The Global Compact for Migration: to sign or not to sign?

Written by

The Global Compact for Safe, Orderly and Regular Migration (final draft of 13 July 2018) is scheduled for adoption at an intergovernmental conference in Marrakesh in December 2018. But in the run-up to this conference, several states, beginning with the United States already in 2017, now followed by Austria, Hungary, the Czech Republic, and others, have announced that they will  not sign the text. Will refusal to sign be relevant in terms of international law? What is the juridical quality of the Compact, which legal consequences does it have, and which normative “ripples” might it deploy in the future? The controversy over the Compact sheds light on the legitimacy of international law-making processes and on the precarious normative power of international law.

A Brief Glance at the Contents

The Compact consists of four parts. Following the preamble, the first part contains, “Vision and Guiding Principles”. The second part, “Objectives and Commitments” contains 23 objectives, proceeded by a part on “Implementation” and the final section “Follow-up and Review”. The Compact purports to set out “a common understanding, shared responsibilities and unity of purpose regarding migration” (para. 9). The purpose is mainly to secure that migration “works for all” (para. 13).

The Compact’s “guiding principles” are, inter alia, people-centeredness, international cooperation, national sovereignty, rule of law and due process, and sustainable development (para. 15). These are well-established and to a large extent also legally entrenched principles. The 23 “objectives” are partly generally recognised such as saving lives (objective 8), respond to smuggling (objective 9), or eradicate trafficking (objective 10). Some mainly correspond to interests of states of origin (such as promoting transfer of remittances, objective 20), others basically satisfy interests of receiving states (such as facilitating return and readmission (objective 21). In substance, the Compact partly repeats international law as it stands or refers to existing instruments (see notably preamble para. 2), partly contains platitudes, and partly contains novel ideas.

A quite strong statement is para. 21: “We commit to adapt options and pathways for regular migration”. This sentence has been chastised by some observers. It has also been criticized that the Compact views migration predominantly under the economic aspect, bracketing resulting cultural problems. Finally, the responsibilities of the states of origin to improve living conditions so as to forestall the desire or need for migration are not mentioned in the Compact; neither is overpopulation.

Recent Withdrawals

The gist of the statements of the withdrawing states is basically that the Compact would force states to admit migrants, would be a pull-factor for migration, would contravene domestic migration policies, and violate the states’ sovereignty. (See for the US: United States Mission to the United Nations, United States Ends Participation in Global Compact on Migration, Press release of 2 December 2017; for Austria: Innenministerium, Erklärung zum Ausstieg aus dem UN-Migrationspakt, Press release of 31 October 2018; for Hungary: Ministry of Foreign Affairs and Trade, The UN Global Compact for migration is endangering the security of the Hungarian people, Press release of 6 November 2018.)

These excuses for standing aside seem pre-textual. The Compact does not contain any language obliging participating states to admit migrants. Quite to the contrary, one of the Compact’s objectives is “to cooperate in facilitating a safe and dignified return and readmission” of migrants. The states of origin “commit to ensure that our nationals are duly received and readmitted” (objective 21, para. 37).

The Migration Compact is no International Treaty

Most importantly, the Compact cannot dictate any migration policy, because it is legally non-binding. Because this point has been hotly debated in parliaments and in the media, it needs some explanation.

The bindingness depends on the intentions of the participating states. Whether they intended to bind themselves legally can be determined by looking at indications ranging from the form over procedures to substantive content. Almost everything points towards a legally non-binding text.

The title of the document is basically inconclusive. There are binding treaties which are called “compact” or “pact”, for example the “Pact of San José” as a nickname for the Inter-American Convention on Human Rights. But more frequently, the title Compact has been chosen for non-binding texts. The best-known is the UN Global Compact on transnational corporations which is a platform for transnational corporations for adopting human rights policies, combating corruption and so on. The ILO also adopted a “Jobs Pact” in the context of the world financial crisis in 2008 which was non-binding and whose non-legal quality was not controversial.

Second, the Migration Compact has been elaborated in a typical conference proceeding on the basis of a series of General Assembly resolutions. These resolutions set out the procedural parameters. They could be used for producing a binding treaty or a non-binding declaration. The Compact will be adopted at a specific intergovernmental conference, and not simply in the course of the General Assembly in New York. The text will thereby enjoy a much higher prominence than ordinary General Assembly resolutions. Still, the process does not give a decisive clue on the legal status of the resulting text.

The only factor that would allow for an understanding as a legal commitment is the repetition of the formula “We commit to ….”. However, all factors point against legal commitments. The most important factor is the text itself. Paras 7 and 15 explicitly say that “[t]his Global Compact presents a non-legally binding, cooperative framework”. The Compact also explicitly “reaffirms the sovereign right of states to determine their national migration policy and their prerogative to govern migration within the jurisdiction” (para. 15).

Moreover, the text does not contain short and clear obligations. It rather resembles key United Nations texts such as the Millennium Declaration of 2000 or the Agenda 2030 of 2015. This fact is however not decisive, because new international legally binding treaties such as the Paris Agreement also contain lengthy and verbose paragraphs with weak commitments, are full of hortatory statements and almost completely lack operational provisions.

A final indication is the negotiating history. The EU always made clear that it did not want a binding document. And further ingredients of an international treaty are also missing. There is no provision on ratification or implementation in domestic law. All this speaks against a binding treaty.

The text is what is usually dubbed “soft law”, in the grey zone between law and non-law, between law and politics. (This qualification of course makes sense only on the premise that it is conceptually not possible and normatively not desirable to draw a bright line between law and non-law. This is indeed my premise, mainly motivated by the insight that a strictly dichotomous view of law versus non-law pushes out of the picture many interesting and important normative phenomena and would make it more difficult to understand what is really going on in global governance). This means that on the one hand, the Compact will not generate legally binding obligations but that it is on the other hand, not legally irrelevant.

Legal and other Functions of the Migration Compact

The legal functions of the Migration Compact might be gathered under the labels “pre-law”-functions, “para-law”-functions, and “law-plus”-functions. A text such as the Compact can be “pre-law”, a forerunner of hard law, paving the way for a formal treaty. It can also give rise to the crystallisation of customary international law. This would of course need an opinio iuris and concomitant practice over some time. When signing states − as here − explicitly say that the commitments are not legally binding, it is difficult to deduce any legal opinion from this.

The situation is therefore quite different from the Universal Declaration of Human Rights of 1948, some of whose rights meanwhile have the force of customary international law.

Another example is the United Nations Declaration on Indigenous Peoples 2007, which is not legally binding. Australia, a country with a history of discrimination and extermination of its indigenous population, voted against this declaration in the United Nations General Assembly, but two years later (in 2009) decided to adopt the text. A process of customisation of some of the provisions of that declaration is ongoing. But in contrast to the Migration Compact, the Indigenous Peoples’ Declaration looks much more like a treaty with short and clearer provisions.

A soft law text can also substitute missing hard law (“para-law”). And indeed, to some extent the Migration Compact is a workaround for the very low status of ratification of the International Convention on the Protection of the Rights of All Migrant Workers and Members of their Families of 1990 which has been ratified so far only by 34 states and among them no European target states of migration.

Finally, soft law texts such as the Migration Compact can serve as a guideline for the interpretation of hard law, can flesh out hard law commitments and make them more concrete (“law-plus function”). For example, the Compact might add some detail to obligations flowing from conventions on nationality (cf. objective 4), and from the UN Convention on Transnational Organized Crime (cf. objective 10). The Compact will also be drawn on by domestic authorities and courts, for the interpretation of national law. And it may factor in as a parameter for the the exercise of discretion by national administrative agencies. So overall, signing the Migration Compact will not be irrelevant in legal terms.

At first glance, a soft law document seems to be a second-best option in comparison to a real and binding treaty. However, the choice of such an instrument has many strategical advantages. The lack of a formal legal quality has the big advantage of flexibility. The elaboration and adoption of such texts is normally easier and quicker. And it is easier to scrap them, too. A recent example is the US-American denunciation of the so-called Iran Nuclear Deal in May 2018. The “Joint Comprehensive Plan of Action (JCPOA)” had been signed in 2015 by the ”E3/EU+3” (China, France, Germany, the Russian Federation, the United Kingdom and the United States, with the High Representative of the European Union for Foreign Affairs and Security Policy) and the Islamic Republic of Iran. As a non-binding, non-legal arrangement, the US Administration could easily walk out in 2018. (One legal issue here that the JCPOA had been “endorsed” by the Security Council in SC Res 2231 (2015) of 20 July 2015, para. 1. but this was not enough to transform the non-binding deal into formally binding law.)

Potential Effectiveness of the Migration Compact

In the sphere of international relations, the difference between formally binding texts and non-binding texts seems smaller than in domestic law. The reason is that even “hard” law can only rarely be enforced by courts (international or domestic ones).

Famous more or less well-functioning institutions did not and do not have a formal international legal status. For example, the GATT 1947 was a treaty that had never entered into force but which was nevertheless “provisionally” applied until 1994. The OSCE, despite the change of the name “Conference on Security and Cooperation in Europe” into “Organisation” by the Budapest decision of 1994 (point 29) is not a formal international organisation. It does not have a founding treaty and does not enjoy international legal personality. Nevertheless, the OSCE plays an important role in security matters in Europe, and it is exactly its flexibility which is seen as an advantage by many participants, notably by the United States.

Many international treaties possess only soft documentation mechanisms such as reporting, discussion groups, and so on. The follow-up and review mechanism of the Migration Compact is not so far away from those. The Compact’s part on implementation mentions three strategies or instruments: a capacity-building mechanism and “network” (para. 45) and biannual reporting of the Secretary-General (para. 46). The follow-up and review consists in an “International Migration Review Forum” (para. 49) which shall meet every four years starting in 2022, and which shall culminate in a “Progress declaration” (para. 49 lit e). Some guide-post about organising these review fora are given in para. 54. The Compact emphasises that review is a “state-led” process but with the inclusion of “all relevant stakeholders” (para. 49).

So is it a good or a bad sign for international law that some governments now shy away from signing the Compact and thus apparently take even soft law seriously? This question is hard to answer. If governments were acting in good faith, their scrupulousness could be taken as a gesture of principled willingness to comply with the “commitments” embodied in the Compact. But another explanation for the sudden reluctance is simply that they seized an opportunity to appeal to partners and publics in their countries who are hostile towards migrants. Therefore, general conclusions about the strategic value of such texts cannot be drawn.

Ultimately, the Compact might in reality become as effective or ineffective as an international treaty, might be complied with and lead to the desired changes of action. So the difference between a “soft” compact and a hard treaty lies not so much in its lower effectiveness but rather seems to lie in a different basis of legitimacy.


The executive branches like soft law not only for acting swiftly and flexibly, but also for keeping out national parliaments.

Does the Migration Compact enjoy a lower degree of democratic legitimacy than a formal treaty? A famous case illustrating this question is the so-called new Strategic Concept of NATO of 1999. Here, NATO member states had adopted a text which foresaw an extension of treaty operations both in geographical scope and in substance. Upon the complaint of a political faction in the German parliament (the Bundestag), the German Constitutional Court (BVerfGE 104, 151, judgment of 22 Nov. 2001 – 2 BvE 6/99) had to determine whether the new strategy was a formally binding treaty (an amendment of the NATO treaty) or not. In this case, the Constitutional Court focused on the will of the parties to bind themselves, to be gathered from indications as mentioned above. It concluded that the new strategy was not a binding treaty. Therefore, the German constitution was not violated by not involving the Bundestag (had it been a formally binding treaty, then the Strategic Concept would have had to be domestically ratified by adopting a formal parliamentary statute).

If the Migration Compact were a formal international treaty, voting in parliament would be needed in some form or another in probably all democratic states in order to receive the treaty in the domestic legal order and make it binding law.

Given the non-legally binding nature of the Compact, formal parliamentary voting on it would be an anomaly. It might however happen. In Switzerland, three parliamentary commissions have requested the government to formally submit the Migration Compact for parliamentary approval, as opposed to just signing it (motions of both chambers’ committees on constitutional politics (motion 18.4093 SPK-NR of 19th Oct. 2018 and motion 18.4103 SPK-SR of 8 Nov. 2018) and of the upper house’s foreign policy committee (motion 18.4106 APK-SR of 12 Nov. 2018)). In other words, the parliamentarians wish to treat the Compact like a formal treaty. This would go beyond the extant legal obligation to consult the Swiss parliament. This obligation to consult explicitly encompasses the consultation on soft law texts (see Notiz des Sekretariats der Aussenpolitischen Kommissionen, ‘Mitwirkungsrechte der eidgenössischen Räte im Bereich Aussenpolitik’, p. 2; relating to Art. 152(3) of the Swiss Act on Parliament). The Swiss debate about the proper parliamentary involvement with regard to the Migration Compact is ongoing at the time of writing.

In other states, too, intense parliamentary debates took place about the Compact Can such open discussions on the Compact properly be seen as an exercise of deliberative democracy? Or are these not sufficient to confer some modicum of democratic legitimacy on the text? At this point it is worth recalling that the hearings organised by the United Nations were open to parliamentarians, and that these forms of transparency and participation might to some extent mitigate the absence of formal democratic voting in national parliaments. In various rounds, civil society organisations were heard, and in total seven regional hearings were conducted. But as far as I can see, no explicitly migration-hostile groups were present. These hearings are in some way closer to the ideal of a global democracy than votes in national parliaments, because they are not confined to specific national publics but have a global scope. However, as long as they are – due to weak media attention, time constraints, and language barriers – in reality far removed from concerned citizens, they can fulfill this democratic function only weakly.

The more or less sudden refusal by some governments to sign the Compact (besides being motivated by the domestic political situation and desire to satisfy populist and xenophobic sentiments of populations by showcasing restrictive immigration policies) could indicate that the drafting process was not so transparent after all.

The expectations in this regard have much changed in the last decades. While it used to be a diplomatic prerogative to conduct secret negotiations, an actual transparency turn has taken place. The current trend is to involve the public more in the processes of elaborating international law − both hard and soft. For example, the adoption of the Paris Agreement was fully under public scrutiny. Transparency politics need to strike a delicate balance. Too early or too much transparency (participation of the public and participation of civil society organisations) can bring negotiations to fail. But too little transparency can also lead to non-acceptance − which might be the case with the Migration Compact. Remember the Transatlantic Trade and Investment Partnership (TTIP) which was negotiated behind closed doors between the United States and the European Union. When the text was leaked in 2016, a public uproar was heard in Europe. The stalling of the TTIP-negotiations was however less due to misapprehensions of the public than a decision of the new US presidential administration.

Whatever short term effects the current public debate on the Migration Compact will have for this text itself, it will not terminate the global efforts to regulate migration, because the need for global regulation ─ preceded and prepared by open and trans-national deliberations ─ will not go away.

Print Friendly, PDF & Email

Leave a Comment

Comments for this post are closed


Markus Krajewski says

November 21, 2018

Many thanks, Anne, for this insightful post and contribution to a rational debate on such an important topic. Just one point: While the number of parties to the Migrant Workers' Rights Convention is regrettably low, it is not as low as stated in the post: The ICMW has 54 state parties, see

Erasmo Lara Cabrera says

November 22, 2018

A very insightful post, most timely. A couple of comments:

1. The reluctance of some governments to join the Compact is, as you mentioned, mainly due to political reasons and national debates. At the negotiations, Hungary brought her Foreign Minister (and local TV). He made colorful statements, coincidentally in the context of local national elections. The legal character of the document, I believe, is being used for those purposes and not necessarily because there is a real concern as per its normative value.

2. As per soft-law-making, the process was indeed extremely transparent, at the negotiations, with stakeholder sessions, regional consultations and so on. I believe that the argument of lack of transparency and clarity as to the process and aims of the document would not be a substantiated one. Also at the negotiations countries expressed the view that they would bring the outcome document to their constituents and no complaints were raised as to a lack of transparency.

3. The document will be adopted at an Intergovernmental Conference in December. Some say they will not join (a clear minority, even within Europe). Would it not be better for such countries to go to the conference, call for a vote, and clearly register their opposition, should the legal nature of the Compact be their main concern?

Indeed it has been a very rich and interesting process and your post is quite appreciated. I liked the reference to the "para-law" role of the Compact as per the UN Migrant Workers Convention (would it become a more "lively" document after the Compact is adopted?).

Vielen Dank!

John Morss says

November 23, 2018

Thankyou for this detailed analysis.
It does seem slightly disconcerting after several centuries of treaty-making to be discussing what is and what is not a treaty.
.. (and whether treaties make 'CIL'...)
On the point about Australia and a history of extermination of indigenous people, which seems a little parenthetical given the gravity of the issue, I met recently with ex-foreign minister of Oz, Alexander Downer (we met, I hasten to add, on an escalator in the London Underground... was this a summit or its opposite?! No communique was issued) which prompts me to point out that extermination was surely a British colonial policy/practice/
negligence (eg compare German imperial policy with Herero, etc) -- not 'Australian' (unless you refer to the Stolen Generation?). (By the way I am both British and Australian so I share the guilt either way. And to 'discrimination',
no defence.)

Oskar says

November 25, 2018

Very useful. Thanks. I am from Slovakia and our MOFA was chairman of the UN GA. He faces serious backlash from some political parties here as a result so he may not even be allowed to travel to Marrakech..why an irony.

anne peters says

November 30, 2018

The German parliament (Bundestag) has on 29 Nov voted positively on the Migration Compact (372 in favour, 153 against, 141 abstensions). This gives the compact a sufficient democratic basis in Germany. (Had it been a treaty, a formal parliamentary statute would be needed).

Actually, the compact will not be "signed" but most likely adopted by consensus so the post's title was misleading ... I thank Ralph Janik for pointing this out.
See his post in German:

Dear colleagues,
thank you for lucid comments and for pointing out error and typo (ratification status of the MWC - thank you markus).


Ron Max says

December 3, 2018

Thanks for this detailed elaboration. National democracy till date is a simulation of democracy and in this sense global democracy is a façade. There are several international instruments / conventions already in place, so why was there a necessity to draft this pact? Given the fact that the pact lacked awareness creation via media or otherwise, it is a deliberate innovative attempt to murder democracy. Whether the pact will or will not mitigate efforts to regulate migration again depends on the individual country consciousness, in this sense, pact is a notion or rather a weapon providing future scope for nations to exploit its interpretations as per their will especially in countries where International Law is directly binding.