- Program in Refugee and Asylum Law
- Additional Resources
The Israeli Supreme Court recently handed down concomitantly two decisions that would further restrict the employment opportunities of asylum seekers in Israel. In one decision, the Court ruled that a 20% tax, levied on employers who employ “foreign workers,” would apply to employers who employ Eritrean and Sudanese nationals. The court ruled so, despite the fact that Eritreans and Sudanese have been under a temporary protection (TP) regime in Israel for over a decade; and despite the fact that there is a governmental declared policy of turning a blind eye to employers who employ Eritrean and Sudanese nationals, notwithstanding the fact that the latter do not get working permits under the TP regime. In the second decision, the Court ruled that the Ministry of Finance’s directive, according to which employers who employ “foreign workers” cannot bid in governmental and municipal tenders, also applies to employers of Eritreans and Sudanese.
Between 2006 and 2013, around 60,000 Eritrean and Sudanese asylum seekers entered Israel through its then open southern border (around 40,000 are still in the country). Israel applied a TP regime to them from the beginning, but did not allow them to apply for asylum until around 2013. Since then, several thousand Eritrean and Sudanese nationals have filed asylum applications and are waiting for a decision; thousands of others applied for asylum and were rejected, and several thousand who did not formally apply for asylum at all (for various reasons) are simply covered by the TP regime. Very few – around nine Eritrean and Sudanese nationals – were granted refugee status. The Court’s aforementioned two decisions do not apply to the lucky nine, yet they apply to all other “classes” of Eritrean and Sudanese enumerated above.
In this short article, I wish to highlight only one aspect of the Court’s decisions, which is the (in)applicability of the Refugee Convention to the cases. I think this matter is of special importance since, while similar measures apply in other countries, their compatibility with the Refugee Convention has never been judicially reviewed.
Indeed, similar types of “foreign worker taxes” are levied on employers of non-citizens in several Asian jurisdictions including Singapore, Malaysia, Thailand and Hong Kong. Moreover, evidence shows that the “foreign worker tax” is levied in these territories on employers of asylum seekers and migrant workers alike. However, none of these jurisdictions is a signatory to the Refugee Convention. Therefore, the legitimacy of such measures was never analyzed in these countries in light of the Refugee Convention. The Supreme Court of Israel seems to be the first judiciary in the world to have done so. Alas, it got it wrong.
The Israeli Court remarked that the appellants (employers of Eritrean and Sudanese nationals) erred for positioning that all Eritreans and all Sudanese nationals are refugees, because such an assumption “undermines the authority vested in the State to determine—according to international law principles—who will be given refuge within its territory.” However, in order for Articles 29(1) or 17(2) of the Refugee Convention – the articles that the appellants referred to – to be applicable, the Eritrean and Sudanese asylum seekers need not be recognized as refugees by the Israeli authorities. According to Hathaway’s well-known theory of “levels of attachment,” which is grounded in the Convention’s language and its travaux preparatoires, some of the Convention’s provisions apply not only for recognized refugees, but also for asylum seekers, particularly those who wait for decisions in their applications for a long time.
As mentioned above, some of the Eritrean and Sudanese nationals in Israel are in this exact situation. Things might be different for those Eritreans and Sudanese nationals who simply rely on the long-standing Israeli TP regime. It is legally debatable if they are also entitled to the protection of the aforementioned articles of the Convention. On the one hand, the Convention does not regularly apply to adult migrants who never applied for asylum. On the other hand, the fact that some of them registered with UNHCR upon arrival to Israel, and were barred for years from applying for asylum with the Israeli RSD Unit, might justify treating them as if they had applied for asylum. This matter requires more in depth legal investigation.
However, the court failed to delve into these pertinent questions. It instead ruled that even if arguendo all Eritreans and Sudanese nationals in Israel were entitled to Article 29(1) and Article 17(2) protection, there was no infringement of these two articles by the taxation imposed on their employers.
Article 29(1) provides as follows:
“The Contracting States shall not impose upon refugees duties, charges or taxes, of any description whatsoever, other or higher than those which are or may be levied on their nationals in similar situations”.
The Court ruled that Article 29 is irrelevant to the taxation under consideration since the taxation is not imposed on refugees themselves and does not affect them. In so ruling, the Court relied on the fact that the tax is levied on the employers, rather than the employees (refugees), and, furthermore, since the statute that imposed this tax explicitly forbids the employer from “rolling” its costs to her employees, no indirect affect is expected. The court disregarded the Article’s broad language, and did not delve into the interpretation of the Article’s words “of any description whatsoever.” The court also refused to consider the realistic view that despite the prohibition, employers, whose prime interest is to maximize their profits, would nevertheless transfer the cost of the tax (partially or in whole) onto their employees.
The Court then swiftly considered article 17(2). This article provides as follows:
“[R]estrictive measures imposed on aliens or the employment of aliens for the protection of the national labour market shall not be applied to a refugee… who fulfils one of the following conditions:
(a) He has completed three years’ residence in the country;
(b) He has a spouse possessing the nationality of the country of residence. A refugee may not invoke the benefits of this provision if he has abandoned his spouse.
(c) He has one or more children possessing the nationality of the country of residence.”
As Hathaway accurately observed, Article 17(2) undoubtedly confers a duty to exempt refugees from restrictive measures “whether the restriction is formally directed at non-citizens themselves, or at [their] employers.” While the plain language of Article 29(1) indeed refers to the imposition of taxes directly on refugees themselves, this cannot be said of Article 17(2). Furthermore, the Court acknowledged that the tax at issue was endorsed to protect the national labor market (minimize the risk of unemployment by Israelis), which is exactly what the plain language of Article 17(2) forbids states from doing, as far as refugees are concerned. Nevertheless, the Court dismissed the applicability of the Article to the case, on the grounds the appellants presented no evidence to support the finding that the tax would adversely affect the Eritrean and Sudanese employees’ financial situation.
The Court observed that Eritrean and Sudanese nationals usually work in jobs that are unattractive to Israelis. Therefore, it argued, these employers have no real option but to employ Eritreans and Sudanese. Hence, the Court concluded that Eritreans and Sudanese nationals would not lose their jobs because of the taxation. As mentioned above, it further ruled that they would not bear the costs of the taxation, simply because the law prohibits the employers from rolling the taxation onto their employees. What brutal logic.
With all due respect, I think the Court’s reasoning is flawed in several ways. First, it seems, as was mentioned before, to ignore Article 17’s plain language. The wording of article 17(2) does not suggest a consequentialist criterion (i.e. a criterion according to which only measures which would be detrimental to refugees are the subject of prohibition). It straightforwardly prohibits the imposition of a measure “for the protection of the national labour,” which means that the purpose of the measure, rather than its outcome (i.e. would asylum seekers be adversely affected by it) is what lies at the heart of the prohibition. One might further argue that the Court is inconsistent in its interpretational method: in interpreting Art. 29, the Court used a strict literary interpretation, while it used a “purposive” interpretation in interpreting Article 17(2).
Second, and more importantly, the Court seems to make three dubious factual assumptions: (a) that only foreigners would (necessarily) perform the “three Ds” jobs (dangerous, difficult and dirty) in Israel, (b) all employers are law abiding, and therefore would not roll the tax costs onto their disadvantaged employees as forbidden by law and, most surprisingly, (c) the tax, which was believed and overtly intended by the legislator to induce employers to prefer Israeli workers over non-Israelis, would not meet its goal. The Court gives no explanation or references to any of these assumptions. Oddly, it did not even consider making the more reasonable opposite presumption, according to which the law imposing the tax would indeed meet its declared goal. Such a presumption would have demonstrated more respect for the Israeli Knesset. At the very least, the Court could have shifted the burden of persuasion onto the Government. Then, the Government would have to persuade the Court that the tax would not achieve its declared goal, in the sense that employers would continue to prefer Eritrean and Sudanese workers notwithstanding the law, and would keep their legal duty not to transfer the tax costs onto them.
Hence, in my opinion the Court erred at minimum in its refusal to apply Article 17(2) of the Refugee Convention to the case. As many Eritrean and Sudanese asylum seekers long await a decision in their asylum applications, and as practically all have waited in Israel for over three years,  Article 17(2)(a) applies to them, and the tax should not be levied on their employers.
Furthermore, in its second decision regarding the exclusion of Eritrean and Sudanese nationals from working for manpower agencies providing services to the governmental and municipal sectors, the Court failed to consider the applicability of Articles 17(2) and 29(1) altogether. In this case too, the Court recognized that the underlying and declared purpose of the Ministry of Finance’s directive was to favor the employment of Israelis over foreigners. Nevertheless, the Court was satisfied with saying that the Eritreans and the Sudanese nationals could still work for manpower agencies which provide services for the private sector. It is unreasonable, however, to assume that such a measure, and especially both measures together, would not hinder the employment opportunities of asylum seekers. At the very least, the burden should have been shifted to the government to persuade the court that this would be the case.
The Israeli Knesset and government has been very creative in the last decade in coming up with new measures to make the lives of asylum seekers less pleasant in the country. Until now, the Israeli Supreme Court refrained from reviewing these measures in light of the “bill of rights” that asylum seekers and refugees are entitled to under Articles 2-34 of the Refugee Convention. It focused instead, on Israeli constitutional law and on the principle of non-refoulement. It is refreshing therefore, that the Court found it apt to consider Articles 17(2) and 29(1) of the Convention in the aforementioned cases. It would have been even better if the Court had done so in more depth and with an open heart. If it had done so, I am convinced the Court would have ruled that at least a subclass of the Eritrean and Sudanese asylum seekers—those who await for decisions in their asylum application for over three years—were exempt (themselves, or their employers) from the restrictions reviewed in the two aforementioned cases.
 In HCJ 6312/10 Kav La’Oved v. The Government (2011) (Isr.) https://perma.cc/7SPP-4DHU (ruling that for the time being the Attorney General’s “policy” of not fining employers of Eritreans and Sudanese under TP for illegal employment strikes a “proper balance” between the interests of the asylum seekers and the interests of the State of Israel).
 Reuven (Ruvi) Ziegler, No Asylum for ‘Infiltrators’: The Legal Predicament of Eritrean and Sudanese Nationals in Israel, 29 J. Immigr., Asylum and Nat’lity L. 172,181 (2015).
 Eritreans who file for asylum based on their defection from the military service are routinely denied asylum based on Israel’s legal position that military defection does not constitute a ground for asylum under the Refugee Convention; various proceedings regarding this position are pending before the Israeli court system. See AdminA 32641-10-16 State of Israel v. Doe (2017) https://perma.cc/BD4A-C53U.
 See supra note 5, at 181.
 Amarjit Kaur, Managing Labour Migration in Malaysia: Guest Worker Programs and the Regularisation of Irregular Labour Migrants as a Policy Instrument, 38 J. Asian Studies L. Rev. 345, 349, 351 (2014); Elizabeth Ruppert, Managing Foreign Labor in Singapore and Malaysia 8, 18 (World Bank, Policy Research Working Paper, 1992); Srawoorth Paitoonpong & Montinee Chaksirinont, A Foreign Levy For Thailand, TDRI Quarterly Review 3 Sept. 2007; Hossein Jalilian & Glenda Reyers, Migrants of the Mekong: Wins and Losses, in Costs and Benefits of Cross-Country Labour Markets in the GMS 1,81 (Hossein Jalilian ed., 2012).
 As seen in the list of states party to the Refugee Convention (and its accompanying 1967 Protocol), Singapore, Malaysia and Thailand never signed these instruments. In regard to Hong Kong, while China is a party to the Convention and the Protocol, it never extended its application to Hong Kong. See https://perma.cc/9M7H-CVDT. It should be noted that the relevant provisions of the Refugee Convention, Articles 17(2) and 29(1)discussed below, are not norms of customary international law. Therefore, countries who have not signed and ratified the Convention and the Protocol are not bound by their decrees under international law.
 Supra note 1, at para. 50.
 Such a recognition would be of a declaratory, rather than a constitutive, nature. A person is a refugee because he or she meets the criteria of the “refugee” definition in the Convention, not because a state has declared him or her to be one. See James C. Hathaway & Michelle Foster, The Law of Refugee Status, 1 (2014).
 James C. Hathaway, The Rights of Refugees Under International Law, 184-85, 192 (2005)
 Article 29(1) applies to “refugees,” i.e. asylum seekers who are in the first level of attachment. Such “level” is met by mere lodging of an asylum application which has not yet been rejected. Article 17(2) also – and unlike Article 17(1) – refers to “refugees.” See Hathaway, supra note 12 at 756. Yet, Article 17(2)(a) refers to refugees who were already present in the receiving country for three years.
 The Court did not explicitly consider, in the Y. B. Si Resources decision, see supra note 3, whether the inability to work in companies provide services to governmental or municipal entities is compatible with Articles 29(1) and 17(2) of the Convention.
 UN Convention on Refugees art. 29(1).
 See Hathaway, supra note 12, at 530; Boldizsar Nagy, Fiscal Charges in The 1951 Convention Relating to the Status of Refugees and its 1967 Protocol: A Commentary 1214, 1222 (Andreas Zimmermann ed., 2011).
 Supra note 1, at para. 51.
 UN Convention on Refugees art. 17(2).
 See Hathaway, supra note 12, at 760-61; See also Nehemiah Robinson, Convention relating to the status of refugees; its history, contents and interpretation; a commentary 115 (1953)(explaining that “the first category may relate to measures taken by the authorities directly at the foreigner… [t]he second group apparently deals with restrictions imposed on the employer… [i]n order to combine all the possibilities, the authors of the Convention combined both cases of restrictions.”)
 See Vienna Convention on the law of treaties art. 31(1), May 23, 1969, U.N.T.S vol. 1155 p. 331 (providing that “[a] treaty shall be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose.”). Here, the Court did not refer to Articles 31-32 of the Vienna Convention for the law of treaties.
 Oddly, the court also did not suggest what would be the purpose of the taxation if it would not meet the legislator’s declared goal of incentivizing employers to prefer Israeli employees to non-Israeli ones.
 Since Israel did not enable Eritrean and Sudanese asylum seekers to apply for asylum until around 2013, an interesting question might arise regarding those who had been living in Israel for over three years, yet filed for asylum less than three years prior to the date which the authorities wished to impose the taxation upon their employers. Even if one argues that the such asylum seekers do not fulfil the “three years” criterion stipulated in Article 17(2)(a), there are many Eritrean and Sudanese asylum seekers who filed for asylum three years prior to the Court’s ruling. I would, however, hesitate to argue that they do not fulfill the criterion, given the government induced confusion regarding the possibility to apply for asylum throughout the years. Finally, for a position supporting taking into account the time the asylum seeker was unable to file for asylum see Alice Edwards, Article 17 in The 1951 Convention Relating to the Status of Refugees and its 1967 Protocol: A Commentary 950, 969 (Andreas Zimmermann ed., 2011) (noting “a good faith reading would require [Article 17(2)(a)] to be read as including any time spent in the reception State since lodging a claim to asylum, or where refugee status determination is not undertaken or is suspended, from the date of entry to the territory.”) See also UN Convention on Refugees art. 17(3).
 See Paul Weis, The Refugee Convention: The Traveaux Prparatoires Analysed with a Commentary by Dr. Paul Weis, 1951 107: “The second paragraph [of Article 17] does not mean that refugees must be granted national treatment. In many countries aliens require a work permit and in this case, it is required of refugees, too, unless they have been specifically exempted from it, but it has to be accorded to them ex officio if they fulfil any of the conditions stipulated in paragraph 2.” See also Hathaway, supra note 12, at 756 (for an explanation that despite the fact that Article 17(2)(a) uses the term “residence”, it simply refers “to a de facto ongoing presence, rather than to legal notions such as the establishment of domicile.”).