17 May 2018

New model treaty to replace 79 existing Dutch bilateral investment treaties

Yesterday, the Dutch Ministry of Foreign Affairs launched an internet consultation in relation to a new draft model Bilateral Investment Treaty (‘BIT’) that is intended to serve as the basis for the re-negotiation of the 79 BITs that are currently in place between the Netherlands and States outside the EU. The Dutch government’s intention to renegotiate and “modernise” each of these 79 BITs may have important implications for companies structured through the Netherlands that have subsidiaries or other investments outside the EU.

Summary

BITs are treaties that seek to protect investments made by companies incorporated in one treaty State (the ‘home State’) into the territory of the other treaty State (the ‘host State’). These treaties require that investments be accorded a certain minimum level of treatment (e.g., investments are not to be expropriated without compensation and are to be treated fairly and equitably), failing which the investor may generally bring an international arbitration claim directly against the host State (rather than having to go through that State’s local courts).

Dutch BITs are the second most frequently-invoked BITs globally, according to statistics by UNCTAD, suggesting that they fulfil a key function in facilitating international investment flows globally. Dutch BITs traditionally offer strong protections for investors, and are generally considered to be the ‘gold standard’ as far as investment protection is concerned. Following criticism from certain NGOs and in public opinion that BITs are too heavily skewed in favour of investors, the new draft Model BIT made available yesterday departs from that tradition to a certain extent, aiming to “create more balance between the rights and duties of host States and investors”.

The new draft Model introduces stricter eligibility requirements for investors to qualify for protection; replaces open norms for treatment of investments with more elaborate definitions that are in some ways more restrictive; and imposes time limitations on investors’ ability to bring arbitration claims against a host State. At the same time, the new draft Model addresses legitimate investor concerns about the efficiency of arbitration proceedings, and introduces other innovative features that have the potential of becoming the new standard going forward.

Timing and implications

For companies structured through the Netherlands that have subsidiaries or other investments outside the EU, the new Model BIT could in the medium to long term have important implications; the most far-reaching of which is that they could no longer be eligible for protection. The terms of the new Model are intended to replace those in all 79 BITs with States outside the EU that the Dutch government plans to renegotiate.

Re-negotiation of existing Dutch BITs will not occur overnight, and is subject to authorisation by the European Commission. It could, however, lead to amended BITs based on the new Model within the next one to two years.

Dutch BITs currently in force generally provide that their terms will continue to be in effect for existing investments for a certain period (typically 10 to 15 years) following termination. In the event of renegotiation, however, a much shorter transitional regime (or no regime at all) is likely to apply.

If the new Model terms are adopted following renegotiation, implications include:

  • A subsidiary or investment made outside the EU will no longer be eligible for protection unless it is held by a company in the Netherlands that either has “substantial business activities” in the Netherlands or is owned or controlled by another Dutch company that meets the substance requirement (Articles 1(b)(ii) and (iii)). The purpose of this provision is to exclude “mailbox companies”. However, the new draft Model does not provide clear criteria, which is needed to give businesses structured through the Netherlands the legal certainty that they are either protected or understand what they need to do to meet the substance requirement.
  • An investment will no longer be eligible for protection unless it has certain characteristics, including the commitment of capital or other resources and the assumption of risk (Article 1(a)). While this will generally not affect traditional forms of investment (such as shares in a subsidiary), it could exclude more exotic forms that currently qualify (such as a pending claim in a local court). Claims to money arising from commercial contracts for the sale of goods or services are excluded from protection (Article 1(a)). Specific limitations apply to the restructuring of public debt (as per a separate protocol).
  • Legitimate expectations by an investor that the regulatory regime in place when the investment was made will not be fundamentally amended, appear no longer protected (Articles 9(2) and (4)). This could mean, for instance, that if the host State terminates or overhauls a feed-in-tariff regime that was designed to attract investment and was the basis for an investment in a renewable energy facility, the host State has no duty to compensate.
  • An indirect expropriation (a governmental measure that has an equivalent effect of an outright nationalisation) will require that “fundamental attributes of property” be taken to be eligible for compensation. This could mean that if, in the example of the feed-in-tariff above, the value of the investment is reduced to zero as a consequence of the measure, yet no “fundamental attributes of property” are taken, the host State has no duty to compensate (Article 12(3)).
  • A number of prescription periods will apply. These include that a claim must be commenced within two years after proceedings before a local court were exhausted or ceased (Article 18(4)(b)) or within five years of the date on which the investor first acquired or should have first acquired knowledge of the infringing measure and of the loss incurred (Article 18(4)(a)).
  • The investor and the host State may no longer each appoint an arbitrator to the arbitral tribunal that will hear a claim under the BIT. All members of the tribunal are appointed by the Permanent Court of Arbitration (PCA) in The Hague or by the International Centre for the Settlement of Investment Disputes (ICSID) in Washington DC (Article 20(1)). However, these institutions are required to “thoroughly” consult with the disputing parties before making the appointment (Article 20(2)), suggesting the use of a ‘list’ procedure (under which parties may veto proposed candidate arbitrators and indicate preference for others) as is currently the practice under the UNCITRAL Arbitration Rules in certain cases. The provisions on arbitration are replaced when a treaty providing for a multilateral investment court comes into force (Article 15).
  • In deciding on the amount of the compensation, tribunals may take into account any non-compliance with the investor of its commitments under the UN Guiding Principles on Businesses and Human Rights as well as the OECD Guidelines for Multinational Enterprises (Article 23).

In a number of respects, the new draft Model improves the position of investors, particularly by facilitating the expeditious conduct of arbitral proceedings. It includes a provision that the arbitral tribunal should generally render a final award within 24 months (Article 22(1)); that the arbitration is in principle not bifurcated between a merits and a jurisdictional phase (Article 20(9)); and that common claims may be consolidated into one arbitration which facilitates small and medium sized businesses (Article 19(7)). Additionally, host States are required to adhere to explicit “rule of law” standards, such as that they are to guarantee principles of consistency, impartiality and transparency in their administrative behaviour, and they must ensure that investors have access to proper mechanisms of dispute resolution (Article 5), albeit that these provisions cannot be enforced in arbitration proceedings.

Policy choices

The new draft Model is a significant overhaul compared to earlier Model BITs (most recently of 2004). In certain areas, it reduces the extent and depth of protection available for investors, while at the same time recognising investor interests in other areas. The policy choices behind the overhaul are the consequence of three forces operating in parallel.

First, the perception of certain NGOs and in public opinion that BITs generally, and Dutch BITs particularly, are skewed in favour of investors to the detriment of host States. While BITs are voluntarily concluded by host States, the argument has been made that open standards in BITs have been too broadly interpreted by arbitrators, beyond what host States would have voluntarily agreed to. Although this argument finds little empirical support (apart from limited exceptions, host States have in large part not terminated or sought to amend BITs following adverse arbitration decisions), this perception has gained some traction with the Dutch government. In the accompanying notes to the draft new Model, it is observed that “the Dutch government acknowledges certain concerns regarding previously concluded investment treaties”.

Second, since the Treaty of Lisbon, the European Commission in principle has the exclusive competence to conclude investment treaties with States outside the EU. Under Regulation No. 1219/2012, EU Member States are nevertheless permitted to renegotiate or conclude such treaties, but only with the authorisation of the Commission. The influence that the Commission has over Member States’ BITs is felt in the draft Model, which copies provisions from treaties negotiated by the Commission (such as CETA) in an apparent effort to ensure that the requisite authorisation can be obtained.

Third, the new Model is a reflection of global shifts in economic power, which have turned States such as the Netherlands which are traditionally capital-exporting home States partly into capital-importing host States. Accordingly, where the Dutch government has traditionally primarily had an eye on the interests of Dutch investors investing abroad, it is now taking into account that it is equally a host State for investors from the State with which the BIT is concluded, and may be the subject of claims from that State’s investors. This is reflected in the accompanying notes which state that “the Dutch government has strived for a more equal balance between incoming and outgoing investments … and the offensive and defensive interests of the Netherlands”.

Reflection

While these considerations are understandable to a large degree, in our view they do not justify certain amendments and omissions in the new draft Model BIT. We take each of the three points above in turn, in reverse order.

First, in the relationships with most States, the Netherlands remains a far more significant exporter of capital than an importer. This is particularly the case in relation to the States that would be the subject of the renegotiated BITs, which exclude major economic powers such as the US, China, Japan, Canada and Singapore, since treaties with these States are subject to (ongoing or failed) negotiations by the European Commission. In relationships with most countries in Africa, Latin America and certain parts of Asia, the Dutch government should not let attention for the legitimate interests of Dutch investors be restrained by the unlikely prospect of claims by investors from those States.

Second, the Dutch government’s desire to facilitate approval from the European Commission for the renegotiated BITs is understandable. Yet copying provisions from EU treaties designed to protect investors in well-developed states like Canada and Singapore (as the model does in various places), will not offer Dutch investors the protections they need in Africa, Latin America and certain parts of Asia. Protection against government measures that deprive an investor of the full value of the investment should be re-introduced in a well-defined manner. The “rule of law” standards that are new (Article 5) should be made enforceable by investors to give them actual meaning and effect.

Third, in our view a number of concerns raised by NGOs and in the public opinion in relation to BITs are justified. These include the lack of transparency of arbitral proceedings under BITs, a need for safeguards to ensure that arbitrators are completely independent and impartial from the disputing parties, and a further definition of open norms for greater legal certainty. The draft Model BIT rightly addresses these issues. We also agree that the use of “mailbox companies” to gain access under Dutch BITs is, as a policy matter, not a practice that should be encouraged when concluding new treaties in current times. However, the substance requirement in the draft Model (which is necessary in order to exclude mailbox companies from having access) introduces significant legal uncertainty in the way it is currently phrased. The draft Model should provide clear criteria, so that investors have certainty that they either qualify under the substance requirement, or know what to do if they wish to qualify.

Finally, the draft Model BIT should go further in addressing a number of additional issues under the current BIT practice. For instance, a limited number of States ordered to pay compensation under BITs have managed to avoid payment for periods of time. Rules of sovereign immunity, which in themselves are legitimate, may in practice make it difficult to collect. The absence of actual recourse following a dispute raises the costs of investing, and may deter investments from being made despite the availability of a BIT. We propose that the draft Model BIT contain a waiver of immunity of enforcement (with an explicit carve-out for diplomatic assets) to allow actual enforcement once an independent tribunal has held, on the basis of clear standards, that compensation is owed to an aggrieved investor under the BIT. The UN Convention on Jurisdictional Immunities expressly provides for the possibility to have such waivers (Article 19(a) of the Convention), and waivers are regularly seen in contracts concluded by States. Both the PCA and ICSID offer model clauses. A provision to this effect would greatly aid the effectiveness of Dutch BITs. It would also provide Dutch BITs with an important distinguishing feature that is capable of facilitating the leading role that Dutch BITs have so successfully played in the facilitation of global investment flows over the past decades.

Businesses and other interested parties can submit comments to the Dutch government’s consultation process until 18 June 2018.