We are testing a new beta website for gov.scot go to new site

Evaluation of the National Mortgage to Rent Scheme

Listen

CHAPTER FOUR SCALE OF ACTIVITY AND NEED

4.1 This chapter considers the numbers of cases applying to the MTR scheme and what has happened to them, over the 5-year life of the scheme, drawing mainly on the central administrative data. It goes on to consider the wider potential need for the scheme, by considering a range of evidence from secondary data sources on homeowners with financial problems and those actually losing their homes. Comparisons across time and space and for different demographic groups raise issues about targeting, knowledge and promotion of MTR, which are further discussed in the light of stakeholder comments.

Overall Numbers

4.2 Table 4.1 shows the number and percentage of MTR cases broken down in two ways, by the year in which they applied to the scheme and by their status as at March 2008. The scheme has been in operation for rather less than five years during which it has handled 1,481 applications. The number of applications in the first full year (2004) was only 164 but this rose to a peak of 405 in 2006, before dropping slightly to 381 in 2007. The figure for the early part of 2008 suggests that there may be a further increase in this year.

Table 4.1: MTR Cases by Year Applied and Current Status 13

Year Applied

Number

Percent

2003

119

8.0

2004

164

11.1

2005

286

19.3

2006

405

27.3

2007

381

25.7

2008 (Q1)

120

8.1

Total

1475

99.6

Missing info

6

0.4

1481

100.0

Status Cats

Settled

624

42.1

Ineligible

469

31.7

Withdrawn

182

12.3

In Process/unknown

206

13.9

Total

1481

100.0

Source: MTR and Allocation of Cases Databases

4.3 Stakeholders believed that demand appeared to have risen although possibly not in line with increasing mortgage arrears and repossessions in recent months. It was thought that higher levels of demand, indicating higher levels of need, may be identified if the scheme was more widely publicised.

4.4 At the time when the data were transferred there were 206 cases 'in process' or whose final status was not known. This is about a half of annual throughput, suggesting crudely that cases may take about 6 months to process (but see below for more detail on processing times). Of all cases taken on, 624 or 42% have been 'settled', meaning that in these cases the applicant has sold their house to a social landlord and become a tenant of that landlord with assistance from the scheme. Just under a third (31.7%) were ineligible, while a further 12.3% withdrew. However, if we discount the cases still in process, we get a fairer picture of the overall 'success rate' of applications. This turns out to be 48.9%, with 36.8% ineligible and 14.3% withdrawing.

4.5 Table 4.2 below looks at the pattern of status for each year of application. This shows a 'success' (i.e. settlement) rate of 54-57% for the first three years, dropping to about 47% in 2006. It is difficult to judge what the final rate will be for 2007, but the 2006 figure suggests some dropping off in the settlement rate, with more cases withdrawing. This trend to more withdrawals looks likely to continue through 2007.

Table 4 2: Current Status by Year Applied (percent for each year)

Status Cat

2003

2004

2005

2006

2007

2008

Total

Settled

57.1

57.3

54.2

45.7

31.8

42.2

Ineligible

31.1

29.9

38.1

37.8

26.2

17.5

31.8

Withdrawn

11.8

12.8

7.3

15.3

15.5

4.2

12.3

In Process/unknown

0.3

1.2

26.5

78.3

13.6

Total

100.0

100.0

100.0

100.0

100.0

100.0

100.0

Source: MTR and Allocation of Cases Databases

Geographical Patterns

4.6 The geographical pattern of take-up of the MTR scheme is interesting for several reasons. It is likely to reflect a mixture of influences, including housing market and population factors which affect the potential extent of problems with mortgage payments and repossessions, but also variations in the penetration of knowledge about the scheme among both the general public and key stakeholders, such as mortgage lenders, social landlords and advice centres. In addition, there may be cultural variations which affect people's willingness to consider this type of solution to their problems, for example willingness to become a social tenant.

4.7 Table 4.3 shows the distribution of applications across all of the local authorities, ranked in descending order in terms of the number of cases. Horizontal lines through the table indicate the 25 th percentile points of the distribution. It is remarkable to note that a quarter of all cases arise in just two local authorities, North Lanarkshire and Glasgow. It is also interesting to note that the local authority with the largest number of cases is North Lanarkshire and not the most populous authority, Glasgow, which is also generally regarded as the most deprived authority in Scotland. The second largest authority, Edinburgh, which arguably has the greatest 'affordability' problems, comes sixth in the list. South Lanarkshire, Fife and North Ayrshire come higher up the list. Other authorities with quite large number of cases relative to their size include Midlothian, Dumfries and Galloway, and East Lothian. At the other end of the scale, one might question why numbers of cases seem so small in such authorities as Inverclyde, Stirling, Angus, Moray or Clackmannanshire.

4.8 We go on to investigate takeup against various independent measures of potential need/demand. However, this table alone suggests that there are apparent anomalies, which may well relate to different levels of awareness of the scheme or willingness to use it. Table 4.3 also shows the percentage of mortgaged owners in each authority (from SHS) and the ratio of the MTR application share to this share. Ratios above 1.0 indicate areas with higher than expected takeup, based on this simple demographic criterion. Areas with a relatively high takeup on this basis include: North Ayrshire, North Lanarkshire, Midlothian, East Lothian, Dumfries and Galloway, West Lothian and South Lanarkshire. Glasgow, Dundee and Perth and Kinross are only slightly above the expected rate. Edinburgh is significantly below the expected share.

4.9 Table 4.4 provides a regional summary of the take-up, using an ad hoc definition of regions to try to capture the broad sectors of the country as well as the urban-rural distinction. This shows that Central Scotland, broadly defined, accounts for 84% of cases, while the rural extremities account for only 8% of cases.

4.10 Table 4.5 looks at the pattern of case status within each region. This shows some significant differences. Lothian has a lot more settled cases and less ineligible. Highlands and Islands (and North East to a lesser degree) have more ineligible cases and less withdrawn cases.

4.11 The qualitative research with stakeholders sheds some light on possible factors behind these variations. The association of MTR applications with wider problems of debt implies a societal problem but one which may be more prevalent in certain areas. The Accountant in Bankruptcy (AiB) reports that bankruptcy problems are concentrated in areas facing more economic difficulty. Funding of advice centres may be geographically constrained to socially deprived areas so this may explain some concentrations of referrals. More wealthy areas may not qualify for MTR on grounds of the ceiling on property values (or total subsidy). Stakeholders believed that property value limits were a key constraint on demand. A significant influence on the distribution of successful MTR applications may be the availability of willing social landlords, which is shown in Chapter 5 to be a significant reason for cases not proceeding. Landlord motivations and concerns are discussed fully in Chapter 8.

Table 4.3: Number and Percentage of MTR Applications and Mortgaged Owners by Local Authority

Local Authority

Frequency

MTR Percent

Cumulative %

Owners w/mort Percent

Ratio

North Lanarkshire

187

12.6

12.6

6.0

2.12

Glasgow City

154

10.4

23.0

8.8

1.19

South Lanarkshire

123

8.3

31.3

6.4

1.30

Fife

93

6.3

37.6

6.8

0.93

North Ayrshire

89

6.0

43.6

2.5

2.36

Edinburgh City

80

5.4

49.0

8.4

0.65

West Lothian

76

5.1

54.2

3.4

1.53

Aberdeen City

62

4.2

58.3

4.0

1.05

Midlothian

62

4.2

62.5

2.3

1.81

Dumfries & Galloway

54

3.6

66.2

2.3

1.60

East Lothian

51

3.4

69.6

1.9

1.78

Aberdeenshire

44

3.0

72.6

4.2

0.71

Renfrewshire

44

3.0

75.6

3.6

0.83

Perth & Kinross

40

2.7

78.3

2.4

1.14

Highland

39

2.6

80.9

3.4

0.78

Dundee City

35

2.4

83.3

2.1

1.12

East Ayrshire

35

2.4

85.6

2.4

0.98

South Ayrshire

32

2.2

87.8

2.2

0.98

Falkirk

31

2.1

89.9

2.6

0.81

East Dunbartonshire

30

2.0

91.9

2.5

0.82

West Dunbartonshire

23

1.6

93.5

1.7

0.91

East Renfrewshire

18

1.2

94.7

2.3

0.52

Inverclyde

16

1.1

95.7

1.9

0.57

Stirling

15

1.0

96.8

2.0

0.51

Argyll & Bute

12

0.8

97.6

1.6

0.50

Angus

8

0.5

98.1

2.2

0.25

Scottish Borders

6

0.4

98.5

2.0

0.21

Comhairle Nan Eilean Siar

5

0.3

98.9

1.6

0.21

Moray

4

0.3

99.1

2.0

0.14

Clackmannanshire

3

0.2

99.3

1.9

0.11

Orkney Islands

2

0.1

99.5

1.5

0.09

Shetland Islands

1

0.1

99.5

1.3

0.05

Source: MTR and Allocation of Cases Databases: Scottish Household Survey (1999-2005))

Table 4.4: Number of MTR Applications by Region

Region

Frequency

Percent

Cumulative %

Clydeside/West C B

758

51.2

51.2

Lothian

269

18.2

69.3

Central-Fife-Tayside

217

14.7

84.0

North East

118

8.0

92.0

Rural South

60

4.1

96.0

Highlands & Islands

59

4.0

100.0

Total

1481

100.0

100.0

Source: MTR and Allocation of Cases Databases

Table 4.5: Case Status by Region (percent within each region)

Status

Clydeside/ West CB

Lothian

Central-Fife-Tayside

North East

Rural South

Highlands & Islands

Total

Settled

36.5

52.4

46.1

45.8

46.7

40.7

42.1

Ineligible

34.8

20.1

29.5

39.0

23.3

45.8

31.7

Withdrawn

12.8

12.3

12.4

11.0

13.3

6.8

12.3

In Process/unknown

15.8

15.2

12.0

4.2

16.7

6.8

13.9

Total

100.0

100.0

100.0

100.0

100.0

100.0

100.0

Source: MTR and Allocation of Cases Databases

Potential Need/Demand

4.12 The project Brief requires us to try to make some estimates of the potential need/demand for the MTR scheme. This is obviously relevant to questions about the overall effectiveness of the scheme in fulfilling its purpose; what proportion of households facing mortgage difficulties and repossession are actually being helped by the scheme? At a more detailed level, such estimates could help to indicate whether the scheme is fully targeting all the areas of need, for example in terms of different geographical areas or in terms of different types of household.

Data Sources

4.13 To get a handle on potential need/demand we have to look at independent data sources, rather than at the MTR administrative records themselves. One approach is to try to use data on the actual numbers of repossession actions in Scotland. For this purpose, we look at data provided by the Scottish Government on the number of court actions for repossession of property in breach / default of a house purchase loan. These data have considerable shortcomings, as has been made clear by those providing the data, but they do offer time series and geographical breakdowns.

4.14 Another source might have been the Council of Mortgage Lenders, who provides data for the whole of the UK. However, they ceased to provide separate data for Scotland some years ago, because of concerns about the accuracy of the previously published data. The underlying problem with this source relates to the problem of locating possession actions or arrears by country, given that the data are collected by lender and there is no simple relationship between lenders and countries. Several major banks are headquartered in Scotland but lend across the UK, and vice versa.

4.15 Therefore, a second approach is to turn to large scale household surveys, such as SHS or SHCS. These can provide indicators of self-reported difficulties with payment, and also measures of high payment to income ratios. They can also provide a profile of the households experiencing these problems, as well as a geographical breakdown. By pooling SHS data across years the sample size can be boosted and some trends over time identified.

4.16 A third type of source is administrative data on households being re-housed as homeless who are former owner occupiers and/or subject to mortgage repossession. SCORE data on new RSL tenants may also be relevant here, although this does not cover the local authority sector. Nevertheless, we report briefly on key numbers which may be inferred from this source.

Arrears, Repossessions and Court Actions

4.17 The most widely cited statistics on mortgage arrears and repossessions are provided by the CML, primarily for the whole of the UK. Trends in these indicators are charted in Figure 4.1, with all indicators expressed as a percentage of the number of mortgages outstanding. This chart shows that as expected movements in arrears and possessions are correlated. The chart shows the very large spike in arrears and possessions in the early 1990s. There is also currently a pronounced increase, although not yet to the levels of 1992

4.18 Court actions for mortgage repossession precede actual possession and can provide an earlier warning of trends and also a wider indicator of numbers of cases affected, as not all cases proceed to possession. Figure 4.2 shows trends in court actions for repossession in England and Wales, including an overall number, the number suspended (to enable households to come to arrangements), and the number 'made', i.e. confirmed. We show England and Wales to capture the wider UK trend, bearing in mind that the legal and courts systems are different in Scotland. In the peak year of 1991 there were 180,000 actions initiated, but this fell to a low point of 60,000 in 2002, before rising again to 132,000, which is between 1992 and 1993 levels.

Figure 4.1

Figure 4.1

Source: Council of Mortgage Lenders.

Figure 4.2

Figure 4.2

Source: Wilcox (2008) Table 53.

4.19 However, the number of actions 'made' fell from 74,000 in 1991 to 16,000 in 2002 before rising again to 47,000 in 2006. The proportion of suspended actions rose, relative to this, up to about 1998, but more recently the proportion of suspended actions has fallen. This is possibly indicative of, firstly, lenders responding to requests and codes of practice which encourage them to exercise forbearance and to use repossession as a last rather than a first resort; but also, more recently, a possibly worrying trend the other way. Possessions have recently been seen to be rising faster than arrears, and one possible explanation of this is the role of sub-prime lenders in moving quickly to the possession stage.

4.20 How does Scotland fare on these numbers? It is only possible to approach this question with considerable circumspection, owing to the data limitations alluded to above. Table 4.6, which should be read with considerable caution, particularly as regards the absolute number of repossessions in Scotland, provides a comparison for repossessions. This table appears to suggest that the number of possessions fell by more than half between 2000 and 2003, from 3,240 to 1,360, paralleling the fall across the UK. However, expressed as a percentage of the number of loans outstanding, it appears that Scotland has a higher rate of repossessions, relative to the population at risk (0.31% vs. 0.19% in 2000; 0.13% vs. 0.07% in 2003). It is possible that the rate may be a better indicator than the number, although we cannot be confident of this.

4.21 It may seem surprising that Scotland has a higher repossession rate (if it does), because it has less conventional affordability problems. But higher repossessions are perhaps to be expected given that Scotland has more relatively 'new' homeowners. Supporting evidence for this is presented below.

Table 4.6: UK-Scotland comparison of repossession numbers and rates 2000-2003
(Note Scotland based on discontinued series, subject to doubts about regional allocation)

Year

UK

No Repos

UK

Repos %

Scotland

No Repos

Scotland

Repos %

2000

21220

0.19

3,240

0.31

2001

18,280

0.16

3,150

0.28

2002

11,970

0.11

2,110

0.21

2003

7,830

0.07

1,360

0.13

Source: CML

4.22 The data available from the Scottish courts are also subject to considerable problems and uncertainties; for example, it is believed that some cases are misclassified and that difficulties exist in identifying disposed cases (the Scottish Court Service is taking action to address these data quality issues). With these provisos in mind we provide a summary in Table 4.7 of the measure most closely related to repossessions, the number of court actions for repossession of property in breach / default of a house purchase loan which are made ('disposed') in the lender's favour, for the period 2003-2007. We group by regions partly to overcome concerns about geography of the courts, particularly in suburban areas around the main cities. We also adjust for a major inconsistency in Glasgow. Therefore, it should be noted that these figures are not exactly the same as those provided by the Scottish Government, but we believe they are consistent.

4.23 The number of court actions appears to have increased sharply in 2005, while subject to only slow growth before and after this point. We cannot explain this discontinuity. Broadly different regions shared this pattern to varying degrees. However, rural areas tended to see a smaller increase and some reduction in the last years.

Table 4.7: Estimated number of ordinary cause Mortgage Lender actions disposed of in favour of the pursuer in the Sheriff Courts, by region, Scotland 2003-2007

Region

2003

2004

2005

2006

2007

Clydeside

1366

1409

2220

2122

2360

Lothian

383

441

618

892

903

Central

489

601

810

845

787

North East

399

298

380

288

209

Rural South

198

201

308

341

262

Highlands & Is

148

91

101

133

163

Scotland

2983

3041

4437

4621

4684

Source: Scottish Government courts data Note:
data for Glasgow adjusted for major inconsistency between earlier and later part of the period in where cases were recorded. Therefore, these estimates and are not identical with those provided by the Scottish Government, but they are intended to be consistent.

Estimating Scottish Repossessions

4.24 From the UK and Scottish data from CML and courts combined, it would appear that on average the numbers of repossessions in a year is equivalent to about 40% of the number of court actions 'made' (disposed in favour of the pursuer). However, it is not clear that this ratio is the same in Scotland as in England and Wales, nor whether it is constant over time (it appeared to be higher, at 73%, in Scotland in 2001 for example).

4.25 If the ratio of 40% had applied in 2003, the number of repossessions in Scotland would have been just under 1,200 in 2003. This figure is quite close to the CML's 1,360 figure in Table 4.6. Noting that the latter figure is also regarded as rather unreliable, we would have to conclude that our best estimate of repossessions in Scotland in 2003 was 'about 1,300'. If the 40% ratio holds, the number may be expected to have risen to 'about 2,000' by 2007. The average over this period, which coincides with the life of the MTR scheme would then have been 'about 1,600'.

4.26 The absolute scale of these numbers may look quite modest, compared with the more highly publicised UK figures, but it should be noted again that the incidence of repossession in Scotland, as a percentage of mortgaged owners, appears to be higher than the UK average. This may also be related to the finding in Chapter 2 that claims for ISMI in Scotland have remained at a significant level and not fallen since the early 1990s.

4.27 On the basis of these estimates it is then possible to make some comment on the relative scale of the contribution made by the MTR scheme. Between late 2003 and early 2008 MTR 'settled' 139 cases per year 14 (i.e. brought them into social sector). If one makes the sweeping simplifying assumption that each of these was a repossession averted, then one can say that the impact of the MTR scheme has been to reduce annual repossessions by 139/(1600+139)=8%. Clearly there is a considerable margin of uncertainty around this precise percentage, but it is a reasonable guide to the order of magnitude of the impact of the scheme.

Scottish Household Survey

4.28 We have analysed data from the Scottish Household Survey covering the period 1999-2005. The most useful indicators in this source are subjective measures based on the questions 'How the household is managing financially these days' and 'How often the respondent has worried about money in the past few weeks'. These are combined with certain other indicators into two composite flag variables, as described in Box 4.1.

Box 4. 1: Defining financial difficulty in Scottish Household Survey

Buyers in Severe Financial Difficult:
Owner occupiers buying their home with a mortgage
'In deep financial trouble'
No savings
Unable to afford to buy locally based on income & houses prices (2005 price basis)
(this indicator available for years 1999-2002 and 2004)

Buyers in Moderate Financial Difficult:
Owner occupiers buying their home with a mortgage
'In deep financial trouble' or 'Have Some financial difficulties' or 'Worry about money almost all the time'
No savings
Unable to afford to buy locally based on income & prices (2005 price basis)
(this indicator only available for 1999 and 2000)

4.29 Table 4.8 shows how incidence of the first of these indicators of difficulty has varied over the time period. This tells a broadly consistent story of improvement up to 2001 followed by some deterioration thereafter. The second indicator was only available for 1999-2000, but shows that the proportion of mortgaged owners with some degree of financial difficulty was quite considerable.

Table 4.8: Households buying with mortgage in severe or moderate financial difficulty by year (percent of mortgaged owner occupiers)

Data year

Severe

Fin Diff

Moderate

Fin Diff

1999

0.50%

3.14%

2000

0.41%

3.59%

2001

0.26%

2002

0.43%

2004

0.29%

Total

0.38%

3.37%

Number of Cases

26322

10533

Source: Scottish Household Survey

4.30 Table 4.9 breaks the scores down by age and household type. The pattern by age is quite different from the profile of MTR clients (discussed in chapter 5). This time the youngest age group shows much higher difficulty scores and the middle aged and older group generally lower scores. However, the incidence of moderate difficulty rises again over the age of 60.

Table 4.9: Households buying with mortgage in severe or moderate financial difficulty by age and household type (percent of mortgaged owner occupiers)

Severe

Fin Diff

Moderate

Fin Diff

Banded age of HIH

16 to 24

1.99%

5.98%

25 to 34

0.33%

3.30%

35 to 44

0.30%

3.06%

45 to 59

0.40%

3.06%

60 to 74

0.30%

5.49%

75 plus

0.00%

3.08%

Household type

Single Adult (younger)

1.06%

5.83%

Lone Parent Family

1.24%

13.82%

2 adults no children

0.12%

2.04%

2 adults 1 child

0.23%

1.74%

2 adults 2 children

0.08%

1.27%

2 adults 3+ children

0.63%

5.01%

3+ adults

0.12%

1.51%

Single Elderly

0.16%

7.28%

2 Elderly

0.23%

4.43%

Total

0.38%

3.37%

Source: Scottish Household Survey; first indicator for period 1999-2004; second indicator for 1999-2000.

4.31 Household types showing a higher incidence of problems are lone parent families, single adults, and larger families. This reflects patterns of income and, in the latter case, outgoings relative to income, across household types.

4.32 Table 4.10 is particularly interesting as it shows the relative incidence of problems in different types of neighbourhood based on the 'Mosaic' classification. It turns out that the incidence of problems is greatest in 'Disadvantaged Council Estates' and 'Families in Council Flats' areas, and in areas with a lot of 'Renting Singles'. It is important to remember that this index is the prevalence of problems among mortgaged owner occupiers in these areas. This evidence suggests that problems of marginal, insecure or unsustainable homeownership and problem debt are most prevalent in areas dominated by social rented housing. In these areas prices are typically low, making these areas of entry for lower income buyers. Many of these buyers may be ex-council tenants and many of the properties involved may be ex- RTB. This confirms some hypotheses suggested earlier, and previously predicted in Bramley et al 2002.

Table 4.10: Households buying with mortgage in severe or moderate financial difficulty by Mosaic neighbourhood type, 1999-2000 (percent of mortgaged owner occupiers)

Mosaic groups (10)

Severe

Fin Diff

Moderate

Fin Diff

Number

of Cases

High income areas

0.08%

1.11%

1595

Middle income owners

0.20%

2.45%

2127

Low income owners

0.36%

3.23%

1239

Better-off council

0.31%

4.38%

1586

Disadvantaged council estates

1.05%

6.54%

818

Families in council flats

1.40%

5.34%

329

Renting singles

0.80%

5.47%

531

Singles and flats

0.60%

3.65%

944

Country dwellers

1.01%

3.35%

801

Institutional areas

0.33%

1.80%

562

Total;

0.46%

3.37%

10532

Source: Scottish Household Survey
Note: Mosaic labels provide descriptive flavour for areas; they do not mean that all the households in those areas have those characteristics.

4.33 It is possible to look at the incidence of problems in SHS by local authority, although sample sizes are rather small in some cases. This shows that some of the same authorities identified in Table 4.3 also feature highly, but there are also some differences, for example in the greater incidence of problems in higher status suburban areas. Value and subsidy limits may tend to exclude people from these areas from using MTR. There may also be differences in the type of financial problems experienced and in local culture and expectations, for example in relation to social renting as an option.

Scottish House Condition Survey

4.34 We can report more briefly on a parallel analysis based on the Scottish House Condition Survey ( SHCS) for 2002. This does not have similar subjective financial difficulty questions but it does have more information on housing costs as well as incomes. It is therefore possible to look at more objective measures of affordability pressures, and again combine these into multi-criteria risk indicators. The limitations of this include the relatively smaller sample (15,000), the slightly dated information (2002), as well as the lack of subjective problem measures. We therefore only report here on headline results.

4.35 Table 4.12 shows that 1.71% of owner occupiers in Scotland were paying more than 50% of their income on housing (mortgage plus repairs), while 2.98% were spending more than 40%. If you combine that with having low savings (under £3000) and not having enough income to buy in the local market at typical lower quartile prices, then you get an incidence of 1.08%. This proportion is similar to the SHS-based moderate difficulty index described above. Including Council Tax and energy cost in a wider housing cost concept and setting the trigger for this ratio at 50% of net income, the incidence rises to 1.66%. Adding a further alternative criterion, of having a mortgage of more than 4.5 times gross current income, further increases the incidence to 1.94%.

Table 4.11: Incidence of Composite Affordability Flags for Owner Occupiers in Scotland, 2002

Variable Name

Criteria

% of Owners

Affprob

Basic ratio hsg cost to net income>0.5

1.71%

Affprob1

Basic ratio hsg cost to net income>0.4

2.98%

Affprob2

AR >0.4, Savings <£3k, Unable to buy mkt

1.08%

Affprob3

AR>0.4 or Wider HCIR>0.5, Savings<£3k, unable to buy.

1.66%

Affprob4

AR>0.4 or Wider HCIR>0.5 or LTIR>4.5, Savings<£3k, unable to buy.

1.94%

Source: Scottish House Condition Survey 2002.

4.36 This shows similar demographic patterns to those found with SHS: higher for single adult and lone parent households, small and large households; higher for ethnic minorities. It is also higher for those expecting to move,

4.38 The pattern by local authority shows some similarities but some further differences. The highest authorities are Glasgow, Inverclyde, City of Dundee, Edinburgh, South Lanarkshire, East Dunbartonshire, East Renfrewshire, East Ayrshire, and Moray. Clearly this embraces a mixture of city, poorer urban areas, and more affluent areas. Again, the appearance of affluent suburban areas in this list prompts some consideration of why they do not feature so much in the MTR caseload, as commented on above.

SCORE data

4.39 The Scottish Continuous Recording System ( SCORE) provides detailed individual data on most RSL lettings, both properties and people. Using data for 2005/06 we can identify that 9.3% of new tenants (1,043) were previously owner occupiers. If we make the assumption that this proportion applied to all lettings to new tenants by local authorities as well as housing associations (total net relets of 52,380, based on Bramley et al 2005 Table 2.2), that would imply a total of 4,875 former owners moving into social renting that year. The main reasons for rehousing of former owners were, in descending order, medical/health (32.5%), support needs (14.2%), relationship breakdown/bereavement (13.9%), financial difficulty (8.1%), 'require independent living' (5.8%), mortgage repossession (3.3%) and disability (3.3%). These data confirm our impression that the main flow of owners into social renting is associated with ageing and associated health/social problems, and that the part that is due to financial mortgage problems is a minority, on this evidence about 11.4% or about 560 households per year for the whole social sector.

4.40 It is also clear that most of these former owners moving into social renting do not go through the homeless route. However, it may be a higher proportion of those in mortgage difficulty, as against a lower proportion for the elderly/health/social cases. In SCORE in that year there were 97 'former owners prior to homelessness' cases - for the whole social sector this would gross up to about 450 cases per year. These numbers are not inconsistent with the homelessness data discussed below.

Homelessness

4.41 We can report and reflect briefly on the headline incidence of mortgage default as a cause of homelessness, based on the Scottish Government homelessness statistics. In 2006/07 mortgage default accounted for 913 homeless applications, compared with 761 in 2003/04. From the viewpoint of homelessness as a whole, this is a very small number, only 1.5% of total applications. However, from the viewpoint of mortgage repossessions, it is a relatively high proportion. Using our earlier estimates (see under 'Estimating Scottish Repossessions') of the number of possessions, we would suggest that around half of these end up as applying as homeless (59% in 2006/07, 46% in 2003/04) 15.

4.42 Not all homeless applicants are assessed as homeless and in priority need (77% of applications overall); and not all of these end up being re-housed by social landlords (overall, about half of priority need cases, or 35% of applications). We would expect that, for this mortgage repossession group, these proportions could be above average, because the group contains more families with children and more potentially vulnerable older groups than the overall homeless applicant population (these groups are more likely to be priority and more likely to be rehoused). However, interview survey evidence reported in Chapter 5 (para 5.65) seems to question this. In addition, some cases may apply because of a threat of mortgage default rather than one which has happened, in which case they would not end up being rehoused. If half of these homeless applicants were re-housed by social landlords, that would amount to about 460 cases per year (25-30% of repossessions). This figure is virtually the same as that derived from SCORE above.

4.43 We suggested earlier that MTR was helping to avoid up to 8% of repossessions per year (139 cases per year). If the cases so helped were 'typical' of all repossessions, then we would expect the impact of this on homelessness to be of the order of 73 less homeless applications per year (8% of 913, based on 2006/07 data), or 36 less households re-housed as homeless. This means that only just over a quarter of the MTR cases helped would actually represent a homeless re-housing averted. At the other extreme, if we assumed that the typical MTR cases were highly likely otherwise to apply as homeless, then it could be a much higher figure. Clearly, the maximum feasible figure would be 139 homeless re-housing cases avoided. That would be 30% of the estimated total number of households re-housed as homeless due to mortgage default (460). However, the true figure is unlikely to be as high as this. The interview survey data suggested that 'half or more' of MTR settled cases avoided homelessness or possession (para 5.61), which is more consistent with the first estimate in this paragraph.

Potential future demand

4.44 How far can we predict future levels of demand for MTR? We are not in a position to operate a formal forecasting model, but drawing on evidence reviewed earlier in this chapter, the comments in Chapter 3 (para's 3.32-3.35), and broader findings from the research, we can make some general statements about future demand. Firstly, demand is likely to increase sharply in the immediate future, because numbers of cases in arrears and going through the Courts appear to be increasing on a national basis. Secondly, it is now clear that, following the Credit Crunch, a significant economic recession is going to take place, a recession which is particularly affecting the housing market. Figures 4.1-4.2 suggest that the number of actions for repossession could rise several-fold in the recession, based on the 1991 experience and the recent upturn. However, that could be unduly pessimistic, given the greater policy emphasis on forebearance (Mortgage Rights Act, MCOB). There is also the important point that in 1989-91 a crucial trigger was high interest rates, while now monetary policy is emphasising low rates.

4.45 In the medium term, demand will be affected by how quickly the economy and the housing market recover from this recession, but also by factors such as how well the scheme is promoted and whether its eligibility criteria are modified. We infer from evidence on local variations in take-up that there may be scope to promote the scheme more in some areas. In addition, current media attention to the repossession issue and actions and statements by government ( UK and devolved) on the issue are raising the profile of MTR. Possible widening of eligibility is discussed further in Chapter 10, picking up comments from Chapter 3 and elsewhere. In the short and medium term values are likely to be lower, with more cases of negative equity. This will tend to increase demand, because the alternative sell/trade down option will be less viable; it will also affect cost because there will be less capital contributions to defray subsidy, although on the other hand the capital values may fall somewhat. A final factor to consider is the point that, following the recent credit crisis, the availability of generous secondary lending secured against home values may become much more restricted. Given that this has been a major source of MTR demand, this could have the effect of reducing demand, offsetting some of the above points.

4.46 Overall, taking account of the above points, it seems very probable that demand for MTR will rise significantly in the near and medium term future. If future demand is going to be significantly higher, this will place more strains upon the existing budget. The Scottish Government would then need to either allocate more resources within its available totals, or apply additional rationing criteria. Possible ways in which eligibility criteria might be restricted are discussed in Chapter 10.

Conclusions

4.47 The MTR scheme has built up to a level of about 400 applications per year, of which just under half successfully convert into settled cases, the majority of the remainder being ineligible. The geographical pattern shows uneven take-up between localities which can only partly be explained by the demographics of the population at risk. We cannot explain all of this variation but suggest that it may reflect a combination of the effects of eligibility criteria (e.g. value limits), differential promotion of/knowledge about the scheme, or local cultural factors.

4.48 Eligible demand tends to be strong from poorer areas which formerly had a lot of social housing, including former RTB properties, and is generally greater in urban central Scotland. MTR cases are older than typical homeowners in financial difficulty. This may be because younger owners have more options and possibilities (e.g. to move geographically, enter private renting, increase their income, or get financial help from older relatives). However, the geographical pattern of takeup fits to some extent with survey data showing a higher incidence of financial difficulty among homeowners in former council areas

4.49 MTR cases may in many instances avert homelessness. However mortgage default only represents a small fraction of total homeless applications (1.5%). Our best estimate is that MTR may, in helping 140 households per year to remain in their homes, have reduced repossessions by 8% in Scotland. This could represent between a quarter and a third of the number of households rehoused as homeless due to repossession in Scotland.

4.50 The demand for MTR is likely to rise significantly in the near and medium term futures. This has implications for resourcing and possibly for rationing.