OPTION VALUES FOR RETIREMENT

1
AGIR DEMWEL Seminar 12-14 June, 2003 ETLA,
Helsinki Options for Retirement Health and Time
Use in a Cross-Section of Countries
2
OUTLINE OF PRESENTATION

• Introduction
• Allocation of time considering total working
  time
• Withdrawal from labour force
• Welfare regimes
• Data
• Methods
• Net replacement rate
• Option value
• Health status
• Results
• Time use
• Net replacement rates
• Option values
• Conclusions

3
TIME USE DATA, AGEING POPULATION, AND LABOUR
MARKET RESEARCH

• Previous work at ETLA
• Ruuskanen, Olli-Pekka Replacement Rates and
  Reservation Wages Considering the Value of
  Household Work and Lost Leisure (Etla Discussion
  Paper 588, 1997)
• Huovinen, Pasi and Hannu Piekkola The Time is
  Right? Early Retirements and Use of Time by Older
  Finns (ETLA B 189 2002)
• Applications of time-use data
• Economic and social accounting (macro-level)
• Descriptive analyses, household production,
  labour market behaviour (micro-level)

4
WITHDRAWAL FROM THE LABOUR FORCE
Table 1. Average ages of withdrawal from the
labour force
Scherer, P. (2002) Age of Withdrawal from the
Labour Force in OECD Countries. Labour Market nad
Social Policy Occasional Papers. No. 49. OECD
5
TIME-GEOGRAPHICAL AUTHORITY CONSTRAINT AND
ALLOCATION OF TIME BETWEEN WORKPLACE AND HOME
MASTERS THESIS MAY 6, 2003 Liisa Harmoinen
6
OBJECTIVES AND RESEARCH QUESTIONS

• Ageing, Health, and Retirement in Europe
  project, ETLA
• Labour market behaviour and time use of the
  ageing

Are there differences in time use btw. groups and
countries? Effect of the value of household work
on labour market behaviour?
7
THEORETICAL FRAMEWORK
Time Geography Constraints to activity
Economic Theory Allocation of time
Allocation of time between workplace and home
Welfare Regimes Social security
8
TIME GEOGRAPHY AND CONSTRAINTS TO ACTIVITY

• Capability constraint
• Physiological and physical characteristics and
  needs
• Coupling constraint
• Places, resources, other people
• Authority constraint
• Legislation, authorities, rules, norms

9
ACTIVITY TRAJECTORY daily vs. lifetime
work
home
10
AUTHORITY CONSTRAINT Welfare Regimes

• Nordic Regime
• Finland, Denmark
• Continental Regime
• Belgium, Germany, Netherlands
• British Regime
• United Kingdom
• Southern Continental
• Portugal

11
WELFARE REGIMES
12
WELFARE REGIMES Time Use Approach
13
STRUCTURALISTIC VIEW OF THE SOCIETY
14
ALLOCATION OF TIME Total Work

• Assume people consider total work time (paid
  workdomestic work) when allocating time between
  work and leisure to maximize utility
• Value domestic work with net wage for this kind
  of work
• Account for increase in domestic work in
  non-employment

Production function for household goods
15
ALLOCATION OF TIME Total Work
Utility function
Production functions
Time constraint
Goods constraint
Ms The effect of social security payments by the
employer Mm The effect of wage tax rate Mk The
effect of consumption tax rate
16
DATA

• MTUS data
• 24 countries, 50 time-use surveys
• Harmonized data
• Original Data
• (Belgium), Denmark, Finland, Germany, The
  Netherlands, UK, (Portugal)
• Finland 1999/2000 follows new Eurostat standards
• one weekend one weekday
• demographic and income variables linked

Table 2. Countries and survey years covered
17
METHODS

• Document time use in household work
• Wage rates, social security payments, tax
  treatment
• Estimating health status
• Replacement rates
• Option values

18
NET REPLACEMENT RATE
Net replacement rate net pension value of
household work net earnings value of
household work
19
OPTION VALUE
20
ESTIMATING HEALTH STATUS

• Finnish data includes self-assessed health
  variables
• unhealth notable health problems
• 5-15 of 45 to 59-year- olds
• ECHP very poor -very good five categories

21
LOGISTIC ESTIMATION OF HEALTH STATUS

• Logistic model to explain health problems in
  Finland, predict health status in other countries
• Socio-economic variables, time use
• income
• education
• family
• working hours

22
LOGISTIC ESTIMATION OF HEALTH STATUS Finland
23
LOGISTIC ESTIMATION OF HEALTH STATUS Finland
24
LOGISTIC ESTIMATION OF HEALTH STATUS Other
Countries
25
MAIN TIME USE RESULTS

• Women (10 hours) and the unhealthy ( 2-5) adjust
  total work supply by less after withdrawal from
  work
• Less work than men and the healthy before,
• more work after withdrawal from work
• Domestic work increases 10 hours in
  non-employment for both sexes
• 5.2 /h2704 /a
• Female shares
• 1/2 of total work
• 2/3 of domestic work
• Non-employed and the unhealthy share all work
  more equally

26
TOTAL WORK Work Domestic Work
Table 4. Total work by gender, employment status,
and health (hours per week)

• Low burden Denmark, High burden Portugal, UK
• Women adjust total work 10 hours less than men
  with respect to employment status
• The unhealthy
• Unhealthy men work less, health has no effect
  among the non-employed
• Unhealthy women work less, employed and
  non-employed

27
DOMESTIC WORK
Table 5. Domestic work by gender, employment
status and health (hours per week)

• Women supply 11 hours more domestic work than men
• Domestic work increases in non-employment
• Healthy 10.1 hours for men and 10.3 for women
• Unhealthy 8.1 hours for men and 2.4 hours for
  women

28
FEMALE SHARE OF DOMESTIC WORK
Table 6. Female share of unpaid work time by
employment status and health

• Women do 2/3 of domestic work
• Unhealthy and the non-employed share domestic
  work more equally

29
PAID WORK
Table 7. Weekly hours of paid work by gender and
female share of paid work

• Men supply 10 hours more paid work than women
• Continental countries equal allocation
• Perhaps household work services bought more than
  elsewhere

30
CONCLUSIONS Time Use

• Men and women contribute equally to total work in
  most countries, allocation of total work between
  paid work and domestic work based on the
  specialisation of work
• Dual burden most relevant for young workers with
  children, not for older workes
• Allocation of additional time towards domestic
  work greater form men after withdrawal from work
• Huovinen and Piekkola (2002) also find that
  allocation of time is a much more important
  determinant for retirement for men than for
  women. This is also shown in the positive effect
  of active time use on the probability to retire.

31
REPLACEMENT RATE

• Value of domestic work while at work on average
• 1/3 of net earnings for men and
• 2/3 of net earnings for women
• After retirement relative value of domestic work
  exceeds annual pension income for both genders
• Changes in domestic work greater for men
• Value of domestic work almost doubles after
  retirement

32
REPLACEMENT RATE

• Replacement rates on average close to 100 percent
  when domestic work is accounted for
• The effect of domestic work greater for women
• 40 percent for men and 45 percent for women
• The average replacement rates (age at last year
  of work 55-70 years)
• Women 101 with domestic work and 69
  without
• Men 93 and 67
• With the exclusion of domestic work slightly
  lower for men than for women replacement rate
  decreases with increasing income level

33
REPLACEMENT RATE Men
Table 8. Net replacement rates accros countries
for men
34
REPLACEMENT RATE Women
Table 9. Net replacement rates accros countries
for women
35
OPTION VALUES FOR RETIREMENT

• Domestic work has two effects on option value
• Replacement rate and pension wealth are higher
• Relative value of continuing to work and the
  accumulation of total income are lower because
  of foregone domestic work

36
OPTION VALUES Belgium

• Additional domestic work after withdrawal from
  work worth
• 1500 a year for men and
• 3200 a year for women.
• Replacement rates are
• 95 for men and
• 125 for women
• Incentives to accumulate pension wealth are low

37
OPTION VALUES Finland

• New system in 2005
• Accrual 1.5 before age 53,
• 1.9 ages 53-62 and
• 4.5 ages 63-68
• Pension wealth at 62
• 13,000 higher for men,
• 14,800 higher for women
• in the new system, as replacement rates are 5
  percentage points higher
• Increase in domestic work
• 10 hours for men and
• 13 hours for women
• Men start at low level of domestic work, marginal
  value one of the highest with hourly net wage
  7.8

• Difference btw net earned income and pension
  income 5,000 after retirement, compensated by
  an equal increase in the value of domestic work.
• Individuals are rather indifferent whether to
  retire at age 62 or at age 68, women especially

38
Germany

• Additional domestic work after withdrawal from
  work worth
• 900
• Replacement rate 10 percentage points higher when
  domestic work included,
• 77-79 at age 62
• the German system can encourage postponement of
  retirement until 65 years of age or beyond for
  wome
• Encourage to work until 63 if individuals cannot
  draw any pensions earlier deduction of pension
  at 3.6 per year if retirement occurs before 65
  and an addition to pension at 6 per year if
  retirement is postponed

39
Netherlands

• Value for additional domestic work after
  retirement is relatively low 4000 for men and
  2900 for women, but from low initial level
• Public pension system with flat pension implies
  low replacement rate around 87 percent
• (46 percent for men and
• 40 percent for women when domestic work is
  excluded)
• Incentives to retire are relatively low for
  low-income earners in the public system
• Optimal retirement age close to the pensionable
  age

40
United Kingdom

• Additional domestic work after withdrawal from
  work worth 4400 a year for men and
• 3100 a year for women
• Low replacement rates,
• 71 for men and
• 81 for women
• Strong incentives to postpone retirement

41
COUNTRY TABLES Belgium
42
COUNTRY TABLES Denmark
43
COUNTRY TABLES Finland
44
COUNTRY TABLES Germany
45
COUNTRY TABLES Netherlands
46
COUNTRY TABLES UK
47
REGIONAL DIFFERENCES

• Denmark most similar to Finland
• Finland similar to Netherlands and UK, partly to
  Portugal.
• UK similar to Portugal
• Central Europe with relatively short paid work
  hours and average domestic work hours Germany
• Small countries and UK with long paid work hours
  and long domestic work hours for men and women
• Exceptions Denmark with relatively little
  domestic work, but greatest similarity to Nordic
  regime.

Table 10. Measures of dissimilarity (D) between
pairs of countries in weekly hours of paid and
household work by employed men and women.

48
REGIONAL DIFFERENCES IN RETIREMENT INCENTIVES

• Three groups
• High replacement rate and flat option value curve
  after the pensionable age Central Europe
  (Germany), Finland and Portugal
• High replacement rate and downward sloping option
  value curve since pensionable age Belgium, the
  Netherlands.
• Low replacement rate and rising option value
  curve after the pensionable age Denmark, UK

49
REGIONAL DIFFERENCES IN RETIREMENT INCENTIVES

• Expected lifetime at age 65 12 years for men and
  17 years for women
• (when considered at age 55)
• Pension accruals since pensionable age must be
  relatively high for the option value to be flat
• With high pension wealth (replacement rate)
  additional income from work is of low importance
• Finland accrual at age 63-67 at 4.5 percent
  annually
• With low replacement rates the option value after
  pensionable age can be upward sloping
• UK and Denmark

50
CONCLUSIONS Retirement

• Accounting for the value of household work yields
  on average 40 percent higher net replacement
  rates and lower option values
• Continental Europe (Germany) high replacement
  rates and flat option value curves after the
  pensionable age
• Small countries are categorized to those with
• High replacement rate and flat option value
  curve after pensionable age (Finland and
  Portugal)
• High replacement rate and downward sloping
  option value curve after pensionable age (Belgium
  and Netherlands)
• Low replacement rate and rising option value
  curve after pensionable age (Denmark, UK)

51
CONCLUSIONS Retirement

• Observed and optimal ages of withdrawal from work
  
• The optimal path for retirement quite close to
  the actual average retirement in countries with
  low option values such as Belgium and Netherlands
  (men).
• In some countries with high option values for
  continuing to work (Germany, Finland), pension
  wealth can, however, be high

52
CONCLUSIONS Time use

• Differences in time use with respect to
  employment status, gender, and country
• Non-employed spend more time at home and supply
  more household work than the employed
• In most countries changes in domestic work after
  retirement are greater for men than for women
• Individual characteristics, such as the health
  status and own perception of life expectancy,
  arguably have a strong effect
• The unhealthy smaller increase in domestic work
  after retirement, shorter life expectancy, less
  effect on option value?

53
OPTION VALUES at age 55 (1000 )
54
OPTION VALUES at age 55 (1000 )
55

• For more information or a copy of the paper,
  please contact
• hannu.piekkola_at_etla.fi
• or
• liisa.harmoinen_at_etla.fi

56
WELFARE REGIMES
Table 3. Welfare regimes according to
Esping-Andersen and Kosonen
Esping-Andersen, Gøsta (1990). The Three Worlds
of Welfare Capitalism. Polity Press,
Cambridge Kosonen, Pekka (1994). European
Integration A Welfare State Perspective.
University of Helsinki Sociology of Law Series
No. 8. Yliopistopaino, Helsinki
57
WELFARE REGIMES time use approach
Table 4. Three alternative service economies
according to Gershuny time-use perspective
Gershuny, Jonathan (2000). Changing Times. Oxford
University Press, Oxford.
58
TOTAL WORK WELFARE REGIMES
Table 11. Total work by gender, employment
status, and welfare regime, 45-64 year-olds
(hours per week)
59
TOTAL WORK Welfare Regimes
Table x. Total work and female share of total
work by welfare regime
60
DOMESTIC WORK Welfare Regimes
Table x. Domestic work and female share of
domestic work by welfare regime
61
PAID WORK Welfare Regimes
Table x. Paid work and female share of paid work
by welfare regime
62
EMPLOYMENT WELFARE REGIMES
Table 12. Employment rates by country and welfare
regime, 45-64 year-olds (reported time-use in
paid work).
63
TIME-GEOGRAPHICAL AUTHORITY CONSTRAINT AND
ALLOCATION OF TIME BETWEEN WORKPLACE AND HOME
MASTERS THESIS MAY 6, 2003 Liisa Harmoinen
64
OBJECTIVES AND RESEARCH QUESTIONS

• Ageing, Health, and Retirement in Europe
  project, ETLA
• Labour market behaviour and time use of the
  ageing

Are there differences in time use btw. groups and
countries? Effect of the value of household work
on labour market behaviour?
65
THEORETICAL FRAMEWORK
Time Geography Constraints to activity
Economic Theory Allocation of time
Allocation of time between workplace and home
Welfare Regimes Social security
66
TIME GEPGRAPHY AND CONSTRAINTS TO ACTIVITY

• Capability constraint
• Physiological and physical characteristics and
  needs
• Coupling constraint
• Places, resources, other people
• Authority constraint
• Legislation, authorities, rules, norms

67
AUTHORITY CONSTRAINT Welfare Regimes

• Nordic Regime
• Finland, Denmark
• Continental Regime
• Belgium, Germany, Netherlands
• British Regime
• United Kingdom
• Peripheric Regime
• Portugal

68
WELFARE REGIMES
69
WELFARE REGIMES Time Use Approach
70
STRUCTURALISTIC VIEW OF THE SOCIETY
71
MAIN RESULTS

• Non-employed and women supply more household work
  than employed and men
• Value of household work affects incentive to
  retire
• Higher net replacement rate
• Lower option value
• Differences between countries in functioning of
  authority constraint, some core periphery
  consistency

72
HOUSEHOLD WORK
73
PAID WORK
74
NET REPLACEMENT RATES
75
OPTION VALUES Belgium
76
OPTION VALUES Denmark
77
OPTION VALUES Finland
78
TIME USE AND WELFARE REGIMES