Impact of Demographic Projections on Sub-national Planning
Submitting Institution
University of ManchesterUnit of Assessment
SociologySummary Impact Type
TechnologicalResearch Subject Area(s)
Economics: Applied Economics
Studies In Human Society: Demography, Human Geography
Summary of the impact
Research on demographic projection methods, undertaken at the University
of Manchester (UoM), provided the basis for POPGROUP, a software package
recognised as the industry standard for local demographic planning in
Great Britain and adopted by the Local Government Association (LGA) in
2010. Three impacts emerge from POPGROUP. Firstly, in estimating how
migration will be restricted or stimulated according to the provision of
housing, the software permits local government to assess a range of
house-building scenarios, enabling the effective implementation of, most
recently, the `National Planning Policy Framework' (2012). Secondly, by
extending demographic projections to areas other than local authority
boundaries, it enables national statistical agencies to project
demographic demand (e.g. in National Parks). Thirdly, the ubiquity of
POPGROUP has led to the increasing use of demographic tools within the
commercial sector.
Underpinning research
This programme of research originated in the late 1990s when Professor
Ludi Simpson (1992-2008, now Honorary Professor of Population Studies)
worked in the local authority sector (Bradford Council) and at UoM on the
`Estimating with Confidence Programme' (ESRC/ALCD). At UoM the research
has been led by Simpson, with contributions from Dr Christian Brand
(Research Associate, 2003-2006, 2007-2012), Susan Lomax (Research
Assistant, 2007-2009) and Dr Alan Marshall (Research Associate, 2010-2011,
2012-).
Research on demographic projection methods underpins the POPGROUP
software; developed to forecast population, households and the labour
force, for areas and social groups. Demographic projections
calculate future population size and structure based on assumptions about
future trends in fertility, mortality and migration. The distinctive
nature of the research contribution rests on the improved use of
cohort component models (which use information on fertility,
migration and mortality trends within age-sex cohorts) for strategic
planning, through better integration of population estimates within
demographic projections. POPGROUP was designed by Simpson in 2000 to
implement a cohort component projection model underwritten by six local
authorities. It has been updated twice (v.2 in 2002; v.3 in 2005), with a
further generalised Derived Forecast model issued in 2010, accommodating
forecasts of any population characteristic related to age and sex,
including households and the labour force. A fourth version of POPGROUP is
due in 2014.
Emerging from a consideration of complex forecasting strategies, the
research aimed to produce software of sufficient rigour that would
convince advanced users, yet be practical and functional for
non-specialist staff. Based on Excel, it builds upon users' existing
spreadsheet skills, with alternative assumptions developed as scenarios.
The user remains in control of the area to be forecast, of data inputs,
and of analytical outputs additional to the software's own flexible
reporting and graphical routines. Additionally, software development to
improve quantitative training in POPGROUP methods was supported by the
ESRC (2003-2008). This training recognised that previous software packages
had attempted to fill the same requirements yet had either been focused on
temporary needs, or had become inaccessible through inadequate support.
POPGROUP's methodological innovations include incorporating the
mathematical equivalence of social groups and geographical areas,
thereby satisfying both a longstanding need for local area forecasting
and a growing interest in the social dimensions of population dynamics
as they relate to ethnic diversity. The first version of the
software facilitated user input of base populations, births, deaths and
migrants, and assumptions about future rates of fertility, mortality and
migration, providing tabular and graphical outputs of population structure
and demographic indicators [D]. Subsequent research incorporated
into the software integrated forecasts of population, housing and the
labour force [E], with further methodological innovations implementing
improved algorithms for expectation of life, decomposition of population
and household change, dynamic population pyramids and comparison of
forecast scenarios.
Housing provision and economic development impose constraints on
populations which are also resolved through migration; hence the
translation of migrants into households is a key algorithm in the
POPGROUP approach. This permits a range of complex social factors to
be taken into account when constructing projections: such as the age and
sex composition of migrants and the existing population; the propensity of
residents of different ages to form households; the proportion of the
population in communal establishments; and the proportions of vacant
dwellings, shared dwellings and second homes.
A key feature of the research presented here is that it has been
simultaneously incorporated into the software, and disseminated via
academic outputs [A][E] (alongside updated software manuals).
Specific research enabled by the software includes: an ageing population
and the restricted housing of National Parks [B]; changing ethnic
composition [C]; and estimation of local demographic characteristics [A],
the arena where the social impact of POPGROUP is most evident.
References to the research
(all references available upon request — AUR)
In addition to academic articles, a dissertation and a book, a range of
software manuals were also produced between 2002 and 2011 and are
available via the CCSR website
(http://www.ccsr.ac.uk/popgroup/).
[A] (2011) Simpson, L. and Snowling, H. "Estimation of Local Demographic
Variation in a Flexible Framework for Population Projections" Journal
of Population Research 28(2) 109-127 (Special issue on small area
projections) doi:10.1007/s12546-011-9060-7
[B] (2009) Marshall, A. & Simpson, L. "Population Sustainability in
Rural Communities: the Case of Two British National Parks" Applied
Spatial Analysis and Policy 2(1) 107-127
doi:10.1007/s12061-008-9017-1
[C] (2009) Finney, N. and L. Simpson 'Sleepwalking to Segregation'?
Challenging Myths about Race and Migration. Bristol, Policy Press
(REF 2014) (AUR)
[D] (2007) Simpson, L. "Fixing the Population: From Census to Population
Estimate" Environment and Planning A 39(5) 1045-1057 (RAE 2008)
doi:10.1068/a38141
[E] (2004) Simpson, L. "Integrating estimates within population
forecasts" Applied Population and Policy 1(2) 89-104 (AUR)
Details of the impact
Context: As population projection software has become an
increasingly essential tool in planning and policy formulation within the
UK, POPGROUP has become the leading software for population planning. The
value of the POPGROUP software derives from its flexible framework which
can be tailored to each user's own environment, with impact delivered
through:
- The widespread adoption of the POPGROUP software by government and
commercial planning agencies.
- The recognition of POPGROUP as the industry standard for local
planning.
Users include local and national governments, health agencies, commercial
planning companies and universities, with the number of organisations
using the software increasing from 40 in 2008, to 111 by July 2013 (when
the most recent survey took place) [1][6]. In October 2012 two-year
teaching licences were issued for the first time [1]. The software
continues to develop, with the addition of the generalised Derived
Forecast facility in 2010, immediately utilised to implement forecasts of
long-term illness and disability in 2011 [2]. Additionally, since 2010
Data Modules have been provided to fill demographic models with Government
(DCLG) — and other authoritative projections from within the UK — from
which users can develop in-house `alternative scenarios'. These
developments have led to a range of impacts:
1. POPGROUP has become the industry standard. Now "recognised
as the `industry standard' population forecasting model" [8], the
Head of Population Statistics at the Welsh Assembly notes how it has "contributed
to the development of the demographic knowledge base within local
authorities" [5]. It was adopted in 2010 by the LGA, through an
agreement with the Steering Committee of software users, and is presently
used by 81 Local Authorities [1][6]. Simpson, in collaboration with these
users, has taken the lead in designing each subsequent software
development, and is now technical manager to the Steering Committee.
2. POPGROUP has facilitated the implementation of policy in socially
dynamic contexts. In particular, POPGROUP addresses the UK
Government's `National Planning Policy Framework' (2012) that requires
every local authority to make a `Strategic Housing Market Assessment',
supported by robust demographic projections. This is exceptionally
important, as the co-ordination of such assessments by regional bodies was
scrapped in 2010, as part of the UK Government's `localism' agenda. As the
co-chair of Local Authorities on the Central and Local Information
Partnership notes:
"POPGROUP allows Local Government demographers the flexibility to
examine population, household and labour force projections at a level
appropriate to their needs... alternative scenarios can be quickly
generated, by manipulating any of the components, and compared with
previous sets. This facility is particularly useful when looking at
alternative Planning scenarios for House Building as required by Local
Spatial Strategies; with the abolition of Regional Spatial Strategies,
this has become a key requirement for Local Authorities" [4].
The success of POPGROUP in providing an accessible technical framework
for local planning has emerged directly from the underpinning research
cited, with its algorithms and understanding of common local needs across
diverse circumstances providing an accessible and user-controlled planning
tool.
3. POPGROUP has been used throughout the nations of the UK in
socio-geographically diverse situations. The national statistical
agencies of Wales, Scotland and Northern Ireland have made use of POPGROUP
for their production of sub-national demographic forecasts. The Welsh
Assembly Government used the software for their council area population
forecasts (2010 and 2013); noting that "POPGROUP has proved to be a
useful tool in enabling government bodies such as the Welsh Government
to produce demographic projections at a local authority level — this
includes Unitary Authorities and National Parks" [5]. National
Records Scotland also used the software for national park population
forecasts (2011), working with Simpson and Lomax to develop local ward
projections. Their Head of Population and Migration Statistics confirms
the value of this work:
"The project was in two phases the first of which was the production
of projections for multimember wards for Fife Council area and the
second phase was to work with National Records of Scotland (then General
Register Office for Scotland) to develop the same projections in-house
and to compare different methods using more detailed data. Comprehensive
recommendations and guidance was produced to allow others to more easily
produce their own projections for small areas. Feedback from Councils on
the project and on the guidance has been very favourable. Subsequent
recent follow-up work to produce user guidance for the derived forecast
model was also produced to very short deadlines and within budget."
[7]
4. POPGROUP is increasingly used by commercial planning agencies.
As of July 2013, six commercial organisations have purchased POPGROUP
(Barton Willmore, Ecorys UK, GVA Grimley, Nathaniel Lichfield and
Partners, ORS and Regeneris Consulting Ltd.) These organisations advise
local authorities and housing development companies in Britain, and have
turned to POPGROUP software as a `standard' since 2010. GVA, a large
consultancy firm working both for local authorities and private
development agencies, underlines this recognition:
"POPGROUP is becoming increasingly recognised as the Industry standard
population forecasting model. This is evident through the requirement to
have POPGROUP expertise in a majority of the recent public sector tender
opportunities we have worked on. This requirement has also stimulated
the commercial sector's use of the model to interrogate and challenge
published projections elsewhere. From our own experience we have used
POPGROUP both to define local housing requirements as well as challenge
those previously set. Our work using POPGROUP has been used to
successfully justify an increase in local housing requirement in support
of major housing development, by demonstrating a range of scenarios
where higher housing numbers would be required." [8]
In large part, it is POPGROUP's accessible Excel platform that has
assisted wider uptake.
Ongoing engagement: In addition to these impacts, UoM's annual
training course in POPGROUP — in addition to self-training POPGROUP
practical modules — reflects breadth of impact. In 2013, the 18
participants came from: local authorities (9), Universities (4), national
statistical agencies (3) and commercial companies (2) [9]. Moreover,
international interest has been reflected in users in the United States
(U.S. Census Bureau), Finland, Singapore and Malaysia, and the inclusion
of POPGROUP in a 2011 review of three demographic software packages
commissioned by the Latin American Population Association [10]. The
influence of POPGROUP is ongoing, as upon the retirement of Simpson in
2009, Edge Analytics Ltd won a tendering process to support users and to
develop, distribute and market the software. According to Edge Analytics'
Managing Director, "the contract helped the company to develop and
enlarge from one to five staff by 2013, each of whom have had
significant workloads related to POPGROUP" [3]. Simpson remains the
technical manager.
Sources to corroborate the impact
(all claims referenced in the text)
[1] (2013) POPGROUP List of Users & Register of Work (2005-2010)
& (2012) POPGROUP Educational License (October)
[2] (2011) `Limiting Long Term Illness (LLTI) & Disability
Projections: Data Modules for the Derived Forecast Model (Edge Analytics)
(passim)
[3] Testimonial from Director, Edge Analytics (13th June 2013)
[4] Testimonial, Joint Chair (Local Authority), Central and Local
Information Partnership's `Population Statistics' Sub-Group (25th
April 2013)
[5] Testimonial from Head of Population Statistics, Welsh Assembly
Government (9th May 2013)
[6] (2013) Local Government Association `Report of the POPGROUP User
Survey' (July)
[7] Testimonial from Head of Population Statistics, National Records
Scotland (3rd May 2013)
[8] Testimonial from Senior Consultant, GVA Consultancy (25th
April 2013)
[9] Demographic Forecasting with POPGROUP — Future Training Flyer &
Registration on recent course (16th - 17th May 2013)
[10] (2010) ALAP Evaluation `Informe sobre Programas Informáticos para
Proyecciones Demográficas' (16th November) (pp.44-52,
Spanish — English Intro)