The analysis of Households wealth with and beyond national accounts

1
The analysis of Households wealth with and
beyond national accounts

• WPFS, 13 October 2008, OECD, Paris
• Denis Marionnet, Banque de France

2
Introduction (1/3)

• The most common use of national accounts to
  analyse HHs wealth consists in combining
• data from financial accounts
• outstanding amounts
• financial transactions
• revaluation flows (when available)
• and/or data from total wealth accounts
• land
• housing
• These data allows analysing
• the evolution in time of the structure of total
  portfolio or financial portfolio (stocks),
• financial transactions and arbitrages between
  financial instruments,
• growth rates of different assets, indices on
  outstandings,
• international comparisons,
• With various tables and graphs

3
Introduction (2/)

• Examples taken from the Banque de France Bulletin
  Digest No 148, April 2006

• http//www.banque-france.fr/gb/publications/telech
  ar/bulletin/148etud1.pdf

4
Introduction (2/2)

• Examples taken from the ECB Monthly Bulletin,
  November 2007

• http//www.ecb.int/pub/pdf/other/pp75-87_mb200711e
  n.pdf

5
Overview

• Brief presentation of 4 papers that propose ways
  to supplement the traditional analysis of HHs
  wealth
• 1. Enhancing the analysis by combining micro data
  from Wealth Surveys with national accounts
• The composition of household wealth between 1997
  and 2003, P. Girardot and D. Marionnet,
  Quarterly Selection of Articles of the Banque de
  France, n12, Summer 2008
• 2. Enhancing the analysis by making financial
  intermediation transparent
• La destination finale de lépargne des ménages,
  A. Rincon, Bulletin de la Banque de France,
  n167, November 2007
• The final financial investment of French
  households, D. Marionnet, Irving Fisher
  Committee Bulletin, n25, Bank for International
  Settlements, March 2007
• 3. Enhancing the analysis by taking into account
  implicit social security and pension wealth in
  HHs assets
• Implicit social security and pension wealth in
  households assets in the US and France, D.
  Durant and M. Reinsdorf, paper presented at the
  International Association for Research in Income
  and Wealth, August 2008

6
1. National accounts data combined with Wealth
Surveys

• The idea was to combine in the same article
• National financial and non-financial accounts
  data drawn up by INSEE and the Banque de France
• With a particular look at valuation effects on
  both real estate and financial assets
• Households Wealth Surveys data conducted by
  French NSI, INSEE, in 1998 and 2004, adjusted
  against these macroeconomic data
• This Survey looks at the changes in the
  distribution of household wealth (real-estate,
  financial and professional) and the holding rates
  of the various assets. It also includes very
  detailed information on the factors underlying
  households investment behaviour family and
  professional biographies, inheritance and
  donations, income and financial position, motives
  for holding/ not holding a certain type of asset.
• The article The composition of household wealth
  between 1997 and 2003, P. Girardot and D.
  Marionnet, Quarterly Selection of Articles of the
  Banque de France, n12, Summer 2008
  (http//www.banque-france.fr/gb/publications/telec
  har/bulletin/qsa/qsa12etud_6.pdf)

7
1. National accounts data combined with Wealth
Surveys

• Methodological aspects
• In order to bring the scope of the financial
  accounts in line with that of the INSEE Wealth
  Survey, some assets hardly held by households
  were not included in the analysis.
• In addition, for comparison purposes, financial
  assets were grouped together at the macroeconomic
  level in a different manner than in the usual
  presentation of financial accounts.
• Because the survey data tend to underestimate
  outstanding amounts relative to national
  accounting data, the results presented in this
  article are based on the amounts adjusted against
  wealth outstanding amounts in the national
  accounts .

8
1. National accounts data combined with Wealth
Surveys

• Between 1997 and 2003, gross household wealth
  increased rapidly (average annual increase of
  8.6 compared to 4.1 for their gross disposable
  income)
• Mainly as a result of a boom in real-estate
  prices
• Real estate accounted for 58 of households
  private wealth (excl. professional wealth) in
  1997, compared with 66 in 2003.
• Contributions to the increase of total wealth
• rise in real estate assets 78
• of which increase in the value of housing and
  land 61
• of which increase in investment flows 17
• increase in households financial assets 22
• of which cumulated annual investment flows 21)
• of which increase in the value of financial
  assets 1

9
1. National accounts data combined with Wealth
Surveys

• Illustration

10
1. National accounts data combined with Wealth
Surveys

• Wealth inequalities increased between 1997 and
  2003
• The Gini index (concentration indicator
  calculated on the basis of private wealth) rose
  from 0.614 in 1997 to 0.629 in 2003.

• Thus, the private assets held by the 10 least
  wealthy households amounted to less than EUR
  2,110 in 2003, compared with EUR 1,930 in 1997,
  whereas those held by the 10 wealthiest
  households stood at over EUR 450,060 in 2003, as
  against EUR 296,400 in 1997.
• Homeowners posted a very significant rise in
  their gross private wealth, while on average
  households holding only financial assets have
  benefited less.

11
1. National accounts data combined with Wealth
Surveys

• Illustration detailed tables presenting HHs
  total, real estate or financial wealth broken
  down according to households characteristics

12
1. National accounts data combined with Wealth
Surveys

13
1. National accounts data combined with Wealth
Surveys

• Five types of more or less diversified financial
  portfolios drawn up using a hierarchical
  ascending classification method

14
1. National accounts data combined with Wealth
Surveys

• This typology can be linked to the structure of
  private wealth by decile as households are more
  likely to diversify their portfolio when their
  personal environment is favourable stable family
  situation, high income, no risk of unemployment.

15
1. National accounts data combined with Wealth
Surveys

• Risky financial assets higher holding rates in
  2003 than in 1997
• HHs financial savings increasingly channelled
  into risky investments the share of their risky
  assets rose from 14.0 in 1997 to 19.2 in 2003,
  after having peaked at 23.8 in 2000.
• In a context characterised by large fluctuations
  in financial market prices, it is useful to
  define the characteristics of holders of risky
  financial assets and to attempt to explain their
  behaviour using an all other things being equal
  analysis.
• The qualitative model used measures the impact of
  each household characteristic (age,
  socio-occupational category, type of household,
  whether the parents were holders of risky assets,
  income, etc.) on the probability of holding a
  risky financial asset.
• What is measured is the gap between the
  probability of holding a risky financial asset in
  the reference situation and in the situation
  under review. The reference situation is that of
  households between 40 and 49, in an
  intermediate-grade occupation in the public
  sector, with two children and in the 5th decile
  of private wealth and disposable income.
• Main findings
• The head of the typical household holding risky
  assets has a high income and considerable wealth.
  His parents also held risky assets he has no
  children.
• Similar considerations seem to prevail among
  managers in 2003, they hold risky assets
  slightly more often than the other social
  categories.
• In general, households investment behaviour is
  carried over from one generation to the next.
• Overall, households wealth, level of income and
  degree of information on financial investments
  are key determinants of their propensity to hold
  risky assets.

16
2. Making financial intermediation transparent

• As the proportion of intermediated instruments in
  HHs financial wealth increases in many
  countries, the interest in X-raying their
  financial investment gets increasingly relevant.
• Using a method that makes financial
  intermediation transparent is one way of doing
  so, making possible to complement the use of
  financial accounts for the analysis of HHs
  financial wealth. Indeed, SNA93 and ESA95
  nomenclatures do not always provide sufficient
  detail for a complete analysis of HHs wealth and
  the risks they bear.
• This method was first used by M. Boutillier et
  al. in Placements des ménages en Europe le
  rôle des intermédiaires financiers se transforme
  en profondeur, Economie et Statistiques, n354,
  pp. 85-102, 2002.

17
2. Making financial intermediation transparent

• This approach consists in looking through
  financial institutions (FI) and re-allocating to
  households the final assets that FI hold on
  other sectors
• deposits,
• debt securities,
• quoted shares,
• unquoted equity,
• Available, if possible, with different breakdowns
  in order to enhance the analysis
• less or equal to 1 year / over 1 year maturity,
• national currency / foreign currencies,
• resident counterparties / non-resident
  counterparties,
• securities issued by general government / issued
  by other sectors,
• etc,
• For the re-allocation process, the structure of
  investment of each FI is applied to households
  assets held with FI.
• As FI also invest part of their assets with other
  FI, this method has to be applied several times
  so as to completely eliminate intermediated
  investments from the structure applied to HHs
  intermediated assets.
• Indeed, life insurance corporations invest in
  mutual fund shares and mutual funds are allowed
  to invest into other mutual fund shares.
  Therefore, after one round of re-allocation,
  mutual fund shares remain in HHs assets. Thus,
  the remaining amount of investment in MFS should
  be replaced by the structure of their investment,
  giving again a residual amount invested in
  MFS,...
• This can be solved by matrix stepwise
  calculations please refer to the articles

18
2.1 Making transparent financial assets held with
all FI in 2006

• In this case, 4 final instruments
• deposits (including loans and currency),
• debt securities,
• quoted shares,
• unquoted equity (AF512AF513),
• Available with the following breakdowns
• national currency / foreign currencies,
• resident counterparties / non-resident
  counterparties,
• General Government financing / private sector
  financing,
• For the year 2006
• Financial intermediaries made transparent
• Credit institutions and investment corporations
  ( banks)
• Money market mutual funds
• Non-money market mutual funds
• Insurance Corporations
• The article La destination finale de lépargne
  des ménages, A. Rincon, Bulletin de la Banque de
  France, n167, November 2007 (http//www.banque-fr
  ance.fr/fr/publications/telechar/bulletin/etu167_2
  .pdf)

19
2.1 Making transparent financial assets held with
all FI in 2006

• The initial structure of HHs financial
  investment (1)

At end 2006, 87 of HHs financial assets were
placed with financial intermediaries and 13 were
directly invested in final assets
20
2.1 Making transparent financial assets held with
all FI in 2006

• Financial intermediaries structure of investment
  (1)

• Each FI has its own structure of investment

21
2.1 Making transparent financial assets held with
all FI in 2006

• The final destination of HHs savings

• A different view of HHs financial investment

The transparency making process modifies
significantly the structure of investment by
instrument of HHs financial wealth
22
2.1 Making transparent financial assets held with
all FI in 2006

• The final destination of HHs savings

• Financial intermediation contributes to
  international diversification of HHs financial
  investment

23
2.1 Making transparent financial assets held with
all FI in 2006

• The final destination of HHs savings

• Financial intermediation contributes to
  international diversification of HHs financial
  investment

24
2.1 Making transparent financial assets held with
all FI in 2006

• The final destination of HHs savings

• HHs financial investment after transparency
  making process largely finances General
  Government, French or Euro area

25
2.1 Making transparent financial assets held with
all FI in 2006

• The final destination of HHs savings

• HHs financial investment after transparency
  making process largely finances General
  Government, French or Euro area

26
2.2 Making transparent financial assets held with
all non-banking FI during 1994-2005

• In this case, 5 final instruments
• deposits (including loans and currency),
• debt securities,
• quoted shares,
• unquoted equity (AF512AF513),
• and additional real-estate related financial
  instruments.
• Available with the following breakdowns
• less or equal to 1 year / over 1 year maturity,
• national currency / foreign currencies,
• resident counterparties / non-resident
  counterparties.
• Over the period 1994-2005
• The article The final financial investment of
  French households, D. Marionnet, Irving Fisher
  Committee Bulletin, n25, Bank for International
  Settlements, March 2007 (http//www.bis.org/ifc/pu
  bl/ifcb25o.pdf)

27
2.2 Making transparent financial assets held with
all non-banking FI during 1994-2005
Initial structure of HHs' financial investments

• Deposits, loans and currency predominant but
  declining (from 45 to 35)
• Life insurance the most rapid and constant
  growth (from 19.9 to 34.4)
• Unquoted equity significant increase (from 12.3
  to 19.6)
• Mutual fund shares significant decrease in MMFs
  shares (5.7 to 1), slighter decrease of other
  MFs shares (from 11.1 to 8.7)
• Quoted shares some fluctuations around 4
• Debt securities sharp decrease in direct holding
  (from 6.4 to 1.5)

28
2.2 Making transparent financial assets held with
all non-banking FI during 1994-2005
Compared to direct holding, debt securities is
the final instrument which increases the most,
followed by quoted shares while unquoted equity
and deposits present a limited increase.
HHs' financial investments in final instruments

• Deposits, loans and currency still predominant
  despite a decline (from 49 to 38)
• Debt securities fluctuations around 30
• Unquoted equity increasing though in spite of
  some fluctuations due to market valuation
• Quoted shares fluctuations due to stock market
  prices evolution
• Additional real-estate related assets residual

29
2.2 Making transparent financial assets held with
all non-banking FI during 1994-2005
Direct and intermediated HHs holdings of
(in of all final holdings of the instrument)
(in of all final holdings of the instrument)

• At Dec05, 78 of HHs final investment in debt
  securities are made via life insurance, 17 via
  mutual funds and 5 directly.

• Quoted shares, as a final instrument, are held
  in a fairly even manner over the period.
• At Dec05, 42 are held via ICs, 30 directly and
  28 via IFs.

• Unquoted equities are in average held directly at
  more than 85 over the decade

30
2.2 Making transparent financial assets held with
all non-banking FI during 1994-2005

• risky assets assets highly sensitive to price
  fluctuations
• French HHs tend to increasingly hold risky
  assets clear upward trend (in spite of
  sensitivity to stock market fluctuations).

• Risky assets predominant (77) when housing
  assets are included.
• The trend towards more risk exposure has been
  reinforced by the rapid rise in housing prices
  since 2000.

31
2.2 Making transparent financial assets held with
all non-banking FI during 1994-2005

• Liquidity ability to sell relatively rapidly
  the underlying instruments to obtain cash.
• HHs financial asset is less liquid than that of
  IFs and life ICs.
• Indeed, life ICs invest in marketable securities
  whereas HHs assets in life insurance may not be
  considered as liquid products (tax-exemption on
  realised capital-gains occurs after 8 years).
• A different picture would appear if MFIs were
  made transparent as they hold an important share
  of non-liquid assets (loans mainly) on their
  asset side.

32
2.2 Making transparent financial assets held with
all non-banking FI during 1994-2005

• French financial accounts do not distinguish
  euro-denominated assets from foreign currency
  denominated assets for the instruments other
  mutual fund shares and life insurance
  contracts. gt currency risk exposure can only be
  calculated after transparency.
• The share of assets denominated in foreign
  currencies after transparency lies a little bit
  above 5 before the euro changeover and around
  4.5 after. gt the euro changeover permitted
  geographical diversification while reducing the
  need to bear currency risk.
• French HHs bear a low and relatively stable
  foreign currency risk.

33
2.2 Making transparent financial assets held with
all non-banking FI during 1994-2005

• Before transparency, the share of HHs wealth
  invested in assets involving NR counterparts is
  quite low reflecting the fact that HHs primarily
  transact with resident intermediaries and do not
  easily invest directly with RoW counterparts.
• After transparency, geographical diversification
  comes from financial intermediaries and is
  increasing. The share of HHs wealth involving NR
  issuers or counterparts started to rise in 1998,
  climbing from 11.5 up to 21.9 in Dec. 2005.
• This increase has been made possible by the euro,
  which allows geographical diversification without
  currency risk (ICs are constrained by regulatory
  rules in their capacity to incur such a risk).

34
3. Enlarging households wealth for pensions (1/5)
Households pension wealth in social security and
ICPF in the supplementary tables of the revised
SNA a test exercise for France
35
3. Enlarging households wealth for pensions (2/5)
Insurance corporations (mainly) and PERCO
(marginally) are in charge of private pension
plans in France
36
3. Enlarging households wealth for pensions (3/5)
Compilation of actuarial value of pension funds
in the US
37
3. Enlarging households wealth for pensions (4/5)
A renewed view on international comparison of
households assets (in of corrected disposable
income social security by Prost) United states
France
38
3. Enlarging households wealth for pensions (5/5)
The risk on social security may be measured by
the discounted future financing gaps
Assets and value at risk (in grey) for HHs in
France (as a of gross disposable income)
The article Implicit social security and
pension wealth in households assets in the US
and France, D. Durant and M. Reinsdorf, paper
presented at the International Association for
Research in Income and Wealth, August 2008
(http//www.iariw.org/papers/2008/durant.pdf)