Shared equity schemes: evaluation reports

Reports on the evaluation of shared equity schemes.


8. Additionality

This chapter assesses the 'additionality' delivered by the three shared equity schemes. For the purposes of this study, additionality has been defined as including the following elements:

  • Demand-side additionality being the number of households buying a home with a shared equity scheme, who would not have been able to do so in the same broad area in the absence of the scheme; and
  • Supply-side additionality being the number of new homes built as a direct result of the schemes that would not otherwise have come forward in a given time period.

The assessment of additionality draws on demand and supply-side evidence considered over the preceding chapters of the report. This chapter also draws on more detailed survey analysis to consider any differences in additionality across key buyer groups and over the life of the shared equity schemes, and the extent to which 'additional' buyers are distinct in terms of their characteristics and purchase choices.

Our approach to assessing additionality has been adapted from that used in the evaluations of Help to Buy in England and Wales that were commissioned by the UK Government and Welsh Government respectively. This assessment approach is primarily focused on identifying direct effects of shared equity schemes in terms of the proportion of sales that would not have taken place otherwise. It is important to note that this may not fully capture 'indirect' effects of the schemes, such as any additional sales as a result of increasing confidence amongst buyers, developers and lenders.

Context to the assessment of additionality

The Scottish Government's three shared equity schemes were introduced at different points in the housing market and economic cycle and were developed to address different barriers to home ownership. NSSE and OMSE were introduced in 2007 to improve equity of access to home ownership for moderate-income households who had been disadvantaged by house price inflation, whereas HtB was introduced in 2013/14 to support the recovery of the house building industry and help households to overcome increased deposit requirements introduced following the 2008 global financial crisis.

As noted in chapter 4, access to home ownership in Scotland has improved in the period since 2015 with less restrictive mortgage lending criteria and deposit requirements. This means it is important to consider the continuing case for the Scottish Government's shared equity schemes as they are currently operating. This is particularly so for HtB, given the upturn in new build activity and improvements in mortgage availability and affordability, factors that were a key part of the rationale for its introduction. Estimates of additionality and any evidence of changes over time in the level of additionality generated by the schemes, are important aspects of the evidence base which will inform decisions on the future of the schemes.

Evidence on operation of Help to Buy

Programme information supplied by the Scottish Government included usable records for 15,984 HtB completed purchases. As outputs in appendix 2 show:

  • Around 70% of HtB completed sales were in 10 local authority areas. The five local authorities with the largest number of HtB completed sales were North Lanarkshire (12%), Glasgow (12%) South Lanarkshire (9%), Edinburgh (8%) and Renfrewshire (6%).
  • Two thirds of HtB sales were developed by a pool of 12 private developers, with close to half being developed by three major builders; namely Persimmon Homes Limited, Taylor Wimpey UK Limited and BDW Trading Limited (Barratt).
  • Between 2013-14 and 2018-19, equity loans totalled more than £490 million, supporting the purchase of over £2.8 billion worth of property. Unlike the UK Government scheme, interest is not charged on HtB equity loans. As a result, the Scottish Government may be more exposed to the risk of another downturn in the housing market, although this would be mitigated if Scottish Government loans are based on a smaller equity share.
Figure 37: Lower quartile, median and upper quartile HtB prices for Scotland, 2013 to 2019
Figure 37: Lower quartile, median and upper quartile HtB prices for Scotland, 2013 to 2019

Source: Scottish Government HtB administrative records

Reliable price data for NSSE and OMSE is not available but Figure 37 summarises prices for HtB sales. It confirms that prices for HtB sales have reduced over time, consistent with the lowering of the house price cap. It also shows that the range of HtB sale prices in any given year have also reduced over time, again largely influenced by the lowering of the price threshold. For instance, the difference between the lower and upper quartile HtB price in 2016/19 was £39,000, down from £77,650 in 2013/14. The variation is likely to reflect differences in the size, composition and location of the properties bought with a HtB equity loan, but data limitations preclude analysis to document this.

As HtB was intended to reduce the barriers to homeownership caused by restrictive deposit requirements, it was not surprising to find that just under half (48%) of HtB purchasers had secured an equity loan of 20%, most of whom had accessed HtB in its first two and a half years of operation. A similar proportion had an equity loan of 15%, most of whom had secured the loan from April 2016 onwards, reflecting changes made since 2016. This suggests that the majority of buyers have opted to take the maximum value of equity loan possible.

Correspondingly, the average cash value of the equity loan secured has fallen back whilst variations in the cash value of the equity loans issued in any given year have narrowed (see figure 38). In the first half of 2019, for instance, the median equity loan stood at £26,250, down from a peak of £37,199 in 2014. Over the same period the 'gap' between equity loans in the lower and upper quartile of the distribution fell from £17,200 to £6,150. By way of comparison, evidence from UK Finance indicate that the average deposit made by first time buyers between June 2015 and June 2019 declined from £29,500 to £27,000. This suggests changes to the HtB price cap have brought the scheme more closely into line with 'market practice' in respect of first-time buyers, albeit HtB purchasers must make a 5% deposit in addition to the equity loan.

Figure 38: Cash value of HtB equity loan in year it was secured, 2013-2019
Figure 38: Cash value of HtB equity loan in year it was secured, 2013-2019

Source: Scottish Government HtB administrative records.

Assessing demand-side additionality

Chapters 7 and 8 considered qualitative findings for private developers, RSLs and lenders involved in the three shared equity schemes. This included some findings which point to the additionality generated by the schemes.

Private developers involved in the HtB scheme and RSLs involved in the NSSE scheme indicated a common perception that most buyers using these schemes would have been unable to buy a suitable home without some assistance. Private developers generally felt that most households buying under the current HtB £200,000 price cap would have been unable to access the new build market otherwise, especially those buying in higher value markets. In support of this, some referred to an increase in the proportion of first-time buyers using the scheme in recent years.[21] RSLs also generally perceived that means testing and economic assessment had ensured that the great majority of NSSE buyers could not have accessed the market otherwise. RSLs typically suggested that NSSE had supported households 'caught' in a position of having insufficient priority to access social rented housing, and insufficient income or equity to buy.

On the other hand, lenders and some developers were more cautious about the extent to which HtB had generated demand-side additionality. Lenders tended to observe that HtB purchasers included households that were likely to have been able to buy without assistance in the foreseeable future, or could have otherwise purchased a home in the resale market. Some developers also conceded that prior to the first price cap reduction in October 2014, substantial proportions of buyers may have been able to buy without assistance especially in lower value markets in the west of Scotland. Some also suggested that easier access to mortgage finance in the last 1-2 years may have increased the proportion of buyers that could have accessed the housing market without assistance. This perception is supported by evidence set out in chapter 4, showing an increase in Loan-to-Value ratios since 2015.

The views of housing developers, RSLs and lenders were compared to survey results regarding buyers' views on how their use of a shared equity scheme had shaped their housing outcomes. Drawing on the approach used in evaluation of schemes in England and Wales, our assessment of additionality was based on buyer agreement or disagreement with the following statements:

  • I would have been able to buy a property I wanted anyway without the scheme
  • I would have been able to buy the same property without the scheme
  • I would have been able to buy a smaller property that was still suitable for my needs without the scheme
  • I would have been able to buy a similar property that was not new build without the scheme (for HtB and NSSE buyers only)

Although our approach to the assessment of demand-side additionality is necessarily based on the views of those using the schemes, this brings a risk of over or under-estimation of the impact of shared equity schemes. For example, those who have used the schemes may feel an obligation to report that they benefited from the assistance provided. Buyers may also be unaware of wider systemic market impacts, such as whether their own improved access to the market had limited access for others, or if schemes have improved lender and/or developer confidence and thus improved access more widely. Consequently, we drew on outputs from our earlier analysis of housing market activity and the profile of shared equity buyers to refine our assessment of the scale of additionality.

Considering the options

The statements noted above can be used to develop various approaches to assessing additionality, dependent on the interpretation of additionality used and how confident we can be that survey responses accurately describe buyers' ability to access the market without assistance. For example, a 'strict' interpretation of additionality would be to assume that only buyers who felt that they 'definitely' could not have bought without assistance are genuinely additional. The most 'generous' interpretation would be to assume that only those who felt that they 'definitely' could have bought without assistance should be considered non-additional (i.e. those who felt that they 'probably' could have bought without assistance would still be counted as additional).

The approach used by the UK Government falls between these extremes; those who felt that they 'definitely' or 'probably' could not have bought without assistance were counted as additional. Our assessment introduced a fourth option, which also counted those who gave a neutral or unclear view as additional. This recognised that buyers are likely to vary in how well they recall the circumstances around their house purchase, and may also vary in terms of how 'optimistic' they are about their ability to buy without assistance.

The additionality estimates calculated using these four methods are summarised in Table 9 over the page. As this shows, there is considerable variation between the 'strict' (option 1 estimating additionality at 6% of those using schemes) and 'generous' (option 4 estimating 76%) estimates. The scale of the difference between these options is also broadly consistent across the three schemes.

Option 2 (the UK Government approach) and option 3 (option 2 plus 'neutral' answers) provide similar additionality estimates, of 23% and 28% respectively. However, the difference between these two estimates is larger for NSSE and OMSE buyers; an 8% and 7% difference respectively, compared to a 4% difference for HtB buyers.

Table 9: Overview of alternate additionality estimates
Option 1 Option 2 Option 3 Option 4
Help to Buy 3% 16% 20% 73%
NSSE 9% 31% 39% 82%
OMSE 14% 40% 47% 83%
ALL 6% 23% 28% 76%

Option 1: Only those who answered 'definitely not' (or 'strongly disagree') to questions about whether they could have bought without assistance.
Option 2: Option 1 plus 'probably not' or 'tend to disagree' answers.
Option 3: Option 2 plus neutral or unclear answers.
Option 4: Option 3 plus thus who answered 'probably' (or 'tend to agree') to questions about whether they could have bought without assistance.

Our assessment is that options 1 and 4 above provide unrealistic estimates of additionality. In part this reflects that neither option is consistent with qualitative findings discussed in previous sections on developer and lender estimates of additionality. More detailed analysis of the profile of additional buyers as identified by the more 'generous' option 4 also suggests that a substantial proportion of these buyers are likely to have been able to access the market without assistance in the foreseeable future. This analysis suggests that option 4 over-estimates additionality for HtB buyers in particular, but that the NSSE and OMSE schemes have been more effectively targeted at households who would otherwise be unable to buy:

  • Nearly half (48%) of additional buyers (based on option 4) may have been able to buy a similar property in a less convenient or desirable area. Those buying through HtB were most likely to have indicated this (56% compared with 41% for NSSE and 32% for OMSE).
  • 17% of additional buyers (based on option 4) were homeowners at the time of buying. Nearly all of these buyers used HtB scheme, accounting for 24% of the option 4 additionality estimate for HtB. Some of these buyers may have been moving due to relationship breakdown or other changes in their personal circumstances, but this group also appears to include some owners using the schemes to access better housing options within the sector and who would not be considered additional.
  • Survey results suggested considerable overlap in the income distribution of additional and non-additional buyers. In the case of HtB, the median income of additional buyers (based on option 4) was only 2% lower than the median income for non-additional HtB buyers. By comparison, for OMSE the median incomes of additional buyers were 7% lower than for non-additional OMSE buyers.
  • A large majority of additional HtB buyers (based on option 4) paid upwards of £150,000 for their current home. This was above the average price paid by all first-time buyers in Scotland; in the five years to the end of 2018 the average price paid by first time buyers fluctuated between £137,400 and £145,200. In contrast, the majority of additional NSSE and OMSE buyers paid less than £150,000, and often considerably less.

Options 2 and 3 provide broadly similar estimates, and our analysis suggests that both are reasonable overall estimates of additionality. However, the more significant differences in option 2 and 3 estimates for NSSE and OMSE buyers raise some concerns regarding the extent to which option 2 accurately describes the impact of these schemes.

A larger proportion of NSSE and OMSE buyers provided neutral or unclear responses to the additionality statements than was the case for HtB buyers. This is likely to be due at least in part to the larger proportion of NSSE and OMSE buyers who bought their property three or more years ago, including some who bought their property more than five years ago. For these buyers, neutral responses may reflect a lack of recall, and are not necessarily a reliable indicator that buyers would have been able to access the market without assistance. This is supported by survey findings which suggested that these 'neutral' respondents do not differ significantly in profile when compared with buyers identified as additional under option 2. On this basis, we have taken option 3 as our 'central' estimate of demand-side additionality.

'Central' estimate of demand-side additionality

Figure 39 summarises our central estimate of the additionality generated by shared equity schemes. This estimates that the overall proportion of buyers using Scottish Government shared equity schemes who can considered to be 'additional' is 28%.

Figure 39: 'Central' estimate of demand-side additionality for shared equity schemes
Figure 39: 'Central' estimate of demand-side additionality for shared equity schemes

Source: Survey of buyers. Note: the final statement ("I would have been able to buy a similar property that was not new build without the scheme") was presented only to HtB and NSSE buyers.

Within this overall estimate, the proportion of 'additional' buyers is estimated to be considerably smaller for HtB buyers (20%) than for NSSE (39%) and OMSE buyers (47%). The smaller proportion of HtB buyers assessed as additional is primarily due to a larger proportion of buyers indicating they would have been able to buy a smaller property that met their needs without assistance; 70% of all HtB buyer respondents indicated this, compared with 49% of NSSE and 41% of OMSE buyers.

Figure 40: Estimate of demand-side additionality by scheme
Figure 40: Estimate of demand-side additionality by scheme

Source: Survey of buyers.

Survey results indicate that the rate of additionality has been fairly stable for HtB over the scheme's lifetime at around 20%, with a small jump to 25% in 2019 (see Figure 41). There has been greater volatility in the additionality rate for OMSE and NSSE, with no apparent upward or downward trend.

To control for change in the HtB price cap, the assessment was restricted to apply only to those buying properties under £200,000 throughout the life of the scheme. This group of HtB buyers also show an apparent increase in additionality, from 19% of those buying under £200,000 prior to 2018 to 26% of those buying in 2019. However, this increase is only evident for those buying in 2019, and survey evidence is insufficient to assess whether this is a genuine ongoing upward trend.

Figure 41: Estimate of demand-side additionality by year of purchase
Figure 41: Estimate of demand-side additionality by year of purchase

Source: Survey of buyers.

Even for buyers who are not counted as additional based on answers to the four statements, the schemes could have had an effect by bringing forward these purchases in time. Indeed, a large majority of non-additional buyers across the three shared equity schemes reported that the schemes allowed them to buy a property sooner than they could otherwise have done (HtB: 82%, OMSE: 89%, NSSE: 73%).

Profile of additional buyers

Table 10 below summarises the profile of additional buyers based on our central estimate. This includes all additional buyers across the three schemes; additional tables for individual schemes are provided at the annex to this report.

This profile indicates differences in the 'additionality rate' across key buyer groups. Lower income households, those in social or private rented accommodation, single parents, those buying smaller and lower value homes, those aged 40+ and first-time buyers were most likely to be additional. Figure 42 summarises the estimated additionality rates for each of these groups. This shows that the additionality rate is over 30% for each of these groups.

Figure 42: Estimate of demand-side additionality for key buyer groups
Figure 42: Estimate of demand-side additionality for key buyer groups

Source: Survey of buyers.

The profile of additional buyers also suggests that the shared equity schemes have had some success in supporting households who typically have more difficulty accessing home ownership. For example, the larger proportion of low income households and those living in rented accommodation classed as 'additional' is consistent with the schemes assisting those who commonly struggle to raise the deposit required to access the market.

Table 10: Profile of additional buyers (central estimate)
Additionality rate1 % of additional buyers2 % of non-additional buyers3
Shared equity scheme
Help to Buy 20% 50% 78%
NSSE 39% 5% 3%
OMSE 47% 45% 19%
First-time buyers
First-time buyer 32% 78% 70%
Existing owner 24% 22% 30%
Age of respondent
Under 30 25% 20% 25%
30-34 27% 24% 26%
35-39 30% 23% 22%
40-44 32% 15% 13%
45+ 34% 18% 14%
Household type
1 adult, no children 31% 30% 28%
1 adult, with children 40% 11% 7%
2 adults, no children 22% 22% 32%
2 adults, with children 31% 32% 29%
Other 34% 5% 4%
Tenure prior to buying
With parent/relative 24% 20% 28%
Social renting 43% 17% 10%
Private renting 35% 45% 36%
Owner occupier 20% 13% 22%
Other 31% 5% 5%
Household income
Up to £20,000 49% 18% 8%
£20,001 to £30,000 34% 30% 24%
£30,001 to £40,000 30% 26% 25%
£40,001 to £50,000 23% 15% 21%
More than £50,000 17% 12% 23%
Purchase price
Under £100,000 38% 12% 6%
£100,000-149,999 30% 39% 30%
£150,000-199,999 20% 33% 43%
£200,000 or more 21% 17% 21%
Property type bought
House 28% 70% 80%
Flat 37% 25% 19%
Other 52% 4% 2%
Property size bought
1 or 2 bedrooms 37% 41% 31%
3 bedrooms 28% 43% 49%
4+ bedrooms 25% 16% 21%

1 Buyers in category who are additional as a % of all buyers in category.
2 Buyers in category who are additional as % of all additional buyers.
3 Buyers in category who are non-additional as a % of all non-additional buyers.

Supply-side additionality

Consideration of supply side additionality was confined to HtB. This was partly due to the small numbers of NSSE sales at the Scotland wide level and partly because the survey of developers was focused on those participating in the HtB scheme. The main elements to our assessment of supply-side additionality were based on:

  • The views of developers on the impact of shared equity schemes on new build development brought forward and the ability of households to access the housing market; and
  • Applying our demand-side additionality estimate to new build output over the period of the shared equity schemes, in order to estimate the proportion of new build completions that can be considered additional.

The developer survey found that:

  • 21 of 26 developers agreed that HtB had increased demand for new build, rising to 24 of 26 who agreed it had increased sales,
  • 13 of 26 concurred that HtB had enabled them to build in new locations whilst 11 of 26 said it had enabled them to deliver larger developments.

Qualitative findings indicated that developers typically felt that in the initial years, HtB had played an important role in unlocking buyer demand by supporting access to lending and boosting the confidence of developers to bring developments forward. In contrast, developers felt that HtB had played a smaller if still significant role in sustaining consumer demand in the last two to three years, although it continued to have a positive impact on developer confidence, partly by easing cashflow constraints. In particular, developers said that HtB had enabled sites to be built out more quickly and that the resulting improved cash flow had boosted their pipeline programme.

In terms of impact on the wider housing market, none of the developers that completed the survey felt that HtB had increased new build sale prices, although these responses appeared to be influenced by the lowering of HtB price cap in 2014 and again in 2016.

Developers and other participants were not always able to offer a view on impact of HtB on local house prices and the housing market in general. Amongst those who did express a view, HtB was felt to have exerted downward pressure on new build house prices in some localities where properties built for the entry level of the market would otherwise sell at around £200,000-£250,000. In such instances, developers said they had to make a decision whether to bring new units under the price cap to ensure demand.

In terms of the perceived impact on built form, construction prices and land prices:

  • Only 25% of developers indicated that HtB had affected the property type and size profile of their developments. This was supported by qualitative findings, which indicated developers were mainly building semi-detached and terraced houses for the HtB market, especially outside of city locations.
  • Some suggested that the current £200,000 HtB price cap had started to limit options for houses in last 1-2 years, particularly in higher value markets - in those areas, some suggested that flats are the only feasible option.
  • Few developers had looked to develop alternative products for the lower end of the market either before the crash or in an attempt to unlock the market before HtB was introduced.
  • There was little perception that HtB had played a significant role in inflating land or construction costs. Aside from the reality that HtB units tended to have been built on sites developers had already acquired, those who expressed a view pointed to the difficulties of separating out the impact of HtB from the impact of the Scottish Government's Affordable Housing Supply Programme on the price of new land acquisitions and construction costs.

The views of developers suggest that HtB has generated supply-side additionality benefits. On the other hand, the qualitative nature of this evidence does not support the estimation of supply additionality. Indeed, it remains very difficult to place a figure on the extent to which HtB sales have been additional in spite of the fact they comprised a large share of new build sales, especially in its first 3 full years of operation (see chapter 4). As with demand, it had initially been hoped to examine whether "hard" data supported the views of developers and other stakeholders. For the most part, however, small area house price and construction data would be required to assess whether HtB supply had been additional and had not crowded out other development. In the absence of access to such data, we drew on survey results to derive an estimate of supply additionality.

Our analysis involved combining the estimated demand side additionality rates for HtB with data on the volume of new build sales delivered through the scheme. This measure does not take account of, for example, 'indirect' additionality generated through increased developer confidence in bringing forward development. Likewise, it does not take account of any negative effects such as HtB displacing private sector development that would have occurred without the scheme, or whether the scheme has inflated prices thereby reducing access for other prospective buyers.

On the basis of our 'central' demand-side additionality estimate for HtB (20%), we estimate the direct impact of HtB on new build supply at 5% of all private new build completions from 2013/14 to 2018/19. This is equivalent to around 3,000 new build sales.

Table 11: Modelling supply-side additionality
Number(% of all new build)
All private new build completions 2013/14 to 2018/19 64,118
Help to Buy transactions 2013/14 to 2018/19 15,030 (23%)
Central estimate of 'additional' Help to Buy (20% of all Help to Buy transactions) 3,006 (5%)

Housing market indicators of additionality

In order to contextualise and these estimates, further analysis of housing market activity was carried out in relation to a number of key market indicators.

To begin, we looked at whether HtB had led to a significant shift in housing construction away from properties above the price threshold, towards properties below the threshold. Figure 43 measures changes in new build transactions by selected upper price thresholds from 2012/13 (index 100), the year prior to the introduction of HtB. It shows that:

  • There has been a growth in new build transactions across the price distribution but the growth in sales has been much more pronounced above £250,000 than in sales with a value of under £250,000.
  • Since 2016/17 new build transactions under £230,000 have fallen back slightly, suggesting developers are increasingly developing larger and more expensive homes, in more desirable areas or a combination of both.
  • New build sales suggest that lowering of the HtB price threshold in the period to 2014/15 -2016/17 may have encouraged more development at levels close to the HtB thresholds that might otherwise been case.
Figure 43: Growth in New Build transactions in Scotland, 2012/13 to 2018/19
Figure 43: Growth in New Build transactions in Scotland, 2012/13 to 2018/19

Source: Analysis of residential property transactions recorded by Registers of Scotland.

One anticipated consequence of the lowering of the price cap would be the bunching of new build transactions around the price threshold. To explore this further, Figure 44 over the page looks at the share of new build transactions in selected price bands over the period since 2012/13. It shows:

  • An increase in number and share of sales in the £175,000-£200,000 price band in the last 2 years, consistent with lowering of the price threshold to £200,000. However, transaction levels in the adjacent price bands suggest this has at least in part been driven by a decline in sales under £175,000.
  • Sales in the £230,000-£250,000 price band declined and £200,000-£230,000 sales increased coinciding with the short period where the price cap was £230,000.
  • Comparison of new build sales and open market resales indicates that the proportion of new build sales under £200,000 decreased from 58% in 2012/13 to 39% in 2018/19. In contrast, the proportion of resales under £200,000 fell by only 5 percentage points over the same period. This suggests that new build transaction prices have been increasing rather more rapidly than resale prices, although some of this increase is likely to be due to changes in the mix of new homes being transacted.

Housing market trends also indicate that new build sales as a share of all sales over the HtB price cap have typically increased at a more consistent rate than the share under the price cap. This is evident in consistent growth in new build sales as a share of all sales within the £250,000-£275,000 price band. Figure 45 also shows new build sales as a share of all sales under and over the HtB price cap, taking account of the reduction in the price cap between 2012/13 and 2016/17. This shows a small increase in the share of new build sales over the price cap (from 20% of all sales prior to 2016, to 24% from 2016/17 onwards), alongside a fall in the relative role of new build sales under the price cap (from 10% of all sales prior to 2017, to 7% in 2018/19). That suggests that the new build market not targeted by HtB has been performing similarly, and potentially more strongly, than the section of the market that has been supported by HtB.

Figure 44: Trends in new build sales volume, as a share of all sales
Figure 44: Trends in new build sales volume, as a share of all sales

Source: Analysis of residential property transactions recorded by Registers of Scotland.

Figure 45: Trends in new build sales as a share of all sales, up to HtB price cap
Figure 45: Trends in new build sales as a share of all sales, up to HtB price cap

Source: Analysis of residential property transactions recorded by Registers of Scotland.

Contact

Email: socialresearch@gov.scot

Back to top