House Prices and Land Prices under the Microscope:

A Property-Level Analysis
Morris A. Davis, Rutgers University
Stephen D. Oliner, AEI and UCLA
Edward J. Pinto, AEI

Presented at the Third International Conference on Housing Risk

American Enterprise Institute, Washington, DC
September 17, 2014

Note: The views expressed are ours alone and do not represent those of the institutions with which we are
affiliated. We thank FNC, Inc., and Marshall & Swift, a CoreLogic company, for providing the data for the paper.

Introduction
Key lesson from the financial crisis: better information is needed to
evaluate risk in housing and mortgage markets
The research presented today helps fill that gap by quantifying
collateral risk at a fine level of geography
Focus on land value, as land is the risky part of house price
Estimate land value at the property level; no previous study has done so on
such a large scale
Characterize the recent boom/bust cycle in land prices and house prices for
Washington, DC metro area, supplemented by results for Boston and Miami

Separately, have constructed mortgage risk indexes based on a

rigorous stress test. Available at www.housingrisk.org.

Key Takeaways
Land prices were more volatile than house prices, especially in areas
with inexpensive land and a low land share in 2000.
In those areas, the land share jumped the most through 2006. This
increase was a useful predictor of the later crash in house prices.
Movements in land prices and land shares had predictive information
that was not evident from looking at house prices themselves.
These results highlight the value of focusing on land for assessing
house-price risk.

Scope of the Paper

Focuses on detached single-family homes
Main results are for Washington, DC metro area
City of Washington, DC
Virginia: Arlington, Fairfax, Loudoun, and Prince William Counties, plus City of
Alexandria
Maryland: Montgomery and Prince Georges Counties

Compare DC area results to preliminary results for Boston and Miami

Boston: Norfolk, Suffolk, and Middlesex Counties
Miami: Dade County

Snapshot of the Washington, DC Area

County/city
Arlington Co., VA

Population

Median
Household
Income

% of Adults w/
Bachelors
Degree

% Black or
Hispanic

220K

$101K

71%

24%

1.1 mil

$108K

58%

25%

Loudoun Co., VA

340K

$122K

57%

20%

Prince William Co., VA

430K

$95K

37%

40%

City of Alexandria, VA

150K

$82K

61%

38%

Montgomery Co., MD

1.0 mil

$95K

57%

35%

Prince Georges Co., MD

880K

$72K

30%

78%

Washington, DC

630K

$65K

52%

59%

314 mil

$52K

29%

29%

Fairfax Co., VA

Memo, U.S. total

Note: Population and percent Black or Hispanic are as of 2012. Median household income and percent with bachelors degree are averages over
2010-2012; median household income is in 2012 dollars. Source: U.S. Census Bureau.

Compared to US average, Washington area as a whole is affluent, with a well educated

and ethnically diverse population. But there are differences across the area.

Arlington, Fairfax, and Loudoun: high income and education; low Black and Hispanic share

Prince Georges: lower income and education, high Black and Hispanic share

Others have a blend of these characteristics. In Montgomery, diversity within the county.
5

Key Data Sources

Property-level data
House characteristics, location, year built, sales history, and AVM value:
FNC, Inc.
Reconstruction cost as new: Marshall & Swift (M&S), a CoreLogic company

Zip-level data
House price indexes: FNC
Construction cost indexes: M&S

DC area dataset contains about 680,000 detached single-family

homes across 142 zip groups. About 90% coverage of universe.

Estimating Land Value

Create a market-based stake in the ground for every home

Consists of an estimate of total property value broken into land value and
structure value.

Estimate land value for the stake in the ground from sale price of
new homes in the zip group minus reconstruction cost at sale date
Land is a shorthand for amenities of a given location

Stake in the ground is in 2000 for a majority of zip groups

Is in a later year when there are too few sales of new homes in 2000 alone.

Creating Time Series

Given stake in the ground, move AVM from that period through time
with house price index for homes zip.
Move structure value through time with construction cost index for
homes zip.
Back out land value for each home in each period as AVM minus
structure value.
Take average of house-level results within each zip group.

Average AVM, 2013:Q3

House prices are relatively low in Prince George's County and most outlying areas,and relatively high in Fairfax,
Arlington, and Alexandria, as well as the affluent parts of Montgomery County and DC.

20872, 21771
20180

20871
20882, 21797

Loudoun County, VA

20197
20129
20158

20132

20876

20838, 20839
20841, 20842

20176

20874

20837

20878

20175
20165

20147
20117, 20184, 20130, 20135, 20141

20105

20152
20151
20120

20109

20143, 20155
20136

By quintile*
Lowest

Second
Third

Fourth

20121

20110

20832

20855

20850

22066
20164, 20166

20879

20877
20880

20854

20148

20169, 20137

20886
20879

20853
20851

20895
20896

20814

20119, 20181

20902
20901
20910

20815

20015

20012

20904

20783
20903
20912

20707
20866

20705

20708

20740
20770

20706
20737 20784
20001
22102
20781
20008 20009
20190, 20191
20784
22101
20710
22182
20016
20010 20017
20171
20018 20712
20011
20722 20785
22181
20007
22043
22213, 22207
22027
20002, 20003
22124
22180
20019
22046 22205 22209, 22201
20743
22033
22203
22042
22031
20020
22206
22041
22202
20747
22204
22044
22030
22301 20032
22003
20746
22305
20748
22311
22032
22302
22312 22304
20745
22151
22314
20124
22015
22303, 22310
22307
22152 22150
22306
22315
20744
20623, 20735
22153
22039
20812
20816

20111

22060, 22309

Prince William
County, VA

20868
20905

20906

20852

20817
20818

20170 20194

Montgomery County, MD

20777, 20833
20860, 20861
20862

22192

20112

22079

20782

20715

20769

20720

20721

20716

20774

20772

Prince George's
County, MD

22308

20601, 20607

20608, 20613

22193
22191
22134, 22172

22025
22026

Fairfax County, VA

Highest

*Range of average AVM values by quintile (rounded to the nearest $1000) is $128,000 to $304,000, $304,000 to $436,000, $436,000 to $554,000, $554,000 to $670,000,
and $670,000 to $1,657,000. Source: Authors' calculations using data from FNC, Inc.

Average Lot Value, 2013:Q3

Lot values largely mirror the pattern for house prices.

20872, 21771
20180

20871
20882, 21797

Loudoun County, VA

20197
20129
20158

20132

20876

20838, 20839
20841, 20842

20176

20837

20117, 20184, 20130, 20135, 20141

20878

20165

20190, 20191
20171

20152
20151

22033

20120

20109

20136

By quintile*
Lowest

Second
Third

Fourth

22182

22181
22124

20895
20896

22027
22180

22030

20124

22015

22151
22152

22311
22312

22150

22153

22039

20111

22101

22003

22032

20814

22304

20119, 20181

22192

20112

20910
20012

22079

22301
22305
22302

22303, 22310
22315

22306

22060, 22309

Prince William
County, VA

20901

20015

20812
20816

20904

20902

20815

20707
20866

20783
20903
20912

20707
20866

20708

20705
20740
20770

20706
20737 20784
20781
20784
20710

20782

20001
20008 20009
20016
20010 20017
20018 20712
20011
20722 20785
20007
22043
22213, 22207
20002, 20003
20019
22046 22205 22209, 22201
20743
22203
22042
20020
22206
22041
22202
20747
22204
22044

22102

22031

20121

20110

20851

20817
20818

20868
20905

20906

20852

20170 20194

20105

20853

20850

22066
20164, 20166

20143, 20155

20855

20854

20148

20169, 20137

20832

20877
20880

20175
20147

20879

20886
20879

20874

Montgomery County, MD

20777, 20833
20860, 20861
20862

20032

20745

22314

20748

20715

20769

20720

20721

20774

20746

20772

22307
20744

20716

20623, 20735

Prince George's
County, MD

22308

20601, 20607

20608, 20613

22193
22191
22134, 22172

22025
22026

Fairfax County, VA

Highest

*Range of average lot value by quintile (rounded to the nearest $1000) is $0 to $134,000, $134,000 to $201,000, $201,000 to $289,000, $289,000 to $456,000, and $456,000 to $1,012,000.
Source: Authors' calculations using data from FNC, Inc. and Marshall & Swift, a CoreLogic company.

10

Average Lot Size, 2013:Q3

As expected, lots tend to be small in close-in areas and larger further out.

20872, 21771
20180

20871
20882, 21797

Loudoun County, VA
20197
20129
20158

20132

20876

20838, 20839
20841, 20842

20176

20886
20879

20874

20837

20117, 20184, 20130, 20135, 20141

20165

20105

20152
20151
20120

20109

20143, 20155
20136

By quintile*
Smallest
Second
Third

Fourth

20121

20170 20194

20814

22015

22150

22153

22039

20111

22152

Prince William
County, VA

20901
20910

20815

22192

20112

22079

20015

22303, 22310

22315

22306

22060, 22309
20119, 20181

20904

20902

20012

20783
20903
20912

20707
20866

20707
20866

20708

20705
20740
20770

20706
20737 20784
20001
22102
20781
20008
20190, 20191
20009
20784
22101
20710
22182
20016
20010 20017
20171
20018 20712
20011
20722 20785
22181
20007
22043
22213, 22207
22027
20002, 20003
22124
22180
20019
22209,
22201
22046 22205
20743
22033
22203
22042
22031
20020
22206 22202
22041
20747
22204
22044
22030
22301 20032
22003
20746
22305
20748
22311
22032
22302
22312 22304
20745
22151
22314
20812
20816

20124

20110

20817
20818

20868
20905

20906

20851

20895
20896

22066
20164, 20166

20853

20852

20854

20148

20169, 20137

20832
20855

20850

20175
20147

20879

20877
20880

20878

Montgomery County, MD

20777, 20833
20860, 20861
20862

20782

20715

20769

20720

20721

20774

20772

22307
20744

20716

20623, 20735

Prince George's
County, MD

22308

20601, 20607

20608, 20613

22193
22191
22134, 22172

Fairfax County, VA

22025
22026

Largest

*Range of average lot size by quintile (rounded to the nearest 1000 square feet) is 4,000 to 9,000, 9,000 to 13,000, 13,000 to 16,000, 16,000 to 28,000, and 28,000 to 183,000.
Source: Authors' calculations using data from FNC, Inc.

11

Average Standardized Land Price per Square Foot, 2013:Q3

After controlling for differences in lot size, the price of land is highly correlated with house prices.
That is, house prices reflect the amenity value of a given location.

20872, 21771
20180

20871
20882, 21797

Loudoun County, VA
20197
20129
20158

20132

20876

20838, 20839
20841, 20842

20176

20886
20879

20874

20837

20117, 20184, 20130, 20135, 20141

20165

20105

20152
20151
20120

20109

20143, 20155
20136

By quintile*
Lowest

20121

Fourth

20814

20812
20816

22015

22150

22153

22039

20111

22152

20119, 20181

20901
20910

20815

22192

20112

22079

20015

20012

20783
20903
20912

20707
20866

20708

20705
20740
20770

20782

22303, 22310 22307

22315

20715

20769

20720

20721

20744

20716

20774

20772

22306

22060, 22309

Prince William
County, VA

20904

20902

20707
20866

20706
20737 20784
20001
22102
20781
20008
20190, 20191
20009
20784
22101
20710
22182
20016
20010 20017
20171
20018 20712
20011
20722 20785
22181
20007
22043
22213, 22207
22027
20002, 20003
22180
22124
20019
22209,
22201
22046 22205
20743
22033
22203
22042
22031
20020
22206 22202
22041
20747
22204
22044
22030
22301 20032
22003
20746
22305
20748
22311
22032
22302
22312 22304
22314
20745
22151

Prince George's
County, MD

20623, 20735

22308

20601, 20607

20608, 20613

22193

Second
Third

20817
20818

20170 20194

20124

20110

20851

20868
20905

20906

20895
20896

22066
20164, 20166

20853

20852

20854

20148

20169, 20137

20832
20855

20850

20175
20147

20879

20877
20880

20878

Montgomery County, MD

20777, 20833
20860, 20861
20862

22191
22134, 22172

Fairfax County, VA

22025
22026

Highest

*Range of average land price per square foot for a quarter-acre lot by quintile (rounded to the nearest dollar) is $0 to $11, $11 to $17, $17 to $25, $25 to $39, and $39 to $106.
Source: Authors' calculations using data from FNC, Inc. and Marshall & Swift, a CoreLogic company.

12

Average Land Share of Property Value, 2013:Q3

The land share tends to be high in areas with high land prices and vice versa.
But there are outliers, some of which could arise from limitations of the data.

20872, 21771
20180

20871
20882, 21797

Loudoun County, VA
20132

20197
20129
20158

20876

20838, 20839
20841, 20842

20176

20886
20879

20874

20837

20117, 20184, 20130, 20135, 20141

20165

20190, 20191
20171

20152
20151

22033

20120

20109

20143, 20155
20136

By quintile*
Lowest

Second
Third

Fourth

20895
20896

22182

22181
22124

22027
22180

22030

20124

22015

22151
22152

22311
22312

22150

22153

22039

20111

22101

22003

22032

20812
20816

22304

20119, 20181

22192

20112

20901
20910

20815

20015

20012

20783
20903
20912

20707
20866

20708

20705
20740
20770

20706
20737 20784
20781
20784
20710

20782

22079

20032

22301
22305
22302

22314

22303, 22310 22307

22315

22306

22060, 22309

Prince William
County, VA

20904

20902

20814

20707
20866

20001
20008 20009
20016
20010 20017
20018 20712
20011
20722 20785
20007
22043
22213, 22207
20002, 20003
20019
22046 22205 22209, 22201
20743
22203
22042
20020
22206
22041
22202
20747
22204
22044

22102

22031

20121

20110

20817
20818

20170 20194

20105

20851

20868
20905

20906

20852

22066
20164, 20166

20853

20854

20148

20169, 20137

20832
20855

20850

20175
20147

20879

20877
20880

20878

Montgomery County, MD

20777, 20833
20860, 20861
20862

20745

20748

20715

20769

20720

20721

20716

20774

20746

20772
20744

20623, 20735

Prince George's
County, MD

22308

20601, 20607

20608, 20613

22193
22191
22134, 22172

22025
22026

Fairfax County, VA

Highest

*Range of average land share of property value by quintile (rounded to the nearest 1%) is 0% to 37%, 37% to 46%, 46% to 58%, 58% to 71%, and 71% to 80%.
Source: Authors' calculations using data from FNC, Inc. and Marshall & Swift, a CoreLogic company.

13

House Price Increase, 2000-2006

Substantial variation across zip groups. Some tendency for increases to have been largest in outlying
areas and smallest in "band of affluence" along the Potomac River in DC, Maryland, and Virginia.

20872, 21771
20180

20871
20882, 21797

Loudoun County, VA
20197
20129
20158

20132

20876

20838, 20839
20841, 20842

20176

20886
20879

20874

20837

20878

20165

20105

20152
20151

20169, 20137

20120

20109

20136

By quintile*
Smallest
Second
Third

Fourth

20121

20817
20818

20170 20194

Prince William
County, VA

20904

20902
20901
20910

20815

20015

20012

20783
20903
20912

20707
20866

20707
20866

20708

20705
20740
20770

20706
20737 20784
20001
22102
20781
20008 20009
20190, 20191
20784
22101
20710
22182
20016
20010 20017
20171
20018 20712
20011
20722 20785
22181
20007
22043
22213, 22207
22027
20002, 20003
22180
22124
20019
22046 22205 22209, 22201
20743
22033
22203
22042
22031
20020
22206 22202
22041
20747
22204
22044
22030
22301 20032
22003
20746
22305
20748
22311
22032
22302
22312 22304
22314
20745
22151
22015

22150

22153

22039

20111

22152

22192

20112

22079

20782

22303, 2231022307

22315

20715

20769

20720

20721

20716

20774

20772

22306

22060, 22309
20119, 20181

20868
20905

20906

20814

20812
20816

20124

20110

20851

20895
20896

22066
20164, 20166

20853

20852

20854

20148

20143, 20155

20832
20855

20850

20147
20117, 20184, 20130, 20135, 20141

20879

20877
20880

20175

Montgomery County, MD

20777, 20833
20860, 20861
20862

20744

20623, 20735

Prince George's
County, MD

22308

20601, 20607

20608, 20613

22193
22191
22134, 22172

22025
22026

Fairfax County, VA

Largest

*Range of house price increase by quintile (rounded to the nearest 1%) is 89% to 127%, 127% to 131%, 131% to 147%, 147% to 165%, and 165% to 204%.
Source: Authors' calculations using data from FNC, Inc.

14

House Price Decline, 2006-2012

Sharper geographic differentiation than in the boom period. House prices fell the most in Prince George's County
and in outlying areas. Prices fell much less in the affluent, close-in areas.

20872, 21771
20180

20871
20882, 21797

Loudoun County, VA
20197
20129
20158

20132

20876

20838, 20839
20841, 20842

20176

20886
20879

20874

20837

20117, 20184, 20130, 20135, 20141

20165

20164, 20166

20152
20151
20120

20109

20143, 20155
20136

By quintile*
Lowest

Second
Third

Fourth

20121

20817
20818

20170 20194

20814

22015

22150

22153

22039

20111

22152

Prince William
County, VA

20901
20910

20815

22192

20112

20015

20012

20783
20903
20912

20707
20866
20708

20705
20740
20770

20782

22303, 2231022307

22315

22306

22060, 22309
20119, 20181

20904

20902

20707
20866

20706
20737 20784
20001
22102
20781
20008 20009
20190, 20191
20784
22101
20710
22182
20016
20010 20017
20171
20018 20712
20011
20722 20785
22181
20007
22043
22213, 22207
22027
20002, 20003
22180
22124
20019
22046 22205 22209, 22201
20743
22033
22203
22042
22031
20020
22206 22202
22041
20747
22204
22044
22030
22301 20032
22003
20746
22305
20748
22311
22032
22302
22312 22304
22314
20745
22151
20812
20816

20124

20110

20851

20868
20905

20906

20895
20896

22066

20105

20853

20852

20854

20148

20169, 20137

20832
20855

20850

20175
20147

20879

20877
20880

20878

Montgomery County, MD

20777, 20833
20860, 20861
20862

20715

20769

20720

20721

20774

20772
20744

20716

20623, 20735

Prince George's
County, MD

22308

22079

20601, 20607

20608, 20613

22193
22191
22134, 22172

22025
22026

Fairfax County, VA

Highest

*Range of house price decline by quintile (rounded to the nearest 1%) 2% to 13%, 13% to 21%, 21% to 31%, 31% to 41%, and 41% to 51%. Bottom quintile includes zips
20007, 20008, 20016, 20815, 22201, 22203, 22204, 22205, 22206, 22209, and 22207 where prices increased.
Source: Authors' calculations using data from FNC, Inc.

15

Land Prices over Time

For DC-area zips grouped into quintiles by quarter-acre land price in 2000:Q1.
Very large swing for low-price zips, much less for high-price zips.
Index, 2006=100

110
100

110
100

90

90

Highest quintile

80

80

Fourth

70
60

70
60

Third
50

50

Second

40
30

40
30

Lowest quintile
20

20

10

10
2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

Note: Observations for 2013 represent average of Q1, Q2, and Q3.
Source: Authors' calculations using data from FNC, Inc. and Marshall & Swift, a CoreLogic company.
16

Prices and Construction Costs: 2000-2006

In DC area, land prices rose more than house prices, especially in zips with low
land prices. Construction costs accounted for little of the rise in house prices.
700

Percent change

600
Land Price

House price

Construction cost

500

400

300

200

100

0
Lowest quintile

Second

Third

Fourth

Highest quintile

Zips grouped into quintiles by quarter-acre land price in 2000:Q1

Source: Author's calculations based on data from FNC, Inc. and Marshall & Swift, a CoreLogic company.
17

Prices and Construction Costs: 2006-2012

In DC area, land prices fell more than house prices, especially in zips with low land
prices, where the declines were staggering. Construction costs continued to rise.
Zips grouped into quintiles by quarter-acre land price in 2000:Q1
Lowest quintile

Second

Third

Fourth

Highest quintile

20
10
0
-10
-20
-30
-40
-50
-60
Land Price

House price

Construction cost

-70
-80
Percent change
Source: Author's calculations based on data from FNC, Inc. and Marshall &Swift, a CoreLogic company.
18

Land Shares over Time

For DC-area zips grouped into quintiles by quarter-acre land price in 2000:Q1. Large
swing in land share for low-price zips, while share is much more stable for high-price zips.
80%

80%

Highest quintile

70%

70%

Fourth

60%

Third

50%

50%

Second

40%

60%

40%

30%

30%

Lowest quintile
20%

20%

10%

10%
2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

Note: Observations for 2013 represent average of Q1, Q2, and Q3.
Source: Authors' calculations using data from FNC, Inc. and Marshall & Swift, a CoreLogic company.
19

Predictive Power of Changes in Land Share

In DC area, house prices fell more from 2006 to 2012 in zips
where the land share rose a lot during 2000-06.
20

House price change, 2006 to 2012, percent

10

0
10

15

20

25

30

35

40

45

50

55

-10

-20

-30

-40

-50

R2 = .31
-60

Maximum rise in land share from 2000 to any year through 2006, percentage points

Source: Authors' calculations using data from FNC, Inc. and Marshall & Swift, a CoreLogic Company.
20

Boston Prices and Construction Costs: 2000-2006

Same general pattern as in DC, though the increases in land prices
and house prices were not as large.
250

Percent change

200
Land Price

House price

Construction cost

150

100

50

0
Lowest quintile

Second

Third

Fourth

Highest quintile

Zips grouped into quintiles by quarter-acre land price in 2000:Q1

Source: Author's calculations based on data from FNC, Inc. and Marshall & Swift, a CoreLogic company.
21

Miami Prices and Construction Costs: 2000-2006

Same general pattern as in DC, though the increases in land prices
and house prices were larger.
1000

Percent change

900
800
700
Land Price

House price

Construction cost

600
500
400
300
200
100
0
Lowest quintile

Second

Third

Fourth

Highest quintile

Zips grouped into quintiles by quarter-acre land price in 2000:Q1

Source: Author's calculations based on data from FNC, Inc. and Marshall & Swift, a CoreLogic company.

22

Boston Prices and Construction Costs: 2006-2012

Same general pattern as in DC, though the price declines were not as large
(except for land prices in the lowest quintile).
Zips grouped into quintiles by quarter-acre land price in 2000:Q1
Lowest quintile

Second

Third

Fourth

Highest quintile

20
10
0
-10
-20
-30
-40
-50
Land Price

House price

Construction cost

-60
-70
-80
Percent change
Source: Author's calculations based on data from FNC, Inc. and Marshall &Swift, a CoreLogic company.
23

Miami Prices and Construction Costs: 2006-2012

Same general pattern as in DC, though the price declines were larger.
Zips grouped into quintiles by quarter-acre land price in 2000:Q1
Lowest quintile

Second

Third

Fourth

Highest quintile

20

-20

-40

-60

-80

-100
Land Price

House price

Construction cost

-120
Percent change
Source: Author's calculations based on data from FNC, Inc. and Marshall &Swift, a CoreLogic company.
24

Boston Land Shares over Time

Same general pattern as in DC, with huge gap in shares between lowest and highest
quintiles and much larger swing in lowest quintile.
80%

80%

70%

70%

Highest quintile
60%

60%

Fourth
Third

50%

50%

40%

40%

Second
30%

30%

20%

20%

Lowest quintile
10%

10%
2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

Note: Observations for 2013 represent average of Q1, Q2, and Q3.
Source: Authors' calculations using data from FNC, Inc. and Marshall & Swift, a CoreLogic company.
25

Miami Land Shares over Time

Same general pattern as in DC, except that share for lowest quintile turns negative.
We will be exploring the source of this negative share in future work.
80%

80%

70%

70%

Highest quintile
60%

60%

50%

50%

Fourth
40%

40%

30%

30%

Third
20%

20%

Second
10%

10%

0%

0%
2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

-10%

2012

2013
-10%

Lowest quintile
-20%

-20%

Note: Observations for 2013 represent average of Q1, Q2, and Q3.
Source: Authors' calculations using data from FNC, Inc. and Marshall & Swift, a CoreLogic company.
26

Boston Predictive Power of Changes in Land Share

Same general pattern as in DC area.
10

House price change, 2006 to 2012, percent

0
5

10

15

20

25

30

35

40

45

-10

-20

-30

R2 = .46
-40

-50

Maximum rise in land share from 2000 to any year through 2006, percentage points

Source: Authors' calculations using data from FNC, Inc. and Marshall & Swift, a CoreLogic Company.
27

Miami Predictive Power of Changes in Land Share

Same general pattern, though less predictive power than in DC.
-10
10

15

20

25

30

35

40

45

50

House price change, 2006 to 2012, percent

-20

-30

-40

-50

-60

R2 = .23
-70

Maximum rise in land share from 2000 to any year through 2006, percentage points

Source: Authors' calculations using data from FNC, Inc. and Marshall & Swift, a CoreLogic Company.
28

Robustness Checks for Price Results

Under way:
Use a market-based stake in the ground in the middle of sample
period instead of near the beginning
Little effect on results for DC
Currently being done for Boston and Miami

Still to be done:
Estimate hedonic land-price indices for comparison
Method 1: Use estimated land prices for zips with large volume of newly-built
homes since 2000
Method 2: Use MLS data on residential land sales

Comparison to hedonic indices important after 2006-07 for areas

with little or no construction
29

What Lies Behind the Boom/Bust Cycle?

We have described what happened to house and land prices over
the recent boom/bust cycle.
What accounts for the observed price patterns?
We have begun to assess the role of fundamentals and credit supply
Fundamentals include household income, availability of jobs, crime, quality of
schools

Today will present initial results for the DC area on the correlation
of household income and availability of jobs with land prices

30

Employment per Working-age Person within a Driving Radius of 10 Miles, 2000

"Job-rich" zips are concentrated in bands running north and west from downtown DC.

20872, 21771
20180

20871
20882, 21797

Loudoun County, VA
20197
20129
20158

20132

20876

20838, 20839
20841, 20842

20176

20874

20837

20878

20175
20165

20147
20117, 20184, 20130, 20135, 20141

20164, 20166

By quintile*
Lowest

Second
Third

Fourth

20832
20855

20850

20853
20851

20895
20896

20817
20818

20170 20194

20868
20905

20906

20852

22066

20814

20902
20901
20910

20815

20812
20816

20015

20012

20904

20783
20903
20912

20707
20866

20707
20866

20705

20708

20740
20770

20706
20737 20784
20001
22102
20781
20008 20009
20190, 20191
20784
22101
20710
22182
20016
20010 20017
20171
20018 20712
20011
20722 20785
22181
20007
20152
22043
22213, 22207
22027
20002, 20003
22180
22124
20151
20019
22046 22205 22209, 22201
20743
22033
22203
22042
22031
20020
22206 22202
22041
20120
20747
22204
22044
22030
22301 20032
22003
20746
22305
20748
22311
22032
22302
20121
22312 22304
22314
20745
20109
22151
20124
22015
22303, 2231022307
22152 22150
22306
22315
20136
20744
20110
20623, 20735
22153
22039
20111
22308

20105

20143, 20155

20879

20877
20880

20854

20148

20169, 20137

20886
20879

Montgomery County, MD

20777, 20833
20860, 20861
20862

20782

20715

20769

20720

20721

20716

20774

20772

Prince George's
County, MD

22060, 22309

Prince William
County, VA

20119, 20181

22192

20112

22079

20601, 20607

20608, 20613

22193
22191
22134, 22172

22025
22026

Fairfax County, VA

Highest

*Range of values by quintile (rounded to the nearest integer) is 0 to 4, 5 to 8, 9 to 13, 14 to 24, and 25 to 59.
Note: Working age defined to be 18-64 years old.
Source: Authors' calculations using data from the Census Bureau.

31

Adjusted Gross Income per Household, 2001

High-income zips are also mainly north and west from downtown but extend further into the suburbs than the "job-rich" zips.

20872, 21771
20180

20871
20882, 21797

Loudoun County, VA
20197
20129
20158

20132

20876

20838, 20839
20841, 20842

20176

20874

20837

20878

20175
20165

20147
20117, 20184, 20130, 20135, 20141

By quintile*
Lowest

Second
Third

Fourth

20853
20851
20852

20895
20896

20817
20818

20170 20194

20868
20905

20906

20814

20902
20901
20910

20815

20812
20816

20015

20012

20904

20783
20903
20912

20707
20866

20707
20866

20705

20708

20740
20770

20706
20737 20784
20001
22102
20781
20008 20009
20190, 20191
20784
22101
20710
22182
20016
20010 20017
20171
20018 20712
20011
20722 20785
22181
20007
20152
22043
22213, 22207
22027
20002, 20003
22180
22124
20151
20019
22046 22205 22209, 22201
20743
22033
22203
22042
22031
20020
22206 22202
22041
20120
20747
22204
22044
22030
22301 20032
22003
20746
22305
20748
22311
22032
22302
20121
22312 22304
22314
20745
20109
22151
20124
22015
22303, 2231022307
22152 22150
22306
22315
20136
20744
20110
20623, 20735
22153
22039
20111
22308

20105

20143, 20155

20832
20855

20850

22066
20164, 20166

20879

20877
20880

20854

20148

20169, 20137

20886
20879

Montgomery County, MD

20777, 20833
20860, 20861
20862

20782

20715

20769

20720

20721

20716

20774

20772

Prince George's
County, MD

22060, 22309

Prince William
County, VA

20119, 20181

22192

20112

22079

20601, 20607

20608, 20613

22193
22191
22134, 22172

22025
22026

Fairfax County, VA

Highest

*Range of values by quintile (rounded to the nearest $1000) is $25,000 to $47,000, $48,000 to $58,000, $59,000 to $72,000, $73,000 to $88,000, and $89,000 to $216,000.
Note: Number of households is proxied by number of tax returns. Data are unavailable for zip group 22025.
Source: Authors' calculations using data from IRS Statistics of Income.

32

Change in Employment within a Driving Radius of 10 Miles per Working-age Person, 2000 to 2012
Job gains concentrated in DC and Virginia. Maryland counties have lagged behind.

20872, 21771
20180

20871
20882, 21797

Loudoun County, VA
20197
20129
20158

20132

20876

20838, 20839
20841, 20842

20176

20874

20837

20878

20175
20165

20147
20117, 20184, 20130, 20135, 20141

By quintile*
Lowest

Second
Third

Fourth

20853
20851
20852

20895
20896

20817
20818

20170 20194

20868
20905

20906

20814

20902
20901
20910

20815

20812
20816

20015

20012

20904

20783
20903
20912

20707
20866

20707
20866

20705

20708

20740
20770

20706
20737 20784
20001
22102
20781
20008 20009
20190, 20191
20784
22101
20710
22182
20016
20010 20017
20171
20018 20712
20011
20722 20785
22181
20007
20152
22043
22213, 22207
22027
20002, 20003
22180
22124
20151
20019
22046 22205 22209, 22201
20743
22033
22203
22042
22031
20020
22206 22202
22041
20120
20747
22204
22044
22030
22301 20032
22003
20746
22305
20748
22311
22032
22302
20121
22312 22304
22314
20745
20109
22151
20124
22015
22303, 2231022307
22152 22150
22306
22315
20136
20744
20110
20623, 20735
22153
22039
20111
22308

20105

20143, 20155

20832
20855

20850

22066
20164, 20166

20879

20877
20880

20854

20148

20169, 20137

20886
20879

Montgomery County, MD

20777, 20833
20860, 20861
20862

20782

20715

20769

20720

20721

20716

20774

20772

Prince George's
County, MD

22060, 22309

Prince William
County, VA

20119, 20181

22192

20112

22079

20601, 20607

20608, 20613

22193
22191
22134, 22172

22025
22026

Fairfax County, VA

Highest

*Range of values by quintile (rounded to the nearest 0.1) is -1.1 to 0.2, 0.3 to 0.6, 0.7 to 1.4, 1.5 to 3.0, and 3.1 to 21.9.
Note: Working age defined to be 18-64 years old.
Source: Authors' calculations using data from the Census Bureau.

33

Percent Change in Adjusted Gross Income per Household, 2001-2011

Gains in income did not follow a crisp geographic pattern. Every jurisdiction had zip groups with small increases and others with large increases.

20872, 21771
20180

20871
20882, 21797

Loudoun County, VA
20197
20129
20158

20132

20876

20838, 20839
20841, 20842

20176

20874

20837

20878

20175
20165

20147
20117, 20184, 20130, 20135, 20141

By quintile*
Lowest

Second
Third

Fourth

20853
20851
20852

20895
20896

20817
20818

20170 20194

20868
20905

20906

20814

20902
20901
20910

20815

20812
20816

20015

20012

20904

20783
20903
20912

20707
20866

20707
20866

20705

20708

20740
20770

20706
20737 20784
20001
22102
20781
20008 20009
20190, 20191
20784
22101
20710
22182
20016
20010 20017
20171
20018 20712
20011
20722 20785
22181
20007
20152
22043
22213, 22207
22027
20002, 20003
22180
22124
20151
20019
22046 22205 22209, 22201
20743
22033
22203
22042
22031
20020
22206 22202
22041
20120
20747
22204
22044
22030
22301 20032
22003
20746
22305
20748
22311
22032
22302
20121
22312 22304
22314
20745
20109
22151
20124
22015
22303, 2231022307
22152 22150
22306
22315
20136
20744
20110
20623, 20735
22153
22039
20111
22308

20105

20143, 20155

20832
20855

20850

22066
20164, 20166

20879

20877
20880

20854

20148

20169, 20137

20886
20879

Montgomery County, MD

20777, 20833
20860, 20861
20862

20782

20715

20769

20720

20721

20716

20774

20772

Prince George's
County, MD

22060, 22309

Prince William
County, VA

20119, 20181

22192

20112

22079

20601, 20607

20608, 20613

22193
22191
22134, 22172

22025
22026

Fairfax County, VA

Highest

*Range of values by quintile (rounded to the nearest 1%) is -1% to 20%, 21% to 26%, 27% to 31%, 32% to 41%, and 42% to 73%.
Note: Number of households is proxied by number of tax returns. Data are unavailable for zip group 22025.
Source: Authors' calculations using data from IRS Statistics of Income.

34

Level of Land Prices Across Zip Groups

Run regression of quarter-acre land price per square foot on
employment per working-age person within 10-mile driving radius and
avg. household AGI.
Land price and employment data are for 2000; AGI data are for 2001
Employment and AGI both have very significant positive effects
Strong explanatory power regression explains about one-half of the
variation in the level of land prices across zip groups
Results using data for 2012 are quite similar

35

Change in Land Prices Across Zip Groups

Run parallel regression to explain the change in quarter-acre land
price per square foot from 2000 to 2012.
Explanatory variables are change in employment per working-age
person within 10-mile driving distance and change in average
household AGI
Changes in AGI are highly significant, changes in employment less so
Regression explains up to one-half of the variation in land-price
changes across zip groups

36

Summary of Results
Housing submarkets are important in Washington, DC area.
House-price swing was:
Mildest in affluent, close-in areas
Considerably more severe in places where land was relatively cheap

Land prices were more volatile than house prices everywhere, but
especially in areas with initially inexpensive land
Large rise in land share was a useful predictor of sharp house price
correction later on
Results highlight the value of focusing on land for assessing
house-price risk
Regarding fundamentals, initial results point to important
connection between income, job availability, and land prices
37

Next Steps
Additional work on fundamental factors (income, employment,
crime, school quality, rents)
Assess contribution of high-risk mortgage lending to explosion in
land prices in certain areas
Extend analysis to other metro areas across the U.S.
Data collection and analysis under way for seven other metros (LA, Phoenix,
Seattle, Chicago, Detroit, Memphis, Oklahoma City)
Eventually, intend to expand coverage to include all of the largest metro
areas and a broad sampling of mid-size cities

38