Would a regional market in tradable refugees1 quotas (TRQs) solve the inefficiency of the asylum system in the European Union (EU)? I argue that a market in refugee quotas should not exist in the EU insofar as it does not increase protection efficiency relative to the Dublin System. I measure protection efficiency in terms of the maximisation of the number of refugees who receive protection from persecution (Bauböck 2018). This is to be contrasted with an allocative efficiency, or the maximisation of allocations of responsibility of refugees to member states. In order to defend this thesis, I firstly present Moraga and Rapoport’s TRQs proposal (2015), a response to what they consider the flaws of the Dublin System. I then argue that because TRQs change countries’ incentives and do not deal with the striking inequalities of treatment of refugees across the EU, they ultimately do not produce an increase in protection efficiency. After evaluating one objection to my argument, I offer an alternative to TRQs, which shows a stronger commitment to the protection of refugees while avoiding some of the issues caused by the TRQ system.

Moraga and Rapoport’s (2015) TRQs proposal is motivated by the failures of the Dublin System. First, the authors argue that by assigning the responsibility for asylum-seekers to their country of first entry, the Dublin System puts an unfair burden on some countries more geographically accessible to refugees or disproportionately preferred by them. For instance, 87% of refugees in the EU live in Germany, France, Greece, Austria and Italy only (Moraga and Rapoport 2015). Secondly, the first-entry rule prevents refugees from choosing their country of asylum and thus fails to take into account refugees’ preferences. The authors then argue that the Dublin System is unable to deal with the striking inequality of refugee treatment across the EU, operates in a costly and inefficient manner, and does not deal with the integration prospects of asylum seekers.

In order to deal with these issues, the authors propose to introduce TRQs according to which each member state of the EU is assigned a quota for refugees. This quota can then be traded in a market in which countries in excess of refugees receive a price for accepting one additional refugee while countries that accept fewer refugees than their assigned quota have to pay a price. Furthermore, this market is sensitive to countries’ preferences over refugees and to refugees’ preferences over countries thanks to a matching mechanism (Moraga and Rapoport 2015). This distributive system firstly allows countries to state their preferences over refugees. Refugees then accept their favourite offer amongst member states willing to take them in and reject any offer from countries that they refuse to go to. Countries which have been rejected then make new offers to those refugees who have not rejected them while refugees with multiple offers can chose their preferred one. The matching mechanism continues until member states no longer have any visas to offer. This is supposed to make hosting refugees more attractive for member states that can now express their preferences over refugees’ characteristics, such as language or skills. The matching system also takes into account refugees’ preferences over countries thereby allowing them to participate in the very decision-making process shaping their lives. Hence, the TRQs proposal challenges some of the Dublin System’s flaws insofar as it takes into account refugees’ agencies and has the potential to operate in a less costly manner2.

However, TRQs fail to necessarily create an efficient market whereby the protection from persecution of refugees is increased compared to the Dublin System. This is because it is possible that under TRQs some countries still receive a disproportionate number of refugees compared to other member states. Indeed, wealthier member states in which hosting refugees is politically unpopular will have more autonomy to pay penalties while transferring responsibility for refugees to other countries. These member states might in turn see a financial incentive in taking refugees in, without any real commitment to the provision of protection from persecution. This is because the market format of TRQs can lead member states to think about their obligations towards refugees either as a revenue source or as a burden that can be gotten rid of through a monetary exchange (Sandel 2013), shifting incentives away from the obligation to provide protection.

I believe this claim to be plausible, in light of the already prevalent political concerns over the number of asylum seekers which have led many member states to constrain or reverse movements of mobility rather than to ensure protection in the face of lost rights (Long and Crisp 2010). This is demonstrated clearly by the several measures of deterrence or non-entrée undertaken in the EU in the past two decades (Gammeltoft-Hansen and Hathaway 2014). Deterrence includes increasing border control, the construction of fences but also more infamous measures such as the ending of the Operation Mare Nostrum in October 2014. Despite having saved over 100,000 lives from drowning over the course of a year, this search-and-rescue mission near the cost of Libya was cancelled by Italy on the grounds that it encouraged more migrants (amongst them refugees) to risk their lives attempting to cross the sea (Parekh 2020). These deterrence policies in the Mediterranean Sea have been shown to have increased the number of deaths at the border of the EU (Tinti and Reitano 2016). Hence, many member states have demonstrated a reluctance to let refugees in, a desire that can be facilitated by the TRQs system.

Thus, TRQs might create an incentive for wealthier countries to simply buy their way out of the TRQ system and a financial incentive for poorer countries to host refugees. This in turn can lead to an inefficient system that does not lead to an increase of protection from persecution compared to the Dublin System. If refugees are disproportionately allocated to some financially motivated countries, they risk being exposed to countries that show a lesser commitment to the protection of refugees and that have poor humanitarian standards. This risk arises because once a state has bought out of their legal obligation, they have no incentive to ensure that other countries uphold their humanitarian commitment. Furthermore, this exposure can be explained by the fact that the TRQs do not deal with the inequality of refugee treatment across the EU. Indeed, the TRQs mechanisms for allocation of responsibility of refugees seem to ‘rely on further legal fictions that [member states] form a uniform expanded area of protection, such that transfers of asylum seekers may be assumed to be safe’ (Costello 2015: 233).

However, this is not the case, as refugees themselves testify. For instance, Bulgaria often detains refugees asking for asylum, in violation of international law (Parekh 2020). Even more strikingly, the very rate of recognition of asylum claims varies across countries with an average for Syrian refugees of 95% in Europe, contrasting with 50% in Estonia and 43% in Slovakia (European Stability Initiative 2015), discrediting claims of standardised treatment. This inequality in treatment across member states was actually recognised by the European Court of Human Rights, which ruled in 2011 that a refugee in Belgium could not be sent to Greece because it would account as a breach of the prohibition of ill treatment (McDonald-Gibson 2016). Hence, by assuming a homogenous treatment of refugees across the EU and by financially incentivising some countries and not others to take in refugees, TRQs fail to maximise refugees’ protection from persecution compared to the Dublin System.

Yet, one might argue that TRQs will always provide better protection from persecution than the Dublin System because the market will ensure that every refugee is allocated to a country thanks to the matching mechanism. Prima facie, the matching mechanism would incentivise countries to take in refugees according to their own preferences and it would allow refugees to refuse to go to some countries with poor humanitarian records, thereby increasing protection from persecution. However, this line of argument assumes that refugees have a genuine choice to refuse a country’s offer when the alternatives (refugee camps, return to home country) are as dire, if not more, than the prospect of going to a country with low humanitarian standards. Secondly, taking into account countries’ preferences over refugees might lead to an increase of protection of some refugees but not others if member states’ preferences are biased towards one kind of refugee, such as educated males without children. This claim remains to be empirically tested beyond the small data on member states’ preferences over refugees provided by the European Relocation from Malta program (Moraga and Rapoport 2015). 

Finally, note that an increase in allocative efficiency is not tantamount to an increase in protection efficiency. Indeed, persecution often remains for refugees, even after a member state has taken responsibility over them. For instance, refugees in France report being on the receiving end of cruel and brutal treatment by the police (Fine 2019). A Syrian refugee in Bulgaria declared having to sleep on the street while waiting for his asylum application to be processed without receiving any information about the progress of said application (Parekh 2020). Similarly, the chief of the Hellenic Police declared that the police had to make asylum seekers’ lives miserable in order to show them that they were not free to do as they wished in Greece (Fili 2016). Thus, even after being put under the responsibility of a member state, refugees still face unfair or cruel treatment over long periods of time. If TRQs increase allocative efficiency, they still do not tackle refugees’ inability to access their human rights (Long and Crisp 2010). TRQs do not challenge the lack of efficiency in protection from persecution and this phenomenon is only further emphasised by the shift of incentives of member states introduced by the market nature of TRQs. Hence, there should not be a market in TRQs in the EU.

Instead, I propose that the EU introduces fixed quotas based on countries’ specific characteristics, combined with the matching system proposed by Moraga and Rapoport (2015). Furthermore, if no satisfactory outcome is achieved through this mechanism (because refugees’ biases are oriented disproportionately towards a country or vice versa), the allocation of responsibility should then adhere to the five criteria delineated by the New Pact on Immigration (European Commission 2020), to be considered in the following order:

(1) The child’s best interest.
(2) Presence of a family member who has been granted international protection or is an applicant in a member state.
(3) Member State that has issued a residence document or a visa.
(4) Member State that has issued a diploma or qualification.
(5) Member State of irregular entry.

However, my proposal is contingent on the EU’s commitment to the upmost uniform treatment of refugees, one directed towards the provision of protection from persecution and of dignified means of subsistence. Furthermore, the EU needs to develop serious sanction mechanisms, both for member states which do not uphold these humanitarian standards and for member states which do not respect their fixed quotas. Proper sanctions, fixed quotas, and a matching mechanism could then create an incentive for member states to become attractive countries in a bid to appeal to their preferred refugees. This proposal would increase the efficiency of the system by providing more protection from persecution than the Dublin System while avoiding exposing refugees to the poor humanitarian records of some financially motivated countries.

I have argued that there should not be a market in TRQs on the grounds that it does not maximise the provision of protection from persecution of refugees compared to the Dublin System. Indeed, TRQs can change the incentives of member states from an obligation to provide protection to seeing refugees as either a revenue opportunity or as monetarily tradable. Furthermore, the TRQs proposal does not deal with the striking inequalities in refugees’ treatment across the EU and therefore risks to actually expose some refugees to persecution in financially motivated states with poor humanitarian records. After exploring one objection to my argument, I offered an alternative to the Dublin System and TRQs: fixed quotas combined with a matching mechanism and better sanctions. I believe this would create a more efficient system in the EU: one committed to providing human rights to those who have lost them.


1 Refugees are defined following the 1951 UN Convention definition of refugees; see (United Nations High Commissioner for Refugee 1951)

2 The discussion of the cost efficiency of TRQs is beyond the scope of this paper, but see (Moraga and Rapoport, 2015) for more.


Bauböck, R. 2018. ‘Refugee Protection and Burden-Sharing in the European Union’. Journal of Common Market Studies 56(1): 141–156.

Costello, C. 2015. Defining Protection: Refugeehood, and Human Rights, The Human Rights of Migrants and Refugees in European Law. Oxford: Oxford University Press.

European Commission. 2020. ‘New Pact on Migration and Asylum’. URL: https://ec.europa.eu/info/sites/default/files/1_en_act_part1_v7_1.pdf

European Stability Initiative. 2015. 'The Refugee Crisis through Statistics'. URL: https://www.esiweb.org/pdf/ESI%20-%20Refugee%20Statistics%20Compilation%20-%2017%20Oct%202015.pdf

Fili, A. 2016. The Continuum of Detention in Greece. URL: https://www.law.ox.ac.uk/research-subject-groups/centre-criminology/centreborder-criminologies/blog/2016/05/continuum

Fine, S.2019. ‘II—Refugees, Safety, and a Decent Human Life’. Proceedings of the Aristotelian Society 119(1): 25–52.

Gammeltoft-Hansen, T. and Hathaway, J. C. 2014. ‘Non-Refoulement in a World of Cooperative Deterrence’. Columbia Journal of Transnational Law 53(2): 235–284.

Long, K. and Crisp, J. 2010. ‘Migration, mobility and solutions: an evolving perspective’. Forced Migration Review (35): 56–57.

McDonald-Gibson, C. 2016. Cast Away: True Stories of Survival from Europe’s Refugee Crisis. New York: New Press.

Moraga, J. F.-H. and Rapoport, H. 2015. ‘Tradable Refugee-admission Quotas and EU Asylum Policy’. CESifo Economic Studies 61(3/4): 638–672.

Parekh, S. 2020. No Refuge: Ethics and the Global Refugee Crisis. Oxford: Oxford University Press.

Sandel, M. J. 2013. ‘Market Reasoning as Moral Reasoning: Why Economists Should Re-engage with Political Philosophy’. The Journal of Economic Perspectives 27(4): 121–140.

Tinti, P. and Reitano, T. 2016. Migrant, refugee, smuggler, saviour. London: Hurst & Company.

United Nations High Commissioner for Refugees. 1951. 'Convention and Protocol Relating to the Status of Refugees'. URL: https://www.unhcr.org/protection/basic/3b66c2aa10/convention-protocol-relating-status-refugees.html .