Memorandum
City of
Planning Department
|
TO: |
Dave Corliss, |
|
FROM: |
Planning Staff |
|
CC: |
|
|
Date: |
|
|
RE: |
Population
Projections and Estimates |
Planning Staff
recently undertook an effort to determine if the Department has been using best
practice methods of projecting future population for
Staff also tested
Horizon 2020 population projections using different methods than were used for
Horizon 2020. In addition, staff is in
the process of testing the US Census population estimates for
I. Existing
Planning Department Methods
Projections
Population projections are forecasts of population at some future date.
o
Horizon
2020 provides three projections: low, medium and high. The
Low projection model uses a simple linear regression with decennial census data
from 1950 to 2000. The Middle projection
model uses a simple linear regression with data from the Census estimates for
the years 1990 to 1999. The High
estimate model uses the average decade growth rate from 1950-2000 of 24
percent.
Estimates
Population estimates are an annual estimate
of population.
o
The
Planning Department estimates the population of
II. Population
Projection Best Practice Literature Review
Further Thoughts on Simplicity and Complexity in Population Projection
Models, Smith,
o
No
evidence to suggest that the use of complex methods has led to consistently
greater accuracy in forecasting total population than can be achieved with
relatively simple extrapolation techniques. Population projections for horizons
longer than 20 years are of dubious predictive value. (p. 12-13)
o
Combining
projections from different models can be beneficial. Combining can be done in many ways: using
different models or techniques, different specifications of the same model or
technique, simple averages vs. weighted averages, historical weightings vs.
subjective weightings, and so forth. (pp. 15-16)
Local Government Finance: Capital Facilities Planning
and Debt Administration, Steiss, Alan Walter, http://www-personal.umich.edu/~steiss/page49.html
o
For
purposes of capital facilities planning, it is necessary to develop fairly
detailed estimates and projections of population by age-cohorts. Such detail is particularly applicable to the
identification of capital facility needs associated with certain age groups
(such as school facilities, health care facilities, housing for the elderly,
recreation needs, etc.)
o
Population
change involves three separate components: births, deaths, and migration. Component models consider the separate effects of each of these
factors and requires more comprehensive and detailed data than usually
are available to local planners.
o
The
simpler methods have a wider range of application and may be used to produce
population projections at almost any geographic level. Econometric and cohort component models have
more extensive data demands and may be less appropriate for smaller geographic
areas.
III.
Community Methods in Practice
Population
Projections
o
o
Comp
Plan projects 363,000 people in the county by 2025 (250,291 base in 2000) and
527,000 by 2050. This projection was
reached using an assumed growth rate of 1.5 percent per year throughout the 50
year period. That rate is more than the
historic rate of 1.2% (since the 1960’s) but mirrors recent growth trends.
o
The Comp
Plan anticipates that the rate of growth will be routinely monitored.
o
o
o
o
The Comp
Plan forecasts a county population of 567,000 in 2030 (from a base of 435,000
in 1997. Population forecasts for
o
o
The
o
o
Process
not outlined in Comp Plan. Data used was
local housing demographics, 1990 Census and building permit information.
o
o
o
o
o
o
Population
estimates for 2020 prepared by a consulting firm.
o
o
Used a
cohort-component method to project a county population. Municipal populations were projected using
four simple methods and averaging the results.
The county projection was used as the control for the municipal
projections.
Population Estimates
o
o
Uses a
building permit method to estimate a yearly population. Uses occupied housing units, average
household size, and persons in dorms, institutions and group homes.
o
o
Provides
population estimates for
o
State
of
o
The
State of
While there is a variety of population projection and estimating
techniques utilized by different communities, the techniques used by
IV. Method
Testing
Population Projection Testing
Staff ran a number of projection methods to test the existing Horizon
2020 population projections. Numerous bivariate curve fitting methods were used including
polynomial regression models, exponential models, and logarithmic regression
models. Results indicate the Horizon
2020 mid-range projections are on target.
Population Estimate Testing
The method of population estimating used by the Planning Department is
a method that is well established and used by many jurisdictions around the
country. The Planning Department
population estimate for
Staff is also in the process of testing the recent US Census population
estimate numbers for
Staff will continue to test the US Census
numbers. If Staff determines with
confidence the US Census is underestimating