the pre-RDP population to 30 percent for RDP Job duration for the initial job did not significantly differ between the two comparison groups 4.9 months - on average for employed RDP participants versus 4.7 months for the pre-RDP group. If subsequent jobs are added to the job duration statistic, the pre-RDP group shows longer duration (6.3 months compared to 5.2 months for RDP clients). Pre- and post-RDP wages, based on data from welfare records and case management files, were similar, about $4 per hour or less. Based on welfare data, the pre-RDP employed refugee had Pre-RDP welfare grant savings due to employment-related The Oregon Refugee Early Employment Project (REEP) The Oregon Refugee Early Employment Project (REEP), which began September 16, 1985, integrates the delivery of cash assistance with case management, social services, and employment services within the private non-profit sector in an effort to increase refugee employment and reduce reliance on cash assistance. Encompassing a tri-county area surrounding Portland, where 85 percent of all refugees in Oregon initially settle, REEP's objectives are to place: (1) 75 percent of all employable participants in full-time, permanent employment within 18 months of their arrival in the U.S.: (2) 50 percent of employable participants within 12 months of their arrival; (3) 25 percent of employable participants within 6 months of their arrival reducing the aggregate 18-month dependency rate for these clients from 80 percent to 50 percent; and (4) to enable 63 percent of all participants (including minor children) to graduate from the demonstration within 18 months of their arrival because their family earnings exceeded program income standards. - The project has been serving needy refugees who do not meet the AFDC or SSI categorical requirements (i.e., members of twoparent families, couples without children, and single individuals) during their initial 18 months in the United 38 States. The target population includes both new arrivals and secondary migrants. Refugees who normally are eligible for assistance under AFDC continue to be eligible for that program and do not participate in REEP. The project is continuing operations through FY 1990. The Refugee Policy Group (RPG), under contract to the State of Oregon to evaluate the effectiveness of REEP, reported that by the end of the third year of REEP-operation, the project reached its objective of placing at least 75 percent of employable adults in permanent, full-time employment within 18 months of their arrival. The median hourly wage for these refugees was $4, and 56 percent were employed for at least 90 days. By the end of the third year of REEP, 274 cases (30 percent) had been closed due to economic self-sufficiency while 181 cases (20 percent) lost eligibility because of time-expiration and 242 cases (26 percent) left the Portland area. Discretionary Social Service Initiatives Over the last seven years, the Administration has committed 03 discretionary funds to support several initiatives aimed at aiding refugees to achieve self-sufficiency in the shortest possible time. During FY 1989 and FY 1990, ORR is providing a substantial amount of its discretionary funds under two program announcements entitled (1) Key States Initiative and (2) Job Links. (1) Key States Initiative: In FY 1987, ORR established the Key States Initiative (KSI). This initiative was undertaken to reduce cash assistance utilization by refugee families in States with high welfare utilization. Five States entered into Cooperative Agreements under the auspices of the Key States Initiative and one of these subsequently dropped out of the program. In the other four States, implementation of KSI has involved changes in the State systems for refugee services, as well as within the system of service providers and the refugee communities. These changes are now in various stages of implementation. Approximately $2.65 million was awarded in FY 1988, and about $2.3 million was awarded each year during FY 1989 and FY 1990 for continuation of four of these initiatives. (2) Job Links: As part of its two-tiered approach to assist, in different ways, States with both high and low welfare utilization, ORR in FY 1989 and 1990 offered non-KSI States an opportunity, competitively, to enter into cooperative agreements with ORR through a discretionary program named "Job Links." As the name suggests, under Job Links, custom-designed 39 linkages are created through development of service programs in States with good job opportunities and refugees needing jobs. Job Links aims to provide services that will encourage refugees to remain in such favorable States, and hopefully to attract other refugees from less favorable States which have high welfare dependency. Job Links places strong emphasis on such services as employment services, skills training, and support services aimed at enhancing refugee employability. ORR expects to award about $3 million to States under Job Links in FY 1990. Planned Secondary Resettlement Another initiative, aimed at improving economic adjustment opportunities, is the Planned Secondary Resettlement (PSR) Program. PSR grants assist clearly defined groups of refugees who are experiencing severe and protracted unemployment and public assistance dependency to achieve accelerated economic self-sufficiency. This is done through carefully planned relocation to communities offering favorable resettlement opportunities. Eligible grantees include States and public and private non-profit organizations that have had demonstrated experience in the provision of services to refugees, such as mutual assistance associations (MAAs) and national and local voluntary agencies. As of the end of FY 1989, there were six PSR grantees: four mutual assistance associations and two voluntary agencies. From FY 1983 to FY 1989, 226 families (1,150 individuals) have relocated from high welfare areas to self-sufficient communities through the PSR program. In FY 1989, the average cost of resettling families through the PSR program was $10,000 per family while average welfare cost savings to the government were estimated at $987 per month per family. At this rate, PSR families, on average, repay the cost to the government in just 10 months. Refugee Self-Help Initiatives Strengthening refugee/entrant mutual assistance associations (MAAS) is a priority goal of ORR. MAAS have become a great resource to communities because of the refugees' firsthand experiences and because they set examples of successful adjustment. Since 1980, ORR has allocated about $28.8 million for initiatives which strengthen the MAAS' role in the resettlement of new refugees and provide incentives for States to fund MAAS as social service providers. For fiscal year 1990, 46 States and the District of Columbia will be providing funds directly to MAAS. It is anticipated that MAAS - 40 will increasingly become active, viable community resources with broad public and private support. Population Composition. The demographic characteristics of refugees to be admitted to the U.S. in FY 1991 are likely to be similar to those of refugees admitted in FY 1989 and 1990 while varying somewhat from those of refugees admitted in the previous few years. For example, the higher proportion of Soviet and European refugees in the flow since 1989 has meant more adults and fewer children compared to the early 1980s, when refugees from Asia predominated. At the same time, refugee populations will continue to vary in age and sex composition from the resident American population as well as from each other. Indochinese refugees, for example, are younger on average than the resident U.S. population. The median age of the Indochinese arrivals from 1975 through 1989 was between 20 and 21 years. Nearly 40 percent were under age 15, compared to 23 percent of the American population. In this same period, the percentage who were age 65 and over remained roughly constant at less than 2 percent. About 55 percent of the arriving Vietnamese in recent years were males, in contrast to the general U.S. population in which about 49 percent are males. Because of the large proportion of children among the Indochinese refugees, the ratio of dependents will continue-to-be higher than in the general U.S. population, although the low proportion of elderly Indochinese is a significant mitigating factor. The young age structure means that a large number of young children from these refugee families will be entering the schools each year for some years to come. Soviet refugees were the largest single nationality group to arrive in FY 1989 and one of the oldest, with a median age of 30. New refugees from Ethiopia, Iran, and Eastern Europe averaged in their middle to late twenties in 1989, while those from Afghanistan were only twenty on average. Cuban refugees were the oldest, with a median age in FY 1989 of 36. Male refugees outnumber females in most groups, comprising 52 to 60 recent of arrivals. The Soviets are an exception, with a slight female majority. Because these nationality groups differ from each other in their background characteristics, any change in the source countries of the refugee flow means a corresponding change in the demographic impact of the refugee population. During the first seven months of FY 1990, 75 percent of newly arrived refugees have been placed in ten States. The same ten States received 76 percent of the FY 1989 placements (Table III). Five of these ten States are receiving a higher share of the arrivals in FY 1990 than in FY 1989. The most notable trend is the increase from 19 to 25 percent in the proportion of refugees going to New York coupled with the declining share going to California. The California proportion has fallen from 45.6 percent in FY 1988 to 29.1 percent in FY 1989 to 23.8 percent in the first seven months of FY 1990. Much of this change has occurred because the large numbers of Soviet Armenians who arrived in FY 1988 and settled near relatives in the Los Angeles area have been replaced by Soviet Jews settling in New York. Together, California and New York continue to receive nearly half of all newly arriving refugees. Table IV displays the number of refugee arrivals in every State during the first seven months of FY 1990. Generally, refugee communities in areas of current resettlement will continue to grow with admission of additional family reunification cases. As Table V shows, the geographic distribution of the Indochinese refugees is now well established, with nearly 40 percent living in California, and the rest distributed widely. Table VI The more than 315,000 refugees who have arrived since FY 1980 from areas outside Southeast Asia have a residential distribution different from the Indochinese. This has tended to diffuse the impact of refugee arrivals upon local communities. Large numbers of these other refugees have been resettled in cities in the Northeast and the Midwest. shows the placement locations of the non-Indochinese refugees who arrived during the most recent seven fiscal years. California received the largest number, more than 62,000, with New York in second place at more than 41,000 and growing rapidly. This table also shows the wide variations in ethnic composition of States' refugee populations. While 40 percent of the arriving refugees were non-Indochinese, thirteen States and the District of Columbia resettled a refugee population composed of more than 50 percent non-Indochinese; and six States and Guam resettled a population of 90 percent or more Indochinese. Among the States with the largest refugee populations, Florida, Illinois, Michigan, New Jersey, and New York have majorities of non- Indochinese refugees. As the ethnic composition of the arriving refugee population shifts in response to new needs, so will the geographic placement patterns of the new arrivals. During the first seven months of FY 1990, only 30 percent of the arrivals have been Indochinese, compared with 60 percent during the |