families-and-households/2011/stb-families-households.html> ... Statistics, accessed online 19 July 2012,
More than just homes
Measuring Moat’s true social value contribution By Brian Johnson, Chief Executive and Angelo Sommariva, Public Affairs and Policy Manager
Moat thought leader essay
INTRODUCTION At a time of limited public subsidy, affordable housing providers face the challenge of proving the true value of their outputs in order to justify continued funding from the state. This scenario has urged many housing associations – including Moat – to make the case that the sector delivers significant value beyond the development of new homes. In recent years, the provision of secondary and ‘added value’ services – that is, activities above and beyond the creation of a physical home – have become increasingly important. This will continue to be the case as the sector intensifies its efforts to encourage greater social mobility among its residents. However, the onus of proof as to the benefit of such spending remains firmly with housing associations, forcing the development of more accurate and robust measurement tools. Moat has carried out significant work in this area, which is summarised in this publication. Although we are able to provide evidence for the majority of assumptions made to date, we consider that this is by no means a finished work, and would welcome further testing.
In any year, Moat’s performance is expressed as the sum of the increase in social NPV of investments plus in-year social value created by service delivery. In order to use this formula, it was necessary to decide which products and services could be accurately measured. These are referred to as ‘strands’ throughout this paper, and are derived from three types of outputs: •
Products, or the creation of a ‘tenure unit’ These items are core to the business of housing associations. They sometimes involve delivering a new building (such as physically developing an Affordable Rent home), whilst other times it involves non-physical products, such as providing a mortgage rescue.
•
Secondary services These are services which cannot be considered core tenure units, but which are necessary to the overall operation and success of the organisation. They may include services which housing associations have a duty to undertake.
•
‘Added value’ services This includes services above and beyond the basics of providing a tenancy through which rent is paid, and a property kept in good repair. An example may involve referring a resident to an employment provider, or providing financial advice.
Intended audiences for this work We have identified the following main audiences for our work on the measurement of social value: •
•
Board and management – This work creates a structure around which Board and management can judge the magnitude of benefit delivered from different products and services. An understanding of how and why investments and activities can deliver value to society is important alongside an understanding of their financial impact on the organisation. Providers of public subsidy – This includes the main arms of government from which subsidy is received, and for whom justification of value for money and financial return on investment is critical.
Our methodology Social value, as measured by savings to, or enhancements of the public purse, can be expressed in two ways: 1.
For investments in buildings, it is expressed as the net present value (NPV) of future benefits. This is because the future benefits are essentially secured at the point of build. The basic assumptions used to calculate this NPV are in line with those in the Green Book and Magenta Book issued by the Treasury around government spending. 2.
Social value created as a consequence of service delivery is expressed as pounds in the year of the delivery of the service.
Following an extensive analysis of the products and services delivered by Moat across each of these areas, we concluded that the following are underpinned by sufficient evidence to make accurate measurement possible: •
The value of discounted rent
•
Homeownership backing up the welfare system
•
Homeownership as insurance against poverty
•
Employment created through building new homes
•
Reduction of fuel poverty
•
Employment advice and/or referral
Many other strands were explored, but discarded on the basis of a lack of evidence. Although we hope to be able to include some of those strands in future, this will require further research to find empirical evidence. Ultimately, we aim to create a framework that would be of benefit not only to Moat, but also the wider sector.
THE MEASUREMENT STRANDS
The following is a summary of all strands – products, secondary services, and ‘added value’ services’ – which we believe are underpinned by sufficient evidence to make accurate measurement possible.
Products
The value of discounted rent This strand is based on the assumption that if subsidised housing tenures were not available, that
residents would be housed in open market rented housing. This, in turn, would lead to a higher proportion of people in need of greater state assistance – through Housing Benefit or an equivalent subsidy. For products offering a rent lower than market rent, the assumption made is that: •
if these products were not available, residents would be housed in open market rented housing, and that;
•
for the proportion who would need Housing Benefit in this situation, there is a reduction in expenditure by the state equal to the difference between the subsidised cost of the housing and open market rents.
Other than the local difference between subsidised rent and open market rent, which is based on real local data for Moat, the other key assumption is the proportion of residents who would require Housing Benefit if they were housed in open market rented accommodation. The proportions assumed for different housing types range from 95% for social/affordable rented, sheltered, and supported, to 35% for shared ownership. The gap between the subsidised cost of housing and the open market cost of housing is assumed to increase with RPI. Housing under-supply in England suggests that the gap is likely, in reality, to inflate faster than this. Once the home is built, the benefit is accrued each year, so the value of discounted rent is given as a net present value into the future. Employment created through building new homes This strand uses the assumption that each new home built by Moat incrementally increases the total homes built in England by one and, as a result, either retains existing jobs or creates new ones. This benefit is especially relevant at a time when the construction sector is depressed. This is calculated by applying analysis undertaken by Savills, which found that building 100,000 new homes creates 228,000 direct construction jobs and a further 228,000 supply-side jobs. This produces a base figure of 4.56 jobs created for every home built.1 This employment creates a saving in welfare benefits that would otherwise have to be paid. This amounts to £12,000 per person-year of employment. In addition to this saving, an increase in tax income of £2,000 per person-year of employment is generated. Based on these figures, the formula for calculating social value is: Number of new homes, multiplied by 4.56, multiplied by (£12,000 + £2,000). Therefore, the social value delivered by Moat in 2011/12 through the development of 415 homes is £26,493,600. This benefit is delivered once only for each home, at the point of building the home. It applies to all tenure types where a new home is constructed.
Homeownership as insurance against poverty This strand looks heavily into the theories of assetbased welfare. Academics John Doling and Richard Ronald, who have written extensively on the subject, suggest that the position of housing in national welfare systems is ‘much more complex than its role simply as providing physical shelter.’2 One aspect of this complexity is the impact that ownership can have on pensions in old age. Our analysis largely relies on the work of professor Francis Castles, who qualified the link between home ownership and the generosity of old-age pensions. His thinking is based on the hypothesis that individuals who own their own homes outright are able to get by on smaller pensions, and are therefore less likely to fall into poverty in retirement. This is known as the Castles trade-off. For a large percentage of owners, the process of home purchase is complete by the time of retirement. This creates a net benefit equivalent to the rent otherwise payable on the property, minus outgoings for maintenance and property taxes. In other words, when individuals own their own homes, they can ‘get by on smaller pensions.’3 This hypothesis does, however, come with a warning: asset-based welfare can only work when a critical mass of people own their homes in retirement. Australia serves as a good example of how this can work. Even though state pensions for older Australians are low by international comparisons, over 80% of retirees own their homes (with most being outright owners). This largely mitigates the potential for higher levels of poverty among this group.4 Our intention from this strand is to estimate the number of people who own a home as a result of Low Cost Home Ownership (LCHO) schemes vs the number of these people for whom ownership would otherwise not have been possible (without these schemes). Based on our experience in delivering LCHO products, we have estimated that this is true in at least 50% of cases. Although we are confident that this figure is a reasonable estimate, we intend to further test this to ensure that the margin of error is further reduced. We then use a further assumption to determine outright ownership. Therefore, of the 50% of people described above, 70% are assumed to staircase to 100% by retirement age. This percentage was also formulated based on Moat experience with LCHO products. In coming months, we expect to refine this figure and base it on a yearly-revised percentage. It is also worth noting the following points. Firstly, the benefit derived from this strand flows primarily to the individual homeowner rather than to the state. The state is an indirect beneficiary, as the burden of higher pensions and other welfare payments is significantly reduced among the homeowner group. We therefore treat this strand as insurance against the risk of
poverty, high state pensions and possible welfare dependency.
to the NHS has been worked out using the following calculation: £14.4 billion divided by 26.3 million households8 equals £547.52 per household. This results in a combined cost of £796.47 per household.
Secondly, current LCHO products have not been designed in such a way as to ensure staircasing to 100% by retirement age. We are therefore unclear whether current products will accurately reflect the 70% staircasing assumptions. What is clear is that future products will have to be engineered to ensure full ownership by retirement age; they are therefore likely to exceed 70% ownership levels in the long-term.
Our methodology then seeks to place a value on the risk of a household falling into fuel poverty, as this allows for a savings per household calculation. In order to do this, pre-existing definitions on fuel poverty and the Standard Assessment Procedure (SAP) ratings framework have been used. We have estimated that the highest possible risk reduction (expressed as a percentage), achievable by an SAP rating of 100 (the maximum number possible), is 20%. This seems entirely realistic given that a home with an SAP rating of over 90 is designed to remain within the ‘safe temperature’ range under normal winter temperatures, even without the aid of heating.9
Homeownership backing up the welfare system This strand looks at the potential benefits created through the release of equity in retirement. It will be measured on the basis that outright homeownership delivers the ability to release equity, which can then be used to fund care, support or income needs – therefore reducing the burden on the state.
Using an SAP rating of 50 as the neutral value, the maximum saving per household which could be attributable is £159.60 per annum (20% of £796.47 for an SAP rating of 100). Equally, an SAP rating of 0 would incur a loss of £159.60 per annum. The aim for housing associations should therefore be to improve the SAP rating in order to derive a higher a social value figure in pound terms. For example, an average SAP rating of 50 would derive a social value of £0pa per household, whereas a rating of 75 would derive around £80pa per household. This calculation is illustrated below, using actual Moat data (with an average SAP rating of 60 across 6861 of its properties).
The assumptions associated with this benefit are that 30% of residents will release 50% of the capital in their homes 30 years in the future. This is calculated as a net present value on this basis. This applies to shared ownership and shared equity products only.
Secondary services
Reduction of fuel poverty Much has been written on the need to alleviate fuel poverty across the UK, and a great deal of public money has been spent on initiatives with this aim. It is considered that fuel poverty leads to significant problems associated with negative health outcomes and a poorer quality of life.5
Moat’s SAP rating of 60 means that there is an average saving of £31.92 per household. Multiplied by 6861 households, the total social value delivered per annum equals £219,003.
This strand draws on the evidence that people in fuel poverty are considerably more likely to experience cold indoor temperatures in their homes. This leads to a heightened risk of respiratory and cardiovascular disease – both of which have been evidence-linked.
Value-add services
Employment advice and/or referral In order to measure this strand, it is necessary to collect accurate data on residents who find work following the provision of employment advice. Where housing associations provide such services, it is typical to collect data for a period of several weeks following
The cost of respiratory disease per UK household is worked out using the following calculation: £6.6 billion6 divided by 26.3 million households7 equals £250.95 per household. Likewise, the cost of cardiovascular disease
SAP rating 0
50
60
100
% Saving Upper negative impact limit
Neutral
Upper positive impact limit
Saving pa / per household (£) £159.60
£0
£31.92
£159.60
work placement (typically a minimum of 13 weeks). This is also standard across the welfare-to-work sector, where the new payment-by-results structure has forced providers to tighten data collection for customers placed into work. Where a housing association enters into partnership with an employment provider to offer a referral service, it will be necessary to obtain data from that provider on referred residents in order to accurately measure this strand. As job outcome payments are first made by DWP after 13 and 26 weeks in employment,10 providers will have accurate data on customers in employment for this period of time. An assumption will then be made regarding the saving to the Treasury of moving an individual resident into employment – derived from benefit savings. This figure is currently placed at £12,000 per person-year of employment (the same figure used for the Employment created through building new homes strand). In addition to this saving, an increase in tax income of £2,000 per person-year of employment is generated. This benefit is delivered once only for each person referred, and who finds employment, and who remains employed for 13 weeks. At the 13 week stage, as the drop-out rate reduces dramatically, we will assume that that person will remain in employment for a full year. The formula for this strand can be illustrated as follows: Number referred and employed for 13 weeks, multiplied by (£12,000 + £2,000). Therefore, if Moat successfully refers and finds employment for 30 residents in a given year, this will equate to a social value contribution of £420,000 in that year.
CONCLUSIONS AND OUTCOMES TO DATE Using the measures that we are comfortable with to date – that is, all strands outlined in this publication – Moat’s social value contribution in 2011/12 was £55.2m for a capital grant input of £28.8m. This is the sum of in-year value derived from activities and NPV of future value generated through asset creation. Recycled capital grant was also treated as if it were new grant. We consider that our thinking is much further developed on the value of investments side. In the areas of service delivery, it is hoped that some pieces of work underway will create more robust means of quantification. Some highlights from Moat’s numbers for 2011/12 are as follows: •
The savings in Housing Benefit payments to Treasury equal £21.2m through the creation of subsidised rented housing. Perhaps unsurprisingly, this number is of a similar magnitude to the capital subsidy used for this housing in 2011/12 (£26.2m). With a harder line
now being taken on not putting subsidy into S106 homes, it is anticipated that this equation will look more positive in 2012/13. •
The future savings created through LCHO products is estimated as £6.0m compared to a subsidy input of £2.6m.
•
In addition to the above, Moat’s development programme as a whole generated up to £26.5m in welfare savings and taxation revenue through the creation or maintenance of employment. The importance of this benefit is heightened during a downturn.
•
The service work that resulted in 40 people going into employment (on the basis of employment for 1 year) brought £0.6m benefits.
Above all, the discipline of having to think through carefully where further social value might be delivered, and working through the numbers associated with this, has already proven useful and constructive for Moat.
Footnotes Savills, The Case for Housing: Impact of Housebuilding on the UK Economy, Market Insight, spring/summer 2010, p.3. 1
John Doling and Richard Ronald, Home ownership and asset-based welfare, 8 December 2009, p.2.
2
Caroline Dewilde and Peter Raeymaeckers, Castles’ trade-off revisited: Individual and institutional determinants of old-age poverty, Research Unit on Poverty, Social Exclusion and the City (OASeS), Department of Sociology, University of Antwerp. 3
John Doling and Richard Ronald, Home ownership and asset-based welfare, 8 December 2009, p.6.
4
John Hills, Fuel Poverty: the problem and its measurement, interim report of the Fuel Poverty Review, October 2011, p.9. 5
Only primary care costs have been included in this calculation as the secondary care cost figures are less robust.
6
Office for National Statistics, accessed online 19 July 2012, 7
Office for National Statistics, accessed online 19 July 2012, 8
The World Health Organisation considers an indoor temperature of between 18-24°C to be a safe temperature range. This information outlined in more detail in John Hills, Fuel Poverty: the problem and its measurement, interim report of the Fuel Poverty Review, October 2011, p.84. Various sources including WHO materials were used by the Review. 9
Department for Work and Pensions (DWP) press release, Grayling: first Work Programme data shows promising signs, 9 July 2012, accessed online 15 July 2012 10
www.moat.co.uk September 2012
