISBN 1-55775-558-2 1. Jordan — Economic policy. 2. Jordan — Social policy. 3. Debts, External — Jordan 4. Jordan — Economic conditions. I. Maciejewski, Edouard. II. Mansur, Ahsan S., 1951-. III. Alonso-Gamo, Patricia. IV. Occasional paper (International Monetary Fund) ; no. 136. HC415.26.J67 1996

Price: US$15.00 (US$12.00 to full-time faculty members and students at universities and colleges)

Please send orders to: International Monetary Fund, Publication Services 700 19th Street, N.W., Washington, D.C. 20431, U.S.A. Tel: (202) 623-7430 Telefax: (202) 623-7201 Internet: [email protected]

recycled paper

©International Monetary Fund. Not for Redistribution Contents

Page Preface vii I Introduction I M Overview of Macroeconomic Performance and Structural Reforms Since Late 1988 2 Edouard Maciejewski andAhsan Mansur Structure of the Economy 2 Origins of Economic Difficulties 5 Broad Strategy During the 1970s to Mid-1980s 5 Macroeconomic Outcome: An Overview 6 Structural Weaknesses 8 Adjustment and Structural Reform Efforts, 1989-94 9 Performance During 1989-90 9 Performance Since 1992 10 Macroeconomic Stabilization and Outcome 10 Structural Reforms 11 III Macroeconomic Environment and Factors Underlying Growth and Investment 13 Christopher McDermott Economic Developments During 1976-94 13 Boom of the 1970s 13 Economic Slowdown in the Early 1980s 13 Economic Crisis in the Late 1980s 14 Economic Recovery, 1989-94 15 Factors Influencing Long-Term Growth 15 Macroeconomic Stability and Growth 16 Total Factor Productivity Residuals 17 Factors Influencing Investment 18 Methodology and Data 18 Empirical Observations 19 Concluding Observations 20 IV Public Debt Dynamics and Fiscal Policy 21 Etienne de Callatay and Ahsan Mansur Historical Background and Buildup of Public Debt 21 Origin of the Problem 21 Debt Accounting Framework 24 Debt Dynamics and Sustainability of the Debt, 1975-88 25 Shift in Fiscal Policy Stance, 1989-94 25 Debt Dynamics and Sustainability of the Debt Ratio 28 Revenue Mobilization 29

iii

©International Monetary Fund. Not for Redistribution CONTENTS

Taxes on Foreign Trade 29 Taxes on Domestic Transactions 30 Taxes on Income and Profits 30 Nontax Revenue 30 Expenditure 30 Recent and Forthcoming Structural Reforms in the Fiscal Area 31 Appendix Methodology of Debt Dynamics Accounting Framework 32

V Financial Liberalization and Monetary Reforms 34 Jean-Pierre Chauffour Background and Developments 34 Operation of Monetary Policy, 1970-88 34 Monetary Developments 34 Credit Expansion Driven by the Public Sector 36 Use of Monetary Instruments Until the Late 1980s 36 Monetary Policy Under Government Stabilization Programs, 1988-94 37 Use of Monetary Instruments 37 Banking Supervision and Regulatory Measures 38 Development of AFM 3 8 Ongoing and Prospective Financial Reforms 41 Appendix I Money Demand Function 41 Appendix II Money Multiplier and Reserve Money, 1980-94 42 Appendix III Monetary Program Under Indirect Monetary Control 45

VI External Debt Strategy 49 Patricia Alonso-Gamo andAhsan Mansur Emergence of the External Debt Crisis 49 Debt Management Strategy After the Debt Crisis 49 Paris Club Agreement, 1989 49 Negotiations with Commercial Bank Creditors 50 Changes in Debt Strategy 50 Paris Club Rescheduling Agreements, 1992 and 1994 50 Revised Debt Strategy vis-a-vis Commercial Banks 51 Negotiations with Other Official Creditors 52

VII Liberalization of Trade and Exchange Systems 53 Patricia Alonso-Gamo Structure of Trade 5 3 Trade Regime and Trade Policy Reforms Since Late 1988 54 Trade Regime Prior to the Reforms 54 Policy Changes in the Trade Regime 55 Opening Up the Economy 56 Regional Developments and Future Challenges 57 Developments in the Exchange System 57 VIII Social Aspects of the Adjustment Program: Strengthening the Social Safety Net 58 Ahsan Mansur Poverty Profile 58 Poverty and Income Distribution 58 Dimensions and Distribution of Poverty, 1991-92 58 Why Poverty Deteriorated 60 Poverty Alleviation Policies 61 Temporary Social Safety Net Instruments 61 Food Subsidy Schemes 61

©International Monetary Fund. Not for Redistribution Contents

National Aid Fund 62 Health Care Subsidies 63 Other Welfare Programs 63 Permanent Social Safety Net Arrangements 63 Public Education System 63 Public Health System 63 Other Safety Net Arrangements 64 Overview of Recent Developments and Medium-Term Strategy 64

Boxes 5.1. Banking Crisis of 1989-90 39 8.1. Social Security Corporation 59

Tables 2.1. Social and Demographic Indicators 2 2.2. Selected Economic and Financial Indicators 3 3.1. Macroeconomic Performance 14 3.2. External Financing and Investment 14 3.3. Summary of Estimated Panel Regression Coefficients from a Growth Accounting Framework 16 3.4. Impact of Macroeconomic Policy on Real GDP Growth 17 3.5. Contributions to Economic Growth by Factors of Production 18 3.6. Estimates of Private Investment Equation, 1977-94 19 4.1. Central Government Operations 22 4.2. Fiscal Sustainability 26 4.3. Fiscal Impulse 27 5.1. Comparative Performance of Emerging Capital Markets, 1994 40 7.1. Direction of Foreign Trade 54 8.1. Selected Poverty Indicators 60 8.2. Factors Contributing to Change in Poverty 61 8.3. Food Subsidies 62 8.4. Selected Countries: Expenditure on Education 64 8.5. Health Indicators, 1991 64 Al. Sensitivity of the Monetary Program to Selected Parameters 47

Charts 2.1. Structure of the Economy 6 2.2. Movement in Monetary Aggregates, Prices, and the Exchange Rate 6 2.3. Central Government Operations 7 2.4. Central Government Deficit, Revenue, and Expenditure 7 2.5. External Sector 8 2.6. Exchange Rate Indices 8 2.7. External Debt and Debt Service 9 3.1. Real GDP Growth and Investment 15 3.2. Real GDP Growth, Fiscal Balance, and the Current Account 15 4.1. Fiscal Balance 21 4.2. Expenditure 23 4.3. Tax Revenue 23 4.4. Nonmilitary Current Expenditure 23 4.5. Debt-to-GDP Ratio 24 4.6. Evolution of Debt-to-GDP Ratio 28 4.7. Fiscal Impulse Breakdown 28 4.8. Implicit Interest Rate and Economic Growth Rate 29 4.9. Revenue and Grants 29

©International Monetary Fund. Not for Redistribution CONTENTS

5.1. Monetary Aggregates 36 5.2. Prices and Interest Rates 37 Al. Broad Money Demand 42 A2. Income Velocity of Money 42 A3. Quarterly Demand for Money 43 A4. Currency Outside Banks 43 A5. Banks' Cash in Vaults and Reserve Requirements 44 A6. Excess Liquidity and Money Multiplier 45

Figures 5.1. Banking System 35

The following symbols have been used throughout this paper: ... to indicate that data are not available; — to indicate that the figure is zero or less than half the final digit shown, or that the item does not exist; n.a. to indicate that the item is not applicable; between years or months (e.g., 1991-92 or January-June) to indicate the years or months covered, including the beginning and ending years or months; / between years (e.g., 1991/92) to indicate a crop or fiscal (financial) year. "Billion" means a thousand million. Minor discrepancies between constituent figures and totals are due to rounding. The term "country," as used in this paper, does not in all cases refer to a territorial entity that is a state as understood by international law and practice; the term also covers some territorial entities that are not states, but for which statistical data are maintained and provided internationally on a separate and independent basis.

©International Monetary Fund. Not for Redistribution Preface

The papers presented in this study were prepared by staff members who have par- ticipated in cooperative exchanges between Jordan and the IMF over the past three years. The authors acknowledge the excellent research assistance provided by Use- Marie Fayad and secretarial support by Irene Carpenter, Susan Jones, and Mary Ann Miles. The authors are also grateful to Elisa Diehl of the External Relations Depart- ment, who edited the paper for publication and coordinated production. The views expressed here, as well as any errors, are the sole responsibility of the authors and do not necessarily reflect the opinion of the Government of Jordan, the Executive Directors of the IMF, or other members of the IMF staff. The study was completed in August 1995 and is based on information available at that time.

vii

©International Monetary Fund. Not for Redistribution This page intentionally left blank

©International Monetary Fund. Not for Redistribution I Introduction

ordan has made major progress over the past sev- supported by several World Bank sectoral lending J eral years in macroeconomic stabilization and programs, extended debt reschedulings from Paris transformation in its economic structure. Its impres- Club official bilateral creditors and commercial sive performance has been the result of a persistent, banks, and financial assistance from the European courageous, and difficult process of adjustment and Union. structural reform, which started in 1989 and intensi- This paper analyzes recent developments, macro- fied in 1992. The IMF and the World Bank have economic policies, and structural reforms in Jordan been associated closely with this process and at the and focuses on several central aspects of the coun- request of the Jordanian authorities continue to assist try's recent experience. Included are discussions on them in these efforts, particularly in the context of a the factors underpinning growth in the form of struc- rapidly changing regional environment. The authori- tural reforms in key areas, such as public finance, ties' efforts to date have been supported by two the financial system, the trade and exchange regime, stand-by arrangements (1989 and 1992) from the Jordan's external debt-management strategy, and the IMF, and by a three-year extended arrangement dynamics of the public debt and its sustainability under the extended Fund facility (EFF) that was ap- (taking into account the recent stance of fiscal proved in May 1994. Jordan's efforts have also been policy).

©International Monetary Fund. Not for Redistribution II Overview of Macroeconomic Performance and Structural Reforms Since Late 1988

Edouard Maciejewski and Ahsan Mansur

n recent years, Jordan has made major progress I toward achieving macroeconomic stabilization Table 2.1. Social and Demographic and sustained economic growth. The country's Indicators remarkable success in achieving high real growth with continued price stability and a strengthening of Area (thousand sq. km.) 89.2 its balance of payments is attributable to the stabi- Total population (1993; millions) 4.2 lization and reform strategy the authorities adopted Population per sq. km. 21.0 in the wake of the balance of payments crisis of the Rate of population growth late 1980s. The strategy was further strengthened in (1983-93; percent a year) 6.0' the aftermath of the August 1990 regional crisis Population per physician 1.9 Population per hospital bed 1.8 when 300,000 Jordanians (about 10 percent of the Life expectancy at birth (years) 66.6 population) returned from neighboring countries, Crude birthrate (per thousand) 33.9 leading to a sharp increase in unemployment, a Crude death rate (per thousand) 3.0 disruption of trade, a loss of remittances, and a Infant mortality rate (per thousand) 41.2 stoppage of aid flows from other countries in the Population age structure (percent) 0-14 years 41.9 region. 15-64 years 55.5 This section provides a background analysis cov- Population distribution (percent) ering the structure of the economy, the sources of Urban 78.0 Jordan's economic difficulties, and recent adjust- Rural 22.0 Access to safe water (percent of population) 97.0 ment and structural reform efforts. The analysis fo- Access to electricity (percent of population) 99.0 cuses on the emergence of large financial imbal- School enrollment rate ances in the 1980s and the adjustment and structural Primary 97.0 reform efforts undertaken since late 1988. Secondary 79.0 Pupil-teacher ratio Primary 22.0 Secondary 19.0 Structure of the Economy GNP per capita (1993; in U.S. dollars) 1,334.0 Jordan is a lower-middle-income country of about Source: Department of Statistics, Government of Jordan, Stot/'s- ticalYearbook 1993 (Amman, 1993). 4 million inhabitants and an annual per capita in- Note: East Bank only. Most recent estimates are given. come estimated at $1,370 in 1994 (Tables 2.1 and 'Includes the effect of Jordanians returning from abroad in the 2.2). Its economic structure is dominated by trade- aftermath of the 1990 regional crisis. and service-related activities (including government services); these account for more than two-thirds of GDP at factor cost (Chart 2.1), and manufacturing, agriculture, mining, and construction account for the rest. Construction has been the driving force during However, the public sector is not engaged in manu- periods of strong economic growth. Because of its facturing activities; it is primarily limited to provid- very narrow production base, the economy is highly ing basic services (health and education), public util- dependent on imports, which represent nearly 60 ities (water and electricity), and infrastructural percent of GDP. Workers' remittances from the support (mainly in the areas of transportation, com- neighboring oil-exporting countries and processed munications, and irrigation). Central government mining-based exports are the primary sources of Jor- expenditures account for about 35 percent of GDP, dan's foreign exchange earnings. although at about 30 percent of GDP the total ratio The size of the public sector in Jordan is large in of revenue to GDP is also high in comparison with relation to the level of domestic economic activity. other non-oil middle-income countries.

©International Monetary Fund. Not for Redistribution Structure of the Economy

Table 2.2. Selected Economic and Financial Indicators

Est. Est. 1989 1990 1991 1992 1993 1994

In millions of Jordan dinars National income Nominal GDP at market prices 2,372.2 2,668.3 3,855.0 3,493.0 3,882.5 4,266.2 Nominal GDP at factor cost 2,109.6 2,324.5 2,505.6 2,960.9 3,301.2 3,622.1

Annual changes in percent Real GDP at market prices -10.6 1.9 1.8 16.1 5.8 5.8

In percent of GDP at factor cost

Sectoral shares Agriculture 6.6 8.1 8.5 8.3 8.3 8.0 Mining 7.3 6.4 5.0 4.4 3.3 3.1 Manufacturing 12.1 14.9 13.7 13.7 13.7 14.4 Electricity and water 2.5 2.3 2.5 2.2 2.1 2.1 Construction 4.8 4.5 5.0 7.3 7.6 7.4 Trade 8.6 9.3 10.2 9.4 9.5 9.7 Transport and communications 17.0 15.6 15.3 15.2 15.0 15.7 Government services 20.4 19.3 18.9 18.7 19.5 19.2 Other services 20.6 19.6 20.9 20.7 21.0 20.4

In percent of nominal GDP

Consumption 95.0 99.0 97.4 98.5 98.8 96.7 Gross national savings 8.6 3.3 5.0 13.4 14.2 14.6 Gross fixed investment 23.4 26.0 23.7 30.0 31.8 27.4 Annual changes in percent Prices GDP deflator 21.0 11.4 5.2 5.4 5.1 3.9 Cost of living index 25.7 16.2 8.2 4.0 3.3 3.5 Import price index 45.1 36.6 0.2 -5.5 1.6 2.7 Export price index 56.1 19.0 11.3 -3.7 1.0 4.8

In millions of Jordan dinars Government budget Total revenue and foreign grants 869.5 1,026.5 1,052.8 1,306.4 1,332.4 1,389.8 Revenue 565.2 734.8 805.2 1,169.0 1,165.8 1,244.5 Foreign grants 304.3 291.7 247.6 137.4 166.6 145.3 Total expenditure and net lending 1,053.8 1,223.8 1,300.6 1,288.6 1,381.6 1,496.7 Current expenditure 890.0 976.2 1,028.1 1,100.1 1,154.9 1,244.3 Capital expenditure 196.4 159.5 167.7 203.1 248.7 271.6 Net lending 16.2 2.2 -35.8 -15.2 -22.0 -19.2 Extrabudgetary expenditure and discrepancy -48.8 85.9 140.6 Overall balance (excluding grants)' -488.6 -489.0 -495.4 -1 19.6 -215.8 -252.2 Overall balance (including grants)2 -136.1 -197.3 -247.8 92.8 25.3 -12.8 Financing 136.1 197.3 247.8 -92.8 -25.3 12.8 Foreign (net) 90.6 210.5 24.0 39.2 152.7 41.1 Domestic (net) 17.2 -101.9 -61.4 100.8 -13.8 -3.8 Banking system 21.1 -122.5 -90.7 85.8 -28.8 Nonbank sources -3.9 20.6 29.3 15.0 15.0 Change in overdue obligations 28.3 88.7 285.2 -123.9 -20.2 In percent of GDP at current market prices Total revenue and foreign grants 30.6 38.4 36.9 37.6 34.8 32.4 Revenue 23.8 27.5 28.2 33.7 30.0 29.2 Foreign grants 12.8 10.9 8.7 3.9 4.1 3.4 Total expenditure and net lending 44.4 45.9 45.6 36.9 36.4 35.1 Current balance -13.7 -9.0 -7.8 2.0 0.3 Overall balance (excluding foreign grants)1 -20.6 -18.3 -17.4 -3.2 -5.7 -5.9 Primary balance (excluding foreign grants)1 -I I.I -8.0 -6.8 5.9 0.7 -0.4

©International Monetary Fund. Not for Redistribution II OVERVIEW

Table 2.2 (continued)

Est. Est. 1989 1990 1991 1992 1993 1994

Annual changes in millions of Jordan dinars

Money and quasi money 327.6 150.9 763.9 308.6 288.93 359.7 Money 145.7 105.6 179.7 105.8 14.13 16.0 Foreign assets (net) 319.1 259.7 778.4 12.0 -50.83 65.6 Domestic assets (net) 8.5 -108.8 -14.5 296.6 339J3 294.1 Claims on government (net)4 21.2 -122.8 76.6 85.8 -28.33 -57.3 Claims on public entities/enterprises5 -13.3 -0.8 14.2 -6.0 43.53 45.0 Claims on specialized credit institutions -3.6 -2.0 -7.8 -2.5 Claims on financial companies 7.6 -16.5 -9.5 10.1 35.63 17.0 Claims on private sector 101.5 135.0 89.2 236.3 289.83 453.7 Other items (net)6 -104.9 -75.1 -177.2 -27.1 -I.23 -164.3

Changes in percent of beginning period stock of money and quasi money

Money and quasi money 12.4 5.1 24.5 7.9 6.93 8.0 Foreign assets (net) 12.1 8.7 24.9 0.3 -I.23 1.5 Domestic assets (net) 0.3 -3.7 -0.5 7.6 8.13 6.6 Claims on government (net)4 0.8 -4.1 2.5 2.2 -0.73 -1.3 Claims on public entities5 -0.5 0.5 -0.2 I3 1.0 Claims on private sector 3.8 4.5 2.9 6.1 6.93 10.1 Other items (net)6 -3.8 -3.2 -6.2 -0.5 0.83 -3.3

In millions of U.S. dollars

Current account -104 -754 -712 -741 -650 -399 Goods and services -733 -1,176 -897 -1,247 -1,025 -725 Trade balance -1,288 -1,668 -1,439 -2,119 -2,293 -1,950 Exports, f.o.b. 1,110 1,064 1,132 1,220 1,248 1,424 Imports, c.i.f. 2,398 2,732 2,571 3,339 3,541 3,374 Foodstuffs 369 609 613 612 628 586 Oil and oil products 415 466 430 446 454 430 Other 1,614 1,657 1,528 2,281 2,459 2,358 Services (net) 555 493 543 873 1,268 1,225 Remittances (net) 533 430 389 741 962 1,000 Receipts 623 500 450 800 1,040 1,093 Payments 91 71 61 86 78 93 Travel (net) 126 176 35 112 218 188 Receipts 547 512 317 462 563 582 Payments 422 336 282 350 345 394 Investment income (net) -385 -402 -362 -335 -310 -315 Receipts 40 67 114 112 99 73 Payments 425 470 477 447 409 388 Other (net) 282 289 481 381 398 306 Receipts 693 936 1,036 985 1,011 629 Payments 412 647 555 604 613 323 Unrequited transfers 629 422 185 506 375 326 Private 30 29 21 158 35 2 Public 599 393 164 348 340 324

Capital account -399 -362 73 151 -147 -54 Public sector -380 -431 -672 -337 -415 -271 Disbursements 574 439 135 335 155 242 Payments 954 -871 -806 -672 -570 -513 Private sector (net) -19 69 26 45 20 26 Costs of restructuring — — -158 232 -35 Transfers of workers'savings 112 719 600 480 226

Errors and omissions 82 243 372 102 226 17

Overall balance -421 -760 -267 -488 -571 -436

Current account balance7 -2.5 -19.1 -17.0 -14.4 -11.6 -6.5 Overall balance -10.3 -18.9 -6.4 -9.5 -10.2 -7.1

©International Monetary Fund. Not for Redistribution Origins of Economic Difficulties

Table 2.2 (concluded)

Est. Est. 1989 1990 1991 1992 1993 1994

In millions of U.S. dollars, unless otherwise indicated

Gross official reserves8 End of period 48.0 221.0 825.0 769.9 595.2 431.2 Months of imports 0.2 1.0 3.8 2.4 2.2 1.5 Outstanding external public debt End of period 6,412.0 7,591.0 6,872.0 6,625.0 6,658.0 6,651.0 Percent of GDP 157.0 188.8 163.9 128.9 118.8 108.9 External public debt service 1,328.4 1,351.3 1,290.5 1,096.5 1,027.0 831.0 Percent of exports of goods and services 24.9 23.2 38.7 22.4 19.9

Exchange rate U.S. dollar/Jordan dinar Period average 1.7532 1.5089 1.4689 1.4712 1.4434 1.4312

Percentage change in real effective exchange rate9 (-depreciation) -15.6 -3.0 I.I 0.8 -0.2 0.6

Sources: Government of Jordan, Department of Statistics, Statistical Yearbook 1993; Jordanian authorities; and IMF staff estimates. 'Commitment basis. 2Cash basis. 3Based on the revised sectoral classification recommended by the IMF's Statistics Department; because of reclassification, the changes shown under this column may not be meaningful. 4Consists of central government operations: net general budget and net own budget. 5Also includes claims on municipalities and local governments and on the Social Security Corporation. includes claims on specialized credit institutions (for the period up to 1993 under the old classification) and financial institutions. 7Excluding grants from the Gulf Crisis Financial Coordination Group and transfers of workers' savings. 8Readily usable foreign exchange of the Central Bank of Jordan; excluding foreign exchange deposits by residents at the Central Bank of Jordan, gold, and claims on the Central Bank of Iraq; data are for end of period. 9Period averages.

The quality of services provided by the public sec- prospects. These problems were compounded by the tor in Jordan is generally good and social indicators adverse effects of the regional crisis that erupted in are satisfactory. Literacy and enrollment rates at pri- August 1990. mary and secondary school levels are high, and nu- trition and health conditions are generally satisfactory (Table 2.1). Origins of Economic Difficulties Prudent fiscal (or public sector) expenditure programs, combined with the availability of foreign Broad Strategy During the 1970s to grants and the regional economic boom associated Mid-1980s with the sharp rise in oil prices in the mid-1970s and early 1980s, led to economic prosperity in Jordan From the early 1970s until recently, Jordan's during the 1970s and the first half of the 1980s. Sub- strategy was to develop itself as a provider of sequently, however, as the regional economies en- skilled manpower and trade-related services for the tered into a recessionary period in the wake of Arab countries in the region. Accordingly, the falling petroleum prices by the mid-1980s,1 Jordan's authorities had chosen a strategy aimed at edu- underlying financial imbalances came to the fore. By cating Jordan's youth to prepare them for employ- 1987-88, these imbalances had become unsustain- ment in and around the region. At the same time, able and led to a severe balance of payments crisis with the sizable amounts of workers' remittances that seriously undermined the economy's growth and aid received from abroad, Jordan was able to maintain income and consumption at levels that exceeded those that could be expected from the 'These also provided for Jordan's exports of goods and nonfac- available production capacity in the domestic tor services. economy.

©International Monetary Fund. Not for Redistribution II OVERVIEW

Chart 2.1. Structure of the Economy Chart 2.2. Movement in Monetary (In percent of GDP) Aggregates, Prices, and the Exchange Rate

GDP Deflator, Prices, and the SDR-Jordan Dinar Exchange Rate {Percent chances)

Sources: IMF, International Financial Statistics, and various Recent Economic Development reports. 'Including government services.

Macroeconomic Outcome: An Overview Jordan's economic growth during the 1970s and through the mid-1980s was robust. Domestic prices Sources: IMF, International Financial Statistics, and various Recent were generally stable, with inflation averaging Economic Development reports. 5 percent during the decade through the mid-1980s, reflecting in part prudent monetary policy (Chart 2.2).2 As a result of high recurrent expenditures, however, Jordan incurred large fiscal deficits (Charts 2.3 and 2.4) that, excluding foreign grants, averaged more than 20 percent of GDP through the sector (averaging 10 percent of GDP). Investment mid-1980s. These were largely financed by readily remained at comfortable levels (at more than available foreign grants. Fiscal revenues remained 20 percent of GDP), but much of the domestic pri- highly dependent on international trade-based taxes, vate sector investment was directed to housing con- even though the revenue-to-GDP ratio increased to struction and was foreign financed. Despite an un- 25 percent of GDP by the mid-1980s. The domestic usual import boom, the external current account savings rate was relatively modest, with private sec- remained in virtual balance during this period tor savings offsetting large dissavings of the public (Chart 2.5). However, because of inappropriate incentives and price signals, efficient import-competing and ex-

2The average rate of inflation was contained at about 5 percent port-oriented domestic manufacturing activities during the decade through the mid-1980s. could not develop. Moreover, by the mid-1980s, the

©International Monetary Fund. Not for Redistribution Origins of Economic Difficulties

Chart 2.3. Central Government Chart 2.4. Central Government Deficit, Operations Revenue, and Expenditure (In percent of GDP) (In percent of GDP)

Sources: IMF, International Financial Statistics, and various Recent Economic Development reports.

flow of foreign grants from traditional regional countries and inflows of workers' remittances started to decline in the aftermath of the oil price collapse. The authorities responded to these developments by (1) turning to external commercial and domestic bank borrowing to finance unsustainable levels of domestic aggregate demand and increasingly large budgetary deficits—with claims on the Central Government by the domestic banking system increasing by more than 300 percent during 1986-88; and (2) maintaining domestic price stability through control of administered prices. As a result, there was a rapid acceleration of the external debt-service burden, which led to the emergence of serious financial imbalances. Moreover, the sharply higher public sector debt-service payments further accentuated Jordan's medium-term fiscal imbalance. In response to an easing of the credit stance Sources: IMF, International Financial Statistics, and various Recent and a large devaluation, the inflation rate started to Economic Development reports. accelerate to double-digit levels by 1988. At about this time, commercial banks started to accumulate nonperforming assets, leading to bank failures by the late 1980s; and access to external borrowing virtually ceased by 1988. 1987, was depreciated by 21 percent in real effec- To address these difficulties, the authorities initi- tive terms in 1988. By that time, however, Jordan ated corrective macroeconomic policies, including a had become one of the most heavily indebted coun- large devaluation of the Jordan dinar (Chart 2.6): tries in the world, with its ratio of external debt to the exchange rate, which had remained fixed to the GDP increasing to 193 percent of GDP by 1990 SDR for more than twenty years through the end of (Chart 2.7).

©International Monetary Fund. Not for Redistribution II OVERVIEW

Chart 2.5. External Sector Chart 2.6. Exchange Rate Indices (In percent of GDP) (1980= 100)

Sources: IMF, International Financial Statistics,.and various Recent Economic Development reports.

Sources: IMF, Information Notice System, and International Financial Statistics.

Structural Weaknesses Jordan's economic performance was handicapped not only by its limited resource base but also by pol- In the agriculture sector, subsidized water and do- icy-induced structural weaknesses in various sectors. mestic producer price support contributed to the in- The tax system in Jordan—although effective in mo- efficient use of scarce water resources, while pro- bilizing resources for government operations—suf- duction controls on major vegetable crops prevented fered from inefficiencies resulting from cascading, higher production and export of high-value widespread exemptions, a multiplicity of taxes and products4 in which Jordan appeared to have some tax rates, and inefficient tax administration. On the comparative advantage. The energy sector also suf- expenditure side, high military expenditure, exten- fered from inadequate pricing policies for petroleum sive subsidy programs (covering basic foods, energy, products and electricity, institutional inefficiency in agricultural production and inputs, and transporta- the associated public sector enterprises, and the lim- tion) and a growing external debt-service burden ited role of the private sector in energy sector devel- complicated efforts to reduce the fiscal deficit to a opment projects.5 In the financial sector,6 the avail- sustainable level. ability of financial instruments was limited, and Moreover, Jordan's trade regime was character- instruments for indirect control had not been devel- ized, until the late 1980s, by a notionally high and oped, although in 1988 reserve ratios started to be complex tariff structure, nontariff barriers, wide- more actively used to absorb excess liquidity in the spread exemptions, and institutional inadequacies, banking system (generated in part by rapidly in- which adversely affected its export and industrial creasing bank borrowing by the Government). sector performance.3 4For more details on structural issues related to Jordan's agri- 3The maximum tariff rate was 318 percent, while, owing to culture sector, see Jordan: Towards an Agriculture Sector Strat- widespread exemptions, the bulk of imports were imported duty egy, Report No. 7547-50 (Washington: World Bank, 1990). free. Nearly thirty institutions accounting for 51 percent of im- 5For more details on structural issues related to Jordan's energy ports into Jordan were exempt from import duty. Nontariff barri- sector, see Jordan: Energy Sector Study, World Bank Report No. ers in the form of quantitative restrictions were also widespread; 7984-JO; Vols. 1 and 2 (Washington: World Bank, 1990). about 40 percent of imports were subject to quantitative restric- 6Four main direct monetary instruments were used: banks' re- tions in 1988. Combined with these nontariff barriers, the high serve requirements, the liquidity ratio, credit guidelines, and in- degree of variations in the tariff rates (between zero and 318 per- terest rate ceilings, with legal reserve ratios being kept low (effec- cent, with a standard deviation of 26.1) led to a high degree of tively in the range of 7—8 percent of local currency deposits) variation in effective rates of protection. during most of the period through 1987.

©International Monetary Fund. Not for Redistribution Adjustment and Structural Reform Efforts, 1989-94

lar, the expansions in the net domestic assets of the Chart 2.7. External Debt and Debt Service banking system, the trade deficit, and the external current account deficit were kept below their respec- tive program targets, allowing the Central Bank of Jordan's foreign exchange reserves to exceed the program target significantly in 1989. The budgetary situation also improved, although, owing to higher- than-expected interest payments on foreign debt, the overall budget deficit exceeded the program target. On the structural front, following the exchange rate depreciation in 1988, the authorities introduced a number of reforms in the exchange and trade systems and interest rate policy. These included, in particular, abandoning the dual exchange rate system, which they had introduced as a temporary emergency measure in July 1989 to address acute balance of payments difficulties; unifying the official and interbank rates at the level of the more depreciated interbank rate, leading to a further nominal effective depreciation of 8.5 percent; adopting flexible ex- Source: Jordanian authorities. change rate management, with the Jordan dinar pegged to a basket of currencies; abandoning the policy of supporting the interbank rate through intervention; freeing up interest rates, with both deposit Adjustment and Structural Reform and lending interest rates allowed to be freely deter- Efforts, 1989-94 mined by market forces; and phasing out nontariff barriers in stages and replacing them with tariffs ini- To address the rapidly growing imbalances, in late tially, and reducing tariff barriers as a whole subsequently, to lower effective protection and foster a 1988 and early 1989, the Government adopted a 7 medium-term growth-oriented adjustment program broadly neutral system of incentives. supported by the IMF and the World Bank, which re- The Government also adopted a strategy aimed at sulted in some progress toward the reduction of obtaining debt relief through the rescheduling and macroeconomic imbalances and the introduction of lengthening of the maturity structure of debt and re- structural reform measures in several areas. How- ducing the debt burden in relation to GDP in a ever, following the August 1990 regional crisis, real medium-term context (see Section VI for details). income declined; unemployment rose sharply—as a Consistent with these objectives, the Government result of the reflux of Jordanians working in neigh- (1) concluded a restructuring agreement with Paris boring countries; trade was disrupted; and aid flows Club official bilateral creditors in July 1989 and reg- from countries in the region ceased. These adverse ularized relations with other official bilateral credi- developments made it impossible for Jordan to tors along terms similar to those obtained from the achieve the program targets. Paris Club; (2) initiated negotiations with the com- When the regional crisis ended in early 1991, the mercial banks to obtain a multiyear restructuring of authorities re-evaluated and resumed their adjust- obligations falling due to banks, with provisions for ment and structural reform efforts. This led to the new money and options for debt conversion; (3) lim- adoption of a second medium-term adjustment and structural reform program beginning in 1992, supported by the IMF and the World Bank and other of- 7In August 1988, 30 categories of products had their import ficial donors and creditors. bans removed and replaced with tariff protection. The percentage of imports subject to quantitative restrictions was reduced from 40 percent in early 1988 to less than 7 percent in 1990. Import Performance During 1989-90 prohibition on 11 consumer luxury items, which were banned on a time-bound basis for balance of payments reasons, was elimi- Jordan's economic and financial performance dur- nated in January 1990. For the remaining 93 percent of the items ing 1989-90 was encouraging. All the policy actions subject to tariffs, the range between the maximum and minimum contemplated in the authorities' program were im- tariffs was reduced to 55 percent at a first stage, by lowering the maximum tariff rate to 60 percent and increasing the minimum plemented, which, in combination with debt relief tariff rate to 5 percent. This narrowing of the range reduced the and bilateral grants, enabled Jordan to meet all its variation in tariffs (measured in terms of the standard deviation) quantitative performance criteria in 1989. In particu- to 17.5 in 1990 from 26.1 in 1988.

©International Monetary Fund. Not for Redistribution II OVERVIEW

ited all new borrowings to longer-term maturities solid recovery in the agriculture, trade, and manu- and, in many instances, at concessional interest facturing sectors. A favorable supply response and rates; and (4) canceled most new commercial bor- import prices, coupled with prudent demand-man- rowings that were in the pipeline. agement policies, contributed to a sharp deceleration The authorities implemented various structural of inflation to the 4-5 percent range during this pe- measures in 1990, including tariff reforms and fur- riod, much below the program targets. The unem- ther interest rate liberalization, and pursued flexible ployment rate declined to 12-15 percent during interest and exchange rate policies. However, a num- 1993-94 from a peak of 25 percent in 1990, despite ber of adverse developments during 1990, including, high labor force growth. in particular, the regional crisis, pushed the economy The overall fiscal deficit (excluding foreign off track. grants) declined from nearly 18 percent of GDP in 1991 to less than 4 percent of GDP in 1992, after taking into account the effect of nonrecurrent rev- Performance Since 1992 enue sources; this result compares with the program The broad objectives of the authorities' programs target of about 14 percent of GDP. This sharp reduc- since 1992 are to reduce the macroeconomic imbal- tion in the fiscal deficit reflected the effect of discre- ances significantly; continue the process of struc- tionary fiscal measures (2.5 percent of GDP) tural reform, despite considerable uncertainties adopted in the context of the 1992 budget; strict ex- about trade and aid relations between Jordan and its penditure control; and buoyant revenue collection, neighbors; and achieve balance of payments viabil- including nonrecurring revenues (3.5 percent of ity at an early date while attaining satisfactory GDP). When the effect of nonrecurring revenues is growth performance in the context of stable domes- excluded, the underlying fiscal deficit is estimated to tic prices and increasing the role of the private sector have declined to about 7 percent of GDP. The au- in the economy. thorities' fiscal policy stance was further tightened The key policy elements of the program are the during 1993-94, with the underlying fiscal deficit following: (excluding nonrecurring factors) declining to less • Reduce the fiscal deficit substantially, through than 6 percent of GDP in 1994 as a result of further revenue enhancements and containment of cur- discretionary measures and strong economic growth. rent expenditures; The favorable fiscal outturn and cautious credit • Contain credit expansion consistent with the ex- policy stance contained the growth of the net domes- ternal and inflation objectives and with achiev- tic assets of the banking system to less than 8 percent in 1992, below the program target of 8.5 per- ing interest rates that are positive in real terms; 8 • Maintain external competitiveness; cent. Supported by the improved fiscal position, • Carry forward the structural reforms initiated in monetary expansion remained well below the in- 1989, including the second phase of tariff reform crease in nominal GDP during 1992-94. and the reform of the indirect tax system through The external current account deficit declined to the introduction of a general sales tax; and some 14 percent of GDP (excluding workers' sav- • Pursue reforms in the agriculture and water and ings and grants from the Gulf Crisis Financial Coor- energy sectors for implementation over 1992-94. dination Group (GCFCG)) in 1992, although imports were much higher than programmed because of the economic expansion. The deficit declined by Macroeconomic Stabilization and Outcome an additional 3 percentage points to less than 12 per- Jordan's macroeconomic performance during cent of GDP in 1993 and thereafter by a further 1992-94 was solid. The economy's real growth, in- 5 percentage points to 6.5 percent of GDP in 1994. flation, and fiscal adjustment performance were This outcome was attributable to the continuing much better than anticipated at the inception of the strong performance of remittances and nontradi- program; and despite a surge in imports associated tional exports and to the strong recovery of tourism with the large influx of returnees, the balance of pay- receipts. These receipts largely offset the shortfall in ments adjustment has been stronger than projected. traditional mineral-based exports caused by prob- After several years of decline or virtual stagnation, lems in shipping phosphate rocks under the UN- economic activity gained momentum in 1992. Real imposed sanctions, the loss of markets, and the col- GDP growth (at market prices) exceeded 16 percent lapse in world market prices, as well as the higher in 1992 (with real GDP growth at factor cost at 12 growth of imports associated with buoyant eco- percent), against a program target of 3 percent; and growth averaged 6 percent a year during 1993-94. This turnaround is largely attributable to strong ac- 8For more details on monetary developments and the corrective tivity in construction, but was also facilitated by a measures, see Section V.

©International Monetary Fund. Not for Redistribution Adjustment and Structural Reform Efforts, 1989-94 nomic activity in 1992. Nontraditional exports were surcharge and other fees, border taxes were reduced strong throughout the period. to 70 percent or less, and the effective tariff rate was With these developments and large transfers of reduced to 21 percent. In the agriculture and water workers' savings (primarily repatriation of foreign sectors, control of cropping patterns was abolished, financial assets of Jordanians who had previously and the average tariff on agricultural imports was re- worked abroad), the overall balance of payments duced to 30 percent, as part of the reform effort sup- deficits were sharply reduced in both 1992 and 1993, ported by the World Bank Agriculture Sector Adjust- and gross official reserves (that is, the Central ment Loan (ASAL). Bank's foreign exchange reserves) exceeded the pro- Under the ongoing reform of the energy sector op- gram targets in both years. Rescheduling of debt ser- erations supported by the World Bank Energy Sector vice and overdue obligations to Paris Club official Adjustment Loan (ESAL), the authorities are seek- bilateral creditors in February 1992 and receipts of ing to achieve certain financial targets in the opera- exceptional assistance from the GCFCG signifi- tions of the Jordan Electricity Authority (JEA) and cantly contributed to the balance of payments fi- to formulate a detailed financial and institutional re- nancing during 1992-93. However, uncertainties in structuring plan for the power sector, including the the region stemming from the peace process and the incorporation of the JEA and the establishment of a potential redemption of Jordan dinars circulating in new oil and gas company. Furthermore, as part of its the West Bank caused gross official reserves to de- public sector enterprise reform and privatization cline sharply in the first half of 1994. They improved program, the Government has sold a large part of its significantly in the second half of 1994, in response shares in the Jordan Hotels and Tourism Company to a substantial tightening of the stance of monetary and completed the commercialization of the Alia policy, but remained below the program target by the Gateway Hotels and duty-free shops at the Amman end of 1994. International Airport. The Telecommunications Cor- Jordan also made significant progress in normaliz- poration (TCC) is allowing the private sector to pro- ing payments relations with its external creditors and vide complementary services, and a strategy to pri- in reducing outstanding payments arrears. In addi- vatize the TCC is expected to be completed by the tion, the authorities took steps to liberalize the ex- end of 1995 under the World Bank Telecommunica- change system by relaxing restrictions on foreign tions Project. exchange for current account transfers. The annual In September 1993, the authorities started to issue amount of foreign exchange that residents and non- central bank certificates of deposit (CDs) denomi- residents can transfer abroad was gradually in- nated in Jordan dinars as a new indirect instrument creased, to JD 35,000 (equivalent to $50,000), by of monetary control. Since then, there has been a early 1994; and the Central Bank abolished the regu- steady strengthening of indirect monetary control lation under which export proceeds had to be repatri- operations and other supporting reforms, with the ated within a specified period. Central Bank implementing monetary policy through a reserve money program and using an indirect monetary control mechanism to effectively in- Structural Reforms fluence domestic interest rates, without resorting to Under its medium-term economic reform pro- ad hoc and arbitrary use of credit limits. Commercial gram, Jordan intended, in particular, to (1) proceed banks have been responding efficiently to the inter- with structural reforms initiated in late 1988, specifi- est rate policy the Central Bank is pursuing through cally, tariff reform, reform of the indirect tax sys- open market operations. tem—through the introduction of a general sales In June 1994, to broaden the domestic tax base tax—and the direct tax system; (2) introduce indirect and improve the elasticity and efficiency of the tax monetary control and a number of other financial re- system, the authorities replaced the consumption tax form steps; (3) move to external current account system with a general sales tax at the manufacturing convertibility; (4) implement a number of reforms in and import stages. Its main features are a basic rate the agriculture and water and energy sectors; and of 7 percent, with a rate of 20 percent applying to 30 (5) improve the regulatory framework for domestic items (mainly luxury goods); a tax credit for inputs and foreign investments. to remove cascading of taxes at different stages of The tariff reform program initiated in 1988 was production; and a limited number of exemptions. concluded in early 1992, with a further reduction of Progress was also made in liberalizing the ex- the maximum tariff rate (by 10 percentage points) to change system. In particular, following several steps 50 percent (except for a few luxury items). Further to liberalize current account restrictions, in February progress was made in late 1994: basic tariffs that ex- 1995 Jordan accepted the obligations of Article VIII, ceeded 50 percent were reduced to 50 percent or less Sections 2, 3, and 4 of the IMF's Articles of Agree- (except on three items); as a result, together with the ment (see Section VII).

©International Monetary Fund. Not for Redistribution II OVERVIEW

Further progress in structural reform was achieved (mining, industry, hotels, and hospitals), 35 percent with the recent passage of additional reform mea- for banks and financial institutions, and 25 percent sures and their implementation by the time of the for all other companies; encouraging capital accu- Amman economic summit in late October 1995. The mulation by imposing a withholding tax of 10 per- package consists of several amendments to the gen- cent on distributed profits; and broadening the tax eral tax system, a reform of corporate and personal base by reducing and simplifying exemptions and income taxes, and a new investment law. applying uniform standard deductions for all wage The measures to reform the general sales tax in- earners. clude increasing the basic rate to 10 percent; allow- The new Investment Law, which replaces both the ing tax rebates for inputs to production if the final Encouragement and Investment Law of 1978 and the good is exported tax exempt; extending the sales tax Law Governing Arab and Foreign Investment of to services according to an extended positive list; al- 1992, provides for equal treatment of all investors, lowing taxpayers below a certain threshold to register eliminates the distinction between projects in the voluntarily; and introducing a supplementary duty on same sector, rationalizes the incentive structure, and selected luxury products, alcohol, and tobacco. opens the financial market to all investors. The authorities improved the direct tax system by The package is expected to be complemented later eliminating tax holidays (except for investments in this year by a law governing the operations of the less developed areas); limiting tax deductibility to Amman Financial Market (AFM), further measures net interest payments; reducing the number of tax to deepen the financial markets, and a further tariff rates, including maximum tax rates, for both per- reform. Overall, the reforms that have already been sonal and corporate income taxes from 50 percent to adopted or are under consideration are consistent 35 percent; rationalizing corporate income tax rates, with the authorities' medium-term private sector and with a view to treating all corporate sectors equally export-led strategy and should enable Jordan to ben- by establishing three flat corporate tax rates of efit from prospective regional developments and 15 percent for companies in "encouraged" sectors new trade and financial opportunities.

©International Monetary Fund. Not for Redistribution Ill Macroeconomic Environment and Factors Underlying Growth and Investment

Christopher McDermott

his section provides background information on transportation and mineral-based processing sectors; Thow Jordan's adjustment and reform programs such investments therefore had limited direct impact and the resulting improved macroeconomic policy on the future growth potential of the economy. Jor- environment have influenced investment and growth dan's growth and investment depended heavily on in recent years. The basic indicators of the macro- external financing. Private construction activity was economic environment that are used to examine the largely financed through workers' remittances, while potential effect of adjustment programs on invest- grants and loans from oil-exporting countries in the ment and growth are the inflation rate, the fiscal bal- region accounted for a large part of the financing of ance, and the external current account balance. Jordan's public sector investment. Movements in the terms of the trade, the inflow of The indicators of macroeconomic policy describe workers' remittances, and changes in the real ex- a mixed situation during 1976-80: inflation aver- change rate are basic indicators of external shocks aged about 12 percent, and the fiscal deficit of the affecting growth and investment. Central Government (excluding foreign grants) av- The empirical analysis presented in this section eraged 29 percent of GDP; however, the external and inferences from other empirical works suggest a current account—reflecting receipts of sizable for- number of conclusions. First, real GDP growth is at- eign grants—remained virtually in balance. The tributable more to increases in physical capital than large budget deficits were the result of the authori- to increases in the labor force. Second, despite im- ties' efforts to improve public utilities and services, pressive growth performance, total factor productiv- together with higher defense expenditure and subsi- ity has declined in recent years because of the need dies for certain basic goods. to absorb into the economy some 300,000 Jordani- ans returning from abroad. Third, the most signifi- Economic Slowdown in the Early 1980s cant impact on real GDP growth has been achieved through fiscal policy; accordingly, the negative im- By the early 1980s, the terms of trade had deteri- pact of the high fiscal deficits of the past has been orated, resulting in the subsequent deterioration in declining with the progress made in reducing the the external current account balance to a deficit of deficit, despite falling foreign grants. Fourth, fiscal some 5 percent of GDP. Furthermore, as oil prices policy has affected real GDP growth primarily by and related revenues started to decline, the level of slowing down the rate of capital accumulation by the external aid (primarily from regional oil-producing private sector. Finally, the most important determi- countries) dropped from about 17 percent of GDP to nant of private investment in Jordan is the inflow of 10 percent of GDP, which in turn led to a reduction external financing in the form of workers' remit- in public sector investment as a share of GDP from tances and transfers of savings. nearly 19 percent of GDP to less than 14 percent (Table 3.2). While public sector investment bore the brunt of the shock, private sector investment re- Economic Developments During mained steady, at approximately 16 percent of GDP. 1976-94 The decline in investment and inflow of workers' remittances adversely affected the economy, with real GDP growth falling to an average of 6 percent Boom of the 1970s during 1981-85. The fiscal deficit (excluding for- Between 1976 and 1980, real GDP rose by eign grants) declined by more than 10 percentage 9.5 percent a year, and investment averaged 35 per- points, although at 18.5 percent of GDP by 1985 it cent of GDP (Table 3.1). During this period, domes- remained very high; the reduction in the fiscal tic private sector investment was directed partly to deficit was largely attributable to cuts in develop- residential housing construction and partly to the ment expenditure. Inflation was contained to an av-

©International Monetary Fund. Not for Redistribution Ill MACROECONOMIC ENVIRONMENT

Table 3.1. Macroeconomic Performance

1976-80 1981-85 1986-90 1991-94 1976-94

Real GDP growth1 9.5 6.2 -0.9 7.4 5.4 Gross fixed investment2 35.0 31.0 22.3 28.7 111 Inflation rate1 11.7 5.4 9.7 4.8 7.8 Fiscal balance2 Excluding grants -29.0 -18.5 -20.1 -10.8 -20.0 Including grants -12.4 -8.5 -12.8 -6.0 -10.1 Current account balance3 0.2 -5.2 -3.3 -12.5 -4.8 Terms of trade1 (1992= 100) 85.3 80.1 91.9 98.8 87.8 Real exchange rate1 (1992= 100) 171.4 142.7 117.3 99.9 135.6

Sources: Department of Statistics, Government of Jordan; Central Bank of Jordan; and IMF staff estimates. 'Period average. 2ln percent of GDP. 3ln percent of GDP, excluding foreign grants.

erage annual rate of 4 percent, reflecting the author- inflation. Jordan's economy also remained highly ities' policy of maintaining a stable nominal ex- vulnerable to the unfavorable fluctuations in income change rate and a moderation in the world rate of in neighboring countries as a result of declining oil inflation. prices and associated revenues. The decline in workers' remittances and grants and loans from Economic Crisis in the Late 1980s neighboring oil-exporting countries led to a growing external imbalance and an associated rapid In the second half of the 1980s, debt-service buildup of debt and debt service. By 1988, this ex- obligations and the fiscal imbalance increased ternal imbalance had reached unsustainable propor- sharply. Jordan's debt service was some 45 percent tions. General financial instability in this period in of exports of goods and services in 1989, while the turn led to bank failures and a reduction in private average fiscal deficit (excluding foreign grants) in- capital inflows. Reflecting this deteriorating macro- creased to 20 percent of GDP. Faced with growing economic environment, aggregate investment de- budget deficits, the authorities resorted to borrow- clined to some 22 percent of GDP, with private sec- ing from both domestic and foreign commercial tor investment averaging only 7 percent of GDP. banks. The monetization of the deficit led to higher Real GDP declined steadily over 1986-90, with Jor-

Table 3.2. External Financing and Investment (In percent of GDP)

Central Government Central Private Workers' Capital Government Period Investment1 Remittances Expenditure External Aid

1976-80 16.0 20.7 18.9 16.6 1981-85 16.2 22.3 13.9 10.0 1986-90 7.1 14.2 15.3 7.3 1991-93 16.3 16.3 13.1 4.8 1976-94 13.7 18.5 15.4 9.9

Sources: Department of Statistics, Government of Jordan; Central Bank of Jordan; and IMF staff estimates. 'Private investment is constructed from total gross fixed investment less central government capital expenditure.

©International Monetary Fund. Not for Redistribution Factors Influencing Long-Term Growth

Chart 3.1. Real GDP Growth and Chart 3.2. Real GDP Growth, Fiscal Investment Balance, and the Current Account (Percent changes) (Percent changes)

Sources: Department of Statistics, Government of Jordan; and Sources: Department of Statistics, Government of Jordan; and Central Bank of Jordan. Central Bank of Jordan.

dan's worst recession in recent history occurring in effective absorption into the Jordanian economy of 1989 (Chart 3.1). some 300,000 Jordanians from abroad (about 10 percent of the population and a higher proportion of the Economic Recovery, 1989-94 labor force), who also brought with them substantial savings and technical know-how to revitalize the Since late 1988, interrupted only by the August private sector. 1990 regional crisis, Jordan has been implementing a medium-term structural adjustment program. Since 1991-92, the macroeconomic policy indicators have Factors Influencing Long-Term Growth improved significantly. The inflation rate declined to 3.5 percent in 1994, and the average rate of inflation Recent developments in the growth literature have for 1989-94 was less than 5 percent. The fiscal emphasized that economic conditions and implemen- deficit declined sharply, averaging about 11 percent tation of appropriate public policies influence the rate of GDP during this period (Chart 3.2). Including for- of economic growth. Barro and Barro and Sala-i- eign grants and the effect of external debt relief, cen- Martin1 link growth to fiscal variables in their en- tral government operations on a cash basis were in dogenous growth models, which were tested empiri- surplus during most of the years. The improvement cally by Cashin2 for member countries of the in macroeconomic indicators also led to a recovery Organization for Economic Cooperation and Devel- in investment to about 29 percent of GDP, despite a opment (OECD). Fischer3 extends the notion that significant decline in external official financing compared with earlier periods; and real GDP ex- 'R.J. Barro, "Government Spending in a Simple Model of En- panded by more than 7 percent a year. The strength- dogenous Growth," Journal of Political Economy, Vol. 98 (Octo- ening of Jordan's macroeconomic performance since ber 1990), pp. SI03-25; and R.J. Barro and X. Sala-i-Martin, late 1991 is attributable to three key factors: (1) the "Public Finance in Models of Economic Growth," Review of Eco- nomic Studies, Vol. 59 (October 1992), pp. 645-61. implementation of effective macroeconomic poli- 2 cies, such as restrained fiscal and monetary policy Paul Cashin, "Government Spending, Taxes, and Economic Growth," Staff Papers, International Monetary Fund, Vol. 42 and prudent exchange rate management; (2) the reg- (June 1995), pp. 237-69. ularization of external debt-service obligations, a 3Stanley Fischer, "Development, Macroeconomics, and prudent debt-management policy that contributed to Growth," NBER Macroeconomics Annual, 1991 (Cambridge, a sharp reduction in the debt and debt-service ratios, Massachusetts: National Bureau of Economic Research, 1991), pp. 329—64; and Stanley Fischer, "The Role of Macroeconomic and substantial external financial support from offi- Factors in Growth," Journal of Monetary Economics, Vol. 32 cial bilateral and multilateral creditors; and (3) the (December 1993), pp. 485-512.

©International Monetary Fund. Not for Redistribution Ill MACROECONOMIC ENVIRONMENT

Table 3.3. Summary of Estimated Panel Regression Coefficients from a Growth Accounting Framework

Regressors Changes in Black market Dependent Inflation Fiscal the terms exchange rate Variable rate balance1 of trade premium

Real GDP growth -0.039 0.228 0.043 -0.017 (-4.65) (-4.49) (-2.71) (-2.76)

Productivity growth -0.018 0.137 0.038 -0.006 (-2.49) (-3.23) (-2.60) (-1.17)

Rate of capital accumulation -0.037 0.075 0.008 -0.019 H-77) (-1.61) (-0.62) (-3.56) Labor force growth -0.002 -0.007 0.0009 0.0003 (-1.14) (-0.53) (-0.29) (-0.22)

Source: Stanley Fischer, "The Role of Macroeconomic Factors in Growth," Journal of Monetary Economicsyo\. 32 (December 1993), pp. 485-512. Note: Twenty-two countries are in the panel: Argentina, Chile, Colombia, Cote d'lvoire, Dominican Republic, Ecuador, Ghana, Greece, India, Indonesia, Jamaica, Kenya, Korea, Malawi, Mexico, Morocco, Pakistan, Paraguay.Thai- land,Turkey,Venezuela, and Zambia. Figures in parentheses are t-statistics. 'In percent of GDP.

governments can influence growth by creating a sta- Macroeconomic Stability and Growth ble macroeconomic framework, which can be created In examining the links between growth and indi- through macroeconomic policies that are conducive 4 cators of macroeconomic stability, Fischer estimated to growth. The key macroeconomic policies are con- several sets of panel regressions. All of the coeffi- sidered to be monetary, fiscal, and exchange rate cients from the estimated equation of the growth rate policies designed to keep inflation low and pre- of real GDP from 1961 to 1988 on a set of macro- dictable, a stable and sustainable fiscal balance, ap- economic indicators and external shock indicators propriate real interest rates, a competitive and pre- are significantly different from zero (Table 3.3). The dictable real exchange rate, and a viable balance of results imply notably that (1) a country that has an payments. The signaling effect government manage- inflation rate 100 percentage points higher than an- ment has on the private sector is one of the key mech- other will have a growth rate that is 3.9 percentage anisms through which macroeconomic policies influ- points lower; (2) a country with a fiscal balance that ence growth. To measure this signaling effect, is in greater surplus or lower deficit by 1 percent of Fischer recommends using basic indicators of macro- GDP will have a growth rate that is 0.23 percent economic policy, such as the inflation rate, the gov- larger; (3) a country with a higher black market ex- ernment fiscal balance, and the current account bal- change premium will grow slower than a country ance. In particular, Fischer used a regression analog with a lower black market exchange premium, other of growth accounting to identify the channels through things being equal; and (4) negative shocks in the which macroeconomic variables affect economic terms of trade reduce growth. Overall, the results in- growth; the idea behind this growth accounting dicate that, when the measures of macroeconomic framework is to measure the relationship between policies indicate stability, the prospects for growth growth and several key macroeconomic variables. are favorable. Growth can be attributed to changes in the supplies of factors of production (labor and capital) or to changes These findings are used to examine the effects of in total factor productivity (reflecting changes in the macroeconomic policies on Jordan's growth perfor- efficiency with which these factors are used). mance during 1976-94 (Table 3.4). First, the negative impact of inflation on growth for Jordan is about a third of a percentage point a year during 1976-94. However, most of the impact came during 1976-80 4Other public policies conducive to growth include the enact- ment and enforcement of private property rights through the legal and 1986-90. Had Jordan been able to maintain system and the maintenance of national security. price stability in those two periods, the growth rates

©International Monetary Fund. Not for Redistribution Factors Influencing Long-Term Growth

Table 3.4. Impact of Macroeconomic Policy on Real GDP Growth (Percent change)

1976-80 1981-85 1986-90 1991-94 1976-94

Policy indicators Fiscal balance Excluding grants -6.61 -4.22 -4.58 -2.46 -4.56 Including grants -2.83 -1.94 -2.92 -1.37 -2.30 Inflation rate -0.46 -0.21 -0.38 -0.19 -0.30 Changes in terms of trade -0.49 0.16 0.25 0.44 0.38

Source: IMF staff estimates. Note:The impact on real GDP growth is calculated by multiplying the period mean of the policy indicator (see Table 3.1) by the relevant estimated coefficient in Table 3.3.

would have been nearly half a percentage point growth. The approach is a regression analog of higher. However, even in the slump years, the re- growth accounting and allows for identification of straining effect of inflation on growth is signifi- the channels through which macroeconomic policies cantly less in Jordan than in the other countries in affect economic growth. Growth can arise through the sample. This reflects the fact that Jordan experi- increased efficiency, which shows up as increased enced relatively moderate rates of inflation over the positive residuals in an estimated production func- entire period. Second, the changes in the terms of tion, or through increases in factors of production. trade had a limited and mostly positive impact on For this exercise, two specific types of production Jordan's growth performance. function have been specified and estimated. The first This analysis shows that the stance of fiscal policy one is a standard Cobb-Douglas production function, represents the greatest impediment to Jordan's in which output is a function of labor and physical growth performance. Over the entire period capital, while the second also includes a measure of 1976-94, the fiscal deficit (including foreign grants) human capital—measured by school enrollment—as restrained growth by an average of over 2 percent- an additional factor of production. The restriction age points a year. Had there been no foreign that the two factors' shares in the Cobb-Douglas pro- grants—which have been used largely to finance duction function are equal and satisfy constant republic investment—growth would have been returns to scale cannot be rejected using a standard hy- duced by nearly 5 percentage points. Jordan's re- pothesis test. Based on the extended production liance on grants from oil-exporting countries in the function, the estimated factor shares are 44 percent region to finance public sector investment—a signif- for physical capital, 23 percent for labor, and 33 per- icant source of growth—diminished over the period. cent for human capital. For the first type of produc- If the grants received during 1976-80 had not mate- tion function, total factor productivity estimates, rialized, growth would have been reduced by an average of nearly 7 percentage points. However, if the Solow residuals = real GDP growth - 0.5 physical grants received during 1991-94 had not material- capital accumulation - 0.5 labor force growth; ized, growth would have been slower by an average of just over 2 percentage points. and for the second set of production function, total factor productivity estimates,

Total Factor Productivity Residuals Mankiw-Romer-Weil (MRW) residuals = real GDP growth - 0.44 physical capital The World Bank conducted a study of sustainable accumulation - 0.23 labor force growth growth in Jordan that included an analysis of total - 0.33 human capital. factor productivity.5 The approach uses a production function to examine the relative contribution to Both the Solow residuals and the MRW residuals growth of factors of production and productivity show that total factor productivity declined over the period 1981-94, with the largest decline occurring 5 Jordan: Consolidating Economic Adjustment and Establish- during the period 1991-94 (Table 3.5). Estimates ing the Base for Sustainable Growth, Report No. 12645-JO from the simple two-factor production function (Washington: World Bank, 1994). show that the growth component provided by extra

©International Monetary Fund. Not for Redistribution Ill MACROECONOMIC ENVIRONMENT

Table 3.5. Contributions to Economic Growth by Factors of Production (Period averages; percent change in real terms)

Growth Contribution of

Physical Labor Human Solow MRW Period GDP capital force capital Residuals1 Residuals1

1976-80 9.5 6.5 1.3 1.7 1981-85 6.2 7.0 2.7 -3.5 1986-90 -0.9 0.3 0.3 -1.5 1991-94 7.9 10.2 11.9 -14.2

1976-94 5.4 5.5 3.2 -3.3

1976-80 9.5 5.7 0.6 1.2 2.0 1981-85 6.2 6.1 1.3 1.2 -2.4 I986-872 5.0 3.1 I.I 0.9 -0.1

1976-872 7.4 5.4 1.0 1.2 -0.2

Source: IMF staff estimates based on World Bank, Jordan: Consolidating Economic Adjustment 'A measure of total factor productivity changes; see text for definition. 2Data available only until 1987.

labor is more than offset by the decline in total factor function of (1) private external inflows measured by productivity, implying that increased output from a workers' remittances; (2) a public investment financ- higher labor force was virtually offset by a loss of ing constraint measured by the excess of central productivity. However, estimates from the three-fac- government capital expenditure over the Central tor production function, which allows decomposi- Government's available external financing; (3) a fis- tion between an increase in the quantity of labor and cal policy indicator, measured by the fiscal deficit the quality of labor, show that for an increase in eco- excluding foreign grants; (4) an indicator of the sus- nomic activity attributable to labor, about 50 percent tainability of financial policies measured by gross can be attributed to an increase in quality. A striking official reserves; and (5) an indicator of macroeco- feature of the results is that most of the growth in nomic uncertainty, measured by changes in the real economic activity was generated by increases in exchange rate. The rationale for using external in- physical capital. The manner in which macroeco- flows, a government financing constraint, crowding- nomic policy affects investment and, thus, overall out effects, and uncertainty over policies and macroeconomic growth is examined below. economic conditions as key factors influencing investment in Jordan is as follows. First, in many countries an important element in Factors Influencing Investment investment decisions is the availability of credit for financing the investment. However, for Jordan, in- In view of the importance of physical capital investment financing mainly took the form of external vestment in generating growth in Jordan, this sub- inflows of private capital—particularly through the section conducts a detailed time-series study of fac- transfers of workers' remittances and savings. Ac- tors influencing private investment in Jordan. These cordingly, private investment in Jordan fluctuated factors include some of the policy indicators that with the flows of workers' remittances, which in turn were shown to be important determinants of growth depended heavily on movements in oil prices in the in the previous subsection, as well as some specific region's oil-exporting countries. variables found to be empirically important for pri- Second, the possibility that public investment and vate sector investment behavior in Jordan. fiscal policy may in general crowd out private investment—as discussed in Easterly, Rodriguez, and 6 Methodology and Data Schmidt-Hebbel —arises from government prefer-

The analysis is based on empirical estimates of a 6William R. Easterly, Carlos A. Rodriguez, and Klaus Schmidt- private investment function for Jordan during Hebbel, eds., Public Sector Deficits and Macro economic Perfor- 1977-94. Private investment is specified to be a mance (New York: Oxford University Press, 1994).

©International Monetary Fund. Not for Redistribution Factors Influencing Investment

Table 3.6. Estimates of Private Investment Equation, 1977-94 (Dependent variable: private investment/GDP)'

Independent Coefficient Standard Variables Estimate Errors2 t-statistic

Constant 4.16 3.34 1.25 Workers' remittances/GDP 0.85 0.23 3.64 Government financing constraint3 -0.81 0.18 -4.49 Official reserves (in months of imports) 1.94 0.53 3.68 Real exchange rate (percent change) -0.23 0.10 -2.25 Fiscal balance/GDP4 0.50 0.12 4.22

Summary and diagnostic statistics5 Number of observations 18.00 R squared 0.71 Durbin-Watson statistic 1.38 Breusch-GodfreyTest for AR(I) errors6 1.77 (0.18) Breusch-Pagan Test for heteroscedasticity6 From workers' remittances 4.91 (0.03) From government financing constraint 1.18 (0.28) EngleTest for ARCH errors6 1.11 (0.29)

Sources: IMF staff estimates based on data provided by the Central Bank of Jordan and the Department of Sta- tistics, Government of Jordan. 'Private investment was constructed from gross fixed investment less central government capital expenditure. 2Standard errors were computed using White's heteroscedasticity-consistent variance-covariance matrix estimator. 3The government financing constraint is defined as central government capital expenditure in relation to GDP less central government external financing in relation to GDP. 4Excluding foreign grants. 5p-values shown in parentheses. 6X2 test with one degree of freedom.

ential access to credit at administered interest rates; ered with caution, given the low degree of freedom or increasing lending rates resulting from a higher afforded by the small sample available.7 absorption of funds by the public sector; or lower private rates of return because of competition be- Empirical Observations tween the Government and the private sector for investment opportunities. In Jordan, such potential The estimates suggest a very strong relationship crowding out is represented by the excess of central between the availability of remittance income and government capital outlays over available external private sector investment. The marginal propensity financing. to invest from workers' remittances is estimated to Third, uncertainty about policy and the general be 0.85 and is statistically highly significant (Table macroeconomic environment gives investors the in- 3.6). This finding shows the importance of external centive to wait until the uncertainty has been refinancing for Jordan: when the flow of remittances moved before committing themselves to an irre- dries up—as happened in the late 1980s and early versible decision. Uncertainty is not a measurable entity, but it is assumed that investors use some indicators to gauge its level. In the case of Jordan, two such indicators are likely to be the level of gross of- 7A11 the series entering the investment equation were expressed in scaled form (as a ratio of GDP or imports) or in rates of change ficial reserves and the real exchange rate. to obtain a meaningful economic interpretation free of spurious The data used to estimate the private investment correlations caused by unrelated trends in the data. Because a bat- equation are all annual time series covering tery of diagnostic tests indicated the possible presence of het- 1977-94, making a total of only 18 observations. eroscedasticity, all standard errors are calculated using a het- Accordingly, while this econometric exercise pro- eroscedasticity-consistent variance-covariance matrix estimator. The diagnostic procedures do not indicate any other problems in vides some interesting information, any conclusions the estimation, and the estimated model explains 71 percent of derived from it are tentative and should be consid- the movements in private investment.

©International Monetary Fund. Not for Redistribution Ill MACROECONOMIC ENVIRONMENT

1990s—private investment is constrained and shows that there is a large crowding-out effect and growth is inhibited. that large fiscal deficits are the most important indi- An equally strong but inverse relationship was cator of macroeconomic policy performance. A lack found between private investment and the excess of of fiscal discipline has been a major impediment to public investment over available foreign grants. The private investment in Jordan in the past. These find- estimated coefficient on this variable was -0.81, ings are consistent with the strategy underpinning nearly equal but of opposite sign to the estimated the emphasis on fiscal adjustment and the gross offi- coefficient on workers' remittances. Furthermore, an cial reserve buildup in Jordan's recent structural ad- F-test of whether the estimated coefficients of the justment programs. overall financing flows (workers' remittances plus the financing constraint) sum to zero is also statistically acceptable. The coefficient estimates and tests Concluding Observations show that there is an exact one-to-one correspon- dence of public to private investment. Virtually all The empirical analysis presented above shows Jordanian investment is financed through external that there is a strong relationship between the macro- sources. If the Government invests more than its economic environment and growth in Jordan. The own internal public savings, then it must use financ- basic indicators of the macroeconomic environment ing that would have been available to the private are found to be the inflation rate, the fiscal balance, sector. The size of the financing constraint means and the external current account balance, with the that the cost of public investment is the forgone op- fiscal balance having the greatest influence on real portunity in the private sector and that this cost will GDP growth. Furthermore, real GDP growth is at- be fully reflected. tributable more to increases in physical capital than The estimated coefficients on gross official re- to increases in the labor force, highlighting the role serves and the real exchange rate show that these in- of savings in achieving higher growth. An analysis dicators of uncertainty provide a strong signal to of private investment shows that the inflow of exter- investors about the health of the economy, and this nal financing in the form of workers' remittances in turn has a large impact on investment decisions. and transfers of savings is the most important chan- The estimated coefficient on official reserves is 1.94 nel for private investment in Jordan. Fiscal deficits and is statistically highly significant. The implication have adversely affected growth by crowding out pri- is that an improved reserves position would clearly vate investment. In this sense, Jordan's remarkable enhance the country's prospects for private invest- fiscal adjustment in recent years is the single most ment. The estimated coefficient on the real exchange important macroeconomic policy instrument that has rate is -0.23 and is significantly different from zero. favorably influenced its investment and growth per- The implication is that a devaluation is considered a formance. Fiscal policy seems to have worked sign of a weak economy and thus deters private in- through two channels: the crowding-out effect on investment, while maintaining real exchange rate sta- vestment and thus growth; and the effect of signal- bility would encourage confidence and thus invest- ing whether or not the Government is in control of ment. These results indicate that Jordan's continued the economy. Overall, the empirical analysis high- vulnerability to external shocks and the related per- lights the importance of sustained fiscal and external ception of uncertainty in the macroeconomic envi- adjustments in creating a stable macroeconomic en- ronment tend to discourage domestic investment. vironment and in fostering growth and investment The estimated coefficient on the fiscal balance is and the importance of large transfers of workers' 0.5 and is highly significant. The parameter estimate savings in easing the adjustment process.

©International Monetary Fund. Not for Redistribution IV Public Debt Dynamics and Fiscal Policy

Etienne de Callatay and Ahsan Mansur

he public sector in Jordan has traditionally been Tvery large, with central government expendi- Chart 4.1. Fiscal Balance tures averaging about 43 percent of GDP during (In percent of GDP) 1972-94 (Chart 4.1). The public sector also encom- passes local authorities, decentralized agencies, financial and nonfinancial public enterprises, and, until 1991, several extrabudgetary funds. Although no quantitative estimation is available for the operations channeled through these public sector entities, they are believed to be of significant magnitude. Purchases of military equipment through the De- fense Fund, financed by foreign assistance, are not included in the central government accounts. Be- cause of a lack of adequate data, the analysis in the following section will concentrate on the budgetary operations of the Central Government. Although foreign grants from neighboring oil- exporting Arab countries were very large during 1975-88, accounting for about 12 percent of GDP, Sources: IMF, International Financial Statistics, and various Recent recourse to foreign and domestic debt financing was Economic Development reports. also large, averaging about 10 percent of GDP a year during the same period (Table 4.1). In the second half of the 1980s, with rising interest rates and later the depreciation of the Jordan dinar, fiscal policy turned out to be unsustainable, requiring an urgent in the region, Jordan allocated substantial resources and comprehensive fiscal consolidation effort. to national defense, which was extremely burden- some given the narrow production base of the economy. Military expenditures accounted for more than 1 Historical Background and Buildup of 10 percent of GDP during the 1970s and the 1980s, Public Debt accounting for much of the fiscal deficit and accumulation of foreign debt. Domestic fiscal revenue— Origin of the Problem 21 percent of GDP on average between 1972 and 1988—was too low to finance such a high level of Jordan's high fiscal expenditure until the early expenditure; moreover, its structure was unbalanced, 1980s—equivalent to almost half of GDP—was with taxes on foreign trade providing nearly two- largely attributable to two major components—capi- thirds of total tax revenue (Chart 4.3). tal expenditure and military outlays—which to- Jordan was able to sustain such a large public ex- gether accounted for about 62 percent of total central penditure program because of large receipts of for- government outlays and 32 percent of GDP over eign grants, which despite some year-to-year fluctu- 1973-83 (Chart 4.2). Capital expenditures—a large ations remained high until the regional crisis of part of which was financed from external sources— remained extremely buoyant between 1973 and ] 1983 (averaging 15 percent of GDP), and then de- The figure does not include pension payments for military personnel or purchases of military equipment through the De- clined to a more sustainable level of 9 percent on av- fense Fund, generally secured from external sources under vari- erage between 1984 and 1988. Like most countries ous bilateral military aid programs.

©International Monetary Fund. Not for Redistribution 2 I V PUBLI C DEB T DYNAMIC S AN D FISCA L POLIC Y Table 4.1. Central Government Operations (In percent of GDP)

Average Average 1975 1980 1985 1988 1989 1990 1991 1992 1993 1994 1975-88 1989-94

Total revenue 22.3 19.2 21.8 24.0 23.8 27.5 28.2 33.7 30.7 29.2 21.5 28.8 Tax revenue 13.8 11.5 12.2 11.3 11.5 14.8 14.7 18.5 16.6 16.3 12.1 15.4 On income and profits 2.2 2.4 2.7 1.9 2.2 4.3 3.3 3.1 3.1 3.2 2.3 3.2 On domestic transportation 4.0 1.7 2.9 3.5 4.4 4.5 4.6 5.2 5.8 6.4 2.5 5.1 On foreign trade 7.6 7.5 5.9 5.2 4.4 4.4 4.8 8.2 6.1 5.2 7.1 5.5

Nontax revenue 8.5 7.6 9.6 12.7 12.3 12.7 13.5 15.2 14.1 12.9 9.3 13.4

Total expenditure1 54.0 43.5 36.7 41.2 45.8 42.6 41.9 37.3 37.0 35.5 41.6 40.0 Primary expenditure 52.9 42.3 32.8 34.4 36.3 32.2 31.3 28.2 30.5 30.0 39.3 31.4 Current primary expenditure 31.9 27.2 24.3 25.3 28.0 26.2 25.5 22.4 29.0 23.7 26.5 25.0 Civil current primary expenditure 19.2 17.0 13.2 14.0 17.3 16.6 16.0 14.1 15.7 15.2 14.9 15.8

Current expenditure 33.0 28.5 28.3 32.1 37.5 36.6 36.0 31.5 30.5 29.2 28.8 33.5 Wages and salaries 5.5 6.5 6.5 6.1 6.5 6.0 6.4 6.9 3.2 6.4 Goods and services 2.5 2.4 2.0 1.6 2.1 2.0 2.6 2.0 1.4 2.0 Interest payments I.I 1.2 4.0 6.8 9.5 10.3 10.5 9.1 6.5 5.5 2.3 8.6 Subsidies 2.2 4.0 1.7 0.8 3.1 3.3 2.1 1.6 1.4 1.0 1.8 2.1 Military outlays 12.7 10.2 II.1 11.4 10.6 9.6 9.4 8.3 8.3 8.5 11.7 9.1 Other outlays 3.6 4.3 5.7 5.6 5.2 4.4 5.4 5.3 2.0 5.3

Capital expenditure 21.0 15.0 8.5 9.1 8.3 6.0 5.9 5.8 6.5 6.4 12.8 6.5 Fiscal balance, excluding grants1 -31.7 -24.3 -14.9 -17.2 -22.0 -15.0 -13.7 -3.6 -6.3 -6.4 -20.1 -11.2 Primary balance -30.7 -23.1 -10.9 -10.4 -12.4 -4.7 -3.1 5.5 0.2 -0.9 -17.8 -2.6 Net lending and extrabudgetary operations2 2.2 6.2 6.2 -1.4 3.3 3.7 -0.4 -0.6 -0.5 2.4 0.7 Fiscal balance, excluding grants -31.7 -26.5 -21.1 -23.5 -20.6 -18.3 -17.4 -3.2 -5.7 -5.9 -22.5 -11.9 Primary balance -30.7 -25.3 -17.1 -16.7 -II.1 -8.0 -6.8 5.9 0.7 -0.4 -20.2 -3.3 Grants 15.9 17.7 13.0 8.6 12.8 10.9 8.7 3.9 4.1 3.4 12.6 7.3 Fiscal balance, including grants -15.9 -8.8 -8.1 -14.9 -7.8 -7.4 -8.7 0.7 -1.6 -2.5 -9.9 ^.5

Sources: Jordanian authorities; and IMF staff estimates. 'Excluding extrabudgetary expenditure and net lending. 2Net lending data are available from 1978 onward and extrabudgetary data from 1985 onward.

©International Monetary Fund. Not for Redistribution Historical Background and Buildup of Public Debt

Chart 4.2. Expenditure Chart 4.4. Nonmilitary Current (In percent of GDP) Expenditure (In percent of GDP)

Sources: IMF, International Financial Statistics, and various Recent Economic Development reports. Sources: IMF, International Financial Statistics, and various Recent Economic Development reports.

1990; and a low average interest rate on public debt (significantly below the growth rate of the econ- Thereafter, interest payments increased steadily in omy). Notwithstanding the sizable receipts of for- response to the rapid debt buildup and increases in eign grants, the overall fiscal deficit (including world interest rates. They peaked to an unsustainable grants) averaged 10 percent of GDP during 1972-88 level of nearly 11 percent of GDP during 1990-91, and made external borrowing necessary. Through on a commitment basis. The rapid increase in inter- 1983, interest payments on the external public debt est payments after 1988 also reflected in part the were relatively modest, accounting for less than sharp depreciation of the Jordan dinar over 1988-89, 2 percent of GDP during the period (Chart 4.4). which, together with other policy measures, was designed to address the balance of payments crisis that had developed. An important feature of the economic environ- Chart 4.3. Tax Revenue ment of the 1980s was the combination of high inter- (In percent of GDP) est rates in the world market and moderate economic growth in Jordan, which resulted in a more rapid growth in the debt-to-GDP ratio than would otherwise have been the case. This characteristic of the environment also meant that the Government's sol- vency constraint—the condition that the value of outstanding public debt be no greater than the sum of the present discounted values of expected primary surpluses and high-powered money—became binding on the operation of fiscal policy. For example, in 1988, just before the balance of payments crisis, tax revenues amounted to some 11 percent of GDP and nontax revenues, to about 13 percent, while total expenditure remained at 41 percent of GDP, of which 7 percentage points were attributable to debt service and 9 percentage points to capital expenditures. With foreign grants equivalent to almost 9 per- Sources: IMF, International Financial Statistics, and various Recent Economic Development reports. cent of GDP, the overall fiscal deficit, including foreign grants, was 15 percent of GDP. Reflecting the cumulative effect of such policies, the public debt

©International Monetary Fund. Not for Redistribution IV PUBLIC DEBT DYNAMICS AND FISCAL POLICY

a period in which actual and potential outputs are as- Chart 4.5. Debt-to-GDP Ratio sumed to be equivalent and the fiscal stance is neutral. The fiscal stance, in this sense, measures the magnitude of the stimulus injected (withdrawn) through the budget into (from) domestic aggregate demand and beyond the primary deficit implied by the structure of budgetary operations in the base period. The exchange rate effect matters in countries where the proportion of public debt denominated in foreign currencies is large and where the exchange rates vis-a-vis currencies in which the external debt is denominated have fluctuated significantly. Change in the stock of debt (£>) on account of central government operations can be decomposed as follows:

D,-D,_{ =-(ta- ga)YP, - ta-{Yt - YPt) + r,D,.x

Sources: IMF, International Financial Statistics, and various Recent + FD,_y{(ERtJER,)-\]-lB,, (1) Economic Development reports. where ta and ga are, respectively, the average ratio of revenue and primary expenditure to GDP during 1975-88; Yt and YPt are nominal and potential GDP at current prices in year t\ rt is the nominal implicit interest rate on public debt in year t\ Dt is the net ratio reached almost 100 percent of GDP in 1988 stock of public debt at the end of year t\ FDt is the (Chart 4.5). net stock of public debt denominated in foreign currency at the end of year t\ ERt is the average nominal Debt Accounting Framework exchange rate (per unit of domestic currency); and IBt is the discretionary primary balance in year t. The growth of the public debt ratio can be decom- The appendix provides a detailed methodology. posed into five major components based on the stan- The debt dynamics can be expressed in relation to 2 dard budget financing identity: (1) the structural GDP as: component, as determined by the underlying structural fiscal balance depending on trends in revenue DtIYt - Dt_{IYt_x =-(ta- ga)-YPt/Yt ~ ta

reflecting movements in domestic economic activity -lBJYt- [g,l(1 +gt)} compared with the trend output; (3) the evolution of {(1 + r,)-A-i/*M + (FDt/Yt_{).[(ER^/ERt) - 1]}, (2) interest payments; (4) the exchange rate effect; and (5) the annual discretionary fiscal impulse. The with gt representing the nominal rate of GDP growth structural component is defined as the average pri- in year t. The last two components of the right-hand mary deficit adjusted for variations in domestic outside of equation (2) are essentially adjustment terms put. The output variation component of the budget because output in the denominator relates to differ- deficit reflects variations in revenue resulting from ent years. These adjustment terms vanish if actual deviations in economic activity from its potential current output is used as the common denominator. trend level.3 The fiscal stance can be defined as ex- The right-hand side of equation (1) reflects the pansionary (contractionary) relative to the base year five aforementioned components, that is, the struc- if the actual primary deficit exceeds (falls short of) tural component, the output variation effect, interest the cyclically adjusted deficit, where the base year is payments, the exchange rate effect, and the annual discretionary component. The change in the debt ratio can, in a reduced form, be expressed as the op- 2Peter Heller, Richard Haas, and Ahsan Mansur, A Review of posite of overall fiscal balance (Bt, corrected for the the Fiscal Impulse Measure, Occasional Paper No. 44 (Washing- exchange rate effect). In turn, the fiscal balance can ton: International Monetary Fund, 1986). be decomposed along the same line, as follows: 3Primary expenditures are assumed not to be affected by cyclical fluctuations—a reasonable assumption given the absence of an unemployment insurance scheme in Jordan. Instead of using a Bt = (ta ~ ga)-YPt + ta-{Yt - YPt) - VtDt_X + IBt. (3) "base-year" approach, as is most often the case when estimating trend values, the methodology used here is based on average The fiscal balance reflects the difference between values. structural revenue and primary expenditure, the

©International Monetary Fund. Not for Redistribution Shift in Fiscal Policy Stance, 1989-94 cyclical output gap, the debt-service burden, and the quired primary balance will depend on the differen- yearly fiscal impulse. tial between the implicit interest rate on the public debt and the growth rate, both in nominal terms, and Debt Dynamics and Sustainability of the Debt, the current debt ratio. If the differential is positive, 1975-88 stabilization of the debt-to-GDP ratio requires a primary surplus of a magnitude proportional to the pre- Between 1975 and 1988, the ratio of outstanding vious year's debt-to-GDP ratio and to the size of this central government debt to GDP increased from 68 differential. If the differential is negative, stabiliza- percent to 95 percent (Chart 4.5; Table 4.2).4 A de- tion can be achieved even with some primary deficit. composition of annual fiscal balances indicates that, In Jordan, the implicit interest rate increased of the average annual fiscal deficit (excluding for- sharply during the 1980s because of a substantial eign grants5) of 20 percent of GDP during 1975-88, shift in the composition of foreign debt toward debt 17.6 percent of GDP was attributable to the struc- on commercial terms. In 1980, over 80 percent of tural primary deficit (Table 4.3, Charts 4.6 and 4.7). the outstanding external debt was owed to official The injection of expansionary stimulus was particu- bilateral creditors, primarily on concessional terms. larly strong up to 1980, and declined substantially By 1988, however, the share of external public debt during the early 1980s. In response to the rapid accu- owed to foreign banks and commercial companies mulation of the debt burden and external financing had increased to 20 percent from 7 percent in 1980, constraints, the authorities started to tighten the as the availability of funds on more attractive terms stance of fiscal policy in the mid-1980s; accordingly, from official creditors fell short of the Government's the fiscal impulse turned significantly contractionary financing needs. In contrast, interest rates on domes- from the mid-1980s. Interest payments on the out- tic debt were broadly stable. standing public external debt, which was low (at The differential between the implicit interest rate about 1 percent of GDP) during the 1970s, increased and the growth rate was negative in the 1970s and significantly in the 1980s in line with the growing through 1984 (Chart 4.8). Simultaneously, although debt burden, accounting for 7 percent of GDP by the actual primary deficit remained high, it declined 1988 and 10 percent of GDP by the end of the steadily through 1984 and remained stable thereafter decade. Based on these considerations, the observed through 1988. The combination of those two factors deterioration in the fiscal position between 1985 and led to a contraction of the debt-to-GDP ratio be- 1988 would be attributable to the continued large tween 1979 and 1983 and a steady increase from structural primary deficit, economic slowdown, and 1984 onward. In 1988, the debt-to-GDP ratio markedly higher interest charges, rather than to dis- jumped by 36 percentage points to 95 percent of cretionary fiscal policy. GDP. The large exchange rate adjustment that took The same budget identities can be used to measure place in 1988 explained about 80 percent of the in- the primary balance required to stabilize the ratio of crease in the debt ratio in 1988. public debt to GDP and to compare it with the observed primary balance. Stability of the debt-to-GDP ratio will be achieved if the primary surplus in per- Shift in Fiscal Policy Stance, 1989-94 centage of GDP is as follows: The authorities drastically shifted the stance of PB,IY, = [(r, - g,)/(l + gdi-D.JY^ fiscal policy beginning in late 1988—in the context + (FA-i/^-i)-[l/(l + gdHWt-i'ER,) - I). (4) of IMF-supported adjustment programs—which helped reduce the Central Government's fiscal Assuming no exchange rate effect on the valua- deficit, excluding foreign grants, by almost 18 per- tion of the outstanding stock of foreign debt, the re- centage points to less than 6 percent of GDP in 1994. Two-thirds of the fiscal consolidation was attribut- 4For the purpose of this empirical analysis, central government able to expenditure restraint and the remainder to an debt is defined to be limited to obligations arising from fiscal increase in revenue. Central government finances deficits accumulated over a certain period of time. Accordingly, started to improve in 1989, as adjustment policies the stock of outstanding debt and its accumulation—based on the (including discretionary fiscal measures) were im- analysis presented in this section—are different from the corresponding figures noted in other sections of this paper. plemented. Despite the adverse effects of the 1990 5Foreign grants, seen as a financing item, are excluded from regional crisis, which depressed economic activity total revenue. This can bias the fiscal balance downward, given and increased pressures on public services and social that part of the grants have financed expenditure—in particular, expenditure programs as a result of the large number capital expenditure—which would not have taken place other- of returnees, the fiscal deficit was reduced further in wise. Extrabudgetary operations, however, are excluded because no information prior to 1985 is available. Net lending and equity 1990 and 1991. In response to a strong economic re- participation are also excluded. bound and adoption of sizable discretionary revenue

©International Monetary Fund. Not for Redistribution 2 6 I V PUBLI C DEB T DYNAMIC S AN D FISCA L POLIC Y Table 4.2. Fiscal Sustainability

Average Average 1975 1980 1985 1988 1989 1990 1991 1992 1993 1994 1975-88 1989-94

In millions of Jordan dinars

Observed GDP (Y) 379.1 1,180.3 2,020.2 2,264.4 2,372.1 2,668.3 2,855.1 3,493.0 3,882.5 4,266.2 n.a. n.a. Public debt12 257.8 658.6 992.7 2,154.2 2,980.0 3,376.0 3,581.2 3,517.5 3,660.1 3,980.6 n.a. n.a. Domestic debt 65.4 197.8 347.2 903.6 976.7 1,019.2 1,021.4 1,004.2 1,059.1 1,059.1 n.a. n.a. Foreign debt 192.4 460.8 645.5 1,250.6 2,003.3 2,356.8 2,559.8 2,513.3 2,601.0 2,921.5 n.a. n.a. Public debt variation 60.9 78.1 38.9 918.8 825.8 396.0 205.2 -63.7 142.6 320.4 n.a. n.a. Fiscal balance2 -60.2 -78.1 -38.9 -195.5 -216.9 -109.2 -143.0 10.5 -84.4 -126.1 n.a. n.a. Primary balance -116.2 -272.9 -221.2 -236.0 -295.0 -124.9 -89.8 191.0 6.4 -36.9 n.a. n.a. Interest charges 4.1 14.5 80.3 154.3 226.2 276.0 300.8 317.9 251.0 234.5 n.a. n.a. Grants 60.1 209.3 262.6 194.8 304.3 291.7 247.6 137.4 160.2 145.3 n.a. n.a. Exchange rate effect3 -0.7 0.0 0.0 -723.3 -608.9 -286.8 -62.2 53.2 -58.2 -194.3 n.a. n.a. In percent of GDP

Public debt 68.0 55.8 49.1 95.1 125.6 126.5 125.4 100.7 94.3 93.3 57.8 II 1.0 Domestic debt 17.3 16.8 17.2 39.9 41.2 38.2 35.8 28.7 27.3 24.8 18.8 32.7 Foreign debt 50.7 39.0 32.0 55.2 84.5 88.3 89.7 72.0 67.0 68.5 38.9 78.3 Public debt variation 16.1 6.6 1.9 40.6 34.8 14.8 7.2 -1.8 3.7 7.5 9.9 11.0 Fiscal balance2 -15.9 -6.6 -1.9 -8.6 -9.1 -4.1 -5.0 0.3 -2.2 -3.0 -7.6 -3.8 Primary balance -30.7 -23.1 -10.9 -10.4 -12.4 -4.7 -3.1 5.5 0.2 -0.9 -17.8 -2.6 Interest charges I.I 1.2 4.0 6.8 9.5 10.3 10.5 9.1 6.5 5.5 2.3 8.6 Grants 15.9 17.7 13.0 8.6 12.8 10.9 8.7 3.9 4.1 3.4 12.6 7.3 Exchange rate effect3 -0.2 0.0 0.0 -31.9 -25.7 -10.7 -2.2 1.5 -1.5 -A.6 -2.3 -7.2 In percent Implicit interest rate (r) 2.1 2.5 8.4 12.5 10.5 9.3 8.9 8.9 7.! 6.4 4.8 8.5 Output growth (g) 26.2 20.3 2.0 2.5 4.8 12.5 7.0 22.3 11.2 9.9 16.1 11.3 Interest-output difference (r - g) -24.1 -17.8 6.5 10.0 5.7 -3.2 1.9 -13.5 -4.0 -3.5 -11.3 -2.8

Change in debt-to-GDP ratio2 2.5 -3.4 1.0 39.2 30.5 0.9 -I.I -24.7 -6.4 -1.0 2.1 -0.3 Primary balance effect -30.7 -23.1 -10.9 -10.4 -12.4 -4.7 -3.1 5.5 0.2 -0.9 -17.8 -2.6 Grants effect 15.9 17.7 13.0 8.6 12.8 10.9 8.7 3.9 4.1 3.4 12.6 7.3 Interest-output difference effect -12.5 -8.8 3.1 5.4 5.2 -3.6 2.3 -13.8 -3.6 -3.0 -5.4 -2.8 Exchange rate effect -0.2 0.0 0.0 -31.9 -25.7 -10.7 -2.2 1.5 -1.5 -4.6 -2.3 -7.2

Source: Jordanian authorities; and IMF staff estimates. 'Reconstructed from 1989 debt figure,annual deficits, and exchange rate effect. Excluding net lending and extrabudgetary expenditure. 3Negative if depreciation.

©International Monetary Fund. Not for Redistribution Table 4.3. Fiscal Impulse

Average Average 1975 1980 1985 1988 1989 1990 1991 1992 1993 1994 1975-88 1989-94

In millions of Jordan dinars

Observed GDP (Y) 379.1 1,180.3 2,020.2 2,264.4 2,372.1 2,668.3 2,855.1 3,493.0 3,882.5 4,266.2 n.a. n.a. Potential output (YP)' 430.4 1,092.0 2,078.3 2,779.5 2,601.5 3,021.5 3,268.4 3,398.7 3,560.0 3,687.1 n.a. n.a. Output gap (Y-YP) -51.3 88.3 -58.1 -515.1 -229.4 -353.2 -413.3 94.3 322.5 579.1 n.a. n.a. Fiscal balance2 -120.3 -287.4 -301.5 -390.3 -521.2 -400.9 -390.6 -126.9 -244.6 -271.4 n.a. n.a. Primary balance -116.2 -272.9 -221.2 -236.0 -295.0 -124.9 -89.8 191.0 6.4 -36.9 n.a. n.a. Structural primary balance -76.7 -194.7 -370.6 ^195.6 -67.2 -78.0 -84.4 -87.8 -91.9 -95.2 n.a. n.a. Cyclical component —11.0 18.9 -12.5 -110.6 -66.2 -101.9 -119.2 27.2 93.0 167.0 n.a. n.a. Interest charges 4.1 14.5 80.3 154.3 226.2 276.0 300.8 317.9 251.0 234.5 n.a. n.a. Discretionary balance -28.4 -97.1 161.8 370.1 -161.6 55.0 113.8 251.6 5.3 -108.7 n.a. n.a. In percent of GDP

Fiscal balance2 -31.7 -24.3 -14.9 -17.2 -22.0 -15.0 -13.7 -3.6 -6.3 -6.4 -20.1 -11.2 Primary balance -30.7 -23.1 -10.9 -10.4 -12.4 -4.7 -3.1 5.5 0.2 -0.9 -17.8 -2.6 Structural primary balance -20.2 -16.5 -18.3 -21.9 -2.8 -2.9 -3.0 -2.5 -2.4 -2.2 -17.6 -2.6 Cyclical component -2.9 1.6 -0.6 -4.9 -2.8 -3.8 -4.2 0.8 2.4 3.9 0.3 -0.6 Interest charges I.I 1.2 4.0 6.8 9.5 10.3 10.5 9.1 6.5 5.5 2.3 8.6 Discretionary balance -7.5 -8.2 8.0 16.3 -6.8 2.1 4.0 7.2 0.1 -2.5 -0.6 0.7 CPI increase (in percent) 12.0 II.1 3.0 6.6 25.8 16.1 8.2 4.0 4.7 3.6 7.4 10.4

Sources: Jordanian authorities; and IMF staff estimates. 'Assuming potential output annual growth rate is equal to the average observed growth rate over 1975-88; potential output in 1994 is such that the sum of output gaps is zero. Excluding net lending, extrabudgetary expenditure, and grants. Shif t i n Fisca l Polic y Stance , 1989-9 4 2 7

©International Monetary Fund. Not for Redistribution IV PUBLIC DEBT DYNAMICS AND FISCAL POLICY

Chart 4.6. Evolution of Debt-to-GDP Ratio Chart 4.7. Fiscal Impulse Breakdown

Sources: IMF, International Financial Statistics, and various Recent Economic Development reports. Sources: IMF, International Financial Statistics, and various Recent Economic Development reports. Note: The structural balance and the cyclical effects have been estimated for two subperiods, 1975-88 and 1989-94. As a result, there is a break in the series between 1988 and 1989.

measures, the Central Government's public finances improved dramatically in 1992. The underlying fiscal deficit (excluding foreign grants) declined to 1992, reflecting strong fiscal efforts and economic about 7 percent of GDP, and, taking into account growth, the effect of the Government's discretionary nonrecurrent revenue gains, the deficit decreased fiscal measures, and some favorable transitory fac- even further.6 Fiscal consolidation continued during tors. Overall, the primary fiscal deficit had almost 1993, albeit at a much slower pace. vanished by 1992, compared with 21 percent of Reflecting the discontinuation of grants from most GDP on average through 1988. Despite higher inter- countries in the region after the Gulf crisis, foreign est rates, the differential between the average inter- grants declined by 5 percentage points over 1988-94 est rate and nominal output became negative every to 3.4 percent of GDP in 1994. As a proportion of year except in 1991, when output growth was total central government receipts, grants decreased sharply lower in the aftermath of the 1990 regional from 26 percent in 1988 to 10 percent in 1994. crisis. The significant improvement in the debt ratio occurred despite a substantial reduction in the inflow of foreign grants and the adverse effects originating Debt Dynamics and Sustainability of from exchange rate depreciations in 1993 and 1994 the Debt Ratio (cumulatively, about 6 percent of GDP), as the Jor- The evolution of the debt-to-GDP ratio largely dan dinar was de facto pegged to the U.S. dollar. mirrored the fiscal consolidation process that was Foreign grants declined steadily, from about 13 per- started in late 1988. The debt ratio on account of cent of GDP in 1989 to 3 percent in 1994 as tradi- central government operations, which peaked in tional sources virtually stopped providing grants to 1990 at 126 percent of GDP, declined sharply in Jordan after the regional crisis. 1992 to 101 percent of GDP, and to 93 percent of Overall, the turnaround toward a primary budget GDP in 1994. A reversal in the interest-output differ- balance and the negative interest-output differential ential, which turned significantly negative beginning both indicated that, beginning in 1990, with the fun- in 1992, contributed to the observed debt reduction. damental shift in the stance of fiscal policy, the debt The primary fiscal balance sharply improved be- ratio became sustainable and exhibited a marked de- ginning in 1990 and became significantly positive in clining trend. The structural improvement in the primary balance was attributable to favorable developments in revenue and expenditure, reflecting efforts Equivalent to 3.5 percent of GDP, mainly as a result of the introduction of a customs duty on cars imported by Jordanians re- to improve the elasticity of the tax system, broaden turning from abroad. the tax base, and contain recurrent outlays.

©International Monetary Fund. Not for Redistribution Shift in Fiscal Policy Stance, 1989-94

Chart 4.8. Implicit Interest Rate and Chart 4.9. Revenue and Grants Economic Growth Rate (In percent of GDP) (Percent changes)

Sources: IMF, International Financial Statistics, and various Recent Economic Development reports. Sources: IMF, International Financial Statistics, and various Recent Economic Development reports.

Revenue Mobilization points of GDP. Also behind the revenue increase was the removal in January 1990 of the ban, im- Total revenue increased from 24 percent of GDP posed in November 1988, on imports of some high- in 1988 to almost 30 percent of GDP in 1994, with duty luxury items. tax revenue increasing by 5 percentage points (Chart While maintaining their efforts to mobilize rev- 4.9). The increase essentially reflected frequent dis- enue, the authorities also launched in 1989 a pro- cretionary revenue-raising measures, as the structure gram of gradual tariff reform that aimed at a lower, of the tax system remained inelastic with respect to more uniform, and nondiscriminatory structure of income. The revenue structure, however, improved, protection among economic sectors. As part of this with a decline in relative terms of taxes on foreign reform, they reduced the maximum cumulative trade; and administrative procedures for collecting import tax rate to 50 percent from more than 300 taxes and arrears were tightened. Receipts from post percent for all nonluxury imported goods. At the and telephone services and profits from the Central same time, however, to reduce the revenue effect of Bank of Jordan were the most buoyant nontax rev- this reform, they increased the minimum effective enue sources. tariff rate by levying an additional tax at a 5 percent rate—instead of the standard 3 percent rate—on imports that were exempted from other import Taxes on Foreign Trade taxes. Apart from the positive effects on the value of The authorities launched a second round of tariff imports of the 1988 and 1989 exchange rate depre- reform in November 1994. Customs tariff rates on a ciations, the increase of receipts from customs du- limited range of products, in particular, in the trans- ties resulted from higher import volume associated portation sector (new and used cars, trucks, and with the acceleration in the pace of economic activ- spare parts), were reduced. The tariff range was low- ity and a tariff reform. Collection of import taxes ered to 70-200 percent from 125-320 percent for was also boosted in the early 1990s by a shift in the new cars; and to about 50-150 percent from composition of imports in favor of consumer 115-310 percent for used cars. (The ban on imports durable goods, on which higher duty rates were of used cars more than five years old has been main- levied. This shift was associated with the reflux of tained for environmental and other considerations.) Jordanians from neighboring countries after August Trucks, for which duties varied from 15 percent to 1990. In addition, a temporary duty was levied on 30 percent, are now exempted. The customs duty on returnees' cars from November 1991 to December spare parts was reduced to 10 percent from 30 per- 1992, boosting revenues by about 2 percentage cent. As a result of the various trade reforms under-

©International Monetary Fund. Not for Redistribution IV PUBLIC DEBT DYNAMICS AND FISCAL POLICY

taken in recent years, the maximum customs tariff Taxes on Income and Profits rate declined to 50 percent (except that on cars, cigarettes, and alcoholic beverages).7 Receipts from personal and corporate income taxes, which were less than 2 percent of GDP in 1988, increased to 3.2 percent of GDP by 1994. In Taxes on Domestic Transactions 1994, about 60 percent of direct taxes were collected Between 1988 and 1994, the revenue from taxes from corporations; individuals (partnerships and on domestic transactions as a proportion of GDP in- professionals) contributed 25 percent and salaried creased to 6.4 percent from 3.5 percent. While the employees 15 percent. Dividends, interest, and capi- annual nominal growth rate of GDP averaged 11 tal gains were exempt from taxes. percent over the period, revenue from these taxes in- Although the tax base was somewhat broadened creased by 23 percent on average. in November 1989 when exemptions and deductions Revenue from the consumption tax and excise du- were reduced, the personal income tax remained ties—consolidated under the Consumption Tax Law characterized by numerous exclusions of income of November 1988—almost doubled between 1988 from tax and liberal exemptions, with the effect that and 1994, reflecting several increases in tax rates relatively few individuals were actually subject to and extensions of the coverage of goods subject to income taxation. There were ten income tax brack- taxation. The total yield of other taxes on domestic ets, with a top marginal tax rate of 45 percent, and transactions, although of limited magnitude, grew five different corporate tax rates based on activity rapidly over the same period as a result of increases (industrial, financial, other) and ownership (public, in the property transfer tax rate and higher airport private). departure fees. The coverage of the consumption tax was broad- Nontax Revenue ened in 1992, and with 21 items added in June 1993, the total number of items subject to the consumption Nontax revenue as a proportion of GDP remained tax rose to 106. In addition, the existing consump- broadly stable, at about 13 percent of GDP during tion tax on certain imported items was extended to 1988-94. Fees, licenses, and charges for government their equivalent produced domestically, and certain services were increased substantially in 1986 and specific rates were converted into ad valorem rates. again in the subsequent three years. A royalty tax on Despite the coverage extension, the structure of the phosphate and potash production was introduced in consumption tax remained heavily concentrated on a 1989. The operating surpluses from post and tele- few products, with excises on cigarettes alone ac- phone services expanded continuously after 1988 counting for 44 percent of consumption tax receipts; because of increased coverage and rate increases. the latter, together with receipts from excises on cement, soft drinks, steel bars, and lubricating oil, ac- Expenditure counted for 80 percent of consumption tax receipts. With a view to broadening the domestic tax base During 1989-94, the average increase in central and making the tax system more elastic and more ef- government expenditure and net lending was limited ficient, the authorities replaced the consumption tax to 5.7 percent, and, accordingly, total expenditure as with a general sales tax on June 1, 1994. It is levied a proportion of GDP declined by 12 percentage on all imports, all manufactured goods, and some points to 35 percent in 1994. This reduction was services produced or rendered locally by providers achieved notwithstanding strong pressures on expen- whose annual sales exceed JD 200,000. ditures from several sources. In particular, exchange The standard ad valorem rate of the general sales rate adjustments during 1988-89 led to the subsi- tax was 7 percent at its inception in 1994; a 20 per- dization of food staples and higher interest payments cent rate is levied on a limited number of goods, on foreign debt, and the 1990 regional crisis led to such as perfume and electrical appliances. Specific strong pressures on recurrent expenditures. Supple- rates continue to be levied on tobacco products, al- mentary expenditures had to be incurred during coholic beverages, soft drinks, cement, and a few 1990-91, to respond to the needs of the large num- other products. Exports are zero rated. The recent ber of Jordanian workers returning from neighboring amendment to the general sales tax law broadened countries (especially for housing and education, and the tax base significantly and increased the basic rate emergency stockpiling of food supplies) and in- to 10 percent. creased military preparedness. However, the authorities succeeded in accommodating these pressures without significantly hindering ongoing efforts to 7This rate, however, excludes various fees and surcharges. In- cluding these, the maximum effective tariff rate would be 70 per- contain expenditures. Extrabudgetary expenditure cent, except for a few luxury items. was stopped, net lending became negative, and both

©International Monetary Fund. Not for Redistribution Recent and Forthcoming Structural Reforms in the Fiscal Area current and capital expenditures decreased as a pro- one based on broader domestic consumption. On portion of GDP.8 the expenditure front, further action is needed to Since 1992, current expenditure has been fully contain the civilian and military wage bill (includ- covered by total domestic revenue, whereas a signif- ing pension liabilities) and food subsidies. A num- icant shortfall existed in the late 1980s. Progress in ber of key policy actions have recently been changing the composition of expenditure, however, adopted by the Jordanian Parliament: has been mixed. On the one hand, military outlays • In September 1995, the Parliament adopted sev- have been reduced by about 3 percentage points of eral amendments to the general sales tax, with a GDP, and food subsidies have also declined. Gross view to improving the efficiency of the tax sys- lending to public enterprises and decentralized agen- tem. The main amendments increase the stan- cies has been significantly curtailed. Capital expen- dard rate to 10 percent; replace the positive list diture was reduced to 6.4 percent of GDP in 1994— of services subject to taxation by a negative list a level broadly in line with the absorption capacity with limited exemptions; allow for voluntary of the Jordanian economy. On the other hand, the registration of taxpayers; and provide for intro- wage bill increased from 6.5 percent in 1988 to ducing a supplementary duty on selected luxury, 6.9 percent in 1994, to compensate for the sharp de- or socially undesirable, products in order to pro- cline in real wages during 1988-91; the share of tect revenue in the context of the next stage of public pensions in total outlays also increased signif- external tariff reform. icantly during this period. • The direct tax system was also reformed in Sep- Wages and salaries, which constitute the largest tember 1995. The improvements included elim- component of civilian expenditure, accounted for inating tax holidays (except for investments in 25 percent of total current fiscal outlays and almost less-developed areas); limiting tax deductibility 7 percent of GDP in 1994. Between 1991 and 1994, to net interest payments; reducing the number of civilian employment increased by 15 percent to tax rates and the maximum tax rates for both 118,500 employees. Over the same period, the total personal and corporate income taxes; rationaliz- wage bill grew by almost 60 percent in nominal ing corporate income tax rates, with a view to terms and by 40 percent in real terms. Pension out- treating all corporate sectors equally, by estab- lays for both civil servants and military personnel in- lishing three flat corporate tax rates of 15 per- creased by 1 percentage point of GDP between 1988 cent for companies in "encouraged" sectors and 1994, owing to increases in the number of bene- (mining, industry, hotels, and hospitals), 35 per- ficiaries and the average level of pension. The long- cent for banks and financial institutions, and term sustainability of pension arrangements also re- 25 percent for all other companies; encouraging mains questionable. capital accumulation by imposing a withholding tax of 10 percent on distributed profits; and broadening the tax base by reducing and simpli- Recent and Forthcoming Structural fying exemptions and applying uniform stan- Reforms in the Fiscal Area dard deductions for all wage earners. The current level of total domestic revenue in • In the next stage of external tariff reform, numer- relation to GDP (almost 30 percent) is broadly satis- ous customs fees and charges will be combined factory and is comparatively high among develop- into one consolidated tariff structure; the number ing countries at about the same level of develop- of tariff rates will be reduced; nontariff barriers ment. There is a need, however, for further decisive will be lowered through the elimination of im- efforts to enhance revenue elasticity and the effi- port-licensing requirements and import monopo- ciency of the tax system and to reduce dependence lies; and the maximum consolidated tariff rate on nontax revenues. Additional structural improve- will be further reduced. The intended introduc- ments in the tax system would help avoid the fre- tion of supplementary (excise) duties within the quent and politically difficult discretionary revenue framework of the general sales tax will pave the measures implemented to offset the inelasticity of way for reducing maximum duties without re- the existing tax system. They would also help reori- ducing tax revenue. These supplementary duties ent the tax system to the changing structure of the should be applied equally on both domestic and economy away from a system based on imports to imported goods and imposed on a limited range of items selected on the basis of equity, health, and environmental considerations. 8Until 1991, total expenditure included extrabudgetary expenditure, of an unknown nature, measured as the difference between Further structural reform measures are to be the financing available to the treasury and the recorded cash implemented in several other areas, including, in deficit. particular, food subsidization. Better targeting,

©International Monetary Fund. Not for Redistribution IV PUBLIC DEBT DYNAMICS AND FISCAL POLICY

cross-subsidization, a higher cost-recovery rate for ITt = discretionary revenue in year t, subsidized prices, and the complete elimination of IEt = discretionary primary expenditure in year t, some subsidies, in combination with favorable inter- IBt = discretionary primary balance in year t, national commodity price developments, helped rt = nominal implicit interest rate on public lower the budgetary cost of subsidies from 3.3 per- debt in year t, cent of GDP in 1990 to 1.0 percent in 1994. How- gt — nominal rate of GDP growth in year t, and ever, generalized price subsidies have persisted for ERt = average nominal exchange rate (per unit of wheat, and the targeting of means-tested subsidies domestic currency). needs to be improved. A significant increase in commodity prices combined with the steady population Instead of a "base-year" approach, which is used growth rate could jeopardize the savings achieved most often to estimate trend values, a methodology between 1990 and 1994. The authorities are thus based on "average values" has been designed. In the considering reforming the current generalized food absence of reliable proxies, such as the utilization subsidy program by charging full prices to con- rate of fixed investment or the unemployment rate, sumers while protecting low-income households potential output is assumed to grow at a constant through direct cash transfers based on family in- rate, equal to the average growth rate observed from come. Such a scheme should enable the authorities 1975 to 1988. Once the growth rate is known, the to generate net budgetary savings (about 0.5 percent initial level of potential output must be determined. of GDP, based on current costs) and also prevent A solution would have been to assume that potential waste and smuggling of subsidized products. output was equivalent to actual output in the base year 1974. However, because of the concave shape of actual output function over time, this would have led to an almost permanent excess of actual output Appendix: Methodology of Debt over potential output, disturbing the cyclical compo- Dynamics Accounting Framework nent. Therefore, the level of potential output in 1974 has been adjusted in such a way that the undis- Jordan's successive large fiscal imbalances have counted sum of the differences between actual and led to a large stock of debt. To analyze the debt dy- potential output became zero. namics, the growth of the public debt ratio can be decomposed into five elements: Decomposition of Debt Dynamics • the structural trend in revenue and expenditure, • cyclical fluctuations, The starting point is that the change in the stock • the discretionary fiscal impulse, of debt in domestic currency is the opposite of the • the evolution of interest payments, and budget balance, which has a positive value if in sur- • an exchange rate effect. plus. In addition, a correction must be made for exchange rate variations, which affect the value of the The purpose of this appendix is to describe the inherited stock of foreign-financed debt. Exchange methodology used in assessing the relative weights rate variations also affect the budget balance of factors behind debt dynamics in Jordan, and the through various channels, such as foreign interest magnitude of the fiscal consolidation effort required charges or external grants, but these effects are not to stabilize the ratio of public indebtedness to output. singled out in the following analysis. To ignore the Notation will be as follows: exchange rate effect, as debt dynamics studies often Y = nominal GDP at current prices in year t, do, is a major drawback when significant exchange t rate adjustments have occurred. The exchange rate YPt = potential GDP at current prices in year t, D = net stock of public debt at the end of year t, pertinent to this analysis is the average one vis-a-vis t currencies in which foreign liabilities are denomi- FDt = net stock of foreign-financed public debt at the end of year t, nated, using weights of the outstanding foreign debt. Bt = overall balance of year t (positive value if surplus), PBt = primary balance of year t (positive value if A - A-i =~Bt + FDt^[{ERt^IERt) - 1], (A.I) surplus), Tt = total revenue in year t, Then the budget balance is decomposed into two ta = average ratio of revenue to GDP during parts: the primary balance and interest payments. In- 1975-88, terest charges depend upon the outstanding stock of Et = total primary expenditure in year t, debt and the current average interest rate on that ga = average ratio of primary expenditure to debt. It is assumed that the current deficit does not GDP over 1975-88, lead to interest payments during the current year.

©International Monetary Fund. Not for Redistribution Appendix

Bt PBt rr D't-\> (A.2) DtIYt - Dt-XIYt-x = - {fa - ga)'YPtIYt - ta'(Yt - YPt)IYt - lBt/Yt + rt-DtJYt_{ + (FA-1/^-1) In turn, the primary balance can be decomposed [ h into its components: total revenue and total primary UER,JER,)-\] Hi''0+8,))] expenditure. {(l+r,)-A-i/^-i + (^A/^r-i) [(ERt.{/ERt)-l]}. (A.9) PBt = Tt-Et. (A3) The last two components of the right-hand side of equation (A.9) are adjustment terms attributable to Tax revenue during a given year can be broken the fact that output in the denominator relates to dif- down into a structural component, a cyclical effect, ferent years. The adjustment term vanishes if actual and a discretionary component. In the absence of current output is used as a common denominator. discretionary measures, tax revenues are assumed to be a function of the observed output and the average tax pressure ratio. To isolate the structural compo- Stabilization of Indebtedness nent from the cyclical effect, potential output can be The same budget identities, with the same limited substituted for observed output to capture the struc- data requirements, can be used to measure the pri- tural tax revenue, and the difference is deemed to be mary balance required to stabilize the indebtedness the cyclical effect. ratio and to compare it with the observed primary balance. Tt = ta-Yt + ITt = ta-YPt + ta-(Yt - YPt) + IT, (A.4) Substituting equation (A.2) into equation (A.I) gives: Primary expenditure can be decomposed in a sim-

ilar way, but it is assumed that there is no cyclical ef- A DV-i PBt + rt-D7-J fect, that is, that noninterest expenditures are, in the + FDt_v[{ERtJERt) 1] (A. 10) absence of discretionary measures, a function of potential output and the average expenditure-to-GDP Dividing all terms by Y,, or, equivalently, by ratio. y,_i(l + gt), and rearranging leads to:

DtIY, - D,JYt.i = - PB,IY, + [{r, - g,)/( 1 + g,)] D.JY,., Et = gaYPt + IE, (A.5) + (FD.JY^ylW + g,)] Substituting (A.4) and (A.5) into (A.3) and replac- [(ER,-i'ER,)-\]. (A.ll) ing the difference between the discretionary tax and Therefore, stability of the debt-to-GDP ratio will expenditure elements by a discretionary fiscal bal- be achieved if the primary surplus in percentage of ance element yields the following result: GDP is as follows:

PBt = (ta - ga>YPt + ta-(Yt - YPt) + IB, (A.6) PBl/Yl=[(rt-gt)/(\+g,)}-DrJYl_l + (FA-i/F,.,) [l/(l+g,)H(HWfl?,)-l]. (A.12) In turn, substituting (A.6) into (A.2) gives: Assuming no exchange rate effect on the valua-

Bt = (ta - ga)-YPt + ta(Yt - YPt) + IBt - rt-Dt_{. (A.7) tion of the outstanding stock of foreign debt, if the differential between the implicit interest rate on pub- Finally, substituting (A.7) into (A.I), the result is: lic debt and the growth rate, both in nominal terms, is positive, stabilization of the debt-to-GDP ratio re-

A - A-i =-(ta- ga)-YPt - ta-(Yt - YPt) - IBt quires a primary surplus whose magnitude is propor-

+ r,A-i + FDt-.x• [(ERt-i/ER,) - 1]. (A.8) tional to the preceding year's debt-to-GDP ratio and to the size of this differential. If the differential is The five components of the right-hand side of negative, stabilization can be achieved even with equation (A.8) are the five aforementioned elements: some primary deficit. (1) the structural component, (2) the cyclical effect, The time series for public debt has been con- (3) the discretionary component, (4) the interest structed by adding the successive cash deficits. The charges, and (5) the exchange rate effect. implicit interest rate on public debt is equal to inter- The debt dynamics can be expressed in ratios to est charges paid during a given year divided by the GDP: stock of debt at the end of the previous year.

©International Monetary Fund. Not for Redistribution V Financial Liberalization and Monetary Reforms

Jean-Pierre Chauffour

ordan's financial system has expanded consider- The banking system played an effective role in ably over the last two decades, as evidenced by the intermediating short-term savings and long-term fi- ; increase in the ratio of money and quasi money to nancing requirements.2 Total outstanding credit fa- GDP from 60 percent in 1970 to over 100 percent in cilities of commercial banks increased from 20 per- the late 1980s. Like the rest of the economy, the fi- cent of GDP in 1970 to more than 60 percent of GDP nancial system in Jordan was adversely affected by in 1994, reflecting a deepening of private sector fi- the financial crisis of the late 1980s. Thereafter, it nancing. Simultaneously, the role of the specialized strengthened significantly as a result of the imple- credit institutions as major intermediating institu- mentation of structural reforms, a general rebound- tions expanded considerably. Their total loans and ing of economic activities after 1992, and the associ- advances to the private and public sectors stood at ated increased profitability of the commercial banks. 20 percent of GDP in 1993, compared with less than Although the financial system is deeper than before, 7 percent of GDP in the early 1970s. it is still dominated by a small number of domestic The payments system also improved considerably, banks; the variety of financial instruments for users with the commercial banks' clearing operations and providers of funds is rather limited; the capital reaching 7.2 million checks in 1993, compared with market—with the exception of the stock market—is less than 500,000 in 1970. The shift from cash hold- extremely thin. In general, like in many other devel- ings toward time and savings deposits was reflected oping countries, the financial system faces the chal- in an unprecedented growth of resident deposits in lenge of funding the most efficient investments the banking system, which increased from 20 per- while managing risks and integrating with world fi- cent of GDP in 1970 to about 90 percent of GDP in nancial markets. 1993 (Chart 5.1). This rapid monetization of the economy resulted in a downward trend in the income velocity of money during the 1980s (Appendix I). Background and Developments Performance of the financial system was further enhanced by the establishment in 1976 of the In 1970, the Jordanian banking system was under- Amman Financial Market (AFM) for trading in se- developed and comprised eight commercial banks, curities. Since its creation, the AFM has expanded of which four were branches of foreign banks. The its operations rapidly, making it one of the most ac- largest domestic bank and the largest foreign bank tive emerging markets in the Middle East. accounted for half the total branches of domestic and foreign banks in Jordan, respectively. Twenty years later, the number of banks has increased threefold;{ the total number of branches has risen from 41 to Operation of Monetary Policy, 1970-88 more than 300 and are spread all over the country (Figure 5.1). The deepening of the financial system Monetary Developments during these two decades encompassed many areas: Jordan's inflation during the 1970s and through banking intermediation has been enhanced, the payments system improved, and an active stock ex- the late 1980s was generally low, reflecting in part a change established.

traditionally, the financial depth of an economy is measured •The banking system in Jordan consists of the Central Bank of by the ratio of liquid liabilities to GDP, assuming that the size of Jordan, 6 foreign commercial banks, 16 domestic commercial the financial intermediary sector is positively correlated with the banks (of which 1 operates on the basis of Islamic financial prin- provision of financial services. Total liquid liabilities of the finan- ciples), 7 financial institutions, a number of financial institutions cial system may be proxied by the ratio of money and quasi that do not accept deposits, including the Social Security Corpo- money to GDP. In Jordan, this ratio has doubled since 1970, to ration, and 18 insurance companies. reach 125 percent in 1993. D

©International Monetary Fund. Not for Redistribution Operation of Monetary Policy, 1970-88

Figure 5.1. Banking System

1964 Central Bank of Jordan

National Banks Foreign Banks

Commercial Banks

1930 Aral? Bank PLC 29 40 1949 British Bank of the Middle East 5 0 1956 Jordan National Bank 34 4 1951 Arab Land Bank 19 0 1960 Bank of Jordan 37 24 1957 Rafidain Bank 3 1 1960 Cairo Amman Bank 20 13 1969 ANZ Grindlays Bank 14 0 1977 Jordan Kuwait Bank 17 7 1974 CitiBank 1 0 1978 Jordan Gulf Bank 27 1989 Arab Banking Corporation (Jordan) 9 0 1991 Business Bank 9 0

Islamic Banks

1979 Jordan Islamic Bank for Finance and Investment 29 2

Investment Banks 1 1978 Arab Jordan Investment Bank 8 7 1989 Jordan Investment and Finance Bank 4 0 1990 Amman Bank for Investments 4 3 1991 Union Bank for Savings & Investment1 10 3 1993 Philadelphia Investment Bank2 3 1 1993 Middle East Investment Bank 18 1

Specialized Credit Institutions

Public Ownership Joint Ownership 1 1 1959 Agricultural Credit Corporation 16 0 1965 Industrial Development Bank 3 0 1965 Housing and Urban Development 1968 Jordan Cooperative Organization 10 7 Corporation 1 0 1974 Housing Bank 109 7 1966 Cities and Villages Development Bank 2 0

Other Financial Institutions

Money Changers Contractual Credit & Saving Companies

1984 Beit Al-Mal Saving and Investment for Housing In Amman 44 Outside Amman 28 Company, Ltd. (Beitna) 6 0

Representative Offices 2

Note: Year of establishment is given. Resulted from the merger of DARCO company and Jordan Finance House. • Number of branches, including headquarters. Previously known as REFCO. [ I Number of minibranches.

©International Monetary Fund. Not for Redistribution V FINANCIAL LIBERALIZATION AND MONETARY REFORMS

during 1970-88. However, beginning in the early Chart 5.1. Monetary Aggregates 1980s, a shift took place in the relative composition of the domestic credit expansion. The rapid expansion in claims on the private sector started to moder- Deposits with Commercial Banks ate in 1979, decelerating from about 25 percent (as a (In percent of GDP) percentage of the beginning period stock of broad money) to 2.8 percent in 1987, whereas credit to the public sector started to rise at a double-digit rate.

Credit Expansion Driven by the Public Sector As a result of the expansionary fiscal policy stance from the mid-1980s, Jordan's public finances deteriorated rapidly during 1987-88. As receipts of foreign grants declined steadily, recourse to domestic bank borrowing to finance the growing budget deficits increased sharply. Claims on the Govern- ment increased by 130 percent in 1987 and by 50 percent in 1988. This development together with a Domestic Credit (As a percent of beginning stock of broad money) concomitant rapid increase in credit expansion to public sector enterprises resulted in a sharp expansion in the net domestic assets of the banking system. In the event, the annual rate of expansion of broad money (money and quasi money) averaged 13 percent during 1987-89, a period during which the real economy was stagnant or in recession. Thus, the economy was driven by government demand, which resulted in a large transfer of savings from the private to the public sector, crowding out private sector activities, and, ultimately, weakened the role of the banking system.

Use of Monetary Instruments Until the Sources: Department of Statistics, Government of Jordan; Late 1980s Central Bank of Jordan. From the early 1980s through 1988, money and credit policies were essentially geared toward stimu- lating domestic economic activity. Accordingly, commercial banks' reserve requirements, liquidity 3 ratios, credit guidelines, and interest rate ceilings prudent stance of monetary policy. Money and were adjusted frequently to support bank liquidity quasi money expanded at an annual average rate of and encourage domestic lending. Some of the instru- 18 percent during 1970-88, mainly as a result of the ments used to this effect included lowering the ceil- expansion in the net domestic assets of the banking ings on balances held abroad in relation to their cur- system; these increased by 14 percent (as a percent- rent foreign commitments, extending credit against age of the beginning period stock of broad money) securities, lowering the maximum lending rate on average during the period. By contrast, the rate of charged by the commercial banks and financial com- expansion in net foreign assets was limited to 4 per- panies and the interest rate ceilings on time and sav- cent on average during 1970-88. The bulk of the do- ings deposits, and expanding and easing the use of mestic liquidity expansion took the form of credit to the export credit facility. In 1987, the Central Bank the private sector, which rose by about 10 percent on introduced further measures to bolster bank liquidity average during the period. Borrowing by the Central and thereby facilitate domestic lending activity. Government and the public entities remained rela- These measures included lowering the amount that tively small, rising on average by 6 percent a year commercial banks were required to invest in stocks of public shareholding companies to 9 percent from 3The average rate of inflation was contained at about 6 percent 15 percent of their capital and reserves; reducing the during the decade through the late 1980s. amount of treasury bonds that the commercial banks

©International Monetary Fund. Not for Redistribution Monetary Policy Under Government Stabilization Programs, 1988—94 were required to hold against demand deposits to 5 percent from 9 percent; and, subsequently, elimi- Chart 5.2. Prices and Interest Rates nating the requirement that commercial banks hold (Percent changes) treasury bills and government bonds against deposits. Nominal and Real Interest Rates

Monetary Policy Under Government Stabilization Programs, 1988-94

Use of Monetary Instruments In late 1988, the authorities began to tighten monetary policy significantly in the wake of the emerging balance of payments crisis and intensified pressures on the exchange rate and domestic prices. To encourage domestic savings mobilization, banks removed ceilings on deposit rates, and banks and financial institutions increased their maximum base lending rate in September 1988 and February 1989. Comparative Inflation Performances The commercial banks' interest rate structure was completely freed in February 1990, following the removal of ceilings on the base lending rate, at which time the base lending rate (including commission) edged upward to about 12 percent in September 1990, and the effective lending rate increased to 12-13 percent. The Central Bank also raised its rediscount rate and the rate on advances to commercial banks to 7 percent from 5.75 percent in September 1988, further increasing the rates to 8.5 percent by August 1989 and restoring positive real interest rates by 1991 (Chart 5.2). In line with the increase in the interest rate structure, in August 1989, the Sources: Department of Statistics, Government of Jordan; Central Bank also raised the rate of the export Central Bank of Jordan. financing facility to 8 percent for exports to Arab countries that had trade agreements with Jordan and to 6 percent for exports to other countries. The rates on government securities were also raised dur- 1989.4 To attract foreign currency deposits to the ing 1989-90. The rates on treasury bills were in- banking system, residents were allowed, as of July creased from 5 percent in 1988 to 6.25 percent in 1988, to deposit JD 50,000 with commercial banks September 1990, and the interest rate on govern- and financial institutions, up from JD 30,000; this ment bonds with a maturity of six years was in- limit was raised to JD 150,000 in September 1990.5 creased to 9 percent. In October 1990, a reserve ratio of 5 percent was In addition to raising the interest rate structure, imposed on investment banks, and the minimum the Central Bank began, in late 1988, to tighten do- ratio of capital and reserves that banks and finance mestic credit expansion. It raised the required re- companies were required to hold in public share- serve ratios on time and savings deposits in Jordan holding companies was raised to 20 percent from 15 dinars to 9 percent from 6 percent—the same ratio percent. A measure to tighten credit entailed pro- that was applied to dinar demand deposits. The hibiting commercial banks from extending credit Central Bank further increased the required reserve against foreign currency deposits and restricting ratio on dinar deposits to 11 percent effective Octo- ber 1989. Moreover, commercial banks were required to place on deposit with the Central Bank 35 4An internationally competitive interest rate is payable on these percent of their foreign currency deposits (of resi- foreign currency deposits and a penalty of 10 percent a year (payable in foreign currency) is required on the amount of the dents and nonresidents), including advance import shortfall. deposits denominated in foreign currency by June 5Jordanians working abroad are exempt from these limits.

©International Monetary Fund. Not for Redistribution V FINANCIAL LIBERALIZATION AND MONETARY REFORMS

their overdraft facilities to 20 percent of their credit In 1994, monetary policy was geared toward facilities. Financial institutions were prohibited maintaining the attractiveness of financial assets de- from extending credit in excess of JD 200,000 to nominated in Jordan dinars, in support of the ex- any one client without prior approval from the Cen- change rate policy, and building up gross official re- tral Bank. Ceilings were also imposed on credit ex- serves to a more comfortable level. Consistent with pansion to banks' board members and employees, as this policy, interest differentials in favor of assets de- well as to limited private shareholding companies. nominated in Jordan dinars were increased to Following a tightening from late 1988 through 2.7-2.9 percentage points vis-a-vis assets of similar 1990, the credit policy stance remained broadly un- maturity denominated in U.S. dollars. The effect of changed in 1991 in the aftermath of the regional cri- this high interest rate policy was to limit broad sis and was further tightened during 1992-94. Infla- money expansion to 8 percent, below the nominal tion (as measured by the consumer price index) GDP growth of 9.9 percent. The increase in the net decelerated sharply during 1989-94 and returned to domestic assets of the banking system was limited to its historic low levels, which were comparable with 6.6 percent of the beginning period stock of broad inflation performances in industrial countries and money, partly because of the tight credit stance but considerably lower than in many countries in the more directly because of net repayments to the bank- Middle East region. In view of the credit expansion ing system by the Central Government and the con- to the private sector (by over 7 percent of the begin- tractionary effect of other items (net) associated with ning period stock of broad money) in the first quar- the accumulation of reserves and capital by commer- ter of 1992 and persisting excess liquidity in the cial banks. banking system, the Central Bank adopted a series of restrictive measures. It raised reserve requirements by 2 percentage points, to 13 percent on commercial Banking Supervision and Regulatory Measures banks and to 7 percent on investment banks in April Bank supervision was strengthened in 1989 in the 1992, and by an additional 2 percentage points in wake of the financial problems facing two major do- early 1993, with a view to siphoning off excess li- mestic commercial banks (Petra Bank and the Jor- quidity. Furthermore, the Central Bank used moral dan Gulf Bank). The insolvency of Petra Bank sur- suasion with commercial banks to restrain lending. faced in August 1989, following the bank's inability In May 1992, the Central Bank implemented addi- to comply with the directive requiring all banks to tional measures. First, it opened a new six-month de- place 35 percent of their foreign currency deposits posit facility for commercial banks with an interest with the Central Bank of Jordan. The authorities rate of 4 percent.6 Second, it introduced, for the first acted promptly to prevent a banking crisis by time, a capital adequacy ratio of 1 to 10 of capital shoring up Petra Bank's reserves so that it could (including reserves) relative to credit for all com- meet the run on its deposits and settle some of its mercial banks. Third, it established a new maximum foreign obligations. Financial support was provided loan-to-deposit ratio of 90 percent. All commercial through an overdraft facility with the Central Bank. banks were required to observe both the capital ade- Moreover, the Economic Security Council created a quacy ratio and the loan-to-deposit ratio. Commer- special committee to oversee the operations of the cial banks that were in violation of any of these ra- Petra Bank. Finally, the Central Bank issued a num- tios had to consult immediately with the Central ber of additional regulatory measures governing Bank on a timetable to achieve compliance. loan loss provisioning, capital adequacy require- The credit rationing resulting from the use of di- ments, and portfolio management of the commercial rect instruments for credit control prevented the de- banks (see Box 5.1). velopment of proper market instruments to allocate scarce financial resources efficiently. The money multiplier increased in the early 1980s, reflecting the Development of AFM fall in the public's cash-to-deposit ratio, and it decreased during 1987-90, partly because of the re- Together with the Istanbul Stock Exchange, the gional financial and political turmoil (see Appendix AFM is one of the leading capital markets of the II). The money multiplier had recovered to its pre- Middle East.7 It ranked fifth out of 45 emerging mar- crisis level by 1991 and remained at about this level kets in terms of total return index in 1992. The ratio thereafter. of Jordan's market capitalization to GDP, which

6Previous facilities were for a duration of up to three months, 7For more details, see Mohamed El-Erian, and Manmohan S. and carried interest rates of 1-2 percent. Reserves held at the Kumar, "Emerging Equity Markets in Middle Eastern Countries," Central Bank under the previous facilities amounted to about 6 Staff Papers, International Monetary Fund, Vol. 42 (June 1995), percent of the money stock. pp.313-43.

©International Monetary Fund. Not for Redistribution Development of AFM

Box 5.1. Banking Crisis of 1989-90

The collapse of a major bank in August 1989, the exposed, with high levels of foreign obligations. In sev- third-largest banking institution in Jordan, together with eral cases, speculative operations in the form of posi- the emergence of financial difficulties in six other finan- tions between the official and parallel exchange mar- cial institutions linked to it jeopardized the integrity of kets resulted in the loss of almost the entire foreign Jordan's banking system.1 The banking crisis emerged assets of the banks. for three main reasons: (1) inadequate banking regula- To preserve the security of the entire banking system tions, (2) overexposure of the banking system to the real from the systemic risk of chain failures, the Central estate market, and (3) imprudent speculation on foreign Bank had to inject $400-500 million (about 10 percent exchange. of GDP in 1989) into the banking system through its One of the main reasons for the emergence of a overdraft facility to meet the run on insolvent banks' major banking crisis in the late 1980s was the lack of deposits and to settle some of their foreign obligations. proper banking supervision and prudential regulation. This very large injection of liquidity was reflected in The specialized financial institutions, which are ori- both the Central Bank's balance sheet and the monetary ented toward long-term sectoral projects and are thus survey through the growth of reserve money and likely to be exposed to risky intermediation, were not money and quasi money aggregates. Both monetary ag- under the surveillance of the Central Bank but were su- gregates increased by more than 15 percent in 1989 and pervised by their own sector ministries. The Central contributed to fuel inflation, which peaked at 26 per- Bank was limited to ensuring that commercial banks cent, at a time when real GDP was contracting by 13.5 complied with five critical operating ratios and to ap- percent. Since 1990, the Central Bank has gradually proving any credit in excess of JD 500,000 granted by a begun to recover a part of these unsecured assets; by commercial bank. However, this prudent credit man- mid-1994, it had recovered 20 percent of the total cost agement seems to have discouraged the commercial of the bailout operations. banks from doing the proper risk credit analysis in their In the wake of the banking crisis, the Central Bank loan allocations. More important, Jordan's accounting issued a number of additional regulatory measures regulations for loan loss provisions were inefficient in aimed at improving its surveillance over the banking identifying nonperforming assets early because deci- system. Specific measures governing loan loss provisions to build provisions were usually delayed until sioning, capital adequacy requirements, and portfolio after the final court decision on the destitute asset was management of commercial banks were enacted. As rendered. These weaknesses in the regulatory frame- regards provisioning for bad loans, the guideline is- work resulted in widespread fraudulent practices and sued in 1992 required provisioning of 50 percent for mismanagement that the monetary authorities failed to loans for which banks had not received any payments detect. for more than 180 days but less than one year, and Loans collateralized with real estate property became 100 percent of the loan amount for loans for which the nonperforming during the real estate market slump of banks had not received payments for more than one the late 1980s. In the aftermath of the second oil shock, year or for which the banks had determined that no re- Jordan registered large inflows of workers' remittances, payment would be expected. Regarding general provi- which reached an all-time high of $1.3 billion (about 25 sioning for ordinary bank credit, 1-2 percent would be percent of GDP) in receipts from abroad in 1984; this required for direct credit facilities and 0.25-0.50 per- injection of demand led to a rapid expansion of private cent for indirect credit facilities. The guideline also sector construction activities, and exposure of the bank- streamlined the procedure for notifying the Central ing system. As speculative real estate activities col- Bank about loan losses. The regulations governing lapsed in the late 1980s, together with the collateralized capital adequacy requirements of the commercial values of real estate loans, several banks were highly banks issued in 1992 essentially applied the Basle exposed, with numerous nonperforming assets in their Convention regarding capital adequacy to Jordan's balance sheets. banking system. Banks' excessive exposure on their foreign exchange A subsequent effect of the banking crisis has been to position also contributed to this crisis. Before 1987, the shift the composition of the assets held by the private Jordan dinar was strictly pegged to the SDR in nominal sector, in either cash or deposits, in favor of assets de- terms. In 1987, with the weakening of Jordan's balance nominated in foreign exchange. In the event, the com- of payments position, a parallel market for foreign ex- bination of low nominal interest rates, high anticipated change emerged, and financial institutions were over- inflation, and the perception that further depreciation of the Jordan dinar may follow contributed to a fall in the demand for holdings of Jordan dinars. Thus, by 1989, !The major bank accounted for more than 10 percent of the about 30 percent of the increase in broad money was in total assets of the banking system in 1989. Total bad assets the form of an increase in residents' foreign currency were believed to represent up to 20-30 percent of the total as- deposits, which in part reflected the valuation gains sets of the banking system. arising from the dinar's depreciation.

©International Monetary Fund. Not for Redistribution V FINANCIAL LIBERALIZATION AND MONETARY REFORMS

Table 5.1. Comparative Performance of Emerging Capital Markets, 1994

Number of Price- Largest Market Market Domestic Earnings Price-Book Dividend Market Capitalization1 Capitalization2 Companies Ratio Value Ratio Yield Share3

Middle Eastern countries Egypt 4,263 9.4 700 Iran, Islamic Republic of 2,770 3.7 147 Jordan 4,594 81.6 95 20.77 1.71 2.44 45.8 Morocco 4,376 14.1 61 Tunisia 2,561 16.1 21 Turkey 21,605 17.1 176 30.99 6.35 3.57 48.6 Other emerging markets Argentina 36,864 13.1 156 17.71 1.42 2.91 41.7 Chile 68,195 130.7 279 21.36 2.54 2.41 46.4 Colombia 14,028 21.1 113 19.54 1.38 1.70 61.2 Greece 14,921 15.6 216 10.42 1.89 4.61 37.2 India 127,515 45.5 7,000 26.67 4.17 0.98 19.4 Nigeria 2,711 6.6 177 6.00 1.59 8.37 49.3 Thailand 131,479 91.8 389 21.16 3.70 1.98 35.6 Industrial markets Japan 3,719,914 81.0 2,205 97.30 2.17 0.70 United Kingdom 1,210,245 1 13.1 2,070 14.80 2.20 4.20 United States 5,081,810 75.4 7,770 16.90 2.61 2.90

Source: Emerging Stock Markets: Factbook (Washington: International Finance Corporation, Central Capital Markets Department, 1995). 'In millions of U.S. dollars. 2As a percentage of GDP. 3Share of market held by the 10 largest stocks.

stood at about 74 percent in 1992, exceeds that of clined by 30 percent, to JD 245 million in 1993 and most emerging markets and is similar to that of in- to JD 134 million to 1994. Since 1990, the increase dustrial countries (Table 5.1). Instruments traded on in the index has been broadly based, with the indus- the stock exchange include private corporate equi- trial sector registering the best performance, fol- ties and bonds, Jordanian government bonds and lowed by insurance, banking, and other services. treasury bills, and development bonds. The bull or Unlike other emerging capital markets, the AFM is bear markets of the 1980s were determined by de- dominated by domestic commercial banks and is velopments in workers' remittances (which were characterized by low foreign participation. Although themselves closely correlated with movements in in- the influence of foreign investors is increasing, most ternational oil prices). The impressive performance market participants are Jordanian (85 percent in of the AFM was also attributable to the fact that it 1992). Residents of some neighboring Arab coun- did not have to compete in an environment of per- tries also participate in the stock market. Although verse yield curve, such as rates of return on fixed in- repatriation of investment income in any convertible terest deposits well in excess of the rate of inflation, currency out of Jordan by Arab and non-Arab in- resulting from government intervention. Market cap- vestors is free of restriction, foreign participation by italization stood at $4.7 billion as of the end of De- residents of non-Arab countries has so far been cember 1994, and the number of shares traded in limited. 1994 amounted to 134 million for a total value of A range of economic and institutional circum- $620 million. stances have contributed to the successful develop- Since 1990, the performance of the market has ment of the Jordanian equity market since 1990, improved steadily, with the AFM index registering considerably enhancing investors' confidence in Jor- annual gains averaging 25 percent and the index of dan's economic potential. These circumstances in- total returns improving markedly. However, during clude Jordan's recent gains in macroeconomic stabi- 1993-94, the AFM was somewhat less buoyant than lization, and the progress it has made in reducing the in 1992 because of a tightening of credit since the external debt and debt-service burden, normalizing second half of 1993; the number of trade shares de- payment relations with external creditors, and liber-

©International Monetary Fund. Not for Redistribution Appendix I alizing the exchange system. As regards institutional Second, in line with the steady implementation of factors, Jordan has embarked on comprehensive fi- indirect monetary control operations, the Central nancial liberalization aimed at better mobilizing and Bank streamlined certain facilities and is improving allocating domestic and foreign resources. In partic- existing facilities by (1) replacing the deposit facili- ular, the Jordanian authorities decided to improve ties by CDs as the term deposits mature; (2) elimi- the trading, reporting, and accounting systems; nating the credit-to-deposit ratio once the indirect strengthen legal procedures; and liberalize regula- monetary control program becomes effective—on tions governing foreign direct and portfolio invest- prudential grounds, the authorities have already ments, including ownership, market access, and adopted the Basle standards regarding the risk- repatriation of capital, dividends, and profits. based capital-to-assets ratio; and (3) viewing the liquidity ratios as having a useful prudential function but not as monetary instruments. The Central Ongoing and Prospective Bank is also (1) gradually reducing reliance on Financial Reforms obligatory reserve requirements as a monetary policy instrument; (2) improving the operations of Although the Jordanian authorities have not issued reserve requirements by gradually unifying and re- a detailed blueprint for further financial reform, they ducing the reserve requirements for all banks; and have already either implemented some elements or (3) eliminating or reducing preferential credit facili- referred to them in official pronouncements. The au- ties and providing subsidies explicitly through the thorities' strategy, in general terms, entails (1) con- budget. trolling monetary conditions using the new instru- Third, the authorities are planning to remove the ments of indirect monetary control (Appendix III); current impediments to the development of the inter- (2) maintaining all interest rates positive in real bank market. To this end, they eliminated the prac- terms and achieving much more flexibility in inter- tice of double reserve requirements on interbank de- est rates than in the past, with a view to achieving ef- posits in 1994. ficient market-based mobilization and allocation of loanable funds; (3) using central bank CDs denominated in Jordan dinars in a manner that would enable Appendix I: Money Demand Function the authorities to respond to shifts in the flow of foreign assets and movements in the exchange rate, and Broad money demand (money and quasi money) to maintain the liquidity of the banking system con- is usually analyzed using the income elasticity o( sistent with domestic price stability; (4) eliminating money demand or, more directly, the income veloc- redundant monetary control instruments, particularly ity of money. following the effective introduction of indirect mon- The transactions demand, the precautionary de- etary controls, and improving the efficiency of the mand, and the speculative demand for money are remaining necessary monetary instruments; (5) re- usually proxied in the money demand function with moving impediments to the development of the in- key macroeconomic variables, in particular, real terbank market; and (6) improving efficiency GDP, consumer prices, and other variables such as through enhanced competition among commercial interest rates or real wealth. The estimation of the banks and specialized credit institutions, such as the money demand function in Jordan is complicated by Housing Bank, the Industrial Development Bank, a number of economic factors (such as the policy of and the Agriculture Credit Corporation. direct monetary control, the recent switch toward in- The authorities have made considerable progress direct control, erratic inflows of labor remittances, recently in a number of areas of structural reform. and transfers of workers' savings) or political factors First, the Central Bank has held weekly auctions (such as the August 1990 regional crisis, and recent of CDs since September 1, 1993 and, since then, has developments and prospects in the West Bank and increasingly used this indirect instrument to exercise Gaza Strip). However, a rough estimated relation- control over reserve money. For the indirect system ship for 1980-93 suggests a long-term income elas- of control to have its desired impact, the Central ticity close to unity and that real demand for money Bank must maintain reserves close to their required increases with inflationary trends and thus is sensi- minimum level at all times by developing proper tive to money illusion (Chart Al). monetary programming techniques, including test- Income money velocity, defined as the ratio of ing, and by improving reserve forecasting. The auc- nominal GDP at market prices to outstanding money tions of CDs are being coordinated with the auctions and quasi money at the end of the period, decreased of treasury bills in terms of timing and settlement, steadily during most of the 1980s, from 1.2 in 1981 eligible holders and bidders, and the range of maturi- to 0.8 in 1989 (Chart A2). This downward trend re- ties offered. flected essentially the monetization of the economy

Chart Al. Broad Money Demand Chart A2. Income Velocity of Money (In billions of Jordan dinars) (In percent)

Regression output Constant -2.51476 Source: IMF, International Financial Statistics. Standard error of Y estimate 0.04127 R squared 0.99425 Number of observations 12 Degrees of freedom 10

Demand GDP CPI Deposits Y coefficients 0.933301 0.642500 0.171518 Standard error Appendix II: Money Multiplier and of coefficients 0.1 16819 0.123982 0.029012 Reserve Money, 1980-94 t-statistics 7.989295 5.182207 5.911975 Ex post, the money multiplier (k) is simply the Sources: Central Bank of Jordan; and IMF staff estimates. ratio of the stock of money and quasi money (MQM) to reserve money (RM): k = MQM/RM. Ex ante, the money multiplier is derived consistent with the cash- to-deposit ratio (C/D), the ratio of banks' cash in vaults to deposits (VC/D), the ratio of required re- during the 1970s and 1980s. Since 1989, the ratio serves to deposits (RR/D), and the ratio of excess re- has fluctuated within a narrow range and has been serves (excluding cash in vaults) to deposits (ER/D): relatively stable. k = (C/D + 1)/(C/D + VC/D + RR/D + ER/D). This In the absence of quarterly national accounts, the expression is derived from the following three quarterly path for money demand can be captured identities: with the seasonal information content in available historical data. Chart A3 presents the quarterly de- (1) MQM=C + D, mand for money as a percentage of the annual (2) RM = C + VC + RR + ER, and change in money demand. It appears that the largest (3) MQM = k* RM. expansion in money demand generally occurs during Identity (1) means that the money supply equals the first two quarters of the year, in line with the sea- the sum of cash outside banks (C) and deposits of sonal pattern for credit and investment. On average, the public in the banking system (D). Identity 75 percent of the year's money demand is reached (2) means that reserve money equals the sum of by midyear, and the last two quarters account for cash holdings outside of the banking system and only 25 percent.8 Except during 1988-92, when spe- total bank reserves (including banks' cash in cial factors affected the financial system, this sea- vaults). A simple manipulation substituting (1) and sonal pattern appears rather robust. (2) into (3) yields the ex ante definition of the money multiplier. The money multiplier, that is, the relationship be- 8On average, the annual distribution for money demand is tween reserve money and broad money, was affected 30 percent for the first quarter, and 45 percent, 15 percent, and 10 by four factors during 1980-94: the public's cash percent, respectively, for each subsequent quarter. holding over time, the licensed banks' cash in vaults,

Chart A3. Quarterly Demand for Money Chart A4. Currency Outside Banks (As a percent of annual change in money demand) (As percent of deposits)

Annual Distribution

Sources: Central Bank of Jordan; and IMF staff estimates. Quarterly Distribution

the legal rate for reserve requirements (which is the weighted average of legal rates applicable to licensed banks9 on various deposits), and the licensed banks' behavior toward constituting excess reserve (excluding cash in vaults). Accordingly, the following key constituents are analyzed for 1980-94.

Cash-to-Deposit Ratio The pattern of cash holdings is reflected in the change in the ratio of currency outside banks to total deposits in the banking system. This ratio has steadily decreased since 1980, except during Sources: Central Bank of Jordan; and IMF staff estimates. 1988-90 when unusual events (including the collapse of Petra Bank, the bankruptcy of several financial institutions, and the August 1990 regional crisis) temporarily undermined confidence in the banking system. This downward trend is the result 1980s and was sensitive to developments in infla- of various institutional and financial innovations, tion (Chart A4).10 including in particular the proliferation of means of Furthermore, the analysis of the quarterly cash-to- payments other than cash (such as credit cards or deposit ratio reveals a strong seasonal pattern associ- banking transfers); and increasing confidence in the ated with the seasonal consumption, or demand, of banking system. The cash-to-deposit ratio, which the private sector. On average, the ratio declines dur- exceeded 55 percent in 1980, declined to less than ing the first quarter, rebounds during the second and 30 percent in the early 1990s. The estimated equa- third quarters, and decreases sharply in the fourth tion indicates that the cash-to-deposit ratio declined quarter. However, the stability of this seasonality at an annual average rate of 1.7 percent during the was disrupted in the late 1980s and in 1990 as a re-

9Twenty-two licensed banks (including the Housing Bank) 10This is apparent from the following regression for the cash-to- have liabilities in the form of sight and time deposits payable on deposit ratio during 1980-93: C/D = -0.017*Time + 0.05infla- demand, transferable by check, or otherwise usable for making tion + Dummies 1988-90 + 34.6 (-17.8) (5.1) SER = 1.3%; R2 = payments. 0.98.

suit of the unusual events mentioned. Since 1991, posits, and by the exemption of advances for expon however, this seasonal pattern appears to be holding, promotion toward reserve requirements. although it is less pronounced. Banks' excess reserves at the Central Bank of Jor- dan (excluding remunerated deposits at the Central Ratio of Cash in Vaults to Deposits Bank) are less predictable than legal reserves because they are related to (1) the microeconomic situ- Because of improvements in banking manage- ation of each individual bank, (2) the interest rate ment and a deepening of the financial system, the and credit control (credit ceiling) policies, and ratio of cash in vaults of licensed banks to deposits (3) individual constraints on commercial banks' bal- decreased during the 1980s, from 1.3 percent in ance sheets. Three different phases ca.> be identified 1980 to 0.8-0.9 percent of deposits in the early in the evolution of the ratio of licensed banks' excess 1990s (Chart A5). Although banks' cash in vaults, reserves to deposits during the 1980s 'Chart A6). like other deposits at the Central Bank of Jordan, From 1980 to 1987, the ratio declined to 1 percent represents immediate and final payment, it (i.e., banks' cash in vaults) cannot currently count toward satisfying the reserve requirements. Until the treatment of cash holdings is changed so that they count as reserve assets, cash in vaults is independent of Chart A5. Banks'Cash in Vaults and other reserve assets in Jordan's monetary program. Reserve Requirements (As a percent of deposits) Legal and Excess Reserves Ratios The money multiplier is highly sensitive to the Banks' Cash in Vaults ratio of banks' required and excess reserves at the Central Bank to deposits. Although legal reserves are supposed to be easily predictable because they are known institutional instruments, the existence of different reserve requirements complicates the pre- dictability of this component. Legal reserve requirements are currently higher for commercial banks (14 percent) than for investment banks (9 percent) because of the authorities' belief that investment banks are better able to forecast their liquidity needs.11 In addition, with a legal reserve requirement of 5 percent, the Housing Bank was granted an exemption because of its involvement in long-term financing.12 As a result, the ratio of effective reserves to deposits is a weighted average of the legal rates for commercial and investment banks, and the fluctuations within the range reflect substitutions in the Legal and Actual Reserve Requirements public deposit behavior between these banks (Chart A5).13 In addition, projecting the actual rate for the legal reserve requirement is complicated by the requirement that reserves be held against interbank de-

nUntil 1988, legal reserve requirements for Jordan dinars dif- fered according to the nature and maturity of deposits (demand, time, or savings deposits). 12Since 1994, the Housing Bank is subject to a 15 percent legal reserve requirement on its commercial activities (which represent less than 1 percent of its total activities); and a 5 percent legal reserve requirement (which gradually increased to 9 percent in 1994) on its traditional activities. 13In 1993, deposits in commercial banks (excluding those of the Housing Bank) represented 71.4 percent of total deposits in the banking system. The total deposits of the Housing Bank and Sources: Central Bank of Jordan; and IMF staff estimates. other investment banks represented 18.3 and 10.3 percent, respectively, of total deposits in the banking system.

the Central Bank in 1992.14 In 1993, this component Chart A6. Excess Liquidity and Money of excess liquidity appears to have been largely ab- Multiplier sorbed, mainly through the newly introduced CDs denominated in Jordan dinars. It disappeared completely in 1994 as the remunerated deposits of three Licensed Banks' Excess Liquidity and six months matured. (As a percent of deposits) Money Multiplier The money multiplier is defined as the ratio of money and quasi money (M2 in the case of Jordan) to the narrow concept of reserve money.15 Consis- tent with the trends in the above constituents, the money multiplier increased in the early 1980s, up to 3 by 1987 (echoing the fall in the cash-to-deposit ratio), and decreased thereafter to 2.6 by 1990, partly in response to the regional financial and political turmoil. The money multiplier recovered to its pre-crisis level by 1991 and remained at about this level thereafter, reflecting declines in both the cash-to-deposit ratio and the licensed banks' excess reserves.

Money Multiplier Appendix III: Monetary Program Under Indirect Monetary Control As part of their medium-term structural reform program, the authorities moved from a system of direct credit controls16 to more market-based and indirect instruments of monetary control. In this new monetary environment, the program aims at forecasting the Central Bank's balance sheet in two stages. The first stage consists in ascertaining reserve money, consistent with the estimated money demand resulting from the overall macroeconomic baseline scenario (targeted growth, inflation, fiscal and external adjustments). The second stage essen-

Sources: Central Bank of Jordan; and IMF staff estimates. 14In 1991, the Central Bank promoted its three- and six-month remunerated accounts to sterilize the large inflows of foreign assets generated by labor remittances and the transfer to Jordan of workers' or returnees' lifetime savings from neighboring countries. These deposits accumulated to more than JD 350 million out from 5 percent, notwithstanding sharp fluctuations of total deposits of JD 700 million at the Central Bank in 1992. 15 around this trend. In 1988, the ratio collapsed and Money and quasi money includes currency in circulation and demand, saving, and time deposits of residents, denominated in turned negative, reflecting the emergence of the Jordan dinars and foreign currencies, with both the Central Bank banking system crisis. By the early 1990s, the ratio and the licensed banks (including the Housing Bank). Reserve had recovered to its pre-crisis level, and in the after- money is defined as the sum of currency in circulation (currency math of the August 1990 regional crisis, the excess outside banks and banks' cash in vaults) and demand deposits of reserves accounted for 2.5 percent of total deposits. licensed banks at the Central Bank (excluding remunerated deposits and CDs denominated in Jordan dinars). In line with the This latter level seemed to have been the historical recommendations of the IMF Statistics Department, a broader average level targeted by the licensed banks to pre- concept of reserve money would include all demand deposits at serve themselves against an unexpected shortfall in the Central Bank, particularly those of the financial institutions, public nonfinancial institutions, and the private sector. liquidity. However, it should be noted that the effec- 16 tive level of excess liquidity should also include the The system of direct monetary controls consisted of a maximum loan-to-deposit ratio of 90 percent and the 10 to 1 ratio of banks' remunerated deposits at the Central Bank, credit to capital. These ratios were reinforced by limits on the ex- which represented half the total banks' deposits at pansion of individual bank credit introduced in late July 1993.

tially forecasts the rest of the Central Bank's balance recent developments and prospects in the West Bank sheet, in order to project the volume of CDs denomi- and Gaza Strip. nated in Jordan dinars that the Central Bank should auction to manage total liquidity in the banking sys- Other Balance Sheet Items of the tem. The sequence in which the program is worked Central Bank out is illustrated by the following relationships: For the indirect system of control to have its de- (1) RM = MQM (targeted)/m (projected), sired impact, the monetary authorities must monitor (2) NDA = RM - NIR (targeted), and the Central Bank's NDA closely,18 in particular, gov- (3) CDs = CREDg (targeted) + CREDbs ernment borrowing from the Central Bank and its re- (projected) + OIN (net) (projected) - NDA. discount credit facilities, consistent with the targeted The first equation indicates the level of RM con- or desired level for net international reserves.19 On sistent with the projected money multiplier (m) and this basis, the Central Bank would control the liquid- desired broad money supply MQM. The second ity of the banking system by auctioning dinar- identity gives the level of the Central Bank's net denominated CDs and by intervening in the sec- domestic assets (NDA) consistent with the projected ondary market. reserve money and targeted (or desired) level of net Of the constituents of the Central Bank's net do- international reserves (NIR). The third identity indi- mestic assets, net claims on the Central Government cates central bank intervention in the CD market (CREDg)20 are projected in line with the contem- through its sales and purchases consistent with plated fiscal adjustment and required central bank fi- (1) the Central Bank's projected NDA; (2) the tar- nancing,21 and the seasonal pattern, which has been geted government borrowing at the Central Bank associated with government borrowing and deposits (CREDg); (3) the estimated amounts of rediscount at the bank in the past. at the Central Bank's credit facility windows Consistent with the program objective of building (CREDbs); and (4) the projected changes in other up official foreign exchange reserves, the Central items (net) (OIN). With respect to the conduct of monetary policy, the sequence is reversed: through its intervention in the CD market and its control over 18The Central Bank's net domestic assets are defined as the difference between the sum of (1) claims on the Central Government government borrowing and the rediscount credit fa- (general and own budgets), (2) claims on the Social Security Cor- cilities, the Central Bank tries to manage its net do- poration, (3) claims on municipalities and local government, 17 mestic assets consistent with the targeted floor of (4) claims on public nonfinancial institutions, (5) claims on the net international reserves, so as to keep reserve private sector, (6) claims on licensed banks, (7) claims on other money at levels conducive to the country's desired financial institutions, and (8) unclassified assets; and the sum of (1) dinar-denominated CDs, (2) licensed banks' remunerated de- money supply. posits with the Central Bank, (3) deposits of the Central Govern- ment (general and own budgets), (4) other public sector deposits, (5) deposits of financial institutions, (6) deposits of the private Reserve Money and Money Multiplier sector, (7) unclassified liabilities, and (8) capital accounts. Alter- natively, the net domestic assets of the Central Bank are defined The money multiplier is estimated in line with the as the difference between reserve money and the Central Bank's projections for public cash holdings, currency out- net international reserves. side banks, banks' cash in vaults, and legal and ex- 19Jordan's net international reserves consist of (1) gold, (2) its cess reserve requirements. The quarterly pattern IMF reserve position, (3) purchases from the IMF, (4) SDRs, (5) treasury bills and securities in foreign currencies, (6) balances should reflect the projected seasonal pattern of cash and deposits in foreign currencies, (7) bilateral accounts (net), to deposits, all changes in the legal rates of reserve and (8) other deposits in foreign currencies less deposits of inter- requirements, and the expected changes in commer- national institutions. Net international reserves are equivalent to cial banks' excess reserves at the Central Bank. Con- the Central Bank's net foreign assets, adjusted for purchases and repurchases from the IMF. sistent with the adjustment program in terms of 20 These consist mainly of advances to the Government, devel- growth, inflation, and fiscal and external adjust- opment bonds, and treasury bills. ments, the projected reserve money remains highly 2'By law, the Central Bank may grant an interest-free ordinary dependent on the quarterly demand for broad money. advance to the Government of up to 20 percent of domestic rev- The money demand function may be temporarily enues as estimated in the general budget law. It may also extend a profit advance to the Government from its share of net profits. As disrupted by the financial liberalization as well as by regards issuance of government bonds and treasury bills, the Cen- tral Bank is not committed by law to buy any amount of the public debt issues. (The nominal value of bonds issued by the Gov- 17Net claims on government, claims on licensed banks, and ernment cannot exceed 20 percent of the actual capital CDs denominated in Jordan dinars are the largest components of expenditures of the previous year; that of treasury bills issued the Central Bank. Other components are either known with cer- cannot exceed 25 percent of the average domestic revenues col- tainty (for example, capital accounts) or easily predictable (remu- lected in the past three fiscal years, or of total currency in circula- nerated deposits, other items (net), and net claims on nonbanks). tion, whichever is larger.)

Table Al. Sensitivity of the Monetary Program to Selected Parameters

Money Change in Change in Sales of Multiplier Reserve Money Money Demand CDs (net)

Baseline scenario 3.02 5.70 12.3 84 Elementary simulation1 Income elasticity of money demand2 — 1.0 I.I -12 Cash-to-deposit ratio3 -0.04 1.7 — -27 Reserve requirement4 0.08 -2.5 — 40

Source: IMF staff estimates. 'Different from baseline scenario. 2lncome elasticity of money demand 0.1 point higher. 3Cash-to-deposit ratio I point higher. 4Reserve requirement (legal or excess) I point lower.

Bank should gradually increase the stock of gross Sensitivity of the Monetary Program to foreign exchange reserves.22 Selected Parameters Claims on the banking system (CREDbs) consist of the credit facilities at the Central Bank's rediscount Under indirect instruments of monetary control, windows and the unsecured advances attached to the the monetary program is likely to be highly sensitive bailout of bankrupt financial institutions during the to its achievements in terms of growth, inflation, and 1980s. First, concerning the rediscount facilities, the fiscal and external adjustments; financial liberaliza- Central Bank operates a standard rediscount facility at tion and financial innovations; and monetary and an interest rate of 8.5 percent; because of the commer- banking developments and prospects in the West cial banks' excess liquidity, they have not drawn on Bank and Gaza Strip. In particular, the money multi- this facility since 1989. However, the shift toward mar- plier (and thus the broad money supply) may become unstable because of public behavior with re- ket-based instruments for monetary control will proba- 25 bly make this facility more attractive to the banks when spect to holding cash or the banks' behavior with the excess liquidity is absorbed. The Central Bank also respect to building excess reserves. In addition, as operates two preferential credit facilities. Although experienced by countries that have recently switched to more market-based instruments of monetary con- they are of limited quantitative significance at pres- 26 ent,23 the Central Bank must exercise strict quantitative trol, the demand-for-money function may not re- control over access to its refinance and rediscount fa- main stable, at least during the transition period. The cilities, both on standard and preferential credit facili- use of the Jordan dinar as legal tender in the West ties, and rationalize the interest rates on these facilities. Bank and Gaza Strip and the opening of banks deal- Second, with respect to the unsecured advances to ing in Jordan dinars may contribute to this instability those banks that experienced serious difficulties during of the demand for money. In this regard, some fluc- the 1980s,24 the Central Bank may expect to gradually tuations in the money multiplier and in reserve recover part of these unsecured assets. money may result from variations in three selected variables: the income money velocity, the ratio of The amount of dinar-denominated CDs to auction cash to deposits, and the ratio of reserve require- is residually determined so as to balance the balance ments to deposits (Table Al). sheet of the Central Bank. A money demand income elasticity that is 0.1 point higher than the baseline scenario would re- 22Defined as net international reserves excluding gold, foreign quire an additional increase in money supply of exchange deposits of residents at the Central Bank, and claims on the Central Bank of Iraq. 23The Central Bank charges 3.5 percent for its export credit fa- 25The currency-to-deposit ratio is likely to vary for economic cility on documentary credit (90 percent of the value of the credit reasons, including, in particular, the opportunity cost of holding may be rediscounted for a period of up to 18 months). It also oper- cash, but also for political reasons, especially confidence in the ates a credit facility for specialized credit institutions (such as the peace process and the development of a sound banking system in Industrial Development Bank, the Housing Corporation, and the the West Bank and Gaza Strip. Agricultural Credit Corporation) for a period of up to five years. 26See for example, W. Tseng and R. Corker, Financial Liberal- 24Total injection of liquidity toward the restructuring of failed ization, Money Demand, and Monetary Policy in Asian Coun- banks amounted to $400-500 million, of which the Petra Bank tries, IMF Occasional Paper No. 84 (Washington: International alone accounted for $300 million. Monetary Fund, 1991).

about 1.1 percentage points, a supplementary in- about 1.7 percentage points and would require the crease of reserve money of about 1 percentage purchase of JD 27 million in CDs compared with point, and, consistent with the underlying level of the baseline scenario. More important, a decrease in net international reserves, the repurchase of some the ratio of reserves to deposits (in either legal or JD 22 million in CDs from the baseline scenario. excess reserves) would imply lower reserve money Similarly, a ratio of cash to deposits that is 1 per- growth of about 2.5 percentage points (compared centage point higher by the end of the year would with the baseline) and the auction of an additional allow for an additional increase in reserve money of JD 40 million in CDs.

Patricia Alonso-Gamo and Ahsan Mansur

ordan's economic expansion accelerated in the receipts from services; about one-half of the external 1970s in the wake of the regional economic boom, current account receipts in Jordan derived from ser- which opened up opportunities for Jordan's exports vices, mostly remittances and receipts from tourism and for employment of Jordanians in the area; these and travel. Arrears on external debt-service pay- developments led in turn to a markedly higher in- ments began to accumulate in 1989. flow of external grants from neighboring oil-exporting countries. This process continued through the mid-1980s, even when the regional economies Debt Management Strategy After the started to experience recessions resulting from the Debt Crisis collapse of petroleum prices. As payments difficulties emerged, and as part of an overall strategy to achieve external current ac- Emergence of the External Debt Crisis count adjustment, the authorities initiated an external debt-management policy in 1989 aimed at ob- Despite the ensuing substantial reductions in re- taining debt relief through rescheduling, lengthening mittances and foreign grants, the authorities at- the maturity structure of debt, and reducing the debt tempted to maintain their domestic economic poli- burden in relation to GDP over the medium term. cies unchanged during 1984-88. Instead of adjusting This strategy has four major elements. First, the au- to the lower financing available, Jordan resorted in- thorities concluded a restructuring agreement with creasingly to foreign borrowing on commercial Paris Club creditors in July 1989 and regularized re- terms; as a result, its external outstanding public or lations with other bilateral creditors along terms publicly guaranteed debt built up steadily during this similar to those obtained from the Paris Club. Sec- period to $8 billion by the end of 1988 (127 percent ond, they initiated negotiations with the commercial of GDP), while its outstanding short-term debt banks to obtain a multiyear rescheduling of obliga- reached $400 million. By that time, with the slow- tions falling due, with provisions for new money and down in economic activity and high real interest the option for debt conversion at a discount. Third, rates in the world market, the debt burden had they limited all new borrowings to longer-term ma- reached unsustainable proportions. turities, mostly at concessional interest rates. Along In response to these developments, Jordan's exter- these lines, they also converted all short-term debt to nal scheduled debt-service payments rose to nearly medium-term maturities and incurred no new short- $1.4 billion in 1989; at the same time, receipts from term or nonconcessional debt in the maturity range remittances declined and exports slowed markedly, of 1-12 years. Finally, Jordan canceled most new so that the scheduled debt service as a percentage of commercial borrowings that were in the pipeline; the exports of goods and services increased to 46 per- cancellation of a military aircraft purchase agree- cent in 1989, about five times the corresponding ment reduced the debt burden by $1 billion. ratio in 1983.l Jordan's external vulnerability was exacerbated, and its creditworthiness impaired, by the high debt-to-GDP ratio and high dependence on Paris Club Agreement, 1989 About half of Jordan's public and publicly guaranteed external debt was owed to Paris Club credi- 'This compares with 16 percent for the developing countries in tors, in the form of either official development assis- general and 27 percent for the heavily indebted countries as ob- tance or officially guaranteed suppliers' credits. served in 1989. The ratio of external debt to exports of goods and services reached 213 percent in 1989, compared with 290 percent Another 15 percent of the debt was owed to other for the heavily indebted countries and 135 percent for the devel- official creditors, such as Arab countries and the for- oping countries as a group. mer Soviet Union. At the request of the Government

of Jordan, Paris Club creditors rescheduled Jordan's working abroad. In the event, Jordan was unable to external debt on July 19, 1989. The 18-month con- discharge its external debt-service obligations and solidation period (July 1, 1989-December31, 1990) made payments only to official creditors, primarily coincided with the term of the first stand-by to multilateral organizations. Cumulative overdue arrangement. obligations by the end of 1991 amounted to almost Under the Paris Club agreement, all arrears and $800 million, and the amount falling due in 1992 amortization falling due during the consolidation pe- was estimated at that time at $1.1 billion. riod on account of debt from before the cut-off date2 The authorities were aware that Jordan was unable and six months of interest payments were subject to to normalize payments relations with its creditors rescheduling. Total debt relief from Paris Club credi- through regular debt servicing in cash and that ex- tors was estimated to be $587 million over the 18- ceptional financing through restructuring of the debt month period.3 Jordan was also required to offer and debt-service payments was necessary. It also be- other bilateral official creditors terms not better than came clear that capitalization of the unpaid debt-ser- those offered to Paris Club creditors. This provision vice obligations would lead to a buildup of external allowed Jordan to receive additional debt relief debt and that growth in debt service would exceed worth $645 million during the same period. the growth potential for the external current account receipts, making the overall situation unsustainable in the medium to long run. Negotiations with Commercial Bank Creditors Moreover, the new medium-term outlook implied Unlike many other heavily indebted developing that, in addition to limiting new borrowing to only countries, Jordan's debt to commercial banks was concessional terms and longer-term maturities, the small in relative terms, accounting for about 15 per- revised debt-management strategy would require cent of the outstanding public and publicly guaran- more concessional rescheduling terms from Paris teed external debt. Initially, the authorities sought a Club official bilateral creditors. Jordan would have multiyear rescheduling agreement with commercial to undertake operations to reduce debt and debt ser- bank creditors, extending the repayment period over vice through commercial bank creditors and several a longer term. An informal agreement was reached smaller official bilateral creditors. with the steering committee of commercial banks in October 1990, under which Jordan agreed to pay all interest obligations in arrears through the end of Paris Club Rescheduling Agreements, March 1990, and the banks agreed to reschedule or 1992 and 1994 refinance all remaining interest and all amortization Immediately after the approval of a new 18-month obligations. In the event, however, agreement could stand-by arrangement with the IMF, Jordan con- not be reached by the time the regional crisis erupted cluded the second rescheduling agreement with the in August 1990. Moreover, as operations to reduce Paris Club in February 1992. Under the agreement, debt and debt service were soon gaining momentum Jordan obtained debt relief through rescheduling of in negotiations between the commercial banks and a all interest and amortization on debt incurred before number of heavily indebted developing countries the cut-off date that would fall due during an (including Argentina and Brazil), the multiyear 18-month period beginning on January 1, 1992, and rescheduling agreement was abandoned. all amortization arrears and 50 percent of interest arrears accumulated up to the end of 1991. The terms allowed for longer repayment and grace periods than Changes in Debt Strategy the 1989 agreement. Overall, the debt relief from Paris Club creditors was estimated at $771 million The August 1990 crisis severely aggravated Jor- over the 18-month period, and the amount for 1992 dan's external situation because of the loss of export (including the rescheduling of arrears) was esti- markets for goods and services in neighboring coun- mated to be $603 million. The agreement provided, tries and of foreign grants. Moreover, there was a on a voluntary and bilateral basis, for selling or ex- sharp reduction in remittance flows, largely resulting changing debt, in the framework of debt for nature, from the massive return of Jordanians who had been debt for aid, debt for equity swaps, or other local currency debt swaps. The consolidation period was further extended to correspond to the extension of 2The cut-off date for Jordan was January 1, 1989, which was the stand-by arrangement. six months prior to the beginning of the consolidation period. After the IMF's Executive Board approved on 3The maturity period for the rescheduled amortization obligations was 9.3 years with 4.8 years of grace, and for rescheduled May 25, 1994 a new three-year program for Jordan interest obligations, the maturity period was 10.3 years with 5.8 under the extended Fund facility (EFF), the Jordan- years of grace. ian authorities sought further debt rescheduling from

Paris Club creditors for debt payments falling due services for developing countries facing debt-servic- during the program period. In recognition of Jor- ing problems. dan's efforts at macroeconomic stabilization and The increasing emphasis on operations to reduce structural reform, creditors granted Jordan more fa- debt and debt service was also motivated by a num- vorable rescheduling terms than under the 1992 ber of other considerations. First, given the balance rescheduling agreement. They rescheduled all debt- sheet situation of some commercial banks, the as- service obligations that had not already been sessment at that time was that it could prove difficult rescheduled (interest and principal) on debt incurred to arrange adequate interest refinancing through before the cut-off date; and current maturities (prin- concerted new money loans.6 Second, apart from the cipal and interest) as a result of the 1989 agreement financing implications, the potential absence of such falling due during July 1, 1994-May 31, 1997 on concerted new financing raised comparability prob- graduated repayment terms (over a period of 18 lems vis-a-vis other official creditors. Third, there years).4 In addition, all interest payments under the was growing evidence that many commercial banks 1992 Paris Club agreement falling due during July 1, were willing to dispose of their claims on Jordan, 1994-June 30, 1995 were also deferred, with repay- often at substantial discounts. This shift in the over- ments over five years, beginning December 31, all debt strategy also gained momentum in Jordan in 1997. Total debt relief obtained as a result was esti- 1992 when, following the record level of repatriation mated at about $1.2 billion (including $50 million in of returnees' savings after August 1990, Jordan's of- deferred interest payments) over the consolidation ficial foreign exchange reserves increased markedly. period. Assuming similar terms from other official By the end of December 1992, the foreign assets of bilateral creditors, overall debt relief during the EFF commercial banks amounted to $1.4 billion, of period would amount to about $1.4 billion. which some $600 million was deposited with the Central Bank of Jordan.7 Revised Debt Strategy vis-a-vis With a view to improving Jordan's debt-servicing Commercial Banks capacity and creditworthiness in the longer term, the authorities revised their strategy further, seeking a After the regional crisis, it became evident that, to more comprehensive restructuring of the debt stock attain medium-term balance of payments viability, on the basis of a market-based financing menu.8 Jordan should undertake operations to reduce mar- Given the favorable circumstances, in early 1993, ket-based debt and debt service. There were growing the Jordanian authorities stepped up their efforts to concerns regarding the impact of the external debt buy back debt. By then, Jordan was able, through overhang on Jordan's growth potential.5 While re- cash buybacks, to extinguish external debt owed to peated principal rescheduling and interest refinanc- commercial banks for a total amount of about $600 ing (through concerted new money loans) could pro- million at a cost of $115 million. In addition, the vide the necessary short-term cash-flow support, country devoted $39 million during the earlier part medium-term projections indicated that the resulting of 1993 to buybacks from commercial banks, for debt stock and debt-service ratios would remain debt with a face value of $100 million. high, well above the average level of some 40 per- Negotiations proceeded in parallel and, finally, on cent of GDP and 27 percent of exports of goods and June 30, 1993, the Jordanian authorities reached an agreement in principle with London Club creditors on the terms for restructuring Jordan's external debt 4These terms were being applied for the first time to a lower- to foreign commercial banks. Jordan continued to middle-income country. Specifically, foreign and domestic assessment of country risk may not improve when the stock of contractual debt surpasses 6As reflected in the case of other major debtor countries such economic agents' perceptions of the debtor country's capacity to as Mexico, Brazil, and Argentina, because of more stringent regu- service the debt. These uncertainties are perceived to increase the latory provisioning requirements on developing country exposure cost of capital in excess of the normal return and vitiate the envi- and the prospects of seemingly endless episodes of concerted new ronment for investment over a prolonged period. Issues related to money packages based on existing bank exposure, the erosion of debt overhang, linkage between domestic and foreign indicators cohesion within the banking community intensified and made it of country risks, and impediments to access to credit have been extremely difficult to implement new money packages. covered in Michael P. Dooley and others, Debt Reduction and 7The proportion of commercial banks' foreign assets held with Economic Activity, IMF Occasional Paper No. 68 (Washington: the Central Bank of Jordan was higher than the legal requirement International Monetary Fund, 1990); Hoe E. Khor and Liliana (35 percent) because the banks found it a preferred option, given Rojas-Suarez, "Interest Rates in Mexico: The Role of Exchange the excess liquidity prevailing in the economy at that time. Rate Expectations and International Creditworthiness," IMF 8The menu icluded such options as reduced interest par ex- Working Paper 91/12 (Washington: International Monetary Fund, change, discount exchange, and buyback at a discount, which al- 1991); and Joseph Stiglitz and Andrew Weiss, "Credit Rationing lowed the authorities, in particular, to bring the contractual value in Markets with Imperfect Information," American Economic Re- of commercial bank claims and other debt more in line with Jor- view, Vol. 71 (June 1981), pp. 393^10. dan's debt-service capacity.

pay 30 percent of interest due until the closing date, tions by Paris Club creditors, the stock of arrears December 23, 1993. The agreement covered $736 outstanding has been gradually reduced from a peak million in principal and about $121 million in inter- of $780 million at the end of 1991 to about $120 est arrears. A recalculation of interest arrears using million at the end of 1992 and $11 million at the end the lower interest rates that prevailed at the time of 1993. The authorities have reported progress in saved Jordan about $33 million. Banks opted to formalizing rescheduling agreements with some of- exchange 32.8 percent of the debt ($243 million) ficial bilateral creditors not covered by the Paris for discount bonds9 and 67.2 percent ($493 million) Club agreements. for par bonds;10 only a small fraction of the debt Finally, in line with the revised debt strategy, in ($0.25 million) was bought back in the context of the 1992 Jordan completed an agreement with Russia, debt- and debt-service-reduction package, as the whereby $614 million of its debt to the former So- buyback price of 39 cents per dollar was well below viet Union was extinguished at a large discount. The the prevailing market price. On the closing date, Jor- first payment for the buyback was made in cash for dan made a cash payment of 10 percent of interest $88 million in 1992; further installments were set- on eligible debt to be exchanged for par bonds and tled during 1993 and 1994 in the form of commodity 50 percent of interest on eligible debt to be ex- exports. Similarly, official debt to Brazil was bought changed for discount bonds. The balance of interest back at a large discount. arrears was exchanged for floating-rate, past due in- As a result of its sound external debt management, terest bonds (PDIs).11 The overall cost of the pack- geared to restoring access to world financial mar- age was about $150 million, financed by a draw- kets, Jordan's creditworthiness has continued to im- down of gross official reserves and the issuance of prove. By limiting new borrowing and pursuing debt CDs denominated in foreign currency.12 buybacks at substantial discounts, Jordan reduced its ratio of debt to GDP from 193 percent in 1990 to less than 110 percent at the end of 1994. Successive Paris Negotiations with Other Official Creditors Club reschedulings lowered debt service as a share The authorities have also sought rescheduling or of exports of goods and services from 45 percent in refinancing agreements from all other external credi- 1990 to 25 percent at the end of 1994. Moreover, in tors (including suppliers, and official bilateral credi- the wake of the peace process, $800 million of debt tors other than the Paris Club) on terms no more fa- forgiveness was granted to Jordan, mostly by the vorable than those accorded to Paris Club creditors. United States ($702 million). Debt-for-equity and As a result of the rescheduling of overdue obliga- debt-for-nature swaps with Switzerland and Ger- many have already been concluded, and more are likely to follow. 9A collateralized discount bond option, at 65 percent of face Nevertheless, the debt-to-GDP ratio would still value, with a bullet payment after 30 years, and interest at a floating rate of six-month LIB OR (London interbank offered rate) remain high—96 percent by the end of 1998—even plus 13/i6 a year. Principal was fully collateralized, in addition to a under the assumption of continuing rapid growth of six-month rolling interest guarantee. exports, yearly remittances of $1.1 billion, and sub- 10Collateralized par bonds, with a 30-year bullet payment. The stantial foreign direct investment. Whether these exchange bonds pay interest at a fixed rate of 4 percent for the would materialize depends crucially on private sector first four years, 5 percent in year five, 5.5 percent in year six, and 6 percent thereafter. The principal amount and six-month rolling perceptions of the debt and prospects for servicing interest guarantee were fully collateralized. the debt: the existing large debt overhang, however, 1 •Repayable in 19 semiannual installments with a three-year could act as a disincentive to private sector invest- grace period, and carrying interest at six-month LIBOR plus l3/i6 a ment. To reassure the markets, Jordan would need year. The PDI bonds were not collateralized. 12As follows: $21 million in interest collateral ($6 million for not only to continue its sound debt-management poli- discount and $15 million for par bonds); $97 million for principal cies, but also to obtain sufficient new concessional fi- collateral ($24 million for discount and $73 million for par nancing and, possibly, further debt relief. Continued bonds); $29 million as a cash down payment for interest arrears strong adjustment and proper incentives through ($21 million for discount and $8 million for par bonds); and eliminating distortions would also contribute to the about $3 million in transaction fees. For the balance of interest arrears ($91 million), PDI bonds were issued. The cost of the buy- attainment of sustainable export-oriented growth and back was marginal. help improve Jordan's creditworthiness.

Patricia Alonso-Gamo

n late 1988, in response to the balance of pay- of merchandise exports; in particular, workers' re- I ments crisis, Jordan embarked on a reform of its mittances were equivalent to 77 percent and travel external tariff and trade system. Trade liberalization receipts to 41 percent of exports of goods. gained momentum in 1990, when tariff and nontariff Export diversification has continued to increase trade barriers were unilaterally reduced.1 These mea- in recent years. Based on the Standard International sures were complemented by a reform of the En- Trade Classification (SITC) of commodities, Jor- couragement of Investment Law and institutional re- dan's exports in 1988-91 were dominated by tradi- forms in support of the industrial sector. tional goods, mainly raw materials such as potash The ongoing reforms seek to better integrate the and phosphates. Other important exports are phar- Jordanian economy with the rest of the world by maceuticals, detergents, and fertilizers, together improving external competitiveness to gain market with other manufactures, food, and live animals. share in foreign markets and by promoting a more But phosphates and potash, which accounted for efficient allocation of resources—especially toward 34.1 percent of exports in 1989, accounted for only tradable goods—to widen the country's narrow 19.4 percent of exports in 1994. By contrast, the export base. On both counts, Jordan has registered share of nontraditional exports has continued to in- a notable degree of success so far. Its external crease: manufactures and chemicals accounted for competitiveness (as measured by the real effective 29 percent of exports in 1989 and 34 percent in exchange rate) has improved considerably since 1994. As for imports, there has been a shift toward 1988, as evidenced by the growth of nontradi- capital goods and raw materials: the share of capital tional exports and the diversification of its trading goods increased from 21 percent in 1989 to 26 per- partners. As for internal competitiveness, the trad- cent in 1994, allowing for sustained growth with ables sector has become more competitive, with lower imports. domestic production shifting toward the tradables Jordan's exports and imports are not geographi- sector since 1988.2 The continuing liberalization of cally concentrated, and there has been some further Jordan's trade system is a cornerstone of its out- geographic diversification of exports over the last ward-oriented growth strategy, and will enable the five years (Table 7.1). The principal destination of country to benefit from regional developments and exports in 1994 was other Arab countries (42.4 per- new opportunities arising in the wake of the peace cent), particularly Saudi Arabia, and India (11.1 per- process. cent), the European Union (EU) (5.1 percent), and Japan (1.6 percent). The relatively low share of the EU and Japan may be partly attributable to the Structure of Trade still-dominant role of barter and protocol trade, the limited list of manufacturing products, as well as Jordan's trade is characterized by an unusually institutional barriers to access to those markets. Iraq large imbalance: the merchandise trade deficit remains one of the most important sources of amounted to 31.9 percent of GDP in 1994 (down imports, especially for crude oil, along with the from 40.9 percent of GDP in 1993). In fact, in 1994, EU (33.2 percent of the total) and the United States the value of receipts in the services account (factor (9.9 percent), whereas Japan's share (4 percent) is plus nonfactor) was almost twice (191 percent) that relatively small. Although more than 40 percent of Jordan's exports are directed toward Arab countries, only 22 percent of imports come from Arab 'Jordan's effort to liberalize the trade regime was supported by countries. Jordan maintains a trade surplus with an Industry and Trade Policy Adjustment Loan from the World Saudi Arabia and India, but has very large bilateral Bank. 2The share of tradables in GDP rose by 25 percent during trade deficits with Iraq, the EU, and the United 1986-92, and that of nontradables declined by 8 percent. States.

Table 7.1. Direction of Foreign Trade

1989 1990 1991 1992 1993 1994"

In millions of Jordan dinars

Domestic exports, f.o.b. 534.1 612.3 598.6 633.8 691.3 793.9 Arab countries 241.3 258.9 172.3 222.4 285.3 337.0 European Union 25.1 22.1 18.5 19.1 27.9 40.4 India 94.9 129.1 109.6 96.4 65.9 88.1 Japan 18.2 13.0 10.6 12.1 9.8 12.6 Eastern European countries 42.7 24.4 27.8 15.7 32.9 13.7 China 11.0 18.0 32.6 14.0 16.5 8.2 Other countries 100.9 146.8 227.2 254.1 253.0 293.9 Imports, c.i.f. 1,230.0 1,725.8 1,710.5 2,276.0 2,453.6 2,362.6 Arab countries 337.6 348.4 338.6 524.0 498.8 523.1 European Union 365.7 494.1 512.2 650.1 753.8 794.7 France 72.6 97.9 73.6 78.1 98.5 II 1.8 Germany 80.8 101.2 133.2 186.7 202.8 184.4 Greece 6.0 6.9 15.9 9.4 15.2 9.4 Italy 51.2 67.6 73.6 110.8 134.9 138.6 United Kingdom 74.0 89.4 77.8 108.4 127.9 120.8 Others 81.1 131.1 138.1 156.6 174.5 229.7 Japan 45.7 54.3 61.1 132.2 123.6 93.6 United States 170.8 299.5 178.2 246.2 311.5 232.9 Eastern European countries 55.8 85.3 85.6 129.7 151.9 137.9 China 21.9 25.0 29.1 52.1 50.6 62.4 Other countries 232.5 419.2 505.7 541.7 563.4 518.0 In percent of total Domestic exports, f.o.b. 100.0 100.0 100.0 100.0 100.0 100.0 Arab countries 45.2 42.3 28.8 35.1 41.3 42.4 European Union 4.7 3.6 3.1 3.0 4.0 5.1 India 17.8 21.1 18.3 15.2 9.5 II.1 Japan 3.4 2.1 1.8 1.9 1.4 1.6 Eastern European countries 8.0 4.0 4.6 2.5 4.8 1.7 China 2.1 2.9 5.4 2.2 2.4 1.0 Other countries 18.9 23.0 38.0 40.1 36.6 37.0 Imports, c.i.f. 100.0 100.0 100.0 100.0 100.0 100.0 Arab countries 27.4 20.2 19.8 23.0 20.3 2.1 European Union 29.7 28.6 29.9 28.6 30.7 33.6 France 5.9 5.7 4.3 3.4 4.0 4.7 Germany 6.6 5.9 7.8 8.2 8.3 7.8 Greece 0.5 0.4 0.9 0.4 0.6 0.4 Italy 4.2 3.9 4.3 4.9 5.5 5.9 United Kingdom 6.0 5.2 4.5 4.8 5.2 5.1 Others 6.6 7.6 8.1 6.9 7.1 9.7 Japan 3.7 3.2 3.6 5.8 5.0 4.0 United States 13.9 17.4 10.4 10.8 12.7 9.9 Eastern European countries 4.5 4.9 5.0 5.7 6.2 5.8 China 1.8 1.4 1.7 2.3 2.1 2.6 Other countries 18.9 24.3 29.6 23.8 23.0 21.9

Source: Central Bank of Jordan. 'Preliminary actuals.

Trade Regime and Trade Policy institutional inefficiencies that severely hindered its Reforms Since Late 1988 export and industrial sector performance. Although the average tariff rate on the value of total imports was about 19 percent, because of widespread ex- Trade Regime Prior to the Reforms emptions, private firms and households in Jordan Jordan's trade regime in the late 1980s was char- were shouldering a high burden of import taxation. acterized by high tariff and nontariff barriers and by The maximum tariff rate was 318 percent, while the

©International Monetary Fund. Not for Redistribution Trade Regime and Trade Policy Reforms Since Late 1988 bulk of imports were imported duty free. Nearly those of the second category, thereby reducing the thirty institutions were exempt from paying import share of domestic manufacturing protected by such duty, which implied that about 51 percent of imports restrictions from 40 percent in early 1988 to less into Jordan were not subject to customs duty. Non- than 7 percent in 1990. In 1995, in the context of the tariff barriers were widespread, with about 40 per- World Bank Agriculture Sector Adjustment Loan cent of domestic manufacturing activities being pro- (ASAL), the Government eliminated the remaining tected through quantitative restrictions in 1988. The restrictions (with the exception of those on some es- high degree of variation in the tariff rates (between sential food items). zero and 318 percent, with a standard deviation of Pricing policy also plays an important role in re- 26.1 percent), combined with these nontariff barri- stricting imports. The Ministry of Supply sets distrib- ers, led to a high degree of variation in effective ution margins on imports of consumption goods and rates of protection. on 19 consumption goods produced in Jordan. The As a result of the protective system, the overall in- authorities set retail prices for fresh food and vegeta- centive structure favored production for the domes- bles. Price interventions cover 11 percent of domestic market, despite its limited size. There were no tic consumption. The Government also provides con- significant export incentives to even partly offset the sumer subsidies on sugar, rice, powdered milk, and bias against exports. By the late 1980s, the long- wheat, which help reduce the domestic prices of term implications of the extensive protection of the these goods and may discourage production. domestic economy became apparent. The import- • Tariff reform. In designing the reform, the au- substitution policy promoted and protected an indus- thorities were constrained by the need to protect rev- trial sector that was generally inefficient, and the enues from customs duties because domestic tax po- growth potential of the nontraditional export base re- tential was impaired by the underdeveloped nature of mained unexploited. This trade policy stance, to- the tax system and administrative capacity. They gether with an appreciating exchange rate in real ef- therefore lowered high tariffs while raising low tar- fective terms, contributed to a low and stagnant level iffs so that the average tariff was broadly unchanged. of nontraditional exports. According to some esti- The effect of this act was to reduce the discrimination mates at that time, the protection regime imposed a implied by a more differentiated structure, producing heavy burden on consumers and industries that used a more uniform tariff regime. Two groups of com- the products of the protected sectors (equivalent to modities required special treatment. First, tariffs on a 9.6 percent of GDP).3 short list of goods, primarily mass consumption items amounting to 6.5 percent of dutiable imports, were kept at zero in the face of the overall tariff re- Policy Changes in the Trade Regime form. The main items on this list included capital • Phasing out of quantitative restrictions. To fos- goods (3.6 percent of dutiable imports), cereals ter a broadly neutral system of incentives, the au- (1.8 percent), books and newspapers (0.6 percent), thorities lowered effective protection by shifting and gold for financial transactions (0.5 percent). Cap- from quantitative restrictions to tariffs and, subse- ital goods were included as an integral part of the lib- quently, by reducing tariff barriers as a whole. In Au- eralization of investment controls (announced in gust 1988, they removed import bans on thirty cate- 1989), under which all investment licensing was re- gories of products and replaced them with tariff moved and a cumbersome discretionary tariff exemp- protection. Following this shift, the remaining quan- tion approval process was replaced with a uniform titative restrictions can be classified into four cate- tariff of zero on capital goods. Second, for luxury gories: (1) imports of 5 products—tomato paste, goods the total tariff was split into a basic tariff and a fresh milk, certain dairy products, mineral water, and consumption tax, the tariff being subject to a statu- table salt—remained banned (cigarettes were subject tory legal maximum. The consumption taxes on im- to quota); (2) 11 imported luxury, predominantly ports were collected at the import stage, while taxes consumer, items were banned on a time-bound basis on similar domestically produced items were intro- for balance of payments reasons; (3) products whose duced in connection with the general sales tax in imports were regulated for a variety of reasons, in- 1994. cluding health and safety; and (4) mass consumption For the remaining 93 percent of the items subject goods imported by the Ministry of Finance to protect to tariffs, the range between the maximum and mini- the poor. At the next stage, the authorities eliminated mum tariffs was reduced to 45 percentage points in a large number of quantitative restrictions, including two further stages over a two-year period. First, the maximum tariff rate was reduced to 60 percent, and

3 the minimum tariffs (except those on a few basic Jordan: Policies and Prospects for Small and Medium-Scale Manufacturing Industries, Report No. 6848-JO (Washington: items) were increased to 5 percent in 1990; this nar- World Bank, 1988). rowing of the range reduced the variation in tariffs

measured in terms of the standard deviation to 17.5 traders who would not require an import license compared with 26.1 in 1988. Second, in early 1992, once registered, is under consideration. the maximum rate was reduced to 50 percent, and • Promotion of exports. Exports are generally not the standard deviation further declined to 15.9. Be- subject to taxes, and there are no explicit restric- cause of measures to increase the minimum tariff tions on exports from Jordan. However, to over- rate to 5 percent, the revenue loss resulting from come the hidden bias against exports, a number of these two rounds of tariff rationalization was less initiatives have been undertaken since 1989 to pro- than JD 1 million. mote exports: a drawback system for import duty In June 1993, as part of the move toward the in- and taxes paid on inputs used in exports was intro- troduction of the general sales tax, the coverage of duced; measures streamlining administrative proce- the existing consumption tax was expanded while dures were implemented; the requirements for ad- certain specific unit taxes were converted into ad mission of intermediate inputs were eased; and the valorem, and other rates were raised to 10-20 per- arrangements for export finance were rationalized, cent. In October 1994, the Government lowered all as were incentives for investments in general. The tariffs of more than 50 percent to 50 percent (to- Government upgraded the existing export-promo- bacco, alcohol, and cars) or less, affecting 21 per- tion institution—the Jordan Commercial Center cent of imports. It also announced the reduction in Corporation (JCCC)—giving it primary responsibil- tariff bands to six rates (5, 10, 20, 30, 40, and 50 ity for export and investment promotion. Moreover, percent). These measures were motivated mainly by to rationalize and improve export financing, the the required synchronization with the newly intro- Central Bank of Jordan improved the operation of duced general sales tax, which is levied, unlike the its export rediscount facility in 1990 and 1992 by consumption tax, on amounts that include customs extending both the coverage and duration of export duty. Further tariff reductions are being undertaken credits. At the same time, it reduced the interest rate in the context of the new extended arrangement subsidy available to exporters and allowed the inter- with the IMF est rate for export finance to vary in line with com- The trade reforms initiated in late 1988 resulted in mercial rates. a more uniform tariff regime, with reduced variance in tariff rates: the import-weighted average tariff was reduced from 34.4 percent in 1987 to 25 percent in Opening Up the Economy 1994; and the tariff coefficient of variation was reduced from 167 in 1987 to 88.5 percent in 1994.4 Jordan has initiated negotiations on accession to The increased uniformity was achieved mainly by the World Trade Organization (WTO) and for a trade reducing the combined tariff and surcharges at the agreement with the EU. Numerous benefits may de- upper end of the spectrum. In May 1995, under the rive from WTO membership: most-favored-nation ASAL, the Government eliminated the remaining treatment, the use of established settlement proce- quantitative restrictions except on a few essential dures for disputes, active participation in the formu- food items, and converted them to tariffs. Import- lation of multilateral trade rules, and access to vari- licensing requirements were eliminated (except for ous organizations that facilitate international trade. It protocol trade and product standards).5 The intro- would also help the authorities lock in trade liberal- duction of an import registration system, switching ization measures and resist domestic protectionist from product-based licensing to a list of approved pressures, because rules of the General Agreement on Tariffs and Trade (the WTO's predecessor) foster the use of nondiscriminatory price-based instru- 4In 1994, of a total of 6,735 tariff lines, 330 (4.9 percent) were ments (tariffs); require equal tax treatment for do- zero rated, covering 16.7 percent of imports; 1,302 (19.3 percent) mestic and imported goods; and constrain the use of had a tariff of 5 percent and accounted for 24.6 percent of all im- quantitative restrictions and trade measures—such ports; 672 (10 percent) had tariffs of 10 percent and accounted for 13.2 percent of all imports; 1,229 lines (18.2 percent) had rates as surcharges—for exclusively balance of payments between 15 and 30 percent, accounting for 20.3 percent of all im- reasons. ports; 1,926 (28.6 percent) had tariffs between 35 and 45 percent In the context of a new Mediterranean strategy, and accounted for 12.8 percent of imports; 999 (14.8 percent) had the EU is working toward establishing a Euro- tariffs of 50 percent and accounted for 7.1 of imports; and 277 had tariffs in excess of 50 percent and accounted for 5.3 percent Mediterranean Economic Area, creating a 33-coun- of all imports. Imports tend to be in commodities with lower try free trade zone within 12 years. Agreements have tariffs and production in sectors with higher tariffs. Weighted av- already been signed with Morocco, Tunisia, and Is- erage nominal rates of protection were quite low in agriculture rael. Against this backdrop, Jordan has initiated its (8.9 percent) and mining (1.0 percent) relative to manufacturing (24.1 percent), but there were wide variations within manufactur- negotiations with the EU. To facilitate the economic ing industries. and social adjustment that this process entails, the 5As provided in Import and Export Regulation No. 74 of 1993. EU will provide financial support.

Regional Developments and purposes, the remaining restrictions on current invis- Future Challenges ibles were not binding for bona fide transactions and did not hinder foreign investments. Peace has created a number of opportunities for Further liberalizing steps were taken during 1993 Jordan, including the possibility of an investment- and early 1994. In 1993, banks were authorized to led boom in the region that could increase the de- offer foreign currency accounts to residents. Li- mand for Jordanian exports of goods and services. censed banks and financial companies were autho- But peace also poses challenges. First, although the rized to offer investment portfolio management ser- accord with Israel offers Jordan additional trade op- vices to nonresidents in major foreign currencies. portunities, a set of complex bilateral issues prevents They were also authorized to offer margin trading Jordan from benefiting immediately from the ex- services for nonresidents through their nonresident panding trade and investment opportunities in the re- foreign currency accounts credited with foreign gion. Second, there is potential for an expansion of means of payment coming from abroad, subject to a trade between Jordan and the West Bank and Gaza 30 percent margin. From January 1994, licensed Strip, but it may be some time before it materializes. banks and financial companies were authorized to On the other hand, if a free trade area covering only open accounts in foreign currencies for nonresidents Israel and the West Bank and Gaza Strip were con- without restriction. Withdrawals and transfers from stituted, it would represent competition for Jordan. A nonresident accounts were permitted freely. At the major restructuring of trade patterns in the region is same time, the annual amount that residents were also likely to ensue from the prospective early liber- permitted to transfer abroad to meet current pay- alization of trade between several regional ments for invisibles, without prior approval or justi- economies and the EU. For Jordan to remain com- fying documents, was raised to JD 35,000 from petitive, it would need to proceed with its efforts at JD 20,000. Transfers by residents in excess of that swift trade liberalization. amount were allowed with supporting documen- tation, and authorizations were granted liberally. As a result of these measures, current account Developments in the Exchange System convertibility was virtually achieved. The crucial step came in February 1995, when the Jordanian au- In the wake of the balance of payments crisis in thorities formally accepted the obligations of Arti- 1989, the authorities tightened the payments and ex- cle VIII, Sections 2, 3, and 4 of the IMF's Articles change system initially to cope with the intensifying of Agreement. balance of payments difficulties. As the balance of Jordan continues to take steps to further liberalize payments position improved, however, they gradu- its exchange system. A draft law to regulate foreign ally liberalized the exchange system, and by the end exchange transactions has been submitted for parlia- of 1992 Jordan had reverted to the liberal exchange mentary approval. The new law, which will super- system that it had enjoyed in the past. The specific sede the Transitory Foreign Currency Supervision measures adopted to liberalize the exchange system Law, guarantees complete current account convert- are summarized below. ibility for residents and nonresidents and capital ac- In February 1991, Jordan reduced advance import count convertibility for nonresidents; in particular, deposit requirements markedly and eliminated them no restrictions are imposed on the repatriation of in May 1992. Jordan's Central Bank doubled the profits and dividends for foreign investors. The law amounts of foreign exchange that residents and non- also allows residents to engage in a number of capi- residents could take abroad in November 1991 and tal account transactions and to take out any funds again in November 1992. It also increased substan- originally brought into Jordan through the banking tially the limits for resident and nonresident holdings system. Under the new legislation, the Central Bank of foreign exchange deposits in Jordan. As a result of will allow interbank lending in foreign currency, these liberalization steps, by mid-1992 the exchange thereby fostering the creation of a foreign currency system had became quite liberal; for all practical interbank market.

©International Monetary Fund. Not for Redistribution VI11 Social Aspects of the Adjustment Program: Strengthening the Social Safety Net

Ahsan Mansur

In addition to maintaining the traditional family- Poverty and Income Distribution based social arrangement and the formal pension system that operates through the Social Security Cor- Available information on per capita output growth poration for the private sector (see Box 8.1) and the and inequality (as measured by the Gini coefficient) government pension plans, Jordan has strengthened suggests that during 1973-80, per capita income in- the safety net system for the poorer segments of the creased by 7.9 percent a year while income inequal- population since the 1989 economic crisis. Before ity increased, especially in rural areas; and, during 1989, Jordan operated a system of administered 1980-86, per capita income was flat, but income dis- prices, primarily involving an overvalued exchange tribution improved, especially in rural areas. Based rate. The massive devaluation of the Jordan dinar in on the 1986-87 income and expenditure survey, 1988-89—although an important component of Jor- World Bank estimates indicate that 100 percent of dan's effort to ensure sustainable long-term growth Jordanian households were above the severe poverty and eventual improvement in overall living stan- or general poverty lines. Using a much higher level dards—in the short term hurt the poor as well as of annual per capita income (JD 148) than that of the workers receiving inadequate wage adjustments. World Bank to define what he called the "absolute poverty line," Al-Saquor found that 3 percent of the As a result of the economic crisis of the late population fell below this line.1 The depth of 1980s, per capita real income and consumption poverty, measured in terms of the poverty gap,2 was dropped sharply through a slowdown in domestic 1.9 percent for the lowest decile and 4.8 percent for economic activity and a decline in the inflow of the lowest 20 percent. Based on Al-Saquor (1990), workers' remittances and official transfers. The the depth of poverty below the absolute poverty line poverty profile deteriorated markedly, following was 0.3 percent. sustained improvements during the 1970s and the first half of the 1980s. The actions taken to reduce domestic imbalances—particularly, containment of Dimensions and Distribution the public sector wage bill and other current expen- of Poverty, 1991-92 ditures, exchange rate realignment, and price adjustments—also lowered household income in real Poverty worsened during the balance of payments terms in the short run. crisis and the regional conflict. Regardless of the measure chosen, a comparison of data indicates that poverty was deeper and more widespread and that there was greater inequality among the poor. In Poverty Profile 1991, 8.7 percent of the population was estimated to be below the severe poverty line, and 19.8 percent Associated with the economic boom, poverty in below the general poverty line compared with zero Jordan declined steadily during 1980-86, with the percent in 1986.3 The poverty gap also deteriorated share of the population below the poverty line declining from 24 percent in 1980 to less than 3 percent in 1986-87. However, as the economic situation 1M. Al-Saquor, "The Poverty Line in Jordan," in Income Dis- faltered, the poverty situation deteriorated sharply in tribution in Jordan, ed. by K.A. Jaber, M. Buhbe, and M. Smadi (Boulder, Colorado: Westview Press, 1990). the late 1980s, worsening further after the 1990 re- 2Based on a concept that reflects the percentage of national in- gional crisis (Table 8.1). At the same time, Jordan's come by which the expenditures of the poor would have to be in- rapid population growth has strained the availability creased to bring them up to the poverty line. and quality of social services. The strong economic 3The expenditure data for 1991 were unreliable because the av- turnaround during 1992-93 is believed to have erage expenditures reported in the survey were only 76 percent of those in the national income accounts, compared with 94 percent helped somewhat in reducing the level of unemploy- for the 1992 income and expenditures survey. Accordingly, the ment and alleviating poverty. 1991 estimates for poverty are likely to be biased upward.

Box 8.1. Social Security Corporation The Social Security Corporation (SSC) was set up in and vocational diseases, the benefit is equivalent to 60 1980 to provide social insurance for old-age pensions, percent of the last salary; for survivor pensions, it is 50 survivor pensions, disability pensions, work accidents, percent of the last salary. and vocational diseases for nongovernment workers. For the time being, the financial position of the SSC The scheme is mandatory for companies with five em- appears to be strong (see table), largely because it is a ployees or more and voluntary for others. About 9,000 new establishment, with a ratio of contributors to ben- companies are enrolled, totaling some 275,000 contrib- eficiaries of more than 10 to 1. Current arrangements, utors. Contributions are equivalent to 15 percent of in particular, regarding early retirement and the mini- gross wages, of which 10 percent is paid by the em- mum pension provision, may appear in the long run to ployer and 5 percent by the employee. The overall be too generous or may require transfers from the bud- number of beneficiaries is currently about 23,000. get to maintain full funding. If the SSC turns out to be The benefit formula for old-age pensions is based on only partially funded, changes in demographic trends, the average salary over the last two years of activity, in particular, a lower population growth rate and in- multiplied by 2 percent times the number of years of creased life expectancy, will affect the long-term sus- contributions. Once pensions are paid out, there is no tainability of the current scheme. With the increase subsequent cost of living adjustment. There is a mini- over time in the average number of contributing years, mum pension of JD 50 a month. Because the SSC is the relative weight of the minimum pension provision relatively recent, the average number of contributing should decrease. However, given the absence of cost years is still limited; therefore, about 90 percent of the of living adjustment, pressures to increase the mini- recipients take advantage of the minimum pension pro- mum pension amount are likely to remain strong. An- vision. Old-age pensions are supplemented with family other source of concern is related to the return on in- allowances equivalent to 10 percent for the spouse and vestment. The SSC scheme is based on a so-called 5 percent for each dependent child under the age of 26, defined-benefit principle; that is, beneficiaries do not with a maximum of two. The retirement age is 60 for share the risk on investment. In this regard, investment men and 55 for women. Early retirement, with a small in operations owned or controlled by the Government penalty, is allowed from the age of 45 if the applicant (particularly if the public enterprises are not efficient) has contributed for at least 15 years. For work accidents may not be judicious.

Assets and Liabilities of the Social Security Corporation (Millions of Jordan dinars)

1987 1988 1989 1990 1991

Cash 0.4 0.1 0.9 0.2 0.4 Deposits with banks and financial institutions 49.0 61.8 88.4 135.8 163.8 Development and corporate bonds 25.8 31.5 29.6 29.5 40.0 Companies' shares 14.5 30.8 50.4 49.5 52.1 Investment in full ownership of an institution 24.3 24.8 25.7 24.8 25.5 Loans 82.2 94.9 97.2 104.6 128.5 Advances and claims on participants 13.6 17.6 19.9 21.5 19.5 Fixed assets 3.3 3.4 3.8 5.1 7.7 Other assets 10.0 5.3 5.3 10.5 8.7 Assets = Liabilities 223.1 270.2 321.2 381.5 446.2 Balance of net accruing revenues at the beginning of the year 178.4 221.9 269.1 317.4 372.4 Net revenue for the year 43.5 47.2 48.3 55.0 64.3 Creditors 0.9 0.5 0.4 0.5 0.8 Other 0.3 0.6 3.4 8.6 8.7

Source: Social Security Corporation.

sharply, widening to between 1.9 percent and However, the extent of improvement was perhaps 5.0 percent. somewhat less than what appears from the data Based on the 1992 income and expenditure sur- analysis because of potential weaknesses in the vey, poverty was somewhat less severe in 1992. 1991 data. Although poverty appeared to have de-

Table 8.1. Selected Poverty Indicators

Poverty Line 1 2 (Jordan dinars per capita Head Count Index Poverty Gap a year at current prices) (In percent)

Income and expenditure survey, 1986-87 Al-Saquor's absolute poverty line 148 3.0 0.3 World Bank Severe poverty line 59 General poverty line 79 Maximum income Lowest 10 percent 197 10.0 1.9 Lowest 20 percent 246 20.0 4.8 Income and expenditure survey, 1992 Al-Saquor's absolute poverty line 261 14.9 3.7 World Bank Severe poverty line 104 0.5 0.1 General poverty line 139 1.8 0.4 Maximum income Lowest 10 percent 223 10.0 2.3 Lowest 20 percent 294 20.0 5.3 Employment, unemployment, returnees, and poverty survey, 1991 World Bank Severe poverty line 102 8.7 1.9 General poverty line 137 19.8 5.0

Sources: World Bank, Jordan: Poverty Assessment, Country Sector Study Report No. l2675,Vol. I (Washington, 1994); M.AI-Saquor, "The Poverty Line in Jordan," in Income Distribution in Jordan, ed. by K.A.Jaber, M. Buhbe, and M. Smadi (Boulder, Colorado: Westview Press, 1990). 'Measures the percentage of the population for whom per capita expenditures are below the poverty line. 2Measures the depth of poverty—that is, the percentage of national income by which the expenditure of the poor would have to be increased to bring them up to the poverty line.

clined in 1992 more than had been envisaged, all in- elimination of many food subsidies.4 For the bottom dicators point to a marked deterioration compared 20 percent, average food consumption—accounting with the pre-crisis situation. In particular, based for about half of their total expenditures—dropped on Al-Saquor's absolute poverty line, the propor- by 36 percent during the corresponding period. tion of the population below the absolute poverty A breakdown in the change in poverty in terms of line was almost 15 percent, that is, a fivefold in- underlying factors, such as growth and redistribu- crease from the pre-crisis level based on the same tion, indicates that between 1986 and 1992 the head measure. count index (which measures the percentage of the population for whom per capita expenditures are Why Poverty Deteriorated below the poverty line) increased by 11.9 percent, about 70 percent of which could be explained by the During the past two decades, Jordan's poverty decline in economic growth (Table 8.2). However, profile has been directly related to its economic unlike previous episodes or in other countries in growth performance. As real expenditures per capita similar circumstances, redistribution of expenditures declined by 22 percent in real terms (from JD 530 to (an indicator of income)—particularly for the urban JD 415 in constant 1986 prices) between 1986 and areas—partly mitigated the rise in poverty. The rural 1992, real expenditures declined for all groups. For population suffered a greater deterioration, which the poorest 20 percent, the decline in real expendi- was attributable both to a decline in income and ture was 36 percent, whereas for the top 20 percent, expenditure and increasing income inequality near the decline was limited to 11 percent. During the poverty line. Poverty alleviation policies—which 1986-92, food prices measured in terms of nominal market prices increased by 78 percent because of the 4As noted, a part of the increase was mitigated through the in- devaluation of the Jordan dinar and the reduction or troduction of a food coupon system in 1990.

Table 8.2. Factors Contributing to Change in Poverty (In percent)

Change in Head Source of Change Count Index Growth Redistribution Residual

World Bank poverty line, 1980-86' All Jordan -23.6 -14.7 -15.8 6.9

Al-Saquor poverty line, 1986-922 All Jordan 11.9 8.3 -1.3 5.0 Urban sector 10.4 7.1 -2.1 5.4 Rural sector 17.3 12.5 1.4 3.4

Source: World Bank, Poverty Assessment. Note:The negative sign indicates that income distribution around the poverty line became more nearly equal. 'Defined in Table 8.1. The head count index was 24.5 percent in 1980 and zero in 1986. 2Defined in Table 8.1.The head count index was 3.0 percent in 1986 and increased to 14.9 percent in 1992.

are more focused on urban areas because of better program—particularly the generalized food sub- accessibility—could have contributed to this effect. sidy—from its peak in 1990 by issuing food coupons directly to the poor. In addition to providing food subsidies, the safety net system alleviates poverty Poverty Alleviation Policies through cash or in-kind assistance. The main programs are cash transfers through the NAF, health The authorities have been responsive to the social care subsidies, employment through projects that and human resource dimensions of its development generate income, and in-kind grants. The Govern- strategy. Accordingly, expenditures on education ment is the main provider of food, cash, and health and health have always accounted for a large seg- care subsidies, whereas nongovernment organiza- ment of Jordan's total budgetary outlays. This strat- tions (NGOs) and the United Nations Relief Works egy has paid dividends in terms of marked improve- Agency (UNRWA) are dominant in in-kind assis- ments in literacy, health, and other social indicators. tance and income-generating projects. Following the economic crisis in the late 1980s, in addition to maintaining its commitments to health and education sectors, Jordan has also increased its Food Subsidy Schemes expenditures on welfare programs. A large part of To cushion the poor from the effect of the large this welfare expenditure has been in the form of sub- depreciation of the Jordan dinar in 1988-89, the au- sidized staple commodities. Moreover, programs thorities initially left unchanged the domestic prices that provided direct income support to the extremely of a range of food items sold through the Ministry of poor—such as through the newly established Na- Supply. Consequently, budgetary outlays for food tional Aid Fund (NAF)—were also introduced at subsidies rose to a peak of JD 83.5 million in 1990 that time (1989). (3.1 percent of GDP) (Table 8.3). Until September 1990, no targeting mechanism was in place, and the income distribution effect of general subsidies was Temporary Social Safety Net Instruments not satisfactory.5 Moreover, the subsidies repre- The authorities' expenditure on safety net pro- sented an open-ended budgetary commitment, which grams primarily consisted of generalized food subsi- was unsustainable given the rapid population growth dies, food coupons, transfers through the NAF, and health cards. In 1990, total food subsidies amounted 5In principle, only inferior goods justify a general subsidy be- to 3.5 percent of GDP, accounting for almost 10 per- cause of demand patterns; most of the benefits are self-targeted to cent of central government current expenditure. De- the poor. Ahmad (1991) estimated on the basis of the 1987 in- spite the Government's best intentions, the general- come and expenditure survey, that, of the subsidized goods, only flour could be considered an inferior product. Thus, it was likely ized food subsidy was both expensive and inefficient that until September 1990 the generalized subsidy system made from a distributional point of view. Accordingly, the substantially greater transfers to the rich and the middle classes authorities reduced the budgetary cost of the welfare than to the poor.

Table 8.3. Food Subsidies

1988 1989 1990 1991 1992 1993 1994

In millions of Jordan dinars

Wheat -6.8 -36.2 -40.5 -34.7 ^3.3 -43.3 -39.5 Barley -3.6 -7.2 -6.9 -7.0 -8.6 -9.0 -3.3 Sugar 0.7 -15.6 -27.4 -2.4 1.9 2.7 -0.5 Rice -0.2 -7.3 -11.9 -5.2 -5.8 -6.7 -4.8 Milk -4.4 -3.9 -6.7 -4.3 -3.0 Meat1 7.3 -5.6 -4.8 — — — — Sorghum2 2.2 -2.8 — — — Maize2 ^.3 -7.3 — — — Other -4.0 -1.2 18.9 2.0 5.2 7.0 7.9 Total subsidy (-) -6.6 -73.1 -83.5 -61.2 -57.3 -53.6 -43.2 In Ipercent of GDP

Total subsidy (-) -0.3 -3.1 -3.1 -2.1 -1.6 -1.4 -1.0 Generalized subsidies -0.3 -3.1 -3.4 -1.8 -1.5 -1.3 -1.0

Source: Jordanian authorities. 'Subsidies for meat were eliminated in 1990. Subsidies for sorghum and maize were eliminated in 1992.

and the fiscal adjustment required under the authori- Although the general subsidy on wheat was costly ties' stabilization program. and there was scope for achieving better results, the To contain or reduce the growing budgetary costs program was maintained partly because of its distri- of the general food subsidy scheme, a scheme was butional impact. According to the 1987 household introduced in September 1990 to ration sugar, rice, income and expenditure survey, wheat and bread and powdered milk through food coupons; it has re- were the main food expenditure items, accounting mained in effect since then. The food coupon system for about 15 percent of expenditure for the poorest aims at ensuring the availability of fixed quantities 20 percent of the population, compared with only 2 of these food items at reasonable prices for all Jorda- percent for the richest 20 percent. Thus, even with nians. Arrangements were also made to ensure the similar savings in nominal terms, the relative gains availability of additional quantities of these com- to the poor households were much higher. Some modities at a price generally higher than the full modest price adjustments were made in 1993 to con- cost. By introducing the rationing scheme, the au- tain the budgetary costs of the wheat subsidy. thorities essentially limited the open-ended budgetary commitment implied by a generalized subsidy. Under the food coupon system and through the National Aid Fund implicit cross-subsidization scheme, subsidies on Since its establishment in 1987, the NAF—an au- milk, sugar, and rice were reduced from JD 43.8 mil- tonomous government agency—has steadily in- lion in 1990 to JD 8.3 million in 1994, notwithstand- creased its operations to provide direct income sup- ing a significant increase in the population, particu- port for extremely poor families with financial larly after the 1990 regional crisis.6 Other operations support from the Government. In 1994, the NAF of the Ministry of Supply resulted in budgetary sav- provided direct income support to about 31,000 ings because the prices of products were generally extremely poor households with 147,000 persons maintained above their import costs. In 1994, when (3.5 percent of the population) at or below the means testing was introduced for food coupons, poverty level. In that year, the NAF budget amounted households with a member earning more than to JD 12.7 million, which was divided among three JD 500 were excluded from the benefits. major categories: cash transfers (JD 11.1 million), the rehabilitation program (JD 1.0 million), and administration (JD 0.4 million). Despite the continued fiscal adjustment, the authorities increased the bud- 6A large part of these budgetary savings, however, reflected declining import prices for sugar and rice (these prices declined by getary support to the NAF from JD 8 million to about 30 percent and 20 percent, respectively, over 1990-92). JD 10 million in 1993 and to JD 12.7 million in

1994. This increased budgetary support allowed for Public Education System a significant broadening of NAF operations by providing an additional JD 10 a month for each family Like in most other countries, poverty in Jordan is and an increase in the number of families covered inversely related to the rise in the level of people's under the program by 8,000, to 31,000.7 Currently, education. Illiterates account for 22 percent of the population 21 and older, representing about 42 per- an average of JD 6.3 a month for each person is 8 being transferred to the beneficiaries; this compares cent of the lowest 10 percent. Because the likeli- with the absolute poverty level of JD 11.5 a month hood of being poor decreases with education, and for each person. given that Jordan has achieved remarkable progress in reducing illiteracy and establishing good educational standards, the prospect for further poverty re- 9 Health Care Subsidies duction through education is good. In spite of the economic crisis during the late 1980s, growth in the The Ministry of Health gives cards to the poor, number of teachers was more than twice the increase granting them lower fees for health care, and also in enrollment over the last decade; the number of waves fees on a case-by-case basis. People with in- classrooms increased by 50 percent during the comes below JD 200 a month are eligible for health 1980s, and every village with ten or more children of cards, and disabled persons are eligible irrespective school age has a school. Accordingly, the illiteracy of income. About 50,000 health cards—valid for two rate has declined from 54 percent in 1965 to less years—were issued during 1991-92; the total num- than 20 percent currently, and the absolute number ber of beneficiaries was estimated to be 322,000 in of illiterate adults has decreased despite the high 1992. The amount of income transferred (subsidies rates of population growth. Overall, Jordan main- beyond the general subsidy of the Ministry of tains a high-quality comprehensive public education Health) was rather modest, at JD 3.5 a person a year, system that is cost-effective (Table 8.4). and the total estimated transfer on this account was also modest (JD 1.1 million). Virtually universal primary school enrollments suggest that public spending on education has reached the poor. Public education costs are 95 per- Other Welfare Programs cent subsidized, and poor students (about 15 per- Other income transfer programs, funded by private cent) are not required to pay school fees. Reflecting contributions and UNRWA, also provide income sup- these developments, the poorest 20 percent of the port to a large number of households. Under an population spent about 0.8 percent of total expendi- UNRWA program for special hardship cases, more ture on education, compared with 3.5 percent for than 30,000 individuals qualified to receive predomi- nonpoor households (1986 income and expenditure nantly in-kind assistance equivalent to a monthly survey). transfer of JD 9.2 a person. The Zakat Fund—financed through private contributions but adminis- Public Health System tered by the ministry charged with religious charities and institutions (Aw'qaf)—supports 3,000 house- Although Jordan does not have a formal public holds on a regular basis (with an average monthly health insurance system, health indicators suggest transfer of JD 7.7 to each household in 1991). that the coverage of the current health system is quite comprehensive (Table 8.5). All military and civil employees and their dependents pay a small Permanent Social Safety Net Arrangements monthly premium and are covered under government health insurance plans.10 The insurance plan In addition to setting up the temporary arrange- covers free treatment at public hospitals and very ments, the authorities have also alleviated poverty low fees in primary health care facilities. Only the and contributed to human resource development eligible poor and disabled receive the benefits free through a number of permanent programs. Through of charge. All other households have access to pri- these, education, health, pensions, and social security are available to all eligible residents without means tests. Traditionally, the authorities' contribu- 8At the other end, people with vocational or secondary educa- tions or higher represent only 11 percent of the lowest 10 percent tions to these programs have been high, and priori- while comprising 34 percent of the population 21 and older. ties for social sectors have been maintained. 9More than 70 percent of the poor and very poor heads of households are older than 40 and grew up before primary education was universal. 7The NAF's monthly assistance to households amounts to 10For military personnel and their dependents, the premium is JD 25 for the head of the household, and JD 5 for the spouse and 1.1 percent of the basic salary with a ceiling of JD 1.5 a month; each additional dependent (up to four, because there is a ceiling for civil servants, the premium is 2 percent of the basic salary of JD 50 a family). with a ceiling of JD 8 a month.

Table 8.4. Selected Countries: Expenditure on Education (In percent, unless otherwise indicated)

Educational Quality Students advancing Ratio of Unit Cost Illiteracy rate, to final year of pupils to Spending on Education (1990) of Primary ages 15-19 primary school teachers Share of total Share of Education (1990) (1987) (1989) government spending GDP (1990)'

Algeria 12 90 28 212 Egypt 36 95 24 13.4 4.9 138 Iran, Islamic Rep. of 21 87 24 21.7 3.8 142 Jordan 3 96 28 14.8 5.4 150 Morocco 21 67 26 17.0 4.8 175 Syria 18 89 26 7.4 2.2 157 Tunisia 5 72 30 17.0 5.1 178 Yemen (1987) 31 45 15.5 6.0 181

Sources: World Bank, Social Indicators of Development (Washington, 1992); and IMF, Government Finance Statistics (Washington, 1992). 'In U.S.dollars.

mary health care centers and are charged subsidized Overview of Recent Developments and fees somewhat higher than those public sector em- Medium-Term Strategy ployees are charged. While pursuing stabilization and structural reform programs involving large price adjustments, the au- Other Safety Net Arrangements thorities have launched new programs and strength- The Social Security Corporation (SSC) provides ened existing ones to protect the vulnerable segments retirement pensions and medical expenses for work- of the population; they have also maintained the qual- related injuries. The SSC is financed through taxes ity and availability of education and health care ser- on private, formal sector employment, with contribu- vices for all Jordanians. In the first two years tions by enterprises and employees. The scheme is (1989-90), however, in the absence of a comprehen- mandatory for companies with five employees or sive strategy for a social safety net, the program was more and voluntary for others. About 9,000 compa- not well targeted and the costs were not sustainable. nies are currently covering some 275,000 contribu- To address this problem and improve basic social as- tors (about one-third of the labor force). Contribu- sistance, the authorities expanded the operations of the tions amount to 15 percent of gross wages, of which NAF, introduced a coupon system for several food 10 percent is paid by the employer and the remainder items, raised certain prices, and introduced cross-sub- by the employee. The total number of beneficiaries is sidization; through these initiatives, they have suc- currently estimated to be about 23,000 (see Box 8.1). ceeded in significantly reducing the cost of the safety

Table 8.5. Health Indicators, 1991 (Per 10,000 population)

Arab Developing Industrial Jordan Nations Countries Countries

Doctors 17.1 8 7 27 Nurses, midwives 8.7 II II 70 Dentists 3.8 3 2 9 Pharmacists 5.7 4 2 8

Source: World Bank, Poverty Assessment.

©International Monetary Fund. Not for Redistribution Overview of Recent Developments and Medium-Term Strategy net program in recent years while improving the tar- In addition to maintaining good social infrastruc- geting of benefits. In particular, the total budgetary ture through expenditure programs on education cost of the authorities' food subsidy program declined and health, Jordan has succeeded in establishing a from a peak of 3.5 percent of GDP in 1990 to 1.4 per- wide network of welfare programs ranging from cent of GDP in 1993. Following further price adjust- generalized subsidies to targeted means-tested in- ments in 1994, the cost declined further in 1994 to come support programs. The poverty reduction ef- about 1 percent of GDP; most of the decline was on fect of the generalized subsidy program, however, is account of the generalized subsidy on wheat. believed to be rather limited. In contrast, targeted In addition to contributing significantly to a sus- programs under the NAF and the health card have tained reduction in the fiscal deficit, this consolida- benefited a sizable segment of the poorest house- tion of the subsidy program made it possible to in- holds. A redistribution of the resources away from crease the budgetary transfers to the NAF for direct generalized subsidies to the targeted programs income support for the poor. NAF expenditure ac- could further improve the efficacy of the authori- counted for 23 percent of total targeted safety net ties' welfare program. transfers in 1992, and this ratio increased further in Sustainability of the cost of the authorities' safety 1993. Both benefits and coverage have been ex- net program—currently at about 1.6 percent of panded significantly since 1993. By providing per GDP—needs to be addressed in a medium-term capita benefits of JD 75 a year, this program suc- context. Notwithstanding the remarkable progress ceeded in bringing about 3.5 percent of the popula- achieved in recent years, continued efforts are neces- tion (who were below the absolute poverty line) sary to further consolidate and sustain the resulting close to the poverty line. The food coupon system gains. The recent reduction in the level of wheat and served the majority of the population, although indi- other subsidies was partly attributable to a decline in vidual benefits were relatively low (totaling JD 6.5 a international commodity prices, which may reverse year). It is believed that the low- and middle-income over time. High population growth (at a rate of more households—who either do not qualify or avoid than 3 percent a year) would also continue to exert means-tested transfer programs—receive benefit pressures on food subsidies and other expenditure transfers only through food coupons and the general- programs, such as pensions, education, and physical ized subsidies. The health card system provides infrastructure. The authorities' ongoing medium- fairly comprehensive health care coverage to more term adjustment and reform program may imply than 8 percent of the population, at a reasonable cost short-term costs to certain socioeconomic groups (JD 3.5 a person a year). that would then have to be addressed.

Recent Occasional Papers of the International Monetary Fund

136. Jordan: Strategy for Adjustment and Growth, edited by Edouard Maciejewski and Ahsan Mansur, with contributions from Patricia Alonso-Gamo, Jean-Pierre Chauffour, Etienne de Callatay, and Christopher McDermott. 1996. 135. Vietnam: Transition to a Market Economy, by John R. Dodsworth, Erich Spitaller, Michael Braulke, Keon Hyok Lee, Kenneth Miranda, Christian Mulder, Hisanobu Shishido, and Krishna Srinivasan. 1996. 134. India: Economic Reform and Growth, by Ajai Chopra, Charles Collyns, Richard Hemming, and Karen Parker with Woosik Chu and Oliver Fratzscher. 1995. 133. Policy Experiences and Issues in the Baltics, Russia, and Other Countries of the Former Soviet Union, edited by Daniel A. Citrin and Ashok K. Lahiri. 1995. 132. Financial Fragilities in Latin America: The 1980s and 1990s, by Liliana Rojas-Suarez and Steven R. Weisbrod. 1995. 131. Capital Account Convertibility: Review of Experience and Implications for IMF Policies, by staff teams headed by Peter J. Quirk and Owen Evans. 1995. 130. Challenges to the Swedish Welfare State, by Desmond Lachman, Adam Bennett, John H. Green, Robert Hagemann, and Ramana Ramaswamy. 1995. 129. IMF Conditionality: Experience Under Stand-By and Extended Arrangements. Part II: Background Pa- pers. Susan Schadler, Editor, with Adam Bennett, Maria Carkovic, Louis Dicks-Mireaux, Mauro Mecagni, James H.J. Morsink, and Miguel A. Savastano. 1995. 128. IMF Conditionality: Experience Under Stand-By and Extended Arrangements. Part I: Key Issues and Findings, by Susan Schadler, Adam Bennett, Maria Carkovic, Louis Dicks-Mireaux, Mauro Mecagni, James H.J. Morsink, and Miguel A. Savastano. 1995. 127. Road Maps of the Transition: The Baltics, the Czech Republic, Hungary, and Russia, by Biswajit Baner- jee, Vincent Koen, Thomas Krueger, Mark S. Lutz, Michael Marrese, and Tapio O. Saavalainen. 1995. 126. The Adoption of Indirect Instruments of Monetary Policy, by a Staff Team headed by William E. Alexan- der, Tomas J.T. Balino, and Charles Enoch and comprising Francesco Caramazza, George Iden, David Marston, Johannes Mueller, Ceyla Pazarbasioglu, Marc Quintyn, Matthew Saal, and Gabriel Sensen- brenner. 1995. 125. United Germany: The First Five Years—Performance and Policy Issues, by Robert Corker, Robert A. Feldman, Karl Habermeier, Hari Vittas, and Tessa van der Willigen. 1995. 124. Saving Behavior and the Asset Price "Bubble" in Japan: Analytical Studies, edited by Ulrich Baumgart- ner and Guy Meredith. 1995. 123. Comprehensive Tax Reform: The Colombian Experience, edited by Parthasarathi Shome. 1995. 122. Capital Flows in the APEC Region, edited by Mohsin S. Khan and Carmen M. Reinhart. 1995. 121. Uganda: Adjustment with Growth, 1987-94, by Robert L. Sharer, Hema R. De Zoysa, and Calvin A. McDonald. 1995. 120. Economic Dislocation and Recovery in Lebanon, by Sena Eken, Paul Cashin, S. Nuri Erbas, Jose Martelino, and Adnan Mazarei. 1995. 119. Singapore: A Case Study in Rapid Development, edited by Kenneth Bercuson with a staff team comprising Robert G. Carling, Aasim M. Husain, Thomas Rumbaugh, and Rachel van Elkan. 1995. 118. Sub-Saharan Africa: Growth, Savings, and Investment, by Michael T. Hadjimichael, Dhaneshwar Ghura, Martin Muhleisen, Roger Nord, and E. Murat Uger. 1995. 117. Resilience and Growth Through Sustained Adjustment: The Moroccan Experience, by Saleh M. Nsouli, Sena Eken, Klaus Enders, Van-Can Thai, Jorg Decressin, and Filippo Cartiglia, with Janet Bungay. 1995. 116. Improving the International Monetary System: Constraints and Possibilities, by Michael Mussa, Morris Goldstein, Peter B. Clark, Donald J. Mathieson, and Tamim Bayoumi. 1994.

115. Exchange Rates and Economic Fundamentals: A Framework for Analysis, by Peter B. Clark, Leonardo Bartolini, Tamim Bayoumi, and Steven Symansky. 1994. 114. Economic Reform in China: A New Phase, by Wanda Tseng, Hoe Ee Khor, Kalpana Kochhar, Dubravko Mihaljek, and David Burton. 1994. 113. Poland: The Path to a Market Economy, by Liam P. Ebrill, Ajai Chopra, Charalambos Christofides, Paul Mylonas, Inci Otker, and Gerd Schwartz. 1994. 112. The Behavior of Non-Oil Commodity Prices, by Eduardo Borensztein, Mohsin S. Khan, Carmen M. Reinhart, and Peter Wickham. 1994. 111. The Russian Federation in Transition: External Developments, by Benedicte Vibe Christensen. 1994. 110. Limiting Central Bank Credit to the Government: Theory and Practice, by Carlo Cottarelli. 1993. 109. The Path to Convertibility and Growth: The Tunisian Experience, by Sal eh M. Nsouli, Sena Eken, Paul Duran, Gerwin Bell, and Ziihtu Yucelik. 1993. 108. Recent Experiences with Surges in Capital Inflows, by Susan Schadler, Maria Carkovic, Adam Bennett, and Robert Kahn. 1993. 107. China at the Threshold of a Market Economy, by Michael W. Bell, Hoe Ee Khor, and Kalpana Kochhar with Jun Ma, Simon N'guiamba, and Rajiv Lall. 1993. 106. Economic Adjustment in Low-Income Countries: Experience Under the Enhanced Structural Adjustment Facility, by Susan Schadler, Franek Rozwadowski, Siddharth Tiwari, and David O. Robinson. 1993. 105. The Structure and Operation of the World Gold Market, by Gary O'Callaghan. 1993. 104. Price Liberalization in Russia: Behavior of Prices, Household Incomes, and Consumption During the First Year, by Vincent Koen and Steven Phillips. 1993. 103. Liberalization of the Capital Account: Experiences and Issues, by Donald J. Mathieson and Liliana Rojas- Suarez. 1993. 102. Financial Sector Reforms and Exchange Arrangements in Eastern Europe. Part I: Financial Markets and Intermediation, by Guillermo A. Calvo and Manmohan S. Kumar. Part II: Exchange Arrangements of Previously Centrally Planned Economies, by Eduardo Borensztein and Paul R. Masson. 1993. 101. Spain: Converging with the European Community, by Michel Galy, Gonzalo Pastor, and Thierry Pujol. 1993. 100. The Gambia: Economic Adjustment in a Small Open Economy, by Michael T. Hadjimichael, Thomas Rumbaugh, and Eric Verreydt. 1992. 99. Mexico: The Strategy to Achieve Sustained Economic Growth, edited by Claudio Loser and Eliot Kalter. 1992. 98. Albania: From Isolation Toward Reform, by Mario I. Blejer, Mauro Mecagni, Ratna Sahay, Richard Hides, Barry Johnston, Piroska Nagy, and Roy Pepper. 1992. 97. Rules and Discretion in International Economic Policy, by Manuel Guitian. 1992. 96. Policy Issues in the Evolving International Monetary System, by Morris Goldstein, Peter Isard, Paul R. Masson, and Mark P. Taylor. 1992. 95. The Fiscal Dimensions of Adjustment in Low-Income Countries, by Karim Nashashibi, Sanjeev Gupta, Claire Liuksila, Henri Lorie, and Walter Mahler. 1992. 94. Tax Harmonization in the European Community: Policy Issues and Analysis, edited by George Kopits. 1992. 93. Regional Trade Arrangements, by Augusto de la Torre and Margaret R. Kelly. 1992. 92. Stabilization and Structural Reform in the Czech and Slovak Federal Republic: First Stage, by Bijan B. Aghevli, Eduardo Borensztein, and Tessa van der Willigen. 1992. 91. Economic Policies for a New South Africa, edited by Desmond Lachman and Kenneth Bercuson with a staff team comprising Daudi Ballali, Robert Corker, Charalambos Christofides, and James Wein. 1992. 90. The Internationalization of Currencies: An Appraisal of the Japanese Yen, by George S. Tavlas and Yuzuru Ozeki. 1992. Note: For information on the title and availability of Occasional Papers not listed, please consult the TMF Publications Catalog or contact IMF Publication Services.