16.The National Audit Office highlighted that in 2015, HM Revenue & Customs (HMRC) and Concentrix handled the renewals process differently to the 2016 renewals process (discussed in paras 7 to 8 of this report). In 2015, following a request from Concentrix to delay the terminations of 49,000 tax credits awards to claimants who did not renew their claim by the annual deadline of 31 July, HMRC stopped these payments in batches over a six-week period. In contrast, during the 2016 renewals process HMRC stopped 45,000 payments to claimants in one week in August 2016.
17.Concentrix told us that it had expected that HMRC would once again stagger the terminations of awards in 2016 over a longer time period. It believed the situation would have been different if HMRC had done so, and would thereby have helped Concentrix to handle the high volumes of calls; in August 2016 Concentrix answered just 35% of calls in five minutes against a target of 90%. HMRC disagreed with this view, however, suggesting that staggering the terminations would have only made a “marginal difference”, as it was only one of a number of issues that contributed to the fall in performance in August 2016.
18.We asked HMRC about the reasons for stopping payments more immediately in 2016 than it did in 2015. HMRC explained that its normal practice is to stop payments soon after the 31 July deadline to avoid claimants building up overpayments that they must repay. It said that there was no reason for Concentrix to think it would differ from its normal process. HMRC agreed to stop awards over a longer period in 2015 as a “one-off exception” following Concentrix’s request.
19.HMRC told us that there were clearly issues of communication about the way the termination of awards in 2016 were managed. HMRC said that it would have looked at staggering terminations in 2016 had Concentrix asked it to do so, but Concentrix did not ask it to do anything different to its normal practice. Concentrix, however, explained that HMRC’s decision to terminate awards in one week was a surprise to them, with no prior discussion about the planned approach in the daily, weekly and monthly calls with HMRC. Concentrix said it had expected HMRC to instigate discussions about managing the termination of awards.
20.We also asked HMRC what other lessons it had drawn about communication in managing the contract. It said that it needed to test contingency arrangements more thoroughly to ensure that they are robust to circumstances that might arise. It also highlighted that there are lessons about how issues are escalated through the organisation to senior decision-makers who can resolve them. HMRC explained that it had started to address this particular lesson through a seminar with its senior staff to reinforce the importance of escalating issues in contracts and other areas of its own administration.
21.HMRC used a payment-by-results model to pay Concentrix commission as a percentage of savings from reducing error and fraud in tax credit claims. HMRC calculated savings based on the value of both recovering incorrect payments already made and preventing future overpayments to claimants. Concentrix would not be paid for any cases where its decision was overturned. The Social Security Advisory Committee noted in its July 2016 report, however, that there was an incentive for Concentrix staff not to overturn decisions given it would impact negatively on its revenue.
22.We asked HMRC how a payment-by-results model could have been made to work in this type of contract. HMRC agreed that it got the balance wrong between focusing on savings and thinking about claimants. HMRC’s pilot had identified issues with call-handling, as the provider answered just 14,333 (or 68%) or the 21,114 calls received during the pilot. The pilot focused on whether the private sector was capable of providing additional capacity to conduct tax credits compliance checks. HMRC said that it does not intend to involve the private sector in a full payment-by-results contract for compliance activity again.
23.Concentrix also said the contract was not designed in the right way to deliver what was required, raising concerns about commercial viability of the contract from the outset. HMRC told us that the contract was not generating the revenue that Concentrix had expected, or the performance levels that HMRC had intended. To try and address the situation, HMRC and Concentrix renegotiated the contract in October 2015. This included a higher rate of commission, from the basic level of 3.9% to 11% of savings made and different performance measures and monitoring arrangements. HMRC explained that it renegotiated the contract to make it work financially for Concentrix and to improve quality and customer service levels.
24.HMRC acknowledged that its commercial function was not fit for purpose to be able to avoid the mistakes that it made in contract letting. It highlighted that “HMRC has struggled over a number of years to get the kind of commercial function it needs”. It told us that it went to the market “to get a commercial partner to run the commercial organisation”. HMRC said that it is now re-evaluating its commercial and contract management function to make sure it has the required skills and resources in the future.
25 , paras 3.30–3.31
26 ; , para 3.14
30 , para 2.19
32 ; , para 1.16
33 ; , paras 3.3, 3.11
4 April 2017