The Scottish Government

Source: CML
Notes

1. The number of loans in Scotland is approximately 8-9% of the UK total. Loans have grown in Scotland by 25.8% between 1993 and 1999.

2. Lenders were asked to supply information on primary loans on properties only, and if they had two loans on the same property to count this as one. Thus, the number of loans given here should relate, in the main, to mortgages, rather than secondary secured loans.

3. The CML estimates of numbers of loans are based on figures supplied by lenders they surveyed.

4. # There are no figures available for 2000 in Scotland but the increase in CML's recorded repossessions between 1998 and 1999 is not large. Given the fall in debt servicing costs over the last year it seems unlikely that there will have been an increase and there may even have been a drop. CML has published the number of repossessions for the UK as a whole. For 2000 these figures show a 26% fall from 1999.

5. A general point to clarify is that CML 'UK' figures include Scotland and Northern Ireland.

3.11. The figures presented above show

• a 26% increase in the total number of loans in the period 1993-99 and of 22% in the period 1994-99
• a 33% increase in the total number of repossessions in the period 1993-99 and of 70% in the period 1994-99
• an increase in the rate of repossessions as a percentage of all loans of 7% in the period 1993-99 and of 41% in the period 1994-99.

3.14 These figures illustrate the nature of the trends involved in an interesting way. Despite the large rise in the number of repossessions, it appears that it is partly a result of statistical effects as well as underlying tendencies that might be a function of administrative, social or economic factors. A significant part of the increase can be attributed to the increase in the size of the loan market over the period. Reasons for this interpretation are set out below.

• a point to note is that the changing propensity for loans to result in repossession is not measurable in terms of the changing numbers of repossessions. The indicator of propensity is the proportion of loans resulting in repossession
• trends in numbers of repossessions are likely to appear volatile, because the propensity indicator is the coefficient of a large total number of loans. The 1994-99 increase of 70% in the number of repossessions is a function of a rise in the indicator of only 0.09 percentage points. Such a small change in the indicator could result from causes that were not strikingly evident
• other factors being equal, the more loans extant at any time, the more repossessions there will be. Some of the 1994-99 increase in the number of repossessions can therefore be explained simply by the increase in the number of loans over that period. The residual underlying increase, not accounted for by this factor, is 881 or 48%, rather than the 70% actual increase
• this 48% increase is the true 1994-99 outcome of the change in propensity measured at only 0.09%.

3.15 The small change in propensity is where explanation of the steep rise in repossessions must be sought. Many influences could be involved, but it is clear that an important factor must be the redistribution of levels of loan maturity resulting from the rising market. In the market illustrated in Table 3.3 old loans have been replaced at a much faster rate by new loans. It is known that most repossessions occur when loans are relatively recent and the debt burden is likely to be higher, so the characteristics of the population of loans will have changed to such an extent that a rise in the propensity indicator of 0.09% seems feasible.

3.16 While this must be an important component of the increase, it is unlikely to tell the whole story. The loans market has also grown in England, but there is a downward trend in repossession. This does not alter the fact that changes in the size and maturity of the market will have been influential, but it does imply that the other factors underlying the trend in England must have been even stronger than would otherwise be apparent. (One factor may be that house price inflation in the 1990s was higher in England than in Scotland, which offered borrowers more chance of trading out of difficulty and avoiding repossession.)

3.17 Unfortunately, data on the age distribution of loans and repossessions in Scotland in 1994 and 1993 are not available to support an estimate of the proportion of the rise that can be accounted for in this way. Some illustrative calculations have been undertaken by the research team, but these have an insufficiently factual basis for presentation here. However, a reasonable working assumption is that a significant proportion, perhaps about half, of the residual 48% increase when the effect of the increase in the total number of loans has been taken out, might be accounted for in this way.

3.18 It therefore appears that the apparent increase of 70% in repossessions between 1994 and 1999 would be much reduced once statistical inevitabilities have been taken into account. While this may serve to assuage some of the very understandable concerns that the increase has given rise to, it must be acknowledged that there is probably a residual upward trend which is smaller, but still of concern.

3.19 That said, the figures in Table 3.3 are cautionary. They suggest that social policy development should be based on wider consideration of the issues than statistics (as, of course, it is). The trends are volatile and do not seem susceptible to interpretation, given the information available. If the 2000 figures for Scotland were to show a drop in repossessions (as currently seems probable), then the 1993-2000 trend might be very different from the steep upward trend 1994-99, especially if account is taken of the increased number of loans.

3.20 The above comment is in no way intended to suggest that social policy initiatives aiming to reduce the level and impact of repossession are superfluous: the adverse impact on all parties to repossession is certainly a problem worth addressing.

civil justice statistics (cjs)

3.21 Table 3.3 shows the number of decrees in Scotland as reported by the CJS.

Table 3.3 JS figures for mortgage repossession decrees from 1994 to 2000

Note

Figures in relation to repossession of property by mortgage lenders have been collected by the courts since a revised system of collecting Civil Judicial Statistics was instituted on 1 January 1994.

3.22 CJS figures reflect the number of repossession decree cases in the Scottish courts. They record all cases that have been initiated, not only those granted decrees. Neither actions entered nor decrees made against a borrower necessarily end up as a possession. The research discovered that most lenders take possession in roughly half of successful repossession court cases (see chapters 5 and 6). Therefore, court decrees do not indicate the true number of repossessions.

3.23 A valid comparison of the CML and CJS figures becomes impossible on those grounds alone. Additionally, CJS figures exclude settlements that did not require forced repossession and any other cases not reaching the courts.

3.24 There are also reasons why trend information might be incompatible. A change in the number of court decrees is not necessarily an indicator of equivalent change in repossessions. Lenders may vary their policies concerning the treatment of arrears and the routes to repossession selected and this could mean that trends in CJS data would not be equivalent to the CML data trends.

3.25 Interviews with lenders and legal advice taken by the project suggested that there was quite a variation in routes taken by different lenders and that some lenders have made changes in the legal procedures they use. For instance, some now go straight for court decrees instead of using a calling-up notice. This would cause a rise in repossession decrees, without a corresponding rise in repossessions.

3.26 Other reasons covered later in the report (Chapter 4) also show that CJS figures must be interpreted cautiously. These include

1. Case classification errors by court employees (confusion has resulted in repossession cases being coded under 'Land/Heritable - Recovery of debt out of heritable estate' instead of 'Mortgage Lender - Repossession of property in breach/default of a house purchase loan')

2. Difficulty in distinguishing between commercial and residential repossessions

3. Cases with individuals, local authorities or Scottish Homes as pursuers were not thought to be residential repossessions and were removed from our analysis of court records. They would not be comparable with the CML figures, which do not cover public sector loans. However, in a discounted RTB sale, a local authority would have a standard security over a property and any action taken would be recorded in the CJS. The number of such repossessions may not be large, but they would tend to widen the gap between the data sources.

Homelessness applications to local authorities and housing applications to housing associations

3.27 Other indicators of the levels of repossession in Scotland were examined to see if another source could throw any light on the differences between the CML and CJS figures. Data on the number of households presenting themselves as homeless to local authorities were examined.

3.28 From local authority homeless applications, it is possible to identify cases where people are homeless because of a 'court order - mortgage default'. The figures show that there has been a decline in applications for local authority housing for reasons of mortgage default as a percentage of total applications (Table 3.4). The reason for this decline is not known but, in the light of comments from the borrowers we interviewed, it may be because public-sector housing is no longer seen as desirable. It was precisely because of the perceived unattractiveness of public-sector housing that some of our interviewees bought their own homes. Re-housing in a private-sector rented home was seen by some as the only acceptable option after they had lost their owner-occupier homes.

Table 3.4 Applications for local authority housing as loss/threatened loss of accommodation due to a court decree on mortgage default in Scotland

Source: Scottish Executive Statistical Bulletin. Housing Series: HSG/2001/5
Note

Figures have been rounded to the nearest hundred.

3.29 Housing associations also log all applications for accommodation. These can also be classified according to the reason for the application, with 'mortgage arrears/repossession' specifically recorded. The number of relevant applications is too small to use as a trend indicator with any confidence, but they have more than doubled in six years, rising from 27 in 1993/4 to 69 in 1999/2000 (see Table 3.5). This should be interpreted while bearing in mind the growth in the importance of housing associations vis-à-vis local authorities as promoters of social housing. Data from housing associations are consistent with an increase in repossessions, though the information could not be used to determine the rate of increase.

Table 3.5 Mortgage arrears/repossession as the reason for requiring housing in housing association applications

Note

All of the data is gathered on our financial year basis running between 1 April and 31 March.

Source: an extract of SCORE database, provided by Scottish Homes.

3.30 Collated housing applications figures did show a tendency to rise, although there were year on year variations. Some broad similarities between the trend in the CML figures and the Local Authority (LA) applications were apparent. However, the examination of these figures has not contributed to explaining the discrepancies between CJS and CML figures. Nevertheless, it has been useful in demonstrating how many people sought places in public-sector housing after repossession.

trends in cml and cjs figures

3.31 It has been made clear that CML, CJS and local authority housing applications all measure different things. Nevertheless, they have been brought together in Figure 3.1 to show the broad trends in these three sets of figures.

Figure 3.1 Trends in CML, CJS and LA housing applications

3.32 The CJS and the CML figures both show an increase in repossessions, although the CJS increase is much larger. Local authority applications are stable throughout the same period. Although the considerations presented above suggest that the figures are not comparable, the overall interpretation must be that the 1994-1990 period does incorporate an increase.

Implications

3.33 The implications of the findings presented in this chapter will be brought together in the synthesis in Chapter 8. The principal points to be taken forward there are

• CJS records repossession orders, and CML figures represent actual repossessions. The CJS figures are therefore indicators, rather than measurements of repossession, i.e. at most, they may be used to identify trends
• reliable trend indication using CJS will depend on there being a background of consistent practice by lenders and a reasonably consistent legislative and policy background. In particular, the CJS figures would misrepresent repossession trends if the proportion of orders acted upon varied over time
• our estimate of the proportion of orders leading to repossession is 40%. This goes a long way to explain the discrepancy between the CJS and the CML figures

• our postal survey of lenders recorded a higher amount of repossession than the CML data. This difference, which amounts to about 30%, is not easily explained, but may be to do with the inclusion of secondary lenders and non-CML lenders in our survey, while CML do not include secondary lenders. It is unlikely that double counting of repossessions is an explanation as lenders said they would only count repossessions they actioned themselves

• despite the discrepancy between the CML and research surveys, there is no reason to doubt that the upward trend shown in the CML figures is genuine. A significant amount of the increase can be accounted for by the increase in market size over the period in question
• repossession is most common in the early years of mortgages. An important factor responsible for the increase in repossessions must be the redistribution of levels of loan maturity resulting from the rising market.

« Previous | Contents | Next »