CY) t CO wo t t_ * E CN UU s fim OH b F O S > O i > > N~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~j The Impact of Future Labor Policy Options on the Palestinian Labor Market By Elizabeth Ruppert Bulmer* Social and Economic Development Group Middle East and North Africa Region Email: Eruppertbulmer@worldbank.org June 2001 Discussion papers are not formal publications of the World Bank. They represent preliminary and often unpolished results of country analysis and research. Circulation is intended to encourage discussion and comments; citation and the use of the paper should take account of its provisional character. The findings and conclusions of the paper are entirely those of the author and should not be attributed to the World Bank, its affiliated organizations, or to members of its Board of Executive Directors or the countries they represent. Table of Contents Summary The Impact of Future Labor Policy Options on the Palestinian Labor Market I. Introduction .............................................................1 II. Stylized facts of the Palestinian labor market .............................................................1 i. Segmentation between domestic employment and employment in Israel .2 ii. Constrained labor mobility .2 iii. High unemployment rate negatively correlated with employment in IsraeL .2 iv. Rising wages concurrent with falling unemployment .3 v. Persistent wage gap between domestic and Israeli jobs .4 vi. Unskilled Palestinian andforeign workers in Israel are substitutes .4 Synthesis of the stylized facts .............................................................5 III. Equilibrium model of the Palestinian and Israeli labor markets .............................................5 IV. Exogenous policy shocks and predicted quantitative effects ....................................................7 Calibrating the model .............................................................8 Results .............................................................9 Sensitivity analysis .............................................................11 V. Significance of the results ............................................................ 12 Figures Figure 1: Palestinian Employment by Location ................... ..........................................2 Figure 2: Palestinian Unemployment and Underemployment .............................................................3 Figure 3: Unemployment by Region .............................................................3 Figure 4: Correlation between Palestinian Employment in Israel and Unemployment ............ ............3 Figure 5: Change in Employment in Israel and Unemployment ...........................................................3 Figure 6: Average Daily Wage (NIS) .............................................................4 Tables Table 1: Initial Parameter Values .............................................................8 Table 2: Predicted Results of Labor Policy Changes ............................................................ 10 Table 3: Sensitivity Analysis ............................................................ 11 Table 4: Sensitivity Analysis for Large Parameter Changes ............................................................ 12 References ............................................................ 15 Annex I ............................................................ 16 Annex II ............................................................ 17 w;9i 193leI JAXII J L .O:l Ci' Ji >1 4 ji . )11& bA L5. 21 n1,1, .LYA1 ,A7 kjllLW1 LSj i syj ZIssIj oWI @t_)lolvljsa; a 8JIaX Z.J LW uu1j.xN la 4 sY jv~~~' IJl.t j i LJ .Rj Ei jil 1LJ L).O -.-JlXl ;1.3 iL: g 3-;o ) L~L,llJl ^-A L.j.; tLlj ;i jill Z .l i 1; 9 . l Z411 ! L U J 91 XXS MojUL.dl jLwi L 3jj! tJu ol;y Cu_ Asy JLxI L s. ^ ,;: IJ s; i IillJ-al l _J eiLeIS i S llXl;59 < l IyY lli Z1W_11 (,,I JAJ LI.n ?A xUli _1 4 , . F4N!Ce1R9JIl lel ZIA oLjl*;t i _W1;I> 9A*J9.o%1? l le1 i J-_A Ae .6jJ LjIe d-vs. ..1 I lell ;iJ c u ilJ JillF3 ptLi F il 1<. L';1L e)LA 41-rLS I >A j. ggj5 1 v _.-U| L s*y * We >1 ll LI 4Ai ;L]| SL:UL L 218j 1)_ il) He ;<>X (J;~Lz1 2 A 0e J5le9 1 |es,;L. e Resume Le pr6sent document cherche a quantifier les resultats du travail imputables aux changements de la politique du travail dans le contexte d'un accord 6ventuel de statut definitif entre I'autorite palestinienne et le gouvernement israelien. Un modele d'6quilibre du travail donne le cadre theorique pour tester les diff6rentes options de la politique du travail pour les march6s du travail palestinien et israelien fortement int6gr6s. L'analyse d6crit les principaux facteurs qui affectent les d6cisions d'offre et de demande de la main-d'ceuvre en Cisjordanie et dans la Bande de Gaza et en Israel - a savoir, 1'importance de la demande de main-d'aeuvre israelienne pour des travailleurs palestiniens, les ecarts salariaux persistants entre einplois domestiques et emplois isra6liens mieux pay6s qui ne sont disponibles qu'en quantit6s limitees, et le ch6mage structurel observ6 en Cisjordanie et a Gaza. Ces distorsions sont saisies dans un modele qui s'articule sur l'offre fixe de main-d'ceuvre palestinienne a Israel en raison de la politique d'immigration isra6lienne. Le modele sert a caract6riser l'impact des changements exogenes dans les flux de main-d'ceuvre palestinienne a Israel en cons6quence des changements de la politique des permis, des contr6les de securite et des fermetures des frontieres en Israel. Les resultats du modele sont calibres en utilisant les donnees disponibles pour produire des estimations quantitatives de l'impact a long terme sur les niveaux d'emploi, le ch6mage et les salaires dans les 6conomies palestinienne et israelienne. Les resultats statiques comparatifs indiquent qu'une augmentation des flux de main-d'oruvre a Israel accroit le salaire domestique palestinien car des travailleurs d6ja employes chez eux sont attir6s par des emplois mieux pay6s en Israel qui sont devenus plus ais6ment disponibles et a un cout moins 6leve de transport. L'augmentation de l'emploi ext6rieur palestinien n'est par consequent que partiellement compensee par une chute du ch6mage palestinien, et l'emploi palestinien domestique devrait chuter selon les previsions. L'ecart de salaire qui en resulte entre emplois domestiques et en Israel diminue quelque peu, mais l'ecart qui reste continue a affecter les d6cisions d'offre et de demande de main- d'ceuvre. * L'auteur tient a remercier le Bureau central des statistiques de l'Autorite palestinienne pour les donnees et l'assistance qui lui ont ete fournies ainsi que Sebastien Dessus, Peter Fallon, Radwan Shaban et un examinateur anonyme pour leurs commentaires. Summary This paper seeks to quantify the labor outcomes arising from changes to labor policy in the context of a potential final status agreement between the Palestinian and Israeli governments. An equilibrium labor model provides the theoretical framework to test different labor policy options for the highly integrated Palestinian and Israeli labor markets. The analysis describes the main factors affecting labor supply and demand decisions in the West Bank and Gaza and Israel - namely, the importance of Israeli labor demand for Palestinian workers, the persistent wage gaps between domestic employment and better-paying Israeli jobs that are available only in limited supply, and the structural unemployment observed in the West Bank and Gaza. These distortions are captured in a model that hinges on the fixed supply of Palestinian labor to Israel due to Israeli immigration policy. The model is solved to characterize the impact of exogenous changes in Palestinian labor flows to Israel as a result of changes in Israeli permit policy, security controls and border closures. The model's results are calibrated using available data to generate quantitative estimates of the long-run impact on employment levels, unemployment, and wages in the Palestinian and Israeli economies. The comparative statics results indicate that an increase in labor flows to Israel raises the domestic Palestinian wage as some workers already employed domestically are drawn to higher-paying Israeli jobs which have become more readily available and at a lower commuting cost. The increase in external Palestinian employment is therefore only partially offset by a decline in Palestinian unemployment, and domestic Palestinian employment is predicted to decline. The resulting wage gap between domestic and Israeli jobs narrows somewhat, but the remaining gap continues to affect labor supply and demand decisions. * The author gratefully acknowledges the data and assistance provided by the Palestinian Central Bureau of Statistics, and comments from S6bastien Dessus, Peter Fallon, Radwan Shaban, and an anonymous reviewer. The Impact of Future Labor Policy Options on the Palestinian Labor Market I. Introduction Palestinian and Israeli policy makers are grappling with an array of policy questions in the context of future final status negotiations to define the terms of a permanent political accord between Israel and an independent Palestinian state. The final political decisions will have direct and indirect economic implications, and labor policies will be central. The Israeli and Palestinian economies are highly integrated, especially the two labor markets; under conditions of relatively open but controlled borders - that is, prior to the 2000/01 crisis - more than 100,000 Palestinian workers, or one-fifth of the total labor force, commute daily to relatively high-paying jobs in Israel. Although Palestinian labor flows to Israel are constrained through permit requirements and border closures, employment in Israel provided one-third of total Palestinian labor earnings in 1999. Labor mobility to Israel affects employment levels (both domestically and in the Israeli unskilled sector), relative wages, and growth outcomes in both economies, implying that the potential impact of final status policies regarding Palestinian labor flows is significant. Any change in policy that affects labor supply and demand, namely through tightening or easing Palestinian labor mobility across borders, will lead to different labor allocations in the Palestinian economy and in Israel, especially vis-a-vis Israel's unskilled foreign labor force. Potential final status economic arrangements - both broadly and with respect to labor policy - will fall somewhere in the range between complete separation (i.e., closed borders) and full economic integration (i.e., open borders). The objective of this analysis is to quantify the impact of potential final status labor policies on Palestinian employment, unemployment and wage outcomes in the long run. To this end, the analysis develops a theoretical model that characterizes the labor market in the West Bank and Gaza (WBG) and its links to the unskilled labor market in Israel. The analytical framework captures the main factors affecting labor supply and demand decisions in the West Bank and Gaza and Israel - namely, the importance of Israeli labor demand for Palestinian workers, the persistent wage gaps between domestic Palestinian employment and better-paying Israeli jobs that are available only in limited supply, and the structural unemployment observed in the West Bank and Gaza. By calibrating the model using labor force survey data from the Palestinian Central Bureau of Statistics (PCBS) and other available data, the analysis generates concrete estimates of the possible long-term impact of labor policy changes on employment, relative wages, and unemployment. The paper is organized as follows. Section II presents an overview of the Palestinian labor market and recent trends in employment and worker flows to Israeli jobs. These stylized facts are reflected in the equilibrium labor market model described in section III. In section IV, the model is solved to determine the impact of exogenous policy changes on domestic and Israeli employment and wages, and the predicted effects are calibrated to generate quantitative estimates. The final section draws conclusions from the results of the analysis. II. Stylized facts of the Palestinian labor market The Palestinian and Israeli econornies have a long history of interdependence that pre-dates the current political and economic arrangements. Until the late 1980s, Palestinians and Israelis could move freely between each other's territory. Israel introduced permit requirements in 1991 to control the movements of Palestinian workers, and the 1993 Oslo accords marked a fundamental shift in Israeli labor policy, as permit controls and other mobility restrictions (i.e., temporary border closures) were strictly enforced. This effectively discouraged Israeli demand for Palestinian labor, which was 2 gradually replaced by foreign workers. Quarterly labor force surveys conducted by PCBS since 1995 provide data on trends in Palestinian employment (both domestic and in Israel), unemployment, and wages. The Palestinian labor market exhibits six main features, or stylized facts: (i) segmentation between domestic employment and employment in Israel; (ii) constrained labor mobility; (iii) high unemployment rate negatively correlated with employment in Israel; (iv) rising wages concurrent with falling unemployment; (v) a persistent wage gap between domestic and Israeli jobs; and (vi) unskilled Palestinian and foreign workers in Israel are substitutes. i. Segmentation between domestic employment and employment in Israel Israeli demand for Palestinian labor has a large Figure 1 impact on aggregate Palestinian employment outcomes. In Palestinian Employment by 1999, nearly 23 percent of employed Palestinians held jobs Location in Israel or Israeli settlements (averaging 126,000 out of a total employment level of 555,000, as shown in Figure 1).1 5o 0000 This represents an increase from the 15-16 percent average 5s 000.i _ in 1995 and 1996, but is below historical levels around 30 percent observed during the 1980s and into the early 1990s. 400.000 ii. Constrained labor mobility 200000 no .000 IhM i Palestinians entering Israel for work must hold valid work permits in order to pass through border controls. The o qualifying criteria for a permit are selective (i.e., based on 1996 1997 1998 1999 gender, age, and marital status), and workers are prohibited [ Ls rael a WBG from spending the night in Israel. Border controls are very strict along the perimeter of the Gaza Strip, but monitoring is less rigorous in the West Bank due to its porous borders. Even workers commuting between Gaza and the West Bank, or between separated areas of the West Bank, are subject to security checks and limits on their movements. The situation improved somewhat with the introduction of a "safe passage" between Gaza and the West Bank in 1999, although access remained subject to certain qualifying criteria. Since 1993, there have been periodic episodes of severely restricted labor mobility imposed by Israel. Under "comprehensive" closures, Palestinians are effectively constrained to remain in their home villages.2 Although large numbers of Palestinians manage to cross illegally into Israel, total labor flows are reduced sharply during closures, bringing average daily labor flows down to between 50,000 and 60,000 (based on historical PCBS data). iii. High unemployment rate negatively correlated with employment in Israel Border closures and other security controls on Palestinian labor flows during 1995-1997 coincided with high rates of unemployment within WBG which averaged nearly 24 percent in 1996. Unemployment declined substantially in 1998 and 1999, falling to half its 1996 level (see Figure 2). Underemployment followed a similar pattern.3 Disaggregating the data by region, Gaza suffers much higher rates of unemployment relative to the West Bank (depicted in Figure 3). Figure 4 illustrates I Data are from PCBS Labor Force Surveys, and reflect full employment only (excluding underemployment). 2 During October 2000-June 2001 (the time of this writing), mobility restrictions reached their most extreme level to-date; border closures were in effect total, and in some areas Palestinian residents were subject to extended periods of round-the-clock curfews. 3 A worker is defined to be underemployed if his/her employment is inadequate relative to alternative employment. This includes those who work less than 35 hours per week but desire additional hours of work. 3 Figure 2 Figure 3 Palestinian Unemployment Unemployment by Region and Underemployment 35% 400/ 35% 6o 25% 25% 9 1W 20% ___ ~~~~~~~15% 10% o -_ -. ,A l 0%- % m . ii ELI P 0% o 5 T 'a-- 1995 1996 1997 1998 1999 1995 1996 1997 1998 1999 L Unerrphoyment Eo UndererTployment - West Bank ----Gaza Strip the close link between Palestinian employment in Israel and domestic Palestinian unemployment; when Palestinian employment in Israel is low, unemployment is high, reflected by a correlation coefficient of -0.9. Quarter-to-quarter fluctuations in Palestinian labor flows and unemployment Figure 4 Figure 5 Correlation between Palestinian Change in Enrployment in Israel Emiployment in Israel and and Uneffployment Unenfployment 80.000 200,000 60,000 150,000 ; 40,000 ;. o 20,000 0 100,000 : 0 50,000 .00F -40.0 00 50,000 100,000 150,000 200,000 60,000 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Employed In Israel --d(irpl. in Israel) +- d(Unerrployrmnt) (depicted in Figure 5) suggest that in the short run, sharp reductions in Israeli jobs lead to large increases in unemployment, as domestic Palestinian wages do not adjust immediately to absorb returning Palestinians. iv. Rising wages concurrent with falling unemployment According to PCBS labor force survey data, Palestinian wages have increased in both nominal and real terms since 1997, as depicted in Figure 6 by the fact that nominal wages increased faster than inflation. These findings are surprising given that unemployment declined markedly over the same period, contradicting the theory that declining unemployment and rising employment drive wages down. This evidence implies that job creation was not accommodated by downward wage adjustment to absorb additional job seekers, but rather that pent-up demand for labor was eased through additional supply. 4 v. Persistent wage gap between domestic and Israeli jobs Although average wages earned within WBG appreciated more than those earned by Palestinians working in Israel during the past 4 years, a significant gap persists (see Figure 6). In 1999, wages earned by Palestinians employed within WBG averaged nearly 60 percent of those earned by Palestinians working in Israel and the settlements (according to PCBS data). Wage gaps of comparable magnitude were observed prior to the interim period as well (Angrist 1996). Several factors might explain higher wages in Israel: higher average productivity in Israel, excess Israeli demand for unskilled labor, or minimum wage and other labor legislation and payroll tax financing (relevant to formal sector workers only). In 1999, the marginal returns to working in Israel compared to working domestically were over 90. percent, controlling for individual factors.4 vi. Unskilled Palestinian andforeign workers in Figure 6 Israel are substitutes Average Daily Wage (NIS) It is useful to examine the features of the 10 Israeli unskilled labor market in order to understand 100 the dynamics of Palestinian employment and wages. 60 The non-Palestinian portion of the unskilled labor 60 ' force in Israel consists primarily of foreign workers, 40 far outnumbering Palestinians. This reflects a 20 fundamental composition shift since the early 0 1990s, when the number of-foreign workers in A , c (b , b c° cb , , Israel was negligible and Palestinian commuters o o o Co o ol3 o exceeded 100,000 (Shaban 1999). Israeli labor demand for unskilled workers was accommodated gNomina wage gsrael NominalwageWBG by immigration policies that granted easier access to foreign rather than Palestinian workers, and at a lower unit labor cost to the employer, effectively institutionalizing discrimination against Palestinians in favor of foreigners. For workers in the formal sector, payroll tax rates for Palestinians range from 16 to 29 percent (Farsakh 1999), compared to an effective rate of 7-9 percent for foreign workers (Eckshtein and Prizat 2000). Although foreign work permits have a fixed duration (typically I year), most foreign workers remain longer - either through permit renewal or in informal jobs - suggesting an unconstrained supply of foreign labor. This is evidenced by the fact that illegal foreign workers are at least equal in number to legal foreigners, and by s'ome estimates outnumber them three-to-one. Data on wages of foreign workers in Israel is unavailable, but anecdotal evidence and international experience suggest that foreign workers have a low reservation wage that reflects sending-country conditions. Moreover, this reservation wage may be significantly below the prevailing wage paid to Palestinians in Israel's formal sector, as implied by the lower unit labor cost and the continued inflow of foreigners to meet increasing Israeli labor demand (Bartram 1998). Given that jobs in the unskilled sector have low value-added, Palestinian and foreign workers are close technical substitutes, evidenced by the fact that Israeli employers have to a large extent replaced Palestinians with foreigners since 1993. They may not be perfect substitutes, however, in light of continued Israeli demand for Palestinian workers despite their higher cost compared to foreigner labor. Although possible explanations include language ability or skills specialization which give Palestinians a comparative advantage (Farsakh 1999), these employment trends are surprising, and suggest that political factors play some role. 4 See annex I for regression results using micro-level labor force survey data from the second quarter of 1999. The determinants of wage earnings were estimated using Mincerian specifications that control for individual characteristics including age, gender, educational attainment, occupation and sector of employment. 5 Synthesis of the stylized facts This combination of stylized facts points to -the dynamic processes and pressures that shape current Palestinian labor market outcomes. Decisions of labor supply and demand can be broadly characterized as follows: Palestinians can choose to work domestically within WBG, but wages are lower than those offered by Israeli employers. Jobs in Israel are not readily available, however, due to permit requirements, uncertain access as a result of security controls and border closures, and high transportation and search costs. Nevertheless, the prospect of higher wages leads many Palestinians to seek jobs in Israel despite the associated costs and risks of unemployment and income loss. The fact that some portion of the Palestinian labor force is effectively searching or queuing for a job in Israel rather than accepting less remunerative work in the domestic economy implies that high Israeli wages raise the market wage in the Palestinian labor market, thereby reducing the level of domestic employment. Because permit requirements and border controls constitute limits on the supply of Palestinian labor to Israel, wages are not market-clearing. This rigidity vis-a-vis Palestinian labor supply leads to a persistent wage gap and gives rise to structural unemployment. In other words, Palestinian unemployment and domestic wages respond to external job opportunities and wages in Israel. And although supply controls on Palestinian labor flows are at least partially binding, those on foreign workers are not, suggesting a potentially infinite supply of foreign labor. The model developed in this analysis reflects these features (see annex II for details of the model). III. Equilibrium model of the Palestinian and Israeli labor markets The observed labor market segmentation in Israel and wage rigidity in the Palestinian economy supports the use of a theoretical approach based in labor market dualism and structural unemployment. Policy changes vis-A-vis labor flows between the Palestinian and Israeli economies will have far-reaching implications: although Israeli employers can substitute foreign workers for Palestinian labor (albeit at a higher social cost), the employment and earnings implications on the Palestinian side will be significant. The model presented here describes the equilibrium allocation of labor and wages under existing arrangements implemented during the 1990s, and is subsequently used to examine the impact of positive or negative changes in labor mobility. Whereas the model incorporates the interaction of labor demand in Israel and the West Bank and Gaza, the model abstracts away from aggregate labor demand in Israel in order to focus on unskilled labor only.5 Much of the literature on dual labor markets applies to the segmentation between the formal and informal sectors and the associated wage differentials (e.g., Dickens and Lang 1985). This analysis, in contrast, addresses the segmentation that arises in the Palestinian labor market between workers employed domestically and in Israel. The model examines total demand for Palestinian labor by incorporating the relevant part of aggregate Israeli labor demand (namely, the demand for unskilled workers), accounting for substitution effects between foreign and Palestinian workers. By assuming that Palestinian and foreign workers are substitutes, changes in labor supply and demand are transmitted to the Israeli unskilled labor market and throughout the Palestinian labor market. S The Israeli labor market is highly segmented between skilled and unskilled labor, where skilled jobs are typically held by Israelis, and unskilled jobs are held by Palestinians or foreign workers (in this analysis, "foreign" refers to non-Palestinians). Palestinian labor is assumed to have no role in the Israeli skilled sector, such that there is no substitution between skilled Israelis and all non-Israeli workers. There are, of course, skilled Palestinian and foreign workers employed in high productivity occupations in Israel, but this simplifying assumption enables the analysis to focus on the Israeli unskilled sector. 6 The model developed in this analysis (see annex II for details) captures the main factors contributing to Palestinian labor market outcomes, namely limited labor supply of Palestinians to Israeli jobs, effectively unlimited supply of foreign workers to Israel's unskilled sector, exogenously determined foreign wages, domestic Palestinian labor demand, the relative unit labor costs of Palestinians and foreigners working in Israel and Palestinians employed in WBG, and structural unemployment. The model uses a static equilibrium framework to assess the long-run impact of exogenous shocks to Israeli immigration policy on the distribution of the Palestinian labor force among domestic employment, Israeli employment, and unemployment. The size of the Palestinian labor force is treated as constant, such that the model does not account for labor force entry and exit due to demographic factors or the discouraged worker effect. In addition, feedback effects from employment and income growth on labor demand are not considered. The labor model also assumes a production fuAction of fixed capital costs, thus abstracting away from substitution between labor and capital., The level of Palestinian employment in Israel is determined by labor demand on the part of Israeli employers, the number of permits issued, and the effective control of labor flows across borders. Palestinians are drawn to better-paying Israeli jobs, but are limited by mobility constraints that affect legal and illegal workers alike. In the model, therefore, Palestinian labor supply to Israel is exogenously fixed below its market-clearing level (i.e., LP, = DLn ), and the demand for Palestinian labor in Israel is assumed to exceed Palestinian labor supply to Israel. The persistence of Palestinian unemployment and the lack of downward adjustment in the domestic Palestinian wage described in section n suggest some wage stickiness, which cannot be due to labor regulation since there is little, and thus must be due to the proximity of Israeli jobs which pay considerably higher wages. By fixing the level of Palestinian labor flows to Israel, labor supply cannot respond fully to demand and therefore real wages are not equilibrated. However, the model assumes that expected wages are equilibrated through changes in labor supply and demand.6 Specifically, a Palestinian worker is assumed to choose between a domestic job, which he/she can obtain with certainty but at a lower wage, or an Israeli job, which is available with some probability less than one, depending on Palestinian labor mobility and the unemployment rate. That is, when the number of permits is restricted and Palestinian unemployment is high, the competition for obtaining an Israeli job increases sharply, reducing the probability of earning the higher Israeli wage. Furthermore, Palestinians working in Israel incur additional costs for transportation, job search, and permit requirements (or permit evasion efforts), all of which contribute to lower the Israeli wage in real terms. The expected wage for Israeli employment is therefore discounted by these costs as well as the reduced probability of obtaining the job. In equilibrium - that is, at the margin - expected earnings are equalized when a worker is indifferent between working domestically or seeking employment in Israel. In the model, the probability of obtaining an Israeli job is expressed as a linear function of the employment rate, or 1 minus the unemployment rate, (1-up), and is less than 1. Note that by defining employment as full employment (i.e., those who worked more than 35 hours per week during the reference period), the residual unemployment rate, up, includes underemployment as well. This is consistent with the fact that underemployment represents excess labor supply, given that workers who are underemployed would, by definition, prefer to be fully employed. 6 This approach was delineated in the migration literature by Harris and Todaro (1970) to describe the wage gap between rural and urban jobs. More recently, Ruppert (1999a,1999b) employed a different application of expected wage equalization to analyze labor markets with structural unemployment. 7 Domestic Palestinian wages are denoted by WpD, and wages earned by Palestinians in Israel are denoted wpl. The costs associated with working in Israel (e.g., for transportation, search and permits costs) are captured in the model by the fraction ,, such that in real terms, Palestinian workers in Israel earn ,Bwp, (,<1), and 1 depends positively on the degree of Palestinian labor mobility and thus the magnitude of labor flows to Israel, Lp,. In other words, as the number of permits increases and security controls are eased, the transportation and search costs decline and the share 13 of the actual wage wp, increases (namely, P'(Lp1>O). The equilibrium wage condition that describes the wage gap between domestic Palestinian wages and those earned by Palestinians employed in Israel is given by: WPD = (1 -UP) *,l) * Wpi Since O<(l-up)1. The size of (D depends negatively on the magnitude of Palestinian labor flows into Israel; as more Palestinians are allowed to cross into Israel and security restrictions are eased, the risk of absenteeism diminishes and (D declines (i.e., (D'(LPl) < 0). The model assumes that wages of foreign workers are determined by labor market conditions in the sending, or labor-exporting, country, and that the supply of foreign workers is effectively infinite, as discussed above in the stylized facts. Foreign wages therefore do not respond directly to changes in Israeli labor demand. Palestinian and foreign workers in Israel are assumed to be close technical substitutes, but have different unit labor costs. The distortions reflected in the stylized facts and captured by the model - namely the limits on Palestinian labor supply to Israel (LPI ), the costs of commuting (f3), the uncertainty of finding an Israeli job, the Israeli employer's cost of Palestinian absenteeism (0), and the difference in wages paid to Palestinian and foreign workers in Israel - create wedges that affect outcomes in equilibrium by preventing full employment and wage adjustment to equate labor supply and demand. IV. Exogenous policy shocks and predicted quantitative effects The model is solved to assess the impact of changes in Palestinian employment in Israel on the long-run allocation of Palestinian and foreign labor in Israel, domestic Palestinian employment, domestic and Israeli wages, and Palestinian unemployment. Comparative statics exercises are used to determine the effect of exogenous changes in Lp1 as a result of new labor policies under a final status 7 This assertion is supported by evidence from the 1999 labor force survey that a disproportionate share of unemployed Palestinians previously held jobs in Israel or the settlements. 8 agreement, and the results are calibrated to the Palestinian context using available data (discussed below). The comparative statics results summarized here indicate the direction of impact on the model's endogenous variables (reflecting calibrated results): (a) dw 0 dL1 (b) dLPD < dL., (c) dwp Po dLP, (d) d <0 dLv, (e) dLFI 0 dLz1 Calibrating the model The model is calibrated by assigning parameter values using available data (summarized in Table 1) to generate quantitative estimates of the impact of changes in labor mobility. Data on labor force size, Table 1: Initial Parameter Values employment levels, underemployment, Parameter- Value Data Source and unemployment comes from PCBS P quarterly labor force surveys for 1999. L 671,307 PCBS labor force surveys 1999 Employment data on foreign workers is LPD 428,918 PCBS labor force surveys 1999 less reliable; official Israeli estimates of Lp, 126,253 PCBS labor force surveys 1999 LFI range from 150,00-160,000, including Up 116,136 PCBS labor force surveys 1999 both legal and illegal workers, while other up 17.3% PCBS labor force surveys 1999 estimates lie in the 200,000-250,000 range L (Oxford Analytica 1996, Arnon et al. LFI 160,000 Israeli Ministry of Labor - ~~~~~~~~250,000 Oxford Analytica 1996, 1997, Kav La'Oved 2000). To address Am0on et al. 1997 the uncertainty surrounding the magnitude wp1 100 index of LFI, the model iS calibrated for 5 of- L.,temdli clbae o WPD 52 author's regression estimates on different values, two of which are reported PCBS labor force data Q2 1999 below. Wages of foreign workers are WFI 90 author's estimates estimated to be 10 percent less than 0 authorim ates Palestinian wages, which reflects 0.63 equilibrium wage equation anecdotal evidence vis-a-vis informal co 1F.I author's estimates workers, and the maximum difference in o 0.3 international experience2 payroll tax rates for legal foreign workers a 1000 author's estimates compared to legal Palestinians (Farsakh < 1.15 author'sestimates 1999). Palestinian wages in Israel wpj are indexed to 100. The wage of es -0.20 author's estimates domestically-employed Palestinians wpD Y 1.00 author's estimates relative to wpl is given by the equilibrium TDefined in annex 2 Hamermesh 1993, Guasch et al. 1997, World Bank 1997 9 wage condition, and depends on wage, unemployment, and underemployment data from PCBS labor force surveys. Using regression analysis on the determinants of earnings, the marginal returns to working in Israel averaged 91 percent in the second quarter of 1999 (as reported in annex I), implying a domestic Palestinian wage equivalent to 52 percent of the Israeli wage. By substituting values for WPD, Up and wpj into the equilibrium wage equation, D is calculated to be 0.63. This value of f3 is more or less consistent with observed transport costs equivalent to one-third of daily income (World Bank 2000). The degree to which ,B responds to shifts in labor flows is unknown; for simplicity, ep - the elasticity of i with respect to Lp, - is assumed to be equivalent in magnitude to e, the elasticity of D with respect to Lp1 (discussed below). The elasticity of domestic Palestinian labor demand, denoted by o in the model, is assigned a value of 0.3, which lies within the [0.15, 0.75] range identified by Hamermesh (1993) as reflecting "normal" values based on studies measuring aggregate constant-output labor demand elasticity for homogenous labor in industrial countries. Evidence from Argentina and Morocco suggest that employers in developing countries exhibit similar tendencies, with values falling between 0.3 and 0.75 (see Guasch et al. 1997, World Bank 1997). In the Palestinian case, o is set in the low end of the range to reflect the domestic economy's resource constraints and resulting limited ceapacity. Assigning a value to c - the elasticity of substitution between Palestinian and foreign workers in Israel's unskilled sector - is problematic due to lack of data on foreign workers, especially wage data. By assuming the two types of labor are very close technical substitutes, ca takes a high value. The model is calibrated for ca equal to 1,000 (near-perfect substitutes), and sensitivity to this assumption is tested below. The non-wage premium for absenteeism D paid by Israeli employers of Palestinians represents an implicit cost that cannot be easily quantified. Data on the gap between foreign and Palestinian net wages is needed, in addition to information on Israeli demand for unskilled labor and the actual size of the foreign labor force. If Lp1 and LFI are close substitutes, then the observed demand for Palestinian workers despite cheaper labor alternatives suggests that the premium is not very large. For this calibration, it is assigned a value of 1.15, implying a 15 percent premium. As Lp1 increases, (D adjusts downward; the degree to which cD responds to shifts in labor flows is estimated to be moderate (namely, elasticity Eo is set equal to -0.2) to reflect the size of variation in Lp1 observed in the past. Although the magnitudes of 'D and 0i are unlikely to move linearly with changes in Lp1, as implied by assuming co and sp constant, the lack of data precludes a more complicated assumption. Finally, -because the model uses production function g(Lp,LFI) to describe the marginal conditions of employing particular types of labor, it is necessary to specify the functional form of production. For purposes of generality, a CES production function is used. Results The model is calibrated for three different labor policy changes, resulting in (i) marginally improved Palestinian labor mobility, (ii) greatly improved labor mobility and therefore large Palestinian inflows into Israeli jobs, and (iii) sharp reductions in Palestinian employment in Israel. The three scenarios are discussed separately, and results are reported in Table 2 below (note that because the model is linear, the results for case 2 and case 3 are directly proportional to case 1). Case 1: Improved Palestinian labor mobility to Israel. In case 1, Palestinian labor mobility is assumed to improve slightly under a future final status arrangement, but essentially reflects the 10 current policy of limited Palestinian access to the Israeli labor market through permit requirements and border controls. The results reported in Table 2 below illustrate the impact of an increase in Palestinian labor flows by 10,000 workers. In colurnns A and B (for LFI equal to 160,000 and 250,000, respectively), the inflow of 10,000 additional Palestinians into Israeli jobs displaces 5,000- 8,000 foreign workers from employment, leading to a net increase in Israel's unskilled employment (Lul) by 2,100-5,000. Although the unit cost to Israeli employers of Palestinian labor wp1(P declines slightly, due to the decline in '1 as more Palestinians are given access to Israeli jobs, the wage itself (wp,) is predicted to rise by 1.6 percent, drawing more Palestinians to search for Israeli jobs. This effectively raises the domestic Palestinian wage wpD by 4 percent, and reduces domestic Palestinian labor demand. Domestic employment LPD is estimated to fall by 5,200. The resulting wage gap is narrower, which is consistent with lower unemployment and lower commuting costs (, rises by 1 point to 0.64, as shown in the bottom of Table 2). In net terms, the pool of unemployment (including underemployed workers) declines by 4,800 workers - considerably less than the 10,000 new Palestinian jobs in Israel - and the unemployment rate falls from its initial rate of 17.3 percent to 16.6 percent. Table 2: Predicted Results of Labor Policy Changes Case 1 Case 2 Case 3 dLp,>O dLpl>>O dLp< 0, f"(LPD) <0. Competitive Palestinian firms maximize profits by solving the following: (3) Max f (LPD) - WPDLPD - CD LPD which leads to first order condition (4) f'(LPD) = WPD . Notice that prices are normalized to 1, implying that wpD represents real wages. The elasticity of domestic (i.e., within WBG) labor demand for Palestinian workers LPD at wage wpD is given by: (5 (dLpDwPD' 0 d WPD LPD The level of Palestinian employment in Israel is determined by labor demand on the part of Israeli employers, the number of permits issued, and the effective control of labor flows across borders. PalestiniansFare drawn to better-paying Israeli jobs, but are limited by mobility constraints that affect legal and illegal workers alike. In the model, therefore, Lp, is exogenously fixed below its market-clearing level (i.e, LPI = LPI ), and the demand for Palestinian labor in Israel is assumed to exceed Palestinian labor supply to Israel. A wage gap arises in equilibrium when expected wages are equal at the margin, given by: (6) WPD = (1 - up),8(LpI) WPi, 18 Where (1-up)O). The sensitivity with which ,B responds to changes in Lp1 is given by elasticity £p: (7) d/fl= (L i) *_ >O_ . Al dLni f(L"1) Palestinian employment in Israel Lp1 depends on Israeli demand for unskilled workers, Lu1, the supply of foreign workers, and the elasticity of substitution between Palestinian and foreign workers. Total Israeli demand for unskilled labor is given by: (8) LuZ= LPI + LFI The Israeli unskilled sector is assumed to produce a single non-tradable good g(Lpl,LFI) using two types of labor inputs, Palestinian and foreign. Non-labor inputs are assumed to have fLxed cost CI. Let g(LpI,LFH) be continuously twice differentiable, with dg (.)/dLp, = gp > 0, dg(-)/dLFn = gF > 0, cross-partial derivatives dgp/dLFn = gPF = dgl/dLi = gFP > 0, and second derivatives dgp/dLp, = gPP < 0 and dgF/dLF, = gFF < 0. Palestinians in the Israeli unskilled sector earn wage wp1, but employers incur an additional non-wage cost 1 (>>l) associated with Palestinian workers, where (D represents a premium linked to absenteeism due to closures and other security constraints. This non-wage premium is endogenous and depends on the magnitude of Palestinian labor flows into Israel; as more Palestinians are allowed across the border, (D declines (i.e., V'(LPt) <0). The sensitivity with which D responds to changes in Lp1 is given by elasticity E': (9) co d4(Lp,) L, ) Foreign workers earn wage wFn, which is exogenously determined by labor market conditions in the sending, or labor-exporting, country (i.e., WFI = WFI ). The supply of foreign workers is effectively infinite, and foreign wages do not respond directly to changes in Israeli labor demand. However, there may be other costs borne by employers, denoted in the model by the parameter y (yt1), where y depends on LFI. Competitive Israeli firms maximize profits by solving: (10) Max g(LP, LFI) - Wn, 4(LP) LP - y(LF) WFILF - C, LFI generating the following first order conditions: (l 1) gF = )rWF dcD(L~,) (12) gp = Wp,4)(LP,) + WpjLp, dLP, Note that the price of g (LP,LPI) is also normalized to 1. 19 In order to solve the model for the equilibrium labor allocation and wages in the various sectors, two additional pieces of information are needed: the elasticity of Israeli labor demand for Palestinians, Ti, and the elasticity of technical substitution between Palestinian and foreign labor, a. These are defined in equations 13 and 15 respectively (Hamermesh 1993). (13) dLv Wp#D(Ln) = -(1- s)a, d (wP,(t(L,)) LPI where s is the cost-share of input LPI in total revenues: (14) s WPAD(LP,)LP1 O O1 (15) / ~ >0. d ln(g' ) Nine variables are endogenously determined: LPD, LFI, LUI, up, wpD, wpi, ,D and r. Exogenous variables are: Lnp , WFI, CD, Cl, , O, £, 64, and C. The main comparative statics results are as follows: dwn WpI ( 1 (a) = - _ =-I + dLPD Ln , Lp, ((1- s) -jL - _LP 1g+)1 (b) dL-0 - LI n (1-S)o j-1 dLn 1+ 6LPI (1-UP) -( 5P _; -L _ (E+)1 (c) _dwPD - up)fi(Lpi) LPD L- (1- .). - 1ij dLn, SLPD 1+ D (1-UP) r a~ AL ( 1 1 du [LPD _- _ E; + ( s))1j (d) dLnL 1+ L LPD (1- UP) 20 (e) dLF, dL,1 £ +4) (l-PXP ap , 2p_ 1-p s(~~a1 1-p gp i9Ppp +gfp] O 0LFI g-P F-Pg MENA Working Paper Series No. 1 Has Labor Migration Promoted Economic Integration in the Middle East? June 1992. Nemat Shcftk, The World Bank and Georgetown University. No. 2 The Welfare Effects of Oil Booms in a Prototypical Small Gulf State. September 1992. Ahmed Al-Mutuwa, United Arab Emirates University and John T. Cuddington, Georgetown University. No. 3 Economic and Social Development in the Middle East and North Africa. October 1992. Ishac Diwan and Lyn Squire, The World Bank. No. 4 The Link Between Trade Liberalization and Multi-Factor Productivity: The Case of Morocco. February 1993. Mona Haddad, The World Bank. No. 5 Labor Markets in the Middle East and North Africa. February 1993. Christopher A. Pissarides, The London School of Economics and Political Science. No. 6 International Competitiveness of the Private Industry and the Constraints to its Development: The Case of Morocco. June 1993. Hamid Alavi The World Bank. No. 7 An Extended RMSM-X Model for Egypt: Quantifications of Market-Oriented Reforms. September 1993. Karsten Nimb Pedersen, The World Bank. No. 8 A Report on the Egyptian Tax System. October 1993. Mark Gersovitz, Roger H. Gordon and Joel Slemrod, The World Bank. No. 9 Economic Development and Cooperation in the Middle East and North Africa. November 1993. Ishac Diwan and Lyn Squire, The World Bank. No. 10 External Finance in the Middle East: Trends and Prospects. December 1993. Ishac Diwan, John Underwood and Lyn Squire, The World Bank. No. 11 Tax Incidence on Agriculture in Morocco (1985-1989). April 1994. Jean-Paul Azani, CERDI, University of Auvergne, Clermont-Ferrand (France), et CSAE, Oxford (U.K.). No. 12 The Demographic Dimensions of Poverty in Jordan. August 1994. Chantal Worzala, The World Bank. No. 13 Fertility and Family Planning in Iran. November 1994. Rodolfo A. Bulatao and Gail Richardson, The World Bank. No. 14 InvestientEfficiency, Human Capital & Migration A Productivity Analysis of the Jordanian-Economy. May 1995. Gaston Gelos, Yale University, DepartniPnnt of Economics. No. 15 Tax Effects on Investment in Morocco. August 1995. David Sewell, Thomas Tsiopoulos and Jack Mintz, The World Bank. No. 16 Reconstruction in Lebanon: Challenges for Macroeconomic Management. April 1999. Daniela Gressani and John Page, The World Bank. No. 17 Towards a Virtuous Circle: A Nutrition Review of the Middle East and North Africa. August 1999. Regional HNP Knowledge Management, The World Bank. No. 18 Has Education Had a Growth Payoff in the MENA Region? December 1999. Lant PrLtchett, The World Bank. No. 19 Rationalizing Public Sector Employment in the MENA Region. December 2000. Elizabeth Ruppert Bulmer, The World Bank. No. 20 Achieving Faster Economic Growth in Tunisia. March 2001. Auguste T. Kouame. The World Bank. No. 21 Trade Options for the Palestinian Economy: Some Orders of Magnitude. March 2001. Claus Astrup and S6bastien Dessus. The World Bank. No. 22 Human Capital and Growth: The Recovered Role of Educational Systems. April 2001. SebastienDessus. The World Bank. No. 23 Governance And The Business Environment In West Bank/Gaza May 2001. David Sewell. The World Bank. i Copies of MENA Working Papers may be obtained from Ms. Gena Gorospe, MNA VP, Ph: (202) 473-2623.