Transcript EIU proposal

The Future of Global Real Estate
A subscription service
uncovering the future of
global property values
Economist Intelligence Unit
Country and Economic Research
Winter 2009
1
Our proposed methodology
2
A new dawn for real estate?
• Economic boom of the last six years was largely characterised by:
- huge increase in credit and liquidity
- high demand for assets – equities, bonds, commodities, property
• Nevertheless, cheap credit was not the only driver of property prices
- demographic trends
- changes in incomes
Long-term “fundamentals”
- pace of urbanisation
- macroeconomic environment
• But in many markets property prices rose far above a level which could be
justified by these long-term drivers, i.e. above “valuation based on
fundamentals”
• Recent credit crunch accompanied by a steep decline in property prices
3
What about existing real estate research?
• Not many ‘global’ products as such
- different consultancies focussing on different regions
- e.g. Global Insight & Moody’s for US, Jones Lang LaSalle for
separate regions
- coverage mostly for developed / OECD economies
• Many survey based forecasts
- short-term forecasts; limited country coverage
- e.g. PwC “Emerging Trends in Real Estate”
• Modelling based on macroeconomic fundamentals seems
restricted to academic research and international
organisation working papers
- e.g. International Monetary Fund’s (IMF) World Economic
Outlook, 2008; OECD Economic Outlook No.78, 2005
4
Our methodology
• Theoretical background:
- IMF, WEO 2004: “House prices in Australia, UK, Ireland and Spain
exceeded their predicted values by 20 pc”
- IMF, WEO 2007: “During 1997 to 2007 […] house prices were [up to] 30
pc higher than justified by the fundamentals”
- OECD, Economic Outlook
2005:”To address [overvaluation]
it is necessary to relate these
prices to their putative underlying
determinants”
5
Our methodology
• Econometric analysis to arrive at a real estate ‘true value' price
equation
- based on a regression which best explains past price fluctuations given
historical economic and financial data
- determine what should have happened to prices given the path of
economic fundamentals in the past and determine the positive or negative
'price gap‘
• Forecasts: calculate price equation based on our robust in-house
macroeconomic forecasts
- determine the future path of ‘true value' prices of real estate in light of
future macroeconomic conditions
- EIU’s forecasting approach will combine long-term economic forecasting
with property specific factors and will ensure that price forecasts take
appropriate account of the state of the economy and income levels
6
Why the Economist Intelligence Unit?
Independent, long-run perspective required
Some property specialists will forecast property prices based on historic
trends and industry specific factors (such as availability of planning
permits etc). But a truly insightful long run property forecast requires
much more than this - it needs to be rooted in a deep understanding of
the broader national and international economic context.
This is an area in which the EIU has a proven track record. Therefore the
EIU’s forecasting approach, which combines long-term economic
forecasting with property specific factors, is designed to ensure that our
forecasts take appropriate account of the state of the economy and
income levels. Many of the mistakes in forecasting property prices in the
past have arisen because these factors were not taken sufficiently into
account.
7
Why the Economist Intelligence Unit?
World leader in country analysis and forecasting.
For over 60 years we have provided business intelligence that corporate
executives, government officials and academics require to understand
developments around the world.
We cover more than 200 countries, providing economic forecasts on
the world's 150 largest markets.
A truly insightful long run property forecast needs to be rooted in a deep
understanding of the broader national and international economic context.
This is an area in which the EIU has a proven track record.
It is our analytical framework and forecasting methodology that
gives us our competitive edge.
Our approach combines the best in analysis–drawing on the country
expertise of our specialists–and the best in forecasting, grounded in
tested models, carefully vetted data and a quality–control process that
ensures both accuracy and consistency.
8
Our methodology – variables to test
Price equation variables
Dependent variable
Change in real residential/commercial property price
Explanatory variables
Explanation / Hypothesis
Lagged change in real price
‘Persistence’ effect
Price divided by personal income per capita
‘Reversion’ effect or affordability indicator
Growth in personal income per capita
Reflects growing wealth and propensity to buy property
Income and corporation tax rates
Act as downward pressures on the propensity to buy real estate
Short-term interest rate (real and nominal; current and
lagged)
To reflect cost of borrowing for home-owners
Long-term interest rate (real and nominal; current and
lagged)
Reflects long-term financing costs for commercial property development
Change in stockmarket prices
Potential substitute for speculative investment
Population growth
Creating higher demand and upward pressure on prices
Growth in the number of households
Creating higher demand and upward pressure on prices
Population aged 20-39 divided by total population
Reflecting pool of potential first-time buyers of property
Growth in supply of credit as percentage of GDP
To account for credit conditions which influence ability to finance property
acquisition
Unemployment
Business cycle indicator and potential pool of consumers/labour force
Residential/commercial rental yield
To account for buy-to-let investors; also to account for rental market substitute
Global /regional real estate prices
Relative domestic price to global prices, reflecting decision to buy/sell in other
regions
9
Our methodology – UK residential case study
We are already able to accurately model quarterly UK residential property
prices:
1.06
Real house price growth
(Source: DCLG)
EIU model estimate 1.04
1.02
.010
1.00
.005
0.98
Model 1 drivers:
- Income growth
- Previous growth in price
(speculator effect)
- Interest rates
- Population growth
- Growth in domestic credit
- Labour market conditions
.000
-.005
-.010
94 95 96 97 98 99 00 01 02 03 04 05 06 07 08
Residual
Actual
Fitted
But what would have
happened if prices were
driven only by economic
fundamentals?
10
Our methodology – UK residential case study
Annual UK property prices based on ‘fundamentals’:
.3
Real house price growth (Source:
DCLG)
.2
.1
.0
.15
EIU fair price model
estimate
.10
-.1
Model 2 drivers:
-
Income growth
Interest rates
Population growth
Economic development
Labour market conditions
-.2
.05
Actual prices rose faster
than the economic
fundamentals since 1997
.00
-.05
-.10
82 84 86 88 90 92 94 96 98 00 02 04 06 08
Residual
Actual
Fitted
But undervalued from
1990 to 1996
11
Our methodology – Spain residential case study
Again, controlling for fundamentals, residential prices in Spain rose above the
price level explained by the fundamental drivers from 2003. During the
economic downturn, we expect actual prices to converge towards these
“correct” levels and even undershoot based on past trends.
Spain house price index, 2002=100
180
Real house price
160
EIU fair price
140
Price
gap
Model 3 drivers:
- Income growth
- Interest rates
- Population growth
- Labour market conditions
120
100
80
Source: Banco de Espana; Economist Intelligence Unit
60
estimates
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
12
Our methodology – UK commercial case study
We have also applied our approach to commercial property values. The
preliminary results are shown below. Changes in key economic variables are
able to explain much of the change in commercial property prices.
Model 4 drivers:
- Income growth
- Interest rates
- Population growth
- Labour market
conditions
- Residential prices
13
Our proposed research products
14
A new dawn for real estate?
Individual forecasting models of residential and commercial property
prices in a comprehensive group of countries and cities that ascertains
the underlying price level based on long-term fundamentals for each
market.
An exciting research service that will provide subscribers with insight
into the real estate market around the world.
• In which countries is real estate overvalued and how
low are prices likely to fall?
• When can we expect a recovery?
• Which markets are relatively undervalued and where
will the next investment opportunities occur?
15
What will our research provide?
•
•
•
•
There are numerous benefits arising from subscribing to our research
service:
Access key price, economic and financial data for
over 50 countries and 75 cities delivered through
functional Microsoft Excel workbooks
Identify which markets are over- or undervalued and
target your investments effectively
Download exclusive forecast data for residential and
commercial property prices to 2020
Understand the key economic fundamentals driving
real estate market prices around the world
16
Our Residential Property Forecasting Service
1. Real estate database
Access comprehensive data on
residential real estate prices for 53
countries and 65 cities, annual and
quarterly, including latest available
data and historical time series
2. ‘Drivers’ database
Access the Economist Intelligence
Unit’s premium economic and
financial indicator and forecasts
database, updated quarterly
through the Excel workbooks
3. Forecasts and scenario testing
Interactive forecasting models in
Excel format with residential price
projections to 2020 with adjustable
parameters for various forecast
scenarios
4. Briefing papers
Textual analysis on the economic
and political outlook for each
country that guide our overall
residential property forecasts
17
Our Residential Property Forecasting Service
Geographical coverage
Countries – over 50
Americas
Argentina
Canada
Colombia
USA
Western
Europe
Austria
Belgium
Cyprus
Denmark
Finland
France
Germany
Greece
Iceland
Ireland
Italy
Luxembourg
Malta
Netherlands
Norway
Portugal
Spain
Sweden
Switzerland
UK
MEA
Israel
South Africa
United Arab Emirates
Central and Eastern
Europe
Bulgaria
Croatia
Czech Republic
Estonia
Hungary
Latvia
Lithuania
Poland
Serbia
Slovak Republic
Slovenia
Ukraine
Asia Pacific
Australia
China
Hong Kong
India
Indonesia
Japan
Malaysia
New Zealand
Philippines
Singapore
South Korea
Taiwan
Thailand
18
Our Residential Property Forecasting Service
Geographical coverage
Cities – 65
Americas
Boston
Chicago
Denver
Las Vegas
Los Angeles
Miami
New York
San Diego
San Francisco
Washington
Toronto
Montreal
Vancouver
Buenos Aires
Bogota
Western
Europe
MEA
Amsterdam Dubai
Athens
Tel Aviv
Berlin
Birmingham
Brussels
Copenhagen
Dublin
Frankfurt
Helsinki
Lisbon
London
Madrid
Manchester
Milan
Munich
Oslo
Paris
Rome
Stockholm
Vienna
Central and Eastern
Asia Pacific
Europe
Belgrade
Bratislava
Budapest
Kiev
Kosice
Krakow
Ljubljana
Prague
Riga
Sofia
Talinn
Vilnius
Warsaw
Zagreb
Bangkok
Delhi
Jakarta
Kuala Lumpur
Makati
Mumbai
Seoul
Shanghai
Taipei (tbc)
Tokyo
Sydney
Melbourne
Auckland
Wellington
19
Our Commercial Property Forecasting Service
1. Real estate database
Access comprehensive data on
commercial property real estate
prices for 46 countries and 75
cities, annual and quarterly,
including latest available data and
historical time series
2. ‘Drivers’ database
Access the Economist Intelligence
Unit’s premium economic and
financial indicator and forecasts
database, updated quarterly
through the Excel workbooks
3. Forecasts and scenario testing
Interactive forecasting models in
Excel format with residential price
projections to 2020 with adjustable
parameters for various forecast
scenarios
4. Briefing papers
Textual analysis on the economic
and political outlook for each
country that guide our overall
commercial property forecasts
20
Our Commercial Property Forecasting Service
Geographical coverage
Countries – 46
Americas
Western Europe
MEA
Argentina**
Brazil*
Canada
Mexico**
USA
Austria
Belgium
Denmark
Finland
France
Germany
Ireland
Italy
Luxembourg**
Netherlands
Norway
Portugal
Spain
Sweden
Switzerland
United Kingdom
South Africa
Central and Eastern
Europe
Bulgaria**
Croatia**
Czech Republic*
Estonia**
Hungary**
Latvia**
Lithuania**
Poland*
Serbia**
Slovakia**
Slovenia**
Asia Pacific
Australia
China*
Hong Kong
India*
Indonesia**
Japan
Korea
Malaysia**
New Zealand
Philippines**
Singapore
Taiwan**
Thailand**
*composite average of main cities
**principal/capital city only
21
Our Commercial Property Forecasting Service
Geographical coverage
Cities – 75
Americas
Western Europe
Buenos Aires
Sao Paulo
Rio
Mexico City
Atlanta
Boston
Chicago
Dallas/FW
Houston
Los Angeles
Miami
New York
Philadelphia
San Francisco
Seattle
Washington
Amsterdam
Athens
Berlin
Birmingham
Brussels
Copenhagen
Dublin
Frankfurt
Helsinki
Lisbon
London
Madrid
Manchester
Milan
Munich
Oslo
Paris
Rome
Stockholm
Vienna
Zurich
Central and Eastern
Europe
Bulgaria (Sofia)
Croatia (Zagreb)
Czech Republic (2 cities)
Estonia (Tallinn)
Hungary (Budapest)
Latvia (Riga)
Poland (8 cities)
Serbia (Belgrade)
Slovakia (Bratislava)
Slovenia (Ljubljana)
Asia Pacific
Australia (3 cities)
China (3 cities)
India (6 cities)
Indonesia (Jakarta)
Japan (Tokyo)
Malaysia (KL)
New Zealand (Auckland)
Philippines (Manila)
Taiwan (Taipei)
Thailand (Bangkok)
Singapore
22
Fees and project team
23
Fees
• Subscriptions to our Residential Property Forecasting Service and
our Commercial Property Forecasting Service will be available from
December
• The annual fee for a subscription to our Residential Property
Forecasting Service with quarterly updates of the forecasts will be
£10,000/US$16,000
• The annual fee for a subscription to our Commercial Property
Forecasting Service with quarterly updates of the forecasts will be
£10,000/US$16,000
• The annual fee for subscriptions to both services with quarterly
updates of the forecasts will be £16,000/US$25,500
For more information, please contact Catherine Wallen at
[email protected]
24
The team
•
•
•
•
•
Project management team
Andrew Williamson, Global Director Economic Research
Gavin Jaunky, Senior Economist
Robert Metz, Senior Economist
John McNamara, Senior Economist
Harald Langer, Economist
Economics team
• Robin Bew, Editorial Director and Chief Economist
• Robert Ward, Director, Global Forecasting
• Chris Pearce, Director, Economics Unit; Director, Data Services
•
•
•
•
•
•
Regional teams
Charles Jenkins, Regional Director, Western Europe
Pat Thaker, Regional Director, Africa
Laza Kekic, Regional Director, Central & Eastern Europe; Director, Country Forecasting
Services
Justine Thody, Regional Director, Latin America
Gerard Walsh, Regional Director, Asia
David Butter, Regional Director, MENA
25
Our economic forecasting record
26
Predicting 2007 GDP growth in the US
Average forecasting error
0.5
0.4
0.3
0.2
0.1
0.0
Global Insight
Consensus average
Economist Intelligence Unit
Root mean squared error, forecasts made in 2006/07 for 2007 annual real GDP growth figure. Root mean squared error is a measure of
average forecasting error and a commonly used standard in assessing forecasting accuracy It is calculated by taking the square root of the
sum squared of each deviation of the forecast from the actual of each observation divided by the number of observations to arrive at a
standardised score.
27
Predicting 2007 GDP growth in the Euro area
Average forecasting error
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
Global Insight
Consensus average
Economist Intelligence Unit
Root mean squared error, forecasts made in 2006/07 for 2007 annual real GDP growth figure. Root mean squared error is a measure of
average forecasting error and a commonly used standard in assessing forecasting accuracy It is calculated by taking the square root of the
sum squared of each deviation of the forecast from the actual of each observation divided by the number of observations to arrive at a
standardised score.
28
Predicting 2007 GDP growth in Asia
Average forecasting error (Malaysia, Thailand, Indonesia,
Taiwan)
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
Global Insight
Consensus average
Economist Intelligence Unit
Root mean squared error, forecasts made in 2006/07 for 2007 annual real GDP growth figure. Root mean squared error is a measure of
average forecasting error and a commonly used standard in assessing forecasting accuracy It is calculated by taking the square root of the
sum squared of each deviation of the forecast from the actual of each observation divided by the number of observations to arrive at a
standardised score.
29
Predicting 2008 global GDP growth
Average forecasting error
1.0
0.8
0.6
0.4
0.2
0.0
IMF
Consensus average
Economist Intelligence Unit
Root mean squared error, forecasts made in 2007 for 2008 annual real GDP growth figure. Root mean squared error is a measure of
average forecasting error and a commonly used standard in assessing forecasting accuracy It is calculated by taking the square root of the
sum squared of each deviation of the forecast from the actual of each observation divided by the number of observations to arrive at a
standardised score.
30
Our long-run forecasting methodology
APPENDIX I
Growth projections
The main building blocks for the long-term forecasts of key market and macroeconomic
variables are long-run real GDP growth projections. We have estimated growth
regressions (based on cross-section, panel data for 86 countries for the 1970-2000 period)
that link real growth in GDP per head to a large set of growth determinants. The sample is
split into three decades: 1971-80, 1981-90 and 1991-2000. This gives a maximum of 258
observations (86 countries for each decade); given missing values for some countries and
variables, the actual number of observations is 246. The estimation of the pooled, crosssection, panel data is conducted on the basis of a statistical technique called Seemingly
Unrelated Regressions. (SUR) to allow for different error variances in each decade and for
correlation of these errors over time.
The regressions, which have high explanatory power for growth, allow us to forecast the
long-term growth of real GDP per head for sub-periods up to 2030, on the basis of
demographic projections and assumptions about the evolution of policy variables and
other drivers of long-term growth.
31