A Look Backward at the Economic Development Objectives of

Report
A Components of Growth Analysis of
a Small,Vibrant Metropolitan Area:
Spokane Washington Case Study
Dr. Roger Coupal
Agricultural and Applied Economics
University of Wyoming
Framework: Components of Income approach
Based upon: Smith, G. (1996): Garnick, D.(1990)
TPI/N = E/N + P/N + T/N + IA/N
Components of Income approach:
E/N = H/J + E/H + J/N
N = Population
TPI = Total personal income
E = Earnings
P = Property income
T = Transfer payments
IA = Income adjustments
H = Hypothetical earnings
J = Number of Jobs (full and part-time
H/J = Industry mix component
E/H = Differential Earnings Component
J/N = job / population Ratio
Energy Booms
High Tech Mfg bust
hart
Energy Bust
Y
USE/J
H/J
E/J
50,000
Chart
40,000
Y
Spokane Earnings / Job,
Spokane Hypothetical
Earnings / Job, and US
Earnings / Job
• Tends to trend the nation but at
a lower level.
• Do lower energy prices mean
higher growth?
• Growth during the high tech
growth but flat during the
energy boom.
• Energy bust: an uptick.
Y
USE/J
H/J
E/J
50,000
30,000
2010
2008
2006
2004
2002
2000
1998
1996
1994
1992
1990
1988
1986
1984
1982
1980
1978
1976
1974
1972
20,000
1970
Y
40,000
year
2010
2008
2006
2004
2002
2000
1998
1996
1994
1992
1990
1988
1986
1984
1982
1980
1978
1976
1974
1972
20,000
1970
30,000
5/15/13 9:23 AM
Job to population ratio, Spokane and the United States
Chart
Y
0.70
J/N
USJ/N
0.65
0.60
Ratio
0.55
0.50
• Generally increased except
during the lead up to the
current recession
• Spokane J/N tracked slightly
higher since the mid-90’s
• As a pct of the national
0.45
0.40
2010
2008
2006
2004
2002
2000
1998
1996
1994
1992
1990
1988
1986
1984
1982
1980
1978
1976
1974
1972
1970
0.35
year
Chart
110.0%
Job to population ratio as a percent of the US ratio
Job to Pop Ratio
105.0%
100.0%
95.0%
90.0%
1973
1978
1983
1988
1993
year
1998
2003
2008
Industry Mix Component of relative per capita income and compared with earnings
per Job, Pct Change from the preceding year
Summary of Fit
RSquare
0.58965
RSquare Adj 0.568606
Root Mean Square Error 0.013718
Mean of Response 0.017351
Observations (or Sum Wgts) 42
Parameter Estimates
Term
Estimate
Intercept
0.0025275
IndMixGrth
0.031583
IndEarnGrth
0.5352067
Std Error
0.002962
0.158376
0.082951
Residual by Predicted Plot
t Ratio
0.85
0.20
6.45
Prob>|t|
0.3986
0.8430
<.0001*
Job ratio component of relative per capita income growth
Summary of Fit
RSquare
0.591299
RSquare Adj
0.57034
Root Mean Square Error 0.013691
Mean of Response
0.017351
Observations (or Sum Wgts) 42
Parameter Estimates
Term
Estimate
Intercept
0.0138919
DifEarnGrwth
0.3862331
J/NGrwth
0.7684218
Std Error
0.002226
0.146913
0.103109
t Ratio
6.24
2.63
7.45
Prob>|t|
<.0001*
0.0122*
<.0001*
Differential Earnings Growth Component of Relative PCI, Spokane
5/15/13 10:07 AM
Chart
Y
DifEarnGrwth
PCPIGrth
0.10
Pct Change
0.05
0.00
year
Summary of Fit
RSquare
0.233941
RSquare Adj
0.194656
Root Mean Square Error 0.018744
Mean of Response
0.017351
Observations (or Sum Wgts) 42
Parameter Estimates
Term
Estimate
Intercept
0.0110054
DifEarnGrwth
0.4417462
IndMixGrth
0.7060462
Std Error
0.003577
0.215804
0.208765
t Ratio
3.08
2.05
3.38
Prob>|t|
0.0038*
0.0474*
0.0016*
2010
2008
2006
2004
2002
2000
1998
1996
1994
1992
1990
1988
1986
1984
1982
1980
1978
1976
1974
1972
1970
-0.05
Ù
pcpi t = m + f pcpit -1 - pcpit - 2
Model: AR(1)
Model Summary
DF
41
Sum of Squared Errors
RSquare
RSquare Adj
MAPE
MAE
Stable :
Invertible:
15881057.7
0.91296883
0.91084611
1.74134482
580.072929
Yes
Yes
Parameter Estimates
Term
Lag Estimate
Std Error
t Ratio
AR1
Intercept
0.009
4695.124
101.39
7.43
1
0
0.993
34896.543
Prob>|t|
<.0001*
<.0001*
Constant
254.408251
ÑPCPI = f (ÑIndMix,ÑE / J, ÑJ / N,ÑUSJ / N)
Summary of Fit
RSquare
0.675913
RSquare Adj 0.640877
Analysis of Variance
Parameter Estimates
Term
Estimate
Intercept
0.0061186
IndMixGrth
-0.114912
E/JGrwth
0.554309
J/NGrwth
0.6939885
Std Error
0.022885
0.161848
0.144011
0.097906
USJ/N
0.042908
0.0075217
t Ratio
0.27
-0.71
3.85
7.09
0.18
Prob>|t|
0.7907
0.4822
0.0005*
<.0001*
0.8618
Conclusions and Discussion
• Slower growth in earnings per job than the national setting
• Continued divergence between national hypothetical and local earnings
• Less connection between the national and local conditions
• More reliance on growth factors (J/N, etc.)
• They type of national growth may or may not affect a metro area
Methodological / Geographic considerations
• We don’t know how other metro
areas perform.
• What is perhaps more important is
whether capital flows facilitate more
startups.
• Firms within groups of like
industries tend to cluster. What an
economic development initiative
would like is separate from what it
can actually accomplish, through no
fault of its own.
Jeffrey City, Wy: Boeing branch plant location?
Vet School, Medical School? Anything, please!!

similar documents