Transcript Demographic Change and the Demand for Housing in England

Housing Markets,
the Economy,
and the Crisis
Piet Eichholtz
Maastricht University
Housing market crises and economic crises
• IMF recently investigated 122 recessions in 21 OECD
countries 1960 and 2007
• 28 recessions coincided with a severe housing market
crisis
– Recessions with housing market crisis are deeper and longer
• Cumulative output loss 3.7% vs 2.3% for normal recession
• Last one quarter longer
– Housing investments shrunk 8 to 12 quarters
• Effects far deeper than for equity market crises
– GDP loss housing market crisis: 3,4% (pre) to 0,8% (post)
– GDP loss equity market crisis: 4,0% (pre) to 2,6% (post)
• How is that possible?
Mechanism is saving and consumption
• House prices effects on consumption growth larger than
equity price effects
– House price rises seem(ed) more permanent
– Banks are more willing to finance more and/or home owners willing
to save less
• 2001 - 2005 US housing wealth growth explains part of
consumption growth
– Effect is especially strong for young households. Older people are
more careful
• Development housing market explains quarter of Dutch
economic growth 1995 - 2000 (DNB, 2003)
How does home ownership affect household behavior?
Consumption effect housing much bigger than financial wealth
• Lettau and Ludvigson (2004)
– Unexpected wealth shocks must be perceived as permanent to affect
consumption spending
• Case, Quigley, and Shiller (2005)
– Look at macro data across US states and across OECD countries
– Changes in housing wealth affect consumption more than changes in
financial (i.e. stock market related) wealth
• Bostic, Gabriel, and Painter (2005)
– US housing wealth growth (2001-2005) explains 12% consumption growth
– Effect especially big for non-durable consumption
– Effect strongest for young people. Older people more careful
• Engelhardt (1995) and Skinner (1996)
– Effect of US house wealth changes on saving is asymmetric
– Gains in wealth decrease savings by 0.4 cent per dollar
Housing markets kept the world economy
afloat between 2001 and 2003
• Crisis equity markets 2000-2003 comparable to ’29 …
– US: loss in equity wealth 2000-2001 = $ 8 trillion
– NL: loss in equity value 2000-2003 = 65% or € 574 billion
• … but that recession was relatively mild
• Housing markets delivered wealth compensation
– US: gain in housing wealth 2001-2004 = $ 9.6 trillion
– NL: gain in housing wealth 2000-2005 = 21%, or € 228 billion
But this came at a cost
Credit growth and house price changes, EU cross section
And this is the result
Outstanding mortgages in percentage of GDP
Selected countries, 1983 and 2007
What happens when house prices go down?
• Consumption effect is unfortunately larger for price falls
than for price growth
• Saving goes up substantially
• Engelhardt (1996): asymmetric effect of house prices on
US consumption in 1980s
– Only price falls matter
• Skinner (1996): every dollar of house price fall leads to 10
cents in extra savings in the US
• Dutch Central Planning Bureau (2006) found effect of
price falls to be 2.5 to 5 times as strong as price rises
Will European house prices fall across the board?
• Structural price falls sofar mostly in countries and regions
with large over supply and/or massive fall in demand
– (US), Spain, Ireland, UK
• “Every housing downturn begins with an excess supply of
available dwellings” (Case and Quigley, 2008)
A scenario for years to come?
Netherlands early ’80s: real price fall 48%
400,00
Nominale index
Reële index
300,00
200,00
100,00
0,00
1965 1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989
House price developments in Europe (1)
Mostly booms until 2007 …
House price developments in Europe (2)
… but not all booms turn into busts …
… and one market does not boom at all
House prices relative to income give better
perspective on booms …
… and Germany is still the great outlier
A housing market with history: Amsterdam
Volatile with booms and busts, but stationary in real terms
350
300
250
200
150
100
50
0
1650
1700
1750
rents (1650=100)
1800
1850
1900
1950
house prices (1650=100)
2000
And how about the U.S. housing market?
Just had the biggest boom/bust in history
Components of crisis differ strongly by country
• Market sentiment still bad in many countries
– Large hardware spending postponed, consumption still
reasonable in many countries
• Demografics is getting problematic in many regions in
east and south
– Central Europe, Club Med
– Without large scale intervention, these markets are doomed
• Economies may get back into recession
– Unemployment varies strongly across countries
• Interest rates (euribor, IRS) still very low
– Liquidity premiums are falling
• Oversupply problematic in some countries
– But not in most of Netherlands
Expected demographics: change in total population
Large differences across Europe until 2050
Change total pop.
2010-2050
Demographics, Netherlands 1995-2050
800.000 households to go
Aantal huishoudens (+ prognose)
8500
8000
x1000
7500
7000
6500
6000
1995
2000
2005
2010
2015
2020
2025
2030
2035
2040
2045
2050
Bron: CBS
Demografics also has positive side
Older home owners ‘consume more house’
Demand increases with “human capital”
– Higher education, income, and better health lead to more housing demand
Demand increase with age
– Households only consume less house at end of life (80+)
Bron: CBS
Economy, construction and jobs
Dutch labor market still very strong
•
•
Housing construction still very low
Job loss in construction sector (estimated): 50.000
– Inclusiding secondary job losses
– Started late 2009
•
•
•
OECD’s biggest worry about Dutch labor market was structural
tension
Unemployment has not gone above 6%
Labor population set to shrink, further reducing probability of structural
unemployment
Long interest rate seems very low …
Dutch 10-yr government bond yield, 1980-2010
14
12
10
8
6
4
2
0
80
82
84
86
88
90
92
94
96
98
00
02
04
06
08
10
Bron: Datastream
… but is it really that low?
Dutch rate averaged 4.4% from 1814 to 1994
10%
8%
6%
4%
2%
19
94
19
74
19
54
19
34
19
14
18
94
18
74
18
54
18
34
18
14
0%
Source: Eichholtz and Koedijk (1996)
Credit crisis, Interest Rate and Construction
• Credit crisis influences both financing of new supply and
demand
– So effect on prices is unclear
• Financing of new construction
– Rate is still low because of low government bond rate …
– … and despite very high spreads
– Availability of credit is the larger problem
Construction of new dwellings Netherlands
Still dramatically low and falling again
Conclusions
• Housing markets can dampen crises, but can also make
them more severe
– Housing wealth drives consumption and saving
• The current housing crisis is not global or even European
• Is the Dutch housing market in a crisis?
– Yes: no liquidity, no construction, no activity
– No: prices have not really fallen
• Will prices fall in a big way? I think it’s unlikely …
– Net supply is low
– Mortgage rates not likely to go up much
• … but on the other hand
– Demographics are turning sour
– Government policy very uncertain
– But economic news
Reading suggestions
•
Hernando de Soto
–
•
The Mystery of Capital, Bantam Press, 2000
Ed Glaeser
–
Triumph of the City: How Our Greatest Invention Makes Us
Richer, Smarter, Greener, Healthier, and Happier , Penguin
Press HC, 2011
Extra: Housing supply and spatial policy
Scarcity housing is not linked to population density
• International (EU) relationship between population density
and ratio housing cost / income is 0.06
• Price elasticity of Dutch housing supply is zero
• Scarcity of housing land is political choice
– Supply housing land 1981-2003 grew 39% in Belgium; 12% in
Netherlands
– Ratio housing land / agricultural land (2000)
• Netherlands: 9.5%
• Denmark: 11.9%
• Belgium: 12.8%
– If the Netherlands would use the Belgian ratio, the supply of
housing land would rise 31%
Be
Ne lg
de ië
r la
nd
Pr
ov
Vl
in
cie aan
An der e
tw n
er
p
Fr en
ies
lan
Dr d
en
Gr th
on e
in
ge
Ze n
ela
n
Fl
ev d
ol
an
Br d
us
Ov sel
er
ijs
se
L
No
im l
or
d- burg
Br
a
Ge ban
t
Zu lder
i d l an
-H d
ol
la
n
No U d
or tre
d- ch
Ho t
lla
nd
prijs (€/m2)
And what are the consequences?
Residential land prices in Belgium and Holland, 2005
600
500
400
300
200
100
0
Supply and scarcity
Housing production and prices
Cost components of new dwellings
The Dutch Experience