Refugee Status Appeals Authority 

REFUGEE APPEAL NO. 72668/01

AT WELLINGTON

Before:                                  R P G Haines QC (Chairperson)
                                            GJX McCoy QC (HK) (Member)

Representing the Appellant:    JS Petris and R Woods

Appearing for the NZIS:           M Hodgen and B Keith

Date of Hearing:                     23 July 2001

Date of Ruling:                       5 April 2002


RULING ON LEGAL ISSUES

INDEX

BACKGROUND

BURDEN OF PROOF

Meaning of burden of proof
Non-adversarial model preferred
Non-adversarial proceedings and the burden of proof
Burden of proof - international survey
Conclusions from international survey

BURDEN OF PROOF IN NEW ZEALAND

POSITION PRIOR TO 1 OCTOBER 1999

The Terms of Reference
The Authority’s case law under the Terms of Reference
Rising levels of abuse
Withholding of co-operation
The limits to the refugee inquiry
Refugee Appeal No. 523/92
Burden of proof and the Court of Appeal
The Immigration Amendment Act 1999

THE 1999 PROVISIONS

Relevance of the power to inquire
Authority not a Commission of Inquiry
Authority not obliged to inquire

THE CASES ON BURDEN OF PROOF

Butler
Ally Hassan Mohammed
C
Don
T

RELEVANCE OF NON-REFUGEE CASES

CONCLUSION ON THE BURDEN OF PROOF ISSUE

WELL-FOUNDED FEAR

Jurisprudence of the RSAA
An objective test
High Court cases
There is no balance of probabilities test
Conclusion on well-foundedness

THE MEANING OF PERSECUTION

CONCLUSION

BACKGROUND

[1] The appellant, a citizen of the Socialist Republic of Vietnam, arrived in New Zealand on 20 April 1997.  A year later, on 6 April 1998 he lodged a claim to be recognised as a refugee in New Zealand.  Unsuccessful at first instance, he appealed to this Authority, that appeal being heard on 15 February 2000 by a panel comprising C Parker and CM Treadwell.  In a decision delivered on 3 August 2000, the appeal was dismissed.

[2] On an application for judicial review in the High Court, Wellington the decision of the Authority was set aside and the appeal remitted back to the Authority for reconsideration: TN v Refugee Status Appeals Authority (High Court Wellington, CP 212/00, 10 May 2001, Chisholm J).

[3] The appeal was set down for rehearing on 11 July 2001.  On 4 July 2001 counsel for the appellant filed written submissions in support of the appeal.  Three potentially significant propositions of law were advanced:

(a) On the question of onus or burden of proof it was submitted that in terms of s 129P(1) of the Immigration Act 1987 the responsibility on a refugee claimant is no more than a responsibility to establish what the claim is.  The authority cited in support of this proposition was T v Refugee Status Appeals Authority [2001] NZAR 749 (Durie J);
(b) The term “persecution” in the Refugee Convention was to be given a “dictionary” meaning in preference to the “sustained or systemic violation of basic human rights” approach articulated by Professor James C Hathaway in The Law of Refugee Status (Butterworths, 1991) 104-105 and approved in DG v Refugee Status Appeals Authority (High Court Wellington, CP 213/00, 5 June 2001, Chisholm J);

(c) Notwithstanding the Authority’s well established jurisprudence which follows and applies authoritative decisions of the High Court of Australia, particularly Minister for Immigration and Ethnic Affairs v Guo (1997) 191 CLR 559, the Authority ought to accept that speculation is an element of the real chance test.  Reliance was placed on SWH v Refugee Status Appeals Authority (High Court Wellington, CP 203/97, 7 April 1998, Gallen J) where (it was claimed) the decision of the Full Court of the Federal Court of Australia in Guo was adopted in preference to the later decision of the High Court of Australia which actually reversed the Federal Court.

[4] Given the potentially far reaching implications of the submissions advanced in support of the appeal the Authority, with the consent of counsel for the appellant, proposed that the assistance of Crown Counsel be obtained on behalf of the New Zealand Immigration Service.  This would allow the Authority to have the benefit of full argument before making a decision.  It was anticipated that the issues would necessarily have to be dealt with comprehensively as it was conceivable that either the appellant or the Minister of Immigration would challenge the decision by way of judicial review.  The Authority has a duty to assist the High Court by addressing the issues in some depth.

[5] In the circumstances more fully described in the Minute of 11 July 2001, Mr Hodgen and Mr Keith advised that they would present submissions on behalf of the New Zealand Immigration Service.

[6] It was common ground that the role of Crown Counsel was to assist the Authority only in relation to the legal issues raised by the appellant.  In accordance with longstanding practice the Crown understandably did not wish to be heard on the ultimate issue, namely whether on the facts the appellant satisfies the definition of “refugee” in the Refugee Convention.

[7] To assist counsel to focus more meaningfully on the issues raised, the Authority in its Minute of 11 July 2001 identified in general terms the legal issues to be addressed.  They were:

First, the meaning and effect of s 129P(1) of the Immigration Act 1987.

Second, how the word “persecution” in the refugee definition is to be interpreted and applied.

Third, the interpretation of the “well-founded” element of the refugee definition.

Fourth, how the Authority, as an inferior tribunal, is to decide which conflicting line of High Court authority is to be followed.

Fifth, whether the decision of Durie J in T was decided per incuriam in that the decision does not address Butler v Attorney-General [1999] NZAR 205, 213 (CA).

 Whether it was intended that Part VIA of the Immigration Act 1987 (as inserted by the Immigration Amendment Act 1999, s 40) consolidate pre-existing law.

[8] In accordance with the timetable set by the Authority, written submissions were filed.  Oral argument took place on 23 July 2001.  The Authority has been greatly assisted by all counsel.

[9] The delay in delivering this ruling is very much regretted.  One of the members of the panel has been unavoidably absent overseas for a protracted period and in addition the time available to the panel to give the issues proper consideration has been reduced by the need to address the fraudulent abuse of the Authority’s jurisdiction by a large group of Thai nationals, a matter to which we will return.
 
 

BURDEN OF PROOF

Meaning of burden of proof

[10] As so much of the first part of this decision is concerned with the burden of proof, it would be as well that the Authority explains what it means by “burden of proof”.  It has always understood that in the refugee context the burden of proof means the burden of establishing the facts and contentions which support the claim that the individual meets the requirements of the refugee definition.  Compare 17 Halsbury’s Laws of England 4th ed, para 13.

Non-adversarial model preferred

[11] The procedure prescribed by Part VIA of the Immigration Act 1987 for determining refugee status in New Zealand is an administrative one, comprising a first instance decision by a refugee status officer followed by an appeal to the Refugee Status Appeals Authority.

[12] Both at first instance and on appeal the respective decision-makers are free, subject to the constraints imposed by the Act, the Immigration (Refugee Processing) Regulations 1999 (SR 1999/285) and to the requirements of fairness, to determine their own procedures: s 129G(7) and Schedule 3C, para 8.  The Authority also has the powers of a Commission of Inquiry under the Commissions of Inquiry Act 1908: Schedule 3C, para 7.  It is not bound by any rules of evidence: Schedule 3C, para 9(1).  The procedures at both levels are informal and non-adversarial.  They can be described as investigative or inquisitorial: Practice Note No. 2/99 (1 October 1999), para 6.1 and Refugee Appeal No. 70656/97 Re KB (10 September 1997).  This is the preferred model of refugee adjudication.  See for example W. Gunther Plaut, Refugee Determination in Canada (1985) 120-123 and Professor James C Hathaway, Rebuilding Trust - Report of the Review of Fundamental Justice in Information Gathering and Dissemination at the Immigration and Refugee Board of Canada (December 1993) 5-7.  The following passage has been taken from Gunther Plaut, Refugee Determination in Canada at 121-122:

In the refugee context, the “judicial” or “adversarial” model appears to be inappropriate.  In the United States, asylum hearings are of the strictly adversarial mode and the Immigration judges to whom I spoke were quite dissatisfied with them.  They felt that the system is not conducive to obtaining all the necessary information and that justice would be better served if all parties shared in the attempt to establish the facts rather than opposed one another.

Further, the adversarial system assumes that there are conflicting interests to be resolved by an impartial judge.  In refugee determination, there is not (or should not be) an “adversary” to the refugee.  There do not exist, as in a civil suit, two parties with conflicting financial or other interests; nor are there, as in a criminal proceeding, the interests of the state confronting the accused.

[13] However, not all State Parties to the Refugee Convention have adopted the non-adversarial system and even in Canada the Convention Refugee Determination Division of the Immigration and Refugee Board is assisted by a refugee hearing officer who may call and question any person who claims to be a Convention refugee and any other witnesses, present documents and make representations: Immigration Act 1985 (Can), s 68.1.

[14] In New Zealand the non-adversarial nature of proceedings before this Authority is unique in the immigration context and may be compared with appeals to the Residence Appeal Authority and the Removal Review Authority where there is a statutory duty to receive information, evidence and submissions from the chief executive of the Department of Labour.  See ss 18F(3) and 50(3) of the Immigration Act 1987.  There are no comparable provisions in the refugee context.

Non-adversarial proceedings and the burden of proof

[15] In common law countries the investigative or inquisitorial model is not, of course, unique to the refugee context.  It is well known in the context of administrative tribunals and Commissions of Inquiry.  The contrast between the procedures followed by tribunals of this kind and those followed in ordinary civil litigation and in criminal trials was remarked on in Mahon v Air New Zealand [1984] AC 808, 814E (PC).  Applying this decision, the High Court of Australia in Minister for Immigration and Ethnic Affairs v Wu Shan Liang (1996) 185 CLR 259, 282-283 (Brennan CJ, Toohey, McHugh & Gummow JJ) stated:

Submissions were made at the hearing of the appeal as to the correct decision-making process which it would have been permissible for the delegates to adopt.  These submissions were misguided.  They draw too closely upon analogies in the conduct and determination of civil litigation.

 Where facts are in dispute in civil litigation conducted under common law procedures, the court has to decide where, on the balance of probabilities, the truth lies as between the evidence the parties to the litigation have thought it in their respective interests to adduce at the trial.  Administrative decision-making is of a different nature. [Mahon v Air New Zealand Limited cited].  A whole range of possible approaches to decision-making in the particular circumstances of the case may be correct in the sense that their adoption by a delegate would not be an error of law.  The term "balance of probabilities" played a major part in those submissions, presumably as a result of the Full Court's decision.  As with the term "evidence" as used to describe the material before the delegates, it seems to be borrowed from the universe of discourse which has civil litigation as its subject.  The present context of administrative decision-making is very different and the use of such terms provides little assistance.

[16] It has also been said that where proceedings are not adversarial, an applicant does not carry any burden of proof: McDonald v Director-General of Social Security (1984) 6 ALD 6 at 9 (FC:FC).  But there is no inflexible rule, as can be seen from Commissioner of Police v Ombudsman [1988] 1 NZLR 385 (CA) at 391 line 25-50 per Cooke P and Patel v Chief Executive of the Department of Labour [1997] NZAR 264, 272-273 (CA).  In the latter case, the Court of Appeal, while accepting that a reference to onus of proof in the administrative context was not fully apt, nevertheless upheld the validity of the prescription by the Minister of Immigration of a standard of proof in Government residence policy, a standard which required “conclusive” proof.  In Australia, the limitations to McDonald were also recognised in Barker v Australian Telecommunications Commission (1990) 95 ALR 72, 79-80 (Einfeld J):
It seems that, however labelled, and whatever nomenclature is used, it is perfectly proper to approach administrative proceedings in terms of responsibilities to call evidence and affirmatively to persuade the tribunal of the point of view being advocated.
[17] General statements of principle as to the appropriateness of an onus of proof in administrative decision-making can only be taken so far in the abstract.  Ultimately the issue will turn on the particular statutory setting as ascertained from the text and purpose of the legislation.  In the present case the operative provisions specifically provide that it is the responsibility of the refugee claimant “to establish the claim”: ss 129G(4) and 129P(1) Immigration Act 1987.  The interpretation of these provisions in Part VIA of the Act follows shortly.

[18] However, it might be helpful to first briefly survey the burden of proof issue at the international level.

Burden of proof - international survey

[19] There is wide acceptance of the principle that a burden of proof lies on the person submitting a refugee claim.  There is equally wide acceptance that this does not mean importing into refugee determination the attendant complexities which attach to adversarial notions of the onus of proof, complexities which were recently highlighted in the New Zealand context by Janet November in Burdens and Standards of Proof in Criminal Cases (Butterworths, 2001) and illustrated in the civil context by Robertson, “Limitations and Burdens: Humphrey v Fairweather (1992) 6 PRNZ 450” [1994] NZLJ 203.  As Ms November explains (see particularly para 1.2.2), policy considerations dictate that the burden and standard of proof in the criminal context is different to that in the civil context.  Likewise, in our view refugee determination is sui generis and specific recognition must be given to the policy factors which shape decision-making in this unique context.  The point is cogently made by the UNHCR in the Handbook on Procedures and Criteria for Determining Refugee Status, paras 196 and 197:

196. It is a general legal principle that the burden of proof lies on the person submitting a claim. Often, however, an applicant may not be able to support his statements by documentary or other proof, and cases in which an applicant can provide evidence of all his statements will be the exception rather than the rule. In most cases a person fleeing from persecution will have arrived with the barest necessities and very frequently even without personal documents. Thus, while the burden of proof in principle rests on the applicant, the duty to ascertain and evaluate all the relevant facts is shared between the applicant and the examiner. Indeed, in some cases, it may be for the examiner to use all the means at his disposal to produce the necessary evidence in support of the application. Even such independent research may not, however, always be successful and there may also be statements that are not susceptible of proof. In such cases, if the applicant's account appears credible, he should, unless there are good reasons to the contrary, be given the benefit of the doubt.

 197. The requirement of evidence should thus not be too strictly applied in view of the difficulty of proof inherent in the special situation in which an applicant for refugee status finds himself. Allowance for such possible lack of evidence does not, however, mean that unsupported statements must necessarily be accepted as true if they are inconsistent with the general account put forward by the applicant.

[20] It is significant for present purposes to note that para 197 specifically accepts that the burden of proof lies on the person submitting a claim.  More recently, addressing the specific context of the Western European jurisdictions, the UNHCR in An Overview of Protection Issues in Western Europe: Legislative Trends and Positions Taken by UNHCR (September 1995) at 32 under the heading “Burden and Standard of Proof” stated:
It is generally accepted that, in the refugee status determination process, certain basic legal principles and minimum procedural safeguards should be observed because of the potentially serious consequences for individuals and the implication for states of whatever decision is taken.  Thus, refugee status determination procedures are to some extent governed by the rules of civil procedure according to which a person makes an assertion or submits a claim is obliged to adduce evidence in support of it.  The claimant bears the burden of proof which is discharged when the required standard of proof has been attained.
[21] It is significant that the context of this recognition of the burden of proof is the Western European or Continental legal system which is often (but at times erroneously) held up as representing the “ideal” of the inquisitorial approach.  It is necessary to be wary of the dangers of over-simplifying the dichotomy between adversarial and inquisitorial procedures.  See the discussion by Aronson & Dyer in Judicial Review of Administrative Action 2nd ed (LBC, 2000) at 403-404.  As noted by Leroy Certoma in “The Non-adversarial Administrative Process and the Immigration Review Tribunal” (1993) 4 PLR 4, 7, the modern non-adversary system in Europe is not a pure inquisitorial system but combines both adversarial and non-adversarial features.

[22] One of the leading academic commentators on refugee law, Professor Guy S Goodwin-Gill, one time Legal Adviser at the Office of the United Nations High Commissioner for Refugees, similarly asserts in The Refugee in International Law 2nd ed (Clarendon Press, Oxford, 1996) 34-35, 349 that the onus is on the refugee applicant to establish his or her case.

[23] Similarly, Dr Paul Weis who was long associated with the work of the International Refugee Organisation and the Office of the United Nations High Commissioner for Refugees states in “The Concept of the Refugee in International Law” 87 J. du Droit Int’l 928, 986 (1960):

The burden of proof is, according to general principles of law, on the applicant “[E]i incumbit probatio qui dicit non qui negat”, “actori incumbit onus probandi”.  The normal rules of evidence are, however, difficult to apply in proceedings for the determination of refugee status.  The applicant may call witnesses in support of his statements and he may sometimes be able to present documentary evidence.  But it follows from the very situation in which he finds himself as an exile, that he will rarely be in a position to submit conclusive evidence.  It will essentially be a question whether his submissions are credible and, in the circumstances, plausible.  The principle “in dubio pro reo” should be applied mutatis mutandis, ie, where there is, in the absence of conclusive evidence, doubt about the facts the applicant alleges, he should be given the benefit of the doubt.
[24] Both authors, while recognising that care is necessary in applying the onus in an unthinking way, see the imposition of an onus as entirely unexceptional, both in common law systems and in the legal systems of (Western) Europe.  It has not been claimed that the imposition of an onus of proof is inconsistent with the inquiry mandated by the Refugee Convention.

[25] In some jurisdictions the onus is placed on the refugee applicant by statute, in others by case law:

(a) As far as Australia is concerned, the High Court of Australia in Minister for Immigration and Ethnic Affairs v Guo (1997) 191 CLR 559, 570 accepted that the onus of proof is on the refugee claimant:
An applicant for refugee status who has established a fear of persecution must also show that the persecution which he or she fears is for one of the reasons enumerated in Art 1A(2) of the Convention.
Prior to this decision different approaches had been adopted in the Federal Court, though without causing difficulties in practice: Mary Crock, Immigration and Refugee Law in Australia (Federation Press, 1998) 138-139.  This is attributed to the fact that the Refugee Review Tribunal recognises the special considerations which both the UNHCR Handbook, Professor Goodwin-Gill and Dr Weis have noted.

(b) In Canada, at common law a refugee claimant bears the burden of establishing he or she is a Convention refugee: Adjei v Canada (Minister of Employment and Immigration) [1989] 2 FC 680, 682 (FC:CA).  See also Canada (Attorney General) v Ward [1993] 2 SCR 689, 751 (SC:Can) and Orelien v Canada (Minister of Employment and Immigration) [1992] 1 FC 592, 603-605 (FC:CA).  Section 8 of the Immigration Act 1985 (Can) also states that the burden of proof for a person seeking to enter Canada rests on that person, a point noted in Ward at 707.

(c) In the United Kingdom it is well established in case law, at least since R v Secretary of State for the Home Department; Ex parte Sivakumaran [1988] AC 958 (HL) that the onus rests on the refugee claimant.  See for example Ponnampalam Anandanadarajah v Immigration Appeal Tribunal [1996] Imm AR 514, 519 (CA) and Abdul Aziz Faraj v Secretary of State for the Home Department [1999] INLR 451, 456 (CA).

(d) In the United States of America the burden of proof is statutory.  The following summary is taken from Deborah E Anker, Law of Asylum in the United States 3rd ed (Refugee Law Center, 1999) at 14 (footnote citations omitted):

The statute and regulations provide that the individual applicant has the burden of proof both for asylum and withholding protection; this rule is consistent with international legal principles [UNHCR Handbook para 196 cited].  To establish eligibility under the asylum qualification provisions of the INA, an applicant bears the burden of proving herself a refugee as defined in INA Section 101a(42)(A)...  To establish eligibility under the Section 241(b)(3) withholding provision, a person must demonstrate that, if returned to her country of origin or last habitual residence, her “life or freedom would be threatened” for one of the same five reasons”.
(e) In Western Europe the survey in Carlier, Vanheule, Hullmann & Galiano eds, Who is a Refugee: A Comparative Case Law Study (Kluwer, 1997) shows that there are some countries in which the onus is clearly placed on the refugee claimant, while in others the position is less clear.  The page references which follow refer to the text cited.  The countries which recognise an onus of proof on the claimant are Switzerland (142), Germany (263), Denmark (320), France (390), Luxemburg (477) and Portugal (545).  From the limited information in the text it is not entirely clear what the position is in Austria, Belgium, Spain, Greece, Italy and the Netherlands.  The French, German and Swiss jurisprudence is discussed in greater detail by Professor Walter Kälin in “Well-Founded Fear of Persecution: A European Perspective” in Bhabha & Coll eds, Asylum Law and Practice in Europe and North America: A Comparative Analysis (Federal Publications, 1992) 21 at 33-34.  No challenge is made to the principle that the burden of proof lies on the refugee claimant.
Conclusions from international survey

[26] There is widespread acceptance across a range of jurisdictions of the principle that the burden of proof lies on the refugee claimant.  It is significant that major refugee receiving countries fall into this category, namely France, Germany, the United Kingdom, Canada, the USA and Australia.  Significantly, the legal principle is acknowledged and accepted by the Office of the United Nations High Commissioner for Refugees and by leading academics (Weis, Goodwin-Gill, Kälin).

[27] Acceptance of the principle is a different issue to the question as to how the principle is to operate in practice.  As the passages from the Handbook, Weis and Goodwin-Gill demonstrate, the very nature of refugee determination requires a contextual understanding of  how the principle is to be applied.  As Professor Kälin points out in “Well-Founded Fear of Persecution: A European Perspective”in Bhabha & Coll eds, Asylum Law and Practice in Europe and North America: A Comparative Analysis (Federal Publications, 1992) 21 at 21-33 the danger is not in the burden of proof, but in imposing too strict a standard of proof.

[28] It is sufficient for present purposes to note from the international survey that there is nothing inherently objectionable to a burden of proof in the refugee context.  Nor do the common law cases discussed at paras [15] and [16] above condemn a burden of proof as being necessarily inconsistent with administrative decision-making.

[29] This background is a convenient starting point for an examination of the burden of proof in New Zealand.

BURDEN OF PROOF IN NEW ZEALAND

POSITION PRIOR TO 1 OCTOBER 1999

[30] Any consideration of the burden of proof in New Zealand must take account of both the case law and of the evolving sources of the Authority’s jurisdiction and powers.  The latter will be addressed first.

The Terms of Reference

[31] Following the October 1990 general election the incoming administration, on 17 December 1990, approved new procedures for the determination of applications for refugee status.  Those procedures included the setting up of the Authority.  On 11 March 1991 the procedures were incorporated into Terms of Reference and the Authority heard its first appeal early in June 1991.  See further Singh v Refugee Status Appeals Authority [1994] NZAR 193, 198-199 (Smellie J).  Although the Terms of Reference were subsequently modified on three separate occasions, the basic outline of the procedures remained unchanged.  In chronological order the Terms of Reference were Terms of Reference (March 1991); Terms of Reference (1 April 1992); Terms of Reference (in force on 30 August 1993) and the Rules Governing Refugee Status Determination Procedures in New Zealand (in force from 30 April 1998).

[32] The unusual feature of the refugee determination system in its original form was that it operated on an extra-statutory basis.  The view taken in the High Court was that the procedures were the creature of the prerogative but nonetheless amenable to judicial review: Benipal v Ministers of Foreign Affairs and Immigration (High Court Auckland, A Nos.  878/83, 993/83 & 1016/83, 29 November 1985, Chilwell J) at 264-273 (appeal by the Crown dismissed on other grounds in Minister of Foreign Affairs v Benipal [1988] 2 NZLR 222 (CA)); Singh v Refugee Status Appeals Authority [1994] NZAR 193, 209-212 (Smellie J); Khalon v Attorney-General [1996] 1 NZLR 458, 461 (Fisher J); B v Refugee Status Appeals Authority (High Court Auckland, M 1600/96, 23 July 1997, Giles J) at 3-4.  However, the Court of Appeal twice expressed reservations as to both the reviewability of the Authority’s decisions and as to the appropriateness of the procedures being extra-statutory:  Butler v Attorney-General [1999] NZAR 205, 218-220 (Richardson P, Henry, Keith, Tipping & Williams JJ); S v Refugee Status Appeals Authority [1998] 2 NZLR 291, 294 (Henry, Keith & Blanchard JJ).  Legislative reform was finally enacted in 1999 in the form of the Immigration Amendment Act 1999.  Section 40 of that Act inserted a new Part VIA into the Immigration Act 1987.  Section 129N(1) of the principal Act provides that the Authority is “continued” as a body.

[33] The first Terms of Reference of March 1991 were the essence of simplicity and brevity, comprising two pages only.  The final Terms of Reference of 30 April 1998 were considerably expanded (fourteen pages).  In essence, however, they remained the same in that the Authority was required to consider the first instance decision, to interview the appellant and to deliver a decision.  None of the Terms of Reference addressed the issues of burden of proof and standard of proof.  The Authority was, however, given power to regulate its own procedure.

The Authority’s case law under the Terms of Reference

[34] In the initial years the Authority avoided ruling on the issue of burden of proof though as to the standard of proof it has from the outset understood the “well-founded” standard as requiring an assessment as to whether there is a real chance of the claimant being persecuted.  See the Authority’s first decision in Refugee Appeal Nos. 1/91 Re TLY and 2/91 Re LAB [1992] NZAR 542, 546.  The Authority tried to mould the procedures around the paradigm of  the inarticulate, traumatised and honest refugee claimant with whom the Authority would embark on a joint inquiry as to refugee status.

[35] Unfortunately, the Authority was unable to sustain this approach in the face of:

(a) Rising levels of abusive claims;

(b) The withdrawal by a significant number of claimants from the co-operative inquiry process.

Rising levels of abuse

[36] In the period 1991 to 1996 the percentage of successful appeals ranged between 16% and 49%, as the table which follows shows.  The sometimes substantial fluctuations are partly explained by the (then) dysfunctional decision-making process at first instance (see for example Refugee Appeal No. 2226/94 Re LRR (16 October 1996)) - a criticism no longer valid since the introduction of the statutory regime - and partly by a disturbing level of abuse.  >From the financial year 1996-97, however, the approval rate has averaged approximately ten percent:

Refugee Status Appeals Authority
Appeals Allowed and Dismissed - 1991 to 2002


Total number of decisions
Number allowed
Percentage allowed
Number dismissed
Percentage dismissed
1991/92
57
17
29.8%
40 
70.2%
1992/93
166
28
16.8%
138
83.2%
1993/94
409 
109
26.7%
300
73.3%
1994/95
509
250
49.1%
259
50.9%
1995/96
491
132
26.9%
359
73.1%
1996/97
522
54
10.3%
468
89.7%
1997/98
436
33
7.6%
403
92.4%
1998/99
405
42
10.4%
363
89.6%
1999/2000
517
74 
14.3%
443
85.7%
2000/01
642
52 
8.1%
590
91.9%
2001/02*
420
14
3.3%
406 
96.7%
Total No/ average %
4574
805 
17.6%
3769
82.4%

 [* covers nine months from July 2001 to March 2002]

[37] The degree of abuse of the refugee procedures is illustrated by Refugee Appeal No. 70002/96 Re BS (7 May 1996) and was further documented in Refugee Appeal No. 70951/98 (5 August 1998).  In the latter decision the Authority noted that in the period from 1 April 1995 to 31 March 1996 some 63% of all new appeals were repeat appeals.  Of all second Punjabi appeals heard and decided by the Authority from October 1994 to 30 April 1998, and there were approximately 300 of such cases, only eight succeeded.  In percentage terms, the success rate of these appeals was 2.67%.  In the period July 1996 to June 1997 thirty-two cases were determined to be prima facie manifestly unfounded and therefore to be dealt with on the papers.  This represented 8.2% of total appeals.  However, in the period July 1997 to June 1998 the number rose to 120.  The percentage of manifestly unfounded  appeals as a ratio to the total number of appeals decided was 27.40%.  In the result, the percentage of appeals allowed by the Authority dropped from a historic high of 49% in the financial year 1994/95 to 26.9% in the financial year 1995/96 and from 10.3% in the financial year 1996/97 to 7.6% in 1997/98.  The Authority recorded that considerable resources were being diverted to deal with manifestly unfounded cases and the principal victim of the abuse was the genuine refugee claimant.

[38] When abuse of this magnitude occurs the genuine refugee claimant is not the only victim.  The refugee determination system itself is imperilled and in Refugee Appeal No. 70951/98 (5 August 1998) the Authority went to some lengths to explain the underlying policy imperatives which dictated the approach it resolved to take in addressing the abuse of its procedures.

[39] The regrettable fact is that because refugee status trumps all immigration control and immigration policy, abuse of the system will always be present and certainly continues in New Zealand.  It is simply not possible to prevent the unscrupulous from lodging abusive claims.  All a system can do is to ensure that such claims are fast-tracked in order to remove incentives created by delay.  The claims nevertheless represent a very substantial drain on resources, both at first instance and on appeal.  The most recent manifestation of abuse has been the blatant manipulation of the system by a large group of Thai nationals who lodge refugee applications simply to secure the open work permit which is usually granted to refugee claimants, or in the alternative, to secure time in New Zealand while the claim is processed.  In the period 1 April 2001 to 15 March 2002 some 401 refugee applications by Thai nationals were received by the Refugee Status Branch.  All were declined.  Some 232 almost identical appeals were lodged with the Authority by Thai nationals in the period 5 June 2001 to 7 March 2002.  There is no credible evidence to support these refugee claims.  Yet even when the claims are rejected (at first instance and later on appeal), many simply re-lodge the claims at first instance to postpone their removal once again.  As at 15 March 2002 some 146 claims had been re-lodged a second time, 22 claims a third time and two for a fourth time.  One of the ploys is to insist that the evidence of the claimant be given not in his or her own language (Thai), but in Pali, a language which most, if not all of the applicants do not understand or speak.  But they know that there are no Pali interpreters in New Zealand.  The Authority addressed such claims in Refugee Appeal No. 72752/01 (15 November 2001).  The facts recorded in that decision make for depressing reading.  For an example of a Thai national who made three repeat refugee claims (and appeals) in the space of eighteen months reference should be made to Refugee Appeal No. 73417/02 (11 March 2002).  In that case a fourth refugee application was lodged while the third appeal was being determined.

[40] The Authority’s experience has been that endemic abuse of its procedures has the potential of consuming all of the Authority’s slender resources, thereby bringing the entire system to its knees.  The delay in delivering this decision is due in no small measure to the diversion of resources to deal with the Thai claims.  Since the first years the Authority has been on a steep learning curve.  In particular it has had to face hard choices when confronted with the need for the system to protect itself from abuse while ensuring that the New Zealand refugee determination procedures will, in an expeditious and fair manner, recognise those who truly are refugees within the meaning of Article 1 of the Refugee Convention.

Withholding of co-operation

[41] It would seem an unexceptional proposition that a genuine refugee claimant would wish to co-operate in every way with the decision-maker to ensure that the refugee inquiry is unhindered, the ultimate aim being to arrive at the truth, namely that the claimant is a credible witness and that the facts establish the criteria prescribed by the Convention’s inclusion clause.  However, there has unmistakably emerged an attitude that the responsibility of a refugee claimant is not to co-operate willingly in the ascertainment of the full facts of the claim, but rather to pursue the case in an opportunistic if not legalistic manner, cards held close to the chest, to be produced reluctantly and then only if there is no other palatable alternative.

[42] The joint inquiry envisaged by the paradigm is by this unfortunate attitude more often than not turned into a contest of wills, the claimant producing information selectively, the decision-maker asking more and more questions in order to get to the “truth”.  Legal stratagems and objections more appropriate to curial proceedings are employed to prevent lines of inquiry, as for example where members of the same family lodge applications on similar grounds.  It is said that the Authority may not, when hearing one case, know what has been said in the related claim brought by the other family member(s).  See for example the facts in Refugee Appeal No. 70385/98 (10 December 1998).  Stratagems have been employed to endlessly adjourn cases, as in Refugee Appeal No. 112/92 (10 October 1996). The fast track hearing of appeals by those in custody has on occasion met significant resistence by claimants who have other agenda to pursue, as in the group of hunger-strikers discussed in Refugee Appeal No. 71735/99 (25 January 2000) - Minute and in the later substantive decision in Refugee Appeal No. 71735/99 (12 September 2000).  In the actual hearings there is widespread deceit and fraud, as indicated by the large number of cases which fail on credibility grounds.  The fraud is at times sophisticated, as illustrated by Refugee Appeal No. 72491/00 (24 September 2001) (false identity, passport swapping), Refugee Appeal No. 71051/98 (26 August 1999) (long history of false refugee claims and use of fake passports), Refugee Appeal No. 71634/99 (19 August 1999) (admitted serial lies in the course of two refugee applications in New Zealand - five different versions of the facts advanced).  The lengths to which refugee claimants will go to construct a claim to refugee status are illustrated by Refugee Appeal No. 2254/94 Re HB (21 September 1994) and Refugee Appeal No. 70100/96 (28 September 1997).  On the question of proof, submissions have come close to saying that it is sufficient for the refugee claimant to simply lodge the claim.  It then becomes the responsibility of the relevant decision-maker, without the willing and co-operative assistance of the claimant, to determine whether the refugee claim is made out.  Quite apart from the issue of resources (more of which later), this is a submission that the claimant carries no burden of proof and owes no obligation of goodwill, co-operation, candour and disclosure.

[43] The inquiry process is itself inhibited by legal and practical constraints.

The limits to the refugee inquiry

[44] For obvious reasons, it is imperative that refugee claims be dealt with confidentially.  Disclosure may expose either the claimant, family members, associates or similarly situated persons to a risk of persecution in the home country.  The making of inquiries into the circumstances of the refugee claim is therefore fraught with danger.  Indeed the making of inquiry can in some circumstances expose the refugee status officer and the Authority to the risk of prosecution for breach of s 129T of the Act which imposes a statutory obligation of confidentiality.  At times that obligation may require confidentiality as to the very fact or existence of a claim:

129T. Confidentiality to be maintained—

(1) Subject to this section, confidentiality as to the identity of the claimant or other person whose status is being considered under this Part, and as to the particulars of their case, must at all times, both during and subsequent to the determination of the claim or other matter, be maintained by refugee status officers, the Authority, other persons involved in the administration of this Act, and persons to whom particulars are disclosed under subsection (3)(a) or (b).

(2) Compliance with subsection (1) may in an appropriate case require confidentiality as to the very fact or existence of a claim or case, if disclosure of its fact or existence would tend to identify the person concerned, or be likely to endanger any person.

(3) Subsection (1) does not apply to prevent the disclosure of particulars—

(a) To a person necessarily involved in determining the relevant claim or matters; or

(b) To an officer or employee of a Government department or other Crown agency whose functions in relation to the claimant or other person require knowledge of those particulars; or

(c) To the United Nations High Commissioner for Refugees or a representative of the High Commissioner; or

(d) In dealings with other countries for the purpose of determining the matters specified in section 129L(d) and (e) (whether at first instance or on any appeal); or

(e) To the extent that the particulars are published in a manner that is unlikely to allow identification of the person concerned, whether in a published decision of the Authority under clause 12 of Schedule 3C or otherwise; or

(f) If there is no serious possibility that the safety of the claimant or any other person would be endangered by the disclosure in the particular circumstances of the case.

(4) Nor does subsection (1) apply to prevent the disclosure of particulars in relation to a particular claimant or other person to the extent that the claimant or person has, whether expressly or impliedly by their words or actions, waived his or her right to confidentiality under this section.

(5) A person who without reasonable excuse contravenes subsection (1), and any person who without reasonable excuse publishes information released in contravention of subsection (1), commits an offence.

[45] Inquiries also lead to inevitable delay.  In addition there are practical difficulties involved in locating a trustworthy person in the country of origin to make the inquiry and this will also usually involve finding a trustworthy interpreter to accompany the investigator.  There are also significant problems in conducting an inquiry at a distance in that the inquirer will not be equipped with sufficient information to guide the questions in one direction or the other, according to the answers given.  Then there is the matter that the question or issue investigated may not have been fully or properly answered.  The results may be enigmatic or ambiguous at best.  Corruption is rife in many countries of origin and it is often difficult to know whether the information provided is reliable.  These and other factors inhibiting the inquiry process received explicit recognition in AB v Refugee Status Appeals Authority [2001] NZAR 209, 222-225 (Nicholson J).  Largely because of the considerations mentioned and discussed in that case the Authority’s general approach to refugee claims is to focus primarily on the credibility of the refugee claimant as assessed against publicly accessible information.  Expressed another way, the inquiry into a refugee claim is severely constrained by practical difficulties, budgetary constraints and the limit to the number of hours that can be devoted to each claim.  It is simply not possible to conduct an open-ended roving inquiry both in New Zealand and overseas to unearth “the truth”.  The practical imperatives spoken of in AB v Refugee Status Appeals Authority necessarily bring the focus to bear on the credibility of the refugee claimant and this, in turn, highlights the obligation of the claimant to co-operate in every way with the refugee decision-maker who is by force of circumstances precluded from making meaningful inquiry into the claim beyond the general issue of credibility assessed against available country information.

[46] It is against this background that it is possible to turn to Refugee Appeal No. 523/92 Re RS (17 March 1995) in which the Authority grasped the nettle and held that the burden of proof rests on the claimant.

Refugee Appeal No. 523/92

[47] In Refugee Appeal No. 523/92 Re RS (17 March 1995) the Authority held that a refugee claimant carries a burden of proof to establish his or her claim.  In the present context the following points in Refugee Appeal No. 523/92 may be noted:

(a) The Authority took the view then (and remains of the view) that any person who claims that New Zealand owes him or her international protection obligations under the Refugee Convention must act in good faith and act honestly and co-operatively in the refugee determination process;

(b) The requirement that a refugee claimant proves his or her claim to New Zealand’s international obligation to provide surrogate protection is not to place on him or her an unreasonable obligation.  Otherwise the door will be open to abuse, with claimants doing no more than lodging an application for refugee status unsupported by any account of the facts, and expecting the decision-maker to carry out an investigation without the claimant’s assistance;

(c) Resting the burden of proof on a claimant does not impose an unreasonable requirement as it is mitigated by three factors, namely the low standard of proof, the liberal application of the benefit of the doubt principle and finally, by the fact that the non-adversarial nature of the proceedings means that the enquiry is shared between the claimant and the decision-maker.

[48] The relevant passages from Refugee Appeal No. 523/92 Re RS (17 March 1995) at 17-20 follow:
We can now address the issue whether a refugee claimant carries a burden of proof to establish his or her claim.

 ....

A person who claims the right to be recognized as a refugee under the Refugee Convention must necessarily be aware of the circumstances which justify the assertion that he or she holds a well-founded fear of persecution for one of the five Convention reasons.  By making a claim to refugee status, that person must shoulder the obligation of establishing the claim as the facts on which it is based lie peculiarly within the knowledge of the claimant.

This is a basic proposition which would ordinarily require no articulation given that a person in fear of persecution would be expected to make every effort to establish his or her claim.

It is also inherent in a claim on New Zealand's international obligations under the Refugee Convention that claimants must act in good faith.  The requirement that they prove their claim to the surrogate protection afforded by the Refugee Convention is not, therefore, to place on them an unreasonable obligation.  Otherwise the door will be opened to abuse, with claimants doing no more than lodging an application for refugee status unsupported by any account of the facts, and expecting the decision-maker to carry out an investigation without the claimant's assistance.  It is only a small step from there to say that refugee status is established if the decision-maker is unable to prove that the claimant is not a refugee.  This would be an absurd state of affairs.  In fairness to the appellant, it must be said that no such claim on his behalf was advanced.  However, the submission that a non-party to the appeal (the NZIS) carried a burden of proof in the relocation context cannot be separated from the general issue of the burden of proof in refugee applications.  A submission that a claimant does not carry the burden of proof as to relocation is not very different from a submission that the claimant carries no burden at all to establish the refugee claim.  Certainly the appellant's submission did not address the issue as to why there should be two separate burdens, or why the burden as to relocation only fell on the NZIS if it appeared at an appeal hearing.  In short, there was a distinct absence of logic and merit to the appellant's submission.

Our holding that a refugee claimant carries the burden of proving his or her claim does not break new ground.  The UNHCR Handbook on Procedures and Criteria for Determining Refugee Status para 196 opines:

"It is a general legal principle that the burden of proof lies on the person submitting a claim."
In Canada the principle is enshrined in s 8(1) of the Immigration Act 1985, a fact noted in Canada (Attorney-General) v Ward [1993] 2 SCR 689, 707 (Can:SC).

Resting the burden of proof on a claimant does not impose an unreasonable responsibility as it is mitigated by three principal factors:

1. The standard of proof.  As will be shown in the next section, the standard of proof in the Inclusion clause context is that of a "real chance" of persecution.  This is a low threshold and adequately addresses the concerns expressed by Grahl-Madsen in The Status of Refugees in International Law (Vol 1) (1966) 145-146 where, in addressing the issue of proof refugeehood, he observes:

"In one respect, however, a liberal attitude is called for outright, in order that full effect may be given to the provisions of the Refugee Convention and the purposes for which they are intended: it is a well-known fact that a person who claims to be a refugee may have difficulties in proving his allegations.  He may have left his country without any papers, there may be nobody around who may testify to support his story, and other means of corroboration may be unavailable.  It would go counter to the principle of good faith if a contracting State should place on a suppliant a burden of proof which he, in the nature of things, could not possibly cope with."
2. The benefit of the doubt principle is to be applied liberally, as decisions of this Authority will show.  The principle is that if a decision-maker is unable to make up his or her mind as to whether the claimant is a refugee, a decision in favour of the claimant is to be given as it is inherent in such a situation that the claimant's account could be true.

3. The non-adversarial nature of the proceedings means that the enquiry is shared between the claimant and the decision-maker.  This ameliorates any disadvantage a claimant might face in the elucidation of the facts, especially information relating to the human rights conditions in the country of origin.  This information is relevant both to the overall issue of credibility and to the objective or "well-founded" aspect of the claim.  However, the fact that the claimant and decision-maker each have a responsibility to ascertain the facts does not relieve the claimant of the legal burden of proof to establish the claim.

We believe that very much the same is said in the UNHCR Handbook on Procedures and Criteria for Determining Refugee Status para 196:

"It is a general legal principle that the burden of proof lies on the person submitting a claim.  Often, however, an applicant may not be able to support his statements by documentary or other proof, and cases in which an application can provide evidence of all his statements will be the exception rather than the rule.  In most cases a person fleeing from persecution will have arrived with the barest necessities and very frequently even without personal documents.  Thus, while the burden of proof in principle rests on the applicant, the duty to ascertain and evaluate all the relevant facts is shared between the applicant and the examiner.  Indeed, in some cases, it may be for the examiner to use all the means at his disposal to produce the necessary evidence in support of the application.  Even such independent research may not, however, always be successful and there may also be statements that are not susceptible of proof.  In such cases, if the applicant's account appears credible, he should, unless there are good reasons to the contrary, be given the benefit of the doubt."
However, as Grahl-Madsen points out in The Status of Refugees in International Law (Vol 1) 146, the liberal attitude to refugee determination adopted by the Authority should not lead to:
"... an uncritical acceptance of any and all allegations made by suppliants."
[49] Much of this passage was cited with approval in C v Refugee Status Appeals Authority (High Court Auckland, M 1365-SW00, 4 May 2001, Nicholson J) at [60] and [61].

Burden of proof and the Court of Appeal

[50] In Butler v Attorney-General [1999] NZAR 205 (CA) one of the submissions made was that the Authority was under an obligation to address and investigate a particular matter even though it had not been raised by Mr Butler.  The matter was the reasonableness element in the internal flight alternative/relocation/internal protection alternative context.  The Court of Appeal dealt with the challenge by holding at 213-215:

(a) The burden of establishing the elements of the refugee claim rests on the claimant:
A person claiming refugee status has the burden of establishing the elements of the claim.  That rule should however not be applied mechanically.  Those making a decision which may put an individual’s right to life at risk and Courts reviewing any such decision, have a special responsibility to see that the law is complied with, for example Bugdaycay v Secretary of State for the Home Department [1987] AC 514, 531, 537.
(b) While the Authority may be required of its own motion to investigate an issue, it is not an error of law to fail to rule on a matter not referred to by a claimant and if it does not stand out as requiring a decision.  See p 215:
At this stage we conclude that given the way the case was presented to the RSAA it cannot be said that it committed an error of law in not separately addressing a distinct reasonableness element.  No such element was presented to it as arising from the facts.  Indeed so far as we understand the facts it would have been very difficult for the appellant to have done that.  This is not the kind of case where either the law or the factual situation before the Authority requires it of its own motion to take up any such additional element; see the statements to similar effect of Black CJ and Whitlam J in Randhawa v Minister for Immigration, Local Government and Ethnic Affairs (1994) 124 ALR 265, 270-271, 280 (FCA).
It cannot be an error of law for a Tribunal considering a matter (here location) which is properly before it to fail to rule on some particular aspect of that matter if the particular aspect is not referred to by the interested party and if it does not stand out as requiring decision.
[51] In our view the holding of Butler is that the burden of proof rests on the refugee claimant.

[52] There is a remarkable similarity in wording in the language used by the Court of Appeal and that used by the Authority:

Appeal 523/92 - the obligation of establishing the claim.

Butler - the burden of establishing the elements of the claim.

The significance of this point will be returned to shortly.

[53] The only point that remains to be made is that prior to 1 October 1999, being the date on which Part VIA of the Immigration Act 1987 came into force, there was no challenge to either Refugee Appeal No. 523/92 or to Butler.

The Immigration Amendment Act 1999

[54] The language of Refugee Appeal No. 523/92 and of Butler is found in two provisions of the new Part VIA of the Act as enacted by the Immigration Amendment Act 1999, s 40.

[55] First is s 129G(5) which prescribes how a claim is to be made and handled at first instance by a refugee status officer:

It is the responsibility of the claimant to establish the claim, and the claimant must ensure that all information, evidence, and submissions that the claimant wishes to have considered in support of the claim are provided to the refugee status officer before the officer makes a determination on the claim.
[56] Second is s 129P(1) which in the context of an appeal to the Authority provides:
It is the responsibility of an appellant to establish the claim, and the appellant must ensure that all information, evidence, and submissions that the appellant wishes to have considered in support of the appeal are provided to the Authority before it makes its decision on the appeal.
[57] The submission for the appellant in the present case is that “responsibility to establish the claim” does not impose a burden to prove the claim.  It is no more than a responsibility to establish what the claim is.  The decision relied on is T v Refugee Status Appeals Authority [2001] NZAR 749 at [23] and [38] (Durie J).  On one interpretation this represents a substantial departure from the Butler holding that the responsibility is to establish the elements of the claim.  Before addressing the decision in T an analysis of the statutory provisions is required.

THE 1999 PROVISIONS

[58] While the background to the legislation is relevant, it is largely a question of statutory interpretation whether the responsibility to “establish the claim” requires more of a claimant than to merely establish what the claim is.  The two provisions in question are ss 129G and 129P:

129G. How claim made and handled—

(1) A claim is made as soon as a person signifies his or her intention to seek to be recognised as a refugee in New Zealand to a representative of the Department of Labour or to a member of the Police.

(2) Once a claim is made, the claimant must, on request by a refugee status officer, confirm the claim in writing in the prescribed manner.

(3) A claimant must as soon as is possible endeavour to provide to an officer all information relevant to his or her claim, including—

(a) A statement of the grounds for the claim; and

(b) An indication of whether any other members of the claimant's immediate family who are in New Zealand are also seeking recognition as refugees and, if so, whether any such claim is on different grounds.

(4) A claimant must provide an officer with a current address in New Zealand to which communications relating to the claim may be sent and a current residential address, and must notify the officer in timely manner of a change in either of those addresses. The officer may rely on the latest address so provided for the purpose of communications under this Part.

(5) It is the responsibility of the claimant to establish the claim, and the claimant must ensure that all information, evidence, and submissions that the claimant wishes to have considered in support of the claim are provided to the refugee status officer before the officer makes a determination on the claim.

(6) For the purpose of determining a claim, an officer—

(a) May seek information from any source; but

(b) Is not obliged to seek any information, evidence, or submissions further to that provided by the claimant; and

(c) May determine the claim on the basis of the information, evidence, and submissions provided by the claimant.

(7) Subject to this Part and to any regulations made under it, and to the requirements of fairness, an officer may determine his or her own procedures on a claim.
 

129P. Procedure on appeal—

(1) It is the responsibility of an appellant to establish the claim, and the appellant must ensure that all information, evidence, and submissions that the appellant wishes to have considered in support of the appeal are provided to the Authority before it makes its decision on the appeal.

(2) The Authority—

(a) May seek information from any source; but

(b) Is not obliged to seek any information, evidence, or submissions further to that provided by the appellant; and

(c) May determine the appeal on the basis of the information, evidence, and submissions provided by the appellant.

(3) An appellant must provide the Authority with a current address in New Zealand to which communications relating to the appeal may be sent and a current residential address in New Zealand, and must notify the Authority in timely manner of a change in either of those addresses. The Authority may rely on the latest address so provided for the purpose of communications under this Part.

(4) In its consideration of an appeal or other matter under this Part, the Authority may request the chief executive of the Department of Labour to seek and provide relevant information.

(5) The Authority may dispense with an interview of the appellant or other affected person only if both—

(a) The appellant or other affected person has been interviewed by a refugee status officer in the course of determining the relevant matter at first instance or, having been given an opportunity to be interviewed, failed to take that opportunity; and

(b) The Authority considers that the appeal or other contention of the person affected is prima facie manifestly unfounded or clearly abusive.

(6) Despite subsection (5), the Authority may determine an appeal or other matter without an interview if the appellant or other person affected fails without reasonable excuse to attend a notified interview with the Authority.

(7) If a summons is issued by the Authority under section 4D of the Commissions of Inquiry Act 1908 in respect of a person detained in custody, the Superintendent or other person in charge of the relevant penal institution or other approved premises, or other person having custody of the detained person, must produce, or allow the production of, the person as directed in the summons.

(8) The Authority may decide the order in which appeals or other matters are to be heard, and no decision on an appeal or other matter is to be called into question on the basis that the appeal or other matter ought to have been heard or decided earlier or later than any other appeal or matter or category of appeal or matter.

(9) In any appeal involving a subsequent claim, the claimant may not challenge any finding of credibility or fact made by the Authority in relation to a previous claim, and the Authority may rely on any such finding.

[59] The word “claim” is defined in s 129B as a claim in New Zealand to be recognised as a refugee in New Zealand:
 
Claim” means a claim in New Zealand to be recognised as a refugee in New Zealand.
[60] A claim is made as soon as a person signifies to a representative of the Department of Labour or to a member of the Police his or her intention to seek to be recognised as a refugee in New Zealand: s 129G(1).  Once a claim is made, the claimant must, on request, confirm the claim in writing in the prescribed manner: s 129G(2).  As to this requirement the Immigration (Refugee Processing) Regulations 1999 (SR 1999/285), Reg 3, directs the lodging of a form in which the claimant discloses extensive detail about his or her personal history and background and about the claim itself.  Regulation 3(4) expressly provides that:
In so far as reasonably obtainable, the following details and documents are also to accompany confirmation of the claim:

(a) Evidence of identity (including a recent photograph of the claimant);

(b) Evidence of country of origin;

(c) Any evidence supporting the fact or likelihood of persecution;

(d) Where available, documents indicating by whom the persecution or potential persecution is alleged and the reason for that persecution;

(e) Details of persons who can be contacted to support or verify the claim (if any).

[61] Additionally s 129G(3) of the Act imposes a mandatory obligation on the claimant to endeavour to provide (inter alia) all information relevant to his or her claim, including a statement of the grounds for the claim:
A claimant must as soon as is possible endeavour to provide to an officer all information relevant to his or her claim, including—

(a) A statement of the grounds for the claim; and

(b) An indication of whether any other members of the claimant's immediate family who are in New Zealand are also seeking recognition as refugees and, if so, whether any such claim is on different grounds.

[62] It is in this context that s 129G(5) then provides that it is the responsibility of the claimant to establish the claim and to ensure that all information, evidence and submissions the claimant wishes to have considered in support of the claim are provided to the refugee status officer before the officer determines the claim:
It is the responsibility of the claimant to establish the claim, and the claimant must ensure that all information, evidence, and submissions that the claimant wishes to have considered in support of the claim are provided to the refugee status officer before the officer makes a determination on the claim.
[63] Reading s 129G(5) together with s 129B, the responsibility to establish the claim means the responsibility to establish the claim in New Zealand to be recognised as a refugee in New Zealand.  Subsections (3) and (5) of s 129G make it clear that the “claim” [to be recognised as a refugee in New Zealand] is to be distinguished from:
(a) The confirmation of the claim, in particular the form (and by inference its contents) prescribed by s 129G(2) and Regulation 3;

(b) The “information” relevant to the claim and the statement of the “grounds” of the claim referred to in s 129G(3);

(c) The information, evidence and submissions the claimant wishes to have considered in support of the claim as referred to s 129G(5).

[64] In summary s 129G distinguishes “the claim” from:
Information relevant to [the claim] - s 129G(3).

The grounds for [the claim] - s 129G(3)(a).

Information, evidence and submissions in support of [the claim] - s 129G(5).

[65] The conclusion to be drawn from these statutory provisions is that the “claim in New Zealand to be recognised as a refugee in New Zealand” is clearly separate from the information, grounds, evidence and submissions provided by the claimant in support of the claim.  The latter establish what that claim is.  Given the language of the statute we cannot accept the appellant’s submission that the responsibility to establish the claim means a responsibility to establish what the claim is.  On the appellant’s argument the words “responsibility of the claimant to establish the claim” become mere surplusage, adding nothing to the duties imposed by the balance of s 129G.  As noted by Professor JF Burrows in Statute Law in New Zealand 2nd ed (Butterworths, 1999) at 202, tautology apart, it is difficult to “ignore” entire clauses and this will only be done in extreme circumstances.  Here the language is clear and free of tautology and there are no extreme circumstances.  Applying s 5 of the Interpretation Act 1999 the text must be interpreted as a whole and in the light of the purpose of the legislation.  No sensible reason was offered as to why Parliament would want to impliedly overrule Butler and Refugee Appeal No. 523/92, while employing virtually the ipsissima verba which those decisions used to impose the burden of proof in the first place.  Both the plain reading of the text and the historical background (the “external context” discussed by Professor Burrows at op cit 155, 158-159) to s 129G require the words “responsibility to establish the claim” to be understood as imposing on the refugee claimant the burden of establishing that he or she is a refugee within the meaning of Article 1 of the Refugee Convention.

[66] While the provisions discussed so far address specifically the procedures at first instance, the refugee determination process must be seen as an integrated whole.  The specific provision in s 129P(1) that on appeal it is the responsibility of an appellant to establish the claim must receive the same interpretation.  In this regard, s 129G(5) and s 129(1) are not materially different.

[67] In so far as T v Refugee Status Appeals Authority [2001] NZAR 749 is concerned, none of the foregoing points are addressed and the obiter comments made by Durie J are, with respect, the weaker as a result.

Relevance of the power to inquire

[68] The further submission by the appellant was (in effect) that as a refugee claimant has no obligation beyond establishing what the claim is, it falls to the Authority, as a Commission of Inquiry, to fully investigate the claim.  This submission comes very close to a contention that the burden of proving or disproving the claim lies on the Authority.  Again the decision in T v Refugee Status Appeals Authority [2001] NZAR 749 was relied upon.  The relevant passages are at [22], [23] and [38]:

[22] Section 129P(2) does not detract from the primacy of an inquisitorial approach, in my view, and instead, the section must be read in the context of the inquisitorial function.  In short, the Authority is generally obliged to inquire, but in appropriate cases may be relieved from so doing.

[23] What then is meant by 'the responsibility of an appellant to establish the claim' in s 129P(1)?  First, as I have mentioned, the reference is not to a burden of proving the claim.  I think it confuses when an inquisitorial body substitutes words more at home in a Court and for that reason alone I think it is unwise, in this instance, to substitute for words that the legislature itself has chosen.  But more than that, in looking at the scheme as a whole, I think the responsibility referred to in the subsection can be no more than a responsibility to establish what the claim is.  The provision cannot by itself deprive the Authority of its role as a Commission of Inquiry with all the attendant duties to fully inquire into such claims as are presented to it.

[38] ...  Further, 'the responsibility of an appellant to establish the claim' can not in itself relieve the Authority from conducting its own inquiry into the claim, as established, unless for other reasons it considers there is no need to do so.

[69] Quite apart from the statutory interpretation issues already referred to, there are at least two fundamental objections to the appellant’s submission.

Authority not a Commission of Inquiry

[70] First, the Authority is not a Commission of Inquiry.  It has, by virtue of Schedule 3C, paras 7 & 9(1), some of the powers of a Commission of Inquiry.  But that does not make it a Commission of Inquiry and it should not be expected to act as if it was: Attorney-General v Moroney [2001] 2 NZLR 652 at [44] (Rodney Hansen J).  This is underlined by the fact that the Authority’s jurisdiction is confined to New Zealand.  It has no jurisdiction or power to require inquiries to be made overseas, and it is in the very nature of refugee claims that virtually all avenues of inquiry will be overseas.  All the Authority is empowered to do by the Immigration Act 1987 is to request the chief executive of the Department of Labour to seek and provide relevant information: s 129P(4).  It cannot require the chief executive to seek and obtain information.  The reasons for this are self-evident and are largely those examined by Nicholson J in AB v Refugee Status Appeals Authority [2001] NZAR 209, 222-225.

[71] There are also resource issues.   A substantial number of refugee claims are received by the Refugee Status Branch each year.  Figures for the past five to six years follow:

Refugee Status Branch
Refugee Applications Received - 1996 to 2002


Financial Year
Applications received by RSB
2001/02 (to 31 March 2002)
1103
2000/01
1694
1999/2000
1393
1998/99
2003
1997/98
1608
1996/97
1534
Total 
9335

[72] The refugee claimants came from seventy-nine countries.  In the period July 2001 to March 2002 alone there were sixty different nationalities.

[73] The Authority, in turn, in the same five to six year period has received a substantial number of appeals:

Refugee Status Appeals Authority
Appeals Received - 1996 to 2002


Financial Year
Appeals received by RSAA
2001/02 (to 31 March 2002)
877
2000/01
640
1999/2000
574
1998/99
578
1997/98
410
1996/97
458
Total
3537

[74] If the appellant’s submission is correct, none of these individuals (each of whom must necessarily be aware of the circumstances which justify his or her assertion to hold a well-founded fear of being persecuted for one of the five Convention reasons) need do more than state what the claim is and then require the refugee status officer and the Authority to investigate fully the claim in the country in question.  The contention is, with respect, unrealistic and untenable.

[75] The resource implications of the appellant’s strained and unnatural interpretation of the statutory provisions are considerable and must be taken into account.  >From time to time various figures appear in the press.  For example, it has been reported that each refugee claimant costs about NZ$30,000, including welfare assistance, but not including health or education: Eugene Bingham, “Refugee frauds cost us millions”, Weekend Herald, Saturday, November 24, 2001.  More recently it has been said that the cost to the New Zealand taxpayer of both  resettlement refugees and asylum-seekers is more than NZ$34 million in their first year in New Zealand and there are ongoing costs in benefits, health and education: Jonathan Milne, “Refugees First-Year Cost $34m” Christchurch Press, January 30, 2002.  Enquiry by the Authority with the New Zealand Immigration Service has disclosed that there are no precise figures, only estimates.  The Authority is told by the Immigration Service that the estimated cost for the average refugee claimant is approximately NZ$12,427, a figure which includes welfare payments while the claim is being processed, health screening, limited ESOL costs, the cost of the initial determination at first instance, the cost of any appeal to this Authority and in some instances, removal costs.  Applying this figure to the 1,694 refugee claimants who lodged their claims in the 2000/01 year, the estimated cost to the taxpayer for this group is NZ$21.05m.  In the same year the estimated cost of the quota refugee resettlement programme (750 persons) was NZ$22.45m.  Whether taken singly or combined together (NZ$43.50m), these figures represent considerable expenditure.  By way of comparison the budget for the New Zealand Immigration Service in the 2000/01 year was NZ$81.08m.  This figure embraces Policy Advice, Visa and Permit Management, Border and Investigations, Support Services - Appeal Authorities, Refugee Services, Settlement Services, Settlement Information, Language Fees Refund and Support for groups working with refugee claimants.  The budget for the Ministry of Foreign Affairs and Trade to provide policy advice and representation directed to the management of New Zealand’s membership of, and interests in, international institutions was NZ$26.4m in the 2000/01 year.

[76] Other examples could be given and the value of the comparisons is admittedly debatable.  But the point is that government revenue is finite and departmental budgets are keenly contested.  Parliament must be taken to be aware of these facts.  The less persuasive, therefore, is any statutory interpretation argument which not only strains the language of the statute, but also demands expenditure of considerable, if not unquantifiable, sums and to require the Authority to pursue what in practice is the unobtainable.  In the result, the submission for the appellant that the Authority must “fully investigate” refugee claims proceeds on an untenable reading of the statute and sets (at significant cost) a task impossible to achieve.

[77] And as will be seen from the next section, the statute, possibly anticipating these difficulties, specifically provides that a refugee status officer and, in turn, the Authority, is not obliged to seek any information, evidence or submissions further to that provided by the claimant.

Authority not obliged to inquire

[78] It is in this context that it is possible to turn to the second fundamental objection, namely the statutory limitation on the duty to inquire.  Section 129G(6) provides:

For the purpose of determining a claim, an officer—

(a) May seek information from any source; but

(b) Is not obliged to seek any information, evidence, or submissions further to that provided by the claimant; and

(c) May determine the claim on the basis of the information, evidence, and submissions provided by the claimant.

Section 129P(2) contains identical provisions relating to the Authority:
The Authority—

(a) May seek information from any source; but

(b) Is not obliged to seek any information, evidence, or submissions further to that provided by the appellant; and

(c) May determine the appeal on the basis of the information, evidence, and submissions provided by the appellant.

[79] Refugee status officers do not have the powers of a Commission of Inquiry.  That is the sole privilege of the Authority.  Yet the statutory limits to the duty of a refugee status officer to inquire and to the duty of the Authority to inquire are expressed in language which is materially indistinguishable.  The fact that the Authority has certain powers under the Commissions of Inquiry Act 1908 does not change the meaning of s 129P(2).  Those powers can only be exercised “within the scope of [the Authority’s] jurisdiction and “subject to Part VIA”.  See Schedule 3C:
7.  Authority to be Commission of Inquiry—

The Authority has the powers of a Commission of Inquiry under the Commissions of Inquiry Act 1908 within the scope of its jurisdiction, and, subject to Part VIA and any regulations made under it, all the provisions of that Act except sections 11 and 12 (which relate to costs) apply to the Authority as if it were a Commission of Inquiry.

[80] In the result, the powers of the Authority under the Commissions of Inquiry Act 1908 are not freestanding or autonomous.  They are expressly subject to Part VIA of the Act and to any regulations made under the Act and the powers must be exercised within the scope of the Authority’s jurisdiction.  The powers under the Commissions of Inquiry Act 1908 are facultative only and do not impose a duty to inquire.  This is emphasised by the clear statement in ss 129G(6) and 129P(2) that there is no obligation on the part of the decision-maker to seek information.

[81] This limitation has a parallel in respect of the Authority’s obligations under the rules of fairness.  Its duty to inquire is not unconfined. The facts of the present case do not require an examination of the nature and extent of the Authority’s duty to inquire under the rules of fairness.  It is sufficient to note that any common law duty to inquire must now be read subject to s 129G(6)(b) and (c) and s 129P(2)(b) and (c).  The statement that a refugee status officer and the Authority is not obliged to seek any information, evidence or submission further to that provided by the claimant and may determine the application on the basis of the information, evidence and submissions provided by the claimant is an express statutory limitation on the duty to inquire and the decisions of SWH v Refugee Status Appeals Authority (High Court, Wellington, CP 203/97, 7 April 1998, Gallen J) and of AB v Refugee Status Appeals Authority [2001] NZAR 209 at [52] (Nicholson J) must now be read with some caution as they both predate Part VIA of the Act.  The resource implications of an unconfined duty to inquire would also have to be taken into account in the fairness context: PP Craig, Administrative Law 4th ed (Sweet & Maxwell, 1999) 416:

“In deciding upon the application of natural justice or fairness the court will, as noted above, balance between, on the one hand, the nature of the individual’s interest, and on the other, the likely benefit to be gained from an increase in procedural rights and the costs to the administration of having to comply with such process rights.”
[82] Importantly Professor Craig points out at op cit 417 that the existence of judicial balancing should not lead to a conclusion that all such balancing is necessarily premised on the same assumptions.  That is, premises which underpin an essentially law and economics approach to natural justice or fairness may be far removed from those which underlie a more rights-based approach to process considerations. There is also force in his observation at op cit 419 that:
“To denominate certain interests as rights for the purposes of procedural protection, and to take no account of other factors in determining the nature of this protection, is implausible given that the costs of such protection have to be borne by society.  As Mashaw states,
‘... we cannot sustain a vision of the world in which rights ring out true and clear, unencumbered by the consideration of conflicting claims of others to scarce resources.  It is the fundamentally compromised nature of social life that interest balancing recognizes and confronts.’
This same point has been recognized by Dworkin, who notes that in both the criminal and civil process the individual is provided with less than the optimum guarantee of accuracy, and that ‘the savings so achieved are justified by considerations of the general public welfare.’”
[83] This small digression into the rules of fairness merely underlines the point that there is nothing objectionable per se to Parliament restricting the duty to inquire in the refugee context.  The restriction in ss 129G(6) and 129P(2) is clear recognition that in this area of the law one cannot approach the task of fact finding with the assumption that there is no limit to the nature and extent of the inquiries which must be undertaken.  Given the infinite variety of circumstances which can give rise to a refugee claim and further given what Professor Hathaway has termed the “ever-present evidentiary voids” and a duty to prognosticate potential risks rather than simply to declare the more plausible account of past events (Professor James C Hathaway, Rebuilding Trust - Report of the Review of Fundamental Justice in Information Gathering and Dissemination at the Immigration and Refugee Board of Canada (December 1993) 6), the work of a refugee decision-maker would never be done.  The decision in Don v Refugee Status Appeals Authority [2001] NZAR 343 (Chambers J) illustrates the point.  In that case one claim was that the inconsistencies in the claimant’s account ought to have been more fully investigated to ascertain whether the explanation lay with the perceived inadequacies of the interpreter.  Further it was submitted that as there was medical evidence suggesting that the claimant suffered from post-traumatic stress disorder the degree to which that disorder (if proved) explained the inconsistencies was also a matter which ought to have been investigated.  The Authority was criticised for failing to make further inquiry.  A further submission was that if the Authority was not satisfied with a claimant’s evidence, it was obliged to seek further evidence.  All the submissions were rejected by Chambers J on the basis of s 129P(2).

[84] Had the 1999 amendments not included ss 129G(6) and 129P(2), cases like Don would raise potentially difficult issues.  As the requirements of fairness vary according to the circumstances of each specific case, it would be difficult for a refugee decision-maker to know with any certainty whether the point had been reached beyond which there was no further duty to inquire.  It is clear that Parliament enacted ss 129G and 129P with the express purpose of providing a clear and specific statement as to the nature and extent of the duty.  The decision-maker has a discretion to seek information, but is not obliged to do so and may determine the appeal on the evidence before it.

[85] It is in this context that it is possible to turn to the submission for the appellant and in particular, his reliance on the statement in T v Refugee Status Appeals Authority [2001] NZAR 749 at [25] that:

Nor do I think that s 129P(2)(b) and (c) provide a general exception to the inquisitorial function.  There will be occasions when at one extreme, a claim is so complete that the pursuit of further material is not needed.  On other occasions the claim may be so frivolous or vexatious, or the answer so plain, that the pursuit of further material is simply not warranted.
[86] The statement that s 129P(2)(b) and (c) do not provide a general exception to the inquisitorial function is largely unexplained.  For the reasons set out above, s 129G(6) and s 129P(2) were explicitly intended to avoid an open-ended duty to inquire and to avoid inevitable arguments as to where, under the rules of fairness, the duty to inquire began and ended.  It is a recognition that in refugee determination the decision-maker seldom has enough “facts” and further, that “facts” alone will not determine whether the person is a refugee.  The decision whether a person is a refugee is not a question of hard fact but of evaluation: Karanakaran v Secretary of State for the Home Department [2000] 3 All ER 449, 477a (CA) per Sedley LJ.

[87] It is now necessary to examine the case law on the burden of proof issue.

THE CASES ON BURDEN OF PROOF

[88] The first three cases (Butler, Mohammed and C) predate the Immigration Amendment 1999, only two (Don, T) postdate the amendment.

Butler

[89] Butler v Attorney-General [1999] NZAR 205 (CA) is of the first importance because it not only explicitly recognises that the burden of establishing the elements of the refugee claim rest on the claimant, the language of the decision is reflected in the text of ss 129G(5) and 129P(1).  No discussion of the burden of proof in the refugee context is complete without Butler.

Ally Hassan Mohammed

[90] In Ally Hassan Mohammed v Refugee Status Appeals Authority (High Court Auckland, M 500/95, 21 December 1995, Tompkins J) at [6] counsel conceded that the burden of proof rested on the refugee claimant, the Authority having referred expressly in its decision to para 196 of the UNHCR Handbook on Procedures and Criteria for Determining Refugee Status.  The decision accordingly contains no discussion of the issue.

C

[91] In C v Refugee Status Appeals Authority (High Court Auckland, M 1365-SW00, 4 May 2001, Nicholson J) the Court at [60] - [62] referred expressly to Refugee Appeal No. 523/92 Re RS (17 March 1995) and stated that the Authority in that case had correctly found that the burden of proof was on the refugee claimant.

[92] It can be seen that prior to the legislative amendments the burden of proof seldom arose as a significant issue before the Authority or before the High Court.  In any event, Butler settled the issue definitively.  Since 1 October 1999 the issue has surfaced only once before the Authority (Refugee Appeal No. 71729/99 [2001] NZAR 183 at [80] - [82]) and twice before the High Court, namely in Don v Refugee Status Appeals Authority [2001] NZAR 343 (Chambers J) and in T v Refugee Status Appeals Authority [2001] NZAR 749 (Durie J).  It is, in practice, not a significant issue, certainly not in Auckland where nearly all refugee hearings take place and where there is an active Refugee Bar.

Don

[93] In Don v Refugee Status Appeals Authority [2001] NZAR 343 (Chambers J) there is no discussion of the burden of proof, the judgment merely recording in passing at [4] that:

There was no dispute that the plaintiff had the responsibility of establishing his claim.  Section 129P expressly so provides.
T

[94] As will have been noted, the appellant’s submissions rely heavily on T v Refugee Status Appeals Authority [2001] NZAR 749 for the proposition that a refugee claimant does not have the burden of proving the claim.  It is easy to overlook the actual holding or ratio of T which is that while the Authority had (in the view of Durie J) erroneously stated that the burden of proof was on T, it had in fact disposed of the matter correctly, adopting a “suitably broad approach”.  See paras [40] & [43] - [44].

[95] Counsel for the appellant and for the New Zealand Immigration Service conceded that the comments made by Durie J on the burden of proof issue were obiter.

[96] It was also conceded by counsel for the appellant that the obiter comments in T are in direct conflict with Butler.  Counsel further stated that he did not seek to distinguish Butler on the basis that it was decided prior to the enactment of the Immigration Amendment Act 1999.  In our view this approach was entirely proper given the close relationship between the formula used in Butler and that subsequently employed in ss 129G(5) and 129P(1).

[97] There can be no question that the Authority is bound by Butler and in that sense it is not necessary to address T any further.  However, given the substantial degree to which the appellant’s submissions rested on T (at least initially - there was a discernable shift during oral argument) it is best that we summarise the difficulties the Authority has in reconciling T to Part VIA of the Act.  The list which follows does not categorise the points in any particular hierarchy.  It is simply an assembly of the points made earlier in this decision.  Unless otherwise indicated, the paragraph numbers refer to the paragraphs of this decision:

(a) The decision in T erroneously assumes at [28] and [29] of that decision that a burden of proof is inconsistent with an investigative or inquisitorial decision-making model.  Case law in Australia and New Zealand demonstrates that this is not necessarily the case.  See earlier in this decision at paras [15] and [16].

(b) No account has been taken of the fact that there is widespread acceptance across a range of jurisdictions and legal systems of the principle that the burden of proof lies on the refugee claimant.  That legal principle is acknowledged and accepted by the Office of the United Nations High Commissioner for Refugees and by leading academics.  See paras [19] to [27] above.

(c) No account has been taken of the fact that the attempt by the Authority to avoid imposing a burden of proof failed in the face of alarming levels of abuse and the withdrawal by a significant number of refugee claimants from the co-operative inquiry process.  See paras [34] to [42] above.

(d) No account has been taken of the statutory obligation of confidentiality.  See para [44] above.

(e) No account has been taken of the practical limits to the refugee inquiry.  See para [45] above.

(f) Apart from a fleeting reference by Durie J at para [31] of his decision, no account has been taken of Refugee Appeal No. 523/92 and the reasons given by the Authority in that case for finding that a refugee claimant must shoulder the obligation of establishing the claim.  Nor has account been taken of the three factors identified by the Authority in that case which mitigate resting the burden of proof on the claimant.  See paras [47] to [49] above.

(g) Notably the decision in T fails to address Butler, notwithstanding that we are told by counsel that the Court of Appeal decision was cited in argument before Durie J.  See paras [50] to [52] above.

(h) The decision fails to read s 129G(5) with s 129B and the rest of s 129G, particularly s 129G(3).  These provisions show that the “claim in New Zealand to be recognised as a refugee in New Zealand” is clearly separate from the information, grounds, evidence and submissions provided by the claimant in support of the claim.  On the interpretation favoured by Durie J the words “responsibility of the claimant to establish the claim” become mere surplusage.  See paras [59] to [67] above.

(i) The decision in T erroneously assumes that the Authority is a Commission of Inquiry and that it has the resources and funding to investigate refugee claims in the manner proposed by Durie J.  See paras [68] to [76] above.

(j) The decision fails to give proper meaning and effect to the express stipulation in ss 129G(6) and 129P(2) that neither a refugee status officer nor the Authority is obliged to seek any information, evidence or submissions further to that provided by the claimant.  See paras [77] to [79] above.

(k) The decision fails to recognise that the powers of the Authority under the Commissions of Inquiry Act 1908 are not freestanding or autonomous and are expressly subject to Part VIA of the Act and regulations and must be exercised within the scope of the Authority’s jurisdiction.  They are facultative and do not override ss 129G(6) and 129P(2).  See paras [80] to [86] above.

[98] As mentioned earlier, it was conceded before the Authority that the obiter comments in T are in direct conflict with Butler and counsel did not seek to distinguish Butler on the basis that it was decided prior to the enactment of the Immigration Amendment Act 1999.  The concessions were properly made but the Authority would, in any event, have ruled that Butler was binding on the Authority, not T.  In these circumstances there is no reason for the Authority to address issue 4 identified at para [7] above, namely how the Authority, as an inferior tribunal, is to decide which competing line of High Court authority is to be followed.  Nor is there any need to address issue 5, namely whether the decision in T was decided per incuriam in that the decision does not address Butler.  There is no need to add unnecessarily to this already lengthy decision.

[99] Next, it is necessary to address the High Court decisions cited in argument which relate  not to Part VIA of the Act, but to the jurisdiction of the Removal Review Authority under Part II.

RELEVANCE OF NON-REFUGEE CASES

[100] The Authority has been referred to various High Court decisions which address the burden of proof issue in the immigration (ie non-refugee) context.  The decisions relate to the humanitarian appeal jurisdiction of the Removal Review Authority under the former s 63B of the Immigration Act 1987 (now s 47) and the jurisdiction of the Deportation Review Tribunal in relation to the deportation of residence permit holders convicted of serious offences (s 105).

[101] In Bajao v Chief Executive of the Department of Labour [2000] NZAR 185 (Wild J), Hullia v Chief Executive Department of Labour [1999] NZAR 412 (Wild J) and De Borja v Removal Review Authority [1999] NZAR 471 (Gendall J) it was held that in a humanitarian appeal to the Removal Review Authority there is a responsibility or onus on the appellant to satisfy that Authority that the statutory criteria exist.  In Faavae v Minister of Immigration (No. 2) (High Court Auckland, M 1434/96, 9 May 1997, Fisher J) it was held that an appellant before the Deportation Review Tribunal carries a burden of proof to satisfy the Tribunal that the statutory criteria stipulated in s 105 of the Act are made out.  It was submitted to us that in so holding Fisher J was not entirely in accord with the earlier decision of Anderson J in Faavae v Minister of Immigration [1996] 2 NZLR 243.  This is not necessarily so.  The comments made by Anderson J on the burden issue were in the context of a judgment pointing out that the Deportation Review Tribunal had erred in assuming that the applicant had to rebut a presumption that the decision by the Minister to deport was a correct decision.  The decision of Fisher J is more to the point that the obligation is on the applicant to satisfy the Tribunal that the conditions required for a successful appeal exist.  If interpreted in this way, the High Court cases are in accord and there is a symmetry between all of the provisions of the Act dealing with residence, removal, deportation hearings and refugee applications.

[102] In the circumstances of the present appeal it is not, however, necessary for the Authority to attempt a definitive reconciliation of the High Court decisions.  The refugee provisions are distinct, if not unique in the Immigration Act 1987 in that s 129G not only imposes a duty to provide information relevant to the claim and a statement of the grounds of the claim, subs (5) specifically states that it is the responsibility of the claimant to establish the claim.  Section 129P(1) is in the same terms.  There is no such provision in the context of residence and removal appeals or of appeals to the Deportation Review Tribunal.

[103] In respect of residence and removal appeals, the provisions stipulate only that it is the responsibility of the appellant to ensure that all information, evidence and submissions the appellant wishes to have considered in support of the appeal are received by the relevant appeal authority within the stipulated appeal period: ss 18F(2)(a) and 50(2)(a).  Neither Authority is obliged to consider any material supplied after that period: ss 18F(2)(b) and 50(2)(b).  The exceptions to this rule are not material in the present context.

[104] In contrast to the Part VIA refugee procedures, neither the residence nor the removal appeal provisions stipulate that “it is the responsibility of the appellant to establish the claim”.  The same is true of the Deportation Review Tribunal provisions in ss 22 and 105.

[105] The unique feature of ss 129G(5) and 129P(1) is that they remove from any doubt the burden of proof issue in the refugee context.

[106] The fact that even in the non-refugee context Wild, Gendall and Fisher JJ have separately held that there is a burden or responsibility of proof underlines this Authority’s own ruling on the burden issue.  The simple point as expressed by Fisher J in Faavae (No.2) at 15-16 is that:

... when it comes to the question whether there was an ‘onus’ on the appellant, no-one has every denied that pursuant to s 105 the Tribunal had to be ‘satisfied’ that certain conditions were established before the appeal could succeed.  The Tribunal had to be ‘satisfied’ that it ‘would be unjust or unduly harsh to deport’ and that ‘it would not be contrary to the public interest to allow the appellant to remain’.  So the appellant carried the risk that if it could not produce or point to the evidence necessary to so satisfy the Tribunal, he or she would lose.  Some lawyers and judges may be uncomfortable in labelling that responsibility as an ‘onus of proof’.  For my own part I have always myself thought that the expression ‘onus of proof’ was customarily applied to a responsibility of that nature.  Indeed I find it difficult to understand how one could accept that the standard of proof required of the appellant in that situation was proof ‘on the balance of probabilities’ (with which no-one seems to have any difficulty) unless one first accepted that there was an onus of proof to which the standard could be applied.

But in any event one should not waste too much energy upon labels.  Everybody agrees that unless and until sufficient evidence is advanced before the Tribunal to satisfy it as to the conditions required for a successful appeal under s 105 the appeal will fail.  So wherever the requisite evidence came from, the burden of persuasion in the end finally rested with this appellant.

CONCLUSION ON THE BURDEN OF PROOF ISSUE

[107] In Refugee Appeal No. 523/92 the Authority held that the burden of proof rested on the refugee claimant, “burden of proof” being understood as the burden of establishing the facts and contentions which support the claim that the individual meets the requirements of the refugee definition.  The language in which this was expressed in Refugee Appeal No. 523/92 was “the obligation of establishing the facts” on which the claim to refugee status is based.  In Butler the burden of proof was described as “the burden of establishing the elements of the claim”.

[108] For the reasons given, the statutory obligation imposed by ss 129G(5) and 129P(1) “to establish the claim” simply re-states the law as it was prior to the Immigration Amendment Act 1999.  While the statute does not use the term “burden of proof” the difference in label is immaterial.  As Fisher J observed in Faavae (No. 2) at 16, one should not waste too much energy on labels.  Wild J noted in Bajao at 189 that the two terms are essentially interchangeable.

[109] In the result, the law remains that as stated in Butler at 213:

A person claiming refugee status has the burden of establishing the elements of the claim.  That rule should however not be applied mechanically.
[110] While for the purpose of this extended analysis the Authority has used the term “burden of proof”, it is acknowledged that in the post-1999 context it is preferable to use the language of the statute, namely “the responsibility to establish the claim”.
 
 
WELL-FOUNDED FEAR

[111] The submission for the appellant, reduced to simple terms, is that the Authority should follow and apply the decision of Einfeld J in Guo v Minister for Immigration and Ethnic Affairs (1996) 135 ALR 421, 439-442 (FC:FC) (Beaumont, Einfeld and Foster JJ).  Expressly and impliedly, the judgment of Einfeld J proceeds on the basis that, in considering the question of a well-founded fear of persecution for a Convention reason, a decision-maker should begin with the hypothesis that there is a real chance of persecution for a Convention reason, examine whether the facts, including “foreseeable future speculation and ... the potentialities” point to the hypothesis and, if so, examine “whether it is negated by other compelling facts”.

[112] The judgment of Einfeld J was emphatically and comprehensively rejected and condemned on appeal in Minister for Immigration and Ethnic Affairs v Guo (1997) 191 CLR 559, 573-577, 592 (HCA) (Brennan CJ, Dawson, Toohey, Gaudron, McHugh, Gummow, Kirby JJ).

[113] Before us counsel for the appellant made no attempt to challenge the comprehensive reasons given by the High Court for holding that the decision given by Einfeld J was wrong in law.  Instead the Authority was simply asked to follow Einfeld J in preference to the High Court of Australia.  The only reason given was that the judgment of Einfeld J had “found favour” with Gallen J in SWH v Refugee Status Appeals Authority (High Court Wellington, CP 203/97, 7 April 1998) at 3.  Counsel’s argument was also notable for the fact that he cited none of the Authority’s leading decisions on the meaning of “well-founded fear”.

[114] The decision of Gallen J in SWH is not authority for the proposition for which it has been cited and there is no reason for the Authority to abandon its long established jurisprudence or to adopt the judgment of Einfeld J in preference to the seven member High Court Bench in Guo.

[115] Because only a handful of cases in the High Court of New Zealand have touched on the well-founded element of the refugee definition and because none have taken account of the Authority’s own jurisprudence, it is necessary to explain why the Authority has followed a line of decisions of the High Court of Australia in understanding the “well-founded” element as requiring no more than a real chance of persecution.  This is the most “liberal” interpretation of well-foundedness which has been accepted in the common law jurisdictions.  Clarity, simplicity and faithfulness to the language, object and purpose of the Refugee Convention are the strengths of the real chance approach.  In view of the misconceived submissions we have received, the merits of the real chance approach need to be re-emphasised.

Jurisprudence of the RSAA

[116] From the time the Authority first sat in 1991 it has adopted and applied the decision in Chan v Minister for Immigration and Ethnic Affairs (1989) 169 CLR 379 (HCA) (Mason CJ, Dawson, Toohey, Gaudron and McHugh JJ) that a well-founded fear of being persecuted is established when there is a real chance of such persecution occurring.  This means that there may be less than a 50% chance of persecution occurring and the chance can be as low as 10%.  But the chance must be substantial as distinct from a remote chance.  The principal decisions of the Authority addressing the issue are Refugee Appeal No. 523/92 Re RS (17 March 1995) 23-27; Refugee Appeal No. 70074/96 Re ELLM [1998] NZAR 252, 260-263 and Refugee Appeal No. 71404/99 (29 October 1999) paras [23] - [40] & [62].  What follows draws together the main strands of these decisions.

[117] Atle Grahl-Madsen in The Status of Refugees in International Law Vol 1 (1966) at 180 postulates the following example:

“Let us for example presume that it is known that in the applicant’s country of origin every tenth adult male person is either put to death or sent to some remote ‘labour camp’, or that people are arrested and detained for an indefinite period on the slightest suspicion of political non-conformity.”
[118] The question posed by this example is whether a one in ten risk, or to express the issue in percentage terms, a ten per cent chance of persecution will qualify as a “well-founded” fear.  In answering this question in the affirmative, Grahl-Madsen goes on to state at op cit 180:
“In such a case it would be only too apparent that anyone who has managed to escape from the country in question will have ‘well-founded fear of being persecuted’ upon his eventual return.  It cannot - and should not - be required that an applicant shall prove that the police have already knocked on his door.”
[119] In further addressing this risk, Grahl-Madsen at op cit 181 goes on to state:
“If the risk is not so clear for all to see as in the above-mentioned example, the determination as to whether there exists ‘well-founded fear’ will be more difficult.  But the real test is the assessment of the likelihood of the applicant’s becoming a victim of persecution upon his return to his country of origin.  If there is a real chance that he will suffer persecution, that is reason good enough, and his ‘fear’ is ‘well-founded’.”
[120] The possible tests of “likelihood” and “real chance” can be seen in this last paragraph.

[121] The test of when a fear of persecution is well-founded was considered by the Supreme Court of the United States of America in Immigration and Naturalization Service v Cardoza-Fonseca (1987) 94 L Ed 2d 434, 447.  Stevens J, delivering the majority opinion, rejected a balance of probability standard for ascertaining the well-foundedness of a fear of persecution.  In specifically referring to the Grahl-Madsen illustration, he stated:

“That the fear must be ‘well-founded’ does not ... transform the standard into a ‘more likely than not’ one.  One can certainly have a well-founded fear of an event happening when there is less than a 50% chance of the occurrence taking place.”
[122] And at 453 he pointed out that there is no room in the definition for concluding that because an applicant only has a ten per cent chance of being shot, tortured, or otherwise persecuted, that he or she has no well-founded fear of the event happening.  He then went on to adopt the “reasonable possibility” standard:
“... so long as an objective situation is established by the evidence, it need not be shown that the situation will probably result in persecution, but it is enough that persecution is a reasonable possibility.”
[123] By specifically referring to the Grahl-Madsen illustration the Supreme Court of the United States clearly accepted that even a one-in-ten chance of anticipated persecution occurring will suffice under the well-founded fear standard.  This is possibly the only positive feature of Cardoza-Fonseca.  The balance of the holdings (which are not relevant in the current context) have been trenchantly criticised.  See James C Hathaway & Anne K Cusick, “Refugee Rights are Not Negotiable” (2000) 14 Geo. Immigra. LJ 481, 485, 515.

[124] When the issue was considered by the House of Lords in the following year in R v Secretary of State for the Home Department, Ex parte Sivakumaran [1988] AC 958 (HL) their Lordships preferred to formulate the test as one requiring the establishment of “a reasonable degree of likelihood” of persecution.  See Lord Keith at 994 (Lords Bridge, Griffiths and Goff agreeing).

[125] Almost immediately the issue was considered by the High Court of Australia in Chan v Minister for Immigration and Ethnic Affairs (1989) 169 CLR 379 (HCA).  The High Court, having given consideration to Cardoza-Fonseca and Sivakumaran, held that an applicant for refugee status would satisfy the definition if he or she showed a genuine fear founded on “a real chance” of persecution for a Convention reason, applying the formulation suggested by Grahl-Madsen.  Mason CJ at 388-389 explained why “real chance” was chosen in preference to the USA and English approaches:

“... I prefer the expression ‘a real chance’ because it clearly conveys the notion of a substantial, as distinct from a remote chance, of persecution occurring ... If an applicant establishes that there is a real chance of persecution, then his fear, assuming that he has such a fear, is well-founded, notwithstanding that there is less than a 50 per cent chance of persecution occurring.”
[126] Toohey J, after referring to the passage from Grahl-Madsen’s text cited above, explained at 407 the advantage of the “real chance” approach:
“The test suggested by Grahl-Madsen, ‘a real chance’, gives effect to the language of the Convention and to its humanitarian intendment.  It does not weigh the prospects of persecution but, equally, it discounts what is remote or insubstantial.  It is a test that can be comprehended and applied.  That is not to say that its application will be easy in all cases; clearly, it will not.  It is inevitable that difficult judgments will have to be made from time to time.”
[127] In Canada the preferred formulations are “reasonable chance” or “good grounds”: Adjei v Canada (Minister of Employment and Immigration) [1989] 2 FC 680, 683-684 (FC:CA).

[128] In New Zealand, the “real chance” formulation has been adopted by this Authority because, in its experience, it is a test which is more readily comprehended and applied (the point made by Toohey J in Chan) and because of its clarity in conveying the notion of a substantial, as distinct from a remote chance, of persecution occurring (the point made by Mason CJ in Chan).  While it may be true in one sense to say that the “reasonable possibility” (USA), the “reasonable degree of likelihood” (UK) and the “real chance” (Australia, New Zealand) tests amount to the same thing, we remain of the view that the real chance test is to be preferred.  The danger inherent in a test formulated in terms of “possibilities” and “likelihoods” is that these terms mean different things to different persons and often shed more confusion than light, as illustrated by the following passage taken from R v Gough [1993] AC 646 (HL) in which the administrative law test for bias was considered.  Lord Goff (with whom Lords Ackner, Mustill, Slynn and Woolf agreed) stated at 670E:

“I prefer to state the test in terms of real danger rather than real likelihood, to ensure that the court is thinking in terms of possibility rather than probability of bias.”
[129] The recent modification of this test in Porter v Magill [2002] 1 All ER 465 at [102] (HL) does not affect the point.  It should also be mentioned that one of the most experienced Canadian refugee law practitioners has described the Sivakumaran test as imposing a higher standard to that adopted in Adjei.  See Barbara Jackman, “Well-Founded Fear of Persecution and Other Standards of Decision-Making: A North American Perspective” in Bahabha & Coll, Asylum Law and Practice in Europe and North America: A Comparative Analysis (Federal Publications, 1992) 37, 44.

[130] It must be remembered, however, that the words used in Article 1A(2) of the Refugee Convention are “well-founded” and that to use the real chance test as a substitute for the Convention term is to invite error.  This is the point made in Minister for Immigration and Ethnic Affairs v Guo (1997) 191 CLR 559, 572; (1997) 144 ALR 567, 576 (HCA).  In that case the High Court of Australia (Brennan CJ, Dawson, Toohey, Gaudron, McHugh & Gummow JJ, Kirby J agreeing) made helpful observations as to when a fear of persecution is well-founded.  The point stressed was that conjecture or surmise has no part to play in determining whether a fear is well-founded.  The majority stated that:

“No doubt in most, perhaps all, cases ... the application of the real chance test, properly understood as the clarification of the phrase ‘well-founded’,  leads to the same result as a direct application of that phrase...  Nevertheless, it is always dangerous to treat a particular word or phrase as synonymous with a statutory term, no matter how helpful the use of that word or phrase may be in understanding the statutory term.  In the present case, for example, Einfeld J thought that the ‘real chance’ test invited speculation and that the tribunal had erred because it ‘has shunned speculation’.  If, by speculation, His Honour meant making a finding as to whether or not an event might or might not occur in the future, no criticism could be made of his use of the term.  But it seems likely, having regard to the context and his Honour’s conclusions concerning the tribunal’s reasoning process, that he was using the term in its primary dictionary meaning of conjecture or surmise.  If he was, he fell into error.  Conjecture or surmise has no part to play in determining whether a fear is well-founded.  A fear is ‘well-founded’ when there is a real substantial basis for it.  As Chan shows, a substantial basis for a fear may exist even though there is far less than a 50% chance that the object of the fear will eventuate.  But no fear can be well-founded for the purpose of the Convention unless the evidence indicates a real ground for believing that the applicant for refugee status is at risk of persecution.  A fear of persecution is not well-founded if it is merely assumed or if it is mere speculation.  In this and other cases, the tribunal and the Federal Court have used the term ‘real chance’ not as epexegetic of ‘well-founded’, but as a replacement or substitution for it.  Those tribunals will be on safer ground, however, and less likely to fall into error if in future they apply the language of the Convention while bearing in mind that a fear of persecution may be well-founded even though the evidence does not show that persecution is more likely than not to eventuate.”
[131] With this statement we respectfully agree.

An objective test

[132] In every decision given by the Authority specific reference is made to the refugee definition in Article 1A(2) of the Refugee Convention and in particular to the phrase “owing to well-founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group or political opinion”.  The Authority then identifies the principal issues as:

1. Objectively, on the facts as found, is there a real chance of the appellant being persecuted if returned to the country of nationality?

2. If the answer is Yes, is there a Convention reason for that persecution?

This formulation of the issues simultaneously addresses itself to the specific language of the Convention and discloses the decision-maker’s understanding of what a well-founded fear is.

[133] It is quite clear that the adjectival phrase “well-founded” qualifies both the word “fear” as well as the word “persecuted” and thus decisively introduces an overriding objective test for determining refugee status.  An unbroken line of case law establishes that the focus of the Convention is not on the facts as subjectively perceived by the refugee claimant, but on the objective facts as found by the decision-maker.

[134] Before the Convention criteria can be satisfied, there must be a well-founded fear of persecution.  As explained by Lord Keith in R v Secretary of State for the Home Department, Ex Parte Sivakumaran [1988] AC 958, 992G (HL):

"... the question whether the fear of persecution held by an applicant for refugee status is well-founded is likewise intended to be objectively determined by reference to the circumstances at the time prevailing in the country of the applicant's nationality.  This inference is fortified by the reflection that the general purpose of the Convention is surely to afford protection and fair treatment to those for whom neither is available in their own country, and does not extend to the allaying of fears not objectively justified, however reasonable these fears may appear from the point of view of the individual in question."
[135] Lord Goff made the same point at 1000D :
"In truth, once it is recognized that the expression "well-founded" entitles the Secretary of State to have regard to facts unknown to the applicant for refugee status, the expression cannot be read simply as "qualifying" the subjective fear of the applicant - it must, in my opinion, require that an enquiry should be made whether the subjective fear of the applicant is objectively justified.  For the true object of the Convention is not just to assuage fear, however reasonably and plausibly entertained, but to provide a safe haven for those unfortunate people whose fear of persecution is in reality well-founded."
[136] Lord Templeman at 996D concurred:
" ... in order for a "fear" of "persecution" to be "well-founded" there must exist a danger that if the claimant for refugee status is returned to his country of origin he will meet with persecution.  The Convention does not enable the claimant to decide whether the danger of persecution exists.  The Convention allows that decision to be taken by the country in which the claimant seeks asylum."
[137] The significance of Sivakumaran lies in the paramount importance given to the objective element of the definition.  The subjective element, in the view of the House of Lords, is of marginal relevance.

[138] In so holding the House of Lords expressly rejected the suggestion in paras 37 and 42 of the UNHCR Handbook on Procedures and Criteria for Determining Refugee Status (1979) that determination of refugee status primarily requires an evaluation of the applicant's statements rather than a judgment on the situation prevailing in the country of origin.  As noted by McHugh J in Chan v Minister for Immigration and Ethnic Affairs (1989) 169 CLR 379, 429 (Mason CJ at 385 agreeing):

"In Sivakumaran the House of Lords, correctly in my view, held that the objective facts to be considered are not confined to those which induced the applicant's fear.  The contrary conclusion would mean that a person could have a "well-founded" fear of persecution even though every one else was aware of facts which destroyed the basis of his or her fear."
[139] We have from the outset been of the view that the Sivakumaran decision should be followed in New Zealand on the issue of the objective component of the refugee definition.  We have been fortified in this view by the fact that the primacy of the objective element has also been recognized by the Supreme Court of the United States in Immigration and Naturalization Service v Cardoza-Fonseca (1987) 94 L.Ed 2d 434 and by the High Court of Australia in Chan v Minister for Immigration and Ethnic Affairs (1989) 169 CLR 379.  Our view has been strengthened by the fact that Professor Hathaway himself has recently pointed out that the Convention does not require consideration of the applicant’s subjective mental state, an opinion he reinforces by reference to the French language text of the Convention (the English and French texts being equally authentic).  See James C Hathaway & Anne K Cusick, “Refugee Rights are Not Negotiable” (2000) 14 Geo. Immigra. LJ 481, 521:
There is no historical basis, however, for the assertion that investigation of a well-founded fear of persecution requires consideration of the applicant’s subjective mental state.  The better view is that, like the French language text, the word “fear” was used by the drafters of the Convention simply to denote a well-founded forward-looking assessment of risk.  The test is therefore fundamentally objective” [Emphasis in original text]
[140] This has long been the Authority’s approach to the Convention.  See Refugee Appeal No. 70074/96 Re ELLM [1998] NZAR 252, 260-263.

[141] The only qualification we need add is that on the issue of the standard of proof of the well-founded fear, we have preferred, for the reasons given, the "real chance" test adopted by the High Court of Australia in Chan to the "reasonable possibility" test of Cardoza-Fonseca and the "reasonable likelihood" test of Sivakumaran.  A helpful discussion of the issues is to be found in James C Hathaway, The Law of Refugee Status (1991) 75-80.

High Court cases

[142] Only in a small number of cases has the well-founded issue surfaced in the New Zealand High Court.  In chronological order they are SWH v Refugee Status Appeals Authority (High Court Wellington, CP 203/97, 7 April 1998, Gallen J); K v Refugee Status Appeals Authority (High Court Auckland, M No. 1586-SW99, 22 February 2000, Anderson J) and DG v Refugee Status Appeals Authority (High Court Wellington, CP 213/00, 5 June 2001, Chisholm J).  Apart from SWH, none refer to the question other than in passing.

[143] In SWH v Refugee Status Appeals Authority (High Court Wellington, CP 203/97, 7 April 1998, Gallen J) it appears to have been accepted by Gallen J at 10-14 that the real chance approach to the well-foundedness element of the refugee definition is the correct approach.  Certainly Gallen J at 14 correctly noted that in Butler v Attorney-General [1999] NZAR 205, 212, 213 (CA) there was no criticism of the “real chance” approach.  For reasons which are not clear, his Honour did not himself employ the term, but that is without significance.  The important point made by Gallen J at 14 is (in effect) that if it is accepted that there is a real chance of persecution taking place, but equally a real chance of persecution not taking place, the fear of persecution remains well-founded.  Plainly that is correct and the Authority has always proceeded on that basis.  The relevant passage reads:

In determining the likelihood or otherwise of the factual material on which the applicant relies to establish the objective nature of the fears, all that is necessary for the applicant to establish is that there is a real possibility that what he or she fears will occur, will occur.  It is wrong to consider that in terms of a balance of probabilities.  In determining the question, the Authority is obliged to consider potentialities, but ought not to determine that out of a range of possibilities, one is more likely than another.  If one which justifies the concerns of the applicant has a real possibilities of occurrence, then that is sufficient.
[144] It is clear from the context that the reference in the penultimate sentence to “a range of possibilities” means “a range of real possibilities”.

[145] The difficulty with the appellant’s submissions is that they require the adjective “real” to be erased from “real chance” and “real possibility”.  It is telling that the submissions rely entirely on the judgment delivered by Einfeld J in Guo Wei Rong v Minister for Immigration and Ethnic Affairs (1996) 135 ALR 421, 442 where the learned Judge stated that the real chance test “is in essence a test of possibilities” and that the decision-maker must engage in speculation of the future (p 440).  This unfortunately conflates a real chance or real possibility with “a” possibility.  In addition, the “practical approach”suggested by Einfeld J at 442 and urged on us by counsel for the appellant is pre-eminently impractical:

In summary, therefore, there will in the present and most other cases be a five stage process:

(1) Identify the hypothesis.  In most cases this will simply be whether there is a real chance of persecution on a Convention ground.

(2) Note the relevant evidence.

(3) Engage in foreseeable future speculation and note the potentialities.

(4) Address the question of whether or not the evidenced facts point to the hypothesis.

(5) If so, examine whether it is negated by other compelling facts.

The advantage of this practical approach is that it would turn the tribunal’s mind first to speculation and a consideration of all the evidence, and reduce the risk that the tribunal will weigh up alternatives and adopt a balance of probabilities test.

[146] On appeal by the Minister of Immigration in Minister for Immigration and Ethnic Affairs v Guo (1997) 191 CLR 559 (HCA) the High Court of Australia pointed out that there are several difficulties with this approach.  We mention two only:

[147] First, in the context, the reference by Einfeld J to speculation was not to making a finding as to whether or not an event might or might not occur, but to conjecture or surmise.  In this he was in error.  This is explained by the High Court at 572.  There, after referring to the statement by Mason CJ in Chan that a real chance of persecution will exist even if there is less than a 50% chance of persecution occurring and the statement by McHugh J in the same case that a real chance of persecution excludes a far-fetched possibility of persecution but that as little as a ten percent chance of persecution may constitute a well-founded fear of persecution, the Court stated:

Chan is an important decision of this Court because it establishes that a person can have a well-founded fear of persecution even though the possibility of the persecution occurring is well below 50 percent.  But to use the real chance test as a substitute for the Convention term “well-founded fear” is to invite error.

... it is always dangerous to treat a particular word or phrase as synonymous with a statutory term, no matter how helpful the use of that word or phrase may be in understanding the statutory term.  In the present case, for example, Einfeld J thought that the “real chance” test invited speculation and that the Tribunal had erred because it “has shunned speculation”.  If, by speculation, his Honour meant making a finding as to whether or not an event might or might not occur in the future, no criticism could be made of his use of the term.  But it seems likely, having regard to the context and his Honour’s conclusions concerning the Tribunal’s reasoning process, that he was using the term in its primary dictionary meaning of conjecture or surmise.  If he was, he fell into error.  Conjecture or surmise has no part to play in determining whether a fear is well-founded.  A fear is “well-founded” when there is a real substantial basis for it.  As Chan shows, a substantial basis for a fear may exist even though there is far less than a 50 percent chance that the object of the fear will eventuate.  But no fear can be well-founded for the purpose of the Convention unless the evidence indicates a real ground for believing that the applicant for refugee status is at risk of persecution.  A fear of persecution is not well-founded if it is merely assumed or if it is mere speculation.  In this and other cases, the Tribunal and the Federal Court have used the term “real chance” not as epexegetic of “well-founded” but as a replacement or substitution for it.  Those tribunals will be on safer ground, however, and less likely to fall into error if in future they apply the language of the Convention while bearing in mind that a fear of persecution may be well-founded even though the evidence does not show that persecution is more likely than not to eventuate.

[148] This passage was expressly adopted by the Authority in Refugee Appeal No. 71404/99 (29 October 1999) at [37] and Refugee Appeal No. 71729/99 (22 June 2000) at [61].

[149] Second, the High Court of Australia stigmatised the “practical” approach suggested by Einfeld J as “ingenious” and unsupported by the terms of the Convention or the proper approach to administrative decision-making in this context.  The approach of Einfeld J was said by the High Court to owe much to the statutory system dealing with the grant of pensions to war veterans.  See the following passage at 573-574:

Expressly and impliedly, the judgment of Einfeld J proceeds on the basis that, in considering the question of a well-founded fear of persecution for a Convention reason, the Tribunal should begin with the hypothesis that there is a real chance of persecution for a Convention reason, examine whether the  facts, including “foreseeable future speculation and ... the potentialities” point to the hypothesis and, if so, examine “whether it is negated by other compelling facts”.

Ingenious as his Honour’s approach may be, it is not supported by the terms of the Convention or the proper approach to administrative decision making in this context.  It is an approach the genesis of which can be found in the decisions of this Court dealing with the very different question of the grant of pensions in cases concerning injuries to war veterans.  But the law governing those cases - s 120 of the Veterans’ Entitlements Act 1986 (Cth) - expressly provided that the relevant tribunal shall be satisfied beyond reasonable doubt that there is no sufficient ground for determining a causal connection between certain matters if, after a consideration of the whole of material before it, it is of the opinion that the material before it does not raise a reasonable hypothesis connecting those matters.  However, the legislation governing refugee cases has no statutory counterpart to s 120.  To approach refugee cases in the way that his Honour suggests is to assume that there is always a well-founded fear of persecution unless the facts negate it.  For the reasons that we have already given, the Act does not support such an approach.  Nor do the general principles of administrative law which underpin the AD (JR) Act give any support for that approach.

[150] Neither of these two fundamental, if not unanswerable, objections are even noted by Gallen J in SWH.  Indeed there is not even recognition at p 12 of the decision that the case of “Minister for Immigration and Ethnic Affairs v Guo and Another [1997] 144 ALR 567” is linked to the case of “Guo Wei Rong v Minister for Immigration and Ethnic Affairs and Another [1996] 135 ALR 421”.  In these circumstances we cannot accept the submission that Gallen J preferred the decision of Einfeld J to that of the High Court of Australia which expressly overruled Einfeld J and strongly criticised his approach.  This is reinforced by the fact that Gallen J did not adopt “speculation” as the test, nor did he adopt the “practical approach” advocated by Einfeld J, as can be seen from the passage from p 14 of SWH already set out above which does not refer to speculation, but to a “real possibility”.

[151] The additional point which must be made is that there is no significance in the use by Gallen J of “real possibility” rather than “real chance”.  The latter expression has respectable lineage, originating with Atle Grahl-Madsen and adopted by the High Court of Australia in Chan.  It has been adopted by this Authority for the reasons given earlier and it is the most “liberal” explanation of the well-founded requirement.  Experience has shown that it is also the expression least likely to be misunderstood by decision-makers.  It also removes refugee determination from the complexities, at times bordering on befuddlement, inherent in using curial terms such as “likelihood”, “probability” and “possibility” when determining the chance of something occurring in the future.  In the result SWH provides no basis for adopting “real possibility” in favour of “real chance”.

There is no balance of probabilities test

[152] The misguided notion that “facts” in the refugee context must be proved to a balance of probability standard has never been accepted by the Authority.  In this the New Zealand jurisprudence as developed by this Authority has avoided the problems which have bedevilled decision-makers in Australia and the United Kingdom.  The Authority has deliberately avoided adopting a balance of probabilities test for “facts” and has consciously refrained from using the term “probabilities”.  The reasons are those elegantly stated by Sedley LJ in Karanakaran v Secretary of State for the Home Department [2000] 3 All ER 449, 479b (CA):

The question whether an applicant for asylum is within the protection of the convention is not a head-to-head litigation issue.  Testing a claim ordinarily involves no choice between two conflicting accounts but an evaluation of the intrinsic and extrinsic credibility, and ultimately the significance of the applicant’s case.  It is conducted initially by a departmental officer and then, if challenged, by one or more tribunals which, though empowered by statute and bound to observe the principles of justice, are not courts of law ... Such decision-makers, on classic principles of public law, are required to take everything material into account.  Their sources of information will frequently go well beyond the testimony of the applicant and include in-country reports, expert testimony and - sometimes - specialised knowledge of their own (which must of course be disclosed).  No probablistic cut-off operates here: everything capable of having a bearing has to be given the weight, great or little, due to it.  What the decision-makers ultimately make of the material is a matter for their own conscientious judgment, so long as the procedure by which they approach and entertain it is lawful and fair and provided their decision logically addresses the convention issues.  Finally, and importantly, the convention issues from first to last are evaluative, not factual.  The facts, so far as they can be established, are signposts on the road to a conclusion on the issues; they are not themselves conclusions.  How far this process truly differs from civil or criminal litigation need not detain us now.
[153] We also respectfully agree with the following passage from the judgment of Sedley J at 479j:
While, for reasons considered earlier, it may well be necessary to approach the convention questions themselves in discrete order, how they are approached and evaluated should henceforward be regarded not as an assault course on which hurdles of varying heights are encountered by the asylum seeker with the decision-maker acting as umpire, nor as a forum in which the improbable is magically endowed with the status of certainty, but as a unitary process of evaluation of evidential material of many kinds and qualities against the convention’s criteria of eligibility for asylum.
Conclusion on well-foundedness

[154] The phrase “real chance” best captures and clarifies the meaning of the phrase “well-founded”.  However, as explained by the High Court of Australia in Guo at 572, the phrase is not synonymous with the Convention term.  It is an epexegetic of “well-founded”, not a replacement or substitution for it.  In determining whether a fear of persecution is well-founded an assessment must be made as to whether or not an event might or might not occur in the future.  In this limited sense speculation is required.  But conjecture or surmise has no part to play in determining whether a fear is well-founded.  A fear is “well-founded” when there is a real substantial basis for it.  A substantial basis for a fear may exist even though there is far less than a 50 percent chance that the object of the fear will eventuate.  But no fear can be well-founded for the purpose of the Convention unless the evidence indicates a real ground for believing that the applicant for refugee status is at risk of persecution.  A fear of persecution is not well-founded if it is merely assumed or it if is mere speculation.
 
 

THE MEANING OF PERSECUTION

[155] The third argument originally advanced by the appellant was that a “dictionary” approach to the interpretation of “persecution” should be adopted in contrast to the more holistic approach articulated by Professor James C Hathaway in The Law of Refugee Status (Butterworths, 1991) 104-105, namely that persecution may be defined as the sustained or systemic violation of basic human rights demonstrative of a failure of state protection.  However, at the hearing on 23 July 2001 this submission was abandoned and the Authority was told that the appellant does not challenge the decision of Chisholm J in DG v Refugee Status Appeals Authority (High Court Wellington, CP 213/00, 5 June 2001) which endorsed the “sustained or systemic violation of basic human rights” interpretation.

[156] In the these circumstances there is no need to address the persecution issue, thereby making an already long decision even longer.

[157] We would only add that the Authority’s approach to the persecution issue was most recently set out in Refugee Appeal No. 71427/99 [2000] NZAR 545 at [43] to [55].  The reasons for not adopting a “dictionary” interpretation are explained in these paragraphs.  We would reinforce these reasons by reference to the point made by Professor JF Burrows in Statute Law in New Zealand 2nd ed (Butterworths, 1999) at 189-191 namely that to give words their “ordinary meaning” is not always the inexorable guide to a decision it is made out to be.  First, there are not many words in the English language that have only one ordinary meaning.  A glance at any dictionary reveals that most words have several shades of meaning, all of them perfectly “ordinary”.  The sense in which such a word is being used in a statutory provision depends on the context and purpose of the provision.  The meaning one is seeking is the natural and ordinary meaning of the word read in context, and in the light of the purpose of the provision.  This is precisely the point made by the Authority in Refugee Appeal No. 71427/99 at [43] and [44] applying the Vienna Convention on the Law of Treaties, 1969, Article 31(1), the international analogue to the Interpretation Act 1999.

[158] Secondly, as Professor Burrows points out at op cit 190, language is not of mathematical precision.  Sometimes what to one judge is clearly within the ordinary meaning of a word will not be to another.  A graphic illustration provided by Professor Burrows is R v Maginnis [1987] AC 303 (HL), a case in which the House of Lords divided over the ordinary meaning of the word “supply”.

[159] While the decision in DG v Refugee Status Appeals Authority is welcome support for the Authority’s approach, the decision is necessarily confined by the narrow terms in which the issue arose for determination by the High Court.  The decision does not explore the rationale for the dynamic and purposive interpretation articulated by Professor Hathaway.  For this reference must be made to the Authority’s own jurisprudence.  The High Court has approved the principle.  The decisions of the two jurisdictions must be read alongside each other.  Counsel appearing before the Authority must necessarily be aware of the Authority’s jurisprudence and of the underlying reasons for the approach it has taken to the meaning of the word “persecution”.  It is a progressive approach, finding increasing favour overseas.  It is regrettable that the Authority should have to spend so much time defending a jurisprudence which counsel overseas would wish their own decision-makers to adopt.  But perhaps counsel for the appellant is unaware of this fact.

CONCLUSION

[160] All of the preliminary issues raised by the appellant having been disposed of it is appropriate that the appeal be set down for hearing.  The Secretariat will liaise with counsel for the appellant in the usual way.  For the reasons given earlier in this decision, counsel representing the New Zealand Immigration Service do not, for good reason, wish to be involved in the hearing of the merits of the refugee claim.
 
 

"RPG Haines"
 [Rodger Haines QC]
 Chairperson