Vietnamese Asylum Seekers: Refugee Screening Procedures Under the
Comprehensive Plan of Action (Letter Report, 10/21/96, GAO/NSIAD-97-12).

Pursuant to a congressional request, GAO reviewed the implementation of
the Comprehensive Plan of Action (CPA) for the resettlement of asylum
seekers from Vietnam and southeast Asia, focusing on: (1) whether Hong
Kong and Indonesia implemented CPA refugee status determination
procedures in accordance with international standards and criteria; (2)
alleged corruption in the program; (3) whether asylum seekers who
returned to Vietnam encountered persecution; and (4) U.S. and United
Nations High Commissioner for Refugees' (UNHCR) costs associated with
CPA implementation.

GAO found that: (1) both Hong Kong and Indonesia agreed to follow
international standards for processing asylum seekers, and their
procedures met CPA basic structural requirements and minimal procedural
requirements set by UNHCR; (2) although there were some implementation
errors, Hong Kong's screening process had sufficient checks and balances
to ensure fair treatment of the 60,275 asylum seekers, and there were no
allegations of corruption in Hong Kong's program; (3) a review indicated
that the Hong Kong government made reasonable screening decisions in
nine of ten selected cases; (4) UNHCR had no hard evidence of widespread
rumors of corruption in the Indonesia program, which screened 18,131
persons; (5) UNHCR reviews of 486 cases upheld the screening decisions
in 481 cases, and found that corruption, while not substantiated in case
files, could have inflated the number of positive decisions; (6) a
review indicated that Indonesia made reasonable screening decisions in
six of ten selected cases; (7) the Department of State did not
adequately monitor the screening processes but relied on UNHCR reports
on CPA implementation; (8) the UNHCR program for monitoring the
voluntary repatriation of returnees was well-structured and implemented;
(9) there was no evidence of persecution among the returnee population;
and (10) UNHCR spent $8.8 million, including about $1.6 million for
1996, to maintain its monitoring operation in Vietnam and about $444
million in CPA and other funds between 1989 and 1995.

--------------------------- Indexing Terms -----------------------------

 REPORTNUM:  NSIAD-97-12
     TITLE:  Vietnamese Asylum Seekers: Refugee Screening Procedures 
             Under the Comprehensive Plan of Action
      DATE:  10/21/96
   SUBJECT:  Political refugees
             International relations
             International organizations
             Foreign governments
             Ethical conduct
             Eligibility determinations
             Immigration or emigration
             Administrative costs
             Political corruption
IDENTIFIER:  Dept. of State Orderly Departure Program
             UNHCR Comprehensive Plan of Action
             Hong Kong
             Indonesia
             Asia
             Vietnam
             
******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO report.  Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved.  Major          **
** divisions and subdivisions of the text, such as Chapters,    **
** Sections, and Appendixes, are identified by double and       **
** single lines.  The numbers on the right end of these lines   **
** indicate the position of each of the subsections in the      **
** document outline.  These numbers do NOT correspond with the  **
** page numbers of the printed product.                         **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
** A printed copy of this report may be obtained from the GAO   **
** Document Distribution Center.  For further details, please   **
** send an e-mail message to:                                   **
**                                                              **
**                                            **
**                                                              **
** with the message 'info' in the body.                         **
******************************************************************


Cover
================================================================ COVER


Report to the Chairman, Subcommittee on International Operations and
Human Rights, Committee on International Relations, House of
Representatives

October 1996

VIETNAMESE ASYLUM SEEKERS -
REFUGEE SCREENING PROCEDURES UNDER
THE COMPREHENSIVE PLAN OF ACTION

GAO/NSIAD-97-12

Vietnamese Asylum Seekers

(711155)


Abbreviations
=============================================================== ABBREV

  ASEAN - Association of Southeast Asian Nations
  CPA - Comprehensive Plan of Action
  INS - Immigration and Naturalization Service
  ODP - Orderly Departure Program
  UNHCR - United Nations High Commissioner for Refugees

Letter
=============================================================== LETTER


B-274161

October 21, 1996

The Honorable Christopher H.  Smith
Chairman, Subcommittee on International Operations
 and Human Rights
Committee on International Relations
House of Representatives

Dear Mr.  Chairman: 

The United States has a continuing special interest in the protection
and welfare of many Vietnamese and other ethnic groups in Southeast
Asia (including former employees of the U.S.  government and others
closely associated with the U.S.  presence in Vietnam before 1975),\1
and family members of persons in the United States.  Accordingly,
since 1975 the United States has resettled hundreds of thousands of
asylum seekers from the region, including about 12,900 Indo-Chinese
refugees under the Comprehensive Plan of Action (CPA).\2 The CPA was
an arrangement, or nonbinding agreement, for achieving a durable
solution to the continuing flow of asylum seekers in the Southeast
Asia region, adopted by the governments represented in the
International Conference on Indo-Chinese Refugees in Geneva on June
13-14, 1989. 

In response to your request, we reviewed the implementation of the
CPA.  Specifically, we (1) determined whether the first-asylum
countries of Hong Kong and Indonesia implemented CPA refugee status
determination procedures in accordance with international standards
and criteria,\3 (2) developed information about alleged corruption in
the program, and (3) ascertained whether asylum seekers returning to
Vietnam had encountered persecution.  We also identified U.S.  and
United Nations High Commissioner for Refugees (UNHCR) costs
associated with implementing the CPA. 

Regarding the refugee screening process and corruption issues, the
scope of our review was limited to Hong Kong and Indonesia,\4 where
we were provided access to documents contained in UNHCR case files. 
However, we were not permitted to review Hong Kong or Indonesian
government files, nor did we interview individual asylum seekers in
these countries.  Moreover, as with most human endeavors such as
this, actual refugee status decisions involved considerable judgment
on the part of interviewing and reviewing officials in determining
the merits and credibility of the applicants' claims, particularly
when many asylum seekers lacked evidence and documentation to
substantiate their claims.  We do not believe the information in
UNHCR's files was sufficient for us to make conclusive determinations
on asylum seekers' refugee status; however, we believe it was
sufficient to assess whether status determination decisions made by
country and UNHCR officials appeared reasonable based on
international refugee adjudication standards.  We examined the Hong
Kong and Indonesian screening processes within this context. 

At the request of Congressmen Robert K.  Dornan, Benjamin A.  Gilman,
and Thomas M.  Davis, III, and Congresswoman Zoe Lofgren, we are
currently examining other aspects of the CPA's implementation in Hong
Kong and Indonesia, as well as in Malaysia and the Philippines.  This
work involves about 235 individual asylum seeker cases and focuses
primarily on an analysis of whether the cases may be of special
interest to the United States.  Our emphasis is on cases involving
family unity and victims of violence, which are not addressed
directly in this report, and on some cases involving refugee status
determination.  However, the respective governments restricted our
access to asylum seekers and documents belonging to the first-asylum
countries.  We will be reporting separately on this work. 


--------------------
\1 The United States withdrew its remaining military forces and
civilian presence from Vietnam when the South Vietnamese government
fell in April 1975. 

\2 The United States has also resettled over 445,000 Vietnamese of
special interest from inside Vietnam under the continuing Orderly
Departure Program (ODP), established in 1979 to provide a safe, legal
alternative means of leaving the country, other than clandestinely by
boat.  (The ODP program also serves to relieve the flow of asylum
seekers into first-asylum countries.)

\3 Hong Kong is a British Dependent Territory.  Refugee status
determination procedures are often referred to as refugee screening
procedures.  Those given refugee status are referred to as screened
in, and those denied refugee status are screened out. 

\4 Other primary countries of first-asylum not included in this
review were Thailand, Malaysia, and the Philippines. 


   BACKGROUND
------------------------------------------------------------ Letter :1

Between 1975 and 1989, over 2 million Indo-Chinese, many of them
Vietnamese, left their countries of origin seeking asylum elsewhere
in the region.  Most were ultimately resettled in Western countries,
including the United States and Canada.  However, by the late 1980s
an unabating outflow, large numbers of asylum seekers already in
asylum camps, and continuing resettlement placed increasing burdens
on neighboring countries and territories and Western countries, and
the need for a durable solution to the problem became critical. 

In June 1989, 75 countries, including the United States, Great
Britain, Association of Southeast Asian Nations (ASEAN) countries,\5
and Vietnam adopted by declaration the CPA agreement to deal with the
problem.  The agreement's primary objectives were to deter
clandestine departures, provide temporary refuge to all asylum
seekers, establish procedures for determining their refugee status,
resettle in third countries those found to be refugees, and
repatriate those determined not to be refugees. 

The agreement stipulated that refugee status determination screening
was the responsibility of the national authorities of individual
first-asylum countries, in accordance with the 1951 U.N.  Convention
relating to the Status of Refugees and its 1967 Protocol,\6 bearing
in mind the 1948 Universal Declaration of Human Rights, and taking
into account the special situation of asylum seekers and the need to
respect the family unit.  Refugee status determination procedures
were to be structured in accordance with international norms and were
to employ internationally accepted refugee status determination
criteria.  International screening procedures include prescreening
counseling for applicants, first-instance screening, and an appeal
process for denied applicants. 

Notwithstanding the responsibilities vested with the first-asylum
countries under the CPA, the agreement placed a number of important
responsibilities on UNHCR.  The CPA's drafters envisioned the
agreement's implementation as a dynamic process that would require
continued coordination and possible adaptation to respond to changing
situations.  To ensure its effective implementation, the plan, among
other things, established that

  -- UNHCR, with the financial support of the donor community, would
     be in charge of continuing liaison and coordination with
     concerned governments and intergovernmental and nongovernmental
     organizations to implement the agreement;

  -- UNHCR was to participate in the refugee status determination
     process in an observer and advisory capacity; and

  -- UNHCR would be responsible for (1) providing training to first-
     asylum country officials to help ensure fairness and consistency
     in the screening process; (2) coordinating the timely
     resettlement of those found to be refugees; and (3)
     administering a safe, dignified repatriation program for those
     found to be nonrefugees.\7

Appendix I discusses UNHCR's responsibilities under the CPA relative
to its U.N.  chartered authorities and responsibilities. 

The Charter of the Office of High Commissioner authorizes UNHCR to
make its own determinations about individuals' refugee status.  This
authority is commonly referred to as mandate authority.  In what is
essentially an additional procedure in the refugee status
determination process, UNHCR has examined under mandate authority
many asylum seekers that were denied refugee status by national
authorities.  Some first-asylum countries, such as Hong Kong, have
recognized UNHCR's mandate authority; others have not. 

Although not a requirement of the CPA, UNHCR has also carried out a
returnee monitoring program designed to ensure that asylum seekers
returning to Vietnam did so under conditions of safety and dignity
and in conformance with Vietnamese and international law. 

According to State Department statistics, over 120,000 Vietnamese
were screened under the CPA.  As of March 1996, about 33,200 had been
screened and resettled in third countries.  About 75,500 had returned
to Vietnam.  A total of 36,623 remained in asylum camps in the region
as of March 1996.  (See fig.  1.)

   Figure 1:  Vietnamese Asylum
   Seekers in Southeast Asia Camps

   (See figure in printed
   edition.)

   Source:  Department of State.

   (See figure in printed
   edition.)

In March 1996, the Steering Committee of CPA member countries
announced the CPA's formal closure as of June 30, 1996.  The
Committee reaffirmed that the only viable option for Vietnamese
determined to be nonrefugees was for them to return to Vietnam,
either voluntarily or under a mandatory repatriation program.  The
Steering Committee directed UNHCR to, therefore, phase down its care
and maintenance activities and staffing in the first-asylum countries
as of that date.  In the case of Hong Kong, mindful of the special
circumstances prevailing in the territory and the large number of
asylum seekers still in the camps, UNHCR was to make other
appropriate arrangements to resolve the asylum seeker problem as soon
as possible.\8

The application of international and U.S.  refugee screening criteria
can differ somewhat and could result in different screening
decisions.  For example, under U.S.  criteria an asylum seeker could
be properly considered a refugee solely on the basis of past
persecution,\9 but under international criteria, this is only one of
several elements to be considered.  Other elements include whether
(1) the past persecution was distant in time, (2) the individual was
subsequently able to lead a normal life, and (3) it is reasonably
likely that the individual will be persecuted on return to his or her
country.  Therefore, individuals considered refugees under U.S. 
criteria may not be refugees under international criteria.  (See app. 
I for a more complete discussion of these differences.)


--------------------
\5 Participating ASEAN countries included Brunei Darussalom,
Indonesia, Malaysia, the Philippines, Singapore, and Thailand. 

\6 Of the five first-asylum countries that adopted the CPA, only the
Philippines was a signatory to the 1951 Convention or the 1967
Protocol, but it had not enacted comprehensive legislation to
implement the Convention and Protocol. 

\7 As discussed later, UNHCR took a more active role than enumerated
in the CPA agreement in Indonesia's screening process. 

\8 As of June 30, 1996, first-asylum camps in Malaysia were closed
and all remaining asylum seekers in the camp were repatriated to
Vietnam; the CPA and UNHCR's involvement in the camps in the
Philippines were terminated; and Indonesia and Thailand extended
their involvement with the CPA for 90 days. 

\9 While under U.S.  law a person who was actually persecuted in the
past on account of race, religion, nationality, membership in a
particular social group, or political opinion may qualify as a
refugee, an Immigration and Naturalization Service (INS) officer may,
at his or her discretion, deny refugee status if conditions in the
person's country have changed to such an extent that the alien's fear
of persecution is no longer well-founded.  (INS Basic Law Manual;
Matter of Chen, Int.  Dec.  No.  3104 BIA 1989.)


   RESULTS IN BRIEF
------------------------------------------------------------ Letter :2

The CPA agreement stipulated that the first-asylum countries' refugee
screening procedures be carried out in accordance with established
international screening criteria and procedures, including
prescreening counseling for applicants, first-instance screening, and
an appeals process for denied applicants.  Our examination in Hong
Kong and Indonesia indicated that both programs met the CPA's basic
structural requirements for refugee adjudication criteria and
screening procedures.  Both countries agreed to follow international
criteria recognized in the 1951 U.N.  Convention relating to the
Status of Refugees and its 1967 Protocol, and both countries'
procedures met minimal procedural requirements endorsed by the
Executive Committee of UNHCR. 

On the basis of our review of Hong Kong's screening process,
notwithstanding acknowledged errors during its implementation, we
believe the process contained sufficient checks and balances to
provide reasonable assurances that asylum seekers' cases could be
heard and errors could be identified and corrected.  We received no
allegations of corruption in the program in Hong Kong. 

The Hong Kong Immigration Department was responsible for first-
instance screening, and an independent Refugee Status Review Board
was established to hear denied applicants' appeals.  Depending upon
the circumstances of their cases, those denied refugee status at
appeal had further opportunity to have their cases reviewed in the
Hong Kong courts.  All asylum seekers denied refugee status had their
cases examined by UNHCR for mandate refugee review.  All asylum
seekers had legal counsel available to them--through UNHCR, privately
retained attorneys, and possibly the Hong Kong Legal Aid Department. 
(Hong Kong's screening process is discussed in more detail in app. 
II.)

Hong Kong government and UNHCR officials acknowledged errors, such as
misinterpretations of language, questionable interviewing practices
and techniques, and inadequate assessments of applicants' claims,
during the screening process--particularly early in the process.\10
However, the officials told us they had worked to correct the errors
and that the process had improved over time. 

In a limited test of Hong Kong's screening procedures and the quality
of the screening decisions, we relied on UNHCR case files to examine
the decisions reached in 10 selected cases.\11 (Hong Kong authorities
denied us access to their case files.) On the basis of documentation
contained in UNHCR's case files, we believe the Hong Kong government
and UNHCR officials' decisions appeared to be reasonable in 9 of the
10 cases we examined.  In the 10th case, we identified information
that we believe should have been factored into the decision.  UNHCR
agreed and subsequently granted the individual mandate refugee
status.  A UNHCR official in Hong Kong attributed the initial
(negative) screening decision to human judgment by the reviewing
officials. 

In Indonesia, we believe the large majority of asylum seekers with
strong claims for refugee status were screened in--due in part to
UNHCR's heavy involvement in the screening process.\12 In fact, there
is evidence that an undetermined number of nonmeritorious cases were
also screened in by the Indonesian authorities.  UNHCR recommended
that these cases be screened out but did not challenge the
authorities' decisions.  Corruption in the Indonesia process likely
contributed to undeserving cases' gaining refugee status; however, it
is unlikely that strong cases were denied refugee status due to unmet
corruption demands. 

The process in Indonesia provided for both first-instance screening
and an appeals procedure involving Indonesian authorities and UNHCR
legal consultants; however, few asylum seekers received individual
legal assistance in preparing for screening or drafting appeals. 
According to UNHCR, its staff provided group prescreening counseling,
and its legal consultants provided screened-out applicants
counseling, mostly in group sessions, on their right of appeal and
appeal procedures.  However, no legal assistance was provided in
drafting appeals.  According to UNHCR, the Indonesian government did
not allow nongovernmental organizations to participate in asylum
seeker counseling, and asylum seekers did not have access to private
attorneys or judicial review. 

According to UNHCR, it concentrated its efforts in Indonesia on
trying to ensure that those with well-founded refugee claims were
screened in.  It did this by performing an initial interview of all
asylum seekers and making recommendations on their refugee status to
Indonesian authorities.  According to UNHCR, the Indonesian
government accepted all but 22 of the more than 3,000 cases it
recommended for refugee status. 

Rumors of corruption in Indonesia's screening process began at the
inception of the screening process.  Nongovernmental organizations,
asylum seeker advocacy groups, some Members of Congress, and others
have severely criticized the process, alleging widespread corruption
and denial of refugee status to asylum seekers with well- founded
claims due to their unwillingness or inability to accede to bribery
demands or demands for sexual favors.  For example, Boat People
S.O.S.  provided us a list of seven cases in Indonesia that it said
were strong cases improperly screened out due to unmet corruption
demands. 

Reports, files, and documents provided to us by nongovernmental
organizations, advocacy groups, former asylum seekers, UNHCR, and
other knowledgeable sources indicated widespread corruption in the
process.  UNHCR officials in Jakarta and at the asylum camp in
Galang, an asylum seeker spokesman in Galang, and other reliable
sources at Galang also readily acknowledged corruption in the
process. 

We did not conduct an independent investigation of allegations of
corruption provided us.  UNHCR officials had no hard evidence with
which to demonstrate corruption, how and with whom it took place, and
how it affected the screening process.  Indonesian authorities we
spoke with denied the existence of corruption in the process. 
Therefore, we cannot definitively state that corruption occurred or
that weak or well-founded cases were screened in or out due to
corruption or failure to comply with corruption demands. 

However, on the basis of documentation available to us, discussions
with UNHCR officials in Geneva and Indonesia, and the results of a
series of reviews of the process by UNHCR in 1995, it appeared that
corruption in the process was likely and asylum seekers with weak
cases participated in the corruption in order to gain otherwise
unattainable refugee status.  Moreover, we cannot rule out the
possibility that bonafide (ultimately screened in) refugees were
victimized by unscrupulous Indonesian officials and possibly by
locally hired UNHCR legal consultants during the screening process. 
But, we did not identify well-founded cases that were screened out on
the basis that they were unable to meet corruption demands.  (In our
examination of the seven cases referred to us by Boat People S.O.S.,
five appeared to be reasonable decisions.  The decisions in the
remaining two cases appeared questionable; however, UNHCR had
reviewed the cases and accorded the claimants mandated refugee status
as part of its 1995 review of the Indonesia process.)

UNHCR monitors in Vietnam have reported that they found no evidence
of persecution of returned asylum seekers by Vietnamese authorities. 
Our review of 12 returnee cases referred to us by Boat People S.O.S. 
and Asia Watch as cases of persecution or special concern also
revealed no evidence of persecution.  We found that UNHCR had
implemented a comprehensive monitoring program that provides
reasonable assurance that returnees were not persecuted.  Our
discussions in Vietnam with various knowledgeable officials from the
United States, European Union, the British Embassy, and
nongovernmental organizations with activities in Vietnam, and
interviews with 22 returnees in the North and South reinforced the
monitors' reports.  Nevertheless, Vietnam remains an authoritarian
communist government that does not tolerate political dissent, and
returnees risk persecution should they become involved with political
activities not approved by the government. 

UNHCR spent an estimated $444 million on the CPA from 1989 through
1995,\13 excluding unpaid obligations of $139 million in Hong Kong. 
U.S.  contributions to UNHCR for the CPA during the period were an
estimated $151 million.  The United States also contributed about $9
million to voluntary agencies in support of CPA activities. 


--------------------
\10 Refugee screening in Hong Kong actually began in June 1988, a
year before the inception of screening under the CPA.  Hong Kong
modified its process somewhat to conform to CPA requirements. 

\11 Five of the cases were referred to us by Boat People S.O.S., a
Vietnamese asylum seekers advocacy group.  We selected the remaining
five randomly from a list of cases referred to UNHCR by another
nongovernmental organization. 

\12 Seven cases were not screened in during screening but were later
given mandate refugee status by UNHCR. 

\13 This figure excludes general program expenditures in Thailand. 
General program funds provide for UNHCR's in-country operations and
include administrative expenses and care and maintenance costs for
in-country refugee populations. 


   HONG KONG'S PROCEDURES AND
   CRITERIA CONSISTENT WITH CPA
   GUIDANCE
------------------------------------------------------------ Letter :3

A total of 60,275 persons were screened under the CPA in Hong Kong.
Table 1 shows the results of first-instance and appeals screening and
the UNHCR mandate review. 



                                Table 1
                
                 Hong Kong Refugee Status Determination
                   Results (March 1989-December 1995)

                                 Allowed        Refugee
                                 refugee         status        Percent
                                  status   not approved       approved
-------------------------  -------------  -------------  -------------
First-instance screening           6,974         53,301           11.6
Appeal\a                           2,821         46,078            5.8
UNHCR mandate                      1,553         47,374            3.1
======================================================================
Total                             11,348         48,927           18.8
----------------------------------------------------------------------
\a Some screened out at first instance did not appeal their
decisions. 

Hong Kong government reports stated that refugee screening procedures
were carried out under UNHCR guidelines in accordance with
internationally accepted refugee status determination practices. 
According to Hong Kong Immigration Department officials, CPA
screening was conducted in accordance with the UNHCR Handbook on
Procedures and Criteria for Determining Refugee Status, and no
additional criteria were developed. 

Hong Kong Immigration Department officials stated that selection of
interviewing officers was based on rank, experience, education, and
language and writing skills.  The selected interviewing officers were
experienced naturalization officers, visa officers, and Hong Kong
entry point supervisors in the Immigration Department. 

UNHCR reports and interviews with UNHCR and government officials in
Hong Kong indicate that UNHCR took an active role in training
government officials involved in the screening process.  The
officers' training program included a 9-day seminar of lectures and
workshops and an extensive list of readings, including articles and
books on Vietnam and human rights topics.\14 Immigration Department
officials told us most of the interviewing officers were also sent to
Vietnam for 7-day familiarization visits.  We were told that an
extensive library of materials on conditions in Vietnam was
established for the officers' use.  At the height of CPA screening
there were approximately 220 interviewing officers. 

According to UNHCR officers in Hong Kong, informal monthly staff
meetings were held during the initial phases of the screening program
at which Hong Kong's procedures were discussed and critiqued.  We
were told that if issues or points were raised that needed to be
brought to the government's attention they were passed on to
Immigration Department officials. 

The UNHCR officials also said that in accordance with a 1988
agreement with the Hong Kong government, UNHCR monitored and advised
the government on its refugee screening procedures and provided the
government periodic formal assessments of its procedures.  According
to UNHCR officials, these assessments identified various problems and
suggested improvements.  Early problems with Hong Kong first-instance
screening procedures that UNHCR brought to the government's attention
included inadequate interpretation of applicants' statements during
screening interviews, questionable interviewing practices and
techniques by interviewing officers, and inadequate assessments of
applicants' claims by interviewing officers.  Hong Kong government
officials with whom we met also acknowledged such problems with
screening procedures.  UNHCR officials told us the Hong Kong
government was very responsive to the problems raised and endeavored
to correct them.  The Chairman of the Refugee Status Review Board
told us that the quality of interviewing officers' first-instance
screening decisions varied widely initially but that the officers had
good training and learned quickly. 

To gain greater insights into Hong Kong's screening procedures, and
as a limited test of them, we examined 10 selected cases in which
asylum seekers had been screened out by the Hong Kong authorities and
denied mandate by UNHCR.  Five of the cases were provided to us by
Boat People S.O.S.; we selected the other five at random from a list
of cases another advocacy group had requested that UNHCR reexamine. 

Among the first five cases, one had been accepted by the United
States on family reunification grounds and was awaiting
resettlement.\15 Another had been offered resettlement in the United
States but refused the offer unless his son, who was screened
separately as an adult and denied refugee status, could join him.  In
two other cases, we concluded, based on the documentation in UNHCR's
files, that the applicants had failed to establish a well-founded
fear of persecution.  In the fifth case, the applicant was screened
out due to a lack of plausibility about portions of his claims. 
However, when we raised questions about portions of his claims and
pointed out to UNHCR the strength of those aspects of the claim where
plausibility had not been questioned, UNHCR reexamined his case and
mandated refugee status for him.  A UNHCR official acknowledged
judgmental errors in this case.  According to the official, any
system dependent upon human judgments will sometimes produce errors. 

In the other five cases, documentation in UNHCR's case files
indicates that the applicants could not establish a well-founded fear
of persecution for any of the reasons specified in the 1951
Convention.  On the basis of available documentation, it appears
these cases were reasonably adjudicated. 


--------------------
\14 Later extended to 2 weeks. 

\15 UNHCR Handbook on Procedures and Criteria for Determining Refugee
Status (Handbook) indicates that at a minimum the spouse and minor
children benefit from the principle of family unity.  In practice,
other dependents, such as aged parents of refugees, are normally
considered if living in the same household. 


   INDONESIA'S PROCEDURES AND
   CRITERIA CONSISTENT WITH CPA
   GUIDELINES
------------------------------------------------------------ Letter :4

Both Indonesian government and UNHCR officials stated that refugee
screening procedures were generally carried out under UNHCR
guidelines, in accordance with internationally accepted refugee
status determination criteria.  Our review of the process indicated
that it met the basic structural requirements of the international
guidelines, such as provision for appeal of negative first instance
decisions and the application of international refugee status
determination criteria.  UNHCR trained both P3V and appeals level
officials in the application of the procedures, according to UNHCR
officials.\16

A total of 18,131 persons were screened under the CPA in Indonesia. 
Appendix III describes Indonesia's screening process.  Table 2 shows
the results of first-instance and appeals screening. 



                                Table 2
                
                 Indonesia Refugee Status Determination
                  Results (March 1989-September 1993)

                                 Allowed        Refugee
                                 refugee         status        Percent
                                  status   not approved       approved
-------------------------  -------------  -------------  -------------
First-instance screening           5,083         13,048             28
Appeal                             2,759          9,463           22.5
UNHCR mandate                          -              -              -
Total\a                            7,842         10,289           43.3
----------------------------------------------------------------------
\a Some screened out at first-instance did not appeal their
decisions. 

Of 13,048 persons screened out at first-instance, 12,222 filed
appeals.  The appeals process in Indonesia consisted of a Review
Committee, which reviewed all appeals, and an Appeals Board, which
reviewed those cases on which the Review Committee could not reach a
decision.  The Review Committee decided the large majority of the
appeals, confirming 9,392 negative decisions and reversing 2,592
(about 21.6 percent).  The Committee could not reach a decision on
238 cases, and these were referred to the Appeals Board.  The Board
reversed 167, or about 70 percent of the cases referred to it, to
positive decisions. 

UNHCR officials in both Jakarta and Galang (the island asylum camp in
northern Indonesia) told us they were most concerned with ensuring
that deserving asylum seekers were screened in and concentrated their
efforts to that end.  They said that in their view Indonesian
officials were too generous in screening in weaker cases, but UNHCR
(in keeping with its policy) did not challenge the Indonesians'
decisions on asylum seekers they believed to be undeserving.  A UNHCR
official told us it was in the Indonesian government's interest to
screen in and have resettled in third countries as many applicants as
possible, because those remaining in the camp represented a problem
for the government. 


--------------------
\16 "P3V" is the Indonesian inter-ministerial task force created to
deal with Vietnamese boat people. 


   CORRUPTION APPARENT IN
   INDONESIA'S SCREENING PROCESS
------------------------------------------------------------ Letter :5

Rumors of corruption began appearing in Indonesia at the inception of
the status determination process.  The first documented reporting of
corruption appeared in June 1991, when a UNHCR official at Galang in
a memorandum to his office in Jakarta reported rumors of P3V
officials soliciting sexual favors from female asylum seekers and
offerings of such services by the asylum seekers in exchange for
favorable screening decisions.  Boat People S.O.S.  provided us
numerous statements, affidavits, and referrals of individuals who had
been asked for or had paid bribes or other demands for themselves or
family members in exchange for favorable screening decisions.  The
alleged corruption involved not only Indonesian government officials
in Galang and Jakarta, but two Indonesian legal consultants employed
by UNHCR, as well.  The allegations included assertions that asylum
seekers with genuine refugee claims were screened out due to their
inability to met corruption demands. 

Affidavits and interviews with individuals referred to us by Boat
People S.O.S.  and discussions with UNHCR officials in Jakarta and
Galang and inhabitants of the Galang camp indicated potentially
widespread corruption in the screening process.  UNHCR officials told
us that there was undoubtedly corruption in the camps during the
screening process.  And, residents in the camp indicated that
corruption existed.  A Vietnamese camp leader told us, for example,
that P3V officials actively solicited bribes and asylees willingly
offered payment in exchange for help in the screening process. 

UNHCR officials told us that despite widespread rumors of corruption
they had no hard evidence of it.  A senior UNHCR official told us
that UNHCR had no authority to investigate Indonesian government
officials' activities and that UNHCR could only urge Indonesian
authorities to pursue the matter.  The official told us that
Indonesian authorities regularly conducted investigations at Galang
and that a number of cases were overturned as a result.  He said,
because P3V officials were removed and cases were overturned, he
concluded that corruption was involved.  He said he never saw a
documented corruption case, however, and Indonesian authorities did
not acknowledge corruption.  Indonesian officials also denied to us
that any corruption existed in the refugee determination process. 

UNHCR officials in Indonesia told us they took various steps to
combat corruption by pressuring the Indonesian government to minimize
it.  These included mentioning allegations to Indonesian authorities,
discussing rumors with high-level foreign government officials during
consultations and visits to the Galang camp, and publishing
allegations in quarterly operations reports, all in the attempt to
embarrass the government into acting. 

UNHCR officials acknowledged there were no posters or leaflets
distributed in the camps urging residents not to participate in
corruption and to report any bribery solicitations.  There were no
UNHCR notices or warnings to camp officials.  There were also no
provisions, such as hot lines, established for residents to report
anonymously. 

A UNHCR official at Galang told us that on his own initiative he
conducted a 15-month investigation into corruption in the camp.  The
official, who for a time lived in the camp and had his own network of
informants, was unable to obtain proof of corruption.  During his
investigation, the official compiled a list of about 40 corruption
cases, all instances of individuals paying bribes but being screened
out nonetheless.  The official said he heard of these cases only
because the individuals were complaining in an attempt to recoup
their bribe money.  According to the official, none of the cases in
his investigation involved individuals with strong refugee claims
being screened out because they could not pay bribes.  The official
concluded that asylum seekers were active participants in corruption
and were unwilling to provide direct evidence because it benefited
those with weak claims. 

In response to intense criticism by nongovernmental organizations and
some members of the U.S.  Congress in late 1994 and early 1995, UNHCR
launched a series of four reviews of screening procedures and the
impacts of corruption on the screening caseload in Indonesia.  The
reviews, conducted by UNHCR attorneys and senior program personnel
from outside Indonesia, examined a total of 486 nonrandomly selected
cases, representing about 10 percent of the remaining camp
population.\17 The selected cases included high- visibility cases,
such as the 22 cases screened positive by UNHCR but negative by
Indonesian officials, all the cases examined by UNHCR's two
Indonesian legal consultants who had corruption allegations lodged
against them, and 21 cases previously referred from the U.S. 
Department of State and nongovernmental organizations.  The reviews
also included both positive and negative cases on which UNHCR
consultants and Indonesian officials had reached the same decisions. 

The reviews upheld the screening decisions in 481 of the 486 cases
reviewed.  The remaining five cases were given refugee status under
UNHCR's mandate authority, although this status had not yet been
recognized by the Indonesian government or communicated to the asylum
seekers at the time of our field work in January 1996.  The reviews
concluded that, overall, Indonesia's screening procedures properly
identified and screened in cases with serious protection concerns,
but that a small number of borderline cases could possibly have
benefited from more sympathetic application of the screening
criteria.  The reviews found that corruption was a factor that
impinged on screening procedures, but was rarely, if ever,
substantiated in the case files, and that the chief effect was to
inflate the number of positive decisions by also screening in weak
cases.  The reviews did not support the assertion that strong cases
failed because they could not pay corruption demands. 

The reviews also concluded that UNHCR did not take adequate steps to
detect or prevent corruption in the screening process, allowing
corruption to undermine the integrity of the process.  They also
indicated that UNHCR took no actions with regard to the two
Indonesian legal consultants alleged to have been involved in
corruption.\18

In a limited test of Indonesia's processing procedures and screening
decisions, we examined a sample of 10 cases.  All were cases alleged
by Boat People S.O.S.  to have been improperly screened out and
included seven cases where the individuals were allegedly screened
out because they could not pay bribes demanded of them.  Our review
of UNHCR case files in the branch office near Galang (the files did
not contain P3V or appeals data) indicated that in five of the seven
cases allegedly screened out due to the applicants' inability to pay
bribes, the decisions reached appeared to be reasonable and
consistent with screening criteria.  In two of the cases, however,
the applicants appeared to have reasonable claims and the screening
decisions appeared to be incorrect.  Both cases had been examined by
UNHCR as part of its 1995 review, however, and both were among the
five cases mandated during the review. 

Of the three remaining cases, one appeared to have a reasonable claim
for refugee status but was screened out at first instance.  However,
he fled Galang for Australia before his case was appealed.\19 UNHCR
officials told us that had he been screened out on appeal, he
probably would have received UNHCR mandated refugee status.  The
second case appeared to be reasonably adjudicated.  In the third
case, we raised questions that resulted in UNHCR's mandating the
individual.  In this case, Indonesian authorities screened the
individual out at both first instance screening and upon appeal. 
UNHCR representatives had recommended that the individual be screened
in at first instance, but UNHCR concurred in the appeal decision. 
UNHCR also rejected the individual at mandate review, on the bases
that elements of the individual's claims did not constitute
persecution and additional important representations presented at
appeal lacked credibility.  UNHCR reconsidered and mandated the case
after we raised questions about whether elements of the individual's
and his family's treatment after 1975, taken together, could afford
the individual the benefit of a doubt--even without considering the
information provided at appeal. 


--------------------
\17 Cases do not equate to numbers of individuals:  A case can
involve several individuals. 

\18 UNHCR dismissed the officials in 1994. 

\19 The individual was granted refugee status in Australia. 


      DEPARTMENT OF STATE FAILED
      TO MONITOR SCREENING
      PROCESSES
---------------------------------------------------------- Letter :5.1

As a participating country and a major contributor to the CPA, the
United States has a strong interest in the CPA's effective
implementation.  The Department of State has responsibility for
ensuring that U.S.  interests in international programs that receive
U.S.  funding are protected. 

Despite widespread concerns that corruption affected the refugee
screening process, State Department officials told us they had only a
general awareness of the problem in Indonesia and elsewhere in the
region.  Our review indicated that while State Department's
Washington and embassy officials were familiar with the CPA program
and had in-depth knowledge of asylum seekers' care and maintenance
and resettlement issues, they possessed only cursory knowledge of the
first-asylum countries' screening processes and weaknesses in them. 
It was not until advocacy groups began petitioning State and UNHCR in
mid-1994 (after the screening process had essentially concluded) to
look into charges of corruption that the scope of the alleged
corruption became known.  It appears that until that time, State
relied on UNHCR reporting to determine how the screening process was
being implemented and confined its oversight and management to care
and maintenance and resettlement issues.  State officials
acknowledged to us that the Department failed to become aware of the
problems in the camps early enough and that the Department did not
adequately monitor the screening processes of the countries of
first-asylum to ensure their integrity. 


   UNHCR'S RETURNEE MONITORING
   PROCEDURES APPEARED EFFECTIVE
------------------------------------------------------------ Letter :6

UNHCR's monitoring efforts in Vietnam began with the voluntary
repatriation of 75 persons from Hong Kong in March 1989, thereby
pre-dating the June 1989 CPA.  The legal basis for UNHCR's program is
a Memorandum of Understanding between UNHCR and Vietnam, signed in
December 1988, which committed the Vietnamese government to
".  .  .  ensure that the voluntary return from the countries of
first-asylum will take place in conditions of safety and dignity in
conformity with national and international law." This translated, in
practice, to ensuring that no voluntary returnees would be subjected
to punitive or discriminatory measures related to their illegal
departures, stays abroad, or returnee status, and that they would not
be subjected to intimidation or harassment.  The Memorandum of
Understanding also stipulated that UNHCR would have full access to
voluntary returnees.  Involuntary returnees were not covered by the
Memorandum of Understanding, but subsequent Memoranda between the
Vietnamese and first-asylum governments relating to repatriation of
nonrefugees provided the same assurances for involuntary returnees
that the UNHCR/Vietnam agreement afforded voluntary returnees.\20

We found UNHCR's monitoring program to be well structured and
implemented, achieving broad national coverage and contact with large
numbers of returnees.  Together with a network of contacts made up of
western governments, nongovernmental organizations, journalists, and
other interested individuals, UNHCR was able to maintain visibility
over a significant portion of the returnee population, thus helping
to monitor the Vietnam government's adherence to internationally
agreed terms of returnee admission, reception, and reintegration.  As
of January 1996, UNHCR monitors had visited all 53 provinces in
Vietnam, 360 of the 363 districts with resident returnees, and about
20,000 (about 26 percent) of the approximately 77,000 returnee
population.  UNHCR and other officials we spoke with in Vietnam told
us that they had no knowledge of any political persecution among
returnees.  Appendix IV describes UNHCR's returnee monitoring
process. 


--------------------
\20 UNHCR was a party to the Memoranda of Understanding involving
Thailand, Malaysia, Indonesia, and the Philippines:  UNHCR was not a
party to Vietnam's Memorandum of Understanding with Hong Kong and
United Kingdom. 


      MONITORING EFFORTS
      IDENTIFIED NO EVIDENCE OF
      PERSECUTION AMONG RETURNEES
---------------------------------------------------------- Letter :6.1

Both UNHCR and British monitoring officials told us they had found no
evidence of persecution among the returnee population.  We also met
with officials of the U.S.  Embassies in Hanoi and Bangkok and
Immigration and Naturalization Service officials who administer the
Orderly Departure Program; the European Union and several
nongovernmental organizations who were conducting returnee assistance
programs; Asia Watch; and numerous individuals with personal
knowledge of returnee cases who had traveled independently in
Vietnam.  They told us they were not aware of any instance of
persecution among the returnees. 

In a limited test of the status of returnees, we examined 12 cases
referred to us by Boat People S.O.S.  and Asia Watch as examples of
persecution or cases for special concern.  Our discussions on these
cases with UNHCR monitoring officers and examination of UNHCR's case
files indicated that none of the individuals had been subjected to
persecution by the government related to their departure.  According
to UNHCR monitoring officers, three of the individuals were in prison
for predeparture criminal offenses, such as murder and armed robbery. 
Three others were briefly arrested for criminal activities but were
released from prison and had re-integrated well, as had the other
individuals. 

In addition, we interviewed 22 returnees from northern and southern
Vietnam and discussed their experiences since returning.  All of the
returnees told us that they had not experienced problems with
government officials since their return.  Our discussions indicated
that all had assimilated well and had been welcomed back into their
communities.  The returnees' most common complaints were economic
difficulties and establishing family registration when attempting to
relocate to new provinces. 

The absence of protection-related problems for returnees does not
mean that there are not concerns over the Vietnamese government's
human rights conduct.  Political opposition is not tolerated by
government authorities, even in the present mood of reforms. 
According to Asia Watch, those who publicly question the authority of
the Communist Party have been detained and imprisoned, be they
proponents of multiparty democracy, advocates of civil and political
rights, or religious leaders seeking greater autonomy from official
control.  UNHCR officials told us that returnees who were forcibly
returned or were politically active and outspoken in the camps of
first-asylum often faced extended questioning by government security
officials upon return.  Though a matter of concern, an Asia Watch
official told us that in their opinion these interrogations do not
rise to the level of persecution.  We were told by UNHCR officials
that the government questioning ends soon after the individuals'
return to their homes. 


      MONITORING COSTS
---------------------------------------------------------- Letter :6.2

UNHCR's monitoring effort in Vietnam is unprecedented:  There has
been no other refugee situation where UNHCR has conducted individual
case monitoring.  UNHCR has expended $8.8 million, including an
estimated $1.6 million for 1996, to maintain its monitoring operation
in Vietnam. 

In addition, between 1993 and 1995, UNHCR allocated $34.4 million to
support the economic and social reintegration of returnees.  This
assistance included cash grants of $240 to $360 to each returnee, the
operation of two reception centers, in-country travel costs of
returnees, and additional assistance to unaccompanied minors and
other vulnerable returnees. 

The assistance also included approximately $9 million in micro-
projects (such as the construction of schools, health clinics, water
delivery systems, and bridges).  The micro-projects were designed to
benefit local populations that absorb the returnees as well as
returnees themselves.  UNHCR views the projects as an important
mechanism for cultivating goodwill among local officials who assist
in the monitoring effort and prevents resentment among the local
populations who are often jealous of returnees' reintegration
allowances. 


   UNHCR AND U.S.  CPA
   EXPENDITURES
------------------------------------------------------------ Letter :7

In annual funding appeals for the CPA program, UNHCR requested
voluntary contributions of about $640 million between 1990 and 1995. 
It received substantially less in contributions, expending an
estimated $444 million in Special Program (CPA) and General Program
(regional) funds between 1989 and 1995.  (See table 3.)



                                     Table 3
                     
                        Schedule of UNHCR CPA Budgets and
                              Expenditures (1989-95)

                              (Dollars in millions)

             1989     1990     1991     1992     1993     1994     1995    Total
--------  -------  -------  -------  -------  -------  -------  -------  =======
Budgets:
Special         -   $69.56   $90.49  $102.41   $85.97   $78.41   $67.03  $493.87
 programs
General         -    38.94    28.15    28.47    24.82    14.97    10.51   145.86
 programs
================================================================================
Total           -   108.50   118.64   130.88   110.79    93.38    77.54   639.73
Expendit
 ures:\a
Special    $12.25    44.64    58.88    58.75    55.24    50.78    34.91   315.45
 programs
General     33.54    20.79    19.86    19.06    14.81    10.12  10.51\c   128.69
 program
 s\b
================================================================================
Total      $45.79   $65.43   $78.74   $77.81   $70.05   $60.90   $45.42  $444.14
                                                                              \d
--------------------------------------------------------------------------------
\a According to UNHCR officials, CPA expenditures were limited to
available voluntary contributions. 

\b We excluded general program expenditures in Thailand because
UNHCR's program in Thailand includes various refugee-related
activities in addition to CPA activities, and UNHCR did not have a
breakout of General Program expenditures for CPA versus other
activities. 

\c Budget estimate.  Complete expenditure data were not available at
the time of our review. 

\d These figures exclude the costs of UNHCR monitoring activities in
Vietnam. 

Table 4 shows U.S.  contributions of $150.83 million to UNHCR's
General and Special Programs in support of the CPA during fiscal
years 1990-96. 



                                      Table 4
                      
                      Schedule of U.S. Contributions to UNHCR
                       CPA Programs for Fiscal Years 1990-96,
                                as of February 1996

                               (Dollars in millions)

        1990      1991      1992      1993      1994      1995      1996     Total
--  --------  --------  --------  --------  --------  --------  --------  ========
Sp    $11.54    $18.10    $22.35    $16.98    $12.00     $8.50    $10.00    $99.47
 e
 c
 i
 a
 l
 p
 r
 o
 g
 r
 a
 m
 s
Ge        \b      5.13     13.62      9.00     12.61     11.00        \b     51.36
 n
 e
 r
 a
 l
 p
 r
 o
 g
 r
 a
 m
 s
 \
 a
==================================================================================
To    $11.54    $23.23    $35.97    $25.98    $24.61    $19.50  $10.00\c   $150.83
 t
 a
 l
----------------------------------------------------------------------------------
\a According to State officials, contributions to UNHCR's General
Program were for Asia, and a portion could have been used for
elements of other UNHCR programs in the region. 

\b The State Department made no general program contributions in
support of the CPA during these periods. 

\c The State Department stipulated that none of the 10 million fiscal
year 1996 contribution could be used for any movements of asylum
seekers under involuntary repartriation programs. 

The United States also contributed $9.05 million to various
nongovernmental organizations for CPA-related activities between
fiscal years 1992 and 1996.  Recipient organizations included

  -- Save the Children Federation ($3.40 million),

  -- World Vision ($1.46 million),

  -- International Catholic Migration Commission ($2.23 million),

  -- Southeast Asia Resource Action Center ($1.77 million),

  -- Hmong National Development ($0.13 million), and

  -- InterAction ($0.06 million). 


   AGENCY COMMENTS
------------------------------------------------------------ Letter :8

The Department of State and UNHCR generally agreed with this report. 
(UNHCR comments are reprinted in app.  V; State did not provide
written comments.) The Immigration and Naturalization Service (INS)
provided technical clarifications (which we incorporated as
appropriate), but did not provide written comments. 


   SCOPE AND METHODOLOGY
------------------------------------------------------------ Letter :9

We conducted our review at the Department of State and INS in
Washington, D.C., and at UNHCR Headquarters in Geneva, Switzerland;
in Hong Kong and Indonesia; and in Vietnam. 

At all these locations, we examined available program records and
files and interviewed knowledgeable officials involved with the
program.  We did not examine in depth all of the CPA first-asylum
countries' programs but limited our review to Hong Kong and
Indonesia.  We selected Hong Kong because of the large asylum seeker
population, which included Vietnamese from both the North and South,
and Indonesia because its distinctly southern Vietnamese asylum
seeker population was representative of the other first-asylum
country populations, and because it was one of the countries with
alleged corruption in the screening process. 

Our review of the refugee status determination process was limited in
that we were not permitted to review Hong Kong or Indonesian
government files, nor did we interview asylum seekers in these
countries.  However, we did review documents contained in UNHCR
files, and took notes on them.  UNHCR also provided us with written
summaries of the cases.  We do not believe the information contained
in UNHCR's files was sufficient for us to make conclusive,
independent determinations on the asylum seekers' refugee status. 
However, we determined whether the decisions made by Hong Kong,
Indonesia, and UNHCR officials on the cases appeared reasonable based
upon international refugee adjudication standards and information
available to us in the files.  We assessed the cases on these bases. 

Our scope was limited to examining how the CPA was implemented in
these countries and did not include an examination of subsequent
resettlement activities involving UNHCR and the resettlement
countries.  This included resettlement activities in which the U.S. 
Department of State and the INS may have been involved. 


      STATUS DETERMINATION PROCESS
---------------------------------------------------------- Letter :9.1

We took several steps to assess whether CPA refugee status
determination procedures were implemented in accordance with
international standards and criteria.  We interviewed and reviewed
documents and testimonies by critics of the asylum screening
process--Lawyers Committee for Human Rights, Refugees International,
United States Catholic Conference, Refugee Concern Hong Kong, among
others, to obtain an understanding of the problems they alleged were
prevalent.  To obtain an understanding of the program and its alleged
weaknesses, we interviewed UNHCR officials currently involved in the
CPA program and numerous officials who had first-hand knowledge of
the status determination process from their tours of duty in Geneva
and the countries of first-asylum during the early stages of the CPA. 
In these meetings we obtained detailed information on each step of
the screening process, including training of first-asylum country
screening officials, prescreening counseling provided asylum seekers,
appeals counseling, case monitoring by UNHCR, and other procedures to
ensure the fairness of the process. 

To obtain U.S.  and international refugee determination standards and
criteria, we interviewed U.S.  and UNHCR officials and examined
pertinent U.S.  legislation and U.N.  documents, which establish
refugee determination criteria and describe adjudication procedures. 
The latter included INS Refugee Processing Guidelines and the UNHCR
Handbook on Procedures and Criteria for Determining Refugee Status. 

In Hong Kong, we met with government officials responsible for each
phase of the status determination process and obtained pertinent
documents descriptive of the process.  In these meetings, we inquired
into how the screening process was designed and implemented and
evaluated those processes against criticisms levied by
nongovernmental organizations that represent asylee interests.  We
met with a prominent local attorney and an official from the refugee
legal advocacy organization, LAVAS, who provided legal counsel to
asylum seekers to obtain their insights, evaluations, and criticisms
of the screening process. 

As a limited test of both screening procedures and screening
decisions, we examined 10 cases represented by the Lawyers Committee
for Human Rights (five cases) and Boat People S.O.S.  (five cases) as
having been wrongly screened out.  We reviewed these cases from both
the merits of the claim and the soundness of the process perspective. 
However, our access to case files was limited.  The government of
Hong Kong denied us access to its screening files.  UNHCR officials
reviewed their case files with us and allowed us to take notes.  They
also provided us with written summaries of the cases.  They did not
permit us to make copies of material contained in the case files. 

UNHCR files did not contain transcripts or detailed interview
write-ups of the first-instance screening interviews by Hong Kong
screening officials or transcripts from the Refugee Status Review
Board.  Also, we did not interview asylees.  However, UNHCR's case
files contained the Hong Kong authorities' decisions and the stated
reasons for them; the asylum seekers' biographical profiles;\21
written appeals of first-instance screening decisions by the asylum
seekers and/or their UNHCR-provided or private legal counsellors; and
statements by UNHCR officials reviewing the cases for mandate status. 
The latter included the asylum seekers' claims, summaries of the
applicants' claims made to the Hong Kong authorities and their
decisions, and UNHCR's independent assessments and recommendations
regarding mandate status.  The files also sometimes contained other
documentation, such as reports from UNHCR monitors in Vietnam
verifying some aspect of asylum seekers' claims. 

In Indonesia, we met with government, UNHCR, and U.S.  Embassy
officials, the chairman of the refugee camp committee, and other
individuals with first-hand experience with the screening process in
order to determine how the process was designed and implemented.  To
test the screening process, we examined 10 cases represented by Boat
People S.O.S.  as examples of cases wrongly screened out due to
corruption or inappropriate application of refugee criteria.  In
conducting our case reviews, we reviewed documentation made available
to us by UNHCR and discussed the merits of the cases and the
adjudication process with UNHCR officials.  Similar to Hong Kong, our
access to case files was limited.  As in Hong Kong, UNHCR officials
in Indonesia reviewed their case files with us and allowed us to take
notes.  They also provided us with written summaries of the cases. 
The Indonesian government denied us access to their screening files
and UNHCR's files did not contain P3V documents or documents related
to the appeals process.  Therefore, we could not make definitive
determinations on the accuracy of the Indonesian authorities'
screening decisions.  We reviewed the decisions to determine whether
they appeared reasonable based on information in the UNHCR case files
and UNHCR reviewers' assessments of the claims. 


--------------------
\21 The profiles contain such information as the asylee's family
history; school, work and military experience; periods of time in
re-education camps or New Economic Zones; and details supporting
claims of persecution. 


      ALLEGATIONS OF CORRUPTION
---------------------------------------------------------- Letter :9.2

To develop information about alleged corruption in the CPA program,
we interviewed State Department, UNHCR, and nongovernmental
organization officials, and other persons knowledgeable of the
screening process and camp life.  From Boat People S.O.S., we
obtained documents, affidavits, and reports that Boat People S.O.S. 
represented to us as proof of corruption in the screening process. 
We did not verify the authenticity of the information provided to us
by Boat People S.O.S.  We reviewed documents and files at the
Department of State and at UNHCR offices in Geneva and Indonesia for
reporting on corruption in an attempt to determine what actions were
taken to respond to corruption charges.  We also interviewed seven
former residents of first-asylum camps in the region who said they
had first-hand experience with corruption in their status
determination proceedings. 

In Indonesia, we examined seven cases represented to us by Boat
People S.O.S.  to have been wrongly screened out due to corruption in
order to determine whether (1) the individuals presented strong
refugee claims and (2) there were any indications of corruption.  We
also reviewed three other cases represented to us by Boat People
S.O.S.  to have been improperly screened out.  As in Hong Kong, we
did not interview the asylum seekers whose cases we examined. 

To obtain additional insight into corruption allegations, we
interviewed a UNHCR official who conducted a 15-month investigation
on corruption in Galang camp.  We also examined the results of a
series of UNHCR case reviews conducted in 1995, which included cases
handled by UNHCR's Indonesian legal consultants alleged to have
engaged in corruption.  While UNHCR officials did not provide us the
full reports or case files on their 1995 reviews, they discussed the
results with us and provided us a brief summary of the review.  We
were denied access to Indonesian government files, and high level
Indonesian government officials we met with denied the existence of
corruption in the screening process. 


      TREATMENT OF RETURNEES TO
      VIETNAM
---------------------------------------------------------- Letter :9.3

To determine whether those who returned to Vietnam were subject to
persecution by the government, we examined UNHCR's program for
monitoring returnees and the skills, access, and independence of
their monitors.  We met with U.S.  Embassy officials in Bangkok and
Hanoi and officials from the INS who administer the Orderly Departure
Program and were familiar with conditions in Vietnam.  We inquired
into their freedom of movement and access to returnees in Vietnam and
whether the government of Vietnam was abiding by its commitments not
to persecute returnees.  We met with the human rights group, Asia
Watch, to obtain names of returnees of concern to them and reviewed
that organization's reporting on Vietnam.  We met with officials from
the British Embassy in Hanoi who monitor persons forcibly returned to
Vietnam from Hong Kong and obtained their assessment of the
Vietnamese government's treatment of this caseload.  We also met with
officials from several nongovernmental organizations who work with
the returnee population, and with numerous individuals who had
traveled independently within Vietnam and had personal knowledge of
some returnee cases to determine how the returnees were faring as a
whole and whether they were aware of any instances of persecution. 

While in Vietnam, we interviewed 22 returnees from both the north and
south of Vietnam to determine what challenges they have faced since
returning home and whether they have experienced any problems from
the government.  We also observed UNHCR officials as they (1) met an
Orderly Return Program flight from Hong Kong and (2) conducted a
monitoring mission to observe their freedom of movement, access to
returnees, and whether the presence of Vietnamese authorities
inhibited the candidness of returnees' statements.  We also examined
UNHCR's returnee monitoring database and obtained information on 12
returnee cases presented to us from Boat People S.O.S.  and Asia
Watch as examples of persons being persecuted or of special concern. 


      COST DATA
---------------------------------------------------------- Letter :9.4

To determine costs of the CPA, we obtained financial data from the
UNHCR and the Department of State.  We did not verify the accuracy of
the data provided. 

We conducted our review from November 1995 to August 1996 in
accordance with generally accepted government auditing standards. 


---------------------------------------------------------- Letter :9.5

We are sending copies of this report to the Chairmen and Ranking
Minority Members of the House and Senate Committees on
Appropriations, the House Government Reform and Oversight Committee,
the Senate Governmental Affairs Committee, and other interested
committees; the Secretary of State and the Attorney General; the
United Nations High Commissioner for Refugees; and others upon
request. 

If you or your staff have any further questions concerning this
report, please contact me at (202) 512-4128.  Major contributors to
this report were David R.  Martin, Assistant Director, and Patrick A. 
Dickriede, Senior Evaluator. 

Sincerely yours,

Harold J.  Johnson, Associate Director
International Relations and Trade Issues


THE COMPREHENSIVE PLAN OF ACTION
STIPULATED NATIONAL PROGRAMS USING
INTERNATIONAL SCREENING STANDARDS
=========================================================== Appendix I

The Comprehensive Plan of Action (CPA) agreement called for a
consistent, regionwide refugee status-determination process to be
conducted by first-asylum countries' national authorities in
accordance with established international refugee criteria and
procedures.  The United Nations High Commissioner for Refugees
(UNHCR) was to participate in the process in an observer and advisory
capacity and was to institute a comprehensive regional training
program for the national officials involved in the process. 

The refugee screening criteria to be followed were those recognized
in the 1951 Convention relating to the Status of Refugees and its
1967 Protocol, bearing in mind the 1948 Universal Declaration of
Human Rights and the special circumstances and needs of the asylum
seekers.  The UNHCR Handbook on Procedures and Criteria for
Determining Refugee Status was to serve as an authoritative and
interpretative guide in developing and applying CPA screening
criteria.\1

The UNHCR Handbook quotes the 1951 Convention definition of a refugee
as any person who

     .  .  .  owing to well-founded fear of being persecuted for
     reasons of race, religion, nationality, membership of a
     particular social group, or political opinion is outside the
     country of his nationality and is .  .  .  unwilling to avail
     himself of the protection of that country; or who, not having a
     nationality and being outside .  .  .  of his former habitual
     residence .  .  .  is unable or, owing to such fear, is
     unwilling to return to it. 

The Handbook includes detailed discussions of the international
instruments defining a refugee; the criteria for determining refugee
status (including exclusion clauses, special cases, and the principle
of family unity); and the procedures for determining refugee status
(including principles and methods for establishing the facts in a
case, and the application of benefit of the doubt). 

CPA screening procedures to be followed by the governments of the
countries of first-asylum were to be in accordance with those
endorsed by the Executive Committee of UNHCR, as follows: 

  -- The competent interviewing official shall have clear instruction
     for dealing with the case and should act in accordance with the
     principle of non-refoulement.\2

  -- The applicant should receive the necessary guidance as to the
     screening procedures to be followed. 

  -- There should be a clearly identified authority--whenever
     possible a single control authority--with responsibility for
     examining requests for refugee status and making a decision in
     the first instance. 

  -- The applicant should be given the necessary facilities,
     including the services of a competent interpreter, for
     submitting his or her case to the proper authorities.  The
     applicant should also be given the opportunity to contact a
     UNHCR representative. 

  -- If recognized as a refugee, the applicant should be informed
     accordingly and provided written documentation certifying his or
     her refugee status. 

  -- If not recognized as a refugee, the applicant should be given a
     reasonable time to appeal for a formal reconsideration of the
     decision. 

  -- The applicant should be permitted to remain in the first-asylum
     country pending a decision on his or her initial request, and
     any appeal to higher authority. 


--------------------
\1 Basic procedures for determining refugee status were endorsed by
the Executive Committee of UNHCR, 28th Session, 1977.  At the same
session, the Executive Committee requested that UNHCR consider
issuing a handbook for the guidance of governments, relating to
procedures and criteria for determining refugee status.  The UNHCR
Handbook was initially published in 1979. 

\2 Participating states shall not expel or return a refugee to the
frontiers or territories where his or her life or freedom would be
threatened on account of race, religion, nationality, membership in a
particular social group, or political opinion. 


      UNHCR'S RESPONSIBILITIES
      UNDER THE CPA
------------------------------------------------------- Appendix I:0.1

The Office of the United Nations High Commissioner for Refugees was
established by statute adopted by the U.N.  General Assembly on
December 14, 1950, as an annex to U.N.  Resolution 428 (v).  The
statute stipulated that the work of the High Commissioner is
humanitarian and social in nature and of an entirely nonpolitical
character and that the High Commissioner is to provide international
protection, under the auspices of the United Nations, to refugees
falling within the competence of his or her office.  In establishing
the Office, the General Assembly called upon governments to cooperate
with the High Commissioner in the performance of the Office's
functions concerning refugees falling under its competence.  While
recognizing that the Office's effectiveness depended upon the
cooperation of U.N.  member and nonmember states, the statute gave
UNHCR no directional or enforcement authority over them.  The statute
states that the High Commissioner should follow policy directives of
the General Assembly or the Economic and Social Council.  The
Executive Committee of the High Commissioner's Program, a 49
member-state advisory committee on refugees, approves and supervises
the material assistance program of the High Commissioner's Office and
advises the High Commissioner on the implementation of the
Commissioner's functions under the statute. 

UNHCR's responsibilities under the CPA were consistent with those
envisioned by the Office's enabling statute.  The agreement charged
UNHCR with coordinating the CPA's implementation with involved
countries and devising procedures for monitoring its progress.  The
CPA stipulated that first-asylum countries were to be responsible for
screening asylum seekers for refugee status, and that UNHCR was to
participate in the process in an observer and advisory capacity. 
UNHCR's responsibilities in the screening process were to include: 

  -- assisting the first-asylum countries in designing and
     implementing their screening processes,

  -- providing comprehensive training for national officials involved
     in asylum seeker screening, and

  -- advising asylum seekers about the screening processes and
     procedures. 

UNHCR was also charged under the agreement with coordinating with
first-asylum and third countries the timely resettlement of those
found to be refugees and with administering a safe, dignified
repatriation program for those found to be nonrefugees. 

A UNHCR senior legal officer in Geneva told us that, with regard to
refugee status determinations, the CPA's authors recognized that the
first-asylum countries were sovereign decisionmakers, and the CPA
authors did not expect UNHCR to ensure a proper decision in every
case.  The CPA's authors, he said, did not view UNHCR as the final
arbiter in refugee determinations, and UNHCR was not the
decision-making body.  As previously noted, the Statute of the Office
of High Commissioner authorizes UNHCR to make its own determinations
about individuals' refugee status.  This authority is commonly
referred to as mandate authority.  In what was essentially an
additional procedure in the refugee status determination process,
UNHCR could examine screened out asylum seeker cases for mandate
(refugee) status.  The CPA agreement did not make specific reference
to UNHCR's mandate authority, but Hong Kong nonetheless recognized
this authority.  The other countries of first-asylum had not formally
recognized this UNHCR authority as of June 1996, but some, such as
the Philippines and Indonesia, indicated to us that UNHCR's mandates
would be honored. 


      APPLICATION OF INTERNATIONAL
      AND U.S.  REFUGEE SCREENING
      STANDARDS CAN DIFFER
------------------------------------------------------- Appendix I:0.2

The U.S.  Immigration and Nationality Act\3 as a amended, defines a
refugee (in part) as

     .  .  .  any person who .  .  .  is unable or unwilling to
     return to .  .  .  [his or her country of nationality or
     habitual residence] because of persecution or [underscoring
     added] a well-founded fear of persecution on account of race,
     religion, nationality, membership in a particular social group,
     or political opinion .  .  .  . 

The Immigration and Naturalization Service's (INS) 1983 refugee
processing guidelines state that the burden of proof of refugee
status rests with the applicant.  According to INS' guidelines, the
applicable burden to be met by the applicant is a well-founded fear
of persecution, based on one of the reasons for persecution in the
refugee definition. 

In determining an applicant's refugee status, INS officers must
evaluate whether the applicant's statements and feelings lead to a
finding of well-founded fear of persecution based on their
consistency and credibility in light of known conditions and
practices in the country from which refuge is sought. 

In addition, Public Law 101-167\4 states that certain categories of
aliens (including Vietnamese who are members of categories of
individuals determined by the Attorney General to be targets of
persecution) may for purposes of refugee status determination
establish that they have a well-founded fear of persecution based on
one of the five reasons for persecution by asserting such a fear and
asserting a credible basis for concern about the possibility of such
persecution. 

Categories of Vietnamese found by the Attorney General to be of
interest to the United States include former South Vietnamese
government officials and members of the military; persons formerly
closely affiliated with the U.S.  or Western institutions; those sent
to re-education camps, or to New Economic Zones because they were
considered politically or socially undesirable; members of certain
ethnic or religious groups; and family members of the above. 

According to INS guidelines, these categories of aliens shared common
characteristics that identified them as targets of persecution in
their countries.  The facts surrounding the establishment of the
categories were sufficiently known and established to permit what
was, in essence, "judicial notice" to be taken without further proof. 
Thus, according to INS guidelines, if applicants clearly proved
themselves to come within one of the established categories they
would have also established themselves likely targets of persecution. 

INS guidelines state that if in interviewing an applicant the INS
officer is satisfied that he or she falls within a designated
category, a strong likelihood will have been established that the
applicant is qualified for refugee status.  The interviewing officer
should then obtain information regarding the applicant's actual
persecution or well-founded fear of persecution.  Statements
indicating that category applicants fled their country for fear of
persecution or due to actual acts of persecution, if credible, are
generally sufficient for findings of refugee status, according to INS
guidelines. 

Notwithstanding the view by some that the U.S.  refugee definition
fully replicates the international definition, both U.S.  and UNHCR
officials acknowledged that the definitions--and their
applications--can differ. 

According to senior INS officials, the international definition is
forward-looking "owing to well-founded fear of being persecuted,"
while the U.S.  definition can look to the past "persecution or a
well-founded fear of persecution."

Thus, under the U.S.  definition an applicant can be adjudicated a
refugee on the basis of past persecution.\5 Indeed, INS guidelines
state that refugee status may be based upon persecution suffered in
the past or upon the likelihood of future persecution.  The presence
of either is sufficient:  Both conditions are not required. 

According to UNHCR officials, refugee screening under international
standards assesses and takes into account past persecution, but also
when the persecution took place and ended, and the reasonable
likelihood of future persecution upon the individual's return to his
or her country of origin.  The officials stressed that while past
persecution is an element in establishing whether an individual has a
well-founded fear of persecution upon return, it is not sufficient to
establish refugee status.  They added, however, that past persecution
can be the basis for refugee status on humanitarian grounds when the
persecution was so egregious that the individual could not reasonably
be expected to resume a normal life upon return. 

UNHCR officials stated that changing conditions in asylum seekers'
countries-of-origin must also be taken into account in assessing
individuals' well-founded fear of persecution.  UNHCR's assessment of
conditions in Vietnam, based on reports from nongovernmental
organizations, governments, and UNHCR's monitors in Vietnam indicated
significantly improved conditions in the country.  UNHCR reports on
the CPA stated, for example, that what may have been a borderline
case in the past could well fail to establish a well-founded fear of
persecution now in light of improved conditions in Vietnam. 


--------------------
\3 Section 101(a) (42), codified as 8 U.S.C.ï¿½1101 (a) (42). 

\4 Public Law 101-167, Section 599D, 103 Statute 1261. 

\5 According to the INS' Basic Law Manual, which provides guidance on
refugee and asylum adjudications, an applicant for refugee status may
establish his claim by demonstrating past persecution in lieu of
evidence of a well-founded fear of persecution.  If it is determined
that an applicant has established past persecution, he shall be
presumed to have a well-founded fear of persecution.  If, however, a
preponderance of the evidence shows that conditions in the
applicant's country have changed to such an extent that the alien's
fear of being persecuted is no longer well-founded, an INS officer
may deny refugee status in the exercise of discretion if the officer
is satisfied that an alien who has suffered persecution in the past
is no longer in danger of persecution.  Nevertheless, if it is
determined that the applicant has demonstrated compelling reasons for
being unwilling to return arising out of the severity of the past
persecution, he may be granted refugee status--Matter of Chen, Int. 
Dec.  No.  3104 (BIA 1989). 


ASYLUM SEEKER SCREENING PROCESS IN
HONG KONG
========================================================== Appendix II

All Vietnamese migrants arriving in Hong Kong before June 16, 1988,
were automatically given refugee status, making them eligible for
resettlement.  However, according to Hong Kong government documents,
it became increasingly clear during the mid-1980s that the large
majority of arriving Vietnamese migrants did not have a well-founded
fear of persecution in Vietnam and thus were not entitled to refugee
status.  According to the documents, departures from Vietnam appeared
to be motivated principally by the desire for resettlement, rather
than asylum, and resettlement countries were becoming increasingly
reluctant to take people from Hong Kong.\1 The Hong Kong government
therefore decided that it could no longer grant automatic refugee
status, and on June 16, 1988, initiated refugee screening procedures. 
These procedures were amended in September 1988 following an
agreement with UNHCR, and again in June 1989, following
implementation of the CPA agreement.  Between March 1989 and December
1995, 60,275 asylum seekers in Hong Kong were screened for refugee
status under the CPA.  About 11,300 (18.8 percent) were screened in
and resettled.  The remaining 48,900 were determined not to be
refugees and were either voluntarily or involuntarily repatriated to
Vietnam or remained in Hong Kong asylum camps. 

Hong Kong government and UNHCR officials acknowledged mistakes in the
implementation of Hong Kong's CPA processing procedures, particularly
early in the process, but said the process improved as it matured. 
We also noted errors.  However, we believe the process was generally
implemented in accordance with CPA guidelines. 


--------------------
\1 During the 1980s, the balance between north and south Vietnamese
arriving in Hong Kong began to change, with the proportion of
northerners steadily increasing.  In 1984, 28 percent came from the
northern part of Vietnam, whereas by 1989 the proportion from the
north had increased to 86 percent. 


   HONG KONG'S PROCEDURES AND
   CRITERIA WERE CONSISTENT WITH
   CPA GUIDANCE
-------------------------------------------------------- Appendix II:1

We interviewed UNHCR and Hong Kong government officials and reviewed
various documents they provided us in developing a description of
Hong Kong's refugee screening procedures.  Refugee screening in Hong
Kong was a multitiered process involving both government bodies and
UNHCR.  The process included initial screening and an appeals process
by the government, mandate review by UNHCR, and possible judicial
review through the Hong Kong courts.  The process is depicted in
figure II.1. 

   Figure II.1:  Screening
   Procedures in Hong Kong

   (See figure in printed
   edition.)

According to Hong Kong government reports, asylum seekers were
informed upon interception in Hong Kong waters that they were
entering the territory illegally and if they stayed would be subject
to screening to determine their refugee status.  They were told that
those accorded refugee status would be resettled and those found not
to be refugees would be detained pending repatriation. 

Prior to initial, or first-instance, screening, both the Hong Kong
Immigration Department and UNHCR were to provide asylum seekers
orientation and process familiarization interviews and materials. 
UNHCR legal consultants and Australian lawyers (who were working
under a Jesuit Refugee Service project) were to provide asylum
seekers legal advice and assistance prior to first-instance
screening.  Private Hong Kong attorneys entitled to practice in Hong
Kong, when hired by asylum seekers, could also provide legal advice
at this stage. 

At first-instance screening Immigration Department assistants, using
UNHCR-approved questionnaires, interviewed asylum seekers to collect
biographical information.  This included the individuals' personal
and family histories; military service data; involvement in political
parties or resistance organizations; religious activities; any denial
of economic and social rights; prosecution; and motives for leaving
and not wishing to return to Vietnam.  There was also provision for
any additional points or comments the individual wished to make. 
Immigration officers, assisted by government-provided interpreters,
interviewed the asylum seekers, using the completed questionnaires as
baseline data, and recorded their claims to refugee status.  The
officers made preliminary refugee status determinations from this
data.  These decisions were reviewed by senior immigration officers
before the applicants were notified of the screening results. 
Notices of determination explaining the decisions were to be given
all applicants.  If screened in as refugees, the applicants were
moved to a refugee camp to await resettlement.  If screened out,
applicants were advised of their right to appeal. 

Screened out applicants could appeal to the Refugee Status Review
Board, a body composed of government and nongovernment officials. 
The review process was to be initiated by submitting the cases to the
Review Board within 28 days of receipt of the initial status
determination. 

On the day applicants were notified of negative determinations,
copies of their files were to be given to UNHCR legal
consultants--Agency for Volunteer Service counselors,\2 who met and
advised them on the merits of their cases.  If the counselors
believed particular claims warranted reconsideration, the counselors
prepared the submissions to the Review Board.  If not, they provided
applicants counseling and guidelines for submitting their own written
submissions.  According to UNHCR and Review Board officials, Agency
for Volunteer Service consultants filed about 25 percent of the
approximately 49,000 appeals filed. 

Private Hong Kong attorneys could also file appeals on behalf of
their clients.  Review Board officials estimated that about 8 percent
of those whose cases were not appealed by UNHCR counselors retained
private attorneys who submitted their appeals for them. 

The Review Board was organized into two-person panels, each of which
heard cases.  No legal representation or oral evidence could be given
at the Review Board, although according to the Board chairman about
25 percent of the asylum seekers making appeals were re-interviewed
by Board members.\3 A positive decision by either panel member could
overturn negative Immigration Department decisions. 

The Review Board provided the asylum seekers written notification of
its decisions, with reasons for decisions.  If screened in at this
review, the asylum seekers were transferred to refugee camps to await
resettlement.  If screened out, they were informed that they would be
permitted to remain in Hong Kong pending their repatriation. 

According to a representative of the Hong Kong Attorney General's
Office, under Hong Kong administrative law asylum seekers could seek
judicial review of the Review Board's negative decisions in the Hong
Kong courts if they believed the decisions were unreasonable or
illegal, or involved procedural improprieties.  Judicial review was
not a review of the cases' merits but, rather, the procedures
employed in deciding them. 

Asylum seekers apply for judicial review of their cases through the
Hong Kong Legal Aid Department, which determines the merits of the
applicants' claims.  The Department can reject the applications,
request the Refugee Status Review Board to reconsider (rescreen) the
cases, or refer them to the courts for judicial review.  According to
a Legal Aid Department official, 3,651 asylum seekers had applied for
assistance as of July 1996.  Of that number, 450 had been granted
Legal Aid Department assistance, and another 1,354 applications were
under active consideration. 

The Legal Aid Department official could not provide us the number of
cases rescreened by the Review Board or heard by the courts. 
According to the official, those statistics were not kept.  However,
the official said that few cases actually go to court as the
Department usually processes deserving cases through the Review
Board.  According to the Attorney General's Office representative we
spoke with, there was only one instance where the court ordered a
decision by the Review Board set aside.  In that case, the court
ruled that both the first-instance and Review Board decisions were
flawed by procedural irregularities.\4

In what became essentially another level of review, the Hong Kong
government recognized UNHCR's right to grant refugee status under its
own authority.\5 This meant that in cases the Hong Kong government
rejected but UNHCR considered meritorious, UNHCR was able to
recognize persons as refugees under its own mandate.  UNHCR viewed
recognition of its mandate as an important safety net for ensuring
that no persons with valid claims were improperly screened out and
returned to Vietnam. 

According to UNHCR officials in Hong Kong, all screened-out
applicants' cases, about 46,000, were examined for the possibility of
mandate and about 11,000 were selected for mandate review.  UNHCR
granted mandates (refugee status) to about 1,550.\6


--------------------
\2 The Agency for Volunteer Service was a local nongovernmental
organization created and financed by UNHCR to provide legal counsel
to denied applicants. 

\3 UNHCR could monitor the re-interviews. 

\4 The individual was later rescreened, and was screened in, by Hong
Kong authorities. 

\5 The Statute of the Office of the UNHCR authorizes UNHCR to make
its own determinations about refugee status.  This authority is
commonly referred to as mandate authority. 

\6 As of January 1996. 


ASYLUM SEEKER SCREENING PROCESS IN
INDONESIA
========================================================= Appendix III

Refugee screening in Indonesia was a two-phased process that met
basic international structural screening standards but did not
contain some of the checks and balances we found in the Hong Kong
process.  UNHCR played a major role in the process, although the
Indonesian authorities retained decision-making responsibilities. 


   INDONESIA'S PROCEDURES AND
   CRITERIA WERE CONSISTENT WITH
   CPA GUIDANCE
------------------------------------------------------- Appendix III:1

Refugee screening in Indonesia began in September 1989 and ended in
September 1993.  All asylum seekers arriving after March 17, 1989,
were screened for refugee status.  (Those arriving before that time
were automatically accorded refugee status and the opportunity to
resettle in third countries.) The process in Indonesia was
essentially a two-stage (first-instance and appeals) system,
involving both Indonesian government authorities and UNHCR.  This
process is depicted in figure III.1. 

   Figure III.1:  Screening
   Procedures in Indonesia

   (See figure in printed
   edition.)

First-instance screening was conducted at Galang, the island asylum
camp located several hundred miles north of Jakarta, the capital. 
Upon arrival at the camp, asylum seekers registered and provided
basic biographical information to P3V, the Indonesian
inter-ministerial task force created to deal with Vietnamese boat
people.  Later, UNHCR staff provided group prescreening counseling
and materials and explained screening procedures and criteria.  UNHCR
staff, assisted by Vietnamese camp volunteers, then administered a
more comprehensive questionnaire, designed to elicit more detailed
information about the individuals' claims for refugee status.  We did
not evaluate the adequacy of the information provided to asylum
seekers during prescreening counseling.  However, an independent
observer with long experience at Galang told us UNHCR provided the
asylum seekers ample information on the screening process and that
asylum seekers were well prepared for the screening process. 

At first-instance screening, preliminary interviews were conducted by
UNHCR legal consultants, attorneys trained in international refugee
determination procedures and criteria.  The consultants were assisted
by UNHCR interpreters.  The consultants prepared written case
evaluations and screening recommendations, on the basis of the
applicants' biographical data and the interviews.  The consultants
worked under the supervision of a senior UNHCR legal officer, who was
responsible for reviewing the consultants' case evaluations. 

The asylum seekers' files, containing their biographical data, the
UNHCR questionnaires, and the consultants' evaluations and screening
recommendations were then passed to P3V.  Teams of P3V officials,
consisting of an interviewing officer and an interpreter, interviewed
the asylum seekers to assess their credibility and verify the
accuracy of the information in the UNHCR consultants' preliminary
interviews.  (UNHCR representatives were free to monitor the
interviews, but only spot checked them, according to UNHCR
officials.) The interviewing officers then prepared the
first-instance screening decisions. 

The UNHCR consultants' interviews were, in effect, the substantive
first-instance screening interviews.  According to UNHCR officials,
P3V officials placed strong credence in the consultants' evaluations,
and rarely disagreed with their recommendations.  A UNHCR official
told us that of the more than 3,000 cases screened positive by UNHCR
at first-instance, P3V disagreed with the consultants'
recommendations in only 22 cases.  A UNHCR official informed us that
some P3V interviewing officers asked only a few questions of
applicants, relying on the consultants' evaluations in reaching their
decisions. 

Written screening decisions were served to each applicant. 
Screened-in cases were processed for resettlement in third countries;
those screened out were informed of their right to submit a written
appeal. 

Screened out applicants had 15 days from the date of notification to
appeal their decisions.  UNHCR legal consultants counseled
applicants, mostly in group sessions, on how to submit their appeals. 
UNHCR-trained camp-resident volunteers assisted applicants in
drafting their appeals.\1

Applicants' appeals were considered by the inter-ministerial Review
Committee in Jakarta.  The Committee based its deliberations on the
applicants' appeals submissions and case files, which had been
forwarded from Galang.  It did not reinterview asylum seekers,
according to UNHCR officials.  The Committee could uphold or reverse
negative first-instance decisions depending upon whether it believed
the appeals submissions provided additional credible information and
whether the cases had been screened in conformance with international
refugee status determination criteria.  Most appeals were decided at
the Review Committee level.  These decisions were considered final;
however, according to UNHCR, many of the positive decisions were
delivered to the applicants as late as 1 year or more after the
decision was made.  This was contrary to the CPA's guidance
concerning prompt notification, and, according to a UNHCR program
officer, provided an opportunity for corruption to occur. 

The Committee had the option of referring questionable or unclear
cases to a 10-member Appeals Board for final determination.  The
Committee referred only those cases on which it could not make a
clear determination. 

A UNHCR representative attended Review Committee and Appeals Board
meetings as an observer and advisor.  The representative was also
responsible for presenting UNHCR's views on cases at appeal.  UNHCR
officials involved in the screening process told us applicants' case
files and appeals were forwarded to the UNHCR office in Jakarta at
the same time they were sent to the Review Committee.  They said
UNHCR in Jakarta reviewed all asylum seekers' appeals before they
were discussed by the Review Committee to determine those with merit. 
They said that while P3V at Galang had almost always accepted UNHCR
positive screening recommendations at first instance, there were
occasionally cases that P3V denied despite the recommendations.  The
Galang consultants noted those cases for Jakarta's review.  The
officials told us they instructed their representative to present
UNHCR's views on the cases it felt strongly about to the Review
Committee and/or Appeals Board. 

According to UNHCR officials, UNHCR and the Review Committee or
Appeals Board reached consensus on every case UNHCR believed merited
refugee status.  The Indonesian authorities accepted UNHCR's positive
first-instance recommendations on all but 22 cases, screening out
those cases during the appeals process.  During the appeals process
UNHCR concurred with the Review Committee's and/or Board's negative
decisions.  Thus, according to UNHCR officials, because no
meritorious claims remained in question, UNHCR had no need to
exercise its mandate authority.  (Later, however, UNHCR mandated
seven cases as a result of its reviews of screening decisions.)


--------------------
\1 Indonesia did not permit nongovernmental organizations to
participate in applicant counseling, as the Hong Kong government did
with Agency for Volunteer Service counselors. 


UNHCR MONITORING OF RETURNEES TO
VIETNAM
========================================================== Appendix IV

The CPA charged UNHCR with administering the repatriation of asylum
seekers determined not to be refugees but did not specifically charge
the agency with the monitoring mission.  However, U.N.  Secretary
General Javier Perez de Cueller did so in a September 1990 letter to
the High Commissioner for Refugees, when he requested that the High
Commissioner, independent of the mandate as High Commissioner for
Refugees and on an exceptional basis, serve as the Secretary
General's Special Representative to coordinate and monitor the
returnees program to Vietnam.  In his letter, the Secretary General
noted that the CPA's success depended upon a balanced implementation
of all its aspects, including that of the return and reintegration of
persons determined not to be refugees, and that humanitarian
consideration argued strongly for the United Nation's continued
involvement in the matter. 

As of February 1996, UNHCR employed seven international staff as
monitoring officers (four based in Hanoi and three in Ho Chi Minh
City) and a contingent of national administrative and support staff
to implement its monitoring program.  All of the monitoring officers
were fluent in Vietnamese--several were also fluent in regional
dialects.  Five of the monitoring officers had worked in first-asylum
country programs prior to taking positions as monitors in Vietnam. 
None of the monitoring officers needed or used interpreters in their
work. 

The UNHCR monitoring system is built around and keyed to priority
cases, returnees of special significance or concern or those who
specifically request monitoring upon their return.\1 UNHCR offices in
the first-asylum countries are to notify the Hanoi office of pending
returnees, both priority and routine cases.  This information is to
be placed in returnee databases in UNHCR's Hanoi and Ho Chi Minh City
offices. 

Returnees, who usually arrived by air in Hanoi or Ho Chi Minh City,
were to be met by UNHCR officers and first-asylum country embassy
representatives,\2 in addition to Vietnamese authorities.  If the
flights comprised involuntary returnees from Hong Kong, they were
also to be met by British Embassy officials, as Hong Kong's
involuntary return program, called the Orderly Return Program, was
being carried out under a bilateral agreement between the British and
Hong Kong governments and Vietnam.  British Embassy monitoring
officers were to independently monitor Hong Kong's involuntary
returnees, including priority cases.  British and UNHCR officers
viewed their dual monitoring of these cases as positive reinforcement
of each other's efforts.  The British viewed their monitoring effort
as necessary to be consistent with their orderly return Memorandum of
Understanding with the Vietnam government, notwithstanding UNHCR's
monitoring program. 

Returnees were to be transferred from the airports to reception
centers, where they were to undergo registration and reintegration
processing by government agencies and foreign nongovernmental
organizations, such as Nordic Assistance to Returning Vietnamese. 
They were also to be visited by UNHCR monitors, who were to answer
any questions, pass out business cards with addresses and telephone
numbers where they could be reached,\3 and take names and addresses
of individuals who requested priority monitoring. 

From the returnee databases, UNHCR planned monitoring missions--field
trips designed to visit specific priority cases.  These cases were
supplemented with names of routine returnees in the area.  The
monitors notified the government authorities of their monitoring
mission destinations (by province, district, and village) and in the
south provided lists of all returnees in the provinces to be
visited.\4

According to UNHCR monitoring officials this approach had several
advantages.  It

  -- ensured government authorities of UNHCR's openness, while
     securing the assistance of local officials as guides in finding
     individual returnees;\5

  -- ensured systematic planning so that all priority cases would be
     visited;

  -- was more efficient than random visits; and

  -- did not specify individual priority cases to be visited. 

This approach also enabled UNHCR to visit with local government
officials and help ensure that local officials were adhering to
government commitments not to persecute returnees. 

Monitoring missions, which normally consisted of two monitoring
officers in the south and one in the north, were designed to contact
100 to 150 returnees per mission.  Monitoring officers had conducted
363 monitoring missions as of our visit in February 1996.  As they
traveled, the monitors stopped at villages to ask village leaders and
local citizens how local returnees were faring as a whole.  They also
visited outlying areas to ask similar questions.  The monitors then
visited individual returnees.  UNHCR monitors told us that through
these practices they gained a good knowledge about a much larger
proportion of returnees than the numbers visited would indicate. 

If the officers sensed a returnee was reluctant to speak candidly
they employed the technique of either occupying government officials
with one officer while the other took the individual aside, or simply
taking the returnee aside, to discuss the situation.  According to
UNHCR officials, as well as our observations while interviewing
returnees in Dong Nai province, government officials normally did not
attend the interviews but rather waited in places such as coffee
shops until the monitoring visits were completed.  Monitors, as well
as others with extensive experience in Vietnamese culture told us,
however, that Vietnamese speak their minds freely and would not be
reticent to voice their concerns of treatment in front of government
officials. 

The government's attitude and attention toward monitoring has evolved
over time; returnee monitoring initially drew considerable government
attention, according to UNHCR monitoring officers.  They said their
monitoring missions drew entourages of 10 to 12 persons, including
officials from the Ministry of Interior and the Ministry of Labor,
War Invalids, and Social Affairs; but, over time the government's
attitude grew more relaxed.  Over the last several years UNHCR
monitoring officers were usually accompanied only by an official from
the Labor and Social Welfare Department under the provincial Peoples'
Committee and the Labor Ministry.  These departments are UNHCR's
counterparts and hosts in the provinces.  They organize and
administer the distribution of UNHCR's individual cash grants and
support the planning and implementation of UNHCR's micro-projects. 



(See figure in printed edition.)Appendix V

--------------------
\1 Examples of priority monitoring include individuals who may have
been politically active in first-asylum camps or who may have been
facing prosecution or imprisonment for criminal acts committed before
their departure from Vietnam. 

\2 According to UNHCR officials, embassy representatives usually met
only Orderly Return Program, or involuntary returnee, flights. 

\3 Unrestricted communication in Vietnam has led to increased direct
contact by returnees with UNHCR monitoring officers.  In 1995, for
example, UNHCR's Ho Chi Minh City office had 1,168 walk-in visits. 

\4 In commenting on a draft of this report, UNHCR said that it does
not provide government authorities lists of returnees to be monitored
in the north. 

\5 According to UNHCR monitors, there are no street signs in many
Vietnamese villages, and individuals must be located by local
officials. 


COMMENTS FROM THE UNITED NATIONS
HIGH COMMISSIONER FOR REFUGEES
========================================================== Appendix IV



(See figure in printed edition.)



(See figure in printed edition.)


The following are GAO's comments on the United Nations High
Commissioner for Refugees letter dated September 9, 1996. 

GAO COMMENTS

1.  We agree that the CPA Steering Committee's March 1996
announcement reaffirmed language in the CPA that asylum seekers found
not to be refugees should be repatriated, and we revised our report
accordingly. 

2.  Hong Kong Immigration Department officials we spoke with stated
that refugee screening was conducted in accordance with UNHCR
guidelines and that no additional screening criteria beyond that
stipulated in the CPA were applied. 

3.  While our draft report referred to mandates resulting from
UNHCR's 1995 reviews, we modified the report to reflect the number of
mandates resulting from all UNHCR reviews in Indonesia. 

4.  We modified the report to incorporate this suggestion. 

*** End of document. ***