Report No. 39201-JO 39201 Hashemite Kingdom of Jordan Resolving Jordan's Labor Market Paradox of Concurrent Economic Growth and High Unemployment December 23, 2008 Social and Economic Development Group Middle East and North Africa Region Document of the World Bank HashemiteKingdomofJordan Resolving Jordan's Labor Market Paradox o f Concurrent Economic Growth and HighUnemployment Table of Contents ACKNOWLEDGMENTS ................................................................................................................... IV EXECUTIVESUMMARY .................................................................................................................... i CHAPTER1 JORDAN'S . LABOR MARKET PARADOX .................................................................... 1 A. JORDAN'SEMPLOYMENTSTRATEGYAND RECENTECONOMICRESULTS........................................... 1 B. UNEMPLOYMENTDEFINED ............................................................................................................. 6 C. CHARACTERISTICS THE UNEMPLOYED OF ....................................................................................... 7 CHAPTER 2 THREEMISMATCHES INTHE JORDANIANLABOR . MARKET ................................ 13 A. GEOGRAPHY:THEDISTANCEBETWEENNEWJOBSANDTHE UNEMPLOYED................................... 13 B. EMPLOYABILITY:TECHNICALSKILLSAND EXPERIENCE . WORKPLACEBEHAVIORAND................... VS PRODUCTIVITY ............................................................................................................................. 15 C. EXPECTATIONS: FALSE OPTIMISMINA CHANGINGECONO ........................................................ 17 CHAPTER 3 LABOR . DEMAND: THEQUESTFORHIGH-WAGE JOBGROWTH ......................... 21 A. RECENTSECTORAL PATTERNS OFJOB GROWTH ............................................................................ 21 B. NEWBUSINESSES: KEYTOGREATER CREATION.............................................................. THE JOB 24 c. DIVERSIFYING EXPORTS GENERATETO HIGH-WAGEJOBS ............................................................ 30 D. REMOVING DISTORTIONS SPURJOBCREATION........................................................................ TO 38 CHAPTER 4 LABOR . SUPPLY: INVESTINGINTHE EMPLOYABILITY OF THE WORKFORCE AND REALIGNINGEXPECTATIONS ....................................................................................................... 43 A. INVESTING INEMPLOYABILITY...................................................................................................... 43 B. REALIGNING EXPECTATIONS ......................................................................................................... 48 CHAPTER 5 CONCLUSIONS, POLICY OPTIONS, AND RECOMMENDATIONS . ............................. 57 REFERENCES ................................................................................................................................. 59 LIST OF BOXES BOX 1.1. SOURCES OF DATA USED INTHIS REPORT ....... ...5 BOX 3.1. SELECTED LABOR LAWS AND REGULATIONS IN JORDAN ........................................................... 30 Box3.2. JORDAN'SQUALIFIED INDUSTRIAL ZONES(QIZs) ... BOX 3.3. FEMALE EMPLOYMENT INQIZS BOX 3.4. SECTORSFAVOREDBY JORDAN'SINVESTMENT PROMOTION LAW ............................................ 39 Box4.1. EMIGRATION FROM JORDAN ............................ ................. BOX 4.2. WORK SEARCH AND JOB ACCEPTANCE POLICIES INAUSTRALIA'SPROGRAM OF .................................. 53 UNEMPLOYMENT INSURANCE .................................................................................................... 55 FIGURE1.1. INCOMEFROMEMPLOYMENTACCOUNTSFORMOSTHOUSEHOLDINCOME.................................. LIST OF FIGURES 1 FIGURE1.2. CHANGING AGEDISTRIBUTIONOFTHEPOPULATION................................................................... 2 FIGURE1.3. THEPARADOX: STRONGINVESTMENTAND GDPGROWTH, BUTHIGHUNEMPLOYMENT .............3 FIGURE1.4. INTERNATIONALCOMPARISONS OFLABOR FORCESTATUS.......................................................... 7 FIGURE1.5. AGEAND LABOR FORCESTATUS(EMPLOYED, NOTEMPLOYED, NOTINLABOR FORCE), 2005 .....8 FIGURE1.6. AGEAND UNEMPLOYMENT, 2005 ............................................................................................... 9 FIGURE1.7. CHARACTERISTICS OFYOUTH UNEMPLOYMENT,2005.. ......................................................... 9 FIGURE1.8. EDUCATIONAND UNEMPLOYMENT, 2005 ..................... ....................................................... 10 FIGURE1.9. LABOR .................... FIGURE1.10.GENDER AND UNEMPLOYMENT................................... FORCESTATUS BY GENDER, 2005 FIGURE 2.1. UNEMPLOYMENT RATEBY GOVERNORATE, 2000AND FIGURE 2.2. EDUCATIONAL ACHIEVEMENT OFJORDANIAN AND FOREIGNWORKERS, 2004 .......................... 16 FIGURE2.3. UNEMPLOYMENT RATESBY UNIVERSITY SPECIALIZATION, 2000-05 ........................................ 17 FIGURE 2.4. UNIVERSITY SPECIALIZATIONBY GENDER ................................................................................ 18 FIGURE 2.5. SHAREOFUNEMPLOYED JORDANIANS WHO WOULD ACCEPT ONEOFTHECOMMONLY . AVAILABLE JOBS AT PREVAILING WAGES ....................................................................... FIGURE 3.1. AVERAGE ANNUAL NET JOB CREATION BY SECTOR, 200-05 .................................. FIGURE 3.2. GDPAND JOB GROWTHBY SECTOR, 2000-2004 ....................................................................... FIGURE 3.3. AVERAGE ANNUAL NET JOB CREATION (JORDANIANS ONLY), BY OCCUPATION, 2000-05 .........22 23 FIGURE 3.4. GROWTH INLABOR PRODUCTIVITY 1990-2003 ........................................................................ 24 FIGURE 25 FIGURE3.6. GROSSNATIONALSAVINGS AND GROSS DOMESTICSAVINGS, 2000-06.......... 3.5. PERFORMANCEINATTRACTING EXTERNAL FINANCING, 2001-06 ............................................. ..........26 FIGURE 3.7. ISSUES OR MAJOR CONSTRAINTSTO DOINGBUSINESS INJORDAN............. IDENTIFIED AS SEVERE BY SIZE OF ENTERPRISE................................................................. FIGURE 3.8. EASE STARTING A BUSINESS................. OF ........................... .28 FIGURE 3.9. EASE HIRING AND FIRING..................................................................................................... 29 OF FIGURE 3.10.JORDAN'SEXPORTPERFORMANCE,2000-05 ... ...................................................... FIGURE 1.INCREASES INEXPORTINTENSITY PEREMP 3.1 ERAGE ANNUAL .............. RATE OF EMPLOYMENT GROWTH (1995-2004) ......................................................................... 32 FIGURE3.12.CHANGES INAPPARELAND OTHER GOODSEXPORTS, 1998-2006 ............. ...................... .34 FIGURE3.13. EMPLOYMENTQIZS,2 0 0 1 ~.............................................................. IN 6 FIGURE 3.14. EVOLUTIONOFTHE TECHNOLOGICALSTRUCTUREOFEXPORTS, SELECTED COUNTRIES ............... FIGURE 3.15. GROWTH INMEDIUMAND HIGHTECHNOLOGY EXPORTS BETWEEN 1990AND 2004..................36 AND REGIONS,1980-2004 ......................................................................................................... 36 FIGURE 3.16. BRAINDRAIN INJORDANAND SELECTEDOTHER COUNTRIES, 2006 .............. FIGURE 4.1. LABOR FORCEPARTICIPATIONBEFOREAND AFTER MARRIAGE ..................... FIGURE4.2. PREFERENCEFORHIRINGMENOR WOMEN.................................................................... FIGURE4.3. PERCAPITA REALGROSSNATIONAL DISPOSABLEINCOME, 1976-2005 ......... FIGURE4.4. CIVILSERVICEAPPLICATIONS AND APPOINTMENTS, 2001-04....................... FIGURE4.5. QUALIFICATIONS OFCIVIL SERVICE APPLICANTS DEMONSTRATESINACCURATE INFORMATIONSO FIGURE4.6. RELATIONSHIPOFUNEMPLOYEDPERSONSTO HEADOFHOUSEHOLD, 2004 .............................. 50 FIGURE 4.7. EXPENDITURE QUINTILE OF THE FAMILIES OF THE UNEMPLOYED .............................................. 51 FIGURE 4.8. REMITTANCESACCRUE PREDOMINANTLY TO WEALTHY HOUSEHOLDS. ..................... 52 LIST OF TABLES TABLE1.1.ALLOFFICIAL DATA SOURCESTELL THESAMESTORY: MANY NEWJOBSBUTMOSTFILLED.............. BY FOREIGNWORKERS TABLE2.1.GOVERNORATE SHARESINJOB CREATIONAND UNEMPLOYMENT AT BEGINNINGOFTHE PERIOD 13 ...................................................................................................................... 4 TABLEGEOGRAPHIC DISTRIBUTIONOFJORDANIAN POPULATIONAGED 15 AND OLDER,.......................... 15 TABLE3.1.AVERAGE GROWTHRATEOFLABOR PRODUCTIVITYAND LABOR COST. BY SECTOR. 2000-03 ......24 TABLE3.2.INCOME TAX RATESAND INVESTMENTINCENTIVES...................................................................... 40 TABLE3.3.WORK-PERMIT FEES(JD/YEAR) .................................................................................................... 41 ACKNOWLEDGMENTS This report was prepared by the Economic and Social Development Unit of the Middle East Department, Middle East and North Africa Region of the World Bank, led by Mustapha Nabli, who has provided overall guidance and supervision. The core team was led by Susan Razzaz and included Niti Bhutani, Jad Chaaban, Mohamed Chemingui, Yusuf Mansur, Claudia Nassif, and Peter Walkenhorst. Additional contributions were provided by Sebnem Akkaya, Chadi Bou Habib, StephenKaram, and John Nasir. The team would like to thank our many counterparts in Jordan for candid discussions and access to data. These include our colleagues from the Ministry of Planning and International Cooperation, the Ministry of Trade and Industry, the Ministry of Labor, the Department of Statistics, the National Center for Human Resource Development, the Chamber of Commerce, the Chamber of Industry, and Dajani ConsultingCompany. The report was peer reviewed by Arup Banerji and Julia Devlin. Valuable guidance and comments were also received from Osman Ahmed, Randa Akeel, Sebnem Akkaya, Andras Bodor, Dipak Dasgupta, Sebastien Dessus, Farmkh Iqbal, Indermit Gill, John Nasir, Soren Nelleman, Miria Pigato, Aleksandra Posarac, Joseph Saba, Haneen Sayed, Carlos Silva-Jauregui, John Speakman, Zafiris Tzannatos, and Sophie Warlop. Administrative assistancewas expertly provided by Muna Salim, IsabelleChaal-Dabi and May Ibrahim. Steven Kennedy edited the report. EXECUTIVE SUMMARY Jordan's Labor Market Paradox: Economic Growth with High Unemployment 1. Reducing unemployment i s a top priority of the Government of Jordan, as expressed in its principal strategies, "We Are All Jordan" and the National Agenda. Consistent with those strategies and with the recommendations of donors, the Government's approach to reducing unemployment has been to createjobs by attracting investments that promote GDP growth. ii. Inrecentyears,Jordanhassuccessfullyattractedinvestment, achievedstrongGDP growth, and created many new jobs. Private foreign investments have increased dramatically since 2001 while GDP growth averaged 6 percent during 2000- 2005. And, consistent with the expectation of the government's strategy for reducing unemployment, GDP growth has generated a substantial number of new jobs: from 2000 to 2005, Jordan created between 23,900 and 44,500 additional jobs per year.' However, unemployment (among nationals) did not fall as expected, but has remained relatively steady near 14percent. ... 111. In this context, the strategy for reducing unemployment needs to be reconsidered. A continuation of the policies of the past five years may lead to strong growth in investment, GDP, and evenjobs, but is unlikely to reduce unemployment. Figure 1. The paradox: strong investment and GDP growth, but high unemployment J 9 1 301 Eg1 2 - c 1 0 - gx E=8 - 6 - 4 - 2L 0 2001 2002 2003 2004 2005 2006 1999 2000 2001 2002 2003 2004 2005 1999 2000 2001 2002 2003 2004 2005 Foreigndirect investment, 2001-06 GDP growth rate, 1999-2005 Unemployment rate, 1999-2005 Sources: CentralBank of Jordan,World Development Indicators,Employmentand UnemploymentSurveys iv. A close look reveals that many newly created jobs were not filled by Jordanians but by foreign workers. Although specific numbers vary depending on the data source used, the story is consistent. Estimates based on the DOS Employment and Unemployment Survey and the Ministry of Labor's database of worker permits find that 63 percent of newly createdjobs were filled by foreign workers. Estimates based on DOS's "blended" data set2finds that 52 percent of 1 Estimates of job creation vary considerably depending on the source of data. In this study we report a range, taking into consideration several official sources. See Box 1.1 for a detailed description of data sources. The DOS "blended" data set combines the data from the Employment and Unemployment Survey (EUS) with data from several enterprise-based surveys including the Employment in Enterprises (EIE) survey and surveys of the agriculture and construction sectors. The DOS "blended" data set does not include foreigners working in household-based activities, such as house-keepers. We used this data source at the 1 newly created jobs were filled by foreign workers. As a result of this trend, foreign workers in Jordan now number 250,000 to 300,000, far more than the number of unemployed Jordanians (175,000 in2005). Table 1. All official data sourcestell the same story: many new jobs but more than halfhave been filled by foreign workers Job creation based on DOS's EUSandMOL data 2001 2002 2003 2004 2005 Average Jordanians 200 9,900 6,400 42,900 23,300 16,500 Foreigners 26,000 -9,400 21,600 69,900 31,600 27,900 Total 26,200 500 28,000 112,800 54,900 44,500 Job creation based on DOS's"blended" data 2001 2002 2003 2004 2005 Average Jordanians 29,400 14,400 22,000 500 -9,400 11,400 Foreigners 4,400 4,400 20,700 28,800 2,900 12,500 Total 33,800 18,800 42,700 29,300 -6,434 23,900 Note: Totals and averages may not addup due to rounding. See Box 1.1 for description of sources. V. Labor market mismatches. The current situation is not so much an inadequate number of jobs as it i s a mismatch between the available jobs and the preferences of unemployed Jordanians. Understanding why so many new jobs have been filled by foreign workers and why other jobs remain unfilled is essential to aligning economic policies with the goal of creatingjobs for Jordanians. vi. Debate not sufficiently grounded in evidence. Hypotheses have been offered to explain the paradox of higheconomic growth and highunemployment. Common explanations cite factors such as geography, a "culture of ha me,"^ inadequate skills, job insecurity, and wages and working hours unacceptable to meet Jordanian standards of living. The debate is, however, fragmented and often not grounded in evidence. This report presents the available evidence and enumerates the most important impediments to reduction of unemployment. The report covers the period 2000 to 2005. vii. Coverage of report. This report also recommends actions to further the Government's goal of reducing ~nemployment.~To accomplish this we cover industrial policy, fiscal policy, regional development, education, and social protection programs. Although this report comments on employment aspects of several policies and programs, truly comprehensive analysis of these policies and programs i s beyond its scope. Rather, these findings and recommendations should be considered in the context of ongoing work by the Government, the World Bank, and other donors request of the Ministry of Planning and International Cooperation. For more information, please refer to footnote 13. The term "culture of shame" refers to the idea that certain types o f work activities are dishonorable and that performing these jobs would lower one's social status. Anthropologist Ruth Benedict coined the term in her 1945 book, The Chrysanthemum and the Sword. She proposed that Japan had a "shame culture" rather than the "guilt culture" that is common in the West. Shame cultures are heavily influenced by concepts of honor, obligation, and duty in a way that i s less true in individualistic Western culture. 4 The report emphasizes recommendations for government due to its role as an economic policy maker, information provider as well as a major employer and provider of safety nets. Resolution of the unemployment problem will require cooperation of others as well, especially entrepreneurs and labor. 11 in the various policy and program areas. That work includes analytical work and investments in the areas of (a) public expenditure, (b) financial sector functioning, (c) investment climate, (d) poverty mapping, and (e) pension reform. It also includes projects in (a) higher education development, (b) education reform for the knowledge economy, (c) social protection, (d) employer-driven skills development, and (e) regional and local development. ThreeMismatchesin theJordanian Labor Market: Geography,Employability,and Expectations viii. ... Three crucial mismatches explain the simultaneous existence of increasinglabor demand and high, sustained unemployment inJordan. 0 Geography.Newjobs and prospective workers are far apart. 0 Employability. Although Jordanian workers have sufficient education, vocational training, and job experience, employers often prefer foreign workers because of their workplace behavior and productivity. 0 Expectations.Jordanians maintain a false optimism about their employment prospects and earning potential. ix. Geography. New jobs are often created far from where unemployed workers live. About 65 percent of the jobs created between 2000 and 2004 were in Amman, although less than 40 percent of the unemployed population lived there. Other governorates did not see new jobs commensurate with their share of the unemployed. For example, Irbid, home to 17 percent of the unemployed, saw a net reduction injobs filled by J~rdanians.~ x. Government actions, including the investment promotion law, and locations selected for the industrial estates and free zones, have steered economic activity away from Amman. However, the effect of these actions on business decisions has been minor. And, while jobs did not move sufficiently toward the concentrations of unemployed Jordanians, neither did the unemployed move toward job opportunities. The percentage of working age people living in Amman grew less than 1percent between the 1994 and 2004 census rounds. xi. Employability. Jordanian workers are similar to foreign workers in their education, vocational training, and job experience. But they are often considered lacking in terms of their employability skills. When employers prefer foreign to Jordanian workers, surveys show that their preference can be traced to what mightbe called "employability" skills. xii. In a recent survey by the University of Jordan's Center for Strategic Studies: employers said that Jordanian workers work less, assume less responsibility, and are absent more often than foreign workers. When asked "Are Jordanian workers as valuable as foreign workers and, if not, why not?" the majority responded that Jordanian workers were less valuable than foreign workers. Only 6 percent believed that this was due to less experience. Rather, employers cited workplace behavior+ffort, responsibility, attendance, and flexibility. In addition to being more The figures in this paragraph include unemployment of Jordanians and jobs filled by Jordanians. They do not include foreign workers. Center for Strategic Studies, University of Jordan (2006b). expensive per unit of time, the survey suggests that Jordanian workers are considerably more expensive per unit of output. xiii. ... Expectations. The most important mismatch may be between the high expectations of the unemployed and the prevailing low wages of available jobs. Many proposed explanations (e.g., "culture of shame," wages that erode living standards, job insecurity, unacceptable hours) are alternate ways of stating the same truth: current jobs are seen as not worth taking, so many stay home. Available evidence confirms that there is no "culture of shame" in an absolute sense; even unpleasant jobs would be acceptable if they paid well enough. While this finding is illuminating, the question remains: are unemployed Jordanians willing to accept today's jobs at theprevailing wages? xiv. The evidence suggests that fewer than half of unemployed Jordanians say yes. When asked about their willingness to accept jobs from a list of available jobs at prevailing wages, 46 percent were willing to accept at least one of the jobs. The majority did not want to work in the existing jobs. Labor economists refer to the second group as "voluntarily unemployed." The Government can reduce unemploymentmore effectively if it adopts a separate approach for the voluntarily unemployed: traditional approaches, such as increased investment, may not work for this group. Priority:Assist the Unemployedbut Willing xv. Assisting the 46 percent minority willing to accept existing jobs is a clear priority, particularly because so many in this group are likely to come from poor households. Relevant programs typically include a combination of employment services, training, and income support, but if such programs are to be effective policy makers need to know the relative importance of geography, employability issues, and poorly functioning labor markets. xvi. Focus training programs on employability skills. Training programs - through the school system and the TVET system - should focus more strongly on the employability skills employers seek. Expansion of programs like INJAZ will support this goal. Reform of the training system must include a redefinition of the role of the public sector so it focuses only on tasks that the private sector cannot do. Priority: Createan Environment to Promote Growthof High-WageJobs xvii. Encourage the creation of high wage jobs - or try to lower expectations. Policymakers are also concerned with the voluntarily unemployed because their unemployment foments dissatisfaction and could have social and political repercussions. To a large extent, assisting the voluntarily unemployedhinges on this question: CanJordan competeinternationally in high-wage products? If the answer i s yes, the Government should eliminate distortions that affect the types ofjobs that are created. If the answer i s no, unemployment will decline among the voluntarily unemployed only if this group lowers its expectations to match the kind of production inwhich Jordan can compete. xviii. Eliminate legal and regulatory distortions. Existing policies encourage entrepreneurs to focus on low-wage activities. For example, current labor laws have created benefits targeted at agriculture and the garment industry. The tax system, too, provides large advantages to some sectors, such as agriculture, at the expense of others, including many "knowledge economy" services such as communications and finance. iv xix. Set immigration and emigration policy consistent with the vision for Jordan's future. Jordan's policy of easy entry for uneducated foreign workers and active encouragement of the emigration of educated Jordanians affects the composition of the labor force. For entrepreneurs, the pool of available labor appears to have plenty of unskilledworkers, but many fewer skilled workers, a combination that favors business activities that emphasize unskilled labor. The impact of the distorted pool of labor i s reinforced by differences in the employment contracts of foreign workers vis-h-vis Jordanian workers (temporary vs. permanent) and different degrees of enforcement of labor regulations that make foreign workers cheaper than similarly qualified Jordanians. xx. Encourage creation of new businesses. International evidence shows that newly created businesses are crucial for job creation as well as economic discovery and export diversification. Because new businesses are often small, addressing constraints faced by small businesses is important. The World Bank's recent Investment Climate Survey identified the most important constraints for small businesses as political and macroeconomic instability, high tax rates, corruption, and difficulties with business licensing and permits. Improving labor policy can also encourage creation of new businesses. Although labor policy in Jordan is generally very good, employment-protection legislation-associated with low job creation for women and youth-is an area that needs attention. xxi. The Government can give the economy its best chance to compete internationally in high-wage products in three ways: (i) removing legal and regulatory distortions; (ii)discouraging use of foreign labor and encouraging skilled Jordanians to contribute their skills to the Jordanian economy; and (iii) encouraging creation of new businesses. Priority: Increase Employability and Incentives toAccept Existing Jobs xxii. Increase employability by linking quality education with employer needs. Inthe long term, Jordan's comparative advantage depends on how well the education system provides Jordanians with relevant skills. For many years, Jordan has been more educated than neighboring countries. Recent reforms focused on educational quality mean that Jordan can become competitive in terms of workplace behavior and productivity as well. Emphasis on linking high- quality education with employer needs (e.g., through employability skills, internships, and career counseling) would effectively leverage these achievements. The right policies outside the education system can ensure use of educational investments and thereby increase the productivity of the labor force. Centralized financing of maternity leave benefits, for example, has been found to reduce discrimination against women and lengthen their tenure inthe labor force. xxiii. Ensure that social-protection reforms increase incentives to work. Long-term efforts to balance labor demand and supply will create the best new jobs for Jordanians. The mix of jobs, however, changes gradually. In the short term, reducing unemployment among the voluntarily unemployed depends on increasing incentives to accept existing jobs. Existing social-protection programs-the National Aid Fund (NAF), pensions, and income support for the unemployed-do not appear to create major disincentives to work. However, as the existing programs are reformed, care must be taken to preserve or amplify incentives to work. Introducing unemployment insurance and improving the targeting of the NAF (to cover the working poor) are both valuable reforms, yet potentially introduce disincentives to work. Mechanisms that have been found to work well in other countries include linkingbenefits to participation in other labor market programs, such as employment service programs. V xxiv. Remove the disincentives caused by civil service hiring practices. Government recruiting and hiring practices should reduce disincentives to accept existing jobs. By accepting applications even when no jobs are available, the Civil Service Bureau sends a misleading message. About 20 percent of the unemployed have registered with the Civil Service Commission, a sign that they expect that a civil servicejob may be forthcoming. Eliminating this practice and redoubling efforts to minimize new hiring will reduce the voluntary unemployment created by misplaced expectations. xxv. Encourage realistic labor market expectations through social marketing and broad dissemination of labor market data. Social marketing (as has been used successfully in Jordan on health issues) can be usedto promote the notion that all types of work are respectable. Among the most important roles of government in labor markets is the formal collection and broad dissemination of information (e.g., on occupational data, job locations or prevailing wages) for businesses andjob-seekers. Although Jordan i s among the best in the region in this area, more needs to be done to improve the collection of information on the opportunities available to Jordanian workers and to ensure the general public's perceptions of the labor market are accurate. Conclusion xxvi. Reducing unemployment requires a focus on its causes. Separate approaches to both the involuntarily and voluntarily unemployed will increase the efficacy of Government actions. xxvii. The involuntarily unemployed are a clear priority. Employment among those who actively seek work can be supported through increased focus on employability skills. xxviii. A strategy to help the voluntarily unemployed. In the long term, voluntary unemployment can be reduced by aligning policies with the goal of making Jordan a knowledge economy. Such alignment requires (i) removing legal and regulatory distortions, (ii)discouraging use of foreign labor and encouraging skilled Jordanians to contribute their skills to the Jordanian economy, and (iii)encouraging creation of new businesses. In the short term, voluntary unemployment will require increasing incentives to accept existing jobs by (i) ensuring social protection reforms increase incentives to work, (ii) removing disincentives caused by civil service hiring practices, and (iii)encouraging realistic labor market expectations through social marketing and broad dissemination of labor market data. vi CHAPTER JORDAN'S LABOR 1. MARKET PARADOX Some 175,000 Jordanians are unemployed-nearly 15 percent of the workforce. At the same time, at least 300,000 non-Jordanians are working inthe kingdom. These numbers bring a critical paradox into sharp focus: demand for labor is high, but job vacancies are filled by foreigners or remain unfilled, while many unemployed Jordanians remain so, often for long periods. This situation implies that additional investment and new jobs are will not be enough to cure unemployment. Other steps are needed to needed to address mismatches between jobs and workers. A. Jurdan :rEinyloymentSfrutegy andRecentEconuinikResukr 1.1. Improving living standards through employment is a top priority of the current Government. The goal of the Kulna AI-Urdun(We Are All Jordan) program, in accordance with the National Agenda, i s to "iniprove the quality of life of Jordanians through the creation of income-generating opportunities, the improvement of standards of living, and the guarantee of social welfare.'" Figure 1.1. Income from employment accounts for most household income (richest) quintile 5 1 quintile 3 (poorest)qmr e 1 __7 0 20 40 60 EO 100 % share of income employment 0 rents 13transfers Source: HouseholdIncomeandExpenditureSurvey 2002. 1.2. Income from employment is the principal mechanism through which GDP growth is distributed among the population. Jordanians rely strongly for their incomes on their major asset-labor. Even the richest 20 percent of Jordanians derive more than 55 percent of their income from employment (Figure 1.I). King Abdullah launched the National Agenda initiative in 2005 as a process for setting guidelines, programs, and strategies to reform the political, educational, social welfare, and state financial systems over the coming decade. The 26-member drafting committee, established by royal decree, was chaired by Deputy Prime Minister Marwan Muasher and included former ministers, politicians, academics, and businessmen. The "We Are All Jordan Commission" was formed in 2006 to follow up on the National Agenda and the recommendations of the We Are All Jordan Forum. The resulting action plan i s a detailed, costed plan for moving forward on an integrated set of reforms. 1 1.3. Taking advantage of its "demographic gift" is one of Jordan's main challenges. As Jordan now enters the third stage of its demographic transition,' it experiences a bulge in the working-aged population relative to the size of the child and elderly population. Whereas in 1979 less than half of the population was of working-age, by 2020 more than two-thirds of the population will be of working-age (Figure 1.2). The demographic gift presents Jordan with an opportunity for high rates of savings and investment. Of course, the demographic gift becomes a curse if the bulge in working-aged population remains unemployedor outside of the labor force. Figure 1.2. Changingage distributionof the population 1979 1994 2004 2020 children 0working-agedadults 0 elderly Source: DOS, Social Trends inJordan, Volume 3, July 2007 1.4. Because wage income is so important to the population's financial well-being, the Government gives priority to reducing unemployment. The Government's main strategy has been to attract foreign investment and generate growth to create 46,000 additional jobs per year, with the expectation that this will reduce unemployment to 6.8 percent by 2017 (National Agenda 2005). Estimates vary on the number of new jobs needed each year to reach the target, and on the extent of investment growth necessary to create thosejobs.gIn any case, the strategy i s the same: a strong focus on investment and growth will reduce unemployment. But more jobs may not be enough to reduce unemployment, for the reasons set out inthis report. 1.5. Measured in terms of investment and GDP, the Jordanian economy has performed quite well over the past five years. Private foreign investment, in particular, has increased dramatically since 2001, triggered by soaring oil profits among Gulf neighbors and the immigration of wealthy Iraqis. Annual GDP growth increased from 3.5 percent during 1996-2000 Like other developing countries, Jordan has been going through a demographic transition. In the first stage of the transition, population growth i s stable with highrates of both births and deaths. In the second stage population increases rapidly as death rates decline (due to improved health care, food supply and sanitation) while birth rates remain high. The second stage i s characterized by a large bulge in the population of children. In the third stage, population growth levels off as birth rates drop. The demographic gift i s temporary and eventually the population bulge becomes elderly and i s dependent on a smaller working-aged population. See Birdsall et. al. (200I). The Department of Statistics estimates that the Jordanian labor force increaser by 40,000 persons per year. Thus, a minimum of 40,000 net new jobs are neededjust to keep unemployment from rising. See DOS (2007). 2 to 6 percent during 2000-05 (Figure 1.3)." That strong performance was underpinned by comprehensive structural reforms duringthe 1990s, which improved the environment for private investment, liberalized the trade regime, supported a robust privatization program, and initiated modem regulatory and institutional frameworks in energy and water. Indeed, in the mid-I990s, Jordan ranked as one of the best reformers, not only in the region but also compared with other middle-income countries (World Bank 2006d). 1.6. Consistent with the expectation of the government's unemployment reduction strategy, GDP growth has generated a substantial number of new jobs. From 2000 to 2005, the Jordanian economy created between 24,000 and 44,000 additional jobs per year. However, unemployment did not fall as expected, but has remained relatively steady near 14.5 percent. In this context, the strategy for reducing unemployment needs to be reconsidered. Figure 1.3. The paradox: strong investment and GDP growth, but high unemployment ?n 25i / 2L 0 2001 2002 2003 2004 2005 2006 1999 2000 2001 2002 2003 2004 2005 1999 2000 2001 2002 2003 2004 2005 Foreign direct investment,2001-06 GDP growth rate, 1999-2005 Unemploymentrate, 1999-2005 Sources: CentralBank of Jordan, World DevelopmentIndicators,Employmentand UnemploymentSurveys. 1.7. A close look reveals that many newly created jobs were not filled by Jordanians but by foreign workers." Although specific numbers vary depending on the data source used, the story is consistent (Table 1.1). Estimates based on the DOS Employment and Unemployment Survey and the Ministry of Labor's database of worker permits find that 63 percent of newly createdjobs were filled by foreign workers. Estimates based on DOS's "blended" data set finds that 52 percent of newly created jobs were filled by foreign workers. As a result of this trend, foreign workers in Jordan now number 250,000 to 300,000, far more than the number of unemployedJordanians (175,000 in 2005). lo Measured inper capita terms, GDP growth increased from about 0.5 percent in 1996-2000 to more than 3.5 percent in 2000-05. See World Bank 2006d. l1 Calculations of job growth in Jordan are based on jobs filled by both Jordanian and foreign workers. Calculations of Jordanian unemployment, however, are based only on those jobs filled by Jordanians. If we were to calculate the unemployment rate of the labor force currently living in Jordan (including the foreign workers) the unemployment rate would be significantly lower than the unemployment rate of Jordanians. The Government of Jordan rightly focuses on the unemployment rate of Jordanians because that is the population whose welfare the Government aims to increase. 3 Table 1.1. All official data sourcestell the same story: many new jobs but most filled by foreign workers Job creation based on DOS's EUSand MOL data 2001 2002 2003 2004 2005 Average Jordanians 200 9,900 6,400 42,900 23,300 16,500 Foreigners 26,000 -9,400 21,600 69,900 31,600 27,900 Total 26,200 500 28,000 112,800 54,900 44,500 Job creation based on DOS's"blended" data 2001 2002 2003 2004 2005 Average Jordanians 29,400 14,400 22,000 500 -9,400 11,400 Foreigners - 4,400 4,400 20,700 28,800 2,900 12,500 Total 33,800 18,800 42,700 29,300 -6,434 23,900 Note: Totals and averages may not addup due to rounding. See Box 1.1 for descriptionof sources. 1.8. If all of the new jobs created from 2001 to 2005 had been filled by Jordanians, the country would be on track to meet its unemployment goal.'* With less than half of newjobs being taken up by Jordanians, even strong job growth cannot relieve Jordan's high rates of unemployment. 1.9. The co-existence of high job growth with persistent unemployment among Jordanians indicates the presenceof mismatches in the labor market. This report investigates these mismatches and examines policy options that might help in reducing them. The report i s organized in three main sections. Section I1describes three mismatches in the Jordanian labor market arising from considerations of geography, workplace behavior, and workers' expectations. Section I11examines policy alternatives in the area of labor demand-focusing on investment climate, trade, and labor policy. Section IV turns to labor supply-xamining quality of education and vocational training, women's attachment to the labor force, and the high expectations of unemployedJordanians. 12 Reliance on Jordanian rather than foreign labor has other advantages as well. The Jordanian economy benefits more from "learning by doing" when skills are acquired by Jordanians rather than transient foreign workers. And, because Jordanian workers consume a far higher share of their earnings within Jordan, the multiplier effect i s higher from jobs filled by Jordanians. 4 Box 1.1. Sources of data usedinthis report Several official sources of data have been used for this report. EUS and Census data. The Department of Statistics (DOS) conducts a quarterly household-based employment and unemployment survey (EUS). The EUS provides detailed information on the employment status of various segments of the population. Where total numbers o f individuals, rather than proportions are needed, we combine the EUS with Jordanian census data from 1994 and 2004. Year-on-year changes in the number of employed persons are then used as an estimate of netjob creation. MOL data. The EUS does not report data on foreign workers. For that reason, we have supplemented the EUS with Ministry of Labor data on registered foreign workers-officially 260,000 in 2005. Those data have several limitations. For one thing, it i s widely recognized that many unregistered foreign workers are present in Jordan. Some estimates suggest that their true number of foreign workers is closer to 600,000. Moreover, the ministry's database contains few variables, which limits the possible analysis that can be performed. DOS "blended" data. The DOS "blended" data set combines the data from the EUS with data from several enterprise-based surveys including the Employment in Enterprises (EIE) survey and surveys of the agriculture and construction sector^.'^ The DOS "blended" data set does not include foreigners working in household-based activities, such as house-keepers. We used this data source at the request of the Ministry of Planning and International Cooperation. The various sources each have their advantages and disadvantages. The EUS i s useful because it provides estimates of employment that are consistent with the official estimates of unemployment. (DOS's blended data set sometimes finds employment trends that contradict unemployment trends.) Enterprise-based surveys, including the EIE, construction and agriculture surveys (which are used in DOS's blended data set) provide reliable information on sectors of work. (EUS information is generally taken from housewives who may not be well informed about the details of family members' employment.) The M O L database of registered foreign workers suffers from three types of inaccuracies, with an ambiguous net effect. First, by definition, the database excludes illegal workers, therefore underreporting the true number of foreign workers. Second, the number of permits issued is slightly larger than the number of permitted workers due to duplicate permits. While the first problem tends toward underreporting, the second problem tends toward over-reporting. Third, the M O L periodically has campaigns to issue permits to foreigners previously working illegally. These campaigns increase the accuracy of the database in terms of number of foreign workers, though it may show an increase in the number of foreign workers when, in fact, the additional workers were not new but were simply becoming legalized. Enterprise-based surveys, including the EIE, construction and agriculture surveys (which are used in DOS's blended data set) tend to underreport the number of foreign workers becauseemployers 13In the DOS "blended" data set numbers of Jordanian people working inthe following sectors were taken from the EUS: agriculture; construction; some manufacturing industries; wholesale and retail trade; hotels and restaurants; transportation storage and communication; and other community services. The EIE was used for the numbers of Jordanian people working in sectors covered by large samples including: some manufacturing industries; public administration; education; health; financial mediation; real estate activities; and electricity, gas and water supplies. The numbersof foreign workers were taken from the EIE except for agriculture and construction. Agriculture surveys were used to estimate the numbers of foreign workers in the agricultural sector. Construction surveys were used to estimate the numbers of foreign workers in construction. Because the data on the organized construction sector was considered inaccurate, the total number of foreign workers in construction sector i s estimated as twice that o f the unorganized construction sector. 5 may be afraid of being caught hiring un-permitted foreign workers. As the attention paid by the Government to foreign workers has increased in recent years, such underreporting is likely to have increased over time and appears to be large: DOS's blended data set indicates a total of about 160,000 foreign workers (including both legal and illegal workers) in Jordan during 2005 -far less than the 260,000 legal foreign workers registered with the MOL. Data on Jordanians working outside of Jordan, were such data available, would have allowed a much more thorough treatment of the labor market, because of the very large number o f well-educated workers who take jobs outside of Jordan. As with foreign workers inside Jordan, opportunities for work outside the country have a significant impact on labor supply inside the country. It i s unfortunate that such data are not collected. The above mentioned data sources were supplemented with macroeconomic data from the National Accounts and Central Bank o f Jordan reports. Throughout this report we also refer to findings from the World Bank's Investment Climate Survey (2006~).This survey was conducted through face-to-face interviews with 503 enterprises in Amman, Irbid and Zarqa during 2006. The sample included enterprises with five or more employees. In addition to interviews with enterprise management, interviews' were conducted with 10 workers per enterprise, controlling for job category. For more information on the World Bank's Jordan Investment Climate Survey, see httt,://iresearch.worldbank.org/InvestmentClimate/. Where useful, we provide comparisons with other countries and regions. The countries used as comparators are those with GDP per capita (in purchasing power parity terms) similar to Jordan's, as identified in Iqbal (2006). B. UnempZoymentDefzned 1.10. Individuals of working age (age 15 and over) are either (a) employed, (b) unemployed, or (c) not in the labor force. By definition, both the employed and the unemployed are included in the labor force. 1.11. According to the Government's Department of Statistics, an individual i s unemployed if he or she i s without work, able to work, available for work, and looking for work." The distinction between being unemployed and not in the labor force i s not as clear cut as it may appear. On one side, many of those counted as unemployed are not actively searching for work; an individual i s counted as searching for work even if he or she has simply asked a relative or friend for a job during the past month. On the other side, many of those counted as not in the labor force are in fact available for work but are not seeking work because they believe no work i s available, are tired of seeking work, or for other reasons. These individuals are sometimes referredto as "discouraged workers." 1.12. The unemployment rate is defined as the number of unemployed individuals as a share of the number of individuals in the labor force. A third of Jordanian adults were employed in 2005, while 6 percent were unemployed (Figure 1.4). Because both groups are in the labor force, the labor force participation rate i s 33 percent plus 6 percent-that is, 39 percent. The unemployment rate, therefore, was 6 percent divided by 39 percent-that is, 15.4 percent. 1.13. Jordan has an unusually high rate of unemployment and an unusually high proportion of the population that is not economically active. Both the female and male labor force participation rates are among the lowest in the w0r1d.l~ l4 The EUS reports the male labor force participation rate in 2005 at 64 percent. This compares to 89 percent in Syria, 84 percent in Morocco, 78 percent in Tunisia, 77 percent in Egypt. The EUS reports the female labor force participation rate in 2005 at 12 percent. This compares to 39 percent in Syria, 29 percent in Morocco, 30 percent in Tunisia and 22 percent in Egypt. (Data for countries other than Jordan i s 6 Figure 1.4. International comparisonsof labor force status EAP Paraguay Philippines LAC Jamaica ECA Syria SAR Sri Lanka Guatemala Morocco MENA El Salvador Egypt Jordan c- 1-- I I --7 0 10 20 30 40 50 60 70 BO 90 100 employed (in the labor force) unemployed(inthe labor force) C3 not in labor force Sources World DevelopmentIndicators2006, Department of Statistics Employment and UnemploymentSurvey 2005. Note: Thc World DevelopmentIndicatorsestimateJordan's employmentat 48 percent of the adult population, unemploymentat 7 percent of the adult populationand45 percent of the adult populationas not inthe labor force. These figures placeJordan close to the MENA average. Note: EAP=East Asia and Pacific, LAC=Latin America and Caribbean,ECA=EuropeandCentral Asia, SAR=South Asia, MENA=Middle East and North Africa. C Characterzktzksofthe Uizemployed 1.14. Before moving onto the main body of the report, we finish this section by studying the characteristics of the unemployed by education, age, and gender. Policy must be tailored to the groups among the unemployed that are (a) the most numerous and (b) have the highest rates of unemployment. These two criteria do not always coincide, which complicates the task of policy makers. 1.15. Age. Unemployment is a phenomenon of youth, both in the percentage of the unemployed and in unemployment rates (Figures 1.5 and 1.6). Unemployment rates are high among those aged 15 to 19 years; however, since only 12 percent of this age cohort participate in the labor force, this group constitutes only 16 percent of the unemployed. By contrast, the group aged 20 to 24 years constitutes a much larger share (38 percent) of the unemployed. Although their unemployment rate is somewhat lower (29 percent compared to 39 percent for the younger cohort), the labor force participation rate of 20 to 24 year olds i s much higher (45 percent compared to only 12 percent for the younger cohort). Unemployment i s relatively rare among for 2005 or closest year, uses the same age category as the EUS, and is taken from the World Development Indicators.) 7 those aged 40 years and over, a group that constitutes just 7 percent of the unemployed. Policy making must focus on youth. 1.16. Youth unemployment is not just a problem of finding a first job. Although high unemployment among young people may be related to lack of experience, 60 percent of the unemployed have previously heldjobs (Figure 1.7), though far fewer young women than young men have done so. Youth unemployment i s related to frequent episodes of job gains and losses, known as "churning." Half of the unemployed have been unemployed for 6 months or less; 80 percent for 12 months or 1e~s.l~Churning usually indicates a healthy economy adjusting to changing circumstances through job creation and destruction.16It i s detrimental, however, if high turnover destroys incentives for workers and employers to invest in education and training or if churning pushes labor into less productive sectors. Younger workers are more susceptible to churning than older workers becausethey have had less time to develop enterprise-specific skills. Figure 1.5. Age andlabor force status (employed, not employed, not inlabor force), 2005 65- years o o 55-64 {ears c c 1 42-54 {ears o o 25-39 years old 20-24 years o 15-19 years old 0 20 40 60 80 100 % population ekeemployed (in the labor force) 0 unemployed (in the labor force) not in labor force Source. EUS 2005. 15An examination of data from 2000 to 2005 shows no apparent trend in previous job experience of the unemployed or in the duration of unemployment. l6About 20 percent of enterprises are created and destroyed each year in many countries, involving 10-20 percent of the workforce (World Bank 2005b, 2006b). 8 Figure 1.6. Age and unemployment,2005 Unemployment rates by age Age distribution of the unemployed population , 65+ years oia a 2 55-64 years oia_r65+ years old 0% 55-64 years old 4 40-54 years oia 5 25-39 years oia 20-24 years old 29 0 5 10 15 20 25 30 35 40 45 unemployment rate Source: EUS 2005 and Census 2004. Figure 1.7. Characteristicsof youth unemployment,2005 Previous work experience of unemployed youth Duration of unemployment among youth - .. 25 +6% Months lena e 033 I1Month 13-24Months .I. .. " male 73 2-3 Months 18% L , 16% 1 0 10 20 30 40 50 60 70 80 % of unemployed Source:Census 2004 and EUS 2005. 1.17. Education. The highest rate of unemployment is found among the 10 percent of the adult population that holds a bachelor's degree (Figure 1.8). Despite this fact, most unemployed individuals have no more than secondary education. Because both the less-educated and more-educated face difficulties, policy makers need to address the problems of both groups. 9 Figure 1.8. Education and unemployment,2005 Unemployment rates by educational level Educational status distribution of the unemployed population Illiterate 1% IntermediateDiploma 144 Secondary LessThan Secondary 15 1 Illiterate 9 2 Secondaw L ___ 14% ~ 0 5 10 15 20 Yo population Sources: EUS 2005 andCensus 2004 1.18. Gender. The unemployment rate for women (26 percent) is twice as high as that for men (13 percent). Nevertheless, women comprise only 27 percent of the unemployed population because of their very low labor force participation rate (12 percent compared to 64 percent for men, Figure 1.9). Women who identify themselves as housewives are considered to be outside the labor force.17 Figure 1.9. Labor force status by gender, 2005 0 20 40 60 80 100 % population employed (in the labor force) 0unemployed(inthe laborforce) 0not inthe laborforce Sourre: EUS 2005. 1.19. Women's participation in the labor force is strongly correlated with education (Figure 1.10a). Few women with a secondary education or lower participate inthe labor force. By contrast, nearly two-thirds of women with a university education do so. Although on the whole women make up just 27 percent of the unemployed population, they constitute the majority l7 Housewives are considered not to be in the labor force because they are not engaged in production of goods and services that are exchangeable in the market. For further discussion of nonmarket work (sometimes referred to as the "care economy"), see World Bank (2004a). I O among unemployed persons with a university education (Figure 1.10~).Therefore, any policy designed to addressunemployment among the most educated must pay attention to women. Figure1.10. Gender andunemployment a. Labor force participation by gender and education 100 - g 90- ..................... -Z % + 80- 70- illiterate less than secondary intermediate bachelors secondary diploma degree+ ......male -female Source: EUS 2005. b. Unemploymentrate by gender and education female, bachelor'sdegree + 30 female, intermediatediploma 25 female, secondary female, less than secondary female, illiterate 0 male, bachelor's degree + 12 male, intermediate diploma 9 male, secondary male, less than secondary 15 male, illiterate 9 0 5 10 15 20 25 30 unemployment rate Source: EUS 2005. 11 c. Gender and education distribution of the unemployed population male, illiterate female, bachelor's r 1% male, bachelor's male, less than secondary 42% intermediate diploma male, intermediate female, less than female, secondary: I secondary 3% I 3% male, secondary) 11% Source: EUS 2005 and Census 2004 12 CHAPTER 2. THREEMISMATCHES INTHE JORDANIAN LABOR MARKET 2.1. Several hypotheses have been advanced to explain the coexistence of high growth and high unemployment-and the mismatch between new jobs and unemployed Jordanians. Among the components of those hypotheses are geography, inadequate skills, unacceptable working hours, job insecurity, a prevailing wage that does not provide a reasonable standard of living, and Jordan's "culture of shame" (the idea that certain types of work are dishonorable and demeaning). The debate around these hypotheses, however, i s fragmented and often not grounded in evidence. Here we marshal evidence to explain Jordan's labor market paradox in terms of three mismatches: geography, employability, and-most important- expectations. A. Geugraphy: TheDzktancebetweenNew Jobsandthe Unemployed 2.2. Many of the new jobs created in recent years have not been created where the unemployed reside. There is a clear dichotomy between Amman, the capital city, and the rest of the country. 55 percent of the jobs created between 2000 and 2005 were in Amman (Table 2.1, column B), which i s home to 38 percent of the unemployed (column A). In most other governorates, job creation was typically not commensurate with the governorate's share in total unemployment. Amman's disproportionate share of job creation i s even starker when we look only at jobs that have been filled by Jordanians (column C). For example, Irbid, home to 17 percent of the unemployed, saw 11 percent of all new jobs, but a net loss in jobs filled by Jordanians." As a result of the mismatch in location of new jobs, unemployment rates fell in Amman and Zarqa between 2000 and 2005, but rose in all other governorates (Figure 2.1). In some cases the increases were large: in Aqaba, Irbid, Tafilieh, Maan, and Karak unemployment rates increased by 5 percentage points or more. Table 2.1. Governorate shares injob creation and unemployment at beginningof the period % ofjob creation % of unemployed % ofjob creation 2000-2004 % ofjob creation Jordanians in2000 2000-2004 (total) (Jordanian) 2000-2004 (foreign) (a) (b) (c) (4 Amman 38 55 67 48 Irbid 17 11 -1 17 Zaraa 15 14 22 10 Balaa 7 9 6 10 Mafraq 5 4 3 4 Karak 5 3 -2 6 Madaba 3 2 2 3 Jarash 3 1 1 1 Ajloun 2 0 0 0 Tafileh 2 1 0 1 Ma'an 2 1 0 I Aqaba 1 0 3 -2 Total 100 100 100 100 Sources: EUS 2000 and 2004, census 1994 and 2004, and MOL database. l8A large share of the jobs created in Irbid were in Qualified Industrial Zones (QIZs), which tend to hire few Jordanians. For more information on QIZs, see Section 3 of this report. 13 Figure 2.1. Unemployment rate by governorate, 2000 and 2004 Amman Zarqa Balqa Aqaba Ajloun Madaba lrbid Jarash Mafraq Maan Tafiela Karak 0 5 10 15 20 25 unemploymentrate Sources: EUS 2000 and2004, Census 1994, and2004, and MOL database. 2.3. Why have the unemployednot moved toward the newly created jobs? Recent census data show surprisingly low mobility among working-age Jordanians (Table 2.2). Low mobility i s likely due to strong attachments to home-and to high housing prices in Amman. Mobility appears to be particularly constrained for women. A recent study by the Center for Strategic Studies (2006b) at the University of Jordan found that distance was the most important reason why unemployed women rejectedjob offers (28 percent of female respondents).l9 2.4. To promote job creation in disadvantaged locations, Jordan has been implementing a policy of offering tax incentives to businesses that locate in selected districts.*' This policy has been only partially successful. Jobs have been created in some of the Industrial Estates in disadvantaged areas, though fewer jobs have been created in the population centers of those areas; in many cases unemployed Jordanians are unwillingto commute the necessary distances to the Industrial Estates. Despite large investment and a great deal of business activity in the Aqaba Free Zone, unemployment in the Aqaba governorate increased from 10 percent in 2000 to 15 percent in 2005. A study by the U.S. Agency for International Development (USAID 2004) concluded that the influence of the available incentives on location of investment was minor. An apparently more successful approach has been the Social Productivity Program (SPP) of the Ministry of Planning and International Cooperation. Through the SPP, the Ministry builds on existing economic activities and the skills of the communities, and avoids introducing new activities (Jordan Business 2007). l9 This is consistent with studies from other countries, which find that female workers are less mobile than *'maleworkers because of their household responsibilities (see, for example, Assaad and Arntz 2005). For more on the investment promotion law, see Section 111. 14 Table 2.2. Geographicdistribution of Jordanianpopulationaged 15and older, by governorate, 1994and 2004 1994 2004 Change Ailoun 2 2 0.1 Amman 38 38 -0.4 Aqaba 2 2 0.1 Balqa 7 7 0.2 Irbid 19 19 0.3 Jerash 4 5 0.5 Karak 2 2 0.1 Ma'an 4 4 -0.1 Madaba 1 1 0.1 Mafraq 4 5 0.5 Tafilah 1 1 0.1 Zaraa 16 15 -1.0 Total 100 100 0.0 Urban 79 78 -0.6 Rural 21 22 0.6 Total 100 100 0.0 Source: DOS, 1994 and 2004 census. 2.5. The alternative to the current policy of encouraging employers to move jobs to where the unemployed reside is to encourage the unemployed to relocate to areas where demand for labor is high. This option takes advantage of agglomeration, which i s an important driver of productivity and cost savings. The efforts of the Greater Amman Municipality to create morebuilding space for residentialunits will contribute to this option by making housing in Amman more affordable. Policies such as taxing vacant land and government-guaranteedmortgageshavebeendiscussedas well.*' B. Employabz22y: TechnicalSkzlhandExperiencevx WorkplaceBehavzorand Productivz2y 2.6. Jordanians are not at a disadvantage vis-&-vis foreign workers in terms of quantity of education or vocational training. The majority of unemployed Jordanians have educational qualifications that closely match those of foreign workers (Figure 2.2).22Moreover, Jordanians have an advantage in having more on-the-job experience than their foreign counterparts. Jordanian workers have more experience both in their currentjob (an average of 5.1 years vs. 3.9 years for foreign workers) and inprior relevant experience (4.1 vs. 3.8). These facts are important given the widespread perception that foreign workers can be replaced by Jordanian workers as soon as Jordanian workers receive training. 21 See Mansur 2007 and al-Asad 2007. 22 The figure likely overstates the educational status of foreign workers, because it includes only foreign workers in enterprises and excludes the many less-educated foreign workers with informal contracts. 15 Figure 2.2. Educational achievementof Jordanian and foreign workers, 2004 Foreign Employees Jordanian Unemployed I 3 Jordanian Employees 0 20 40 60 80 100 % of category W More Than Secondary Secondary Vocational Apprenticeship 0 Less Than Secondary Sources: Employmentin EnterpriseSurvey 2004; EUS 2004 for unemployed. 2.7. Although foreign workers do not have an educational advantage over Jordanians, employers often prefer to hire them because foreign workers are perceived to be more "employable"-that is, to be more responsible, reliable, courteous, and attentive. According to the 2006 survey by the Center for Strategic Studies, 61 percent of employers believed that Jordanian workers were less valuable than foreign workers because they put in less effort, took less responsibility, worked less, and were absent more than foreign workers.23 2.8. Foreign workers may be more productive. The survey also indicates that foreign workers are perceived by employers to be more productive than Jordanian workers. Most evidence suggests that foreign workers are paid considerably less than Jordanian workers. But, even if foreign workers were paid the same as Jordanian workers, they could still be cheaper per unit of output. A member of the executive board of the Jordan Garment and Textile Exporters Enterprise Association was recently quoted as saying that "Jordanians produce 50 percent less" (Jordan Business, September 2006). If true, this would mean that foreigners would be cheaper per unitof output unless they were paidtwice the wage paid to Jordanian workers. 24 2.9. While Jordanians are not at a disadvantage in terms of quantity of education, concerns persist about the quality of Jordanian education. In addition to deficits in the teaching of initiative and critical thinking, one hears concerns about the relevance of the fields of specialization of Jordanian workers. Among university specializations, unemployment rates in agriculture, veterinary, and educational specializations have remained twice as high as those of engineers and health care workers (Figure 2.3). Some tension between supply and demand i s inevitable in any dynamic economy, because of the time required to obtain skills and because needs for technical skills continue to change. An efficient labor market, however, should 23 The World Bank's Investment Climate Survey (2006~)found that 58 percent of foreign workers worked overtime in the previous month, compared to 40 percent of Jordanian workers. 24Wage data from the EUS and M O L databases indicate that, on average, foreign workers are paid considerably less than Jordanian workers per month. (The vast majority of foreign workers are identified in the MOL database as receiving minimum wage.) On the other hand, the Jordan Garment and Textile Exporters Enterprise Association has argued that foreign workers are slightly more expensive than Jordanians because employers pay for a round-trip plane ticket for the worker and for housing and meals. 16 minimize the duration of such a mismatch. Persistent gaps, however, indicate that students lack accurate information about the current demand for the specializations from which they can choose. Figure 2.3. Unemployment rates by university specialization, 2000-05 medicineand nursing architectureand civil eng engineering math biology fine atis law agnculture hoteVrestauranfftourism social sciences humanities physical sciences computers journalism businessand public admin education 29 0 5 i o 15 20 25 30 . % 01 categoly Source: EUS 2005. 2.10. The skills mismatch among women is more pronounced than among men. Women have a higher propensity to enter specializations in low demand, such as education and humanities, and a lower propensity to enter specializations in high demand, such as engineering (Figure 2.4). These differences are not large, however, and educational specialization explains only a small part of gender differences inunemployment. C Expectations:Fahe0ptzmzj;rlZzn aChangzngEconomy 2.11. Although individuals vary in the importance they place on various job characteristics, a high enough wage can usually compensate for undesirable features. The 2006 study by the University of Jordan's Center for Strategic Studies cited inthe previous section concluded that unemployed Jordanians would be willing to accept unpleasant jobs if they paid well enough. That is, there is no "culture of shame" in an absolute sense. The study surveyed unemployed Jordanians and asked them which jobs they would accept. By varying the hypothetical wages for ajob, the study showed that if paid enough, most unemployed Jordanians would be willing to accept the jobs that have been created over the past five years in industry, agriculture, and construction. 17 Figure 2.4. University specialization by gender engineering and 20 architecture health and medicine a0 math, science, computers 14 education 19 social sclence, 34 Dusinessand ,aw humanities 0 5 I O 15 20 25 30 35 40 %of students female male Source: EUS 2005. 2.12. But are unemployed Jordanians willing to accept the existing jobs at prevailing wages? By combining the survey findings described in the previous paragraph with information on prevailing wage rates for various occupations from labor force surveys, we can estimate the gap between the expectations of Jordanian workers and the compensation offered by available jobs. When unemployed Jordanian workers were asked about their willingness to accept jobs from a list of widely available jobs at prevailing wages, about 46 percent of respondents indicated a willingness to accept at least one of the jobs (Figure 2.5).25 The other 54 percent of unemployedJordanians want to work, but not at available jobs-unless the pay were significantly more than presently offered. Labor economists refer to the second group as "voluntarily unemployed." There i s a limitation to this study. The list of available occupations used was not exhaustive; some of the 54 percent may have been willing to accept some other job at the prevailing wage (e.g., many unemployed persons are willing to accept civil service jobs at the prevailing civil service salary.) Nevertheless, the list does include many of the jobs that have grown strongly since 2000 and the study i s useful in distinguishing between the two groups of unemployedpersons-the voluntarily and involuntarily unemployed.26 2.13. Government actions to reduce unemployment will be more effective if they distinguish between the voluntarily and involuntarily unemployed. A continuation of current investment-promotionpolicies may help the 46 percent who are involuntarily unemployed, but it will not work for the 54 percent who are voluntarily unemployed. To help the latter, a two- 25For example, respondents were asked whether or not they would be willing to accept factory work - the job with the highest acceptance rate -- at various wages. 22 percent of respondents said they would accept at 115JD per month, 43 percent at ISOJD, 61 percent at 180JDand 69 percent at 200JD. According to the EIEsurvey, the prevailing wage for factory work is 154JD. By interpolation, we estimate that 46 percent would he willing to accept factory work at 154JDper month. All other types of jobs had lower acceptance rates. ''The Center for Strategic Studies analyzed differences between men and women and among individuals with various educational qualifications in their willingness to accept jobs. Although the issues are highly relevant and deserve attention, the survey sample size was too small to allon conclusions to be drawn separately by sex or educational qualification. 18 pronged approach is needed. First,the Government should aim to create an enabling environment for the creation of more high-wage jobs. Second, it should focus on enhancing incentives to work. Figure 2.5. Share of unemployedJordanians who would accept one of the commonly availablejobs at prevailing wages cleaner(146JD) construction worker (164JD) fuel station worker (154JD) messenger(146 JD) buildingwatchman (162 JD) waiter (162 JD) I agricultureworker (146 JD) factory worker (154 JD) 46 54 0 20 40 60 80 100 % of respondents willing to accept 0 unwilling to accept Sources: Center for Strategic Studies 2006b andEmploymentin EnterprisesSurcey 2005. 2.14. Focusing on the involuntarily unemployed should be a high pri0rity.2~Most programs to assist the involuntarily unemployed combine employment services, training, and income support. But to set priorities and ensure good targeting, policy makers need more information about this group to determine the relative importance of geography, skills, and expectations intheir decision making. 2.15. Solving the problem of the voluntarily unemployed is more politically contentious and requires more sensitive government actions. Some argue that the government should not waste scarce resources on those who are not needy enough to accept existing jobs, while others argue that this group i s of utmost importance because their unemployment could have serious social and political repercussions. 2.16. The future of the voluntarily unemployed hinges on one question: Can Jordan compete internationally in high-wage products? The existing pattern of trade is the best indication of Jordan's comparative advantage. Jordan has had more success in low-wage, labor- intensive manufacturing than in services or high-tech products. Of course, the existing comparative advantage need not be seen as permanent. In the long term, governments can influence comparative advantage through various policies and by investing in the quality of the workforce. Nevertheless, it should be kept in mindthat change takes time and that other countries 27We do not have direct information on the characteristics of the involuntarily unemployed. If we assume this group coincides with the poorer 46 percent of the unemployed, we can identify certain characteristics that distinguish this group from the 54 percent voluntarily unemployed (assumed to be the wealthier 54 percent of the unemployed). First, the involuntarily unemployed are geographically dispersed, while the voluntarily unemployed are concentrated in Amman and Irbid. Second, the involuntarily unemployed have lower levels of education and are younger. Third, the involuntarily unemployed include a higher proportion of men. 19 are also changing. To be competitive, Jordan must not only improve, but move to the forefront. Even if Jordan can develop a comparative advantage in high-wage products in the long term, adjustingthe expectations of the workforce will be necessary in the short term. 2.17. The next chapter examines the challenges faced by Jordan in its quest for high-wagejob growth. 20 CHAPTER3. LABOR DEMAND: THEQUESTFOR HIGH-WAGEJOB GROWTH 3.1. With a workforce that is relatively highly educated and a decent investment climate, Jordan may be able to build a high-wage, "knowledge economy." The Jordanian Government can give the economy its best chance at competing in high-wage products if it can align all areas of policy with this goal. The main messages of this section are: 0 Increasing aggregate investment will not solve the unemployment problem. Aggregate investment i s already adequate, thanks to the influx of foreign capital. Investment climate policy should be geared instead toward encouraging creation of new businesses, especially small businesses,which tend to be better at producingjobs. 0 A variety of laws and policies draw entrepreneurs into favored sectors considered by the Government to be particularly important. By attracting investment to sectors considered strategically important, these economic incentives reduce investment available to other sectors that have the potential to generatehigh-wagejobs. A. RecentSectoraZPatterns afJb Growtt?8 3.2. From 2000 to 2005, the manufacturing and construction sectors created many jobs, the majority of which were filled by foreign workers (Figure 3. There was a significant increase of foreign workers in agriculture, while the number of Jordanians in the sector declined. The education, health and wholesale and retail trade sectors created a significant number of jobs for Jordanians. Few foreigners work inthose sectors. Throughout this section, we examine net job creation. For lack of data on newly created and newly eliminatedjobs, we use data on alljobs that exist at various points in time and examine the change over time. We use the terms "additional jobs" and "net job creation" interchangeably. 28This section addresses sectors following the International Standard Industrial Classification of economic activities (ISIC Rev 3) rather than in the sense o f formal vs. informal sector. The vast majority of Jordanians work as paid employees, with only 10 percent self-employed and only 2 percent unpaid workers. There was very little change in these percentages between 2000 and 2005. 29The data for figure 3.1 is taken from DOS's "blended" data set as reported in DOS (2007). Analysis of the sectoral patterns of job growth using the EUS and MOL data gives similar conclusions. The major differences are that the EUS and MOL data suggest (i)positive growth in the number of Jordanians in construction and (ii)larger growth in the number of foreign workers in agriculture, as compared to the "blended" data set. 21 Figure 3.1. Average annual netjob creation by sector, 200-05 construction public administration hotelsand restaurants real estate and business sewices other sewices Ql utilities i finance d mining "j -6,000 -4,000 -2,000 0 2,000 4,000 6,000 8,000 10,000 Sources: DOS "blended" data set. as reportedinDOS (2007) 3.3. GDP growth was more strongly correlated with growth injobs for foreign workers than for Jordanian workers. Several of the sectors with the highest GDP growth from 2000 to 2004-including manufacturing, and construction (Figure 3.2)-created a large number of jobs, albeit low-skilled, low-paying jobs that were unattractive to Jordanian workers. The finance sector also grew strongly, and tends to hire Jordanian rather than foreign workers, but did not create any net employment. Agriculture also grew strongly during this period, and created many jobs for foreign workers, though a net decrease injobs for Jordanian~.~' Figure 3.2. GDP andjob growth by sector, 2000-2004 -*O1 + m'n'ng -30 GDP growth Sorirces: DOS's "blended" data set, as reportedinDOS (2007),NationalAccounts 2000-04 30The strength of the relationship between job creation and GDP growth i s affected by the quantities of labor and capital and the efficiency of their utilization. An assessmentof the appropriate balance of inputs in the various sectors would be useful but i s beyond the scope of this study. 22 3.4. Most job growth for Jordanians occurred at the highest and lowest ends of the occupational spectrum, whereas mid-level occupations saw little or negative growth (Figure3.3). The implication of this pattern i s that opportunities are expanding for workers with higher education (bachelor's degree and above) and for those with little education. However, opportunities for individuals with secondary education and intermediate diplomas are declining, relegating them to jobs that require no more than an elementary education. We present data on job growth among Jordanians only, because available data on the occupations of foreign workers use a nonstandard classification and cannot be combined with the data for Jordanians. Figure 3.3. Average annual netjob creation (Jordanians only), by occupation, 2000-05 unskilledoccupations machine operators craft~menitrade~meii service and sales technicians b professionals - 4 000 2 000 0 2 000 4 000 6 000 8 000 10000 Sources: EUS 2000 and 2005, censuses 1994and 2004. 3.5. Adjustment to increased demand for labor has been reflected in wages as well as employment trends. Real wage growth was strongest in the sectors with highest growth in employment (and GDPJ3'Likewise, real wage growth was strongest in the occupations with the highest growth in employment. Recent wage growth i s likely to be unsustainable in some sectors, however, because the gains were not based on increased productivity. Changes in labor productivity are an important indicator of a country's competitiveness and thus of its potential for job and wage growth. The rate of growth of labor productivity in Jordan i s higher than the average for the region but lower than that of many other emerging economies (Figure 3.4) Sustainable wage increases (wage increases supported by productivity increases) were observed in key sectors of manufacturing, construction, wholesale and retail trade, finance, and health (Table 3.1).32However, wage increases in excess of productivity were observed in public administration, education, and hotels and restaurants, indicating unsustainability and decreased competitiveness. "Detaileddiscussionofwagetrendscanbefoundthereport"The EstimationofaWageIndexforJordan: 1994-98 and 2000-03," (Population Council 2006). 32Inthe cases of manufacturing and finance, wage growth was far below productivity growth. 23 Figure 3.4. Growth inlabor productivity 1990-2003 Morocco 0.5 Israel 0.7 Romania 1.2 Tunisia Mexico Egypt Turkey` 2.8 Bulgaria Malaysia Jordan Syria Chile Poland China -1 0 1 2 3 4 5 6 7 8 9 growth rate Source: WDI, staff calculations. Note: Data for Jordanare from 1995-2003; data for Tunisiaare from 1992-2003. Table 3.1. Average growth rate of labor productivity andlabor cost, by sector, 2000-03 Economic activity Labor productivity Labor cost GDPAabor employed JD/worker Mining and quarrying 14.5 9.6 Manufacturing 6.6 2.6 Electricity, gas, and water 3.7 2.7 Construction 10.7 9.4 Trade, repair services, and personal and household goods 4.0 0.7 Hotels and restaurants -20.4 -2.6 Transport, storage and communications 11.3 11.5 Financial intermediation 21.0 9.7 Real estate, renting, and businessactivities 4 . 7 -3.5 Public administration -0.2 2.2 Education -5.2 0.7 Health and social work 6.7 3.1 Other community, social, and personal service activities 2.1 -3.0 Sources: DOSEmployment in Enterprises Survey and National Accounts B. NewBuxzkexxex: TheKey toGreaterfob Creation 3.6. Currently, aggregate investment is very high. Private foreign investment has increased dramatically since 2001 because of soaring oil profits among Jordan's neighbors and immigration of wealthy Iraqis (Figure 3.5).33Arabs provided 96 percent of Jordan's total external 33Recent policy changes have made it easier for Iraqis to invest. In the past, Iraqi investors and entrepreneurs were forced to find Jordanian businesses to avoid the numerous restrictions and exorbitant business licensing fees imposed on them. The Government has launched a program to enable Iraqi 24 financing in recent years.34A recent report by the United Nations Conference on Trade and Development (UNCTAD 2006) ranked Jordan among the top 20 most attractive countries for foreign direct investment (FDI), due to its recent privatization of power, water, transport, and telecommunications services. Figure 3.5. Performance inattracting external financing, 2001-06 30 i 25 4 / 20. ------------_______ a --__ D --____ -_______ ----- 2 1 5 - 8 10 - ---- 2001 2002 2003 2004 2005 2006 --- Remittances -Foreign Direct lnvesmtment Sources: Central Bank of Jordan and Department of Statistics. Note: Data for 2006 are estimatedon the basis of the first three quarters. 3.7. In the longer term, aggregate investment may become an important concern, given Jordan's low savings rate. Gross national savings (which includes remittances) remains positive but has fallen as a share of GDP (Figure 3.6) and is among the lowest in the world (Lopez-Carlos et al. 2006). Gross domestic savings have been negative for many years. Jordan is thus vulnerable to shocks that affect the flow of remittances and FDI. To mitigate that risk, the Government has taken steps to reduce its stock of outstanding debt and its current pend ding.^' Reliance on foreign investment will continue, but care should be taken to minimize vulnerabilities by maintaining highreserves of foreign exchange at the central bank and controllinginflation.36 immigrants to obtain legal permanent residency by depositing substantial funds (typically JD 50,000, which they are not allowed to withdraw) in Jordanian banks (Financial Times, December 3, 2004). 34 According to Jordan Investment Board figures published in the Jordan Times (October 4, 2006), the United States and Canada provided 3 percent and Europeans provided only 1percent. 35 The government has successfully reduced debt from 102 percent of GDP in2001 to 83 percent of GDP in 2005, and has controlled the budget deficit. 36 Jordan i s less vulnerable in the short term than many countries because of its reliance on financing from Arab neighbors whose income from high oil prices i s likely to remain high and because of its reliance on FDI rather than short term borrowing. On the other hand, Jordan's pegged exchange rate makes it more difficult for the Government to adjust to changes in capital flows. 25 Figure 3.6. Gross national savings and gross domestic savings, 2000-06 40 1 -30' --- gross domesticsavings -gross national savings Source: CentralBank of Jordan. 3.8. Aggregate investment in productive (job-creating) activities is within reasonable ranges, though room for improvement remains. A large share of investment goes into activities that do not directly generatejobs. For example, of the FDIreceived in 2005, more than 35 percent consisted of purchases of existing company assets (including the large privatization program) by foreign in`vestors, and about 15 percent was comprised of purchases of land and real estate.37Nevertheless, capital investment-that is, investment in assets (plant and equipment) directly involved in production that can generatejobs-continued to grow, reaching 24 percent of GDP in2005.38 3.9. Creation of new businesses will be crucial to job creation and competitiveness and should receive increased policy attention. New businesses are crucial to job creation as well as economic discovery and export diver~ification.~' In all market economies, businesses enter and exit the market continuously. The contribution of new businesses to job creation is more than proportional to their presence in the economy. Whereas an average of 10 percent of businesses were created or shuttered in any given year in the countries of the Organization for Economic Co- operation and Development (OECD), new businesses contributed a far higher 30 percent of all new jobs (World Bank 2005a).40 37See Central Bank of Jordan 2005. 38 Jordan's gross capital formation as a percent of GDP i s within reasonable ranges as compared to other countries and regions. Though lower than the MENA region as a whole (27 percent) , South Asia's 31 percent, and East Asia's 40 percent, Jordan's GCF i s above that of Europe and Central Asia (22 percent) , Latin America (21 percent) and Sub Saharan Africa (19 percent). Data are drawn from the World Bank's UnifiedSurvey, an internal database. 39Newjobs are created when new businesses start up and when existing businesses adjust their workforce. The share of new jobs arising from start ups vs. existing businesses is related to the business' industry/sector. Successful businesses in high-tech manufacturing industries (ICT, radio and TV) and some service industries (communication and finance) often expand their workforce over time. In contrast, traditional manufacturing industries hire at start up but expand their workforce little after that. See Bartlesmand, Haltiwanger, and Scarpetta (2004). 40The rate is even higher (38 percent) in Eastern Europe and the Russian Federation as they develop their own market economies, and as former government enterprises are moved into the private sector. Interestingly, in Eastern Europe and the Former Soviet Union, foreign privatization i s associated with increased employment and wages in all countries, while domestic privatization i s more likely to be associated with unemployment and wage reductions. 26 3.10. Because business creation and expansion (as well as business closure and reductions) occur more frequently among small than large businesses, and because small businesses create a disproportionate number of new jobs, information on the obstacles faced by small businesses is especially valuable.41 Government can have a powerful effect on job creation by minimizing those constraints. The World Bank's Investment Climate Survey (2006~)asked enterprises whether any of 14 issues were severe or major constraints to doing business in Jordan (Figure 3.7). The most important constraints for small (as well as large) enterprises were political and macro instability, tax rates, corruption, and business licensing/permits. Importantly, all of the 14 issues were considered to be more severe constraints for small enterprises than for large enterprises-in some cases by wide margins. Figure 3.7. Issuesidentified as severe or major constraints to doing businessin Jordan by size of enterprise Labor Regulations Functioning of Courts Inadequately Educated Labor Force Access to Financing Tax Administration Business Licensing and Permits Corruption Tax Rates Macro Instability Political Instability 0 10 20 30 40 50 60 70 % of respondents Source: World BankInvestmentClimate Survey 2006c. 3.11. Corruption can be a severe impediment to starting a business. In particular, entry is dificult if favoritism (wasta) gives well-connected enterprises an advantage over less-well- connected but possibly more economically viable enterprises. Although Jordan generally is considered less corrupt than many other countries, there i s room for improvement, especially with regard to favoritism. Bribe-taking i s less prevalent in Jordan than in most other countries, according to Transparency International, which ranked Jordan 40 out of 163 in its 2006 index, although it had slipped from its previous rank of 37.42The poll of opinion leaders by the Center 41Reallocation of inputs (including hiring and firing of labor and other inputs) i s far higher among young and small businesses than among older and larger businesses. See surveys of evidence in Geroski (1995), Ahn (2000), and Davis and Haltiwanger (1999). 42Transparency International Corruption Perceptions Index 2006. Jordan's 2006 score was 5.3 out of 10 (with 0 being extremely clean and 10being extremely corrupt). In 2005 the score was 5.7. 27 for Strategic Studies (2006a), however, found that more than 60 percent of those interviewed believe that corruption is "rampant" in the public sector, and 46 percent believe it has risen in the past 4 years.43Unsurprisingly, corruption i s considered a more severe constraint among small enterprises (47 percent of enterprises) than among large enterprises (40 percent of enterprises). 3.12. Cumbersome business licensing and permits requirements also make it difficult to start a business (Figure 33).Despite significant efforts to reduce bureaucratic hurdles, Jordan ranks in the bottom quarter of countries inthe ease of starting a business (World Bank 2007a). Figure 3.8. Ease of starting a business Jamaica Paraguay Syria Angola 0 20 40 60 80 100 120 140 160 180 easiest most difficult ease of startinga business ranking(out of 175countries) Source:World Bank 2007a. 3.13. While not the most important constraint faced by businesses, access to financing was identified as severe or major constraint by a third of small businesses.4 Moreover, a of the relative importance of various obstacles in the observed growth of businesses finds finance to be among the few obstacles that are truly bindingon business growth. 3.14. Businesscreation and expansion can be encouraged by making it easier to adjust to changing economic conditions by hiring and firing workers. Labor policy in Jordan compares favorably with other countries (Figure 3.9 and Box 3.1). One area that still requires attention, 43Of those interviewed, 21 percent believe that corruption has remained unchanged over the past 4 years, 19percent said it haddecreased and 13 percent did not know. 44Jordan has a mixed record with regard to access to financing for small and mediumenterprises (SMEs). Although Jordan is better than many countries in terms of days to process SME loans, the minimum size of loans is quite large and considerably larger than in many other developingcountries. See Beck et al. 2006 for detailedinternationalcomparisonsof SME access to financing. 45Although businesses report many obstacles to growth, not all of the obstacles are equally constraining. Some perceived obstacles, while correlated with businessgrowth, do not directly affect business growth. A study of observablebusinessgrowth, based on the World BusinessEnvironmentSurvey of businesses in 80 countries, concluded that only obstaclesrelatedto finance, crime, and political instability directly affect the growth of rate of businesses. This study also confirms the finding of other studies that smaller businesses are more financially constrained than large businesses. See Ayyagari, Demirguc-Kunt, and Maksimovic (2006). 28 however, i s Jordan's strict employment-protection legislation, particularly the severance pay system. Employment-protection legislation like that of Jordan, i s significantly associated with low job creation-specially for women and youth-in the OECD and elsewhere (Garibaldi and Mauro 1999; World Bank 2004b; World Bank 2005a). Data for 19 developed and developing economies suggest that flexible hiring and firing rules are positively associated with higher rates of new businesses, which are often better at harnessing new technologies (World Bank 2004~). The negative impact of employment-protectionlegislationis more damaging in the most dynamic sectors-those exposed to the greatest variability in demand or in which technological change i s most rapid. Both conditions require more frequent adjustments in the workforce (Haltiwanger, Scarpetta, and Schweiger 2006). When businesses are unable to fire workers, they are less likely to hire permanent workers. To avoid burdensome regulations on firing workers, businesses may hire fewer workers than they otherwise would or may hire workers on temporary contracts. Jordan's labor law allows temporary contracts, and it i s likely that employers make heavier use of temporary contracts than they would if firing regulations were less burdensome, but data are not available on trends in use of temporary contracts. The fact that foreign workers are hired on temporary contracts may contribute to their attractiveness over Jordanian workers. Figure 3.9. Ease of hiring and firing Philippines Angola Jamaica Morocco Guatemala El Salvador 0most 1 2 3 4 5 difficult easiest index of ease of hiring and firing Source: World Economic Forum2006. 29 Box 3.1. Selected labor laws and regulations inJordan Minimum wage. There i s currently a national debate on increasing the minimum wage. The debate should take into consideration (a) that the wages of low-skilled workers (elementary occupations) have increased faster than those in other categories of workers in recent years, and (b) that minimum-wage increases generally benefit incumbent workers at the expense of new entrants because they discourage job creation. The minimumwage increased to 110JD per month in June 2006 (see table). Effective date Minimumwage (JD) October 1999 80 Januarv 2003 85 August 2005 95 June 2006 110 Hours. Normal work hours are limited to 8 hours per day, not to exceed 48 hours per week, with at least one day off per week. Overtime i s voluntary and i s paid at 125 percent of regular pay (or 150 percent o f regular pay if performed on the employee's day off). Fourteen days of paid annual leave are provided (21 days after 5 years in the same enterprise) in addition to official holidays. Employees receive 14 days of paid sick leave. Rules affecting temporary workers. Foreign workers require working permits, the duration of which may not exceed one year. Thus, foreign workers are always on temporary contracts. New employees can be placed on probation for three months, during which time they can be terminated without notice or severance Pay. Rules affecting women. Women in enterprises of 10 or more employees are entitled to maternity leave of 10 weeks at full pay and up to a year without pay. During the first year after giving birth women are entitled to one hour off from regular hours for nursing. Childcare must be provided by enterprises that have 20 or more married women and at least 10children of childcare age. Termination. One month's notice i s required before contract termination. Termination i s allowable if the employee does not fulfill obligations or for economic/technical circumstances by permission of the Ministry of Labor. If termination i s determined to be arbitrary, employers must reinstate and compensate workers. Terminated workers receive severance pay in the amount of one month's wage for each year of service. If work i s temporarily suspended for economic/technical reasons, the enterprise must pay the first 10 days' wages at full pay plus 60 days' wages at half pay. Exclusions. The labor law does not cover workers in the public sector, agriculture, domestic servants or unpaid family workers. Source: Ministry of Labor 6: Diversfyzng Exportsto GenerateHzgh-WageJobs 3.15. Jordan's recent trade performance has been encouraging. Jordan has increased its exports dramatically in recent years (Figure 3. lo), while reducing tariffs significantly and making commitments to further r e d ~ c t i o n sOccurring against the backdrop of the relatively low barriers . ~ ~ ~~ ~~ 46 Lower tariffs will cut Jordan's customs revenue, but the country has already adjusted to the cut. Since 1995, the share of import duties in total tax revenues has declined from more than a third to less than a fifth. Over the same time, the ratio of trade taxes to GDP fell from more than 7 percent to less than 4 percent. Once Jordan's additional commitments are implemented, tariff revenues are projected to decrease by a further 36 percent from 2005 levels. 30 that Jordan's exports face,47 the elimination of anti-export biases in the economy produced dramatic results: exports more than doubled in nominal terms between 2000 and 2005.48It should be noted, however, that despite the dramatic increases, Jordan slipped slightly in its relative position, because other countries performedeven better. Figure 3.10. Jordan's export performance, 2000-05 I 40 1 30 35 25 - a :2 0 0 - 8 15 - 10 -I 5-1 0 4 2000 2001 2002 2003 2004 2005 ...... exportsof goods -exports of setvices Source: CentralBank of Jordan. Note: data for 2005 are provisional. 3.16. Trade is also important for employment growth. Employment in Jordan has grown faster in sectors that trade more. Manufacturing sectors that have increased their activity in international markets have experienced above-average employment growth. Increases in export intensity per employee are positively correlated with the average annual rate of employment growth for the period 1995-2005 (Figure 3.11). (A similar relationship can be observed between export growth and wage increases.) An exception to this trend i s the textiles sector, which shrank in terms of overall employment, even as it became more export oriented. But apparel production expanded substantially both in terms of exports per employee and employment, so that the positive correlation between export intensity and employment growth continues to hold if textiles and apparel are taken as a single entity. Over the past decade a shift has occurred within the sector from relatively capital-intensive textiles production to more labor-intensive apparel manufacturing. 47A team of analysts in the World Bank's Research Department recently estimated a Market Access Overall Trade Restrictiveness Index (MA-OTRI), which provides an aggregate measure of foreign barriers. This indicator corresponds to the uniform tariff that, if imposed by all trading partners on exports of a particular country (instead of the tariffs actually applied and nontariff barriers), would leave overall exports of that country unchanged (Kee, Nicita, and Olarreaga 2005). For Jordan in 2001/02, the MA-OTRI amounted to 25.9 percent on agricultural exports and 9.3 percent on manufactured products. For all merchandise trade worldwide, the tariff equivalent of all tariff and nontariff measures was 9.9 percent. Among the 91 countries for which data are available, therefore, Jordan placed in the quartile of countries that are facing relatively low barriers to their exports. Hence, foreign barriers do not seem to be holding back Jordan's exports. 48Jordan's export performance was strong despite a somewhat overvalued real effective exchange rate. The World Bank estimates that the overvaluation has caused a loss of only 3.1 percent of exports (Nabli and Veganzones 2004). 31 Figure 3.11. Increasesinexport intensity per employee(1995-2005) and average annual rate of employment growth (1995-2004) lo i 4 4 apparel --ae 8 - 4 4 4 .C z-6 6 - m 4 beverages Iood and gE 4 - trend textiles 4 -10 0 to 20 30 40 50 60 70 growth in exports per employee (`000USD) Source:World Bank staffbasedon UNComtradedatabase and DOS EmploymentinEnterprises. 3.17. Despite good overall performance, Jordan's export basket remains narrowly based. Diversification of exports is important to countries like Jordan. Jordan's exports are concentrated in three sectors. In 2004, chemicals (fertilizers and pharmaceuticals), clothing, and food products accounted for 30.5 percent, 26.1 percent, and 13.2 percent of total exports. There are three major reasonsto be concerned about that concentration, particularly as it relates tojob creation. First, opportunities in apparel and food are limited. Both are highly contested in world markets and generate narrow profits for producers. World demand i s growing more slowly for these products than the average for all products. Second, Jordan's concentration in clothing (discussed above) has led to only moderate increasesin employment of Jordanians and offers limited promise inthe future. Third, Jordan's food industry depends on climate-sensitive agricultural production. Because Jordan is among the 10 most water-scarce countries in the world, agriculture i s unsustainable as an export base and has been able to continue only with large water subsidies.49Moreover, agriculture i s strongly dominated by foreign workers, and the sector cannot be argued to be of strategic importance as a source of stablejobs. 49Jordan runs a water deficit of 500 million cubic meters per year, with a disproportionate share of available water (63 percent) going to agriculture. Despite its water scarcity, Jordan maintains very low water prices-about US$0.02 per cubic meter compared to US$0.24 for Israel, US$0.15 for Cyprus, US$O.lO for Syria, and US$O.10 for Morocco. Jordan's very low water prices encourage unsustainable agricultural practices, including inefficient irrigation and otherwise unprofitable agricultural exports. See World Bank (2006d) and FAOSTAT 2004. One of Jordan's few areas of possible comparative advantage in agriculture is exporting fresh produce to Europe during the winter months. But the European's continuing duties on fresh produce from Jordan limits that potential. Revealed comparative advantage (RCA) analysis in Nassif and Walkenhorst (forthcoming) indicates that one of Jordan's advantages is agriculture. Conclusions should not be drawn from this finding, however, because the RCA methodology does not distinguish between comparative advantages based on sustainable resources vs. comparative advantage based on unsustainable policies. In the case of Jordan's RCA in agriculture, the advantage i s likely to be due exclusively to subsidized water and other policy factors. 32 3.18. Of Jordan's three major exports, only the chemical industry i s economically sustainable and can provide a sustainable source of jobs for Jordanians. Chemicals i s Jordan's only dynamic export industry, that is, the only industry that has grown faster than the world a~erage.~' In two other dynamic industries, transportation and machinery, Jordan seems to be "de-specializing,'' that is, exporting less than the world average.51 3.19. Jordan's recent trade success via QIZs may not be sustainable. Qualified Industrial Zones (QIZs) have been directly responsible for a large share of the growth in exports andjobs in Jordan since 2000 (Vox 3.2). The share of QIZ exports in total exports grew from 0.2 percent in 1999 to almost 28 percent in 2005 (Figure 3.12). During this same period, the QIZs generated 54,000 new jobs, about one-third of which have been taken by Jordanian workers. Among the Jordanian workers who hold QIZ jobs, nearly two-thirds are women, many of whom are holding their first job (Box 3.3). While these jobs provide significant benefits to the workers' families, the impact on unemployment i s negligible, as these women were unlikely to be in the labor force before being hired. Most QIZ workers are foreigners brought into Jordan under temporary labor contracts. Their number swelled after 2004, when the Government decreased work permit fees for foreign workers employed at the QIZs from the usual JD 300 per year to JD 150 (Figure 3.13).52 The QIZs provide few, if any, linkages to other parts of the economy and, therefore, have had little indirect job-creation effect.53 Box 3.2. Jordan's QualifiedIndustrialZones (QIZs) In 1997, Jordan and the United States signed the QIZ Agreement, which extended the market access privileges of the US.-Israel Free Trade Agreement to approved enclaves in Jordan that produce goods in collaboration with businesses in Israel. To be eligible for free access to the US. market, products from the QIZs have to have at least 8 percent Israeli content, 11.7 percent Jordanian content, and a total content from Israel, Jordan, West Bank and Gaza, and the United States of at least 35 percent. The QIZs presented an important opportunity for textile and clothing manufacturers from other countries to circumvent their quota restrictions. Apparel manufacturing requires little capital investment and permits a great deal of geographic mobility, with factories picking up and moving to take advantage of opportunities. The production processes in the QIZs are primarily cutting, sewing, trimming, ironing, and packaging-all low-investment and labor-intensive processes. 50 In general, chemicals, whether fertilizers or pharmaceuticals tend to be capital intensive and provide few jobs. We do not have sufficient information to determine the job-creation potential of chemicals in Jordan. 51 The implication of de-specializing in these dynamic product categories i s that, by reducing its exports, Jordan is missing out on potentially expanding export and job creation opportunities. Indeed, as exports in these categories contracts, jobs are likely to be lost. 52 The average annual wages at QIZs are JD 1,770, slightly higher than average wages in the textile sector (JD 1,570) but considerably lower than wages in manufacturing (JD 3,070), industrial estates (JD3,020), and industry (JD 2,986). 53 The total output of the QIZs is exported: therefore there are no forward linkages. Even in the case of transportation there i s little benefit as the majority of the QIZ factories ship their products to the United States via Israel instead of Jordan.53There are few backward linkages as there i s little textile or accessory production in Jordan. And, the fact that all local production must be carried out within the QIZ (limiting interaction with other companies) and the simplicity of the production processes limits possibilities for others to learn from the production in the QIZs. Moreover, the direct link with the U.S. market is limited as most distributing and marketing channels are controlled by the Israeli partners who have a better knowledge of U.S. markets. 33 Figure 3.12. Changesinapparel and other goods exports, 1998-2006 3,500 1 3,000 j 2,500 -I I =E 5 2,000 1 I 1 1998 1999 2000 2001 2002 2003 2004 2005 2006 other goods Qclothes Source: Central Bank of Jordan 2007. Figure 3.13. Employment in QIZs, 2001-06 60,000 1 50,000 ' 10,000 - O i 2001 2002 2003 2004 2005 June-06 &I3Jordanian C3foreign Source. Ministry of Trade and Industry, QIZ unit andMinistry of Labor. 34 Box 3.3. Female employment in QIZs The establishment of QIZs and the subsequent development of the clothing industry have provided employment opportunities for many women in Jordan, particularly those with no or little formal education. More than 90 percent of all Jordanian workers in QIZs are female. The approximately 15,000 Jordanian women who were employed in QIZs during 2005 represented about 15 percent of the total female labor force. Most of these women enter their jobs with no former work experience and are often their family's main source of cash income. Female Jordanian workers in QIZs are usually single, very young, and generally from a poor, rural background. Sometimes, they are offered the opportunity to participate in company training programs to help them enhance their knowledge and abilities. However, employment in a QIZ provides only limited career prospects and does generally not develop skills that can be easily transferred to jobs outside the QIZ. Moreover, working conditions have been a concern. Inearly 2006, a report published by the US. National Labor Committee alleged abuse and exploitation of workers in QIZs. These allegations have reinforced beliefs among a segment of the population that believes QIZ employment to be unsuitable for women. Therefore many women are discouraged by their family to take jobs in QIZs. Businesses, as well the Jordanian government, have made efforts to counter these beliefs. Road shows, open-door policies, image campaigns, increased labor inspections, and better transportation have been undertaken to encourage female labor participation in QIZs and to convince the population that abusive practices are not common and certainly not the norm. 3.20. In contrast to the export and job growth related to the QIZs, higher-wage exports including higher-technology products and services have been slow to take off. Jordan's performance in medium and high-technology products has been mixed. In addition to generating high-skilljobs, products containing a medium to high degree of technology are good choices for export promotion because they are among the most dynamic products in world trade.54Although high-tech products have been a relatively high share of Jordan's exports for over 25 years, the country has not been able to expand that share (Figure 3.14). The major shift over the past decade has been toward labor-intensive, low-technology products and a corresponding shift away from resource-based products. In contrast, many other countries have successfully expanded in medium and hightechnology exports duringthe same period (Figure 3.15). Although the shift out of resource-based products has been a positive development, Jordan's employment and wage picture might have been rosier today if Jordanian businesses had moved more forcefully into higher-technology products. 54 Various arguments in the literature underscore the importance of technology intensity and (export) growth. Products with higher technology content are said to represent the most dynamic products in world trade. Technology-intensive industries are less vulnerable to entry by competitors than industries for which scale, skill, and technology requirements are low. Export markets for low-technology products are saturated-over time countries can only sustain export growth by taking shares from other low technology exporters. Technology-intensive activities have a higher learning potential, and eventually lead to faster growth in quality-enhancing capabilities. Technology-intensive export sectors are likely to have larger spillovers to other activities and to the national technology and innovation system. See World Bank (2007b). 35 Figure 3.14. Evolution of the technological structure of exports, selectedcountries and regions, 1980-2004 100 -I 90 j I 80 -I 70 1 8 40- 1I 30 1 2o lo 0- 1990 2004 Source: staff calculation based onUNCTAD 2006 Figure 3.15. Growth in medium and high technology exports between 1990 and 2004 Jordan South East Asia Morocco 10New Membersof EU Lebanon Latin America and Carribbean 7-77 0 50 100 150 200 250 300 % growth from 1990to 2004 Source. Stdff calculation based on UNCTAD 2006. 3.21. Jordanian policy makers have expressed a desire for Jordan to become a knowledge economy, focusing on exports of services and high-tech products that make use of skilled labor, such as the many voluntarily unemployed Jordanian workers discussed in the previous section. Jordan's economy already i s dominated by services, which account for almost 70 percent o f GDP and more than 75 percent of jobs, so service exports are likely to be area of comparative advantage. But service exports account for only 20 percent of total exports. Three subsectors considered key to Jordan's service exports-information and communication technologies (ICT), medical services, and tourism-are underperforming for lack of trained personnel and of opportunities that are sufficiently attractive to keep the best personnel from emigrating from Jordan. According to the World Bank's recent investment climate survey of enterprises (World 36 Bank 2006c), 44 percent of service enterprises believe that an inadequate supply of appropriate personnel i s a constraint on their growth. Only 23 percent of manufacturing enterprises expressed the same belief. 3.22. The Government considers ICT to be a key sector, devoting significant support and attention to its regulatory and policy requirements. Various public initiatives have been designed to address the presently low rates of connectivity, computer penetration, and private sector participation in policy formulation. The Government's proclaimed goal i s for Jordan to become a regional hub for telecommunications outsourcing in the Middle East and North Africa. The country's public spending for ICT as a share of GDP i s among the highest in the world (World Bank 2005~).Nevertheless, Jordan's ICT sector remains relatively small. Its present rate of growth (14 percent) lingers below expectations of the government. Export revenues reached US$79 million in 2004, a feeble 4 percent of total service exports. And only a small share of Jordan's 4,000 ICT graduates each year find jobs in Jordan. The country's ICT industry, with 9,000 workers at the end of 2004, employs only a small fraction of the available human resources. In addition to bridgingthe gap in supply and demand of human resources, Jordan will need to accelerate its efforts to foster domestic demand, enhance sector capabilities, strengthen protection of intellectual property rights, and improve its regime for e-commerce and e-government if it i s to become competitive inthis sector.55 3.23. A growing number of foreigners visit Jordan for medical treatment. The latest available data indicate that more than 29,000 patients, most from Yemen, Libya, and other Arab countries, chose Jordanian hospitals for their qualified medical staff, relatively advanced medical services, and advantageous prices. In 2003 the country's medical services industry generated export revenues of US$500-600 million. The most serious constraints to growth are the industry's inability to expand to meet growing demand from abroad and the lack of trained personnel. A wider array of training opportunities in health management are required to develop the industry's potential. Other problems include untapped opportunities for clustering (linking medical care with dental and plastic surgery services, for example) and weaknesses in laws and regulations related to the quality of services. 3.24. Despite the extensive inventory of modern hotels, tourism lags below its potential. Jordan's tourism industry has grown at rates below the world average and below the average for the region (except for the past three years). The World Tourism Organization (quoted in Jordan's National Tourism Strategy 2004) has found that 9 of 30 emerging tourism destinations are in the Middle East and North Africa. The region's tourism sector has shown great dynamism, with 12.2 percent annual growth in tourist arrivals between 1995 and 2002. Jordan was not among the best performers. In fact, Jordan achieved 5.8 percent annual growth, below the world average (7 percent). Only recently have tourism receipts begun to increase, in part because of Jordan's role as a logistics link in Iraq's reconstruction. Tourism receipts more doubled between 2001 and 2005 and now exceed JD 1 billion, a record. Much room for improvement remains. To sustain growth in the sector, it will be necessary to address supply constraints, such as limited air access, insufficient entertainment facilities, and a lack of concerted tourism marketing. The industry continues to rely on traditional cultural attractions, neglecting other available travel-related segments, such as businessand medical tourism (National Tourism Strategy 2004). 3.25. Jordan's export future is uncertain: QIZs are likely to produce fewer exports and jobs, but it is unclear if other sectors can step in to maintain Jordan's recent trade success. Although the QIZ agreement i s not time-bound, Jordan's free trade agreement (FTA) with the 55According to discussionswith officials in the Ministry of Informationand Communication Technologies. 37 United States will extend the market-access preferences now enjoyed by QIZ-businesses to producers outside the zones by 2012.56 At the same time, the expiration of the Multi-Fibre Agreement (MFA) will limit-but not eliminate-the attractiveness of the QIZs for foreign investors, while increasing international competition in apparel.57Even after full quota removal, apparel exporters from QIZs will retain tariff preferences over their competition from Asia.58In all, Jordan may lose 1,300 jobs and 42 percent of its textile exports (12 percent of total exports) from expiration of the MFA ( L O 2005). Another threat to Jordan's QIZs comes from similar agreements the United States has signed with other countries and from the potential tariff reductions that might be associated with the Doha Round of trade negotiation^.^^ For all these reasons, it appears likely that Jordan will experience no further increases in exports or job creationthrough the QIZs and may experience some reduction. Aware of this possibility, Jordan's policy makers have turned their attention to other sources of export growth andjob creation. D. RemuvzngDhturtiunstuspur Jub Creatiun 3.26. This much is clear: The jobs the Jordanian economy has created over the past five years have not been attractive to a large segment of the Jordanian labor force. Can Jordan's economy compete internationally in exports of products that create the kind of jobs to which Jordanians aspire? Unfortunately, public sector institutions are notoriously bad at picking winners: costly public support for certain industries often brings little benefit. Often, the most valuable contribution the public sector can make i s to remove policy-based distortions that deter the economy from creating desirable jobs. 3.27. While Jordan's policy environment has improved significantly, there remain areas where distortions may be holding back high-wage job creation. Two areas of distortion are agriculture and apparel. The incentives that are available for these sectors encourage a focus on short-term profits and allow access to a virtually unlimited supply of cheap, low-skilled (mostly foreign) labor. Three types of distortions merit attention: the tax system, labor regulations, and migration policy. Other distortions are relevant but are outside the scope of this report. For example, the agriculture sector-which provides few jobs for Jordanian workers-survives in 56 The implementation schedule for the FTA in the clothing sector is back-loaded, such that Jordanian producers of textiles and apparel that are located outside QIZs will not get free access to the U.S. market until 2012, at the very end of the transition period. Also, differences in rules of origin will remain. In particular, the FTA provides for a minimum-content requirement of 35 percent of combined value-added from Jordan and the United States, while QIZ exports must consist of at least 35 percent combined value- added from Jordan, Israel, the West Bank and Gaza, and the United States. 57The quotas imposed under the MFA were phased out according to the schedule of the WTO's Agreement on Textiles and Clothing at the beginning of 2005, leading to a surge in clothing exports from China and other Asian producers that triggered the re-imposition of temporary safeguard quotas after January 2006. Once these safeguards expire at the end of 2008, the intensity of competition in the U.S. apparel market will increase sharply. 58I t has been estimated that higher labor costs and the additional expenses of overseas procurement of raw materials put Jordan's apparel producers at a cost disadvantage of about 10 percent vis-&vis the major competitors from Asia (Saif 2006). The tariff preferences for clothing that QIZ-producers enjoy in the U.S. market exceed 10percent for many products, so that the QIZs do not seem to be in immediate danger. 59A successful conclusion of the Doha Round of multilateral trade negotiations could trigger a world-wide reduction in tariffs that would also reduce protection in the US clothing market. Moreover, the USA has signed an increasing number of preferential agreements that provide partner countries with similar market access preferences as Jordanian exporters. Recent examples include the QIZ Agreement that was concluded with Egypt in 2004, as well as the CAFTA-DR Agreement. These agreements put an increasing number of competitors into the same market access position as Jordanian QIZ producers and, hence, tend to erode the latter's profit margins. 38 large part due to heavily subsidized water. In this section, we address other aspects of the agriculture sector's favored status but leave water subsidies to be addressed by more qualified analysts. 3.28. The tax system is designed to provide strong incentives to agriculture and industry. The multi-layered tax system that confronts entrepreneurs is the result of the overlay of three sets of laws: conventional tax laws, the investment promotion law (IPL),and the so-called Industrial Estates Law. The income tax laws differentiate tax rates by sector, the most favored sector being agriculture, which is subject to no company tax at all (Table 3.2). Companies operating in mining, manufacturing, hotels, hospitals, transportation, and construction pay a 15 percent rate of income tax. All other sectors pay rates of 25 percent or more. Projects supported by the IPL in a narrow set of favored sectors (detailed in Box 3.4) enjoy three major privileges: (a) reduction in the company income tax for a period of 10 years, with the size of the reduction based on geographic location; (b) exemption from import duties; and (c) a one-stop shop that significantly reduces the time and effort requiredfor business licensing, registration, and work permits for foreign workers. The Jordan Industrial Estates Law provides benefits to activities located in public industrial estates. The benefits include a two-year exemption from income tax and a permanent exemption from property taxes, as well as subsidized land, infrastructure, and a range of business services. Combined, these three sets of laws provide persuasive incentives to agriculture, followed by industry and some categories of services, at the expense of other activities, including communications, insurance and financial services, and most other types of services. Some have suggested that a more effective incentive system would be based on business performance (in terms of job creation or innovation) rather than on one-time investments. AMIR (2004) provides a detailed discussion of the incentive effects of the tax system. Box 3.4. Sectors favored by Jordan's investment promotion law Agriculture Industry Hotels (three stars or more) Hospitals (but not clinics and other health services) Conventions and exhibition centers Leisure and recreational compounds Maritime transport and railways Pipeline transportation and distribution services for water, gas, and petroleum derivatives Other sectors that the Council of Ministers may elect to add Source:Jordan InvestmentBoard. Note: Despite appearances, the focus on hospitalsdoes not support medicaltourism. Some 70 percent of medical touristsuse clinics andmedicalcenters rather than hospitals. 39 Table 3.2. Income tax rates and investment incentives I Tax rate without incentives Sector or after Tax rates inincentive zones (percent) incentives expire (percent) A B C Industrial Estates Agriculture 0 N/A NIA N/A N/A 3.75 Zero for 2 years+ A Industry, including miningand 11.25 7.50 for B C manufacturing 15 for 10 for 10 11.25 7.50 3.75 years years years for 10 for 10 for 10 years years years 11.25 3.75 Hotels and hospitals 15 for 10 for 10 7.50 for N/A years years vears 3.75 Maritime transport and railways (MTR) 15 for Zero for 2 years NIA NIA IO +3.75for 10years Construction, electricity and gas, transportation excluding MTR 15 NIA N/A N/A N/A Leisure and recreational compound for 25 10+ N/A N/A vears 18.75 Conventions and exhibition centers for 25 10+ NIA NIA NIA Transport and distribution of water, gas, and oil 25 Defined case by case N/A Other sectors not listed elsewhere 25 N/A N/A NIA NIA Banks and financial companies, excluding Insurance 35 N/A NIA NIA N/A Source: IncomeTax Law of 1985 and IPL 1995 (andits amendments).Cited in AMIR (2004). Note: Zone C includesdistricts strongly targetedfor investment andgiven maximumbenefits,Zone B includesdistricts targetedfor investment andreceivingmoderatebenefits,Zone A includesall other districts. A complete listing of districts andzones canbe found inthe InvestmentPromotionLaw at www.iordaninvestment.com; N/A = not applicable 40 3.29. Labor regulations reinforce the tax distortions by making labor less expensive for the agriculture and the largely garment-based QIZs than for other sectors. Through January 2007, the garment industry was exempted from the increase inthe minimumwage that took effect inother sectors inJune 2006, a small but politically important concession. The wage in all sectors i s now JD 110 per month. Agriculture, the garment industry, and Q I z s also pay much lower fees to obtain work permits for foreign workers (Table 3.7). Table 3.3. Work-permit fees (JD/year) Non-JordanianArabs Other foreigners Agriculture 60 120 QlZs (1st 3 years) 150 150 QlZs (4th year and after) 175 175 Other sectors 180 300 Source: Ministry of Labor. 3.30. Other policies create unintended distortions emphasizing low-wage work. Immigration and emigration policy creates an economy more conducive to low-wage than to knowledge-based. The combined effect of immigration and emigration has profound implications on the kinds of economic activities attractive to businesses. An explicit and consistent policy should be formulated by the Government. Labor and migration policies have the practical effect of encouraging entrepreneurs to focus on economic activities that employ low- wage, low-skill workers. From the entrepreneur's point of view, Jordan offers a virtually unlimited supply of low-wage, low-skilled workers, most of them foreign, which undoubtedly affects their decisions.60Foreign workers can be less expensive to employers not only because they may accept lower wages, but also because foreign workers are on temporary contracts and therefore are not entitled to social security or severancepay.61 3.31. There i s nothing inherently wrong with admitting foreign workers. The impact of immigration depends critically on the extent to which migrants and natives are complements or substitutes for one another; that is, whether they are competing for the same jobs or operating in segmented markets. When foreign workers provide skills unavailable in the domestic labor market they may complement native workers to the benefit of all. Borjas (2003), studying the United States, finds that immigration harms the prospects and lowers the wage of competing native workers up to college level. A 10percent increase in supply reduces wages by 3 4 percent. Those with the lowest skills are hardest hit. College graduates, who are often complements to foreign workers, may actually experience a wage increase as a result of immigration.62InJordan, regulations that make foreign workers more desirable than Jordanians, combined with immigration policies that permit virtually unlimited supplies of foreign workers, make Jordan appear, from the entrepreneur's perspective, to resemble a low-skill, low-wage economy and create incentives for economic activities that may be more suited to the characteristics of the 6o Other factors, too, impact the skill mix of available labor. The mismatch between the education system and the labor market, for example, was discussed in Section I1of this report. 61 Entry of non-Arab foreign workers, who must be issued work permits before entering the country, i s controlled by the Ministry of Labor. However, entry of Arabs, who do not require visas to enter Jordan, i s not regulated. Although Arabs officially require permits to work, a large number are currently working without permits owing to the difficulties of enforcement once they are inside Jordan. For a fuller discussion of the impact of immigration on local workers, see World Bank (2006b). 41 foreign workers than to the characteristics of Jordanians. Unfortunately, analysis of the benefits and costs of the current implicit migration policy i s seriously undermined by the lack of data on this sizeable component of Jordan's labor force-estimated to represent at least one in five workers.63 3.32. Since the 1970s, many educated Jordanians have left their country for better opportunities abroad. More information on emigration from Jordan i s offered in Section IV, but two issues related to the demand for labor are worth noting here. First, remittances from emigrants have been found in recent studies to cause significant appreciation in the real exchange rate, making Jordanian exports less competitive. Second, if highly skilled workers are complements to low-skilled workers in production, the brain-drain effect of emigration can lead to an increase in unemployment. In the World Bank's investment climate study (World Bank 2006c), 44 percent of enterprises in Jordan's service sector identified an inadequately educated labor force as a major or very severe constraint to operations and growth. The problem i s not as great among manufacturingbusinesses, which make less use of educated labor: only 23 percent of manufacturing businesses identified it as a constraint. Although the short-term benefits of emigration are clear (in terms of remittances and relieving pressure on local labor markets), recent research has begun to question the relative costs and benefits of emigration. In particular, recent research (Schiff 2005) has argued that the long-term benefits of experience gained overseas are realized only if emigrants return home-a phenomenon that i s far from inevitable- and that, with emigration, the social return on investment in education i s lost. We will have more to say about emigration and remittances in the next section. The World Economic Forum identified brain drain as one of Jordan's notable competitive disadvantages (Figure 3.16). Figure 3.16. Brain drain inJordan and selectedother countries, 2006 Philippines Egypt Paraguay Jamaica Sri Lanka Jordan Morocco Angola El Salvador Guatemala 0 1 2 3 4 index of brain drain Source: World EconomicForum2006. Note: the brain drain index i s constructedusingdata from the WEF's annualExecutiveOpinion Survey. Respondents were asked to describethe extent to which their country's most talentedpeople leaveto pursueopportunitiesin other countries. A score of 1meansthe most talented peoplenormally leaveto pursueopportunitiesin other countries, while a score of 7 meansthe most talented people almost always remainin the country. 63 The EUS and census report 939,986 Jordanians employed. The Ministry of Labor reports 218,756 registered foreign workers. Even accepting this clear undercounting (see footnote 44), foreign workers make up 19 percent of all workers. If we use a higher figure, such as the 300,000 commonly referred to, foreign workers would account for 24 percent of the labor force. 42 CHAPTER 4. LABOR SUPPLY: INVESTING INTHE EMPLOYABILITY OF THE WORKFORCE AND REALIGNINGEXPECTATIONS 4.1. In the previous section, we examined the types of jobs the Jordanian economy has created over the past five years and the policies that have given Jordan its comparative advantage in those areas. Whatever Jordan's current comparative advantage, we argued that the government can influence its future by better aligning its economic policies with its aspirations. 4.2. We turn now to issues related to labor supply-Jordan's abundant resource. In addition to enhancing crucial aspects of the investment climate, the government can take actions on the labor supply side to reduce unemployment. These actions include investing in the employability and realigning the expectations of the workforce. The main messages of this section are: 0 Jordan's future success as a high-wage knowledge economy will require excellence inquality as well as quantity of education. 0 Improving links to employment can make technical and vocational education and training (TVET) a tool for reducing unemployment among the 40 percent (the involuntarily unemployed). 0 Jordan can take better advantage of past investments in education by increasing women's ability to balance family and work. A centrally financed system of maternity leave could raise employment among women and increase the educational level of the workforce. 0 Unemployment cannot be reduced in the short and medium term without a realignment of expectations. Expectations of civil service jobs and opportunities for emigration prevent the unemployed from accepting available jobs. New efforts are neededto control the growth of the civil service and-just as importantly-to dispel illusions that civil service jobs will be widely available. A large share of the unemployed can afford to remain unemployed, supported by their families (with remittance income from other family members playing a crucial role). In addition to their positive impacts, remittances increase unemployment and reduce Jordan's competitiveness. Careful consideration of both positive and negative impacts of emigration i s needed. Although existing social protection programs do not appear to create large disincentives for work, care must be taken to ensure reforms do not introduce such disincentives. A. Investzhgzn Empluyabz22y 4.3. In Section 111, we argued that unemployment is shaped by the type of investment as well as its quantity. Here we argue that the type of investment in humancapital matters as well. In the long term, the composition of jobs in Jordan will be determined by the level and quality of I education of the labor force, as well as the presence or absence of distortions in the business environment. The educational system can reduce unemployment among the voluntarily unemployed if it imparts skills commensurate with the wages Jordanians aspire to. In the short term the existing composition ofjobs must be taken as given. Short technical training can reduce unemployment among the involuntarily unemployed if the training responds to the needs of businesses-including employability skills. This can be done by making fuller use of previous investments in education and training. To take fuller advantage of women's education and early work experience, Jordan's policy on maternity leave should be reformed. 43 4.4. Jordan's future success as high-wage, knowledge economy will require excellencein quality as well as quantity of education. Jordan has made enormous progress over the past generation in expanding enrollment rates and other measures of the quantity of education, but quality remains a challenge. In its assessment of the Jordanian educational system, the Jordan Human Development Report 2000 confirmed that the country's primary school students were faring well compared to international counterparts in knowledge but not in problem-solving skills. The same weaknesses were apparent in tests of secondary students that included analytical questions. The report also identified lack of career counseling and poor linkages with the labor market as serious deficiencies. It emphasized the need for an appropriate balance of academic and technical-vocational subjects, as well as mechanisms to link students to the labor market and so provides a smooth transition from school to work. The importance of the quality of education will be even more important tomorrow than it i s today. Rapid changes in the world economy will imply the need for constant retraining and skill upgrading. Workers with a higher quality of education are more able to learn new skills and adapt new technology (de Ferranti and others 2002). Although more educated individuals earn higher wages than less educated individuals (which provides an incentive to invest in education), high rates of unemployment among the highly educated individuals reduce the benefits of (and incentive to invest in) education. Recent researchprovides`strongevidence that cognitive skills of the population-rather than mere school attainment-are powerfully related to individual earnings and economic growth. Moreover, this research finds that teacher quality i s more important than class size or spending on schools, and that quality matters as much at low skill levels as it does at high skill levels (Hanushek and Woessmann 2006). 4.5. Continuing the efforts already started will be necessary, with particular emphasis on career counseling and links to work. The educational reforms of the early 2000s were designed to remedy some of the deficiencies of Jordan's educational system. The Education for the Knowledge Economy program, for example, emphasized access to computers and development of analytical skills. It has made progress since 2003 in development of a new curriculum, emphasizing problem-solving skills and teamwork (rather than rote learning as in the previous curriculum), trained 13,000 teachers in use of the new curriculum, provided computer labs in over 1,000 schools and expanded enrollment in kindergartens by 50 percent. But additional efforts are needed to enhance the linkages between the education system and the labor force. 4.6. Having recognized the importance of job market skills as well as technical skills, the Government included "enhancing pioneer, risk-taking, and willingness skills" in the plan of action presented at the We Are All Jordan Forum. The plan recognizes that secondary and university students must develop realistic expectations of labor market demand for various specializations and skills and must become more employable while still in school. 4.7. Career counseling based on accurate information and programs that provide early work experiences (through internships, for example) may be particularly useful additions to the ongoing efforts of the ministries of education and higher education. International experience shows that school-to-work programs involving a greater integration of school and work though part-time work, internships, workplace-based training, and the promotion of youth entrepreneurship among graduates can be successful in reducing youth unemployment. Such programs have been developed in several countries including Australia, the Czech Republic, Germany, Egypt, South Africa, the United Kingdom, and the United States (Kolev and Saget 2005). 44 4.8. Jordan's INJAZ program i s an important step in the right direction and will be increasingly useful as it i s expanded. The program aims to "enhance the capacity and competitive advantage of Jordanian youth to enable them to enter the job market as qualified employees and business owners." It does this by increasing the involvement of the Jordanian private sector inthe education system and linking youth to the job market. INJAZ provides courses in the public school system for students grade 8-university. Courses include topics including career options, communication, teamwork, customer service, business ethics, preparing for job interviews, and personal financial habits. INJAZ i s a nonprofit organization and provides courses for more than 42,000 each school year.64 4.9. In the short and medium term, the existing composition of jobs must be taken as given. Improving the links to employment can make TVET a tool for reducing unemployment, although this will not be a cure due to the fact that the advantage of foreign workers does not lie in greater training or experience. Rapid changes in the global economy have created a need for frequent retraining and thus for effective vocational education and training systems. Enterprises are not likely to train the existing workforce if they are already faced with competitive cost , pressures and may not benefit from investment in training as mobile workers leave for other enterprises. Under such circumstances businesses will seek access to a global pool of workers with the desired skills to substitute for the existing workforce. Such behavior places even greater pressure on the public TVET system to provide industry-specific skills previously provided by enterprise-level training (World Bank 2004~).A recent European Union report on Jordan's TVET system (European Training Foundation, 2006) concluded that its immediate priorities should be stronger labor market linkages (including the continued development of labor market information), deeper employer involvement upstream in the training process, and greater flexibility and responsiveness to market changes. The report argued that labor market information needs to be expanded to include data on labor demand and training supply, and that the links between policy making and labor market information need to be strengthened. Jordan's Vocational Training Corporation has made progress in involving employers in apprenticeship programs, but the TVET system as a whole remains highly supply driven, overly specialized, and focused on technical skills rather than employability issues, such as teamwork (ETF2006). 4.10. A reform effort supported by the King and the current Government is beginning to reassess TVET-and skill development more generally-to refocus it on employers rather than the public sector. The reform effort will require strong support at the highest levels to ensure adherence to its announced goals. Special attention must be paid to the collection, analysis, and broad di~semination~~of labor market information, and to assessing needs, designing curricula, and arranging internships. Public employment services in many countries contract for training services from private sector providers linkingpayments to output indicators, such as the number ofjob placements, rather than input indicators, such as the number of workers trained (World Bank 2003). The substantial potential of workplace-based training has not been tapped in Jordan. Care must be taken, however, to ensure that workplace-based training does not deteriorate into a simple wage subsidy program. Many such programs have only temporary effects on unemployment.66Training tends to provide better results for women than for men, and 64For more information on INJAZ ,see www.Injaz.org.jo. 65 Students lack information on the labor market (e.g., wages, working conditions and the likelihood of finding employment) to help them decide among training programs. Trainers and program developers lack information to help them design appropriatecurriculum. 66Studies of wage subsidy programs have found that after the subsidy ends participants are less likely to remain employed and more likely to earn less than control groups. One notable success was the US. Job Training Partnership Act program, where subsidies were combined with carefully monitored on-the-job 45 better results for youth than for older workers. That i s good news for Jordan, given Jordan's high unemployment among women and 4.11. Jordan can take better advantage of past investments in education by increasing women's ability to continue working. Jordan's women are an insufficiently utilized resource. Jordanian women tend to have high levels of education and are much less likely to emigrate than their male counterparts, implying that a large supply of well-educated women i s available for employment. Jordanian women often become employed but leave the labor force permanently when they marry or have their first child (Figure 4.1). Many others never find work for various reasons-among them educational specializations that do not match the needs of employers, incompatibility of work with family obligations, and discrimination. Indeed, small businesses, which we have argued are key to job creation, have a strong preference against hiring women (Figure 4.2). Figure 4.1. Labor force participation before and after marriage never married 55 9 0 20 40 60 80 % of group male 0female Source: EUS 2005. training activities. Employment subsidies arc usually for disadvantaged groups, although for some countries (Belgium, France,the Netherlands)they apply to all low-paid work. Most wage subsidy programs in the OECD have little positive impact and suffer from substantialdead weight (workers would have been employed without the subsidy) and substitution (subsidized workers displace nonsubsidizedworkers). See World Bank (2006b). 67A successful example of training for youth is the "Jovenes" programs of Argentina, Chile, Peru, and Uruguay, which target disadvantagedyouth. These programs, which combine training and work experience with other services, include the private sector and are financed by tripartite levy-grant schemes or by the government. They have substantial and positive impacts on employment and earnings, but often they are small in scale and may not be cost-effective. Their participants may displace other workers. See World Bank (2006b). 46 Figure 4.2. Preference for hiring men or women preferencefor hiring women no preference preferencefor hiring men 0 20 40 60 80 % of respondents Source: World Bank InvestmentClimate Survey 2006c. 4.12. A centrally financed system of maternity leave could raise employment among women and increase the overall educational level of the workforce.68Jordan already mandates generous maternity benefits. The problem lies with how the benefits are financed. Employers presently pay maternity leave benefits directly to their female employees, making women more expensive to hire, leading employers to avoid hiring them, or recouping the cost of maternity benefits by offering women lower wages. The result is a vicious cycle in which employers prefer not to hire women (Figure 4.2), while women feel unappreciated and thus may lack commitment to the labor force. The vicious cycle can be broken by centralizing the financing of maternity leave benefits. Insuch systems a central body (often the social security agency) collects payments from enterprises based on the number of employees and then pays wages during maternity leave either directly to women or indirectly through their employer. Although the cost to employers remains the same as under noncentralized systems (there i s no cost to the public budget), centralized financing eliminates the direct link between the benefit and the individual employee. In addition to increasing women's productivity, by making greater use of the education and work experience they gain before having children, centralized maternity leave financing can increase the productivity of the labor force as a whole by reducing the inefficiencies caused by dis~rimination.~~Recently the Social Security Corporation (SSC) has proposed a plan for centralizing maternity leave benefits. Under the proposal, the SSC would collect contributions from private firms (0.5 percent of the wage bill) and from workers (0.25 percent of their wage) and then pay beneficiaries during maternity leave. If enacted, this proposal could take an important step toward raising employment among women and increasing the overall educational level of the workforce. 68 In Jordan, female labor force participation increases sharply with her education. (See Section I.) As increasing numbersof women enter the labor force, the most educated among them are likely to enter first. And, because women and men have nearly equal levels of education, an increasing female participation i s likely to result in an increase in the overall education of the workforce. 69 Discrimination i s generally economically inefficient because it prioritizes gender of the worker over productive capabilities. For more on this topic, see World Bank (2001) and World Bank (2004b). 47 B. ReakgnzngExpectations 4.13. Unemployment cannot be reduced in the short and medium term without a realignment of expectations. Investments in the productivity of the labor force take time to become effective, and the results are uncertain. Whatever the long term future of the Jordanian economy, reducing unemployment in the short and medium term will require realigning the expectations of the workforce. The education system has an important role to play, as we discussed above. So too does the Civil Service Administration, emigration policy, social protection programs and access to labor market information. 4.14. Jordanians have high expectations for increases in living standards. Dramatic increases in living standards during the 1970s and 1980s created expectations of continuous growth. And, recent high growth has reinforced that belief, with real disposable income finally matching previous levels.70(Figure 4.3). High expectations, on the other hand, weathered the crisis, in part thanks to spotty information on the prevailing wages and working conditions. The most visible information many Jordanians have about the labor market i s the experiences of friends and relatives. Many focus on opportunities to work outside Jordan or on civil service jobs, often obtained and maintained through influence or pull (wasta) rather than qualifications and hard work. Recent efforts to eliminate wasta and improve the availability of labor market information could improve incentives and make expectations more realistic. Figure 4.3. Per capita real gross national disposableincome, 1976-2005 1600 1 1400 1 88FL 100011jj I 1200 j 4 800 8 400 0 4 1975 1980 1985 1990 1995 2000 2005 Source: World Bank data. 4.15. Expectations of civil service jobs and opportunities for emigration prevent the unemployed from accepting available jobs. Many of the unemployed harbor high expectations for comfortable, secure jobs overseas or as with Government. Although the pace of hiring has slowed, Jordan's civil service continues to expand. New hires crept up from about 7,000 in 2001 to nearly 9,000 in 2004 (Figure4.4). Meanwhile, far greater numbers of people filed applications for employment with the Civil Service Bureau, implying that the hope, if not the expectation, of obtaining a civil service job remains very high. About 20 percent of all unemployed individuals, 'O The currency crises and debt-service default of the late 1980s brought the 1980s boom to an abrupt halt. Since then income had been recovering very gradually. Only recently, has the economic boom brought levels back up to those experienced before. 48 and 45 percent of unemployed women, apply to the civil service bureau. There is a clear mismatch between the qualifications of the average applicant and those of appointees (Figure 4.5), indicating a lack of realistic expectations on the part of applicants. Those with intermediate diplomas, for example, account for 33 percent of applicants but only 25 percent of those hired. Women account for 68 percent of applicants, butjust 50 percent of those hired. 4.16. The government should avoid creating unreasonable expectations. By continuing to accept applications that are not associated with specific job vacancies, the Civil Service Bureau perpetuates the unrealistic expectations of applicants. New efforts are needed to control the growth of the civil service and-just as importantly-to dispel illusions that civil service jobs will be available for any but the most qualified candidates. 4.17. How do the unemployed afford to remain without wage income while waiting for civil service and emigration opportunities to open up? A large share of the unemployed are supported by their families (with remittance income from other family members playing a crucial role). Unlike in many countries, social protection policies do not currently act as a disincentive to work. However, as Jordan enhances its social protection policies, there i s a strong risk of introducing disincentives. Figure 4.4. Civil service applications and appointments, 2001-04 180,0001 168,100 160,000 - 2001 2002 2003 2004 applicants W appointments Source:NCHRD (no date). 49 Figure 4.5. Qualifications of civil service applicants demonstrates inaccurate information more than bachelor'sdegree ~ university 73 less than university !--- I _____- v - 0 ---~ 42 20 40 60 80 Yo of group applicants Elappointments Source: NCHRD (no date) 4.18. More than three infour unemployed Jordanians (76 percent) live with their parents and only one in four belongs to the lowest expenditure quintile (Figure 4.6), and significant numbers of unemployedindividuals belongto Jordan's wealthiest families (Figure 4.7). Figure 4.6. Relationship of unemployed persons to head of household, 2004 sibling,2%; 11 child, 76% Source: NCHRD (no date). 50 Figure 4.7. Expenditure quintile of the families of the unemployed quintile5 (richest) quintile 1 Source: DOS HIES, 2002-03. 4.19. International evidence shows that emigration and remittances have both benefits and costs. The benefits of emigration and remittances are well known. Remittances increase (and smooth) consumption in recipient households and are a stable source of foreign exchange. For a country with high unemployment, emigration reduces the size of the labor force, possibly reducing unemployment rates. And many emigrants eventually return home with savings and valuable experience. In recent years, however, economists have begun exploring the negative consequences of remittances. Most thoroughly studied to date i s the potential for remittances to push up the country's real exchange rate, widen the current-account deficit, weaken monetary control, contribute to inflationary pressures, and result in sectoral misallocation of investment. Evidence from Latin America and Eastern Europe indicates that remittances create disincentives to work, cause appreciation of the real exchange rate, and often fail to have much of an impact on poverty.71In Jamaica, remittance income explains a large part of that country's simultaneously "A studyof 10LatinAmericancountries (WorldBank2006a) foundthatremittancesreducedlaborforce participation of recipient households and the number of hours worked by members of recipient households in 9 out of the 10 countries. The study controlled for a variety of factors, including education, the presence of young children, home ownership, and place of residence. In Mexico, for example, 90 percent of individuals in nonrecipient households participated in the labor force, compared with only 60 percent in recipient households. Separate studies o f Jamaica found that receipt o f remittances significantly raised the wage needed to induce an unemployed person to accept a job offer (Le., their reservation wage) and, in a general equilibrium model, led to an appreciation o f the real exchange rate (Kim 2007; Bussolo and Medvedev 2006). Macroeconomic evidence also points to real exchange rate appreciation as a result of remittances in seven of the eight countries examined in the same World Bank study. These countries experienced appreciation in their real exchange rates (increases in the equilibrium real exchange rates and overvaluation), falling exports, and rising imports. The study also found that the poverty-reducing effects of remittances were far from universal. In cases where most emigrants were poor, remittances reduced poverty, but not in cases where most emigrants were from wealthier families, as i s the case in Jordan. The studies also found that remittances lowered the household savings rate in wealthy recipient households, while increasing it in poor recipient households. Remittances had a mixed impact on entrepreneurship. More generally, recent empirical studies (Chami, Fullencamp, and Jahjah 2005; IMF2005) relying on longitudinal data find a negative or insignificant effect of remittances on economic growth. 51 high labor demand and high unemployment-a phenomenon Jamaica shares with Jordan (Kim 2007). 4.20. Evidence for Jordan shows that remittances reduce labor supply, cause the real exchange rate to appreciate, and are likely to have minimal effects on poverty. Remittance income - as with any unearned income - is expected to reduce work efforts of the recipients. Probit regressions of the determinants of labor force participation and unemployment confirm that remittance income significantly decreases labor force participation and significantly increases unemployment. A recent study by the IMF (IMF 2006) finds that remittances have had a significant impact on Jordan's real exchange rate.72This effect i s likely to be partially offset by the inflow of inexpensive labor, which tends to lower the price of nontraded services. Zaqqa 2006 found that remittances to Jordan had a positive impact on construction activity and imports, but not on industry. 4.21. InJordan, remittanceincomeaccruesprimarily to middle- and upper-classfamilies. Jordanians who emigrate tend to be well educated and from prosperous families.73As a result, remittances are greater for wealthier households than for poor ones, not only in their absolute value but also as a percentage of household income. Infact, the wealthiest quintile gets 14percent of its income from remittances (Figure 4.8). There may be cause for concern that the entire population must bear the negative consequences of remittances, whereas primarily the wealthy receive their benefits. Figure 4.8. Remittances accrue predominantly to wealthy households quintile 1 (poorest) 3 i quintile 2 4 quintile 3 5 qJ nlile 4 7 i quintile 5 (richest) 14 I 0 2 4 6 8 10 12 14 16 % of income '*Toput the magnitude of the impact in perspective, the IMFreport shows that grants have had twice as much impact on exchange rates as remittances. In recent years, the impact of remittances i s likely to have decreasedbecauseremittances have decreased as a share of GDP. 13According to Adams (2006), more than 55 percent of Jordanian emigrants have 13 or more years of schooling; only 16 percent have fewer than eight years. Zaqqa (2006) estimates that 44 percent have postsecondary education; 27 percent have only primary education. 52 Box 4.1. EmigrationfromJordan Although reliable statistics on emigration from Jordan are scarce, a recent estimate suggests that 450,000 have left the country, with the majority sending homeremittances (Azzam2006). Although estimates vary, there is consensus that emigrants tend to be among the most educated Jordanians. A survey of emigrants found that 65 percent (45,200 individuals) of those residingabroad planned to return to Jordan. Of these, 61 percent (about 27,600 individuals) planned to return prior to retirement (Zaqqa2006). Estimates of Jordaniansworking abroad, 1975-2006 1975 1980 1985 1995 2001 2003 2006 Arab countries 264,700 261,500 276,000 200,000 215,200 265,500 - Other countries 39,500 43,900 52,000 21,700 26,100 28,500 - Total 304,200 305,400 328,000 221,700 241,300 294,000 450,000 Sources: Zaqqa (2006). Datafor 2006 are fromAzzam inJordan Times, December 11, 2006. -Datanotavailable. 4.22. The policy options for addressing the negative consequences of remittances are limited. The best of the lot, but no easy task, is to make it attractive for would-be emigrants to remain at home. Several piecemeal options have been advanced as well. One i s to adjust fiscal policy to counteract real exchange rate appreciation, but the adjustment needed to stabilize the real exchange rate may be quite large and therefore infeasible for reasons of political economy.74 Sterilization of remittance flows i s equally unrealistic, for two reasons. First, the magnitude of sterilization would make their quasi-fiscal costs untenable. Second, sterilization could push up domestic interest rates, which would then attract other types of inflows in pursuit of the high yields. Those inflows would put more pressure on the exchange rate. A third option, taxing remittances, would have the effect of driving remittance flows into informal channels. A fourth option is to shift from payroll taxes to value-added or sales taxes. Reducing payroll taxes would increase demand for labor while maintaining the government balance and sterilizing most of the negative labor-supply effect of risingremittances (Bussolo and Medvedev 2006). 4.23. Jordan should study the effects of emigration and remittances and ensure that the findings are reflected in policy making. In few countries do remittances make up as large a share of GDP as inJordan,75and in few countries i s such a large share of the educated population working outside the country. Given the economic and social magnitude of emigration and remittances, therefore, Jordan should begin to collect data on emigrants, returned emigrants, and remittances. Such data are needed to analyze the positive and negative consequences of both phenomena. 4.24. Existing social protection programs do not appear to create major disincentives to work. 0 National Aid Fund (NAF).Some have suggestedthat NAF, social security, and severance pay policies may create disincentives to work. Although these programs have problems, including some related to incentives, we did not find them to be keeping the unemployed from accepting existing jobs. The categories of individuals eligible for NAF tend to be outside the labor force. Thus there are few unemployed beneficiaries, even among 74Estimates from World Bank (2006e) indicate that increases in the ratio of government consumption to GDP are associatedwith appreciationsinthe realexchangerate, but at amuchlower rate than remittances. 75Jordanranks sixth inthe world inremittancesas a percentageof GDP (WorldBank 2006b). 53 dependent^.^^ Probit regressions of the determinants of labor-force participation and unemployment find that receiving NAF benefits does not have a significant effect on labor-force participation and does not raise the likelihood of unemployment. Details of the regressions are provided inthe Appendix. Pensions. Many consider the social security pension system to be unsustainably generous in terms of its overall benefit level and early retirement provisions that allow beneficiaries to retire at age 45 with only minor reductions in their monthly pension. This policy clearly creates disincentives for work; a sharp decline in employment occurs at age 45. The pension system does not, however, have a large impact on unemployment: the evidence shows that unemployment rates are very low for those over 40 and increase only slightly at age 45.77 Income supportfor the unemployed. Workers who are laid off are entitled to a lump-sum severance payment.78Although data are not available for Jordan, severance pay policies are considered unlikely to create significant disincentives to accepting employment because the total received in severance pay i s unrelated to the duration of ~nemployment.~~Severance payments do, however, reduce job creation and can encourage quitting if, as inJordan, benefits are available for voluntary resignation (World Bank 2004b). See also section 2 regarding impact of severance pay. 4.25. Although existing social protection policies do not appear to create large disincentives, care must be taken to ensure that reforms of existing programs preserve or amplify incentives to work. Reforms of the social protection system are currently under discussion by the Government. New policies must not create disincentives that could exacerbate already highunemployment. 4.26. The NAF is a prime candidate for reform. At present the NAF i s badly targeted: many households in poverty do not receive benefits, while many beneficiaries have income well above the poverty line. Now omitted from the NAF altogether are households whose heads are working or could be working. The Ministry of Social Affairs i s considering alternative eligibility criteria, including proxy means testing, in an attempt to reach the working poor and able bodied. Once such households become eligible, it will be necessaryto find ways to avoid creating disincentives '15The presence of unemployed individuals in the household negatively affects the benefit amount of even the primary beneficiary. 77 The prevalence of early retirement programs, popular in the 1970s, has declined in the 1990s as the programs proved very costly without freeing upjobs for younger workers, as expected. If younger workers are complements and not substitutes for older workers, early retirement programs may even have a negative effect on the employment of younger workers. In addition, the additional financial burden of supporting pensions of early retirees may contribute to an increase in social security contribution rates, thereby raising the cost of labor and thus contributing, via a reduction in demand for labor, to a higher rate of unemployment. 78 Workers who participate in the social security system receive the lump sum through the Social Security Corporation, while workers who do not participate in the system but who work for enterprises with five or more employees, receive the lump sum directly from their employer. Although the rules vary slightly between the two systems, the basic principle is the same. Workers in enterprises with fewer than five employees are not covered. 79 Although severance pay does not create a moral hazard by curtailing or delaying the job search, it does affect incentives to enter unemployment and hence creates another moral hazard. Severance pay can create incentives for workers to leave employment if workers who quit are eligible for payments. Jordan's labor law does not require severance pay if the employee quits, but it is unclear the extent to which employees and employers are aware of the specifics of the law. 54 for work. Mechanisms that have been found to work well in other countries include linkingsafety net benefits to other labor market programs, such as employment services or training programs. 4.27. The Government is also considering the creation of an unemployment insurance program that could substitute for the existing severance pay policy (for workers covered by social security). Severance pay systems reduce layoffs (by making them costly), but they also discourage hiring, as employers hesitate to make permanent commitments.80Moreover, because severance pay systems tend to reduce employee turnover, they reduce opportunities for new, younger workers. International evidence shows that severance pay systems tend to decrease aggregate employment and to increase youth unemployment (Vodopivec and Raju 2002). By replacing severance pay, unemployment insurance has the potential to reduce aggregate unemployment inJordan. 4.28. Although unemployment insurance is preferable to severance pay in overall economic efficiency, it can lower the incentive of unemployed individualsto search for and .accept jobs. There i s considerable evidence that unemployment insurance programs increase unemployment and that the effects are stronger the longer the period of eligibility and the greater the benefit (Vodopivec and Raju 2002).*' The disincentives can be minimized by careful links to programs that permit monitoring of beneficiaries' job search and their acceptance of job offers (Box 4.2). International evidence demonstrates that monitoring moves people out of unemployment insurance and into employment (OECD 2000). The necessary links, however, require significant administrative capacity. Box 4.2. Work search andjob acceptance policies inAustralia's program of unemployment insurance Australia's experience with work-search requirements-and with determinations of "suitable" jobs for unemployed workers- i s informative. With increases in unemployment beginning in the 1970s, Australia's work-search requirements became progressively more formal, with the unemployed being required to supply evidence of an active search for employment. Changes effective in 1991 required both the short- and long-term unemployed to satisfy an activity test. For those unemployed for fewer than 12 months, the activity test includes an active work search or participation in training. The long-term unemployed must participate in an activity agreement (which may include unpaid volunteer work) intended to secure reemployment but tailored to individual circumstances. Further changes in the activity test became effective in 1995, with increased emphasis was placed on early identification of individuals who are likely to be unemployed long term. Before the large increase in unemployment in the 1970s, unemployed Australians were expected to accept only "suitable" work. As unemployment outpaced vacancies, the definition of suitable work was changed. After 1976 the unemployed were required to accept work even if it meant a reduction in wages or status. B y 1989, this definition had been further modified to require acceptance of casual, part-time, or temporary work. Source: Vodopivec and Raju (2002). Reduced labor market flows tend to hinder labor-force adjustment and the reallocation of jobs. In so doing they also slow down aggregate productivity growth (Davis and Haltiwanger 1999). 81 Vodopivec and Raju (2002) summarize the evidence. The disincentive effects of unemployment insurance can be significant. Increases in the duration of benefits create a greater disincentive than do changes in the benefit level. 55 4.29. In addition to the aforementionedgovernment actions for realigning expectations of the workforce, improving the functioning of the labor market is necessary. The Government has a limited but crucial role in the labor market. The Government can play an important role by integrating employment services $e., job search assistance) with social protection programs (such as the NAF and unemployment insurance), TVET programs, and the education system. To improve on public provision of employment services, the Government can incorporate private providers into public programs (World Bank 2003). Moreover, by increasing the transparency of civil service hiring (public job advertisements and appointments based on objective criteria) the Government reduce the sway of wustu in public employment while improving recruiting efficiency. Finally, the Government has a crucial role to play in collecting, analyzing, and disseminating labor market information. Areas requiringparticular attention are the following: 0 Data collection and analysis to distinguishbetween the involuntarily unemployed and the voluntarily unemployed for the purpose of focusing public resources on the constraints faced by each group. 0 Dissemination of information to help prospective students make informed choices among programs and to develop realistic expectations of the labor market. 0 Improvement of the links between the Department of Statistics and policy makers, by developing capacity in line ministries (such as the Ministry of Labor) to interpret and use data indeveloping labor policy. 0 Development of reliable statistics on productivity and unit labor costs to permit policy makers to track movements inthese quantities. 4.30. In this section we discussed several actions related to labor supply that government can take to reduce unemployment. In the long term, if the quality of Jordan's education system can match its success in terms of quantity, Jordan may be able to develop a comparative advantage as a high-wage knowledge-based economy. In the short and medium term, increased focus on links to employment in the vocational training system can increase the employability of the involuntarily unemployed. Increasing women's attachment to the labor market through appropriate maternity leave policies will make fuller use of past investments in education and training. Unemployment of the voluntarily unemployed will not be reduced in the short and medium term without a realignment of expectations. Policies of civil service, emigration, social protection and access to labor market information all have crucial roles to play. 56 CHAPTER 5. CONCLUSIONS, POLICY OPTIONS, AND RECOMMENDATIONS 5.1. In recent years, Jordan has produced a large number of new jobs, but many have been taken up not by Jordanians but by foreigners. Many Jordanians remain unemployed. That there are now more foreign workers in Jordan than there are unemployed Jordanians suggests mismatches in the labor market. 5.2. Thisreport examined three mismatches: Expectations. Notwithstanding assertions that a "culture of shame" prevents Jordanians from accepting work they deem to be beneath them, we found that if wages are high enough, Jordanians are willing to accept the types of jobs now available in the labor market. Unfortunately, the wages at which most unemployed Jordanians would accept suchjobs are higher than prevailing wages. The majority of Jordan's unemployed can be considered voluntarily unemployed by this definition. Only 46 percent are willing to accept existingjobs at prevailing wages. Skills. There also i s a mismatch of skills, but not in the sense commonly believed. It i s widely believed that Jordanian workers will fill manyjobs now filled by temporary foreign workers once they receive training. The evidence, however, indicates that most unemployed Jordanians already have education that i s at least as good as that of foreign workers, as well as more vocational training and on-the-job experience than most foreign workers. The apparent mismatch of skills has more to do with responsibility, effort, and other aspects of employability. Geography. A disproportionate share of new jobs has been created in Amman. Employers have not moved to where the unemployed live, nor have the unemployed moved to areas where demand for labor i s high. 5.3. Expectations. In the short and medium term, unemployment will remain high- especially among the voluntarily unemployed-until prospective workers adjust their unrealistic expectations about available jobs. Four categories of actions are recommended. First, the Government can encourage the necessary adjustment by reforming civil service policy to reduce government hiring and eliminating the practice of accepting applications that are not tied to specific jobs and specific qualifications. Second, although existing social protection policies do not appear to create disincentives for work, care must be taken as these policies are reformed to avoid introducing such disincentives. Policy debates are ongoing for the reform of Jordan's income assistance program (the National Aid Fund) and for the possible introduction of unemployment insurance. Both types of programs have the potential to provide important assistance to poor and unemployed individuals, but both have their risks. International experience offers useful insights into mechanisms that can provide assistance while minimizing disincentives for work. Third, social marketing (as has been used successfully in Jordan on health issues) can be usedto promote the notion that all types of work are respectable. Finally, the Government has an important role to play in providing and disseminating accurate information about the labor market. Armed with information about trends in wages and expanding occupations, Jordan's citizens can develop more realistic expectations and betterjob-hunting strategies. 5.4. No panacea in training. In the short and medium term, programs that provide technical and vocational education and training (TVET) can reduce unemployment to the extent that they increase employability. The ongoing TVET reform i s an opportunity to realign the goals of this system. TVET should not, however, be seen as a stand-alone solution to unemployment or as the 57 magic pill that will allow Jordanians to replace foreign workers, because better training is not what gives foreign workers their advantage over Jordanian workers, as shown inthis report. 5.5. Rolling back policies that encourage emigration and creation of low-wage jobs. Jordan's economy and business environment have proven successful in creating low-wage jobs. The factors that made this possible are clear: (a) a tax system and labor policy that provide incentives to agriculture and industrial-estate-based manufacturing; and (b) immigration and emigration policies that keep the labor force inJordan at a lower educational level than that of the average Jordanian citizen. The educational level of the labor force in Jordan i s lower than the average of Jordanian citizens because so many of the best-educated Jordanians leave and so many less-educatedforeign workers enter. A continuation of existing policies will produce a situation in which (a) educated Jordanians work primarily outside Jordan, sending home remittances that their relatives spend inside Jordan, and (b) less-educated Jordanians compete with foreign workers for low-wagejobs. 5.6. Moving toward a high-wage, knowledge-based economy. The alternative scenario-to move toward a high-wage, knowledge-based economy-would require significant changes in the business environment. Cultivating Jordan's competitiveness in high-wage services and high-tech products will require policies that encourage business creation and that remove the distortions that presently distract entrepreneurs from focusing on Jordan's comparative advantage: its relatively well-educated citizens. In the long term Jordan can enhance its competitiveness as a high-wage, knowledge-based economy by emphasizing the quality of education (including problem-solving and teamwork) and including in the curriculum links to the labor market (includingaccurate informationon career options and employability skills). 5.7. Understandingemigration. Emigration has clearly provided Jordan with benefits, chief among them increasing (and smoothing) consumption among emigrants' families and providing a stable source of foreign exchange. Less i s known, however, about the costs of systematic emigration. Recent research has begun to explore those costs, which include a reduction in competitiveness (though an increase in the country's real exchange rate) and an increase in the "reservation wage" (which reduces job acceptance among the unemployed and thus labor force participation). Further study and discussion are needed on this important issue. The necessary precondition to developing an understanding of emigration-and of the labor market in general- i s the collection of pertinent data. 5.8. Boosting women's participation in the labor force. Jordan can increase the educational level of its work force and take advantage of past investments in women's education by encouraging women to join and remain in the labor force. One way to do this is by ensuring benefits that allow women to combine work and family (e.g., maternity leave) without creating disincentives for enterprises to hire women. Centrally financed maternity leave could help. 5.9. Integration of findings in Government-World Bank work program. Our findings and recommendations should be considered in the context of ongoing work by the Government in conjunction with the World Bank. That work includes analytical work and investments in the areas of public expenditure, financial sector functioning, investment climate, poverty mapping, and pension reform. It also includes projects in higher education development, education reform for the knowledge economy, social protection, employer-driven skills development, and regional and local development. 58 REFERENCES Ahn, S. 2000. "Firm Dynamics and Productivity Growth: A Review of Micro Evidence from OECD Countries" OECD Economics department Working Paper 297, Paris. Al-Asad, Mohammad. 2007. "To tax or not to tax: the case of Amman's empty land plots." Jordan Times, July 5. AMIR. 2004. "Reformulating the Tax Incentive Program in Jordan: Analysis and Recommendations." Prepared by Chemonics International, Inc., for the Achievement of Market-Friendly Initiatives and Results Program (AMIR), with support from the U.S. Agency for International Development. Assaad, R., and Melanie Arntz. 2005. Constrained Geographical Mobility and Gendered Labor Market Outcomes Under Structural Adjustment: Evidence from Egypt, World Development 33 (3): 431454. Ayyagari, Meghana, Asli Demirguc-Kunt, and Vojislav Maksimovic. 2006. "How Important Are Financing Constraints?" World Bank Policy Research Paper 3820. Washington, DC. January. Azzam, Henry. 2006. "The Growing Importance of Remittances to the Jordanian Economy." Jordan Times, December 11. Bartlesmand, Eric, John Haltiwanger, and Stefan0 Scarpetta. 2004. "Microeconomic Evidence of Creative Destruction from Industrial and Developing Countries," IZA Discussion Paper Series 1374, Institute for the Study of Labor, Bonn, Germany. Beck, Thorsten, Asli Demirguc-Kunt, and Miarai Soledad Martizez Periea. 2006. "Banking Services for Everyone? Barriers to Bank Access and Use Around the World." Policy ResearchWorking Paper 4079. World Bank, Washington DC. Benedict, Ruth. 1946. The Chrysanthemum and the Sword. Boston: Houghton-Mifflin. Birdsall, Nancy, Allen Kelly and Steven Sinding Eds. 2001. Population Matters: Demographic Change, Economic Growth and Poverty in the Developing World. Oxford: Oxford University Press. Borjas, George. 2003. "The Labor Demand Curve I s Downward Sloping: Reexamining the Impact of Immigration on the Labor Market." Quarterly Journal of Economics 118(4): 1335-74. Bussolo, Maurizio, and Denis Medvedev. 2006. "Do Remittances Have a Flip Side?" A General EquilibriumAnalysis of Remittances, Labor Supply Responses, and Policy Options for DC. Jamaica. Unpublished paper. Development Research Group, World Bank, Washington, Center for Strategic Studies. 2006a. "Corruption Issues in Jordan: A Public Field Survey." Jordan University, Amman, Jordan. www.css-jordan.org. Center for Strategic Studies. 2006b. "Study on Employment in Jordan: Replacement of Foreign Labor." Economic Studies Unit, Jordan University. September. http://www .jcss.org/UploadEvents/ 18.pdf. Central Bank of Jordan. 2005. Annual Report. Amman. 59 Chami, R., C. Fullenkamp, and S. Jahjah. 2003. "Are Immigrant Remittance Flows A Source of Capital for Development?" IMF Working Paper 189, International Monetary Fund, Washington, DC. Davis, Steven, and John Haltiwanger. 1999. "Gross Job Flows." In David Card and Orley Ashenfelter (eds.), The Handbook of Labor Economics. Amsterdam: North Holland. Department of Statistics (DOS) of Jordan. 2007. "Draft Study on Changes Related to Workers, 1995-2004." Unpublished. de Ferranti, David, Guillermo Perry, David Ledermean and William Maloney. 2002. "From Natural Resources to the Knowledge Economy: Trade and Job Quality." Latin American and Caribbean Studies, World Bank, Washington, DC. European Training Foundation. 2006. Technical and Vocational Education and Training in Jordan: Areas for Development Coordination. European Training Foundation, Turin, Italy. Garibaldi, Pietro, and Paolo Mauro. 1999. Deconstructing Job Creation, IMF Working Paper WP/99/109. Geroski, Paul. 1995. "What Do We Know about Entry?" International Journal of Industrial Organization 13: 421-440. Haltiwanger, John, Stefan0 Scarpetta and Helena Schweiger. 2006. "Assessing Job Flows Across Countries: The Role of Industry, Firm Size, and Regulations." World Bank Policy ResearchWorking Paper 4070. Washington, DC. November. Hanushek, Eric, and Ludger Woessmann. 2006. "The Role of Education Quality in Economic Growth." Policy Research Working Paper 4122. World Bank, Washington, DC. Hashemite Kingdom of Jordan and World Bank. 2004. "Jordan Poverty Assessment," main report, vol. 2. Amman. ILO (International Labour Organization). 2005. "The Endof the Multi-Fiber Arrangement and Its Implication for Trade and Employment." Employment Strategy Paper 2005116. Geneva. IMF(InternationalMonetary Fund).2005. World Economic Outlook: Globalization and External Imbalances (chapter 2). Washington, DC. April. Iqbal, Farmkh. 2006. Sustaining Gains in Poverty Reduction and Human Development in the Middle East and North Africa. Orientations in Development Series. Washington, DC: World Bank. Jordan Business. 2006. "Catch 22." September 6. http://www.jordan- business.net/magazine/index.php?option=com~content&task=view&id=38&Itemid=45 . Jordan Business. 2007. "Uneven Development Reversed." May. Kee, H. L., A. Nicita, and M. Olareagga. 2005. "Estimating Trade Restrictiveness Indices." World Bank Working Paper, Washington, DC. Kim, Namsuk. 2007. "The Impact of Remittances on Labor Supply: The Case of Jamaica." Policy Research Working Paper 4120. World Bank, Washington DC. Kolev, Alexandre, and Catherine Saget. 2005. "Toward a Better Understanding of the Nature, Causes and Consequences of Youth Labor Market Disadvantage: Evidence for South- East Europe" Social Protection Discussion Paper Series 0502. World Bank, Washington, DC. 60 Lopez-Claros, Augusto, Michael E. Porter, Xavier Sala-i-Martin, and Klaus Schwab. 2006. Global Competitiveness Report 2006-2007. Published for the World Economic Forumby Palgrave Macmillan: Hampshire, England. Mansur, Yusuf. 2007. "Fixing the rent law." Jordan Times. August 28 Nabli, Mustapha, and Marie-Ange Veganzones. 2004. "How Does Exchange Rate Policy Affect Manufactured Exports inMENA Countries?' Applied Economics 36(19). NationalAgenda. 2005. The Jordan We Strive For. www.nationalagenda.jo. National Tourism Strategy 2004-20 10. http://www.tourism.j oPDFs/NTS%20Book.pdf. Nassif, Claudia, and Peter Walkenhorst. Forthcoming. "Trade, Competitiveness and Employment in Jordan." MENA Social and Economic Development Working Paper Series, World Bank, Washington, DC. NCHRD (National Center for Human Resource Development). N o date. "Estimates of Demand for Labor During the Year 2005 Through Job Advertisements in Al-Rai Newspaper." http://www .almanar.jo/AlManarWeb/Default.a~px?tabid=79. OECD (Organisation for Economic Co-operation and Development). 2000. Employment Outlook. Paris. Population Council. 2006. "The Estimation of a Wage Index for Jordan: 1994-98 and 2000-03." Produced for the National Center for Human Resources Development (Jordan) by the Population Council, Cairo. Saadi-Sedik, Tahsin, and Martin Petri. 2006. "To Smooth or Not to Smooth-The Impact of Grants and Remittances of the Equilibrium Real Exchange Rate in Jordan." IMF Working Paper WP/O6/257, International Monetary Fund, Washington, DC. Saif, Ibrahim. 2006. "The Socio-Economic Implications of the Qualified Industrial Zones in Jordan." Economic Studies Unit, Center for Strategic Studies, Jordan University. http://www.jcss.org/UploadEconomic/55.pdf. Schiff, Maurice, 2005. "Brain Gain: Claims about its Size and Impact on Welfare and Growth are Greatly Exaggerated." Policy Research Working Paper 3708. World Bank, Washington DC. Transparency International. 2006. "Transparency International Corruption Perceptions Index 2006." www.transparency.org UNCTAD (United Nations Conference on Trade and Development). 2006. World Investment Report 2006. New York and Geneva: UnitedNations. UNDP (United Nations Development Programme). 2000. Jordan Human Development Report. New York and Geneva: UnitedNations. USAID. 2004. "Reformulating the Tax Incentive Program in Jordan: Analysis sand Recommendations." Report 512.02.01. USAID, AMIR Program, Jordan. Vodopivec, Milan and Dhushyanth Raju. 2002. "Income Support Systems for the Unemployed: Issues and Options." Social Protection Discussion Paper 0214, World Bank, Washington, DC. World Bank. 2001. "Engendering Development." Policy Research Report. World Bank, Washington DC. 61 World Bank. 2003. "Public Employment Services: Functions and Innovations," Employment Policy Primer Note 3. Washington, DC. September. http://www.worldbank.org/labormarkets. World Bank. 2004a. "Gender and Developmentin the Middle East and North Africa." MENA DevelopmentReport, WorldBank,Washington, DC. World Bank. 2004b. "Unlocking the EmploymentPotentialin the Middle East and North Africa: Toward a New SocialContract." MENA Development Report, World Bank, Washington DC. World Bank.2004c. World Development Report 2005: A Better Investment Climatefor Everyone. Washington, DC. World Bank. 2005a. Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union. Washington, DC:World Bank. World Bank. 2005b."Workers and Labor Markets." In World Development Report 2005. Washington, DC. World Bank.200%. WorldDevelopment Indicators. Washington, DC. World Bank. 2006a. Close to Home: The Development Impact of Workers' Remittances in Latin America. WorldBank, Washington, DC. World Bank. 2006b. Global Economic Prospects 2007- Managing the Next Wave of Globalization. Washington, DC. World Bank. 2006c. "Investment Climate Survey Online." http://iresearch.worldbank.org/InvestmentClimate/ World Bank 2006d. "Jordan Country Assistance Strategy 2006-10." Report 35665-50. Washington DC, World Bank. World Bank 2006e. "The Development Impact of Worker's Remittances in Latin America." Report 37026, Washington, DC. World Bank.2007a.Doing Business 2007: How to Reform. Washington, DC. World Bank. 2007b. "Export Diversificationin Egypt, Jordan, Lebanon, Morocco and Tunisia." Social and Economic Development Group, Middle East and North Africa Region. Washington, DC. Zaqqa, Nadim. 2006. "Economic Development and Export of Human Capital: A Contradiction? The Impact of Human CapitalMigration on the Economy of Sending Counties: A Case Study of Jordan." PhDthesis, Kassel University, Germany.www.upress.uni-kassel.de. 62