House of Commons - COMPASS: Provision of asylum accommodation

COMPASS: Provision of asylum accommodation - Public Accounts Committee Contents

1 Contract and relationship management

1. On the basis of a report by the Comptroller and Auditor General, we took evidence from the Home Office (the Department) and its three contractors, G4S, Serco and Clearel, on the provision of asylum accommodation, following the introduction of the new COMPASS contracts in March 2012.[1]

2. The Department provides accommodation for asylum seekers and their families who are assessed as being destitute, while their cases are being processed. As at April 2013, the Department provided accommodation for around 23,000 asylum seekers, although this number can fluctuate in response to world events. The cost of providing this accommodation in 2011-12 was £150 million.[2]

3. In March 2012 the Department (through the then UK Border Agency) signed six new contracts for the provision of asylum accommodation in England, Scotland, Wales and Northern Ireland for a five-year term with a possible two-year extension. This saw a shift away from 22 separate contracts, involving a mix of 13 different suppliers (consisting of private providers, local authorities and the voluntary sector), to six regional contracts provided by three larger contractors—G4S, Serco and Clearel. However this approach is at variance with wider Government effortsto increase the use of small and medium size enterprise (SMEs) and to reduce dependency on a handful of larger suppliers.[3]

4. The Department told us that alongside securing savings through economies of scale and improving the standard of accommodation,it had sought to transfer risk and responsibilities to contractors. However, the Department accepted that in seeking to pass more risks over to contractors, it has created new risks that could impact poorly on what is a vulnerable group of service users. In particular, by opting to contract with fewer, larger contractors,the Department reduced the diversity of provision, stripped away expertise and now has fewer options availableshould a contractor fail.[4]

5. The Department told us that it had required contractors to minimise the disruption to service users as the contracts changed hands, for example, by ensuring that as few asylum seekers as possible were moved. About 19,000 asylum seekers and their families were affected by the transition between contracts, of whom 90%(approximately 17,000 people) stayedin their existing accommodation and about 2,300 had to move to new properties.[5]It was vital that the Department helped to broker the handover between outgoing and incoming contractors and with local authorities, especially given that Serco and G4S had no experience of providing accommodation for asylum seekers.[6] However the Department adopted a hands-off approach. For example, it told us it had not completed a survey on the quality of housing stock run by outgoing contractors, and it did not provide the incoming contractors with liaison points with the outgoing contractors to ease the handover.The Department acknowledged that the imminent expiry of the previous contracts, called Target contracts, led to it rushing the mobilisation and transition stages of COMPASS, which brought about additional challenges to already complex contracts.[7]

6. The incoming contractors told us that when they bid for these contracts they had intended to inspect the properties before taking them over from the outgoing providers;however, G4S and Serco both failed to do this.[8]G4S acknowledged this was a mistake on their part and that they should have asked for more time. Serco, however, maintained that this failure was not wholly its fault, and was due in part to a lack of access to information on housing quality during the tendering process. Both contractors stated that they had not had the time to inspect all properties. They told us that, had they done so, they would not havetaken on all the properties from the former providers because the accommodation did not meet minimum standards. G4S and Serco both had to invest in raising the standards of accommodation they inherited from the outgoing contractors—G4S claimed to have spent £1.9 million and Serco £572,000 to renovate properties before they became fully operational.[9]

7. G4S and Serco both said that had they asked for more time, that would have delayed the transition and it could have resulted in many more asylum seekers being moved, which the Department had been seeking to avoid.[10] Serco maintained that its failure to inspect the properties it was taking over was because:of the lack of time available;it did not have rights of entry to inspect properties; and it did not have access to the outgoing contractors' exit plans. It rejected the suggestion that the reason had been the cost of undertaking the inspections.[11] While the incoming contractors should have carried out their own checks as part of their due diligence process, the Departmenthad notensured that information from the outgoing contractors was requested and shared with their replacements. The Department told us that it was looking at how it could improve arrangements when contracts ended so that in the future outgoing suppliers assisted with a smooth handover at the end of their contracts.[12]

8. All three contractors told us that assumptions, on which their bids for the COMPASS contracts were formed, were based on data provided by the Department at the outset of the tendering process. The quality of this data has subsequently attracted much criticism from contractors.[13] They claimed that the inaccuracies within this data had led to costly flaws in some of their key assumptions and that this, in turn, had resulted in some of the difficulties they have subsequently faced in running the service.[14]Serco alleged that itsfailure to manage the transition effectively was due to flawed assumptions based in part on its ownmarket research, but also based on inaccurate information contained within the Department's tender documentation.[15]

9. Each new property used to house asylum seekers must be approved by the relevant local authority, and an example of inaccurate information related to the ease and speed with which this would occur. Serco told us that the Department's bid documentation had stated that it would take 48 hours for a local authority to give approval. However, Serco said that it had found that in 48% of cases in 2013 it took local authorities more than 48 hours to approve, and in those cases it took an average of 11 days.[16] Clearel, the only current contractor to have worked in this sector before, told us that in its experience approval from local authorities was not always provided within 48 hours, and sometimes it could take much longer. Clearel had therefore not relied on a particular timescale when preparing its bid.[17]

10. Restrictions apply on where asylum seekers can be housed, and agreement must be sought from local authorities up to a defined 'cluster limit' due to social cohesion and resourcing pressures, such as on school places. This limit is generally defined as no more than one asylum seeker per 200 residents, although there is some variation.[18]Contractors made assumptions about how easy it would be to find accommodation given the cluster limits applied by local authorities; Serco told us that it had not realised how difficult it would be to find suitable properties. During the early months of the contract Serco did not appear to understand well enough the challenges local authorities faced in managing the impact of overconcentration of asylum seekers on social cohesion.[19]G4S and Sercoalleged that their flawed assumptions and lack of transparency with data contributed to the delays in them becoming fully operational.[20]The Department accepted that in the early phase of the contract it should have shared with contractors its data on the estimated flow of asylum seekers into the UK—information that, while difficult to predict, could have helped contractors better plan the amount of accommodation they might need—and it confirmed it was now doing so.[21]

11. After signing the contracts in March 2012, the contractors entered a three month mobilisation period up to May 2012, during which the Department expected the contractors to establish supply chains and get staff in place.Following the mobilisation period, the Department planned to have a four month transition period from June to September 2012, leading to the contracts being operationally ready in all areas by early October 2012. However, the Department extended the transition period from September to December 2012to ensure continuity of service when it became aware that the original timetable was overly ambitious because of the poor standards of accommodation, the limited availability of properties, and problems that suppliers were having establishing their supply chains.[22]

12. During the extended transition period (June to December 2012) the Department had not put in place an effective incentive regime which meant the contractors were under no financial pressure to become operational on time and deliver the standards of accommodation required. The Department told us it had not applied penalties for failing to meet key performance indicators during those seven months, only invoking penalties post transition (from January 2013). The Department was unable to explain to us its reasons for not applying penalties during the transition.[23]

13. Clearel, the only pre-existing contractor, was the only COMPASS contractor to complete transition on time, becoming fully operational across both of its new contracts from late September 2012. G4S and Serco both took a further three months to become fully operational (late December 2012) and acknowledged that the contract was more complex than they had anticipated at the outset.[24]Despite the delays, transition to COMPASS was completed before the extensions the Department had negotiated with the outgoing contractors expired.[25]

1 C&AG's Report, COMPASS contracts for the provision of accommodation for asylum seekers, Session 2013-14, HC 880, 10 January 2014 Back

2 C&AG's Report, paragraphs 1, 1.3 and 1.14 Back

3 Qq105-106,181; C&AG's Report, paragraphs 1.11-1.12 Back

4 Qq105-106, 109-110, 125, 181 Back

5 Q127; C&AG's Report, paragraphs 2.16 and 2.18 Back

6 Qq1-7 Back

7 Qq127, 147, 149, 157-8 Back

8 Qq61, 156;C&AG's Report, paragraph 2.15 Back

9 Qq58, 68 Back

10 Qq36, 69-72 Back

11 Qq59, 68, 69; C&AG's Report, paragraph 2.10 Back

12 Qq147, 169 Back