Middle Eastern Studies, Vol. 42, No. 3, 409 ­ 421, May 2006

UNRWA and Moral Hazard
FRED GOTTHEIL
There's a sobering adage that reads `No good deed goes unpunished.' Like most adages, it contains both a kernel of truth and more than a kernel of exaggeration. Among the good deeds that invite punishment are those that, although inspired by good purpose, quickly and unexpectedly provoke perverse behaviour on the part of intended beneficiaries that, in the end, not only sabotages the deed but inflicts punishment on its architects and possibly on third parties. Examples of `no good deed goes unpunished' abound in the worlds of politics, law, and economics. The punishment outcomes are described in the economic literature as moral hazard.1 To economists, moral hazard stems from the willingness of individuals, governments or non-governmental organizations to engage in activities more hazardous than they would otherwise undertake because they are assured that other individuals, governments or non-governmental organizations would assume the potentially negative consequences that may flow from their more hazardous activities. International lending institutions such as the IMF and the World Bank provide classic examples of moral hazard generation. They offer loans at favourable rates of interest to developing country governments who know, by practice, that accountability is lax and liability can be shifted to others. Reliance on such shifting excites perverse behaviour on the part of the borrowing governments. In many cases, little attention is paid by them to the worthiness of their investment projects. Hazardous ventures are undertaken that result, not unexpectedly, in failure. And expectedly, the costs associated with the borrowers' inability to meet their obligations are borne not by the borrowers but by the lending institutions. In other words, as critics of the IMF and the World Bank see it, `no good deed goes unpunished'.2 While the economic circumstances associated with IMF-related or World Bankrelated moral hazard may seem far removed from any circumstance associated with Middle East refugees, the fact of the matter is that they have much in common. UN, and specifically UNRWA (United Nations Relief and Works Agency) policy with respect to Middle East refugees generated moral hazard outcomes for over five decades that are structurally similar to those generated by the international lending institutions. Their moral hazard similarities are uncanny. Both trigger perverse incentives and behaviour that sabotage hoped for outcomes. Both generate parasitic stakeholders that add layers to moral hazard. Many of the intended beneficiaries of UNRWA policy ­ the Middle East refugees ­ end up as ancillary recipients while other agents ­ UNRWA staff, Middle East governments, other governments, and NGOs ­ shape and perpetuate UNRWA policy to their advantage. All this occurs at other people's expense.
ISSN 0026-3206 Print/1743-7881 Online/06/030409-13 ª 2006 Taylor & Francis DOI: 10.1080/00263200500521263

410

F. Gottheil

Ray Wilkinson notes in the UN journal Refugees: `Mass flight is nothing new. From the dawn of history, entire populations have been periodically forced to flee their homes and their countries during times of conflict.'3 Of course he states the obvious. In the twentieth century, as in centuries before, world wars, regional wars, and civil wars triggered massive human displacements. Virtually no community, country, or continent was immune. Hundred of millions of people assumed refugee status; many were subsequently repatriated and still many more chose or were forced to resettle elsewhere. In the 1940s alone, three unrelated conflicts generated approximately 40 million refugees. How these 1940s refugees came into being, were assisted, and their refugee status finally resolved varied considerably and underscores the uniqueness of the Middle East refugee experience. It also explains why moral hazard emerged in the case of Middle East refugees and not in others. During the Second World War, tens of millions of Europeans were uprooted from their homes and homelands. The UN's International Refugee Organization (IRO), the refugee agency created in 1947 to succeed the United Nations Relief and Rehabilitation Administration (UNRRA), was assigned the task of facilitating their repatriation and resettlement. Refugee status was understood to be a temporary state.4 It was in the interest of the refugees, of the European governments, and of the IRO that the move toward refugee resolution was made `with all deliberate speed'. People simply wanted to get on with their lives. By force of circumstance, moral hazard never became an issue. Tens of millions of other refugees were created during a multiplicity of regional and civil wars that erupted in Asia, Africa, Latin America, and the Middle East in the 1940s and early 1950s. The most notable among them were the estimated 12 to 20 million refugees who, following the partition of India in 1947 fled east and west to and from India and newly-created Pakistan. Despite world expressions of concern, the International Committee of the Red Cross's (ICRC) appeal to the world community evoked no more than a paltry response. The expiration of their refugee status followed a difficult, but natural course of refugee absorption. Here, too, people simply wanted to get on with their lives. These two refugee-creating events, among others in the 1940s, set the backdrop to an analysis of the refugee flow that followed the partitioning of Palestine in 1948. Like the India­Pakistan partition, the Palestinian one erupted in conflict and triggered substantial dislocations of populations. Palestinian Arabs fled from homes and communities within newly created Israel to other parts of Palestine and to adjacent Arab states while Jews fled from their homes and communities within Palestine and from Arab states to Israel. While the causes that precipitated the flow of Arab refugees and their numbers are still hotly debated issues, the fact that over 600,000 Arabs fled Israel and a somewhat smaller number of Jews fled their homes is generally accepted.5 What was radically different in this Palestinian case was the creation of an international refugee agency focused on a single-country ­ UNRWA (United Nations Relief and Works Agency) ­ to facilitate the resolution of the refugee problem.6 In truth, these several hundred thousand Palestinian Arab refugees were no different in character and circumstance from the millions of other world refugees torn in conflict from their homes and homelands. Why, then, was an agency created specifically for Palestinian Arabs while all the other refugees in the world were placed

UNRWA and Moral Hazard

411

under the trust of the United Nations High Commissioner for Refugees, UNHCR? One reason, suggested by Guy S. Goodwin-Gill and others, was the interjection of Middle East politics. UNHCR was essentially non-political, designed strictly to assist refugees as refugees. Middle East governments and Palestinian elites, on the other hand, had a very distinct political agenda for Palestinian refugees.7 With hindsight, had UNRWA been folded into the wider-ranging UNHCR during UNRWA's formative years and Arab Palestinian refugees subjected to the same UNHCR rules and guidelines that were applied by it to all other refugees, the process toward resolution of Arab Palestinian refugee displacement would probably have taken a different course. And perhaps the creation of what was to become UNRWA-based moral hazard might never have arisen.

That UNRWA was designed to be temporary was clearly spelled out in its 1951 Report.8 It expected to withdraw from intervening directly in the lives of Palestinian refugees within a few years after its birth.9 Noteworthy, it appreciated the hazardous outcomes that could obtain with continuing direct involvement. In the first article of its findings, the Report warns that a potential for moral hazard existed unless refugee resolution is pursued with all deliberate speed. In its own words: `There must be a firm goal of terminating relief operations. Sustained relief operations inevitably contain the germ of human deterioration.'10 This point is reemphasized in the second article of the findings: `There is now considerable agreement among governments that refugees cannot continue indefinitely in their present conditions.'11 While UNRWA set no definitive time frame to the process of repatriation and resettlement, it placed ultimate ownership of the Palestinian refugee solution on Middle East governments. It saw its own role as being twofold: to address the immediate needs of the refugees during its short-lived tenure and during that tenure to arrange for the shift of responsibility from itself to Arab governments. Again, in its words: `The Agency, as rapidly as feasible, should move out of operations into the role of financial and technical assistance to sovereign governments.'12 UNRWA's 1951 Report detailed the particulars of a three-year refugee-support programme that would, if not put an end to the problem, then at least provide a `road map' to its resolution. Among the UNRWA goals were: `an end to refugee camp life', the granting to refugees `adequate rights of citizenship and work within individual countries', and an effort to `facilitate their freedom of movement among countries'.13 While not discouraging repatriation ­ in fact the Report specifically noted that `repatriation and compensation must not be prejudiced by any Agency programme',14 the Report nonetheless focused its attention on what it clearly understood as doable: resettlement. It focused on resettlement largely because UNRWA's primary focus was the well-being of Arab Palestinian refugees. The language of the Report was clear. As with refugees and refugee agencies elsewhere, UNRWA's view was that every effort should be made to encourage Palestinian refugees to get on with their lives. It didn't take UNRWA very long before it realized that it hadn't the authority, the muscle, or the will to resolve the refugee problem. The window of opportunity for resettlement closed quickly on UNRWA. At first, Egypt, Iraq, Lebanon, Syria,

412

F. Gottheil

Libya, and Jordan expressed some willingness to absorb some of the refugees. But nothing of substance materialized. UNRWA-generated large and small-scale works projects in the Arab countries were created for Palestinian refugees with the expectation that these short-run employment opportunities would, in the long run, anchor the refugees into resettlement. While cooperating with UNRWA and accepting these UNRWA-funded projects as part of its own development schemes, none of the host governments ­ Jordan excepted ­ was willing to accept refugees as a matter of policy. Their resistance to resettlement, at least from their point of view, was well reasoned. The 1949 armistice notwithstanding, Arab governments still did not accept Israel's legitimacy and to agree to resettlement as a resolution to the refugee problem would be tantamount to acknowledging the permanence of Israel.15 Self preservation was another factor. In the view of the 1954 U.S. Special Study Mission to the Near East: ` . . . any Arab political leader suggesting an alternative to repatriation in what was formerly Palestine would have been ousted from office and, perhaps, have run the risk of assassination.'16 Furthermore, some Arab governments feared that absorption of refugees could well undermine their own political stability. The Lebanese government, for example, believed that adding the large number of Palestinian refugees already in Lebanon to its citizen population base ­ most of whom were Sunni ­ would undermine Lebanon's delicate political sectarian balance.17 The Lebanese concern about internal security was not unique. Historian Benny Morris, commenting on the 1948­ 49 negotiations concerning repatriation and resettlement argued that the Arab states regarded the refugees as a potential Fifth Column.18 As far as Israel was concerned, its own agenda ruled out any sizeable repatriation. Its attention was focused on the absorption of refugees from Europe and the Middle East. Also, it understood as the Arab states did that the 1949 armistice was simply that: an armistice, not an end to the regional conflict. In its view, Israel's military had to contend not just with the armed forces of the Arab states but with contingents of Palestinian militia. The fact of the matter was that it saw its own existence then as quite precarious.19 Under these circumstances, anything amounting to repatriation in large numbers was out of the question.20 These facts on the ground left UNRWA completely checkmated. Its only viable function ­ albeit, an important one ­ was to provide direct relief in the form of food rations, shelter and social services. That is to say, the Palestinian refugees' refugee status was, by force of Middle East politics, frozen. What had been an agency designed to assist refugees back to normal life ­ as was UNHCR's role in every other of the world's refugee cases ­ became an agency denied that principle charge. The seeds of UNRWA-based moral hazard had been sown. Perhaps the most radical response UNRWA could have taken to alleviate its untenable position as a refugee agency was to allow its three-year mandate to expire.21 After all, denied authority to pursue either repatriation or resettlement, it had lost control over the essential purpose of its mission. By allowing its mandate to lapse, it would have served notice on the Arab governments that, whatever their national and regional political agendas, they had no choice but to cope with the problem of resolving the refugee condition, as other refugee-hosting countries had under UNHCR.22

UNRWA and Moral Hazard

413

Instead, UNRWA reinvented itself. It became strictly a caretaker agency. In its own words: `UNRWA and UNHCR are both UN agencies mandated by the international community to do specific jobs for refugee populations. UNRWA deals specifically with Palestinian refugees and their unique political situation. One reason for the distinction is that in the main the UNHCR is mandated to offer refugees three options, namely local integration, resettlement in third countries, or return to their home country. These are not feasible for Palestinian refugees as the first two options are unacceptable to the refugees and their host countries and the third is rejected by Israel.'23 This UNRWA distinction between itself and UNHCR is not entirely accurate. The government of Jordan offered to resettle its Palestinian refugees by granting them Jordanian citizenship. Many became citizens and many more, for a variety of reasons, declined the offer. The Jordanian exception aside, that UNRWA was compelled to deal with the `political situation', as UNRWA described it, was indeed unique. No other refugee agency ­ not UNHCR nor any of the other refugee agencies that preceded it ­ allowed itself to be ensnarled by and to be defined by the hosting governments' political agendas. UNRWA's reinvention as a caretaker agency also influenced the growth patterns of its refugee population and in this respect distinguished itself, once again, from UNHCR. The refugee population that UNHCR serves, at any time, is the number who fled their homelands minus those refugees repatriated or resettled. Because there was virtually no repatriation or resettlement among UNRWA's refugee population, its size includes not only those who fled their homes but also during the course of over a half-century and in considerably larger numbers their children, grandchildren, and great grandchildren, regardless of where and under what social, political, and economic conditions they live. Another distinction between UNRWA and UNHCR on population counts is this: Palestinians who had fled their homes from one location within Palestine to another location within Palestine ­ say, from a village in what became Israel to a location in the West Bank ­ are nonetheless defined by UNRWA as refugees, even though they had not fled their homeland. By UNHCR reckoning, they are not refugees.24 And counted as well among the Palestinian refugees are descendants of refugees born, raised, and living elsewhere in the Middle East and abroad, who, never having seen the Palestinian homeland, are free nonetheless to return to it and to live there permanently but choose not to do so. Their decision to reject repatriation to the Palestinian homeland had nothing to do with the principles of non-refoulment since persecution of returnees was at no time a perceived threat.25 They do not satisfy UNHCR's definition of refugee. Like almost every other people, Arab Palestinians live just about everywhere. Approximately a quarter million found their way to the American continents, and about the same number resettled in the European Union. Several thousand made Australia their home. Yet the overwhelming majority of the 9.7 million Palestinians in 2003 still lived in the Middle East, 50 per cent of them within the pre-1948 boundaries of British Mandatory Palestine. The controversy over initial estimates of Palestinian refugees and of the criteria used to assign refugee status notwithstanding, the data UNRWA accepts and admits as registered Arab Palestinian refugees are shown in Tables 1 and 2.

414

F. Gottheil

The totals of Table 1 include Arab Palestinian refugees living in the Palestinian homeland (excluding the million Arab Palestinians living in Israel who are Israeli citizens) who account for close to 40 per cent of the total refugee population.26 The remaining 60 per cent under UNRWA auspices reside in Jordan, Lebanon, and Syria. Approximately a third of all the refugees live in refugee camps administered by UNRWA, the remaining two thirds live in and around major cities ­ e.g., Damascus, Beirut, Amman, Jericho ­ and are also recipients of UNRWA support. This refugee population increased by over 450 per cent ­ averaging more than 3 per cent per year ­ during the 1953­2002 period of Table 1. During the 12 years, 1990­2002, the population growth increased by an exceptional average annual rate of 4.1 per cent. It is noteworthy that these additions to the refugee population were offspring of refugees who themselves were at least a generation removed from those actually displaced from their homes. See Table 2. A rather striking statistic is the 24 per cent of all UNRWA registered refugees that were born since 1985. The oldest of these refugees was born 36 years after the
Table 1. Registered Palestine refugees in camps and as a percentage of the total registered refugees, 1953­2002 Total refugee population 870,158 1,136,487 1,445,022 1,863,162 2,466,516 3,737,494 3,973,360 Refugee population in camps 300,785 409,223 500,985 613,149 697,709 1,211,480 1,262,867 Refugee population in camps (%) 34.6 36.0 34.7 32.9 28.3 32.4 31.8 Total refugee population living in Palestinian homeland (%) * * 41.0 37.5 37.6 37.7 37.7

Year 1953 1960 1970 1980 1990 2000 2002

Source: Adapted from Report of the Commissioner General of the United Nations Relief and Works Agency for Palestine Refugees in the Near East, General Assembly Official Records, United Nations, New York, various years. *Until 1967, the West bank of Jordan was administered as an integral part of the Jordan field.

Table 2. Registered Palestine refugees by age group: 2000 Refugee population Born Born Born Born Born Born after 1996 after 1985 after 1975 after 1965 after 1955 pre-1955 481,873 1,363,818 2,072,674 2,645,210 3,022,434 715,060 Total refugee population, % 13 24 56 71 81 19

Source: Adapted from Report of the Commissioner General of the United Nations Relief and Works Agency for Palestine Refugees in the Near East, General Assembly Official Records, United Nations, New York, various years. *Until 1967, the West bank of Jordan was administered as an integral part of the Jordan field.

UNRWA and Moral Hazard

415

partition of Palestine and the creation of UNRWA, and 33 years after UNRWA had expected its mandate to have been fulfilled. The percentage of all registered refugees born since 1975 increases to 56 per cent of the total. Of these registered refugees, 40 per cent are still living in the Palestinian homeland. These characteristics of the UNRWA-registered Arab Palestinian refugee population are the unavoidable consequence of the unmanageable circumstances that befell UNRWA. They also contribute to ­ and explain ­ the inevitability of UNRWA-based moral hazard. To be strictly a caretaker refugee agency for a specific group of refugees for five years is one thing. To be strictly a caretaker refugee agency for the same group of refugees for over 50 years is quite another matter. In the former case, the refugees recognize that the agency's function is to provide them assistance during their critical period of transition. In the latter case, the refugees come to regard the agency's assistance as a form of permanent entitlement. UNRWA's moral hazard problem, it appears, originated precisely because it was made to accept ­ and accepted ­ the role of providing long-term permanent entitlements to Arab Palestinian refugees. The result was the creation of a perverse set of incentives among refugees that discouraged many from pursuing viable options to their long-term refugee status. It also encouraged many non-refugees in the region to attempt to register for refugee status or at least to take advantage of the entitlements UNRWA offered. Finally, UNRWA's half-century tenure as a caretaker agency helped create a relatively large and influential bureaucracy that, as stakeholders in the provision of entitlements, pursued self-serving agendas that tended to perpetuate the Palestinian refugee condition rather than its resolution. Consider first the moral hazard outcomes that are associated with UNRWAregistered refugees who choose to live in refugee camps. Over the course of their 50-year residence, housing within the camp areas radically improved from canvas tents to permanent structured housing that compared not unfavourably with housing afforded by many non-refugee Arabs living in the host countries.27 The major difference between housing in refugee camps and housing for many nonrefugees in the host countries is that approximately 70 per cent of the refugees living in UNRWA camps owned their own homes,28 and those who didn't paid no rent, no municipal taxes, and had access to free water and sanitation services. The refugees' access to goods that reflect levels of material comfort is shown in Table 3. Registered refugees ­ essentially refugee residing in camps ­ own these `comfort goods' in surprisingly high percentages that are similar to the percentages owned by
Table 3. Selected durable goods owned, by refugee status, 1997, % Private car Non-refugee Registered refugee Non-registered refugee 22.1 17.8 21.7 Refrigerator 80.8 80.7 75.6 Washing machine 72.3 75.5 65.6

Source: L.B. Jacobsen, Finding Means: UNRWA's Financial Crisis and Refugee Living Conditions, FAFO Report 427, Vol.1, 2003, p.73.

416

F. Gottheil

non-camp refugees. UNRWA, although not the sole provider of healthcare within the camps, provided still a variety of vital medical services, among them prenatal care, medicine, and vaccination.29 According to conventional health indicators, the outcome for Palestinian refugees was good and considerably better than most found in developing countries.30 But the most enriching form of entitlement offered to registered refugees by UNRWA was education. Elementary and secondary education in camps generated literacy rates ­ 80 per cent for men and 72 per cent for women ­ that were even higher than rates achieved by non-refugees. In fact, rates for Palestinian refugees in Jordan turned out to be more akin to those in Southern Europe than in the Middle East.31 Without prejudice, consider what kind of incentive schemes these UNRWAsupplied amenities created for refugees. How powerful are the incentives to move out of the refugee camps when real income there ­ employment income earned either in the camp economy or beyond plus UNRWA entitlements and transfer payments from family members working elsewhere ­ may match or even exceed the real incomes earned by large subsets of the host countries' populations? Herein lay the seeds of moral hazard. UNRWA's benefactors ended up absorbing the approximately $250 million annual cost ­ for over 50 years ­ of financing entitlements that, having once served to assist displaced refugees, now serve to perpetuate through these entitlements a strong disincentive for Palestinian refugees to shed their refugee camp status. And because repatriation and resettlement of refugees had been frozen for 50 years, the registered refugee camp population had mushroomed from 300,000 in 1953 to over 1.2 million in 2002 ­ by 4.6 per cent per year since 1990 ­ placing enormous pressure on the voluntary-financed UNRWA budgets and forcing UNRWA to seek year-after-year emergency funding from its same funding sources. That is to say, by the force of demography and by the persistence of perverse disincentives, the costs associated with UNRWA-based moral hazard grew year by year while the actual entitlements given to each of the increasing numbers of refugee camp families were necessarily reduced.32 Still, the fact that many non-Palestinians in host countries left their villages to find employment, housing, and hoped-for fraudulent registration as a refugee in a Palestinian refugee camp is evidence that the standard of living within the refugee camps was ­ at least compared to the standards enjoyed by some non-Palestinians in the host economies ­ anything but inferior.33 It also implied that to the extent that these non-refugees were successful in gaining access to UNRWA entitlements, another layer of moral hazard ­ based on another set of disincentives ­ ended up being borne by both UNRWA and its benefactors. The 50-year longevity of these moral hazards had been bolstered from the start by the Palestinian refugee leadership whose political agenda was to preserve the demographic strength of the camps.34 Arab academics professing to champion the cause of the Palestinian refugees pressed for the same outcome. Hassan Elnajjar, for example, argued that UNRWA played a subversive role by providing refugees with free education and vocational training. He elaborated: `Palestinian higher education has been observed to have its own disadvantages because of its relationship with emigration. First, it leads to the loss of the highly educated. Second, it depends on alien institutions to train the nation's human resources. Third, it is relevant to the Arab job market, not to the Palestinian needs.'35

UNRWA and Moral Hazard

417

Elnajjar's rather perverse conclusion that UNRWA-supplied education and vocational training are detrimental to the well being of refugees follows only because he ­ not the refugees who, given choice, chose otherwise ­ placed greater importance on the political viability of the refugee camp than on refugees securing for themselves a better economic future. He faults UNRWA for having `contributed to the dispersion of about one third of the refugees in the 1960s and 1970s.'36 Whatever effect UNRWA provision of free education had on the refugees' ability and ultimate decision to quit the camps, as Table 1 shows, approximately 70 per cent of UNRWA-registered refugees over the 50-year period 1953­2003 elected to live and work in the open economies of the Middle East. The vast majority of them opted for the non-Israel part of the Palestinian homeland, Jordan, Syria, and Lebanon, and entered those labour markets along with non-UNRWA-supported Palestinians, Jordanians, Syrians, and Lebanese. That they managed on their own to achieve a standard of living comparable to the standards enjoyed by others in the area is shown in Table 4. The Table 4 data show that Palestinian economic achievement has been comparatively attractive. With the exception of Jordanians, the Palestinian per capita income living and working in the Palestinian homeland exceeded those of non-refugee Arabs, and in some cases by substantial percentages. That is to say, allowing for sufficient time, economic self-interest ­ as opposed to political ideology ­ secured for the Palestinian refugees a relatively tolerable non-refugee standard of life. While UNRWA continues to confer refugee status on the 70 per cent of Palestinians refugees not living in refugee camps, these Palestinians have in fact and by their own choice reintegrated into non-refugee productive environments. Were it not, then, for the 30-plus per cent of refugees still in camps, the raison d'etre for an UNRWA would have long ceased to exist.37 But even the 1.2 million ^ refugees remaining in camps, as Tables 3 and 4 show, have acquired standards of
Table 4. Per capita income: Palestinian, Palestinian camp Refugees, Syrian, Jordanian, Lebanese, and Egyptian: 1990s Per capita income: US$ Jordan Lebanon Syria Egypt Palestine Homeland (Syrian Refugee Camps) (Lebanon Refugee Camps) (Jordan Refugee Camps) 1,755 1,500 790 715 1,684 (456) (794) (616) % of Palestinian per capita income 104.2 89.1 46.9 42.5 ­ ­ ­ ­

Source: Background Note: Jordan, US Bureau of Near Eastern Affairs, US Department of State, November, 2003, p.1; Background Notes: Lebanon, US Bureau of Public Affairs, US Department of State, January 1994, p.1; Background Notes: Egypt, U.S. Bureau of Public Affairs, U.S. Department of State, March, 1995, p.1; Statistical Abstract of the United States: 2002, U.S. Department of Commerce, Washington, D.C., p.833; Israel and The Occupied Territories, The Economist Intelligence Unit, U.K., 1996-7, pp.16 and 61. L.B. Jacobsen, Finding Means: UNRWA's Financial Crisis and Refugee Living Conditions, Fafo Report 427, Vol.1, p.148.

418

F. Gottheil

living that rival many of those in the open economies of the Middle East and continue to be conspicuously superior to the standards of living associated with the millions of UNHCR's refugee populations and even the hundreds of millions of non-refugee populations in Asia, Africa, and Latin America.38 So what explains the durability of UNRWA? Herein, again, lies moral hazard. UNRWA, as an entitlements-generating institution has, over its 50-year tenure, evolved into what Professor Abraham Ashkenasi has called a `state within a state.'39 That the 1948 or `original' Palestinian refugees have long since become a population of diminishing proportions and as such of diminishing concern for UNRWA, or that succeeding generations of UNRWA-defined refugees, to varying degrees, have acquired employment and incomes comparable to their non-Palestinian neighbours seems to be of no account to UNRWA. By all measures of refugee need ­ compared, for example, to the needs of the millions of refugees in UNHCR's domain ­ UNRWA had outlived its purpose. But its survival continues, nurtured and assured by a politically adept set of stakeholders ­ its area staff, its international staff, NGOs, Middle East governments, and others. And this, too, translates into moral hazard. Consider this: in 2001, UNHCR assisted 20 million refugees, UNRWA 4 million. Yet UNHCR's staff totalled 5,000 compared to UNRWA's 23,000 staff. That is to say, UNHCR's refugee-staff ratio was 4,000 compared to UNRWA's 174. What is significant about 50 years of UNRWA is not that it was a refugee agency that served the Arab Palestinian refugee population with much affect, but that it continues to do so despite the fact that the majority of Palestinians have reintegrated into the open economies of the Middle East and elsewhere de facto, and that most of those who still remain in refugee camps ­ after 50 years ­ do so in the Palestinian homeland. By all accounts, the refugee status of the overwhelming numbers of Palestinian refugees should have expired somewhere along that 50-year range. But it continues. And therein lies the essence of its moral hazard. UNRWA was reinvented to serve political agendas unrelated to its initial and honourable mission. Forced to abandon the pursuit of assisting refugees to get on with their lives ­ repatriation or resettlement ­ it became strictly a caretaker agency, dispensing entitlements to refugees who, by UNHCR standards, would not be so defined. All this at enormous cost. Its over $250 million annual budgets represent, minimally, a continuing moral hazard. Even more so is the moral hazard associated with the set of disincentives built into UNRWA ­ political and monetary ­ that discourages refugees from seeking economic betterment. In the end, UNRWA cannot accomplish what it set out to do and is blamed for and must pay for what it ends up doing. As the adage reads: `No good deed goes unpunished.' Notes
1. Moral hazard is defined in the Encyclopedic Dictionary of Economics as `an effect of economic institutions arranged so that individuals have an incentive to maximize consumption at a social cost to others because they do not bear the full cost of their consumption.' (Guilford, CT: The Dushkin Publishing Group, Inc.), 4th edn., 1991), p.164. Example: The financial security afforded by fire insurance to the insured may affect the insured's behaviour and incentives. Vigilance, a high personal priority before insurance, loses some of its value with insurance. Proper home maintenance, for example, becomes compromised because the insured is protected against the possibility of major financial loss.

UNRWA and Moral Hazard

419

2.

3. 4. 5.

6.

7.

8.

9. 10. 11. 12.

13. 14.

15.

16.

The consequences are perverse. Although providing financial security to the insured, fire insurance may actually invite a higher incidence of fire damage. Example: In India, a programme paying people a bounty for each rat pelt handed in was intended to exterminate rats. Instead it led to rat farming. `All insurance policies, IMF or otherwise, entail some degree of moral hazard. That is in their nature.' A. Haldane and A. Taylor, `Moral Hazard: how does IMF Lending Affect Debtor and Creditor Incentives?' Financial Stability Review (June, 2003). See also B. Eichengreen, `Can the Moral Hazard caused by IMF Bailouts be Reduced?' Geneva Reports on the World Economy Special Report 1, (2000). `The Changing Face of Protection', Refugees, The UN Refugee Agency (UNHCR), Vol.3, No.132 (2003), p.6. IRO came to an end in 1952. It was superseded in 1951 by the United Nations High Commissioner for Refugees (UNHCR), a refugee agency that is neither country nor event specific. UNWRA's 1950­51 attempt at creating an accurate refugee count failed due to fraud, bribery, intimidation, and lack of cooperation from hosting Arab governments. The census data finally arrived at represented a political compromise between UNRWA and its Arab clients. See B. Schiff, Refugees Unto The Third Generation: UN Aid to Palestinians (Syracuse: Syracuse University Press, 1995), pp.22­5. In Nov. 1948, The UN established The United Nations Relief for Palestinian Refugees (UNRPR) to assist the refugees. UNRPR was replaced by United Nations Relief and Works Agency (UNRWA) in Dec. 1949. Goodwin-Gill writes: `The competence of the High Commissioner in the political issues surrounding the Palestinian question was once thought incompatible with the proclaimed non-political character of UNHCR's work' The Refugee in International Law (Oxford: Claredon Press, 2nd edn, 1996), p.91. Goodwin-Gill's view was shared by others. For example: `The stark truth is that it has been in the crude political interests of every state in the area to manipulate the status of the refugees in their own interests.' Situation of the Palestinian Refugees,' Report by the Committee on Migration, Refugees and Demography, (Rapporteur: Mr. Atkinson), Council of Europe, 1988, p.54. See also A. Ashkenasi, `The International Institutionalization of a Refugee Problem: The Palestinian and UNRWA,' The Jerusalem Journal of International Relations, Vol.12, No.1 (1990), pp.63­4. `Assistance to Palestine Refugees: Special Report of the Director and Advisory Commission of the United Nations Relief and Works Agency for Palestine Refugees in the Near East' (Paris: General Assembly, Sixth Session, Supplement No.16A (A/1905/Add.1, 1951). `Every effort should be made by the Agency and the governments to arrange for the transfer of relief administration to the governments no later than 1 July 1952.' Ibid., p.2. Ibid., p.2. My italics. Ibid., p.2. Article 10 of its findings, Ibid., p.2. My italics. UNRWA describes what it means by `as rapidly as possible'. In the first of its seven recommendations it states: `Every effort should be made by the Agency and the governments to arrange for the transfer of relief administration to the governments not later than 1 July 1952.' Ibid., p.2. Ibid., p.2. The Report mentions the possibilities of refugee absorption in Jordan and Egypt. Ibid., p.2. In December 1948 the UN General Assembly passed Resolution 194 (III). Paragraph 11 declares that `The refugees wishing to return to their homes and live in peace with their neighbors should be permitted to do so at the earliest practicable date, and that compensation should be paid for the property of those choosing not to return and for the loss of or damage to property which, under principles of international law or in equity, should be made good by the Governments or authorities responsible.' By mid 1950s, it was clear to Israel that `living in peace' was not on the Arab agenda and clear to the Arabs that repatriation was not an option Israel was prepared to consider. `[The Arab governments] regarded repatriation as the `just' solution and, incidentally, as one that could help undermine the Jewish State, to whose continued existence they all objected.' B. Morris, The Birth of the Palestinian Refugee Problem 1947­1949 (New York: Cambridge University Press, 1987), p.285. `The Arab Refugees and Other Problems in the Near East, Report of the Special Study Mission to the Near East (Washington: U.S. Government Printing Office, 1954), p.3. The subsequent assassinations of King Abdullah of TransJordan in 1951 who had expressed a willingness to accept the permanence of Israel and President Anwar Sadat of Egypt in 1981 who was the first among Arab leaders to sign a peace treaty with Israel indicated that the Mission's view was not entirely without substance.

420

F. Gottheil

17. United States Committee on Refugees, Country Report: Lebanon (2001), p.1. See also, `Identity Crisis: Palestinians in Post-War Lebanon', World Refugee Survey (1997), p.37. 18. Morris, op.cit., p.285. 19. In Alan Dowty's view: ` . . . it is hard to identify any significant voluntary repatriation that took place while the causes of the outflow remained unchanged; the problems in returning Tamils to Sri Lanka illustrate the point.' A. Dowty, `Return or Compensation: The Legal and Political Context of the Palestinian Refugee Issue', World Refugee Survey (1994), p.29. For a similar view with reference to UNHCR and the problems in the Great Lakes region, see J. Crisp, `Refugees and International Security: An Introduction to Some Key Issues and Policy Challenges,' Paper Prepared for the 4th International Security Forum (Geneva: November, 2000), p.8. 20. Reacting to U.S. diplomatic pressure, Israel ultimately proposed to accept 100,000 Arab refugees as part of an overall peace settlement. The idea was rejected by the Arab states. Another proposal Israel considered negotiable was the `Gaza Plan' which would have transferred sovereignty of the Gaza Strip from Egypt to Israel and transferred as well the approximately 250,000 Arab Palestinian refugees living in it. Egypt rejected the proposal. See Morris, particularly chapters 4, 5, and 9. 21. UNRWA has been functioning on a series of three-year mandates that has been repeatedly renewed by the UN General Assembly since 1949. Its current three-year mandate expired in June, 2005. 22. The precedents were clear. In Europe following World War 2, no large ethnic groupings of refugees had been repatriated. Following the partition of India, resettlement rather than repatriation of refugees was the norm. 23. `Setting The Record Straight.' UNRWA publication, http://www.un.org.unrwa/myths/index.html., p.4. My italics. 24. UNHCR distinguishes between `home' and `homeland,' restricting the designation `refugee' to people who had actually fled their homeland. Those who fled their homes to other parts of their homeland are defined by UNHCR as `internally displaced.' Referring to Eritea: `So long as Eritea remained an Ethopian territory, Eriteans residing in Ethopia that had fled from conflict-torn areas were not refugees because no internationally recognized border had been crossed.' `Case: Eritera, The 1993 Referendum on Independence from Ethiopia,' (IOM) International Organization for Migration (2003), p.2. This `home' and `homeland' distinction is applied as well to Palestinian refugees. Ziad Abu Zayyad, Rashid Khalidi, Salim Tamari, and Abbas Shiblaq distinguish between the `right of return to home' in principle but not in practice from the `right of return to homeland' in both principle and practice. See Z. Abu Zayyad, `The Palestinian Right of Return: A Realistic Approach,' Palestine­ Israel Journal 2 (1994), p.77; R. Khalidi, `Toward a Solution', Palestinian Refugees: Their Problem and Future (Washington: Center for Policy Analysis on Palestine, 1994); S. Tamari, Palestinian Refugee Negotiations: From Madrid to Oslo II (Washington: Institute For Palestine Studies, 1996); and A. Shiblaq, `Commentary,' Shaml Newsletter 3, (1996), p.4. 25. The principle of non-refoulment prohibits the return of an individual to a country where they will be tortured or persecuted. 26. The Palestinian homeland is a post-1922 designation that refers to the entire region west of the Jordan River. 27. E. Marx, `Palestinian Refugee Camps in the West Bank and Gaza', Middle Eastern Studies, Vol.28, No.2 (1992), p.289. 28. `Many camp refugees report that they own their own dwellings (between 70 and 90 per cent), although there is no regulatory framework surrounding ownership, buying and selling.' L.B. Jacobsen, Finding Means: UNRWA's Financial Crisis and Refugee Living Conditions, Fafo Report 427, Vol.1 (2003), p.60. 29. Ibid., pp.156­95. 30. Ibid., p.160. See also, Marx, op.cit., p.284. 31. Jacobsen, op.cit., p.85. 32. See, for example, J. Bennet, `Agency Aiding Palestinians is Strapped,' New York Times, Feb. 11, 2003. 33. One estimate put the percentage of non-refugee occupants of housing in the refugee camps of Gaza at 25 per cent. Marx, op.cit., pp.284­8. 34. A. Plascov, The Palestinian Refugees in Jordan: 1948­1957 (London: Frank Cass, 1981), pp.62­3. 35. H. Elnajjar, `Planned Emigration: The Palestinian Case,' International Migration Review, Vol.27, (1993), p.34. My italics. Elia Zureik, reviewing Benjamin Schiff's Refugee Unto the Third Generation: UN Aid to Palestinians makes the same argument: `Palestinian refugees have at times blamed the

UNRWA and Moral Hazard

421

36. 37. 38.

39.

organization for working for their resettlement, and training them for jobs which will facilitate their emigration from the occupied territories and absorption in the host countries.' Fofognet Digest, 24­4 March (1996). See also I. Abu Lughod, `Educating a Community in Exile: The Palestinian Experience,' Journal of Palestine Studies, Vol.2, Issue 3 (1973), p.111. Ibid., p.34. Noteworthy, by UNHCR standards, the 70 per cent of Palestinians not living in camps but defined by UNRWA as refugees would not be regarded as refugees. A multiplicity of UNHCR and U.S. Committee for Refugees periodicals, notably Refugees, UNHCR Global Reports, and World Refugee Survey, as well as research papers, such as USCR's Refugee Reintegration in War-Ravaged Eritea chronicle the racial, religious and gender abuse of its refugee populations as well as the near starvation levels they are forced to endure. The World Bank's Human Development Report (various years) provides country by country human development indices that show living standards for most of Africa considerably below those of Palestinians in refugee camps. Ashkenasi, op.cit., p.63.