INTRODUCTION
Magyar Farming Company Ltd, Kintyre Kft and Inicia Zrt v. Hungary is a decision of layered significance in international investment law. Rendered under the UK-Hungary Bilateral Investment Treaty of 1987, the EUR 7.1 million finding against Hungary is less consequential than the doctrinal work the tribunal performs in reaching it1. This review focuses on the tribunal’s treatment of the police power defence and its relationship to the expropriation finding. The police power doctrine sits at the site of deepest incoherence as the doctrine is invoked frequently and defined inconsistently producing unpredictable results. Magyar Farming demonstrates with precision what careful reasoning within this framework can produce and what it cannot resolve.
FACTUAL BACKGROUND
A. Genesis of the Dispute
In 1997, British farming professionals led by Mr. Andrew Hunter incorporated Magyar Farming Company Ltd as a UK holding company and, through it, acquired Inicia Zrt, a Hungarian agricultural enterprise holding a twenty-year lease over 760 hectares of State-Owned land in Ikreny. Under the Arable Land Act, Inicia held a statutory prelease right, entitling it to match any third-party offer for the land upon expiry of the lease, which forms the cornerstone of the claimant’s long term investment security. In 2011, the Fidesz government pursuing a land reform policy directed at redistributing State-owned farmland to local family farmers, passed the 2011 Amendment which precluded the exercise of statutory prelease right in connection with National Land Agency Tenders without providing any transitional arrangements for incumbent lessees, compensation, or protection for the vested rights of incumbent lessees. In 2015 Inicia Zrt was forcibly evicted from the land with the State Secretary of the Ministry of Agriculture vowing to “protect Hungarian agricultural land under all circumstances, not only legally, but physically as well2.” The claimants initiated ICSID arbitration in 2017.
B. The Expropriation Finding
The tribunal grounded its expropriation analysis, through the doctrine of acquired rights which was affirmed in Aramco as one of the fundamental principles of public international law3.It further drew on the distinction drawn in Encana v. Ecuador between a general statutory power(which the state may alter prospectively) and a right already vested upon a private party’s reliance on that power(which attracts protection from retrospective extinguishment)4. Inicia’s prelease right had crystallized on conclusion of the lease agreement. The 2011 Amendment extinguished it retrospectively, without carve-out or compensation. The relevant question the tribunal focused on was functional, that is in a market where Inicia couldn’t hold land, the prelease right was the “functional alternative to ownership” in terms of the investor’s expectations of legal certainty and stability5.The award’s most critical finding is that compensation was owed even if the state acted with a legitimate public purpose, evenhandedly and with due procedure of law which directly refutes the proposition of the police power defence that regulatory legitimacy is self-sufficient defence against expropriation.
THE POLICE POWER DEFENCE: SEQUENCING, INCOHERENCE AND THE STANDARD OF REVIEW
Having found expropriation, the tribunal turned to whether the 2011 Amendment was nevertheless a non-compensatory exercise of the police powers. Citing Saluka v.Czech Republic the tribunal expressly acknowledged that “there is no comprehensive test that may be used to distinguish regulatory expropriation, for which compensation is required, from an exercise of police/regulatory powers”6.Rather than resolving this gap, the tribunal constructed a two-category framework of non-compensable measures, that is the enforcement of actions targeting investor wrongdoing( criminal sanctions, license revocations),and measures abating threats posed by investor activities to public health, environment or public order7.Since, Hungary’s land redistribution fell in neither, the tribunal concluded there was “no rationale that would justify exempting Hungary from its duty to pay compensation”8. However, this framework invites critical scrutiny. As per Article 6 of the UK-Hungary BIT, the “equivalent effect” language has been used9.Read under VCLT Article 31(1)10,it anchors this inquiry into effect. The police power doctrine particularly applied through Magyar Farming’s two category framework is departing from the textual basis of the Article 6 itself by focusing on the purpose and category of the measure rather than its effect. The tribunal finds expropriation, then asks whether the purpose-based police power defence overrides it. This creates a structural disconnect between the treaty’s text and the Customary International law derived defence, a disconnect the tribunal does not acknowledge.
A. The Sequencing Problem: Police Power as Backstop or Independent Test?
The first and most fundamental question is whether the police power defence, functions as an independent test or as a confirmatory backstop once expropriation is already established. Ranjan’s analysis of Philip Morris v. Uruguay identifies this as the central structural deficiency of the doctrine as currently applied11. In Philip Morris, the tribunal articulated the police power rule in two distinct formulations12.In paragraph 295, it stated that reasonable, bona fide regulatory measures adopted for public welfare purposes do not constitute expropriation and absolve the State from compensation notwithstanding economic damage caused13.In a later paragraph, it restated the rule as requiring measures to be bona fide, non-discriminatory and proportionate14.These are not equivalent formulations as the first requires only good faith and public purpose while the second adds non-discrimination and proportionality as cumulative requirements which the tribunal never reconciled. Ranjan also notes that the tribunal concluded, using the word “also” and “additional”, that the police power defence provided further reason to reject the expropriation claim after finding no substantial deprivation15.The significance of this sequencing is that the police power defence was doing no independent work and was operating as backstop to a finding that had already been made on other grounds. The question of what the defence would do if substantial deprivation had been found was left unanswered.
The Magyar Farming tribunal does not fully escape this problem. It finds expropriation first, then addresses police power separately which is structurally cleaner than Philip Morris. But then the two-category framework it applies does not supply a rule capable of operating independently of the expropriation finding. More critically, the tribunal’s engagement with Saluka is selective in a manner that undermines its own reasoning. The tribunal cites Saluka paragraph 263 to show that no comprehensive test exists and justify its own two-category framework. Yet it doesn’t engage with Saluka paragraphs 255 to 259, where the tribunal held that measures taken in a non-discriminatory manner and directed at general welfare, or in pursuit of a State’s “political, social or economic ends”, are non-compensable16. Hungary’s land reform could be argued under Saluka’s own substantive formulation. By citing it only for the gap and not for the content of the test that fills it, the Magyar Farming tribunal treats the absence of a bright line as license to construct a framework narrower than the very jurisprudence it invokes and does so without acknowledging the inconsistency. This is precisely what an absence of universal evaluative criteria produces, not bad faith but selective engagement that is camouflaged as principled reasoning.
B. The Proportionality Gap
The reliance on categorical classification thus marginalizes the possibility of making proportionality inquiries. Henckels argues that proportionality, properly applied, requires three discrete and sequential stages which are suitability(a rational connection between the measure and its objective), necessity(whether an alternative less restrictive measure was available to achieve the same objective) and strict proportionality (whether the public benefit outweighs the cost to the investor)17.In the case of Tecmed, the tribunal stated it would afford ‘due deference’ to Mexico policy objectives, but then proceeds directly to a strict proportionality assessment without engaging with suitability or necessity18.This approach of jumping to the third stage without completion of the first two is methodologically flawed and produces outcomes that are too deferential(finding measures proportionate without testing whether less restrictive alternatives existed) and insufficiently deferential(applying stringent scrutiny to the State’s choice of regulatory instrument)19.
The Magyar Farming tribunal does not apply proportionality analysis at all. Its two-category framework operates as a threshold classification that is either the measure falls within a recognized category of non-compensable police powers, or it does not, without any structured engagement with suitability, necessity or balancing. This is not merely an alternative methodology but an abandonment of structured proportionality in favor of a categorical rule. The consequence is that the tribunal never asks whether Hungary could have achieved its objective through arrangements that preserved the vested prelease rights of lessees while reallocating State-owned land as those leases expired. As SD Myers v. Canada established, authorities enacting measures impacting foreign investment are required to select the measure with the least restrictive effect on that investment20.Magyar Farming reaches no conclusion on this question, as the question was never asked.
C. The standard of Review Gap
The arguments above connect to the deeper question of standard of review. Investment tribunals have not developed a coherent approach, lying along a continuum from complete deference to de novo review. The LG&E tribunal set a high threshold, finding measures with a social or general welfare purpose must generally be accepted without liability except where the State’s action is obviously disproportionate21.The Tecmed tribunal, despite professing deference, applied what was effectively de novo scrutiny. Magyar Farming applies neither proportionality nor explicit standard of review. Its two-category framework implicitly affords zero deference to Hungary’s assessment that agricultural land redistribution constitutes a legitimate exercise of police power, without explaining why. This is precisely the structural deficiency Angin identifies that is the absence of universal evaluative criteria produces outcomes that are principled in their particulars but not generalized in their implications22.A tribunal can reach a defensible conclusion on specific facts while providing no guidance for the next case involving economic regulation that serves a genuine but uncategorized public interest.
CONCLUSION
Magyar Farming v. Hungary makes a durable contribution to the vested rights doctrine. Its finding that compensation is owed even for legitimate regulatory action interfering with vested rights directly refutes the proposition that regulatory legitimacy is a self-sufficient defence. But its police power analysis illustrates the persistence of three structural deficiencies as seen above. The sequencing problem is deferred, proportionality is abandoned and the standard of review is implicit. It equally demonstrates that without universal evaluative criteria, a coherent standard of review and structured proportionality analysis, the police power doctrine will continue to produce results that are defensible in isolation and incoherent as a system. The path forward, lies not in better outcomes in individual cases but in the adoption of a coherent, appropriately deferential analytical structure, one that applies proportionality in its proper sequence, articulates an explicit standard of review and provides universal criteria whose absence Angin identifies as the system’s central pathology. Until that framework exists, Magyar Farming will stand as a well-reasoned decision in an unreasoned system.
*****
Footnotes
1. Magyar Farming Company Ltd,Kintyre Kft and Inicia Zrt v. Hungary, ICSID Case No. ARB/17/27.
2. Magyar Farming Company Ltd,Kintyre Kft and Inicia Zrt v. Hungary, ICSID Case No. ARB/17/27,46.
3. Saudi Arabia v. Arabian American Oil Company (Aramco),Ad hoc Arbitration, Award(1963).
4. Encana Corporation v. Republic of Ecuador,LCIA Case No. UN3481,UNCITRAL,Award (Feb. 3,2006).
5. Magyar Farming Company Ltd,Kintyre Kft and Inicia Zrt v. Hungary, ICSID Case No. ARB/17/27,360.
6. Magyar Farming Company Ltd,Kintyre Kft and Inicia Zrt v. Hungary, ICSID Case No. ARB/17/27,365.
7. Magyar Farming Company Ltd,Kintyre Kft and Inicia Zrt v. Hungary, ICSID Case No. ARB/17/27,366.
8. Magyar Farming Company Ltd,Kintyre Kft and Inicia Zrt v. Hungary, ICSID Case No. ARB/17/27,367.
9. Agreement between the Government of the United kingdom of Great Britain and Northern Island and the Government of the Hungarian People’s Republic for the Promotion and Reciprocal Protection of investments art.6,Mar.9,1987.
10. Vienna Convention on the Law of Treaties art.31,May 23,1969,1155 U.N.T.S.331.
11. Prabhash Ranjan, Police Powers, Indirect Expropriation in International Investment Law and Article 31(3)(c) of the VCLT:A critique of Phillip Morris v. Uruguay, 9 ASIAN J.INT’L K.98(2019).
12. Philip Morris Brands Sarl, Philip Morris Products S.A. and Abal Hermanos S.A. v. Oriental Republic of Uruguay ,ICSID Case No.ARB/10/7,Award (Jul.8,2016).
13. Ranjan,supra note 11,at 110.
14. Ranjan,supra note 11,at 107.
15. Ranjan,supra note 11,at 108.
16. Saluka Investments B.V. v. The Czech Republic,UNCITRAL,Partial Award (Mar.17,2006),253-259.
17. Caroline Henckels, Indirect Expropriation and the Right to Regulate: Revisiting Proportionality Analysis and the Standard of Review in Investor-State Arbitration, 15 J. INT’L ECON. L. 223 (2012).
18. Técnicas Medioambientales Tecmed, S.A. v. The United Mexican States, ICSID Case No. ARB(AF)/00/2, Award (May 29, 2003).
19. Henckels,supra note 17,at 223.
20. S.D. Myers, Inc. v. Government of Canada, UNCITRAL, Partial Award (Nov. 13, 2000).
21. LG&E Energy Corp., LG&E Capital Corp., and LG&E International, Inc. v. Argentine Republic, ICSID Case No. ARB/02/1, Decision on Liability (Oct. 3, 2006).
22. Berfu Beysulen Angin, The Right to Regulate vs Investment Protection: Unveiling the Causes of Imbalance and the Limits of Current Reform Efforts in International Investment Law, 40 ICSID REV. 11 (2025).