The Survey on Household Income and Wealth: a tool for policy and research

1
The Survey on HouseholdIncome and Wealtha tool
for policy and research

• Ivan Faiella
• Banca dItalia
• Servizio Statistiche Economiche e Finanziairie
• 21 Ottobre 2008

2
Bank of Italy SHIW Survey on Household Income and
Wealth
3
SHIW the questionnaire

• core sections
• Composition Of Household At End Of Survey Year
• i.e. characteristics of household members (sex,
  age, education,)
• Employment and Income Sources
• i.e. job status of hh members aged 15 or more.
  Income sources of all hh members ,
• Payment Instruments, Financial Assets and
  Liabilities
• i.e. use of current accounts, debit/credit/prepaid
  cards, financial assets and liabilities,
• Principal Residence, Unincorporated Business and
  Other Tangible Assets
• i.e. principal residence characteristics and
  value, mortgages,
• Non Durable And Durable Consumer Goods
• i.e. expenditure, alimonies, perceived
  well-being,
• Forms Of Insurance
• i.e. life and health insurances, private pensions
  ,
• Information To Be Provided By The Interviewer
• i.e. assessment of the interview

monographic sections Households Relationships
With Banks Intergenerational Transfers Housework
And Care For Family Members Household Consumer
Behaviour Job condition judgments / expectations
and many more topics
4
Distribution of household income and
wealth (thousands of euro, 2004)
5
Distribution of household wealth (percentages
2004 euro)
6
Assets/liabilities held by households
(percentages of households)
7
Household wealth and expenditure (as fraction of
income)
8
Possession of financial assets (percentages, 2004)
9
Reasons for not buying goods and services over
the Internet (percentages, 2004)
10
Head of households holding Credit/Debit
cards (percentages, 2004)
11
House-related debts (percentages, 2004)
12
Time value of money (percentages, 2004)
() Imagine you won the equivalent of your
households net annual income. The sum will be
paid to you in a years time. However, if you
give up part of the sum you can have the rest
immediately. To get the money right away would
you give up?
13
The panel dimension (percentages, 1991 and 2000
waves)
Source Faiella and Neri (2004)
14
SHIW sampling and nonsampling errors

• A well-known problem of surveys it is their
  tendency to yield estimates of income and wealth
  that are biased downward. This can be caused by
• high concentration of wealth ? low probability
  of including the wealthiest households in the
  sample
• interviewees tend to under-report, consciously
  or not, their wealth (and sometimes income)
• selection bias because affluent households are
  less willing to participate in sample surveys

15
SHIW adjusted data
According to Financial Accounts, in 2002 the
Italian households held 1,453 billions of euro
(456 according to SHIW unadjusted data, 1,231
after the adjustment)
16
SHIW nonresponse
Evidence of selection bias (nonresponse) in the
SHIW (DAlessio and Faiella , 2002)
17
SHIW other methodological issues

• In In the years a number of methological issues
  have being issued concerning SHIW design and
  results
• panel attrition and estimators bias
• ...regional (small area) estimation of income and
  wealth
• ...nonresponse behaviour in the SHIW
• ...the process of weighting in the SHIW
• ...accounting for sampling design in the SHIW
• ...measurement error in the SHIW

18
SHIW data use

• SHIW data are used
• to appraise the financial behaviour of households
  and their attitudes towards payment instruments
• to support fiscal policy and evaluate income and
  wealth distribution
• to explore the relations among socio-economic
  features of households and their economic
  preferences
• for research purposes, inside and outside the
  Bank of Italy (Italian an foreign universities,
  private and public research institutes,
  think-tanks,)

19
SHIW data using (papers written using SHIW data,
by topic)
20
SHIW data using an example

• The Importance of Household Finance (HF)

• HF is about the use financial markets by
  households to attain their objectives
• Is analogous to Corporate Finance which deals
  with firms reliance on financial markets to
  attain their objectives
• Finance investment
• Constrain managers discretion by appropriate
  choice of capital structure (resolve agency
  problems)
• Manage liquidity etc.
• Growing importance and interest in household
  finances
• Households are much more deeply involved in
  financial decisions
• Rely much more on financial markets
• Are more exposed to financial market developments

NB This part relies on Guiso (2007)
21
What is HF about?

• Household make a number of decisions
• How much to consume
• How much to work
• How much to invest in education
• Whether and when to buy a house
• How much to accumulate for retirement and in
  what form
• and many others
• For achieving their objectives they rely on
  financial markets

22
What financial markets are used for

• Financial markets can be used to
• Decide how to save ones wealth that is to
  choose how to transfer resources from the
  present to the è portfolio choice
• Decide whether how much and what type of debt to
  raise that is the media to get resources now
  from the future è debt choice
• Manage ones liquidity and transactions media è
  money demand and demand for transactions media
• Provide insurance against some risk è insurance
  demand

23
Does this matter for a Central Bank?

• Yes for at least two reasons
• Because how central bank policies spread out
  depend on how households decisions are affected
  by CB policy instruments, which typically alter
  financial markets equilibrium
• Money demand
• Portfolio reallocations
• Wage claims and inflation expectations
• Consumption decisions (non durable and durable)
  
• Because households ability to achieve their
  objectives depends on how well financial markets
  work, in turns affected by regulatory policies
  enacted/enforced by CB

24
Households do not behave as theory predicts

• But households do not behave as theory predicts
• Few participate in financial markets
• Those who do, hold very different portfolios
• Portfolios are poorly diversified
• Risk tolerance important but many other variables
  matter for behavior
• Differences in wealth are key
• The big gap is between the uniformity predicted
  by theory and the heterogeneity of behavior

25
SHIW conclusions
Household wealth surveys (as all hh surveys)

• a very powerful tool to gather every kind of
  information, but
• you need to ask the right question
  (questionnaire design)
• it is difficult to get the right answer
  (data quality is a key issue)
• the setup is complex (questionnaire design,
  field activities, weighting, data processing,
  point estimates and their variability)
• but you can refer to the established best
  practices at international level e.g., Groves et
  al. (2004), Survey Methodology)

26
SHIW conclusions
But hh wealth is heavily skewed!
27
SHIW conclusions
Household wealth surveys what next?
28
SHIW conclusions
Household wealth surveys what next?