The Economic Psychology of Saving and Debt

1
The Economic Psychology of Saving and Debt

• Paul Webley
• Talk to Economic Psychology seminar group, Autumn
  2002

2
Outline of talk

• 1. Some background (Keynes, Katona, Thaler etc)
• 2. Two routes into problem debt (a panel study in
  Holland)
• 3. What lies behind saving motives? (a
  questionnaire study in the UK)
• 4. Saving and individual differences (a
  questionnaire study in the UK)
• 5. Saving, individual differences and reference
  groups (a questionnaire study in the UK and
  Italy)

3
Some background from Economics

• Saving is an area where Economists have made an
  explicit appeal to psychology to explain
  behaviour debt and borrowing have been of less
  interest - they have just seen as the opposite of
  saving.
• Keynes was not only a great economist, he was
  also a reasonable psychologist and his list of 8
  motives for saving has stood the test of time
  (only one has been added)

4

• ? To build up a reserve against unforeseen
  contingencies
• (the precautionary motive)
• ? To provide for the anticipated future
  relationship between income and needs (the
  life-cycle motive)
• ? To enjoy interest (the inter-temporal
  substitution motive) K said this was
  unimportant
• ? To enjoy a gradually improving expenditure (the
  improvement motive)
• ? To enjoy a sense of independence and power to
  do things (the independence motive)
• ? To secure a masse de manoeuvre to carry out
  speculative or business projects (the enterprise
  motive)
• ? To bequeath a fortune (the bequest motive)
• ? To satisfy pure miserliness (the avarice
  motive)

5

• Most economic theories of saving concentrate on
  motive 2 (the most obvious reason for saving
  today is to spend tomorrow)
• Modern theories e.g. Carrolls buffer-stock
  model also focus on motive 1 (the need to have a
  reserve for emergencies)
• Either way, just about all economic models of
  saving assume optimisation.or utility
  maximisation over the life-span. The life-cycle
  hypothesis says that saving at any stage of a
  persons life-cycle can be predicted from his
  current income and wealth, expectation of future
  income and life expectancy, by finding the stream
  of consumption that will maximise utility. So
  youd expect something like this ..

6
Schematic diagram of hump saving
7
Saving motives across the life-span

• Some psychological data that are relevant to
  these economic ideas come from the Dutch
  socio-economic panel
• This shows that being prepared for emergencies
  (having a buffer) is indeed the most important
  motive for saving at nearly all ages

8

• Note The relative unimportance of saving for
  old age probably reflects the quality of pension
  provision in Holland

9
What Economic Psychologists say

• By contrast economic psychologists have suggested
• (i) Saving is not a unitary phenomenom (Katona
  makes distinctions between discretionary and
  contractual saving and between voluntary and
  involuntary saving)
• (ii) most people have a self-control problem -
  saving is actually difficult to do
• (iii) saving involves forming expectations about
  the future (especially future income) - which is
  not straightforward
• (iv) the assumption of fungibility (that all
  wealth is equivalent) is wrong. Thaler has
  proposed that people have different mental
  accounts, and based on this, a behavioural
  life-cycle model of saving.

10
Katonas contribution

• Pointed out that an individuals definition of
  saving differed from that of an economist - in
  particular drew distinction between discretionary
  and contractual saving
• Discretionary an active decision to save during
  the current accounting period Contractual a
  decision made previously (e.g. to enter a pension
  plan that involves regular payments)
• Another useful distinction is between involuntary
  and voluntary saving - involuntary saving comes
  about when there is no active decision to save
• Katona noted that, year by year, people plan to
  save more than they actually do

11
Saving involves self-control - which is difficult!

• Basic fact animals and children (and some
  adults), when given a choice between a reward
  that is available immediately and one that is
  available with some delay tend to choose the
  small immediate reward
• This preference is not, in itself, irrational - a
  number of factors make current consumption
  objectively more attractive than saving (death,
  risk of inflation, risk of default etc).
• But the extent of the preference defies rational
  analysis - normal adults say that they would
  prefer 5 now to 10 in two months - which could
  only be rational if the rate of inflation were
  several thousand percent
• So we combat this by developing techniques for
  self-control (e.g. pre-commitment,
  money-management techniques)

12
Saving and expectations about the future

• People have a tendency to be over-optimistic in a
  wide variety of domains (e.g. students
  over-estimate their post-graduation spending
  power). If this is true for expectations about
  future income, saving will be too low and
  borrowing too high
• BUT the very few studies that have been carried
  out (on Dutch households) suggest that people
  under-estimate their future incomes
• SO it may be better to think of people as using
  optimism (or pessimism) as a strategy, rather
  than being optimistic per se. Optimistic
  expectations about when one will finish a project
  may act as a motivator - pessimistic expectations
  about future income may lead to good money
  management

13
Thaler - the concept of mental accounts

• Thaler proposes that people have a number of
  mental accounts that operate independently of
  each other
• Money in different mental accounts is spent
  differently propensity to spend is highest in
  the current income MA and lowest in the asset
  MA
• This means that people may borrow to buy a car
  whilst they have savings - the bank will ensure
  that they repay the loan and this keeps their
  savings intact
• The idea of mental accounts, coupled with the
  notion that the individual is an organisation
  containing a far-sighted planner and a myopic
  doer, explains quite a lot of the financial
  behaviour seen.

14
Some recent studies

• Debt - a panel study in the Netherlands (Webley
  and Nyhus, 2001)
• Why do people save? - a study in the UK (Canova,
  Manganelli and Webley, 2002)
• Individual differences and saving (Daniels and
  Webley, 2000)
• Survey study of saving in UK and Italy (Webley
  and Burlando 2002)

15
Debt study

• Steady stream of recent research on debt but it
  is very static (only gives a snap-shot) and based
  on mail surveys with very low response rates
• Need to be clear about distinction between debt,
  default and borrowing.
• Borrowing is planned and intended (note that
  mortgages are not seen as debts but as savings by
  most people)
• Debt is an obligation that a person is unable or
  unwilling to discharge
• Default - simply not paying money owed
• Here focus is on debt

16

• Assume that there are two routes to debt
  (a) the personality route where
  relevant enduring dispositions (time orientation,
  lack of self-control), which have been fostered
  by particular parenting styles, lead people into
  debt (b) the rational route - if people
  experience what they see as a temporary drop in
  income, or a temporary increase in income, they
  may decide to run up (temporary) debts
• Used data from three waves of the CentER panel
  (from1994, 1995, 1996). This is a representative
  sample of the Dutch public. Individuals complete
  large numbers of questionnaires over computers so
  there is a vast amount of information available
  on them.
• Measure of debt was based on whether people were
  in arrears, had bank debt, reported being in
  debt, had numerous credit arrangements

17
Results

• Descriptive findings same as previous studies -
  debtors were younger, had lower income, rented
  accommodation, had more children, had less
  unfavourable attitudes towards debt
• Psychological variables add to our ability to
  predict debt. There are 7 predictors Income,
  Age, No of children, Presence of partner,
  attitude to debt, obesity, use of money
  management techniques, self-control
• Many differences between never debtors and mild
  debtors
• Debtors had more variation in income, higher
  5-year income expectations, more income
  uncertainty, low life expectancy (the rational
  route) and were less conscientious and had less
  self-control (the personality route)

18
How far is debt a short term problem?

• Of the stayers (those in the panel for all 3
  years), 66 were in debt in 1994. One-third stayed
  in debt for the next two years, one-third have
  got out and stayed out of debt
• Chronic debtors have lower incomes, less
  self-control, shorter time horizons
• For 5 psychological variables looked at, the
  correlations between debt status in 1994 and
  psychological variables in 1996 were higher than
  psychological variables in 1994 and debt status
  in 1996. So differences in psychological
  variables may be a consequence rather than a
  cause of debt - e.g. obesity may be a result of
  comfort eating

19
Debt status in 1994 and 1995
20
Why do people save?

• A total of 141 British adults answered a
  mail-questionnaire. 97 participants (69) stated
  that they intended to save during the next 12
  months.
• Those intending to save had to provide 4 reasons
  why they wanted to save, then explain why these
  are important, and then asked to give further
  justifications.
• There were no significant differences between
  savers and non-savers concerning sex, age,
  marital status, education, occupation, type of
  accommodation, dependant children, and source of
  income.
• Savers had a higher overall income than
  non-savers, were more positive about the economic
  situation of the UK and are more optimistic.

21
(No Transcript)
22
Saving Goals

• The placing of goals on the vertical axis follows
  the relative ordering implied by the abstractness
  ratio. Self-esteem and Self-gratification are
  the highest-order goals Holidays / Hobbies and
  Purchases are the lowest-order goals. Arrows go
  from goals that function as sources of motivation
  to goals that serve an intermediary or end-state
  objective.
• 3 general orientations can be discerned. One of
  these deals with ways of avoiding debt and of
  achieving a certain security in life. The second
  is reflected in the desire for self-gratification,
  which can be reached by means of holidays etc.
  The third focuses on old age. For these
  respondents, saving for retirement is important
  to guarantee gratification.. The 3 super-ordinate
  goals are also linked security leading to
  autonomy, self-esteem and self-gratification are
  reciprocally connected.

23
Individual differences and saving

• Purpose of this study is
• 1. to explore the relationship between
  individual differences and saving
• 2. to look at the relative contribution of
  economic and psychological factors (do
  psychological factors improve our ability to
  predict saving?)
• 3. to look at the whether household saving
  behaviour is best explained using psychological
  data from both household members or just from the
  main decision maker

24

• 1. Questionnaires posted to 530 household (1060
  individuals) in Exeter and Plymouth. Household
  selected from electoral register which comprised
  two individuals with same name and opposite
  gender note biases the sample towards
  conventional households . Each member of couple
  asked to complete questionnaire
• 2. Completed questionnaires returned by 110
  households (195 individuals)
• 3. Questionnaire covered
• (i) demographic and economic variables (income,
  number of children, occupation, sex, age,
  housing)
• (ii) psychological variables (impulsiveness,
  Consideration of Future Consequences scale,
  self-control, time preference, economic
  socialisation)
• (iii) measures of saving (total household
  savings, regular saving)

25
Results

• What IS the relationship between individual
  differences and saving?
• 1. The exact pattern of the relationship depends
  on whether one is looking at total or regular
  savings, and using data from just decision makers
  or both household members
• 2. time preferences, self-control etc all
  associated with saving in the expected direction
  (e.g. more impatient, less saving) except that
  higher impulsiveness was associated with more
  saving
• 3. Economic socialisation variables were not
  important - but only measured with two very
  simple items.

26
DO psychological factors improve our ability to
predict saving? YES

• 1. Hierarchical regression used with variables
  entered in order socio-economic, individual.
  diffs, economic socialisation
• 2. Total savings best predicted by age and time
  preference
• 3. Regular saving best predicted by financial
  situation, self-control, CFC scale, impatience
• 4. Note that this fits with Lunt and Livingstone
  findings that for total savings, socio-economic
  variables were more important and that for
  recurrent savings, psychological variables were
  more important

27
IS household saving behaviour best explained
using data from both household members? YES

• 1. There was a positive correlation between the
  scores of spouses on most psychological variables
  (impulsiveness .45, CFC .24, delay of
  gratification, .36) though not all. Correlations
  were higher for those couples who had been
  married longer
• 2. Using psychological information from both
  couples improves our ability to predict both
  total savings and regular saving. Using only
  psychological variables, the R2 are
• Total savings Amount saved regularly
• Data from both spouses .32 .43
• Data from decision maker .23 .29

28
Explaining cross-national differences in saving

• 1. There are very large differences in the
  savings rates across countries. Japan has a very
  high saving rate, Italy quite a high rate and the
  UK and USA relatively low rates. Why?
• 2. Economists attempts to explain these
  differences using economic indicators (growth
  rates, social security systems, tax incentives
  etc) have been largely unsuccessful
• 3. May they stem from cultural differences
  between countries? Carroll et al show that there
  are differences in immigrants saving behaviour by
  country of origin but that these do not match up
  with the differences in national savings rates.

29
Cross-national questionnaire study of savings

• Study tried to shed some light on the causes of
  the international differences in savings through
  a consideration of motivations for saving, income
  expectations, reference group membership and
  trust in government.
• Questionnaires covered all of these issues plus
  financial planning. Five different measures of
  saving were used.
• Questionnaires distributed in Bristol, Exeter,
  Turin, Asti, n347.

30
Cross-national differences in independent
variables.

• Compared to those in the UK those in the Italian
  sample were (i) more inclined to want to save as
  much as possible (ii) had a longer time horizon
  (mode is 5-10 years for IT, next couple of years
  for UK), and (iii) found it easier to control
  their spending. Italians save more than UK sample
  (on all saving measures). The importance of
  different reasons for saving was similar though
  it is clear that medical care and childrens
  education are more important saving motives for
  the Italians and that saving for future
  consumption (e.g. holidays) is more important for
  the English.
• The preferred size of a buffer is much higher in
  the Italian sample (over 34,000, as opposed to a
  mean of 8,962 in the UK sample).

31
Results

• The three psychological variables (time horizon,
  planning, control) are all related to saving. So
  too was general trust in government (those who
  trust more save more), income, age and perceived
  financial status compared to reference group.
• To examine whether these relationships were
  independent, the data were analysed using
  multiple regression. A hierarchical form of
  analysis was used, which looked at the effects of
  a series of groups of variables in turn. The
  variables were entered into the analysis in the
  order economic, demographic, social
  psychological, psychological. This analysis shows
  that whilst economic factors matter (a higher
  income leads to more saving), so too do social
  psychological and psychological factors. The
  impact of reference groups is particularly
  notable
• The best set of variables for predicting saving
  behaviour included income, housing, social
  comparison, self-control, planning and time
  horizon.

32
THE END

• Unless you want some policy implications ...

33
Some policy implications

• Saving is hard and getting into debt is easy.
  Policy needs to be based on a firm understanding
  of why people save (and why they get into debt)
• Education matters. What evidence there is
  suggests that learning to wait begins in
  childhood. So - from a long term perspective -
  we need to foster parenting and School practices
  that encourage saving
• The main reason for saving in the UK seems to be
  - ultimately - hedonistic (e.g. to have the
  holiday of a lifetime. People in the UK have
  trust in the government and still believe (unlike
  most Italians) that the system (the state,
  their employee) will provide them with the
  pensions they are entitled to. So for Brits,
  saving for old age is well important. So one way
  - oddly- to encourage UK citizens to save would
  be to undermine their faith in the system.

34
Some policy implications (continued)

• Believing that you are financially better off
  than members of your reference group is a good
  predictor of saving. But most Brits (strangely)
  believe that they are worse off than people like
  them. This belief may be open to manipulation
• Those who have longer planning horizons save more
  (and are in debt less). Thinking more about the
  future is also something that can be encouraged
• There is probably value in disseminating good
  (simple and effective) money management
  techniques. Most people are surprisingly bad at
  managing money!