Transcript Forecasting Employment, Housing, and Population

Forecasting
Employment, Housing, and
Population
VisionWest Consortium Meeting
April 14, 2016
Dean Bangsund
Nancy Hodur, PhD
Department of Agribusiness and Applied Economics
North Dakota State University
2010-2014 Period
• Ever changing landscape with substantial uncertainty
• Lacked data (quality and quantity) for traditional
population forecasting
• Multiple economic drivers
• Developed non-traditional methods and models
Previous Forecasting Efforts
Employme
nt
Housing
Population
Retaining all of our previous employment
modeling
All new modeling with new data sets and
capabilities
So What is New for Our Project?
• Workforce by place of work and place of residency
• Official fertility and mortality data
• Net migration (permanent pop)
• Housing supply data
• Non-employer data
• Official petroleum industry employment
• Updated co-hort population model
• Housing needs assessment (NDSU) and general
population forecast (Commerce-Census)
• Workforce characteristics
Longitudinal Employer-Household Dynamics -Origin-Destination Employment Statistics (LODES)
Where do people work who live in
Williams County?
Where do people work who live in
Williams County?
Where do workers live that are employed
in Williams County
Trends in Workforce Logistics
Employment and Local Workforce
Source: Labor Market Information Center, Job Service North Dakota, Local Area Unemployment Statistics (LAUS) Unit
What does this mean?
• Employment at a specific location may/may not translate
to residents of that location
• Employment in one location can affect population in
another location
• Workforce is not limited to those residing in the immediate
area
• Our modeling will be incorporating these locational
dynamics!!
Fertility and Mortality
Traditional Components of Population Change
• Confidential data at the age cohort level, but available to
include in our modeling
• We can say
• Fertility rates are increasing (total and in %)
• Mortality rates are decreasing (total and in %)
• Take Away—Williams County will experience population
change without changes in employment
Fertility and Mortality
Net Migration
In Migration less Out Migration
Permanent Residents Moving into the County
Housing Supply
NOT American Community
Survey data—no multi-year
averages.
Estimates are based on
“Administrative Data”
Census estimates of
permanent housing are
being ground-truthed as
part of the Housing
Needs Assessment study.
While preliminary, initial
observations are that
Census estimates (based
on administrative data)
and actual supply are
fairly close.
Source: U.S. Census Bureau, Population Division, Annual Estimates of Housing Units for the United States, Regions,
Divisions, States, and Counties: April 1, 2010 to July 1, 2014, May 2015.
Nonemployer Data
• Census and Bureau of Economic Analysis Data Sets
• Self-employment
• Contractors
• Consultants
• Farms/Ranches
• Individual practice
Source: U.S. Dept of Commerce, Bureau of Economic Analysis.
Source: U.S. Dept of Commerce, Bureau of Economic Analysis.
Petroleum Industry Direct Employment
Source: Job Service North Dakota. Actual count of jobs, by place of
work, based on federal and state unemployment insurance
classifications. Available for 2013, 2014, and 2015 (June release).
Petroleum Industry Direct Employment
Co-hort Model
• Traditional approach to population modeling
• Population (new yr) = population (old yr)
+ births – deaths + in-migration – out migration
Co-hort Component Population Modeling Dynamics
Out Migration
Deaths
In Migration
Age Group
Mortality
Rates
Migration
Rates
Fertility
Rates
By age and
sex
By age and
sex
By age
cohort
Male
Female
85 +
80-84
75-79
70-74
65-69
60-64
55-59
50-54
45-49
40-44
35-39
30-34
25-29
20-24
15-19
10-14
5-9
0-4
6
Births
(period 1)
4
2
0
Percentage of
Population
2
4
6
Births
(period 2)
Period 1
Period 2
Period 3
Supplemental Resources
• ND Dept of Commerce-Census Data Center
• Three population scenarios (generic)
• Housing needs not included
• ND Housing Finance Agency – Statewide Housing Needs
Assessment
• One population scenario (generic)
• Identify housing needs (quantity and affordability)
Methods
How Does All This Work?
Step 1: size of current workforce
Resident
Population
• Participation Rate
• Employment Rate
Resident
Workforce
Total
Workforc
e
Commuter
s
• Outside the region
• Delineated by
segment of industry
NonReside
nt
Workforce
Step 2: how much workforce do we need
Future Scenario using
Employment Model
Demand
Supply
Future Labor
Requirements
Existing
Labor Pool
Residents
Greater
Less
Equal
Resident
Workforce
Employment
Change
combined with
Workforce
Characteristics
Nonresidents
Greater
Nonresident
Less
Workforce
Equal
Step 3: determine inputs into co-hort model
Labor Demand is Greater than Local Labor Supply
Implications for Co-hort Analysis
Effects on Population
Increase in net migration
(permanent residents)
Permanent population increases
Increase in nonresident
commuters (not local residents)
Service population increases
Both actions
Growth in permanent and service
populations
Labor Demand is Less than Local Labor Supply
Implications for Co-hort Analysis
Effects on Population
(negative) net migration
Population loss
Labor participation rate declines
Little if any effects on population
Unemployment rate goes up
Little if any effects on population
in short term, long term would
lead to out migration
Step 4: incorporate labor force
(needs/changes) into co-hort model
Start (2015)
Employment
Workforce
Needed
Migration
Period 2
Period 3
Employment
Employment
Workforce
Needed
Workforce
Present
Workforce
Needed
Workforce
Present
Needs/changes
Needs/changes
New Population
New Population
Available
Workforce
Available
Workforce
New Population
Available
Workforce
Step 5: estimate housing requirements
• residents
• commuters
Employment
Population
• permanent
• service
• permanent
• temporary
Housing
Requirements
Housing Need
Housing Needs
Assessment
Single Family
Permanent
Housing
Units
Resident
Population
Apartment
Mobile Homes
Traditional
Relationships
(previous modeling)
Temporary/Non
traditional
NonResident
Workforce
Workforce
Characteristics
Total Housing
Current Geographic Scope
Economic/Employment Scenarios
• Employment, housing, and population (permanent and
service) for the following based on crude oil prices:
• Low price
• Medium price
• High price
• Cyclical
Other Scenarios?
• Considerations relevant to VisionWest Consortium
members and interests
• We will consider additional scenarios but would like
feedback to formulate additional alternatives
Final Outputs
(Late Summer)
• (1) Webinars—each county
• 25 to 35 slides covering key elements
• (2) Written Report
• Discussion of data, methods, and results
• Appendix material for tabular data
Questions/Discussion
• Dean Bangsund at 701-231-7471 or
[email protected]
• Nancy M. Hodur at 701-231-7357 or
[email protected]