Creating Capabilities
James Heckman
University of Chicago
and Human Capital and Economic Opportunity Working Group,
Institute For New Economic Thinking
RIETI
October 8th, 2014
This draft, September 23, 2014 5:11pm
Creating Capabilities, September 23, 2014 5:11pm
2 / 122
.45
.5
Figure 1: Trends in Family Income Inequality in Japan
nsfers
d Tra
an
Taxes
.35
.4
Before
sfers
.25
.3
After Taxes and Tran
1985
1990
1995
Before Taxes and Transfers
After Taxes and Transfers
Year
2000
2005
2010
OLS Slope=0.005, p-value=0.000
OLS Slope=0.001, p-value=0.043
Source: Bradley J. Setzler (2014)
Creating Capabilities, September 23, 2014 5:11pm
3 / 122
Measures of Wealth Inequality (NSFIE)
0.66
Equalized Net Financial Wealth (GINI)
0.64
0.62
0.6
0.58
0.56
0.54
0.52
1984
1989
1994
1999
2004
Source: Lise etal., 2014
Data: National Survey of Family Income and Expenditure (NSFIE)
Source: Lise et al., 2010
Creating Capabilities, September 23, 2014 5:11pm
4 / 122
Poverty Rates in Japan
70
60
50
40
30
20
10
0
1985
1988
Poverty rate (%)
1991
1994
Child poverty rate (%)
1997
2000
2003
2006
2009
2012
Poverty Rate Single Parent (18-65) with Children Under 17
Source: Comprehensive Survey of Living Conditions
Note: Poverty rate is calculated according to OECD guideline ( The poverty rate is the ratio of the number of people who fall below the poverty
line and the total population; the poverty line is here taken as half the median household income)
Creating Capabilities, September 23, 2014 5:11pm
5 / 122
Figure 2: Percentage distribution of the number of Japanese households
by self-assessed living-conditions, for all the households and specific
households; Figure
2010
19: Percentage distribution of the number of households by self-assessed living-conditions, for all
the households and specific households
2010
Very comfortable
Very difficult
All households
Difficult
(59.4%)
Somewhat difficult
27.1
Somewhat comfortable
Normal
32.3
4.1
35.8
0.7
(51.5%)
Aged households
21.3
4.0
30.2
44.0
0.5
(65.7%)
Households with
children
3.8
31.0
34.7
0.3
30.1
(85.6%)
Mother-child
households
0.5
50.5
0%
10%
20%
35.1
30%
40%
50%
60%
70%
13.9
80%
90%
100%
Source: FromFrom
Ministry of Health,
Labour, and Welfare
Source
Ministry
of Health, Labour, and Welfare.
Creating Capabilities, September 23, 2014 5:11pm
6 / 122
A Basic Fact on Gender Inequality in Hourly Wage in Japan
Table 1: Gender Inequality in Employment Status and Hourly Wage
(2006 Wage Census in Japan)
Source Yamaguchi (2014).
Creating Capabilities, September 23, 2014 5:11pm
7 / 122
Intergenerational Mobility and Inequality:
The “Gatsby Curve”
IGE: ln Y1 = α + βln Y0 + ε
|{z}
|{z}
Income of
parents
Italy
UK
USA
France
Japan
Germany
NZ
Sweden
Canada
Australia
Norway
Finland
Denmark
0
Intergen. Elasticity of Earnings (IGE)
.1
.2
.3
.4
.5
Income in current
generation
.2
.25
.3
Gini Coefficient, After Taxes and Transfers
Corak's Chosen IGE
.35
OLS Slope=2.05, p-value=.001
Source: Bradley J. Setzler (2014)
Note:
Inequality is measured after taxes and transfers.
Gini index defined on household income.
IGE measured by pre-tax and transfer income of individual fathers and sons.
•
•
•
Creating Capabilities, September 23, 2014 5:11pm
8 / 122
.5
Figure 3: Changing to Pretax Family Income
USA
Intergen. Elasticity of Earnings (IGE)
.1
.2
.3
.4
Italy
UK
France
Japan
NZ
Canada
Norway
Germany
Australia
Sweden
Finland
0
Denmark
.34
.38
.42
Gini Coefficient, Before Taxes and Transfers
Corak's Chosen IGE
.46
OLS Slope=1.36, p-value=.093
Creating Capabilities, September 23, 2014 5:11pm
9 / 122
The traditional approach to inequality is “alms to the poor”
or “redistribution.”
Creating Capabilities, September 23, 2014 5:11pm
10 / 122
Skills are Major Determinants of Inequality
Creating Capabilities, September 23, 2014 5:11pm
11 / 122
• A Strategy Based on Creating Capabilities:
Capabilities are “Capacities” or “Skills”
• Capabilities: The Capacities to Act and to Create
Future Capacities
Capabilities are defined as the real f reedoms people have
to achieve and the beings and doings that they value and
have reason to value
Sen, 1979, 1985, 1992; Nussbaum and Sen, 1993.
Creating Capabilities, September 23, 2014 5:11pm
13 / 122
Eight Broad Lessons from the Recent Research Literature on
Creating Capabilities
Creating Capabilities, September 23, 2014 5:11pm
14 / 122
1. Multiple Skills
Multiple skills vitally affect performance in life across a variety of
dimensions. A large body of evidence shows that cognitive and
noncognitive skills affect labor market outcomes, the likelihood of
marrying and divorcing, the likelihood of receiving welfare, voting,
and health.
2. Gaps in Skills
Gaps in skills between individuals and across socioeconomic groups
open up at early ages for both cognitive and noncognitive skills.
Many measures show near-parallelism during the school years across
children of parents from different socioeconomic backgrounds, even
though schooling quality is very unequal.
Creating Capabilities, September 23, 2014 5:11pm
16 / 122
3. Capabilities Can Be Created by Investment
The early emergence of skill gaps might be interpreted as the
manifestation of genetics: Smart parents earn more, achieve more,
and have smarter children. A body of strong experimental evidence
shows the powerful role of parenting and environments, including
mentors and teachers in shaping skills.
Genes are important, but skills are not solely genetically determined.
The role of heritability is exaggerated in many studies and in
popular discussions. Genes need sufficiently rich environments to
fully express themselves. There is mounting evidence that gene
expression is itself mediated by environments. Epigenetics informs us
that environmental influences are partly heritable.
Creating Capabilities, September 23, 2014 5:11pm
17 / 122
4. Critical and Sensitive Periods in the Technology of Skill
Formation
There is compelling evidence for critical and sensitive periods in the
development of a child. Different capacities are malleable at
different stages of the life cycle. For example, IQ is rank stable after
age 10, whereas personality skills are malleable through adolescence
and into early adulthood. A substantial body of evidence from
numerous disciplines shows the persistence of early life disadvantage
in shaping later life outcomes. Early life environments are important
for explaining a variety of diverse outcomes, such as crime, health,
education, occupation, social engagement, trust, and voting.
5. Family Investments
Gaps in skills by age across different socioeconomic groups have
counterparts in gaps in family investments and environments.
Children from disadvantaged environments are exposed to a
substantially less rich vocabulary than children from more
advantaged families. At age three, children from professional
families speak 50% more words than children from working-class
families and more than twice as many compared to children from
welfare families. There is a substantial literature showing that
disadvantaged children have compromised early environments as
measured on a variety of dimensions. Recent evidence documents
the lack of parenting knowledge among disadvantaged parents.
Parenting styles in disadvantaged families are much less supportive
of learning and encouraging child exploration.
Creating Capabilities, September 23, 2014 5:11pm
19 / 122
6. Resilience and Targeted Investment
Although early life conditions are important, there is considerable
evidence of resilience and subsequent partial recovery. To our
knowledge, there is no evidence of full recovery from initial
disadvantage. The most effective adolescent interventions target the
formation of personality, socioemotional, and character skills through
mentoring and guidance, including providing information. This
evidence is consistent with the greater malleability of personality and
character skills into adolescence and young adulthood. The body of
evidence to date shows that, as currently implemented, many later
life remediation efforts are not effective in improving capacities and
life outcomes of children from disadvantaged environments. As a
general rule, the economic returns to these programs are smaller
compared to those policies aimed at closing gaps earlier.
Creating Capabilities, September 23, 2014 5:11pm
20 / 122
However, workplace-based adolescent intervention programs and
apprenticeship programs with mentoring, surrogate parenting, and
guidance show promising results. They appear to foster character
skills, such as increasing self-confidence, teamwork ability,
autonomy, and discipline, which are often lacking in disadvantaged
youth. In recent programs with only short-term follow-ups,
mentoring programs in schools that provide students with
information that improves their use of capacities have also been
shown to be effective.
Creating Capabilities, September 23, 2014 5:11pm
21 / 122
7. Parent-child/Mentor-child Interactions Play Key Roles in
Promoting Child Learning
A recurrent finding from the family influence and intervention
literatures is the crucial role of child-parent/child-mentor
relationships that “scaffold” the child (i.e., track the child closely,
encourage the child to take feasible next steps forward in his or her
“proximal zone of development,” and do not bore or discourage the
child). Successful interventions across the life cycle share this
feature. The child as an “emergent” system.
Creating Capabilities, September 23, 2014 5:11pm
22 / 122
8. High Returns to Early Investment
Despite the generally low returns to interventions targeted toward
the cognitive skills of disadvantaged adolescents, the empirical
literature shows high economic returns for investments in young
disadvantaged children. There is compelling evidence that
high-quality interventions targeted to the early years are effective in
promoting skills. This is a manifestation of “dynamic
complementarity”.
Creating Capabilities, September 23, 2014 5:11pm
23 / 122
Predistribution, Not Just Redistribution
A Comprehensive Understanding of Capability Formation
Creating Capabilities, September 23, 2014 5:11pm
25 / 122
• To effectively produce capabilities, we should take a more
comprehensive approach to understanding the economics of
skill development.
• Need to formulate policies that clearly recognize what skills
matter, how they are produced and at what stage of the life
cycle it is most productive to invest, and how we should
prioritize public policy toward producing skills.
• Doing so avoids fragmented and often ineffective approaches to
public policy that miss the pervasive importance of skills.
• The skills problem is at the core of many social and economic
problems that plague societies around the world.
Creating Capabilities, September 23, 2014 5:11pm
26 / 122
Fragmented solutions are often not effective.
Creating Capabilities, September 23, 2014 5:11pm
27 / 122
Fragmented Solutions
Creating Capabilities, September 23, 2014 5:11pm
28 / 122
Prevention, not just remediation.
Creating Capabilities, September 23, 2014 5:11pm
29 / 122
The Ingredients of Effective Capability Formation Strategies
Creating Capabilities, September 23, 2014 5:11pm
30 / 122
Modern Understanding of Human Development
Creating Capabilities, September 23, 2014 5:11pm
31 / 122
• The family lives of children are the major producers of
cognitive and socio-emotional skills.
Creating Capabilities, September 23, 2014 5:11pm
32 / 122
• Supplementing the family and its resources, engaging it in
enriching the life of the child, in supporting the child in school,
and in giving sound advice to children, are effective policies. So
are policies that enhance the skills of parents to be parents.
(i) If society intervenes early enough and in a consistent fashion
over the life cycle of a child, it can promote cognitive and
socioemotional capabilities, as well as the health and wellbeing
of children born into disadvantage.
(ii) Through multiple channels, these effects percolate across the
life cycle and across generations.
(iii) For example, high-quality early interventions reduce inequality
by promoting schooling, reducing crime, and reducing teenage
pregnancy.
(iv) They promote health and healthy behaviors.
(v) They also foster workforce productivity.
Creating Capabilities, September 23, 2014 5:11pm
33 / 122
(vi) These interventions have high benefit-cost ratios and rates of
return. They pass efficiency criteria that any social program
should be asked to pass.
(vii) Quality early childhood policies are among the rare social
policies that face no equality-efficiency tradeoff.
(viii) What is fair is also economically efficient.
(ix) Early interventions that build the capability base of children
have much higher economic returns than later remediation and
prevention programs, such as public job training, convict
rehabilitation programs, adult literacy programs, tuition
subsidies, or expenditure on police to reduce crime.
Creating Capabilities, September 23, 2014 5:11pm
34 / 122
• Universal ingredient of all successful interventions–by
families, schools, and mentors in the work place
• “Scaffolding”
• Monitoring and mentoring the child, taking stock of
where they are and taking them to the next step.
Interactions and interplay are at the heart of all
successful skill development approaches.
Creating Capabilities, September 23, 2014 5:11pm
35 / 122
The Importance of Cognitive and Character Skills
Creating Capabilities, September 23, 2014 5:11pm
36 / 122
Table 2: The Big Five Domains and Their Facets: OCEAN
Big Five Personality Factor
Facets (and correlated trait adjective)
Openness to experience
Fantasy (imaginative)
Aesthetic (artistic)
Feelings (excitable)
Actions (wide interests)
Ideas (curious)
Values (unconventional)
Competence (efficient)
Order (organized)
Dutifulness (not careless)
Achievement striving (ambitious)
Self-discipline (not lazy)
Deliberation (not impulsive)
Warmth (friendly)
Gregariousness (sociable)
Assertiveness (self-confident)
Activity (energetic)
Excitement seeking (adventurous)
Positive emotions (enthusiastic)
Trust (forgiving)
Straight-forwardness (not demanding)
Altruism (warm)
Compliance (not stubborn)
Modesty (not show-off)
Tender-mindedness (sympathetic)
Anxiety (worrying)
Hostility (irritable)
Depression (not contented)
Self-consciousness (shy)
Impulsiveness (moody)
Vulnerability to stress (not self-confident)
Conscientiousness
Extraversion
Agreeableness
Neuroticism/Emotional Stability
Creating Capabilities, September 23, 2014 5:11pm
37 / 122
Ever Been in Jail by Age 30, by Ability (Males)
Ever been in jail by age 30, by ability (males)
olarization
.15
rgument
kills
.10
ritical and
ensitive
eriods
Probability
vidence
.05
nvironment
ntuitive
stimates
lustration
.00
0 – 20
21 – 40
41 – 60
61 – 80
81 – 100
Percentile
ummary
Cognitive
Noncognitive
Note: This figure plots the probability of a given behavior associated with moving up in one ability distribution for someone after
integrating out the other distribution. For example, the lines with markers show the effect of increasing noncognitive ability after
integrating the cognitive ability.
Note: This figure plots the probability of a given behavior associated with moving up in one
ability distribution for someone after integrating out the other distribution. For example, the
lines with markers
the Stixrud,
effect and
of increasing
Source:show
Heckman,
Urzua (2006).noncognitive ability after integrating the
cognitive ability.
Source: Heckman, Stixrud, and Urzua (2006).
Creating Capabilities, September 23, 2014 5:11pm
38 / 122
Probability of Being Single With Children
(Females)
Probability of being single with children
olarization
.10
gument
kills
.08
itical and
ensitive
eriods
nvironment
Probability
vidence
.06
.04
tuitive
stimates
ustration
.02
0 – 20
21 – 40
41 – 60
61 – 80
81 – 100
Percentile
ummary
Cognitive
Noncognitive
Note: This figure plots the probability of a given behavior associated with moving up in one ability distribution for someone after
integrating out the other distribution. For example, the lines with markers show the effect of increasing noncognitive ability after
integrating the cognitive ability.
Note: This figure plots the probability of a given behavior associated with moving up in one
ability distribution for someone after integrating out the other distribution. For example, the
lines with markers
theStixrud,
effectand
of Urzua
increasing
Source:show
Heckman,
(2006).noncognitive ability after integrating the
cognitive ability.
Source: Heckman, Stixrud, and Urzua (2006).
Creating Capabilities, September 23, 2014 5:11pm
39 / 122
Probability of being a 4-year college graduate by age 30
Probability and
Confidence Interval (2.5-97.5%)
ii. By Decile of Cognitive Factor
iii. By Decile of Personality
1
1
0.8
0.8
0.6
0.6
0.4
0.4
0.2
0.2
0
1
2
3
4
5
6
Decile
7
8
9
10
0
1
2
3
4
5
6
7
8
9
10
Decile
Notes: The data are simulated from the estimates of the model and our NLSY79 sample. We use the standard convention that higher deciles are associated with higher values of the variable.
The confidence intervals are computed using bootstrapping (200 draws).
Creating Capabilities, September 23, 2014 5:11pm
40 / 122
The GED illustrates the power of non-cognitive skills
Creating Capabilities, September 23, 2014 5:11pm
41 / 122
Figure 4: Distribution of Cognitive Ability by Educational Status (No
College Sample, All Ethnic Groups)
Dropout
GED
Secondary Graduate
Source: Reproduced from Heckman et al. (2011), based on data from the National
Longitudinal Survey of Youth, 1979 (NLSY79).
Creating Capabilities, September 23, 2014 5:11pm
42 / 122
Figure 5: Hourly Wage Differences of GED Recipients and Traditional
Graduates Compared to Uncertified Dropouts—Ages 20–39
(b) Hourly Wage
0
2
4
6
8
10
Annual Earnings
Abil
5−29
Raw Abil
Raw Abil
Raw Abil
Raw Abil
Raw Abil
Raw Abil
30−34
35−39
20−24
25−29
30−34
35−39
.15
) Hours Worked
GED
(d) Employment
H.S.
+/− S.E.
0
.05
.1
Source: Reproduced from Heckman et al. (2014), which uses data from the National
Longitudinal Survey of Youth, 1979 (NLSY79).
Creating Capabilities, September 23, 2014 5:11pm
43 / 122
Figure 6: Distribution of Non-Cognitive Skills by Education Group
Non-Cognitive Skill
Dropout
GED
Secondary Graduate
Source: Reproduced from Heckman et al. (2011), which uses data from the National
Longitudinal Survey of Youth, 1979 (NLSY79).
Creating Capabilities, September 23, 2014 5:11pm
44 / 122
Evidence on Noncognitive Skills for Japan
Creating Capabilities, September 23, 2014 5:11pm
45 / 122
Figure 7: Standardized Regression Coefficient associated with Years of
Schooling in Japan
Source: LEE SunYoun and OHTAKE Fumio, RIETI Discussion Paper Series 14-E-023, May
2014
Note: The figure displays standardized regression coefficient from multivariate of years of
schooling completed on the personality trait and parental education, controlling for age and
age-squared and gender. The darker rectangular bars are the estimates with the control of
parental educational background and the line bars represent robust standard errors.
Creating Capabilities, September 23, 2014 5:11pm
46 / 122
Figure 8: Standardized Regression Coefficient associated with Earnings in
Japan
Source: LEE SunYoun and OHTAKE Fumio, RIETI Discussion Paper Series 14-E-023, May
2014
Note: The figure displays standardized regression coefficient from multivariate of annual
income on the personality trait and ones own educational attainment, controlling for potential
experience and its squared, gender, occupation, type of employment, industry, company size,
and years of work experience at the current work place. The darker rectangular bars are the
estimates with the control of parental educational background and the line bars represent
robust standard errors.
Creating Capabilities, September 23, 2014 5:11pm
47 / 122
Figure 9: Adjusted R-2 associated with Earnings in Japan and the US
Source: LEE SunYoun and OHTAKE Fumio, RIETI Discussion Paper Series 14-E-023, May
2014
Note: Adjusted R 2 s for linear regressions for annual income (log). Total indicates the
Adjusted R 2 when Big 5, total years of schooling, and behavioral characteristics are all
included into the wage equation.
Creating Capabilities, September 23, 2014 5:11pm
48 / 122
• Can these traits be reliably measured?
Creating Capabilities, September 23, 2014 5:11pm
49 / 122
Developing a More Comprehensive Measurement Systems:
A Task-Based Framework for Identifying and Measuring
Skills
Creating Capabilities, September 23, 2014 5:11pm
50 / 122
Personality traits are the relatively enduring patterns of
thoughts, feelings, and behaviors that reflect the tendency
to respond in certain ways under certain circumstances.
(Roberts, 2009, 140)
• All psychological measures are performance on tasks
Creating Capabilities, September 23, 2014 5:11pm
52 / 122
Figure 10: Determinants of Task Performance
Incentives
Effort
Character
Skills
Task
Performance
Cognitive
Skills
Creating Capabilities, September 23, 2014 5:11pm
53 / 122
Figure 11: Decomposing Variance Explained for Achievement Tests and
Grades into IQ and Non-Cognitive Skills: Stella Maris Secondary School,
Maastricht, Holland
Source: Borghans et al. (2011).
Note: Grit is a measure of persistence on tasks (Duckworth et al., 2007).
Creating Capabilities, September 23, 2014 5:11pm
54 / 122
Gaps Open Up Early
Creating Capabilities, September 23, 2014 5:11pm
55 / 122
0
.5
Mean cognitive score
1
Trend in mean by age for cognitive score by maternal education
3
5
College grad
8
18
Age (years)
Some college
HS Grad
Less than HS
Each score standardized within observed sample. Using all observations and assuming data
missing at random. Source: Brooks-Gunn et al. (2006).
Creating Capabilities, September 23, 2014 5:11pm
56 / 122
Rank onbehavior
Anti-Social
AverageAverage
percentilePercentile
rank on anti-social
score, by income
Behavior
Score,
by
Income
Quartile
quartile
ation
55
ent
and
ve
nment
Score Percentile
ce
50
45
40
35
30
e
25
tes
20
tion
4 Yrs
6 Yrs
8 Yrs
Age
ary
Lowest Income Quartile
Second Income Quartile
Third Income Quartile
Highest Income Quartile
10 Yrs
12 Yrs
Lexical Ability of Children by Annual Household Income
35
30
Vocabulary Scores
25
20
15
10
5
3 year olds
4 year olds
< 5million yen
5 to 9 million yen
5 year olds
>9 million yen
Source: recreated from Hamano 2010
Note: Lexical Ability is measured by Picture Vocabulary Test (PVT). PVT measure the
degree of development of language comprehension, in particular, basic “vocabulary
comprehension.” The test is used internationally and asks children to perform such tasks as
choosing the most appropriate picture from a set of four for a word given by the examiner.
Creating Capabilities, September 23, 2014 5:11pm
58 / 122
Figure 12: Mother’s educational attainment and level of academic ability,
Mathematics
Source: Akabayashi et al. (2013).
Creating Capabilities, September 23, 2014 5:11pm
59 / 122
Figure 13: Family income and academic ability, Mathematics
Source: Akabayashi et al. (2013).
Creating Capabilities, September 23, 2014 5:11pm
60 / 122
How to Interpret This Evidence
Creating Capabilities, September 23, 2014 5:11pm
61 / 122
• Evidence on the early emergence of gaps leaves open the
question of which aspects of families are responsible for
producing these gaps.
• Is it due to genes?
• Family environments? Neighborhood and community effects?
• Parenting and family investment decisions?
• The evidence from a large body of research demonstrates an
important role for investments and family and community
environments in determining adult capacities above and beyond
the role of the family in transmitting genes.
• The quality of home environments by family type is highly
predictive of child success.
Creating Capabilities, September 23, 2014 5:11pm
62 / 122
“It is said that heaven does not create one man above or below
another man. Any existing distinction between the wise and the
stupid, between the rich and the poor, comes down to a matter of
education.” – Fukuzawa Yukichi
Creating Capabilities, September 23, 2014 5:11pm
63 / 122
Variation in Family Environments
Creating Capabilities, September 23, 2014 5:11pm
64 / 122
Mothers’ Speech and Child Vocabulary: Hart & Risley, 1995
Children enter school with “meaningful differences” in vocabulary
knowledge.
1. Emergence of the Problem
In a typical hour, the average child hears:
Family
Status
Welfare
Working Class
Professional
Actual Differences in Quantity
of Words Heard
616 words
1,251 words
2,153 words
Actual Differences in Quality
of Words Heard
5 affirmatives, 11 prohibitions
12 affirmatives, 7 prohibitions
32 affirmatives, 5 prohibitions
2. Cumulative Vocabulary at Age 3
Cumulative Vocabulary at Age 3
Children from welfare families:
Children from working class families:
Children from professional families:
500 words
700 words
1,100 words
Creating Capabilities, September 23, 2014 5:11pm
65 / 122
Relation between the Mother's Education Beckground and her
Educational Awareness and Behavior (%)
Junior or Senior High School Graduate
Vocational College or Junior College
University or Graduate School
100
90
80
70
60
50
40
30
20
10
0
Reads picture Takes the child
Provides
Talks to the Many books at Mother reads
story books to
to visit
breakfast for child about the
home
books
child
museums
the child every
news
day
Mother visits
museums
Parents are
aware that
children should
be exposed to
foreign cultures
Source: Hamano 2010
Creating Capabilities, September 23, 2014 5:11pm
66 / 122
0.05
Proportion of Children in Family Type
0.10
0.15
0.20
0.25
0.30
Children Under 18 Living in Single Parent Households by Marital Status of Parent
1976
1980
1984
1988
Divorced
Widowed
1992
1996
Year
2000
2004
2008
2012
Married, Spouse Absent/Separated
Never Married/Single
Source: IPUMS March CPS 1976‐2012
Note: Parents are defined as the head of the household. Children are defined as individuals under 18, living in the household, and the child of the head of household. Children who have been married or are not living with their parents are excluded from the calculation. Separated parents are included in “Married, Spouse Absent” Category
Creating Capabilities, September 23, 2014 5:11pm
67 / 122
Children Under 20 Living in Single Parent Households by Marital
Status of Single Mother (Japan)
8
7
6
Perventage
5
4
3
2
1
0
1998
2001
2004
Divorced
2007
Widowed
2010
2013
Unmarried
Source: Comprehensive Survey of Living Condition
Creating Capabilities, September 23, 2014 5:11pm
68 / 122
Single Parent Households Among Households with Children under 20 in
Japan
8
7
6
Percentage
5
4
3
2
1
0
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
single mother
single father
Source: Comprehensive Survey of Living Condition
Creating Capabilities, September 23, 2014 5:11pm
69 / 122
• Interpreting effects of family income: U.S. / Japan / Denmark
Creating Capabilities, September 23, 2014 5:11pm
70 / 122
College attendance by AFQT and Family Income Quartiles (1997)
Source: Belley and Lochner (2007).
Creating Capabilities, September 23, 2014 5:11pm
71 / 122
Figure 14: In Japan: Family income plays a key role in determining
students’ path following high school graduation
Percent of High School Graduates
Creating Capabilities, September 23, 2014 5:11pm
72 / 122
An Informative Comparison: US vs. Denmark
Creating Capabilities, September 23, 2014 5:11pm
73 / 122
High School Completion and College Attendance by Mother's Education Denmark and US
High School Completion
College Attendance
0.7
0.9
0.8
0.6
0.7
0.5
0.6
0.5
0.4
0.4
0.3
0.3
0.2
0.2
0.1
0.1
0
0
No High School
High School
Denmark
College
No High School
US
High School
Denmark
College
US
Source: Rasmus Landers
Note: Observation Denmark 33,954; CNLSY 3,268
Source:
James Heckman and Rasmus Landersø (2014).
Note: Observation Landersø Denmark 33,954; CNLSY 3,268.
Creating Capabilities, September 23, 2014 5:11pm
74 / 122
Figure 15: High school completion by mother’s education
Source: Rasmus Landersø (2014).
Creating Capabilities, September 23, 2014 5:11pm
75 / 122
Figure 16: College attendance by mother’s education
Source: Rasmus Landersø (2014).
Creating Capabilities, September 23, 2014 5:11pm
76 / 122
• What is the role for capabilities in explaining gaps?
Creating Capabilities, September 23, 2014 5:11pm
77 / 122
Figure 17: Cognitive skills at age 15-16 and parental income / wealth
(a) CNLSY
(b) Denmark
Source: James Heckman and Rasmus Landersø (2014).
Creating Capabilities, September 23, 2014 5:11pm
78 / 122
Figure 18: College attendance and parental income / wealth
(a) CNLSY
(b) Denmark
Source: James Heckman and Rasmus Landersø (2014).
Creating Capabilities, September 23, 2014 5:11pm
79 / 122
Figure 19: High school completion by parental income and wealth — θC ,
θN,C
(a) CNLSY
(b) Denmark
Source: James Heckman and Rasmus Landersø (2014).
Creating Capabilities, September 23, 2014 5:11pm
80 / 122
Figure 20: College attendance by parental income and wealth — θC ,
θN,C
(a) CNLSY
(b) Denmark
Source: James Heckman and Rasmus Landersø (2014).
Creating Capabilities, September 23, 2014 5:11pm
81 / 122
Genes, Biological Embedding of Experience,
and Gene-Environment Interactions
Creating Capabilities, September 23, 2014 5:11pm
82 / 122
DNA methylation and histone acetylation patterns in young and old twins
Methylation patterns in young and old twins
Manel Esteller
Source: Fraga, Ballestar et al. (2005).
Creating Capabilities, September 23, 2014 5:11pm
83 / 122
Interventions
Early childhood interventions targeted to disadvantaged children are
effective in overcoming these gaps.
• They provide evidence against a purely genetic argument.
• These provide supplementary parenting for disadvantaged
children.
• A primary avenue through which they operate is personality and
noncognitive skills.
• Did not boost IQ.
Creating Capabilities, September 23, 2014 5:11pm
84 / 122
Cognitive Evolution through Time, Perry Males
Male Cognitive Dynamics
105
Treatment
100
95
IQ
90
85
Control
80
75
Treatment
79.2
94.9
95.4
91.5
91.1
88.3
88.4
83.7
Control
77.8
83.1
84.8
85.8
87.7
89.1
89.0
86.0
Entry
4
5
6
7
8
9
Age
Creating Capabilities, September 23, 2014 5:11pm
10
85 / 122
Yet the Perry Program has a statistically significant annual rate of
return of around 6%–10% per annum—for both boys and girls—in
the range of the post–World War II stock market returns to equity
in the U.S. labor market, estimated to be 6.9%.
Creating Capabilities, September 23, 2014 5:11pm
86 / 122
• It worked primarily through noncognitive and character
channels.
Early interventions reducing problem behavior lower the probability
of engaging in unhealthy behaviors in adulthood.
Creating Capabilities, September 23, 2014 5:11pm
87 / 122
.8
.4
.6
density
.6
.4
.2
.2
0
0
density
.8
1
1
Figure 21: Mechanisms: Externalizing Behavior, Males
2
2.5
3
3.5
(a) Control
4
4.5
5
2
®
2.5
3
3.5
4
(b) Treatment
4.5
5
®
Data: Perry Preschool Program.
Source: Heckman, Pinto, Savelyev (2013).
Creating Capabilities, September 23, 2014 5:11pm
88 / 122
Figure 1: Decompositions of Treatment Effects,
Males of Treatment Effects, Males
Decomposition
zation
ment
CAT total*, age 14(+)
Employed, age 19 (+)
Monthly Income, age 27 (+)
nce
No tobacco use, age 27 (+)
al and
tive
ds
# of adult arrests, age 27 (-)
onment
ve
ates
ation
mary
Jobless for more than 2 years, age 40 (-)
Ever on welfare (-)
Total charges of viol.crimes with victim costs, age 40, (-)
Total charges of all crimes, age 40 (-)
Total # of lifetime arrests, age 40 (-)
Total # of adult arrests, age 40 (-)
Total # of misdemeanor arrests, age 40 (-)
Total charges of all crimes with victim costs, age 40 (-)
Any charges of a crime with victim cost, age 40 (-)
0
10
20
30
40
50
60
Cognitive Factors
Socio-Emotional State
Personal Behavior
Other Factors
70
80
Creating Capabilities, September 23, 2014 5:11pm
90
100%
89 / 122
Long-Term Health Effects of Perry: Males
The Abecedarian Intervention
Source: Campbell, Conti, Heckman, Moon, Pinto, and Pungello (2014)
Creating Capabilities, September 23, 2014 5:11pm
91 / 122
ABC Health Effects Mid 30s: Males
Data: Abecedarian Program. Source: Campbell et al. (2014)
Creating Capabilities, September 23, 2014 5:11pm
92 / 122
ABC Mechanisms: IQ
Creating Capabilities, September 23, 2014 5:11pm
93 / 122
Attachment, Engagement
Toward a Deeper Understanding of Parenting and Learning
• In both Perry and ABC (and many other interventions) a main
channel of influence is on parent-child interactions.
• Enhanced attachment and engagement of parents.
• This has important implications for how we model family
influence.
Creating Capabilities, September 23, 2014 5:11pm
94 / 122
Dewey:
“Successful schools do what successful parents do”
—Dewey (1915)
Recent analyses would change this paraphrase to:
“Successful interventions to promote capabilities at any
age do what successful parents and mentors do”
Creating Capabilities, September 23, 2014 5:11pm
95 / 122
Mechanisms—producing effects
(a) Information
(b) Changing preferences of parents
(c) Parental response to child’s curiosity and interest induced by
participation in the program
Creating Capabilities, September 23, 2014 5:11pm
96 / 122
Figure 22: Parental Warmth, Perry Preschool
.
.4
Density
.3
.2
.1
0
-2
-1
0
1
Parental Warmth
2
3
Pooled
Control
Treatment
Note: This figure presents the densities –pooled and by treatment status– for a single factor summarizing a set of questions in
the Perry questionnaire attempting to measure how much affection the child gets from the parent(s).
Creating Capabilities, September 23, 2014 5:11pm
97 / 122
Figure 23: Family Conflict, Perry Preschool
.
.8
Density
.6
.4
.2
0
-2
0
2
4
Family Conflict
Pooled
Control
Treatment
Note: this figure presents the densities –pooled and by treatment status– for a single factor summarizing a set of questions in
the Perry questionnaire attempting to measure family conflict in the household.
Creating Capabilities, September 23, 2014 5:11pm
98 / 122
Figure 24: Parental Authority, Perry
.
.4
Density
.3
.2
.1
0
-4
-2
0
2
Parental Authority
Pooled
Control
Treatment
Note: this figure presents the densities –pooled and by treatment status– for a single factor summarizing a set of questions in
the Perry questionnaire attempting to measure how much discipline the child is subject to from the parent(s).
Creating Capabilities, September 23, 2014 5:11pm
99 / 122
Figure 25: Spending per student on pre-primary education was low in
Japan in 2009
Note: The bars show public (bottom part) and private (top part) education spending in US
dollars, adjusted for price level differences across countries, for children too young for primary
school. Annual spending is based on the number of students, calculated on a full-time basis.
Source: OECD (2012f), OECD Education at a Glance 2012.
Creating Capabilities, September 23, 2014 5:11pm
100 / 122
Understanding the Dynamics of Skill Formation:
Skills Beget Skills
Synergisms: Skills Enhance Each Other
Creating Capabilities, September 23, 2014 5:11pm
102 / 122
Creating Capabilities, September 23, 2014 5:11pm
103 / 122
Figure 26: Framework for Understanding Skill Development
Prenatal
Inherited
Traits
Prenatal Investments
Skills
Birth
Skills
Childhood
Skills
Adulthood
Parenting,
Environment, School
Parenting,
Environment, School
Higher Education
Earnings
Crime
Health
Creating Capabilities, September 23, 2014 5:11pm
104 / 122
Static complementarity: Having higher level of a skill boosts
productivity of other skills in investment and performance
Dynamic complementarity: Investing today boosts the skill base
for tomorrow
Dynamic complementarity increases with age
Creating Capabilities, September 23, 2014 5:11pm
106 / 122
Critical and sensitive periods
Creating Capabilities, September 23, 2014 5:11pm
107 / 122
But Early Life Conditions Are Not the Full Story:
Resilience, Recovery, and Repair
Many Later Remediation Efforts Targeted to the Less Able
are Costly and Often Ineffective
But Some Adolescent Policies are Effective Mentoring and
Information has a Powerful Effect
Creating Capabilities, September 23, 2014 5:11pm
110 / 122
The policies that are effective for adolescents provide
mentoring and integrate schooling and work. At the core of
effective mentoring is what is at the core of effective
parenting: attachment, interaction, and trust. Effective
policies focus on developing social and emotional skills,
teaching conscientiousness.
Creating Capabilities, September 23, 2014 5:11pm
111 / 122
Mentoring can be effective—workplace-based intervention
shape noncognitive skills.
Creating Capabilities, September 23, 2014 5:11pm
112 / 122
What about promoting education?
Creating Capabilities, September 23, 2014 5:11pm
113 / 122
Early development is as important as education in promoting
wages, employment, and health.
Creating Capabilities, September 23, 2014 5:11pm
114 / 122
Disparities by Education (Post-compulsory Education)
• Education, Wages, Employment, and Health
Note: Conti and Heckman (2010). Author’s calculations using British Cohort Study, 1970.
Creating Capabilities, September 23, 2014 5:11pm
115 / 122
Schooling promotes cognitive and noncognitive abilities
Creating Capabilities, September 23, 2014 5:11pm
116 / 122
Expected value of te
0
Expected value of test
Expected value of test
Expected value of te
0
0
0
Figure
27: Causal Effect of Schooling−.5 on ASVAB Measures of Cognition
−.5
−.5
Less than 12
12
13 −15
16 or more
Less than 12
12
13 −15
16 or more
Years of completed schooling at test date
Years of completed schooling at test date
0.5
0
−0.5
Expected value of test score, covariates fixed at mean
1
1.5
1
.5
0
−.5
Less than
12 than 12
Less
12
Years ofYears
completed
schooling
at test date
of completed
schooling
at test date
of test score, covariates fixed at mean
(c) (c)
1.5
1.5
1
1
.5
0.5
0
0
−.5
−0.5
13 −15 13 −1516 or more
16 or more
12
f test score, covariates fixed at mean
(b) Math
(b) Knowledge
(b)
1.5
Mean value of test score, covariates fixed at mean
Mean value of test score, covariates fixed at mean
1.5
Expected value of test score, covariates fixed at mean
(a) Paragraph (a)
Comprehension
(a)
−.5
13 −15
16 or more
Less than 12
12
13 −15
16 or more
Less than 12
12
Years of completed schooling at test date
Years of completed schooling at test date
13 −1513 −15 16 or 16
12 12
more
or more
Years of
completed
schooling
at testatdate
Years
of completed
schooling
test date
Less than
Less12
than 12
(d) (d)
1.5
Source: Heckman et al. (2006).
Notes: Mean effect of schooling on components of the ASVAB. The first four
1
1
components are averaged to
create
males with average ability. We standardize
the test scores to have within-sample mean zero and variance one. The model is
estimated using the NLSY79
sample. Solid lines depict average test scores, and
.5
.5
dashed lines, 2.5%–97.5% confidence intervals. Regressors are fixed at means.
Creating Capabilities, September 23, 2014 5:11pm
117 / 122
Expected value of test score
.2
Figure 28: Causal Effect of Schooling on Two0 Measures of
Non-Cognitive Skill
−.2
0.8
0.6
0.4
0.2
0
−0.2
Less than 12
12
13 or more
Years of completed schooling at test date
Less than 12
12
13 or more
Years of completed schooling at test date
(d) Rosenberg Self-Esteem Scale
(a)
Expected value of test score, covariates fixed at mean
Expected value of test score, covariates fixed at mean
(c) Rotter Locus of Control Scale
0.8
0.6
0.4
0.2
0
−0.2
Less than 12
12
13 or more
Years of completed schooling at test date
f test score, covariates fixed at mean
Source: Heckman et al. (2006). (a)
Notes: Effect of.8 schooling on socioemotional scales for males with average ability, with 95% confidence bands. The locus of
control scale is based on the four-item abbreviated version of the Rotter Internal-External Locus of Control Scale. This scale
is designed to measure the extent to which individuals believe they have control over their lives through self-motivation or
.6
self-determination (internal control), as opposed to the extent to which individuals believe that the environment controls their
lives (external control). The self-esteem scale is based on the 10-item Rosenberg Self-Esteem Scale. This scale describes a
degree of approval
.4 or disapproval toward oneself. In both cases, we standardize the test scores to have within-sample mean
zero and variance one, after taking averages over the respective sets of scales. The model is estimated using the NLSY79
sample. Solid lines depict average test scores, and dashed lines, 2.5%–97.5% confidence intervals. Regressors are fixed at
.2
means.
Creating Capabilities, September 23, 2014 5:11pm
118 / 122
Summary
Creating Capabilities, September 23, 2014 5:11pm
119 / 122
Returns to a Unit Yen Invested
Rate of Return to Investment in Human Capital
Prenatal programs
0
Programs targetted toward the earliest years
Preschool programs
Schooling
Job Training
Prenatal
0–3
4–5
School
Post-Schooll
• For the disadvantaged, spending in most societies is almost in
reverse order.
• This diagram and its policy message have to be carefully
digested.
• It presents the rate of return to a unit of investment in
parenting at the beginning of the life of the child.
Creating Capabilities, September 23, 2014 5:11pm
121 / 122
• Predistribution, not just redistribution.
• Prevention, not just remediation.
Creating Capabilities, September 23, 2014 5:11pm
122 / 122