BLOG POST

The Costs of Hosting Refugees in OECD Countries and Why the UK Is an Outlier

In 2023, the OECD’s Development Assistance Committee (DAC) countries recorded a huge $29 billion of Official Development Assistance (ODA) spent on hosting refugees and asylum seekers within their own borders. 

Here, we put forward new analysis, showing that the cost each country reports per person arriving differs wildly. 

While seven countries report under $1,000 per head, and most (23 of 31) report under $10,000, the UK reports the highest costs per refugee of any country, at nearly $26,000. This is over 30 percent higher than the next highest country, Ireland; and a massive 150 percent higher than the next highest G7 country, Canada. Switzerland reports the third highest figure overall. Other countries, particularly the US, are seeing large numbers of refugees and reporting substantial refugee-hosting costs for the first time, but are reporting a more modest per head cost (currently $7,500 per head). 

The UK is a clear outlier in how much it reports—both in comparison to the rest of the G7 and to its wider OECD peers—and it is setting a poor example for others to follow. 

Refugee numbers and costs rise steeply

The war in Ukraine and instability in Afghanistan, Sudan, Syria and Venezuela have led to a steep increase in the number of people moving to other countries to avoid war or persecution, with the total number of refugees tripling in the last decade to over 43 million. A quarter are hosted in high-income countries, most of whom have fled the war in Ukraine. It is, of course, important that these people are properly supported and, since the mid-2000s, some high-income countries have significantly increased the costs recorded as ODA. Countries can report nearly all the costs of refugee hosting in the first year (though the methodology and the DAC chair both emphasize the need for a conservative approach.) It remains controversial that this is within the rules—the core principle of ODA is that it should be spent to support the welfare and economic development of developing countries, and before 2014, reported costs had never exceeded $5 billion.    

There are nine countries who reported costs of over $1 billion in 2023; in a total of $29 billion. To properly compare these figures, we need to consider how many refugees were hosted. To do so, we combine UN figures on asylum applications by country, with additional data on Ukrainian arrivals which do not appear in the asylum application figures (especially in the EU where most have been granted temporary protection). We consider the figures over the two latest years available (2022 and 2023) to ensure we capture the large spike in Ukrainian arrivals in 2022 and any subsequent costs falling in 2023.  

Comparison of spending per refugee or asylum seeker

The below chart sets out the amount of ODA reported to the DAC per refugee and asylum seeker in 2022 and 2023. The mean spend was $6,100, though 18 of 31 countries spend less than this; and the median country, Japan, reported $3,300 per head. 

Three out of the 31 countries did not report any costs at all in 2022 and 2023 (Australia, Hungary, and Luxembourg); and a further four reported costs of just $1,000 per head or less. 

The UK reports the highest per head cost, with Ireland second and Switzerland third. 

Figure 1: ODA per refugee / asylum seeker 

In-donor refugee cost (IDRC) ODA per refugee / asylum seeker across DAC countries in 2022 and 2023 (2022 USD thousands)

Bar graph of in-donor refugee cost per country with the United Kingdom having the highest cost perperson

Source: Authors' analysis of: OECD DAC1 Table; UNHCR Refugee Data Finder; Eurostat, Decisions granting temporary protection by citizenship, age and sex - quarterly data (migr_asytpfq); Home Office & UK Visas and Immigration, Ukraine Visa Schemes: visa data (updates from 5th January 2023 & 4th January 2024)

Note: The number of recipients of IDRC ODA from DAC countries is calculated as (a) the number of people who have applied for asylum (from ODA-eligible countries), plus (b) Ukrainians who (i) were granted temporary protection, in the case of EU countries, or (ii) arrived under the Ukraine Scheme, in the case of the UK. The chart presents results for the two years 2022 and 2023: this will reduce bias from the spike in the number of recipients in 2022 whose ODA-eligible first-year costs largely fall in 2023.

US reporting significant costs for the first time

The US has seen a large inflow of people in absolute terms (see Figure 2 below), and is recording a relatively modest $7,500 per head for a total of $7.8 billion in 2022 and $6.2 billion in 2023. Before 2021, the US had never reported over $2.5 billion in relation to hosting refugees. Still, it has done so against a backdrop of a steep increase in the number of refugees it hosts and significantly increasing its overall ODA spend—from $40 billion in 2020 to nearly $64 billion in 2023. Many European countries have cut overseas programmes, as they record more costs of refugee hosting as ODA. 

The UK an outlier

Compared to its peer countries, the UK is a notable outlier. It reported significantly more ODA per refugee and asylum seeker ($25,600) than all other DAC countries in 2022 and 2023. It recorded two-and-a-half times more than its next G7 peer, Canada ($10,100); over three times more than Italy ($7,800) and the United States ($7,500); and over four times more than France ($6,400) and Germany ($6,100).

Does the UK need to report these costs?

Giving evidence to Parliament in April this year, the Home Office permanent secretary  Matthew Rycroft said the Government had “no choice” but to report the spend as ODA. As we’ve seen, several countries do not report these costs at all. Indeed, the UK itself did not do so until 2009 despite ongoing inflows of refugees, and before 2021 reported costs were well-under $1 billion.

Are costs per person related to the number of people arriving?

With the significant increase in refugee numbers, is it the case that countries receiving large numbers see the highest costs?

The below chart compares costs reported per recipient with the total number of recipients. This doesn’t suggest any particular relationship; though it highlights that the UK (ISO code: GBR) does not support an unusually high number of recipients compared to other DAC countries; and the largest hosts Germany, Poland and the US, report relatively low costs.

Figure 2: Costs per person vs people supported

IDRC ODA per refugee / asylum seeker in 2022 and 2023 (2022 USD thousands) vs the annual average number of refugees / asylum seekers supported over 2022 and 2023 (thousands) 

Scatter plot of costs per person vs number of people supported with UK having the highest cost of a little more thank 25K per person

Source: Authors’ analysis of: OECD DAC1 Table; UNHCR Refugee Data Finder; Eurostat, Decisions granting temporary protection by citizenship, age and sex - quarterly data (migr_asytpfq); Home Office & UK Visas and Immigration, Ukraine Visa Schemes: visa data (updates from 5th January 2023 & 4th January 2024)

Note: The number of recipients of IDRC ODA from DAC countries is calculated as (a) the number of people who have applied for asylum (from ODA-eligible countries), plus (b) Ukrainians who (i) were granted temporary protection, in the case of EU countries, or (ii) arrived under the Ukraine Scheme, in the case of the UK. The chart presents results for the two years 2022 and 2023: this will reduce bias from the spike in the number of recipients in 2022 whose ODA-eligible first-year costs largely fall in 2023.

Are costs per person related to price levels in each country?

We might expect the costs of accommodation, food and services to be higher in some countries than others. Do countries with higher price levels report more ODA per refugee and asylum seeker?

The answer is yes, as Figure 3 shows. But price levels don‘t explain all the variation in the amount of ODA reported per person. There are a number of outliers, the most obvious of which is the UK. It reports more than two-and-a-half times as much ODA per person as we might expect from its price level. Ireland also reports more than twice as much as its price level suggests. On the other hand, we see that Switzerland’s high level of spend per person can largely be explained by its high price level.

Figure 3: Costs per person vs price levels

IDRC ODA per refugee / asylum seeker in 2022 and 2023 (2022 USD thousands) vs relative price levels in 2023 (US=1)

Scatter plot of costs per person vs price levels per country

Source: Authors’ analysis of: OECD DAC1 Table; UNHCR Refugee Data Finder; Eurostat, Decisions granting temporary protection by citizenship, age and sex - quarterly data (migr_asytpfq); Home Office & UK Visas and Immigration, Ukraine Visa Schemes: visa data (updates from 5th January 2023 & 4th January 2024); World Bank, World Development Indicators (PA.NUS.PPPC.RF)

Note: The line of best fit excludes countries which choose to report zero or very little IDRC: Australia, Greece, Hungary, Luxembourg, Portugal, Slovak Republic, and South Korea (which each reported under $1 thousand per refugee / asylum seeker in 2022 and 2023; see Figure 1).

Policy choices at Chancellor Reeves’ UK budget

The UK’s new Labour chancellor will set Departmental budgets for this and next fiscal year (24-25) at her budget on 30 October. Her Conservative predecessor Chancellor Hunt provided an extra £2 billion last year to help cope with additional refugee costs; and those costs do not appear to have subsided this year. Excluding refugee hosting costs, the UK’s aid budget recovered to 0.42 percent of GNI last year; and without additional funding, it seems likely to fall to around 0.36 percent in 2024.  Each is well below the reduced 0.5 percent target on “overseas aid” that Parliament voted on.

As ever, other countries will be watching the UK closely to shape their own development approach. 

Where next on hosting refugees and costs?

The purpose of having a common measure of ODA across countries is to enable a consistent approach to reporting and a measure that can be compared across countries and over time. The inconsistent inclusion of refugee costs undermines both those aims. Hosting refugees and incurring costs to ensure they are supported is a worthwhile activity which can be seen as a (global) public good that all countries benefit from. The DAC should agree that it is captured in the OECD’s wider measure of Total Official Support for Sustainable Development (TOSSD) which explicitly aims to measure contributions on international public goods. 

Still, the Development Assistance Committee seems incapable of such reforms. It lacks recipients in decision-making, and in any case the consensus-based approach means nearly any country can veto any proposed narrowing. 

For countries like the UK, the only constraint on reporting costs that displace aid remains international norms. The UK is well outside those norms. The new Government’s manifesto committed it to regaining Britain’s leadership on development. It faces an early test.

Disclaimer

CGD blog posts reflect the views of the authors, drawing on prior research and experience in their areas of expertise. CGD is a nonpartisan, independent organization and does not take institutional positions.


Image credit for social media/web: pbnash1964 / Adobe Stock