Nudges and Networks: How to use behavioural economics to improve the life cycle savings-consumption balance

1
Nudges and NetworksHow to use behavioural
economics to improve the life cycle
savings-consumption balance

• Professor David Blake
• Director
• Pensions Institute
• d.blake_at_city.ac.uk
• May 2012

2
Agenda

• Most people are not rational life cycle financial
  planners
• Identifying behavioural barriers
• How behavioural economics can help overcome
  barriers
• Speedometer Plans
• How networks can help
• How to implement A life-cycle fund corporate
  platform

3
Most people are not rational life cycle financial
planners
4
Most people are not rational life cycle financial
planners

• This would require people to accurately forecast
• Total career income
• Total available retirement resources
• Asset returns
• Interest rates
• Tax rates
• Inflation
• Longevity
• Medical and health costs
• It would also require people to have the
  commitment to start and maintain a very long-term
  savings and investment programme

5
Identifying behavioural barriers
6
What needs to be recognised

• In reality, individual decisions subject to
• Bounded rationality
• Certain types of problems are too complex for
  individuals to solve on their own
• Many people have a poor sense of the time
  dimension of their lives
• Bounded self-control
• Individuals lack willpower to execute plans
• In view of these limits on optimising behaviour,
  we need to change our understanding of individual
  economic decision making
• Especially long-term savings decisions
• Such as those involved in accumulating and
  decumulating assets in a pension plan

7
Pre-retirement behavioural barriers
8
Starting to save

• Procrastination and inertia are bad for saving
• Employees fail to join pension plans where they
  are required to opt in
• Retirement saving means reducing consumption now
  in order to have a comfortable income in future
• Requires self-control - not always easy
• Similar to losing weight or giving up smoking

9
How much to save

• It is difficult to know how much to save for
  retirement
• Members may be anchored by irrelevant information
• Default saving rate is 5 - that must be the
  right rate?
• Cognitive polyphasia
• People can think about the same issue in
  contradictory terms in different situations
• I know I should be saving 15 of my income if I
  want a good pension
• I think I will be able to live on much less when
  I retire, so I do not need to save as much as I
  thought, which means I can spend more today

10
What to invest in

• Most DC members dont want to choose funds
• 90 take-up of default funds
• There are many behavioural biases relevant to
  investment
• Regret
• Loss aversion
• Mental accounting
• How people keep track of financial activities
• Framing
• You are aware equities are risky
• Leads to reckless conservatism
• You are aware that equities tend to generate
  higher returns in the long run, despite
  short-term volatility

11
What to invest in

• There are many behavioural biases relevant to
  investment
• Choice overload/anxiety (problem of complexity)
• Too many investment funds means no decision at
  all
• When faced with difficult choices, individuals
  often employ simplifying heuristics (simple rules
  of thumb)
• E.g., choose default option on grounds that
  someone else must have thought that it was good
  idea
• Herding follow the herd
• Weak investment preferences (problem of
  complexity)
• Individuals can easily be led
• Evidence suggests menu design influences fund
  choice

12
When to retire

• Time inconsistency
• When you are young, you believe that you will be
  able and willing to work longer if necessary to
  compensate for inadequate pension savings
• Even if someone tells you that you will probably
  not feel like that when you are older
• When you are old, you regret not saving enough,
  because you do in fact want to retire earlier
• Another example of the poor understanding of the
  time dimension of peoples lives

13
At-retirement behavioural barriers
14
What retirement income strategy

• Effective retirement saving needs an optimal
  decumulation strategy as well as optimal
  accumulation
• Need to deal with
• Human spenders
• Spend too quickly in retirement
• Human hoarders or squirrels
• Spend too slowly in retirement
• Wish to guarantee inheritance for their children
• Properly designed retirement income strategy can
  help both

15
People have a poor understanding of longevity risk
Expected distribution of deaths male 65
Expected distribution of deaths male 85
Most likely ageat death 86
Most likely ageat death 90
Life expectancy 86.6
10
5
Life expectancy 91.6
9
8
4
7
6
3
deaths at each age
deaths at each age

• 1 in 3 will reach 93 and 5 will reach 100

5
25
4
25
2
3
Idiosyncratic risk
1
2
Random Variation Risk
Random Variation Risk
Idiosyncratic risk
1
0
0
65
70
75
80
85
90
95
100
105
110
85
90
95
100
105
Age
Age
Source 100 PNMA00 medium cohort 2007
15
16
What retirement income strategy

• Many scheme members dislike the idea of buying an
  annuity
• Annuities are perceived as poor value
• This may be because members underestimate how
  long they (and their spouse) are likely to live
• There are many behavioural biases relevant to
  decumulation
• Illusion of control
• People like to feel in control of their capital
  but annuitisation leads to a loss of control
• Framing
• Expressed using a consumption frame, annuities
  are desirable
• Expressed using an investment frame, annuities
  are risky

17
What retirement income strategy

• There are many behavioural biases relevant to
  decumulation
• Regret/loss aversion rather than risk aversion
• Annuities are a gamble
• Probability of dying very soon after purchasing
  annuity is very low, but this probability is
  likely to be overestimated
• So loss perceived to be high
• Dying AND losing all your capital too!
• Conversely the significant probability of
  out-living ones resources if one doesnt
  annuitise is underestimated
• So gain perceived to be low
• Hence gain to annuitising will give small
  utility benefit, while loss of dying early may
  have large utility loss

18
How behavioural economics can helpovercome
barriers
19
Behavioural economics

• Combines economics, finance, psychology and
  sociology
• Recognises
• Individuals do try to maximise personal welfare
• But limits to the extent they can do this
• Individuals are Humans not Econs and need nudging
  towards optimal solutions
• Recognises
• Importance of social norm groups and social
  networks in helping individuals improve outcomes
• People like us
• Comes out of the US, so needs to be adapted to
  other countries

20
Overcoming pre-retirement barriers
21
Starting to save

• Behavioural traits have been exploited to design
  pension schemes that increase long-term pension
  savings.
• Classic example is Save More Tomorrow (SMT or
  SmarT) plan
• Thaler and Benartzi (2004)
• Scheme member agrees to start or increase savings
  on regular basis
• Not now but on future significant date
• E.g., date of next pay rise

22
Starting to save

• SMT plans deal with a number of behavioural
  traits
• Accept individuals have self-control problems
• And benefit from using pre-commitment devices
• Auto-enrolment with payroll deduction
• Auto-escalation
• Withdrawal restrictions
• creating psychological and financial barrier to
  accessing funds
• Utilise inertia
• Since, once signed up, workers typically do not
  cancel payroll deduction facility
• Uses herding behaviour constructively
• A worker will join if other workers are joining

23
How much to save

• Importance of an appropriate default contribution
  rate
• Contribution matching by employers provides a
  powerful incentive
• Once enrolled, members tend not to alter
  contribution rate
• Unless automatic annual increases are in place
• Again exploits inertia positively

24
Impact of pre-commitment devices and inertia
Savings rates in SMT plans
25
What to invest in

• To deal with choice overload/anxiety (problem of
  complexity)
• Have only a small number of investment funds to
  cover the range of risk tolerances
• Individuals want to know what a fund does, not
  what its asset mix is
• To deal with simplifying heuristics
• Have a well-designed and low cost default fund
• To deal with inertia
• Use lifestyle investment strategies
• De-risking near retirement is automatic

26
Overcoming at-retirement barriers
27
What retirement income strategy

• Overcoming the illusion of control
• All-or-nothing annuitisation likely to be
  suboptimal as well as undesirable
• Gradually purchasing annuities over time might be
  better
• Deals with
• Interest rate risk hedges interest rate cycle
• Possibility that investment returns might be
  higher in future
• Possibility that mortality rates might be higher
  in future
• Possibly long period of retirement and not
  wanting to be locked into a low-yielding
  bond-like investment

28
What retirement income strategy

• Overcoming regret/loss aversion
• Any pooling of mortality needs to be perceived to
  be fair by the public
• Currently, this is not true!
• At younger ages annuity mortality cross-subsidy
  or survivor credit gives poor value to those
  dying early
• Solutions
• Money-back or capital-protected annuity
• Impaired life annuity

29
It is not a question of IF but WHEN pensioners
should annuitize
Limited value from annuitization Death benefit
seen as more valuable.
Annuitization essential to provide income for life
Survivor credit
Level of survivor credits
Age
Equities
Bonds / Annuities

Investment split - Equities Bonds/Annuities
Source Own analysis 100 PNMA00 2010 plus
improvements in-line with CMI_2009_M 1.00
Survivor credit qx / (1 - qx)
29
30
Death benefits under a money-back annuity
On death any excess of the original purchase
price over the gross annuity payments already
received is returned to the annuitants estate

ACCUMULATED INCOME
DEATH BENEFIT
Purchase price
Accumulated
gross payments
65
67
69
71
73
75
77
79
81
83
85
Age at death
Source Own calculations 100 PNMA00 2010 plus
improvements in-line with CMI_2009_M 1.00
31
What retirement income strategy

• To deal with framing
• Pose problem in a way that generates the optimal
  outcome for most people
• Talk about the income stream generated by the
  annuity rather than the loss of the lump sum
• Explain the annuity in a consumption frame
  (which makes an annuity look safe) rather than an
  investment frame (which makes an annuity look
  risky)
• Emphasize risk of living in poverty in old age,
  rather than giving up the lump sum
• Studies show people with annuities are happier
• they can spend their annuity payments knowing
  they have full longevity risk protection
• show series of photos of decreasing bundles of
  goods that can be purchased due to inflation

32
Speedometer Plans
33
Spend More Today Safely
Using Behavioural Economics to Improve
Retirement Expenditure Decisions

• David Blake Tom Boardman
• February 2012
• (pensions-institute.org/workingpapers/wp1014.pdf)

34
SPEEDOMETER retirement expenditure plan

• Spending Optimally Throughout Retirement
• First, make a plan
• Second, secure 'essential' income
• Third, have insurance and a 'rainy day' fund to
  cover contingencies
• Fourth, secure 'adequate' income
• Fifth, achieve a 'desired' standard of living and
  make bequests
• A universal plan for all retirees

34
35
First, make a plan

• Either ... by using an on-line or telephone-based
  service providing generic financial advice
• Or ... if wealth permits, involving a financial
  adviser whose role is to assist with making and
  implementing the plan and conducting annual
  reviews

35
36
Second, secure 'essential' income

• Plan manages all assets and income sources
  holistically to secure essential income
• Defined as the minimum, core inflation-protected
  income sufficient to meet the retirees
  essential needs for the remainder of their (and
  their spouses) life.

36
37
Third, have insurance and a 'rainy day' fund to
cover contingencies

• Use insurance solutions, when available and cost
  effective, to cover contingencies,
• Where appropriate, rely on state support
• Where possible, maintain flexibility by holding
  sufficient assets to meet uninsurable shocks
  (i.e., a rainy day fund)

37
38
Fourth, secure 'adequate' income

• Secure an adequate level of life-long income
  above the minimum if there is sufficient wealth
• Adequate income defined as that needed to
  achieve the minimum lifestyle to which the
  pensioner aspires in retirement.

38
39
Fifth, achieve a 'desired' standard of living and
make bequests

• The plan uses a simplified choice architecture
  for managing any residual wealth
• Aim of achieving a desired standard of living
  in retirement, while allowing part of the
  remaining wealth to be bequested at a time of the
  retirees choosing.

39
40
How a SPEEDOMETER plan deals with behavioural
traits

• Use of commitment devices and inertia
• Use of defaults
• The plan NOT the member deals with complexity of
  decumulation
• Use of money-back annuities
• Use of phasing
• Positive norming via effective communication
• The slogan spend more today safely to reinforce
  the idea that buying an annuity is a smart thing
  to do.

40
41
The SPEEDOMETER plan involves just four key
behavioural nudges

• First, make a plan
• Automatic phasing of annuitization
• Capital protection in the form of money-back
  annuities
• The slogan spend more today safely to reinforce
  that buying an annuity is a smart thing to do.

41
42
How networks can help
43
How networks can help

• Now beginning to be recognised that nudging is
  more effective in networks
• Employment-based networks are most effective for
• Encouraging pension savings
• Helping to pay-off debt via pay-roll deduction
  with payments used to create positive savings
  once debt paid off
• Social networks
• Family, friends and neighbours
• Internet-based networks
• Daily Dollar - Daily Budgeting Facebook App
  (facebook.com/LiveSolid)
• brings to life the notion that small lifestyle
  changes can add up to big savings.   
• publish the results on your profile

44
Age-based networks
Age range Description
Baby Boomers (1946 64) Gilt Edge Lifestyles Mid Life Affluence Modest Mid Years Advancing Status Ageing Workers
Generation X (1965 81) Successful Starts Happy Housemates Surviving Singles On The Breadline Flourishing Families
Generation Y (1982 95) Happy Housemates Surviving Singles On the Breadline
45
Networks based on personality types
Personality type in retirement Description
Empowered Reinventors (19) Can easily adapt to change welcome adventure and new challenges
Carefree Contents (19) Optimistic about coping with change but do not seek adventure or new challenges
Uncertain Searchers (22) Recognise change could be fulfilling and satisfying, but still trying to make sense of change
Worried Strugglers (40) Worried, bored or saddened after the change. Lack of planning and preparation play a role here
Source The New Retirement Mindscape (Ameriprise Financial, January 2006) Source The New Retirement Mindscape (Ameriprise Financial, January 2006)
46
How to implement A life cycle fund corporate
platform
47
First, get em young

• Savings is a habit that needs to be engendered
  from a very early age
• Four boxes for pocket money

Immediate spending (Instant gratification)
Charity Spending on other than self (Feel good)
Short-term saving for specific item (Deferred
gratification)
Saving for an unspecified purpose
(Precautionary and long-term savings)
When grown up, this becomes the rainy day fund
and the pension fund
48
Life-cycle fund

• Manages savings and loans around key life events
• Paying off student loans and future debt
  management
• Tax-efficient short/medium term savings
• ISA (individual savings account)
• Share incentive plans
• House purchase
• Marriage
• Children school fees
• Holidays
• Retirement
• Inheritance and tax planning
• Long-term care

Life/health assurance
49
Implemented using (corporate) wealth management
platforms/wraps

• Employer as facilitator
• Exploiting one of the most effective networks
• But how much choice and flexibility should be
  offered?
• Econs like lots of choice and flexibility
• Many people, especially the young, claim to like
  choice and flexibility
• Especially the flexibility to delay starting a
  long-term pensions savings programme!
• So provide lots of self-selection?
• Humans do not really like that much choice and
  flexibility
• They like well-designed defaults
• So provide segmented information and products?
• Based on effective client profiling

50
Conclusions
51
Lessons from behavioural economics

• Assume nothing (or very little)
• Design products and marketing strategies with
  abilities of less sophisticated, less experienced
  population in mind
• guiding choice, choice-editing
• Wherever possible, work with human biases not
  against
• Nudging will help if the product design is good
• Networks can help support and reinforce good
  individual behaviour

52
Well-designed pension plans recognise

• Need to help both Human Spenders and Human
  Hoarders
• People need reassurance that it pays to save
• Pension death benefits need to be as generous for
  annuities as they are for income drawdown
• Phasing into annuitisation may be more acceptable
• Annuity products with equity linking might be
  valuable for those who are sufficiently risk
  tolerant

53
Better communication and education alone will not
work

• Need good design of default option
• Education no substitute for good default
• David Laibson, Harvard University
• Pioneer Investments European Colloquia 2007
• Because vast majority of individuals will not be
  able to design their own retirement income
  programme
• Who wants to go into a car show room and be
  offered a choice of car kits to self assemble?

54
Thank you!
55
References

• Mitchell, O, and Utkus, S. (2004) Pension Design
  Structure New Lessons from Behavioral Finance,
  Oxford University Press, Oxford
• Thaler, R, and Benartzi, S (2004) Save More
  Tomorrow Using Behavioral Economics to Increase
  Employee Saving, Journal of Political Economy,
  112, S164-S187
• Thaler, R, and Sunstein, C (2008) Nudge
  Improving Decisions about Health, Wealth
  Happiness, Yale University Press, New Haven
  London