Title: Projecting transient populations - pragmatism or technical correctness? BSPS Conference Sep 2004 Richard CooperResearch team Nottinghamshire County Council
1Projecting transient populations - pragmatism
or technical correctness?BSPS Conference
Sep 2004Richard Cooper Research
team Nottinghamshire County Council
2(No Transcript)
3Joint Structure Plan housing figures
- Regional Planning Guidance (1996-based) 49,000
dwellings 2001-21 - Joint Structure plan accepts total
- Distribution to sub-areas
- South Nottinghamshire 37,000
- Nottingham City supply of 18,500
4Population projections for Nottingham City
- Basis is a set number of dwellings
- Early projections were dwelling-led but
- Migration levels varied widely
- Migration-led projection needed
- More robust output
- More up-to-date information available
- City wanted age / gender projection
5Knowns and unknowns
- How many houses but not types of house,
household or occupants - Age/gender of residents and migrants but not
future migrants - Characteristics of residents but not how those
may change
6Modelling the population
- Changing housing provision (e.g. more flats)
- Assumptions that data in the model will still
pertain - - the migration profile remains the same
- characteristics (fertility, household generation,
etc.) of population remain same for age, gender
relationship
7Nottingham City 2001 Census
8Nottinghamshire (rest of Plan Area) 2001 Census
9Age profile of some JSP districts
10(No Transcript)
11Nottingham city projection no transient
population
12Effects of ignoring the transient population
- Age structure would have many more adults 35-44,
(and fewer 15-24) - ageing through fertile and household creation
ages - For a set number of dwellings (18,500)
- 8,000 fewer (30 less growth)
- For a certain migration level
- 2,500 more dwellings
13But why is this a problem (to Notts!) ?
- Decision to use Patient Register data
- From ONS used in mid-year estimates
- More up-to-date
- More complete than the Census (includes students)
- 3 years data - 1997-2000
- More accurate?
14A problem ? (2)
3,000 more net in-migrants 15-19 but are these
all students?
15(No Transcript)
16A problem ? (3)
- Transient population used where migration data
does not handle flows adequately (1991 Census) - If migration data complete there is no need for a
transient population - in theory OK - However, results did not show sensible outcome
it appeared that some student migrants were being
excluded
17Determining a transient population
- Needed a reality check
- Thought that transient population in CPHM was
wrong for application to Patient Register
migration - How do you decide on a transient population when
some information is missing? - What should the relevant (20-24) population be
doing? - It does not remain absolutely constant, even
though student numbers may do so so how does it
change? - Look at births 20 years ago, not for absolute
numbers but for trends
18What the 20-24 year old projectionss should be
showing
19Changes to 20-24 yr old population 2001 - 2016
Original projection has no transient
adjustment May 2003 resulted from City suggestion
in setting transient population Mar 2004 accounts
for latest information and migration-led
projection
20How does it compare? (1)
21How does it compare? (2)
NB 1996 trend-based, so unusable for Structure
Plan Methodology incorporated separate student
adjustment
22How does it compare? (2)
Main difference is higher 35-54 population in JSP
23(No Transcript)
24(No Transcript)
25(No Transcript)
26(No Transcript)
27(No Transcript)
28(No Transcript)