Supplementary memorandum by the Office
of the Deputy Prime Minister (DEC 01(b))
1. INTRODUCTION
1.1 At the Oral Evidence session on 9 December
ODPM agreed to provide the Committee with information on the following:
— spend and outcomes between 1997 and
2001, and 2001 to 2004; and
— the outcome of the exchange between
ODPM and the PAC on the value for money in stock transfer arising
from the PAC report on transfer.
2. SPEND ON
PFI TO DATE
2.1 The Housing Revenue Account Private
Finance Initiative in social housing has had two bidding rounds,
a pathfinder round of eight schemes, and a second round of eight
schemes—totalling an allocation of £760 million PFI
credits. A third bid round has recently been announced, with a
total allocation of £600 million PFI credits.
2.2 Two schemes have been signed to date—one
in Islington and one in Manchester. The total spent to date by
these two HRA PFI schemes is £4,616,292, (£3,520,356
in Islington out of a total allocation of £74.7 million and
£1,095,936 in Manchester) from their allocation of £36
million which covers a 30-year period.
3. SPEND AND
OUTCOMES BETWEEN
1997-2001 AND 2001-04
3.1 Capital investment by local authorities
covers work not only to reduce the number of non-decent homes
but also work to prevent other homes falling into non-decency
and works outside the decent homes standard such as environmental
works. During 1997-2001 total capital spend by LAs and RSLs was
£10 billion adjusted to 2003 prices using the BCIS index
of building cost inflation. From the English House Condition Survey
we estimate that this reduced the number of non-decent homes by
410,000 with a further reduction of 90,000 from Right to Buy sales.
During that period the average annual spend was £2.5 billion
per year, resulting in an average annual reduction of around 102,000
non-decent homes per year.
3.2 For the period post 2001 we only have
outturn data to March 2003. In 2001-03 LAs and RSLs spent a total
of £5.6 billion on their stock (adjusted to 2003 prices).
LA and RSL returns reported that this delivered a net reduction
in the number of non-decents of around 260,000 with a further
50,000 reduction from Right to Buy sales. The average annual spend
is £2.8 billion resulting in an annual reduction of 130,000
non-decent homes per year.[13]
This is expected to increase as the Arms Length Management Organisation
funding comes on stream. ALMO expenditure is expected to increase
to around £350 million in the current year, compared with
£56 million in 2002-03, with further rises in later years
subject to the timing of ALMOs passing inspection.
4. VALUE FOR
MONEY IN
STOCK TRANSFER—ODPM
RESPONSE TO
THE PAC REPORT
ON TRANSFER
4.1 The National Audit Office report, "Improving
Social Housing Through Transfer" stated that delivering improvements
through housing transfer provided poorer VFM than carrying out
the works under local authority ownership. This was based on a
hypothetical scenario. The figures which resulted in this conclusion
assumed a more rapid increase in rents post-transfer, resulting
in higher Housing Benefit cost. They also assumed higher borrowing
costs for RSLs than those borne by local authorities, and they
included the transaction costs involved in setting up a transfer.
They assumed that both the RSL and local authority would deliver
identical levels of performance and improvements for the same
amount of resources invested, with no allowance for efficiencies
in management costs.
4.2 ODPM provided the PAC with an updated
analysis using the same scenario but using the revised methodology,
set out in the 2003 Treasury Green Book "Appraisal and Evaluation
in Central Government". This uses a real public sector discount
rate of 3.5% and requires risk, in particular "optimism bias"
to be accounted for separately.
4.3 When higher and lower optimism bias
factors were applied to the transfer of 1 million homes over five
years, the impact of transfer to the government, compared to LA
renovation and ongoing investment, was a saving of £4.5 billion
and a cost £0.5 billion respectively.
4.4 Based on these figures, and taking into
account the non-quantifiable benefits of housing transfer, such
as greater tenant involvement, wider regeneration activities and
transfer of risk from the public to the private sector, ODPM is
content to reaffirm the VFM of housing transfer.
4.5 The full text of what ODPM provided
is in the attached extract from a Supplementary Note that ODPM
provided to the PAC following the Select Committee hearing. This
note sets out revised calculations of the vfm of transfer and
was accepted by the PAC.
Annex A
VFM OF STOCK TRANSFER—APRIL 2003
INTRODUCTION
1. This paper assesses out the value for
money of the stock transfer programme taking into account guidance
in the 2003 Treasury Green Book "Appraisal and Evaluation
in Central Government" which took effect on 1 April 2003.
2. The Green Book's revised guidance is
based on a comprehensive review of the appraisal methodology.
It separates out the discount rate and the risk factors associated
with any option. It has a real public sector discount rate of
3.5% which represents the social time preference rate and requires
risk, in particular "optimism bias", to be accounted
for separately. Therefore with a lower discount rate, it is now
essential that other appraisal issues, which would otherwise have
been accounted for implicitly by using a higher rate, are dealt
with explicitly.
OPTIMISM BIAS
3. Optimism Bias (OB) is the tendency for
a scheme's costs and duration to be underestimated and/or benefits
to be overestimated. At an early stage of an appraisal, the Green
Book suggests that maximum optimism bias uplifts for particular
cost and income elements (capital expenditure, rents, etc) should
be applied. OB assumptions should then reduce as the scheme firms
up. For the purposes of the stock transfer the Office has assumed
any assessment is just prior to transfer. At this point optimism
bias for transfer is assumed to be zero, as the transfer price
received by the public sector is fixed and there should be no
changes to it in the future. However for direct investment optimism
bias will exist as there remains scope for change to the cost
to the public sector into the future. For this exercise two scenarios
are considered:
— where the level of risk implied by
the New Green Book is not mitigated—ie optimism bias is
at an upper bound—the higher scenario;
— where some actions are taken to mitigate
risk—so reducing optimism bias to a level based on judgement
and supported by existing evidence—the lower scenario.
4. These two scenarios are considered across
performance in four areas:
— initial investment and life cycle
costs where for both optimism bias uplifts of 24% and 13% are
assumed based on guidance within the Green Book;
— management and maintenance costs
where evidence from PFI schemes suggests uplifts of 11% and 6%
appropriate given the length of time local authorities have been
managing and maintaining the stock; and
— rent collection where 2001-02 Best
Value Performance Indicator LA rent collection, after voids, suggests
uplifts of 3.8% and 1.0%.
In reality there are likely to be a combination
of factors which will apply to different degrees in each case.
However for this exercise they are assumed to average out.
5. As with previous vfm analyses all costs
and benefits are at present values, over 30 years, comparing the
cost to the government sector of renovation through transfer and
local authority renovation with a base case ongoing investment
by a local authority at current expenditure levels. Again a hypothetical
programme of improving 1 million dwellings over five years is
used. The present value of total investment assumed, using a 3.5%
discount rate, is £45 billion at 2002-03 prices when there
is zero optimism bias.
RESULTS
6. When applying the higher OB factors to
the transfer of 1 million homes over 5 years the cost of transfer
to the government sector is £4.5 billion less over 30 years
than local authority renovation and ongoing investment. When the
lower OB factors are applied to the transfer of 1 million homes
over 5 years the cost to the government sector is £0.5 billion
more over 30 years than local authority renovation and ongoing
investment.
7. The Green Book makes clear that to apply
only the 3.5% discount rate with no OB is only half the vfm calculation,
unless the evidence suggests that risk can be entirely mitigated.
As risk cannot be entirely mitigated in the local authority housing
sector any vfm calculation must make allowance for OB. However
applying only the 3.5% discount rate and no OB to a transfer of
1 million homes over 5 years would give a cost of transfer to
the government sector of £5.2 billion more over 30 years
than local authority renovation.
SOCIAL COSTS
AND BENEFITS
8. In addition the Green Book requires consideration
of the social costs and benefits of a project. It emphasises the
need to identify and value the benefits and on ensuring their
realisation. However in doing so it recognises that some benefits
may be impossible to value or that the process of valuation would
be too costly or too time consuming to be worthwhile. In terms
of delivery there is much evidence of the improved performance
that can be secured through housing transfer which sits alongside
the broader benefits that transfer can be seen to bring in terms
of delivering local sustainability and renewal.
PERFORMANCE EVIDENCE
9. There is a substantial body of research
that examine housing transfer and demonstrate the benefits that
are generated through the transfer process. Some of the more recent
are included. The 2001-02 benchmarking exercise of transfer landlords
by the Chartered Institute of Housing and National Housing Federation
Housemark service showed 81% of tenants being satisfied with their
repairs and maintenance service and 85% finding their landlord
to be good or very good at keeping them informed. The equivalent
figures for local authorities were 74% and 77%.
10. The Office's 2000 evaluation report
of stock transfer showed that:
— management costs are 18% lower on
average in transfer landlords than in local authorities;
— 34% of transfer landlord tenants
said that the management of their home had improved since transfer
which compares favourably against only 16% of local authority
tenants nationally saying that the management of their homes had
improved over the previous 5 years;
— 77% of transfer tenants believed
the rent they paid was good value for money of whom 29% thought
it was very good value for money;
— 85% of transfer tenants said they
were satisfied with their landlord compared to 79% as a comparative
for council tenants nationally;
— 67% of transfer tenants said they
were satisfied with the repairs service compared to 64% as a comparative
for council tenants nationally and
— 38% of tenants said that overall
service had improved since transfer and 34% of tenants said that
they expected the service improve.
11. The National Housing Federation 2002
survey to which 80 transfer RSLs responded demonstrated that:
— transfer landlords relet their properties
1½ weeks speedier than local authorities;
— transfer landlords' rent collection
rates are 1% higher than local authorities;
— an estimated 1.5 million people living
in transferred homes are benefiting from on average £7,400
on investment in their homes since transfer;
— the percentage of repairs done on
time has improved on average by 12.4% after transfer.
12. In considering the benefits of transfer
Ministers have also placed importance on the role housing transfer
has in bringing forward the improvement of sub-standard local
authority housing when public funding could not be made available
and the opportunity it offers to access private finance to support
public services. Transfer has also delivered greater tenant choice
and participation, the separation of landlord and strategic housing
functions and the removal stock from authorities where performance
has been weak.
13. This evidence complements the Green
Book based vfm assessment and further underlines why successive
governments have supported housing transfer as a significant plank
of housing policy.
13 These figures are not strictly comparable with those
for 1997-2001 which are from the EHCS and not individual LA and
RSL returns as LAs and RSLs were not asked to measure decent homes
until after April 2001. Back
| 
