S4ibblc iExst Art No-rth Africa 1' LE Discussion Document Paper de Series Travail No.1 I June 1992 ulr Juin 1992 Wom pa aewnot Ion" pubU"amn of t Wotd Bank. Theaet pr4niay andotenm polishd reults of a and research. Circuation is bIended to encourage dscussion and conmmet citti and the us of the pap shoudd take eacownt of its provisona characte. The findis and conclswIons of the paper ame nirey thof the autoral and shoudd not be attribed to the Wod Bank. _ts affilia wtegations, or to nmwer of ika oard of Exwutw DOrecto or the conties they rPnt._ Has Labor Migration Promoted Econonmic Intogration in the Middle East? Nemat Shafik The World Bank and Georgetown University June 1992 This paper has been prepared for the seventeenth annual symposium of the Center for Contemporary Arab Studies on 'Arab Integration: A Critical Evaluation' at Georgetown University, April 9-10, 1992. The views expressed here are those of the author and do not reflect the views of the World Bank or its affiliated institutions. i Abstract In terms of trade ail 4 capital flows, the Middle East is one of the least economically integrated regions of the world. The major exception is labor mobility where intraregional migration flows are extensive and remittances now exceed the value of regional trade iiA goods as well as official capital flows. The explanation for this pattem lies in the extreme differences in factor endowments across the region and, perhaps more importantly, the development policies adopted by both labor importing and exporting countries. In the case of nontradables, such as construction, education, government and domestic services, the oil economies had no alternative but to import labor if local demand was to be met. In the case of tradables, the labor-importing countries tended to have very cutward-oriented trade policies while the labor exporters tended to be inward-oriented. Thus, the demand for tradable goods from the oil economies tended to be met from the world market (particularly the OECD) rather than from regional trading partners. Because the obstacles to trade in goods have been greater than the obstacles to migration, labor mobility and its associated capital flows has been the most important mechanism through which the benefits of the oil windfall have been spread to the poorer states of the region. There is evidence that incomes across the Middle East have become more equal. But, without efforts to solidify regional economic ties on the basis of the efficiency and mutual self-interest, labor migration will remain an anomaly in an otherwise fragmented region. irriY 2i U `s V ri ;9 1 r q" Si -1t lb ~~~~~~~~~~~~R?" . W lc-n;kl 1N11 U~~'' - 9 F¢ IFm'V 'iHz 'r" r' V ¢lUstnss1U IKbY5 '"¢l< Gl-v ) Tf Im ITV,,!fSlXr IGr9 r' £e I! *J ISf¶ imPs ITe nTP '¢C IfX r?c sr' 1W1 Fr I(' i' 1 1 -mn r9 ;Cy(r e 6ns IfT i" rr n IKf '8 tci' ITmITa '1h war"' 'is u lri--nr rr i ;n@ I cIt llnlh-C, rP: 1n 11- II 04l; *c le 3C ¢Uor11m;? Ic>¢T vem 11 mro- imo r' S}- .mfllv 61 1t- +1^' c v5 le : 'r" 1 cilt¢To13 ' irrl k' < r' i im * pr 1, P irrer ff;--2T 6r->r lpert;ric O im-u STI'"-lp irTy-4 irer oftm - rc P -r-h irIP. i m-- -nr , rnm , -u. ,r ° ,rir- ,H- ,r--c im ff Irirmn'- rr i-er L L iii ARREGE En termes de commerce et de flux do capitaux, le Moyen-Orient est l'une des r6glons du monde la moins 6conomiquement int6gr6e. La principale exception etant enregistr6e dans le domaine de la mobilit6 de la main-d'oeuvre oui les flux migratoires intra-regionaux sont importants et ou les transferts de fonds excedent actuellement la valeur du commerce r6gional de marchandises ainsi quo les flux officiels de capitaux. Ce sch6ma s'explique par !es ecarts extremes enregistr6s dans les dotations en facteurs pour l'ensemble do la r6gion et peut 6tre, de manlare plus signiicative, par los politiques de d6veloppement adop.6es tant par les pays importateurs que par les pays exportateurs de main-d'oeuvre. En ce qui conceme les biens non 6changeables, tels que l'enseignement, les services administratifs et intemes, les 6conomies p6trolieres n'ont eu d'autre altemative que l'lmportation de main-d'oeuvre pour satisfaire a la demande locale. En ce qui conceme les biens 6changeables, les pays importateurs de main- d'oeuvre ont eu tendance a s'appuyer sur des politiques commerciales fortement ouvertes sur l'ext6rieur tandis qus les pays exportateurs de main-d'oeuvre se sont repli6s sur eux-m6mes. Ainsi, la demande de biens 6changeables provenant des 6conomies p6trolires a-t-elle 6t6 g6n6ralement satisfaite par le march6 mondial (en particulier par les pays de l'OCDE) plut6t que par les partenaires commerciaux r6gionaux. Etant donn6 que les obstacles au commerce de marchandises ont ete plus importants que les obstacles & la migration, la mobilit6 <'e la main-d'oeuvre et les flux do capitaux qui y sont associ6s ont constitue le principal instrument de transfert des b6n6flces p6troliers exceptionnels aux pays les moins favorises de la r6gion. Une r6partition plus 6quitable des revenus a donc 6t6 constat6e pour l'ensemble du Moyen-Orient. Cependant, si des efforts ne sont pas entrepris en vue de consolider les liens 6conomiques r6glonaux basis sur l'efticacit et l'int6rOt mutuel, le ph6nomene de la migration de la main-d'oeuvre demeurera une anomalle dans une r6gion par ailleurs fragment6e. Has Labor Migration Promoted Economic Integration in the Middle East? The idea of an economically integrated Arab world has been part of the region's political discourse for decades. While the idea is compelling to many, the Middle East is in some ways one of the least integrated regions in the world, despite decades of attempts to give economic mean-ng to the notion of Arab unity. The major exception is labor mobility where intraregional migration flows have been extensive in recent decades. Remittances from migrant labor now exceed the value of regional trade in goods as well as official capital flows. This paper explores the characteristics of economic integration in the Middle East and analyzes why labor flows have been the major channel through which intra-regional economic ties have been forged. The paper starts, with a brief summary of the motives for regional integration. The degree to which the region is integrated-in terms of trade in goods and movement of factors rf production--is then analyzed. The special role that labor migration has played in regional integration is discussed in light of the region's endowment and trade policies. The Middle East's pattern of labor-based economic integration is fairly unique. Whether it has left the region better or worse off is considered in the final section. Why Re?ional Integration? The appeal of economic integration is based on the circumstances of the region and, increasingly, on the changing characteristics of the world economy. At the regional level, there have been two interrelated motives for promoting integration. One was purely political and had its origins in the ideology of Arab nationalism. An economically integrated Arab world would provide the region with critical support for sustained and meaningful political integration. The 2 second motive was more economic and stemmed from the view that the complementarity of crdowments across the region could be the basis for enhanced development. The high income oil exporters were capital-rich, but poor in labor and productive land. Countries like the Sudan, Egypt, Syria and Morocco had substantial agricultural potential and low wage labor, but lacked capital. Lebanon and Jordan had surpluses of skilled labor. The more diversified economies, such as Algeria and Iraq, could also benefit from imports of labor and capital from abroad. Moreover, regional integration would enable all countries to take advantage of economies of scale in production, distribution and resource use. In the Middle East, there is the added dimension of regional conflict. Many argue that a region that is more economically interdependent and prosperous is less likely to experience political and military conflict.' The economic motive has become increasingly important as most countries in the region experienced econorric stagnatiotn in the lat2 1980s and the early 1990s. Per capita income growth in the Middle East was respectable between 1960-73 and the period of the oil booms, 1973-81, was one of accelerated growth. With the collapse in oil prices in 1986, most countries in the region experienced negative per capita income growth. Only three countries - Morocco, Turkey, Tunisia and Yemen - in the region had positive growth rates in the second half of the 1980s. The challenge for the future is even greater because of the region's young, urbanized and rapidly growing population. The Middle East, along with Sub-Saharan Africa, has the highest rate of population growth in the world. Total population in the Middle East was about ' The example of the European Coal and Steel Community formed after World War II is often cited as an example where encouraging economic ties laid the groundwork for peaceful coexistence and eventual integration in the form of the European Community. 3 260 million in 1990, 51% of which live in urban areas and 43% of which are under 15 years of age.2 About one-third of the population in the region's developing countries lives in poverty (less than $1 per day).3 This combination of growing populations and deteriorating living standards has reinforced the imperative for regional economic development. Changes in the wor'd economy, and particularly in world trade, have also reinforced the need for integration. Intraregional trade has been growing faster than world trade in recent years. The emergence of three major trading blocks in the world economy - dorninated by the United States, Japan, and Europe - has resulted in fears that countries outside such regional arrangements could suffer a fall in exports as the adverse substitution effects of regionalization outweigh the favorable income effects. Such fears have provided a stimulus to renewed attempts at regional trading arrangements all over the world - in Latin America, Central America, Africa and Asia. The conventional economic view on regional integration is that it is desirable where the trade creation effects are greater than the trade diverting effects. Thus, agreements among countries that would tend to trade with each other anyway would result in greater welfare gains than those that divert trade to higher cost suppliers. This is consistent with the evidence that agreements tend to emerge among countries where there is already intense trade. In many cases, the static gains from more efficient resource allocation are overshadowed by the dynamic efficiency gains that result from competition. There is also evidence of the benefits of such agreements. Bigger countries tend to grow faster - lending support to the view that economies 2 World Bank (1992). 3 Ravallion, Datt and Chen (1992). 4 of scale and efficiency gains associated with integration result in higher incomes.4 The numerous attempts to promote economic integration in the Arab world have been analyzed extensively elsewhere.5 Since the creation of the Arab League in 1945, economic integration has been on the regional agenda. In 1953, a multilateral trading agreement was signed under the auspices of the Arab League which exempted Arab agricultural commodities from tariff barriers and reduced tariffs on some industrial goods. The Arab Economic Unity Agreement signed in 1956 sought full economic union between Egypt, Iraq, Jordan, Kuwait and Syria. The same countries, with the exception of Kuwait, formed the Arab Common Market in 1964 which sought the gradual elimination of tariff and non-tariff barriers over a ten year period. These and numerous other attempts to promote integration failed largely because there was no willingness to subsume national interests to regional ones. Protectionist interests in all countries often secured exemptions to more open trading arrangements that undermined regional integration efforts. Coordination of other regional policies was also often undermined by local interest groups. The failure of economic integration efforts was not unique to the Middle East. The majority of such efforts failed at promoting integ. .'-n in virtually every developing region.6 The oil boom in the 1970s spurred the growth of institutions to transfer resources from the high income oil exporters to the poorer states in the region. The national funds established 4 Lachler (1989). s See Waterbury and Mallakh (1978), Makdisi (1979), Ghantus (1982). 6 For a survey of attempts at regional integration in other parts of the world, see Lachler (1989). 5 by Kuwait (the Kuwait Fund for Arab Economic Development) and Abu Dhabi (the Abu Dhabi Fund for Arab Economic Development) actually preceded the first oil price increase in 1973, while those established by Saudi Arabia and Iraq were both created in 1974. Regional institutions wei- also established - such as the Arab Fund for Economic and Social Development (1968), the Islamic Development Bank (1974), the Arab Bank for Ecosiomic Development in Africa (1973), the Arab-African Technical Assistance Fund (1974), Arab-African Oil Assistance Fund (1974), the Special Fund for Arab Non-Oil Exporting Countries (1974), and the Arab Monetary Fund (1976). Most of these provided concessional financing for projects or balance of payments support. Economic integration increasingly came to mean transfers from rich to poor states, rather than the more solid ties of genuine economic interdependence. How Economically Integrated is the Middle East? Economic integration can take many forms. These include free trade areas, customs unions, joint-ventures, preferential payments arrangements, favored trading status or common markets. The focus here will be on those dimensions that have actually been realized in the Arab world. In particular, the extent of integration in terms of trade in goods and in factor flows will be assessed before considering the special role of labor movements. Trade The most striking feature of trade patterns in the Middle East is how little the countries of the region trade with each other. Table I provides data on Middle Eastern exports and 6 imports in world trade. About two-thirds of all the region's trade is with the industrial countries. Intra-regional Middle Eastern trade accounts for only between 6-7% of total imports and exports. The Middle Eastern cou.atries trade more with Asia and with Eastern Europe and the former Soviet Union than they do with each other. Table 1. ltra-regonal Middle Eastern Exports and Imgorts as a share of Total Exports and Imgo,g4. J985-90 S&poPs Impor 1985 1986 1987 1988 1989 1990 1985 1986 1987 1988 1989 1990 Percent distribution Industrial countries 60.6 61.9 58.3 56.9 59.0 63.9 68.1 69.3 66.3 66.3 67.4 71.8 Deve;ping countries 31.7 28.9 31.7 33.0 31.S 28.:! 26.5 25.7 27.2 28.0 _8.3 25.3 Africa 1.8 1.8 1.9 2.1 1.9 1.5 .8 1.0 .9 1.0 1.1 1.1 Asia 13.2 11.0 14.0 15.1 I5.0 14.6 8.4 8.9 9.7 10.8 10.6 10.9 Empe 6.1 5.5 5.9 5.6 4.7 4.1 7.3 7.3 7.4 73 6.7 5.0 Middle East 6.2 6.7 6.3 6.8 6.7 5.0 7.6 6.4 7.3 6.7 7.6 6.1 Western Heisphere 4-3 4.0 3.6 3.4 3.1 3.1 2.4 2.1 1.9 2.2 23 2.2 Former U.S.S.R and selected othtercounAriesn.i.e. .8 1.1 1.0 1.0 1.5 1.8 1.2 1.6 1.6 1.4 1.2 1.1 Annual percent change World -12.1 -19.9 12.9 -1.4 22.3 23.9 -17.8 -9.0 1.2 11.5 -0.9 16.9 Industrial countries -10.8 -18.3 6.5 -3.8 26.8 34.3 -20.7 -7.3 -3.0 11.5 .6 24.4 Developingcountries -15.5 -26.8 23.7 2.6 16.5 11.1 -13.0 -11.9 7.4 14.6 .... 4.4 Africa -17.5 :20.2 20.7 11.2 10.0 -2.5 *25.2 18.5 -12.7 24.4 9.4 14.6 Asia .20.1 -33.4 43.5 6.3 21.4 20.6 -14.8 -4.3 11.1 23.4 -2.8 20.5 Europe -9.7 -27.9 22.0 -7.1 3.0 7.5 -13.8 -8.8 3.5 103 -10.2 -12.6 Middle East -14.7 -14.1 6.6 6.3 21.0 -8.3 -9.7 -23.9 15.4 2.5 13.2 -6.3 Western Hemisphere -7.8 -26.0 1.3 -6.3 11.6 21.3 -9.0 -19.8 -9.2 28.9 3.4 11.8 Former U.S.S.R. aWn selected othercountriesn.i.e. 18.1 6.4 .... 3.7 82.7 48.1 14.6 19.9 -2.8 -1.8 -12.0 1.6 Soume: MlF. Direction of Trade Statistcs (1991). When intraregional trade is decomposed by country (Table 2), the pattern is even more stark. The major share of intra-regional trade is dominated by three oil economies - Bahrain, Saudi Arabia and the United Arab Emirates. This is because the data on trade does not exclude re-exports -- such as when Saudi Arabia "exports" oil to Bahrain for refining which Bahrain then 7 "reexports." Such trade involves little value added and the products are not destined ultimately for the regional market. If such reexports were excluded from the data, regional trade would be even less than that reported in Tables I and 2. What is the explanation for such a low level of intra-regional trade? Some of it must lie ir. the composition of regional imports and exports. The exports of the Middle East remain dominated by primary products, particularly oil. Petroleum and petroleum products constitute over 90% of total exports for all the Gulf countries, as well as for Algeria and Libya. The Middle East, along with Africa, has the smallest share of world trade in manufactures.7 The region is also a net importer of food. Thus the imports of the region, which consist largely of food, manufactures and capital goods, are not major exports of the region. But the explanation for the composition of trade lies not only in endowment, but in the policies adopted by the governments of the region. Many of the labor surplus economies (such as Egypt, Sudan, and Syria) that could have been meeting the region's demand for rood and manufactures have followed import substitution policies for decades. Meanwhile, the capital surplus economies have tended to have very open trade regimes, which enable them to import from anywhere in the world. The protected production of regional neighbors could not compete in terms of quality or price with world markets. The evidence of the enormous divergence in trade policy between the labor surplus and the capital surplus countries of the region is presented in Tables 3-5. Three different measures ' World Bank (1992), p. 17. 8 Table 2. Intm -re-ional Exports and Imgorts of Middle Eastern Countrfes. 1985-90 (US$ million) Exrpors Imports 1985 1986 1987 1988 1989 1990 1985 1986 1987 1988 1989 1990 Bahrain 1,224 942 1.189 1,029 1.309 1,390 444 253 258 274 326 351 Egypt 231 197 196 226 230 231 167 131 143 212 271 253 1ron, Islamic Republic of 143 152 220 268 372 292 823 327 414 227 292 274 Iraq 557 483 501 645 824 478 266 337 362 449 529 504 Israel 229 230 140 146 168 5 8 59 27 28 22 7 Jordan 638 479 S15 565 497 587 383 329 851 391 465 427 Kuwait 298 322 389 479 588 324 780 605 684 770 843 435 Lebanon 144 t0o 0OS 152 253 287 267 251 274 349 261 258 Libya 19 24 13 16 30 40 285 104 143 1 2 4 Oman 43 66 394 472 576 547 28 27 31 38 38 38 Qatar 212 230 111 132 188 187 211 160 158 200 238 226 Saudi Arabia 783 818 634 776 903 903 2,406 1.956 2,084 2,386 2.501 2.527 Syrian Arab Republic 810 281 394 122 152 124 75 128 162 249 512 646 United Arab Emirates 859 961 988 1,071 1,397 1.405 1,216 919 866 1,028 1.190 1,048 Yemen Arab Republic 323 281 130 212 169 176 31 21 34 42 41 43 Yemen. P.D. Republic 201 194 226 220 251 264 7 9 15 21 15 16 Middle East not speciftied 24 19 20 20 23 21 28 31 11 14 15 16 Source: iMF. Ditectory of Trade Statistics (1991). of openness are reported because there is considerable controversy over which measures are the most appropriate.' The structure adjusted trade intensity ratios in Table 3 represent the share of imports and exports in GDP adjusting for structural characteristics of the economies such as size, per capita income, and oil endowment. Table 4 reports average import charges by category and Table 5 reports the frequency of non-tariff barriers. The conclusions are consistent across all measures of outvard orientation - the capital surplus oil economies tend to be very open while the labor surplus economies of the region tend to be very closed. Political alliances have added another dimension to regional trade patterns. The composition of Egypt's trade during the late 1960s and early 1970s and Syria's trade until recently were dominated by the Eastern bloc. Extensive trade with Eastern Europe and the 'For a discussion of the methodological debate, see Pritchett (1991). 9 Table 3. Structure adjusted trade intensity ratios. 198S. by rank. Middle Eastern countries OveroU Manufacturing Agrkulture Resources LDCs * Rank % Rank % Rank % Rank Bahrain 69.1 4 21.3 6 2.7 26 44.7 3 Jordan 27.2 12 11.6 16 2.1 28 9.8 14 Egypt 22.5 17 20.8 7 7.6 11 -6.3 73 Algeria 9.6 25 8.3 23 0.1 37 1.0 26 Moroeco -2.1 40 *0.8 43 *0.9 42 *0.5 33 Sudan -2.9 41 1.0 32 0.7 33 -4.2 57 Tunisia .5.0 44 -0.7 42 -6.4 83 3.1 22 Syria Arab Republic -12.4 60 4.1 56 -0.9 41 -7.4 75 Turkey -19.2 75 -9.2 72 -6.7 84 -2.1 42 Yemen. Arab Republic of -21.2 77 -6.7 62 -6.1 81 -7.6 77 United Arab Emirates -23.0 79 -26.4 92 *2.9 58 7.2 IS Kuwait -37.7 90 -22.2 87 -1.6 50 -13.1 89 Oman -43.4 93 -10.8 73 -1.6 49 -31.2 93 oe: The structure adjusted trade intensity ratios are derived from a regression of trade intensity (imports plus exports as a share of GDP) on population, land area, GDP per capita, transportation costs and oil endow..ment. The esulting residual is an indicator of the openness of the economy taking into account structural chamaeteristics. Rank refers to where a particular country is relative to 93 other countries in the sample. source: Adapted from Pritchett (1991). Table 4. UNC_AD data on mean total imoart ghqe by magr poee in geent and rank Overall Manufacturing Agieuture Resources b Rank % Rank % Rank % Rank Saudi Arabi 3.7 4 4.1 6 1.4 3 4.4 10 Qatar 4.3 5 4.0 S 5.4 7 4.0 7 United Aab Emirates 4.3 6 4.7 7 1.5 4 5.9 13 Kuwait 6.5 7 3.9 4 2.1 6 23.1 48 Bahrain 7.2 8 7.6 8 7.6 8 5.0 12 Algeria 18.2 21 22.! 27 15.5 20 2.4 S Syria, Arab Republic of 24.5 34 25.2 33 23.4 33 22.8 47 Jordan 27.1 39 32.2 48 16.3 23 12.4 31 Tunisia 27.5 40 28.0 40 27.8 42 10.7 30 Moroeco 34.6 51 35.1 51 29.8 46 37.5 67 Egypt 41.4 62 42.6 61 57.3 68 16.0 37 Turkey 44.8 65 46.9 65 37.3 58 26.7 54 Sudan 47.0 66 49.4 66 54.6 65 2S.5 52 Iran 70.1 72 80.4 74 69.2 70 20.4 42 Note: Rank refers to where a particular country is relative to 75 other countries in the sample. Source: Adapted from Pritchett (1991). 10 Table S. bWNCTAD data on the frequency of nontariff barriers. by major aUr te Overaul Manufacturing Agriculture Resources Ratio Rank Ratio Rank Ratio Rank Rado Rank United Arab Emirates O.S 2 0.3 3 I.S 4 0.1 9 Qatar 1.2 3 1.2 4 1.5 5 0.0 1 Babrain 3.5 4 0.0 1 7.2 10 0.0 3 Oman 4.0 6 5.2 9 1.5 3 0.1 14 Kuwait 7.9 8 7.2 14 15.1 12 0.3 19 Sudan 8.0 9 8.4 16 12.2 14 0.0 7 Saudi Arabia 8.4 11 8.8 18 14.4 17 0.1 11 Libya 9.4 12 10.0 23 14.3 16 0.0 8 Jordan 16.8 28 7.1 13 66.5 45 0.1 12 Egypt 38.6 40 35.4 42 46.4 37 42.8 35 Morocco 39.7 41 23.0 36 66.6 46 84.4 So Algera 68.4 52 60.1 50 86.6 56 87.4 54 Tunisia 77.6 55 71.7 54 84.0 55 94.1 58 Turkey 90.6 59 97.8 60 79.7 50 70.2 40 Iran 98.8 62 98.6 62 94.2 60 100.0 73 Syria, Arab Republic of 100.0 63 100.0 63 100.0 63 100.0 62 Yemen 100.0 71 100.0 71 100.0 71 100.0 70 Note. Rank refers to where a particular country is relative to 75 other countrics in the sample. Source: Adapted from Pritchett (1991). former Soviet Union was a reflection of political alliances, not market incentives. Similarly, the United States has emerged as one of Egypt's major trading partners in the 1980s, not because of comparative advantage, but because American aid is tied to U.S. source restrictions. Cagl Rows The pattern of regional capital flows is clearly from the capital surplus oil exporters to the labor surplus countries. Because of this pattern, the size of regional capital flows is closely tied to developments in the oil market. The evidence on capital flows, both official aid and private unrequited transfers (largely remittances) is presented in Table 6. The size of regional 11 capital flows tended to be greatest during the oil boom of the 1970s and fell considerably after the oil price collapse in 1986. Three economies - Jordan, Yemen, and Egypt -- emerge as highly dependent on regional capital flows in Table 6.9 For example, over two-thirds of the Jordan's GDP came from regional transfers in 1979 -- about half of GDP came from Arab aid and one-fifth of GDP was remittances from Arab countries. In the case of Yemen, remittances have been between one- third and one-half of GDP since the 1970s. Egypt, like Jordan, was also a major recipient of Arab aid until the Camp David accords in 1979 when the country was isolated politically and economically by Arab governments. But private capital flows continued to grow rapidly and remittances emerged as Egypt's major source of foreign exchange in the 1980s and 1990s. Comprehensive data on other private capital flows, such as Arab investment in other Arab countries, is not available. Data from Egypt on the nationality of private investors under Law 43, the investment promotion legislation, gives some indication of the importance of such flows. On average, Arab investors constituted one-half of all foreign investment in both inland and free zone projects under Law 43 between 1977-1989.b0 The remaining foreign investment came from the European Community, the United States and other countries. Arab investors were obviously the most important foreign investors to Egypt, but were fairly small compared to Egyptian investors who constituted over 60% of total capital under Law 43 projects between 1977-1989. More importantly, the levels of private capital flows from the richer to the poor 9 Lebanon may also be highly dependent on regional capital flows, but the data is too poor to draw any conclusions. 10 Calculation based on Isfahani (1990). 12 stnes in the region are a very small fraction of their total assets held abroad. This reflects the low expected return on regional investments because of risk and the economic policies of the poorer states in the region. Remittance levels have far exceeded official aid in recent years. In 1987, private unrequited transfers were five times greater than intra-Arab official aid flows. This is largely the result of the reductions in aid after the collapse in oil prices in 1986. Remittances from labor migration are now the largest source of capital flows in the region. Labor has been migrating in the Arab world for centuries, but it was the oil boom of the 1970s that saw a manyfold increase in the scale of the phenomenon. In 1975, there were an estimated 1.6 million migrant workers in the labor importing countries of Bahrain, Kuwait, Libya, Oman, Qatar, Saudi Arabia, and the United Arab Emirates. About 1.1 miiLo^. of those workers were of Arab origin." Projections by Serageldin et al (1983) indicated that the number of migrant workers would double to about 3.5 million by 1985. Subsequent estimates of the actual number of migrants in 1985 are closer to 8 million, implying a fivefold increase since 1975.12 It is difficult to assess the scale and composition of labor migration in the 1980s and 1990s because of the absence of comprehensive data. In earlier periods, major studies of migration were conducted by Birks and Sinclair in conjunction with the Economic Commission "Seregeldin et al (1983). 12 Klinov (1991). 13 for West Africa conference on migration in the Arab world in 1981 and later by the World Bank project on Manpower and International Labor Migration in the Middle East in 1983. Since then, numero'js national studies of emigration have become available, but no comprehensive data collection effort for the region is available for the 1980s and 1990s. A recent survey of the available evidence on migration patterns during the 1980s found that, despite the collapse of oil prices and the recession in the region, aggregate migration levels did not fall, although there was a redistribution of labor across the region."3 Many of the migrant workers were in essential sectors an4 were retained despite the collapse in oil prices. Remittance levels did not decrease and in many labor exporting countries, actually increased. The major change in the 1980s was the emergence of Iraq as a major labor importer. With much of its male labor force in arms during its war with Iran, Iraq became especially important as a destination for Egyptian workers. There were at least 1.25 million Egyptian workers alone in Iraq. This Iraqi demand was particularly timely as the growth of labor demand in the Gulf slowed after the oil price collapse in 1986 and there was increasing substitution of Asian for Arab labor. About one in every three Egyptian migrants went to Iraq in the 1980s. There was also some increase in demand for low skilled Egyptian workers from Jordan which was experiencing a construction boom from the remittances of its own higher skill migrants in Gulf. Fears that Asian labor would supplant Arab labor during the 1980s appear to have been unwarranted in the aggregate. Although Asian labor is often lower cost and less politically threatening to labor importing countries, the share of Arab labor has remained fairly stable since 13 Feiler (1991) provides a country-by-country survey of the available evidence on migration patterns during the 1980s. 14 1975 at about 55%. The explanation again lies in Iraq which tended to import mainly Arab labor, particularly Egyptians. This demand from Iraq appears to have offset the rise in the share of Asian labor in the Gulf. Political reasons as well as wage differentials have played an important role in determining both the level and composition of labor migration." The level of migrant workers relative to the indigenous workforce has long preoccupied the governments of the labor importing countries.` The increasing importance of Asian labor was clearly a response to fears about long-term Arab migrants demanding greater political rights. Moreover, shifting political alliances in the region have affected whether Egyptian, Iranian, Jordanian or Palestinian workers were welcome in the Gulf, Iraq or Libya. The Gulf war has also fundamentally altered migration patterns in the region. The most obvious consequence was the repatriation of hundreds of thousands of Arab workers from both Iraq and Kuwait as well as the repatriation of Yemenis from Saudi Arabia. The reconstruction of Kuwait is likely to require the restoration of migration flows, but Arabs, particularly Palestinians, are likely to be a smaller proportion of the Kuwaiti workforce. In the case of Iraq, ,.re-war migration levels are unlikely to be restored anytime in the near future, and the Egypian govemment has begun to look toward Libya to absorb some of the country's surplus labor. '4 For a survey of the role of politics in migration policy, see Feiler (1991) and Russell (1989). 1S The proportion of foreigners in the labor force during the 1980s was about 80% in Kuwait, 60% in Saudi Arabia, 70% in the United Arab Emirates, and 80% in Bahrain. Feiler (1991) 15 Why this gattern of integmration in the Middle East? The pattern of regional integration that exists in the Middle East is fairly unique. In most parts of the world, extensive trade in goods acts as the engine for regional integration. This was certainly the case with the European Community, East Asia and the North American Free Trade Agreement. The extensive movement of goods across borders increases the benefits of coordinated policies on tariffs and non-tariff barriers as well as standards and other policies that govern economic relationships. Labor movement is usually the final, and often the most controversial, feature of regional integration. In the Middle East, labor flows, and the remittances of capital associated with migration, have been the most important feature of regional integration. The explanation lies in the extreme differences in factor endowments across the region and, perhaps more importantly, the development policies adopted by both the labor importing and exporting countries. The distinction between tradable and nontradable goods and services is crucial in explaining the role of factor endowments. Many of the oil exporting, capital surplus economies are characterized by structural labor shortages. In the case of tradable goods, these shortages are not problematic because local demand can be met through imports from world markets. In the case of nontradable goods and services, such as construction, education, health, government and domestic services, there is no alternative but to import labor if local demand is to be met.16 Thus it is not surprising that the vast majority of migrant workers in the oil exporting countries 16 The absence of strong unions in the oil exporting countries also meant that the importation of labor did not face opposition from local interests. This is in contrast to the role of unions in opposing the migration of low cost labor in the North American Free Trade Agreement and in the European Community. 16 are employed in the nontradable sectors of the economy. But why were the oil exporters not importing tradable goods from their neighbors? The explanation lies in the trade orientation of the region's regimes described above. In general, the oil exporters adopted very outward oriented trade policies in order to meet local demand for tradable goods through the world market. In contrast, the labor surplus economies that could have been meeting the regional demand for food and manufactures adopted inward-oriented import substitution policies that discouraged the production of tradable goods.'" Because these import substitution policies tended to favor capital-intensive production (through interest rate subsidies, favorable tariffs on capital goods imports, overvalued exchange rates and skewed public investment programs), unemployment and underemployment were persistent problems. Thus the migration of labor became a convenient mechanism for labor surplus economies to export their unemployment problems. Remittances were also important for the balance of payments, which was the Achilles heel of import substitution strategies that produced little that would generate foreign exchange from export markets. Increased regional trade in goods was also undermined by the selective protection policies that made some tradable sectors de facto nontradables. Few regional migrants worked in the tradable sectors, but where they did, there were substantiai inefficiencies. The case of Saudi Arabian agriculture, which was highly subsidized through a system of input subsidies and price support to meet food self-sufficiency goals, provides an example of the disincentives to trade. The incentives provided by the Saudi government ensured that it was more advantageous for 17 The exception here is Turkey which liberalized its trade regime relatively early and was able to benefit from exports to the Gulf during the oil boom. 17 landowners to import labor to tend wheat fields in Saudi Arabia than it was to import Egyptian wheat. This is despite the fact that Saudi Arabian wheat was produced at five times the world market price and that wheat was sold domestically for less tnan the world price for much of the period. Of course because of the low procurement prices offered to farmers in Egypt, there was no wheat for the country to export. The beneficiaries of Egypt's food pricing policies were urban consumers who were more likely to threaten the regime than a dispersed and disorganized peasantry. This convoluted set of incentives ensured that Egyptian farmers continued to migrate to produce, often less efficiently, in Saudi Arabia or in Iraq. Those who received rents from the status quo - urban wheat consumers in Egypt and rural land holders in Saudi Arabia - would oppose any moves to achieve a more economically rational distribution of production which rtiight also result in greater regional trade. In contrast, the obstacles to labor mobility were far less than those governing trade in goods. Most of the labor surplus economies in the region actively encouraged migration through a variety of mechanisms. In Egypt, emigration became a constitutional right in 1971, exit visas were abolished in 1973, and a Ministry of Emigrant Affairs was established to address the needs of Egyptians abroad. The officially tolerated "own exchange' market provided a channel through which remittances could enter the country at the parallel exchange rate. The government also exempted migrants from paying taxes on income earned abroad and abolished a law requiring migrants to transfer a minimum of 10% of earnings to Egypt at the over valued official exchange rate." In Jordan, the government allowed migrants to postpone their military 1Ibrahim, 1982. 18 service until the age of 37 if they obtained a work permit from another country." The policies of the labor importers were intended to reduce the long run dependence on migrant labor, but were generally unsuccessful. A number of policies have been put into place to reduce the dependence on foreign workers -- including the use of more capital intensive technology, expenditures on training, requirements that nationals be in senior positions in all sectors of the economy, and, in some cases, encouraging women to enter the workforce. Despite these efforts, foreigners still constitute the vast majority of the labor force in these economies. Recognizing the limits of indigenous labor substitudon, the oil economies also severely restrict the duration of migrants stay through visas, work contracts and other policies that prevent foreign workers from becoming permanent residents. These policies also insured that migrants did not become eligible for the benefits, such as housing, education and health care, associated with nationality. ' Feiler (1991). Table . In6ra-rfeg&onaj ALrb capital lowsfficia and grivate. 1973-87 a. Total gfficial Arab assistance as a er,cenfage of GNP of tArab aid recipien countres. 197J-87 l73 1974 1975 1974 177 197* 17 1t5 19"1 1953 19. 195 i$ 195 owl Cdmii A,b MMI) Ent 3.3 5.3 4.7 5.3 4.7 3.9 6.1 5.6 4.8 3.1 2.7 2.1 1.5 3.7 1.1 3.3 bakumin 6.1 20.1 20.2 39.6 10.9 7.4 10.1 5.3 4.6 3.4 7.3 7.0 3.6 4.8 4.1 5.9 Imq 0.1 -0.0 0.2 4.0 4.0 0.1 40.0 .0.0 0.0 0.7 0.5 0.1 0.1 0.1 -0.0 0.1 Jesd. 133. 26.8 28.9 32.2 17.3 16.6 47.5 35.5 29.3 20.1 18.4 17.2 12.2 I0.5 8.2 20.6 1_bfiwa 0.1 3.9 0.3 0.5 2.1 5.2 1.9 4.9 9.6 3.1 0.3 -0.0 0.6 0.1 0.8 2.5 Oam* 1.2 10.7 3.4 6.1 6.1 2.1 6.1 4.9 3.4 1.3 0.6 0.7 0.7 0.9 -0.3 2.2 S3ri 8.9 11.0 9.6 6.3 9.5 8.1 16.1 32.1 9.7 5.6 4.6 3.3 3.6 3.5 2.5 6.7 Yam AabRepeb3 3.2 12.8 IS.9 17.1 12.8 10.6 6.1 10.4 .8. 5.3 4.S 4.0 3.5 2.6 1.a 6.2 Ya. ?.D.R. 4.3 9.2 35.2 35.6 22.2 9.5 4.6 9.1 3.9 14.1 3.5 5.3 4.9 0.1 4.7 8.1 Arb Afda 2.8 4.6 6.9 4.7 S.0 3.3 1.6 1.5 1.4 0.9 0 7 0.3 0.8 0.3 0.2 1.1 A34 0.8 0.1 0.7 0.1 0.9 0.2 0.1 0.1 0.3 4.5 0.0 40.1 0.Q .0.0 4.0 0.1 Earn 7.4 14.5 22.6 11.9 I3.5 9.9 1.2 0.0 *0.3 .0. 1 * 0.3 -0.1 *0. 0.2 0.2 3.0 bMwA&m 30.7 31.1 4.1 40<. 21.1 10.1 13.2 22.7 35.4 16.7 10.1 9.9 10.9 IA -0.2 4.6 mocoa 0.0 0.2 1.2 1.2 Si. 2.3 I.8 3.8 6.4 4.1 1.1 0.6 4.6 0.6 *0.2 2.4 SOMA& 3.0 13.9 34.7 .4 13.2 30.S 9.2 8.7 3.2 6.9 3.8 0.6 1.3 *0.2 0.3 4.8 Sun Q0.6 6.9 4.6 6.9 3.2 2.2 5.5 4.4 2.4 2.8 S.9 1.3 3.S 2.0 1.6 I.4 TA%6 0.2 0.6 1.4 1.6 1.9 0.6 1.6 1.2 0.8 0.6 0.2 0.9 0.5 0.0 0.7 0.8 Toel Aidpleoc. 2.9 4.9 6.0 4.9 4.9 3.6 I.S 3.2 2.8 1.9 1.6 1.1 1.2 0.9 0.6 2.3 ,S wede Sooade1990. b. PrIi'e uneauited transfers as a percwtwae of GP of Arab aid rei ,en countries. 19737-8 1973 1974 1I7S *74 1#977 s9on 1979 195 91 1983 1953 1954 1 9S 986 fps? Cw Ve Aub Mbtd EA 0.9 0.9 0.4 3.1 3.8 4.5 4.0 3.S 2.9 2.5 2.S 2.2 1.4 1.1 1.1 2.3 hbaum 0.0 0.0 -53A -14.3 .13.1 .14.4 -9.2 -1.2 *3.2 -3.4 -2.9 -3.6 -6.6 -9.2 -7.2 -6.3 3mq 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Jhda S.2 k0.4 16.6 30.3 26.6 22.5 20.1 20.3 25.9 24.7 23.6 27.7 21.9 22.3 15.9 22.3 Lab.... 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 00 am" 0.0 -9.2 -12.2 -10.3 -9.4 -8. .7.5 -6.8 .7.0 .8.1 -9.9 -I.5 -10.2 311.2 -9.3 9 4 ,Syd 1.1 t0. 0.5 0.7 1.2 6.8 8.2 5.9 4.1 2.7 2.4 2.0 1.9 1.3 1.0 2.8 Yea Anb Iapab3i 0.0 U8.0 28.7 54.6 60.3 37.1 30.5 30.1 19.8 IS.6 20.5 19.7 16.S 11.9 35.3 22. Yams. P.D.R. It.9 12.9 19.0 28.9 35.3 4J.7 42.9 42.8 44.5 47.0 45.9 44.4 40.S 32.1 32.2 38.5 ArabAbio 2.5 2.7 3.1 3.7 3.4 4.6 4.7 4.6 4.1 4.0 S.l S. 4.3 4.1 4.7 4.3 Agcb 3.7 2.4 2.3 2.2 1.3 1.1 1.0 0.7 0.? 0.8 0.5 O4 0.7 1.3 0.8 1.0 lE 1.3 3.5 4.1 6.5 7.0 12.9 13.3 13.0 10.3 10.4 14.0 13.9 10.2 1.9 11.0 10.2 AMa -5.5 .3.6 4.9 -5.2 4.1 *3.4 S.0 4.0 .2.4 -3.7 -3.4 -2.8 -2.9 .2.7 -2.2 -3.S Uowe= 3.4 3.9 5.4 5.S S.0 S.4 S.8 5.8 7.0 5.9 7.0 ?.S 8.7 9.9 9.9 6.7 SOmA6 0.6 0.7 0.3 0.2 0.2 6.9 2.8 2.9 3.8 o.s 0.9 5.1 0.9 1 4 2.4 2.1 So" 0.3 0.1 0.0 0.7 0.6 0.9 1.3 2.3 2.6 1.2 3.1 2.8 3.2 0.9 1.2 3.6 T" l3.5 3.1 3.1 2.9 3.1 3.5 3.9 3.5 4.1 4.6 4.4 3.9 3.3 4.2 5.2 J.9 ToelAidRc3mai 2.0 2.0 2.1 3.5 3.5 4.6 4.4 4.1 3.6 3.3 40 3.8 3.1 29 3.2 3.5 Sowm93f5 4caeO Dboow ( ). 20 Labor Mgriog n - A Ste&Mny Stone or a S_bae_ute fir Reginal InteSaal Because the obstacles to trade in goods have been greater than the obstacles to labor movements in the Middle East, labor has been the first, and most successful, element of regional economic integration. Labor mobility and its associated capital flows has been the most important mechanism through which the benefits of the oil windfall have been spread to the poorer states of the region. Labor migration has not been a substitute for regional trade where nontradable goods are concerned because there are few alternatives for meeting demand. In the case of tradable goods, where far fewer migrants are employed, the role of labor migration has been more complex. Those few migrants employed in the tradable sectors could have been more efficient producing at home and exporting to the oil economies. This was especially the case where there are strong externalities associated with domestic production, many of which are not exploited in the oil economies where labor turnover is rapid-' But few of the labor exporting countries were characterized by policy regimes that would have produced such tradable goods, so the migrant workers were generally better off being employed abroad than being unemployed at home. Is this pattern of integration in the Arab world desirable? It is necessary to distinguish between private and social interests. Migration obviously benefits the individuals involved - migrants earn higher wages and their employers benefit from access to low cost labor whose taining costs they usually do not incur. At the level of society, the assessment is necessrly more complex. Labor importing countries benefit from the production of migrants but they incur costs in terms of political and social stability as well as questions about the long run sustainability of their dependence on foreign workers. Labor exporters benefit from less 20 There is a growing consensus that there are substantial external benefits associated with productive human capital. See Romer (1986) and Lucas (1988). The impact of integration on such productive extemalities is explored by Rivera-Batiz and Romer (1991). 21 unemployment and from remittances, some of which is invested in the home economy2", but often suffer from selective skill shortages and the loss of the external benefits associated with having workers producing domestically. In some countries migration has had damaging effects - -- such as in Sudan where much of the scarce human capital of that country is abroad. But for countries such as Egypt, Jordan and Yemen, migration was, given the policy regime and the large labor surplus, a benefit to society. In general labor migration has not been a substitute for greater regional trade because it has been concentrated in nontradable sectors. Migration has also been a stepping stone to greater regional integration. Labor movements have provided a mutually beneficial mechanism for sharing the oil wealth across the region while taking advantage of underutilized human resources. The evidence on income distribution indicates that, contrary to the popular perception, incomes across the Middle East have become more equal. Figure 1 shows Lorenz curves for the region for 1970, 1981 and 1989 that irdicate that the distribution of income across the Middle East has moved toward the 45 degree line of equality. This tendency toward convergence of per capita incomes has held in years of oil booms (1981) and in periods of low oil prices (1970) and moderate oil prices (1989). The Lorenz curves say nothing about income distribution within countries, about which the data is very poor in the Middle East. Nevertheless, migration has obviously played an important and effective role in spreading the region's wealth across countries. But migration may not be the most desirable stepping stone. The macroeconomic policies of the Middle Eastern countries reduced the scope for greater integration of trade and investment flows. Greater trade in goods may be especially important for taking advantage of dynamic gains from greater competidon and learning by doing. Migration may also be a weak stepping 21 For evidence on the investment of remittances by migrant laborers, see Adams (1991). 22 stone, especially when the political sensitivities of the oil economies are considered as well as the substantial scope for substituting Arab labor with other nationalities. It seems clear that, without efforts to solidify regional economic ties on the basis of efficiency and mutual self- iiiterest, labor migration will remain an anomaly in an otherwise fragmented region. 23 References Adams, R. (1991) "The Economic Uses and Impact of International Remittances in Rural Egypt," Economic Development and Culttural Change, volume 39, number 4, July. Feiler, G. (1991), "Migration and Recession: Arab Labor Mobility in the Middle East, 1982- 89," Population and Development Review, volume 17, number 1, March. Ghantus, E. (1982), Arab Industrial Integration: A Strategy for Development, London: Croom Helm. Haseeb, K and S. Makdisi (1981), Arab Monetary Inte rration, London: Croom Helm. Ibrahim, S. (1982) in Rich and Poor States in the Middle East, edited by M. Kerr and S. El-Yassin, Boulder: Westview Press and Cairo: AUC Press. International Monetary Fund (1991), Direction of Trade Statistics, Washington, D.C.: International Monetary Fund. Isfahani, H. (1990), 'The Experience of Foreign Investment in Egypt under Infitah," Faculty Working paper number 90-1710, Bureau of Economic and Business Research, University of Illinois, Urbana-Champaign. Kilnov, R., (1991) "Recent Trends in Migration-for-Work from Middle Eastern Countries," paper presented at the Conference on "The Economics of Labor Mobility in the Middle East," Kennedy School of Government, Harvard University, Cambridge, ahusets, February 7-8, 1992. Lachler, U. (1989), "Regional Integration and Economic Development," Industry and Energy Department Working Paper number 14, The World Bank Washington, D.C. Lucas, R. (1988) "On the Mechanics of Economic Development,' journal of Monetary Economics 22. Makdisi, S. (1979), "Arab Economic Cooperation: Implications for the Arab World and World Economies in R. Albioni, editor, Arab Industrialization and Economic Integration, New York: St. Martin's Press. Pritchett, L. (1991) "Measuring Outward Orientation in Developing Countries: Can it be Done?" PRE Working Paper number 566, The World Bank, Washington, D.C. Rivera-Batiz, L. and P. Romer (1991), "Economic Integration and Endogenous Growth," Quarterly Journal of Economics, May. Romer, P. (1986) "Increasing Returns and Long-Run Growth," Journal of olitical Economy, 94:5. 24 Ravallion, Chen and Datt (1992), "New Estimates of Poverty in the Developing World," mimeograph, World Bank. Russell, S.S. (1989), "Politics and Ideology in Migration Policy Formulation: The Case of Kuwait," International Migration Review, volume XXII, number 1, spring. Sayigh, Y. (1982) The Arab Economy: Past Performance and Future Prospects, Oxford: Oxford University Press. Serageldin, I., J. Socknat, S. Birks, B. Li, and C. Sinclair (1983), Manpower and Intenational Labor Migration in the Middle East and North Africa, Washington, D.C.: The World Bank. van den Boogaerde, P. (1990), "The Composition and Distribution of Financial Assistance from Arab Countries and Arab Regional Institutions," Working Paper 90/67, Intemational Monetary Fund, Washington, D.C. World Bank (1992) Global Economic Prospects, Washington, D.C.: World Bank. World Bank, (1992) World Development Report, Washington, D.C.: World Bank.