Can the Work Programme work for all user groups? - Work and Pensions Committee Contents

2  The implications of lower than expected job outcome performance

15.  Official Work Programme job outcome performance data, covering the period from June 2011 to the end of July 2012, were published for the first time in November 2012; some 17 months after delivery began in June 2011. DWP had planned to publish official job outcome data, together with referral data, on a six-monthly basis thereafter. The Minister informed us that lack of data was an issue for him: "I do wish that we had more data from the outset. That is something that we are tackling now [...] I am very keen that we use data to talk about the benefits of the scheme, and also to enable us to manage the scheme and drive through continued performance improvements."[5] DWP informed us, after we had finishing taking evidence, that it now plans to publish the next set of data on 27 June 2013 and on a quarterly basis thereafter.[6]

16.  As was much publicised at the time, the November 2012 data showed that job outcome performance in the first 12-14 months of delivery was significantly lower than had been anticipated by DWP. This chapter examines the reasons for low performance in year one of the programme; the implications for subsequent years of the contracts; and the options open to DWP in the event of continuing poor performance.

Job outcome performance against Minimum Performance Levels

17.  Job outcome performance is measured as the percentage of participants referred to the Work Programme who achieve a 13 or 26 week job outcome in a 12-month period. Participants can achieve a job outcome in a single job or in a series of short-term jobs which together satisfy the definition.

18.  DWP imposed contractual job outcome Minimum Performance Levels (MPLs) on primes for three of the nine payment groups in the Work Programme: Jobseekers Allowance (JSA) claimants aged 18-24 years old; JSA claimants aged 25 and over; and new Employment and Support Allowance (ESA) claimants. MPLs were not set for the remaining payment groups, which include ex-Incapacity Benefit (IB) claimants, Income Support claimants, JSA claimants with severe barriers to work and prison leavers. DWP set out indicative performance targets for these groups rather than a contractual MPL, as it was thought that job outcome performance amongst these groups was too unpredictable.[7]

19.  The contractual MPL for the three mainstream payment groups were set as follows:[8]

Table 2: Work Programme Minimum Performance Levels (MPLs)
Payment Group Job outcomeMPL year 1 MPL year 2MPL year 3 MPL year 4MPL year 5 MPL year 6MPL year 7
JSA 25+26 weeks 5.5%33% 33%33% 33%28% 6%
JSA 18-2426 weeks 5.5%44% 44%44% 44%33% 11%
ESA new claimants13 weeks 5.5%17% 17%17% 17%11% 6%

The first official release of job outcome data in November 2012 showed that: there were 31,000 job outcome payments paid to providers from 1 June 2011 to 31 July 2012, from 878,000 referrals, representing job outcome performance of 3.5% across all payment groups and job outcome performance of 3.2% in relation to the three mainstream groups during the first 14 months of the Work Programme. The overall performance is lower when calculated for the 12-month period from 1 June 2011 to 31 May 2012: 18,000 job outcome payments from 785,000 referrals, representing overall job outcome performance of 2.3% and job outcome performance of 2.1% in relation to the three mainstream groups.[9] Job outcome performance was therefore significantly below DWP's expectations. None of the 18 primes reached the contractual year one MPL of 5.5% in the first 12-14 months of delivery—they were all technically in breach of contract.

20.  As we highlighted in 2011, the MPLs were considered by experts to be extremely challenging. For example, Dave Simmonds, Chief Executive of the Centre for Economic and Social Inclusion (Inclusion) and a leading commentator on welfare-to-work programmes, told us in 2011 that the performance challenge for Work Programme providers was "very severe indeed".[10]

21.  Expert witnesses to this inquiry agreed that MPLs had been set at unrealistic levels. Ian Mulheirn of the Social Market Foundation was "baffled" by how DWP had calculated the performance targets. His view was that the MPLs set for the Work Programme were "macho" targets, designed to show that DWP was being tough on private sector providers. He believed, drawing on the analysis of the National Audit Office (NAO), that a large margin was added to the MPLs on the basis of "pure faith that the Work Programme was brilliantly designed". His view was that this was a "huge dose of wishful thinking". He noted that DWP's record on setting welfare-to-work performance targets was poor, arguing that targets set for previous programmes, such as the Flexible New Deal, had been "even more absurd".[11]

22.  A number of witnesses highlighted that the Work Programme MPLs had been set when forecasts for the economy were more positive.[12] Inclusion's analysis was that the MPLs should be revised down by around 15% to reflect the worse than expected economy. While overall employment levels have held up remarkably well in the economic downturn since 2008, recent research by Inclusion found a strong correlation between economic growth, measured in Gross Domestic Product (GDP), and the employment prospects of long-term jobseekers.[13] Tony Wilson of Inclusion argued that future Work Programme performance targets need to be more responsive to changing economic forecasts, and perhaps even recognise actual economic conditions retrospectively.[14]

23.  Sean Williams of G4S, a prime operating in three CPAs, believed that the Work Programme's failure to meet the contractual MPLs in year one of the contracts highlighted the inadequacy of DWP's targets, rather than a deficiency in the programme itself:

If the Work Programme is not measuring up against the minimum performance levels, there is clearly either a problem with the Work Programme or there is a problem with the minimum performance levels. I would suggest that the minimum performance levels are completely inadequate for the task that they have been set to do. They were set in completely different macroeconomic conditions and, even at the time, leading figures in the industry said that those minimum performance levels were just not realistic.[15]

24.  Beyond the worse than expected economic situation, a number of other reasons have been put forward for low performance in the early months of the Work Programme. The Work Programme was commissioned very quickly; delivery began just over a year after the announcement of the policy and within six months of the publication of the final Invitation to Tender. The NAO has highlighted that, in contrast, the procurement phase of the Flexible New Deal took some 15 months and that some previous welfare-to-work programmes were introduced over a four-year period.[16]

25.   It should also be noted that the Work Programme represents a wholesale restructuring of the sector; some 230 existing welfare-to-work contracts were either terminated or not renewed.[17] A number of primes had to set up operations in areas of the country in which they had never worked before. This involved moving offices and personnel and inevitably led to some delays in service delivery.[18]

26.  The Minister for Employment has acknowledged that Work Programme performance was poorer than expected, citing in mitigation the difficulties providers were facing in achieving sustained job outcomes for participants and problems in tracking, and claiming payment for, the outcomes which they did achieve. However, he has stated that the Government's "aspiration" for the job outcomes to be achieved in future years of the contracts "remains the same".[19]

27.  Primes told us that they still expected to achieve, or get very close to, the MPLs for the three mainstream payment groups in subsequent years of the contracts.[20] Kirsty McHugh, Chief Executive of the industry body, the Employment Related Services Association (ERSA), agreed that this was likely, although she noted that the unpredictability of economic circumstances made it uncertain.[21]

28.  We are concerned about the appropriateness of the Work Programme Minimum Performance Levels (MPLs) and how they were calculated by DWP. They do not appear to be sufficiently responsive to the actual economic conditions in which providers are operating. The lack of realistic MPLs results in realistic assessments of performance being difficult and makes sanctioning primes, and therefore delivering and incentivising improved performance, harder to achieve. For the next round of Work Programme contracts, we recommend that DWP devise, in collaboration with independent experts, a new method of calculating and setting MPLs which are more responsive to the state of the economy; can be more transparently calculated and applied; and can be reviewed periodically during delivery.

29.  On the day before the release of the official job outcome statistics, ERSA published its own unofficial and unverified Work Programme performance data, based on internal information collected from each of the 18 primes. ERSA's data related to "job starts"—the number of Work Programme participants who had achieved at least one day in work—in the 16-month period from June 2011 to the end of September 2012. ERSA's unofficial data showed that 207,381 Work Programme participants had achieved a job during the period. Although there are no official referral data covering the entire period, ERSA's data suggest that around 20% of participants referred to the programme had achieved at least one day in work by the end of September 2012. The data are broken down to show the percentage of each monthly cohort of referrals which had achieved a job start by the end of the period. Some 29% of participants referred to the Work Programme in June 2011, the first month of delivery, had achieved a job start by the end of September 2012. ERSA estimates that 65-85% of job starts will convert into sustained 13 or 26-week job outcomes. Some witnesses were cautious about the reliability of unverified data produced by the industry, but ERSA believes that its figures indicate a reasonably positive picture of performance "building in the pipeline", at least within the mainstream JSA payment groups.[22]

30.  Our scrutiny of the Work Programme in the early months of service delivery has been hampered by a lack of transparent official data. Official job outcome data were not made available until some 17 months after service delivery began. After many months of being unable to provide figures, because DWP claimed they would be unverified and therefore unreliable, the publication of the first official figures was accompanied by the publication of unofficial figures by the Employment Related Services Association (ERSA). Although unofficial, these were frequently referred to by Ministers as an alternative to the official figures. Such an approach to statistics is unhelpful.

31.  We were concerned that DWP had planned to release official Work Programme referral and job outcome data on a six-monthly basis, which we believed would have been inadequate for a programme of this size and importance. We therefore welcome DWP's recent decision to move from six-monthly to quarterly reporting of Work Programme referral and performance data.

32.  At the beginning of May 2013, after we had finished taking evidence, the UK Statistics Authority published a report on official statistics relating to DWP employment programmes, including statistical releases on the range of pre-Work Programme provision and the official Work Programme job outcome data published in November 2012. The main findings of the report were that:

  • There may be scope to extend the published statistics to allow comparisons by claimant and provider characteristics
  • The presentation of the official statistics on job outcomes in the November 2012 Work Programme statistical release "could have been clearer"
  • There is scope for greater coherence across statistical releases and research
  • There is scope to improve compliance with the Code of Practice for Official Statistics, including by reviewing how statistical releases meet the needs of users.[23]

33.  The UK Statistics Authority has recently made a number of recommendations about improvements it would like to see DWP make to the statistics it publishes on the Work Programme and pre-Work Programme employment provision. We support the Authority's recommendations. We accept that it may not be possible for DWP to take all of the recommendations into account in publishing the next tranche of official Work Programme statistics in June 2013. However, we expect the clarity, interpretation and usefulness of subsequent releases to improve in response to the Authority's report and our own recommendations.

Options open to DWP in the event of continuing poor performance

Market share shift

34.  The official data released in November 2012 showed that primes' job outcome performance in the first 14 months of delivery varied between 2.2% and 5%.[24] Currently primes operating in the same CPA receive an equal number of randomly referred jobseekers. Where there are significant differences in job outcome performance between primes in the same CPA, DWP has the option to shift 5% of new referrals from the worst performing prime to the best performing. The trigger for a potential shift of referrals is a 3 percentage points or greater difference in job outcome performance. The market share shift can be applied from two years into the programme (i.e. June 2013) and subsequently every 12 months.[25]

35.  Sean Williams of G4S, a relatively highly-performing prime, argued that shifting market share would bring about swift performance improvements, as G4S had demonstrated in relation to its own subcontractors. He believed that the market share shift mechanism should be implemented by DWP as soon and as "boldly" as possible.[26]

36.  However, expert witnesses urged caution in the application of market share shift. Tony Wilson of Inclusion believed, based on international evidence, that it was likely to bring about short-term improvements in performance but that there were also a "huge host of risks". Principal among these were risks associated with moving towards monopoly provision, which could damage competition and service innovation in the longer term. Ian Mulheirn also highlighted that shifting market share would result in a diminishing proportion of referrals going to the relatively poorly performing provider, which could further impact on that provider's performance. His view was that this would "condemn" some jobseekers to an underperforming provider faced with an increasingly severe performance challenge and reduced funding.[27]

37.  Any shift of market share will be applied on the basis of job outcome performance at prime level. Subcontractors expressed concern about how shifting market share might impact on them. Wheatsheaf Trust, a subcontractor operating in Hampshire, told us that DWP had not made clear whether subcontractors—which could be performing relatively well despite being attached to a poorly performing prime—will be protected by DWP from potential adverse impacts of actions taken against primes for poor performance.[28]

Termination of prime contracts

38.  The then Minister for Employment told the Committee in 2011 that primes would not be allowed to "just sink down and cruise along at a very low level. They will be in breach of contract then and we will remove them."[29] The current Minister reiterated this point, stating that he was "absolutely serious" about contract termination. He told us that DWP had "looked at this scenario" and was confident that, if a prime was consistently underperforming in breach of contract, it was feasible for its contract to be terminated and for a replacement prime to take over its caseload of jobseekers.[30]

39.  We believe that shifting market share from the lowest performing to the highest performing prime in the same Contract Package Area could boost provider performance and should be implemented during 2013. However, the "market share shift" mechanism will need to be actively, carefully and transparently managed and applied by DWP. We recommend that, in response to this Report, DWP provide further information about the scenario modelling it has undertaken to assess the likely impacts on provider performance and service quality of shifting market share between primes. We also remain unconvinced that the early termination of a prime contract could be achieved without significant disruption to services and request further details about the research DWP has conducted to assess the feasibility of contract termination and its impacts on service delivery.

40.  We are concerned that relatively highly-performing subcontractors may suffer as an unintended consequence of market share shift between primes or in the event of prime contract termination. We recommend that, in its response to this Report, DWP make clear how it will protect highly-performing subcontractors attached to poorly-performing primes in the event that the latter lose market share or have their contract terminated.

The implications of lower than anticipated job outcome payments

41.  Expert witnesses were concerned that, in a largely outcome-based payment model, low job outcome performance in the early months of the programme would impact on investment in future services. Ian Mulheirn believed that, with limited up-front funding, providers would inevitably reduce investment in resources at the frontline. He believed that funding was being reduced at the wrong time and that more funding, not less, was required during a period of low economic growth and relatively high unemployment.[31]

42.  It should be noted that funding for the Work Programme will be further reduced by the planned withdrawal of attachment fees from April 2014. Professor Roy Sainsbury, of the Social Policy Research Unit at the University of York and co-author of the official DWP evaluation of the Work Programme, argued that attachment fees had thus far been a "buffer" that had "kept the whole thing going". His view was that DWP should carefully consider the implications of the withdrawal of attachment fees.[32]

43.  Funding within the Work Programme has been considerably diminished by lower than anticipated levels of job outcome payments to providers in the first 14 months of delivery. Attachment fees are due to end in April 2014. We recommend that DWP review the balance between attachment fees and outcome fees, and consider retaining attachment fees, to protect delivery of services, including by subcontractors.

Provision for unsuccessful participants

44.  Even if primes meet their contractual targets, a large proportion of Work Programme participants will reach the end of their two-year attachment period without finding sustained work. Primes accepted that the proportion was likely to be 60-70% of participants.[33] The first cohorts of participants will begin to come to the end of their attachment to the programme from June 2013. Some witnesses noted with concern that DWP did not appear to have plans in place to further support them.[34] JCP staff in Willesden were aware that DWP was formulating arrangements but were not aware of what form further support might take.

45.  We wanted to know what plans the Government had put in place to support unsuccessful participants. The Minister told us that DWP had piloted separate options for those who had made significant progress towards sustained employment—who had undertaken work experience or started a job which had not been sustained, for example—and for those who were still some way from the labour market. Those who had made progress would be supported in their job-searching by JCP and possibly undertake a "community activity placement". JCP would provide "more intensive" support for those with greater barriers to employment who were still some distance from the labour market.[35]

46.  The first cohort of Work Programme participants will reach the end of their attachment period from June 2013. DWP must set out as a matter of urgency the support that will be in place for participants who are unsuccessful in finding sustained work during their two years on the Programme. We recommend that all unsuccessful participants should have an end of Work Programme assessment. Specialist support must be put in place for those who have made little progress. Those who have made significant progress towards sustained employment should be permitted to voluntarily extend their attachment to the Work Programme. The period in which providers are able to claim job outcome fees in relation to these participants should likewise be extended. We also recommend that DWP consider the practicalities of assigning participants who have completed the programme without success, but are close to work, to one of the other primes covering their area.

5   Q 554 Back

6   Ev 123  Back

7   Committee's 2011 Report, para 109 Back

8   See Committee's 2011 Report, chapter 5 Back

9   DWP, Work Programme Statistical Release, November 2012, p 1; see also Centre for Social and Economic Inclusion, Work Programme performance statistics: Inclusion analysis, November 2012, p 1 Back

10   Committee's 2011 Report, para 100 Back

11   Q 5; See also, National Audit Office, The Introduction of the Work Programme, HC 1701, January 2012 Back

12   See, for example, Q 5 [Ian Mulheirn]; Wheatsheaf Trust, Ev 164 Back

13   Centre for Economic and Social Inclusion, Long-term unemployment in 2012, November 2012 Back

14   Q 37 Back

15   Q 273 Back

16   National Audit Office, The Introduction of the Work Programme, HC 1701, January 2012, p 6; para 2.2 Back

17   Ibid., para 2.12 Back

18   Ev 128 Back

19   HC Deb, 3 December 2012, col 652W Back

20   Q 277 [Sean Williams]; Q 278 [Richard Clifton] Back

21   Q 270 Back

22   Q 43 [Ian Mulheirn]; Employment Related Services Association (ERSA), Job start data, November 2012, pp 2 and 9 Back

23   UK Statistics Authority, Statistics relating to DWP Work Programme and pre-Work Programme, May 2013, Monitoring Review 2/13, pp 2-3 Back

24   Centre for Economic and Social Inclusion, Work Programme performance statistics: Inclusion analysis, November 2012, table 1 Back

25   Committee's 2011 Report, para 56 Back

26   Q 295 Back

27   Q 39 Back

28   Ev 163 Back

29   Committee's 2011 Report, paras 56-57 Back

30   Q 472 Back

31   Q 38 Back

32   Q 72 Back

33   Q 297 Back

34   Q 53 [Tony Wilson]; Q 302 [Kirsty McHugh] Back

35   Qq 485-487 Back

previous page contents next page

© Parliamentary copyright 2013
Prepared 21 May 2013