Memorandum by Professor Glen Bramley (AH
89)
INTRODUCTION
This evidence has been prepared in some haste
because by some oversight I was not aware of this Inquiry and
call for evidence until the day before its deadline for submission.
However, I felt that it might be useful if I were able to pass
some comment on some of the questions raised in the Press Notice
of 11 October, as I have been engaged in research on some of these
questions over the last 15 years or so. This research has been
supported at different times by a range of bodies, of which particular
mention should be made of the Joseph Rowntree Foundation, ODPM,
Scottish Executive and the Housing Corporation. While grateful
for their support I must emphasise that of course the views and
interpretations expressed are my own.
In view of both the shortage of time and the
Committee's expressed wish for brevity in submissions, I have
endeavoured to keep these comments brief and to the point. However,
in view of the wide-ranging scope of the Committee's questions,
I realise that I have not succeeded. ("If I had had longer
I could have written something shorter"). References are
made as appropriate to fuller published output from the relevant
research.
BENEFITS OF
AND SCOPE
TO PROMOTE
GREATER HOMEOWNERSHIP
Much has been written on this topic over the
last three decades, some theoretical, some more pragmatic, much
tinged with ideology and much again tinged with special interests.
I shall not attempt to review this vast literature, but merely
point to some particular points on which I can offer some evidence.
This question really divides into two parts,
the first on benefits and the second on means of promotion. The
first part overlaps with the Committee's second question, about
tackling inequalities and poverty. It is widely and rightly observed
that homeownership is a popular aspiration, and has become more
so over time. However, this is not a particularly good reason
for the Government to promote this tenure, (a) because if it is
popular promotion is unnecessary, and (b) because there should
surely be some wider "public good" benefit to justify
public subsidy and policy effort. Also, there are potential dangers
from excessive public promotion of home ownership, if for example
this leads to people of modest means being exposed to excessive
risk of indebtedness, or to take another example if it simply
serves to inflate house prices even more.
What is the basis for arguing that there is
a wider public good to be derived from greater homeownership?
One long-standing argument is that homeowners have a greater incentive
to look after, maintain and improve their homes, thereby helping
to ensure that the major asset of our housing stock is sustained
and enhanced rather than being degraded. There is a sound economic
logic here, but this should not be overstated. There is a good
deal of evidence, from official surveys of house condition and
more targeted surveys, the homeowners often do not have a good
appreciation of the structural condition of their properties and
do not necessarily prioritise work which is in the best long term
interest of maintaining good structural condition, partly because
of incentives within the housing market. The Government's reform
of the house purchase process (Home Sellers Packs, etc) is partly
intended to address this problem.
A second line of argument, which is receiving
increased attention, is that homeownership promotes attitudes
and behaviour which are more consistent with desirable social
outcomes in terms of, for example, stability of local communities,
engagement with local community affairs, and more orientation
to future goals for themselves and their children. The rather
loose term "social capital" has come to be associated
with this view, and it has been argued from a range of evidence
that homeownership is associated with more positive manifestations
of social capital. One specific aspect of this is the role of
social capital in supporting local schools and education, and
this is something about which we have carried out some recent
research which may be of interest. Our study (Bramley & Karley
2005b and c) investigated the role of homeownership in affecting
educational attainment, alongside the many other factors which
are known to be influential, including individual and neighbourhood
poverty. Drawing on three distinct strands of evidence we concluded
that homeownership does have an independent, positive influence
on school attainment, even after allowing for these other factors.
Children of homeowners themselves do better, even allowing for
poverty, parental qualifications, etc; but there is also a significant
effect from schools having a greater proportion of homeowners
(as well as a lower proportion of children from poor households),
and this can spill over to benefit children whose parents are
not themselves homeowners. In view of the central role of educational
achievement in promoting a much wider set of social goals of inclusion,
employment, reduced crime, etc this finding is of some significance.
A third and closely related line of argument
concerns neighbourhood social mix. Tackling neighbourhood concentrations
of poverty and deprivation is a central part of the Government's
overall anti-poverty strategy, and this recognises that such concentrations
can worsen a range of social outcomes (of which education is but
one). Housing policies and social trends of the last 30 years
have led to a situation where social rented housing is a relatively
"residualised" tenure which contains a predominance
of households on low incomes, a majority of whom are not in work.
Some would say that this is partly a byproduct of the promotion
of homeownership, but other factors are involved as well and whatever
the reasons it is now a deeply-established pattern (Stephens et
al 2005). It follows that we should be using our housing,
planning and regeneration policies to promote socially mixed communities
rather than reproducing estates which are segregated by tenure,
income and demography in the way that past policies tended to
do. One part of such policies is that, when regenerating existing
social housing estates, there should be active promotion of owner
occupation on a significant scale within these areas, both through
new building but also through sales of existing stock. Specific
low cost home ownership schemes, like shared ownership, have a
role to play in this. Another part of such policies is the requirement
that most new housing schemes promoted by the private sector should
be required to contain a proportion of "affordable housing",
as is increasingly achieved through the use of s106 planning powers.
We are currently investigating neighbourhood-level
housing market outcomes across the country in a project supported
by the Joseph Rowntree Foundation, with a particular view to assessing
the impact of different types of new housing investment. Preliminary
findings support the contention that new private investment improves
outcomes in lower demand and poorer areas; the impact of new social
investment is less clear (Bramley et al 2005). This partially
confirms findings of an earlier unpublished study looking at greater
Glasgow, where new private housing was clearly associated with
positive outcomes in a way which new social housing was not.
The scope to promote greater homeownership is
as noted above a different question. This is partly a normative
question, about the socially optimal level of homeownership; partly
a forecasting question, about over what time horizon and under
what conditions this might be achieved; and partly a question
of means and mechanisms which might help to achieve it.
I think that, for the kinds of reasons given
above, it is reasonable to argue that there is a socially optimal
level of homeownership which is rather higher than the current
level in England. However, I certainly do not think that this
level is anything like 100%. There many people in our society
for whom homeownership is not the right solution, for example:
younger mobile people who have not settled down to a particular
career in a particular place and in a particular relationship;
people with insufficient level or stability of income to afford
and sustain ownership; people with other debts and liabilities;
people with a range of vulnerabilities or problems which unsuit
them to the responsibilities of ownership, including some older
people. Beyond this, the sustainable level of homeownership depends
on certain structural economic conditions, particularly the overall
distribution of income and wealth and the supply and price of
housing. Broadly speaking, a more even distribution and a more
generous, affordable supply of housing opportunities would increase
the quantitative scope for homeownership. Notwithstanding the
Government's efforts to reduce poverty, the overall distribution
of income is now much more unequal than in the 1970s, and powerful
forces continue to lead towards greater inequality despite the
impact of certain policies intended to moderate this. The position
on housing supply and cost will depend substantially upon whether
the agenda set by the Barker reports (Barker 2003, 2004) is capable
of being implemented.
Given this big picture, Government can do a
certain amount to support additional homeownership at the margins.
I do not think that blanket subsidies on the demand side are the
way to do this. As is clear from economic analysis, from the evidence
reviewed by Barker on supply inelasticity, and from specific studies
including my own (Bramley 1993), a general subsidy (like the former
mortgage interest tax relief) would mainly have the effect of
raising prices. I think that there is considerable merit, however,
in more targeted assistance to marginal first time buyers, of
the kind represented by the Government's various programmes of
"low cost home ownership" (shared ownership, Homebuy).
In various evaluation reports (Bramley et al 2002, 2005) we showed
that: (a) there is substantial potential demand in this so-called
intermediate sector, particularly in the south of England; (b)
such subsidised provision can be good value for public money in
the longer term and good value for purchasers; (c) such schemes
can be reasonably well targeted; (d) they can also fit well with
some approaches to new provision using planning powers and regeneration
schemes. Past provision has been insufficient in scale to really
do much more than dent the problems of affordability and access
to an inflated market, and has not always been as good VFM as
it might. I think that possible models involving lender-financed
equity loans have more potential than has so far been realised,
and that the Government may have missed part of this opportunity
through a reluctance to adopt what I identified as the optimal
model involving time-limited revenue subsidies (Bramley 2004).
I understand that the Committee is proposing
a further inquiry on this topic, which will follow on from a current
NAO study, so I do not pursue this issue here.
Another form of low cost home ownership assistance
is the Right to Buy/Right to Acquire for social sector tenants.
This has been extremely controversial over the years. However,
consistent with the views given above, I would support these policies
subject to (a) the discounts not being set so low as to represent
a significant cost to the public purse, and (b) the discounts
not being paid for by effectively "raiding" the balance
sheets of social housing providers.
Another form of support to the margins of home
ownership which would be potentially significant is the issue
of "safety nets" for home owners who experience a fall
in income and the threat of repossession. Others such as Janet
Ford and Steve Wilcox at York University have investigated this
subject much more extensively than I have. In short, a more adequate
safety net, whether based in insurance, social security benefits
or a widened housing allowance, would reduce the risks of homeownership
for a substantial group of households who might otherwise be (rightly)
deterred from entering such a commitment.
It is also worth mentioning the financing of
students in higher education in this context. The shift towards
higher participation in HE combined with more use of loan finance
is a significant factor behind the more cautious approach of young
adults towards early homeownership.
ECONOMIC AND
SOCIAL IMPACT
OF CURRENT
HOUSE PRICES
This topic was reviewed very well in the Barker
(2003, 2004) reports, which identified a range of adverse impacts
on macro-economic stability, on the labour market and economic
growth, on housing quality and general levels of welfare, on key
worker availability for public services, on homelessness and housing
need, and on widening gulfs in the distribution of wealth. These
arguments are all valid and telling. Yet in a nation in which
a majority already own their homes and have a stake in high and
rising values, it is difficult to marshall a consensus in favour
of policies which would bring about lower price levels and a lower
rate of future capital growth. This is seen at local level, in
opposition voiced to planned new housebuilding, or at national
level in concerns about triggering an economic recession or about
tax issues such as Council Tax revaluation or Inheritance Tax.
One of the most obvious impacts of higher house
prices is on the affordability of and access to home ownership,
something which I have researched extensively. Recent work (eg
Bramley & Karley 2005) suggested that about 41% of younger
new households could afford to buy a home in England (based on
income) in 2002; ranging from 20% in London to 55% in northern
regions. Allowing for access to family wealth increased these
figures by about 4-5%. By 2004 these figures had deteriorated
by about 7% points due to the continuing rise in house prices
ahead of inflation and incomes. Translating these into annual
need numbers, the 2002 baseline assessment suggested a need for
100,000 additional affordable housing units per year, mainly in
London and southern England. The deterioriation by 2004 would
have increased these numbers by 15,000. As a ready reckoner, raising
prices by 25% reduces the percentage of younger households able
to buy by 12% and increases need by 24,000 units, with a generally
greater impact in the southern English regions.
These need numbers are large, much larger in
total than the current subsidised housing programme. I comment
further on this issue below, but this does raise the question,
what happens to these households who appear to be priced out of
the market (and judging by recent data on the falling number of
first time buyers, are actually being priced out)? There is a
mixture of impacts. A lot of these households have moved into
or remained longer in the private rented sector. While this sector
has expanded to some extent with "But to Let", there
will also have been some displacement of poorer households, who
are more likely to end up homeless. It is clear that homeless
numbers are positively correlated with the strength of the housing
market and we have seen significant increases in recent years,
particularly in London and the south. There will also be a tendency
for new household formation to be delayed and for people to have
to share with other households. One remarkable finding from the
2001 Census was that the average household size in London actually
increased over the previous decade, an almost unprecedented trend.
Another clear byproduct of unaffordable housing, in London and
other high demand areas, is longer commuter journeys. My affordability
model shows the logic of this clearly: a third of the net affordable
need could be met by moving to the cheapest adjacent district.
Clearly, mass longer distance commuting contributes to travel
congestion (on both road and rail), unsustainable levels of travel
and associated emissions and pollution, and general stress on
health and family life.
RELATIONSHIP BETWEEN
HOUSE PRICES
AND HOUSING
SUPPLY, AND
SCALE OF
HOUSING DEVELOPMENT
REQUIRED TO
INFLUENCE HOUSE
PRICES
These two questions, which should be considered
together, go to the heart of the challenge posed by the Barker
inquiry. Barker argued that a very large increase in housing supply
would be needed to achieve her desirable trajectory of house prices
increasing at a rate more in line with the European experience
(1.1% per year, rather than 2.4% per year increases in real terms).
My affordability projections for the Inquiry confirmed that you
would need a reduction of this order to achieve a sustained improvement
in affordability (Bramley & Karley 2005). However, the question
of how much extra housing supply would be needed to create this
virtuous scenario for house prices remains one where the evidence
base, in terms of robust economic models of the housing market,
is rather uncertain. Barker itself relied on some studies by Geoff
Meen which were essentially national and where some of the key
parameters were subject to considerable uncertainty. Subsequently,
the ODPM commissioned a team led by Geoff to build more detailed
models at regional level to quantify the relationship between
housing supply levels and the specific affordability target proposed,
expressed as the ratio of lower quartile house prices to lower
quartile earnings. My understanding is that these models (which
have not yet been published in full) continue to show that large
scale additions to housing supply would be required, particularly
in the southern English regions.
I have myself conducted several studies which
seek to answer this question, based on economic modelling at a
more localised level and taking more explicit account of the planning
system and land supply. Earlier studies such as Bramley (1993,
1999) and Bramley et al (1995) are interesting in various ways
but limited by the reliance on cross-sectional data and the need
to make quite a lot of assumptions. More recently, with Dr Chris
Leishman, I have developed a model which is more satisfactory,
in that it is based upon evidence of the behaviour of local housing
markets over time as well as space. I refer now to results which
has just been published in the journal Urban Studies (Bramley
& Leishman 2005), but would mention that the model is under
further development and testing at the time of writing and it
may be possible to supplement this in oral evidence.
The results show that this price reduction target
(more than halving the rate of real house price growth) could
be achieved by a 39% increase in private housebuilding. However,
to achieve this, it would be necessary to release much more land,
with 175% extra in the higher demand areas and 100% more in the
intermediate areas, and also with corresponding increases in social
housing output. Because our proposed increases in supply are targeted
in higher demand areas, essentially in the south, they would represent
a larger percentage increase there (67% in the high demand areas,
46% in the intermediate areas).
Meen's work for Barker (2003, p 59) suggested
that output would need to nearly double in order to get price
growth down to 1.1%. Our results suggest that these estimates
are rather on the high side and that it would not be necessary
to increase private output quite that much, at least at national
level. Nevertheless, our model points to a general problem with
trying to achieve this through the existing planning system, which
is that a far larger amount of extra land would need to be designated,
not all of which would be taken up in the short term. This could
exacerbate the problems of urban sprawl and infrastructure funding.
For this reason, I would favour more use being made of "direct
delivery vehicles" like Development Corporations for growth
areas. These mechanisms provide more control over the timing of
output (for example through disposing of land under licence) and
also provide a means of funding and coordinating infrastructure.
The more recent work by Meen, Muellbauer and
colleagues for ODPM uses a mixture of simpler and more complex
models, and applies these at regional level. It would be very
helpful to the Committee if this work could be published and discussed.
There are several modelling issues which merit further discussion
and examination, including whether the emphasis should be mainly
on "stocks" or "flows", how the models are
reconciled with household projections, how allowance is made for
housing quality, and what happens in the rented sectors.
OTHER FACTORS
INFLUENCING THE
AFFORDABILITY OF
HOUSING FOR
SALE
I am a bit sceptical about attempts to intervene
in housing construction methods as a way of promoting affordability.
Fundamentally, the price of housing in the market is determined
by the market as whole, new and secondhand, and current price
levels are far above construction costs, even if one were to agree
that construction costs are higher than they might be. It is the
land component in the price of new houses which is grossly inflated,
and this land component is a derivative of the general price level
of housing. So it is more important to bear down on this general
price level by increasing supply than it is to try to reduce the
construction cost element. If you just reduced construction cost
without increasing supply, the land price element would rise to
fill the gap.
However, it might be argued that innovations
in construction methods might enable supply to be increased without
running into bottlenecks in the supply of skilled labour. This
could be a serious issue if the level of output were increased
by the magnitudes proposed by Barker. However, there are many
unanswered questions about the quality and durability of more
industrialised forms of housebuilding, as well as about how this
fits with the organisation of the industry. Barker reiterates
criticisms of the quality of the new housing product in Britain
and challenges the industry to improve its performance. It has
been argued, rightly in my view, that the industry has not focussed
on competing on quality because the route to profitability has
lain in getting hold of scarce land with planning permission.
This suggests, again, that the key to better quality lies in increasing
overall supply so that there is more genuine competition.
Given the overall supply situation and the likely
persistence of relatively high prices in the foreseeable future,
there will remain considerable scope for other measures to deliver
housing which is more affordable than that for sale at the prevailing
market level. The most important current mechanism in this category
is the use of s106 planning agreements to secure a proportion
of "affordable" units on all housing sites. This affordable
housing can and should be a mixture of social rented (RSL) housing
and low cost home ownership schemes, like shared ownership and
homebuy (equity loan), which provide opportunities to buy at levels
significantly below prevailing market prices. As noted above,
such schemes can be relatively cost-effective in terms of subsidy,
particularly in the longer term, and if s106 is used appropriately
then most of this subsidy comes from the land value rather than
the exchequer. Unfortunately, s106 has not been used as effectively
as it might be, due to too great a readiness to mix it with public
subsidy (social housing grant), delays in negotiation and policy
uncertainties. S106 also contributes to another desirable policy
goal, that of mixed and balanced communities, which I identified
earlier as part of the wider public benefit of owner occupation.
S106 is a form of fiscal measure, albeit not
a conventional one. There is a post-Barker proposal to introduce
a general "Planning Gain Supplement", which would represent
a more generalised tax on development gains, but could serve to
limit the scope for s106 agreements at local level. While the
PGS could generate more funding for infrastructure, there is still
a danger that this could act to discourage supply, to a greater
extent than s.106 which is more directly tied to facilitating
specific developments and which can provide local authorities
with direct benefits and thereby motivate them to approve developments.
In my earlier work I conducted some rather crude simulations of
the market effects of generalised development charges related
to land values (Bramley et al 1995). These suggested that
there would be some level of impact on supply (negative) and on
price (positive), but that this would not be too severe. If PGS
enabled the funding of the critical infrastructure required to
deliver the "step change" in supply called for by Barker,
then this would be an acceptable cost within a package whose overall
effects would be positive. I should add that we may be in a position
to make some further estimates of the impact of such charges within
the model which we are currently updating.
A number of other fiscal measures impact on
the housing market, including Council Tax, Stamp Duty, and less
directly Capital Gains Tax, Inheritance Tax, VAT (partly through
exemptions, eg as in recent debates about Personally Invested
Pension Schemes). I do not offer detailed comments on these tax
issues here, but just make a couple of general points. Taxes which
relate to the level of housing consumption or the return on housing
assets will impact on the demand for housing. So, for example,
tying Council Tax more clearly to housing values, through revaluation
and rebanding, could act as a damper on housing demand. Such tax
changes which act on the demand side, without affecting the currently
very inelastic supply, will tend to impact on house prices. To
the extent that tax changes or concessions are selective, they
may impact on particular local or submarkets (eg PIPS on second
home submarkets in city centres, University towns, attractive
rural and coastal areas). Also, tax changes which act to damp
demand may also restrict affordability, particularly where they
impact on the current cash cost to users. More desirable tax changes
would be those which reduced rates and expectations of future
capital gains (the asset or speculative element of demand), and
which altered the distributional incidence in a more progressive
fashion.
THE SCALE
OF THE
GOVERNMENT'S
PLANS
The Government is expected to announce the scale
of its plans for new housebuilding, revised in the light of the
Barker report and subsequent research and analysis, shortly. Previous
plans have been really inherited from the old approach based on
household projections (and hence somewhat "circular",
as I and others argued—see Bramley & Watkins 1995),
subject to ad hoc adjustments relating to the 2003 Sustainable
Communities Plan. Indeed, I would argue, from the review of evidence
conducted within the context of the "Evaluation of English
Housing Policy since 1975" (Stephens et al 2005), that the
actual supply of land through the planning system was actually
falling progressively throughout the 1990s and into the early
2000s, regardless of what the government's stated plans were.
In other words, you have to consider implementation as well as
plans. Figure 1 attached has been constructed as part of our modelling
update, and provides a very graphic illustration of the problem
which Barker was addressing. Effective land supply more than halved
in the south of England between 1992 and 2003, and also halved
in the midlands. Only in London did supply recover after 1997,
and then not to levels of the 1990-92 period. These data were
not officially monitored by government after 1997.
|
