Asylum and refugee support in the UK: civil society filling the gaps?

ABSTRACT The vast majority of asylum seekers in the UK are not permitted to enter the labour market. In the absence of the right to work asylum seekers receive welfare support, which amounts to less than a third of the weekly spend of the poorest 10% of British citizens. This article presents new research on the third sector response to the poverty created by this policy regime. Through a four-pronged methodological design we map the scale of this response, and in doing so offer an alternative critical perspective on the inadequacies of government policy, inadequacies which lead to the human rights of some who are within, or who have been through the system, being breached.


Introduction
The vast majority of asylum seekers in the UK are not permitted to enter the labour market (see Mayblin 2016aMayblin , 2016b. In the absence of the right to work asylum seekers receive welfare support from the Home Office which is delivered independently of the income support payments made to unemployed UK (and, currently, EU) citizens. The level of financial support is lowaround 50% of Job Seekers Allowance payments. Successive UK governments since 2002 have argued strongly that restrictions on both welfare and work are necessary to avoid 'pulling' disingenuous asylum applicants (economic migrants) to the country (Mayblin 2017b). These policieslow levels of asylum support and no access to the labour markethave been subject to extensive criticism from NGOs, charities, and community groups (see Fitzpatrick et al. 2015;Pettitt 2013;Refugee Council 2004), here referred to as 'third sector organisations' (TSOs). However, the Home Office, which has a duty under human rights law to adequately support those going through the asylum system, maintains that support is sufficient to cover all 'essential' living needs, and has defended support rates in a court of law. This article presents new research on the third sector response to the demand for services created by this policy regime. We hypothesise that support rates do not cover all essential living needs and the third sector is filling a significant gap in this area. We argue that this can be explained in relation to minority demand for the provision of unpopular public goods, which the state has a disincentive to adequately deliver on.
While there is a burgeoning body of academic and NGO research on the extent and nature of destitution amongst asylum seekers, refused asylum seekers and refugees in the UK; and some research into the types of support provided to them by TSOs, no study has yet investigated the scale of the third sector response to inadequacies in the asylum support regime. Indeed, the refugee third sector (RTS) within and beyond the UK has been under-researched and under-theorised within academia in recent years. The recent lack of academic attention paid to the RTS and the scale of its activities at the national level in the UK is perhaps in part because doing this work presents a significant methodological challenge: this is a part of the third sector which is hard to map because Refugee TSOs (RTSOs) are often small, local, volunteer run, and operate 'below the radar' (Phillimore and McCabe 2010), filling gaps where capacity exists. This is also the case in other comparable country contexts, which similarly suffer from a dearth of data in this area. It is hard, then, to ascertain the full extent of this civil society intervention. In response to this methodological challenge, we have designed a research approach which brings together four data sources. The result is not a comprehensive picture, but it does present a much better understanding of the scale of the third sector response to gaps in state provision in the UK than previously available.
We theorise the gap-filling role of the RTS in relation to the demand side theory of third sector activity (Weisbrod 1975). The third sector is often theorised in relation to the two core sectors of market economies: the state and the market. Within this context, the third sector can fulfil a range of roles which are complimentary to, or in collaboration with the state (Milbourne 2013). The third sector, then, can offer a 'buffer zone between state and society and mitigates social tensions and political conflicts' (Anheier and Seibel 1990, 14). This is true of the asylum support context: poverty and destitution amongst asylum seekers are accepted by the government to be a social problem, and the UK is signatory to EU reception conditions rules which oblige it to prevent such occurrences. Asylum seekers cannot enter the labour market, thus removing the market from the picture as a solution. Yet our data demonstrates that the state is also failing to address this issue, despite having a legal duty to prevent asylum seeker poverty, and claiming that it is indeed fulfilling that role. In other words, within the context of the state-market-other triad, there should be no need for an 'other' (third) sector to step in.
Nevertheless, as Weisbrod (1975) has argued, in democratic societies, government produces collective goods that are desirable to a majority of the electorate. In order to win or maintain power, then, politicians are seeking majority support in their offer of collective goods. Providing for the essential living needs of asylum seekers is not a popular public provision, creating a strong incentive for the government to fail in its provision to them. RTSOs thus fulfil demand for unpopular public goods (the living needs of asylum applicants), and are largely financed by the voluntary contributions of those minority of voters who favour them. RTSOs might, therefore, be conceptualised as gap-fillers that emerge in response to private demands for collective goods not offered by government or available via the market. As our data will show, this situation is not sustainable.
This research makes three key contributions to the field. First, it offers vital new data on this topic, offering a new methodological design which allows us to reach beyond the existing literature. Second, it offers a new critical perspective on the inadequacies of UK government policy, inadequacies which lead to the human rights of some who are within, or who have been through the system, being breached. Third, it offers insights which can be applied in other, particularly European, contexts, where states are failing to meet their international commitments to asylum applicants, and where increasingly stretched TSOs are stepping in.

Asylum Support: The Policy Context
In common with many wealthy countries, successive UK governments have, over the past two decades, sought to decrease the numbers of asylum seekers who are able to travel to the country, who are able to make an application for asylum in the UK, and whose applications are successful (Mayblin 2017a;Squire 2009). While this has in part involved border controls, since the early 2000s it has also entailed restricting the welfare and working rights of asylum seekers in order to reduce economic 'pull factors' which are thought to attract disingenuous applicants. A series of legislative acts have thus been passed which have removed labour market access, have moved asylum seekers out of the mainstream welfare system, and have steadily decreased the levels of financial support paid to them (see Mayblin 2017b for an overview).
In order to access support individuals must be destitute, and accommodation is offered on a no choice 'dispersal' basis around the UK. The European Council's 2003 Reception Conditions Directive determines that the Home Secretary has a duty to provide support in respect of essential living needs, though what might count as 'essential' is at the discretion of the Minister. Since supporting asylum seekers has been electorally unpopular since the early 2000s, there is a political disincentive to ensure support levels are adequate. Until 2008, increases to the rates of asylum support were made on an annual basis and were broadly in line with increases to income support. In 2008 the link to Income Support ended and from 2009 the separate rate for single adults aged 25 and over was removed. The rate of support has been £36.95 per person per week for all categories since August 2015. This was a substantial reduction in support for single parents and families with children, who previously received a larger sum. Academics tend to describe the different economic rights afforded to different groups on the basis of immigration status as a 'stratified rights regime' (Morris 2001), described in Table 1.
This policy design should mean that all groups, except those whose application for asylum has been rejected, are adequately supported by the state. Indeed, the Home Office maintains that support levels are adequate. 1 Yet qualitative research has found that people are liable to become destitute while in the asylum system, and after being granted leave to remain (Fitzpatrick et al. 2016;Lewis 2007). Such research has also found that those who are in receipt of asylum support are living in poverty and are likely to have additional needs which exceed state provision (Dwyer and Brown 2005). In their review of research into poverty amongst asylum seekers and refugees Allsopp, Sigona, and Phillimore (2014, 20) found poverty 'to be present among some of the most vulnerable parts of the asylum seeker population, including pregnant women and newborn babies […] children […] LGBTI individuals […]and torture survivors'. This is a finding which is supported by a wide range of stakeholders. For example, the 2007 Inquiry into the Treatment of Asylum Seekers by the Joint Committee on Human Rights (Children's Society 2013), the 2013 Parliamentary Inquiry into destitution among asylum-seeking families, and the 2013 Home Affairs Select Committee Inquiry into the asylum system all highlighted poverty and destitution among asylum seekers, refused asylum seekers, and refugees. Most research in this area involves qualitative interviews with small numbers of people in particular cities. The fact that repeated studies in different locations have made similar findings over a number of years does indicate that poverty and destitution are a problem, but a national level analysis within the UK is still missing from research in this field. Furthermore, we have been unable to find any comparable work in other country contexts. One approach to investigating the extent of demand for subsistence support nationally is to research the scale of the third sector response, as charities rarely form, or expand services unless a demand is present.

The RTS response: what do we already know?
There has been a dearth of research on the refugee third (or voluntary) sector in recent years. Internationally, most of the literature on TSOs and migrants focuses on Mexicans in the USA, and there is, at the time of writing, very little English language literature focussing on EU countries outside of the UK (Garkisch, Heidingsfelder, and Beckmann 2017). While we acknowledge that there is no consensus on the definition of the third or voluntary sector, such is its diversity (Alcock 2010), the category 'RTS organisations' (RTSOs) here covers all organisations, of any size who specifically focus their charitable work on supporting those who have been, or are going through, the asylum system. At the smallest end of the scale are refugee community organisations (RCOs) which are refugee led, are often formed around national groupings, and form in response to changing international events, as well as national asylum policy (Dwyer and Brown 2005;Zetter and Pearl 2000). These organisations have historically been the most researched part of the sector but are too small to meet the UK Charity Commission mandatory registration threshold (having an annual income of £5,000) (Charity Commission 2016), which is part of the reason why quantifying them (in order to assess the scale of their activities) is so difficult. Both smaller and larger organisations tend to serve asylum seekers, refugees, refused asylum seekers, or all three, and work on a city or countywide scale. These are easier to identify (they are usually registered with the Charity Commission) and are often more well established. The largest organisations are nationally based, with larger incomes and operations in multiple urban centres, but their role as national policy ameliorators has received little academic attention.
As noted above, the third sector can fulfil a range of roles which are complimentary to, or in collaboration with the state (Milbourne 2013) but in the case of asylum support in the UK funding opportunities for the RTS have steadily reduced in recent years. The Home Office has contracted out welfare support advice services for those in the asylum system to just one organisation, Migrant Help, and face-to-face advice surgeries have significantly scaled back in recent years. Nevertheless, policies are in place which should mean that those going through or exiting the asylum system should be adequately supported by the UK government (unless their asylum application has been refused).
There is a growing body of knowledge about the types of activities undertaken by RTSOs. The activities of RTSOs have remained consistent over time according to the research; the major changes reported over the past 15 years are in the areas of client demand (increasing) and available funding (decreasing) (Price 2016). Organisations are often small, local, heavily dependent on volunteers, many are either faith based or rely on churches for service provision support, and are often located in asylum seeker dispersal areas (Dwyer and Brown 2005;Fell and Fell 2014;MacKenzie, Forde, and Ciupijus 2012;Petch, Perry, and Lukes 2015;Phillimore and Goodson 2010). The services provided include housing management and provision, legal and welfare advice, financial and other types of subsistence support such as clothes and food banks, and rights-based advocacy (Mayblin 2017b). As those in the asylum system have increasingly been restricted from accessing mainstream welfare benefits, more people are relying on friends, family, TSOs, communities, and local authorities for support despite government assurances that they are adequately supported via the National Asylum Support Service (NASS) (Price 2016). At the same time, available funding for TSOs and local authorities, as for the rest of the third sector, has decreased.
Existing research suggests that funding is not only dwindling, but is also precarious for organisations that provide services for asylum seekers and refugees (Fitzpatrick et al. 2015;Price 2016;Refugee Council 2004). RTSOs find that securing funding and meeting the rising demand for services diverts resources away from other activities, such as campaigning (Refugee Council 2004; Phillimore and Goodson 2010; Price 2016). A key area of provision is housing and housing advice, both for individual refused asylum seekers who are destitute, and those who should by law be supported within the asylum or mainstream systems, such as refugees and families with children at any stage in the asylum process. As well as providing information on accessing housing, RTSOs are providing night shelters and longer term accommodation (NACCOM 2013).
The apparent extent of this civil society response does indicate that there is a significant demand for support amongst these groups of people (asylum seekers, refugees, and refused asylum seekers), and that the state is not, therefore, providing adequate support to them (despite claims to the contrary). The third sector has a long history of stepping in where the state is unable to support those in need (see Knight 1993 for a UK history). Indeed, there has in recent years been a broader trend of declining public spending and an erosion of the state's responsibility for welfare. Since the 1980s the organisation of welfare in Britain has, therefore, increasingly depended on a plurality of forms of delivery, often involving the third sector (Harris 2010;Milbourne 2013). While other policy fields have been the subject of extensive research, the relationship between the state and the RTS has received very little academic attention (as pointed out by McGhee, Bennett, and Walker 2016). It is nevertheless clear that with the introduction of the NASS in the UK and the policy of housing dispersal (introduced by the Immigration and Asylum Act 1999) the Home Office followed the New Labour trend of subcontracting the delivery of some asylum-related services to RTSOs (Briskman and Cemlyn 2005;Zetter, Griffiths, and Sigona 2005).
While some large organisations within the RTS (such as the Refugee Council and Refugee Action) took on large contracts from the Home Office in the 2000s, and took on service delivery roles in this vein, these contracts have all but disappeared since 2010. In this recent period, the Home Office has brought many previously sub-contracted activities in-house. Despite the loss of several large contracts, there are indications that the RTS is nevertheless continuing to support asylum seekers, refused asylum seekers and refugees as their economic rights and entitlements diminish. In this case, then, a large and/or growing third sector which is providing poverty reduction support to those in the asylum system and/or new refugees would point to failures in state provision, despite commitments to the contrary. As Weisbrod (1975) points out, the third sector usually steps in where the state and market cannot or will not. Of course, states do also seek collaboration with TSOs in order to deliver services, but this is not the case here. Since the third sector has little formal role in the asylum support field (in contrast to other contexts, such as Greece for example), looking at the response of the third sector is an alternative, complimentary, approach to critically evaluating whether policies which determine the economic rights and entitlement of asylum seekers, refused asylum seekers and refugees are fit for purpose.
Various sources suggest that demand for services is very high, and cannot be met. For example, a London-based organisation surveyed by Price (2016, 12) described having queues outside its door from 6 am. Another organisation suggested that securing an advice slot with them was a 'golden ticket'. Yet aside from such anecdotal accounts, we do not know from existing research how many organisations there are, how much they are spending on supporting their client groups, or whether they are supporting clients who should be supported by the state. But such data are, of course, central to the question of whether the UK government is providing for the essential living needs of asylum seekers, as it claims, as well as to the issue of how we might conceptualise the role being played by such organisations.

Research methods
As noted above, RTSOs are difficult to count, which is likely why most research in this area takes a qualitative 'deep dive' case study approach in particular locales, or with particular large organisations, and focuses primarily on qualitative research interviews with small sample sizes. This is true of research into RTSOs in the majority of national contexts, as demonstrated by Garkisch et al.'s (2017) systematic review of the field internationally. The only studies that drew on quantitative data that Garkisch et al. were able to find were based in the US and Canada, and those were focussed on national opinion-based surveys. This means that exploring the scale of the third sector response to refugee and asylum seeker needs in the UK presents a significant methodological challenge. It also means that our contribution to this field goes beyond presenting new empirical material to also offering innovative methodological direction which might inspire work in other national contexts.
We have designed a research approach which brings together four datasets: 1. Data from the Charity Commission, which are the most comprehensive dataset (covering England and Wales) on registered charities and their activities available. 2. A survey of member organisations of a national asylum and refugee accommodation network. 3. Management data from the largest NGO working in the UK in this field. 4. Two case studies with small local organisations in England who support asylum seekers, refused asylum seekers, and refugees.
First, we draw on data from the UK's Charity Commission, which is the government regulator of charities. The publication of the Charity Register Statistics contains the financial returns of charities that they regulate. When charities register, they must provide information about their activities and the people they support. We used the Charity Commission's online database to identify all registered organisations who support refugees and/or asylum seekers and/or refused asylum seekers. To meet our inclusion criteria charities must use the word 'asylum' or 'refugee' in their 'activities description' on the database, and describe their activities as working 'for the prevention or relief of poverty' or 'provides accommodation/housing' (from proscribed options). We reviewed each charity and removed religious organisations and charities that do not serve asylum seekers, refused asylum seekers or refugees as their primary group.
The dataset of charities identified includes data on income, expenditure, activities, and client groups. We manually added information on operational locations, year established, and year removed from the register. Charities with an annual income of over £500,000 are legally obliged to provide the Charity Commission with detailed income and expenditure information (Charity Commission 2016). For these charities, we were able to look for information about the total income from charitable activities, donations, and trading, as well as information about staff and volunteers employed by the organisation.
Some charities are not required to register with the Commission. These include charities with an annual income of under £5000 and places of worship with an annual income of under £100,000. Faith-based organisations are a known source of support to migrants, particularly in providing crisis accommodation (Snyder 2014) but they have not been included in our sample. This will mean that our research significantly underestimates the level of support provided to asylum seekers by non-state actors. The omission of small 'below the radar' organisations, which as Phillimore and McCabe (2010) highlight constitute the largest proportion of civil society organisations, also suggests that our calculations are an under-estimate.
Our second method is a survey of member organisations of a national network of charities (henceforth 'NN') who work on preventing homelessness amongst asylum seekers, refugees and other migrants. The network itself has been a national charity since 2015 and an informal network of voluntary organisations since 2006. It has 38 full members, and there are new organisations developing housing initiatives in this field in different parts of the country every year. This is, therefore, a key network within the sector, suggesting data from their members can offer insights into the scale of housing operations and spend on housing and accommodation nationally. In spring 2017 we collaborated with this network to conduct a survey of accommodation projects.
The survey was sent to all 38 network members and was completed by 24 projects. The questions covered a range of financial information, including income and expenditure, funding sources, subsidised costs, staff and volunteer capacity, and the accommodation and other services provided.
The third source of data comes from the largest UK NGO working in this field (henceforth RNGO). RNGO supports refugees, asylums seekers and refused asylum seekers in a wide variety of ways, including offering emergency food, clothes, and cash to those facing severe hardship, and giving advice about how to access services. The organisation co-ordinates projects in hundreds of locations across the UK and routinely collects management information from each project. This includes a great deal of information about beneficiaries, including numbers, and demographic information such as age, gender, nationality, immigration status, and information about the type of support provided. This information can then be disaggregated by location. It is rare for this data to be used by those external to the organisation, but they granted us access to some key data: the organisation's annual budget for UK destitution services in 2015/16; the number of asylum seekers, refused asylum seekers and refugees supported in different locations; the forms that destitution support takes, the geography of destitution services; and the numbers of staff and volunteers providing destitution services. Information on which groups are being supported is vital because the support available to asylum seekers, refugees, and refused asylum seekers is very different and the policy implications, therefore, hinge on this kind of disaggregated data.
The fourth and final data source moves from the national to the local scale. We conducted qualitative interviews and observations with staff and volunteers in two medium-sized charities that are involved in alleviating the poverty and destitution experienced by forced migrants. Though we are interested in the national scale, there are many gaps in the data obtained from the sources discussed above, primarily difficult to quantify resources such as food parcels made up of donated food, and volunteer time spent working at the local level. These case studies offered the opportunity therefore to investigate the extent of such support in two locales: one dispersal city, and one non-dispersal city. We used the data from these two case studies (henceforth Dispersal-NGO and Non-Dispersal-NGO) to identify how local organisations are funded, and the extent of the role of volunteers in local organisations. This gives us vital insights into the ways in which small local organisations are managing the increasing demand placed on them.
How many TSOs supporting asylum seekers and refugees are there?
Using data from the Charity Commission, we have identified a total of 142 RTSOs that focus on alleviating poverty and destitution amongst asylum seekers, refused asylum seekers, or refugees in England and Wales. As noted earlier, the approach fails to capture RTSOs that have an income of under £5000, as such organisations are too small to meet the Charity Commission's mandatory registration threshold. Within the wider voluntary sector, 54% of organisations have annual incomes of less than £10,000 (though they make up only 5.5% of the sector's total income) (NCVO 2012). This means the number of RTSOs could be much higher.
The Charity Commission records the date charities registered, and the date charities that have ceased operating were removed from the register. Figure 1 presents  The increasing number of RTSOs does appear to indicate that there is growing demand for third sector services. While this increase in organisations correlates with the increase in UK asylum applications in the early 2000s, numbers of applications dipped around 2005 and stayed at significantly decreased levels for over a decade. Furthermore, asylum seekers and refugees should not have been placing such demand on the third sector since their support levels were said by the Home Office to be sufficient.
It is possible that the number of RTSOs is increasing in response to a rising population of refused asylum seekers. There is currently no national estimate available of the size of the population of refused asylum seekers in the UK. The most recent estimate was offered  by the National Audit Office in 2005. However, the Home Office have published data about the final outcomes for those who have made an application in a given year since 2004. This data shows 40% of asylum seekers who were refused asylum in 2004 are still not known to have departed. In each year since 2004, around a third of asylum applicants are refused protection, and are not known to have departed the UK. While there is no way of knowing how many asylum seekers departed without being detected by the Home Office, it is clear there is a growing population of refused asylum seekers in the UK. Figure 3 uses these figures to present an illustration of the rate at which the population of refused asylum seekers is growing. Our approach to calculating the cumulative total population of refused asylum seekers is rudimentarywe start from the obviously incorrect assumption that the population of refused asylum seekers in 1990 was zero, and assume that in each subsequent year one-third of initial applicants have remained in the UK following refusal. Despite the simplicity of the approach, the figures are in the same ballpark as a peer-reviewed study, which estimated there to be 280,000 refused asylum seekers living in the UK in 2001, increasing to 500,000 by 2009 (Gordon et al. 2009). Figure 3 also charts our estimate of the population of refused asylum seekers against the total number of RTSOs. The relationship between the population of refused asylum seekers and the number RTSOs is both intuitive and borne out in this illustration. This raises questions about the capacity of RTSOs to continue to effectively respond to destitution among this group. The story is not so simple, however; there are other factors that may contribute to the growth in the number of RTSOs, which we explore below.
Where are RTSOs located?
The Charity Commission register records the location(s) in which a charity operates. As is the case for charities in general, the vast majority of RTSOs operate in one region; and the number of charities in an area largely mirrors the number of people who live there. Densely populated cities such as Birmingham, Leeds, Manchester, Lambeth, Sandwell, Bradford, and Croydon have a high number of RTSOs (see Table 2), and indeed charities in general.  Table 2 records the 10 local authorities in the country with the highest number of RTSOs, and shows the population size in each location. Home Office data record the number of asylum applicants receiving Section 95 support in different areas across the UK. As shown in Table 2, Newcastle-upon-Tyne, Stockton-On-Tees, and Middlesbrough have a high number of applicants supported under Section 95 and also have a high number of RTSOs. Aside from having high numbers of applicants supported under Section 95, there is no obvious reason why these cities have some of the highest number of RTSOs in the country. These are not major urban areas; they are located in the North East of Englandwhich generally has the lowest concentration of charities of any region in the UK; and in areas of high deprivation such as these, we would usually expect to see fewernot more -TSOs (NCVO 2015).
This link between the number of asylum applicants on Section 95 support and the number of RTSOs in an area, while not surprising, potentially indicates that the system of economic support for asylum seekers and refugees is inadequate. The geography of the RTS may, therefore, be directly related to the implementation of asylum policy at the national level, but this inference is only valid if the groups receiving support, and the types of support being received, are putatively already covered by the state. We explore this in the next section.
How many people are relying on charitable support?
In this section, we look at the number of asylum seekers, refused asylum seekers and refugees who received support from organisations involved in our research (RNGO, Dispersal-NGO and Non-Dispersal-NGO) in 2015/16. Nationally, RNGO supported 9138 asylum seekers, refused asylum seekers and refugees, and 4130 dependents in 2015. The number of asylum seekers being supported equates to approximately 25% of all those who are in receipt of Section 95 support that year. The local picture also indicates that much more is going on than that which is provided by this large organisation. In 2015/ 16 there were 2000 visits to the help desk of the Dispersal-NGO; 100 clients were provided with small weekly welfare payments; 62 clients were provided with long-term accommodation; and 49 clients were provided with emergency accommodation. This is just one of many organisations operating in this dispersal city.
We might expect the demand to be very small in non-dispersal cities. Indeed, RNGO supported just 76 people in our non-dispersal city during 2015, handing out cash on 161 occasions; cash for local travel on 23 occasions; and providing advice about destitution on 141 occasions. Yet for the local organisation that we looked at in this same city (non-Dispersal-NGO) there were 2976 visits to the main office in 2015/16; 2321 food parcels were handed out; in total 1029 clients received help; including 88 unaccompanied young asylum seekers and refugees. The number of people receiving help from RTSOs does not necessarily equate to the number of people experiencing destitution. The number of people that an organisation helps may be as much an indication of the organisation's capacity as it is of the demand for that service; and organisations are unlikely to record or publish information about unmet need. Moreover, we cannot be sure how many people who experience destitution do seek help from voluntary organisations. What the data from RNGO do reveal is the proportion of those supported who are asylum seekers, refused asylum seekers and refugees, which is useful to our enquiry. As previously noted, it is only refused asylum seekers who are not co-operating with their own removal who are made purposefully destitute by the government (in an effort to encourage departure). Asylum seekers should be in receipt of Section 95 support which the Home Office report is adequate for covering living needs. Refugees or those with temporary leave to remain have access to mainstream welfare benefits. Refused asylum seekers who are co-operating with efforts to deport them are also granted subsistence payments. This system of economic support should mean that TSOs only experience demand for support from refused asylum seekers who are absolutely destitute.
Yet Figure 4 presents a breakdown of the immigration status of the people supported by the RNGO in 2015. The majority (53%) of people receiving support are asylum seekers; 25% have been granted some form of protection; and just 10% are refused asylum seekers with no further representations to make. The fact that the majority of people that RNGO helps are asylum seekers or refugees confirms the findings of qualitative studies: that people are liable to become destitute at all stages in their asylum journey, including while in the asylum system, and after being granted leave to remain. As presented in Figure 5, in 2015, the majority (61%) of beneficiaries in this, the largest NGO in the sector, were in receipt of statutory support: just 30% were in receipt of no statutory support. Asylum seekers who are in receipt of Section 95 support should not be relying on charities for food, clothing, bus passes, and hardship funds. The fact that so many are is a clear indication that those who are in receipt of asylum support are living in poverty and are forced to rely on charities to meet their basic needs in relation to food, clothing, bus fares, and unexpected financial events. Figure 6 shows the reason that beneficiaries of RNGO support were destitute. Destitution often arises because of errors and delays caused by government service providers. This includes a significant number of people who become destitute when granted refugee status (26%), or as a result of issues with government provided support (16% of respondents). While the organisation Migrant Help is contracted to deliver advice and administrative support in relation to such problems (for those in the asylum system), the Home Office does not fund RNGOs to support their clients while delays are resolved, and no organisations are funded by the government to support new refugees experiencing a gap in support between the asylum system and mainstream benefits system. This is despite the fact that in these cases destitution is caused by failures in the government's own asylum support delivery system, and in the transition from this to the mainstream welfare system.
For asylum seekers and refugees who are unable to access statutory support, a key area of provision is housing and housing advice. A large number of RTSOs provide information on accessing housing. A smaller number of RTSOs are providing night shelters and longer term accommodation. Members of the accommodation network that we worked with (NN) provide accommodation to asylum seekers, refused asylum seekers, and refugees in 28 cities in the UK. Members of this network hosted individuals in 165 houses, 18 flats, 9 church properties, and 6 night shelters (2 permanent and 1 winter only) in 2016. The majority of members manage hosting schemes which match asylum seekers, refused asylum seekers or refugees with host families. In 2016, the number of people accommodated by network members over the year came to 1707, an increase of 29% since 2015. Of these, 808 were refused asylum seekers and 499 were refugees. Member projects were accommodating 789 people per night at the time of the survey, an increase of 34% in that year. Over 12 months, the network estimates that members provide 209,250 nights of accommodation.

Support that is hard to cost
There are a number of types of support provided by RTSOs which cost little or nothing. Indeed, it is in the very nature of the charitable sector to do as much as possible with as little as possible, and to rely heavily on the good-will of volunteers. While volunteer time is one key factor which is difficult to quantify, services such food parcels, clothes banks, advocacy, and advice contribute to the support package offered to clients, which may become necessary because of gaps in statutory provision, but they are also very difficult to quantify.
In 2015/16 Non-Dispersal-NGO handed out 2321 bags of food to asylum seekers and refugees, totalling £30,869 worth of food provided. Access to their food bank is for asylum seekers, refused asylum seekers, and refugees that have insufficient income to feed themselves and their families. After cash, food parcels, clothing vouchers and hygiene packs were the most common types of support RNGO gave out in 2015. In total, this organisation provided 1535 food parcels, 1370 vouchers for second-hand clothing shops, and 1022 hygiene packs. The extent of the local response to these needs is indicative of very high levels of such support nationally.
The volunteer contribution to the refugee third sector is significant. For example, we estimate there to be more than 218 volunteers across the teams working in Dispersal-NGO, not counting volunteer hosts who offer accommodation to homeless people in their own homes. Volunteers provide essential advice, support and stability to asylum seekers. Volunteers with this charity spend on average a total of 463 hours a week volunteering at the time of this researchthis is the equivalent of 13 full-time roles at minimum wage. If volunteers were paid the national minimum wage of £7.50 per hour, the combined wage bill for Dispersal-NGO volunteers would be £700 a day/ £3,472 a week /£180,544 a year not counting national insurance contributions or other overheads. In Non-Dispersal-NGO a snapshot from volunteer data recording in March 2016 showed that 45 volunteers spent a combined total of 189 hours volunteering across the organisation's destitution services -this is the equivalent of 5 full-time roles. If volunteers were paid the national minimum wage of £7.50 per hour, the combined wage bill for this organisation's volunteers would be £300 a day /£1,500 a week /£78,000 a year. Note the differential burden in the dispersal city, in comparison to the organisation which is not operating in a dispersal city.
Accommodation network member organisations answered survey questions about the number of staff and volunteers that support the organisation (see Figure 7). Smaller organisations rely more on volunteers to deliver services.
Significant resource is expended by RTSOs assisting asylum seekers and refugees to access the welfare benefits that they are entitled to. Three of the four most common reasons for visiting the advice service of Dispersal-NGO in 2015/16 were: to get help accessing support as an asylum seeker; because of confusion over asylum status/process; and, difficulty accessing mainstream benefits. Table 3 presents a breakdown of the most common welfare related actions that RNGO took in 2015. The range of entitlements, appeals and requests for more information suggest that the complexity of the system is barrier to accessing benefits entitlements for asylum seekers, refused asylum seekers, and refugees and that the Home Office's own systems of communication are inadequate. As previously stated, only the organisation Migrant Help is contracted to deliver such  advice services, though these findings suggest that third sector organisations are nevertheless doing that work.

The cost of the RTSO response
The Charity Commission data reveal that the total income of our sample of RTSOs in 2015/16 was £33.4 million. In the same year, expenditure stood at £31.8 million, 95% of total income. The sector is dominated by a high number of small and medium-sized charities. Organisations with an annual income of under £5000 are not required to register with the Charity Commission and while some do, it is not possible to estimate here the total number in operation. Within the wider voluntary sector, the majority of organisations are very small, and yet with the 2010 funding changes, noted earlier, this may not be the case in the RTS. Figure 8 presents a breakdown of the number of RTSOs in each income band (over £10,000). Most of the organisations are small and medium sized: 97% have an income of between £10,000 and £500,000. The average income of RTSOs on the Charity Commission Register is £288.3k in 2015/16. Organisations with an annual income of over £500,000 make up only 3% of the total number of RTSOs registered with the Charity Commission, yet account for 70% of the sector's total income. This resembles the wider UK charity sector (NVCO 2012).
Twenty-four members of the accommodation network that we surveyed answered detailed survey questions about the proportion of their income received from different sources. Twenty out of 24 members received individual donations in 2015/16. Of these, organisations received an average of 20% of their income from individual donations. Donations appear to be particularly important to fledgeling organisationsone small organisation established in 2016 reports receiving 100% of their funds from individual donations; another small organisation established in 2017 receives 70% of their income in donations from philanthropic/faith-based organisations. Organisations with an income of over £500,000 are required to submit details of their funding sources to the Charity Commission. Of the eight charities that this applied to in 2015/16, organisations received on average 19% of the income from individual donations. Twenty two out of 24 accommodation network members received grants from charitable trusts or other organisations in 2015/16. Of these, organisations received an average of 50% of their income from charitable trusts or other grants, making grants the largest source of income for the organisations sampled. This is, of course, a competitive and finite funding source, which makes relying upon it risky for organisations. Fifteen out of 22 network members received statutory funds in 2015/16. Of these, organisations received an average of 15% of their income from statutory sources. All but one of these organisations receives statutory funding from the local council.
Government funding has a huge impact on the income of larger RTSOs: of the eight RTSOs with an income over £500,000 that are registered with the Charity Commission, three are operating with a significantly reduced income compared to five years ago, as a direct result of a reduction in statutory funding. One national RTSO lost £3 million of Home Office funding in 2014/15, another lost £7.7 million in 2015/ 16. The loss of funding is often associated with the Home Office moving the administration of a programme internally. These figures alone constitute a 25% decrease in sector income since 2014. Government funding is, therefore, precarious and subject to wider trends in state spending but demand for support from clients nevertheless continues to increase.

Conclusion
We know from existing research that the stratified regime of welfare rights afforded to different groups who are going through or have been through the UK asylum system results in different vulnerabilities to poverty and destitution as people move through the process. The upshot of this patchwork picture of poverty and destitution is that the third sector is playing a significant role in supporting those who have been failed by a state which has a duty to provide for all of their essential living needs. In this article, we have explored the scale of this third sector response. We hypothesised that the third sector is filling a significant gap in this area, and this is borne out by the data. Analysis of Charity Commission data shows that RTSOs are spending at least £33.4 million per year on poverty alleviation. This figure is certainly an underestimation and is also not indicative of demand, which organisations report exceeds capacity.
Our research shows that the demand to which these organisations are responding is being created both by the refused asylum seeker population, who are not supported by the state, as well as by demand from asylum seekers and refugees, both of whom the government claims have sufficient access to support. Therefore, at the same time as withdrawing from a strategic collaboration on asylum support with the third sector, the Home Office have simultaneously rolled out policies which create extra demand for the sector. The numbers of asylum seekers who are, or should be, in receipt of Section 95 support, who are being supported by TSOs is much higher than we might expect if levels of asylum support were adequate for meeting essential living needs. Data from RNGO show that the majority (53%) of people receiving support from this, the largest national charity supporting such individuals are asylum seekers; 25% have been granted some form of protection; and just 10% are refused asylum seekers with no further representations to make. In 2015, 61% of RNGO beneficiaries were in receipt of statutory support.