Helping pensioners: evaluation of the Income Support pilots
A hard copy of this report summary can be obtained by contacting Paul Noakes [E-Mail: Paul.Noakes@dwp.gsi.gov.uk] or by writing to him at the 'Social Research Division, Department for Work and Pensions, 4th Floor, Adelphi, 1-11 John Adam Street, London WC2N 6HT'.
Research Report No. 105
By Nicola Croden, Paddy Costigan and Grahame Whitfield
This report presents findings from the pensioner Income Support (IS) pilot exercise conducted by the DSS, and the follow-up survey undertaken by the National Centre for Social Research. The pilot was designed with two main objectives; to examine the most effective and cost-efficient ways of identifying IS Entitled Non Recipients (ENRs); and, having identified them, to examine the most effective and efficient ways of encouraging them to claim.
The main findings were:
- Almost one quarter (23 per cent) of pensioners in the pilot were entitled to IS but not claiming it (IS ENRs). This figure greatly differed across the three data-matches; in the Housing Benefit (HB) sample 37 per cent were IS ENRs, in the Attendance Allowance (AA) sample 33 per cent and in the Retirement Pension (RP) sample 11 per cent.
- Overall five per cent of pilot cases made a successful claim for IS. This varied across the three data sources. In the HB sample 16 per cent of pilot cases made a successful claim, in the AA sample nine per cent and in the RP sample two per cent.
- There was wide variation in the proportion of pensioners responding to the different interventions (two postal methods, home visit and telephone). This can be explained largely by the difference in the nature of the methods. However there was no real difference in the number of successful claims made across the four interventions. Though pensioner ENRs expressed a preference for home visits, it was the long postal form that was the most successful at generating claims from ENRs.
- Over one-third (34 per cent) of those assigned to the telephone intervention either did not have a telephone number, or did not have a traceable telephone number. Identifying a large amount of non-coverage using this method.
- In the pilot 32 per cent of all pilot cases did not respond to the pilot exercise, and a further 15 per cent refused to make a claim.
- When advised of their entitlement during their survey interview; 40 per cent (two in five) said they would 'definitely' make a claim; 18 per cent (one in five) said they would 'probably' claim; and 26 per cent said they would not make a claim.
Part One - The pilot
Introduction
This report presents the findings of the pilot exercise which was set up by the Department of Social Security in May 1998 as a response to the Government's manifesto commitment to poorer older people. The pilot was designed to test whether Departmental records could be used to identify pensioners who appeared to be entitled to Income Support (IS), but not claiming it. As well as identifying likely recipients, different approaches were tested to see if any were effective in encouraging them to submit claims.
The cases were taken from nine areas chosen on the basis of geographical spread; available resources in the local BA office; and the willingness of the local authority to co-operate. The cases which were included in the pilot were identified by three types of data matching, using National Insurance Retirement Pensions Records (RP), Attendance Allowance Records (AA) and Housing Benefit Records (HB).
People were contacted by means of one of four intervention methods which aimed to identify IS Entitled Non-Recipients (ENRs) and encourage them to make a claim for IS. Two interventions were postal methods; one posting a short ‘screening’ form for IS, the other a full claim form. The other intervention methods were a telephone call, and a home visit.
Each pilot case, and the progress after the appropriate intervention method had been taken, was monitored by the BA and recorded on individual control sheets. The monitoring process began at the same time as the pilot exercise – May 1998 – and took one year to complete. Of primary interest was the ultimate outcome of each case, according to sample source and intervention. This outcome analysis is discussed in detail in part one of this report.
The pilot was designed with two main objectives;
- To examine the most effective and cost efficient ways of identifying IS ENRs; and
- Having identified them, to examine the most effective and efficient ways of encouraging them to claim.
In order to fully evaluate the proportion of IS ENRs picked up by the data-matching, assessments of entitlement needed to be carried out on as many pilot cases as possible. It was also important to collect information on the pensioners reactions to the interventions received. The “National Centre for Social Research” (formerly SCPR) were commissioned to carry out survey based research to inform this part of the evaluation. The findings of this survey are detailed in Part Two of this report.
Identifying Entitled Non-Recipients (ENRs)
The evaluation assessed the relative success of both the data matching and the intervention methods used in the pilot. Three-quarters of the cases drawn from the original data match were found to be eligible for the pilot. These cases were; of pensionable age; living within the boundaries of one of the pilot areas; not living within a residential home; not currently in receipt of IS and (if within the telephone intervention) have a traceable telephone number.
Success of the data matching
The success of the data matching exercise was determined by the proportion of ENRs identified within each sample. The survey investigated this in further detail and found that, overall, almost a quarter (23 per cent) of the cases were ENRs. The HB sample was the most successful at identifying ENRs; 37 per cent compared with 33 per cent of the AA sample and only 11 per cent of the RP sample.
Success of the interventions
There was wide variation in the proportion of pensioners responding to the different interventions, largely due to the nature of the methods. The postal methods required an active response, whilst the telephone and visit methods gained a response if the pensioner was home (at the notified time) when the BA officer called.
Despite this, there was no real difference in the number of successful claims made across the four interventions. Overall, five per cent of pilot cases made a successful claim for IS, this figure was mirrored across the Long form, Telephone and Visit interventions, whilst the Short form produced three per cent successful claims.
Over one third (34 per cent) of those allocated to the Telephone intervention did not have a telephone, or did not have a telephone number which was traceable. This intervention method therefore could not be used on one third of all of its cases. It was important to allocate people to this intervention prior to knowing the proportion without available telephone numbers/ telephones to test the effectiveness of the intervention.
Overall pilot outcomes
Of those assessed as not entitled to IS (39 per cent), most were ineligible due to their income or level of savings. They were younger on average than the overall pilot cases, and were more likely to come from the RP sample. Ten per cent of cases identified themselves as not entitled to IS. These pensioners were older on average than the overall pilot cases.
Of those who refused to participate in the pilot, cases were more likely to come from the Visit methodology. They were also slightly older (on average) than the overall pilot cases, and came mainly from the RP or AA samples.
Those who did not respond to the pilot intervention made up a sizeable proportion of the pilot cases (32 per cent), these cases were more likely to be older, and male pensioners. Those from the HB sample were slightly more likely to respond than both the AA and RP cases.
Five per cent of the overall pilot eligible cases made a successful claim for IS, but there was wide variation between sample sources. The HB sample achieved the highest proportion of successful claimants (16 per cent). Nine per cent of AA pilot cases made a successful claim while only two per cent of the RP cases did so. The RP sample contained younger pensioners, not currently claiming benefits. Successful claimants were on average older than the overall pilot cases and were slightly more likely to be women.
Effect on take-up rate
At the same time that the cases were assigned to one of the four intervention groups, a fifth group were assigned to a control group. The distribution across sample sources was comparable to that of the pilot cases. Two per cent of these cases were recorded as having made a successful claim, compared to five per cent within the pilot exercise. Pilot case successful claims were on average entitled to £18 IS per week, compared to £28.54 in the control cases.
The pilot intervention had an effect in increasing take-up of IS. However, as the survey data shows, the interventions also generated a substantial proportion of claims from non eligible or not entitled people.
The survey findings provide information about the characteristics and background details of both those who did and did not take part in the pilot.
Part Two – The Survey
The survey
Part Two of this report presents the findings of a survey conducted by the National Centre for Social Research as part of the evaluation of the DSS and the BA Income Support pilot exercise, conducted in 1998(1).
The survey involved 2,578 face-to-face interviews with a sample drawn from the population selected for the pilot exercise (of which 2,159 cases are used in this analysis). It is important to note that the survey sample was not intended to be representative of the general population of pensioners, nor of any special sub-group such as the population of potential ENRs. It was designed to be representative of people in the pilot areas who had been selected for the pilot exercise from the three data matching strategies (AA, HB and RP records).
Information from the survey
The survey provides information not available from the pilot itself, as the sample includes people selected for the pilot, but who refused to or who did not respond to the pilot interventions. As some of these people were classified as ENRs during the survey-based IS assessment, the survey is able to provide a fuller picture of the relative success of the data-matching strategies and the pilot intervention methods.
Information is also provided on the reasons people classified as ENRs gave for not having made a claim for IS after the pilot intervention, as well as on a wide range of questions which explored attitudes towards claiming IS and about peoples’ preferred methods of contact with the BA.
The survey-based assessment of IS eligibility and entitlement
The general rules regarding “eligibility” for and “entitlement” to IS are set out in Section 1.2 of Part One of this report. The survey interview included detailed questions about work, income, savings, and expenditure in order to conduct a simulated assessment which replicated (as far as possible) the BA assessment of benefit units’ eligibility for and entitlement to IS(2).
The survey sample consisted of individual adults selected from the pilot exercise. As IS entitlement depends on the circumstances of the benefit unit to which a person belongs, all survey respondents were thus asked questions about the circumstances of any spouse/partner or dependants (where applicable) in order to produce an accurate assessment of IS eligibility and entitlement(3). In Part Two of this report, survey respondents are used as the representative of their benefit unit, as it would not be feasible to talk about ‘benefit unit characteristics’ in terms of a mix of age, sex, and other factors.
The populations used in the analysis
In addition to identifying ENRs, the survey-based IS assessment identified two categories of “non-eligibility”, and a further category of “non-entitlement”:
- ‘work’: the respondent and/or their partner worked for more than 16 hours a week (not eligible);
- ‘savings’: the respondent and/or their partner had savings over £8,000 (not eligible); and
- ‘income’: the benefit unit was eligible for IS, but had a level of income which resulted in them being classified as not entitled to IS.
All findings in Part Two of this report detailing whether people are classified as ENRs, or as not eligible for/entitled to IS, use the survey-based IS assessment. Although some mismatches with the BA assessments (‘pilot outcomes’) were found - these most likely caused by a combination of mis-reporting to the survey interviewer and/or BA assessors – for the great majority (85 per cent) of cases where outcomes could be compared, these concurred.
The data-match: identifying Entitled Non-Recipients
The relative success of the sample sources in identifying ENRs
HB and AA records were similarly successful in identifying ENRs (37 per cent and 33 per cent of people drawn from these sources were ENRs). RP records were less successful, with only one in ten (11 per cent) people from this source found to be ENRs.
Reflecting that all three sample sources were more successful at identifying female ENRs than male ENRs, and that the pilot population contained many more women than men, female ENRs represented a greater proportion of ENRs drawn from all sample sources. As already noted, RP records were much less successful than AA or HB records at identifying all ENRs .
Older people in the survey sample were more likely than younger people to classified as ENRs, this irrespective of sample source, this increase with age being most marked in respect of the sample drawn from HB records. However, reflecting the age profiles of the different sample sources – where people drawn from AA records were on average the oldest - ENRs drawn from AA records were older on average than those from either HB or RP records. Half (53 per cent) of ENRs drawn from AA records were aged 80 years or over, compared with smaller proportions of those drawn from HB 43 per cent) or RP records (33 per cent).
Single people represented a greater proportion of ENRs than married people, irrespective of which sample source they were drawn from. The difference was most marked among ENRs drawn from AA or HB records (respectively, 82 and 79 per cent of ENRs drawn from these sources were single).
Level of estimated entitlement for IS
On the basis of information provided during the survey-based IS assessment, ENRs drawn from AA records were estimated as having higher average entitlement to IS (£31.92 per week) than those from either RP (£23.50) or HB records (£17.91).
This reflects that about half of ENRs drawn from HB (55 per cent) or RP records (49 per cent) were estimated as being entitled to less than £10 per week, compared with only one-fifth (17 per cent) of ENRs drawn from AA records. In contrast, six in ten (61 per cent) ENRs drawn from AA records were estimated as being entitled to over £25 per week.
Reasons why people were classified as not eligible/entitled
The reasons people were classified as being not eligible for/entitled to IS differed across the three sample sources.
People drawn from AA records were more likely than those drawn from HB or RP records to be classified as ‘ineligible due to savings’ (45 per cent, compared with 30 per cent and 11 per cent respectively). People drawn from RP records were the most likely group to be classified as ‘ineligible through being in full-time work’ (11 per cent), although people in this group were still very likely (59 per cent) to be classified as ‘non-entitled due to income’. Similarly, the majority (86 per cent) of ‘non-eligible/entitled ‘people drawn from HB records were classified as not entitled on this basis.
The pilot intervention: encouraging ENRs to claim
Overall, one in seven (14 per cent) of the survey sample said they had gone on to make a claim for IS after the pilot intervention. The remaining 86 per cent had not claimed.
ENRs were “more likely” than those classified as not eligible/entitled to say they had made a claim for IS after the pilot intervention (23 per cent, compared with 11 per cent). However, a greater “volume” of claims was generated from people who were not eligible for/entitled to IS. This is reflected in that only two-fifths of reported claims (39 per cent) were from ENRs, whilst three fifths (61 per cent) were from people classified as not eligible for/entitled to IS.
Comparison of intervention types
The Long and Short Form interventions were most (and equally) successful in resulting in people making claims for IS after the pilot intervention (20 per cent of those contacted by each of these methods said they went on to claim). Home Visits were about half as successful (11 per cent), with the Telephone intervention being about half as successful again (five per cent)(4).
The Long Form was the most successful intervention in generating claims from ENRs. Almost four in ten (37 per cent) of all claims generated by this intervention were from ENRs. This compares with just over two in ten (24 per cent) in the case of the Short Form . These figures reflect that, whilst one third (35 per cent) and one fifth (22 per cent) of ENRs contacted using these methods (respectively) said they had claimed IS, about one-fifth of people contacted using these methods who were classified as not eligible for/entitled to IS had also gone on to make a claim (19 and 16 per cent respectively).
Due to such small numbers of people who were Visited or selected for the Telephone method making claims, the relative success of these interventions in generating claims from ENRs cannot be determined.
There was no difference in the average amount of IS that ENR claimants and non-claimants were estimated as being entitled to. As such, it can be concluded that the amount of entitlement is only one of many factors involved in the decision making process regarding claiming IS.
However, the Long Form generated claims from ENRs that were entitled to about one-third of the total amount of IS entitlement of all ENRs selected for that intervention. This compares with about one-tenth in respect of the other three intervention methods.
Claim outcomes
Of the 14 per cent of people who said they went on to claim IS after the pilot intervention, one-third (33 per cent) said their claim had been successful. Two in ten (18 per cent) said their claim was still ongoing, whilst the remainder (49 per cent) said their claim had been unsuccessful.
Seven in ten (69 per cent) ENRs who had made a claim for IS said it had been successful, whilst one in ten (11 per cent) said it was ongoing. The remaining 20 per cent said their claim had been unsuccessful. However, a small proportion of people classified as not eligible for/entitled to IS (11 per cent) said they had made a successful claim. These latter two figures are unexpected in that ENR claimants would have been successful, and that people who were classified as not eligible/entitled on the basis of the survey-based IS assessment could only have made unsuccessful claims. This situation is most likely related to some combination of mis-reporting to survey interviewers and/or BA assessors.
Attitudes to claiming Income Support
Although most ENRs stated that their “preferred” method of contact would be a Home Visit, the Long Form was in fact the most successful intervention method in terms of eliciting “claims” from ENRs. This strongly indicates that people’s stated preferences can be at variance with their actual behaviour.
The points that people liked most about the Long Form method were the ability to fill it out at their own pace, and the fact that many sections could be skipped. The main objection was to the length of the form. However, people felt somewhat less strongly about the Long Form’s complexity or the difficulty of supplying the information the Form asked for.
While the Home Visit was seen as the most convenient way of supplying the information needed for an IS claim, the Long Form achieved the second highest rating in this respect, ahead of the Telephone method.
When advised of their entitlement during the survey interview, two-fifths (40 per cent) of ENR non-claimants said they would ‘definitely’ make a claim. A further fifth (18 per cent) said they would ‘probably’ claim. Twenty-six per cent of ENR non-claimants said they would not make a claim.
Pensioners expressed quite strong preferences for having specialist advisors to deal with claims from pensioners, and for continuity in the people that they dealt with.
Comparing pilot and survey outcomes
The survey presents findings on the people who either refused to or did not respond to the pilot intervention, on those who were selected for the Telephone intervention but who could not be contacted using this method, and on those who ‘self-assessed’ themselves as being not eligible for/entitled to IS. Finally, it also compares the pilot outcomes with the survey-assessed IS status of survey respondents.
As might be expected, the data-match people were drawn from was not related to the likelihood of them refusing to or not responding to the pilot exercise. However, the intervention type was. Almost half (45 per cent) of people sent the Long Form and a third (33 per cent) of people sent the Short Form did not respond to these interventions. Also, the Short Form and Home Visit were more likely to have generated refusals to participate in the pilot exercise. Despite this level of ‘opt out’, however, it remains the case that the Long Form and Short Form interventions were relatively more successful than either the Visit or the Telephone methods.
Overall, ENRs and people not eligible for/entitled to IS were similarly likely to have refused to, not to have responded to the pilot intervention or not to have been contacted for the Telephone intervention.
Around three-quarters of people who refused to or who did not respond to the pilot were classified as not eligible for/entitled to IS, leaving about a quarter that were ENRs. A similar proportion of people without a phone/a traceable number were also ENRs. The great majority of people (88 per cent) who self-assessed themselves as not eligible/entitled were correct in this assessment according to the survey-based IS assessment.
In almost nine in ten (85 per cent) cases where outcomes could be compared, the survey-based IS assessment and the pilot outcomes recorded by BA concurred. That is, the majority (84 per cent) of people the BA classified as ENRs (i.e. who made successful claims) were also classified as ENRs on the basis of the survey-based IS assessment. A similar proportion (85 per cent) of people classified by BA as having made an unsuccessful claim for IS were also classified as not eligible for/not entitled to IS following the survey-based IS assessment. This level of agreement can only lend reassurance to the accuracy of the survey.
Relevant publications
Boaz, A., Hayden, C. and Bernard, M., (1999) “Attitudes and Aspirations of Older People: A review of the literature” (DSS Research Report No. 101) Leeds: CDS. (Price £34.00)
Costigan, P., Finch, H. Jackson, B., Legard, R. and Richie, J. (1999) “Overcoming Barriers: Older People and Income Support ”(DSS Research Report No. 100) Leeds: CDS. (Price £29.00)
Disney, R., Grundy E. and Johnson, P. (1998) “The dynamics of retirement” (DSS Research Report No. 72) London: The Stationery Office (Price £36.00)
Finch, H. and Elam, G. (1995) “Managing money in later life” (DSS Research Report No. 38) London: HMSO. (Price £22.00)
Flatley, J. (1999) “Helping Pensioners: Contextual Survey of the Income Support Pilot Areas ”(DSS In-house Research Report No. 60) London: DSS
Hayden, C., Boaz, A., Bernard, M. (1999) “Attitudes and Aspirations of Older People: A Qualitative Study” (DSS Research Report No. 102) Leeds: CDS (Price £29.00)
Legrad, R., Ritchie, J. and Finch, H. “Older People and Income Support: Barriers and Triggers to the take-up of Income Support ”(DSS Research Summary No 6) London: DSS.
Shaw, A., Walker, R., Ashworth, K., Jenkins, S. and Middleton, S. “Moving off Income Support: Barriers and Bridges ”(DSS Research Report No. 53) Leeds: CDS (Price £26.00)