It is certainly the case that the initial processing of asylum claims and the initial protection and support of refugees incurs economic costs to Northern states. However, I presented numerous macroeconomic studies that demonstrate that on balance and over different timescales, refugees’ economic contributions substantially outweigh initial costs (Hillier-Smith, 2024a, pp. 5–6). Rindermann (2025) now counters with three studies from Denmark and the Netherlands to conclude there is conflicting data (p. 19). However, Rindermann’s analysis, and the studies he cites (for example, van de Beek et al., 2025) appear to assume that net economic contributions are reducible to income tax minus welfare assistance alone (pp. 17–19). But, a more complete analysis requires also including demographic offsetting of ageing populations; activate participation in the economy increasing demand for goods and services and thereby jobs for citizens; business creation; filling shortages in high-demand sectors in the labour market such as healthcare and social care and in lower-qualification roles, which also allows existing citizens to specialise and increase their own earning power (Hillier-Smith, 2024a, pp. 5–6).9 Moreover, we can add that refugees uphold the very functioning of many Northern economies simply in virtue of the specific roles they perform within them. Refugees (among other migrants) are overrepresented among ‘key workers’ (Fasani and Mazza, 2026). These are persons, especially celebrated during the COVID-19 pandemic, whose roles in, inter alia, transport, agriculture, food supply, cleaning and sanitation, retail, and healthcare, were and remain essential to economic and social functioning. The claim then that refugees are an overall net economic cost to Northern states is almost certainly false, at least controversial, and certainly not established as a factor to be included within a statist case against refugee protection.

Further, to justify refusing protection on statist grounds, it must be shown that the priority-weighted net costs to citizens are not only (1) sufficiently substantial to outweigh the net costs to refugees of being denied protection, but also (2) not reasonably avoidable insofar as there are no alternative means to address them.10 Failure to demonstrate (2) would entail refusing protection and the resultant serious harms to refugees would be needless and avoidable.

Recall, for instance, Oshal’s concerns with potential ‘life-threatening’ economic costs of refugee protection to the domestic poor. This fails to sufficiently acknowledge the array of policy and fiscal tools available to states including reallocating funding priorities to expand public services; raising taxes on corporations or wealthy individuals; migration ‘impact funds’; expanding social housing (through empty property acquisition, state development, or quotas for private developments); boosting incomes of the domestic poor (through minimum wage increases, income support, or grants); increasing social welfare and provisions in kind; and/or simply providing free prescriptions to those with acute healthcare needs and low incomes as many developed states already do (Scotland, for instance, provides free prescriptions to all). Oshal’s identification of potential hardships to the domestic poor provides powerful reasons to adopt these or comparable (re)distributive policies and targeted interventions, not to refuse protection to refugees.

Oshal does acknowledge the in-principle possibility of policy alternatives but suggests the likelihood of state actors adopting them ‘appears bleak’ (4). However, it is illegitimate to treat state actors’ purported unwillingness to pursue available policies as a fixed constraint to then justify their choosing the policy option of refusing protection to refugees. We are assessing the ethics of refugee policy: what, among available alternatives, state actors ought to do. If unwillingness to choose a policy option was treated as a fixed feasibility constraint that thereby precluded that option, this would have the absurd implication that state actors could escape obligations to choose certain policies simply by not being willing to choose them, and hence, for example, that states would be justified in refusing protection to refugees because that is the policy option they would choose.11 This would risk a circularity and poor ethical analysis. Moreover, it is arbitrary to stipulate away readily available policy alternatives. State actors have the capacity to choose among available policy options, and further, to shape the political and policy landscapes they inhabit to increase the range of policy options available. It is especially arbitrary to stipulate away available policy options while endorsing others such as the forced refugee relocation schemes Oshal later proposes (p. 15), which would face far more robust practical and legal obstacles. Last, in practice, states do in fact adopt policy alternatives that simultaneously invest in local services and improve conditions for existing disadvantaged citizens as well as refugees.12 For one example, the European Union’s Cohesion’s Action for Refugees in Europe (CARE) package released billions for local infrastructure, education, healthcare, housing, and urban regeneration.13

What of security costs? Both Oshal and Rindermann are concerned with crime rates. Rindermann (2025) provides shocking details of violent offences (p. 20), and also includes an appendix (pp. 24–27) collating high-profile cases, newspaper reports, and personal anecdotes concerning terrorist or criminal incidents in Germany where asylum seekers, failed asylum seekers, refugees, or other migrants have committed or are alleged to have committed serious violent offences, including multiple homicides.

Although the veracity of some cases can be contested, some refugees, asylum seekers, failed asylum seekers, and migrants have committed serious violent offences. Violent crimes are, of course, seriously wrong, and must be prevented, with perpetrators subjected to criminal sanction. The question is whether such cases are sufficient to demonstrate the general claim that refugees impose net security costs to thereby justify refusing protection to all on statist grounds.

The fear that refugees pose a security threat is common and nothing new. Edmund Burke brandished a dagger during the parliamentary debate on the 1793 Aliens Act, dramatising the supposed danger of migrants. Belgian refugees in Britain were suspected as German spies during the First World War. Refugees fleeing Eastern Europe were feared to be communist sleeper agents during the Cold War. Even Jewish refugees fleeing Nazi persecution were seen as criminals and security threats (Arar and FitzGerald, 2022, pp. 191–92).

Yet, since refugees are fleeing and seeking protection from threats to their own security, they are understandably overwhelmingly averse to physical violence, and studies find no positive association between the location of refugees and outbreaks of civil violence, crime, or terrorism (Bollfrass et al., 2015; do Céu Pinto Arena, 2017; Amuedo-Dorantes et al., 2018; Forrester et al., 2019; Nowrasteh, 2023). The chance of being killed by an asylum seeker in the U.S. between 1975 and 2015 was 1 in 2.7 billion per year, and by a resettled refugee 1 in 3.6 billion (Nowrasteh, 2023). A comprehensive European analysis found the primary terrorist threat is ‘homegrown’ terrorists radicalised online, with no ‘empirical evidence to suggest that increases in the share of immigrants from abroad is significantly correlated with higher rates of terrorism, including immigration from Muslim majority countries’ (do Céu Pinto Arena, 2017).

Studies showing a rise in crime that was attributed to refugee arrivals in Germany (Alkousaa, 2018; BBC News, 2018; Lange and Sommerfeld, 2024) in fact indicate an overreporting rate for migrants and refugees, and point to explanatory causal factors such as lack of access to the labour market, mental health difficulties, and demographic profile (younger men are more likely to commit crime) regardless of nationality or immigration status. Overall crime rates in Germany have in fact fallen since 2014, coinciding with the arrival of large numbers of refugees (Statista Research Department, 2025).

More broadly, one of the most comprehensive longitudinal studies found that ‘long-standing concerns about immigration as a major source of crime are unfounded’ (Flagg, 2018). A meta-analysis by Kubrin and Ousey (2023) found that in comparison to states with low levels of immigration, ‘at the macro-level, the vast majority of studies indicate that places experiencing more immigration have less (or roughly equivalent) crime’ with ‘immigration’s crime suppressing impact [being] most notable for violent crime, especially homicide’; and, at the micro-level, immigrants are far less likely than existing citizens to commit crime (p. 95). In fact, curiously, the Marie and Pinotti (2024) study that Rindermann cites to support a positive association between immigration and crime actually explicitly negates it: despite overrepresentation in incarceration rates, existing studies as well as comprehensive most-recent European data ‘do not, in general, find any detectable causal effect of immigration on local crime rates’ (p. 195). Further, with a significant rise in hate crimes by right-wing groups and violent attacks against Muslims in Germany and across Europe, among whom are many refugees (Bundeskriminalamt, 2025; CLAIM Organisation, 2025), refugees are plausibly far more likely to be victims rather than perpetrators of violent crime.

Moreover, it is worth reflecting on what is meant by ‘security’, and what is valuable about it. One plausible understanding is (the value of) persons’ lives, wellbeing, and human rights being reasonably secure from standard threats. This is plausibly what Rindermann has in mind in highlighting violent crime that threatens citizens. But, if so, then Rindermann, and the appendix he provides, fails to include any positive contributions to the lives, wellbeing, and rights of citizens in Germany by refugees. For example, as mentioned, refugees are overrepresented among key workers across European states, who during the COVID-19 pandemic incurred substantial health risks to perform roles (in healthcare, social care and social work, food supply, and sanitation) essential to protecting the lives, wellbeing, and rights of countless citizens, and to sustaining society while nearly all were under lockdown, which itself was essential to the protection of all those citizens (Gerver, 2022; BBC News, 2020). Consider also, in the German context, those in the voluntary sector, charity, and civil society, whom, reporting a sense of duty to their host state, support the most vulnerable citizens (Easton-Calabria and Alaous, 2022); the thousands working in the care sector supporting the wellbeing of elderly, vulnerable, and sick citizens (about whom Oshal was also especially concerned) (Grunau, 2025; European Commission Directorate-General for Migration and Home Affairs, 2025); those thousands working in nursing and as doctors including the over 6,000 Syrian doctors who make up the largest group of foreign doctors in German healthcare (AP News, 2024; European Commission Directorate-General for Migration and Home Affairs, 2025). The number of citizens whose wellbeing has been vastly improved and/or whose very lives have been saved by these contributions is incalculable (and thereby admittedly difficult to include case by case in an appendix). In terms of security then, German citizens are overwhelmingly more likely to be assisted or have their lives, wellbeing, and rights saved by a refugee than threatened by one.

A final point is the following. It may be tempting to argue that since the state ought to protect its citizens, and since some refugees may pose a security threat, the state ought to deny protection to refugees. But this would be a mistake. In 1940, U.S. President Roosevelt reported that ‘among German refugees […] especially Jewish refugees, they found a number of definitely proven spies.’ As a result of these and comparable fears, many thousands of Jewish people were denied protection and suffered and died under Nazi persecution (Holocaust Encyclopedia, 2022; Gross, 2015). We all now rightly recognise that using the possibility or fact that some members of a group may pose a threat to deny protection to all was a catastrophic moral error. This was because the security concerns were either insubstantial or could be otherwise addressed, and refusing protection exacted an immense cost in the avoidable suffering and death of thousands of innocents.

It would be an equally serious moral error to suggest the violent crimes of a vanishingly small minority of refugees today could justify denying protection to all, and for the same reasons. There are less harmful and more effective crime prevention and reduction strategies,14 and the moral cost would be immense: extensive suffering and death to innocent persons who have nothing to do with any crime or threat, but who are simply seeking protection and are not liable to be harmed. Denying protection to innocent refugees because of purported security costs would thereby violate the foundational principle – the ground norm of criminal justice, and of prohibitions on collective punishment in International Law – not to harm innocent persons for the crimes of others.

To summarise, on statism, states must balance the priority-weighted interests of citizens with the interests of refugees, to determine potential obligations of protection. Careful consideration reveals that the purported net costs to citizens of refugee protection are not substantiated (economic and security costs are contested), reasonably avoidable (with policy alternatives), or not substantial (in being mitigated, offset, or even outweighed by economic and security gains). By contrast, net costs to refugees of being refused protection are substantiated, not reasonably avoidable, and substantial: contemporary refugees face well-documented severe threats to their lives, basic needs, and human rights and endure extensive mental and physical suffering and human rights violations, both in the causes of their displacement and whilst they are displaced, and have no reasonable alternative to secure adequate protection but to seek safety in Northern states (UNHCR, 2026c; Parekh, 2020; Hillier-Smith, 2023). A statist priority-weighted balancing of interests therefore produces one sound outcome: Northern states have obligations to provide, and thereby may not refuse, protection to refugees.15

Oshal (2025, p. 6) argues that if refugees did have a substantially lower IQ (a 12–15-point deficit) than the national average, this would ‘render it much harder [for those refugees] to participate on full and equal terms in society’, because IQ is the ‘most important predictor of, and contributor to, people’s life outcomes including their educational outcomes, socioeconomic success, and occupational status.’

It is not clear what is meant by ‘participating on full and equal terms in society.’ One plausible understanding is having capabilities to access the goods and opportunities deemed markers of social standing, including particular material goods, employment, housing, community participation, political rights, and civic participation; and, in virtue of these capabilities, being viewed and treated by others as having equal social worth (Wolff, 1998; Anderson, 1999). But on this understanding, it is straightforwardly false that a below average IQ precludes full and equal participation. Afterall, almost precisely half of any domestic population has a below-median IQ (with a smaller percentage lower than that). No one seriously suggests that fellow citizens who might have a below average IQ are thereby precluded from material goods, housing, employment, community, political rights and civic participation, nor that they are for that reason inferior as lesser or non-valuable members of society. Alternatively, Oshal may simply mean ‘having less economic or career success relative to others’, but that is contestable (as discussed below), and entirely different from, and should not be equated with, participating on full and equal terms.

More importantly, one might wonder why refugees may face barriers to social participation. The documented barriers are well established: the psychological trauma of displacement, dangerous journeys, and family separation or bereavement; severe disruption to education, employment, and training; language barriers; navigating an unfamiliar institutional environment; insufficient integration programmes; non-recognition of qualifications from states of origin; legal restrictions on access to employment, education, and political participation; and/or institutional and local hostility and discrimination.17 Despite these evidenced barriers, Oshal instead speculates that a supposedly genetically caused lower IQ is the primary limitation to refugees’ social participation (pp. 9–10). Oshal includes an array of material on alleged intelligence differences between racial groups. This material is irrelevant, however, not least because refugees do not belong to one ‘race’ (whatever that is taken to mean);18 the states with the highest rates of displacement under UNHCR’s mandate are various: Venezuela, Syria, Ukraine, Afghanistan, Sudan, and South Sudan.19 But also because the relevant site of assessment is the question of what barriers refugees face to social participation (alleged IQ differences or social factors), not the supposed causes of alleged IQ differences. Indeed, to assess whether IQ, as opposed to other factors, undermined social participation, that factor ought to be held constant, and hence the relevant comparison would be between arrived refugees and existing citizens with comparable IQ scores. If the existing citizens faced identical or comparable difficulties in accessing social goods and opportunities, this would support Oshal’s claim; if they instead faced fewer and less robust barriers and could access social goods and opportunities with less difficulty, this would support the opposite. Such an experiment would likely be unnecessary, however, since the answer to whether existing citizens and refugees face identical or comparable barriers to social participation (all else being equal) seems a forgone conclusion.

Moreover, despite the difficulties and barriers faced, refugees (and migrants from the Global South more broadly) do access social goods and opportunities and fully participate in social and political communities when granted the capabilities to do so (Keita and Dempster, 2020). For instance, in the German context, 61% of Syrian refugees are in full employment, 75% of them in qualified positions, and 25% overqualified for their current roles; and, as mentioned, the over 6,000 Syrian doctors represent the largest group of foreign doctors in German healthcare (AP News, 2024; European Commission Directorate-General for Migration and Home Affairs, 2025). Further, Oshal cites figures showing only 14% of Somali immigrants in the U.S. have a bachelor’s degree or higher (Batalova and Lorenzi, 2022). But that figure appears an odd selection because that same source in fact shows that ‘Sub-Saharan African immigrant adults tend to have higher levels of education than the overall foreign- and native-born populations […] participate in the U.S. civilian labor force at a higher rate […]’ and ‘were almost as likely to be employed in management, business, science, and arts occupations as their U.S.-born counterparts.’ The source used by Oshal to seek to demonstrate that those from certain regions may fail to participate on full and equal terms therefore demonstrates the opposite.

An even more fundamental concern arises, however. Recall, Oshal is seeking to demonstrate the costs to refugees of being protected. Oshal does this by considering the benefits to those refugees of protection ‘compared to a counterfactual scenario where the refugees’ IQs are broadly similar to, or higher than, the host population’s average’ (p. 6). That baseline is bizarrely ad hoc and necessarily fails to demonstrate any costs to refugees. Demonstrating that a person would have benefitted from X more greatly if they had entirely different characteristics does not in any way demonstrate that they did not benefit from X, much less that X was a cost to them. Rather, the correct counterfactual comparative baseline to calculate if protection is a benefit or cost to refugees is between their being protected in Northern states and their not being so protected. Indeed, this is the comparison Oshal employs for all their other calculations: costs to host societies and costs to the Global South are both calculated as a comparison between refugees being protected within Northern states and those refugees not being so protected (pp. 10, 12). It is inconsistent, as well as misleading, to ‘move the goalposts’ here. And, in doing so, Oshal fails to demonstrate any discrete or overall costs to refugees.

Section 1 challenged purported economic and security costs to host societies. This subsection will therefore be brief, considering only other costs that Oshal highlights, including declining social trust, interethnic tensions (including nativist backlashes and violence), and diminishing ‘institutional quality’ (pp. 10–11). On institutional quality, Oshal cites Garrett Jones’ (2022) view that immigrants from societies with lower levels of ‘trust’ and ‘saving’ and higher corruption negatively impact host societies’ institutions (p. 11).

Declining social trust (presumably concentrated against arrived refugees) may stem from out-group bias, xenophobia, misperceptions due to lack of interaction, or unsubstantiated fears of economic and security threats propagated through media and political misrepresentations. Each could and ought to be addressed through integration and community cohesion programmes, and responsible, evidence-based academic, political, and media representations that alleviate rather than inflame public anxieties (see for example Knappert et al., 2021; Dempster and Hargrave, 2017). Anti-refugee violence is a serious problem, but provides a reason to address it, not to refuse protection to refugees, which would wrongfully punish the (would-be) innocent victims.

On ‘institutional quality,’ recent evidence shows that migrants from states with higher corruption tend to welcome the move and be positively predisposed to new institutional environments, exhibiting high rates of trust, political participation, and support for host institutions, with no detrimental effect on ‘institutional quality’ (Aksoy et al., 2025). This is plausibly particularly the case for refugees, who are forced to flee because their own state was unwilling or unable to adequately protect them, and thereby actively seek safety and participation in another state. Last, any purported impact on the composition and direction of political institutions is contingent on domestic policy on political participation: nearly all states, for example, deny refugees the right to vote let alone stand for office.20

Oshal argues that protecting refugees within Northern states could exact catastrophic costs on the Global South. For Oshal, the Global South is dependent on research and innovation in the Global North, including in disease prevention, green innovation, and food production (p. 13). Protecting this innovation constitutes ‘a global priority’ (p. 14), on which the ‘whole of humanity depends for its future welfare’ (p. 17). Yet, protecting large numbers of refugees will impose economic costs on, and diminish the institutional quality of, Northern states, where institutional quality is measured by ‘protection of property rights, market-friendly policies, control of political corruption, and effectiveness of the judiciary’ (p. 13). These impacts will undermine or slow research and innovation with catastrophic results. Oshal emphasises that 9 million die from hunger and between 5.7 and 8.4 million from poor healthcare each year, and that 250,000 additional deaths a year are projected from climate change, to underline that even a modest delay in innovation can exact ‘a tremendous human toll’ (pp. 14–15), presumably meaning suffering and death for thousands if not millions of innocent persons.

To establish that denying protection to refugees within Northen political communities is necessary to avoid these catastrophic consequences, each of the following causal links must be demonstrated: (1) refugees necessarily impose economic costs and weaken institutions in Northern states; (2) they do so in specific ways that diminish property rights, market-friendly policies, control of corruption, and judicial functioning; (3) to a sufficient extent to decrease (funding for) research and innovation; (4) in ways that impact the specific areas essential to the Global South and to global development more broadly; and (5) to a sufficient extent to increase suffering and death associated with global poverty and climate heating for thousands if not millions of persons.

None are so demonstrated. On (1), as shown, it is not established that refugees necessarily impose net economic costs (in fact, the contrary is more evidenced and plausible) or negatively affect institutional quality. On (2), there is no evidence that refugee protection has specifically altered or would alter Northern states’ market-friendly policies, commitment to property rights, judicial functioning, or levels of corruption. On (3), there is no evidence that any purported costs would necessarily be absorbed by cutting research and innovation. On the contrary, research and innovation funding increased substantially across EU states (by 62%) from 2014–2024 to its highest ever level, the precise years of the protection of large numbers of refugees from the Global South, including in Germany and Sweden, which protect the largest numbers per capita, with Sweden the leader in R&D funding as a percentage of GDP.21 Globally, such spending is at its highest ever level and increasing.22 On (4), there is no evidence that any cut to (funding for) research and innovation would be in areas relating to global health and development. On (5), it is doubtful that (increases to) present and future preventable deaths associated with global poverty and climate heating are, and would be, (primarily) attributable to cuts to, or insufficient, Northern research and innovation, rather than, for instance, Northern states emitting the most greenhouse gasses globally (Ritchie, 2019); cuts to development assistance (Cavalcanti et al., 2025 project that recent USAID cuts will cause 14 million additional deaths by 2030, including 4.5 million children); intellectual property rights preventing global access to life-saving products; maintenance of Global South debt, increasing tariffs, domestic subsidies, and unfavourable trade terms that cripple Global South economies; and the perpetuation of a global economic order with resource and borrowing privileges that incentivise corruption and repressive regimes and thereby sustain or exacerbate global suffering, poverty, and refugee displacement itself (Pogge, 2008, especially pp. 118–20; Ashford, 2013; Blunt, 2015).

Oshal is correct that the preventable suffering and death affecting millions of innocents resulting from global poverty and climate heating is an urgent moral concern. But this concern supports addressing the various established contributing factors, not refusing protection to refugees. It is not clear that refusing protection, and presumably allowing innocent refugees to instead suffer and die, would be an effective (notwithstanding ethical) means to prevent additional poverty and climate-related deaths. Fortunately, there are more simple, direct, effective, and ethical alternatives, the most obvious being reducing global emissions and increasing development assistance (Cavalcanti et al., 2025). And if, contrary to the more established contributors to, and means of preventing, these deaths, Oshal is correct, and research and innovation is the most essential, then Oshal has provided an elaborate argument for ring-fencing or increasing R&D funding and/or increasing the share dedicated to addressing these deaths, not for refusing protection to refugees.

Oshal has not demonstrated costs of refugee protection. Yet, Oshal suggests that even if these purported costs would not obtain, Northern states are not required to protect refugees within their political communities since they could instead protect those refugees more effectively through involuntary relocation to safe third states in the Global South and supporting local integration in those regions (p. 15).

There are many concerns with this proposal. First, in practice, an assumption that refugees should be protected in the Global South underpins policies that result in extensive human rights violations and deaths (Fine, 2019). These include the forced containment of refugees in regions where they face extreme poverty and extensive human rights violations, including torture (Hillier-Smith, 2020); forced encampment where millions are housed in camps for prolonged periods and endure extensive human rights violations, in particular sexual violence (Parekh, 2016); and forced relocation or return schemes to ostensibly ‘safe’ states that are in reality less able or unable to adequately protect refugees, resulting in human rights abuses, trafficking, exploitation, and psychological trauma.23 Forced relocation also overrides refugees’ will, deprives their liberty through coercion, and dehumanises them in treating them as undesirable elements to be exported away akin to some toxic waste (see Sandel, 2012). These schemes are also typically vastly expensive and inefficient: the UK-Rwanda scheme, under which four persons voluntarily relocated to Rwanda, cost hundreds of millions (see UK Home Office, 2024).

Second, ostensible regional protection may not adequately address the relevant harms of displacement that refugees require protection from, including extensive mental and physical suffering; violations of autonomy, dignity, and moral status; deprivation of capabilities to enjoy human rights; social dislocation; and loss of membership of a political community and state (Owen, 2020; Parekh, 2020; Cole, 2022; Hillier-Smith, 2024b). This is especially the case in regions close to refugees’ states of origin, mostly low- to middle-income states without the requisite infrastructure or resources, or requisite institutional and financial support from the Global North. Forced relocation and regional protection are therefore not ethical, effective, or feasible policy alternatives.

It must be objected that adequate regional protection is possible in principle. But adequately addressing the relevant harms of displacement would require a radical increase in foreign political, economic, and development assistance as well as transformational institution building. This would be targeted to drastically improve the macroeconomic conditions of low- and middle-income states that host the majority of the worlds’ refugees to build and sustain a vastly improved infrastructure necessary to adequately protect refugees’ security and subsistence, human rights, Refugee Convention rights, autonomous and dignified existence, and inclusion as political and moral equals through integration into the local economy with rights to work, healthcare, housing, education, civil and political rights, and pathways to political membership.24

This in-principle possibility, however, provides no justification for refusing protection in Northern states to refugees who were not in fact so protected. As Kieran Oberman (2025, pp. 570–71) points out, it is not justified to refuse protection to a person in need on the grounds that she could have been otherwise protected (even more cost-effectively) when that protection was not in fact forthcoming, especially if it was one’s obligation to provide that prior protection in the first instance. For example, a hospital may not refuse to treat a patient in desperate need on the grounds that a more cost-effective home treatment was available when the patient through no fault of her own did not in fact receive it, especially if the hospital was obligated but failed to provide the home treatment.

Moreover, Oshal accepts that Northern states do possess obligations to protect refugees, but these can be discharged either through protection within their political communities or via supporting regional protection (pp. 2, 17).25 Therefore, on this view, states may only refuse protection within their political communities if they instead fulfil the then compulsory obligations to provide the radical and transformative regional support required for adequate protection described above. That would be a seismic rupture to the status quo: a fundamental restructuring of the entire global regime of refugee governance with an abolition of current harmful and punitive practices (border violence, forced detention, encampment, and containment) to be replaced with transformational multilateral increases in foreign assistance, infrastructure investment, and grants to host states, all of which to vastly improve the security, subsistence, human rights protection, and societal and political inclusion of tens of millions of refugees worldwide who presently lack it. Only once this point was reached could Northern states, on Oshal’s account, refuse protection within their political communities. Since no Northern state in the world is sufficiently supporting such regional protection,26 none may refuse protection within their political communities. Hence, when applied, in highlighting the in-principle possibility of regional protection, Oshal has not provided an argument against Northern states protecting large numbers of refugees from the Global South within their political communities, but an argument for it.

As is well established (UNHCR, 2026b), refugees have fled their states of origin due to extensive human rights violations, variously including military attacks on civilians, mass violence, torture, rape and gender-based violence, persecution, extrajudicial killing, ethnic cleansing, and potential genocide. The vast majority are then displaced into conditions where again they lack adequate security and subsistence, and face further human violations in camps or urban areas in the Global South, or on perilous journeys where tens of thousands die trying to seek safety in the Global North. For the minority of refugees able to reach it, protection within Northern states (through asylum or resettlement) provides the immediate benefit of (1) security and subsistence, and thereby safety from severe threats to their basic needs, human rights, freedoms, and lives. The vast improvements to wellbeing, enjoyment of human rights, and prospects for survival for each person protected are incalculable. Protection also secures (2) an autonomous existence with opportunities for refugees to live and rebuild their lives according to their own wishes and reunite with loved ones; (3) a dignified existence and decent standard of living with access to employment, social welfare, and social participation; (4) recognition and treatment as moral equals through societal and political inclusion; and (5) pathways to an otherwise deprived membership of a political community and state. These benefits are immense for each person and are entirely orthogonal to prospective IQ: every refugee protected would receive each of these benefits to the same degree no matter their IQ score.

Section 1 already considered selected benefits refugees provide: significant overall economic contributions; demographic offsetting; societal contributions of key workers, healthcare workers, and social care workers; and civil society participants supporting vulnerable groups. These are again orthogonal to IQ, since, as demonstrated in Section 2.1, it is not IQ that precludes economic and social participation and contribution, and the above benefits are the real contributions of contemporary refugees (who allegedly have a below-average IQ).

South-to-North migration yields vast global economic benefits. Some economists estimate that increasing it by a few percentage points would add tens of trillions to the global economy (Clemens, 2011). One significant benefit to the Global South specifically is remittances. In 2024, remittances amounted to 700 billion USD: three times total global development assistance from the Global North (Our World in Data, 2026). The estimated net economic gains to the South of increasing South-to-North migration by a few percentage points would exceed ‘the combined effect of meeting national targets for development aid, cancelling all Third World debt and abolishing all barriers to Third World trade’ (Holtug, 2020, p. 137). This is why the UN’s development agency and the World Economic Forum advocate greater migration (United Nations Development Programme, 2009; World Economic Forum, 2024). Refugee remittance networks specifically are less stable due to the volatile nature of forced displacement and because would-be recipients are often themselves scattered globally, displaced as refugees or internally. Nonetheless, refugee remittances provide life-saving support to those facing destitution in camps or urban areas, risking their lives on dangerous journeys, or internally displaced facing severe threats, and support post-conflict reconstruction (Alix-Garcia et al., 2019; Vargas-Silva, 2017).

Moreover, protection within Northern states enables the eventual return (voluntary repatriation) of refugees who, had they not been protected, would have endured extensive suffering, human rights violations, and/or been killed. It is plausibly better for post-conflict recovery – rebuilding infrastructure, restoring government capacity, stimulating demand for goods and services to build the economy, providing a workforce, and developing a civil society – to have a well-protected refugee population return than one that has greatly suffered and among whom many have been killed. And returnees who have maintained and/or acquired skills, education, and material resources through participation in Northern economies would further contribute.

Last, 68% of the world’s refugees are currently hosted in low- to middle-income states in the Global South (UNHCR, 2026c). These states often lack the resources and capacity required to adequately protect refugees and their own citizens. Lebanon, for instance, a country the size of Wales, with a GDP 50 times smaller than that of the UK, hosts over one million refugees, and, according to the Office of the United Nations High Commissioner for Refugees (UNHCR), cannot adequately meet their basic needs (UNHCR, 2025). If Northern states, among the most affluent and stable on the globe with greater resources and capacity for protection, protected a far greater and fairer share of refugees within their political communities, this would provide an immense benefit to states in the Global South, whose resources are stretched. This would enable those states to better ensure the security and subsistence of their own citizens and better protect remaining refugees (Gibney, 2015; Lutz and Caballero-Vélez, 2026). A fairer distribution of responsibilities of protection would improve the wellbeing and human rights protection for millions of refugees as well as for millions of citizens of the Global South. Again, none of these benefits – remittances to support vulnerable persons, repatriation of protected refugees, or global responsibility-sharing – are contingent on refugees’ IQ.

To summarise, when only the most obvious, direct, and short- to medium-term benefits of refugee protection are taken into account within a more complete cosmopolitan analysis, an entirely different picture emerges from the one painted by Oshal’s incomplete analysis. Considering the interests of all affected globally reveals that the benefits of refugee protection in the Global North are incalculably enormous, not least to refugees themselves, and overwhelmingly outweigh prospective costs. Since these benefits are orthogonal to IQ, this conclusion holds irrespective of such consideration. Therefore, completing the cosmopolitan analysis reveals that, owing to the enormous overall benefits, Northern states have particularly strong obligations to protect great numbers of refugees within their political communities and IQ is irrelevant to this fact. In this way, the more complete cosmopolitan analysis has been able to resolve the disagreement on the relevance of IQ to Northern states’ obligations of refugee protection.