Transcript Sector S14-S15

Towards longer national account
data time series
Joint project of the OECD
Pioneer Investments and UniCredit
Teresa Sbano and Michèle Chavoix-Mannato
Introduction
 The OECD, jointly with the Economic research Unit of Pioneer Global Asset
Management and UniCredit, is working on a project that aims at extending
back the currently available time series of the Financial Accounts for a group of
OECD countries. This is relevant especially for the following reason:
– the introduction of the new method of classification based on the SNA-93
during the 1990s has made it necessary to reconcile past data with the new
series,
– in order to produce as much as possible consistent time series covering a
reasonably long time horizon,
– to be used both for studying the changes in the allocation of wealth over time
and for comparing the main trends across different countries
Page 2
Opportunities arising from the analysis of longer time series:
 Longer time series give the opportunity to better analyse many
relevant issues such as:
– The evolution of the financial structures over the time
– The evolution in the liabilities side of non-financial corporations
– Correlations between economic cycles and financial cycles
– Convergence in the financial structures between European countries, Japan,
US and Canada
– Measures of wealth effects
– Models on the demand of financial assets
– Measures of the degree of openness of the different financial markets
– The evolution and relative importance of financial institutions
Page 3
Summary: available national account data
Country
Period
Sources
 ITALY
from 1950 to 2004
Bank of Italy
 US
from 1950 to 2004
Federal reserve
 CANADA
from 1960 to 2004
OECD. stat
 GERMANY
from 1970 to 2004
OECD.stat/ golden book
 SPAIN
from 1980 to 2004
OECD.stat/ golden book
 FRANCE
from 1980 to 2004
OECD.stat
 JAPAN
 UK
from 1980 to 2004
OECD.stat
from 1987 to 2004
OECD.stat
Page 4
Households- and non financial corporation - two of the five
institutional sectors
Households and nonprofit organization
serving households
Non-Financial
Corporations
Total economy
+
Rest of the world
Rest of the
world
Financial
Corporations
General
Government
Page 5
From the golden book to the SNA 93
The main changes within SNA 93 refer to the definition of institutional sectors and the
classification and method of valuation of financial instruments
SNA 93
Golden Book
Page 6
Conversion table:
In order to merge data from the two sources, we have constructed the following
conversion table, which allows us to obtain a good degree of correspondence both over
time and across the majority of the countries analysed:
Page 7
Reasons for the discrepancies between the two sources:
 Classification of institutional sectors and financial instruments
– Changes in the definitions of Household and non financial enterprises
– A more detailed breakdown in the SNA 93 classification
– A changes in the definition of instruments
 Financial instrument should undergo a market valuation
– ESA 95 methodology establishes that the valuation of financial balance sheets of
stocks of financial assets and liabilities shall be at market price. The items most
affected by this type of valuation are shares and the other equity and to a lesser extent
the item securities other than shares. The application of this valuation rule is
problematic for the instruments where valuation cannot be directly calculated
Page 8
We can use directly the data published by the central banks
US From 1950 to 2005
Federal reserve
JAPAN From 1980 to 2005
OECD.stat
The Federal reserve has been producing the series since 1950
The Federal Reserve makes some adjustments in order to harmonize its data with the
OECD format
Work on harmonization also by the Bank of Japan
The Bank of Japan has begun releasing the retrospective data for the Flow of Funds
Accounts (based on SNA93): from fiscal year 1980 to 1989, and stock data from end of
fiscal year 1979 to end of fiscal year 1988.
CANADA
From 1960 to 1997
Golden Book
From1970 To 2005
OECD.stat
Also from Statistics Canada we have data consistent with SNA93.
From Statistics Canada we have data consistent with SNA93 starting from 1970.
Looking at the relevant 1969-70 period for series continuity:
for non-financial corporations, the data line up fairly well, with
the only significant gap being a re-allocation of sub-instruments
between F5 and F7;
for households, the data also line up fairly well, with the major
difference being a historical revision affecting government unfunded
pension schemes.
Page 9
We can use directly the data published by the central banks
ITALY From 1950 to 2005
The Bank of Italy Research Department has already produced series since 1950
Regarding stocks data (assets and liabilities) for all the institutional sectors
and for the main financial instruments
Bank of italy
FRANCE -
From 1980 to 2005
We have data from 1980 in the OECD dataset, but historical data are not yet
definitive Banque de France is working on the construction of a more consistent
data set
OECD.stat
AUSTRIA -
In the Golden Book the Austrian data are not available.
From1995 To 2005
The Bank of Austria is working on a longer time series (from 1989) and the
dataset will be ready for the end of the year
OECD.stat
UK –
From 1987 to 2005
OECD.stat
In the Golden Book the Uk data are available from 1987 to 1997. However,
information on balance sheet figures by sector exists in the annual
national accounts publications and it is available from 1975 to 1986. In
order to use this longer time series we need to harmonize them
Page 10
Germany analysis
 GERMANY –
 From 1970 To 2005
 Golden book data set
 - From 1970 to 1997
 And OECD.stat
 - From 1991 to 2005
Starting from 1990 data refer to all Germany previous figures
pertain to western Germany alone
The most important difference from the SNA lies in the
delimitation of the sectors "Non-financial enterprises" and
"Households". In the SNA, the sector "Non-financial enterprises"
is confined to corporate and "quasi-corporate" enterprises,
whereas all other enterprises (small- and medium-sized
proprietorships, self-employed persons, farmers, publicly-owned
undertakings) are included among households or public
authorities, as the case may be. In the capital finance account of
the Deutsche Bundesbank, the enterprise sector is much more
comprehensive and includes enterprises of all legal forms (sector
g, "Producing enterprises"). A separate sector includes private
and public housing activities (sector h2, "Housing"). The
heading "Households" virtually covers only the sphere of private
consumption and saving, but includes saving by self-employed
persons only insofar as it does not represent the net retained
income of enterprises.
Page 11
Germany – Sector S14-S15
Important
discrepancy in the
liabilities side of the
Household sector.
After including the
housing sector in
the Sector S14-S15
we filled the gap
between the two
datasets
Page 12
GERMANY
Sector S14-S15
Sector S11
Page 13
Household financial assets as a percentage of GDP
Household financial assets as a % of GDP
400%
France
350%
Germany
300%
Spain
250%
Italy
UK
200%
US
150%
Canada
100%
Japan
50%




20
04
20
02
20
00
19
98
19
96
19
94
19
92
19
90
19
88
19
86
19
84
19
82
19
80
0%
household wealth as a percentage of GDP presented an upward trend in all countries
analysed.
The total financial asset in % of GDP appears to have increased moderately during
the period 1980 to 1990.
First signs of an important growth are evident starting from 1995 mainly in UK and US.
Rankings are apparently relatively stable, but volatilities different, possibly because of
different asset composition.
Source: PGAM estimation on Central Bank data
Page 14
Liabilities of private sector: Households
Household financial liabilities as a % of total GDP
120%
France
100%
Germany
Spain
80%
Italy
60%
UK
40%
US
Canada
20%
Japan
04
20
02
20
00
20
98
19
96
19
94
19
92
19
90
19
88
19
86
19
84
19
82
19
19
80
0%
Household liabilities relative to GDP seem to highlight differing trends.


In the Euro-zone the data shows a moderate increase during the period under review,
with the only exception of Spain which has had an important growth during the last 8
years.
In US and UK we witnessed an impressive growth of household liabilities in the last 5
years
Source: PGAM estimation on Central Bank data
Page 15
Liabilities of private sector: Non-Financial corporations
Non-financial corporation liabilities as a % GDP
500%
France
400%
Germany
Spain
300%
Italy
UK
200%
US
100%
Canada
Japan


20
04
20
02
20
00
19
98
19
96
19
94
19
92
19
90
19
88
19
86
19
84
19
82
19
80
0%
It would appear that for Non-financial Corporations the percentage of liabilities in
relation to the GDP in the countries under analysis has increased steadily between
1995 and 2000 with the only exception of Canada and UK and Japan.
The levels seem structurally different across countries.
Source: PGAM estimation on Central Bank data
Page 16
A dynamic financial structure in all the macro areas with the only
exception of Japan
It would appear that we are witnessing:
•
A sensible increase of the share of Insurance
and pension funds (AF61).
•
Stability in the role of shares and other equity
(AF5)
•
At a more detailed level, the indirect
participation on the financial market through
mutual fund and retirement product in the last
10 years appears to have
increased
significantly
•
An impressive decrease of safe assets
(currency and deposits – AF2) with a
noticeable switch towards more risk oriented
products and long term instruments
•
Structural differences in relative holdings of
risky assets
Source: PGAM estimation on Central Bank data
Page 17
Strong disparities across countries, each of them shows a specific
profile
Despite the common trend (Securitization and long term investing), Europe appears still
a quite heterogeneous region




Spain, Germany and France are characterised by the importance of banking deposits
UK presents a high level of long term investment (Life insurance and Pension)
Italians stand out because of the importance of fixed income securities
In Italy and Spain the share of insurance technical reserves and pension funds still remains well
below average
Source: PGAM estimation on Central Bank data
Page 18
Conclusions
 The aim of this project is to fill a gap in the Financial Accounts time series in
order to help both practitioners and academics to better understand current
trends.
 I would like to express on behalf of the Steering Committee and the research
team our warmest thanks to those who have helped us so far.
 We have been encouraged by the first responses to our work and hope to be
able to construct in a reasonable amount of time a more complete and as far as
possible reliable dataset.
Page 19
CANADA
Sector S14-S15
Sector S11
Page 20
USA
Sector S14-S15
Sector S11
Page 21
JAPAN
Sector S14-S15
Sector S11
Page 22
SPAIN
Sector S14-S15
Sector S11
Page 23
ITALY
Sector S14-S15
Sector S11
Page 24
FRANCE
Sector S14-S15
Sector S11
Page 25
UK
Sector S14-S15
Sector S11
Page 26