Transcript eres2010_210.content

•
From the beginning of the global financial
downturn, the observation of residential
mortgage market drivers has been focused
on their role in triggering the world crisis.
•
Today, we want to analyze how the
residential mortgage market could support
the upturn and, consequently, how it could
help to solve the present situation.
•
What is the impact of the global financial
crisis on European residential mortgage
markets?
•
What differences between European
countries come to light from
investigating support and restraint factors
of the development of European
residential mortgage markets?
•
This study is part of the wide international
debate about the global financial downturn
and, as a first step, aims to provide in-depth
analysis of residential mortgage market trends
in some European countries.
•
As a second step, the study will investigate the
key variables of the residential mortgage
supply with the goal of highlighting analogies
and differences at cross-border level and
pointing out some trends.
•
The survey draws on data on mortgage
stocks from Quarterly Review - European
Mortgage Federation (EMF)
The countries observed are Great Britain,
Germany, France, Spain, Italy and Poland.
• Analysis of the five main European markets
was accompanied by the study of the
Polish market, one of the most dynamic in
recent years, as an example of the trends in
developing mortgage markets.
•
•
The aim was to assess whether mortgage
market trends can be explained by a
model using as independent variables
some of the main macro-economic
variables:
–
–
–
–
–
–
Gross Domestic Product (GDP);
Population;
Unemployment rate;
Inflation rate;
Gross disposable income;
Household spending.

The methodology applied consists of a
non-linear stepwise forward regression
model.

First, we analysed the period 1998-2008,
and then the period 1998-2007, in order
to isolate the effects of the economic
crisis in the year 2008.
Great
Britain
Germany
France
Spain
Italy
Poland
N. of cases
11
11
11
11
11
11
R²
0,97035890
0,90052840
0,97884939
0,99689673
0,99808463
0,99553619
adjusted R²
0,96294862
0,83421400
0,97356173
0,99482789
0,99726376
0,99107239
df
2,8
4,6
2,8
4,6
3,7
5,5
F
130,9478
13,57968
185,1198
481,8614
1215,883
223,0240
P
0,000001
0,003643
0,000000
0,000000
0,000000
0,000007
GDP
0,561
-2,800
1,100
0
0
0,638
Population
0
-0,460
0
2,140
0,403
-0,130
Unemployment rate
0
0
0
-0,180
0
-0,220
Inflation rate
0
-0,760
-0,160
0,097
-0,070
0
Disposable income
0
0
0
-1,300
0
-0,950
Household spending
0,454
4,480
0
0
0,639
1,040
Selected variables
(BETA)

First of all, the analysis confirms that there
are big differences between countries in
terms of the relationships between the
macro-economic variables and
mortgage stocks.

We can also assess the degree of
significance of each independent
variable.

We identified for each independent
variable, the number of countries where
the variable was “very significant” and
“significant”;

We also identified the countries where
the variable shows a different sign
compared to that commonly found in
the literature.
Very
Significant
Opposite sign
significant
GDP
3
1
1
Population
2
2
2
Unemployment rate
2
0
0
Inflation rate
1
3
1
Disposable income
1
1
0
Household spending
3
1
0
•
Every macro-economic variable proved
significant or very significant in explaining the
mortgage market trends in at least two
countries.
•
However, none of the variables was significant
in more than four countries. So, we find varying
degrees of intensity of the relationship between
variables and mortage stocks among the six
countries.
•
Finally, three variables show in at least one
country a significant relationship, but with an
opposite sign compared to what would be
predicted.
•
Has the financial crisis affected the
trends of these relationships?
•
In order to assess the impact of the
global crisis on these relationships the
analysis was repeated without the figures
of 2008, the first year the crisis hit the
financial retail markets.
N. of cases
R²
adjusted R²
Great Britain
Germany
France
Spain
Italy
Poland
10
10
10
10
10
10
0,99910793 0,90062298
0,9998763
0,9643565
0,99897922
6
0,99553619
0,9983974
0,99839428 0,82112136 0,99962891 0,99770325
9
0,95417273
Df
4,5
4,5
6,3
5,4
3,6
2,7
F
1399,987
11,32836
4041,615
782,9161
1870,065
94,6948
P
Selected variables
(BETA)
0,000000
0,010123
0,000006
0,000005
0,000000
0,000009
GDP
0,346
-2,2
0,788
0
0
2,02
Population
0,523
-0,47
-1,3
1,89
0,471
0
Unemployment rate
0,063
0
0,13
-0,27
0,129
0
Inflation rate
0
-0,59
-0,09
0,085
0
0
Disposable income
0,177
0
0,104
-1,25
0
-1,1
Household spending
0
3,88
1,56
1,29
0,673
0

Even in the period 1998-2007 we found that
the relationships between some macroeconomic variables and the mortgage
market trends were anomalous in that the
sign was different compared to what would
be expected:
› GDP and population in Germany;
› Population in France;
› Unemployment rate and disposable income in
Spain.

The reason for anomalous results does not
have anything to do with the crisis.
Very significant variables 1998-2007
Very significant variables 1998-2008
Great Britain
Disposable income
GDP
Germany
Household spending
GDP, household spending
Population, unemployment rate, inflation
GDP
France
Spain
rate, household spending
Population, unemployment rate,
Population, unemployment rate,
disposable income
disposable income
Population, household spending
Population, inflation rate, household
Italy
spending
GDP, disposable income
Polond
Unemployment rate, household
spending
•
The comparison of the significant
variables in the two periods shows small
differences for most countries.
•
These variables are confirmed as
significant in both periods:
– Household spending in Germany;
– Population, unemployment rate and
disposable income in Spain;
– Population and household spending in Italy.

We sought confirmation of these results
through comparison of the total figures
for the six countries:
› for the year 2007 and
› for the year 2008.
2008
2007
N. of cases
6
6
R²
0, 99973100
0,99843823
adjusted R²
0, 99731034
0,99219117
Df
4,1
4,1
F
464,4936
159,8253
P
0, 034784
0,059248
GDP
3,930
2,640
Population
0
0
Unemployment rate
0,404
0,390
Inflation rate
0,090
0
Disposable income
-11,000
-9,900
Household spending
8,300
8,420
Selected variables
(BETA)

The analysis of the total figures confirms that
some macro-economic variables are very
significant independently of the historical
period (economic stability or beginning of
the crisis).

GDP, unemployment rate, disposable
income and household spending are
confirmed as significant variables to explain
mortgage trends in both 2007 and 2008, the
beginning of the crisis.

Legislation needs to take account of the
very strong and stable links between
macro-economic variables and
mortgage markets.

However, as the study has shown, the
relationships show different degrees of
significance and sometimes even
opposite signs in the different countries.

So, incentives (or disincentives) for the
household credit market, and thus house
purchase, require a differentiated
approach for each country.

European wide policies are necessary,
but they must be based on robust
empirical evidence and models
adapted to the country of application.